Gil, mammals share the same dozens of coat-color genes. Synthesizing and distributing melanins within and between cells is a complex process, with many components performing the same functions in the nervous system, which makes perfect evolutionary sense since pigment cells are are derived from the neural crest.
IOW, the fact that homozygous deletions of the dilute (now Myo5a) gene cause neurological lethality in mice and people is perfectly consistent with physiology and evolutionary biology.
I’ve been involved in these kinds of screens three different times. I’ve used two different kinds of yeast cells (S. cerevisiae and S. Pombe) in screens for cell cycle kinases. And I assisted a colleague using human L-cells to screen for enzymes involved in phospholipid synthesis.
There is also the fact that polar bear skin is a deep black, presumably due to high levels of melanin production in the skin. This allows the polar bear to soak up what little heat there is. The mechanisms for melanin production are not only present, but selected for. All that would be needed is for melanin production to be turned on in the cells that make hair.
It’s hard to see how that would be useful in the Arctic summer. Others have hypothesized that it might be a way for the bear to navigate by sun position in low winter visibility.
Or just as likely, melanin is still being made, but just not transferred to the hair; for dilute, the coat-color dilution is not associated with any reduction in melanin synthesis, just transfer in larger blobs. Last time I checked, the precise mechanism of the transfer was still unknown: https://onlinelibrary.wiley.com/doi/full/10.1111/j.1600-0854.2006.00425.x
In the larger scheme, we should all be careful when drawing conclusions about the potential adaptive properties of any feature. Otherwise, Gould’s spirit will haunt us until we say “Spandrels!” three times.