 Reference Base  Conservation genetics of the black rhinoceros (Diceros bi... |
|
|
|
|
|
| |
| Location: |
World |
| Subject: |
Genetics |
| Species: |
Black Rhino |
 |
|
Mitochondrial DNA in Diceros bicornis. Should all the remaining black rhinos be considered as a single population for breeding purposes? This tactic might increase their chances of survival by increasing effective population sizes and thus forestalling stochastic demographic extinctions, inbreeding depression, and loss of the species' existing genetic variability (Soule 1983; Gilpin & Soule 1986; Ralls et al. 1986; Goodman 1987). Alternatively, do different populations (which may or may not coincide with subspecies designations) merit separate conservation as genetically and possibly ecologically distinct units? The latter strategy might prevent outbreeding depression or the production of animals with genetic makeups inappropriate for a given environment (Templeton 1986). These questions should, ideally, be tackled from both an ecological and genetic standpoint. Ecologically, it might be possible to distinguish locally adapted traits. Statistically significant differences in serum vitamin E levels, for example, have been found between Kenyan and southern African samples, which may reflect substantial differences in diet (Dierenfield, personal communication). Ecological differences also distinguish the desert rhinos of Namibia from the highland forest rhinos of Kenya (Du Toit 1987).
In the absence of any clear morphometric differences, IUCN's African Elephant and Rhino Specialist Group has placed a high priority on genetic studies of black rhinos to resolve whether discrete populations could be identified (Du Toit et al. 1987). As a first step in applying molecular genetic techniques to questions of black rhino conservation, we have examined the mitochondrial DNA (mtDNA) of 23 black rhinos representing two morphologically defined subspecies and three geographic populations. We chose mtDNA because its rapid evolutionary rate has shown it to be a useful molecule for determining intraspecific relationships of many animals (e.g., Wilson et al. 1985; Avise & Lansman 1983). If the rhino populations surveyed here have had separate evolutionary histories for a considerable length of time, it should be reflected in the divergence of mtDNA's from animals in different populations.
The mitochondrial genome consists of a closed circular DNA molecule which codes for 13 proteins and a complete set of transfer RNAs. It is extremely conserved in size (about 16,000 base pairs in all mammals that have been examined) and gene arrangement (Brown 1983). It lacks the complicating features of repetitive DNA or introns; therefore, a relatively simple restriction enzyme analysis of the molecule can be undertaken to yield good estimates of genetic relationships among fairly large numbers of individuals. It is maternally inherited without recombination and thus represents an unambiguous marker of maternal phylogeny. Because it evolves 5-10 times more rapidly than single-copy nuclear DNA (Brown et al. 1979) and intraspecific mtDNA variability has been widely demonstrated, this approach seemed the most likely to uncover genetic differentiation among black rhinos, should it exist.
Materials and Methods
With the cooperation of field biologists, wildlife managers, and zoo personnel in both Africa and the United States, we were able to obtain whole blood from both captive and wild-caught black rhinos (Table 1). Our sample included 11 D. b. michaeli of Kenyan origin now kept in U.S. zoos, 11 D. b. minor taken from wild populations in Zimbabwe, and one captive (U.S.) D. b. minor of South African origin. While blood was separated into plasma, red blood cells, platelets, and white blood cells or buffy coats, the latter two components being our primary source of DNA. Total DNA was extracted from white blood cells or buffy coats by standard procedures. We also obtained frozen organ tissue from three animals that died during the period of our study (Table 1). This frozen tissue served as a source of purified mtDNA, which was isolated by the method of differential centrifugation (Lansman et al. 1981 ).
We have used restriction enzymes to survey the black rhinos for mtDNA polymorphisms. Restriction enzymes recognize specific oligonucleotide sequences, usually 4 to 6 base pairs in length, and cleave double-stranded DNA wherever these sequences occur. By surveying mtDNAs with a set of restriction enzymes, we can obtain an accurate estimation of similarity by determining the proportion of restriction fragments and/or restriction sites they share.
Samples of total DNA were digested with 14 restriction enzymes (Bethesda Research Laboratories) having 5 or 6 base pair (b.p.) recognition sites, according to manufacturer's instructions. These enzymes typically cleave mtDNA into 1-7 fragments. The DNA fragments once obtained were separated electrophoreticahy in 1 % agarose gels along with a radioactively labeled (ot-32-P) one-kilobase ladder (Bethesda Research Laboratories), then transferred to GeneScreen-plus membranes (New England Nuclear) by an alkaline blotting procedure (Southern 1975; Reed & Mann 1985). Flurified mtDNA obtained from tissue was then nick-translated with a-32-P labeled nucleotides and was used to 'probe' the southern blots. Membranes were then washed under high-stringency conditions and exposed to Kodak XAR
Additionally, to increase our resolution, we digested the three purified mtDNA samples (one from each population) with four enzymes having 4 b.p. recognition sites. These enzymes cleave the mtDNA into 20-30 fragments and thus have a greater likelihood of revealing differences between individuals. Because each sample contained only purified mtDNA, the resulting restriction fragments could be directly labeled with -32-P (Brown 1980) before being separated electrophoretically on 3.5% polyacrylamide gels. Again, an appropriate radioactively labeled molecular weight/size standard was included in the gel. Gels were subsequently dried under vacuum and exposed to Kodak XAR film.
The proportion of shared restriction fragments was calculated between the observed mtDNA genotypes. The percent sequence divergence between the mitochondrial genotypes was estimated using equation 6b of Upholt (1977). Calculations for restriction enzymes having 6, 5, and 4 b.p. restriction sites were calculated separately, then weighted according to the total number of base pairs recognized by each type of restriction enzyme. This procedure allowed an overall estimate or weighted average of nucleotide sequence divergence, based on the differences revealed by all the restriction enzymes used, between the mtDNA of different in- dividuals.
Table 1. Black rhino samples
Sample Origin Subspecies Tissue Source
1-8 Zimbabwe minor buffy coat Zimbabwe Parks
9 Zimbabwe minor frozen liver do
10, 11 Zimbabwe minor white blood cells Los Angeles Zoo
12 South Africa minor frozen brain Calvin Bentsen, Brownsville
13-16 Kenya michaeli white blood cells Denver Zoo
17-19 Kenya michaeli do St Louis Zoo
20 Kenya michaeli do Atlanta Zoo
21 Kenya michaeli do Busch Gardens
22 Kenya michaeli frozen liver Kansas City Zoo
23 Kenya michaeli white blood cells Detroit Zoo
Results
Each restriction fragment pattern produced by a given enzyme was arbitrarily assigned a letter. The results of the enzymes having 5 or 6 base-pair recognition sites for all 23 animals are listed in Table 2. The results of enzymes for a smaller set of three animals, including enzymes having 4 b.p. recognition sites (i.e., Hinfl, Hpall, MboI, TaqI), are presented in Table 3. The restriction enzymes used in our survey yielded an average of 140 restriction sites per mitochondrial genome. This corresponds to a recognized total of over 630 b.p., or 3.9% of the mitochondrial genome. For 14 out of a total of 18 restriction enzymes, absolutely no mtDNA variability was observed. That is, all rhinos surveyed had the identical restriction fragment pattern (designated as 'A' in Tables 2 and 3) for any one of these 14 restriction enzymes. One enzyme, EcoRI, was found to be polymorphic among the Zimbabwe rhino, with 3 of 11 animals possessing only one EcoRI restriction site instead of the two sites found in the other 8 members of this population. Three enzymes, BcII, HinfI, and TaqI, revealed a difference between the Kenyan population and the Zimbabwe and South African populations. In each case, the result could he interpreted as a single loss or gain of a restriction site.
In total, then, for our sample of 23 animals, only three mtDNA haplotypes could be distinguished:
(1) the Kenyan haplotype with fragment pattern 'B' for BcII, HinfI, and TaqI;
(2) the Zimbabwe haplotype with fragment pattern 'B' for EcoRI; and
(3) the Zimbabwe and South Aftican haplotype with fragment pattern 'A' for all 18 restriction enzymes.
These three mtDNA haplotypes are extremely similar to one another (Table 4), with an average estimated percent sequence difference between any pair of haplotypes and/or populations of 0.17%. The average difference between subspecies was only slightly higher, 0.29%.
Table 2. mtDNA patterns for enzymes with 5 and 6 b.p. sites
Enzyme # sites South Africa Zimbabwe Kenya
n=1 n=11 n=11
AvaI 2 A A A
BamHI 2 A A A
BgIII 1 A A A
ClaI 1 A A A
DraI 5 A A A
EcoRI 2 (1) A A (B) A
EcoRV 2 A A A
HaeII 4 A A A
HindIII 4 A A A
ScaI 7 A A A
XbaI 4 A A A
Table 3. mtDNA patterns for additional enzymes
Enzyme # sites South Africa Zimbabwe Kenya
n=1 n=11 n=11
AvaII 4 A A A
BcII 6 (5) A A B
HincII 7 A A A
HinfI 30 (29) A A B
HpaII 16 A A A
MboI 23 A A A
TaqI 24 (25) A A B
Table 4. Estimated percent sequence divergence between mtDNA types, based on the proportion of shared restriction fragments (Upholt, 1977).
Zimbabwe 1 Zimbabwe 2 S.Africa
Kenya 0.24 0.39 0.24
Zimbabwe 1 - 0.08 0.00
Zimbabwe 2 - 0.08
S. Africa -
Discussion
The results of the mtDNA analysis strongly suggest a very close genetic relationship among all the black rhinos in our survey. Because of the generally rapid rate of mtDNA evolution in mammals, differences observed among rhino populations appear to indicate a very recent common ancestry. If mtDNA evolves at a rate of 2% per million years as suggested (Brown et al. 1979; Wilson et al. 1985), this common ancestry probably dates back no farther than 100,000 years. Indeed, the level of differentiation between the so-called subspecies is well within the range (0-4%) observed among members of other mammalian species (e.g., Avise & Lansman 1983), and even within the range (0-2% ) that has been observed among members of the same local population (Ashley & Wills 1987). Thus, there is no evidence from these data that the black rhinos we sampled represent 'evolutionarily distinct units.'
We chose mtDNA analysis because we thought it would be most likely to uncover genetic differences between the black rhino populations, should they exist. It seems unlikely from our results that significant barriers to successful interbreeding would exist, given what we estimate to he a brief history of separation between the populations in question. However, more information should be obtained before final management decisions are made. The mitochondrial genome represents only a tiny fraction of an organism's genetic makeup, and problems that might arise from interbreeding might not necessarily, be reflected in mtDNA differentiation. For this reason, we are conducting an allozyme survey in our laboratory to determine if the findings regarding the mtDNA hold for nuclear-coded genes as well. The allozyme survey will also be more informative for determining if the black rhinos suffer from reduced levels of genetic variability, as has been reported for some species that have passed through recent population bottlenecks (Bonnll & Selander 1974; O'Brien et al.1983). Thus far, we have found no allozyme polymorphisms within or between populations, despite the fact that (1) we have included in our analysis animals from Kenya, Zimbabwe, and three different populations in South Africa (Etosha, Addo, and Zululand) and (2) our initial surveys have included all three allozyme loci identified by Merenlender et al. (1989) as polymorphic in African rhinos (Amato & Melnick, unpublished data).
Karyotype analysis is also recommended, because chromosomal differences reducing the fertility of hybrids could conceivably exist in the absence of either allozyme or mtDNA differentiation. There have been no known crosses of black rhinos from different subspecies in captivity, which might indicate reduced viability or fertility.
The application of genetics to conservation issues is a practical endeavor and should yield concrete recommendations for management strategies. Black rhinos from the populations included in our study will probably be the ancestors of all future black rhinos, as their successful breeding is the only hope for the survival of the species. Our results provide strong evidence for a very close genetic relationship among these populations. At the national level, the level at which management decisions are currently made, the pooling of black rhinos carries with it little risk of mixing distinct genetic adaptations worthy of separate conservation efforts. This finding should allow managers to aggregate individuals to create larger local populations or demes. Preserving the black rhino in relatively large local populations would have several beneficial effects. These include retarding the rate of loss of genetic variability, buffering each aggregate against the possibility of demographic extinction, restoring previous population densities, and allowing the wildlife managers with limited resources to provide better production against poachers. Taken together, these effects should, in the long run, increase the probability of survival of this critically endangered species.
|
|
|
File available
details
Original text on this topic: