Agouti protein, mahogunin, and attractin in pheomelanogenesis and melanoblast-like alteration of melanocytes: a cAMP-independent pathway

Abstract

Melanocortin-1 receptor (MC1R) and its ligands, alpha-melanocyte stimulating hormone (alphaMSH) and agouti signaling protein (ASIP), regulate switching between eumelanin and pheomelanin synthesis in melanocytes. Here we investigated biological effects and signaling pathways of ASIP. Melan-a non agouti (a/a) mouse melanocytes produce mainly eumelanin, but ASIP combined with phenylthiourea and extra cysteine could induce over 200-fold increases in the pheomelanin to eumelanin ratio, and a tan-yellow color in pelletted cells. Moreover, ASIP-treated cells showed reduced proliferation and a melanoblast-like appearance, seen also in melanocyte lines from yellow (A(y)/a and Mc1r(e)/ Mc1r(e)) mice. However ASIP-YY, a C-terminal fragment of ASIP, induced neither biological nor pigmentary changes. As, like ASIP, ASIP-YY inhibited the cAMP rise induced by alphaMSH analog NDP-MSH, and reduced cAMP level without added MSH, the morphological changes and depigmentation seemed independent of cAMP signaling. Melanocytes genetically null for ASIP mediators attractin or mahogunin (Atrn(mg-3J/mg-3J) or Mgrn1(md-nc/md-nc)) also responded to both ASIP and ASIP-YY in cAMP level, while only ASIP altered their proliferation and (in part) shape. Thus, ASIP-MC1R signaling includes a cAMP-independent pathway through attractin and mahogunin, while the known cAMP-dependent component requires neither attractin nor mahogunin.

Figure 1

Morphological and pigmentary changes induced in melan-a cells by agouti signaling protein (ASIP). Paired phase contrast (left) and bright-field (right, to show eumelanin) images of melan-a cells grown in normal growth medium (C, control); or in the same medium with 10 nM ASIP added for the last 5 days (ASIP). No phenylthiourea (PTU) or extra cysteine was present (see text). (mb): normal murine melanoblasts in a primary neonatal epidermal melanocyte culture, 7 days after explantation, showing similarity to melan-a cells grown with ASIP. Bar = 100 μm.

Figure 2

Marked increases in pheomelanin/eumelanin ratio and tan coloration by agouti signaling protein (ASIP), cysteine and phenylthiourea (PTU). Cells were subcultured as needed to prevent confluency. (A) Eumelanin and pheomelanin content of melan-a cells after 14 days’ growth with 300 μM PTU followed by 14 days’ growth in standard medium [containing 12-O-tetradecanoyl phorbol-13-acetate (TPA)] with the indicated additives. Data show mean ± SEM of three different samples. (ASIP): 10 nM ASIP; (Cys): 2 mM cysteine plus 100 μM mercaptoethanol; (PTU): 100 μM PTU. (B) Pellets of cells after 14 days’ growth with 500 μM PTU followed by 5 days’ growth with the indicated additives. (A/C): 10 nM ASIP, 2 mM cysteine and 100 μM mercaptoethanol; (TPA): 200 nM TPA; (PTU): PTU at 500 μM for 2 days and 100 μM for 3 days. Significances of selected differences by Student’s t-test (two-tailed) are shown: *P < 0.05; **P < 0.01; ***P < 0.001, (NS): P > 0.05, not significant.

Figure 3

Melanoblast-like appearance of melan-Ay and melan-e cells in the absence of CT. Paired phase-contrast (left) and bright-field (right) images of melan-Ay (Ay) and melan-e (e) cells grown with 20 pM CT (CT) or for several passages without CT (0). Some residual eumelanin is seen in some cells of both genotypes without CT. For reference melan-a cells (a) grown with 20 pM CT are also shown; there is little difference from melan-a cells without CT (Figure 1). Bar = 100 μm.

Figure 5

(A) Inhibition of NDP-MSH-induced cAMP rise by both agouti signaling protein (ASIP) and ASIP-YY in melan-a cells. Intracellular cAMP per well was measured in cultures that had been plated and allowed to attach overnight, after stimulation for 30 min with 30 pM NDP-MSH (M), with or without ASIP (A) or ASIP-YY (Y). Figures in parentheses are concentration of ASIP or ASIP-YY in nM. *RPMI1640 medium with 10% FBS was incubated with ASIP or ASIP-YY at 37°C overnight and was used as stimulating buffer, followed by application of NDP-MSH. Shown are mean ± SEM of triplicate wells. (B) Inhibition of NDP-MSH-induced cAMP rise by both ASIP and ASIP-YY in melan-mg and melan-md cells. Assay as above, using 100 nM ASIP (A) or 100 nM ASIP-YY (Y). The reason for the lower apparent stimulation of melan-a cells by NDP-MSH in (B) than in (A) is unknown; it may reflect change in the assay reagents with time. However the relative responses of different genotypes are clear. (C) Reduction of basal cAMP by both ASIP and ASIP-YY. In this experiment only, to increase assay sensitivity, melan-a cells were plated at a higher number of 6 × 10⁴ cells/well, preincubation with isobutylmethylxanthine (IBMX) was for 5 h at 1 μM, and incubation with the stated additives was for 5 h. IBMX alone (1 μM, a low concentration) had no significant effect, whereas both ASIP (A) and ASIP-YY (Y) at 10 nM gave significant reductions in cAMP levels compared with 1 μM IBMX alone (P = 0.013, P = 0.003 respectively). Similar effects were seen under slightly varied conditions (not shown).

Figure 4

Melanoblast-like shape-change, depigmentation and growth inhibition of melan-a cells by agouti signaling protein (ASIP) but not ASIP-YY. (A) phase-contrast images of melan-a cells grown for 5 days in growth medium (0), with 10 nM ASIP (ASIP) or with 100 nM ASIP-YY (ASIP-YY). Bar = 100 μm. (B) Total melanin content of melan-a cells grown in growth medium only, or with ASIP or ASIP-YY for 14 days. Mean and SEM of triplicate dishes. (C) Cell doubling times of melan-a cells subcultured with no addition, ASIP or ASIP-YY for 30 days. Shown are mean ± SEM of triplicates.

Figure 6

Melanoblastic shape-change and growth inhibition of melan-mg and melan-md cells by agouti signaling protein (ASIP) but not ASIP-YY. Melan-mg cells (mg) and melan-md cells (md) were incubated for 5 days in growth medium alone (0), with 10 nM ASIP (ASIP) or with 10 nM ASIP-YY (YY). Bar = 100 μm. No Cys or phenylthiourea (PTU) present. Cell doubling times of melan-mg cells (B) and melan-md cells (C) were measured in the presence of 10 nM ASIP or 10 nM ASIP-YY for 30 days. Shown are mean ± SEM of triplicate cultures. Retardation of proliferation by ASIP was highly significant for both genotypes by Student’s t-test (P < 0.001), while ASIP-YY had no significant effect.

Figure 7

Model to account for responsiveness to cAMP level in activation but not inactivation of eumelanin synthesis. See Discussion for main explanation. (A-YY) ASIP-YY, (Mc) melanocyte, (Mb) melanoblast, (Ub) ubiquitinylation. Melanocortin-1 receptor (MC1R) is shown with some basal activity (left, lost in e mutant), to explain why POMC null, a/a mice are black, and related data (see text). (This might alternatively be due to BD103 or some unknown signal). Agouti signaling protein (ASIP) would not be required for pheomelanogenesis unless Y has been switched on by the MC1R/cAMP pathway, hence inactive MC1R would give a yellow color even with deficient MGRN1, as seen in e/e, md/md mice.