Elizabeth M. Gillet1,2
1 Institut für Forstgenetik und
Forstpflanzenzüchtung,
Universität Göttingen,
2 Institut für Forstgenetik und Forstpflanzenzüchtung,
Bundesforschungsanstaalt für Forst- und Holzwirtschaft,
Email: egillet@gwdg.de
The objective of this compendium is to associate "Purposes" with "DNA markers". Before reading about associations between individual purposes and markers in the contributions, it can be helpful to survey these two topics separately. In this chapter, a systematic classification of the different characteristics of DNA markers that determine their usefulness is introduced.
1. Genetic markers and their characteristics
The concept of genetic marker and gene marker: Consider a given set of individuals of the same species and a set of characteristics of these individuals. The set of characteristics defines a trait in the set of individuals, if each individual possesses exactly one of these characteristics - its trait state or phenotype.
The genetic information possessed by each individual is termed its genotype and can refer to the entirety of its genetic information or a part of it. For each locus that is involved in the expression of the phenotype, the individuals's genotype is the set of genes present at this locus. Since the following considerations apply both to coding and non-coding DNA, the term gene is used here in a wide sense to denote a defined segment of the DNA of an individual as a unit of transmission, and not only in the narrow sense of a "functional gene". A locus corresponds to a set of "transmission homologous" genes (Gillet 1996). Two genes at the same locus that differ in type are called alleles. A haploid locus has one gene, a diploid locus two genes or alleles, and a polyploid locus more than two. (In humans, all nuclear loci that are not located on the X- or Y-chromosomes are diploid. In most tree species, all nuclear loci are diploid. Many agricultural crop plants have been bred to be polyploid at all nuclear loci.)
The following definitions define particular types of relationships between
phenotypes and genotypes (Gillet 1996; Bergmann et al. 1989).
A trait is termed a genetic trait, if any two individuals possessing
the same genotype also have the same phenotype, regardless of the
environmental conditions in which they exist. In order to
determine the relationship between phenotypes and genotypes, it
is necesary to perform an inheritance analysis, i.e.,
determination of the mode of inheritance of the trait states
(see below). After successful inheritance analysis, a genetic trait
is qualified as a genetic marker, if the relationship holds
that each phenotype can be unambiguously assigned to a set of
genotypes at one or more specified loci. Thus for
a genetic marker it holds that, if one observes an individual's
phenotype, then one knows that this individual
possesses one of a defined set of genotypes. If each phenotype
can be unambiguously assigned to exactly one genotype,
then the genetic marker defines a gene marker. By
this assignment, all involved genes become recognizable. Thus a trait
can be called a gene marker if and only if there is a 1:1 relationship
between phenotype and genotype, such that the alleles present
at each of the involved loci are unambiguously specifiable for each
phenotype. This hierarchy of trait types is summarized
in Table 1.
| Type of trait | Definition |
| Trait | Individual ===> Unique phenotype |
| Genetic trait | Genotype ===> Unique phenotype |
| Genetic marker | Phenotype ===> Unique set of genotypes |
| Gene marker | Phenotype <===> Unique genotype |
Genetic markers preceding the development of DNA markers: Since the advent of recombinant DNA technology in population genetics in the mid-1980's, the repertoire of genetic markers available for population genetic studies in a number of tree species has increased enormously. In order to appreciate this fact, it is interesting to consider the history of genetic marker development in forest genetics. Until the beginning of the 1970's, the only genetic markers available in tree species were the rare morphological traits that could be shown to be controlled by alleles at a single gene locus, such as the aurea phenotype in Norway spruce (Langner 1953). Early attempts to interpret the relative or absolute quantities of the different monoterpenes in the resin of conifer trees (measured using gas chromatography) as genetic markers remained inconclusive due to difficulties in determining mode of inheritance and the probable dependence of their expression on environmental conditions (especially pathogen stress).
Isoenzymes: It was not until the early 1970's, when Bartels (1971) and Bergmann (1971) developed enzyme electrophoresis for Norway spruce, that direct products of tree DNA were made accessible to observation. This technique involved separation of functionally equivalent enzyme molecules according to their differing electrostatic charges, sizes, and molecular conformations, followed by their staining. Inheritance analysis of the resulting banding patterns enabled inference of their mode of inheritance (see below) and, consequently, allowed them to be used as genetic markers. Until recently, practically all progress in the experimental population genetics of forest tree species was achieved using multilocus isoenzyme analysis. Isoenzymes, the "electrophoretically separable variants of one enzyme system" (Bergmann et al. 1989), are coded by genes at one or often several loci. Variants that are coded by alleles at the same locus are called allozymes. Multilocus analysis considers the results for various loci belonging to one or, more commonly, a large number of enzyme systems. In fact, isoenzymes are still widely used as genetic markers for reasons that include the following: They are inexpensive compared to DNA markers, the laboratory protocols are well-established in numerous tree species, they are products of structural genes whose roles in metabolism are known in most cases, and, most importantly, their typical levels of variation makes them suitable markers for a number of purposes.
The usefulness of a marker completely depends on its characteristics. In order to appreciate what can be accomplished with DNA markers that couldn't already be done with isoenzymes, it will be helpful to begin by recalling the characteristics of isoenzymes. Enzyme molecules are direct products of genes, and thus of DNA, and play essential roles in the primary and secondary metabolism of their organism. Enzyme molecules are composed of chains of amino acids as determined by the DNA sequences of the coding genes. Differences in the total electrostatic charges of their amino acid sequences indicate the existence of differences in the DNA sequences. (Due to the redundancy of the genetic code that "assigns" amino acids to nucleotide triplets, the opposite is not necessarily true). Allozymes almost always differ due to single nucleotide substitutions at the locus that cause the substitution of single amino acids of different charges; as a rule, isoenzymes coded by different loci differ in size also (F. Bergmann, personal communication). Size differences result from insertions/deletions of nucleotides that lead to a longer/shorter amino acid sequence.
As summarized in Table 2 below, the typical mode of inheritance of isoenzymes is: Transmission by one or only a few nuclear gene loci; Codominance of gene action (with the exception of the null alleles typically found at some isoenzyme loci), which ensures the identifiability of both genes at a locus and thus of heterozygous individuals. An additional characteristic observed at a number of isoenzyme loci is the following: Prevalence of the same one, two, or even three alleles, accompanied by the same suite of rare alleles, in all studied populations even of related species, and a typically low level of differentiation among populations (Gregorius and Bergmann 1995). This has led to the widespread assumption that isoenzymes are selectively "neutral", i.e., their frequency distributions are due to random effects (random mutation and drift). The authors of the above-cited study, however, consider the universal prevalence of the same two or three alleles at a locus to be evidence of "ontogenetic differentiation of enzyme function", while the universal rarity of the other alleles suggests "recurrent deleterious mutation", both of which are forms of selectivity.
2. Systematic classification of the characteristics of genetic markers
What new characteristics are possessed by DNA markers that earlier markers
did not have? These new characteristics determine which previously intractable
purposes can now be treated using genetic markers. In order to
answer this question, the classification scheme in Table 2
for the description of the characteristics of genetic markers will be helpful:
3. The characteristics of the DNA markers developed within this research project
It will be left to the single contributions to explain
how the respective DNA markers are observed in the laboratory. It
will suffice here to list their characteristics (some of which were first
described during the project) according to the above classification, since
it is these characteristics that decide on the suitability of a marker
for a given purpose:
| Marker type |
|
Level of genetic variability | Function | |
| Mode of transmission | Mode of gene action | |||
| AFLP® fingerprint | biparental nuclear, many loci, unknown no. alleles per locus | dominance at some loci, codominance at others | hypervariable, i.e. each individual has unique banding pattern | unknown |
| Nuclear microsatellites | biparental nuclear, few loci, many alleles per locus | codominance, with exception of null alleles at some loci | large variation within populations, low differentiation between populations | non-coding, may contribute to genome stability |
| Chloroplast microsatellites | uniparental (maternal in angiosperms, paternal in conifers), pseudo-haploid, single locus, many alleles per locus | each cytotype is expressed | low variation within populations, large differentiation between populations | non-coding |
| Mitochondrial intron marker | uniparental (maternal), pseudo-haploid, single locus, many alleles per locus | each cytotype is expressed | low variation within populations, large differentiation between regions | non-coding |
| ITS of ribosomal DNA | biparental nuclear, several loci, several alleles per locus | codominance | high variability, even within a single individual | non-coding |
| cDNA markers | biparental nuclear, one to a few loci, few alleles per locus | codominance | low variation within populations, low differentiation between populations | functional differences possible between alleles of a locus |
| Isoenzymes
(for comparison) |
biparental nuclear, 1-5 loci, 1-7 alleles per locus | codominance, with exception of null alleles at some loci | low to medium variation within populations, low differentiation between populations | functional differences possible between alleles of a locus |
4. Links to the contributions treating the different types of DNA marker
Bartels H (1971) Genetic control of multiple esterases from needles and macrogametophytes of Picea abies. Planta (Berl) 99:283-289.
Bergmann F, Gregorius H-R, Scholz F (1989) Isoenzymes, indicators of environmental impacts on plants or environmentally stable gene markers? In: Scholz F, Gregorius H-R, Rudin D (eds.). Genetic Effects of Air Pollutants in Forest Tree Populations. Springer-Verlag Heidelberg, New York, Tokyo, pp. 17-28.
Bergmann F (1971) Genetische Untersuchungen bei Picea abies mit Hilfe der Isoenzym-Identifizierung. I. Möglichkeiten für genetische Zertifizierung von Forstsaatgut. Allgemeine Forst- und Jagdzeitung 142: 278-280.
Gregorius H-R, Bergmann F (1995) Analysis of isoenzyme genetic profiles observed in tree populations. In: Baradat P, Adams WT, Müller-Starck G (eds.). Population Genetics and Genetic Conservation of Forest Trees. Amsterdam: SPB Academic Publishing, pp. 79-96.
Gillet EM (1996) Qualitative inheritance analysis of isoenzymes in haploid gametophytes: principles and a computerized method. Silvae Genetica 45: 8-16.
Gregorius H-R, Roberds J (1986) Measurement of genetical differentiation among subpopulations. Theoretical and Applied Genetics 71: 826-834.
Gregorius H-R (1988) The meaning of genetic variation within and between subpopulations. Theoretical and Applied Genetics 76: 947-951.
Langner W (1953) Eine Mendelspaltung bei Aurea=Formen von Picea abies (L.) Karst. als Mittel zur Klärung der Befruchtungsverhältnisse im Walde. Z. Forstgenetik u. Forstpflanzenzüchtung 2: 49-51.
Vendramin GG, Ziegenhagen B (1997) Characterisation and inheritance of polymorphic plastid microsatellites in Abies. Genome 40: 857-864.
© Institut für Forstgenetik und Forstpflanzenzüchtung, Universität Göttingen, 1999