Mating system analysis in a natural population of Acacia nilotica subspecies kraussiana

Forest Ecology and Management ELSEVIER

Forest Ecology and Management 79 (1995) 235-240

Short C o m m u n i c a t i o n

Mating system analysis in a natural population of Acacia niIotica subspecies kraussiana A.K. Mandal a,*, R.A. Ennos b " Tropical Forest Research Institute, P.O.R.F.R.C.. Jabalpur-482 021, India ~ Institute of Ecology and Resource Management. University of Edinburgh, Edinburgh EH9 3JU, UK

Accepted 8 February 1995

Abstract

The mating system of Acacia nilotica ssp. kraussiana, a naturally occurring autotetraploid species, was analysed. Open pollinated seeds germinated for 6 days were used for starch gel electrophoresis. Allozyme banding patterns were found to be in conformity with tetrasomic segregation. Multilocus estimate of the proportion of viable progeny due to outcrossing (t m) was 0.983. Estimates of outcrossing rate of families were heterogeneous and ranged from 0.08 to 1.99. Results have been discussed in the light of the strategy to be followed for seed collection for breeding and conservation purposes. Keywords: lsozyme; Electrophoresis; Mating system

1. Introduction

The genetic structure of forest trees and the pattern of genetic variation that are maintained in the successive generations, are essentially a function of the mating systems. Thus, plant mating systems play an important role in shaping the genetic composition of progeny generations. Most forest tree improvement programmes use open pollinated seeds in preference to controlled cross seeds for raising plantations, estimating genetic parameters, and predicting genetic gain. In using open pollinated seeds, random mating and absence of inbreeding are generally assumed. However, theoretical expectations are seldom upheld in nature. Many studies, including those of Shaw and Allard (1982),

* Corresponding author.

Moran and Bell (1983), Brown et al. (1985), Cheliak et al. (1985), Perry and Knowles (1990), Murawski and Hamrick (1991) and Muona et al. (1991), have shown significant levels of selfed progeny and variable levels of outcrossing rates. The practical significance of such findings is that inbreeding depression due to selfing decreases survival and growth of plantations, and biases the estimates of genetic parameters. Thus, the design of an efficient tree breeding strategy will be greatly influenced by our knowledge and thorough understanding of the mating systems. The use of electrophoretic marker loci in lieu of morphological markers and the development of suitable statistical genetic models have facilitated the detailed quantitative estimation of plant mating systems. To date, most of the studies have been made with diploid coniferous species, probably in view of their inherent simplicity. However, tropical tree species, particularly those of higher ploidy level have

0378-1127/95/$09.50 © 1995 Elsevier Science B.V. All rights reserved SSD10378-1t27(95)03542 7

236

A.K. Mandal, R.A. Ennos / Forest Ecology and Management 79 (1995) 235-240

not been subjected to such studies. In recent years, the domestication of a number of tropical tree species, mainly belonging to the genus Acacia has been initiated. For most of these, tree improvement programmes are in their infancy, probably owing to a lack of basic knowledge about the species, including mating system and genetic structure. However, recent interest in the study of polyploid and development of appropriate methodology for analysis of mating systems in polyploid (Barrett and Shore, 1987; Ritland, 1990) has opened the way for quantitative studies of the mating system in previously intractable, though economically important species. One genus in which such advances are of great importance is Acacia. Relatively little is known about their reproductive biology and mating systems. So far, only five Australian diploid Acacia species (Acacia decurrens, Philp and Sherry, 1946; Acacia mearnsii, Moffet, 1956; Acacia auriculiformis. Acacia crassicarpa, Moran et al., 1989; Acacia melanoxylon, Muona et al., 1991), studied for mating system pattern, indicate predominant outcrossing nature. We report here an analysis of the mating system in a natural population of Acacia nilotica spp. kraussiana using electrophoretic marker loci.

2. Materials and m e t h o d s

2.1. Study organism Acacia nilotica is a naturally occurring tetraploid species (2n = 4x = 52). It is one of the highly diversified species in the genus, and has evolved at least nine subspecies fitting its ecologically diversified areas in Africa and Asia. The members of the species have bright yellow flowers and appear to promote outcrossing through andromonoecy. Flowering may occur a number of times in a year, as the flowers are produced on current season growth which is dependent on moisture availability. High flower abortion results in low pod set. Floral dimorphism and low pod/flower ratios are common adaptive mechanisms which enhance pollinator activity at minimal energy investment towards seed production (Ross, 1979). Acacia pollen is distributed as composite units (polyads), 16 being the most common number. The perfect correlation between the size of the polyad

and the shape and diameter of the stigma leads to a one polyad-one stigma relationship. This is probably a functional specialisation for speciation, which limits improper pollen transfer (Guinet, 1986). Bees of the family Megachilidae and Anthophoridae are the primary pollinators, but other insects and birds may also be involved. Self-incompatibility is believed to be the principal outbreeding mechanism in Australian species (Kenrick and Knox, 1989). Outcrossing rates of around 90% have been recorded for such species (Moran et al., 1989; Muona et al., 1991). None of the Afro-Asian species have been subjected to investigation on the above noted areas. 2.2. Seed and seed collection site Open pollinated seeds from 25 individual trees (family) of Acacia nilotica ssp. kraussiana were obtained from the Oxford Forestry Institute for use in the present study. Seeds were collected from a localised population in Chantulo area of Mangochi district in Malawi (14°19'S, 38°48'E, altitude 490 m) during June 1992. The population extends north and south of the Salima to Monkey Bay road, about 2 km before Chantulo settlement, and 26 km before the Monkey Bay T junction. Individual trees were scattered and far apart (50-100 m) from each other. The seed analysed from each tree comprised the contents of 300-500 pods that had been bulked. The soils in the area are dark grey deep cracking clays or gleys. The area is liable to flooding and has the elements of savanna woodland with Adansonia / Cordyia species. The area is subject to fires and is grazed by livestock. 2.3. Electrophoresis Electrophoretic analysis was carried out using 20 open pollinated seeds each from 25 individual families. Seeds were scarified to promote germination. Six-day-old germinated seeds were ground over ice in extraction buffer (KH2 P O 4 0.136 g, cysteine 0.36 g, EDTA 0.336 g, sucrose 13.6 g, Tris 2.4 g, sodium citrate 0.38 g, H 2 0 100 ml, pH 7.0) and absorbed onto filter paper wicks for use. Two enzyme systems, esterase (EST) and malate dehydrogenase (MDH) were assayed. Esterase was separated on 11% starch gel (3 h at 40 mA) in 0.0014 M EDTA,

A.K. Mandal, R.A. Ennos / forest Ecology and Management 79 (l 995) 235 240

0.05 m DL-histidine HC1 gel buffer (pH 7.0), and 0.125 M Tris electrode buffer (pH 7.0) systems. Malate dehydrogenase was separated on 11% gel (5 h at 40 mA) in 0.0045 M DL-histidine HCI, 0.0037 M NaOH gel buffer (pH 6.5), and 0.041 M sodium citrate, 0.0003 M citric acid electrode buffer (pH 6.5) systems. Four polymorphic loci were scored and locus designations are Est-2, Mdh-1, Mdh-2, and Mdh-3. The position of the bands on the electrophoresis gel and the relative staining intensities of the enzyme bands were noted. This was essential for assigning putative genotypes to the progeny that had been scored in this tetraploid species. Est-1 was found to be monomorphic. Enzyme systems were stained as follows. Est, fast garnet GBC salt 100 mg, or- and ]3-napthyl acetate 30 mg each, acetone 3 ml, H 2 0 87 ml, and 0.5 M Tris-HCl buffer (pH 7.1) 10 ml; MDH, Nicolinamide adenine dinucleotide (NAD) 50 mg, 3 [4,5-Dimethylthiazol 2-Y1]-2,5 diphenyl tetrazolium bromide (MT) 30 mg, Phenazine methosulphate (PMS) 2 mg, malic substrate solution 10 ml, 0.1 M Tris-HCl buffer (pH 7.0) 10 ml, and H 2 0 80 ml.

2.4. Mating system analysis Mating system analysis was done on the assumption that A. nilotica is tetraploid and shows tetra

(a) m

aaaa

bbbb

aabb

aaab

abbb

(b) - -

D

aaaa

bbbb

cccc

aabb

aaab

abbb

E

aabc

Table 1 Allelic frequencies in Acacia nilotica ssp. kraussiana Locus

Allele

Frequency

Mdh-1

1 2

0.614 0.386

Mdh 2

l 2 3

0.592 0.368 0.045

Mdh 3

1 2 3

0.570 0.310 0.142

Est 2

1 2

0,578 0.422

somic inheritance. Such a mode of inheritance is expected if the species has autotetraploid rather than allotetraploid origin. The assumption of tetrasomic inheritance is consistent with the lack of 'fixed' heterozygotes, and the variation in gene dosage levels, detected at isozyme loci in the study. Quantitative analysis of the mating system made use of the MLTET programme (Ritland, 1990; Murawski et al., 1994), especially written for the analysis of mating system in autotetraploids. Data on putative progeny genotype arrays from the 25 parent trees were used both to infer maternal genotypes, and to jointly estimate pollen allele frequencies and outcrossing parameters. Estimates of single locus outcrossing rate (q), multilocus outcrossing ( t m) a n d multilocus family estimates of outcrossing rate (tm~) were determined. The method makes normal assumptions of mixed mating model, but the programme is modified to account for tetrasomic inheritance. Variance estimates were obtained by conducting 200 bootstraps. Owing to the low number of families assayed (25), bootstrap resampling was performed

Table 2 Estimates of single (t~) and multilocus (t m) outcrossing rate in Acacia nilotica ssp. kraussiana

i

w

237

m

abcc

- -

- -

aacc aaac

i

accc

Fig, 1. Observed variation in isozyme banding pattern for Est-2 amd Mdh-1 (a), and Mdh-2 and Mdh-3 (h), Also shown are the inferred genotypes at these loci.

Locus

Outcrossing rate (t)

SE (t)

Mdh I Mdh-2 Mdh 3 Est-2 Mean single locus Multilocus

1.990 1.990 1.472 0.556 0.971 0.983

0.018 0.002 0.297 0.168 0.145 0.158

238

A.K. Mandal. R.A. Ennos / Forest Ecology and Management 79 (1995) 235-240

Table 3 Estimates of multilocus family (t,,,,) outcrossing rate in Acacia nilotica ssp. kraussiana Family Outcrossing SE(t,.,) rate (t,,,)

Family Outcrossing SE(tm,) rate (tm,)

1 2 3 4 5 6 7 8 9 10 11 12 13

1~ 15 16 17 18 19 20 21 22 23 24 25

1.12 1.29 1.52 1.87 1.42 1.09 1.35 0.32 0.67 1.70 1.14 0.62 0.49

0.33 0.25 0.38 0.25 0.33 0.41 0.30 0.31 0.16 0.22 0.37 0.53 0.31

0.80 1.39 1.60 1.48 1.99 1.20 1.99 0.68 (I.37 0.32 0.93 0.61

0.48 0.28 0.32 0.18 0.09 0.33 0.00 0.17 0.08 0.16 0.26 0.30

within families, The mathematical formulae and their derivation are available in Murawski et al. (1994).

3. Results

Both enzyme systems studied showed a wide variety of banding phenotypes (Fig. 1). Table 1 shows allelic frequencies and Table 2 shows single locus outcrossing estimates (t S) and their standard errors for the Mdh-1, Mdh-2, Mdh-3 and Est-2 loci, mean single locus outcrossing rate (ts), and the multilocus outcrossing estimate (fro) that uses information from all four loci. Single locus outcrossing estimates ranged from 0.556 for Est-2 to 1.990 for Mdh-1 and Mdh-2 and are significantly heterogeneous. The mean single locus outcrossing rate, t s = 0.971 and the multilocus outcrossing rate, t m = 0.983, are similar. Table 3 shows the estimated multilocus family outcrossing r a t e s (tmi) for the 25 trees. These values were calculated on the assumption that pollen allele frequencies are the same for each maternal parent. The values obtained ranged from 0.08 to 1.99 and are statistically significantly heterogeneous.

4. Discussion

Although polypioids constitute a major portion of the world's flora, very' few quantitative studies of mating systems have been made on these species in

comparison with their diploid counterparts (Barrett and Shore, 1987; Murawski et al., 1994). This is largely a reflection of the technical difficulties associated with the analysis of polyploids, difficulties that are well illustrated in the present study. The first problem is in finding suitable genetic markers. Isozyme banding patterns in polyploids are extremely complex, making genetic interpretation of the variation difficult (Ness et al., 1989). In this case the bands were distinguished not only by the position of bands, but also by the relative staining intensity of the bands. Differences in staining intensity were interpreted as differences in allelic dosage in this tetraploid species, where, within an individual, a particular allele may be present in one to four copies (Fig. 1). Apart from the two enzyme systems reported here, two more systems (Alcohol dehydrogenase (ADH) and Phosphoglucomutase (PGM)) were studied but because of poor resolution and inconsistency in banding pattern, they were not included in the present analysis. The second problem is knowing with confidence the mode of inheritance of genetic variants. The general rule that allopolyploids display regular disomic and autopolyploids exhibit polysomic inheritance patterns often met with difficulties, especially in cases where hybridisation takes place between closely related species, or where pairing behaviour of chromosomes is modified subsequent to the formation of the polyploid (Soltis and Rieseberg, 1986). In the present case, the exact nature of the polyploid origin of A. nilotica ssp. kraussiana is not known. Analysis has been made on the assumption that the population is showing tetrasomic inheritance of isozyme variants. This assumption is based on the arguments that if A. nilotica ssp. kraussiana was an allotetraploid, 'fixed' heterozygosity would be detected, and the frequency of segregating allelic variants would be low. Isozyme analysis revealed, however no 'fixed' heterozygosity, and up to three alle]es were found to be segregating at substantial frequencies at the loci scored in the population. Thus, the study indicated that A. nilotica ssp. kraussiana is unlikely to be an allopolyploid, showing digenicdisomic inheritance. The final difficulty with the analysis of mating systems in polyploids concerns the application of a model for outcrossing estimation (Ritland, 1990; Murawski et al., 1994). These models require com-

A.K. Mandal. R.A. Ennos /Forest Ecology and Management 79 (1995) 235-240

paratively more data than similar models for diploids. This is particularly so where maternal genotypes are unknown, and a large number of progeny needs to be scored to allow these genotypes to be inferred with precision. Finally, the models themselves are necessarily complex, involving far more parameters than diploid models, reducing the strength of the conclusions that can be drawn from them. Acknowledging these difficulties and shortcomings, the present study gives a best estimate of outcrossing rate of t m : 0.983 which indicates that the species has a high degree of outbreeding. The values are comparable to those reported for other diploid acacia species (Moran et al., 1989; Muona et al., 1991) as well as for other tropical tree species (Murawski and Hamrick, 1991). All Australian diploid Acacia species investigated to date have been found to be self-incompatible (Kenrick and Knox, 1989). None of the Afro-Asian acacias have been studied for compatibility reaction but the presence of such a mechanism is not ruled out. The high outcrossing rate in A. nilotica ssp. kraussiana is a clear indication in that direction. Though most of the Acacia species have been found to be self-incompatible, Kenrick (1986) noted that there may be a few species which are substantially self-compatible. Thus, the potential for variation in the mating may be much higher between species/subspecies. Such is the case with Acacia nilotica ssp. leiocarpa which has been found to be a partial outbreeder, with an outcrossing estimate of 0.384 (Mandal et al., 1994). In the present study, single locus estimates varied over loci. Variation in outcrossing rate is a well known consequence of single locus estimation. It is widely recorded among outcrossed tree species, including conifers (Furnier and Adams, 1986) and tropical angiosperms (Moran et al., 1989; Mandal et al., 1994). Theoretically, outcrossing rates must be same for all loci as alleles at all loci are transmitted in the same gametes. However, variation may result from nonrandom mating, including any form of inbreeding. Crow (1986) suggested that assortative mating and inbreeding which are associated with high genetic variances can exaggerate single locus variability. The heterogeneity of estimates over loci may not necessarily show that single loci estimation, or mixed mating model is unsatisfactory (Brown et al., 1985).

239

Significantly higher outcrossing estimates above the natural biological range (t~ > 1.0) were noted for three of four loci studied. Similar estimates have been recorded earlier by Furnier and Adams (1986) and Moran et al. (1989). Such a problem arises owing to (i) sampling effects within the context of a valid mixed mating model, and (ii) non-assortative mating (Brown et al., 1985). Significant variation in the family outcrossing rate indicates that the allelic frequencies in outcross pollen received by different trees are significantly different. Such variation was earlier recorded by Olng'otie (1991) in another autotetraploid species, Acacia tortilis. The genetic mechanisms of the mating system in the genus Acacia are largely unknown. The results of this study indicate that Acacia nilotica ssp. kraussiana is a predominantly outcrossing species. The high outcrossing rate, and lack of significant levels of inbreeding appear to be the main components of the adaptive strategy followed by the species. Outcrossing is a mechanism for maintaining a high level of genetic variability, which presumably is important for survival in a forest environment that is spatially and temporally heterogeneous (Furnier and Adams, 1986). Since there is variation in the outcrossing rate in different subspecies of A. nilotica, a separate estimate should be obtained for other subspecies. A seed collection strategy by sampling a large number of pods all over the crown, which will result in a genetically more diverse sample, should be followed for this predominantly outcrossing subspecies. For a genetic conservation programme, sampling more populations with a relatively small number of trees in each site is suggested. Since the study was based on a limited sample size and quantification of the mating system was done only with two enzyme systems, a more detailed investigation by combining field and laboratory studies using more populations and enzyme systems is recommended.

Acknowledgements We thank Dr. R.D. Barnes and C.W. Fagg for supply of seeds. We also thank Dr. Kermit Ritland for the computer programme. The senior author is grateful to the British Council for providing a fellowship during the course of this study.

240

A.K. Mandal, R.A. Ennos / Forest Ecology and Management 79 (1995) 235-240

References Barrett, S.C.H. and Shore, J.S., 1987. Variation and evolution of breeding, systems in the Turnera ulmifolia L. complex. Evolution, 41: 340-354. Brown, A.H.D., Barrett, S.C.H. and Moran, G.F., 1985. Mating system estimation in forest trees: models, methods and mean ings. In: H.R. Gregorius (Editor), Population Genetics in Forestry. Springer, Berlin. Cheliak, W.M., Danick, B.P., Morgan, K., Yeh, F.C.H. and Strobeck, C., 1985. Temporal variation of the mating system in a natural population of Jack pine. Genetics, 109: 569-584. Crow, J.F., 1986. Basic concepts in Population, Quantitative and Evolutionary Genetics. Freeman, New York. Fumier, G.R. and Adams, W.T., 1986. Mating system in natural populations of Jefferey pine. Am. J. Bot., 73: 1002-1008. Guinet, P.H., 1986. Geographic patterns of the main pollen characters in the genus Acacia with particular reference to subgenus Phyllodineae. In: S. Blackmore and I.K. Fergusan (Editors), Pollen and Spores Form and Function. Academic Press, London. Kenrick, J., 1986. A method for estimating self incompatibility. In: E.G. Williams and R.B. Knox (Editors), Pollination' 86. University of Melbourne, Melbourne. Kenrick, J. and Knox, R.B., 1989. Quantitative analysis of self incompatibility in trees of seven species of Acacia. J. Herod., 80: 240-245. Mandal, A.K., Ennos, R.A. and Fagg, C.W., 1994. Mating system analysis in a natural population of Acacia nilotica subspecies Leiocarpa. Theor. Appl. Genet., 89: 931-935. Moffett, A.A., 1956. Genetical studies in Asasias. 1. The estimation of natural crossing in black wattle. Heredity, 10:57 67. Moran, G.F. and Bell, J.C., 1983. Eucalyptus. In: S.P. Tanksley

and T.J. Orton (Editor), lsozymes in Plant Genetics and Breeding, Part B. Elsevier, Amsterdam. Moran, G.F., Muona, O. and Bell, J.C., 1989. Breeding systems and genetic diversity in Acacia auriculiformis and A. crassicarpa. Biotropica, 21: 250-256. Muona, O., Moran, G.F. and Bell, J.C., 1991. Hierarchical patterns of correlated mating in Acacia melanoxylon. Genetics, 127:619 626. Murawski, D.A. and Hamrick, J.L., 1991. The effect of the density of flowering individuals on the mating systems of nine tropical tree species. Heredity, 67: 197-203. Murawski, D.A., Fleming, T.H., Ritland, K. and Hamrick, J.L., 1994. Mating system of Pachycereus pringlei an autotetraploid cactus. Heredity, 72: 86-94. Ness, B.D., Soltis, D.E. and Soltis, P.S., 1989. Autopolyploidy in Heuchera micrantha. Am. J. Bot., 76: 614-626. Olng'otie, P.A.S., 1991. Acacia tortilis: a study of genetic structure and breeding systems. Ph.P. Thesis, Oxford University. Perry, D.J. and Knowles, P., 1990. Evidence of high self fertilization in natural populations of eastern white ceder. Can. J. Bot., 68: 663-859. Philp, J. and Sherry, S.P., 1946. The degree of natural crossing in A. decurrens and its bearing in wattle breeding. J.S. Afr. For. Assoc., 17: 6-58. Ritland, K., 1990. A series of FORTRAN computer programs for estimating plant mating systems. J. Herod., 81: 235-237. Ross, J.H., 1979. A conceptus of the African Acacia species. Mem. Bot. Survey of South Africa, No. 44. Shaw, D.V. and Allard, R.W., 1982. Estimation of outcrossing rates in Douglas fir using isozyme markers. Theor. Appl. Genet., 62: 113-120. Soltis, D.E. and Rieseberg, L.H., 1986. Autopolyploidy in Tolmiea menziesii: Genetic insights from enzyme electrophoresis. Am. J. Bot., 73: 310-318.