The capacity for internal colour change is related to body transparency in fishes

Abstract

Dear Sir, Chromatophores are large, stellate and spectacular pigment bearing cells, typically located in the skin, that generate body colouration. In many animals, melanocytes ⁄melanophores, the melanin containing chromatophores, are also present in various tissues inside the body, for instance in the ear, brain, abdominal cavity, around internal organs and skeleton. The presence of such internal pigmentation in puzzling as it is hidden from sight. While there is an enormous amount of studies and data on skin chromatophores, from the fine details of the motile machinery to animal behaviour (Aspengren et al., 2009), internal melanocytes have historically been largely ignored, until recently (Aoki et al., 2009; Brito and Kos, 2008; Randhawa et al., 2009; Yajima and Larue, 2008). Remarkably little is still known about their possible functions, and this uncertainty is problematic for the more general question of the role(s) of melanin in itself (Aspengren et al., 2009; Boissy and Hornyak, 2006; Braasch et al., 2009; Hill, 2000; Ito, 2009). While internal melanocytes are prevalent, internal erythrophores and xanthophores appear more uncommon. In fish, however, such cells can be found interspersed with melanocytes and reflective chromatophores in the highly pigmented peritoneum (the endothelium that covers the abdominal cavity) (Nilsson Skold et al., 2008). In juveniles, as well as in adults of species with relatively transparent bodies, internal chromatophores may actually contribute to the overall body colouration, as shown in two-spotted goby females, where abdominal trunk biopsies were analysed (Nilsson Skold et al., 2008). In fishes, skin patterns can be rapidly modified by aggregation or dispersion of the pigment-containing organelles inside the chromatophores present in the skin (i.e. physiological colour change) for background adaptation or signalling displays (Aspengren et al., 2009; Fujii and Oshima, 1994). In comparison, colour change in internal chromatophores has been largely ignored as it has been generally considered that they are not responsive (Boissy and Hornyak, 2006). However, melanocytes in the peritoneum and around the skeleton of the ice goby, Leucopsarion petersii, do indeed adapt to the background by pigment translocations in vivo and in biopsies, thus providing the first evidence for internal colour change (Goda and Fujii, 1996). Recent work on biopsies from the two-spotted goby showed that also internal erythrophores and xanthophores can be responsive (Nilsson Skold et al., 2008). As these examples come from relatively transparent species, it is possible that this phenomenon is more common than earlier believed, especially in species with some degree of body transparency. In order to test if a capacity for internal colour change was related to the degree of body transparency, and to reveal a possible function of these cells, we analysed the regulatory capability of peritoneal melanocytes in eight different teleost species, representing five different orders within the large super order Acanthopterygii and in one member of the super order Clupeomorpha, all with different degrees of body transparency. We used epinephrine and melatonin as potential pigment aggregating agents (see Appendix S1 for methodology). A positive relationship between body transparency and rate of internal colour change would suggest a special adaptive role for internal melanocytes in transparent fish species, and thus constitute a novel function for internal pigments. Our results showed that peritoneal melanocytes were present in all investigated fish species. Especially high densities were found in the gobies and in pipefish, plaice and herring (Figure 1A, B). Internal erythrophores and xanthophores were also observed in the gobies, pipefish and in plaice, but not in the other species. A capacity to regulate peritoneal melanocytes by melatonin and epinephrine was detected in six out of the eight species (Figure 1A, B and Table S1). In all species tested, peritoneal melanocytes had dispersed pigment from the start of the experiment. The epidermal melanocytes of sand goby, black goby and wrasse responded within 30 min, whereas plaice and pipefish only responded appreciably after 2 h. In herring, both controls and treated peritoneum biopsies showed a response after 30 min with an only moderate further