by Alexander Micke
Plant breeding using mutations.
Breeding has been practiced since the early human civilization and selection was the first method of breeding, adding the criteria of suitability for man’s use (e.g. larger seed, better taste, easier harvestability) to those of natural adaptation, fitness and offspring. It has been said, that the ultimate source of all heritable variation to select from are mutations. But such a statement leaves open, where the genes to start with and the genetic code came from. Recombination of genes can provide additional genetic variation, if differences exist not only between various genes, but also in form of alleles of particular genes of prospective recombinants. Such alleles derive from mutations. Using mutants in cross breeding requires no in depth knowledge about mutations, because the mutated trait is the object of desire. But when the mutant trait is not inherited as expected, the breeder may begin to think about the actual mutational event, that led to the mutant phenotype.
There are different definitions of the term „mutation“ and this may create the impression, that the term is somewhat woolly. Definitions range from „a sudden phenotypic change in a character of an individual, not due to crossing or segregation“ up to „an alteration in the macro-molecule of DNA“ (where it remains open, whether the alteration leads to a change in gene function or not). Included under the term „mutation“ is also the augmentation of genetic material through nucleotide or gene copies, through additional individual chromosomes, as well as through the multiplication of whole genomes towards polyploidy. In order to speak more clearly about mutations and their potential for crop improvement, it would seem desirable to have different terms at least for (a) the phenotypic alteration and (b) the various underlying molecular and numerical changes. But in any case, a mutation has to be phenotypically expressed to be selectable, all other mutations are only of scientific interest.
We have today a relatively good understanding of the processes of mutation induction by UV, by ionizing radiation and also by certain chemical mutagens, but the causes of „spontaneous“ mutations are still to some extent uncertain (HALL 1990; RASMUSSON and PHILLIPS 1997). In particular, the kind and rate of mutations during the millions of years of evolutionary history are subject to discussion, as current mutation rates can hardly explain the progress of past evolution (OHNO 1970; KIYOSAWA and NOMURA 1988; KOINANGE et al. 1996; KENRICK and CRANE 1997).
Stability vs. differentiation.
In 1865, Gregor MENDEL showed, that characters (traits) of individuals must be based on controlling elements (genes), that are passed-on intact and unchanged from one plant generation to the next. But these controlling elements are also passed-on intact and unchanged from cell generation to cell generation during the individual plant’s development, i.e. from mother cell to daughter cells in every somatic cell division (based on the mechanism of mitoses). MENDEL should be forgiven for not paying attention to this matter. Even today we still have a problem to comprehend the ontogenetic development from a single „totipotent“ cell (zygote) into a complex organism with different tissues, consisting of specialized cell types, forming organs with special functions, and at the same time maintaining an intact „germ line“ for potentially „totipotent“ generative cells. Long after MENDEL the question emerged, how this ontogenetic differentiation could be reconciled with the stability of the heritable elements postulated by MENDEL and confirmed by the discovery of chromosomes as carriers of genes, the meticulously conservative DNA-replication, the course of events during mitosis and meiosis etc. Now we begin to learn about factors determining the ontogenetic fate of particular meristematic cells (GOODRICH et al., 1997) or e.g. the co-ordination between embryo and endosperm during seed development (RAY, 1997).
In spite of booming in-vitro culture business we have difficulties to clone differentiated cells as such, i.e. with maintaining their specialized functions. This should be extremely useful for the industrial production of particular compounds. Still we don’t know really what distinguishes a differentiated cell from a „totipotent“ (embryonic) cell, what is involved in the de-differentiation of „explant“ cells leading to the grwoth of callus, or what is going on when callus cells eventually become organogenic. Isn’t it still surprising, that the profound processes of differentiation leave the genome essentially unaffected, so that by a de-differentiation „totipotent“ cells can be obtained again? Evolution apparently made provisions for high genetic flexibility during ontogenesis and especially for performance under the challenge of environments (a plant cannot migrate!), but at the same time established reliable systems to preserve genetic integrity for subsequent generations. The unnatural conditions of tissue and cell culture, on the other hand, demand apparently too much from the control system and thus result in genetic instability (PHILLIPS et al. 1994).
Gene expression: dominance vs. recessivity.
MENDEL discovered another phenomenon, relevant for the concept of mutations: He observed differences in strength of gene expression and ascribed this to the occurrence of „alleles“ at the gene locus: the stronger ones he called „dominant“, the weaker ones he called „recessive“ (at some gene loci multiple alleles can occur). MENDEL described the phenotypes observed, when alleles of one or two genes interacted in the same (heterozygous) individuum, and he stated the segregation ratio of phenotypes found upon self-pollination of the heterozygous plant. He did not look into possible reasons for the different strength of expression. He also did not mention any observation he might have made about interactions between genes at different loci, influencing the kind and level of their phenotypic expression (epistasis, pleiotropy).
A few years after MENDEL found the fundamental principles of inheritance, Friedrich MIESCHER (1871) discovered the desoxyribonucleic acid (DNA). Although he himself did not realize the genetic role of this compound, these events marked the beginning of a most fascinating research century, focussing at the chemical basis of inheritance, the genetic code, the structure of chromosomes, the analysis of bio-chemical pathways and their control, the possibility for far-reaching genetic manipulations („gene engineering“), and approaching eventually the understanding of „life“. However, in spite of all progress in molecular genetics, we still cannot fully answer the seemingly rather simple questions: What distinguishes different alleles and what causes the difference between dominant and recessive gene expression? The answers should be most relevant for induced mutations and the judgement of their breeding value, because the majority of induced mutations act like changes from a dominant allele into a recessive one, or represent deletions, which in crossing experiments mimic recessive alleles. The question gets more complicated by the functional hierarchy among genes and the location/time/stage-specific regulation of gene expression by other genes: A silenced dominant gene would probably act like its recessive or null-allele; the mutation of a dominant inhibitor gene to its recessive or null-allele may allow one or more (previously silent) genes to express as dominant alleles.
Artificially induced mutations.
The pioneering experiments by MULLER (1927, 1928) have shown, that it is possible to knock down the stability of genes. This new possibility was looked upon with great optimism. MULLER himself thought even about breeding better human beings, but the plant breeder STADLER was sceptical about the usefulness of genetic alterations induced e.g. by x-rays. The scepticism resulted from experiments with cereals (STADLER 1930, 1932), where he was not satisfied with an impressive quantity of heritable variation, but rather looked for the kind of changes, that might lead towards crop improvement. He did not observe any dominant mutations and suspected, that a recessive one was „simply the destruction of a gene“. He advised students to better make use of „natural“ genetic variation for profitable crop improvement.
When after World War II funds became available for „peaceful uses of atomic energy“, many young researchers in developed as well as developing countries embarked on the fashionable technology of using radiation and radioactive isotopes for inducing mutations, often without paying attention to the rather modest results obtained between 1930 and 1945 by using radiations as well as some chemical mutagens. It was confirmed, that mutagen treatments harm the delicate molecular structure of genes, and hardly produce the minor changes in nucleotides, which could be expected to lead to meaningful codes for different (or even new) enzymes (STADLER and ROMAN 1948). Some chemical mutagens were found to act less destructive and to cause a much higher proportion of minor changes. However, even these lead to missense and nonsense mutations and did not contribute much to crop improvement.
However, with international coordination and some financial assistance by IAEA and FAO from 1964 onwards it could be convincingly demonstrated, that ionizing radiations and also certain chemicals, when handled properly, could induce many useful alterations in the genomes of crop plants. Records maintained by the Joint FAO/IAEA Division in Vienna show, that ca. 2000 crop cultivars with one or more useful traits from induced mutations (mainly from x- and gamma-rays) were released worldwide over a period of 35 years. Included in these records are some outstanding examples of cultivars (e.g. „Diamant“ and „Trumpf“ in barley; „IRAT 13“, „Yuanfengzao“ and „Calrose 76“ in rice; „Lumian 1“ and „NIAB 78“ in cotton; „Pervenets“ in sunflower; „Star Ruby“ in grapefruit), which had a remarkable economic impact (Anonymous 1991; MICKE et al. 1980, 1985; FAO/IAEA Mutation Breeding Newsletters 1972 – 1997). Inspite of the number of induced mutants recognized as valuable crop cultivars and used successfully in cross breeding, of course „new genes“ in the strict sense could not be produced (MICKE 1991). Consequently, one would expect much interest in the question, what kind of molecular alteration in the chromosomal or cytoplasmic DNA, or what kind of structural/numerical alteration in the genome happened in those successful mutant cultivars. However, so far there were not many efforts to bring into accord the destruction caused by mutagens and the good performance of so many induced mutants. The answers to this question should be very relevant for any future investment into mutation breeding and gene engineering, to give a reliable forecast of what can be expected.
Induced mutations occur more or less randomly in the genome, even their target cannot be directed. Only one of the (two or more) alleles of a locus is affected, inheritance is almost ever recessive, therefore homozygosity is normally required for proper expression. Accordingly, results were more often useful in self-pollinating plant species. On the other hand, mutant heterosis has been repeatedly reported (MICKE 1968, 1969, 1976; RÖMER and MICKE 1974; MALUSZYNSKI et al. 1987, 1989) and specific mutations e.g. concerning male sterility (DASKALOV and MICHAILOV 1988) or grain quality traits (RÖBBELEN 1990) proved useful in cross-pollinating species. In vegetatively propagated crops, which usually are heterozygous and therefore could be improved also by deletions uncovering existing alleles, success has been tremendous, but mainly in ornamental plants (BROERTJES and VAN HARTEN 1988).
Today, mutation breeding is not anymore based only upon classical physical mutagens like x- or gamma rays or classical chemical mutagens like EMS or NMH, but also upon variation that occurs during in-vitro culture and has been termed „somaclonal variation“ (SKIRVIN, R.M. 1978; OONO et al. 1984; NOVAK et al. 1988; NOVAK 1991; SUKEKIYO and KIMURA 1991). It was pointed out already, that the term „mutation“ is rather woolly defined, but the term „somaclonal variation“ is even worse! Initially, it described the unexpected, sudden occurrence of more or less heritable variants in the somatic or generative offspring of in-vitro cultured plant material (HEINZ and MEE 1969). In the meantime, it became known, that a wide array of alterations in nuclear and cytoplasmic genetic elements contribute to the observed phenotypic variation, and that many of them are of epigenetic nature (D’AMATO 1986; PHILLIPS et al. 1994). Initially directors of tissue culture laboratories provoked the interest of plant breeders and sponsors by promising new and particularly useful genetic variation (OONO 1978; LARKIN and SCOWCROFT 1981; CHALEFF 1983; EVANS and SHARP 1983), but soon one learned, that uncontrollable occurrence of „somaclonal variation“ could ruin valuable genetic stocks (GÖBEL et al. 1986). An interesting feature of somaclonal variation was the repeatedly reported occurrence of ‘homozygous mutants’ (e.g. OONO et al. 1984). At least some of them were later identified as a kind of „Dauermodifikation“, resulting from gene inhibitions, that could pass through meioses and were stable for several generations (OONO 1985). It seems, that also here more research is needed, to clarify the underlying causes of new variation before its use in crop breeding.
The importance of gene regulation: inhibition vs. activation.
During ontogenesis, particular gene products are needed for a particular phase (e.g. the early embryonic development), but hey have to be suppressed during other phases. The original „totipotency“ of meristematic cells has to be restricted for initiating the formation of the different, functionally specialized tissues. Does this happen by activating genes that were inhibited in embryonic cells or by inhibiting genes that were active during early phases of ontogenesis?
From tumor research it is known, that in the adult human body certain cells may regain their „totipotency“ by the mutation or deletion of a so-called „onco-gene“ (SVOBODA 1994). These genes in reality are inhibitory genes, because their dominant alleles repress „totipotency“ and thereby permit the development of differentiated somatic tissues and the function of specialized organs (STANBRIDGE 1990). A recessive mutation in one of such „onco-genes“ was found to activate more than 15 other genes. Also in plants the transition from a „totipotent“ zygote to specialized somatic cells requires the inhibition of a number of genesand the activation of others. Augmentation of certain genomic portions helps to meet the requirements for specific cell functions. Ultimately, forming a plant organism with its highly specialized tissues and organs should result from a delicate interplay of gene inhibition and gene activation. To assure, that in a particular cell and at a particular time only the needed genes function, requires a highly sophisticated epigenetic system of gene regulation with ability to respond quickly to signals from outside and inside (OHNO 1970; CLARK 1997).
Shouldn’t there be a relationship between the transient epigenetic regulation of gene activity for ontogenetic development and for somatic metabolic performance in already differentiated tissues, and the mechanisms of a (truly heritable) control of gene expression by „regulator genes“? Various mechanisms of gene regulation are known (e.g. CLARK 1997). Obviously many factors can play a role as triggers, such as light and temperature, influences by neighbouring tissue and the position in the organism, chemical concentration gradients and special signals. Regulation may be by transcriptional or post-transcriptional inhibition (GALLIE 1993), it may be cell specific (EDWARDS and CORUZZI 1990), tissue specific, or concern the whole organism (GALLIE and BAILEY-SERRES 1997), it may be the result of parental imprinting or of allelic interaction (MATZKE and MATZKE 1993). For the control of gene expression by regulatory genes it seems relevant to keep in mind, that this interaction should be subject to recombination during meiosis, unless regulation is via the promoter region of the gene, or the regulatory gene is very tightly linked to the controlled gene(s).
A rather well studied example of organ specific genetic regulation is the control of seed endosperm composition, where genes for storage compounds (proteins, starches, lipids etc.) function only during seed formation (GOTTSCHALK and MÜLLER 1982). Here regulation is also concerned with the coordination between embryo, endosperm and maternal tissue (RAY 1997). Well studied is also the regulation of anthocyanin formation e.g. in maize (JAYARAM and REDDY 1990). An example of genetic regulation concerning the whole organism would be the photoperiodic reaction: A single gene may decide, whether and when flowering is induced by a particular photoperiod (MICKE 1979) and this is a crucial decision, whether under natural conditions or in farmers field. Resistance against pathogens is another example (often studied but not so well understood). So-called ‘resistance genes’ recognize attack by a pathogen and signal activation to a cascade of host plant genes for defending the organism (e.g. JÖRGENSEN 1988, 1992). Formation of cell wall appositions, “suicide” of already infected cells („hypersensitivity“), toxin degradation, blockade of pathogen migration are defence measures handled „non-specifically“ by a number of ‘downstream’ genes („polygenic resistance“). But these genes have to be switched on by other genes, which are able to recognize specific ‘pathotype’ attributes. These ‘watchtower-genes’ are a very important acquisition for a plant, because it would be too costly (and probably create physiological problems) to keep all defence genes against all potential pathogen attacks active all life time. Unfortunately, the pathogens have acquired the known „gene-for-gene“ camouflage possibilities to trick out the plant’s ‘watchtower-genes’.
The interesting questions are: In which way are genes activated? How is their activity inhibited? How can a mutation interfere with these activation and inhibition processes?
Every gene must possess a „promoter“-sequence, without which the gene will never be transcribed. The promoter can be blocked, apparently also be mutated to be less promoting, its duplication may cause co-repression, but what initiates or terminates the promoter activity? There are external and internal triggers, but any message first has to find a receptor inside the cell, which then will produce a signal to activate and coordinate gene activities within a tissue, within an organism, at a given time, during a particular ontogenetic stage etc. These signals and their way of transmission are still quite a mystery. From artificial organogenesis under the influence of hormones in culture media one can learn something about somatic regulation of ontogenesis, but the in-vitro environment is rather a-typical. We know really still very little about regulation under natural conditions by systems, which reside in the genome, are inherited, and form a genetic network.
Thinking about induced mutations in this context, one has to keep in mind, that any new mutation is a single cell event, and therefore is irrelevant as long as the mutation is not phenotypically expressed in a sizeable tissue, or is part of the germ-line. Since the effect of mutagens is mainly gene inactivation or deletion, removal of an inhibition as well as the elimination of activation should be possible by a mutation in a dominant regulator gene. In the early days, some researchers were interested to revert recessive mutations to normal functioning dominant alleles. Many ‘revertants’ were obtained, but these were not true ‘backmutations’. They were recessive or semi-dominant mutations in other loci and on other chromosomes (REDEI 1969). Apparently recessive mutations can suppress the original defect of recessive mutations (TULEEN et al. 1966). Nowadays, mutations are being used to dissect the network of genes. By eliminating so far unknown genes, their normal functions are identified. In fact, progress in plant genetics as well as plant physiology depends to a large extent on the induction of recessive or „null-allele“- mutations in the model plant Arabidopsis thaliana (ANONYMOUS 1995).
Which genes are dealt with in evolution and in crop improvement?
Plant breeders in the past used to ascribe a particular trait of a plant or the performance in the field to a particular gene or group of genes, which are to be dealt with in crossing like MENDEL demonstrated more than 100 years ago. It was not taken into consideration, that genes looked at by the breeder indeed are either objects of regulation or are regulator genes themselves and that the genes introduced from parents may all be accumulated in the offspring, but do not function all the time in all the tissues, and would require particular provoking environmental conditions or triggers from outside/inside to be expressed.
All living beings, according to the degree of relatedness, have most of their genes in common. These are first of all the fundamental genes evolved some hundred million years ago, which are required for the basic cell compounds, such as amino acids, proteins, sugars, starches, fatty acids and for the basic cell metabolism , concerning energy, growth, coordination, multiplication etc. (OHNO 1970). Related species share most of the genome of a common ancestor, from which they started separation may be 50 million years ago. The genomes of cultivars belonging to the same species of course may differ only in the alleles of a few dozen genes. Geneticists discovered a striking syntheny among plant species, which means, that e.g. in the chromosomes of grasses like rice, barley and maize there are groups of identical genes in much the same order of sequence. Relevant differences mainly resulted from duplications and translocations within the genome of a common ancient cereal progenitor (KURATA et al. 1994; MOORE et al. 1995). Most of the molecular polymorphism detected within genes seems to be neutral, i.e. not altering gene function. For a breeder, however, only genes with functional polymorphism are of interest. These will not be the essential „structural genes“, but also not the basic „regulator genes“, which coordinate the standard cell and organism functions. It also will not be those genes that determine cell walls or turgor, stomata or photosynthesis, the development of vessels or the transport of nutrients, because manipulations in those genes can hardly be tolerated by any plant. Consequently, the breeder will be able to manipulate only the kinf of genes, which influence the course of events during ontogenesis of a crop plant and those, which are involved in the interaction of this organism with a particular environment. Breeding will be able to influence the timing of processes, the distribution of resources, the partitioning of products and a number of morphological traits. One should note, however, that all these characteristics also fall within the range of the plant’s epigenetic regulation flexibilities!
One may conclude, that plant breeders deal only with a certain portion of „regulator genes“, and this would also refer to experimental mutagenesis. It must be admitted, that in reality there might not exist a rigid separation between „structural“ and „regulatory“ genes. Among the former, some may be coding for a product of little importance for the survival of the individuum (e.g. flower pigment) and they could perhaps possess other temporary regulatory roles in the genetic network. The really essential „structural“ and „regulatory“ genes would be expected to belong to the extremely „conservative“ group of genes which exhibit a minimum of deviations (alleles) and so be practically exempt from manipulation by recombination. This group of indispensable genes probably became during the course of evolution well protected against mutations (spontaneous as well as induced), e.g. by duplications, or by particularly effective repair systems.
If one realizes this rather restricted potential manipulation of genomes by plant breeding, one may wonder, what happened during natural evolution. It seems, that valuable genetic systems once developed, were so well protected, that they became virtually „immortal“ (OHNO 1970). Recently the opinion gains ground that much of the natural evolution was based on altered regulations rather then on making new genes (SHUBIN et al. 1997). A most prominent example of how evolution (and domestication) may have worked, is the development of maize from its ancestor theosinte: One single regulator gene (tb1) was responsible for this phenotypically large step(DOEBLEY et al. 1997).
Do epigenetic mechanisms play a role in evolution and plant improvement?
Heritable alterations of regulatory genes by mutations have certainly played a crucial role during evolution. Their effect, however, to some extent resembles the epigenetic control of gene activities in cell metabolism and in the numerous organism functions during life time.
Evolution usually is thought to be based upon preformed random genetic variation, from which the fittest portion is selected “by nature”. However, epigenetic variation in regulatory systems would seem to be the faster, more effective and more flexible way for a plant to respond as an organism to edaphic or climatic changes in the environment, to temporary abiotic stresses, to competition by other plants, to attack by pathogens and pests, while at the same time maintaining and accurately replicating the accumulated most valuable ‘structural’ as well as ‘regulatory’ genes. In contrast to single cell mutations, which become relevant eventually in the next generation, epigenetic alterations almost immediately are expressed in tissues, organs or whole plants, and therefore are selectable in the same generation. Natural selection of better adaptable individuals and populations therefore can act without delay, favouring their survival or number of progeny. Thus epigenetic processes (although by definition not heritable) should have a profound effect on what goes into the next generation and therefore should have a tremendous influence on evolution. Could this influence be measured in comparison to the rather slow and vague process of accumulating spontaneous (undirected!) mutations, which require phenotypic establishment in part of a population before they can make a contribution of a better fitness, adaptation or competitiveness. The plant breeder, on the other hand, does not appreciate epigenetic variation, and when selecting desired phenotypes, routinely performs progeny tests to sort out the heritable variation from non-heritable.
Is there perhaps another connecting point between mutations and epigenetic regulation? The possibility should not be bluntly rejected (CULLIS 1986; PREM DAS and MESSING 1994). SCHWOCHAU and HADWIGER already in 1969 reported, that certain chemicals acting as mutagens were also unspecific inducers of gene activity. They assumed, that these chemical mutagens attack preferentially the promoter region of genes, because the rest of the gene may be covered by histones and only the regulator site readily accessible.
The idea of inheritance of acquired characteristics was discussed many times and was vehemently rejected. With current advances in molecular genetics the question may be brought up again from a different angle: Do epigenetic regulatory changes have a chance, somehow, to be transferred into the genome? Some epigenetic phenomena have already been found to be transmittable into the next generation, e.g. parental imprinting. Should nature really have thrown away the chance to take advantage in evolution of the striking flexibility, a plant population achieves through epigenetic adaptation of metabolic processes, of the plant architecture, of resistance etc. ? If so, epigenetic regulations at least gave a chance of survival and through this another generation of time for a required adaptation by the generally accepted genetic means such as mutation, selection and recombination.
What is the basis for beneficial induced mutations?
There were a number of futile attempts to link the successful performance of crop plant mutants to the mutagen and the applied dose by which they were originally induced, but there are hardly any investigations about the molecular changes of genes in the genomes of improved mutant cultivars, and only few concerning specific crop plant mutants.
Mutation experiments on the seed quality of cereals induced by various mutagens may shed some light on the issue. AMANO and co-workers studied carefully the changes induced in the waxy loci of rice and maize using EMS, UV, thermal neutrons, and gamma-rays. They looked at the expression of thewaxy gene in pollen and in the endosperm (AMANO 1981, 1985; YATOU and AMANO 1991). They have confirmed a destructive effect of all the mutagens used, but the kind and degree of destruction differed between the mutagens applied, from impairment of proper code transcription (mainly by EMS) to total inactivation or even complete deletion of the locus. Another interesting insight can be derived from studies on the effect of chemical mutagens on a particular locus in barley (Mla12), responsible for a „specific“ resistance to Erysiphe graminis (JÖRGENSEN 1996). Mla12 is a dominant gene whose function is to recognize attack by the mentioned fungus. The recognition is restricted to a particular product of the corresponding fungal „avirulence“ gene. Mutagenesis was tried to alter the dominant, resistance activating allele into a recessive or the null-allele and to see what happens. The (somewhat unexpected) results were: (a) mutants with mutations in the Mla12 locus causing reduced, but not completely lost ability to recognize the corresponding specific pathogen type; (b) mutations outside this locus, which reduced gradually the defence effectiveness. These mutations apparently occurred in genes (activated by the intact dominantMla12 gene) whose proper functioning is required for performing an effective defence. (For incomprehensible reasons the mutated genes were named suppressor (sup), although they were originally active defence genes, now to some extent damaged by a mutation).
Accepting it as a fact, that conventional mutagens have destructive effects on genetic material (MALUSZYNSKI et al. 1995), one could postulate, that those successful mutants mentioned before (e.g. listed in the FAO/IAEA Data Base) carry alterations only in genes, where even a „null-allele“ causes tolerable changes in the plant’s ontogenesis and physiology. But this does not make it plausible, why they became useful from the point of view of a crop breeder or farmer.
If one looks into the literature or at one’s own experiments, one would be able to list numerous examples of induced or spontaneous mutants, which could be explained by an alteration or complete loss of a regulation (e.g. MICKE et al. 1980; YAMAGATA 1981; GOTTSCHALK and WOLFF 1983; DONINI et al. 1984; KONZAK 1984; MICKE 1988). Some specific examples could be listed as follows:
– loss of daylength sensitivity in rice, cotton, castor bean (GUSTAFSSON and LUNDQUIST 1976; MICKE 1979; DONINI et al. 1984)
– loss of apical dominance leading to profuse branching in sweet clover (MICKE 1958; SCHEIBE and MICKE 1967)
– altered control over ontogenetic development, leading to short culm and increased tillering in cereals (MALUSZYNSKI et al. 1986)
– loss of control of root nodulation towards non-, inefficiently or abundantly nodulating pea, bean, chickpea, faba bean, soybean etc. (JACOBSEN and FEENSTRA 1984; CARROLL et al. 1985; PARK and BUTTERY 1989; etc.)
– altered control over endosperm composition in cereals (GOTTSCHALK and MÜLLER 1983; IIDA et al. 1993; VON WETTSTEIN 1995; HABBEN and LARKINS 1995)
– altered control over compounds deposited as reserves in cotyledons of oil seeds (GREEN and MARSHALL 1984; RÖBBELEN 1982) and of grain legumes (TAKAGI et al. 1989; BHATTACHARYYA et al. 1993)
– loss of control over regular leaf and floral organ development in sweet clover (SCHEIBE and MICKE 1967), in faba bean and in pea (GOTTSCHALK 1971)
– improved resistance of barley to powdery mildew by loss of Ml-o regulated inhibition of callose apposition (JÖRGENSEN 1992; MICKE 1992)
– increased resistance to rusts by deletion of inhibitor genes in wheat (KERBER and GREEN 1980; DYK 1982; WORLAND and LAW 1991; WILLIAMS et al. 1992)
– altered control over deposition or recycling of secondary metabolic products in lupin (MICKE and SWIECICKI 1988), in sweet clover (MICKE 1962 a and b; SCHÖN 1966), in rape seed (RÖBBELEN 1990).
Pleiotropy (often seen as a typical attribute of induced mutations) and its modification through crosses (GAUL et al. 1968) could reasonably well be explained by mutational changes in genes functioning as regulators for several other genes, but should then be distinguished from „downstream“ effects of mutant genes, acting „early“ in metabolic pathways.
It seems unlikely, that any of the induced mutations used successfully for variety development affected essential structural genes common to almost all forms of living creatures and preserved over millions of years. It seems also unlikely, that the mutations concern any other gene, that has assumed during the course of evolution an essential position in the gene network of the particular species, because such mutations would have resulted in loss of vigor and probably death (e.g. „chlorophyll mutations“). The mutations used by plant breeders most likely concern those genes, which are involved in genetic/environmental interaction, regulating the response to triggers by signalling activation or inhibition to other genes responsible for particular reactions (MICKE 1996). These regulating genes should be the ones categorized usually as „major genes“ because of their predominating role, in contrast to regulated genes which should be many, each of relatively minor effect on the phenotype (usually classified as minor genes or polygenes).
Considering the use of in-vitro culture techniques in mutation breeding, the following might be concluded: All genetic alterations caused eventually by physical or chemical mutagens can be expected also from ‘somaclonal variation’. In addition, one may expect activation of transposons, but this also suspected for some chemical mutagens. ‘Somaclonal variation’ will be compounded with epigenetic effects. This makes the use for crop improvement more difficult, particularly in asexually propagated plants, since crossing and reselection would be required to sort out desired mutations from other alterations. In-vitro selection is handicapped by a-typical gene expression. Use of haploids derived from anther culture has found its best application in the „doubled-haploids-technique“ , which leads faster to homozygosity for more effective selection.
Experimental mutagenesis has now a more restricted, but better defined place among the methods of applied genetics. In addition to the use of induced mutations in crop impropvement, there is now a most valuable use in fundamental genetics and plant physiology. This will benefit breeders through more effective strategies for crop improvement. However, if a breeder wants „new“ genes with „new“ gene products, he would definitely have to resort to gene engineering and use mutagen treatments as a supplementary tool..
Literature: see Mutation breeding references
The full text of this review was published in “Breeding in Crop Plants: Mutations & In Vitro Mutation Breeding”; Editors Bahar A. Siddiqui and Samiullah Khan; Kalyani Publishers, Ludhiana (India) 1999. p. 1-19. ISBN 81-7663-104-3.
A good source of useful information and publications on using induced mutations for crop improvement would be the Plant Breeding and Genetics Section of the Joint FAO/IAEA Division, P.O.Box 5, A-1400 Vienna (Austria) http://www.org/programmes/nafa
Address of the author: Dr. Alexander Micke, Salmannsdorfer Str. 94, A-1190 Vienna (Austria).