|Preservation of melanosomes|
One objection to our interpretation of melanosomes in full ultrastructural detail in 125-million-year-old fossil feathers of birds and dinosaurs from the Early Cretaceous Jehol fossil beds of NE China, could be that these organelles are unlikely to survive the fossilization process. However, as argued elsewhere (Vinther et al. 2008, 2010), there is extensive evidence that melanosomes are highly resistant to chemical and physical degradation and have higher resistance to decay than the keratin substrate of feathers and hairs in a variety of physical environments.
The toughness of melanin and melanosomesMelanin provides not only colour, but also structural strength. It strengthens structures by cross-linking of proteins (Riley 1997). Melanins supply mechanical strength to seed pods in plants and insect cuticles. Further, black hairs and are generally thicker and tougher than white or grey hair and feathers (Liu et al. 2003a, b). Extensive studies of human hair show that bleaching by sunlight or chemical means, with oxidation of melanins, substantially weakens the hair, even though melanins make up only 2% of the hair (Hoting & Zimmermann 1997).
Similar phenomena are seen also in birds. For example, in birds that are primarily white, such as geese, gulls, pelicans, and cranes, the wing tips often carry a flash of black colour, a terminal set of feathers containing intense packing of eumelanosomes, in order to provide strength and protection (McGraw 2006). This idea that melanin produced abrasion resistance in feathers was proposed first by Burtt (1986), and has since been tested experimentally by several authors (McGraw 2006).
It is well documented (Goldstein et al. 2004; Gunderson et al. 2008) that melanized feathers are more resistant to bacterial degradation than non-melanized feathers. It has also been shown (Bonser 1995) that melanized feathers are more abrasion resistant. Liu and Simon (2003a, b) demonstrated that eumelanin is resistant to chemical degradation.
These observations are supported by extensive practical work in pigment materials research. For example, both eumelanosomes and phaeomelanosomes are routinely extracted from inside keratinous substrates such as hair using various methods of acid/ base extraction, indicating their strong resistance to chemical degradation; alternative methods using enzymatic extraction retain not only the physical structure, but alter the chemistry less (Liu et al. 2003b, and references therein).
Recent studies on the taphonomy of hair samples in archaeological contexts (Wilson et al. 2007) also show that melanin and melanosomes are resistant to decay, even more so than keratin: "Indeed, in the most severe cases of microbial destruction to the hair shaft, the complete loss of keratinaceous material was associated with concomitant survival of melanin pigment granules."
Melanosomes, not bacteriaPalaeontologists interested in taphonomy, the preservation of fossil tissues, have had to learn, and re-learn their craft. At one time, unusual soft tissues, such as muscles, were thought to survive, very rarely, in some version of their original form. Then, close study of specimens under the scanning electron microscope (SEM) showed that these tissues, which normally rot away soon after death, were commonly replicated faithfully by bacteria (Wuttke 1983). The bacteria had infested the carcass soon after death, feeding on soft tissues such as muscle, and they were then themselves killed by the development of an anoxic (oxygen-free) environment under water and replicated in calcium phosphate or other chemicals derived from the rotting carcass and the surrounding environment.
Left: Melanosomes in an isolated pennaceous feather (IVPP V15388B). a, Optical photograph; position of area analysed by SEM indicated by arrow. b, c, SEM images (at lower and higher magnification, respectively) of eumelanosomes preserved as moulds inside small areas that are separated from each other by anastomosing ridges of degraded feather (at arrows in c). Scale bars: a, 5 mm; b, 20 µm; c, 5 µm.
Surely then the best interpretation of the masses of sausage-shaped and spherical bodies in our fossil feathers is that they are the remnants of a film of keratinophilic bacteria that coated the surface of the feather during early diagenesis? Both melanosomes and bacteria are generally similar in size (one micrometre or less) and shape (spherical, oblate or elongate), so it is essential to distinguish the two. There are three pieces of evidence that confirm that the microstructures in the Jehol fossils are melanosomes, not replacement bacteria.
1. Melanosomes embedded in keratin structure. First, the bodies occur embedded inside the feathers, and in those feather parts that exhibit melanosomes in modern birds (Durrer 1986; McGraw 2006). In extant birds, melanosomes in the feather barbules are arranged in complex arrays (Durrer 1986). The typical configuration is one or more layers of regularly oriented melanosomes suspended in a β-keratin matrix below a superficial layer of β-keratin; melanosomes can also occur, usually arranged less regularly, medial to such layers (Durrer 1986). Preservation (presumably as primarily an organic remain) of the degraded keratinous matrix occurs locally in some of the Jehol feathers, most obviously where the fossil bodies are exposed as moulds (Benton et al. 2008); the fossil bodies are, like melanosomes, clearly embedded within this matrix, and are not a superficial coating. The integumentary filaments also exhibit this feature. These phenomena occur in the feathers of birds and dinosaurs from the Jehol Group and other localities.
2. Melanosomes occur only in dark bands. Second, it has been shown (Vinther et al. 2008) that eumelanosomes occur only in the dark bands of banded feathers, and not in the light bands: a fossilized biofilm of keratinophilic bacteria would be likely to occur throughout a uniformly preserved structure, and not stop suddenly along an apparent feather stripe. Notably, those parts of a feather that lack melanosomes, the calamus and proximal part of the rachis, are repeatedly absent in Jehol materials (for example, in the figure above) unless preserved in calcium phosphate (Benton et al. 2008). There is no reason to suppose that a film of keratinophilic bacteria would have developed elsewhere over the surface of the feather, but not on these parts, nor could their absence imply that these portions were buried in the skin and so escaped bacterial replacement because most of the rachis would have been exposed.
3. Unusual packing ultrastructures. The third line of evidence for fossil melanosomes comes from Vinther et al. (2008), who showed packing and layering of melanosomes in fossil feathers, identical to ultrastructures seen in modern feathers and in the Jehol feathers, but incompatible with a bacterial origin.