TITLE

RECOMBINANT MICROBIAL CELLS THAT PRODUCE AT LEAST 28%

EICOSAPENTAENOIC ACID AS DRY CELL WEIGHT

This application claims the benefit of U.S. Provisional Application Nos. 61 /740,502 and 61 /740,506, each filed December 21 , 2012, both of which are incorporated herein by reference in their entireties.

FIELD OF THE INVENTION

This invention is in the field of biotechnology. More specifically, this invention pertains to recombinant microbial cells such as recombinant oleaginous yeast cells that can produce at least 28% eicosapentaenoic acid (EPA) as dry cell weight.

BACKGROUND OF THE INVENTION

A variety of different hosts including plants, algae, fungi, stramenopiles and yeast have been and continue to be investigated as means for commercial production of polyunsaturated fatty acids (PUFA). Genetic engineering has demonstrated that the natural abilities of some hosts, even those natively limited to linoleic acid (LA, 18:2 omega-6) or alpha-linolenic acid (ALA, 18:3 omega-3) fatty acid production, can be substantially altered to result in high-level production of various long-chain omega-3/omega-6 PUFAs.

Although the literature reports a number of recent examples whereby various portions of the omega-3/omega-6 PUFA biosynthetic pathway

responsible for eicosapentaenoic acid (EPA) production have been introduced into plants and non-oleaginous yeast, significant efforts have focused on the use of the oleaginous yeast, Yarrowia lipolytica (U.S. Pat. Nos. 7,238,482 and 7,932,077; U.S. Pat. Appl. Publ. Nos. 2009-0093543 and 2010-0317072).

Oleaginous yeast are defined as those yeast that are naturally capable of oil synthesis and accumulation, wherein oil accumulation is at least 25% of the cellular dry weight, or those yeast genetically engineered such that they become capable of oil synthesis and accumulation, wherein oil accumulation is at least 25% of the cellular dry weight. Emphasis has been placed on the development of transgenic oleaginous Y. Iipolytica strains that can produce enhanced amounts of EPA. This focus on EPA production is due in part to the recognized salutary effects of EPA. For example, EPA has been shown to play a role in maintaining brain, retina and cardiovascular health. EPA is also known to have anti-inflammatory properties and may be useful in treating or preventing diseases linked to inflammation, such as cardiovascular disease and arthritis. Thus, the clinical and pharmaceutical value of EPA is well known (U.S. Pat. Appl. Publ. No. 2009-0093543). Similarly, the advantages of producing EPA in microbes using recombinant means, as opposed to producing EPA from natural microbial sources or via isolation from fish oil and marine plankton, are also well recognized. Interest in EPA production in yeast has also been due to the drive to develop sustainable sources of EPA as alternatives to producing EPA from fish, which would help alleviate problems associated with overfishing.

Enhanced EPA production in Y. Iipolytica has been targeted in two general ways. First, attempts have been made to increase the amount of EPA present in the oil produced by Y. Iipolytica. Such oil, which may not necessarily constitute a large percentage of the dry cell weight of Y. Iipolytica biomass, can be purified away from the biomass, then used in EPA dietary supplements and/or used for further concentration for pharmaceutical applications. Attempts have also been made to increase the amount of EPA in the dry cell weight of Y. Iipolytica. This entails trying to (i) increase the level of oil in Y. Iipolytica while also (ii) increasing the amount of EPA present in the oil. The resulting biomass can be used directly in feeding schemes to deliver a high quantity of EPA in the diet while side- stepping issues of oil purification. Of course, such biomass can also serve as a source of oil in EPA supplements and the oil can also be used for further concentration for pharmaceutical applications, requiring less biomass per unit of EPA produced compared to Y. Iipolytica biomass containing a lower amount of oil.

U.S. Pat. Appl. Publ. No. 2010-0317072 discloses a transgenic Y.

Iipolytica strain that produces oil containing 61 .8% by weight EPA of the total fatty acids of the oil. However, this strain contains 26.5% oil on a dry cell weight basis. So, while the EPA content in the oil is high (61 .8%), the EPA content in the disclosed Y. lipolytica strain on a dry cell weight basis is lower at about 16.4%.

A transgenic Y. lipolytica strain is disclosed in U.S. Pat. Appl. Publ. No.

2012-0052537 that produces oil containing 58.7% by weight EPA of the total fatty acids of the oil. This strain contains 38.3% oil on a dry cell weight basis. So, while the EPA content in the oil is high (58.7%), the EPA content in the disclosed Y. lipolytica strain on a dry cell weight basis is lower at about 22.5%.

U.S. Pat. Appl. Publ. No. 2012-0052537 also discloses a transgenic Y. lipolytica strain that produces oil containing 48.3% by weight EPA of the total fatty acids of the oil. On a dry cell weight basis, this strain contains 56.2% oil and an EPA content of about 27.1 %.

These disclosed examples indicate that as improvements are made in developing transgenic Y. lipolytica strains for enhanced EPA and/or oil

production, an inverse correlation arises between the total amount of oil produced and the amount of EPA present in the total fatty acids of the oil.

Strains engineered to produce higher amounts of oil on a dry cell basis generally have lower amounts of EPA as a percentage of the fatty acids in the oil.

Increases in the total amount of EPA produced on a dry cell weight basis have been realized, even though there has been an inverse relationship between oil production and the amount of EPA produced as a percentage of the total fatty acids in oil. Despite this achievement, there is still a need to develop Y. lipolytica strains that can produce greater total amounts of EPA. Achieving this goal will likely entail the development of new strain modifications that enhance the amount of EPA as a percentage of the total fatty acids in oil, while not compromising the total amount of oil produced by the strain.

Polynucleotide sequences encoding the Sou2 sorbitol utilization protein have been identified by others. Information regarding the function of this protein, however, appears to be limited. For example, Jami et al. (2010, Molecular & Cellular Proteomics 9:2729-2744) disclosed that a "probable" Sou2 protein was present in the extra-cellular fraction of the filamentous fungus Penicillium chrysogenum. The characterization of several amino acid sequences in online databases as Sou2 sorbitol utilization protein appears to be based on sequence homology only without the disclosure of functional studies. The amino acid sequence of the Sou2 protein in Candida albicans is about 72% identical to the amino acid sequence of C. albicans Soul protein, which has been disclosed by Janbon et al . (1998, Proc. Natl. Acad. Sci. U.S.A. 95:5150-5155) and Greenberg et al. (2005, Yeast 22:957-969) to be a sorbose reductase required for L-sorbose utilization. However, despite the homology between C. albicans Soul and -2 proteins, Janbon et al. disclosed that the Sou2 protein is not required for sorbose utilization. The roles of the Soul and Sou2 proteins in lipid metabolism, if any, are believed to be unknown.

Studies are disclosed herein detailing the development of Y. lipolytica strains that can produce more than 28% EPA as dry cell weight. The

modifications used to generate such strains included down-regulating a gene encoding Sou2 sorbitol utilization protein.

SUMMARY OF THE INVENTION

In one embodiment, the invention concerns a recombinant microbial cell that produces an oil comprising at least 28 percent eicosapentaenoic acid (EPA) measured as a weight percent of dry cell weight. This cell comprises at least one polynucleotide sequence encoding an acyl-CoA:lysophosphatidylcholine acyltransferase (LPCAT) comprising at least one amino acid mutation in a membrane-bound O-acyltransferase motif, wherein the LPCAT has LPCAT activity, and the polynucleotide encoding LPCAT is operably linked to at least one regulatory sequence.

In a second embodiment, the LPCAT is a Yarrowia lipolytica LPCAT. In a third embodiment, the Yarrowia lipolytica LPCAT comprises mutations at (i) amino acid position 136 changing methionine to a different amino acid, and (ii) amino acid position 389 changing threonine to a different amino acid.

In a fourth embodiment, the recombinant microbial cell further comprises a down-regulation of an endogenous polynucleotide sequence encoding Sou2 sorbitol utilization protein. In a fifth embodiment, this down-regulation is due to a mutation of the endogenous polynucleotide sequence encoding Sou2 sorbitol utilization protein, wherein the mutation is selected from the group consisting of a substitution, deletion and insertion.

In a sixth embodiment, the Sou2 sorbitol utilization protein encoded by the endogenous polynucleotide sequence comprises an amino acid sequence that is at least 95% identical to SEQ ID NO:10.

In a seventh embodiment, the down-regulation of the endogenous polynucleotide sequence encoding Sou2 sorbitol utilization protein decreases the total amount of sugar alcohols produced by the recombinant microbial cell.

In an eighth embodiment, the recombinant microbial cell further

comprises: (a) at least one polynucleotide sequence encoding

phospholipid:diacylglycerol acyltransferase (PDAT), (b) at least one

polynucleotide sequence encoding delta-12 desaturase, and (c) at least one polynucleotide sequence encoding a dihomo-gamma-linolenic acid (DGLA) synthase multizyme; wherein each of the polynucleotide sequences of (a)-(c) is operably linked to at least one regulatory sequence. In a ninth embodiment, the DGLA synthase multizyme encoded by the polynucleotide sequence of (c) comprises a delta-9 elongase linked to a delta-8 desaturase.

In a tenth embodiment, the recombinant microbial cell further comprises:

(a) at least one polynucleotide sequence encoding delta-8 desaturase, (b) at least one polynucleotide sequence encoding malonyl-CoA synthetase (MCS), and (c) at least one polynucleotide sequence encoding acyl- CoA:lysophosphatidic acid acyltransferase (LPAAT); wherein each of the polynucleotide sequences of (a)-(c) is operably linked to at least one regulatory sequence.

In an eleventh embodiment, the oil produced by recombinant microbial cell comprises at least 30 percent EPA measured as a weight percent of the dry cell weight of the host cell. In a twelfth embodiment, the recombinant microbial cell is an oleaginous yeast cell. In a thirteenth embodiment, this oleaginous yeast cell is a Yarrowia cell.

In a fourteenth embodiment, the invention concerns a method of producing a microbial oil comprising eicosapentaenoic acid. This method comprises: a) culturing a recombinant microbial cell of the invention, wherein a microbial oil comprising eicosapentaenoic acid is produced; and b) optionally recovering the microbial oil of step (a).

BIOLOGICAL DEPOSITS

The following biological material has been deposited with the American

Type Culture Collection (ATCC), 10801 University Boulevard, Manassas, VA 201 10-2209, and bears the following designation, accession number and date of deposit. The biological material listed above was deposited under the terms of the

Budapest Treaty on the International Recognition of the Deposit of

Microorganisms for the Purposes of Patent Procedure. The listed deposit will be maintained in the indicated international depository for at least 30 years and will be made available to the public upon the grant of a patent disclosing it. The availability of a deposit does not constitute a license to practice the subject invention in derogation of patent rights granted by government action.

Y. lipolytica strain Y9502 was derived from Y. lipolytica strain Y8412 as described in U.S. Pat. Appl. Publ. No. 2010-0317072, which is incorporated herein by reference. Y. lipolytica strain Z5585 was derived from Y. lipolytica strain Y9502 as described in U.S. Pat. Appl. Publ. No. 2012-0052537, which is incorporated herein by reference.

BRIEF DESCRIPTION OF THE DRAWINGS AND SEQUENCES

Figure 1 : Biosynthetic pathways for producing omega-3 and omega-6 fatty acids in Yarrowia are shown. Figure 2: Diagrammed are (A) the steps of developing Y. Iipolytica strain Y9502 from wild type strain ATCC #20362, and (B) the steps of developing Y. Iipolytica strains Z5627 and Z5585 from strain Y9502. The percent fatty acid (E.g., EPA) values listed under certain strains represent the percentage of the particular fatty acid in the fatty acids of the oil produced by the strain. The percent oil values listed under certain strains represent the oil as a percent of dry cell weight of the strain.

Figure 3: Diagrammed are (A) the steps of developing Y. Iipolytica strain Z6109 from strain Z5627, and (B) the steps of developing Y. Iipolytica strain Z9276 from strain Z5585. The percent EPA values listed under certain strains represent the percentage of the EPA in the fatty acids of the oil produced by the strain. The percent oil values listed under certain strains represent the oil as a percent of dry cell weight of the strain.

Figure 4: Diagrammed are plasmid constructs (A) pYRH55 and (B) pZKSOU2-New.

Figure 5: Genomic DNA sequence (SEQ ID NO:8) containing the Y Iipolytica SOU2 gene locus (GenBank Accession No. XM_503010,

YALI0D18964g) is shown. The Sou2 amino acid coding region (SEQ ID NO:9) is indicated with bold, underlined text. Two adjacent cytosine residues, which are located at positions 2400349 and 2400350 of chromosome D, are shown in large bold font and are just downstream an apparent TATA box (bold). The mutational insertion observed in strain Z3041 impairing Sou2 expression occurred between these two cytosine residues. The sequence indicated between the two opposing triangles was targeted for removal by the 5'- and 3'-homology sequences of plasmid pZKSOU2-New.

Figure 6: Diagrammed is the genetic targeting strategy used to knock-out the endogenous SOU2 gene in Y Iipolytica using construct pZKSOU2-New. The "X"s shown between certain 5'- and 3'- homology arm sequences denotes sites of homologous recombination. The pop-in event resulting from homologous recombination at the 5'-homology arms results in the juxtaposition of a mutated SOU2 allele with the wild type SOU2 allele. Two different pop-out events are shown. The left-hand pop-out event occurs as a result of homologous recombination at the 3'-homology arms of the gene structure formed from the 5'- arm pop-in event. This process results in a mutated SOU2 allele and removal of the wild type allele. The right-hand pop-out event occurs as a result of homologous recombination at the 5'-homology arms. This pop-out event results in the wild type SOU2 allele.

Figure 7: Diagrammed are plasmid constructs (A) pZKADn-SyP298F and (B) pZKMPn-YD58.

Figure 8: Diagrammed are the steps of developing Y. lipolytica strain Z3041 from strain Z1978U. The percent EPA values listed under certain strains represent the percentage of the EPA in the fatty acids of the oil produced by the strain. The percent oil values listed under certain strains represent the oil as a percent of dry cell weight of the strain.

Figure 9: Diagrammed are plasmid constructs (A) pZKT2-ML9DCB and (B) pZKLY-PP2YAP.

Figure 10: Diagrammed is the plasmid construct pY306-N.

Table 1 . Summary of Gene and Protein SEQ ID Numbers

M136N, M136C, M136G, M136H, M136I, M136F,

M136P, M136S, M136T, M136W, M136Y or M136V

mutation in Motif I

Mutant YILPCAT K137X, comprising K137A, K137R,

K137N, K137G, K137H, K137P, K137S, K137T, or 49 K137Y mutation in Motif I (512 aa)

Mutant YILPCAT L138X, comprising L138A, L138N,

L138C, L138G, L138Q, L138H, L138I, L138M, L138F,

L138P, L138S, L138T, L138W, or L138Y mutation in 50 Motif I (512 aa)

Mutant YILPCAT S139X, comprising S139A, S139N,

S139C, S139G, S139H, S139L, S139M, S139F, 51 S139P, S139W, or S139V mutation in Motif I (512 aa)

Mutant YILPCAT S140X, comprising S140N, S140C,

S140H, S140I, S140L, S140F, S140P, S140W, 52 S140Y or S140V mutation in Motif I (512 aa)

Mutant YILPCAT F141X, comprising F141A, F141 N,

F141 G, F141 H, F141 I, F141 M, F141 P, F141 S, 53 F141T, F141W, or F141V mutation in Motif I (512 aa)

Mutant YILPCAT G142X, comprising G142N, G142H,

G142I, G142L, G142M, G142F, G142P, G142T, 54 G142W, G142Y or G142V mutation in Motif I (512 aa)

Mutant YILPCAT W143X, comprising W143A,

W143G, W143H, W143L, W143K, W143P, W143S, 55 W143T or W143V mutation in Motif I (512 aa)

Mutant YILPCAT N144X, comprising N144A, N144R,

N144G, N144H, N144K, N144F, N144P, N144T or 56 N144V mutation in Motif I (512 aa)

Mutant YILPCAT V145X, comprising V145A, V145C,

V145G, V145E, V145H, V145M, V145F, V145P, 57 V145S, V145T, or V145W mutation in Motif I (512 aa)

Mutant YILPCAT Y146X, comprising Y146R, Y146N,

Y146D, Y146G, Y146E, Y146Q, Y146I, Y146L,

Y146M, Y146F, Y146P, Y146W or Y146V mutation in 58 Motif I (512 aa)

Mutant YILPCAT D147X, comprising D147A, D147N,

D147G, D147E, D147Q, D147H, D147F, D147S, or 59 D147T mutation in Motif I (512 aa)

Mutant YILPCAT G148X, comprising G148A, G148N,

G148H, G148L, G148M, G148F, G148S, G148T or 60 G148V mutation in Motif I (512 aa)

Mutant YILPCAT S376X, comprising S376A, S376G,

S376H, S376L, S376F, S376P, S376T or S376V 61 mutation in Motif II (512 aa) Mutant YILPCAT A377X, comprising A377N, A377G,

A377H, A377L, A377F, A377P, A377S, A377T or 62 A377V mutation in Motif II (512 aa)

Mutant YILPCAT F378X, comprising F378A, F378N,

F378C, F378G, F378H, F378L, F378P, F378S, 63 F378T, F378W, or F378Y mutation in Motif II (512 aa)

Mutant YILPCAT T382X, comprising T382A, T382N,

T382G, T382Q, T382H, T382I, T382M, T382P, 64 T382S, T382W or T382Y mutation in Motif II (512 aa)

Mutant YILPCAT R383X, comprising R383A, R383N,

R383D, R383G, R383E, R383Q, R383H, R383I,

R383L, R383K, R383M, R383F, R383P, R383T, 65 R383W or R383V mutation in Motif II (512 aa)

Mutant YILPCAT P384X, comprising P384A, P384R,

P384G, P384H, P384I, P384L, P384K, P384M,

P384F, P384S, P384T, P384W, P384Y or P384V 66 mutation in Motif II (512 aa)

Mutant YILPCAT G385X, comprising G385A, G385N,

G385C, G385G, G385H, G385I, G385L, G385K,

G385M, G385F, G385S, G385T, G385W, G385Y or 67 G385V mutation in Motif II (512 aa)

Mutant YILPCAT Y386X, comprising Y386A, Y386G,

Y386H, Y386L, Y386F, Y386P, Y386S, Y386T or 68 Y386V mutation in Motif II (512 aa)

Mutant YILPCAT Y387X, comprising Y387A, Y387G,

Y387H, Y387L, Y387F, Y387P, Y387S, Y387T, 69 Y387W or Y387V mutation in Motif II (512 aa)

Mutant YILPCAT L388X, comprising L388A, L388G,

L388H, L388P, L388S, L388T, L388W, L388Y or 70 L388V mutation in Motif II (512 aa)

Mutant YILPCAT T389X, comprising T389A, T389C,

T389G, T389H, T389I, T389L, T389M, T389F, T389P, 71 T389S, T389W, T389Y or T389V mutation in Motif II (512 aa)

Mutant YILPCAT F390X, comprising F390A, F390N,

F390C, F390G, F390H, F390L, F390M, F390P, 72 F390S, F390T or F390V mutation in Motif II (512 aa)

Mutant YILPCAT, comprising single mutations in Motif 73 1 and/or Motif II (512 aa)

Mutant YILPCAT, comprising a single mutation in 74 Motif 1 and a single mutation in Motif II (512 aa)

Mutant YILPCAT (M136S_T389C), Y. lipolytica 75 LPCAT containing M136S and T389C mutations (512 aa)

Mutant YILPCAT (M136S_T389S), Y. lipoidica 76 LP CAT containing M136S and T389S mutations (512 aa)

Mutant YILPCAT (M136V_T389C), Y. lipolytica 77

LPCAT containing M136V and T389C mutations (512 aa)

Mutant YILPCAT (N144A_F390S), Y. lipolytica 78

LPCAT containing N144A and F390S mutations (512 aa)

Mutant YILPCAT (G148A_F390S), Y. lipolytica 79

LPCAT containing G148A and F390S mutations (512 aa)

Mutant YILPCAT (G148N_T382I), Y. lipolytica LPCAT 80 containing G148N and T382I mutations (512 aa)

Mutant YILPCAT (G148N_F390S), Y. lipolytica 81

LPCAT containing G148N and F390S mutations (512 aa)

82

MBOAT Motif I of YILPCAT (17 aa)

83

MBOAT Motif II of YILPCAT (15 aa)

Mutant YILPCAT, comprising a mutant Motif I and/or a 84 mutant Motif II (512 aa)

DETAILED DESCRIPTION OF THE INVENTION

All patents, patent applications, and publications cited are incorporated herein by reference in their entirety.

As used herein the term "invention" or "present invention" is intended to refer to all aspects and embodiments of the invention as described in the claims and specification herein and should not be read so as to be limited to any particular embodiment or aspect.

The terms "SOU2", "SOU2 gene" and "endogenous polynucleotide sequence encoding Sou2 sorbitol utilization protein" are used interchangeably herein. The "Sou2 sorbitol utilization protein" (Sou2p) is encoded by the SOU2 gene.

The Sou2 sorbitol utilization protein of Y. lipolytica (SEQ ID NO:10) has about 66% amino acid sequence identity (according to a BLAST alignment) with Candida albicans Sou2 sorbitol utilization protein (SEQ ID NO:30). Thus, a Sou2 sorbitol utilization protein in certain embodiments has at least 60% amino acid sequence identity with SEQ ID NO:30. A Sou2 sorbitol utilization protein may alternatively have at least 65%, 70%, 75%, or 80% amino acid sequence identity with SEQ ID NO:30.

The term "lipids" as used herein refers to any fat-soluble (i.e., lipophilic), naturally-occurring molecule. A general overview of lipids is provided in U.S. Pat. Appl. Publ. No. 2009-0093543 (see Table 2 therein).

The term "oil" as used herein refers to a lipid substance that is liquid at 25 °C; oil is hydrophobic and soluble in organic solvents. In oleaginous organisms, oil constitutes a major part of the total lipids. Oil is composed primarily of triacylglycerols, but may also contain other neutral lipids, phospholipids and free fatty acids. The fatty acid composition in the oil and the fatty acid composition of the total lipids are generally similar; thus, an increase or decrease in the concentration of fatty acids in the total lipids will correspond with an increase or decrease in the concentration of fatty acids in the oil, and vice versa. The terms "oil", "total lipids", "total lipid content", "total fatty acids", and "total fatty acid methyl esters" are used interchangeable herein.

The term "triacylglycerols" (TAG or TAGs) as used herein refers to neutral lipids composed of three fatty acyl residues esterified to a glycerol molecule. TAGs can contain long-chain polyunsaturated and saturated fatty acids, as well as shorter chain unsaturated and saturated fatty acids.

The term "total fatty acids" (TFA or TFAs) as used herein refers to the sum of all cellular fatty acids that can be derivatized to fatty acid methyl esters (FAME or FAMES) by base transesterification of a given sample, which may be biomass or oil, for example. Thus, total fatty acids include fatty acids from neutral lipids (monoacylglycerols, diacylglycerols, TAGs) and polar lipids (e.g.,

phosphatidylcholine, phosphatidylethanolamine).

The term "total lipid content" of cells as used herein refers to a measure of TFAs as a percent of the dry cell weight (DCW) and can be expressed as

"TFAs % DCW"; e.g., milligrams TFA per 100 milligrams of DCW. For example, 50 TFAs % DCW means that 50% of the dry cell weight is lipid or oil. Total lipid content can be approximated as a measure of FAMEs as a percent of the DCW (FAMEs % DCW). The concentration of a fatty acid in the total lipids is expressed herein as a weight percent of TFAs (% TFAs); e.g., milligrams of a given fatty acid per 100 milligrams of TFAs. Unless otherwise specifically stated herein, reference to the percent of a given fatty acid with respect to total lipids or oil is equivalent to the concentration of the fatty acid as % TFAs (e.g., % EPA of total lipids or oil is equivalent to EPA % TFAs). For example, 50% by weight of EPA in the total fatty acids of an oil is expressed as 50 EPA % TFAs.

It can also useful to express the content of a given fatty acid(s) in a cell as its weight percent of the dry cell weight (% DCW). A measure of total EPA production (EPA % DCW), for example, can be determined using the formula: (EPA % TFAs) * (TFAs % DCW)]/100. A measurement of 30% by weight EPA in the dry cell weight, for example, is expressed by 30 EPA % DCW. The content of a fatty acid(s) such as EPA in the dry cell weight can be approximated using the formula: (EPA % FAMEs) * (FAMEs % DCW)]/100.

The terms "lipid profile", "lipid composition", and "fatty acid profile" are used interchangeably herein and refer to the amount of each individual fatty acid contained in the total lipids or oil, wherein the amount is expressed as a wt % of TFAs. The sum of each individual fatty acid present in the mixture should be 100.

The term "oleaginous" as used herein describes those organisms that tend to store their energy source in the form of lipid (Weete, In: Fungal Lipid

Biochemistry, 2nd Ed., Plenum, 1980). An oleaginous microorganism can comprise, or can accumulate or produce, about 25% or more of its dry cell weight as oil (i.e., > 25 TFAs % DCW).

The term "oleaginous yeast" as used herein refers to those

microorganisms classified as yeasts that can produce a high amount of oil.

Examples of oleaginous yeast include, for example, the genera Yarrowia, Candida, Rhodotorula, Rhodosporidium, Cryptococcus, Trichosporon and

Lipomyces. Preferably, an oleaginous yeast can accumulate in excess of about 25% of its dry cell weight as oil. The term "fatty acids" as used herein refers to long-chain aliphatic acids (alkanoic acids) of varying chain lengths, from about C ₁₂ to C ₂₂ . The

predominant chain lengths are between C ₁₆ and C ₂₂ . The structure of a fatty acid is represented by a simple notation system of "X:Y", where X is the total number of carbon (C) atoms in the fatty acid and Y is the number of double bonds.

Additional details concerning the differentiation between "saturated fatty acids" versus "unsaturated fatty acids", "monounsaturated fatty acids" versus

"polyunsaturated fatty acids" (PUFAs), and "omega-6 fatty acids" (n-6) versus "omega-3 fatty acids" (n-3) are provided in U.S. Pat. No. 7,238,482, which is incorporated herein by reference.

The nomenclature used to describe PUFAs herein is given in Table 2. In the "Shorthand Notation" column, the omega-reference system is used to indicate the number of carbons, the number of double bonds and the position of the double bond closest to the omega carbon, counting from the omega carbon. The remainder of Table 2 summarizes the common names of omega-3 and omega-6 fatty acids, abbreviations that will be used throughout the specification, and the chemical name of each compound.

Table 2: Nomenclature of Polyunsaturated Fatty Acids

c/s-5, 8, 1 1 , 14, 17-

Eicosapentaenoic EPA eicosapentaenoic 20:5 n-3

c/s-7, 10, 13, 16-

Docosatetraenoic DTA docosatetraenoic 22:4 n-6

c/s-4, 7, 10, 13, 16-

Docosapentaenoic DPAn-6 docosapentaenoic 22:5 n-6

c/s-7, 10, 13, 16, 19-

Docosapentaenoic DPA docosapentaenoic 22:5 n-3

c/s-4, 7, 10, 13, 16, 19-

Docosahexaenoic DHA docosahexaenoic 22:6 n-3

The term "PUFA biosynthetic pathway" as used herein refers to a metabolic pathway or process that converts oleic acid to omega-6 fatty acids such as LA, EDA, GLA, DGLA, ARA, DTA and DPAn-6 and omega-3 fatty acids such as ALA, STA, ETrA, ETA, EPA, DPA and DHA. This pathway is described in the literature (e.g., U.S. Pat. No. 7,932,077; U.S. Pat. Appl. Publ. No. 2009- 0093543-A1 ). Briefly, a PUFA biosynthetic pathway involves elongation of the carbon chain through the addition of carbon atoms and desaturation of the molecule through the addition of double bonds, via a series of special elongation and desaturation enzymes termed "PUFA biosynthetic pathway enzymes". More specifically, "PUFA biosynthetic pathway enzymes" refer to any of the following enzymes (and genes encoding these enzymes) associated with the biosynthesis of a PUFA, including: delta-4 desaturase, delta-5 desaturase, delta-6

desaturase, delta-12 desaturase, delta-15 desaturase, delta-17 desaturase, delta-9 desaturase, delta-8 desaturase, delta-9 elongase, C14 16 elongase, C16 18 elongase, C18/20 elongase and/or C20/22 elongase. Figure 1 illustrates certain PUFA biosynthetic pathways.

A PUFA biosynthetic pathway may be "engineered" in certain

embodiments. Such a pathway would comprise one or more foreign or heterologous PUFA biosynthetic pathway enzymes. Such enzymes could be expressed in the cell through the introduction of one or more transgenes encoding the enzymes.

The terms "conversion efficiency" and "percent substrate conversion" herein refer to the efficiency by which a particular enzyme, such as a desaturase or elongase, can convert its respective substrate to product. The conversion efficiency is measured according to the following formula:

([product]/[substrate+product]) *100, where "product" refers to the immediate product and all products derived from it. More specifically, since each PUFA biosynthetic pathway enzyme rarely functions with 100% efficiency to convert substrate to product, the final lipid profile of unpurified oils produced in a host cell will typically be a mixture of various PUFAs consisting of the desired omega- 3/omega-6 fatty acid, as well as various upstream intermediary PUFAs.

The term "Ci ₈ to C20 elongation conversion efficiency" refers herein to the efficiency by which C18 20 elongases can convert C ₁₈ substrates (i.e., LA, ALA, GLA, STA, etc.) to C ₂₀ products (i.e., EDA, ETrA, DGLA, ETA, EPA, etc.). These C18 20 elongases can be either delta-9 elongases or delta-6 elongases.

The terms "delta-9 elongation conversion efficiency" and "delta-9 elongase conversion efficiency" herein refer to the efficiency by which delta-9 elongase can convert C ₁₈ substrates (e.g., LA, ALA) to C ₂₀ products (e.g., EDA, ETrA, DGLA, ETA, ARA, EPA). Delta-9 elongase conversion efficiency is referred to herein as

"% Conv." or "d9e CE(%)".

The terms "membrane-bound O-acyltransferase motif and "MBOAT motif are used interchangeably herein. MBOAT motifs are contained in

LPLATs such as LPCAT and play a role in the enzymatic activity of these proteins (Shindou et al ., 2009, Biochem. Biophys. Res. Comm. 383:320-325;

U.S. Pat. No. 7,732,155; U.S. Pat. Appl. Publ. Nos. 2008-0145867 and 2010-

0317882).

The terms "polynucleotide", "polynucleotide sequence", "nucleic acid sequence", "nucleic acid fragment" and "isolated nucleic acid fragment" are used interchangeably herein. These terms encompass nucleotide sequences and the like. A polynucleotide may be a polymer of RNA or DNA that is single- or double- stranded, that optionally contains synthetic, non-natural or altered nucleotide bases. A polynucleotide in the form of a polymer of DNA may be comprised of one or more segments of cDNA, genomic DNA, synthetic DNA, or mixtures thereof. Nucleotides (usually found in their 5'-monophosphate form) are referred to by a single letter designation as follows: "A" for adenylate or deoxyadenylate (for RNA or DNA, respectively), "C" for cytidylate or deoxycytidylate, "G" for guanylate or deoxyguanylate, "U" for uridylate, "T" for deoxythymidylate, "R" for purines (A or G), "Y" for pyrimidines (C or T), "K" for G or T, "H" for A or C or T, "I" for inosine, and "N" for any nucleotide.

The term "isolated" as used herein refers to a polynucleotide or

polypeptide molecule that has been completely or partially purified from its native source. In some instances, the isolated polynucleotide or polypeptide molecule is part of a greater composition, buffer system or reagent mix. For example, the isolated polynucleotide or polypeptide molecule can be comprised within a cell or organism in a heterologous manner.

The term "gene" as used herein refers to a polynucleotide sequence that expresses a protein, and which may refer to the coding region alone or may include regulatory sequences upstream and/or downstream to the coding region (e.g., 5' untranslated regions upstream of the transcription start site of the coding region). A gene that is "native" or "endogenous" refers to a gene as found in nature with its own regulatory sequences; this gene is located in its natural location in the genome of an organism. An "endogenous polynucleotide encoding" a particular protein is an example of an endogenous gene. "Chimeric gene" refers to any gene that is not a native gene, comprising regulatory and coding sequences that are not found together in nature. Accordingly, a chimeric gene may comprise regulatory sequences and coding sequences that are derived from different sources, or regulatory sequences and coding sequences derived from the same source, but arranged in a manner different than that found in nature. A "foreign" or "heterologous" gene herein refers to a gene that is introduced into the host organism by gene transfer. Foreign genes can comprise native genes inserted into a non-native organism, native genes introduced into a new location within the native host, or chimeric genes. The polynucleotide sequences in certain embodiments disclosed herein are heterologous. A

"transgene" is a gene that has been introduced into the genome by a

transformation procedure. A "codon-optimized gene" is a gene having its frequency of codon usage designed to mimic the frequency of preferred codon usage of the host cell.

"Coding sequence" as used herein refers to a DNA sequence that codes for a specific amino acid sequence. "Regulatory sequences" as used herein refer to nucleotide sequences located upstream of the coding sequence's transcription start site, 5' untranslated regions and 3' non-coding regions, and which may influence the transcription, RNA processing or stability, or translation of the associated coding sequence. Regulatory sequences may include promoters, enhancers, silencers, 5' untranslated leader sequence, introns, polyadenylation recognition sequences, RNA processing sites, effector binding sites, stem-loop structures and other elements involved in regulation of gene expression.

A "promoter" as used herein refers to a DNA sequence capable of controlling the expression of a coding sequence or functional RNA. In general, a promoter sequence is 5' upstream of the transcription start site of a gene.

Promoters may be derived in their entirety from a native gene, or be composed of different elements derived from different promoters found in nature, or even comprise synthetic DNA segments. Promoters that cause a gene to be expressed in most cell types at most times are commonly referred to as

"constitutive promoters".

The terms "3' non-coding sequence", "transcription terminator" and

"terminator" as used herein refer to DNA sequences located 3' downstream of a coding sequence. This includes polyadenylation recognition sequences and other sequences encoding regulatory signals capable of affecting mRNA processing or gene expression.

The term "expression" as used herein refers to the transcription from a gene to produce RNA such as messenger RNA (mRNA). This term can also refer to translation of mRNA into a polypeptide.

When used to describe the expression of a gene or polynucleotide sequence, the terms "down-regulation", "disruption", and "inhibition" are used interchangeably herein to refer to instances when the transcription of the polynucleotide sequence is reduced or eliminated. This results in the reduction or elinnination of RNA transcripts from the polynucleotide sequence, which results in a reduction or elimination of protein expression derived from the polynucleotide sequence. Alternatively, down-regulation can refer to instances where protein translation from transcripts produced by the polynucleotide sequence is reduced or eliminated. Alternatively still, down-regulation can refer to instances where a protein expressed by the polynucleotide sequence has reduced activity. The reduction in any of the above processes (transcription, translation, protein activity) in a cell can by about 40%, 50%, 60%, 70%, 80%, 90%, 95%, or 100% relative to the transcription, translation, or protein activity of a suitable control cell. Down-regulation can be the result of a mutation as disclosed herein (e.g., knock-out) or through the use of antisense or RNAi technology, for example.

The terms "control cell" and "suitable control cell" are used

interchangeably herein and may be referenced with respect to a cell in which a particular modification (e.g., over-expression of a polynucleotide, down-regulation of a polynucleotide) has been made (i.e., an "experimental cell"). A control cell may be any cell that does not have or does not express the particular

modification of the experimental cell. Thus, a control cell may be an

untransformed wild type cell or may be genetically transformed but does not express the genetic transformation. For example, a control cell may be a direct parent of the experimental cell, which direct parent cell does not have the particular modification that is in the experimental cell. Alternatively, a control cell may be a parent of the experimental cell that is removed by one or more generations. Alternatively still, a control cell may be a sibling of the experimental cell, which sibling does not comprise the particular modification that is present in the experimental cell. A sibling cell that could serve as a control cell could be a cell in which a plasmid for protein over-expression is inserted, but not expressed, in the sibling cell, whereas the plasmid is expressed in the experimental cell. It is well within the skill in the art to determine if a cell can be a control cell.

The term "increased" as used herein means having a greater quantity, for example a quantity only slightly greater than the original quantity, or for example a quantity in large excess compared to the original quantity, and including all quantities in between. Alternatively, "increased" may refer to a quantity or activity that is at least 1 %, 2%, 3%, 4%, 5%, 6%, 7%, 8%, 9%, 10%, 1 1 %, 12%, 13%, 14%, 15%, 16%, 17%, 18%, 19% or 20% more than the quantity or activity for which the increased quantity or activity is being compared. The terms

"increased", "greater than", and "improved" are used interchangeably herein. The term "increased" can be used to characterize the expression of a

polynucleotide encoding a protein, for example, where "increased expression" can also mean "over-expression".

The term "operably linked" as used herein refers to the association of two or more nucleic acid sequences such that that the function of one is affected by the other. For example, a promoter is operably linked with a coding sequence when it is capable of affecting the expression of that coding sequence. That is, the coding sequence is under the transcriptional control of the promoter. Coding sequences can be operably linked to regulatory sequences in sense or antisense orientation.

The term "recombinant" as used herein refers to an artificial combination of two otherwise separated segments of sequence, e.g., by chemical synthesis or by the manipulation of isolated segments of nucleic acids by genetic engineering techniques. The terms "recombinant", "transgenic", "transformed", "engineered" or "modified for exogenous gene expression" are used interchangeably herein.

The term "transformation" as used herein refers to the transfer of a nucleic acid molecule into a host organism. The nucleic acid molecule may be a plasmid that replicates autonomously, or it may integrate into the genome of the host organism. Host organisms containing the transformed nucleic acid fragments are referred to as "transgenic" or "recombinant" or "transformed" organisms or "transformants".

The terms "microbial cell" and "microbial organism" are used

interchangeably herein and refer to a microorganism capable of receiving foreign or heterologous genes and capable of expressing those genes. A "recombinant microbial cell" refers to a microbial cell that has been recombinantly engineered. The term "expression cassette" as used herein refers to a polynucleotide sequence comprising a promoter and the coding sequence of a selected gene as well as some other regulatory sequences preceding (5' non-coding sequences) and following (3' non-coding sequences) the coding sequence that are required for expression of the selected gene product. Thus, an expression cassette is typically composed of: 1 ) a promoter; 2) a coding sequence (i.e., open reading frame [ORF]); and 3) a terminator that usually contains a polyadenylation site. Different expression cassettes can be transformed into different organisms including bacteria, yeast, plants and mammalian cells, as long as the correct regulatory sequences are used for each host. As used herein, an "open reading frame" refers to a sequence of DNA or RNA that encodes the amino acid sequence of a polypeptide. The open reading frame begins at the translation initiation start codon (ATG) and ends at the codon immediately 5' to the translation termination codon (stop codon).

The terms "sequence identity" or "identity" as used herein with respect to nucleic acid or polypeptide sequences refer to the nucleic acid bases or amino acid residues in two sequences that are the same when aligned for maximum correspondence over a specified comparison window. Thus, "percentage of sequence identity" or "percent identity" refers to the value determined by comparing two optimally aligned sequences over a comparison window, wherein the portion of the polynucleotide or polypeptide sequence in the comparison window may comprise additions or deletions (i.e., gaps) as compared to the reference sequence (which does not comprise additions or deletions) for optimal alignment of the two sequences. The percentage is calculated by determining the number of positions at which the identical nucleic acid base or amino acid residue occurs in both sequences to yield the number of matched positions, dividing the number of matched positions by the total number of positions in the window of comparison and multiplying the results by 100 to yield the percentage of sequence identity.

The Basic Local Alignment Search Tool (BLAST) algorithm, which is available online at the National Center for Biotechnology Information (NCBI) website, may be used, for example, to measure percent identity between or among two or more of the polynucleotide sequences (BLASTN algorithm) or polypeptide sequences (BLASTP algorithm) disclosed herein. Alternatively, percent identity between sequences may be performed using a Clustal algorithm (e.g., ClustalW or ClustalV). For multiple alignments using a Clustal method of alignment, the default values may correspond to GAP PENALTY=10 and GAP LENGTH PENALTY=10. Default parameters for pairwise alignments and calculation of percent identity of protein sequences using a Clustal method may be KTUPLE=1 , GAP PENALTY=3, WINDOW=5 and DIAGONALS

SAVED=5. For nucleic acids, these parameters may be KTUPLE=2, GAP PENALTY=5, WINDOW=4 and DIAGONALS SAVED=4.

Various polypeptide amino acid sequences and polynucleotide

sequences are disclosed herein as features of certain embodiments of the disclosed invention. Variants of these sequences that are at least about 70- 85%, 85-90%, or 90%-95% identical to the sequences disclosed herein may be used in certain embodiments. Alternatively, a variant amino acid sequence or polynucleotide sequence in certain embodiments can have at least 70%, 71 %, 72%, 73%, 74%, 75%, 76%, 77%, 78%, 79%, 80%, 81 %, 82%, 83%, 84%, 85%, 86%, 87%, 88%, 89%, 90%, 91 %, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% identity with a sequence disclosed herein. The variant amino acid sequence or polynucleotide sequence has the same function of the disclosed sequence, or at least about 85%, 86%, 87%, 88%, 89%, 90%, 91 %, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% of the function of the disclosed sequence.

As discussed in the Background section (above), previous work has shown that as improvements are made in developing transgenic Y. lipolytica strains for enhanced EPA and/or oil production, an inverse correlation generally arises between the total amount of oil produced and the amount of EPA present in the total fatty acids of the oil. Strains that have been engineered to produce higher amounts of oil on a dry cell weight basis generally have lower amounts of EPA as a percentage of the fatty acids in the oil. The Examples disclosed herein show that certain modifications in Y.

lipolytica strains maintain or increase total oil content in a manner that does not decrease the EPA content in the total fatty acids of the oil. Interestingly, such modified Y. lipolytica strains are capable of producing oil comprising at least about 28 percent EPA measured as a weight percent of the dry cell weight of each strain (i.e., 28 EPA % DCW).

Thus, one aspect of the disclosed invention is drawn to a recombinant microbial cell that produces an oil comprising at least about 28 percent EPA measured as a weight percent of dry cell weight. The recombinant microbial cell can be a cell of a yeast, mold, fungus, oomycete, bacteria, algae, stramenopile, or protist (e.g., euglenoid). In certain embodiments, the recombinant microbial cell is an oleaginous microbial cell, such as an oleaginous yeast cell. Examples of oleaginous yeast include species of the genera Yarrowia, Candida,

Rhodotorula, Rhodosporidium, Cryptococcus, Trichosporon and Lipomyces. More specific examples of oleaginous yeast include Rhodosporidium toruloides, Lipomyces starkeyii, L. Iipoferus, Candida revkaufi, C. pulcherrima, C. tropicalis, C. utilis, Trichosporon pullans, T. cutaneum, Rhodotorula glutinus and R.

graminis, for example. Examples of fungal cells in certain embodiments include species of the genera Fusarium (e.g., Fusarium lateritium), Mortierella (e.g., Mortierella alpina) and Mucor (e.g., Mucor rouxii and Mucor circinelloides). The microbial cell can be of the genera Entomophthora, Pythium and Porphyridium in other embodiments of the disclosed invention.

A Yarrowia cell can be the oleaginous yeast cell in certain embodiments of the disclosed invention. Examples of Yarrowia cells that can be modified to produce an oil with at least 28% EPA as a percent of dry cell weight include the following wild type Y. lipolytica isolates available from the American Type

Culture Collection (ATCC, Manassas, VA): strain designations ATCC #20362, #8862, #8661 , #8662, #9773, #15586, #16617, #16618, #18942, #18943, #18944, #18945, #201 14, #20177, #20182, #20225, #20226, #20228, #20327, #20255, #20287, #20297, #20315, #20320, #20324, #20336, #20341 , #20346, #20348, #20363, #20364, #20372, #20373, #20383. The recombinant microbial cell in certain embodiments may be one that has been genetically engineered to produce an increased amount of total lipids and/or fatty acids such as PUFAs. For example, a fatty acid or PUFA

biosynthetic pathway, or portion thereof, may be introduced to the organism by inserting coding sequences for certain pathway enzymes, such as fatty acid desaturases and elongases. Examples of PUFA biosynthetic pathways that can be used herein are shown in Figure 1 . One or a combination of the following enzymes may be genetically introduced to the oleaginous yeast cell to provide a PUFA biosynthetic pathway therein: delta-4 desaturase, delta-5 desaturase, delta-6 desaturase, delta-12 desaturase, delta-15 desaturase, delta-17

desaturase, delta-9 desaturase, delta-8 desaturase, delta-9 elongase, CM 16 elongase, C16 18 elongase, C18 20 elongase, C20/22 elongase. One or more of these enzymes may be from a heterologous source. Example PUFA biosynthetic pathways may contain both a delta-9 elongase and delta-8 desaturase (e.g., refer to U.S. Pat. Appl. Publ. No. 201 1 -0055973, herein incorporated by reference), or both a delta-6 desaturase and delta-6 elongase. Alternatively, the recombinant microbial cell may be modified to have increased total lipids and/or PUFA levels by introducing or deleting genes, other than those encoding desaturases or elongases, that regulate fatty acid biosynthesis.

The PUFAs generated by the PUFA biosynthetic pathway expressed by a recombinant microbial cell of the disclosed invention may include omega-3 and/or omega-6 PUFAs; examples of such PUFAs are linoleic acid (LA), gamma- linolenic acid (GLA), dihomo-gamma-linolenic acid (DGLA), arachidonic acid (ARA), alpha-linolenic acid (ALA), stearidonic acid (STA), eicosatetraenoic acid (ETA), eicosapentaenoic acid (EPA), omega-6 docosapentaenoic acid (DPAn-6), omega-3 docosapentaenoic acid (DPAn-3), eicosadienoic acid (EDA),

eicosatrienoic acid (ETrA), docosatetraenoic acid (DTA) and docosahexaenoic acid (DHA). One or more of these PUFAs in the recombinant microbial cell may be produced from a substrate that is endogenously produced by the recombinant microbial cell or exogenously supplied to the recombinant microbial cell. The recombinant microbial cell of the disclosed invention may be a Yarrowia strain containing a PUFA biosynthetic pathway that produces a PUFA in addition to linoleic acid. For example, the Yarrowia cell may contain enzymes for producing alpha-linolenic acid, gamma-linolenic acid, and/or eicosadienoic acid, none of which PUFAs are produced in wild type Yarrowia. Beyond the fatty acid biosynthetic pathway enzymes normally expressed in Yarrowia to generate linoleic acid, a delta-9 elongase, delta-6 desaturase, and/or delta-15 desaturase may be exogenously expressed for production of eicosadienoic acid, gamma-linolenic acid, and/or alpha-linolenic acid, respectively. These three PUFAs can be further modified (desaturation and/or elongation) to produce downstream PUFAs by expressing additional PUFA pathway enzymes.

The recombinant microbial cell used in certain embodiments of the disclosed invention may be one that has been genetically engineered to produce an elevated amount of lipids compared to its wild type form. Examples of such genetically engineered cells are certain Yarrowia strains disclosed in U.S. Pat. Appl. Publ. Nos. 2009-0093543, 2010-0317072 and 2012-0052537, which are herein incorporated by reference.

The recombinant microbial cell in certain embodiments of the disclosed invention comprises or produces an oil comprising at least about 28% EPA measured as a weight percent of the dry cell weight of the microbial cell.

Alternatively, the recombinant microbial cell comprises or produces an oil comprising at least about 30% EPA measured as a weight percent of the dry cell weight of the microbial cell. Still, in other embodiments of the disclosed invention, the recombinant microbial cell produces an oil comprising at least about 28%, 28.5%, 29%, 29.5%, 30%, 30.5%, 31 %, 31 .5%, 32%, 32.5%, 33%, 33.5%, 34%, 34.5%, or 35% EPA measured as a weight percent of the dry cell weight of the microbial cell.

An increase in the level of EPA measured as a weight percent of the dry cell weight (EPA % DCW) of the recombinant microbial cell in certain embodiments may be at least about 1 %, 2%, 3%, 4%, 5%, 6%, 7%, 8%, 9%, 10%, 1 1 %, 12%, 13%, 14%, 15%, 16%, 17%, or 18% over the EPA % DCW of a control cell. In certain embodiments, this increase in EPA % DCW is coupled with a maintenance or increase in oil content (TFAs % DCW) relative to the oil content of a control cell. A maintenance in oil content in certain embodiments refers to less than about a -3%, -2%, -1 %, or 0% change in oil content of the recombinant microbial cell relative to the oil content of a control cell. An increase in oil content in certain embodiments refers to an increase of more than about 3%, 4%, 5%, 6%, 7%, 8%, 9%, or 10% over the oil content of a control cell.

The recombinant microbial cell in certain embodiments of the disclosed invention has an oil content (TFAs % DCW) of at least about 50%, 51 %, 52%,

53%, 54%, 55%, 56%, 57%, 58%, 59%, 60%, 61 %, 62%, 63%, 64%, or 65% of the dry cell weight of the microbial cell.

The recombinant microbial cell in certain embodiments of the disclosed invention has an EPA content in the total fatty acids of the oil (EPA % TFAs) of at least about 50%, 51 %, 52%, 53%, 54%, 55%, 56%, 57%, 58%, 59%, 60%,

61 %, 62%, 63%, 64%, or 65% by weight.

The recombinant microbial cell in certain embodiments of the disclosed invention produces oil that is devoid of gamma-linolenic acid (GLA). By devoid, it is meant that GLA is below the threshold of detection in the oil, or

alternatively, it is meant that the amount of GLA in the oil is less than 0.1 % by weight of the TFAs of the oil.

The recombinant microbial cell in certain embodiments of the disclosed invention has a stearic acid (C18:0) content in the total fatty acids of the oil (18:0 % TFAs) that is less than about 2.5%, 2.4%, 2.3%, 2.2%, 2.1 %, 1 .9%,

1 .8%, 1 .7%, 1 .6%, 1 .5%, 1 .4%, 1 .3%, 1 .2%, 1 .1 %, or 1 .0% by weight of the

TFAs of the oil.

In certain embodiments of the disclosed invention, the dry cell weight of the recombinant microbial cell grown in a culture is at least about 5.5 grams per liter of the culture. Alternatively, the dry cell weight of the recombinant microbial cell grown in a culture is at least 5.0, 5.2, 5.4, 5.6, 5.8, 6.0, 6.2, 6.4, 6.6, 6.8, 7.0, or 7.2 grams per liter of the culture. The dry cell weight of the microbial cell can be measured in certain embodiments by growing the microbial cell in a culture that is nitrogen-limited for a period of about 120 hours. Such a culture may comprise glucose as the only carbon source or as the predominant carbon source (e.g., other carbon sources less than 5% by weight of the culture). The starting inoculum of the culture for determining dry cell weight of the recombinant microbial cell can be from a culture in which the cells have grown to an OD ₆ oo of about 0.3. All the cells of a certain volume (e.g., 6 ml_) of this OD ₆ oo ~0.3 culture are then used to inoculate a volume (e.g., 25 ml_) of a nitrogen-limited culture medium. This culture is then incubated for a period of about 120 hours, after which the dry cell weight of the culture is determined.

An increase in the dry cell weight of the recombinant microbial cell in certain embodiments may be at least about 1 %, 2%, 3%, 4%, 5%, 6%, 7%, 8%, 9%, 10%, 1 1 %, 12%, 13%, 14%, 15%, 16%, 17%, 18%, 19%, 20%, 21 %, 22%, 23%, 24%, 25%, 26%, 27%, 28%, 29%, 30%, 31 %, 32%, 33%, 34%, 35%, 36%, 37%, or 38% over the dry cell weight of a control cell. The dry cell weights can be measured as above, for example, in making this comparison.

The recombinant microbial cell in certain embodiments of the disclosed invention comprises at least one polynucleotide sequence encoding an acyl- CoA:lysophosphatidylcholine acyltransferase (LPCAT) comprising at least one amino acid mutation in a membrane-bound O-acyltransferase (MBOAT) motif, wherein the LPCAT has LPCAT activity. The polynucleotide encoding the LPCAT is operably linked to at least one regulatory sequence. Thus, certain embodiments are drawn to a recombinant microbial cell that produces an oil comprising at least 28 percent eicosapentaenoic acid (EPA) measured as a weight percent of dry cell weight, wherein the cell comprises at least one polynucleotide sequence encoding an LPCAT comprising at least one amino acid mutation in an MBOAT motif, wherein the LPCAT has LPCAT activity.

The term "acyl-CoA:lysophosphatidylcholine acyltransferase" (LPCAT, EC 2.3.1 .23) as used herein refers to an enzyme that catalyzes the following enzymatic reaction: acyl-CoA + 1 -acyl-sn-glycero-3-phosphocholine = CoA + 1 ,2-diacyl-sn-glycero-3-phosphocholine. LPCAT activity has been described in two structurally distinct protein families, namely the LPAAT protein family (Hishikawa et al., 2008, Proc. Natl. Acad. Sci. U.S.A. 105:2830-2835; Intl. App. Publ. No. WO 2004/076617) and the Ale1 protein family (Tamaki et al., Stan I et al., Chen et al., Benghezal et al., Riekhof et al.).

Polynucleotide sequences encoding mutant LPCAT enzymes as disclosed in U.S. Pat. Appl. No. 61 /661 ,623 (incorporated herein by reference) may be used in certain embodiments disclosed herein. The mutant LPCAT is non-naturally occurring.

Membrane-bound O-acyltransferase (MBOAT) motifs are contained in

LPLATs such as LPCAT and play a role in the enzymatic activity of these proteins. Examples of MBOAT motifs that can be mutated in certain

embodiments of the disclosed invention are disclosed in Shindou et al. (2009, Biochem. Biophys. Res. Comm. 383:320-325), U.S. Pat. No. 7,732,155, and U.S. Pat. Appl. Publ. Nos. 2008-0145867 and 2010-0317882, which are incorporated herein by reference. In certain embodiments, either one or two MBOAT motifs of the LPCAT enzyme are mutated. Since the mutant LPCAT has LPCAT activity, the mutation(s) in the MBOAT motif(s) should not significantly reduce the activity of the enzyme.

The mutated LPCAT in certain embodiments of the disclosed invention is a Yarrowia lipolytica LPCAT (YILPCAT, SEQ ID NO:40) that has been mutated. The terms "Motif I" and "Motif II" are used herein to refer to two different

MBOAT motifs of YILPCAT that can be mutated in certain embodiments. Motif I is represented by the amino acid sequence MVLCMKLSSFGWNVYDG (SEQ ID NO:82), which is located at positions 132-148 of SEQ ID NO:40, whereas Motif II is represented by the amino acid sequence SAFWH GTRPGYYLTF (SEQ ID NO:83), which is located at positions 376-390 of SEQ ID NO:40; both these sequences are contained in wild type YILPCAT.

Motif I and/or Motif II of YILPCAT (SEQ ID NO:40) can be mutated in certain embodiments. Alternatively, Motif I and/or Motif II can be mutated and be comprised within an amino acid sequence that is at least 90%, or 95%, identical to SEQ ID NO:40 and has LPCAT activity. The mutations may be, for example, one or more amino acid substitutions, deletions, and/or insertions in Motif I (SEQ ID NO:40 residues 132, 133, 134, 135, 136, 137, 138, 139, 140, 141 , 142, 143, 144, 145, 146, 147, 148) and/or Motif II (SEQ ID NO:40 residues 376, 377, 378, 382, 383, 384, 385, 386, 387, 389, 390). Substitution mutations may be any of those described herein, for example. In alternative

embodiments, mutations in Motif II can be to residues 376 to 378 and 382-390 of SEQ ID NO:40. Preferably, the activity of a mutant LPCAT polypeptide encoded by a polynucleotide in certain embodiments is equal to or greater than the activity of wild type Yl LPCAT (e.g., SEQ ID NO:40). Such activity can be determined by comparing the EPA % TFAs and/or d9e CE(%) in a recombinant microbial cell over-expressing a mutant LPCAT with the EPA % TFAs and/or d9e CE(%) in a control cell.

In the below examples, YILPCAT mutants having equivalent or increased activity were generated by mutating amino acid residues 133, 134, 135, 136, 137, 138, 139, 140, 141 , 142, 143, 144, 145, 146, 147 or 148 within Motif I, thereby demonstrating that only the methionine residue at position 132 of SEQ ID NO:40 may be unable to tolerate variation. Similarly, YILPCAT mutants having equivalent or increased activity were generated by mutating amino acid residues 378, 382, 383, 385, 388, 389 or 390 within Motif II, thereby demonstrating that the serine, alanine, proline and both tyrosines of SEQ ID NO:83 may be unable to tolerate variation. The amino acids at residues 379-381 (i.e., WHG) of SEQ ID NO:40 were not subjected to mutation, since the histidine of other LPCATs corresponding to H380 of YILPCAT has been reported to be a likely active site residue (Lee et al ., 2008, Mol. Biol. Cell 19:1 174-1 184).

Thus, in certain embodiments of the disclosed invention, the mutant LPCAT comprises an amino acid sequence as set forth in SEQ ID NO:84, wherein SEQ ID NO:84 differs from SEQ ID NO:40 (YILPCAT) by at least one amino acid mutation, wherein:

(i) the amino acid mutation is an amino acid substitution at a residue selected from the group consisting of: residue 133, residue 134, residue 135, residue 136, residue 137, residue 138, residue 139, residue 140, residue 141 , residue 142, residue 143, residue 144, residue 145, residue 146, residue 147 and residue 148;

(ii) the amino acid mutation is in an amino acid substitution at a residue selected from the group consisting of: residue 378, residue 382, residue

383, residue 385, residue 388, residue 389 and residue 390; or

(iii) there are at least two amino acid mutations, wherein a first amino acid mutation is an amino acid substitution selected from the group set forth in part (i), and the second amino acid mutation is an amino acid substitution selected from the group set forth in part (ii).

A mutant YILPCAT in certain embodiments can comprise an amino acid sequence as set forth in SEQ ID NO:73, wherein SEQ ID NO:73 differs from SEQ ID NO:40 by at least one amino acid mutation selected from the group consisting of: V133C, L134A, L134C, L134G, C135F, C135D, C135I, M136T, M136N, M136G, M136P, M136S, M136V, K137N, K137G, K137H, K137Y,

L138G, L138I, L138N, L138A, L138H, L138M, S139G, S139N, S139L, S139W, S140Y, S140I, S140N, S140H, S140P, S140W, F141 V, F141A, F141 M, F141 W, G142I, G142V, G142H, W143H, W143L, N 144A, N144K, N144F, N144T, N144V, V145A, V145G, V145E, V145M, V145F, V145W, Y146G, Y146L, Y146M, D147E, D147N, D147Q, D147H, G148V, G148A, G148N,

F378Y, T382Y, T382I, T382P, R383A, R383M, L388H, L388T, L388G, L388Y, T389A, T389C, T389S, F390C, F390G, F390N, F390T and F390S.

A mutant YILPCAT in certain embodiments can comprise a mutated MBOAT motif. Examples of mutated MBOAT motifs are mutated variants of motifs I (SEQ ID NO:82) and II (SEQ ID NO:83). In certain embodiments, a YILPCAT comprises a Motif I having one or more of the following amino acid substitutions: V2C, L3A, L3C, L3G, K6H, K6G, K6N, K6Y, L7A, L7N, L7G, L7H, L7I, L7M, D16Q, D16N, D16H, G17A, G17V and G17N; and/or a Motif II having one or more of the following amino acid substitutions: F15N, F15C, F15G and F15T. The mutated YILPCAT in certain embodiments of the disclosed invention comprises mutations at (i) amino acid position 136 changing methionine to a different amino acid, and (ii) amino acid position 389 changing threonine to a different amino acid. An example of such a mutated YILPCAT is one in which the position 136 methionine is changed to a serine and the position 389 threonine is changed to an alanine (SEQ ID NO:26). Alternatively, the mutated YILPCAT may comprise the amino acid sequence of any one of SEQ ID

NOs:75-81 , which contain other combinations of mutations in both Motifs I and II. Alternatively, a mutated YILPCAT in certain embodiments may comprise an amino acid sequence that is at least about 90%, 91 %, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical to any one of SEQ ID NOs:26, 75-81 (or any other mutant LPCAT disclosed herein) and have LPCAT function (above).

Alternatively still, a mutated YILPCAT may comprise an amino acid sequence that (i) is at least about 90%, 91 %, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical to any one of SEQ ID NOs:26, 75-81 ; (ii) contains both mutations listed in Table 1 for the mutated YILPCAT of (i); and (iii) has LPCAT function (above). For example, a mutated YILPCAT may comprise an amino acid sequence that is at least about 90%, 91 %, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical to SEQ ID NO:26, where the amino acid sequence has a serine at position 136 and an alanine at position 389.

Further regarding mutations in both Motifs I and II, a mutant YILPCAT in certain embodiments can comprise an amino acid sequence as set forth in SEQ ID NO:74, wherein SEQ ID NO:74 differs from SEQ ID NO:40 by at least one of the pairs of mutations set forth in Table 3 (e.g., an L134A mutation in Motif I may be combined with either a T382I mutation, L388G mutation, F390G mutation or F390T mutation in Motif II, thereby generating mutants

L134A_T382I, L134A_L388G, L134A_F390G and L134A_F390T, respectively). Table 3. YILPCAT Double Mutations Demonstrating Equivalent or Improved EPA

% TFAs and/or Equivalent or Improved % Delta-9 Conversion

Notes: Pairs of mutations comprising a first mutation in Motif I and a second mutation in Motif II lacking a superscript (a or b) resulted in equivalent or improved EPA % TFAs and equivalent or improved % Conv.

³ Indicates a pair of mutations comprising a first mutation in Motif I and a second mutation in Motif II that resulted in equivalent or improved EPA % TFAs (but not equivalent or improved % Conv.).

b Indicates a pair of mutations comprising a first mutation in Motif I and a second mutation in Motif II that resulted in equivalent or improved % Conv. (but not equivalent or improved EPA % TFAs). Although certain combinations of LPCAT amino acid mutations are disclosed herein, one of skill in the art would readily recognize that other combinations of the Motif I and Motif II mutations disclosed herein may be combined as well. Accordingly, one or more of the disclosed Motif I mutations may be used in combination with one or more of the disclosed Motif II

mutations in preparing a polynucleotide encoding a mutant LPCAT polypeptide.

In certain embodiments of the disclosed invention, the recombinant microbial cell comprising a polynucleotide sequence encoding an active LPCAT enzyme with at least one amino acid mutation in an MBOAT motif also

comprises:

(a) an amount of at least one long-chain polyunsaturated fatty acid measured as a weight percent of total fatty acids that is at least the same as or greater than the amount produced by a control cell, and/or

(b) a Ci8 to C ₂ 0 elongation conversion efficiency (e.g., delta-9 elongase conversion efficiency or delta-6 elongase conversion efficiency) that is at least the same as or greater than the conversion efficiency of a control cell.

An increase in the amount of the at least one long-chain PUFA (e.g., EPA) measured as a weight percent of total fatty acids of the recombinant cell over- expressing a mutant LPCAT (containing a mutation in Motif I and/or Motif II) may be at least about 1 %, 2%, 3%, 4%, 5%, 6%, 7%, 8%, 9%, 10%, 1 1 %, 12%, 13%, 14%, 15%, 16%, 17%, 18%, 19%, or 20% over the amount of the at least one long-chain PUFA measured as a weight percent of total fatty acids of a control cell.

An increase in the C ₁ 8 to C ₂ 0 elongation conversion efficiency, delta-9 elongase conversion efficiency, and/or delta-6 elongase conversion efficiency of the recombinant cell over-expressing a mutant LPCAT (containing a mutation in Motif I and/or Motif II) may be at least about 1 %, 2%, 3%, 4%, 5%, 6%, 7%, 8%, 9%, 10%, 1 1 %, 12%, 13%, 14%, or 15% over the Ci ₈ to C ₂₀ elongation conversion efficiency, delta-9 elongase conversion efficiency, and/or delta-6 elongase conversion efficiency, respectively, of a control cell. The control cell in certain embodiments may be a wild type or

recombinant cell that corresponds to the recombinant cell, but that does not comprise, or does not over-express, a polynucleotide encoding an active LPCAT enzyme comprising a mutated MBOAT motif. For example, the control cell does not over-express a mutant LPCAT by virtue of not comprising recombinant polynucleotide sequences encoding mutant LPCAT. Also for example, the control cell does not over-express mutant LPCAT by virtue of comprising, but not expressing, a polynucleotide sequence encoding mutant LPCAT. The control cell may be the recombinant cell as it existed before it was modified to over-express a mutant LPCAT polypeptide (i.e., a parent cell), or may be a recombinant cell that has been modified to contain a recombinant polynucleotide encoding mutant LPCAT, but does not over-express the mutant LPCAT polypeptide (e.g., a cell prepared in parallel with the recombinant cell that over-expresses a mutant LPCAT).

The recombinant microbial cell in certain embodiments of the disclosed invention comprises a down-regulation of an endogenous polynucleotide sequence encoding Sou2 sorbitol utilization protein. In other embodiments, the recombinant microbial cell comprises (i) a down-regulation of an endogenous polynucleotide sequence encoding Sou2 sorbitol utilization protein, and (ii) a PUFA biosynthetic pathway. The down-regulation of the endogenous polynucleotide sequence encoding the Sou2 sorbitol utilization protein in certain embodiments increases the lipid content of the recombinant microbial cell and/or decreases the total amount of sugar alcohols produced by the cell.

The Sou2 sorbitol utilization protein is encoded by the SOU2 gene. The terms "SOU2", "SOU2 gene" and "endogenous polynucleotide sequence encoding Sou2 sorbitol utilization protein" are used interchangeably herein.

The Sou2 sorbitol utilization protein (Sou2p) of Y. lipolytica (SEQ ID NO:10) has about 66% amino acid sequence identity (according to a BLAST alignment) with Candida albicans Sou2 sorbitol utilization protein (SEQ ID NO:30). C. albicans Sou2 sorbitol utilization protein was described by Janbon et al. (1998, Proc. Natl. Acad. Sci. U.S.A. 95:5150-5155) as being similar to Soul sorbitol utilization protein (Soul p), but as not being required for sorbose utilization. Soul sorbitol utilization protein has been described by Janbon et al . and Greenberg et al. (2005, Yeast 22:957-969) to be a sorbose reductase required for L-sorbose utilization. The amino acid sequences of C. albicans Soul and Sou2 proteins have about 72% identity with each other. Given this degree of similarity, it was proposed by Greenberg et al. that Sou2 protein "is probably an oxidoreductase which utilizes NADP(H) as a co-factor and overlaps with Soul p in substrate specificity.

Examples of microbial cell Sou2 sorbitol utilization protein sequences that can be down-regulated in certain embodiments include the sequences disclosed in the following GenBank Accession Nos.: P87218 (Candida albicans), EAZ63262 {Scheffersomyces stipitis), CAX42453 {Candida dubliniensis), EHK97934 {Glarea lozoyensis), ZP_08499267 {Enterobacter hormaechei), CCG21852 {Candida orthopsilosis), YP_887907 {Mycobacterium smegmatis), EJT74900 {Gaeumannomyces graminis), EGZ74878 {Neurospora tetrasperma), EFY961 10 {Metarhizium anisopliae), EGX88960 {Cordyceps militaris), EFY85020 {Metarhizium acridum), EGY16179 {Verticillium dahliae), EEH42615 {Paracoccidioides brasiliensis), EHA51546 {Magnaporthe oryzae), EEP78930 {Uncinocarpus reesii), XP_961 192 {Neurospora crassa),

XP_001523181 {Lodderomyces elongisporus), CAK39371 {Aspergillus niger), XP_002556984 {Penicillium chrysogenum), CAG81202 {Yarrowia lipolytica), CAG84844 {Debaryomyces hansenii), CCE43891 {Candida parapsilosis) and ZP_02917979 {Bifidobacterium dentium).

In certain embodiments of the disclosed invention, the down-regulation of the endogenous polynucleotide sequence encoding Sou2 sorbitol utilization protein is due to a mutation of the polynucleotide sequence. This mutation is a substitution, deletion or an insertion in certain embodiments. A deletion in certain embodiments removes (i) one or more nucleotides from an open reading frame encoding the Sou2 sorbitol utilization protein, and/or (ii) one or more nucleotides of a non-protein-coding sequence located within 500 base pairs of the 5'-end of the open reading frame encoding the Sou2 sorbitol utilization protein.

Examples of a deletion in an SOU2 open reading frame are those removing one or two nucleotides, thereby resulting in a frame-shift mutation; the amino acid sequence encoded downstream such a deletion would be different from the endogenous amino acid sequence. One of ordinary skill in the art would understand that other deletions can be made to create a frame- shift mutation (e.g., any deletion removing a number of base pairs that is not divisible by three). Other deletion examples include those removing the entire SOU2 open reading frame or a portion thereof (e.g., a "knock-out" of SOU2). Where a portion of the SOU2 open reading frame is deleted, or removed by virtue of introducing a frame-shift, down-regulation may occur if the deleted amino acids are necessary for proper Sou2 protein function and/or localization. Alternatively, a deletion in the SOU2 open reading frame may affect proper transcription and/or translation of SOU2. In certain embodiments, the deletion in the SOU2 open reading frame is at least 50, 100, 150, 200, 250, 300, 350, 400, 450, 500, 550, 600, 650, 700, 750, or 800 base pairs. The deletion may be made beginning at the first codon or any downstream codon (e.g., a 250-bp deletion could begin at the tenth codon).

Deletions in certain embodiments of the disclosed invention remove portions of the 5'- and/or 3'-regulatory, non-translated sequences of SOU2. Certain deletions may remove sequences from both of these regulatory sequences; such deletions in most instances would remove the entire SOU2 open reading frame. Other deletions may affect one SOU2 regulatory region and the open reading frame (e.g., deletion of certain 5'-regulatory sequence and 5'-end of the open reading frame). Deletions affecting a 5'-regulatory sequence may down-regulate SOU2 by disrupting proper promoter activity, thereby reducing or eliminating SOU2 transcription. Deletions affecting a 3'- regulatory sequence may down-regulate SOU2 by disrupting proper

transcription termination and/or transcript stability. A deletion in certain embodiments removes one or more nucleotides of a non-protein-coding sequence located within 500 base pairs of the 5'-end of the SOU2 open reading frame. Such a deletion removes sequence from the 5'- non-translated region of the SOU2 transcribed sequence and/or the SOU2 promoter, and may down-regulate SOU2 by reducing transcription and/or translation. The deletion in certain embodiments removes base pairs -10 to -1 , -20 to -1 , -30 to -1 , -40 to -1 , -50 to -1 , -60 to -1 , -70 to -1 , -80 to -1 , -90 to -1 , -100 to -1 , -150 to -1 , -200 to -1 , -250 to -1 , -300 to -1 , -350 to -1 , -400 to -1 , -450 to -1 , or -500 to -1 of the non-protein-coding sequence upstream the SOU2 open reading frame, where the -1 position is the nucleotide immediately 5'-adjacent the SOU2 start codon (ATG). A deletion in other embodiments removes about 10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 150, 200, 250, 300, 350, 400, or 450 consecutive base pairs from any one of these aforementioned regions (e.g., 300 consecutive base pairs deleted within the -500 to -1 region).

In other embodiments a deletion removes one or more regulatory elements in the SOU2 promoter such as a TATA box consensus sequence or a TATA-like sequence (Basehoar et al., 2004, Cell 1 16:699-709, which is incorporated by reference). This type of promoter consensus sequence is usually located within 100 base pairs of the non-protein-coding sequence upstream of the SOU2 open reading frame. Where such a deletion is made to the SOU2 gene promoter in a Yarrowia cell, for example, the TATA box consensus could be removed by deleting one or more base pairs of the -79 to - 72 region with respect to the SOU2 start codon (Figure 5).

In those embodiments in which the recombinant microbial cell is a Yarrowia cell, the down-regulated endogenous polynucleotide sequence may encode a Sou2 sorbitol utilization protein that comprises an amino acid sequence that is at least 95% identical to SEQ ID NO:10. In alternative embodiments, the Sou2 sorbitol utilization protein comprises an amino acid sequence that is at least 96%, 97%, 98%, or 99% identical to SEQ ID NO:10, or the amino acid sequence comprises SEQ ID NO:10. In other embodiments, the down-regulated endogenous polynucleotide sequence encoding a Sou2 sorbitol utilization protein comprises a nucleotide sequence that is at least 95% identical to SEQ ID NO:9. The polynucleotide sequence may alternatively comprise a nucleotide sequence that is at least 96%, 97%, 98%, or 99% identical to SEQ ID NO:9, or the polynucleotide sequence comprises SEQ ID NO:9.

Y. lipolytica in the below Examples was genetically modified to have a deletion that removed the first 287 base pairs of an endogenous SOU2 open reading frame, as well as 235 base pairs of non-protein coding sequence immediately upstream the SOU2 start codon (i.e., positions -235 to -1 were deleted). SEQ ID NO:8 (Figure 5) is an example of genomic DNA sequence containing the Y. lipolytica SOU2 gene locus. This sequence contains 1000 base pairs of non-protein coding sequence upstream the SOU2 start codon, 771 base pairs of SOU2 open reading frame (corresponding to SEQ ID NO:9), and 300 base pairs of non-protein coding sequence downstream of the SOU2 stop codon.

Thus, any of the deletions disclosed herein that can be used to down- regulate SOU2 in Yarrowia may be characterized with respect to nucleotide positions in SEQ ID NOs:8 or 9, or a polynucleotide sequence having at least 95%, 96%, 97%, 98%, or 99% identity thereto (accounting for natural sequence variability that might exist across different Yarrowia strains). Such deletions may similarly be characterized with respect to amino acid positions in SEQ ID NO:10, or an amino acid sequence having at least 95%, 96%, 97%, 98%, or 99% identity thereto. For example, a deletion can remove at least nucleotide positions 1 -768 (an entire Yarrowia SOU2 open reading frame removed) or 1 - 287 of SEQ ID NO:9 (i.e., base pairs 1 -287 of a Yarrowia SOU2 open reading frame), or alternatively a deletion can remove at least nucleotide positions 1001 -1768 (an entire Yarrowia SOU2 open reading frame removed) or 1001 - 1287 of SEQ ID NO:8 (i.e., base pairs 1 -287 of a Yarrowia SOU2 open reading frame). As another example, a deletion can remove at least nucleotide positions 501 -1000 of SEQ ID NO:8 (i.e., -500 to -1 with respect to a Yarrowia SOU2 start codon) or 766-1000 (i.e., -235 to -1 with respect to a Yarrowia SOU2 start codon). As yet another example, a deletion can remove at least nucleotide positions 766-1287 of SEQ ID NO:8 (i.e., -235 to +287,

corresponding to the deletion noted in Figure 5).

An insertion in certain embodiments occurs within (i) an open reading frame encoding the Sou2 sorbitol utilization protein, or (ii) a non-protein-coding sequence located within 500 base pairs of the 5'-end of the open reading frame. As used herein, the terms "insertion" and "integration" are used interchangeably herein to refer to one or more consecutive nucleotides inserted into a genetic sequence. If an insertion is in an open reading frame, it will be disrupted during transcription and/or translation. This may result in an altered sequence of amino acids, extra amino acids in a chain, or premature

termination. The newly synthesized protein produced from such a mutated open reading frame may be abnormally short, abnormally long, and/or contain the wrong amino acids, and will most likely not be functional.

In certain embodiments, an insertion in an SOU2 open reading frame adds one or two nucleotides, thereby resulting in a frame-shift mutation; the amino acid sequence encoded downstream such an insertion would be different from the endogenous amino acid sequence. One of ordinary skill in the art would understand that other insertions can be made to create a frame- shift mutation (e.g., any insertion adding a number of base pairs that is not divisible by three). Down-regulation of SOU2 may occur if the amino acids affected by the insertion are necessary for proper Sou2 protein function and/or localization. Alternatively, an insertion in the SOU2 open reading frame may affect proper transcription and/or translation of SOU2. In certain embodiments, the insertion in the SOU2 open reading frame can be of any length that results in down-regulation of the SOU2 gene. Alternatively, the length of the insertion can be at least 50, 100, 150, 200, 250, 300, 350, 400, 450, 500, 550, 600, 650, 700, 750, 800, 850, 900, 950, or 1000 base pairs. The insertion may be made immediately after the first codon or any downstream codon (e.g., a 250-bp insertion immediately after the tenth codon). An insertion in certain embodiments adds one or more nucleotides into a non-protein-coding sequence located within 500 base pairs of the 5'-end of the SOU2 open reading frame. Such an insertion can affect the 5'-non-translated region of the SOU2 transcribed sequence and/or the SOU2 promoter, and may down-regulate SOU2 by reducing transcription and/or translation. The insertion in certain embodiments is within the -10 to -1 , -20 to -1 , -30 to -1 , -40 to -1 , -50 to -1 , -60 to -1 , -70 to -1 , -80 to -1 , -90 to -1 , -100 to -1 , -150 to -1 , -200 to -1 , -250 to -1 , -300 to -1 , -350 to -1 , -400 to -1 , -450 to -1 , or -500 to -1 region of the non-protein-coding sequence upstream the SOU2 open reading frame, where the -1 position is the nucleotide immediately 5'-adjacent the SOU2 start codon (ATG). The insertion in any of these aforementioned regions can be at least 50, 100, 150, 200, 250, 300, 350, 400, 450, 500, 550, 600, 650, 700, 750, 800, 850, 900, 950, or 1000 base pairs, for example (e.g., 300-bp insertion within the -100 to -1 region). Alternatively, the insertion can be of any length that results in down-regulation of the SOU2 gene.

Y. lipolytica in the below Examples was genetically modified to have an insertion between positions -71 to -70 with respect to the ATG start codon of an SOU2 gene. This particular insertion, as well as any of the insertions disclosed herein that can be used to down-regulate SOU2 in Yarrowia, may be

characterized with respect to nucleotide positions in SEQ ID NOs:8 or 9, or a polynucleotide sequence having at least 95%, 96%, 97%, 98%, or 99% identity thereto (accounting for natural sequence variability that might exist across different Yarrowia strains). For example, an insertion between positions -71 to -70 with respect to the ATG start codon of a Yarrowia SOU2 gene can be described as an insertion between nucleotide positions 930 and 931 of SEQ ID NO:8. As another example, an insertion between base pairs 10 and 1 1 of a Yarrowia SOU2 open reading frame can be described as an insertion between nucleotide positions 1010 and 101 1 of SEQ ID NO:8, or alternatively as an insertion between nucleotide positions 10 and 1 1 of SEQ ID NO:9.

Other types of mutations aside from the aforementioned deletions and insertions can be used to down-regulate the endogenous polynucleotide sequence encoding Sou2 sorbitol utilization protein in alternative embodiments of the disclosed invention. For example, one or more point mutations, which exchange a single nucleotide for another (i.e., a nucleotide substitution), may be used. The point mutation may be a transition point mutation (i.e., a purine in place of another purine, or a pyrimidine in place of another pyrimidine) or transversion point mutation (i.e., a purine in place of a pyrimidine, or a pyrimidine in place of a purine). An example transition mutation is where an adenine is in place of a guanine. An example transversion mutation is where an adenine is in place of a cytosine. Any of these mutations may result, for example, in an amino acid substitution that down-regulates the function of the Sou2 protein.

In certain embodiments, the mutation may be a nonsense mutation within the SOU2 open reading frame; such a mutation changes an amino acid codon to a nonsense codon. Depending on the position of the nonsense mutation in the SOU2 open reading frame, the encoded Sou2 sorbitol utilization protein may be truncated at its carboxy terminus by one more amino acids. Such a truncation may remove at least 1 , 5, 25, 50, 75, 100, 125, 150, 175, 200, 225, 250, or 275 amino acids from the carboxy terminus, for example.

The mutation in certain embodiments may be a missense mutation within the SOU2 open reading frame, where a codon is mutated to encode a different amino acid. Such a mutation may down-regulate SOU2 by virtue of reducing or eliminating the wild type function of the Sou2 sorbitol utilization protein. For example, the proper localization and/or enzymatic activity of the Sou2 protein may be impaired.

A mutation in a codon of the SOU2 open reading frame that does not change the amino acid encoded by the codon (i.e., a silent mutation) is not a mutation as described herein that down-regulates SOU2. Nor is a mutation as described herein one that changes the amino acid encoded by a codon to a related amino acid that does not alter the wild type function of the Sou2 protein. Related amino acids in certain embodiments have side groups that share structure and/or charge, and can be grouped as follows: aliphatic (glycine, alanine, valine, leucine, isoleucine), aromatic (phenylalanine, tyrosine, tryptophan), hydroxyl group-containing (serine, threonine), sulfur group- containing (cysteine, methionine), carboxylic acid group-containing (aspartate, glutamate), amide group-containing (asparagine, glutamine), and amino group- containing (histidine, lysine, arginine).

It would be understood by one of ordinary skill in the art that any of the disclosed mutations to the endogenous SOU2 sequence can be determined to constitute a mutation by referring to the endogenous SOU2 sequence in a microbial cell that has not be modified to mutate the endogenous SOU2 sequence. For example, the SOU2 sequence in a modified Y. lipolytica strain can be compared to the endogenous SOU2 sequence of the counterpart wild type Y. lipolytica strain from which the modified strain was derived.

Any of the above deletions and insertions, as well as any other mutation described herein, may be introduced to an endogenous SOU2 sequence of a recombinant microbial cell using any means known in the art. Genetic targeting techniques may be used, for example, such as those described for modifying yeast (Longtine et al., Yeast 14:953-961 ), fungi (Meyer et al., J. Biotechnol. 128:770-775), algae (Zorin et al., Gene, 432:91 -96), and bacteria (Zhong et al., Nucleic Acids Res. 31 :1656-1664). Alternatively, random mutagenesis techniques may be used.

The down-regulation of an endogenous polynucleotide sequence encoding Sou2 sorbitol utilization protein in certain embodiments is a reduction in the transcription and/or translation of the endogenous polynucleotide sequence by at least about 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, 95%, or 100% relative to the transcription and/or translation of a control cell. In other embodiments, the down-regulation of an endogenous polynucleotide sequence encoding Sou2 sorbitol utilization protein is reflected by a reduction of at least about 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, 95%, or 100% in the function (e.g., protein localization and/or enzymatic activity) of the encoded Sou2 sorbitol utilization protein relative to the function of the Sou2 protein in a control cell. The down-regulation of the endogenous polynucleotide sequence encoding the Sou2 sorbitol utilization protein in certain embodiments of the disclosed invention increases the lipid content of the recombinant microbial cell. This increase in lipid content (TFAs % DCW) can be at least about 3%, 4%, 5%, 6%, 7%, 8%, 9%, 10%, 1 1 %, 12%, 13%, 14%, 15%, 16%, 17%, 18%, 19%, 20%, 21 %, 22%, 23%, 24%, 25%, 26%, 27%, 28%, 29%, 30%, 31 %, 32%, 33%, 34%, or 35% over the lipid content of a control cell.

The down-regulation of the endogenous polynucleotide sequence encoding the Sou2 sorbitol utilization protein in certain embodiments of the disclosed invention decreases the total amount of sugar alcohols produced by the microbial cell. The sugar alcohols may comprise arabitol and/or mannitol, for example. The decrease of the total amount of sugar alcohols can be at least about 20%, 30%, 40%, 50%, 60%, or 70% below the total amount of sugar alcohols in a control cell. The decrease of arabitol and/or mannitol can be at least about 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, or 100% below the amount of arabitol and/or mannitol in a control cell. A 100% decrease in certain embodiments represents that arabitol and/or mannitol are below the threshold of detection.

The control cell in certain embodiments has an endogenous

polynucleotide sequence encoding Sou2 sorbitol utilization protein that does not have any of the mutations disclosed herein. Such non-mutation in the control cell can be determined by comparing its polynucleotide sequence encoding Sou2 sorbitol utilization protein with that of a counterpart wild type cell. For example, where the control cell is a particular recombinant Yarrowia cell that has not been modified to have a mutation in an endogenous

polynucleotide sequence encoding Sou2 sorbitol utilization protein, this polynucleotide sequence in the control cell should be the same or very similar to (e.g., containing silent a mutation) the polynucleotide sequence in the counterpart wild type Yarrowia cell from which the control was derived. Other aspects of a control cell that can be used in certain embodiments are described above. A recombinant microbial cell that does not comprise an endogenous polynucleotide sequence encoding Sou2 sorbitol utilization protein should not be considered to comprise a down-regulation of this polynucleotide if the wild type counterpart cell from which the recombinant microbial cell was derived likewise does not comprise an endogenous polynucleotide sequence encoding Sou2 sorbitol utilization protein.

The recombinant microbial cell in certain embodiments of the disclosed invention comprises (a) at least one heterologous polynucleotide sequence encoding phospholipid:diacylglycerol acyltransferase (PDAT), (b) at least one heterologous polynucleotide sequence encoding delta-12 desaturase, and (c) at least one polynucleotide sequence encoding a dihomo-gamma-linolenic acid (DGLA) synthase multizyme. Each of these polynucleotide sequences (a-c) is operably linked to at least one regulatory sequence.

The term "phospholipid:diacylglycerol acyltransferase" (PDAT; EC

2.3.1 .158) as used herein refers to an enzyme that is capable of transferring an acyl group from the sn-2 position of phospholipids such as phosphatidylcholine (PC) and phosphatidylethanolamine (PE) to the sn-3 position of 1 ,2- diacylglycerol (DAG). This reaction results in lysophospholipids such as lysophosphatidylcholine (LPC) and lysophosphatidylethanolamine (LPE).

Although both PDAT and acyl-CoA:diacylglycerol acyltransferases (DGAT; E.C. 2.3.1 .20) are involved in the terminal step of TAG biosynthesis, only PDAT may synthesize TAGs via an acyl-CoA-independent mechanism.

Dahlqvist et al. (2000, Proc. Natl. Acad. Sci. U.S.A. 97:6487-6492) and Oelkers et al. (2000, J. Biol. Chem. 275:15609-15612) were the first to appreciate that TAG synthesis can occur in the absence of acyl-CoA, via the acyl-CoA-independent PDAT enzyme (structurally related to the

lecithin:cholesterol acyltransferase family of proteins). Following this work, U.S. Pat. No. 7,267,976 (incorporated herein by reference) described the cloning, overexpression and knockout of the Y. lipolytica ATCC #90812 gene encoding PDAT (SEQ ID NO:31 herein), which was determined to share 47.1 % amino acid sequence identity with ScPDAT. A single-amino acid insertion variant (SEQ ID NO:15 herein) of this YIPDAT was disclosed in U.S. Pat. Appl. Publ. No. 2012-0052537.

The heterologous polynucleotide sequence encoding PDAT in certain embodiments of the disclosed invention may encode an amino acid sequence comprising a Yarrowia PDAT. Such a PDAT may comprise the amino acid sequence of SEQ ID NO:15 or 31 , for example. Alternatively, a PDAT in certain embodiments may comprise an amino acid sequence that is at least about 90%, 91 %, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical to SEQ ID NO:15 or 31 and have PDAT function (above).

The term "delta-12 desaturase" (D12; EC 1 .14.19.6) as used herein refers to an enzyme that is capable of introducing a carbon-carbon double bond between the 12th and 13th carbons as numbered from the carboxyl end of a fatty acid.

The heterologous polynucleotide sequence encoding delta-12 desaturase in certain embodiments of the disclosed invention may encode an amino acid sequence comprising a delta-12 desaturase as disclosed in U .S. Pat. No.

7,214,491 and U.S. Pat. Appl. Publ. No. 2007-0254299, both of which are incorporated herein by reference. Alternatively, the amino acid sequence may comprise a delta-12 desaturase from Fusarium moniliforme (e.g., SEQ ID

NO:13), Y. lipolytica (e.g., SEQ ID NO:32), Aspergillus nidulans, Magnaporthe grisea, Neurospora crassa, Fusarium graminearium, Aspergillus fumigatus or Aspergillus flavus, all of which are disclosed in U.S. Pat. No. 7,504,259

(incorporated herein by reference). Alternatively, a delta-12 desaturase in certain embodiments may comprise an amino acid sequence that is at least about 90%, 91 %, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical to SEQ ID NO:13 or 32 and have delta-12 desaturase function (above).

The terms "multizyme" and "fusion protein" are used interchangeably in embodiments herein and refer to a single polypeptide having at least two independent and separable enzymatic activities, wherein the first enzymatic activity is preferably linked to the second enzymatic activity (U.S. Pat. Appl. Publ. No. 2008-0254191 , incorporated herein by reference). The multizyme in certain embodiments comprises two independent and separate enzymes. The "linker" between the two independent and separable enzymatic activities may be comprised of a single peptide bond, although the linker may also be comprised of one amino acid residue, such as a proline, or a polypeptide comprising at least one proline. Examples of linkers that can be used in certain embodiments are disclosed as SEQ ID NOs:4-10 in U.S. Pat. Appl. Publ. No. 2008-0254191 .

A "DGLA synthase multizyme" in certain embodiments of the disclosed invention comprises a delta-9 elongase linked to a delta-8 desaturase. The DGLA synthase multizyme can convert LA to DGLA by virtue of containing both delta-9 elongase activity (converts LA to EDA) and delta-8 desaturase activity (converts EDA to DGLA). The DGLA synthase multizyme can also convert ALA to ETA, since its delta-9 elongase activity can convert ALA to ETrA and its delta-8 desaturase activity can convert ETrA to ETA.

Examples of delta-9 elongase amino acid sequences that can be comprised within the DGLA synthase multizyme are disclosed as SEQ ID

NOs:254 (Euglena anabaena D9e, EaD9e), 319 (Euglena gracilis D9e, EgD9e) and 359 (Eutreptiella sp. CCMP389 D9e, E389D9e) in U.S. Pat. Appl. Publ. No. 2008-0254191 . Examples of delta-8 desaturase amino acid sequences that can be comprised within the DGLA synthase multizyme are disclosed as SEQ ID NOs:328 (mutant Euglena gracilis D8, EgD8M), 430 {Euglena anabaena D8, EaD8) and 514 (Tetrueireptia pomquetensis CCMP1491 D8, TpomD8) in U.S. Pat. Appl. Publ. No. 2008-0254191 . Each of these delta-9 elongases and delta-8 desaturases, or a variant thereof having at least about 90%, 91 %, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% amino acid identity thereto and having either delta-9 elongase or delta-8 desaturase function, may be

comprised in the DGLA synthase multizyme, for example.

In certain embodiments, the DGLA synthase multizyme comprises

E389D9eS/EgD8M (SEQ ID NO:17 herein). This multizyme comprises

E389D9e (SEQ ID NO:33 herein) linked to most of the amino acid sequence of EgD8M (SEQ ID NO:19 herein). Specifically, E389D9eS/EgD8M comprises, in the direction of the amino terminus to the carboxy terminus, SEQ ID NO:33 - linker GAG PARP AG L P PATYYDS LAVM GS (SEQ ID NO:34) - positions 2-422 of SEQ ID NO:19. The DGLA synthase multizyme in other embodiments may be any of those disclosed in U.S. Pat. Appl. Publ. No. 2008-0254191 (e.g., EgD9eS/EgD8M, SEQ ID NO:35 herein; EgD9eS/EaD8S; EaD9eS/EgD8M; EaD9eS/EaD8S, SEQ ID NO:36 herein; EgD9e/TpomD8; EaD9e/TpomD8). Alternatively, the DGLA synthase multizyme in certain embodiments may comprise an amino acid sequence that is at least about 90%, 91 %, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical to SEQ ID NO:17, 35, or 36 (or any of the other above DGLA synthases) and have DGLA synthase function (above). The delta-9 elongase is located at the amino-terminus (N-terminus) of the multizyme polypeptide in certain embodiments.

The recombinant microbial cell in certain embodiments of the disclosed invention comprises (a) at least one heterologous polynucleotide sequence encoding delta-8 desaturase, (b) at least one heterologous polynucleotide sequence encoding malonyl-CoA synthetase (MCS), and (c) at least one heterologous polynucleotide sequence encoding acyl-CoA:lysophosphatidic acid acyltransferase (LPAAT). Each of these polynucleotide sequences (a-c) is operably linked to at least one regulatory sequence.

The term "delta-8 desaturase" (D8; EC 1 .14.19.4) as used herein refers to an enzyme that is capable of introducing a carbon-carbon double bond between the 8th and 9th carbons as numbered from the carboxyl end of a fatty acid.

The heterologous polynucleotide sequence encoding delta-8 desaturase in certain embodiments of the disclosed invention may encode an amino acid sequence comprising a delta-8 desaturase sequence as disclosed in U.S. Pat. Appl. Publ. No. 2005-0273885 or U.S. Pat. Nos. 7,550,651 ; 7,256,033;

7,790,156; 7,943,823; 7,863,502; or 6,825,017, all of which are incorporated herein by reference. Alternatively, the amino acid sequence may comprise a delta-8 desaturase from Euglena gracilis (e.g., SEQ ID NO:19) or Euglena anabaena (e.g., SEQ ID NO:37), for example. Alternatively, a delta-8 desaturase in certain embodiments may comprise an amino acid sequence that is at least about 90%, 91 %, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical to SEQ ID NO:19 or 37 and have delta-8 desaturase function (above).

The term "malonyl-CoA synthetase" (MCS, EC 6.2.1 .14) as used herein refers to an enzyme that catalyzes the following enzymatic reaction: malonate + ATP + CoA -> malonyl-CoA + AMP + pyrophosphate. The enzyme was first purified from malonate-grown Pseudomonas fluorescens (1985, Kim and Bang, J. Biol. Chem. 260:5098-5104), although various Rhizobia homologs have since been isolated from bacteroides within legume nodules (e.g., Kim and Chae, 1991 , Biochem. J. 273:51 1 -516; Kim and Kang, 1994 Biochem. J. 297:327-333). By converting malonate into malonyl-CoA, MCS can provide malonyl-CoA substrate for use in fatty acid synthesis. Thus, in addition to reducing the byproduction of malonates in a cell, MCS expression also helps to avoid carbon and energy waste within the cell, reduce the amount of base required to maintain an optimal pH range during the fermentation process, and reduce the amount of byproduct organic acids that require neutralization within the fermentation waste steam.

The heterologous polynucleotide sequence encoding MCS in certain embodiments of the disclosed invention may encode an amino acid sequence comprising an MCS sequence as disclosed in U.S. Pat. Appl. Publ. No. 2010- 0159558, which is incorporated herein by reference. For example, the amino acid sequence may comprise an MCS from Rhizobium leguminosarum (e.g., SEQ ID NO:21 ). Alternatively, an MCS in certain embodiments may comprise an amino acid sequence that is at least about 90%, 91 %, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical to SEQ ID NO:21 and have MCS function (above).

The term "acyl-CoA:lysophospholipid acyltransferase" or

"lysophospholipid acyltransferase" ("LPLAT") herein refers to a broad class of acyltransferases that can acylate a variety of lysophospholipid substrates at the sn-2 position. More specifically, LPLATs include lysophosphatidic acid (LPA) acyltransferase (LPAAT), which can catalyze conversion of LPA to PA; LPC acyltransferase (LPCAT), which can catalyze conversion of LPC to PC; LPE acyltransferase (LPEAT), which can catalyze conversion of LPE to PE; LPS acyltransferase (LPSAT), which can catalyze conversion of LPS to PS; and LPG acyltransferase (LPGAT), which can catalyze conversion of LPG to PG. Various other designations for LPLATs are used in the art. For example, LPAAT has also been referred to as acyl-CoA:1 -acyl-sn-glycerol-3-phosphate 2-O-acyltransferase, 1 -acyl-sn-glycerol-3-phosphate acyltransferase, or 1 - acylglycerolphosphate acyltransferase. LPCAT has also been referred to as acyl-CoA:1 -acyl lysophosphatidylcholine acyltransferase. Certain LPLATs, such as Saccharomyces cerevisiae Ale1 , have broad specificity and thus a single enzyme may be capable of catalyzing several LPLAT reactions, including LPAAT, LPCAT and LPEAT reactions (Tamaki et al., 2007, J. Biol. Chem.

282:34288-34298; Stahl et al., 2008, FEBS Letters 582:305-309; Chen et al., 2007, FEBS Letters 581 :551 1 -5516; Benghezal et al., 2007, J. Biol. Chem. 282:30845-30855; Riekhof et al., 2007, J. Biol. Chem. 282:28344-28352).

The term "lysophosphatidic acid acyltransferase" (LPAAT, EC 2.3.1 .51 ) as used herein refers to an enzyme that catalyzes the following reaction: acyl- CoA + 1 -acyl-sn-glycerol 3-phosphate = CoA + 1 ,2-diacyl-sn-glycerol 3- phosphate.

The heterologous polynucleotide sequence encoding LPAAT in certain embodiments of the disclosed invention may encode an amino acid sequence comprising an LPAAT sequence as disclosed in U.S. Pat. Appl. Publ. No. 2010- 0317882 or U.S. Pat. Nos. 7,189,559 and 7,879,591 , all of which are

incorporated herein by reference. For example, the amino acid sequence may comprise an LPAAT from Mortierella alpina (e.g., SEQ ID NO:38) or Y.

lipolytica (e.g., SEQ ID NO:23). Alternatively, an LPAAT in certain

embodiments may comprise an amino acid sequence that is at least about 90%, 91 %, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical to SEQ ID NO:38 or 23 and have LPAAT function (above).

Thus, a recombinant microbial cell in certain embodiments of the disclosed invention comprises one or more of the following features as described above: (i) at least one polynucleotide sequence encoding an active LPCAT comprising at least one amino acid mutation in a membrane-bound O- acyltransferase motif;

(ii) a down-regulation of an endogenous polynucleotide sequence encoding Sou2 sorbitol utilization protein;

(iii) at least one polynucleotide sequence encoding PDAT;

(iv) at least one polynucleotide sequence encoding delta-12 desaturase;

(v) at least one polynucleotide sequence encoding a DGLA synthase multizyme;

(vi) at least one polynucleotide sequence encoding delta-8 desaturase;

(vii) at least one polynucleotide sequence encoding MCS;

(viii) at least one polynucleotide sequence encoding LPAAT;

wherein each of the polynucleotide sequences of (i) and (iii)-(viii) is operably linked to at least one regulatory sequence. Each of the polynucleotides of (i) and (iii)-(viii) may be heterologous. The recombinant microbial cell in each embodiment may comprise or produce an oil comprising at least 28 percent EPA measured as a weight percent of the dry cell weight. Certain

embodiments of the recombinant microbial cell comprise feature(s):

- (i);

- (ii);

- (i) and (ii);

- (i), (ii), (iii), (iv) and (v);

- (i), (ii), (vi), (vii) and (viii); or

- (i)-(viii).

For example, certain embodiments of the invention that have both features (i) and (ii) refer to recombinant microbial cells (e.g., recombinant Yarrowia cells) that produce an oil comprising at least 28 percent EPA measured as a weight percent of the dry cell weight and that comprise a down-regulation of an endogenous polynucleotide sequence encoding Sou2 sorbitol utilization protein, and at least one polynucleotide sequence encoding an active LPCAT enzyme comprising at least one amino acid mutation in a membrane-bound O-acyltransferase motif. As described in U.S. Pat. Appl. Publ. No. 2010-0317072 (incorporated herein by reference), Y. lipolytica strain Y9502 was derived from strain Y8412, which in turn was derived from wild type strain ATCC #20362. Y. lipolytica strain Z5585 was derived from strain Y9502 as described in U.S. Pat. Appl. Publ. No. 2012-0052537, which is incorporated herein by reference. Certain of the recombinant Y. lipolytica strains disclosed in the below Examples are derived from Z5585, such as the strains listed in Table 13 (e.g. Z9276). Thus, in certain embodiments of the disclosed invention, the recombinant microbial cell is a Y. lipolytica comprising one of, a combination of, or all of the following heterologous features: down-regulation of a Pex protein-encoding polynucleotide (e.g., Pex3), down-regulation of a Sou2p-encoding polynucleotide, at least 3 polynucleotides encoding C16 18 elongase (e.g., ME3), at least 5 polynucleotides encoding delta-9 elongase (e.g., EgD9e), at least 6 polynucleotides encoding delta-8 desaturase (e.g., EgD8, EgD8M, or EaD8), at least 4 polynucleotides encoding DGLA synthase (e.g., E389D9eS/EgD8M, E389D9eS/EgD8M, EgD9eS/EgD8M,

EaD9eS/EaD8S), at least 2 polynucleotides encoding delta-9 desaturase (e.g., YID9), at least 5 polynucleotides encoding delta-12 desaturase (e.g., FmD12), at least 4 polynucleotides encoding delta-5 desaturase (e.g., EgD5M), at least 3 polynucleotides encoding delta-17 desaturase (e.g., PaD17), at least 2 polynucleotides encoding diacylgiycero! cholinephosphotransferase (e.g., YICPT1 ), at least 3 polynucleotides encoding malonyl-CoA synthetase (e.g., MCS), at least 1 polynucleotide encoding choline-phosphate cytidylyl-transferase (e.g., YIPCT), at least 5 polynucleotides encoding acyl-CoA:lysophosphatidic acid acyltransferase (e.g., MaLPAATI or YILPAAT1 ), at least 3 polynucleotides encoding phospholipid:diacylglycerol acyltransferase (e.g., YIPDAT), at least 1 polynucleotide encoding a mutant acyl CoA:lysophosphatidylcholine

acyltransferase (e.g., YILPCAT (M136X/T389X). Alternatively, the cell may comprise one copy of some or all of these polynucleotides.

Constructs or vectors comprising the polynucleotides described herein may be introduced into a host cell by any standard technique. These techniques include transformation (e.g., lithium acetate transformation [Methods in Enzymology, 194:186-187 (1991 )]), biolistic impact, electroporation, and microinjection, for example. As an example, U.S. Patent Nos. 4,880,741 and 5,071 ,764, and Chen et al. (1997, Appl. Microbiol. Biotechnol. 48:232-235), describe integration techniques for Y. Iipolytica, based on linearized fragments of DNA.

Preferred selection methods for use herein are resistance to kanamycin, hygromycin and the amino glycoside G418, as well as ability to grow on media lacking uracil, leucine, lysine, tryptophan or histidine. In alternate embodiments, 5-fluoroorotic acid (5-fluorouracil-6-carboxylic acid monohydrate [5-FOA]) is used for selection of yeast Ura ^" mutants (U.S. Pat. Appl. Publ. No. 2009-0093543), or a native acetohydroxyacid synthase (or acetolactate synthase; E.C. 4.1 .3.18) that confers sulfonyl urea herbicide resistance (Intl. Appl. Publ. No. WO

2006/052870) is utilized for selection of transformants. A unique method of "recycling" a pair of preferred selection markers for their use in multiple

sequential transformations, by use of site-specific recombinase systems, is also taught in U.S. Pat. Appl. Publ. No. 2009-0093543.

It may be desirable to manipulate a number of different genetic elements in the disclosed embodiments that control aspects of transcription, RNA stability, translation, protein stability and protein location, oxygen limitation and secretion from the host cell. More specifically, gene expression may be controlled by altering the following: the nature of the relevant promoter and terminator sequences; the number of copies of the cloned gene; whether the gene is plasmid-borne or integrated into the genome of the host cell; the final cellular location of the synthesized foreign protein; the efficiency of translation in the host organism; the intrinsic stability of the cloned gene protein within the host cell; and the codon usage within the cloned gene, such that its frequency approaches the frequency of preferred codon usage of the host cell.

Promoters useful to drive expression of heterologous genes in microbial host cells are numerous and known to those skilled in the art. Expression can be accomplished in an induced or constitutive fashion. Induced expression can be accomplished by inducing the activity of a regulatable promoter operably linked to the gene of interest, while constitutive expression can be achieved by the use of a constitutive promoter operably linked to the gene of interest. Virtually any promoter (i.e., native, synthetic, or chimeric) capable of directing expression of a gene is suitable, although transcriptional and translational regulatory regions from the host species may be particularly useful.

In general, the terminator can be derived from the 3' region of the gene from which the promoter was obtained or from a different gene. A large number of terminators are known and function satisfactorily in a variety of hosts, when utilized both in the same and different genera and species from which they were derived. The terminator usually is selected more as a matter of convenience rather than because of any particular property. Preferably, the terminator is derived from a yeast gene. The terminator can also be synthetic, as one of skill in the art can utilize available information to design and synthesize a terminator. A terminator may be unnecessary, but it is preferred.

Although not intended to be limiting, preferred promoters and terminators for use in a recombinant microbial host cell of the genus Yarrowia are those taught in U.S. Pat. Appl. Publ. Nos. 2009-0093543, 2010-0068789, 201 1 - 0059496, 2012-0252079, 2012-0252093, 2013-0089910 and 2013-008991 1 , all of which are incorporated herein by reference.

Additional copies (i.e., more than one copy) of the PUFA biosynthetic pathway desaturases, elongases, etc. genes may be introduced into the recombinant microbial host cell to increase EPA production and accumulation. Specifically, additional copies of genes may be cloned within a single expression construct; and/or additional copies of the cloned gene(s) may be introduced into the host cell by increasing the plasmid copy number or by multiple integration of the cloned gene into the genome.

In general, once a DNA cassette (e.g., comprising a chimeric gene comprising a promoter, ORF and terminator) suitable for expression in a recombinant microbial host cell has been obtained, it is either placed in a plasmid vector capable of autonomous replication in the host cell or directly integrated into the genome of the host cell. Integration of expression cassettes can occur randomly within the host genome or can be targeted through the use of constructs containing regions of homology with the host genome sufficient to target recombination with the host locus. Although not relied on herein, all or some of the transcriptional and translational regulatory regions can be provided by the endogenous locus where constructs are targeted to an endogenous locus.

With respect to engineered recombinant Y. lipolytica host cells, the preferred method of expressing genes in this microbial host is by integration of a linear DNA fragment into the genome of the host. Integration into multiple locations within the genome can be particularly useful when high level expression of genes is desired. Preferred loci include those taught in U.S. Pat. Appl. Publ. No. 2009-0093543, for example.

Another aspect of the disclosed invention concerns a method for producing a microbial oil comprising a polyunsaturated fatty acid (PUFA). This method comprises:

a) culturing a recombinant microbial cell as described herein, wherein a microbial oil comprising a PUFA is produced; and

b) optionally recovering the microbial oil of step (a).

In certain embodiments, the microbial oil produced by the method comprises EPA. Depending on the species of the microbial cell used in the method, the oil may be a fungal oil or yeast oil, for example. The oil in certain embodiments may be recovered or obtained from the recombinant microbial cell after about 12, 24, 36, 48, 60, 72, 84, 96, 100, 105, 1 10, 1 15, 120, 125, 130, 135, 140, 145, 150, 160, 170, 180, or 200 hours of culturing the microbial cell.

The recombinant microbial cell of the present disclosure can be grown under conditions that optimize expression of the disclosed polynucleotides and produce the greatest and the most economical yield of one or more PUFAs. In general, media conditions may be optimized by modifying the type and amount of carbon source, the type and amount of nitrogen source, the carbon-to-nitrogen ratio, the amount of different mineral ions, the oxygen level, growth temperature, pH, length of the biomass production phase, length of the oil accumulation phase and the time and method of cell harvest. Fermentation media for growing the recombinant microbial cell described herein must contain a suitable carbon source, such as described in U.S. Pat. No. 7,238,482 and U.S. Pat. Appl. Publ. No. 201 1 -0059204. Preferred growth media include, for example, common commercially prepared media such as Yeast Nitrogen Base, corn steep liquors, or corn steep solids. Other defined or synthetic growth media may also be used. A suitable pH range for the

fermentation is typically between about pH 4.0 to pH 8.0, wherein pH 5.5 to pH 7.5 is preferred as the range for the initial growth conditions. The

fermentation may be conducted under aerobic or anaerobic conditions, where microaerobic conditions are preferred.

Typically, accumulation of high levels of PUFAs in oleaginous yeast cells requires a two-stage process, since the metabolic state must be "balanced" between growth and synthesis/storage of fats. Thus, most preferably, a two- stage fermentation process can be used for the production of EPA in Yarrowia lipolytica. This approach is described in U.S. Pat. No. 7,238,482, as are various suitable fermentation process designs (i.e., batch, fed-batch and continuous) and considerations during growth. In a two-stage approach, the first stage of the fermentation is for the accumulation of cell mass and is characterized by rapid cell growth and division; a standard amount of nitrogen is included in this stage of fermentation. In the second stage of the fermentation, nitrogen deprivation in the culture promotes a high level of lipid production and accumulation. The first stage may be performed for about 12, 24, 36, 48, or 60 hours, while the second stage (oleaginous) may be performed for about 12-150 hours, depending on the desired level of oil production.

The conditions of growing the disclosed recombinant microbial cell may be oleaginous; for example, oleaginous growth conditions for Yarrowia are described in U .S. Appl. Publ . No. 2009-0093543, which is incorporated herein by reference. Oleaginous growth conditions differ from standard growth conditions mainly in that nitrogen is absent or very limited (nitrogen-limited), but while still providing an ample or high amount of a fermentable carbon source. Example fermentable carbon sources are monosaccharides (e.g., glucose, fructose), disaccharides (e.g., sucrose), invert sucrose, oligosaccharides, polysaccharides, alkanes, fatty acids (e.g., 10-22 carbons), esters of fatty acids, glycerol, monoglyce des, diglycerides, and triglycerides. An example of an oleaginous growth medium lacking nitrogen has about 80 g/L glucose, 2.58 g/L KH ₂ PO ₄ and 5.36 g/L K ₂ HPO . Another example is a medium in which no nitrogen-containing salt is directly added when preparing the medium. Since an oleaginous medium is nitrogen-limited, it may have at most about .050, .100, .125 .150, .175, .200, .225, .250, .275, or .300 g/L of a nitrogen-containing salt (e.g., ammonium- containing salt such as (NH ₄ ) ₂ HPO ₄ , (NH ₄ ) ₂ HSO ₄ , NH ₄ NO ₃ , or NH ₄ CI; a nitrate- containing salt such as KNO3 or NaNOs), amino acid, or urea. The amount of glucose in an oleaginous growth medium may be at least about 20, 30, 40, 50, 60, 70, 80, 90, 100, 1 10, 120, 130, 140, 150, 160, 170, 180, 190, or 200 g/L.

EXAMPLES

The present invention is further defined in the following Examples. It should be understood that these Examples, while indicating preferred aspects of the invention, are given by way of illustration only. From the above discussion and these Examples, one skilled in the art can ascertain the essential

characteristics of this invention, and without departing from the spirit and scope thereof, can make various changes and modifications of the invention to adapt it to various uses and conditions.

GENERAL METHODS

Standard recombinant DNA and molecular cloning techniques used herein are well known in the art and are described by Sambrook, J. and Russell, D., Molecular Cloning: A Laboratory Manual, 3rd Edition, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY (2001 ); and by Silhavy, T.J., Bennan, M.L. and Enquist, L.W., Experiments with Gene Fusions, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY (1984); and by Ausubel, F.M. et. al., Short Protocols in Molecular Biology, 5th Ed. Current Protocols, John Wiley and Sons, Inc., N.Y., 2002. Unless otherwise indicated herein comparisons of genetic sequences were performed using DNASTAR software (DNASTAR Inc., Madison, Wl).

Materials and methods suitable for the maintenance and growth of microbial cultures are well known in the art. Techniques suitable for use in the following examples may be found as set out in Manual of Methods for General Bacteriology {P. Gerhardt, R.G.E. Murray, R.N . Costilow, E.W. Nester, W.A. Wood, N.R. Krieg and G.B. Phillips, Eds.), American Society for Microbiology: Washington, D.C. (1994)); or in Manual of Industrial Microbiology and

Biotechnology, 3rd Edition (R.H. Baltz, J.E. Davies, and A.L. Demain, Eds.), ASM Press, Washington, DC, 2010.

All reagents, restriction enzymes and materials used for the growth and maintenance of microbial cells were obtained from Aldrich Chemicals

(Milwaukee, Wl), DIFCO Laboratories (Detroit, Ml), New England Biolabs, Inc. (Beverly, MA), GIBCO/BRL (Gaithersburg, MD), or Sigma Chemical Company

(St. Louis, MO), unless otherwise specified.

The structure of each genetic expression cassette disclosed herein is

represented by the simple notation system of "X::Y::Z". Specifically, X describes the promoter, Y describes the protein-coding sequence, and Z describes the

terminator. X is operably linked to Y, and Y is operably linked to Z.

Transformation and Cultivation of Y. lipolytica

Y. lipolytica strains were routinely grown at 30 °C in several media,

according to the recipes shown below.

High Glucose Medium (HGM) (per liter): 80 g glucose, 2.58 g KH ₂ PO ₄ and 5.36 g K ₂ HPO ₄ , pH 7.5 (do not need to adjust).

Synthetic Dextrose Medium (SD) (per liter): 6.7 g yeast nitrogen base with ammonium sulfate and without amino acids, and 20 g glucose.

Fermentation medium (FM) (per liter): 6.7 g yeast nitrogen base with ammonium sulfate and without amino acids, 6.0 g KH ₂ PO ₄ , 2.0 g K ₂ HPO ₄ , 1 .5 g MgSO ₄ ^» 7H ₂ O, 20 g glucose, and 5.0 g yeast extract (BBL, BD Diagnostic Systems, Sparks, MD).

Transformation of Y. lipolytica was performed as described in U .S. Pat.

Appl. Publ. No. 2009-0093543, which is incorporated herein by reference. In

general, for transformation of Ura3 ^~ cells, cells were transfected with a plasmid or fragment thereof carrying a URA3 gene, and then selected for transformation on plates lacking uracil.

Fatty Acid Analysis of Y. lipolytica

For fatty acid analysis, cells were collected by centrifugation and lipids were extracted as described in Bligh, E.G. & Dyer, W.J. (Can. J. Biochem.

Physiol., 37:91 1 -917 (1959)). Fatty acid methyl esters (FAMEs) were prepared by transesterification of the lipid extract with sodium methoxide (Roughan, G., and Nishida I., Arch Biochem Biophys., 276(1 ):38-46 (1990)) and subsequently analyzed with an Agilent Technologies 6890N gas chromatograph fitted with a 30-m X 0.25 mm (i.d.) SUPELCO Omegawax320 (Agilent Technologies) column. The oven temperature was ramped from 160 °C to 240 °C at 30 °C/min and then held for 3.8 min at 240 °C.

For direct base transesterification, a Y. lipolytica culture (1 ml_) was harvested by centrifugation (13,000 x g) for 1 min. Sodium methoxide (500 μΙ_ of a 1 % solution) was added to the sample, and then the sample was vortexed and rocked for 45 min. Then, 100 μΙ_ of 1 .0 M NaCI and 500 μΙ_ of hexane were added, and the sample was vortexed and spun. The upper layer was removed and analyzed by gas chromatography as described above.

In general, initial fatty acid screening of new transformants of Yarrowia strains was performed as follows. Single colonies that were grown on minimal medium (MM) plates at 30 °C for 5 to 6 days were re-streaked onto MM plates, grown for two days at 30 °C, and then inoculated into liquid MM in a multi-well plate (e.g., 24-well, 3 ml_ MM) and shaken at 250 rpm at 30 °C for 2 days. The cells from each well were collected by centrifugation, resuspended in HGM, and then shaken at 250 rpm for 5 days. Cells were then processed for fatty acid analysis as described above. Transformants exhibiting a desired fatty acid trait were further analyzed by "flask assay" as described below.

Analysis of Total Lipid Content and Composition in Y. lipolytica (Flask Assay)

For a detailed analysis of the total lipid content and composition in a particular strain of Y. lipolytica, flask assays were conducted as follows.

Specifically, cultures were grown at a starting OD ₆ oo of -0.3 in 25 ml_ of SD medium in a 125-mL flask for 48 h. 6 ml_ of the culture was harvested by centrifugation for 5 min at 4300 rpm in a 50-mL conical tube. The supernatant was discarded and the cells were resuspended in 25 mL of HGM in another 125 mL flask; this culture was incubated for 120 hours (except as otherwise noted) in a shaker incubator at 250 rpm and 30 °C. A 1 -mL aliquot of the culture was then used for fatty acid analysis (as described above) following centrifugation for 1 min at 13,000 rpm, and a 5-mL aliquot of the culture was dried for dry cell weight determination. All flask assays referenced herein were performed following this methodology, except those performed using the "one-step flask assay".

For DCW determination, 10 mL of culture was harvested by centrifugation for 5 min at 4300 rpm. The pellet was resuspended in 10 mL of sterile water and re-harvested as above. The washed pellet was re-suspended in 1 mL of water (three times) and transferred to a pre-weighed aluminum pan. The cell suspension was dried overnight in a vacuum oven at 80 °C. The weight of the cells was determined (g/L).

Total lipid content of cells (TFAs % DCW) was calculated and considered in conjunction with data tabulating the concentration of each fatty acid as a weight percent of TFAs (% TFAs) and the EPA content as a percent of the dry cell weight (EPA % DCW). Data from flask assays are presented in table format summarizing the total DCW of the cells, the total lipid content of cells (TFAs % DCW), the concentration of each fatty acid as a weight percent of TFAs (% TFAs) and the EPA content as a percent of the dry cell weight (EPA % DCW). Y. lipolytica Strains Z5627 and Z5585

The generation of Y. lipolytica strains Z5627 and Z5585 is described in U.S. Pat. Appl. Publ. No. 2012-0052537, which is incorporated herein by reference. As described in Examples 1 -4, Z5627 and Z5585 were used to derive certain strains of the disclosed invention.

Y. lipolytica strains Z5627 and Z5585 were derived from multiple genetic modifications of strain Y9502, which in turn was derived after multiple genetic modifications of wild type Y. lipolytica strain ATCC #20362. The modification steps and intermediate strains used for generating strains Z5627 and Z5585 are shown in Figures 2A (ATCC #20362 to Y9502) and 2B (Y9502 to Z5627 and Z5585). The genotype of both strains Z5627 and Z5585 with respect to wild type Y. lipolytica ATCC #20362 is: Ura ⁺ , Pex3 ^~ , unknown T, unknown 2 ^" , unknown 3 ^' , unknown 4 ^' , YALI0E12947g ^~ , unknown 6 ^' , YALI0B21890g ^' , unknown 8 ^' , unknown 10 ^' , unknown 11 ^' , unknown 12 ^' , unknown 13 ^' , unknown 14 ^' ,

YAT1 ::ME3S::Pex16, GPD::ME3S::Pex20, YAT1 ::ME3S::Lip1 ,

FBAINm::EgD9eS::Lip2, EXP1 ::EgD9eS::Lip1 , GPAT::EgD9e::Lip2,

YAT1 ::EgD9eS::Lip2, YAT1 ::EgD9eS-L35G::Pex20, FBAINm::EgD8M::Pex20, EXP1 ::EgD8M::Pex16, FBAIN::EgD8M::Lip1 , GPD::EaD8S::Pex16 (2 copies), YAT1 ::E389D9eS/EgD8M::Lip1 , YAT1 ::EgD9eS/EgD8M::Aco,

FBAINm::EaD9eS/EaD8S::Lip2, DGAT2M::YID9::Lip1 , GPDIN::YID9::Lip1 , GPD::FmD12::Pex20, YAT1 ::FmD12::Oct, EXP1 ::FmD12S::Aco,

GPDIN::FmD12::Pex16, EXP1 ::EgD5M::Pex16, FBAIN::EgD5SM::Pex20, EXP1 ::EgD5SM::Lip1 , YAT1 ::EaD5SM::Oct, FBAINm::PaD17::Aco,

EXP1 ::PaD17::Pex16, YAT1 ::PaD17S::Lip1 , YAT1 ::YICPT1 ::Aco,

EXP1 ::YICPT1 ::Oct, YAT1 ::MCS::Lip1 , FBA::MCS::Lip1 , EXP1 ::YIPCT::Pex16, YAT1 ::MaLPAAT1 S::Pex16, ALK2LM1 ::MaLPAAT1 S::Pex20,

FBAINm::YILPAAT1 ::Lip1 (2 copies), YAT1 ::YIPDAT::Lip1 (2 copies).

The abbreviations listed in the above genotype are as follows: FmD12 is a Fusarium moniliforme delta-12 desaturase gene and FmD12S is a codon- optimized form thereof (U.S. Pat. No. 7,504,259); ME3S is a codon-optimized Mortierella alpina C16 18 elongase gene (U.S. Pat. No. 7,470,532); EgD9e is a Euglena gracilis delta-9 elongase gene and EgD9eS is a codon-optimized form thereof (U.S. Pat. No. 7,645,604); EgD9eS-L35G is a mutant form of EgD9eS (U.S. Pat. Appl. Publ. No. 2012/0226062); EgD8M is a synthetic mutant E.

gracilis delta-8 desaturase gene (U.S. Pat. No. 7,709,239); EaD8S is a codon- optimized Euglena anabaena delta-8 desaturase gene (U.S. Pat. No. 7,790,156); E389D9eS/EgD8M is a DGLA synthase created by linking a codon-optimized delta-9 elongase gene (E389D9eS) from Eutreptiella sp. CCMP389 (U.S. Pat. No. 7,645,604) to EgD8M (U.S. Pat. Appl. Publ. No. 2008/0254191 );

EgD9eS/EgD8M is a DGLA synthase created by linking EgD9eS to EgD8M (U.S. Pat. Appl. Publ. No. 2008/0254191 ); EaD9eS/EaD8S is a DGLA synthase created by linking a codon-optimized E. anabaena delta-9 elongase gene

(EaD9eS) (U.S. Pat. No. 7,794,701 ) to EaD8S (U.S. Pat. No. 7,790,156); YID9 is a Y. Iipolytica delta-9 desaturase gene (U.S. Pat. Appl . Publ. No. 2012/0052537); EgD5M and EgD5SM are synthetic mutant E. gracilis delta-5 desaturase genes comprising a mutant HPGS motif (U.S. Pat. Appl. Publ. No. 2010/0075386); EaD5SM is a synthetic mutant E. anabaena delta-5 desaturase gene comprising a mutant HAGG motif (U.S. Pat. Appl. Publ. No. 2010/0075386); PaD17 is a Pythium aphanidermatum delta-17 desaturase gene and PaD17S is a codon- optimized form thereof (U.S. Pat. No. 7,556,949); YICPT1 is a Y. Iipolytica diacylglycerol cholinephosphotransferase gene (U.S. Pat. No. 7,932,077); MCS is a codon-optimized malonyl-CoA synthetase gene from Rhizobium

leguminosarum bv. viciae 3841 (U.S. Pat. Appl. Publ. No. 2010/0159558); YIPCT is a Y. Iipolytica choline-phosphate cytidylyl-transferase gene (U.S. Pat. Appl. Publ. No. 2012-0052537 and GenBank Accession No. XM_502978);

MaLPAATI S is a codon-optimized M. alpina lysophosphatidic acid

acyltransferase gene (U.S. Pat. No. 7,879,591 ); YILPAAT1 is a Y. Iipolytica lysophosphatidic acid acyltransferase gene (U.S. Pat. Appl. Publ. No. 2012- 0052537); YIPDAT is a Y. Iipolytica phospholipid:diacylglycerol acyltransferase gene (U.S. Pat. Appl. Publ. No. 2012-0052537); YAT1 is a Y. Iipolytica YAT1 gene promoter (U.S. Pat. Appl. Publ. No. 2010/0068789); Pex16 is a Y. Iipolytica Pex16 gene terminator (GenBank Accession No. U75433); GPD is a Y. Iipolytica glyceraldehyde-3-phosphate dehydrogenase gene promoter (U.S. Pat. No.

7,459,546); GPDIN is a Y. Iipolytica GPD gene promoter plus intron (U.S. Pat. No. 7,459,546); Pex20 is a Y. Iipolytica Pex20 gene terminator (GenBank

Accession No. AF054613); Lip1 is a Y. Iipolytica Lip1 gene terminator (GenBank Accession No. Z50020); Lip2 is a Y. Iipolytica Lip2 gene terminator (GenBank Accession No. AJ012632); FBA is a Y. Iipolytica fructose-bisphosphate aldolase promoter sequence, and FBAIN and FBAINm are Y. Iipolytica FBA promoter plus intron sequences (U .S. Pat. No. 7,202,356); DGAT2M is a Y. Iipolytica

diacylglycerol acyltransferase-2 (DGAT2) promoter sequence (U.S. Pat. Appl. Publ. No. 2012-0052537); EXP1 is a Y. Iipolytica export protein (EXP1 ) gene promoter sequence (Intl. Appl. Publ. No. WO06/052870); GPAT is a Y. Iipolytica GPAT promoter (Intl. Appl. Publ. No. WO 2006/031937); Aco is a Y. Iipolytica Aco gene terminator (GenBank Accession No. AJ001300); Oct is a Y. Iipolytica Oct gene terminator (GenBank Accession No. X69988); and ALK2LM1 is a V. Iipolytica n-alkane-hydroxylating cytochrome P450 gene (ALK2) promoter sequence plus N-terminal 66-bp coding region of the Y. Iipolytica ALK2 gene (U.S. Pat. Appl. Publ. No. 2012-0052537).

As shown below in Table 4, which is also disclosed in U.S. Pat. Appl. Publ. No. 2012-0052537, strain Z5627 can produce oil containing about 49.5% EPA in the fatty acids of the oil. This strain can also produce, as a percentage of DCW, about 52% oil and 25.6% EPA.

As shown in Table 4, strain Z5585 can produce oil containing about 49.4% EPA in the fatty acids of the oil. This strain can also produce, as a percentage of DCW, about 56.6% oil and 28% EPA.

Table 4. Total Lipid Content and Composition in Various Recombinant Y. lipolytica Strains bv Flask Assay (U.S.

Pat. Appl. Publ. No. 2012-0052537) ⁹

TFAs % TFAs EPA

DCW % ETr %

Strain (g/L) DCW 16:0 16:1 18:0 18:1 18:2 ALA EDA DGLA ARA A ETA EPA DCW

Z1977 3.8 34.3 2 0.5 1 .9 4.6 1 1.2 0.7 3.1 3.3 0.9 0.7 2.2 59.1 20.3

Z1978 3.9 38.3 2.4 0.4 2.4 4.8 1 1.1 0.7 3.2 3.3 0.8 0.6 2.1 58.7 22.5

Z1979 3.7 33.7 2.3 0.4 2.4 4.1 10.5 0.6 3.2 3.6 0.9 0.6 2.2 59.4 20

Z1980 3.6 32.7 2.1 0.4 2.2 4 10.8 0.6 3.1 3.5 0.9 0.7 2.2 59.5 19.5

Z1981 3.5 34.3 2.2 0.4 2.1 4.2 10.6 0.6 3.3 3.4 1 0.8 2.2 58.5 20.1

Genotype Additions with Respect to Strain Z1978: YILPAAT1, YIPDAT

L250 4.4 51 .5 2 0.7 2.8 6.1 16.7 0.9 3.3 4.9 0.7 0.6 3.2 50.4 26

Genotype Additions with Respect to Strain Z1978: 2 YILPAAT1, 2 YIPDAT

L258 ^a 5 57.1 2.3 0.9 3.4 7.8 18.7 0.9 4 5.3 0.8 0.6 3.2 45.2 25.8

Genotype Additions with Respect to Strain Z1978: 2 YILPAAT1, 2 YIPDAT, EgD8M, MCS, MaLPAATIS

Z5565 ^b 4.8 56.1 2.1 0.8 2.8 6.8 17.3 0.8 3.8 5.2 1.1 0.8 3.4 47.4 26.6

Z5567 ^b 4.9 56.2 1.9 0.7 2.6 6.2 16.7 0.7 3.8 5.6 1.1 1 3.6 48.3 27.1

Z5575 ^b 4.7 53.8 1.8 0.7 2.4 5.7 15.3 0.6 3.6 5.9 1.2 1 3.6 50.4 27.1

Z5576 ^b 4.9 55.6 2.3 0.9 2.8 6.9 16.9 0.7 3.6 5.5 1.2 0.9 3.3 47.2 26.2

Genotype Additions with Respect to Strain Z1978: 3 YILPAAT1, 2 YIPDAT, EgD8M, MCS

Z5620 ^c 4.5 52.8 2.1 0.7 2.8 6.6 16.1 0.7 3.6 5.7 1.1 0.7 3.3 49 25.9

Z5623 ⁰ 4.3 51 .7 2.3 0.8 2.4 6 15.9 0.7 3.8 5.2 1.1 0.7 3.1 50 25.8

Z5625 ⁰ 4.6 52.7 2.1 0.7 2.7 6.2 16.6 0.7 3.9 5.4 1.1 0.8 3.2 49.1 25.9

Genotype Additions with Respect to Strain Z1978: 2 YILPAAT1, 2 YIPDAT, ME3S, MCS, MaLPAATIS

Z5581 ^d 4.7 56.3 1.9 0.7 2.6 6.1 16.5 0.7 3.7 5.6 1.2 1 3.5 48.7 27.4

Z5582 ^d 4.8 55.6 1.9 0.7 2.5 6.1 16.4 0.7 3.7 5.7 1.1 0.9 3.6 48.9 27.2

Z5583 ^d 4.9 56.8 2 0.7 2.6 6.2 16.7 0.8 3.7 5.4 1 1 3.7 48.4 27.5

Z5584 ^d 4.9 55.3 2 0.7 2.7 6.5 16.1 0.7 3.7 5.7 1.1 1 3.6 48.6 26.8

Genotype Additions with Respect to Strain Z1978: 2 YILPAAT1, 2 YIPDAT, YIPCT, YID9, MaLPAATI S

Z5570 ^e 4.8 55 2 0.8 2.5 6.1 16.4 0.7 3.7 5.5 1.2 1 3.4 48.6 26.8

Z5571 ^e 4.8 54.1 2.2 0.8 2.4 6.5 16.7 0.7 3.8 5.5 1.1 0.9 3.2 48.3 26.2

Z5572 ^e 4.9 54 2.1 0.8 2.5 6.5 16.7 0.7 3.7 5.5 1.1 0.9 3.3 48.4 26.1

Z5574 ^e 5 53.8 1.8 0.7 2.4 5.7 15.3 0.6 3.6 5.9 1.2 1 3.6 50.4 27.1

Genotype Additions with Respect to Strain Z1978: 2 YILPAAT1, 2 YIPDAT, YICPT1, YID9 , MaLPAATIS

Z5585 ^f 4.6 56.6 1.9 0.7 2.6 5.6 16.4 0.7 3.5 5.5 1.1 1 3.5 49.4 28

Z5627 ^f 4.8 52 1.9 0.7 2.6 6.2 16.1 0.6 4 5.6 1 .2 0.9 3.2 49.3 25.6

a Strain L258 was used to derive L258U (Figure 2B), which is Ura3 ^~ .

b Each of strains Z5565, Z5567, Z5575 and Z5576 was derived through the one-step introduction to strain L258U of gene cassettes for expressing EgD8M, MCS and MaLPAATI S.

c Each of strains Z5620, Z5623 and Z5625 was derived through the one-step introduction to strain L258U of gene cassettes for expressing YILPAAT1 , EgD8M and MCS.

d Each of strains Z5581 , Z5582, Z5583 and Z5584 was derived through the one-step introduction to strain L258U of gene cassettes for expressing ME3S, MCS and MaLPAATI S.

e Each of strains Z5570, Z5571 , Z5572 and Z5574 was derived through the one-step introduction to strain L258U of gene cassettes for expressing YIPCT, YID9 and MaLPAATI S.

f Each of strains Z5585 and Z5627 was derived through the one-step introduction to strain L258U of gene cassettes for expressing YICPT1 , YID9 and MaLPAATI S.

⁹ The values shown in the table were measured in each strain after growth in HGM for 120 hours.

Table 4 lists other strains beside Z5627 and Z5585 that were derived through multiple genetic modifications of wild type Y. lipolytica strain ATCC #20362. The strains in this table can produce approximately 20%-28% EPA as a percentage of DCW. Several of the strains listed in Table 4, specifically strains L250 on down to Z5627, are all descendants of intermediate strain Z1978. The genetic modifications ("genotype additions") made in each strain with respect to Z1978 are shown in Table 4. Strains Z5565 on down to Z5627 in Table 4 were directly derived through certain one-step genetic modifications of strain L258U, which is a Ura3 ^~ transformant of strain L258 (see table footnotes).

It is apparent from Table 4 that the different genetic modifications made to strain Z1978 to yield each of descendent strains Z5565 to Z5627 raised the total oil content (TFAs % DCW) from 38.3% to a range of 51 .7%-56.6%. This rise in oil content was associated with an overall decrease in the percentage of EPA in the fatty acids of the oil, from 58.7% in Z1978 to a range of 47.2%-50.4% in descendent strains Z5565 to Z5627.

Thus, the increase in the total amounts of EPA produced on a dry cell weight basis obtained in strains Z5565 to Z5627 (25.6 to 28 EPA % DCW, Table 4) through the genetic modifications of strain Z1978 (22.5 EPA % DCW, Table 4) was achieved through substantially increasing oil production. Despite this significant achievement, it was not apparent from the studies disclosed in U.S. Pat. Appl. Publ. No. 2012-0052537 how to further increase total EPA content on a dry cell weight basis. For example, the steps to maintain or increase oil content while increasing the amount of EPA in the fatty acids of the oil, which would boost EPA % DCW, were unknown.

Thus, there is still a need to increase oil production while also increasing EPA content (EPA % TFAs).

Example 1

Generation of Strain Z6109 Producing at Least about 51 .7% EPA of Total Fatty

Acids with at Least about 54.2% Total Lipid Content This Example describes the generation of Y. lipolytica strain Z6109 through genetic modification of strain Z5627. The genetic modification entailed introducing an expression cassette encoding Arabidopsis thaliana caleosin-1 (AtClol ) into Z5627. Figure 3A shows the modification steps and intermediate strain used for generating strain Z6109.

In order to introduce AtClol to Z5627, it was necessary to first render this strain to be Ura3 ^~ for subsequent selection purposes. Z5627 carries an intact URA3 coding sequence within the integrated plasmid construct pZKMP- ML9DCB, which was previously used to introduce sequences allowing for expression of MaLPAATI S, YID9 and YICPT1 (see U.S. Pat. Appl. Publ. No. 2012-0052537). To disrupt the URA3 coding sequence of Z5627, construct pZKUM was used to integrate a Ura3 ^~ mutant sequence into the intact URA3 sequence. The construction and use of plasmid pZKUM to obtain Ura ^~ Y.

Iipolytica cells has been described (U.S. Pat. Appl. Publ. No. 2009-0093543, see Table 15 therein, which is incorporated herein by reference).

Z5627 pZKUM-transformants with a Ura ^~ phenotype were selected on minimal media (MM) plates containing 5-fluoroorotic acid (5-FOA) (U.S. Pat. Appl. Publ. No. 2009-0093543). A total of eight transformants were grown and identified to possess a Ura ^~ phenotype. These transformants were subjected to an initial fatty acid screening process as described above.

Gas chromatography (GC) analyses showed the presence of 33.5%, 35.7%, 35.9% and 34% EPA of TFAs in Z5627 pZKUM-transformants #2, #3, #4 and #6 cells from FOA-plates, respectively. These four transformants were designated as strains Z5627U1 , Z5627U2, Z5627U3 and Z5627U5, respectively, and were collectively designated as strain Z5627U.

Plasmid pYRH55 (Figure 4A, SEQ ID NO:1 ) was generated to integrate one synthetic Arabidopsis thaliana caleosin-1 (AtClol ) gene into the Yarrowia lipase 7 gene locus (GenBank Accession No. AJ549519). The AtClol coding sequence (SEQ ID NO:2, encoding SEQ ID NO:3) in pYRH55 was derived from GenBank Accession No. AEE85247 and is codon-optimized for expression in Y. Iipolytica (see U.S. Appl. Publ. No. 2012-0301932, which is incorporated herein by reference). This codon-optimized AtClol is herein referred to as AtClol S. Table 5 describes the components contained in pYRH55. Table 5. Description of Plasmid pYRH55 (SEQ ID NO:1 )

a RE, restriction endonuclease

The pYRH55 plasmid was digested with Asc\/Sph\, and then used to transform strain Z5627U5. The transformed cells were plated onto uracil-minus MM plates and maintained at 30 °C for 5 to 6 days. Single colonies were grown as described above for initial fatty acid screening.

GC analyses showed that almost all of the selected 72 strains of Z5627U5 transformed with pYRH55 produced more than 49% EPA of TFAs. Eleven strains (#1 , #1 1 , #15, #16, #20, #21 , #30, #34, #51 , #53 and #54) produced about 55.3%, 50.1 %, 50.9%, 51 .7%, 50.8%, 49.7%, 53.4%, 54.8%, 50.3%, 53.9% and 50.6% EPA of TFAs and were designated as Z6103, Z6104, Z6105, Z6106, Z6107, Z6108, Z6109, Z61 10, Z61 1 1 , Z61 12 and Z61 13, respectively.

Knockout of the Lip7 locus in above strains Z6103 to Z61 13 was not confirmed. The genotype of strains Z6109 and its ten siblings with respect to wild type Y. lipolytica ATCC #20362 was: Ura ⁺ , Pex3 ^~ , unknown T, unknown 2 ^" , unknown 3 ^' , unknown 4 ^' , YALI0E12947g ^~ , unknown 6 ^' , YALI0B21890g ^~ , unknown 8 ^' , unknown 10 ^' , unknown 11 ^' , unknown 12 ^' , unknown 13 ^' , unknown 14 ^' , unknown 15 ^' , YAT1 ::ME3S::Pex16, GPD::ME3S::Pex20, YAT1 ::ME3S::Lip1 , FBAINm::EgD9eS::Lip2, EXP1 ::EgD9eS::Lip1 , GPAT::EgD9e::Lip2,

YAT1 ::EgD9eS::Lip2, YAT1 ::EgD9eS-L35G::Pex20, FBAINm::EgD8M::Pex20, EXP1 ::EgD8M::Pex16, FBAIN::EgD8M::Lip1 , GPD::EaD8S::Pex16 (2 copies), YAT1 ::E389D9eS/EgD8M::Lip1 , YAT1 ::EgD9eS/EgD8M::Aco, FBAINm::EaD9eS/EaD8S::Lip2, DGAT2M::YID9::Lip1 , GPDIN::YID9::Lip1 , GPD::FmD12::Pex20, YAT1 ::FmD12::Oct, EXP1 ::FmD12S::Aco,

GPDIN::FmD12::Pex16, EXP1 ::EgD5M::Pex16, FBAIN::EgD5SM::Pex20, EXP1 ::EgD5SM::Lip1 , YAT1 ::EaD5SM::Oct, FBAINm::PaD17::Aco,

EXP1 ::PaD17::Pex16, YAT1 ::PaD17S::Lip1 , YAT1 ::YICPT1 ::Aco,

EXP1 ::YICPT1 ::Oct, YAT1 ::MCS::Lip1 , FBA::MCS::Lip1 , EXP1 ::YIPCT::Pex16, YAT1 ::MaLPAAT1 S::Pex16, ALK2LM1 ::MaLPAAT1 S::Pex20,

FBAINm::YILPAAT1 ::Lip1 (2 copies), YAT1 ::YIPDAT::Lip1 (2 copies),

FBAINm::AtClo1 S::Pex20.

Analysis of Total Lipid Content and Composition by Flask Assay

Cells of strains Z6109 and its siblings were grown and analyzed for total lipid content and fatty acid composition by the flask assay described above. Table 6 summarizes the DCW, the TFAs % DCW, the amount of each fatty acid as a weight percent of TFAs (% TFAs) and the EPA % DCW of strains from Z6103 to Z61 13.

Table 6. Total Lipid Content and Composition in Strain Z6109 and Its Siblings by Flask Assay

DCW TFAs % % TFAs EPA %

Strain (g/L) DCW 16:0 16:1 18:0 18:1 LA ALA EDA DGLA ARA ETrA ETA EPA DCW (g/L/f

Z6103 4.4 51.1 2.2 0.8 2.0 5.5 14.3 0.5 3.7 5.7 1.4 0.6 3.0 52.0 26.6 0.008

Z6104 4.7 55.3 2.1 0.7 2.8 5.9 15.9 0.6 3.6 5.8 1.3 0.7 3.2 49.7 27.5 0.009

Z6105 4.7 55.9 2.0 0.7 2.7 5.9 16.0 0.6 3.7 5.8 1.3 0.7 3.2 49.7 27.8 0.009

Z6106 4.6 53.8 2.0 0.7 2.6 5.5 15.2 0.6 3.4 5.8 1.2 0.6 3.3 51.4 27.6 0.009

Z6107 4.7 54.2 2.1 0.7 2.8 5.8 16.0 0.7 3.6 5.8 1.2 0.6 3.2 50.0 27.1 0.009

Z6108 4.9 55.1 2.0 0.7 2.7 6.2 16.9 0.7 3.7 5.8 1.1 0.6 3.0 49.5 27.2 0.009

Z6109 4.5 54.2 2.2 0.6 2.4 4.8 15.9 0.6 3.9 5.5 1.5 0.7 2.9 51.7 28.0 0.009

Z61 10 4.3 50.4 1.7 0.6 2.1 5.0 12.9 0.5 3.4 6.1 1.1 0.7 3.9 54.0 27.2 0.008

Z61 1 1 4.3 53.2 2.5 0.9 2.5 6.5 16.5 0.6 3.8 5.3 1.5 0.7 2.7 48.4 25.7 0.007

Z61 12 4.5 55.2 2.1 0.8 2.2 5.7 15.0 0.6 3.5 5.7 1.3 0.7 3.2 51.1 28.2 0.009

Z61 13 4.9 54.9 1.9 0.7 2.6 5.9 15.5 0.7 3.9 5.8 1.4 0.7 3.0 49.7 27.3 0.009

The results in Table 6 generally indicate that heterologous expression of AtClol S in strain Z5627 raised oil production (i.e., TFAs % DCW) and the percentage of EPA in the total fatty acids of oil (i.e., EPA % TFAs). Specifically, while strain Z5627 yielded 52 TFAs % DCW and 49.3 EPA % TFAs (Table 4), strains Z6103 to Z61 13 had average TFAs % DCW of 53.9 and EPA % TFAs of 50.7 (Table 6).

Specific strains analyzed in Table 6 had significantly higher oil and EPA levels compared to Z5627. For example, strain Z6109 had 54.2 TFAs % DCW and 51 .7 EPA % TFAs. Z6109 altogether exhibited a 9.4% increase in the total amount of EPA produced (28 EPA % DCW) compared to that produced by Z5627 (25.6 EPA % DCW).

Example 2

Generation of Strain Z6903 Producing at Least about 51 .4% EPA of Total Fatty Acids with at Least about 49.1 % Total Lipid Content and Reduced Sugar Alcohol

By-Products

This Example describes the generation of Y. lipolytica strain Z6903 through genetic modification of strain Z5585. The genetic modification entailed knocking out the endogenous Y. lipolytica gene SOU2, which encodes Sou2 sorbitol utilization protein. The genetic modification further entailed introducing expression cassettes encoding phospholipid:diacylglycerol acyltransferase (PDAT), delta-12 desaturase, and a DGLA synthase multizyme (delta-9 elongase fused to delta-8 desaturase). The development of strain Z6903 was required in order to develop strain Z9276, which is described below in Example 4.

Figure 3B shows the modification steps and intermediate strains

(Z5585U21 and Z5585K2U) used for generating strain Z6903. Strains Z5585U21 and Z5585K2U were generated as follows.

Generation of Strain Z5585U21

Construct pZKUM (above) was used to disrupt the URA3 gene in strain Z5585 that was previously introduced by the plasmid pZKMP-ML9DCB, which carries sequences allowing for expression of MaLPAATI S, YID9 and YICPT1 (see U.S. Pat. Appl. Publ. No. 2012-0052537). A total of eight 5-FOA-resistant transformants were grown and identified to possess a Lira phenotype. These transformants were grown as described above for initial fatty acid screening.

GC analyses showed the presence of 37.7% EPA in the TFAs of pZKUM- transformant strain #6 cells picked from an FOA plate. This transformant was designated as strain Z5585U21 .

Generation of Strain Z5585K2U (Sou?)

Strain Z5585K2U was generated by knocking out the endogenous Y.

lipolytica SOU2 gene, which encodes Sou2 sorbitol utilization protein, in strain Z5585U21 .

The identification of the SOU2 gene as a genetic target to modify lipid production in Y. lipolytica is described below in Example 5. Briefly, during the construction of strain Z3041 (Figure 8), which involved the genetic modification of strain Z1978, it was observed that strain Z3041 produced more DCW, more oil, and less by-products compared to its parent strain, Z2636. Genome walking and sequencing analyses showed that the promoter region of the SOU2 gene (locus YALI0D18964g, GenBank Accession No. XM_503010, Figure 5) was disrupted in strain Z3041 by an insertion occurring within the promoter region of SOU2 (Figure 5). Sorbitol is a sugar alcohol; other sugar alcohols such as mannitol and arabitol are produced as by-products in the Y. lipolytica strains disclosed herein to be engineered for enhanced lipid production.

Plasmid pZKSOU2-New (Figure 4B, SEQ ID NO:4) was used to knock-out a large portion of the SOU2 gene in strain Z5585U21 . Table 7 describes the components contained in pZKSOU2-New. This vector contains 5'- and 3'- homology arms (denoted as ySOU2-5' and ySOU2-3', respectively) containing sequences derived from the endogenous Y. lipolytica SOU2 locus. Stuffer sequence (non-SOU2 sequence) is located between these homology arms. Table 7. Description of Plasmid pZKSOU2-New (SEQ ID NO:4)

The knock-out strategy entailed a "pop-in/pop-out" process as delineated in the diagram and legend of Figure 6. Briefly, the pop-in event occurred as a result of homologous recombination between the 5'-homology arm of pZKSOU2- New and corresponding sequence at the endogenous Y. lipolytica SOU2 locus. This particular integration event resulted in the juxtaposition of a mutated SOU2 allele with the wild type SOU2 allele, and was selected on the basis that pZKSOU2-New integration rendered the cells of strain Z5585U21 (Ura ^~ ) to be Ura

The pop-out event occurred as a result of homologous recombination between the 3'-homology arm of the integrated mutant allele and corresponding sequence at the adjacent endogenous Y. lipolytica SOU2 locus (Figure 6, left- hand pop-out event). Since this pop-out event resulted in removal of the URA3 gene that had been introduced during pop-in, cells in which the pop-out event occurred leaving behind the mutant SOU2 allele could be selected on FOA plates (i.e., cells are Lira ^' ).

Plasmid pZKSOU2-New was used to transform Z5585U21 . A total of 60 Ura ⁺ transformants were grown on MM plates lacking uracil. Polymerase chain reaction (PCR) amplification analyses indicated that transformants #26 and #28 had undergone recombination between the 5'-arm homologous sequences of pZSOU2-New and the endogenous SOU2 gene.

Strain #26 was picked, grown in liquid YPD media, and then plated on FOA plates to select for cells that subsequently became Lira ^' due to a pop-out event (Figure 6). A total of 96 Ura ^' strains were analyzed by PCR amplification to determine which ones had undergone pop-out events involving recombination at the 3'-arm homology sequences which removed the pZKOSOU2-New backbone sequences (AmpR and URA3) thereby leaving behind a mutated SOU2 allele. This PCR analysis was necessary, since certain Ura ^' cells could alternatively have a wild type SOU2 allele if the pop-out recombination event occurred at the 5'-arm homologous sequence (Figure 6). In 9 of the 96 Ura ^' strains, PCR analyses detected recombination at the 3'-arm homologous sequences indicative of a mutant SOU2 allele. Two of these 9 strains were designated as Z5585K2U1 and Z5585K2U2.

The final genotype of both strains Z5585K2U1 and Z5585K2U2 with respect to wild type Y. lipolytica ATCC #20362 is: Ura ^' , Pex3 ^' , unknown T, unknown 2 ^' , unknown 3 ^' , unknown 4 ^' , YALI0E12947g ^' , unknown 6 ^' ,

YALI0B21890g ^' , unknown 8 ^' , unknown 10 ^' , unknown 11 ^' , unknown 12 ^' , unknown 13 ^' , unknown 14 ^' , Sou , YAT1 ::ME3S::Pex16, GPD::ME3S::Pex20,

YAT1 ::ME3S::Lip1 , FBAINm::EgD9eS::Lip2, EXP1 ::EgD9eS::Lip1 ,

GPAT::EgD9e::Lip2, YAT1 ::EgD9eS::Lip2, YAT1 ::EgD9eS-L35G::Pex20, FBAINm::EgD8M::Pex20, EXP1 ::EgD8M::Pex16, FBAIN::EgD8M::Lip1 ,

GPD::EaD8S::Pex16 (2 copies), YAT1 ::E389D9eS/EgD8M::Lip1 ,

YAT1 ::EgD9eS/EgD8M::Aco, FBAINm::EaD9eS/EaD8S::Lip2,

DGAT2M::YID9::Lip1 , GPDIN::YID9::Lip1 , GPD::FmD12::Pex20,

YAT1 ::FmD12::Oct, EXP1 ::FmD12S::Aco, GPDIN::FmD12::Pex16,

EXP1 ::EgD5M::Pex16, FBAIN::EgD5SM::Pex20, EXP1 ::EgD5SM::Lip1 ,

YAT1 ::EaD5SM::Oct, FBAINm::PaD17::Aco, EXP1 ::PaD17::Pex16,

YAT1 ::PaD17S::Lip1 , YAT1 ::YICPT1 ::Aco, EXP1 ::YICPT1 ::Oct,

YAT1 ::MCS::Lip1 , FBA::MCS::Lip1 , EXP1 ::YIPCT::Pex16,

YAT1 ::Mal_PAAT1 S::Pex16, ALK2LM1 ::Mal_PAAT1 S::Pex20,

FBAINm::YILPAAT1 ::Lip1 (2 copies), YAT1 ::YIPDAT::Lip1 (2 copies).

Strain Z5585K2U1 is herein referred to as Z5585K2U. Generation of Strain Z6903

Plasmid pZKADn-SyP298F (Figure 7A, SEQ ID NO:5) was generated to integrate gene cassettes for expressing PDAT (SEQ ID NO:15, PDAT with an extra alanine at position 2 compared to wild type YIPDAT), delta-12 desaturase (SEQ ID NO:13), and a DGLA synthase multizyme (SEQ ID NO:17, delta-9 elongase fused to delta-8 desaturase) into the alcohol dehydrogenase 3 (ADH3) locus (GenBank Accession No. AF175273) of strain Z5585K2U. Table 8 describes the components contained in pZKADn-SyP298F.

Table 8. Description of Plasmid pZKADn-SvP298F (SEQ ID NO:5)

Sal\/Pac\ Yarrowia URA3 gene (GenBank Accession No. AJ306421 ) (1 1 161 -9510)

The pZKADn-SyP298F plasmid was digested with Asc\, and then used to transform strain Z5585K2U. The transformed cells were plated onto uracil-minus MM plates and maintained at 30 °C for 5 to 6 days. Single colonies were grown as described above for initial fatty acid screening.

GC analyses showed that 9 strains of Z5585K2U transformed with pZKADn-SyP298F produced more than 51 .1 % EPA of TFAs. These strains (#10, #14, #15, #61 , #76, #80, #82, #83, #96) produced about 52.4%, 51 .9%, 51 .6%, 53.2%, 52.3%, 52.4%, 51 .9%, 51 .3% and 51 .1 % EPA of TFAs and were designated as Z6897, Z6898, Z6899, Z6900, Z6901 , Z6902, Z6903, Z6904 and Z6905, respectively.

Knockout of ADH3 locus in above strains Z6897 to Z6905 was not confirmed. The genotype of strains Z6903 and its eight siblings with respect to wild type Y. lipolytica ATCC #20362 was: Ura ⁺ , Pex3 ^~ , unknown T, unknown 2 ^" , unknown 3 ^' , unknown 4 ^' , YALI0E12947g ^~ , unknown 6 ^' , YALI0B21890g ^~ , unknown 8 ^' , unknown 10 ^' , unknown 11 ^' , unknown 12 ^' , unknown 13 ^' , unknown 14 ^' , Sou2 ^' , unknown 15 ^' , YAT1 ::ME3S::Pex16, GPD::ME3S::Pex20, YAT1 ::ME3S::Lip1 , FBAINm::EgD9eS::Lip2, EXP1 ::EgD9eS::Lip1 , GPAT::EgD9e::Lip2,

YAT1 ::EgD9eS::Lip2, YAT1 ::EgD9eS-L35G::Pex20, FBAINm::EgD8M::Pex20, EXP1 ::EgD8M::Pex16, FBAIN::EgD8M::Lip1 , GPD::EaD8S::Pex16 (2 copies), YAT1 ::E389D9eS/EgD8M::Lip1 , SPS19LM::E389D9eS/EgD8M::Glo

YAT1 ::EgD9eS/EgD8M::Aco, FBAINm::EaD9eS/EaD8S::Lip2,

DGAT2M::YID9::Lip1 , GPDIN::YID9::Lip1 , GPD::FmD12::Pex20,

YAT1 ::FmD12::Oct, EXP1 ::FmD12S::Aco, ALK2LM1 ::FmD12S::Erp,

GPDIN::FmD12::Pex16, EXP1 ::EgD5M::Pex16, FBAIN::EgD5SM::Pex20, EXP1 ::EgD5SM::Lip1 , YAT1 ::EaD5SM::Oct, FBAINm::PaD17::Aco,

EXP1 ::PaD17::Pex16, YAT1 ::PaD17S::Lip1 , YAT1 ::YICPT1 ::Aco,

EXP1 ::YICPT1 ::Oct, YAT1 ::MCS::Lip1 , FBA::MCS::Lip1 , EXP1 ::YIPCT::Pex16, YAT1 ::Mal_PAAT1 S::Pex16, ALK2LM1 ::Mal_PAAT1 S::Pex20, FBAINm::YILPAAT1 ::Lip1 (2 copies), YAT1 ::YIPDAT::Lip1 (2 copies), SPS19- P3::YIPDAT::Lip1 .

Analysis of Total Lipid Content and Composition by Flask Assay

Cells of strains Z6903 and its siblings were grown and analyzed for total lipid content and fatty acid composition by the flask assay described above. Table 9 summarizes the DCW, the TFAs % DCW, the amount of each fatty acid as a weight percent of TFAs (% TFAs) and the EPA % DCW of strains from Z6103 to Z61 13.

Table 9. Total Lipid Content and Composition in Strain Z6903 and Its Siblings by Flask Assay

EPA

DCW TFAs % % TFAs EPA % Rate

Strain (a' ^L ) DCW 16:0 16:1 18:0 18:1 LA ALA EDA DGLA ARA ETrA ETA EPA DCW (g/L/h)

Z6897 4.4 47.6 2.0 0.7 2.2 5.5 15.9 0.8 3.7 5.1 0.8 0.5 3.6 52.1 24.8 0.0077

Z6898 4.1 41.0 1.9 0.6 2.5 5.5 16.4 0.8 4.0 5.0 0.8 0.6 3.4 50.9 20.9 0.0061

Z6899 4.7 46.1 1.9 0.8 1.9 6.0 15.6 0.8 3.6 5.1 0.7 0.5 3.5 52.5 24.2 0.0081

Z6900 4.8 46.0 1.7 0.7 2.0 5.4 15.2 0.7 3.9 5.6 0.8 0.6 4.0 52.3 24.0 0.0083

Z6901 4.9 48.7 2.0 0.7 2.3 5.9 16.3 0.8 3.9 5.0 0.9 0.5 3.6 51.2 24.9 0.0087

Z6902 4.6 50.9 2.1 0.7 2.5 6.0 16.8 0.9 3.8 4.9 0.8 0.5 3.5 50.6 25.7 0.0084

Z6903 4.9 49.1 2.0 0.7 2.2 5.7 16.4 0.8 3.8 5.0 0.8 0.5 3.6 51.4 25.2 0.0089

Z6904 4.5 48.2 2.1 0.7 2.3 5.9 16.2 0.8 3.8 5.0 0.8 0.6 3.5 51.1 24.6 0.0079

Z6905 4.8 52.8 2.0 0.7 2.7 6.3 17.6 0.9 3.7 4.7 0.7 0.4 3.2 50.0 26.4 0.0091

Table 9 shows that in strain Z6903, DCW was 4.9 g/L, TFAs % DCW was 49.1 , EPA % TFAs was 51 .4, and EPA % DCW was 25.2, which was the third highest EPA % DCW measurement among the analyzed strains. In strain Z6905, DCW was 4.8 g/L, TFAs % DCW was 52.8, EPA % TFAs was 50.0, and EPA % DCW was 26.4, which was the highest EPA % DCW measurement among the analyzed strains.

Strain Z6903 was further analyzed to determine the levels of the sugar alcohol by-products arabitol, mannitol and erythritol (Table 10). This analysis was also made with strain Z6109 (Example 1 ).

Table 10. Sugar Alcohols Produced bv Strains Z6109 and Z6903

Z6109 Z6903

SOU 2 gene + -

Arabitol (g/L) 4.9 0.0

Mannitol (g/L) 3.5 0.0

Erythritol (g/L) 0.8 2.9

Total sugar alcohols (g/L) 9.2 2.9

The results in Table 10 indicate that strain Z6903 produced no mannitol or arabitol by-products, suggesting that the Sou2 protein is essential for mannitol and arabitol biosynthesis in Y. lipolytica. Also, compared to strain Z6109, strain Z6903 produced about 68% less total sugar alcohol by-products. These results indicate that down-regulation of SOU2 expression in Y. lipolytica significantly decreases the level of sugar alcohol by-product production.

Example 3

Generation of Strain Z7418 Producing at Least about 49.8% EPA of Total Fatty

Acids with at Least about 49.3% Total Lipid Content This Example describes the generation of Y. lipolytica strain Z7418 through genetic modification of strain Z6903. The genetic modification entailed introducing expression cassettes encoding delta-8 desaturase, malonyl-CoA synthetase (MCS), and acyl-CoA:lysophosphatidic acid acyltransferase (LPAAT). The development of strain Z7418 was required in order to develop strain Z9276, which is described below in Example 4.

Figure 3B shows the modification steps and intermediate strain (Z6903U) used for generating strain Z7418. Strain Z6903U was generated as follows. Generation of Strain Z6903U

Construct pZKUM (above) was used to disrupt the URA3 gene in strain Z6903 that was introduced by the plasmid pZKADn-SyP298F (above). A total of eight 5-FOA-resistant transformants were grown and identified to possess a Ura ^~ phenotype. Individual transformants were grown as described above for initial fatty acid screening.

GC analyses showed the presence of 35.0%, 32.7% and 37.7% EPA in the TFAs of pZKUM-transformant strains #5, #6 and #7 picked from an FOA plate, which were designated as Z6903U5, Z6903U6 and Z6903U7, respectively. These three transformants were collectively designated as strain Z6903U.

Generation of Strain Z7418

Plasmid pZK16-Myl_8N was used to integrate gene cassettes for expressing a synthetic mutant delta-8 desaturase derived from E. gracilis

(YAT1 ::EgD8M::Pex20; EgD8M is SEQ ID NO:19), a codon-optimized malonyl- CoA synthetase derived from Rhizobium leguminosarum bv. viciae 3841

(FBA::MCS::Lip1 ; MCS is SEQ ID NO:21 ), and a Y. lipolytica acyl- CoA:lysophosphatidic acid acyltransferase (YAT1 ::YILPAAT1 ::Lip1 ; YILPAAT1 is SEQ ID NO:23) into the YALI0B14795p locus (GenBank Accession No.

XM_500900) of strain Z6903U. The construction of plasmid pZK16-Myl_8N has been described (U.S. Pat. Appl. Publ. No. 2012-0052537, see Table 1 1 therein, which is incorporated herein by reference).

The pZK16-Myl_8N plasmid was digested with Asc\, and then used to transform strain Z6903U5. The transformed cells were plated onto uracil-minus MM plates and maintained at 30 °C for 5 to 6 days. Single colonies were grown as described above for initial fatty acid screening.

GC analyses showed that almost all of the selected 96 strains of Z6903U5 transformed with pZK16-Myl_8N produced more than 50.0% EPA of TFAs. Ten strains (#8, #9, #34, #40, #43, #58, #59, #70, #79, #80) produced about 52.1 %, 50.3%, 51 .5%, 51 .1 %, 53.5%, 51 .4%, 50.8%, 52.2%, 51 .2% and 51 .6% EPA of TFAs and were designated as Z7416, Z7417, Z7418, Z7419, Z7420, Z7421 , Z7422, Z7423, Z7424 and Z7425, respectively.

Knockout of the YALI0B14795p locus in above strains Z7416 to Z7425 was not confirmed. The genotype of strains Z7418 and its nine siblings with respect to wild type Y. lipolytica ATCC #20362 was: Ura ⁺ , Pex3 ^~ , unknown T, unknown 2 ^" , unknown 3 ^' , unknown 4 ^' , YALI0E12947g ^~ , unknown 6 ^' ,

YALI0B21890g ^~ , unknown 8 ^' , unknown 10 ^' , unknown 11 ^' , unknown 12 ^' , unknown 13 ^' , unknown 14 ^' , Sou2 ^' , unknown 15 ^' , unknown 16 ^' , YAT1 ::ME3S::Pex16, GPD::ME3S::Pex20, YAT1 ::ME3S::Lip1 , FBAINm::EgD9eS::Lip2,

EXP1 ::EgD9eS::Lip1 , GPAT::EgD9e::Lip2, YAT1 ::EgD9eS::Lip2,

YAT1 ::EgD9eS-L35G::Pex20, FBAINm::EgD8M::Pex20, EXP1 ::EgD8M::Pex16, FBAIN::EgD8M::Lip1 , YAT1 ::EgD8M::Pex20, GPD::EaD8S::Pex16 (2 copies), YAT1 ::E389D9eS/EgD8M::Lip1 , SPS19LM::E389D9eS/EgD8M::Glo

YAT1 ::EgD9eS/EgD8M::Aco, FBAINm::EaD9eS/EaD8S::Lip2,

DGAT2M::YID9::Lip1 , GPDIN::YID9::Lip1 , GPD::FmD12::Pex20,

YAT1 ::FmD12::Oct, EXP1 ::FmD12S::Aco, ALK2LM1 ::FmD12S::Erp,

GPDIN::FmD12::Pex16, EXP1 ::EgD5M::Pex16, FBAIN::EgD5SM::Pex20, EXP1 ::EgD5SM::Lip1 , YAT1 ::EaD5SM::Oct, FBAINm::PaD17::Aco,

EXP1 ::PaD17::Pex16, YAT1 ::PaD17S::Lip1 , YAT1 ::YICPT1 ::Aco,

EXP1 ::YICPT1 ::Oct, YAT1 ::MCS::Lip1 , FBA::MCS::Lip1 (2 copies),

EXP1 ::YIPCT::Pex16, YAT1 ::MaLPAAT1 S::Pex16,

ALK2LM1 ::MaLPAAT1 S::Pex20, FBAINm::YILPAAT1 ::Lip1 (2 copies),

YAT1 ::YILPAAT1 ::Lip1 , YAT1 ::YIPDAT::Lip1 (2 copies), SPS19- P3::YIPDAT::Lip1 .

Analysis of Total Lipid Content and Composition by Flask Assay

Cells of strains Z7418 and its siblings were grown and analyzed for total lipid content and fatty acid composition by the flask assay described above. Table 1 1 summarizes the DCW, the TFAs % DCW, the amount of each fatty acid as a weight percent of TFAs (% TFAs) and the EPA % DCW of strains from Z7416 to Z7425.

Table 1 1 . Total Lipid Content and Composition in Strain Z7418 and Its Siblings by Flask Assay

TFAs EPA

% TFAs

DCW % EPA % Rate

Strain DCW 16:0 16:1 18:0 18:1 LA ALA EDA DGLA ARA ETrA ETA EPA DCW (g/L/h)

Z7416 5.7 48.7 1.9 0.7 2.4 5.8 16.5 0.9 3.7 4.6 1.1 0.8 3.8 50.0 24.3 0.0099

Z7417 5.9 52.2 2.0 0.8 2.2 7.2 17.5 1.0 4.1 4.6 0.9 0.6 3.5 47.7 24.9 0.0106

Z7418 5.9 49.3 1.9 0.7 2.2 6.0 16.4 0.9 3.8 4.7 1.0 0.8 3.8 49.8 24.6 0.0103

Z7419 5.7 49.8 1.9 0.7 2.3 6.1 16.7 0.9 3.8 4.5 1.0 0.8 3.7 49.7 24.7 0.0100

Z7420 5.2 51.1 2.0 0.8 2.4 6.9 17.4 1.0 4.1 4.3 0.9 0.7 3.4 48.5 24.8 0.0092

Z7421 5.0 48.8 1.9 0.6 2.4 5.7 16.2 0.9 3.7 4.5 1.0 0.8 3.7 50.5 24.6 0.0088

Z7422 5.2 52.5 2.1 0.8 2.3 6.9 17.2 0.9 4.0 4.4 0.9 0.7 3.3 48.4 25.4 0.0094

Z7423 6.2 48.2 2.0 0.8 2.3 6.8 17.5 1.0 3.8 4.5 1.0 0.8 3.7 48.2 23.2 0.0102

Z7424 5.9 49.3 1.9 0.7 2.3 5.9 16.6 0.9 3.7 4.6 1.0 0.7 3.8 49.9 24.6 0.0103

Z7425 6.2 48.7 2.0 0.7 2.4 6.0 16.7 1.0 3.6 4.7 0.9 0.6 3.8 49.7 24.2 0.0107

Table 1 1 shows that in strain Z7418, DCW was 5.9 g/L, TFAs % DCW was 49.8, EPA % TFAs was 49.3, and EPA % DCW was 24.6.

Example 4

Generation of Strain Z9276 Producing at Least about 57.5% EPA of Total Fatty

Acids with at Least about 56.9% Total Lipid Content

This Example describes the generation of Y. lipolytica strain Z9276 through genetic modification of strain Z7418. The genetic modification entailed introducing an expression cassette encoding a mutant acyl CoA:

lysophosphatidylcholine acyltransferase (LPCAT). The construction and analysis of this and other mutant LPCATs is described in Examples 6-9.

Figure 3B shows the modification steps and intermediate strain (Z7418U) used for generating strain Z9276. Strain Z7418U was generated as follows. Generation of Strain Z7418U

Construct pZKUM (above) was used to disrupt the URA3 gene in strain Z7418 that was introduced by the plasmid pZK16-MyL8N (above). A total of twenty-four 5-FOA-resistant transformants were grown and identified to possess a Ura ^~ phenotype. These transformants were grown as described above for initial fatty acid screening.

GC analyses showed the presence of 35.7%, 36.1 %, 32.3% 35.4 and 33.6% EPA in the TFAs of B group pZKUM-transformant strains #2, #3, #6, #7 and #8 picked from an FOA-plate, which were designated as Z7418BU1 , Z7418BU2, Z7418BU3, Z7418BU4 and Z7418BU5, respectively. GC analyses also showed the presence of 35.9% EPA in the TFAs of C group pZKUM- transformant strain #6 picked from an FOA-plate, which was designated as Z7418CU1 . GC analyses also showed the presence of 24.2%, 34.9% and 34.0% EPA in the TFAs of D group pZKUM-transformant strains #3, #4 and #7 picked from an FOA-plate, which were designated as Z7418DU1 , Z7418DU2 and Z7418DU3, respectively.

Generation of Strain Z9276

Plasmid pZKMPn-YD58 (Figure 7B, SEQ ID NO:24) was generated to integrate a gene cassette for expressing a double-mutant Y. lipolytica acyl CoA:lysophosphatidylcholine acyltransferase (YILPCAT [M136S/T389A], SEQ ID NO:26) into the D-arabinitol 2-dehydrogenase locus (GenBank Accession No. XP_504895) of strain Z7418U. Table 12 describes the components contained in pZKMPn-YD58.

Table 12. Description of Plasmid pZKMPn-YD58 (SEQ ID NO:24)

The pZKMPn-YD58 plasmid was digested with Asc\, and then used to transform strain Z7418BU1 . The transformed cells were plated onto uracil-minus MM plates and maintained at 30 °C for 5 to 6 days. Single colonies were grown as described above for initial fatty acid screening.

GC analyses showed that nine of the selected 48 strains of Z7418BU1 transformed with pZKMPn-YD58 produced more than 55.0% EPA of TFAs.

These nine strains (#1 1 , #13, #14, #15, #24, #25, #33, #37, #48) produced about 55.9%, 55.9%, 55.0%, 56.3%, 56.1 %, 57.1 %, 55.3%, 55.1 % and 56.1 % EPA of TFAs and were designated as Z9256, Z9257, Z9258, Z9259, Z9260, Z9261 , Z9262, Z9263 and Z9264, respectively.

GC analyses showed that eleven of the selected 60 strains of Z7418BU2 transformed with pZKMPn-YD58 produced more than 54.7% EPA of TFAs.

These eleven strains (#1 , #5, #8, #10, #18, #26, #30, #35, #42, #45, #54) produced about 56.5%, 54.8%, 57.9%, 56.1 %, 56.1 %, 57.3%, 58.8%, 54.9%. 54.7%, 55.4% and 55.5% EPA of TFAs and were designated as Z9265, Z9266, Z9267, Z9268, Z9269, Z9270, Z9271 , Z9272, Z9273, Z9274 and Z9275, respectively.

GC analyses showed that four of the selected 44 strains of Z7418DU3 transformed with pZKMPn-YD58 produced more than 55.8% EPA of TFAs. These four strains (#10, #12, #15, #16) produced about 56.5%, 55.8%, 55.9% and 56.2% EPA of TFAs and were designated as Z9276, Z9277, Z9278 and Z9279, respectively.

Knockout of the D-arabinitol 2-dehydrogenase locus in above strains Z9256 to Z9279 was not confirmed. The genotype of these strains, including Z9276, with respect to wild type Y. lipolytica ATCC #20362 was: Ura ⁺ , Pex3 ^~ , unknown T, unknown 2 ^" , unknown 3 ^' , unknown 4 ^' , YALI0E12947g ^~ , unknown 6 ^' , YALI0B21890g ^~ , unknown 8 ^' , unknown 10 ^' , unknown 11 ^' , unknown 12 ^' , unknown 13 ^' , unknown 14 ^' , Sou2 ^' , unknown 15 ^' , unknown 16 ^' , unknown 17 ^' ,

YAT1 ::ME3S::Pex16, GPD::ME3S::Pex20, YAT1 ::ME3S::Lip1 ,

FBAINm::EgD9eS::Lip2, EXP1 ::EgD9eS::Lip1 , GPAT::EgD9e::Lip2,

YAT1 ::EgD9eS::Lip2, YAT1 ::EgD9eS-L35G::Pex20, FBAINm::EgD8M::Pex20, EXP1 ::EgD8M::Pex16, FBAIN::EgD8M::Lip1 , YAT1 ::EgD8M::Pex20,

GPD::EaD8S::Pex16 (2 copies), YAT1 ::E389D9eS/EgD8M::Lip1 ,

SPS19LM::E389D9eS/EgD8M::Glo YAT1 ::EgD9eS/EgD8M::Aco,

FBAINm::EaD9eS/EaD8S::Lip2, DGAT2M::YID9::Lip1 , GPDIN::YID9::Lip1 , GPD::FmD12::Pex20, YAT1 ::FmD12::Oct, EXP1 ::FmD12S::Aco,

ALK2LM1 ::FmD12S::Erp, GPDIN::FmD12::Pex16, EXP1 ::EgD5M::Pex16, FBAIN::EgD5SM::Pex20, EXP1 ::EgD5SM::Lip1 , YAT1 ::EaD5SM::Oct,

FBAINm::PaD17::Aco, EXP1 ::PaD17::Pex16, YAT1 ::PaD17S::Lip1 ,

YAT1 ::YICPT1 ::Aco, EXP1 ::YICPT1 ::Oct, YAT1 ::MCS::Lip1 , FBA::MCS::Lip1 (2 copies), EXP1 ::YIPCT::Pex16, YAT1 ::Mal_PAAT1 S::Pex16,

ALK2LM1 ::Mal_PAAT1 S::Pex20, FBAINm::YILPAAT1 ::Lip1 (2 copies),

YAT1 ::YILPAAT1 ::Lip1 , YAT1 ::YIPDAT::Lip1 (2 copies), SPS19- P3::YIPDAT::Lip1 , YAT1 ::YILPCAT(M136S/T389A)::Lip1 . Analysis of Total Lipid Content and Composition by One-Step Flask Assay

Cells of strains Z9256 to Z9279, including Z9276, were grown and analyzed for total lipid content and fatty acid composition by the "one-step" flask assay, which is described as follows:

One loop of freshly streaked cells was inoculated into 3 ml_ One-Step Flask medium (recipe described below) and grown overnight at 250 rpm and 30 °C. The OD ₆ oonm of the culture was measured and an aliquot of cells from the culture was added to a final OD ₆ oonm of 0.25 in 15 mL of One-Step Flask medium in a 125-mL flask. After one day in a shaker incubator at 250 rpm and at 30 °C, 10 mL of a high glucose medium (80 g/L glucose, 1 .9 g/L KH ₂ PO , 6.3 g/L K ₂ HPO ₄ , 8.4 g/L NaHCO ₃ , pH 7.2) was added into the same flask. After 5 days in a shaker incubator at 250 rpm and at 30 °C, a 1 -mL aliquot was used for fatty acid analysis and 10 mL dried for dry cell weight determination. The fatty acid and DCW analyses were performed as described above.

One-Step flask assay media: 0.5 g/L urea, 2.5 g/L Yeast extract, 3.0 g/L KH ₂ PO ₄ , 1 .7 g/L Na ₂ PO ₄ i 2H ₂ O, 20 g/L D-glucose, 0.2 ml/L trace metal solution (100X), 0.25 g/L MgSO ₄ 7H2O, 0.15 mg/L thiamine HCI.

Trace metals solution (100X): 10 g/L citric acid, 1 .5 g/L CaCI ₂ 2H ₂ O, 10 g/L FeSO ₄ 7H ₂ O, 0.39 g/L ZnSO ₄ 7H ₂ O, 0.38 g/L CuSO ₄ 5H ₂ O, 0.2 g/L

CoCI ₂ -6H ₂ O, 0.3 g/L MnCI ₂ 4H ₂ O.

Table 13 summarizes the DCW, the TFAs % DCW, the amount of each fatty acid as a weight percent of TFAs (% TFAs) and the EPA % DCW of strains from Z9256 to Z9279.

Table 13. Total Lipid Content and Composition in Strains Z9256 to Z9279, Including Z9276, by Flask Assay

Strain DCW 16:0 16:1 18:0 18:1 LA ALA EDA DGLA ARA ETrA ETA EPA DCW (g/L/h)

Z9256 7.1 61.3 2.1 0.7 2.0 7.3 14.3 0.3 4.9 5.3 1.4 0.5 2.4 52.9 32.4 0.0165

Z9257 7.0 61 .0 1 .9 0.6 2.0 7.3 14.4 0.3 4.8 5.3 1.4 0.6 2.6 52.7 32.2 0.0162

Z9258 7.1 61 .0 1 .9 0.6 2.0 7.5 15.8 0.4 4.9 5.0 1.3 0.6 2.6 51 .2 31.2 0.0158

Z9259 7.0 61.2 2.0 0.6 2.0 7.2 13.5 0.3 4.8 5.5 1.5 0.5 2.4 53.9 33.0 0.0164

Z9260 7.1 61 .5 2.1 0.7 2.1 7.6 14.9 0.3 5.0 5.2 1.4 0.5 2.4 51 .9 31.9 0.0161

Z9261 6.3 62.0 1 .9 0.6 1.9 7.9 13.8 0.3 4.9 5.1 1.4 0.5 2.3 53.5 33.2 0.0150

Z9262 7.1 61 .2 2.0 0.6 2.0 7.7 15.0 0.4 5.0 5.3 1.3 0.5 2.5 51 .9 31.8 0.0162

Z9263 7.3 59.7 2.0 0.7 2.0 7.4 14.7 0.4 4.9 5.3 1.3 0.5 2.5 52.5 31.4 0.0163

Z9264 5.8 49.5 2.3 0.8 1.7 6.8 13.8 0.5 4.2 5.3 0.9 0.7 4.3 51 .7 25.6 0.0106

Z9265 5.1 50.6 1 .9 0.8 1.3 5.6 10.2 0.3 3.6 4.7 1.1 0.7 2.7 60.0 30.4 0.0109

Z9266 6.7 62.1 2.2 0.8 2.0 7.5 14.9 0.5 4.7 4.7 0.9 0.5 2.8 52.2 32.4 0.0155

Z9267 6.2 53.3 2.4 0.9 1.9 5.3 1 1.3 0.4 3.6 4.4 1.2 0.7 2.4 57.7 30.8 0.0135

Z9268 6.2 58.5 1 .7 0.7 1.8 6.9 13.2 0.4 4.3 4.8 0.9 0.7 3.4 54.4 31.8 0.0141

Z9269 5.9 56.1 1 .7 0.7 1.7 6.9 13.4 0.5 4.3 4.8 0.9 0.6 3.3 54.5 30.6 0.0130

Z9270 5.7 52.7 2.0 0.7 1.7 5.9 12.2 0.4 4.1 4.0 1.0 0.7 2.4 57.6 30.3 0.0124

Z9271 5.7 51 .1 2.1 0.8 1.7 4.5 8.8 0.2 3.3 4.3 1.3 0.7 2.2 62.2 31.8 0.0129

Z9272 6.1 56.8 1 .7 0.7 1.7 7.0 12.9 0.4 4.3 4.8 0.9 0.7 3.4 54.5 31.0 0.0135

Z9273 5.8 56.2 2.0 0.8 1.9 6.8 1 1.2 0.3 4.2 5.0 1.0 0.6 3.0 56.8 31.9 0.0133

Z9274 6.1 56.7 1 .7 0.7 1.7 7.1 12.7 0.4 4.2 4.9 0.9 0.6 3.3 55.5 31.4 0.0137

Z9275 6.2 57.0 1 .7 0.7 1.7 6.9 12.8 0.4 4.3 4.9 1.0 0.7 3.4 54.5 31.1 0.0138

Z9276 6.1 56.9 1.8 0.7 1.8 6.8 10.7 0.2 4.3 5.7 1.7 0.5 2.5 57.5 32.7 0.0143

Z9277 6.3 57.6 1 .7 0.6 1.7 7.0 13.6 0.3 5.0 5.2 1.6 0.7 2.6 53.9 31.0 0.0140

Z9278 6.0 56.0 1 .7 0.7 1.6 7.1 14.0 0.3 4.7 5.3 1.6 0.5 2.4 53.7 30.0 0.0129

Z9279 5.7 60.9 2.0 0.8 1.7 7.3 13.7 0.4 4.9 4.7 1.8 0.5 1.8 53.7 32.7 0.0133 avg 6.3 57.5 1.9 0.7 1.8 6.9 13.2 0.4 4.5 5.0 1 .2 0.6 2.7 54.6 31.4 0.0142 a GLA was not detected in the oil produced by any of strains Z9256-Z9279.

Table 13 shows that with the exception of Z9264, all of the strains tested produced more than 30.0% EPA as DCW. The EPA profiles in strains Z9256, Z9259 and Z9276 were notable and are summarized as follows. In strain Z9256, TFAs % DCW was 61 .3%, EPA % TFAs was 52.9%, and EPA % DCW was 32.4% with an EPA productivity of 0.0165 g/L/h. In strain Z9259, TFAs % DCW was 61 .2%, EPA % TFAs was 53.9%, and EPA % DCW was 33.0% with an EPA productivity of 0.0164 g/L/h. In strain Z9276, TFAs % DCW was 56.9%, EPA % TFAs was 57.5%, and EPA % DCW was 32.7% with an EPA productivity of 0.0143 g/L/h.

The average oil content on a dry cell weight basis in the strains listed in Table 13 was 57.5% (TFAs % DCW). This average oil content is higher than the oil content of strain Z5585 (56.6%, Table 4) from which the strains in Table 13 were derived. More notably, the average EPA content in the total fatty acids of the oil (EPA % TFAs) in the strains in Table 13 was 54.6%, which was about 10.5% greater than the same measurement for Z5585 (49.4%, Table 4). This increase in the EPA content in the total fatty acids of the oil, while maintaining the total amount of oil produced relative to Z5585, resulted in a higher average amount of EPA on a dry cell weight basis (31 .4%) in the strains of Table 13 compared to Z5585 (28%, Table 4). This represented an increase of about 12.1 %, which is consistent with the -10.5% increase observed with oil content.

Several strains in Table 13 exhibited significant increases in oil and EPA content relative to strain Z5585. For example, the total oil content (TFAs % DCW) of strains Z9256 and Z9259 increased by about 8.3% and 8.1 %, respectively, compared to the total oil content in strain Z5585 (Table 4). The EPA content in the total fatty acids of the oil (EPA % TFAs) in strains Z9256 and Z9259 was increased by about 7.1 % and 9.1 % compared to the same

measurement in Z5585 (Table 4). These robust increases in the oil content and EPA content in the total fatty acids of the oil resulted in increases of about 15.7% and 17.9% in the total EPA content (EPA % DCW) of strains Z9256 and Z9259, respectively, compared to the total EPA content of Z5585 (Table 4). Another strain in Table 13, Z9276, exhibited a marginal increase in oil content compared to Z5585, but had a 16.4% increase in EPA content in the total fatty acids of the oil (EPA % TFAs) compared to the same measurement in Z5585 (Table 4). This substantial increase in EPA % TFAs resulted in an increase of about 16.8% in the total EPA content (EPA % DCW) of strain Z9276 compared to the total EPA content of Z5585 (Table 4).

The genetic modifications used to produce the strains in Table 13 from strain Z5585 (Table 4) thus resulted in total oil contents (TFAs % DCW) that were mostly similar to or greater than the total oil content of Z5585. More significantly, this maintenance or increase in total oil content was not coupled to a decrease in the EPA content in the total fatty acids of the oil (EPA % TFAs), which has generally been a problem in previous efforts to increase total EPA production. This previous problem is reflected in the data in Table 4, for example, which shows that increases in oil content (TFAs % DCW) through other genetic modifications generally led to decreases in the EPA content in the total fatty acids of the oil (EPA % TFAs). Thus, previous increases in total EPA content (EPA % DCW) were driven largely in part by increasing total oil production. The strains in Table 13 on the other hand exhibited increased total EPA content (EPA % DCW) that was driven in large part by increasing EPA content in the total fatty acids of the oil while maintaining or increasing total oil content.

It was notable that almost all the strains in Table 13 had very low levels of stearic acid as a percentage of total fatty acids in the oil (18:0 % TFAs).

Specifically, with the exception of strain Z9261 , all the strains had 2.0% or less stearic acid by weight of total fatty acids. The average level for all the strains was 1 .8% stearic acid.

The average dry cell weight of the strains in Table 13 (6.3 g/L) was substantially increased compared to the dry cell weights of strain Z5585 (4.6 g/L) and the other strains listed in Table 4. This represents an average increase of about 37% compared to the dry cell weight of Z5585. Certain individual strains in Table 13, such as Z9256, Z9259 and Z9263, exhibited increases in dry cell weight of about 54.3%, 52.2% and 58.7%, respectively, compared to Z5585. In flask assays, previous strain Z5567 (Table 4) was shown to produce about 0.45 g organic acids/g DCW, and about 0.50 g sugar alcohol/g DCW byproducts. Strain Z9276 on the other hand produced 0.27 g organic acids/g DCW, and 0.18 g sugar alcohol/g DCW. Given the increased production of EPA in strain Z9276 (Table 13) versus that of Z5567 (Table 4), and the decreased production of by-products by Z9276, it is apparent that Z9276 has enhanced carbon flux toward EPA production.

In summary, the genetic modifications employed in Examples 2-4 to develop the strains of Table 13 from strain Z5585 were as follows. In Example 2, Z5585 was first modified to down-regulate expression of the Sou2 protein to yield strain Z5585K2U. Gene cassettes for over-expressing PDAT, delta-12

desaturase and a DGLA synthase multizyme were then introduced to yield strain Z6903. In Example 3, gene cassettes were introduced for over-expressing delta- 8 desaturase, MCS and LPAAT, thereby yielding strain Z7418. Finally, in the present Example, a gene cassette for over-expressing a mutant LPCAT was introduced to yield the strains of Table 13. All except one of these strains can produce an oil containing at least 30 percent EPA measured as a weight percent of dry cell weight.

Example 5

Identification of the SOU2 Gene as a Genetic Target to Modify Lipid Production and Sugar Alcohol Production in Y. Iipolytica

This Example describes that the Y. Iipolytica SOU2 gene, which encodes Sou2 sorbitol utilization protein, can regulate the level of lipids and certain sugar alcohols in Y. Iipolytica. Specifically, it was shown that disrupting the SOU2 gene in a Y. Iipolytica strain increased the amount of oil, and decreased the amount of arabitol and mannitol, produced by the strain.

The SOU2 gene was identified as a genetic target to modify lipid

production in Y. Iipolytica during the development of strain Z3041 from strain Z1978. The steps involved in this process involved intermediate strains Z1978U, Z2636 and Z2636U (Figure 8). The development of Y. Iipolytica strain Z1978U is described in U.S. Pat. Appl. Publ. No. 2012-0052537, which is incorporated herein by reference.

Strain Z2636 was produced by transforming Z1978U with plasmid pZKT2- ML9DCB (SEQ ID NO:27, Figure 9A), which contains cassettes for expressing Y. Iipolytica diacylglycerol cholinephosphotransferase (YICPT1 ; U.S. Pat. No.

7,932,077), Y. Iipolytica delta-9 desaturase (YID9; U.S. Pat. Appl. Publ. No.

2012-0052537), and an M. alpina acyl-CoA:lysophosphatidic acid acyltransferase nucleotide sequence that was codon-optimized for expression in Y. Iipolytica (MaLPAATI S; U.S. Pat. No. 7,879,591 ). Construct pZKUM (above) was then used to disrupt the URA3 gene in strain Z2636 that was introduced by plasmid pZKT2-ML9DCB, thereby producing the Ura ^~ strain Z2636U.

Strain Z3041 was produced by transforming Z2636U with Asc\/Sph\- digested plasmid pZKLY-PP2YAP (SEQ ID NO:28, Figure 9B), which contains cassettes for expressing Y. Iipolytica Yap1 (YIYAP1 , GenBank Accession No. XM_504945), Y. Iipolytica 6-phosphogluconolactonase (YI6PGL, U.S. Pat. Appl. Publ. No. 201 1 -0244512), and Y. Iipolytica glucose-6-phosphate dehydrogenase (YIG6PDH, with 440-bp intron; U.S. Pat. Appl. Publ. No. 201 1 -0244512). Z3041 was produced along with sibling strains Z3030-Z3040 and Z3042-Z3050.

Analysis of Total Lipid Content and Composition by Flask Assay

Cells of strains Z3029 to Z3050, including Z3041 , were grown and analyzed for total lipid content and fatty acid composition by the flask assay described above. Table 14 summarizes the DCW, the TFAs % DCW, the amount of each fatty acid as a weight percent of TFAs (% TFAs) and the EPA % DCW of strains from Z3029 to Z3050. Each value represents an average of two measurements. Table 14. Total Lipid Content and Composition in Strains Z3029 to Z3050, Including Z3041 , by Flask Assay

% TFAs

DCW TFAs % EPA %

Strain DCW 16:0 16:1 18:0 18:1 LA ALA EDA DGLA ARA ETrA ETA EPA DCW

Z3029 3.5 41.5 1.7 0.8 1.7 6.1 12.7 0.7 3.7 3.9 0.9 0.6 2.3 55.3 23.0

Z3030 3.3 38.8 1.9 0.5 2.6 5.4 12.4 0.8 3.3 3.2 0.7 0.7 2.5 57.3 22.2

Z3031 3.3 38.7 2.0 0.5 2.6 5.5 12.4 0.8 3.3 3.2 0.7 0.7 2.5 57.1 22.1

Z3032 3.2 38.5 2.3 0.6 2.4 5.3 12.4 0.7 3.3 3.3 0.7 0.6 2.3 57.3 22.1

Z3033 3.2 38.2 2.3 0.7 2.3 5.4 12.2 0.7 3.1 3.4 0.8 0.5 2.3 57.2 21.9

Z3034 3.3 38.8 2.0 0.5 2.6 5.4 12.4 0.8 3.4 3.3 0.7 0.7 2.6 56.8 22.0

Z3035 3.3 39.1 1.6 0.6 1.7 5.3 1 1.9 0.6 3.4 3.9 0.9 0.6 2.3 57.2 22.3

Z3036 2.9 31.3 1.2 0.3 2.2 3.4 8.3 0.5 2.5 3.2 0.5 1.1 3.4 63.7 19.9

Z3037 3.1 40.0 1.6 0.6 1.7 5.1 1 1.9 0.6 3.4 3.9 0.9 0.6 2.3 57.6 23.1

Z3038 3.2 39.8 1.8 0.7 1.7 5.9 12.9 0.7 3.6 3.8 0.9 0.5 2.3 55.5 22.1

Z3039 3.1 39.2 1.7 0.6 1.7 5.4 12.2 0.6 3.5 3.8 0.9 0.5 2.3 56.9 22.3

Z3040 3.0 38.0 1.4 0.6 1.5 4.6 10.9 0.5 3.1 4.2 1.0 0.5 2.4 59.4 22.5

Z3041 3.3 42.1 1.3 0.5 1.5 5.8 12.2 0.7 4.2 3.6 0.9 0.9 2.5 56.8 23.9

Z3042 2.5 34.6 1.9 0.6 2.1 4.4 10.8 0.5 3.1 3.9 0.8 0.6 2.6 58.5 20.2

Z3043 3.0 40.0 1.6 0.6 1.7 5.2 12.0 0.6 3.5 3.9 0.9 0.5 2.3 57.4 22.9

Z3044 2.9 36.1 2.1 0.8 1.6 5.3 1 1.8 0.5 3.3 3.9 1.0 0.4 2.2 56.6 20.4

Z3045 3.0 39.3 1.9 0.5 2.6 5.2 12.1 0.8 3.2 3.2 0.7 0.7 2.5 57.5 22.6

Z3046 2.9 35.5 2.2 0.6 2.0 4.7 10.7 0.6 2.7 3.2 0.8 0.4 2.0 59.7 21.2

Z3047 3.0 38.1 2.2 0.6 2.6 5.2 12.0 0.7 3.2 3.4 0.7 0.5 2.3 57.2 21.8

Z3048 3.1 39.4 1.9 0.5 2.5 5.4 12.4 0.8 3.3 3.3 0.7 0.7 2.5 57.0 22.5

Z3049 2.8 36.5 2.2 0.6 2.2 5.1 1 1.9 0.7 3.0 3.4 0.8 0.5 2.2 57.9 21.1

Z3050 2.9 40.1 2.1 0.6 2.3 5.8 13.0 0.8 3.5 3.4 0.8 0.6 2.4 55.0 22.1 avg 3.1 38.3 1.9 0.6 2.1 5.2 1 1.9 0.7 3.3 3.6 0.8 0.6 2.4 57.5 22.0

Strain Z3041 had the highest oil content (about 42.1 TFAs % DCW) compared to all of its sibling strains (Table 14). The oil content of Z3041 was also notable since most of its siblings had less than 40% oil, and the average oil content was 38.3% (Table 14). The average increase in oil content in strain Z3041 with respect to its siblings was about 10.7%. These

observations suggested that the integrated sequence from pZKLY-PP2YAP in strain Z3041 provided an additional effect beyond the effects provided to all of the other strains in Table 14.

One possibility was that the integration itself provided an additional trait to Z3041 that contributed to oil synthesis. This prospect led to genome walking and sequencing analyses to determine what genetic trait(s) may have been altered in strain Z3041 to endow enhanced oil levels. The additional analyses found that the -10.5 kbp Asc\/Sph\ fragment of pZKLY-PP2YAP, which contains the YIYAP1 , YI6PGL and YI6PDH expression cassettes, had integrated into the promoter of the SOU2 gene (locus YALI0D18964g,

GenBank Accession No. XM_503010, Figure 5). Specifically, the Asc\/Sph\ plasmid fragment was integrated at -70 with respect to the ATG start codon of the SOU2 gene; this location is immediately downstream the presumptive TATA promoter consensus sequence. Given the location of the integration in the SOU2 promoter, the integration was predicted to down-regulate SOU2 gene expression (reduced transcription).

Aside from having increased oil content, strain Z3041 also exhibited lower levels of certain fermentation by-products compared to Z2636.

Specifically, production of the sugar alcohols arabitol and mannitol in Z3041 was eliminated.

Additional studies were conducted to understand the role of the SOU2 gene in regulating oil and sugar alcohol production in Yarrowia (refer to Example 2). Briefly, the SOU2 gene was knocked out in strain Z5585, thereby producing strain Z5585K2U (Figure 3). The knock-out of SOU2 entailed targeting and deleting about 522 base pairs of sequence beginning about 235 base pairs upstream the ATG start site and ending about 287 base pairs downstream the ATG start site (Figure 5, refer to sequence between opposing triangles). SOU2 gene transcription and Sou2 protein expression were thus completely down-regulated in strain Z5585K2U and its descendent strains.

EXAMPLE 6

Synthesis of Plasmid DY306-N Comprising Variant YILPCAT

This Example and Examples 7-9 are disclosed in U.S. Appl. No.

61 /661 ,623, which is incorporated herein by reference.

The wild type Y. lipolytica LPCAT (YILPCAT) polynucleotide sequence and amino acid sequence are represented by SEQ ID NOs:39 and 40, respectively.

The present example describes the construction of a Yarrowia autonomously replicating vector comprising a variant YILPCAT sequence (plasmid pY306-N, Figure 10, SEQ ID NO:42). The variant YILPCAT polynucleotide sequence, designated herein as YILPCAT* (SEQ ID NO:41 ), lacks two Nco\ restriction enzyme sites that are present in the wild type YILPCAT coding region. Removal of these internal Nco\ sites facilitated subsequent cloning procedures. YILPCAT* encodes wild type YILPCAT protein (SEQ ID NO:40).

As a control, the wild type YILPCAT polynucleotide sequence (SEQ ID

NO:39) was cloned into a Yarrowia autonomously replicating vector to result in plasmid pY306 (SEQ ID NO:43), comprising a ColE1 plasmid origin of replication, an ampicillin-resistance gene, an f1 origin of replication and the Y. lipolytica URA3 gene (GenBank Accession No. AJ306421 ).

The variant YILPCAT* sequence was synthesized by GenScript

Corporation (Piscataway, NJ). Two internal Nco\ restriction sites were removed by creation of silent mutations, while Nco\ and Not\ sites were added, respectively, at the 5' and 3' ends of the YILPCAT open reading frame to facilitate cloning. Specifically, an A12T mutation (i.e., a change from adenosine [A] in YILPCAT (SEQ ID NO:39) at position 12 to thymine [T] in the YILPCAT* variant) and a T918C mutation (i.e., a change from thymine [T] in YILPCAT (SEQ ID NO:39) at position 918 to cytosine [C] in the YILPCAT ^* variant) were introduced into the YILPCAT coding sequence. These two nucleotide substitutions were silent with respect to the amino acids encoded by the variant sequence. The nucleotide sequence encoding the variant YILPCAT lacking its internal Nco\ sites (i.e., YILPCAT ^* ) is represented by SEQ ID NO:41 , while the amino acid sequence encoded thereby is represented by SEQ ID NO:40, which is wild type YILPCAT protein.

YILPCAT ^* was subsequently cloned into plasmid pY306, thereby producing pY306-N (SEQ ID NO:42; Figure 10). Construct pY306-N contained the following components:

Table 15. Components of Plasmid pY306-N (SEQ ID NO:42)

Plasmid pY306-N was used to prepare single- and double-mutants of YILPCAT protein, as described below in Examples 7 and 9, respectively.

EXAMPLE 7

Desiqninq and Svnthesizinq Mutant Yarrowia LPCAT Enzvmes with Modified

Motifs

Based on the premise that conserved amino acid motifs within

YILPCAT are likely involved in catalysis, it was concluded that generation of mutants having variant motifs could result in the identification of an LPCAT enzyme having improved functional activity. A series of single amino acid substitutions were designed within the conserved sequence spanning amino acid residues 132 to 148 of SEQ ID NO:40 (i.e., Motif I) and the conserved sequence spanning amino acid residues 376 to 390 of SEQ ID NO:40 (i.e., Motif II). Within Motif I, a total of 195 amino acid substitutions were designed, as shown in Table 16, by creating various substitutions at each of the 17 amino acid residues within the motif.

Table 16. Single Amino Acid Substitutions within Motif I of YILPCAT Protein

Wild type SEQ ID NO

Single Amino Acid Substitutions

residue

M132A, M132N, M132C, M132G, M132Q, M132H, M132I,

M132 M132L, M132F, M132P, M132S, M132T, M132W, M132Y 44

or M 132V

V133A, V133N, V133C, V133G, V133Q, V133H, V133L,

V133 45

V133M, V133F, V133P, V133S, V133T, V133W or V133Y

L134A, L134N, L134C, L134G, L134Q, L134H, L134M,

L134 46

L134F, L134P, L134S, L134T, L134W, L134Y or l_134V

C135R, C135N, C135D, C135G, C135E, C135Q, C135H,

C135 C135I, C135L, C135K, C135M, C135F, C135P, C135S, 47

C135W or C135Y

M136A, M136N, M136C, M136G, M136H, M136I, M136F,

M136 48

M136P, M136S, M136T, M136W, M136Y or M136V

K137A, K137R, K137N, K137G, K137H, K137P, K137S,

K137 49

K137T, or K137Y

L138A, L138N, L138C, L138G, L138Q, L138H, L138I,

L138 50

L138M, L138F, L138P, L138S, L138T, L138W, or L138Y

S139A, S139N, S139C, S139G, S139H, S139L, S139M,

S139 51

S139F, S139P, S139W, or S139V

S140N, S140C, S140H, S140I, S140L, S140F, S140P,

S140 52

S140W, S140Y or S140V

F141A, F141 N, F141 G, F141 H, F141 I, F141 M, F141 P,

F141 53

F141 S, F141T, F141W, or F141V

G142N, G142H, G142I, G142L, G142M, G142F, G142P,

G142 54

G142T, G142W, G142Y or G142V

W143A, W143G, W143H, W143L, W143K, W143P,

W143 55

W143S, W143T, or W143V

N144A, N144R, N144G, N144H, N144K, N144F, N144P,

N144 56

N144T or N144V

V145A, V145C, V145G, V145E, V145H, V145M, V145F,

V145 57

V145P, V145S, V145T, or V145W

Y146R, Y146N, Y146D, Y146G, Y146E, Y146Q, Y146I,

Y146 58

Y146L, Y146M, Y146F, Y146P, Y146W or Y146V

D147A, D147N, D147G, D147E, D147Q, D147H, D147F,

D147 59

D147S, or D147T G148A, G148N, G148H, G148L, G148M, G148F, G148S,

G148T or G148V

Similarly, a total of 134 amino acid substitutions were designed within Motif II, as shown in Table 17, by creating various substitutions within 12 of the 15 amino acid residues within the motif. No substitutions were made at W379, H380 and G381 , since the histidine of other LPCATs corresponding to H380 of YILPCAT has been reported to be a likely active site residue (Lee et al., 2008, Mol. Biol. Cell 19:1 174-1 184).

Table 17. Sinqle Amino Acid Substitutions within Motif II of YILPCAT Protein

Each of the 329 YILPCAT mutants set forth above in Tables 16 and 17 were individually synthesized and cloned into Nco\/Not\-cut pY306-N vector by GenScript Corporation (Piscataway, NJ). EXAMPLE 8

Identifying Single Amino Acid Substitutions in YILPCAT Having Improved

LPCAT Activity

The present example describes the transformation of each of the 329 pY306-N vectors comprising a YILPCAT mutant polynucleotide sequence (Example 7) into Y. Iipolytica strain Y8406U2, followed by analysis of the lipid profiles of the transformants.

Improved LPCAT activity was indirectly evaluated, based on the observations set forth in U.S. Pat. Appl. Publ. No. 2010-0317882-A1 , which is incorporated herein by reference. Specifically, improved LPCAT activity within Y. Iipolytica strain Y8406U2 transformants comprising a mutated YILPCAT was concluded based on an increase in the concentration of EPA as a weight % of TFAs (EPA % TFAs) and/or an increase in the conversion efficiency of the delta-9 elongase, when either factor was compared to the EPA % TFAs or the conversion efficiency of the delta-9 elongase,

respectively, in Y. Iipolytica strain Y8406U2 expressing the parent wild type YILPCAT protein.

Transformation of Y. Iipolytica Strain Y8406U2

Strain Y8406U2 was transformed to individually express one of each of the pY306-N vectors containing a mutant YILPCAT prepared in Example 7. Y8406U2 is a Ura ^~ strain of Y8406. Details regarding the development of strains Y8406 and Y8406U2 are provided in U.S. Pat. Appl. Publ. No. 2010- 0317882-A1 . Following transformation, individual transformants were subjected to an initial fatty acid screening process as described above.

Briefly, single colonies that were grown on MM plates at 30 °C for 5 to 6 days were re-streaked and grown for two days at 30 °C on MM plates. Single colonies were then inoculated into 3 mL MM in a 24-well plate and shaken at 250 rpm at 30 °C for 2 days. The cells from each well were collected by centrifugation, resuspended in HGM, and then shaken at 200 rpm for 5 days. Cells were then processed for fatty acid analysis as described above. Analysis of Lipid Profiles within Yarrowia Transformed for Expression of Single Mutants of YILPCAT

Tables 18 (Batch 1 ), 19 (Batch 2), 20 (Batch 3), 21 (Batch 4) and 22 (Batch 5) below show the fatty acid profiles and delta-9 elongase conversion efficiencies of individual Y8406U2 transformants comprising a plasmid for expressing a particular single-mutated YILPCAT (single amino acid substitution in Motif I or Motif II). These measurements were also made for certain controls: transformants comprising an empty vector (EV) (i.e., a replicating plasmid with no LPCAT gene [Batch #1 only]) or pY306-N (wild type YILPCAT protein expression [WT]).

More specifically, each table summarizes the number of replicates analyzed for each particular transformant (#), the average concentration of each fatty acid as a weight percent of TFAs (% TFAs), the standard deviation for EPA % TFAs (EPA SD), and the delta-9 elongase conversion efficiency (% Conv). The % Conv. was calculated for each transformant according to the following formula.: (EDA + DGLA + ARA + ERA + ETA + EPA) / (C18:2 + C18:3 + EDA + DGLA + ARA + ERA + ETA + EPA) ^* 100.

Comparison of each mutant's performance relative to the wild type YILPCAT control should only be made within the particular batch in which each mutant was analyzed (i.e., comparisons should not be made between Batch #1 and Batch #2, for example). Mutants shown in bold-face font and followed by a "+" were selected for further studies, as discussed below.

Table 18. Lipid Composition and Delta-9 Elongase Conversion Efficiency in Batch #1 Transformants Comprising a Vector

Encoding YILPCAT Having a Single Amino Acid Substitution

Table 19. Lipid Composition and Delta-9 Elongase Conversion Efficiency in Batch #2 Transformants Comprising a Vector

Encoding YILPCAT Having a Single Amino Acid Substitution

Table 20. Lipid Composition and Delta-9 Elonqase Conversion Efficiency in Batch #3 Transformants Comprising a Vector

Encoding YILPCAT Having a Single Amino Acid Substitution

Table 21 . Lipid Composition and Delta-9 Elongase Conversion Efficiency in Batch #4 Transformants Comprising a Vector

Encoding YILPCAT Having a Single Amino Acid Substitution

Table 22. Lipid Composition and Delta-9 Elongase Conversion Efficiency in Batch #5 Transformants Comprising a Vector Encoding YILPCAT Having a Single Amino Acid Substitution

Based on the above data, it was clear that several of the YILPCAT single-amino acid mutants functioned with approximately equal or improved activity when compared to the parent wild type YILPCAT enzyme (SEQ ID NO:40). This conclusion was made based on measuring LPCAT activity as a function of EPA % TFAs and/or % Conv. In fact, all of the mutant YILPCAT transformants had an EPA % TFAs of at least 75% of the EPA % TFAs measured in the control (transformants with wild type YILPCAT). Also, all of the mutant YILPCAT transformants had a % Conv. that was at least 87.6% of the % Conv. measured in the control.

Fifty-six (56) YILPCAT mutants (comprising one of the following mutations with respect to SEQ ID NO:40: L134A, L134C, L134G, C135D, C135I, M136G, M136P, M136S, M136V, K137N, K137G, K137H, K137Y, L138A, L138H, L138M, S139L, S139W, S140N, S140H, S140P, S140W, F141 A, F141 M, F141 W, G142H, W143L, N144A, N144K, N144F, N144T, N144V, V145A, V145G, V145E, V145M, V145F, V145W, Y146G, Y146L, Y146M, D147N, D147Q, D147H, G148A, G148N, T382I, T382P, R383M, L388G, L388Y, T389A, T389C, T389S and F390C) were found to exhibit equivalent or improved EPA % TFAs and equivalent or improved % Conv. An additional 14 YILPCAT mutants were determined to have equivalent or improved EPA % TFAs when compared to the control (but did not have an equivalent or improved % Conv.), including mutants V133C, M136N, L138G, L138I, L138N, S139G, S139N, W143H, G148V, L388H, L388T, F390G, F390N and F390T. An additional 12 YILPCAT mutants were determined to have equivalent or improved % Conv. when compared to the control (but did not have an equivalent or improved EPA % TFAs), including mutants C135F, M136T, S140Y, S140I, F141 V, G142I, G142V, D147E, F378Y, T382Y, R383A and F390S.

A total of 26 YILPCAT mutants, each comprising a single mutation within either Motif I or Motif II and having equivalent or improved EPA % TFAs and/or equivalent or improved % Conv. were selected for further evaluation (below, Example 9): L134A (100.4%, 100.6%), L134G (101 .3%, 100.7%), M136S (104.0%, 104.0%), M136V (102.2%, 103.3%), K137H (107.3%, 104.4%), K137N (101 .8%, 102.0%), S140H (107.3%, 104.3%), S140W (103.2%, 103.8%), F141 M (105.4%, 106.7%), F141 W (101 .2%, 101 .6%), N144A (105.3%, 103.4%), N144T (101 .8%, 101 .6%), V145M (102.0%, 104.0%), V145W (100.4%, 100.5%), D147H (105.3%, 102.3%), D147Q (103.6%, 101 .2%), G148A (101 .3%, 101 .8%), G148N (102.2%, 101 .8%), T382I (102.9%, 102.5%), T382P (100.2%, 100.2%), R383M

(103.6%, 104.0%), L388G (101 .6%, 100.2%), L388Y (100.0%, 99.9%), T389A (102.2%, 101 .2%), T389C (102.1 %, 101 .5%), T389S (101 .9%, 101 .7%), where the first and second percentages in each parenthetical set correspond to the percentage ratio of EPA % TFAs and % Conv.,

respectively, in the mutant YILPCAT transformants relative to the EPA % TFAs and % Conv. in the wild type YILPCAT control transformants. An additional 8 YILPCAT mutants, each comprising a single mutation within either Motif I or Motif II, also were selected for further evaluation (below, Example 9): F378Y (99.6%, 101 .1 %), T382Y (99.8%, 100.8%), P384A (98.7%, 99.0%), P384G (99.2%, 98.6%), L388T (100.5%, 98.3%), F390G (102.4%, 99.8%), F390S (99.4%, 100.5%) and F390T (101 .6%, 99.3%), where the parenthetical sets are as above.

EXAMPLE 9

Identifying Double Amino Acid Substitutions in YILPCAT Having Improved

LPCAT Activity

The present example describes the synthesis of double YILPCAT mutants, wherein the double mutants comprise both a single mutation within Motif I and a single mutation within Motif II. These double mutants were transformed into Y. Iipolytica strain Y8406U2, followed by analysis of the lipid profiles of the transformants. As in Example 8, improved LPCAT activity was indirectly evaluated based on EPA % TFAs and % Conv.

Generation of Double YILPCAT Mutants

Preferred single mutations within Motif I (L134A, L134G, M136S,

M136V, K137H, K137N, S140H, S140W, F141 M, F141 W, N144A, N144T, V145W, V145M, D147H, D147Q, G148A and G148N) were combined with preferred single mutations within Motif II (F378Y, T382I, T382P, T382Y, R383M, P384A, P384G, L388G, L388T, L388Y, T389A, T389C, T389S, F390G, F390S, F390T) to generate various combinations of double-mutant YILPCAT sequences. Thus, for example, a YILPCAT mutant comprising an SHOW mutation within Motif I and a T382I mutation within Motif II is referred to herein as a YILPCAT mutant S140W_T382I. These double mutants were individually synthesized and cloned into Nco\-Not\ cut pY306-N vector by GenScript Corporation (Piscataway, NJ); SEQ ID NO:74 represents the mutant YILPCAT proteins encoded by the cloned sequences.

Transformation of Y. lipolytica Strain Y8406U2 and Analysis of Lipid Profiles within PY306-N Transformants

The plasmids were transformed into Y. lipolytica strain Y8406U2 and transformants were subsequently grown and subjected to lipid analysis, as described in Example 8. Tables 23 (Batch 6), 24 (Batch 7), 25 (Batch 8) and 26 (Batch 10) show the fatty acid profiles and delta-9 elongase conversion efficiencies of individual transformants of Y8406U2. These measurements were also made for control transformants comprising pY306-N (wild type YILPCAT protein expression [WT]). The Tables are formatted as described in Example 8.

Comparison of each mutant's performance relative to the wild type YILPCAT control should only be made within the particular batch in which each mutant was analyzed (i.e., comparisons should not be made between Batch #6 and Batch #7, for example). Mutants shown in bold-face font and followed by a "+" were selected for further studies including flask assays, as discussed below. Table 23. Lipid Composition and Delta-9 Elongase Conversion Efficiency in Batch #6 Transformants Comprising a Vector

Encoding YILPCAT Having Double Amino Acid Substitutions

Table 24. Lipid Composition and Delta-9 Elongase Conversion Efficiency in Batch #7 Transformants Comprising a Vector

Encoding YILPCAT Having Double Amino Acid Substitutions

Table 25. Lipid Composition and Delta-9 Elonqase Conversion Efficiency in Batch #8 Transformants Comprising a Vector

Encoding YILPCAT Having Double Amino Acid Substitutions

Table 26. Lipid Composition and Delta-9 Elongase Conversion Efficiency in Batch #10 Transformants Comprising a Vector

Encoding YILPCAT Having Double Amino Acid Substitutions

Based on the data set forth above, it is clear that most of the 167 YILPCAT double mutants analyzed above functioned with approximately equal or improved activity when compared to the parent wild type enzyme (SEQ ID NO:40). This conclusion was made based on measuring LPCAT activity as a function of EPA % TFAs and/or % Conv.

More specifically, 106 YILPCAT mutants comprising a single amino acid mutation within Motif I and a single amino acid mutation within Motif II were found to exhibit equivalent or improved EPA % TFAs and equivalent or improved % Conv. These mutants were L134A_T382I, L134A_L388G, L134A_F390T M136S_F378Y, M136S_T382I, M136S_T382P,

M136S_T382Y, M136S_R383M, M136S_P384A, M136S_L388Y,

M136S_T389A, M136S_T389C, M136S_T389S, M136V_T382P,

M136V_T382Y, M136V_P384A, M136V_L388Y, M136V_T389A,

M136V_T389C, M136V_T389S, K137H_P384G, K137H_L388G,

K137H_L388T, K137H_F390S, K137H_F390T, K137N_T382P,

K137N_R383M, K137N_P384G, K137N_F378Y, K137N_L388G,

K137N_L388T, K137N_T389A, K137N_T389C, K137N_T389S,

K137N_F390G, K137N_F390S, K137N_F390T, S140H_T382I,

S140H_P384G, S140H_L388G, S140H_L388T, S140H_F390G,

S140H_F390S, S140W_T382I, S140W_T382P, S140W_T382Y,

S140W_R383M, S140W_P384A, S140W_L388Y, S140W_T389A,

S140W_T389C, F141 M_F378Y, F141 M_T382Y, F141 M_R383M,

F141 M_P384A, F141 M_T389C, F141 W_F378Y, F141 W_T382I,

F141 W_T382P, F141 W_T382Y, F141 W_R383M, F141 W_P384A,

F141 W_T389A, F141 W_T389C, F141 W_T389S, N144A_P384G,

N144A_L388G, N144A_L388T, N144A_F390G, N144A_F390S,

N144A_F390T, N144T_F378Y, N144T_T382P, N144T_T382Y,

N144T_R383M, N144T_P384A, N144T_L388Y, N144T_T389A,

N144T_T389C, N144T J389S, V145M_T382I, V145M_R383M,

V145M_T389A, V145M_T389C, V145W_T382I, D147H_T382I,

D147H_L388G, D147H_L388T, D147H_F390S, D147Q_T382I, D147Q_F390S, G148A_F378Y, G148A_T382I, G148A_T382Y,

G148A_R383M, G148A_P384G, G148A_L388G, G148A_L388Y,

G148A_T389A, G148A_T389C, G148A_F390S, G148A_F390T,

G148N_T382I, G148N_L388T, G148N_F390G and G148N_F390S).

An additional 15 YILPCAT double mutants (of the 167 analyzed) had equivalent or improved EPA % TFAs when compared to the control, while an additional 6 YILPCAT double mutants (of the 167 analyzed) were determined to have equivalent or improved % Conv. when compared to the control.

Confirmation of Improved LPCAT Activity by Flask Assay

A total of 23 YILPCAT double mutants, each comprising a single amino acid mutation within Motif I and a single amino acid mutation within Motif II, and having equivalent or improved EPA % TFAs and/or equivalent or

improved % Conv., were selected for further evaluation (these mutants are noted in bold and with a "+" in Tables 23-26). These mutants were:

S140W_T382P, S140W_T389A, M136V_T389A, M136V_T389C,

M136V_T389S, K137N_T389A, K137N_T389C, K137N_T389S,

M136S_T389A, M136S_T389C, M136S_T389S, F141 W_T382I,

L134A_T382I, K137N _F390G, K137H_L388G, K137H_L388T,

S140H_T382I, S140H_L388G, N144A_F390S, D147H_T382I,

G148A_F390S, G148N_T382I and G148N_F390S. Additionally, mutants M136V_F378Y and G148A_L388T, each having slightly diminished EPA % TFAs and slightly diminished % Conv. in comparison to the control, were selected for further evaluation.

Transformants expressing these double mutant YILPCAT proteins were subjected to flask assays for a detailed analysis of the total lipid content and composition. Specifically, the double mutant strains were individually inoculated into 3 mL FM in 15-mL Falcon™ tubes and grown overnight at 30 °C and 250 rpm. The OD ₆ oonm was measured and an aliquot of the cells was added to a final OD ₆ oonm of 0.3 in 25 mL FM medium in a 125-mL flask. After 2 days in a

Multitron shaking incubator at 250 rpm and at 30 °C, 6 mL of the culture was harvested by centrifugation and resuspended in 25 mL HGM in the original 125- mL flask. After 5 days (120 hours) in a shaking incubator at 250 rpm and at 30 °C, water was added to flasks to bring the total volume back to 25 mL (thereby accounting for evaporation). An aliquot was used for fatty acid analysis (above) and 10 mL of the culture was dried for dry cell weight determination (above).

The flask assay results are shown below in Tables 27 (Group I) and 28

(Group II). The Tables summarize the number of replicates analyzed for each particular transformant (#), the average total dry cell weight of the cells (DCW), the average total lipid content of the cells (TFAs % DCW), the average concentration of each fatty acid as a weight percent of TFAs ("% TFAs), the delta-9 elongase conversion efficiency (% Conv.) and the average EPA content as a percent of the dry cell weight (EPA % DCW).

Table 27. Total Lipid Content, Composition and Delta-9 Elongase Conversion Efficiency in Selected Transformants

Comprising a Vector Encoding YILPCAT Having Double Amino Acid Substitutions, bv Flask Assav (Group I)

Table 28. Total Lipid Content, Composition and Delta-9 Elongase Conversion Efficiency in Selected Transformants

Comprising a Vector Encoding YILPCAT Having Double Amino Acid Substitutions, bv Flask Assav (Group II)

Of the 25 YILPCAT double mutants analyzed, each comprising a single amino acid mutation within Motif I and a single amino acid mutation within Motif II, 17 were observed to have both equivalent or improved EPA % TFAs and equivalent or improved % Conv., while the remaining 8 had equivalent or improved % Conv.

Based on the data set forth above, 22 of the 25 YILPCAT double mutants analyzed above functioned with improved activity when compared to the parent wild type enzyme (SEQ ID NO:40).

Also, the over-expression of certain double-mutant LPCAT polypeptides resulted in increased total lipid content (TFAs % DCW) in the recombinant Yarrowia. For example, over-expression of mutant LPCAT polypeptides comprising the S140W_T382P, S140W_T389A, M136V_T389S and

F141 W_T382I, or K137H_L388G mutation pairs resulted in total lipid contents that were 8% or more increased relative to the total lipid content of the control (Tables 27 and 28). Interestingly, certain transformants had both increased total lipid content and EPA % TFAs. For example, transformants that over-expressed LPCATs with S140W_T389A, M136V_T389C, M136S_T389A, or M136S_T389S mutation pairs had at least a 5% increase in total lipid content and at least a -3% increase in EPA % TFAs with respect to control (Tables 27 and 28). This is a significant observation since it had previously been difficult to induce a

simultaneous increase in both total lipid content and EPA % TFAs. Usually, an increase in total lipid content had corresponded with a decrease in EPA % TFAs, and vice versa.

The double mutant YILPCAT polypeptides listed in bold and with a "+" in Tables 27 and 28, i.e., M136S_T389A, M136S_T389C, M136S_T389S,

M136V_T389C, N144A_F390S, G148A_F390S, G148N_T382I and

G148N_F390S, are disclosed herein as SEQ ID NOs:26, 75, 76, 77, 78, 79, 80 and 81 , respectively.