DGA1 (diacylglycerol acyltransferase gene) overexpression and leucine biosynthesis significantly increase lipid accumulation in the Deltasnf2 disruptant of Saccharomyces cerevisiae

Abstract

We previously found that SNF2, a gene encoding a transcription factor forming part of the SWI/SNF (switching/sucrose non-fermenting) chromatin-remodelling complex, is involved in lipid accumulation, because the Deltasnf2 disruptant of Saccharomyces cerevisiae has a higher lipid content. The present study was conducted to identify other factors that might further increase lipid accumulation in the Deltasnf2 disruptant. First, expression of LEU2 (a gene encoding beta-isopropylmalate dehydrogenase), which was used to select transformed strains by complementation of the leucine axotroph, unexpectedly increased both growth and lipid accumulation, especially in the Deltasnf2 disruptant. The effect of LEU2 expression on growth and lipid accumulation could be reproduced by adding large amounts of leucine to the culture medium, indicating that the effect was not due to Leu2p (beta-isopropylmalate dehydrogenase) itself, but rather to leucine biosynthesis. To increase lipid accumulation further, genes encoding the triacylglycerol biosynthetic enzymes diacylglycerol acyltransferase (DGA1) and phospholipid:diacylglycerol acyltransferase (LRO1) were overexpressed in the Deltasnf2 disruptant. Overexpression of DGA1 significantly increased lipid accumulation, especially in the Deltasnf2 disruptant, whereas LRO1 overexpression decreased lipid accumulation in the Deltasnf2 disruptant. Furthermore, the effect of overexpression of acyl-CoA synthase genes (FAA1, FAA2, FAA3 and FAA4), which each supply a substrate for Dga1p (diacylglycerol acyltransferase), was investigated. Overexpression of FAA3, together with that of DGA1, did not further increase lipid accumulation in the Deltasnf2 disruptant, but did enhance lipid accumulation in the presence of exogenous fatty acids. Lastly, the total lipid content in the Deltasnf2 disruptant transformed with DGA1 and FAA3 overexpression vectors reached approx. 30%, of which triacylglycerol was the most abundant lipid. Diacylglycerol acyltransferase activity was significantly increased in the Deltasnf2 disruptant strain overexpressing DGA1 as compared with the wild-type strain overexpressing DGA1; this higher activity may account for the prominent increase in lipid accumulation in the Deltasnf2 disruptant with DGA1 overexpression. The strains obtained have a lipid content that is high enough to act as a model of oleaginous yeast and they may be useful for the metabolic engineering of lipid production in yeast.

Figure 1. Effect of leucine addition on growth and lipid content of the wild-type and Δsnf2 disruptant

The BY4741 wild-type strain (A) and Δsnf2 disruptant (B) were cultured at 30 °C in 50 ml of NLSD medium, to which required nutrients such as 20 μg/ml of histidine, 20 μg/ml of methionine or 20 μg/ml of uracil and the indicated concentration of leucine were added. Dry cell weight (●) and total fatty acids (■) were measured as described in the Experimental section. Lipid content (▲) is expressed as total fatty acids (mg) per dry cell weight (mg)×100. Results are means±S.D. (n=3) for a 50 ml sample of the culture.

Figure 2. Nile Red staining of the Δsnf2 disruptant transformed with pL1091-5/DGA1 and pL1177-2/FAA3

The BY4741 Δsnf2 disruptant (A) and Δsnf2 disruptant transformed with pL1091-5/DGA1 and pL1177-2/FAA3 (B) cultured in NLSD medium for 7 days were stained with Nile Red. The upper panels show a fluorescent image; lower panels show a differential-interference-contrast image. The bar represents 2 μm.

Figure 3. Immunoblotting of Dga1p in the homogenate of yeast transformed with pL1091-5/DGA1-6×His

The cell homogenate was prepared from the wild-type (WT) and Δsnf2 disruptant transformed with pL1091-5/DGA1-6×His and pL1177-2 (or pL1177-2/FAA3) that had been cultured for 4 (4d) or 7 days (7d). Each fraction (2 μg of protein) was subjected to SDS/12.5%-(w/v)-PAGE under reducing conditions, followed by immunoblotting with anti-hexahistidine monoclonal antibody as detailed in the Experimental section. Arrows indicate molecular-mass standards (Bio-Rad). DGAT activity in the homogenate of the Δsnf2 disruptant transformed with pL1091-5/DGA1-6xHis and pL1177-2 was 799 (4 days) or 354 (7 day) pmol/min per mg protein, whereas DGAT activity in the homogenate of the Δsnf2 disruptant transformed with pL1091-5/DGA1-6×His and pL1177-2/FAA3 was 936 (4 days) or 388 (7 days) pmol/min per mg of protein. DGAT activity in the homogenate of the wild-type transformed with pL1091-5/DGA1-6xHis and pL1177-2 (or pL1177-2/FAA3) was less than 2 pmol/min per mg of protein.