Ausgabe 59/6
Total Page:16
File Type:pdf, Size:1020Kb
Spain et. al.·Silvae Genetica (2011) 60-6, 241-249 Genetic consequences of subtropical rainforest fragmentation on Macadamia tetraphylla (Proteaceae) By C. S. SPAIN1),2) and A. J. LOWE3),*) (Received 4th December 2010) Abstract is now a ubiquitous feature of much of the planet’s Habitat fragmentation can bring about a variety of forested areas (YOUNG and MITCHELL, 1994; LAURANCE gene-flow alterations in plant populations, potentially and BIERREGAARD, 1997; HOBBS and YATES, 2003). Empir- threatening adaptive potential and local persistence. It ical studies examining the genetic impacts of habitat is expected that following habitat fragmentation an fragmentation have demonstrated that the process can increased level of inbreeding will be evident. In addition, bring about a variety of gene-flow alterations. Frequent- a reduction in genetic diversity and increased genetic ly, plant populations experience increased levels of differentiation is expected following severe or long term inbreeding and random genetic drift following fragmen- population bottlenecks. We examined population genetic tation (YOUNG et al., 1996; DAYANANDAN et al., 1999; parameters for the subtropical rainforest tree COLLEVATTI et al., 2001; LIRA et al., 2003; CARDOSO et al., Macadamia tetraphylla (Proteaceae) at six field sites throughout its recently fragmented range, using four 2005; LOWE et al., 2005). The magnitude of such gene flow alterations depends on the severity and duration of microsatellite loci. Genetic diversity (HE) of the juvenile cohort was significantly correlated with estimated popu- the reduction in population size (BARRET and KOHN, lation size. No significant difference was observed for 1991). Where deforestation events lead to a reduction of genetic diversity between adult and juvenile cohorts, but gene flow between forest patches, the genetic bottle- juveniles, and not adults, exhibited significant popula- necks experienced in fragmented populations can cause tion differentiation ( = 0.061; P < 0.0001 and = 0.016; the independent loss of alleles from fragments, resulting P = 0.23, respectively). A second, standardised measure in increased population differentiation (JUMP and of differentiation, ´, yielded similarly large differences PEÑUELAS, 2006). between the two cohorts, though higher estimates of dif- ferentiation overall (adults – ´= 0.034, juveniles – To further understand the impact of fragmentation on ´= 0.116). The coefficient of population inbreeding (f) an Australian subtropical rainforest species, we used was significant and positive in all juvenile, and four out microsatellites to estimate genetic parameters for the of six adult, populations, and was significantly positively midstorey tree Macadamia tetraphylla (Proteaceae). correlated with adult tree density, but not adult popula- Populations of M. tetraphylla were selected to represent tion size. Since fragmentation is relatively recent for the range of population sizes exhibited across its patchy this species, the population bottleneck must have been distribution. The study took place in and around the Mt quite severe to have produced the observed patterns of Warning Caldera, central eastern Australia (Figure 1). population differentiation and genetic diversity. Frag- mentation of forest across the study area over the last The vast majority of subtropical lowland rainforest vege- 100+ years has led to the genetic isolation of M. tetra- tation in which M. tetraphylla grows was cleared phylla populations resulting in increased population between 145 and 105 years ago, with clearing beginning divergence and likely eventual loss of genetic variation slightly earlier in more southerly areas (RITCHIE and in future generations. PUGH, 1981). Key words: Macadamia tetraphylla, subtropical rainforest, Specifically, we aimed to determine: (1) whether larger fragmentation, gene diversity, inbreeding coefficient, differenti- or denser populations have higher levels of genetic ation, microsatellites. diversity and lower levels of inbreeding, relative to smaller populations; and (2) whether the post-fragmen- Introduction tation (juvenile) cohort exhibits higher levels of inbreed- ing and differentiation among populations, compared Habitat fragmentation, the reduction of continuous with the pre-fragmentation (adult) cohort. tracts of vegetation to smaller, spatially distinct patches, 1) School of Integrative Biology, University of Queensland, Materials and Methods St Lucia, Qld 4072, Australia. Study Species 2) Current address: Biodiversity Assessment and Management, Suite 11, 2/20 Shore Street West Cleveland, QLD, 4163, Aus- Macadamia tetraphylla is a small to medium sized tralia. mid-storey rainforest tree that is endemic to central 3) Australian Centre for Evolutionary Biology and Biodiversity, eastern Australia. Listed as vulnerable under both State and School of Earth and Environmental Sciences, University of and Federal legislation, concerns have been raised about Adelaide, North Terrace, Adelaide SA 5005; State Herbarium of its viability in the wild, over both the medium and long South Australia, Science Resource Centre, Department of Envi- ronment and Natural Resources, Hackney Road, Adelaide, SA term (GROSS, 1995). The species is patchily distributed 5005, Australia. within the regional landscape matrix, and is poorly rep- *) Author for correspondence: ANDREW J. LOWE. Phone: +61 434 resented in the reserve system. Populations are small 607 705. E-mail: [email protected]. (usually about 5–25 adults), with < 1000 individuals Silvae Genetica 60, 6 (2011) 241 DOI:10.1515/sg-2011-0032 edited by Thünen Institute of Forest Genetics Spain et. al.·Silvae Genetica (2011) 60-6, 241-249 Figure 1. – Field sites used in this study, showing the distribution of all adult trees present within each site. Each dot represents a tree. Site abbreviations are shown in the corner of each box: CV – Crookes Valley; CC – Cave Creek; MO – Mooball; Mullumbimby Creek; LH – Lennox Head; MF – Minyon Falls. estimated to be within conservation areas (PISANU, The level of fragmentation experienced by M. tetra- 2001). Its preservation is considered important, both in phylla in natural populations is relevant to many other terms of biodiversity maintenance, and also because of species in the region, especially endemics with limited the species’ significance to the macadamia nut industry. population size or distribution, and so it can be consid- The species is hermaphroditic, and flowers are borne on ered a suitable case study species to examine the impact long racemes. The European honeybee, Apis mellifera, of fragmentation on the genetic dynamics of threatened and native stingless bees, Trigona spp., are important plant populations in the region. pollinators in Australian macadamia orchards (HEARD, 1993, 1994; HEARD and EXLEY, 1994). In the wild, both of Study sites and spatial mapping these pollinators have been observed on M. tetraphylla, with A. mellifera the more common of the two. However, Six study sites were selected (Figure 1; Table 1), overall pollinator activity appears to be low (PISANU, encompassing a range of adult population sizes. This 2001; pers. obs.), with potential pollen limitation range of population sizes is regarded as representative (PISANU, 2001). M. tetraphylla is estimated to have a of the variety of situations in which M. tetraphylla can lifespan of over 100 years, of which up to six years con- be found in the regional landscape matrix. All six of stitutes the juvenile period (QUEENSLAND CRA/RFA these study sites are bona fide wild populations, not the STEERING COMMITTEE, 1997). result of plantings by early settlers. 242 DOI:10.1515/sg-2011-0032 edited by Thünen Institute of Forest Genetics Spain et. al.·Silvae Genetica (2011) 60-6, 241-249 Table 1. – Site characteristics and stratification for six populations of M. tetraphylla, central eastern, Australia. Sample collection, DNA extraction and microsatellite les (AO) and effective number of alleles (AE) using POP- survey GENE (YEH et al., 1997). For each population, genomic DNA from every adult Inbreeding coefficient (f) and an index of differentia- (6–37) and 15–20 randomly selected juveniles was tion () were calculated following the formula of WEIR extracted using the QIAGEN DNeasy™ Plant Mini Kit. and COCKERHAM (1984), using FSTAT 2.9.3 (GOUDET, The standard protocol was followed, and success of DNA 1995). The significance of obtained global FST (≈ ) val- extractions and quantity of yield was determined by ues were tested by performing 50,000 randomisations of visualisation on 1.8% agarose gels (stained with ethidi- genotypes among samples. P-values were generated by a um bromide), using a 1000 BP ladder (HyperLadder I, simulation of 50,000 random permutations of genotypes Bioline). Based on screening and optimisation experi- among populations. Tests did not assume H-W within ments, four polymorphic loci (Minµs2, Minµs7, MinµS14 populations. The influence of each locus on generating and MinµS74) were selected from Schmidt et al. (2006) global FST estimates was examined by successive jack- for genotyping the six populations. Samples were plated knifing across loci using GENETIX 4.01 (BELKHIR et al., out onto 96 well trays and diluted 1:10. Ten percent of 1998). We also calculated ´, a standardised measure of individuals were repeated in order to estimate the fre- . This was done by transforming our genotype data in quency of genotyping errors. Polymerase chain reaction the utility RECODEDATA (MEIRMANS, 2006)