The level of genetic diversity within the Chittering populations is high for ... the level and structure of genetic variation and the genetic consequences of different.
Aust. J. Bot., 1988, 36, 273-86
Genetic Diversity and Population Genetic Structure in the Rare Chittering Grass Wattle, Acacia anomala Court
David J. Coates Department of Conservation and Land Management, Western Australian Wildlife Research Centre, P.O. Box 5 1, Wanneroo, W .A. 6065.
Abstract There are 10 known populations of Acacia anomala occurring in two small disjunct groups some 30 km apart. The Chittering populations reproduce sexually whereas the Kalamunda populations appear to reproduce almost exclusively by vegetative multiplication. The level and distribution of genetic variation were studied at 15 allozyme loci. Two loci were monomorphic in all populations. In the Chittering populations the mean number of alleles per locus was 2.0 and the expected panmictic heterozygosity (genetic diversity) 0.209. In the Kalamunda populations the mean number of alleles per locus was 1.15 and the expected panmictic heterozygosity 0.079, although the observed heterozygosity of 0.150 was only marginally less than the Chittering populations (0.177). These data support the contention that the Chittering populations are primarily outcrossing whereas the Kalamunda populations are clonal, with each population consisting of individuals with identical and, in three of the four populations, heterozygous, multilocus genotypes. The level of genetic diversity within the Chittering populations is high for plants in general even though most populations are relatively smsll and isolated. It is proposed that either the length of time these populations have been reduced in size and isolated is insufficient for genetic diversity to be reduced or the genetic system of this species is adapted to small population conditions. Strategies for the adequate conservation of the genetic resources of Acacia anomala are discussed.
Introduction The level and partitioning of genetic variation within and among plant populations are known to be influenced by a wide range of factors such as mode of reproduction, mating system and geographic distribution. Species which have wide geographic ranges, long generation times, are wind-pollinated and outcrossing, such as many of the conifers, tend to have very high levels of variation, most of which occur within populations. In contrast, annual herbaceous species which are primarily selfing have lower levels of genetic variation which occur mostly between populations (Brown 1979; Hamrick et al. 1979; Hamrick 1983; Loveless and Hamrick 1984). The mating system has a major influence on the structuring of genetic variation within a species (Gottlieb 1973, 1975; Brown et al. 1975; Levin 1975, 1978; Schaal 1975; Phillips and Brown 1977; Brown and Jain 1979; Moran and Brown 1980; and others). However, much less is known about the effects of other factors, in particular population size, the degree of population isolation and geographic distribution. Population genetic theory predicts that large populations maintain higher levels of genetic variability than small populations and that the greater the isolation between populations the greater the level of between-population variation. Recent studies provide contradictory evidence. Moran and Hopper (1987), in a study on localised mallee eucalypts, found that small populations have the same level of genetic diversity as larger~populations within the same species. However, localised species were found to have generally lower levels of genetic variability than widespread species (see also Hamrick 1983). The 0067-1924/ 88/030273$02.00
D. J. Coates
effects of factors such as population size and population isolation on the level and distribution of genetic diversity within population systems are of particular interest because of implications relating to the conservation of genetic resources and the survival of rare and threatened species which are generally characterised by small isolated populations. The Chittering grass wattle, Acacia anomala Court, is a naturally rare and extremely localised species occurring in two small disjunct groups of populations some 30 km apart. It is not obviously related to any other species within the genus (Court 1978; B. Maslin, personal communication.) The northern populations appear to reproduce sexually, although seed yields are rather low (B. Dell, personal communication; B. Maslin, personal communication), while the southern populations seem to reproduce solely by vegetative means (root suckering), with no fruit production or seed set observed (D. Marshall, personal communication). This study was initiated to investigate the level and structure of genetic variation and the genetic consequences of different reproductive modes in the two isolated population groups. The investigation was also aimed at determining strategies for the conservation of genetic resources of this rare species.
Materials and Methods Population Sampling Acacia anomala is known from 10 populations confined to the edge of the Darling Scarp just outside metropolitan Perth, Western Australia (Fig. 1). All populations occur in Eucalyptus marginata, E. accedens, E. calophylla woodland on deep lateritic soils. Six of these populations occur in the Chittering area north-east of Perth, the other four are 30 km further south near Kalamunda. At the time of collecting material (May-June 1986) for electrophoretic studies only an estimate was made of the numbers of individuals in each population by attempting to count all plants in the area, because it was sometimes difficult to determine the number of individual plants in a clump and be certain all plants were counted in areas of thick scrub. The Chittering populations are in an area covering 20 km2 and all appear to have been substantially reduced in size in the last 20-30 years due to land clearing and disturbance. The Kalamunda populations occur in a much smaller area (1-2 km2), although they still appear to exist as geographically discrete groups of plants. Based on the size of the Chittering populations in both number of plants and area covered each group of plants in the Kalamunda area was considered a population. Thus, although in this study Kalamunda populations 1 and 2 are treated as discrete populations only 50 m apart, it is also feasible to consider them as part of the same population. A single young inflorescence was collected from individual plants in each population and stored at 4°C for 1-2 days prior to electrophoretic assays for isozymes. Following flowering (Aug.-Sept.), observations were made of seed production and seedlings in both groups of populations. Electrophoresis Whole inflorescences were ground in Eppendorf tubes with a stainless steel rod. A modification of the grinding buffer of Systma and Schaal (1985) gave the best results (50 mg ml- PVP, 0.8 mM NAD, 0.4 mM NADP, 1.O mM EDTA, 1 - 0 mM ascorbic acid, 0.1% BSA w /v, 10% sucrose w / v in doubledistilled H20; equilibrate to pH 6.8 with 0.5 m Tris and add 1.5 mg ml-' dithiothreitol). The isozyme methods used were based on the Helena Laboratories cellulose acetate plate electrophoresis system. The system includes plates in which are set a series of sample wells and applicators. The ground samples were transferred to the sample wells and then directly loaded onto the cellulose acetate plates with the applicator. Before loading they were soaked in the running buffer (10-15 min). An 80 mM Tris-EDTA-maleic acid buffer pH 8.2 (80 mM Tris, 1 mM Na2 EDTA, 1 mM MgC1 (2.44 mM maleic acid)) gave the most consistent isozyme resolution for all enzyme systems (for other running buffers see Richardson et al. 1986). After loading, the plates were placed face down in the electrophoresis tank so that each end rested on a blotting paper wick. The time of an electrophoresis run for this buffer was about 30 min. After electrophoresis 4 ml of 3.5% Agar was added to 4 ml of the reaction mixture for the stain and the solution poured over the plate. Staining usually took 5-10 min at room temperature, although up to 1 h was required for some systems. When the staining reaction
'
Genetic Diversity in the Grass Wattle
was complete the agar overlay was washed off the plate which was then soaked in 7% acetic acid for 15 min, washed in water and air-dried. Although the intensity of bands fades with drying, they can easily be observed and photographed on a light box. The staining procedures were similar to those described by Richardson et al. (1986) except 4 ml of stain buffer was used and there was a proportional increase in the quantity of all stain ingredients.
Fig. 1. Geographical distribution of Acacia anomala on the edge of the Darling Scarp near Perth, Western Australia. Eleven enzyme systems were assayed in this way: alcohol dehydrogenase (ADH; E.C. 1.1.1. I), esterase (EST; E.C.3.1.1. I), glutamate dehydrogenase (GDH; E.C. 1.4.1.3), glucose-phosphate isomerase (GPI; E.C.5.3.1.9), isocitrate dehydrogenase (IDH; E.C. 1.1.1.42), leucine aminopeptidase (LAP; E.C.3.4.17.1), malate dehydrogenase (MDH; E.C.1.1.1.37), malic enzyme (ME; E.C.1.1.1.40), menadione reductase (MDR; E.C.1.6.99.22), 6-phosphogluconate dehydrogenase (6PGD; E.C. 1.1.1.44), phosphoglucomutase (PGM; E.C.2.7.5. 1). Menadione reductase was stained as described by Moran and Hopper (1983). In total, 15 zones of activity were scored and each zone was assumed to represent a single enzyme locus. Loci were designated numerically beginning with the most anodal zone. Alleles at each locus were designated alphabetically with the most anodal allele being 'a' and all others lettered in order of decreasing electrophoretic mobility.
D. J. Coates
8
1
8
0 0 0 0
oo-oo
1 1
o-oo
1 1
0 - 0 0
0 0 -
- 0 0
Q
Q
Q
3 0 0 0 0
- 0 0 0 0
a
.-
-a Y
a m a
w
8
S
5"
+i E
s
z
1
o-oo
p'
- N
00 0 -
0 0 0 0
O Q - 0 0
0 0 -
0 0 0 0
"
0 0 - 0 0
228
wwp'
0 0 6 0
0 0 0 0 0
8 '?
8
0
Q
0
5 5
0
~
Q Q
- 0 0
1 1
Q
- 0 0
1 1
0 0 -
- 0 0
Q
5
0 - 0 0
8
Q
0 - 0 0
8 "
0 0 0 0 0
'?
"
8
0 0 0 0 0
.I?
e
T
' I
L3
*
.-
P
-
-
22F
1
0 0 -
1
0 - 0
w 3
g
0 0 0 0
2m m 2m 3 0 0 0 0 0
Genetic Diversity in the Grass Wattle
0 0
11
0 0 -
8
88
2200
Q
O A O O
Q
8
?
?
0 0 0
O A O O
88
??
0 0 0
Q
O A O O
w m w
Q
A 0 0
F?
0 0 0 0
-
m m Pw m
0 0 0
m Pb m X
0
fm
3 O
0 0
P-
s
0 0 0 0
bP-5
m N
?So
$2 5 0 0 0
0 0 0 0
2 2 00-
w m -
0 0 0
0 0 0 0
S
?FS
-3 ? N
z o o
4mv
- a 0
m\D-
"?*
0 0 0 0
4epUfl
D. J. Coates
Genetic Analyses To provide a measure of the level of genetic variation within populations the following statistics were computed: A , the mean number of alleles per locus; P, the proportion of loci that is polymorphic; H,, the observed heterozygosity (averaged over all loci); He, the expected panmictic heterozygosity. The expected panmictic heterozygosity ( H e ) or gene diversity index (Nei 1973) was calculated as 2
He= 1 - C x,, i= 1 where x, is the frequency of the ith allele summed over k alleles. The expected panmictic heterozygosity was then averaged over all loci for each population. Genetic diversity can be considered to have two components, allelic richness and allelic evenness (Brown and Weir 1983). The mean number of alleles per locus A is a measure of allelic richness whereas He is a useful measure of allelic evenness. To further investigate allelic richness within the Chittering and Kalamunda populations the number of allelic variants in four different classes was determined (see Marshall and Brown 1975; Brown 1978; Moran and Hopper 1987). These classes are: (1) common, occurring in at least one population with a frequency > 10%. (2) rare, not occurring in any population with a frequency > 10%. These two categories can then be subdivided into widespread (W), when they occur in two or more populations, or localised (L), when they are in only one population. Wright's fixation index ( F ) which measures the deviation of heterozygote frequencies from Hardy-Weinberg proportions was estimated for each of the Chittering population as F = 1 - Ho/ He . F can range from - 1 to 1 with positive values indicating a deficiency of heterozygotes compared to Hardy-Weinberg expectations. Significance of the deviations from Hardy-Weinberg proportions at each locus was determined by a X 2 test calculated using Levene's (1949) modification for small sample sizes. The partitioning of genetic diversity within and among populations was analysed by using measures proposed by Nei (1973, 1975). Ht the total gene diversity of a group of populations can be defined as Ht = Hs + DSt, where H, is the mean gene diversity within populations and DSt is the mean gene diversity between populations. Ht can be estimated as 2
H t = 1 - E X,, i= 1 where 2 is the mean frequency of the ith allele at a polymorphic locus. Nei (1973) also defines an absolute measure of gene differentiation, Dm,independent of genetic diversity within populations, which is a measure of the net codon differences between populations. The estimate of Dmis D, = s DSt! s - 1 , where s is the number of subpopulations sampled. In addition, one can estimate the proportion of interpopulation differentiation, Gst, as the ratio Gst = Dst/Ht . Nei's (1972) genetic distance (D) was calculated for each pairwise combination of populations at all 15 loci. All statistics except Nei's diversity measures Ht, Hs, Dst, Dm and GStwere determined by using the computer program BIOSYS-1 (Swofford and Selander 1981).
Results Estimates of population sizes ranged from three for Chittering population 3 to 50 for Chittering population 5 and Kalamunda population 4 (Table 3). All populations appeared to have been affected to some extent by land clearing and/or disturbance. Seed set was observed in individuals from all populations in the Chittering area except population 3. Seedlings were observed in populations 1, 2 and 5. No seed set was observed in any individuals in the Kalamunda area. Genetic Variation Within Populations The allelic frequencies at 15 loci for the 10 Acacia anomala populations are given in Table 1. Two loci (Adh-1 and Me-1) are monomorphic in all populations. In the Chittering populations allelic variation exhibits a segregation pattern typical of sexually reproducing individuals whereas each Kalamunda population consists of individuals with identical multilocus genotypes. Three of the four multilocus genotypes in the
Genetic Diversity in the Grass Wattle
Kalamunda materials are characterised by a number of fixed heterozygous loci. For instance, in population 1 all individuals are heterozygous for the same alleles at Pgrn1, Lap-1 and Gdh-1. These data clearly indicate that the Kalamunda populations are clones of genetically identical individuals. The Kalamunda populations are also characterised by alleles at four loci (Pgm-1 , Pgi-1, Lap-1 and Gdh-1), which are either absent or occur at very low frequencies within the Chittering populations. Chittering population 5 is of interest because of the presence of unique rare alleles at Pgm-1, Pgi-1, Est-2 and Mdh- 1. The partitioning of allelic variants into four distributional classes (Table 2) indicates that over 65% of the allelic variants in the Chittering populations are common and widespread while only one allelic variant is common and localised. The remainder are rare with 17 alleles widespread and six localised. In contrast, because of their clonal nature the Kalamunda populations have only common alleles of which over 70% are widespread. Table 2. Average number of alleles and their distribution in the two I . groups of Acacia anomala populations
F
.
A, mean No. of alleles per locus; W, widespread; L, localised Locality
Chittering Kalamunda
Number of: Populations Loci
15 15
6 4
