Spatial analysis of genetic diversity as a tool for plant conservation
Introduction
During the last few decades we have witnessed an intense academic debate between plant conservation biologists about which is the most appropriate way to approach a conservation problem—the so called ecological or the genetic approach (Schemske et al., 1994). In the course of this long debate, several authors have pointed out that not only is knowledge of the amount of genetic diversity critical for a correct diagnosis of the status, threats and viability of populations (Frankham, 1995), but also the spatial distribution of this diversity (Falk & Holsinger, 1991, Dunham et al., 1999). This recognition falls within a more general statement in which geneticists have recognized the importance of the interaction between genome and environment in time and space so as to better understand evolution (Berry, 1989), thus connecting genetics and ecology (Jelinski, 1997). Genetic diversity may appear spatially structured at different scales, such as population, subpopulation or among neighbouring individuals. This spatial distribution is necessarily a product of environmental influences, including human activities (Knowles et al., 1992), life story traits and demographic past history of the plant species (Loveless & Hamrick, 1984, Slatkin, 1985). Thus, knowledge of spatial genetic structures provides a valuable tool for inferring these causal factors and also the underlying genetic processes such as differential selective pressures, gene flow and drift (Nevo et al., 1986, Barbujani, 1987, Epperson, 1993, Bjornstad et al., 1995). Consequently, information about dispersal, pollinator behaviour, breeding system, safe site availability for establishment and other processes operating and structuring populations at such scales can be derived from spatially explicit approaches (Peakall and Beattie, 1995). Knowledge of all these features represents a key priority for conservation managers. Surprisingly, however, most genetic studies on endangered plants either lack explicit spatial considerations or analyse them only at scales where biological interpretations at the population level are not straightforward.
Although there is considerable agreement that there is no single “correct” scale at which to describe populations, since the processes that originate the spatial patterns may operate at different scales (Levin, 1992), most of the available numerical methods used in population genetics have been applied at medium and large scales (population or larger geographical approaches). The spatial association of environmental and genetic variables (Hedrick, 1986) and the spatial patterning of plant genetic diversity (Heywood, 1991) have been extensively studied at these medium or large scales. However, abundant evidence has been reported in which plant populations exhibit genetic micro-differentiation, spatially closer individuals being genetically more alike than individuals at some distance (Sokal et al., 1989, Epperson, 1993). Even at extremely fine scales, spatial genetic structures have been detected in some plant populations (Turkington & Harper, 1979, Epperson & Clegg, 1986, Wagner et al., 1991, Tani et al., 1998), although, in other cases, random or near random distributions have been found (Waser, 1987, Epperson & Allard, 1989, Leonardi et al., 1995). Therefore, the challenge that arises when studying the genetic structure of populations of rare plants is not only choosing the scale of description, but rather recognising that change may be taking place at several scales at the same time (McCue et al., 1996).
There is an urgent need to integrate the knowledge derived from genetic, demographic and ecological approaches to species conservation in order to be able to formulate management strategies that take into account all the different considerations. Spatial analysis techniques are a meeting point for all these three approaches and thereby progress on this direction is likely to facilitate a much sought comprehensive and integrated outlook in conservation biology.
Section snippets
Objectives
The purpose of this work is to present methods that can be applied to the study of spatial genetic structures in endangered plants, both at broad and narrow scales, and that may be helpful in interpreting key biological processes which affect the viability of the populations. A second matter which is also posed is the treatment of genetic spatial variability in the context of hypothesis testing and its use as a tool in conservation biology.
Traditionally genetic concerns have been put on one
Conventional approaches to genetic structure
Traditionally, population genetic structures within a species have been primarily studied by testing departures of allele frequencies from panmictic expectations or testing for heterogeneity in allele frequencies among populations or other spatial subdivisions (Heywood, 1991). This implies that the genetic information at the individual level is necessarily merged into clusters of a different spatial or biological nature (populations, subpopulations…) and that the relative frequencies are
Molecular markers and nature of genetic data
A wide choice of genetic markers is now available to the conservation biologist (see review by Haig, 1998). Since the 1960s, allozymes have been succesfully used to characterise genetic diversity of rare plant species (Hamrick and Godt, 1990), although sometimes they have not shown enough discriminatory power to distinguish between individuals (Brauner et al., 1992, Buso et al., 1998). In the past decade, DNA techniques have gained ground, especially those based on the polymerase chain reaction
Measuring spatial autocorrelation of genetic data
Genetic data frequently present spatial autocorrelation, that is, the association of values of one geographically distributed variable with the values of the same variable at all other localities. In these cases, it is possible to predict the values of a genetic variable at some points of the space from the known values of the variable at other known sampling points.
Hypothesis testing and partialling out spatial information
Theoretical and empirical work is rapidly advancing a set of hypothesis tests that use spatial patterns to study some genetic processes such as gene flow or natural selection (Epperson, 1990). The evaluation of the effect of environmental conditions on spatial genetic structure can be viewed as a problem of spatial covariation (Legendre and Fortin, 1989). This type of covariation can efficiently be approached by means of constrained ordinations considering the spatial location of each
Past and present applications and future possibilities
Spatial statistical methods provide plant conservation biologists with powerful tools for measuring the structure of genetic diversity of target plants. However, a quick review of current literature shows that the use of these techniques is low among genetic conservationists and even among population geneticists. We have reviewed papers approaching spatial genetic structures in plant species and limited our search to articles dealing with spatial autocorrelation and related techniques. The
Acknowledgements
We are grateful to J. Heywood and D. Gömöry for their very useful comments and suggestions. This work has been partially financed by REN 2000-0254-P4-03 project of the Spanish Ministry of Science and Technology.
References (158)
- et al.
Contrasting patterns of genetic structure in the South African species Androcymbium bellum, A. guttatum and A. pulchrum (Colchicaceae)
Biochemical Systematics and Ecology
(1999) - et al.
Spatial autocorrelation of genetic variation in three stands of Ophiopogon xylorrhizus (Liliaceae s.l.)
Annals of Botany
(2000) On genes and geographya landscape perspective on genetic variation in natural plant populations
Landscape and Urban Planning
(1997)Molecular Markers, Natural History and Evolution
(1994)- et al.
Genetic diversity and structure of the narrow endemic Wyethia reticulata and its congener W. bolanderi (Asteraceae) using RAPD and allozyme techniques
American Journal of Botany
(1999) - et al.
Intraspecific genetic structure in a mixed population of Quercus petraea (Matt.) Leibl and Q. robur L
Heredity
(1994) Autocorrelation of gene frequencies under isolation by distance
Genetics
(1987)- et al.
Genetic and evolutionary consequences of small population size in plants: implications for conservation
- et al.
Fine-scale genetic structure of a Turkey oak forest
Evolution
(1995) Ecologywhere genes and geography meet
Journal of Animal Ecology
(1989)