Wednesday, 16 April 2014

Breeding back and dedomestication, Pt. I: How domestication works

I am actually pretty unhappy with the term “breeding-back”, but it is rather widespread and my tentative alternative I transferred from German “Abbildzüchtung”, “effigy breeding”, implies that the resemblance of the resulting animals with their desired wild type is merely of optic, which is not quite the case. In this dedomestication series, I’ll try to explain the necessity of dedomestication in “recreating” aurochs-like or wild horse-like animals, how it works, the connections between genotype, physiology and phenotype and which implications it has for “breeding-back” and dedomestication, and give some feral examples of post-domestic animal populations. Of course this series is mainly aurochs-centred again, because cattle obviously are more domesticated than horses. I am an amateur on genetics, so I cannot guarantee that everything I write is correct here, but I try to avoid errors.
By the way, forgive me for my previous misuse of the term “phenotype” to describe optic and morphologic traits exclusively. The phenotype is everything that is the result of gene expression, therefore also behaviour and physiology.

To better understand how dedomestication works and what is its role in creating new and authentic post-domestic wild types for the aurochs and western wild horses, it is necessary look at the genetic changes caused by domestication, and how these are connected to the novel traits we see in domesticated animals. This is what Part I is about.
Basically, domestication works by 1) snatching out a group of individuals from a species and thereby creating a new subpopulation and then 2) selecting for certain desired traits. The new subpopulation immediately starts to become different from their wild relatives at least on molecular level because of genetic drift caused by the first step. Subsequent selection and also new mutations enforce this process. Domestic animals become shaped after the purposes of the breeders, affecting the behaviour and hormonal system, immune system, diet and appearance. A probably very important factor in domestication is neoteny/paedomorphy, the retention of juvenile traits into adultery. “Relaxed selection” probably causes atrophy of certain morphologic or behavioural traits [3].
Everybody knows that domestic animals are very diverse in appearance. But what is striking is how many features they actually have in common [1]. Deviant colour variants are found in all domestic animals, white spots in almost each species, even fish. All domestic mammals show paedomorphy to a certain degree, resulting in shortened and domed heads, perhaps an elongated trunk or shortened legs, as much as floppy ears and deviant horn shapes in bovines. Horns of domestic bovines tend to curl outwards compared to the wild type, you see that in many sheep and cattle species. A reduced relative brain size might be considered a general indication of domestication [2]. Since similar changes in different domestication events with different species are observed, it might be assumed that a similar basic mechanism comes into play when a wild animal is turned into a domestic form [1].

The so-called Farm fox experiment, running in Russia since 1959, is a scientific long-term study domesticating a population of Silver foxes, Vulpes vulpes, and studies behavioural, optic and genetic changes, providing a model for domestication. The goal is to resolve the relation of these three factors, particularly in domestication. Their results, the animals as well as their findings, are very interesting.  The domestic foxes have been selected exclusively for tame and friendly behaviour towards humans and not any morphological changes (there is a control group which has been selected for aggressive behaviour), yet clearly domestic features shared with other domesticated animals, dogs in particular, emerged that are absent or much rarer in the control group [1]. These include piebald and diluted colour patterns, elongated trunks and shorter legs, floppy ears, curly tails, under- and overbites and behavioural paedomorphisms like tail wagging, barking and licking approaching humans’ hands. [1] The foxes even show earlier maturity, which was not achieved by actively selecting on this particular trait by commercial fur breeders before [1]. The Farm fox experiment seems to confirm the assumption of its initiator Dmitry Belyaev that physiology, morphology and behaviour are interconnected and that the altering of genes controlling the endocrinology and neurochemicals which are responsible for behavioural genes causes a developmental cascade that also has an effect on other aspects of the functionality of the organism through a countless number of pleiotropic effects [1, 2].  Particularly the thyroid metabolism is considered a key to the developmental changes and paedomorphosis caused by domestication [2]. For example, hypothyroidic rats are smaller than usual rats and have floppy ears, two typical domestic traits [2]. The gene expression in the hippocampus of dogs is affected as well [2]. Probably the modifications in the genome causing some of these changes did not take place on the coding but on the non-coding regions which work upon the expression of genes [2]. The domestic foxes do indeed show neurochemical and hormonal changes, such as a lower productive activity of the adrenal glands [1]
Pleiotropy refers to genes that do not affect a single trait but have a number of functions. All genes of an organism interact with each other and even the genes for discrete, monofactorial traits have more than one function, see coat colours. A prime example is the KITLG locus that produces the ligand for tyrosine-kinase (coded by the KIT-locus), which has a function in germ cell, neural cell and blood cell development. Mutations on these two loci are either hyperpigmentation or leucisms [4], responsible for many of the spotted patterns we see in domestic animals [5].  KIT mutations are the cause of the typical white streak along the face or the “star” on the forehead of many domestic animals, and is also displayed by humans having similar mutations (“piebaldism”). White areas are the result of a disordered migration of the melanoblasts from the neural crest where they are. White-faced cattle like Hereford have a greater susceptibility to Cancer Eye or Bovine Squamous Cell Carcinoma [5], and depigmentation in general increases the risk of cancer because melanin is an important barrier for mutagenic UV radiation, which is evident in basically every animal. The Agouti-locus has a role in pigmentation (responsible for dilutions in many mammal species), but also in regulating lipid metabolism in adipocytes [6]. The Dun-locus, a dilution locus as well, has functions in the nervous system and metabolism. Mutations on this locus cause neuromuscular disorders which can result f.e. in arched tails (see dogs and pigs) because of myelin degeneration [7]. The phenylalanine metabolism might be disturbed as well [7]. Not all mutations on these loci are necessarily that disastrous of course, otherwise all domestic animals with deviant colours would be regularly born with serious disorders. There are certainly similar pleiotropic effects in other traits as well, but seemingly these are less well-studied.

Tameness includes a reduced fear/aggression response when handled by humans. Domesticated or semi-domesticated foxes with a fear response delayed in their onotgeny show a change in the plasma levels of corticosteroids, negatively correlated with the degree of tameness. The level of corticosteroids in the blood plasma of domestic foxes is only 25% of that of the control group. The serotonin system changed as well [1]. Therefore domestication results in a change in the timing of the postnatal development and hormonal mechanisms that are the base for social behaviour and reaction to stress [1]. I suspect that the same hormonal changes perhaps are responsible for the reduction of sexual dimorphism observed in domestic foxes and many other domestic animals. Another very interesting fact is that, as mentioned above, the farm foxes also showed earlier sexual maturity and reproduce more than once a year. Fur farmers tried to achieve that by selecting on this trait for century but they were not successful, but the selection on tameness was [1].

I think that the Farm fox experiment brought some important insights to how domestication evolved. Selection on tameness exclusively produced many of the common novel traits through pleiotropy and molecular cascades, so these basic features probably are not the result of coincidental mutations and/or purposeful selection. Considering that, it is likely that the early farmers of the Neolithic did not initially breed their livestock for desirable, advantageous or aesthetic traits. I think it is most likely that they just caught calves or foals, raised them in an enclosure and constantly got rid of those that were the most difficult to handle. The breeders probably didn’t know that selective breeding can change heritable traits, how could they? After some centuries, the isolated populations of tamed herbivores would have probably started to show the typical domestication features outlined above automatically.


[1] Trut, 1999: Early Canid Domestication: The Farm fox experiment.
[2] Dobney & Larson, 2005: Genetics and animal domestication: new windows on an elusive process.
[3] Koler-Matznick, 2002: Origin of the Dog revisited.


  1. It is a complex subject .The majority of the videos of the tame russian foxes does not show foxes with a lot of white.Belyaev did also developed tame rats in a similar project .But those rats still have a wild phenotype (as far as I know )
    Even today people are often fascinated when they see unusual looking animals ,like white or white spotted deer etc.I would assume that the first people that domesticate the animals we keep today where equally if not more interested individuals that deviated from the wild type even in more subtle ways. Animals that are of wild type are very homogeneous ,when managing a group of animals it is very useful to be able to recognize different individuals.
    Also I would assume that if you keep domesticated or semi domesticated animals in an area where where the species live wild as well,you don't have a positive view of the wild type due to failed attempts to ad non domesticated individuals (or crosses) in your domesticated group of animals.

    1. Of course it is a complex subject and I haven't got the impression that I oversimplified it here. Not all domestic foxes show a very derived appearance, but there is a tendency, and the most important aspect are the novel features that surfaced which resemble those of other domdesitc animals.

  2. No you did not .But maybe Lyudmila Trut (or the media )does focus to much on these morphological traits?
    "The novel traits are still fairly rare. Most of them show up in no more than a few animals per 100 to a few per 10,000. Some have been seen in commercial populations, though at levels at least a magnitude lower than we recorded in our domesticated foxes."

    I need to read the study again.

    1. I don't think they focus too much on morphology, I think it is not the quantity of the novel traits that is relevant, but the fact that they did appear. It is the goal of this program to resolve the connection between genetics, behaviour and phenotype in domestication, and the emergance of these novel traits, no matter in which quantity, is what told a lot. At least that's the way I see it.

  3. The piece i quoted is from : Trut, 1999: Early Canid Domestication: The Farm fox experiment.

  4. But the connection seams very weak .If theses "novel traits" had any significant connection with tame behavior .Would it not be reasonable to expect them to increase as the % of very tame foxes rouse to 70 % .

    1. Perhaps weak, but it's there, and it seems that we have a reasonable explanation for why the same kinds of features show up in many unrelated domestic animals now.