Prostaglandin D2 production in FM55 melanoma cells is regulated by α-melanocyte-stimulating hormone and is not related to melanin production

This study shows that prostaglandins in human FM55 melanoma cells and epidermal melanocytes are produced by COX-1. Prostaglandin production in FM55 melanoma cells was unrelated to that of melanin suggesting that the two processes can occur independently. α-Melanocyte-stimulating hormone, which had no effect on melanin production in FM55 cells, stimulated PGD2 production in these cells without affecting PGE2. While cAMP pathways may be involved in regulating PGD2 production, our results suggest that α-MSH acts independently of cAMP, possibly by regulating the activity of lipocalin-type PGD synthase. This α-MSH-mediated effect may be associated with its role as an immune modulator.


Background
Melanocytes produce melanin and have a role in skin pigmentation (1,2). Cutaneous prostaglandins, such as PGE 2 and PGF 2a , may act as mediators in this process because they increase melanocyte dendricity and melanogenesis (3,4). It has been suggested that these prostaglandins arise from keratinocytes (4) but it is possible that they are also produced by melanocytes acting as autocrine factors in the pigmentary response.
However, prostaglandins have a wide range of biological activities, and while some such as PGE 2 act as pro-inflammatory mediators (5), others such as PGD 2 down-regulate immune responses (6). PGD 2 is formed by prostaglandin D synthase (EC 5.3.99.2) (PGDS), an isoform of which, lipocalin-PGDS (L-PGDS), is expressed in pigment cells (7). Expression of L-PGDS is dependent upon microphthalmiaassociated transcription factor (MITF) (7) which is activated via cAMP and is involved in regulating melanogenesis (1,8,9). Thus, it is possible that PGD 2 production is associated with melanogenesis, and the two processes have a common regulatory pathway.

Questions addressed
Is the production of prostaglandins related to that of melanin in pigment-producing cells, and is it regulated by a-MSH?

Experimental design
Prostaglandins were measured in melanin-producing FM55 human melanoma cells (17) and in human epidermal melanocytes. FM55 cells were used as a model system to Re-use of this article is permitted in accordance with the Terms and Conditions set out at http://www3.interscience.wiley.com/author resources/onlineopen.html examine the relationship between prostaglandin and melanin production and the effect of a-MSH. Because their MC1R does not couple to cAMP FM55 cells do not produce melanin in response to a-MSH (18). Their use therefore allowed the possibility of dissociating prostaglandin production from that of melanin.
The lightly pigmented FM55 cells were established from metastatic melanoma nodules (Dr AF Kirkin, Danish Cancer Society, Copenhagen, Denmark). Human epidermal melanocytes were isolated from skin samples obtained with local ethics committee approval and informed consent from donors undergoing elective plastic surgery. Cell culture (19,20), eicosanoid analysis (21), stimulation and measurement of melanin (20,22) and COX-1 ⁄ -2 protein expression (5) were performed as published; L-PGDS was measured using an immunometric kit (Appendix S1).

Results
PGD 2 and PGE 2 were the major prostaglandins identified in human epidermal melanocytes and FM55 melanoma cells ( Fig. 1a, b). Lipidomic analysis did not confirm production of PGF 2a by FM55, as previously reported using a less specific radiometric approach (17). Western blotting analysis revealed that FM55 cells and melanocytes expressed the constitutive isoform of cyclooxygenase (COX-1) but not the inducible isoform COX-2 (Fig. 1c, d).
Increasing melanin production in FM55 had no effect on prostaglandin production (Fig. 2a, b); when prostaglandins were stimulated with arachidonic acid, melanin production was not affected (23 ± 6 and 24 ± 4 lg melanin ⁄ mg cell protein, before and after treatment, respectively).
a-MSH had no effect on PGE 2 production in FM55 cells but increased PGD 2 and PGD 1 with no effect on melanin (Fig. 2c). 3-Isobutyl-1-methylxanthine (IBMX), which raises cAMP levels by inhibiting phosphodiesterase, increased the production of PGD 2 and, in contrast to a-MSH, increased melanin production (Fig. 2c). As shown in Fig. 2d, the effect of a-MSH on PGD 2 production was dose-related, the maximal increase occurring in response to 10 )8 m a-MSH, a dose within the physiological range of concentrations of   Data expressed as mean ± SEM of n = 3 independent experiments. *P < 0.05, **P < 0.01 and ***P < 0.001 comparing data to control (CTR). (f) Schematic outline of the major signalling pathways involved in melanin and PGD 2 production. In FM55 cells, the MC1R does not couple to cAMP as indicated by the cross. As a consequence, a-MSH fails to stimulate melanin production, and the regulation of PGD 2 production may be via a cAMP independent pathway as indicated by the dotted line. AA, arachidonic acid; PGH 2 , prostaglandin H 2 ; MC1R, melanocortin 1 receptor; MITF, microphthalmia-associated transcription factor.
Letter to the Editor a-MSH in the epidermis (2). At 10 )7 m, a-MSH reduced PGD 2 to control levels. Similar changes were seen with PGD 1 . L-PGDS was present in FM55 melanoma cells, but protein levels were unaffected by a-MSH (Fig. 2e).

Conclusion
It has been suggested that prostaglandins have a role in the pigmentary response (3,4). However, we found no such association between prostaglandin and melanin production in FM55 melanoma cells. This dissociation was further demonstrated in experiments with a-MSH. Although this peptide is melanogenic in human melanocytes via the MC1R (10,11), it fails to have this effect in FM55 cells (18), as confirmed here, yet it increased prostaglandin production. Thus, it would seem that in FM55 cells prostaglandins are produced as part of some non-pigmentary function. PGD 2 is a major prostaglandin in both epidermal melanocytes and FM55 cells. a-MSH modulated the production of PGD 2 in a concentration-dependent manner in FM55, producing a bell-shaped dose response curve similar to that observed for melanin (10) and NO (15). As with many of its actions, it seems that a-MSH is a modulator rather than an outright stimulator.
a-MSH may act specifically to regulate PGD 2 synthesis at the level of L-PGDS (Fig. 2f). This is supported by the concomitant stimulation on PGD 1 but lack of effect on PGE 2 , indicating that a-MSH is not acting at the level of COX.
Our findings indicate that a-MSH may affect the activity, but not expression of L-PGDS. Expression of L-PGDS is upregulated by MITF (7), which is under the control of the cAMP signalling pathway (23). This would explain the increase in PGD 2 production observed in response to IBMX-dependent increased cAMP. It is unlikely that a-MSH acts in this way because the MC1R on FM55 cells does not couple to cAMP (18). We therefore propose that a-MSH acts independently of cAMP and activates L-PGDS rather than inducing its expression. Further studies using human epidermal melanocytes and melanoma cells with different degrees of pigmentation are needed to elucidate this effect of a-MSH and determine whether it is a common property of pigment-producing cells. PGD 2 can inhibit growth of human melanoma cells (24) and loss of L-PGDS expression may be important in allowing the tumor to avoid immune surveillance (25). The fact that PGD 2 is a product of immune cells, such as Langerhans cells, mast cells and macrophages (26) emphasizes its importance as an immunomodulator. Melanocytes are another potential source of cutaneous PGD 2 ; this together with the regulation of PGD 2 production by a-MSH underlines their importance as mediators of immune responses in the skin.