So, studies on gray and melanoma. This one (the original study that found the gray allele):
- Pielberg, Gerli Rosengren; Anna Golovko; Elisabeth Sundström; Ino Curik; Johan Lennartsson; Monika H Seltenhammer; Thomas Druml; Matthew Binns; Carolyn Fitzsimmons; Gabriella Lindgren; Kaj Sandberg; Roswitha Baumung; Monika Vetterlein; Sara Strömberg; Manfred Grabherr; Claire Wade; Kerstin Lindblad-Toh; Fredrik Pontén; Carl-Henrik Heldin; Johann Sölkner; Leif Andersson (2008). "A cis-acting regulatory mutation causes premature hair graying and susceptibility to melanoma in the horse". Nature Genetics. 40 (8): 1004–1009. doi:10.1038/ng.185. PMID 18641652.
finds that gray homozygotes are more likely to have melanomas than horses heterozygous for gray. They also find that bay/black at agouti has a weak but highly significant effect, with black at agouti having more risk of melanoma. They speculate that the increased melanoma risk comes from increased MC1R signalling and would like to have tested if chestnut made a difference as well, but not enough of the lipizzaners carried chestnut. When they say 70-80% of gray horses over 15 get melanomas, they cite "Sutton, R.H. & Coleman, G.T. Melanoma and the Graying Horse (RIRDC ResearchPaper Series) 1–34 (Barton, Australia, 1997)." and "Fleury, C.et al.The study of cutaneous melanomas in Camargue-type gray-skinnedhorses (2): epidemiological survey.Pigment Cell Res.13, 47–51 (2000)." As for the mechanism, they think more expression of STX17 and the neighboring gene NR4A3 causes both the gray color and the tumors by causing increased melanocyte proliferation. Probably NR4A3, because that one is involved in regulation of the cell cycle and is linked to carcinogenesis, plus the gray melanomas were expressing more of the cyclin CCND2, which is a target gene for NR4A3. Cyclins regulate the cell cycle and tumors often make more of them. And they cite "Smith, A.G.et al.Melanocortin-1 receptor signaling markedly induces the expressionof the NR4A nuclear receptor subgroup in melanocytic cells.J. Biol. Chem.283,12564–12570 (2008)." for saying increased MC1R signalling causes increased expression of NR4A3, so I guess there's that link. So they propose that increased proliferation of melanocytes in the skin causes melanomas, and in the hair causes depletion of stem cells which leads to it turning white.
This one:
- Elisabeth Sundström; Freyja Imsland; Sofia Mikko; Claire Wade; Snaevar Sigurdsson; Gerli Rosengren Pielberg; Anna Golovko; Ino Curik; Monika H Seltenhammer; Johann Sölkner; Kerstin Lindblad-Toh; Leif Andersson (2 Aug 2012). "Copy number expansion of the STX17 duplication in melanoma tissue from Grey horses". BMC Genomics. doi:10.1186/1471-2164-13-365.
{{cite journal}}
: CS1 maint: unflagged free DOI (link)
by mostly the same authors finds that while gray has a section of DNA that's repeated twice, in melanoma tumors that area is often repeated even more times with the more aggressive tumors having more repeats. So a non-gray horse would have two copies of this DNA, one on each of the pair of chromosomes, a heterozygous gray horse would have 3, and a homozyous gray horse would have 4, but some of the tumors had 5-8 repeats. Unrelated to melanomas they also mention that some connemaras go gray extremely slowly, with a picture of a 14 year old connemara who is still pretty dark, and they checked and found that they have the same gray mutation as any other gray horse.
This one:
- Teixeira; Rendahl; Anderson; Mickelson; Sigler; Buchanan (2013). "Coat color genotypes and risk and severity of melanoma in gray quarter horses". Journal of Veterinary Internal Medicine. doi:10.1111/jvim.12133. PMID 23875712.
tried to replicate the connection between ASIP and melanoma risk in gray horses that the first study found, and also look for a link to MC1R. They used 335 gray quarter horses while the first study had 694 gray Lipizzaners, and basically nothing came out as statistically significant. The homozygous grays were about twice as likely to have melanomas as heterozygous grays, but there were not enough of them for the result to be significant. Then for whether black/bay/chestnut base color and the underlying extension and agouti genes had any effect, they tried looking at it a bunch of different ways but didn't get anything statistically significant, though they did find the relation between agouti and melanomas trended in the expected direction. They think agouti had less effect than in the first study because the chestnut extension allele was much more common. At they end they compare melanoma prevalence and severity across different breeds and say that in the quarter horses they studied, 16% of all the horses had a melanoma, or 52% of only the horses at least 15 years old. This is compared to studies on other breeds which found a prevalence in 31.4% in the Camargue (68% in Camargues at least 15 years old), 50% in the Lipizzaner (75% in Lipizzaners at least 15 years old), and 89.6% in the Pura Raza Española and crosses (100% in PREs at least 10, not 15 but 10, years old). Here's the studies they cite for the other breeds: camargue, PRE, camargue again, lipizzaner.
I'll see if I can find any other studies on gray and equine melanoma later but this is enough for today. Iamnotabunny (talk) 13:14, 20 April 2020 (UTC)
I’m rereading the Pielberg studies. As for the gray and melanoma studies, I already cited one on a Thoroughbred, if you note the diffs to the changes I added. I think we probably should just link to the equine melanoma article to avoid content forking, and if needed, add more refs there explaining how much variation is in the studies. (It’s worth drilling down on, actually, but there, not here.) Here, we could summarize something like prevalence of melanomas in older horses varies from estimates of x% (cite) to x% (cite) with outlier studies showing 0% and 100%.(cite) I think it’s very much worth noting that very few studies had enough horses in them for results to be statistically significant. The Agouti (A) thing sounds very unproven—in particular, while it creates the bay color on a black coat (E), it can occur in chestnuts (ee), masked because of the absence of E. Montanabw(talk) 15:35, 20 April 2020 (UTC)
Oh, and for the “flea bites,” let’s say “pigmented speckles.” That is more accurate. As in “fleabitten gray horses are recognized by the development of pigmented speckles on the body.” We need some sourcing, but on the grays I’ve known over the years, I’d say there’s definitely a link to heterozygosity, and what’s more, they increase as the horse ages...I’ve seen horses that are pure white at age 8 become totally freckled by age 15. Montanabw(talk) 15:35, 20 April 2020 (UTC)
- I've gone ahead and made the fleabitten edits now. The source is again the Pielberg 2008 study that found the gray allele. Iamnotabunny (talk) 16:11, 20 April 2020 (UTC)
- As for how much to put in this article and how much in the equine melanoma one, I think it would be worth having a section here explaining the connection to melanomas in basic terms, a note about how equine melanomas caused by gray are not usually so deadly as human melanomas, and possibly info about how common melanomas are in the different breeds. Details of the molecular biology not related to the coat color should probably go in the equine melanoma article. Iamnotabunny (talk) 16:26, 20 April 2020 (UTC)
- Looking through the article, it hasn’t had a big update in a decade, and it shows. I just did a lot of wikignoming to clean up formatting, fix images, clarify phrasing and so on. I made the melanoma bit into a new subsection, so if you want to build on it some, in terms of balance, I’d say we could add one more paragraph, but I’d prefer not to overwhelm the article with it, and I really don’t get into the weeds on prevalence within breeds here, because I think the studies are preliminary and it’s a minefield. I think we CAN talk about the studies in the melanoma article, because there we can get into the limits of the studies (i.e. “the only study on foo breed said x, based upon a sample size of y, and etc.” The two studies on breed foobar produced consistent or inconsistent results because one studied x and the other looked at y.”) I compare this to the articles on soring and the Tennessee Walking Horse, where one article went into detail, summarized with wikilinks in the other. Montanabw(talk) 17:24, 20 April 2020 (UTC)
- Wow, that was a lot of changes! Lots of small improvements. But, you must have a narrower browser than me, because the images show up all in the wrong places now. Do you think we could make that top-left box shorter? That might help. Iamnotabunny (talk) 21:12, 20 April 2020 (UTC)
- The infobox? It’s on the right on my screen... it’s also standardized to have a summary of info. So we should keep it as is. But, I realized we didn’t really need the photo of the Andalusian, so I tossed it and tweaked the caption on one of the other images. That may help a little. I realize things got strung out when I moved all the images to default sizes and positions, so they might change with different browsers. If we ever wanted to prep this article for WP:GA or something, though, they’d want us to do it to conform to the MOS. There’s exceptions, of course, but the formatting gods of WP really hate “sandwiched” text. Maybe the age stages or the “confused with gray” sections (or both) could be made into a table, like we did for the spotting patterns at Appaloosa and Leopard_complex#Patterns, but I don’t really have the time and energy right now to do that kind of fussy formatting work... meh. But if you want to mess with tables, that’s a possible solution. Montanabw(talk) 01:18, 22 April 2020 (UTC)
- Sorry, left and right are too similar. I've put the start of a table below.Iamnotabunny (talk) 14:23, 22 April 2020 (UTC)