Chemical Methods for Characterizing Eumelanin and Pheomelanin
Previous methods used for the quantification of melanins in pigmented tissues required the isolation of melanins. Moreover, none of those methods were suitable for distinguishing between eumelanin and pheomelanin. Extensive degradative studies provided a number of chemical degradative methods (1, 7). Among them, it should be noted that pyrrole-2,3,5-tricarboxylic acid (PTCA) was obtained from sepiomelanin in 6.5% yield using hydrogen peroxide at pH 7 followed by alkaline hydrolysis (8). Fattorusso et al. (8) reported that permanganate oxidation of pheomelanin yielded 1,3-thiazole-4,5-dicarboxylic acid (TDCA) and 1,3-thiazole-2,4,5-tricarboxylic acid (TTCA) derived from the benzothiazole units (Fig. 2).
Figure 2. . Structures of eumelanin, pheomelanin, products obtained by oxidation of eumelanin and pheomelanin with permanganate or hydrogen peroxide, and products obtained by reductive hydrolysis of pheomelanin with hydriodic acid.
Download figure to PowerPoint
In 1983, we introduced a rapid method for quantitatively analysing eumelanin and pheomelanin in tissue samples, which made the isolation of melanin pigments unnecessary (9). This method is based on the formation of PTCA by permanganate oxidation of eumelanin and of aminohydroxyphenylalanine (AHP) isomers by hydriodic acid (HI) reductive hydrolysis of pheomelanin, respectively. These specific degradation products are determined by HPLC: PTCA is quantified with UV detection while AHP is determined with electrochemical detection. The original method was later improved to increase the sensitivity and to reduce the time for pre-HPLC work-up (10, 11). The improved method requires only 1–5 mg of tissue samples or 106 cultured cells for each PTCA and AHP measurement. Amounts of eumelanin and pheomelanin can be calculated by multiplying the amounts of PTCA and AHP by factors of 50 and 5, respectively. These conversion factors are based on the 2% yield of PTCA from sepiomelanin and the 20% yield of AHP from synthetic 5-S-CD-melanin, respectively (9, 10, 12).
Our HPLC method is relatively simple, fairly rapid and highly sensitive. It has been applied to quantitatively analyse eumelanin and pheomelanin, not only in synthetic melanins, isolated melanosomes, hair, feathers, skin, nevi and melanomas, but also in human epidermis and cultured melanocytes (2, 13, 14). Moreover, the same degradative methodology has been applied to characterize neuromelanin (15–17; described later).
Formation of PTCA was interpreted in terms of the oxidative breakdown of indole units, either linked through the 2-position or bearing a carboxyl group at the same position (18). In our hands, permanganate oxidation of DHI-melanin and DHICA-melanin in acidic medium produced 0.03% and 2.8% PTCA, respectively (2). Similar results were also reported by Nicolaus's group (18). These results indicate that PTCA is a specific product arising from DHICA-derived structures. Amount of DHICA-melanin can thus be calculated by multiplying the amount of PTCA by a factor of 35 on the basis of the 2.8% yield of PTCA from DHICA-melanin.
A drawback in the acidic permanganate oxidation is that it does not produce appreciable amounts of pyrrole-2,3-dicarboxylic acid (PDCA) either from natural eumelanin or from DHI-melanin (4, 10). Napolitano et al. (19) have re-examined conditions for the oxidative degradation of eumelanin: Using alkaline hydrogen peroxide as the oxidizing agent, they improved the yield of PTCA to 6.1% from DHICA-melanin and obtained PDCA in a significant (0.46%) amount from DHI-melanin. It should be noted, however, that a comparable (0.39%) amount of PTCA was also obtained from DHI-melanin.
Spectrophotometric Methods for Analysing Eumelanin and Pheomelanin
Although our melanin assay based on chemical degradation and HPLC determination is relatively simple, it still requires an HPLC system equipped with UV and electrochemical detectors. In addition, PTCA arises from the DHICA-derived units but not from the DHI-derived units present in the eumelanin polymer. Therefore, we have developed a spectrophotometric method that is specific to eumelanin but does not discriminate between the DHI- and DHICA-derived units (20). In this method, hair and melanoma samples were hydrolysed in hot HI to remove pheomelanic components, and the insoluble eumelanic pigments were solubilized in hot NaOH in the presence of H2O2 and were analysed for absorbance at 350 nm. Although much less sensitive, this spectrophotometric method can substitute for the PTCA method for eumelanin assay when substantial amounts of samples are available.
Spectrophotometric measurement of melanin pigments usually employs hot NaOH solution to solubilize melanin from tissue samples. However, the method is not suitable for hair samples because of the insolubility of eumelanin. We found that hot Soluene-350 in the presence of 10% water is able to completely dissolve mouse hair, sheep wool, and human hair samples (21–23). The resulting brown solutions were analysed for absorbances between 400 and 800 nm. It was found that soluene-350 solutions of eumelanin and pheomelanin differ considerably in visible absorption spectra, with eumelanin giving flatter spectra (24). The absorbance at 500 nm (A500) value may serve as an indicator of the total combined amount of eumelanin and pheomelanin. We have also found that when solubilized in Soluene-350, eumelanin (ratio 0.25–0.33) and pheomelanin (ratio 0.10–0.14) in hair show significantly different ratios of absorbances at 650–500 nm (A650/A500). The absorbance ratio can be used as a parameter to estimate the relative ratio of eumelanin to pheomelanin (24).
We measured the A500 values of the Soluene-350 solution from hairs and compared the amounts of eumelanin and pheomelanin calculated from values of PTCA and AHP using conversion factors that fit best (24). We obtained the following conversion factors: for PTCA, factors are 45, 40, and 160 in mice, sheep, and humans, respectively; for AHP, factors are 2.5, 15, and 10. These conversion factors indicate that the yields of PTCA from eumelanin are 2.2, 2.5, and 0.6% in mice, sheep, and humans, respectively. Natural eumelanin is produced by copolymerization of DHI and DHICA in various ratios which appear to be governed chiefly by the activity of Tyrp2 (25, 26). Thus, the low yield of PTCA in human hairs suggests a low activity of Tyrp2 in humans as compared with other species. Alaluf et al. (27) recently reported that the HPLC method underestimates the melanin content of human epidermis by a factor of 3 as compared with the spectrophotometric method (NaOH solubilization). This discrepancy can be solved using a conversion factor of 160 instead of 50. The conversion factors for AHP also suggest that the yields of AHP from pheomelanin are 40, 7, and 10% in mice, sheep, and humans. However, it should be cautioned that what we measure with the spectrophotometric methods (regardless of solubilizing agents used) is not the concentration of the pigment but the intensity of absorbance of the pigment (24). The large difference in the conversion factors for AHP might be ascribed to the difference in the colour intensity (22). In this regard, it is interesting to note that the colours of pheomelanic sheep wool and human hairs are dark red while those of lethal yellow and recessive yellow mouse hairs are yellowish.
The total amount of melanin, as measured by A500, is useful in characterizing synthetic and natural melanins in combination with PTCA and AHP measurements (21–23, 28, 29). For example, the PTCA/total melanin ratio in slaty mouse hair was only one-fifth that of black hair, suggesting that slaty hair melanin contains DHICA-derived units at a level one-fifth that of black (21). This is caused by the decreased activity of Tyrp2 in slaty mice. The AHP/total melanin ratio is also useful in characterizing synthetic as well as natural pheomelanic pigments (21–23, 29).
Although the spectrophotometric method is useful, it has some disadvantages: 1) Background absorbance arises from tissue components such as proteins, as indicated by the low, but significant absorbance from white or albino hairs (21–23), and 2) the viscosity of Soluene-350 requires caution for reproducible measurements.