Role of Elovl4 protein in the biosynthesis of docosahexaenoic acid

Abstract

The disk membranes of retinal photoreceptor outer segments and other neuronal and reproductive tissues are enriched in docosahexaenoic acid (DHA, 22:6n3), which is essential for their normal function and development. The fatty acid condensing enzyme Elongation of Very Long chain fatty acids-4 (ELOVL4) is highly expressed in retina photoreceptors as well as other tissues with high 22:6n3 content. Mutations in the ELOVL4 gene are associated with autosomal dominant Stargardt-like macular dystrophy (STGD3) and results in synthesis of a truncated protein that cannot be targeted to the endoplasmic reticulum (ER), the site of fatty acid biosynthesis. Considering the abundance and essential roles of 22:6n3 in ELOVL4-expressing tissues (except the skin), it was proposed that the ELOVL4 protein may be involved in 22:6n3 biosynthesis. We tested the hypothesis that the ELOVL4 protein is involved in 22:6n3 biosynthesis by selectively silencing expression of the protein in the cone photoreceptors derived cell line 661 w and showed that the ELOVL4 protein is not involved in DHA biosynthesis from the short chain fatty acid precursors 18:3n3 and 22:5n3.

Fig. 27.1

Endogenous expression of Elovl4 in 661W cells; (a) RT-PCR of Elovl4 cDNA from 661W cells. Complementary DNA from the 661W cells was used as a template to amplify the Elovl4 transcript with actin used as a control (data not shown). Representative results are presented. Lanes: 1, 100-bp markers; 2, mouse retina cDNA; 3–4, 661W cDNA; 5, no reverse transcriptase control. (b) A representative Western blot analysis of 30 μg protein from 661W cells and mouse retina confirmed the expression of ELOVL4 in 661W cells. Bottom panel is β-actin loading control

Fig. 27.2

Elongation of 18:3n3 and 22:5n3 in 661W cells; (a) 661W cells were cultured in medium supplemented with increasing concentrations of 18:3n3 or (b) 22:5n3 and grown for 72 h. Total cellular lipids were extracted, converted to fatty acid methyl esters (FAMEs), and analyzed by gas-liquid chromatography. 18:3n3 was elongated and desaturated to 20:5n3, 22:5n3, and 24:5n3. However, there was negligible conversion to 24:6n3 and 22:6n-3. Similarly, 22:5n3 (b) was incorporated into cellular lipids and some was elongated to 24:5n3

Fig. 27.3

siRNA and shRNA knock-down of mouse Elovl4 in 661W cells; (a) Elovl4 knock-down in 661W cells transfected with 100 or 200 nM anti-Elovl4 siRNA smart-pools. Elovl4 knock-down by the siRNA duplexes was assayed by RT-PCR. (b) Western blot analysis of ELOVL4 knockdown in 661W cells transfected with the pool of 4 siRNA duplexes 72 h after transfection. The bottom panel is β-actin loading control. (c) 661W cells transfected and sorted for GFP-positive cells expressing the pSilencer-anti-Elovl4-shRNA under the human U6 promoter and GFP under the CMV promoter. After sorting, the GFP positive cells were expanded and used for subsequent experiments. (d) Western blot analysis of GFP-positive 661W cells stably expressing anti-Elovl4-shRNA and controls post-sorting

Fig. 27.4

Knock-down of ELOVL4 did not affect the elongation of [1-¹⁴C]-18:3n3 to [¹⁴C]-20:5n3, [¹⁴C]-22:5n3 and [¹⁴C]-24:5n3.; (a) Relative percentage radioactivity in ELOVL4 knock-down and wild type 661W cells supplemented with [1-¹⁴C]-18:3n3. No differences in the formation of [¹⁴C]-20:5n3, [¹⁴C]-22:5n3, and [¹⁴C]-24:5n3 were found between control (non-shRNA expressing cells) and the ELOVL4-knock-down cells. This suggests that ELOVL4 is not involved in the elongation of 18:3n3 to 20:5n3, of 20:5n3 to 22:5n3, and of 22:5n3 to 24:5n3 in these cells. (b) GC-FID results of elongation of 18:3n3 in wild type 661W cells and shRNA-D2-GFP positive cells. There were no differences in the elongation products