Keratinocytes in culture accumulate phagocytosed melanosomes in the perinuclear area


Hideya Ando, e-mail:


There are many techniques for evaluating melanosome transfer to keratinocytes but the spectrophotometric quantification of melanosomes incorporated by keratinocyte phagocytosis has not been previously reported. Here we describe a new method that allows the spectrophotometric visualization of melanosome uptake by normal human keratinocytes in culture. Fontana-Masson staining of keratinocytes incubated with isolated melanosomes showed the accumulation of incorporated melanosomes in the perinuclear areas of keratinocytes within 48 h. Electron microscopic observations of melanosomes ingested by keratinocytes revealed that many phagosomes containing clusters of melanosomes or their fragments were localized in the perinuclear area. A known inhibitor of keratinocyte phagocytosis which inhibits protease-activated receptor-2, i.e., soybean trypsin inhibitor, decreased melanosome uptake by keratinocytes in a dose-dependent manner. These data suggest that our method is a useful model to quantitate keratinocyte phagocytosis of melanosomes visually in vitro.


In the basal layer of the epidermis, specialized cells, termed melanocytes, produce the pigment melanin, which is transferred to neighboring epidermal keratinocytes. Melanin is synthesized and deposited in melanosomes, specialized organelles produced only by melanocytes. In the skin and hair, melanosomes are transferred from melanocytes to neighboring keratinocytes to form melanin caps above their nuclei as an internal sunscreen (Hearing, 2005).

Three possible mechanisms of melanosome transfer have been suggested, i.e., (i) pinching off of melanocyte dendrites containing melanosomes by keratinocytes, (ii) direct inoculation of melanosomes into keratinocytes via keratinocyte-melanocyte membrane fusions through nanotubular filopodia, and/or (iii) melanosome release into the extracellular space followed by their phagocytosis by keratinocytes (van den Bossche et al., 2006; Scott et al., 2002; Seiberg, 2001; Singh et al., 2008; Yamamoto and Bhawan, 1994). The finding that activation or inhibition of protease-activated receptor-2 (PAR-2), a seven-transmembrane G-protein-coupled receptor expressed by keratinocytes but not by melanocytes (Seiberg et al., 2000b), regulates melanosome transfer via keratinocyte phagocytosis (Boissy, 2003; Cardinali et al., 2005; Seiberg et al., 2000b; Sharlow et al., 2000) reinforces the notion that exocytosis and phagocytosis, at least in part, are involved in the machinery of melanosome transfer. However, the exact mechanism of the transfer process of melanosomes from melanocytes to keratinocytes remains to be characterized (van den Bossche et al., 2006; Hearing, 2007). To elucidate that mechanism, a few techniques to evaluate melanosome transfer in vitro have been developed (Berens et al., 2005). Melanocyte-keratinocyte co-cultures have been used to evaluate both steps, i.e., exocytosis and phagocytosis, of melanosome transfer in vitro (Minwalla et al., 2001a,b; Virador et al., 2002). Evaluation of the first step of melanosome transfer, i.e., the exocytosis of melanosomes by melanocytes, can be measured by the chemical analysis of melanin released into the culture medium (Virador et al., 2002), while the latter step of melanosome transfer, i.e., the phagocytosis of melanosomes by keratinocytes, can be evaluated by measuring the internalization of latex microsphere beads by keratinocytes (Cardinali et al., 2005, 2007; Virador et al., 2002).

Together with the identification of the importance of keratinocyte expression of PAR-2 in melanosome transfer, several studies have supported the notion that keratinocytes play a pivotal role in the melanosome transfer process. Keratinocytes derived from dark skin markedly stimulate the expression of melanocyte-specific proteins by co-cultured melanocytes, such as microphthalmia-associated transcription factor and tyrosinase, regardless of whether the melanocytes are derived from light or dark skin, indicating that keratinocytes, at least in part, regulate skin pigmentation (Minwalla et al., 2001b; Yoshida et al., 2007). Further, the internalization of latex beads by keratinocyte phagocytosis has been shown to be activated by keratinocyte growth factor which acts only on the recipient keratinocytes (Cardinali et al., 2005, 2007). To evaluate the phagocytosis of melanosomes by keratinocytes, fluorescent latex beads have been used as a surrogate for melanosomes in many studies, and melanosomes have also been used in a few studies (Seiberg et al., 2000a; Virador et al., 2002). However, the spectrophotometric quantification of phagocytosed melanosomes by keratinocytes using isolated melanosomes has not been previously reported. In this study, we developed a visual and quantitative method to evaluate melanosome transfer focusing on the possible mechanism of melanosome incorporation by keratinocyte phagocytosis in that process. Using that method, the time-dependent changes in the distribution pattern of phagocytosed melanosomes in keratinocytes was examined by light and electron microscopy (see Appendix S1).


Establishment of optimal in vitro conditions for melanosome uptake by keratinocyte phagocytosis

Preliminary experiments were performed to determine the optimal quantity of melanosomes that allows visible pigmentation in a pellet of 8 × 105 keratinocytes (data not shown). After incubation with isolated melanosomes, the amount of incorporated melanin content increased in a time-dependent manner (Figure 1A, B). The uptake of melanosomes was almost saturated at 48 h and there was no statistical difference in melanin content between 48 h and 72 h.

Figure 1.

 Uptake of melanosomes by keratinocytes: (A) Keratinocyte pellets after incubation with isolated melanosomes for 0–72 h. The cells at 0 h were washed immediately after the addition of isolated melanosomes. (B) Graph of the amount of incorporated melanin in keratinocytes; data are expressed as absorbance at 490 nm and represent mean values ±SD of triplicate determinations. (C) Light microscopic images (bright field) of keratinocytes after incubation with isolated melanosomes for 0–72 h. Fontana-Masson staining was performed to demonstrate the localization of incorporated melanosomes in keratinocytes. Scale bars, 100 μm.

Light microscopic observation after Fontana-Masson staining revealed that the ingested melanosomes were evenly dispersed in the cytosol of keratinocytes at 24 h, after which the melanosomes were found primarily in the perinuclear area (Figure 1C). The relative intensity and rate of appearance of Fontana-Masson staining was similar to that measured by absorbance at 490 nm.

Electron microscopic observations of melanosome uptake

Many melanosomes exist in the cytosol of MNT-1 human melanoma cells (Figure 2A). In contrast, some mitochondria were found but no melanosomes were observed in the cytosol of cultured normal human keratinocytes prior to incubation with isolated melanosomes (Figure 2B). The melanosome-rich fraction contained a large number of melanosomes (Figure 2C) and when observed at a higher magnification, various stages of melanosomes were observed interspersed with tiny vesicles that are likely ribosomes (Figure 2D). After incubation with isolated melanosomes for 48 h, the localization of melanosomes incorporated was observed in the perinuclear areas of the keratinocytes (Figure 2E), which was similar to the localization seen with the Fontana-Masson staining (Figure 1C). In the cytosol of the melanosome-ingested keratinocytes at higher magnification, phagosomes or phagolysosomes (a kind of secondary lysosome called a multivesicular body) enclosing clusters of melanosomes and their fragments within a monolayer membrane were observed (Figure 2F).

Figure 2.

 Electron microscopic observation of cells and isolated melanosomes: (A) A representative MNT-1 human melanoma cell from which melanosomes were isolated. (B) A representative normal human keratinocyte prior to the incubation of isolated melanosomes. (C) Isolated melanosomes observed in the melanosome-rich fraction at low magnification. (D) Isolated melanosomes at a higher magnification. (E) Incorporated melanosomes localized in the perinuclear area of a keratinocyte. The dense circle in the nucleus is the nuclear body. (F) Higher magnification of the incorporated melanosomes enclosed within a phagosome in the cytosol of a keratinocyte. Note that the ingested melanosomes are surrounded by a monolayer membrane. Scale bars, 2 μm.

Evaluation of an inhibitor of melanosome uptake on keratinocyte phagocytosis

To further evaluate the capability of this method to measure melanosome uptake by keratinocytes, we used STI, a known inhibitor of keratinocyte phagocytosis that has been shown to inhibit PAR-2 activation and to reduce skin pigmentation (Paine et al., 2001). Keratinocytes after incubation with isolated melanosomes in the presence of STI contained less melanin than keratinocytes in the absence of STI in a dose-dependent manner (Figure 3A). The melanin content per protein was statistically decreased by STI compared to the control (0 μg STI/ml) (Figure 3B). Keratinocyte proliferation determined by protein concentration was unaffected by STI over 48 h (Figure 3C).

Figure 3.

 Soybean trypsin inhibitor inhibits keratinocyte phagocytosis: (A) Appearance of keratinocyte pellets after incubation with isolated melanosomes without or with STI (0.5–2 mg/ml) for 48 h. (B) Bar graphs of incorporated melanin/mg protein in keratinocytes, and (C) protein concentration of cell pellets after incubation with isolated melanosomes in the absence or presence of 0.5, 1 or 2 mg/ml STI for 48 h. Data are expressed as a percentage of control cells untreated with STI (100%) and are mean values ±SD of triplicate determinations. Statistical analysis was performed using the Dunnet II test (**P < 0.01 versus the control).


Melanosome transfer from melanocytes to keratinocytes, at least in in vitro, is thought to proceed through melanosome release by melanocyte exocytosis and subsequent melanosome uptake by keratinocyte phagocytosis (van den Bossche et al., 2006; Cardinali et al., 2007; Virador et al., 2002). Co-cultures of melanocytes and keratinocytes are suitable to evaluate the process of exocytosis-phagocytosis simultaneously, whereas the method reported here can evaluate melanosome uptake by keratinocyte phagocytosis independent of melanocyte factors. The method described here demonstrates that exogenously added melanosomes are taken up by normal human keratinocytes in a time-dependent manner, reflecting a possible melanosome transfer process in which melanosomes released into the extracellular space are phagocytosed by keratinocytes. In addition, light and electron microscopic observations of this model system show that the incorporated melanosomes accumulate and are localized in the perinuclear areas of keratinocytes.

The accumulation and localization of phagocytosed melanosomes in the perinuclear area of keratinocytes after more than 48 h incubation is consistent with a similar distribution of fluorescent latex beads or melanosomes transferred in the presence of a cAMP inducer (Cardinali et al., 2005, 2007; Singh et al., 2008). The localization of incorporated melanosomes in the perinuclear area is also similar in distribution to the supranuclear melanin caps made of transferred and phagocytosed melanosomes that are found in keratinocytes in vivo and in cultured reconstructed pigmented epidermis (Boissy, 2003; Byers et al., 2003; Gibbs et al., 2000). The perinuclear aggregation of phagocytosed melanosomes in keratinocytes has been found to be mediated, at least in part, by the intermediate chain of cytoplasmic dynein, a microtubule-associated motor molecule involved in the retrograde transport of membrane-bound organelles (Byers et al., 2003). In that study, the intermediate chain of cytoplasmic dynein was observed to accumulate in the perinuclear area of cultured human keratinocytes and this observation is consistent with the present result that melano-phagolysosomes accumulate in the perinuclear area.

It is also known that keratinocytes play a pivotal role in regulating the distribution patterns of melanosomes in vitro and in vivo, i.e., the melanosomes in keratinocytes from dark skin are distributed individually, while melanosomes in keratinocytes from light skin are distributed in clusters (Minwalla et al., 2001b; Thong et al., 2003). The keratinocytes used in the present study were from moderately pigmented skin and the melanosomes incorporated into keratinocytes were observed in membrane-bound clusters, similar to the previously reported distribution in light skin-derived keratinocytes. Further, melanosomal membranes were observed to be removed due to the detergent treatment during the purification process (Figure 2D), suggesting that the melanosomal membrane is not important/critical for the ingestion by keratinocytes. Whether the existence of a melanosomal membrane could affect the efficiency of melanosome incorporation by keratinocyte phagocytosis remains to be determined. In any case, the ingested melanosomes in keratinocytes are surrounded by the keratinocyte-derived phagolysosome membrane.

To date, agents or factors that can inhibit melanosome transfer have been reported, including RWJ-50353, a serine protease inhibitor that reduces PAR-2 activation (Seiberg et al., 2000a), STI (Paine et al., 2001) and niacinamide (Greatens et al., 2005). Among those, STI, a reported inhibitor of keratinocyte phagocytosis operating through PAR-2 inhibition, has been shown to inhibit melanosome transfer in the keratinocyte-melanocyte coculture and in cultured artificial skin experiments (Paine et al., 2001). In the present study, STI inhibited melanosome uptake by keratinocytes in a dose-dependent manner, indicating that this method can also serve as a useful model to clarify the mechanistic function of an agent that showed the regulatory effect of melanosome transfer in the coculture system or epidermal equivalents containing melanocytes in vitro.

In conclusion, we have developed an alternative and useful method to evaluate melanosome transfer by focusing on keratinocyte incorporation of melanosomes. This method will help to clarify the role of skin phototype in determining the distribution pattern of melanosomes in the keratinocyte cytosol, and may also be useful to explore the mechanisms of melanosome degradation observed in the skin and hair (Boissy, 2003).