Melanin processing by keratinocytes: A non‐microbial type of host‐pathogen interaction?

Abstract

The mechanisms that regulate skin pigmentation have been the subject of intense research in recent decades. In contrast with melanin biogenesis and transport within melanocytes, little is known about how melanin is transferred and processed within keratinocytes. Several models have been proposed for how melanin is transferred, with strong evidence supporting coupled exo/endocytosis. Recently, two reports suggest that upon internalization, melanin is stored within keratinocytes in an arrested compartment, allowing the pigment to persist for long periods. In this commentary, we identify a striking parallelism between melanin processing within keratinocytes and the host‐pathogen interaction with Plasmodium, opening new avenues to understand the complex molecular mechanisms that ensure skin pigmentation and photoprotection.

1 SYNOPSIS

Melanin processing within keratinocytes remains elusive. Recent reports suggest that melanin escapes degradation and remains in an arrested compartment inside keratinocytes for long periods. In this commentary, we highlight the similarities between melanin processing and the host‐pathogen interaction that occurs with Plasmodium during liver infection and point out some of the most important questions yet to be answered.

The skin is the largest organ of the human body and alongside its barrier properties against external aggressions, it has an important ultraviolet (UV)‐protective function. The superficial layer, the epidermis is composed by two main cell types: keratinocytes and melanocytes. Keratinocytes form a stratified epithelium differentiating from the basal layer to the cornified superficial layers. Melanocytes are residents of the basal layer and are responsible for the synthesis, packaging and transfer of the pigment melanin to neighboring keratinocytes. Photoprotection is achieved by the polarization of melanin around the nucleus of keratinocytes, forming a supra‐nuclear melanin cap or “parasol” on the sun‐exposed side, effectively protecting the nuclear DNA from UV‐induced damage.1 Skin pigmentation is, therefore the result of a cross‐talk between melanocytes and keratinocytes, where melanocytes are pigment‐producers and keratinocytes are pigment‐recipients.2

Melanosome biogenesis is initiated at the perinuclear region and undergoes a process of organelle maturation, whereas fully mature melanosomes tend to accumulate at the tips of melanocyte dendrites.3 When fully mature and located at the tips of melanocyte dendrites, melanin is transferred to keratinocytes. The actual mechanism of melanin transfer remains controversial. Recently, we proposed that the predominant model of melanin transfer is coupled exo/endocytosis where the melanin core or “melanocore” is exocytosed by melanocytes and then internalized by keratinocytes.4

Regardless of the transfer mechanism, melanin is processed within keratinocytes and ends up forming a well‐described supra‐nuclear cap, polarized to the sun‐exposed side.5 Surprisingly, little is known about how melanin enters keratinocytes, how it is processed and accumulates in the supra‐nuclear region of keratinocytes and how it is degraded.6, 7 One of the few insights into this process came with the identification of the G‐protein‐coupled receptor, Protease‐activated receptor 2 (PAR‐2) as playing an essential role in melanin internalization.8 Following PAR‐2 stimulation, pigment uptake is enhanced and upon PAR‐2 depletion, melanin content is decreased in keratinocytes.8 More recently, we and others found that upon internalization, melanocores colocalize with endocytic markers, namely early endosome antigen 1 (EEA1) and Rab5, as well as with the late endocytic markers lysosomal‐associated membrane protein 2 (LAMP2) and CD63.9, 10 Furthermore, the vesicles where melanin is stored within keratinocytes are mildly acidic and mildly degradative, suggesting that melanin resides in maturation‐arrested compartments that persist in these cells.

UV exposure of keratinocytes leads to the accumulation of cyclobutane pyrimidine dimers (CPD), which are key UV‐induced mediators of DNA lesions in human skin and a major risk‐factor for the development of skin cancers.11, 12 Another consequence of UV exposure is the production of reactive oxygen species within melanosomes, which also leads to accumulation of CPDs.13 Therefore, melanin is potentially toxic, and cells have developed mechanisms for degrading it. Indeed, autophagy has been proposed to regulate melanin content within keratinocytes, as melanin levels in human skin samples are reduced by inducers of autophagy and enhanced by its inhibitors.14 Moreover, autophagy can be enhanced by UV radiation15, 16 and two recent studies found that UV exposure increases the expression levels of the lysosomal protease Cathepsin V, leading to increased melanin degradation in keratinocytes.17, 18 Together, these findings suggest that damage caused by UV radiation accumulates with time, triggering melanin degradation.

Here, we propose that melanocores behave as a pathogen within its host cell, the keratinocyte (Figure 1). In fact, the processing of pigment by keratinocytes shares many features with survival strategies utilized by intracellular pathogens inside host cells. Several pathogens, upon entering host cells, survive and multiply in a membrane‐bound compartment or phagosome. Because the default survival mode of the cell is to fuse this new organelle with acidic lysosomes for degradation and pathogen elimination, most intracellular pathogens have evolved strategies to avoid or control this fusion. We suggest that a similar process is at play with melanin within keratinocytes. Indeed, some pathogens, like Mycobacterium tuberculosis19 induce phagosome maturation arrest, inhabiting inside a membrane‐bound compartment containing early endosome markers, thus avoiding fusion with degradative lysosomes. Others, like Salmonella, reside in an acidified late‐endosome compartment that is unable to fuse with degradative lysosomes.20 This is mediated by SifA, a Salmonella‐secreted protein that uncouples Rab7 from its effector Rab‐interacting lysosomal protein (RILP).21, 22

Figure 1

Schematic representation of the similarities between melanin processing within keratinocytes and host‐pathogen interaction between hepatocytes and Plasmodium during liver stage infection. Upon internalization by keratinocytes in a PAR‐2‐dependent‐manner, melanocores are processed and acquire first early endocytic markers (EEA1 and Rab5) and then late endocytic markers (LAMP1/2 and CD63), residing in a compartment that is not highly acidic or highly degradative and polarising to the supra‐nuclear region, where melanin protects nuclear DNA from UV‐radiation. We speculate that at this stage, compartments containing harmful UV‐modified melanin acquire autophagic markers like LC3‐II to target it for degradation. In hepatocytes, upon Plasmodium internalization the parasite subverts the late endocytic pathway of the host cell to grow in a hybrid compartment that is positive for late endocytic (LAMP1/2, CD63 and Rab7) and autophagic markers (LC3) but is not able to become highly acidic, allowing the parasite to grow and fulfil its life cycle during the liver stage of infection

A particularly interesting example of a pathogen that is able to avoid fusion with degradative lysosomes is the Plasmodium apicomplexan parasite. Over the past decade, a body of work has shown that the mammalian liver stage of Plasmodium, the parasite that causes malaria, resides in a membrane‐bound compartment, termed parasitophorous vacuole (PV),23 whose membrane serves as the interface between the parasite and the host. Plasmodium berghei parasites, which infect mice and are one of the most commonly used models of the disease, manipulate their environment and avoid fusion of the PV with host early endosomes; yet, the PV is able to fuse with late endosomes and lysosomes.24 While these late endosomes, which contain the markers CD63, Rab7 and LAMP1 have a mildly acidic lumen and fuse with the PV membrane (PVM), the space between the parasite membrane and the PVM does not seem to acidify and maintains a neutral pH throughout infection.24 To date, the molecular mechanism involved in this selective fusion remains elusive. Importantly, fusion of the PV with late endosomes appears to be essential for parasite growth. Indeed, when host endosome acidification is impaired, parasite growth is diminished.24 Moreover, parasite growth is impaired when autophagy is inhibited.25, 26 Together, these studies suggest that the host late endocytic/autophagic pathways are subverted by Plasmodium parasites as a source of nutrients for growth and development.

While most parasites can escape autophagy‐dependent clearance, the mechanisms involved are only now starting to be unravelled. For example, host LC3 is incorporated into the Plasmodium PVM early on during hepatocyte infection.27 LC3 incorporation and inhibition of autophagic degradation is mediated by the parasite protein, Up‐regulated in Infective Sporozoites gene 3 (UIS3). UIS3 interacts directly with host LC3 and competes for the binding of its downstream interacting proteins, effectively blocking lysosomal degradation.28 Conversely, recruitment of late endosomes and cholesterol but not lysosomes to the PVM is inhibited in uis4 ⁻ parasites,29 via an independent molecular mechanism. There are likely many more yet unknown parasite‐secreted proteins, which can modulate different aspects of the parasite‐host interaction.

A large proportion of Plasmodium parasites survive and grow inside hepatocytes but many cannot mature to later stages of hepatocyte infection.24, 25 While the events that lead to this elimination are unknown, it is tempting to speculate that the parasites that have not been able to manipulate their intracellular environment effectively, lose the fight for survival with the host cell. Indeed, very late stage parasites eventually lose autophagy markers from their PVM, which is a required step, since those that do not lose these markers show impaired growth and do not complete liver stage development.25, 30

To fulfil one important physiologic function, namely photoprotection, melanocores need to avoid rapid destruction by the keratinocyte endocytic and autophagic pathways, just as Plasmodium parasites need to do in order to survive. Although both melanocores and Plasmodium parasites are surrounded by CD63‐ and LAMP‐positive late endosomes, they are able to avoid fusion with degradative compartments and resist in neutral pH or only mildly acidic compartments.

One fundamental difference between these two cases is that melanocores are not living organisms like microbes, which produce and secrete proteins that mimic host cell proteins and induce favourable cell responses, as discussed above. Therefore, a critical question yet to be resolved is: what are the molecular mechanisms responsible for melanocore maturation arrest? We propose that melanocores inhibit autophagy within keratinocytes and thus control their own degradation rate. Indeed, in recent years, several groups have reported that PAR‐2 activation leads to autophagy inhibition through the activation of RAC1 and mTOR pathways.31-33 Therefore, it is tempting to speculate that through the activation of PAR‐2, melanocores signal to arrest the maturation of the compartment where they reside, so it can persist within the cell and exert its function. Importantly, Murase et al. described a correlation between levels of autophagic flux and skin colour associated with ethnicity, raising the possibility that autophagy regulates skin pigmentation and thus photoprotection.14 But what are the triggers for melanocore degradation upon UV‐stimulation? Could it be the accumulation of CPDs? Indeed, CPD formation could lead to a shift in the equilibrium between the protective and toxic effect of melanin, consequently targeting it for degradation to prevent the accumulation of UV signature mutations.

Finally, is this a unique example restricted to the epidermis or are we just uncovering the first example of a process that is more widely used in cell physiology? The skin pigmentation system may be ideal to study the molecular mechanisms regulating avoidance of lysosomal destruction and therefore should yield key answers and likely some surprises to come.

The Editorial Process File is available in the online version of this article.

ACKNOWLEDGMENTS

This work was supported by Fundação para a Ciência e a Tecnologia (FCT, I.P.), Portugal, through grant PTDC/BIA‐CEL/29765/2017, PhD studentship PD/BD/114118/2015 to HM and the FCT Investigator Program IF/00501/2014/CP1252/CT0001 to DCB, as well as iNOVA4Health ‐ UID/Multi/04462/2013, a program financially supported by FCT I.P. / Ministério da Educação e Ciência, through national funds and co‐funded by FEDER under the PT2020 Partnership Agreement.