The New Guppy Genetics (Again)

How difficult is it to let go of a cherished illusion?

Let's take the most basic of ideas, the patterns on guppies. Almost all guppy breeders have the idea of guppy patterns totally wrong. And it is not just a matter of minor quibbling. The whole paradigm they unconsciously work with is wrong.

The paradigm is that there are pattern genes, like little blueprints stored in genes that deck out the guppy body with a tuxedo jacket pattern or a snakeskin print. In fact the language we use to describe guppy patterns reinforces this idea: the snakeskin, the Panda, the half-black pattern, called "tuxedo" among Asians. People talk about the snakeskin gene as if the pattern of black and metal cells was stored as a little blueprint in a single snakeskin gene. Or they think there is a gene that coats the guppy peduncle in black, leaving the fins and front of the body wild type grey, (the half-black gene).

Reading much of the research on guppies you would think that scientists also consider guppy "color genes" to be genetic blueprints. They reach back all the way to Winge and his 1927 paper (The location of eighteen genes in Lebistes reticulatus . J. Gen. (3):1-37.)  and its list of eighteen color genes. But scientists are speaking in the abstract, as I do. Modern cell biology is based on a paradigm that Winge was not privy to. A Winge pattern of red spots and blue iridescence cannot possibly be due to a single gene, something Winge suspected but did not have the theoretical basis to deny or confirm.

In fact the guppy hobbyists who howl and close their ears to the idea that there are no guppy pattern genes should give themselves a good dose of modern cell theory, particularly color cell biology down to the molecular level. It will be a painful withdrawal from their cherished paradigm. But it will be worth it.

Consider this: the half-black guppy is in fact not a half-black guppy. It is a full black guppy that cannot express black in the front of the body. This is an observable fact. You can test this hypothesis yourself, as I did. I acquired a half-black green strain from Luke Roebuck that was particularly black. The females were a very dark color.

Half-Black Green guppy. Notice how black the female is.

The front of the male is grey and although the fins of the male were dark, they are largely green in color. Isn't it curious that the female is full body black while the male is only half-black? What this suggests is that black color is largely suppressed in the front of the male. I decided to test this theory by outcrossing a female from this strain to a Blue Moscow. The result was a full black Moscow.

Moscow with the half-black gene.

In fact there are two ways of creating a Black Moscow. One is to use a gene called the Midnight black gene and the other is to cross a Moscow with a half-black guppy. The half-black Moscow is not as deep black as the Midnight Moscow as you can see in the above picture, but it is a Black Moscow. When I related this fact to Luke, he said the "Brazilians created their Black Moscow using this same strain." So my own experiment received some corroboration at that point.

What set me on the path to this shift of paradigm was a comment in a Penelope Nayudu paper from 1979. See Genetic Studies of Melanic Color Patterns and Atypical Sex Determination in the Guppy, Poecilia reticulata, Copeia 1979, 2 pp. 225-231, and Cytological Aspects and Differential Response to Melatonin of Melanophore Based Color Mutants in the Guppy, Poecilia reticulata, Copeia 1979, 2 pp. 232-242. Nayudu in Genetic Studies (p. 230) notes that the intensity of the pattern is affected by other genes, and speculates that these are probably autosomal genes. So the black in the half-black pattern is not due to the same gene as that confining the pattern to the half-body area (it is sex-linked). It is a separate gene. This is a blow against the paradigm of a single half-black pattern gene determining the color of the peduncle. The reference to this black gene is over three decades old. The Y-linked half-black gene (NiI or NiII) might not be black at all.

There is still the matter of the fins. In the Black Moscow seen above the fins are largely white, with some dark streaking. Most people would be unsurprised by this, thinking that the half-black guppy is one with a black peduncle and the fins can be a different color. But if you look at the first descriptions of the half-black guppy, the black extends into the caudal fin. Here is a picture the Russian scientist Kirpichnikov drew of the half-black pattern (nigrocaudatus is the scientific name) three decades ago.

Kirpichnikov's half-black guppy

For a long time I wondered what happened to the black on the caudal fin in the modern guppy. Was it bred out? It was not until I crossed a half-black Yellow guppy with a wild male that I discovered something amazing.

Here is the HB Yellow strain. Notice how yellow the fins are...there is no black.

The HB Yellow strain I used in the cross.

Now look at the F1 male from the cross.

F1 male who is the result of a wild guppy male cross to a half-black yellow female.

Compare this F1 male to the Kirpichnikov illustration. What you see is the pattern of the half-black gene in the caudal fin. It is almost a perfect replica.

There is also some black in the caudal fin. The wild male I used had no black on the body, except the wild type grey color.

The conclusion I draw from this is that the so-called half-black pattern is a "full body gene" in the sense that is has the potential to be expressed anywhere on the fins or the body. It's expression is suppressed in the fins in the case of the half-black yellow and is suppressed in the case of the half-black strains but not the Moscow. To be more precise, I would say that the black color of the half-black complex of genes has the potential to be expressed anywhere in the body.

The idea that there is a single gene determining a guppy pattern does not hold up to close observation. When I talk about patterns in guppies I usually mean what we see with our eyes, not what we think is happening on the cellular or molecular level. There may indeed be a single gene that causes the full body black color in guppies to be expressed as a half-body black pattern. I would speculate that the half-black gene may be a single gene or several genes that suppress black color in the front of the body. In the Moscow this suppression is not working, making the Moscow and half-black guppies similar but perhaps different in the presence of a single suppressor gene. In other words I see a kind of "negative image" of the half-black, not a gene or genes that colors the peduncle, but rather a gene or genes that prevents the black coloring of the front of the body. In the case of strains like the half-black yellow, there is additionally genes that suppress the expression of black in the fins.

This is a more complicated account of the half-black pattern than most people are willing to contemplate so I do not expect it to actually spread like a virus through the guppy community. In fact I have halted work on what will be the most ambitious (read controversial) contribution to the discussion on guppy color and pattern, "The Guppy Color Manual." I want to gather more evidence from my own crossing experiments and corroborate it with current scientific theory. Plus I might spend some more time trying to transcribe into lay terms the scientific research.

But I am aware of the enormous task this entails. To fully grasp the new guppy color genetics you have to subject yourself to a paradigm shift away from the idea of the genetic blueprint to a much more dynamic and complex view. This view often turns questions inside out. Instead of asking "why is there a half-black pattern," you ask "why are black color cells not expressed in the front of the body?" The answer to these questions are found on the molecular level using modern genetic toolkits.

The real question is "what is happening to the color cells?" The unit of inquiry is not the pattern. Color cells or their pigment organelles are the proper object of inquiry. When we look at guppy color we are looking at color cells. And color cells, like all the cells in the body, are controlled and determined by internal cellular processes (proteins in the cell), as well as external signals (signal transduction). We ask: "what signals is the individual color cell receiving or not receiving because of a failure in the normal processes, what genes are being activated in the color cell or elsewhere that affects how a color cell develops and differentiates into the different types?

The classic example is countershading in the guppy. Countershading is the tendency of guppies to have dark dorsals and silver bellies, like the F1 male in the picture above. Scientists have actually discovered the protein (agouti) that suppresses the expression of black color cells in animals. An analogous protein is operative in the guppy suppressing black and other colors in the belly. Ed Chiasson developed his full red guppy by finding and selectively breeding males with red bellies instead of silver bellies. He stumbled upon a defective form of the protein that normally suppresses red in the belly.

Is this defective suppressor gene a "red color gene?" No, because it actually acts to suppress red color, just as black is suppressed in the fins of the half-black yellow. Is there a full body red gene? There is a gene that causes a color cell to develop as a red color cell (i.e. one containing red pigment organelles) rather than a black, yellow, white or metal color cell. But obviously it can be negated by a suppressor gene. See the shift in thinking? Instead of looking for a "full body" gene, you look at what is happening at the individual color cell. Where is it located on the body is an important question to ask. What other mutations does the guppy have that might affect this individual color cell. What does examining the color under a microscope tell us?

In the case of the snakeskin pattern, examination under a microscope tells us that the snakeskin pattern is an arrangement of color cells where the black color cells exist clumped into islands in a sea of iridophores (metal color cells). So the kinds of questions you want to ask about the snakeskin pattern are: why are the black color cells clumped into islands? Why are they not intermingled with the iridophores as is the case of the wild type guppy? Why does a metal mutation (Schimmelpfennig) alter the snakeskin pattern, either suppressing it, or forcing a lace snakeskin pattern into a much coarser snakeskin pattern? (Hint probably because the Schimmelpfennig mutation affects iridophores.)

My book "Theory and Practice of Guppy Breeding" attempts to act as a primer for those brave enough to embrace new paradigms of how guppy color is expressed. I say "brave enough" because championing a new paradigm does have its risks. On one forum I witnessed a comical pseudo debate about my prowess as a breeder of guppies compared to a well-known guru. The "quality" of the guru's guppy was used as evidence that he was the superior breeder. The fact is breeding guppies to an artificial standard of guppy beauty is little more than "paint by number" for me. Deliberately creating your own guppy design is much more interesting, especially when it involves using genetic tools and genetic knowledge. For me the true art of guppy breeding involves the same artistic and intellectual tools as creating a great painting. This is why I am such a great fan of Japanese breeding and show standards. In Japanese guppy show standards the creativity and genetic knowledge of the breeder is rewarded, not just their ability to conform a guppy to an artificial standard. I recognize that there is still a significant population of guppy breeders in the hobby who derive satisfaction from breeding to fixed and artificial standards and only a relatively small number of breeders are interested in really opening their eyes to the mysteries of guppy color. I respect that. But don't go telling me that your guru is a better breeder. That is just nonsense.