Acyltransferases Regulate Oil Quality in Camelina sativa Through Both Acyl Donor and Acyl Acceptor Specificities

Abstract

Camelina sativa is an emerging biotechnology oil crop. However, more information is needed regarding its innate lipid enzyme specificities. We have therefore characterized several triacylglycerol (TAG) producing enzymes by measuring in vitro substrate specificities using different combinations of acyl-acceptors (diacylglycerol, DAG) and donors. Specifically, C. sativa acyl-CoA:diacylglycerol acyltransferase (DGAT) 1 and 2 (which both use acyl-CoA as acyl donor) and phospholipid:diacylglycerol acyltransferase (PDAT, with phosphatidylcoline as acyl donor) were studied. The results show that the DGAT1 and DGAT2 specificities are complementary, with DGAT2 exhibiting a high specificity for acyl acceptors containing only polyunsaturated fatty acids (FAs), whereas DGAT1 prefers acyl donors with saturated and monounsaturated FAs. Furthermore, the combination of substrates that resulted in the highest activity for DGAT2, but very low activity for DGAT1, corresponds to TAG species previously shown to increase in C. sativa seeds with downregulated DGAT1. Similarly, the combinations of substrates that gave the highest PDAT1 activity were also those that produce the two TAG species (54:7 and 54:8 TAG) with the highest increase in PDAT overexpressing C. sativa seeds. Thus, the in vitro data correlate well with the changes in the overall fatty acid profile and TAG species in C. sativa seeds with altered DGAT1 and PDAT activity. Additionally, in vitro studies of C. sativa phosphatidycholine:diacylglycerol cholinephosphotransferase (PDCT), another activity involved in TAG biosynthesis, revealed that PDCT accepts substrates with different desaturation levels. Furthermore, PDCT was unable to use DAG with ricineoleyl groups, and the presence of this substrate also inhibited PDCT from using other DAG-moieties. This gives insights relating to previous in vivo studies regarding this enzyme.

Keywords: Camelina; Kennedy pathway; acyltransferase; diacylglycerol; diacylglycerol acyltransferase; fatty acid composition; phosphatidylcholine; triacylglycerol.

Figure 1

Overview of the de-novo triacylglycerol formation in seeds. Abbreviations for lipids (black letters): G3P, glycerol-3-phosphate; LPA, lyso-phosphoatidic acid; PA, phosphatidic acid; DAG, diacylglycerol; PC, phosphatidylcholine; TAG, triacylglycerol. Abbreviations for enzymes (white letters): GPAT, G3P acyltransferase; LPAT, lysophosphatidylacyltransferase; PAP, Phosphatidate phosphatase; PDCT, PC : DAG cholinephosphotransferase; DGAT, acyl-CoA : DAG acyltransferase; PDAT, phospholipid:DAG acyltransferase.

Figure 2

Acyl-CoA specificities of C. sativa DGAT enzymes in microsomal preparations from yeast expressing the enzymes, when fed [¹⁴C]glycerol labeled di-6:0-DAG and different acyl-CoAs. (A) DGAT1 and (B) DGAT2. Average value shown ± SD, n=3 replicates. Stars over individual bars indicate significant differences according to one-way ANOVA followed by Tukey’s test. ***p ≤ 0.001.

Figure 3

Acyl-CoA and DAG specificities of C. sativa DGAT enzymes in in microsomal preparations from yeast expressing the enzymes, when fed exogenous long-chain DAGs and [¹⁴C]acyl-CoA as indicated in the figure. Background activity (no DAG added) was subtracted in the figure (see Supplementary Table 1 ). (A) DGAT1 and (B) DGAT2. Average value shown ± SD, n = 3 replicates. Stars over individual bars indicate significant differences compared to the other Acyl-CoA species with the same DAG according to one-way ANOVA followed by Tukey’s test performed on the whole dataset. ***p ≤ 0.001. Different letters over DAG species indicate significant differences between DAG species of all Acyl-CoA combinations at the level of p ≤ 0.001. Complete results of the statistical analysis can be found in Supplementary Tables 2 (DGAT1) and 3 (DGAT2).

Figure 4

Substrate specificities of C. sativa PDAT in microsomal preparations of from developing seeds overexpressing the enzyme, when fed [¹⁴C]radiolabeled PC and different DAG. Stars over individual bars indicate significant differences compared to the other PC species with the same DAG according to one-way ANOVA followed by Tukey’s test performed on the whole dataset. *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001. Stars at lines between bars indicate significant differences between selected PC and DAG species further discussed. Complete results of the statistical analysis can be found in Supplementary Table 4 . Average value shown ± SD, n = 3 replicates.

Figure 5

Incorporation of radioactivity into chloroform soluble lipids in microsomal fractions of yeast expressing in the in figure indicated C. sativa genes fed [¹⁴C]glycerol labeled G3P and 18:3-CoA. Incubation time was 60 min. PA, phosphatidic acid; PC, phosphatidylcholine; DAG, diacylglycerol; EV, Empty vector (control); PDCT, PC : DAG cholinephophotransferase. Average value shown ± SD, n = 3 replicates.

Figure 6

Time-course incorporation of radioactivity into chloroform soluble lipids in microsomal fractions of yeast expressing C. sativa PDCT fed [¹⁴C]glycerol labeled glycerol 3-phosphate and different acyl-CoAs. (A) 18:1-CoA, (B) 18:2-CoA, and (C) 18:3-CoA. PA, phosphatidic acid; PC, phosphatidyl-choline; DAG, diacylglycerol. Average value shown ± SD, n = 3 replicates.

Figure 7

Incorporation of radioactivity into chloroform soluble lipids in microsomal fractions of yeast expressing C. sativa PDCT fed [¹⁴C]glycerol labeled G3P and 18:3-CoA and or ricinoleoyl-CoA. (A) 18:3-CoA and ricinoleoyl-CoA (Ric-CoA) as indicated in the figure. (B) Mixtures of 18:3 and Ric-CoA in proportions indicated in the figure. Incubation time was 60 min. PA, phosphatidic acid; PC, phosphatidyl-choline; DAG, diacylglycerol. Average value shown ± SD, n = 3 replicates.