The fundamental building blocks of red human hair pheomelanin are isoquinoline‐containing dimers

Dear Editor, Pheomelanins, the reddish-brown, sulfur-containing subgroup of melanin pigments associated with distinct mutations of the melanocortin-1 receptor (MC1R) gene (Garcia-Borron et al., 2005), have traditionally been implicated in the abnormal susceptibility of red-haired individuals to sunburn and skin cancer, including melanoma (Simon and Peles, 2010). Recent data (Panzella et al., 2010; Ye et al., 2008) supported the view that pheomelanins act as potent photosensitizers leading to production of toxic reactive oxygen species. So far, however, a definitive assessment of the role of pheomelanin in the UV susceptibility trait has been hampered by the lack of detailed knowledge of the fundamental molecular constituents of these pigments. Pheomelanin formation is biochemically sustained by elevated levels of cysteine (Ito and Wakamatsu, 2008) leading to 5-S-cysteinyldopa (5-S-CD) and minor amounts of 2-S-cysteinyldopa which on oxidation lead to benzothiazine intermediates undergoing dimerization, cross-coupling, or ring contraction to benzothiazole units by complex, as yet poorly elucidated mechanisms (Greco et al., 2009; Napolitano et al., 2001; Wakamatsu et al., 2009). Altogether, these molecular components may contribute to the overall chromophoric features and photoactivity of pheomelanins in the UV-visible region (Napolitano et al., 2008; Ye et al., 2008). Chemical degradation of pheomelanins leads to a series of fragments including 4-amino-3-hydroxyphenylalanine (AHP), a benzothiazole carboxylic acid derivative (BTCA), and a thiazolyl-pyridine pentacarboxylic acid (TPCA) (see Figure S1). AHP is formed by reductive hydrolysis of benzothiazine but not benzothiazole units (Wakamatsu et al., 2009), BTCA results from the alkaline H2O2 degradation of benzothiazole or benzothiazine units (Greco et al., 2009; Panzella et al., 2007), and TPCA has recently been shown to derive from permanganate degradation of a dimeric benzothiazolylthiazinodihydroisoquinoline intermediate (BT-TIQ) in the oxidative polymerization of 5-S-CD in the presence of zinc ions (Greco et al., 2011). This latter study underscored the unique structural significance of TPCA, which is the sole two-ring degradation fragment of pheomelanins specifically related to BT-TIQ or structurally close dimeric units. Herein, we report the results of the first three-marker (TPCA, BTCA, and AHP) analysis of red hair pheomelanin and synthetic models and provide quantitative evidence that BT-TIQ-related structures are major building blocks of the human pigment. Data in Table 1 show the relative yields of the three markers from a red human hair sample. Remarkably, like human pheomelanin, also the dimer BT-TIQ gave on degradation all of the three pheomelanin markers, i.e., besides TPCA, also AHP and BTCA (Figure 1). Two synthetic pheomelanins, namely dopa/cysteine melanin and a pigment prepared by 5-S-CD oxidation in the presence of zinc ions (see Appendix S1), were also analyzed for comparison. Both TPCA/AHP and TPCA/BTCA ratios indicated that the human pigment is extraordinarily rich in BT-TIQ-related units and is similar to the 5-S-CD/Zn2+ pigment relative to the dopa-cysteine pheomelanin. Assuming the 5-S-CD/Zn2+ pigment as “pure” pheomelanin, a hair pheomelanin content of 0.6% can be estimated based on the yields of TPCA, in very good agreement with the values estimated on the basis of AHP yields (Ito and Fujita, 1985). The effect of zinc ions on the structure of synthetic pheomelanins prepared by 5-S-CD oxidation as evidenced by the high TPCA yields can be explained considering that Zn2+ ions can increase the proportion of carboxylated benzothiazine intermediates (Napolitano et al., 2001, 2008), which are crucial precursors to BT-TIQ-like units (Greco et al., 2011). Quite surprisingly, TPCA/AHP and TPCA/BTCA ratios for the red hair pigment were higher than for the model dimer itself. This apparent paradox requires further investigation. It is possible that the protein matrix in red hair spares the units from extensive degradation during the oxidative treatment with respect to the pure dimer. It is also likely that the BT-TIQ system in pheomelanin behaves differently with respect to the pure compound because of altered mechanisms of oxidative breakdown when linked to other structural units. Origin of chemical degradation products of the model dimer BT-TIQ. Thiazolyl-pyridine pentacarboxylic acid-generating subunits are marked by solid line. However, it cannot be questioned, from data in Table 1, that red hair pheomelanin gives relatively high yields of TPCA that denotes very high proportions of BT-TIQ-related moieties, also compared to the synthetic pigment. Similar TPCA/AHP and TPCA/BTCA ratios were obtained from different red human hair samples (Table S1) indicating a homogeneous molecular composition. In all cases, hair samples were taken near to the root, to minimize effects of photodegradation of the pigment with associated lowering of the yields of the markers from 5-S-CD-related units (Greco et al., 2009). A set of synthetic pheomelanins prepared under different conditions were also analyzed for TPCA yield (see Table S2). Up to fourfold higher yields of TPCA with respect to controls were obtained for the pigments synthesized in the presence of Zn2+ ions, confirming the presence of BT-TIQ-related units to a more significant extent. The UV-vis absorption spectra of BT-TIQ and a pheomelanin isolated from red hair were compared under acidic conditions (Figure S2). The pigment showed a distinct band around 400 nm very similar to that of BT-TIQ arising from the acid-induced shift of the less defined band at 360 nm at neutral pH (Greco et al., 2011). As this feature does not match with those of any of the known pheomelanin intermediates (Napolitano et al., 2008), it seemed reasonable to invoke BT-TIQ-like structural units as the factor responsible for the absorption band at 400 nm in the human pheomelanin spectrum. In conclusion, we have used for the first time a TPCA-based, three-marker analysis to demonstrate that the dimeric benzothiazolylthiazinoisoquinoline units recently identified in natural and synthetic pheomelanins (Greco et al., 2011) are dominant structural building blocks of red human hair pheomelanin. This finding would prompt a revisit of pheomelanin build-up in terms of coupling of carboxylated benzothiazine intermediates into a specific BT-TIQ-related dimeric building block preceding further polymerization. Thus, BT-TIQ would reproduce all of the main chemical properties and spectral features of human pheomelanin. In this sense, BT-TIQ compares to none of the low molecular weight intermediates so far isolated, including the trichochromes, and should, therefore, be duly considered in future models of pheomelanin structure. MdI acknowledges support from MIUR (PRIN 2008). This work was carried out in the frame of the EuMelaNet project (http://www.espcr.org/eumelanet/). Figure S1. Structure, formation conditions, and structural significance of main pheomelanin markers. Figure S2. UV-visible absorption spectra of human red hair pheomelanin (bold line) and BT-TIQ (light line) in 0.1 M HCl Table S1. Yields of degradation products from red human hair samples. Table S2. Yields of TPCA from alkaline permanganate degradation of synthetic pheomelanin pigments. Appendix S1. Experimental procedures. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.