Arachidonic acid release mediated by OX1 orexin receptors

Abstract

Background and purpose: We have previously shown that lipid mediators, produced by phospholipase D and C, are generated in OX(1) orexin receptor signalling with high potency, and presumably mediate some of the physiological responses to orexin. In this study, we investigated whether the ubiquitous phospholipase A(2) (PLA(2)) signalling system is also involved in orexin receptor signalling.

Experimental approach: Recombinant Chinese hamster ovary-K1 cells, expressing human OX(1) receptors, were used as a model system. Arachidonic acid (AA) release was measured from (3)H-AA-labelled cells. Ca(2+) signalling was assessed using single-cell imaging.

Key results: Orexins strongly stimulated [(3)H]-AA release (maximally 4.4-fold). Orexin-A was somewhat more potent than orexin-B (pEC(50) = 8.90 and 8.38 respectively). The concentration-response curves appeared biphasic. The release was fully inhibited by the potent cPLA(2) and iPLA(2) inhibitor, methyl arachidonyl fluorophosphonate, whereas the iPLA(2) inhibitors, R- and S-bromoenol lactone, caused only a partial inhibition. The response was also fully dependent on Ca(2+) influx, and the inhibitor studies suggested involvement of the receptor-operated influx pathway. The receptor-operated pathway, on the other hand, was partially dependent on PLA(2) activity. The extracellular signal-regulated kinase, but not protein kinase C, were involved in the PLA(2) activation at low orexin concentrations.

Conclusions and implications: Activation of OX(1) orexin receptors induced a strong, high-potency AA release, possibly via multiple PLA(2) species, and this response may be important for the receptor-operated Ca(2+) influx. The response coincided with other high-potency lipid messenger responses, and may interact with these signals.

Figure 4

The effect of Ca²⁺ on AA release in CHO-hOX₁ cells. (A) The cells were stimulated with orexin-A in the normal extracellular concentration of Ca²⁺ (1 mM) or in the reduced concentration (30 µM) in NaBM. Data are normalized as explained in Data analysis and Figure 3. (B) The cells were, before stimulation with orexin-A, exposed to the treatments that reduce the driving force for Ca²⁺ entry (KBM) or inhibit particular Ca²⁺ channel types [TEA-BM (70 mM TEA), SKF-96395 (10 µM) ]. The cells were preincubated for 5 min in the presence of the specific medium/inhibitor before stimulation with orexin-A in the same medium. Data are normalized as explained in Data analysis and Figure 3. (C) The cells were stimulated with orexin-A in the absence (ctrl) or presence of the Ca²⁺-elevating compounds thapsigargin or ionomycin in NaBM. The first comparison (***) is to the basal (i.e. Do thapsigargin, ionomycin and orexin-A stimulate AA release over the basal?) and the second comparison (†† and ns) to thapsigargin or ionomycin (i.e. Does orexin-A stimulate AA release over thapsigargin or ionomycin?); ns (not significant), P > 0.05; *P < 0.05; **P < 0.01; ***P < 0.001. ††P < 0.01.

Figure 3

The effect of the PLA₂ inhibitors MAFP (A) and R- and S-BEL (B) on the orexin-A-stimulated AA release in CHO-hOX₁ cells. The cells were preincubated for 30 min with the inhibitor, changed to fresh NaBM containing the inhibitor and immediately stimulated with orexin-A. The responses are normalized so that each control response (basal, 1 nM orexin-A, 100 nM orexin-A) amounts to 100% (see Data analysis); ns (not significant), P > 0.05; *P < 0.05; **P < 0.01; ***P < 0.001.

Figure 2

Concentration–response relationships for orexin-A and orexin-B for AA release in CHO-hOX₁ cells in NaBM. A two-site fit was significantly better (P < 0.001) than a one-site fit for both orexin-A and orexin-B.

Figure 1

Orexin-A-stimulated ³H-AA (A, C) and ³H-oleic acid (B) release. (A and B) OX₁ receptor-expressing CHO cells (CHO-hOX₁) were used. The release was assessed in NaBM in the presence of 2.4 mg/mL BSA (+BSA) or in the absence of BSA (−BSA). (C) Both wild-type CHO cells (CHO-wt) and CHO-hOX₁ cells were used, both in the absence of BSA. CHO-hOX₁ cells were preincubated for 30 min with SB-334867, changed to fresh NaBM containing the SB-334867 and immediately stimulated with orexin-A. ns (not significant), P > 0.05; *P < 0.05; **P < 0.01; ***P < 0.001; comparisons in all subfigures are to the corresponding basal in C, significances are only calculated for wild-type CHO cells and CHO-hOX₁ with 10 µM SB-334867.

Figure 5

The effect of PKC and ERK activation on AA release in CHO-hOX₁ cells. (A) Cells were preincubated for 30 min with the MEK inhibitors U0126 and 98059 as indicated, changed to fresh NaBM containing the inhibitor and immediately stimulated with orexin-A. Data are normalized as explained in Data analysis and Figure 3. (B) Cells were stimulated with 1 mM TPA for 7 min in NaBM. ns (not significant), P > 0.05; **P < 0.01; ***P < 0.001.

Figure 6

The effect of AA release inhibition on orexin-induced receptor-operated Ca²⁺ influx (A, C) and Ca²⁺ release (B, D) in CHO-hOX₁ cells in NaBM. (A, C) Cells are over-expressing IP₃-5P1, which precludes Ca²⁺ release. In (A) data from counting of responsive and non-responsive cells (n= 458–496 cells/treatment) and in (C) a representative experiment (n= 29–31 cells/trace). Some error bars (C) are omitted for purpose of clarity. (B, D) Cells were stimulated in nominally Ca²⁺-free conditions (no CaCl₂ added to the NaBM), which gives an extracellular [Ca²⁺]≈ 2.5–3.3 µM as measured using CalciumGreen 5N, which is low enough concentration to abolish the Ca²⁺ influx. In (B) data from counting of responsive and non-responsive cells (n= 551–577 cells/treatment) and in (D) the average Ca²⁺ responses from these cells. The first comparisons (*/**/***) in (A, B and D) are to the basal (i.e. Does orexin cause a Ca²⁺ response?) and the second comparison (††† and ns) to the corresponding orexin response (i.e. Does MAFP inhibit the response to orexin-A?); ns (not significant), P > 0.05; **P < 0.01; ***P < 0.001, †††P < 0.001.