Melanin Standard Method: Empirical Formula 2

Natural melanins are composed of two distinct portions; a protein fraction and a chromophoric backbone. There is no unequivocal evidence for covalent bonding between these two fractions, and standard protocols used in protein purification have failed to separate the protein fraction from the chromophoric fraction. In order to study the chromophoric backbone, many workers have resorted to harsh isolation and purification protocols that are now known to degrade and damage the chromophoric portion. These artifactual melanin preparations are poor models for valid chemical, physical, and biological studies. We have developed a mild isolation and purification protocol for melanins that takes into consideration both the particulate nature of natural melanins and the stability characteristics of the chromophoric fraction. Mathematical factoring of the quantitative amino acid data into the elemental analysis was used to obtain the empirical formula of the chromophoric backbone of melanins. The analyses have shown that melanins from various sources have significantly different amino acid compositions and contents, molar C/N ratios, and empirical formulae. This method successfully differentiates melanins from a variety of sources, namely, human hair, Sepia officinalis, Sigma Chemical Company (cat. no. M8631), autoxidation of dopa, and from the feathers of Rhode Island Red chickens. Analytical results from these studies are presented and discussed.     

Spectrophotometric Characterization of Eumelanin and Pheomelanin in Hair

Mammalian melanins exist in two chemically distinct forms: the brown to black eumelanins and the yellow to reddish-brown pheomelanins. They can be quantified by HPLC analysis of pyrrole-2,3,5-tricarboxylic acid (PTCA) and aminohydroxyphenylalanine (AHP). We recently developed a spectrophotometric method for assaying the total amount of eu- and pheomelanins by dissolving melanins in Soluene-350 plus water. In this study, we examined whether absorbance at 500 nm (A₅₀₀) of the Soluene-350 solution reflects the total amount of melanins obtained by the HPLC methods, and whether the ratio of absorbances between 650 and 500 nm reflects the eumelanin/total melanin ratio in mouse hair, sheep wool, and human hair. Our findings were as follows: (1) Total melanin levels calculated from A₅₀₀ values correlate well with those obtained from PTCA and AHP values by multiplying with the following factors: for mice, PTCA × 45 + AHP × 2.5; for sheep, PTCA × 40 + AHP × 15; and for humans, PTCA × 160 + AHP × 10. (2) The A₆₅₀/A₅₀₀ ratios were higher (0.25–0.33) in black to brown hair while they were significantly lower (0.10–0.14) in yellow to red hair. These results indicate that (1) the A₅₀₀ value can be used to quantify the total combined amount of eu- and pheomelanins, and (2) the A₆₅₀/A₅₀₀ ratio can serve as a parameter to estimate the eumelanin/total melanin ratio. The present method provides a convenient way to qualitatively characterize eu- and pheomelanins in melanins produced in follicular melanocytes.     

Chemical Characterization of Melanins in Sheep Wool and Human Hair

The color of hair and wool in mammals and feathers in birds is mostly determined by the quantity and quality of melanins that are synthesized in follicular melanocytes and transferred to keratinocytes. There are two chemically distinct types of melanin pigments: the black to brown eumelanins and the yellow to reddish pheomelanins. Melanins in sheep wool and human hair of various colors were characterized by HPLC methods to estimate 5,6-dihydroxyindole-2-carboxylic acid (DHICA)-derived units in eumelanins and benzothiazine units in pheomelanins. Melanins were also characterized by spectrophotometric methods after differential solubilization in alkalies. It was demonstrated that 1) black wool in Asiatic sheep contains eumelanin with the DHICA content similar to black mouse melanin, while black to brown melanins from human hair contain much lower ratios of DHICA-derived units, comparable to the slaty mutation in mice, 2) dark brown to brown hair in human contains eumelanin whose chemical properties are indistinguishable from those of black hair, 3) dark red wool and red human hair contain pheomelanic pigments whose chemical properties are rather different from those of yellow pheomelanins in mice, and 4) light brown, blonde, and red hairs in human can be differentiated from each other with this methodology.     

Characterization of Melanogenesis in Normal Human Epidermal Melanocytes by Chemical and Ultrastructural Analysis

Chemical and ultrastructural studies were conducted to define the relationship between type of melanogenesis and fine structures of melanosomes in normal human epidermal melanocytes. Chemical analysis of epidermal melanin demonstrated that the ratio of eumelanin/pheomelanin varied individually, ranging from 1.31 to exclusively eumelanic. Ultrastructural analysis of fine structures of melanosomes revealed that spheroid melanosomes were frequently observed in melanocytes of the epidermis whose eumelanin/pheomelanin ratio was less than 5. Conversely, ellipsoid melanosomes predominated in melanocytes of the epidermis whose ratio was more than 10. On the basis of these findings, it seems reasonable to conclude that 1) normal human epidermal melanocytes synthesize both eumelanin and pheomelanin and 2) pheomelanin synthesis may be characterized by the presence of spheroid melanosomes whereas eumelanin synthesis is ascribed to ellipsoid melanosomes.     

Chemical Degradation of Melanins: Application to Identification of Dopamine-melanin

Melanocytes produce two chemically distinct types of melanin pigments, eumelanins and pheomelanins. These pigments can be quantitatively analyzed by acidic KMnO₄ oxidation or reductive hydrolysis with hydriodic acid (HI) to form pyrrole-2,3,5-tricarboxylic acid (PTCA) or aminohydroxyphenylalanine (AHP), respectively. Dark brown melanin-like pigments are also widespread in nature, for example, in the substantia nigra of humans and primates (neuromelanin), in butterfly wings and in the fungus Cryptococcus neoformans. To characterize such diverse types of melanins, we have improved the alkaline H₂O₂ oxidation method of Napolitano et al. (Tetrahedron, 51: 5913–5920, 1995) and re-examined the HI hydrolysis method of Wakamatsu et al. (Neurosci. Lett., 131: 57–60, 1991). The results obtained with H₂O₂ oxidation show that 1) pyrrole-2,3-dicarboxylic acid (PDCA), a specific marker of 5,6-dihydroxyindole units in melanins, is produced in yields ten times higher than by acidic KMnO₄ oxidation, and 2) PTCA is artificially produced from pheomelanins. The results with HI hydrolysis show that dopamine-melanin produces a 1:1 mixture of 3-amino and 4-amino isomers of aminohydroxyphenylethylamine, while the isomer ratio is about 0.2 in melanins prepared from dopamine and cysteine. These results indicate that alkaline H₂O₂ oxidation is useful in characterizing synthetic and natural eumelanins and that reductive hydrolysis with HI can be applied to analyzing oxidation products of dopamine such as neuromelanin.     

Comparative Analysis of Hair Melanins by Chemical and Electron Spin Resonance Methods

Eighteen hair samples from Karakul newborn lambs with various colors were estimated for eumelanin and pheomelanin contents (C_e and C_p, respectively) by electron spin resonance (ESR) spectrometry and by high-performance liquid chromatography (HPLC). Correlation coefficients between the values estimated by the ESR and HPLC methods were 0.96, 0.93, and 0.99 for C_e, C_p, and C_e/C_p, respectively. The high correlation coefficients show that both methods fit well for estimation of relative values of these parameters. The absolute values of C_e and C_e/C_p coincide rather well when C_e is high, but considerable discrepancies appear when C_e is low. The reasons for these discrepancies are discussed. The HPLC method appears to be more sensitive for detection of low concentrations of pheomelanin, while the ESR method fits well for mass selection purposes.     

Effects of Melanogenesis‐Inducing Nitric Oxide and Histamine on the Production of Eumelanin and Pheomelanin in Cultured Human Melanocytes

Melanin pigments produced in human melanocytes are classified into two categories; black coloured eumelanin and reddish‐yellow pheomelanin. Stimulation of melanocytes with α‐melanocyte‐stimulating hormone (α‐MSH), one of several melanogenic factors, has been reported to enhance eumelanogenesis to a greater degree than pheomelanogenesis, which contributes to hyperpigmentation in skin. Nitric oxide (NO) and histamine are also melanogenesis‐stimulating factors that are released from cells surrounding melanocytes following ultraviolet (UV) irradiation. In this study, the effects of NO and histamine on the ratio of eumelanin and pheomelanin were examined in human melanocytes, and then compared with that of α‐MSH. The amounts of eumelanin and pheomelanin were quantified using high‐performance liquid chromatography analysis after oxidation and hydrolysis of melanin. Melanogenesis was induced by the addition of α‐MSH, NO, or histamine to melanocytes. The amount of eumelanin production significantly increased with independent stimulation by these melanogenic factors, especially histamine, while that of pheomelanin significantly increased with α‐MSH and NO, but only slightly with histamine. As a result, the ratio of eumelanin and pheomelanin increased significantly with the addition of NO or histamine. These results suggest that NO and histamine, as in the case of α‐MSH, may contribute to UV‐induced hyperpigmentation by enhancing eumelanogenesis.     

Quantitative Analysis of Eumelanin and Pheomelanin in Humans,Mice, and Other Animals: a Comparative Review

The color of hair, skin, and eyes in animals mainly depends on the quantity, quality, and distribution of the pigment melanin, which occurs in two types: black to brown eumelanin and yellow to reddish pheomelanin. Microanalytical methods to quantify the amounts of eumelanin and pheomelanin in biological materials were developed in 1985. The methods are based on the chemical degradation of eumelanin to pyrrole‐2,3,5‐tricarboxylic acid and of pheomelanin to aminohydroxyphenylalanine isomers, which can be analyzed and quantitated by high performance liquid chromatography. This review summarizes and compares eumelanin and pheomelanin contents in various pigmented tissues obtained from humans, mice, and other animals. These methods have become valuable tools to study the functions of melanin, the control of melanogenesis, and the actions and interactions of pigmentation genes. The methods have also found applications in many clinical studies. High levels of pheomelanin are found only in yellow to red hairs of mammals and in red feathers of birds. It remains an intriguing question why lower vertebrates such as fishes do not synthesize pheomelanin. Detectable levels of pheomelanin are detected in human skin regardless of race, color, and skin type. However, eumelanin is always the major constituent of epidermal melanin, and the skin color appears to be determined by the quantity of melanin produced but not by the quality.     

Tyrosinase-related proteins suppress tyrosinase-mediated cell death of melanocytes and melanoma cells

The synthesis of melanin intermediates through tyrosinase (TYR) involves the production of cytotoxic free radicals. By using recombinant adenoviruses that express TYR, tyrosinase-related protein 1 (TYRP1) or DOPAchrome tautomerase (DCT), we analyzed the biological function of these proteins with regard to melanin production and the growth of melanocytes, fibroblasts, melanoma cells and nonmelanoma cancer cells. High-level expression of TYR produced newly synthesized melanin and induced cell death in all of these cells. However, when TYRP1 or DCT was coexpressed with TYR in melanocytes and melanoma cells, TYR-mediated cell death was clearly decreased. This decrease was not observed in nonmelanocytic cells. Western blot analysis and measurement of enzyme activity revealed that the expression of TYRP1 or DCT had little effect on the amount or activity of cointroduced TYR in either the melanocytic or nonmelanocytic cells. In cells expressing both TYR and TYRP1 or TYR and DCT, the total amount of melanin and/or eumelanin increased substantially more than that in cells expressing TYR alone. On the other hand, the level of pheomelanin was similar in these three cell types. These findings suggest that TYRP1 and DCT play an important role in suppressing TYR-mediated cytotoxicity in melanocytic cells without decreasing TYR expression and/or activity. These biological activities of TYRP1 and DCT may work through the interaction with TYR in melanosomal compartment.     

A Chemist's View of Melanogenesis

The significance of our understanding of the chemistry of melanin and melanogenesis is reviewed. Melanogenesis begins with the production of dopaquinone, a highly reactive o‐quinone. Pulse radiolysis is a powerful tool to study the fates of such highly reactive melanin precursors. Based on pulse radiolysis data reported by Land et al. (J Photochem Photobiol B: Biol 2001;64:123) and our biochemical studies, a pathway for mixed melanogenesis is proposed. Melanogenesis proceeds in three distinctive steps. The initial step is the production of cysteinyldopas by the rapid addition of cysteine to dopaquinone, which continues as long as cysteine is present (1 μM). The second step is the oxidation of cysteinyldopas to give pheomelanin, which continues as long as cysteinyldopas are present (10 μM). The last step is the production of eumelanin, which begins only after most cysteinyldopas are depleted. It thus appears that eumelanin is deposited on the preformed pheomelanin and that the ratio of eu‐ to pheomelanin is determined by the tyrosinase activity and cysteine concentration. In eumelanogenesis, dopachrome is a rather stable molecule and spontaneously decomposes to give mostly 5,6‐dihydroxyindole. Dopachrome tautomerase (Dct) catalyses the tautomerization of dopachrome to give mostly 5,6‐dihydroxyindole‐2‐carboxylic acid (DHICA). Our study confirmed that the role of Dct is to increase the ratio of DHICA in eumelanin and to increase the production of eumelanin. In addition, the cytotoxicity of o‐quinone melanin precursors was found to correlate with binding to proteins through the cysteine residues. Finally, it is still unknown how the availability of cysteine is controlled within the melanosome.     

The Presence of Tyrosinase and Related Proteins in Human Epidermis and Their Relationship to Melanin Type

The present study was carried out to investigate the abundance of tyrosinase and related proteins (TRP-1 and TRP-2) in human epidermis and their relationship to melanin type. Positive immunocytochemical staining was seen for all three proteins in epidermal melanocytes. For each protein the numbers of positively stained melanocytes were similar in all subjects studied irrespective of skin type. Following 5 daily suberythemal doses of UVB the melanocytes were larger, more dendritic, and increased in number. With TRP-1 and TRP-2 the increase in number in response to UVB was unrelated to skin type and, hence, with melanin type but with tyrosinase there was a much greater increase in skin types III and IV than in skin type I and II. The enhanced numbers of tyrosinase-positive melanocytes were accompanied by increased staining intensity, suggesting a greater expression of tyrosinase in the melanocytes from skin types III and IV compared with skin types I and II. This increase in tyrosinase could be related to the greater levels of eumelanin found in skin types III and IV, and this is in keeping with the view that higher levels of tyrosinase are associated with the production of eumelanin than phaeomelanin.     

The Usefulness of 4‐Amino‐3‐hydroxyphenylalanine as a Specific Marker of Pheomelanin

Reductive hydrolysis of pheomelanin with hydriodic acid (HI) gives two aminohydroxyphenylalanine isomers, 4‐amino‐3‐hydroxyphenylalanine (`specific AHP') and 3‐amino‐4‐hydroxyphenylalanine (3‐aminotyrosine, AT), which derive from the oxidative polymerization of 5‐S‐cysteinyldopa, and 2‐S‐cysteinyldopa, respectively. Since we first introduced this analytical method, the combined amount of AHP and AT (`total AHP') has been extensively used as a marker of pheomelanin. However, one problem with using total AHP as a marker is that background levels originate from precursors other than pheomelanin. Considerable and variable amounts of background AT are produced from other sources, most likely nitrotyrosine residues in proteins. In order to overcome this problem, we developed HPLC conditions which enable the direct injection of the HI reduction products into the HPLC system allowing good separation of AHP and AT. In this way we could study the importance of both degradation products separately and their specificity as markers for pheomelanin. The usefulness of the present method is validated using human hair samples of various colours which were divided into dark, fair or red colours. The combined amount of specific AHP and AT shows an excellent correlation with total AHP, and the amount of specific AHP also correlates with the amount of total AHP. We also examined total AHP and specific AHP values against pyrrole‐2,3,5‐tricarboxylic acid (PTCA) values in the human hair samples. These results show that specific AHP measurement gives a more prominent segregation for the ratio of specific AHP to PTCA among hairs of various colours than the ratio of total AHP to PTCA. Thus, we conclude that `specific AHP' is a more specific marker of pheomelanin than is `total AHP'.     

Excess tyrosine rescues the reduced activity of proliferation and differentiation of cultured recessive yellow melanocytes derived from neonatal mouse epidermis

The murine recessive yellow (Mc1r(e)) is a loss-of-function mutation in the receptor for alpha-melanocyte-stimulating hormone, melanocortin receptor 1 (Mc1r) and produces yellow coats by inducing pheomelanin synthesis in hair follicular melanocytes. However, it is not known whether the Mc1r(e) mutation affects the proliferation and differentiation of melanocytes. In this study, the proliferation and differentiation of recessive yellow epidermal melanocytes cultured in dibutyryl cyclic AMP-supplemented serum-free medium were investigated in detail. The melanocytes produced mainly eumelanin in this culture system. The proliferation of recessive yellow melanocytes was decreased compared with that of wild-type at the e-locus, black melanocytes. The differentiation of melanocytes was also delayed and inhibited in recessive yellow mice. Tyrosinase (TYR) activity and TYR-related protein 1 (TRP1) and TRP2 (dopachrome tautomerase, DCT) expressions were decreased and, in addition, the maturation of stage IV melanosomes was inhibited. Excess l-tyrosine (l-Tyr) added to the culture media rescued the reduced activity of proliferation of melanocytes. l-Tyr also stimulated TYR activity and TRP1 and TRP2 expressions as well as the maturation of stage IV melanosomes and pigmentation. These results suggest that the Mc1r(e) mutation affects the proliferation and differentiation of melanocytes and l-Tyr rescues the reduced proliferative and differentiative activities by stimulating TYR activity and TRP1 and TRP2 expressions as well as melanosome maturation.     

Ion‐Exchange and Adsorption of Fe(III) by Sepia Melanin

Sepia eumelanin is associated with many metal ions, yet little is known about its metal binding capacity and the chemical nature of the binding site(s). Herein, the natural concentrations of metal ions are presented and the ability to remove metals by exposure of the melanin granules to EDTA is quantified. The results reveal that the binding constants of melanin at pH 5.8 for Mg(II), Ca(II), Sr(II) and Cu(II) are, respectively, 5, 4, 14 and 34 times greater than the corresponding binding constants of these ions with EDTA. By exposing Sepia eumelanin to aqueous solutions of FeCl₃, the content of bound Fe(III) can be increased from a natural concentration of ～180 ppm to a saturation limit of ～80 000 ppm or 1.43 mmol/g of melanin. Similar saturation limits are found for Mg(II) and Ca(II). Exposure of Sepia melanin granules to aqueous solutions containing Ca(II) results in the stoichiometric replacement of the initially bound Mg(II), arguing that these two ions occupy the same binding site(s) in the pigment. The pH‐dependent binding of Mg(II) and Ca(II) suggests coordination of these ions to carboxylic acid groups in the pigment. Mg(II) and Ca(II) can be added to a Fe(III)‐saturated melanin sample without affecting the amount of Fe(III) pre‐adsorbed, clearly establishing Fe(III) and Mg(II)/Ca(II) occupy different binding sites. Taking recent Raman spectroscopic data into account, the binding of Fe(III) is concluded to involve coordination to o‐dihydroxyl groups. The effects of metal ion content on the surface morphology were analyzed. No significant changes were found over the full range of Fe(III) concentration studied, which is supported by the Brunauer–Emmett–Teller surface area analysis. These observations imply the existence of channels within the melanin granules that can serve to transport metal ions.     

Group Composition and Adult Sex-ratio of Hamadryas Baboons (Papio hamadryas hamadryas) in Central Eritrea

During a survey of the geographical distribution and abundance of hamadryas baboons (Papio hamadryas hamadryas) in central Eritrea, we collected detailed demographic data on six bands at four sites in different ecogeographical zones. The proportions of age-sex classes within the six bands differed only with respect to juveniles. The general social organization of the Eritrean hamadryas baboons is similar to that reported for Ethiopia and Saudi Arabia. Eritrean hamadryas baboons live in a nested fission-fusion system, with one-male units as the basic social entity. Although the baboons were not provisioned, as they are in many places in Saudi Arabia, and habitat quality in Eritrea is lower than that in Ethiopia, sex-ratios and group composition corresponded more to those found in the Saudi Arabian population. Sex-ratios within the study population, in bands and also in one-male units were significantly more female-biased than in Ethiopian ones, and one-male units tended to be larger. Data from Eritrea suggest that these differences are due to a combination of a heavily fluctuating rainfall pattern and differential maturation of the sexes.     

Cyclic flow of electrons within PSII in thylakoid membranes

In photosynthesis, the electrons released from PSII are considered to be shared mainly by carbon metabolism and the water-water cycle. We demonstrated previously that some electrons are utilized in a CO2- and O2-independent manner in leaves of wild watermelon [Miyake and Yokota (2000) Plant Cell Physiol: 41: 335]. In the present study, we examined the mechanism of this alternative flow of electrons in thylakoid membranes, isolated from fresh spinach leaves, by simultaneously measuring the quantum yield of PSII and the flux of the linear flow of electrons. In the presence of the protonophore nigericin, which eliminates the pH gradient across thylakoid membranes, the quantum yield and the flux of the linear flow of electrons were directly proportional to one another. The quantum yield at a given linear flux of electrons was much higher in the absence of nigericin than in its presence, indicating that an additional or alternative flow of electrons can occur independently of the linear flow in the absence of nigericin. In the presence of nigericin, the alternative flux of electrons increased with decreasing pH and with increasing reduction of the plastoquinone pool. Cyclic flow of electrons in PSII appears to be the most plausible candidate for the alternative flow of electrons. The flux reached 280 micromol x e(-) (mg Chl)(-1) x h(-1) and was similar to that of the CO2- and O2-independent alternative flow of electrons that we found in leaves of wild watermelon. The cyclic, alternative flow of electrons in PSII provides a possible explanation for the alternative flow of electrons observed in vivo.     

Characterization of spontaneous mammary gland carcinomas in female baboons

Spontaneous mammary gland carcinomas occurred in five baboons during a 13-year period at Southwest Foundation for Biomedical Research. The affected baboons ranged in age from 21 to 33 years. Menopause in the baboon occurs at approximately 26 years of age. All five animals had typical invasive ductal carcinoma. Morphologically, the tumors were characterized by neoplastic cells arranged from pseudopapillary and cribiform to more poorly differentiated solid cellular growth patterns. Additional features included lack of tubule formation (4/5), marked nuclear pleomorphism (5/5), a high mitotic rate (4/5) and tumor necrosis (4/5). Applying a grading system used for breast cancer in women, four tumors were graded as poorly differentiated carcinomas and one was graded as moderately differentiated. Co-existent ductal carcinoma in situ (DCIS) was observed in three of the mammary tumors. Metastases to the regional lymph nodes were confirmed in two animals, both with histological evidence of lymphovascular invasion in the primary tumor. Distant metastases were observed in only one animal. Immunohistochemical staining for human therapeutic markers revealed 2/5 tumors strongly positive for estrogen receptor, 1/5 strongly positive for progesterone receptor and 4/4 negative for HER2 expression. Although the incidence appears to be low, these five cases of mammary carcinoma in female baboons suggest that when present baboon mammary carcinoma is usually of ductal origin and behaves similar to a human breast carcinoma.     

Cyclic oscillations in melanin composition within hairs of baboons.

The wild-type agouti-banding pattern for hair is well characterized in lower mammals such as mice. The switch between eumelanin and pheomelanin in bands in the hair results from the interaction of alpha-melanocyte stimulating hormone and agouti signal protein through the melanocortin 1 receptor on melanocytes. However, such banding patterns have not been described to date in higher mammals. We now report such 'agouti'-banding patterns that occur in several subspecies of baboons, and characterize those hairs using chemical and immunohistochemical methods. Hair and skin samples were obtained from the dorsa of adult male baboons of different subspecies (Papio cynocephalus hamadryas (PCH) and Papio cynocephalus anubis (PCA)). The hairs were excised with scissors into the gray and the white bands of the PCH subspecies and into the black and the yellow bands of the PCA subspecies, and were analyzed for total melanin, eumelanin, and pheomelanin by spectrophotometric and chemical methods. Hairs in the PCA subspecies oscillate between a eumelanic band (with high melanin content) and a pheomelanic band, while hairs in the PCH subspecies oscillate between a eumelanic band (with low melanin content) and a non-pigmented band. Those chemical data are consistent with the histological appearance of the hair bulbs stained by the Fontana-Masson technique. The difference in the melanin content between PCH and PCA subspecies is most likely related to tyrosinase levels, as suggested by the presence of unpigmented muzzle in the PCH subspecies compared with the black muzzle in the PCA subspecies.     

Small molecule modulators of aggregation in synthetic melanin polymerizations

There are numerous potential applications for melanin-binding compounds, and new methods are of interest to identify melanin-binding agents. A portion of the polymerization to eumelanin, the black to brown pigment in humans, is thought to be supramolecular aggregation of nanoparticles derived from dihydroxyindoles. Starting with chloroquine, a known eumelanin-binding compound, the ability of small molecules to influence aggregation in synthetic eumelanin polymerizations was investigated. Twenty-eight compounds were tested, including pharmaceuticals, dyes, aromatics, and amines. Compounds that either accelerate or delay the appearance of macroscopic particles in synthetic eumelanin polymerizations were uncovered.