Archives
The structure of LO is divided in two
The structure of 5-LO is divided in two domains, the catalytic C-terminal domain and the N-terminal regulatory C2-like domain (C2ld) [59,60]. The C2ld spans the lactone sale 1-114 and is responsible for translocation and binding of calcium and membranes [61–63]. The catalytic domain is primarily an α-helical structure and contains the catalytic non-heme iron, which is crucial for the catalytic activity of 5-LO [62]. Interestingly, all examined protein isoforms lack amino acids in the catalytic domain. To investigate the catalytic activity of the isoforms, we transfected HEK293T cells which were previously reported to be a suitable cell system to study the 5-LO pathway [64]. In order to get a robust and reproducible 5-LO assay system, we carried out stable transfection of HEK293T cells with 5-LO and its alternative isoforms using the sleepy beauty method. Activity assays in transfected HEK293T cells revealed that all isoforms are catalytically inactive. This observation is in agreement with previously published studies in which 5-LO∆13 and 5-LO∆4 were investigated [42,44]. In 5-LOp12, the amino acids 530-553 are deleted and thus, H550 is missing that is crucial for binding of the non-heme iron. Furthermore, the missing part of p12 is adjacent to the other iron binding motif N554. Therefore, it is likely that the iron content of 5-LOp12 is at least reduced if not exterminated, and thus, explains the absence of enzymatic activity of 5-LOp12. A comparison of transient and stable cellular transfection methods showed that coexpression of isoforms with full length 5-LO resulted in varying effects. In stably transfected cells, coexpression of the isoforms stimulated 5-LO activity whereas an inhibition was observed with transient transfection. In agreement with our data, 5-LO∆13 decreased the activity of 5-LO-WT after co-transfection in HEK293 cells which was mainly attributed to decreased cellular activity which could be restored by cell homogenization [42] and only a small reduction of 5-LO-WT expression was observed when the Δ13 isoform was cotransfected [21]. In our hands, the changes in 5-LO activity roughly correlated with changes in 5-LO-WT protein expression. Thus, the isoforms stimulated 5-LO-WT expression in stably transfected cells whereas an inhibition of 5-LO-WT expression was observed with transient transfection. The data suggest that the alternative isoforms might be involved in the regulation of the expression of the WT enzyme (Figs. 4, 7). At present, it is unclear why coexpression of the isoforms inhibits 5-LO-WT expression in transiently transfected cells and stimulates its expression in stably transfected cells. To investigate possible direct interactions between the alternatively spliced isoforms and 5-LO-WT, we checked the cellular localization of the isoforms by confocal microscopy. It is already known that the location of 5-LO-WT is cell type and stimulus dependent [3,23,24,65]. A previous study demonstrated that the isoform 5-LO∆13 is almost exclusively cytosolic [21]. Here, we found by confocal microscopy of stably transfected cells that all isoforms are localized in the cytosol, whereas 5-LO-WT was found in the nucleus. Only 5-LO-WT responded to calcium ionophore at 5μM and translocated to the nuclear envelope which is in agreement with previous observations for 5-LOΔ13 [21]. This is surprising since the domains known for membrane or calcium binding are present in the 5-LO isoforms. This indicates that the alternative isoforms might possess conformational changes due to the missing amino acids so that they are not able to interact with the nuclear membrane. Interestingly, after treatment with a high concentration of ionophore A23187 (25μM) the 5-LO∆13 isoform translocated to the nuclear membrane. As a result of the possible conformational change, it might need a higher calcium concentration for translocation. Phosphorylation at S271 regulates the nuclear localization of 5-LO-WT [53]. After mutation of the serine into an alanine and treatment of the cells with either sorbitol or KN93/SB203580, the 5-LO-S271A mutant was cytosolic, and thus, in the same cellular compartment as the 5-LO isoforms. However, we could not detect an influence of the isoforms on the specific enzymatic activity of 5-LO-S271A, since the increase in LT formation by coexpression of the isoforms correlates with an enhanced expression of 5-LO-S271A in these cells.