Genetic variability and population differentiation inferred from DNA fingerprinting in silvereyes (Aves: Zosteropidae).
One such tool is the DNA fingerprint (Jeffreys el al. 1985a,b), which reveals restriction fragment length polymorphisms at hypervariable minisatellite loci that are dispersed throughout the genome. Each of these loci comprise a head-to-tail tandem array of short (10 to 60 nucleotide base pairs) repeat units, and the length polymorphisms reflect the variable number of repeat units among loci (Jeffreys et al. 1985a). Fingerprinting has proven especially valuable for testing parentage in a variety of wild and captive animals, because of the numerous low frequency variants that generate a pattern unique to an individual (Jeffreys et al. 1985b). The potential of fingerprinting to address population-orientated questions remains relatively unexplored, probably because the DNA sequences involved generally are thought to evolve too rapidly to retain information about the evolutionary history of a population over periods longer than a few generations. However, in small isolated populations, the fixation of variants already existing in the population might outpace the generation of new length variants (e.g., Faulkes et al. 1990; Gilbert et al. 1990; Wayne et al. 1991). This could result in the temporal stability of minisatellite variants that reflect the evolutionary history of the population.
Minisatellite DNA is untranscribed and apparently nonfunctional at the phenotypic level, thus, assumptions of neutral and unconserved variation are not unreasonable (Jarman and Wells 1989). In fact, Jeffreys et al. (1988) have shown that neutral mutation rates appear sufficiently high to maintain the observed levels of hypervariability. Minisatellite loci show a wide range of heterozygosities, from monomorphic to hyperpolymorphic (Jarman and Wells 1989). Although the exact mechanism remains unclear, new minisatellite variants probably arise by mitotic sister chromatid exchange (Jeffreys et al. 1990). The frequency of this form of mutation, which is low enough to enable the use of minisatellites as genetic markers, is relatively high compared with the rates of point mutation assayed by standard restriction fragment length polymorphisms analysis of mitochondrial and low copy number nuclear DNAs (Jeffreys et al. 1985a; Wonget al. 1986; Burke and Bruford 1987; Gyllensten et al. 1989). Assuming that allele size is selectively neutral, the number of alleles that is maintained within a population depends on the rate of generation by unequal crossing-over and the rate of loss by random genetic drift (Jarman and Wells 1989). All of these features combine to make minisatellite loci potentially appropriate to address questions of genetic variation within and between closely related populations.
Only a handful of published terrestrial field studies so far have used DNA fingerprinting to address population rather than parentage questions. A high level of inbreeding was inferred from low minisatellite diversity in the naked mole rat (Heterocephalus glaber; Faulkes et al. 1990; Reeve et al. 1990), in a small, endangered island population of gray wolves (Canis lupus; Wayne et al. 1991), and in populations of blue ducks (Hymenolaimus malacorhynchos) confined to single-river catchment areas (Triggs et al. 1992). In an extreme case, individuals within island populations of the California Channel Islands fox (Urocyn littoralis) were found to be up to 100% similar at minisatellite loci, with fixed differences between the island populations (Gilbert et al. 1990).
In this paper, DNA fingerprinting is used to compare genetic diversity within and between island and mainland populations of silvereyes (Zosterops lateralis). Off the eastern Australian coast, the Capricorn silvereye (Zosterops lateralis chlorocephala) is one of many examples of island colonization among the Zosteropidae, and high density populations exist on several small wooded cays in the Capricorn and Bunker Island groups of the southern Great Barrier Reef (Mees 1969; Kikkawa 1973). The island race is noticeably larger than the mainland race (Zosterops lateralis familiaris; Kikkawa 1976). This morphological divergence may have occurred in an evolutionarily short time, because the original colonization of present-day coral islands on the southern Great Barrier Reef can probably be limited to the time since suggested formation of island vegetation, about 3000 to 4000 yr ago (Hopley 1982). Similarity of mitochondrial DNA (mtDNA) haplotypes found in the island and the mainland races also suggests a recent separation (Degnan and Moritz 1992).
On Heron Island, in the Capricorn group, mainland birds often appear during autumn/winter, yet in two decades of comprehensive surveys of all breeding pairs on the island, there are no records of mainland birds breeding with island birds (Kikkawa 1970, pers. comm. 1992). Silvereyes are monogamous and on Heron Island live up to 11 yr (Kikkawa 1987). The silvereye population on one of the Bunker Islands, Lady Elliot Island, is known to have colonized only within the past two decades. Prior to the 1960s, Lady Elliot Island suffered a loss of forest habitat because of destruction by goats, and there were no silvereyes on the island in the early 1970s (Walker 1986). The recent colonization of silvereyes coincides with a regrowth of forest on the island since the goats have been removed. In 1985-1986, 100 to 200 individuals were recorded as residents (Walker 1986).
The purpose of this study was to evaluate the degree to which variability at minisatellite loci reflected these population-level processes. First, in the Heron Island population, the relationship between many individuals is known from long-term breeding data. Thus, it was possible to undertake a preliminary pedigree analysis of inheritance of silvereye minisatellite loci and to assess the usefulness of DNA fingerprints in quantifying genetic relatedness between randomly selected individuals. Second, the age structure of this population is also known and permitted analysis of the temporal stability of the population fingerprint profile over several generations. Third, the distribution of silvereyes described above permitted a comparison of DNA fingerprint variability within and among avian populations having different demographic structures and different recent evolutionary histories. Genetic variability in the Heron Island population was compared with that in neighboring Capricorn Island populations, with which a low level of exchange of silvereyes occurs (Kikkawa 1987), and with the recently colonized population of the same race (chlorocephala) from Lady Elliot Island (Bunker Islands). In turn, the island populations were compared with a mainland population (Brisbane) of a different race (familiaris). This study then serves to demonstrate whether the pace of generation of new minisatellite variants may be matched, or even outpaced, by that of fixation of existing variants in small isolated populations. If so, the variation observed at minisatellite loci could prove a valuable molecular tool in addressing the genetic effects of population-level evolutionary processes.
MATERIALS AND METHODS
A preliminary evaluation of the patterns of inheritance of silvereye minisatellites was undertaken by fingerprinting six families, including four broods of two young and two broods of three young. Five of the families comprised Heron Island birds and one was a mainland (Brisbane) family. For population analyses, silvereyes were trapped or mist-netted on Heron Island (N = 85), six additional Capricorn Islands (N = 36), Lady Elliot Island (N = 16), and in Brisbane (N = 16). Offspring from known families were excluded from the population samples. For each individual, genomic DNA was extracted from a 20 ||micro~liter~ to 50 ||micro~liter~ nondestructively obtained blood sample, using a standard phenol-chloroform procedure (Sambrook et al. 1989).
DNA fingerprints were generated using wild type single-stranded M13mp11 (Vassart et al. 1987; New England Biolabs) with electrophoretic and hybridization conditions adapted from Westneat et al. (1988) and Chen et al. (1990). In particular, 2.5 ||micro~gram~ to 3.0 ||micro~gram~ of HaeIII-digested silvereye genomic DNA was electrophoresed in 0.6% agarose/TBE, at 40 volts for 40 h to 44 h and transferred onto Hybond |N.sup.+~ (Amersham) membranes using an LKB Vacugene apparatus. Under suction, gels were denatured for 10 min (at 40 cm |H.sub.2~O), neutralized for 5 min at the same pressure, and transferred in 20 X SSC for 1.5 h to 2.0 h (at 25 cm |H.sub.2~O). Membranes were fixed in 0.4 N NaOH.
Typically, 25 ng to 50 ng of single-stranded M13 was 32P-random-prime labeled (Feinberg and Vogelstein 1984) using an Amersham Multi-prime kit and chromatographically purified from unincorporated nucleotides via a Nick-column (Pharmacia). Hybridizations were at low stringency (55 |degrees~ C), with 1-3 X |10.sup.6~ cpm/mL of labeled probe in a phosphate-buffered hybridization solution (Church and Gilbert 1984). Following hybridization, nonspecifically bound probe was removed by two 10-min washes in room temperature 2 X SSC, 0.1% SDS. A more stringent final wash was performed in 2 X SSC at 55 |degrees~ C for 30 min. Autoradiographs (Kodak XAR film) were exposed for 16 h to 96 h with one intensifying screen; usually, more than one exposure was made to ensure accurate determination of bands of differing intensity.
DNA fragments were scored by comparing migration distances against the size standard run on every gel. A band was considered identical in two individuals if it was of similar intensity and had migrated no more than a 1.0-mm difference in the two individuals. This value represented the maximum difference in the distance migrated by any given band in the same individual run several lanes apart. Often, two lanes of the same individual were run on each gel to facilitate comparisons of band identity across several lanes; on several occasions, the same individual was run on different gels to confirm consistency of banding patterns.
An index of similarity (|S.sub.xy~; e.g., Lynch 1990, 1991) between DNA fingerprints was calculated as the number of bands shared between each pair of individuals (|n.sub.xy~) divided by the total number of bands scored for both individuals (|n.sub.x~ + |n.sub.y~),
|S.sub.xy~ = 2|n.sub.xy~/(|n.sub.x~ + |n.sub.y~).
This is equivalent to Jeffreys et al.'s (1985b) value of x for unrelated individuals and identical to Wetton et al.'s (1987) value of D for unrelated individuals. Population means (|S.sub.xy~) and standard errors of pairwise similarity indices were used for within- and between-population comparison of fingerprints. An unbiased estimation of the standard error of |S.sub.xy~ may be defined as
||2S(1 - S)(2 - S)/f(4 - S)~.sub.0.5~,
where S is the mean probability of band sharing, and f is the mean number of fragments scored in each individual (Lynch 1990, 1991).
Similarity of band sharing between populations, corrected for within-population similarity. was assessed using Lynch's (1990) formula, expanded to incorporate terms for each of the seven islands sampled:
|S.sub.ij~ = 1 + |S|prime~.sub.ij~ - |(|S.sub.i~ + |S.sub.j)/2~.
where |S.sub.i~ is the mean proportion of bands shared among individuals of population i, and |S|prime~.sub.ij~ is the mean proportion of bands shared between random pairs of individuals across populations i and j.
Patterns of Inheritance.--In the six silvereye families analyzed, all fragments detected in the offspring matched those in one or both of the putative parents; that is, no novel fragments were observed in offspring. Parents shared on average 0.63 |+ or -~ 0.11 bands with their offspring; within each family, siblings shared an average of 0.68 |+ or -~ 0.15 bands with each other. From this limited sample, the minisatellite fragments appeared generally to represent alleles of heterozygous loci that are stably inherited and segregate in a Mendelian fashion. By identifying fragments that were exclusive to either the male or female parent, but not both, several unlinked loci were identified as fragments that segregated independently in the offspring. Of the 27 maternal- and paternal-specific bands identified in all families, only five were transmitted to all offspring in a family; four of these were in families with only two young. In a family of two offspring, only |2.sup.2~ = 4 possible segregation patterns exist in offspring for a given parental DNA fragment, hence, the small family sizes and the few families analyzed preclude rigorous statistical testing of linkage. Nonetheless, in the present study, all minisatellite fragments in all offspring can be traced back to one or the other parent, and therefore provide a set of stably inherited genetic markers. This was clearly visible in a family spanning three generations (data not shown), in which some bands from a single grandparent could be followed to the offspring of the third generation. Because no novel bands were observed in the family analysis, a mutation rate for these hypervariable loci in the silvereye cannot be estimated.
Band sharing between three sets of Heron Island parents and their own offspring was compared to that between the same parents and non-related offspring. As expected, parents shared more bands with their own offspring than with nonrelated offspring. Using individuals of "known" relatedness, it was possible to assess the degree to which M13 band sharing could further discriminate different levels of relatedness. Coefficients of relatedness, r, were determined directly from breeding records (J. Kikkawa unpubl. data) according to the convention of Wright (1922). Mean fingerprint similarity indices among both first degree and second degree relatives are higher than observed overall in Heron Island. However, the ranges of band sharing in all three relatedness categories overlap substantially, such that the degree of relatedness between any two individuals could not be determined from their similarity index alone. For 60 individuals of known sex, there was no sex linkage observed for commonly occurring fragments.
To assess the stability of band sharing over 5 yr within the Heron Island population, mean band sharing was compared within and between categories defined by cohort (year born). No shift occurred in mean band sharing in the population over the years sampled. Band sharing between cohorts, derived from pairwise comparisons of three individuals from each of five co-horts, showed that the minisatellite size variants observed in new recruits into the Heron Island population remained the same from one season to the next. Individuals shared on average 0.56 |+ or -~ 0.18 bands with birds born in the same year and 0.51 |+ or -~ 0.15 bands with birds born in a different year.
TABLE 1. Band sharing (S, after Lynch 1990, 1991) between Heron Island adults and offspring. Bold type shows band sharing between parents and their own young. Plain type shows band sharing between adults and unrelated young. The value shown for each pair-wise comparison is the mean of the two values generated by that comparison. Only birds on a single gel were compared. Young Adults F13 F14 F15 F37 F38 F30 F31 H60 0.71 0.59 0.84 0.40 0.42 0.50 0.50 H76 0.56 0.63 0.56 0.42 0.33 0.53 0.53 H253 0.42 0.32 0.38 0.73 0.67 0.33 0.40 CB88 0.40 0.53 0.59 0.56 0.47 0.71 0.71 H288 0.46 0.31 0.40 0.38 0.40 0.83 0.67
Population Comparisons.--The average number of scorable bands per individual was consistent among the different populations, with a mean of 6.4 |+ or -~ 1.3. The within-population mean band-sharing similarities (S) ranged from 0.31 to 0.67. Within the Capricorn Island group, the mean level of band sharing between individuals from different islands (0.47 |+ or -~ 0.06, data not shown) was similar to that between individuals within an island population (0.48 |+ or -~ 0.13). The similarity index (from Lynch 1990) between populations of the Capricorn Islands was estimated as 0.99. Together, these two results strongly suggest that these populations are homogeneous, and subsequent analyses, therefore, included only Heron Island as a representative of the Capricorn group.
The lowest within-population band sharing (0.31), indicating the highest genetic diversity, occurred among individuals of the mainland (Brisbane) population. In contrast, the highest band sharing (0.67) was observed among individuals of the Lady Elliot Island population. Populations of the Capricorn group, including Heron Island, also had high proportions of band sharing (0.38 to 0.57) relative to the mainland but were genetically more diverse than the Lady Elliot Island population. On Heron Island and other Capricorn Islands, it is possible to find an individual "a" that shares all of its bands with individual "b," or none of its bands with individual "b." Among Brisbane birds, the former case was never observed, and among Lady Elliot Island silvereyes, no two birds shared less than 0.38 of their bands. Despite mean differences, the distributions of band sharing within the three populations overlapped substantially. Thus, it would not be possible to assign an individual of unknown origin to one or the other of these populations based only on the degree of band sharing between the unknown TABULAR DATA OMITTED bird and birds from each of the three populations. Nonetheless, the populations appear genetically differentiated, as reflected in lower band sharing between rather than within populations. In fact, the degree of band sharing between birds from different populations is similar to or less than that between presumably unrelated individuals on the mainland.
Within the island populations, four fingerprint bands occurred in more than half of the individuals sampled; only one was observed in some (31%) Brisbane birds. None of these common bands were fixed in all island populations, although a 6.2-kb fragment was observed in 100% of Lady Elliot Island birds and in 81% of birds from the Capricorn Islands. An 8.4-kb fragment was observed in 88% of Lady Elliot Island birds, and at a consistently lower frequency among birds of the Capricorn Islands (average 28%). The occurrences of these common bands in the two island populations, when compared using a ||Chi~.sup.2~ contingency test based on the actual numbers of fragments observed, showed only a slightly significant difference among the two populations (||Chi~.sup.2~ = 9.16, df = 3, 0.01 |is less than~ P |is less than~ 0.05). The stability of these four common fragments over 5 yr was assessed in the Heron Island population by comparing their frequencies within and between categories defined by cohort (data not shown), as described above. There were no significant temporal changes observed in the frequency of the common fragments (||Chi~.sup.2~ = 8.39. df = 12. P |is less than~ 0.01).
Pedigree analysis demonstrated that DNA fingerprints provide a set of stably inherited genetic markers in the silvereye, clearly an important prerequisite to a population analysis. Parents shared more bands with their own offspring than they did with unrelated offspring, but the distributions of band-sharing values overlapped considerably between first degree and second degree relatives, and somewhat less between relatives and nonrelatives. As such, band-sharing measures could not be used definitively to quantify genetic relatedness between randomly selected individuals in the Heron Island population, because of the high level of background band sharing. Where band sharing among unrelated individuals in the population is low, it has been shown to correlate highly with relatedness. In such cases, band-sharing data could be used to distinguish distinct classes of relationship; for example, among Serengeti lions, DNA fingerprinting successfully distinguished between kin with a coefficient of relatedness r |is greater than or is equal to~ 0.125, those with r = 0.02-0.06, and nonrelatives (Packer et al. 1991). The present study suggests that in small isolated populations, such as those confined to islands, higher levels of relatedness among most individuals in the population may preclude the use of fingerprinting to estimate the degree of relatedness between particular individuals.
TABLE 3. Occurrence of commonly observed minisatellite restriction fragments in populations of silvereyes. Proportion of individuals 9.5 8.4 6.2 5.0 Population N kb kb kb kb Capricorn Islands 121 0.65 0.28 0.81 0.92 Lady Elliot Island 16 0.44 0.88 1.00 0.81 Brisbane 16 0.00 0.00 0.00 0.31
However, fingerprint band-sharing measures were successfully used to estimate the extent of relatedness within and between the particular silvereye populations studied and revealed differences in minisatellite allelic diversity. Mean fingerprint similarity within the Brisbane (mainland) population was considerably lower than that within any of the island populations. This is consistent with the prediction that a larger population, such as would be expected for a mainland population, will have higher genetic diversity. Among mainland birds, mean band sharing was comparable to the highest levels generally reported for outbred avian populations (range 0.20 to 0.30). However, caution is warranted when comparing results across studies and using them to draw conclusions about inbreeding or mutation rates, because band sharing varies for different minisatellite probes, even when used with the same set of individuals (e.g., Georges et al. 1988; Reeve et al. 1990; Westneat 1990).
Comparisons between populations analyzed under exactly the same set of fingerprinting conditions, however, can be very informative. In this study, the Brisbane population is assumed to represent the "norm" for a large, outbred population of silvereyes. Thus, the higher band sharing observed within the island populations likely reflects their smaller effective population size. Importantly, the estimates of within-population genetic diversity correlate well with population size, a prediction of population genetic theory (Wright 1969). Band sharing among silvereyes on the Capricorn Islands was intermediate between that observed on the mainland and that observed on Lady Elliot Island. The Heron Island population of about 400 individuals (census population size; Kikkawa 1987) is up to twice the size of that on Lady Elliot Island. High levels of band sharing between silvereyes from different islands of the Capricorn Group strongly suggest that the entire island group may constitute a single admixed population, as would be predicted from the occasional movement of birds banded on Heron Island to other Capricorn Islands (Kikkawa 1987, pers. comm. 1992), such that the total census population size may be even larger. In addition, Heron Island has undoubtedly supported a silvereye population for much longer than the present Lady Elliot population, although the time since colonization of Heron Island is unknown. Presumably silvereyes could have colonized Heron Island and/or the neighboring Capricorn Islands at any time since vegetation became established on these islands, in the vicinity of 3000 to 4000 yr ago (Hopley 1982).
Lady Elliot Island had the highest levels of band sharing observed in silvereyes, and is the most recently colonized of the populations studied. The population may have passed through a bottleneck at colonization, and this could have a dramatic effect on genetic variability (Nei et al. 1975). Lady Elliot Island is also the most remote of the island populations sampled. It is separated from the Capricorn Islands by approximately 100 km, from other Bunker Islands by at least 40 km, and from the mainland by approximately 90 km. The lower genetic variability in this population could therefore be attributed to one or a combination of the following factors: (1) recent founding, (2) small (founder) population size, and (3) high degree of geographic isolation and thus low immigration rates. The possibility that reduced variation is caused by selective advantage associated with particular bands is unlikely in view of the presumptive neutral nature of the loci (e.g., Jeffreys et al. 1988). Even if the Bunker group of islands proves with further study to be a single admixture, as appears to be the case for the Capricorn Group, the fewer vegetated cays comprising the Bunker Group (four; fig. 1) means that it will almost certainly support a substantially smaller total population size than that of the Capricorn group (nine cays; fig. 1). Silvereyes on Lady Elliot Island show a slightly higher fingerprint similarity to birds from the Capricorn group than to birds from the mainland, consistent with the recent colonization by island birds (race chlorocephala) and not mainland birds (race familiaris). Similarities in morphology between Lady Elliot Island and Capricorn Island birds support this same conclusion; indeed, silvereyes on all of these islands clearly represent the larger island race and not the mainland race (S. Degnan pets. obs.; J. Kikkawa pers. comm. 1991). Present evidence from this analysis of minisatellite diversity would suggest that the Capricorn Islands may be genetically differentiated from the Bunker Islands. However, this is based upon the study of a single Bunker Island population (Lady Elliot Island), in which the current status of genetic diversity may have been affected by a recent population bottleneck. Further investigation of other populations in the Bunker Island group is required to fully appreciate the relationship between the two island groups.
Fingerprint similarity among Lady Elliot Island birds (0.67 |+ or -~ 0.13, table 2) correlates remarkably well with values estimated among inbred individuals in other studies. For example, Hillel et al. (1989) reported a mean band sharing of 0.64 among Muscovy ducks (Anser anser domesticus) sampled from a highly inbred flock. In a small, inbred population of gray wolves on Isle Royale, founded within the past 50 yr, all 14 individuals were shown to be genetically as similar as captive populations of siblings (Wayne et al. 1991), with mean band sharing of about 0.68. Furthermore, this level of band sharing among Lady Elliot Island silvereyes was the same as that observed among sibling Heron Island silvereyes (0.68 |+ or -~ 0.15), exactly as reported for the gray wolves.
The three populations studied (considering the Capricorn Islands as a single population represented by Heron Island) differ not only in their degrees of band sharing, but also in the frequency of particular minisatellite fragments. This also implies population differentiation. On the islands, two fragments (6.2 kb and 5.0 kb) were observed in more than 70% of birds sampled, and the 6.2-kb fragment was observed in 100% of Lady Elliot Island birds sampled. In contrast, among mainland birds, the 6.2-kb fragment was not seen at all, and the 5.0-kb fragment was detected in only 31% of individuals. The low levels of polymorphism in these bands among island birds again reflects the small size of the island populations. Nonetheless, no specific fragments existed that could be used as diagnostic population markers. The high frequency of these fragments may indicate that they are becoming fixed in the island populations. Among five generations of Heron Island birds, no significant changes in frequencies of the common bands were detectable. As a corollary, an analysis of the total DNA fingerprint profile of the Heron Island population across these same generations showed that the population profile remained very stable over time. This is the first known analysis of the temporal stability of minisatellite variants and provides evidence that the rate of generation of new variants is matched by the rate of fixation of already existing variants, at least in the small island populations. Thus, minisatellite variability may indeed reflect evolutionary histories over time scales greater than just one or two generations.
Indeed, in the absence of significant gene flow between the mainland and the island groups, the populations have apparently become distinct at hypervariable minisatellite loci. This situation is consistent with the observation that occasional winter immigrants of the mainland race disappear by the onset of the breeding season on Heron Island, such that no interbreeding between island birds and mainland birds is known today (Kikkawa 1970, pers. comm. 1992). Differentiation between the populations can be demonstrated in two interrelated ways. First, the mean band sharing between any two populations was less than the mean within populations, reflecting different levels of allelic diversity in each of the three populations. Second, common fragments observed in the fingerprints of most island birds were not observed or were not common in the fingerprints of mainland birds. Differentiation among islands was also demonstrated for fox populations on the Channel Islands, where mean band sharing ranged from 0.75 to 1.00 and was correlated to population size but not to time since colonization (Gilbert et al. 1990). The time scale for colonization of the islands by the foxes is in the same order of magnitude as that proposed for silvereye colonization of Heron Island: that is, in the order of 2000 to 4000 yr. The census sizes of the fox populations are generally much greater than the size of the Heron Island silvereye population. Nonetheless, mean band sharing within the fox island populations was extremely high, and the apparent fixation of particular variants has occurred with much higher frequency than observed in the silvereyes: that is, the four smallest of the six fox populations sampled had unique restriction fragments found in all individuals on the island (Gilbert et al. 1990).
These observations raise some questions about the generality of evolutionary and conservation biology concepts including the impact of inbreeding, random drift, and population fluctuations on small and isolated populations. In the absence of mutation, migration, and selection, genetic variation in small finite populations is expected to become progressively reduced through random drift (Wright 1969). The rate of loss of genetic variation can be estimated from the variance effective population size, and it has been suggested that an effective size of approximately 500 or more may be required to neutralize the effects of both drift and inbreeding (Franklin 1980; Frankel and Soule 1981; Lande and Barrowclough 1987). Demographic and genetic estimates place the effective size of the Heron Island population in the vicinity of 100; even if Heron Island is considered part of a larger panmictic Capricorn Island population, the effective size is predicted to be well below 500 (Degnan and Kikkawa in prep.). The effective size of the Lady Elliot Island population is almost certainly much smaller. The variation observed at minisatellite loci was lower among island birds compared with mainland birds, consistent with stronger drift and higher levels of inbreeding caused by the smaller island populations. Lady Elliot Island showed particularly low levels of minisatellite diversity, as predicted for a population that may recently have passed through a population size bottleneck (Nei et al. 1975). However, in spite of the observed effective population size, considered small in terms of maintaining long-term genetic variability, the island silvereye populations have apparently maintained greater minisatellite variation than fox populations on the Channel Islands. The silvereye is a species prone to active and successful colonization and may respond to reduced population size (and its consequences) in ways not representative of other species. Indeed, colonizing species may provide some clues about the ways in which populations successfully cope with being small and isolated. Further evolutionary, genetic, and ecological studies should be directed towards these questions for a wide range of colonizing taxa.
For the silvereye, significant differences in mitochondrial DNA (mtDNA) haplotype frequencies (Degnan and Moritz 1992) and single-copy nuclear DNA allele frequencies (Degnan 1993; Degnan et al. in prep.) also demonstrate differentiation between the island (Zosterops lateralis chlorocephala) and mainland (Brisbane, Zosterops lateralis familiaris) populations. Because the analysis of mtDNA permits estimations of the evolutionary relationships and divergence between observed haplotypes (e.g., see Avise et al. 1987), it provides additional information that cannot be ascertained from analysis of minisatellite DNA. In the context of little or no gene flow, the low mtDNA sequence divergence between the Heron Island population and the Brisbane population, and the continued existence of at least two founding mtDNA lineages in the island population, indicate a recent derivation of the island race from the eastern Australian mainland (Degnan and Moritz 1992). As such, the substantial morphological differentiation that exists between the two races must have occurred very rapidly in the island race, presumably because of strong selection. Because no new mtDNA haplotypes have arisen in the island race since its divergence from the mainland race (see Degnan and Moritz 1992), rapid morphological differentiation of island and mainland populations has apparently outpaced mtDNA evolution. The rate of generation of new length variants at the rapidly evolving minisatellite loci more closely reflects the rate of morphological change, because several new alleles have apparently arisen in the island populations since they have been isolated from the mainland.
Despite the successful application of DNA fingerprinting to the aims of this study, the technique is not without shortcomings (see also Hanotte et al. 1992). Most of these shortcomings pertain to the multilocus nature of the fingerprint profile that precludes the identification of specific alleles, simply because so many bands can be seen on a single gel, and because fragments of the same mobility are not necessarily isoallelic (Hill 1987). A consequent problem is that because allelism cannot be established, it is not possible to ascertain whether individuals are homozygous or heterozygous. This is important because the degree to which measures of band sharing between individuals reflects the true fraction of shared genes will depend upon the level of homozygosity (Lynch 1988). A realization of the full potential of population fingerprinting may depend more on the identification of specific hypervariable loci (e.g., Gyllensten et al. 1989), either by the use of single locus minisatellite probes or by resolving specific loci in a multilocus profile. The single locus approach would permit a more rigorous analysis of fragment patterns based on population genetics theory rather than the somewhat ambiguous measures of band sharing and has already proven very successful in breeding ecology studies (e.g., Amos et al. 1991: Kempenaers et al. 1992). The precision with which the multilocus band-sharing approach adopted in the present study reflected the evolutionary history of the populations analyzed suggests that screening hypervariable loci within and between populations is worth pursuing.
I thank J. Kikkawa for sharing unpublished data on the Heron Island silvereye population, and J. Kikkawa, C. Edwards, B. Robertson, and B. Degnan for help with field collection of sam-pies. I am very grateful to C. Moritz for providing laboratory facilities. Valuable comments on the manuscript were provided by B. Degnan, J. Kikkawa, C. Moritz, R. Zink, and especially by A. Larson, D. Westneat, and an anonymous reviewer. This research was supported by the Australian Research Council and the National Geographic Society. I am grateful for the support of a University of Queensland Postgraduate Research Award.
Amos, B., J. Barrett, and G. Dover. 1991. Breeding behaviour of pilot whales revealed by DNA fingerprinting. Heredity 67:49-55.
Avise, J. C., J. Arnold, R. M. Ball, E. Bermingham, T. Lamb, J. E. Neigel, C. A. Reeb, and N. C. Saunders. 1987. Intraspecific phylogeography: The mitochondrial DNA bridge between population genetics and systematics. Annual Review of Ecology and Systematics 18:489-522.
Burke, T., and M. W. Bruford. 1987. DNA fingerprinting in birds. Nature 327:149-152.
Chen, P., N. K. Hayward, C. Kidson, and K.A.O. Ellem. 1990. Conditions for generating well-resolved human DNA fingerprints using M13 phage DNA. Nucleic Acids Research 18:1065.
Church, G. M., and W. Gilbert. 1984. Genomic sequencing. Proceedings of the National Academy of Sciences, USA 81:1991-1995.
Degnan, S. M. 1993. The perils of single gene trees--mitochondrial versus single-copy nuclear DNA variation in white-eyes (Aves: Zosteropidae). Molecular Ecology 2:219-225.
Degnan, S. M., and C. Moritz. 1992. Phylogeography of mitochondrial DNA in two species of white-eye in Australia. Auk 109:In press.
Faulkes, C. G., D. G. Abbott, and A. L. Mellor. 1990. Investigation of genetic diversity in wild colonies of naked mole-rats (Heterocephalus glaber) by DNA fingerprinting. Journal of Zoology, London 221:87-97.
Feinberg, A. P., and B. Vogelstein. 1984. A technique for radiolabelling DNA restriction endonuclease fragments to high specific activity. Analytical Biochemistry 132:6-13.
Frankel, O. H., and M. Soule. 1981. Conservation and evolution. Cambridge University Press, Cambridge, UK.
Franklin, I. R. 1980. Evolutionary changes in small populations. Pp. 135-150 in M. Soule and B. Wilcox. eds. Conservation biology: an evolutionary-ecological perspective. Sinauer, Sunderland, Mass.
Georges. M., A. S. LeQuarre, M. Castellie, R. Hanset, and G. Vassart. 1988. DNA fingerprinting in domestic animals using four different minisatellite probes. Cytogenetics and Cell Genetics 47:127-131.
Gilbert, D. A., N. Lehman, S. J. O'Brien, and R. K. Wayne. 1990. Genetic fingerprinting reflects population differentiation in the California Channel Island fox. Nature 344:764-767.
Gyllensten, U. B., S. Jakobsson, H. Temrin, and A. C. Wilson. 1989. Nucleotide sequence and genomic organization of bird minisatellites. Nucleic Acids Research 17:2203-2214.
Hanotte, O., M. W. Bruford, and T. Burke. 1992. Multilocus DNA fingerprints in gallinaceous birds: general approach and problems. Heredity 68:481-494.
Hill, W. G. 1987. DNA fingerprints applied to animal and bird populations. Nature 327:98-99.
Hillel, J., Y. Plotzy, A. Haberfeld, U. Lavi, A. Cahaner, and A. J. Jeffreys. 1989. DNA fingerprints of poultry. Animal Genetics 20:145-155.
Hopley, D. 1982. The geomorphology of the Great Barrier Reef: Quaternary development of coral reefs. Wiley, New York.
Jarman, A. P., and R. A. Wells. 1989. Hypervariable minisatellites: recombinators or innocent bystanders? Trends in Genetics 5:367-371.
Jeffreys, A. J., R. Neumann, and V. Wilson. 1990. Repeat unit variation in minisatellites: a novel source of DNA polymorphism for studying variation and mutation by single molecule analysis. Cell 60:473-485.
Jeffreys, A. J., V. Wilson, R. Neumann, and J. Keyte. 1988. Amplification of human minisatellites by the polymerase chain reaction: towards DNA fingerprinting of single cells. Nucleic Acids Research 16:10953-10971.
Jeffreys, A. J., V. Wilson, and S. L. Thein. 1985a. Hypervariable "minisatellite" regions in human DNA. Nature 314:67-73.
-----. 1985b. Individual-specific "fingerprints" of human DNA. Nature 316:76-79.
Kempenaers, B., G. R. Verheyen, M. Van den Broeck. T. Burke, C. Van Broeckhoven, and A. A. Dhondt. 1992. Extra-pair paternity results from female preference for high-quality males in the blue tit. Nature 357:494-496.
Kikkawa, J. 1970. Birds recorded at Heron Island. Sunbird 1:34-47.
-----. 1973. The status of silvereyes Zosterops on the islands of the Great Barrier Reef. Sunbird 4:30-37.
-----. 1976. The birds of the Great Barrier Reef. Pp. 279-341 in O. A. Jones and R. Endean, eds. Biology and geology of coral reefs 3, biology 2. Academic Press, New York.
-----. 1987. Social relations and fitness in silvereyes. Pp. 253-266 in Y. Ito et al., eds. Animal societies: theories and facts. Japan Scientific Society Press, Tokyo.
Lande, R., and G. F. Barrowclough. 1987. Effective population size, genetic variation, and their use in population management. Pp. 87-123 in M. Soule, ed. Viable populations for conservation. Cambridge University Press, Cambridge, UK.
Lynch, M. 1988. Estimation of relatedness by DNA fingerprinting. Molecular Biology and Evolution 5:584-599.
-----. 1990. The similarity index and DNA fingerprinting. Molecular Biology and Evolution 7:478-484.
-----. 1991. Analysis of population genetic structure by DNA fingerprinting. Pp. 113-126 in T. Burke et al., eds. DNA fingerprinting: approaches and applications. Birkhauser, Basel, Switzerland.
Mees, G. F. 1969. A systematic review of the Indo-Australian Zosteropidae (Part III). Zoologische Verhandelingen, Leiden 102:1-390.
Nei, M., T. Maruyama, and R. Chakraborty. 1975. The bottleneck effect and genetic variability in populations. Evolution 29:1-10.
Packer, C., D. A. Gilbert, A. E. Pusey, and S. J. O'Brien. 1991. A molecular genetic analysis of kinship and cooperation in African lions. Nature 351:562-565.
Reeve, H. K., D. F. Westneat, W. A. Noon, P. W. Sherman, and C. F. Aquadro. 1990. DNA "fingerprinting" reveals high levels of inbreeding in colonies of the eusocial naked mole-rat. Proceedings of the National Academy of Sciences, USA 87:2496-2500.
Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning. A laboratory manual, 2d ed. Cold Spring Harbor Laboratory, Cold Spring Harbor, New York.
Triggs, S. J., M. J. Williams, S. J. Marshall, and G. K. Chambers, 1992. Genetic structure of blue duck (Hymenolaimus malacorhynchos) populations revealed by DNA fingerprinting. Auk 109:80-89.
Vassart, G., M. Georges, R. Monsieur, H. Brocas, A. S. LeQuarre, D. Christophe. 1987. A sequence in the M13 phage detects hypervariable minisatellite in human and animal DNA. Science 235:683-684.
Walker, T. 1986. The birds of Lady Elliot Island. Sunbird 16:73-82.
Wayne, R. K., N. Lehman, D. Girman, P.J.P. Gogan. D. A. Gilbert, K. Hansen, R. O. Peterson, U. S. Seal, A. Eisenhawer, L. D. Mech, and R. J. Krumenaker. 1991. Conservation genetics of the endangered Isle Royale gray wolf. Conservation Biology 5:41-51.
Westneat, D. F. 1990. Genetic parentage in the indigo bunting: a study using DNA fingerprinting. Behavioral Ecology and Sociobiology 27:67-76.
Westneat, D. F., W. A. Noon, H. K. Reeve, and C. F. Aquadro. 1988. Improved hybridization conditions for DNA "fingerprints" probed with M13. Nucleic Acids Research 16:4161.
Wetton, J. H., E. C. Royston, D. T. Parkin, and D. Walters. 1987. Demographic study of a wild house sparrow population by DNA fingerprinting. Nature 327:147-149.
Wong, Z., V. Wilson, I. Patel, S. Povey, and A. J. Jeffreys. 1986. Cloning a selected fragment from a human DNA "fingerprint": isolation of an extremely polymorphic minisatellite. Nucleic Acids Research 14:4605-4616.
Wright, S. 1922. Coefficients of inbreeding and relationship. American Naturalist 56:330-338.
-----. 1969. Evolution and the genetics of natural populations. IV. The theory of gene frequencies. University of Chicago Press, Chicago.
|Printer friendly Cite/link Email Feedback|
|Author:||Degnan, Sandie M.|
|Date:||Aug 1, 1993|
|Previous Article:||The effects of population size and plant density on outcrossing rates in locally endangered Salvia pratensis.|
|Next Article:||Mitochondrial gene genealogy and gene flow among island and mainland populations of a sedentary songbird, the grey-crowned babbler (Pomatostomus...|