The Restoration of the Quagga; 24 years of selective breeding




Дата канвертавання19.04.2016
Памер113.8 Kb.




The Restoration of the Quagga; 24 years of selective breeding.

Eric H.Harley1*, Craig Lardner2, Michael Gregor3, Bernard Wooding3 and Michael H. Knight4

1 Dept of Clinical Laboratory Sciences, University of Cape Town, Observatory 7925, South Africa

2Far Horizons Wine Estate, Windmeul 7630, South Africa.

3Elandsberg Farms, Hermon, 7308, South Africa

4South African National Parks, PO Box 20419, Humewood, 6013, Port Elizabeth, and Centre for African Conservation Ecology, Dept. Zoology, Nelson Mandela Metropolitan University, Port Elizabeth, S Africa
*email address for correspondence: ehharley@gmail.com
Despite the quagga having been assumed extinct for over 100 years, molecular genetic studies confirm that despite its phenotypic differences,the quagga is conspecific with the common Plains zebra, Equus quagga. The quagga project, therefore, is attempting to restore the characteristic pelage of the quagga by selective breeding from selected Plains zebra individuals. The programme has been in progress for 24 years and is now into its fourth generation of progeny, and is producing individuals with a degree of striping reduction shown by none of the original founders and which approximate the striping pattern typical of some museum specimens of quagga.
Key words: Quagga, Plains Zebra, selective breeding.

Running title: The Restoration of the Quagga.

INTRODUCTION

At the present time there are three extant zebra species: Grevys zebra, Equus grevyi, Mountain zebra, E. zebra, and Plains zebras E. quagga (traditionally E. burchelli, but if the quagga is accepted as a conspecific E. quagga takes precedence). The quagga occupied a region south of the Orange river in the Western Cape region of South Africa (Skead 1987). It had greatly reduced striping over the hind part of the body and legs, and the hind part of the body was typically a dark brown in colour, with the legs remaining white. Due to uncontrolled hunting in the nineteenth century the quagga was driven to extinction in the wild, and the last known captive quagga died in Amsterdam Zoo in 1883. All that remain are 23 skins distributed throughout various museums. There are many paintings of quaggas, but the only photographic record (reproduced below) is that of a lone quagga taken at the London Zoo.

Some 30 years ago a taxidermist at the South African Museum, Reinhold Rau, sent a sample of dry tissue from a quagga foal, for biochemical and immunological analysis. Both DNA (Higuchi et al., 1984, 1987) and immunological (Lowenstein and Ryder, 1985) results were consistent with the quagga being conspecific with the extant Plains zebra, results which were later confirmed in a more comprehensive DNA analysis using 13 museum specimens (Leonard et al., 2005). These results were essential to the initiation of the quagga project since there was controversy as to whether the quagga was a species in its own right or whether it was merely a colour variant of the Plains zebra. For example, from measurements of old museum skulls, the quagga was claimed to more closely related to the mountain zebra than to the Plains zebra (Bennet 1980; Klein & Cruz-Uribe 1999). However, the use of skulls from mounted specimens can be highly unreliable, since readily available horse or donkey skulls were sometimes used for mounting zebra skins. From a summary of the morphological evidence, however, Groves and Bell (2004) concluded that the quagga is a genuine variety of the Plains zebra and promoted use of the original Equus quagga as the appropriate designation for the Plains zebra, as suggested earlier (Rau, 1978).

Following from the molecular evidence of conspecificity, a committee was established in 1986 consisting of zoologists, veterinarians, and museum personnel, with the aim of attempting by a selective breeding programme to retrieve the genes responsible for the characteristic coat patterns of the quagga. The parapatric distribution of quagga and Plains zebra and the wide variety of colour patterns seen in living Plains zebra suggested that these genes may well still be present in diluted form in the extant populations of the Plains zebra.

A total of 19 Plains zebra individuals were selected from reserves in Namibia and South Africa, choosing animals with demonstrable reduction in striping over the rear of the body or legs, or with a tendency to a darker background colour. Plains zebra show a gradient of progressive reduction in striping from north to south of their geographical distribution, with the appearance in the south of lighter “shadow” stripes between the dark stripes (Rau, 1974, 1978), providing ample variation from which to select founder animals for the project. In order not to include characters resulting from new mutations, individuals with obviously deviant striping patterns, a not uncommon phenomenon, were avoided.

It was initially estimated that about four generations of such selection should be sufficient to demonstrate whether simple selective breeding was a feasible way of retrieving the quagga phenotype, and the project is now starting to produce a number of fourth generation foals. The first foal was born to the project in 1988 and the growing herd was later distributed among various separate localities in the Western Cape Province, South Africa, to facilitate identification of pedigrees and to select appropriate grouping of individuals. In keeping with expectations offspring showed either heavier striping, lesser striping, or similar striping patterns to their parents. As time progressed, those with heavier striping were removed, thereby providing the selection process necessary for the success of the project. A previous report documented progress up to the third generation of progeny (Harley et al. 2009). Here we bring up to date the progress of this selective breeding, and the results obtained over the 24 years, and four generations, that the programme has been running.


METHODS

All animals in the project are given a unique, sequential reference number (prefixed by 'F' for a foal which dies before breeding). To measure whether there is progress in stripe reduction animals are photographed from both sides and scored in terms of the number of stripes in five separate body areas, following the method of Harley et al. (2009)

A generation (F) number is calculated for each animal. As not all animals are simple progeny of, for example, first generation matings or pure second generation matings, the generation number of an individual was scored as the total steps through all members of the lineage from the original founder animals to the individual, divided by the number of founder individuals. If a founder appeared twice or more in the pedigree each appearance was included in the total founder number. In this way individuals often received a fractional F number, so for some analyses individuals are grouped into integral number generations by rounding their F number up or down accordingly, e.g. individuals with an F number of 2.6, or 3.2, would both be included in the overall 3rd generation category.

Probabilities associated with Student’s t-test, 2-tailed and with 2-sample unequal variance, were used to measure the significance of difference in the striping and colour pattern between the founder population, the four generations of progeny, museum quaggas, and a selection of 18 randomly chosen wild Plains zebras.



RESULTS

In order to compare the merits of individual stallions and mares to the breeding programme their own individual striping scores were compared at intervals (Figure 1) and animals on the leftward half of the plot were considered appropriate for removal from the program, their sale contributing in a major way to the funding of the programme. An additional measure was to determine the mean stripe scores (in hind body and legs) measured in their progeny (Figure 2). This gave a measure of the breeding value of individuals, which assisted in the selection process, especially for individuals with intermediate values for their own stripe scores.

After 24 years of selective breeding the project is now producing its first fourth generation progeny, a total of 10 by the beginning of 2010. This enables striping scores to be compared between wild Plains zebra, the 19 founders (although only 18 produced progeny), the four generations of progeny, and the museum representatives of the original quaggas. Figure 3 plots the changes in striping over the four generations, with Table 1 detailing the significance values for these differences. Significant reduction in stripe number between wild zebras and founders was seen in the hind body and the legs (Table 1), demonstrating effective selection of founders for initiating the selection programme. More interesting, however, is that there is a progressive reduction in striping in these areas over the four generations of selective breeding, with most inter-generational differences being significant, and with the new fourth generation data, for the first time the mean value for foreleg striping is not significantly different from the mean of the quagga data. As expected, there was little of significance between any group in comparative stripe numbers in the neck regions, nor in the forebody region, except for occasional intermediate significance values (<0.5 but >0.1). The background colour estimates showed no progressive change over time between founders and progeny generations.

In order to determine whether there was a correlation between stripe reduction in hindbody and in forelegs, values for these areas were compared for individuals in the project extant in 2009 (Figure 4 (a)). In addition, the mean values in offspring of all stallions and mares that had had four or more foals were compared (Figure 4 (b)). Correlations between stripe number on hindbody with stripe number on legs is quite weak when individual animals values are plotted (R2 = 0.12; n=97), but is markedly better when the mean foal values of individual stallions or mares are plotted (R2 > 0.41; n=38). The same was found to apply to the correlation between the number of stripes on forelegs with those on hindlegs, although the overall correlations there were somewhat stronger, with R2 values of 0.61 and 0.91 respectively



DISCUSSION

The success of the project in reducing striping over the rear part of the body and legs, as was characteristic of the original quagga, is demonstrable not only from the data shown in Figure 3 but more practically in that some third and especially fourth generation individuals now more closely resemble quaggas than typical Plains zebra. One such animal is illustrated in Figure 5 together with a typical Plains zebra, and a photograph of the London Zoo quagga.

Major considerations now for the project are, firstly, to try and define at what point it can be claimed that the quagga phenotype has been recovered, and secondly, even if the external appearance were to be accepted as conforming to the quagga pelage characteristics, whether it is justifiable in referring to such individuals as quaggas.

Addressing the first question, there are now fourth generation animals with a degree of stripe reduction over the hindbody region which is similar to at least some museum quaggas. Illustrations of some of these individuals as well as representative museum quaggas can be seen on the project web site, www.quaggaproject.org, where it can be observed that these museum specimens are quite variable in striping reduction. Although some of the fourth generation animals have zero stripes on the front legs, none have yet achieved a total absence of stripes on the back legs, two or three stripes tending to persist on the hocks. In this context, it is interesting that there is a reference by Darwin (1872) in "The Origin of Species" to a quagga being observed with "...very distinct zebra-like bars on the hocks", so complete absence of stripes on both legs, although desirable, may not be an absolute requirement. Similarly, there is considerable variation in the depth of background colour in museum quaggas; our fourth generation foals have quite a wide variation in background colouration, from almost white to a mid-brown colour, a range which marginally overlaps the range seen in museum quaggas, but does not approach that seen in the darkest quagga specimens. Background colour is the most difficult measure to quantify, with non-genetic factors such as sun exposure, dust, diet, photographic quality and lighting, and age (for museum specimens) complicating objective measurement, so for the purposes of the project stripe reduction is considered the more important quantitative criterion in judging an individual's status.

Addressing the second question, some conservationists believe that even if an animal conforming to the striping and colour patterns of the quagga were achieved, it would not be correct to describe it as a quagga because the original quaggas might have had other undescribed, and perhaps adaptive attributes, which distinguished them from Plains zebra. The counterargument would be that since it is not possible to define what these presumptive attributes were then the quagga can be defined only by its known attributes, which are its coat patterns, and if these are retrieved by selective breeding then describing the animals possessing them as representatives of the quagga phenotype is justifiable. Both these views have their merits, and therefore the project committee has decided that individuals with a degree of stripe reduction such that there are no scorable stripes on the hindbody or legs (there is still debate as to whether a few stripes on the rear hocks are allowable) will qualify for the title of "Rau quagga", the "Rau" qualifier addressing the concerns of the first question discussed above. The background colour, being so difficult to quantify in both museum quaggas as well as living project animals, will not be a defining criterion for the term "Rau quagga", but will continue to be an objective in the breeding process, and animals with the darkest background colouration will claim priority in the ranking of Rau quaggas.

That in as little as four generations individuals have been produced with a degree of stripe reduction seen in none of the founders implies significant variation in the genes affecting pelage characteristics in the founder population. Such a high degree of genetic variation, together with a remarkable lack of genetic differentiation between the different subspecies, has in fact been recently documented (Lorenzen et al. 2008), implying that genes for pelage characteristics are spread widely through the Plains zebra population. The correlation shown in Figure 4 between the number of stripes on the body and the legs suggest that a single gene is at least primarily responsible for the degree of striping in these areas, and the wide variation between project individuals suggests that such a gene, or its control region, is multiallelic, with most individuals being compound heterozygotes for these alleles. The project may therefore provide an opportunity for future molecular genetic studies into the complex genetics of colour patterning.

Currently the project animals are distributed over 13 closely positioned localities, which facilitates selective breeding and pedigree determination. With the recent achievement of a number of fourth generation animals qualifying for the designation of "Rau quagga" the aim is soon to establish these individuals as a consolidated herd in a national park in the Western Cape. This will enable visitors to experience a rendition of the zebra, in the original locality of the quagga, which has not been encountered for over a hundred years.

ACKNOWLEDGEMENTS:

We are indebted to the late Reinhold Rau, the initiator of the project, which he inspired and promoted from its inception. We thank the many participating landowners who have maintained project animals on their properties, and Brenda Larison for drawing our attention to the Darwin reference.


REFERENCES

1) Bennet D (1980) Stripes do not a zebra make, Part I: A cladistic analysis of Equus. Syst Zool 29, 272-287.

2) Darwin C (1872) In: The Origin of Species, 6th Ed. The New American Library of World Literature, Inc. New York. p153.

3) Groves C, Bell C (2004) New investigations on the taxonomy of the zebras genus Equus, subgenus Hippotigris. Mamm Biol 69, 182–196.

4) Harley E, Knight M, Lardner C, Wooding B, Gregor M (2009) The Quagga project: progress over 20 years of selective breeding. S Afr J Wildlife Res 39, 155-163.

5) Higuchi R, Bowman B, Freiberger M, Ryder O, Wilson A (1984) DNA sequences from the quagga, an extinct member of the horse family. Nature 312, 282–284.

6) Higuchi R, Wrischnik L, Oakes E, George M, Tong B, Wilson A (1987) Mitochondrial DNA of the extinct quagga: relatedness and extent of postmortem change. J Mol Evol 25, 283–287.

7) Klein R, Cruz-Uribe K (1999) Craniometry of the genus Equus and the taxonomic affinities of the extinct South African quagga. S Afr J Sci 95, 81–86.

8) Leonard J, Rohland N, Glaberman S, Fleischer R, Caccone A, Hofreiter M (2005) A rapid loss of stripes: the evolutionary history of the extinct quagga. Biol Lett 1, 291-295.

9) Lorenzen E, Arctander P, Siegismund H (2008) High variation and very low differentiation in wide ranging Plains zebra (Equus quagga): insights from mtDNA and microsatellites Mol Ecol 17, 2812–2824.

10) Lowenstein J, Ryder O (1985) Immunological systematics of the extinct quagga (Equidae). Experientia 41, 1192-1193.

11) Rau R (1974) Revised list of the preserved material of the extinct Cape colony quagga, Equus quagga quagga (Gmelin). Ann S Afr Mus 65, 41–87.

12) Rau R (1978) Additions to the revised list of preserved material of the extinct Cape colony quagga and notes on the relationship and distribution of southern Plains zebras. Ann S Afr Mus 77, 27–45.

13) Skead C (1987) Historical Mammal Incidence in the Cape Province Volume 2: The eastern half of the Cape Province, including the Ciskei, Transkei and East Griqualand. Republic of South Africa : Department of Nature and Environmental Conservation of the Provincial Administration of the Cape of Good Hope, Cape Town.



LEGENDS FOR FIGURES:

Figure 1. Example of individuals sorted initially by the number of stripes in the hindbody region, secondly by the number of stripes in the forelegs, and thirdly (although individual values are not displayed) by the number of stripes on the hindlegs.

Figure 2. Individuals that had had four or more foals sorted by the mean number of stripes measured in the hindbody and in the forelegs of their foals.

Figure 3. The mean number of stripes in hindbody, forelegs, and hindlegs in the founder population, and in each of the four generations (F1 to F4) of progeny. Data are from 18 founders, 66 F1, 77 F2, 25 F3 and 10 F4 individuals.

Fig. 4. Correlation of stripe numbers between hindbody and forelegs; (a) individual data from all mares and stallions alive in Jan 2010; (b) data are the mean values in the foals of individual mares or stallions (from all generations) with more than 4 foals (mean number of foals per individual = 8.4).

Fig. 5. A montage comparing a typical Plains zebra (left) with a fourth generation quagga project individual (centre) and one of the only three photos remaining of a living quagga (right), the latter photographed in London in 1870.

Fig 1.


Fig 2.


Fig 3.


Fig 4


a) b)
Fig 5.



Table 1. T-test probabilities for the comparisons between striping scores for the various areas and colour scores between wild zebras, project individuals, and museum quaggas (NS = not significant at the 5% level).


Area 1: Neck

























Founders

F1

F2

F3

F4

Quaggas

Zebras




NS

NS

NS

NS

NS

NS

Founders







NS

NS

NS

NS

NS

F1










<0.05

<0.05

NS

NS

F2













NS

NS

NS

F3
















NS

NS

F4



















NS

























Area 2: Forebody

























Founders

F1

F2

F3

F4

Quaggas

Zebras




NS

NS

NS

NS

<0.05

<0.05

Founders







NS

NS

NS

NS

<0.05

F1










NS

NS

NS

<0.05

F2













NS

<0.05

<0.05

F3
















NS

<0.05

F4



















<0.05

























Area 3: Hindbody

























Founders

F1

F2

F3

F4

Quaggas

Zebras




<0.01

<0.01

<0.01

<0.01

<0.01

<0.01

Founders







NS

<0.05

<0.01

<0.01

<0.01

F1










<0.01

<0.01

<0.01

<0.01

F2













<0.01

<0.01

<0.01

F3
















NS

<0.01

F4



















<0.01

























Area 4: Forelegs

























Founders

F1

F2

F3

F4

Quaggas

Zebras




<0.01

<0.01

<0.01

<0.01

<0.01

<0.01

Founders







<0.01

NS

NS

<0.05

<0.01

F1










<0.01

<0.01

<0.01

<0.01

F2













<0.05

<0.01

<0.01

F3
















NS

<0.01

F4



















NS

























Area 5: Hindlegs

























Founders

F1

F2

F3

F4

Quaggas

Zebras




<0.01

<0.01

<0.01

<0.01

<0.01

<0.01

Founders







NS

NS

<0.05

<0.01

<0.01

F1










<0.05

<0.01

<0.01

<0.01

F2













NS

<0.05

<0.01

F3
















NS

<0.01

F4



















<0.01

























Background colour

























Founders

F1

F2

F3

F4

Quaggas

Zebras




NS

NS

<0.05

NS

<0.05

<0.01

Founders







NS

NS

NS

NS

<0.01

F1










<0.05

<0.05

NS

<0.01

F2













NS

NS

<0.01

F3
















NS

<0.01

F4



















<0.01




База данных защищена авторским правом ©shkola.of.by 2016
звярнуцца да адміністрацыі

    Галоўная старонка