1. INTRODUCTION

1.1. Why it is important to measure and save adaptive genetic diversity in forest tree populations
1.2. Traditional methods to measure adaptive genetic diversity

1.1. Why it is important to measure and save adaptive genetic diversity in forest tree populations

Genetic diversity is the basis of the ability of organisms to adapt to changes in their environment through natural selection. Populations with little genetic variation are more vulnerable to the arrival of new pests or diseases, pollution, changes in climate and habitat destruction due to human activities or other catastrophic events. The inability to adapt to changing conditions greatly increases the risk of extinction. Gene conservation management aimed to save adaptive genetic diversity should be based on the knowledge of the genetic basis of adaptation. The goal of this paper is to describe how adaptive genetic diversity can be measured using new molecular genetic approaches and achievements in forest genomics.

1.2. Traditional methods to measure adaptive genetic diversity

1.2.1. Field Experiments
1.2.2. Molecular genetic markers

1.2.1. Field Experiments

Field experiments (common-garden tests) have been used traditionally to measure adaptive genetic diversity in trees. These tests continue to be used extensively in tree breeding and are very effective in identification of families and clones that are specifically adapted to particular environments or to a broad variety of environments. However, field experiments are very time consuming and relatively expensive, and more importantly, they are based solely on the phenotypes. They can estimate genetic parameters but only on measurable traits, not on individual genes. This method can neither provide information on what particular genes and how many of them are involved in adaptation nor how much of phenotypic variation can be explained by genetic variation in these genes.

1.2.2. Molecular genetic markers

Another, generally complementary, approach for estimating adaptive genetic diversity is to measure genetic variation using molecular genetic markers. However, DNA variation that resides in the non-coding genomic regions or does not lead to a change in the amino acid sequence (for example, so-called synonymous nucleotide substitutions in the second or the third positions in a codon encoding an amino acid) is unlikely to have any significant contribution to adaptation. Many modern genetic markers belong to so-called anonymous DNA marker type such as microsatellites or simple sequence repeats (SSRs), restriction fragment length polymorphisms (RFLPs), random amplified polymorphic DNA (RAPDs), and amplified fragment length polymorphisms (AFLPs). These marker types generally measure apparently neutral DNA variation, and are very useful (with different efficiency, of course) in the analysis of phylogenetic relationships, population structure, mating system, gene flow, parental assignment, introgressive hybridization, marker-aided selection and genetic linkage. They are not useful for measuring adaptive genetic diversity.

Isozymes are another class of genetic markers widely used in forest genetics in the last several decades. Although variation revealed by these markers is caused by amino acid variation, it is unclear whether this variation is selectively neutral or has any adaptive significance. There are many studies showing great adaptive differences (in morphological or phenological characteristics) among populations of forest tree species, but no accompanying differences for the isozyme markers (see references in Boshier and Young 2000).

Markers of all kinds are used now in forest genetics - both anonymous and genic, dominant and codominant, highly and less polymorphic, expensive and inexpensive, supposedly selective and apparently neutral, abundant and less numerous. A classification of genetic markers is offered in Table 1, which takes into account their most important features. Details on the nature of these markers, their advantages and disadvantages and use in different applications are available elsewhere (see the most recent reviews by Cervera et al. 2000; Linhart 2000; Glaubitz and Moran 2000; Savolainen and Karhu 2000; Chapters 12-14 in Mandal and Gibson 1998).

The ideal marker for estimating adaptive variation should meet the following criteria: (1) be directly involved in the genetic control of adaptive traits; (2) have identified DNA sequence and known function; (3) be readily available for genetic analysis, and (4) have easily identifiable allelic variation. No marker fully satisfies all these criteria. However, a promising new marker, expressed sequence tag polymorphisms (ESTPs), seems to satisfy most or all of these criteria, emerged recently as a result of genomic studies.

Table 1: Comparison of commonly used genetic markers

Feature	RFLP	SSR	RAPD	AFLP	Isozymes	STS/EST
Origin	Anonymous/ Genic	Anonymous	Anonymous	Anonymous	Genic	Genic
Maximum theoretical number of possible loci in analysis	Limited by the restriction site (nucleotide) polymorphism (tens of thousands)	Limited by the size of genome and number of simple repeats in a genome (tens of thousands)	Limited by the size of genome, and by nucleotide polymorphism (tens of thousands)	Limited by the restriction site (nucleotide) polymorphism (tens of thousands)	Limited by the number of enzyme genes and histochemical enzyme assays available (30-50)	Limited by the number of expressed genes (10,000-30,000)
Dominance	Codominant	Codominant	Dominant	Dominant	Codominant	Codominant
Null alleles	Rare to extremely rare	Occasional to common	Not applicable (presence/ absence type of detection)	Not applicable (presence/ absence type of detection)	Rare	Rare
Transferability	Across genera	Within genus or species	Within species	Within species	Across families and genera	Across related species
Reproducibility	High to very high	Medium to high	Low to medium	Medium to high	Very high	High
Amount of sample required per sample	2-10 mg DNA	10-20 ng DNA	2-10 ng DNA	0.2-1 µg DNA	Several mg of tissue	10-20 ng DNA
Ease of development	Difficult	Difficult	Easy	Moderate	Moderate	Moderate
Ease of assay	Difficult	Easy to moderate	Easy to moderate	Moderate to difficult	Easy to moderate	Easy to moderate
Automation / multiplexing	Difficult	Possible	Possible	Possible	Difficult	Possible
Genome and QTL mapping potential	Good	Good	Very good	Very good	Limited	Good
Comparative mapping potential	Good	Limited	Very limited	Very limited	Excellent	Good to very good
Candidate gene mapping potential	Limited	Useless	Useless	Useless	Limited	Excellent
Potential for studying adaptive genetic variation	Limited	Limited	Limited	Limited	Good	Excellent
Development	Moderate	Expensive	Inexpensive	Moderate	Inexpensive	Expensive
Assay	Moderate	Moderate	Inexpensive	Moderate to expensive	Inexpensive	Moderate
Equipment	Moderate	Moderate to expensive	Moderate	Moderate to expensive	Inexpensive	Moderate to expensive

RFLP - restriction fragment length polymorphism; SSR - simple sequence repeats (microsatellites); RAPD - random amplified polymorphic DNA; AFLP - amplified fragment length polymorphism; STS - sequence tagged site; EST - expressed sequence tags.