Chapter 25

Cisgenesis
Next Step in Classical Plant Breeding
Evert Jacobsen and Henk J. Schouten

Abstract Plant breeding is an ongoing activity that started many years ago through domestication of crops by farmers selection. After discovery of Mendels Law plant breeding became gradually more science driven. Nowadays, plant breeding is developing very rapidly because of the development of many new technologies and scientific disciplines that can be applied. Approaches for genetic modification (GM) of plants developed quickly in the eighties and nineties of last century, but it is the first technology that has not been widely accepted in the world by NGOs and consumers. GMO-regulations have been developed which are by the strict application obstructing the development of GM-varieties, especially in Europe. These regulations are based on the modification process and on transgenes originating from non-crossable species. These transgenes are a new gene pool for plant breeding. However, it turns out that cisgenes, which are genes from the plant itself or from crossable species, will be more and more available. They belong to the existing breeders gene pool but they are treated in the regulation like transgenes. It is recommended to exempt from the regulation GM-plants that contain cisgenes only. This chapter provides a historical context of cisgenesis. Further, it discusses breeding approaches of autogamous, allogamous and vegetatively propagated crops. Options for cisgenesis in these kind op crops are presented. Some examples are disease resistance in potato and apple using R- and Avr-genes, hybrid seed production using genes for male sterility, or S_RNase genes for changing self-incompatibility. We regard cisgenesis as next important step in introgression breeding, using natural genes spatie. Cisgenesis has also been compared with intragenics and induced mutation breeding. We recommend less stringent oversight for intragenic plants, compared to transgenic plants especially when it concerns RNAi.

E. Jacobsen () and H.J. Schouten Plant Breeding, Wageningen University and Research Center, Wageningen, The Netherland e-mail: evert.jacobsen@wur.nl E. Jacobsen Transforum Agribusiness & Rural Areas, Louis, Pasteurlaan 6, 2700 AB Zoetermeer, The Netherlands S.M. Jain and D.S. Brar (eds.), Molecular Techniques in Crop Improvement, DOI 10.1007/978-90-481-2967-6_25, Springer Science+Business Media B.V. 2010 591

592

E. Jacobsen and H.J. Schouten

25.1 Introduction 25.1.1 Domestication of Crops and Traits

25.1.1.1 Initiation of Domestication Domestication of the different crops has been a long-term process that started for our main staple food crops more than 10,000 years ago in the Middle East, Central America and South-East Asia. Initially, plant parts like fruits, seeds, tubers were collected in nature only. Wild plants around settlements were more frequently used. Some plant species became permanent residents of human settlements (camp followers) and a permanent food source. During a following phase growing conditions of camp followers were improved, as mankind prepared the soil and collected seeds for sowing. Selection of better individual plants started. Seed of domesticated material was collected and stored for next growing season. Many crop species gradually developed by selection for desired traits. During all development stages of agriculture from collection into modern cultivation, including fruit, flower and vegetable production in the field or in glasshouses, selection and adaptation to the new circumstances has been most crucial. Traits of importance for domestication were yield, flowering time, larger seeds or tubers, shattering resistance, absence of thorns or prickles and of bitter, poisonous compounds, uniformity, adaptation to modern agricultural practice, and many other traits. In some crops the genetic diversity is substantially reduced for alleles that have been particularly important for domestication. In maize that seemed to be the case for example for teosinte branched 1 gene controlling tillering and apical dominance (Buckler et al. 2001). 25.1.1.2 Farmers Selection Mankind has selected crop plants during thousands of years. This process of crop improvement was slow. The genetic aspects involved were poorly understood until the end of the 19th century. Farmers selection to local circumstances and needs resulted in so-called landraces. Landraces are still the plant material for many farmers worldwide. An interesting aspect in the evolution of domestication is the detection of natural introgression by farmer selection in different agro-ecosystems. It means that in addition to introgression of genes by humans through interspecific crossing, selection, and back crossings, also unintended natural introgression has occurred. There are indications (Jarvis and Hudgkin 1999) that it occurs both in autogamous (self-fertilizing) and allogamous (cross fertilizing) crops. Also natural translocation has been observed (Schmidt 2002). There are strong indications that the phenomenon of translocation is occurring in wild plants as well as in crop plants (Lim et al. 1998). It means that human introgression as well as (induced) translocation procedures are not principally different from spontaneous processes occurring in nature. At present, farmers selection and management of local crops is still very important.

25 Cisgenesis

593

This activity is imbedded in forms of traditional plant breeding. Over 40% of the worlds agricultural area is still under management and seed supply of small-scale farmers (Jarvis and Hodgkin 1999). 25.1.1.3 Professional Cross Breeding Rediscovery of Mendels laws at the end of the 19th century led to a shift to more science-based plant breeding. It stimulated professionalization of plant breeding and development of seed business. Combining useful traits by crossing and selection on the basis of classical genetics influenced crop improvement strongly, and led to more rapid selection for superior types. The impact of classical genetics on professional plant breeding started in the early twentieth century, but is still ongoing. It has stimulated domestication of crops in many ways. Selection in plant breeding implies reduction of genetic variation. However, also widening of genetic variation is required for novel traits, such as resistances to biotic or a-biotic stress, further yield improvement, hybrid seed production and new quality traits. Additional genetic variation can be found in crossable relatives. The pool of genetic variation can be further extended by techniques such as embryo rescue, protoplast fusion, and recently genetic modification. In addition to domestication of crops by normal cross breeding, in which several traits from both crossing parents will be combined, it is also important to introgress specific or single traits from wild material into crops. Introgression and in some cases induced translocation breeding of specific traits simultaneously introduces so many unwanted alleles from the wild plant that pre-breeding is required, in order to remove the majority of these unwanted alleles. Unwanted alleles that are genetically tightly linked to the desired allele can bring serious problems. The problem of linkage drag with negative traits is frequently the main bottle neck in such approaches. Solving linkage drag problems may require many years of crossing and selection. Molecular markers can nowadays be helpful to reduce linkage drag problems or to speed up the solution, but it never completely removes the linkage drag. An example of linkage drag is insect resistance in lettuce that was closely linked to compact growth and rapid aging. Jansen (1997) described how Marker Assisted Breeding (MAS) was helpful to solve this linkage problem. 25.1.1.4 Cisgenic Breeding A sometimes more efficient solution for introgression of alleles from wild germplasm into crops is cisgenesis (Schouten et al. 2006a). In this approach, the allele from the wild relative is isolated molecularly from the genomic DNA of that wild relative. Subsequently, this allele is transferred to the recipient crop plant through marker free genetic modification techniques (Fig. 25.1). Isolation of the allele from the donor plant is crucial here, as it is not flanked by undesired alleles anymore. By definition, cisgenesis prevents linkage drag. This is the major advantage of cisgenesis.

594

E. Jacobsen and H.J. Schouten

a
RB Gen

The Standard Vector, pMF1000

35S Recombinase 35S CodALB

+ dexamethasone (dex) T-DNA region RB Gene L

b A.tumefaciens without selection genes c Mixed inoculation followed by genetic segregation

Fig. 25.1 Marker-free transformants can be obtained in three different ways (a) Normal transformation on selection medium followed by induced excision of the marker gene and helper genes using e.g. the standard vector pMF 1,000. The diagram is depicting a right and left border (RB, LB), the cisgene of interest, two recombination sites (blue triangles), and between these recombination sites the selection gene NptII and a recombinase construct, both controlled by 35S promoters. After transformation and selection, the whole segment between the recombination sites is removed on recombinase-mediated excision induced by dexamethasone (Dr. F.A. Krens, Plant Research International, Wageningen). (b) A marker-free A. tumefaciens strain with only agricultural gene(s) in the TDNA is used during transformation. Transformants are selected among many regenerants by PCR. (c) Co-inoculation with two A. tumefaciens strains containing the selection marker or the agricultural cisgene(s), respectively. The marker-free transformants can be selected among the segregants after a sexual step by selfing or back crossing with a wild type plant

Further, by definition, the cisgene contains its native introns and is preceded by its native promoter and followed by its native terminator (Schouten et al. 2006a, b). The cisgenic insertion contains only one or a few cisgenes and is not surrounded by other genes of the donor plant, nor by genes from a vectors backbone. Here domestication of the cisgene is obtained by surgical precise isolation of desired alleles from donor plants, whereas in introgression breeding many times uncontrolled meiotic recombinations of linked genes are required for removal of negative side effects. Cisgenesis requires methods of genetic modification that do not leave behind transgenes for selection of transformed cells, for example coding for herbicide tolerance or for resistance to antibiotics. As shown in Fig. 25.1, nowadays different methods are applied to obtain such marker-free plants (Fig. 25.1). The first possibility is the removal afterwards of the selection genes and other helper genes by recombinase-based excision (Schaart et al. 2004). The second approach is transformation without a selection gene, but screening by means of PCR for transgenic regenerants containing the insert (Vetten de et al. 2003). The third approach is infection with two Agrobacterum strains, the first one containing the selection marker and the second one the agricultural cisgene(s) of interest (Yu et al. 2006). After sexual crossing with wild type plants or after selfing, cisgenic plants can be

25 Cisgenesis

595

selected from the segregating offspring. Plants with only the cisgene(s) are selected and the transgenic plants with only the selection gene or with both selection gene and the cisgene(s) are discarded.

25.2 Intragenics, RNAI and Induced Mutation Breeding

Cisgenesis is very restricted regarding the allowed gene sequences, keeping as close to the classical breeding as possible. Another approach is intragenics. In intragenics, different DNA fragments from the breeders gene pool are put together into new combinations, such as the coding sequence from one gene, and a promoter from another gene, and are inserted by means of genetic modification techniques into the recipient plant. By making novel combinations of native coding sequences and native regulatory sequences, the expression pattern of genes can be changed. In this sense intragenics is less strict and not as close to classical breeding as cisgenesis. In case of intragenics, all tDNA is derived from DNA from crossable relatives, even the border sequences. This approach has been described by Rommens (2004) and Rommens et al. (2007). Intragenics is also based on a marker-free approach. Silencing by means of RNAi can also be obtained by means of intragenics, in case the whole tDNA sequence is derived from plant DNA. RNAi is a very important approach mimicking natural or induced loss of function mutations in plants. A major difference between induced mutation breeding and the RNAi approach is that loss of function by means of a mutation is inherited recessively, but RNAi inherits dominantly and can therefore also be used in polyploid and vegetatively propagated crops. The example of simultaneously improving storage and processing characters in processing potatoes is clear (Rommens et al. 2006).

25.3

Regulation of Cisgenesis and Intragenics

As cisgenic plants are very similar to traditional plants, and at least as safe as traditionally bred plants or plants from (induced) mutation breeding or induced translocation breeding, Schouten et al. (2006a, b) and Jacobsen and Schouten (2007) have proposed to exempt cisgenesis of plants from the regulation on deliberate release of GMOs into the environment, therewith clearing cisgenic plants in a timely and cost-effective manner. Rommens et al. (2007) proposed this also for intragenic plants. The main reason is the source of the genetic material, which is within the species or within sexual compatible species, available to conventional plant breeding. The main difference between cisgenesis and intragenics is that cisgenesis used natural genes including their native promoter, whereas intragenics allows also novel combinations of promoters and coding sequences, present in the plant and which are or can be used in traditional plant breeding.

596

E. Jacobsen and H.J. Schouten

Table 25.1 Proposed changes for notification in GMO-regulations with respect of transgenes, cisgenes or intragenes in plants. The transformation process brings per definition a GM-plant. The DNA source determines the way regulation should be applied. Transgenes are partly from non crossable species, but intragenes and cisgenes are from the species itself or from crossable species. A. Baseline is natural complete genes. B. Baseline is natural functional sequences such as promotors, coding parts and terminators GMO regulation Categories Type of genes A B 1 New transgenes Full Full 2 New events in existing gene-crop combination Partial Partial Intragenes Partial Exempted 3 Cisgenes Exempted Exempted

Table 25.1 shows a proposal for deregulation of cisgenic and intragenic plants. Both in cisgenesis and intragenics, the genes used belong to the gene pool of the conventional breeder. The conventional breeding is regarded as the baseline for the GMO regulation. As cisgenesis uses natural genes with their native promoters from the breeders gene pool, it makes sense in treating cisgenic plants in a similar way as traditionally bred plants (Schouten et al. 2006a, b; Jacobsen and Schouten 2007). Intragenics uses the same gene pool too, and therefore should be deregulated too. However, expression patterns may be obtained that are very unlikely through conventional breeding. Intragenics can therefore result into plants that cannot be obtained by means of conventional breeding. Therefore we propose a partial deregulation for intragenics. RNAi is in this sense a special application of intragenics, mimicking natural or induced mutations. Also in this case deregulation is logic. It results in loss of function of genes that may also be obtained by means of induced mutations. In case RNAi is designed for a sequence that appears at more loci of the plants genome, such as a conserved domain of a gene family, it may interfere with a group of genes. This exceeds the effect of a mutation in one locus. Therefore we propose here partial deregulation. In Table 25.1 we show also the situation in which functional sequences as promotors, coding parts and terminators of intragenes are seen as smallest natural units for exemption. In this case intragenes as well as cisgenes would be exempted.

25.4 Autogamous Crops

This group of plants contains many important annual crops such as cereals, sweet pepper, tomato, pulses and lettuce. They can be subdivided into real diploids like tomato, barley, rice, pea, soybean and lettuce and more complex disomic allopolyploids like wheat and oat. We will restrict this part to diploid autogamous crops. The genetic variation in these crops is mainly found within the species, varying from other varieties to wild material or to a restricted number of related species. All these mentioned crops are advanced regarding their breeding development.

25 Cisgenesis

597

25.4.1 Introgression and Pre-Breeding

Improved varieties have to combine more and more traits not only originating from their indigenous germplasm within the species and existing varieties but for an increasing number of traits, like disease resistance, quality traits and male sterility, also from crossable wild relatives. In practice this is introgression breeding by interspecific hybridization followed by back crosses with the cultivated plant under simultaneous selection for the trait of interest as shown in Fig. 25.2. Bottleneck in this approach is frequently linkage drag with neighboring wild alleles flanking the allele of interest (Jansen 1997). Because of the great number of co-introgressing unwanted alleles from the wild germplasm, pre-breeding is often performed. Pre-breeding activities are frequently focused on solving the problem that other negative traits are genetically linked to the gene of interest. MAS is a nice way to speed up this process of domesticating a trait, however, a lot of efforts are needed before it can be applied properly (William et al. 2007). The problem of linkage drag is increasing if simultaneously several traits have to be introduced from different wild plants, specifically from less related species and/or when these traits are linked on the same locus of the same chromosome in different backgrounds.

a
r

Introgression breeding Receptor r x R Donor R

Induced translocation breeding Receptor r r x R Donor R

c
Receptor r r

Cisgenesis R- gene + from crossable plant, in Agrobacterium tumefaciens

Receptor r r x r

F1 R r

Receptor r x r

F1 R r R r R r r

BC1,2, ... BC1 Addition line

R irradiation

BCn R r R r r

BC and/or selfing R r r
TRENDS in Biotechnology

Fig. 25.2 The most important steps in introgression, induced translocation and cisgenic breeding with a resistance (R) gene. These result in (a) introgression, (b) induced translocation of the R-gene with linkage drag, and (c) cisgenic R-gene insertion without linkage drag (Jacobsen and Schouten 2007)

598

E. Jacobsen and H.J. Schouten

The availability of molecularly isolated genes and marker-free transformation methods would help breeding. For seed propagated crops co-inoculation with two Agrobacterium tumefaciens strains followed by genetic segregation of the selection marker and the agricultural traits, is a good option for obtaining marker-free transformation. It is known that T-DNA from both strains are frequently inserted at different chromosomes in the same transformed cell, so that they will segregate independently after crossing (Yu et al. 2006).

25.4.2 Induced Translocation Breeding

In allopolyploid crops, which are more complex, introgression breeding is not always possible. In examples like wheat, cross breeding is usually restricted to wild material from the crop species itself or to one of the parental species. If other species are needed, introgression breeding is frequently more complex. Main reason for that is the allopolyploid nature of this crop plant. The allopolyploid nature is connected to disomic inheritance of traits instead of polysomic inheritance in autopolyploid crops. In wheat disomic inheritance is occurring because of the Ph-gene (Friebe et al. 1996) which is suppressing interspecific chromosome pairing. The presence of this Ph-gene is not only suppressing interspecific chromosome pairing of the different genomes involved in the existing crop but also from other species during the process of introgression breeding. Suppression of chromosome pairing in interspecific hybrids of wheat, with species like Aegilops umbellulata, Agropyron elongatum, Aegilops speltoides (Jacobsen and Schouten 2007) is also obstructing considerably introgression of wild genes into the allopolyploid genome. This problem can be solved in different ways, such as inducing interspecific chromosome pairing by removing the Ph-gene temporally by nulli 5B-tetra 5A compensating genotypes or by selecting for natural suppression of this trait (Friebe et al. 1996). Fig. 25.2 shows induced translocation as another commonly used solution. Basic aspect is irradiation of monosomic addition lines that have been obtained by interspecific hybridization followed by back crosses with wheat under simultaneous selection for the novel trait of interest. The alien monosomic addition line is bearing the gene of interest. After irradiation of the monosomic addition line and selfing, resistant offspring is selected that lack the alien chromosome but is still possessing the functional alien resistance gene, translocated into the wheat genome. This induced translocation event normally comprises of a piece of alien chromosome with many neighboring donor genes, including the desired resistance gene. Linkage drag is also in this case a negative side effect from the method used. If negative traits are linked to the gene of interest, it is not easy to remove them by meiotic recombination, which is suppressed at such a location. The same holds true for the introgressed piece of alien chromosome after suppression of homoeologous chromosome pairing during the absence of the Ph gene.

25 Cisgenesis

599

The cisgenic approach could also in this case be very helpful. It is a one step approach instead of all complicated steps described above. Precondition is the availability of the gene of interest and transformation ability of the variety which has to be improved. In the case of complex allopolyploid crops the advantage of cisgenesis is highly evident. It will improve breeding power of this type of crops considerably.

25.4.3 Cisgenesis and Its Potential Use in Further Breeding

After cloning of the alleles coding for the desired traits, cisgenesis is an alternative for cross breeding as well as introgression and induced translocation breeding. It is simplifying, as indicated in Fig. 25.2, normal introduction, introgression or translocation of such traits. It is specifically of interest if more than one trait, coming from different breeding parents or species, need to be combined at the same moment. As first step an existing transformable top variety, which has to be improved, is needed. If several cisgenes are introduced, using a gene cassette with these cisgenes, then during subsequent cross breeding these cisgenes are genetically tightly linked together. This is another major advantage. In this way, quantitative traits with a polygenic basis can obtain more attention and be more extensively handled in cross breeding.

Table 25.2 Breeding characteristics of some major crops Crop species Mode of reproduction Type of cultivar Apple Vegetative Cultivars Asparagus Allogamous open pollination (dioecious) F1 Hybrids Barley Wheat Cucumber Maize Autogamous Allogamous monoecious Allogamous (monoecious) Vegetative Sexual Allogamous Autogamous Landrace Modern Cultivars F1 hybrids Landrace and open pollination F1 hybrids Inbred lines Cultivars True potato Seed Open pollination Old/modern Cultivars F1 hybrids

Homogeneity Homogeneous Heterogeneous Homogeneous Heterogeneous Homogeneous Homogeneous Heterogeneous Homogeneous Homogeneous Homogeneous Heterogeneous Heterogeneous Homogeneous Homogeneous

Zygosity Heterozygous Heterozygous Heterozygous Homozygous Homozygous Heterozygous Heterozygous Heterozygous Homozygous Heterozygous Heterozygous Heterozygous Homozygous Heterozygous

Potato Rye Tomato Rice

600

E. Jacobsen and H.J. Schouten

25.5 Hybrid Varieties in Allogamous Crops

Many allogamous crops are more and more pushed into the direction of hybrid varieties. The classical way of breeding is based upon open pollination. In the allogamous crops we will restrict to breeding for hybrid seed varieties. Table 25.2 is representing 1. autogamous crops like barley with landraces and modern varieties and tomato with old varieties and F1 hybrids, 2 allogamous crops, a. vegetatively propagated crops like apple and potato with cultivars based on one genotype, and b. seed propagated crops with landraces, open pollinated varieties, F1-hybrids and inbred lines. The allogamous crops can be subdivided in monooecious like cucumber and maize or dioecious like asparagus. For hybrid seed production male sterility, based on GMS (genetic male sterility) or CMS (cytoplasmic male sterility), and/or gametophytic or sporophytic self-incompatibility are important traits.

25.5.1 Male Sterility in Traditional Breeding

In nature many types of male sterility are found. For breeding the most useful form is based on nucleocytoplasmic male sterility (CMS). In, for example, Brassica vegetables like Chinese cabbage, which are grown for their vegetative parts, the hybrid plant can still be male sterile without influencing yield negatively. In other crops like maize and rice, where seeds are the end product, restoration of male fertility is needed. In such cases, restorer genes (Rf) are very important. In practise, these Rf genes are complicating variety development considerably. In recent years a lot of knowledge has been obtained about the molecular genetic base of CMS as well as Rf genes (Wise and Pring 2002). This brings the possibility of manipulating CMS and its restoration by the GM approach very near, specifically when cisgenes could be involved. In several species like maize (T-urf13), bean (pvs-orf239), Brassica (orf138) and Petunia (pcf) the CMS-genes involved have been identified (Pelletier and Budar 2007; Bentolila 2002). These genes are in all cases resulting from complex DNA rearrangements of mitochondrial DNA. It is remarkable that here single mitochondrial genes are involved in male sterility. These natural MS-genes are available for GM-approaches and genetic modification of mt-DNA is the best option, however, mt-DNA based transformation has still to be developed. In seed propagated crops, like maize and rice, hybrid varieties are dependent on the presence of a restoration system. In breeding of hybrid varieties of maize, Rf genes have been used already for many years. However, as shown in Fig. 25.3 it is always complicating the breeding system considerably. In addition to the mother line, production of the hybrid variety by using the CMS mother line in cross combination with the fertile Rf free maintainer line, which contains normal cytoplasm, it is important to have a Rf containing father line for making a

25 Cisgenesis
Production of females

601

[Cyt: S; Nuc: A,rf/rf ] Mother line of hybrid

[Cyt: N; Nuc: A,rf/rf ] Maintainer line A Production of hybrids

[Cyt: S; Nuc: A,rf/rf ] Mother line of hybrid

[Cyt: S or N; Nuc: B,Rf/Rf ] Father line of hybrid

[Cyt: S; Nuc: AB,Rf/rf ] F1 hybrid

Current Opinion in Biotechnology

Fig. 25.3 The production of male fertile F1 hybrid seed by the use of cytoplasmic male sterility in cms Mother lines, in combination with male fertile Maintainer lines with normal cytoplasm and male fertile Father lines with normal or cms cytoplasm which are homozygous for the restorer gene (Rf) (Pelletier. and Budar 2007)

fertile hybrid. Crossing of this father line with the CMS mother line delivers male fertile hybrid seed with CMS cytoplasm but heterozygous for the restorer gene (Rfrf). In recent years, Rf genes of Petunia (Rf-PPR592; Bentolila et al. 2002), rice (Rf1A and Rf1B; Wang et al. 2006) and maize (Rf2-R213; Liu et al. 2001) have been isolated and can be used in cisgenic GM-approaches.

25.5.2 Transgenesis for Introduction of Male Sterility

Male sterility has already been manipulated for a long time by using transgenes. Most important example is the BAR-STAR system (Reynaarts et al. 1993), successfully applied in Brassica napus, and used in combination with herbicide resistance. With the Barnase gene (an RNase gene from Aspergillus oryzae) GMS can be introduced by using a tapetum specific promoter from tobacco. The Barstar gene, (an RNase inhibitor gene from Bacillus amyloliquefaciens) also driven by the tapetum promoter, is inhibiting the Barnase gene and restoring male fertility. The use of both transgenes is enabling hybrid seed production system in cultivated plants, providing Brassica napus a high increase in yield. However, the disadvantage is, that this system is fully based on transgenes.

602

E. Jacobsen and H.J. Schouten

25.5.3 Cisgenesis and Introduction of Male Sterility

Classical hybrid seed production has been applied in more and more crops. As shown above, the use of CMS in combination with restoration of fertility takes a lot of energy and breeding effort.Manipulation of both traits by using cisgenes is possible, as has been shown above by the molecular isolation of natural mt-based ORF-genes causing male sterility, and of natural nuclear Rf-genes restoring male sterility. The GM-approach would bring much more flexibility to the classical breeder by adding both traits directly to existing, good combining, lines. The GM-approach with Rf- cisgenes is already possible now. The introduction of ORF-cisgenes causing CMS into mt-DNA is not possible now, however, introducing them into chloroplast-DNA is today already a realistic option. The first example of engineering male sterility via transformation of the chloroplast genome is expression of a transgene coding for beta-ketoliase (Ruiz and Daniell 2005). A remarkable observation was that reversibility of the CMS phenotype could be obtained under continuous illumination of the transgenic tobacco plant. In the cases where engineering of mt-DNA and/or chloroplast-DNA transformation is involved, a GM-plant is obtained, however, in case the genes used belong to the existing breeders gene pool it should be exempted. In case of the availability of cisgenic Rf genes the same breeding scheme could be followed as shown in Fig. 25.3 however, the Rf containing father line genes could be obtained more easily.

25.5.4 Self-Incompatibility and a Possible Role of Cisgenesis

Self-incompatibility is promoting outcrossing. It is one of the most important ways in plants to prevent self-fertilization and inbreeding. For breeding of cross-fertilizing crops, self-incompatibility can be a serious hindrance and in plant breeding there are mainly two systems of self-incompatibility, i.e. gametophytic (GI) and sporophytic (SI) self-incompatibility. The GI system can be based on one locus (cabbage, red clover) with many alleles or on two loci (grasses). The gametophytic one-locus selfincompatibility is molecularly well investigated. It results from the interaction between the haploid genotype of the pollen grain and the diploid genotype of the style. Inhibition is found in the style. The S-alleles in the style work independently. It is known that polyploidization restores self-compatibility. Plants belonging to the Compositae family do possess the SI system. This is usually based on one gene with multiple alleles. The interaction is dependent on the diploid maternal genotype of the pollinator and the diploid genotype of the style. Inhibition is found on the stigma. Reaction of the pollen is determined by the dominance relationship of the sporophyte and not by the gametophyte. For hybrid breeding the SI system is frequently used in vegetable crops of Brassica oleracea. Molecular research on GI of, for example, Petunia inflata brought new possibilities for influencing self-incompatibility (McClure 2006). The incompatibility system is complicated but is S-RNase based. Petunia transformed with the PiSLF2 cisgene showed in a heteroallelic background, as earlier observed after polyploidization, breakdown of GI but not in a homoallelic background. It is clear from recent

25 Cisgenesis

603

Fig. 25.4 Schematic representation of transformation experiments to ascertain the function of PiSLF. (a) Self-incompatibility behaviour of an S1S1 transgenic plant carrying a single copy of the PiSLF2 transgene. (b) Self-incompatibility behaviour of an S2S3 transgenic plant carrying a single copy of the PiSLF2 transgene. The genotypes of pollen produced, the predicted S-genotypes of the progeny resulting from self-pollination, and inheritance of the transgene are indicated (Sijacic et al. 2004)

literature that unraveling of the whole GI system is complicated but at the other hand the results with the cloned S-alleles in GM-plants of Petunia inflata (Sijacic et al. 2004), as seen in Fig. 25.4, showed that these cisgenes can be used for inhibiting self-incompatibility. This possibility is based on the earlier observation that two different S-alleles present in one pollen grain prevent the incompatibility reaction in the style. It means that for hybrid varieties, inbreeding is more feasible with additional S-cisgenes. If transformation is not an obstacle, these cisgenes could be used for developing inbred lines and/or seed producing hybrid varieties.

25.6 Vegetatively Propagated Crops 25.6.1 Traditional Breeding

There are many important vegetatively propagated crops like potato, cassava and other root and tuber crops, fruit crops like apple, pear, plumb, peach, banana, ornamentals like Chrysanthemum, Alstroemeria and rose, and bulb crops like tulip, lily,

604

E. Jacobsen and H.J. Schouten

onion (shallot). All these crops have in common that varieties are consisting of one genotype which is vegetatively propagated by roots, tubers, bulbs, shoots, rhizomes or grafts. These crops are annual or perennial. They have also in common that these crops are genetically highly heterozygous and that crosses are recombining all combined traits immediately so that the parental genotype is lost and can never come back in the same combination via backcross procedures. In all these crops it is very well known that variety breeding demands increasingly numbers of seedlings in order to be able to find seedlings combining more desired traits. Nowadays, in potato over 1,00,000 seedlings are needed for one new variety. Genetic variation has also to come more frequently from wild species so that pre-breeding for breeding parents is a very important first step in the breeding process using wild plants. Success of pre-breeding is in these cases also highly dependent on linkage drag problems with presence of undesired alien traits. It is important to realize that improvement of existing varieties in traditional breeding is only possible by spontaneous mutations or by mutation induction. This way of improvement is frequently used in fruit crops like apple and in ornamentals like Chrysanthemum, Alstroemeria and rose and bulb species like tulip, crocus and narcissus, where changes in appearance are directly useful. Loss of function of specific genes can also be obtained by the RNAi approach, This is very powerful as has been shown, for example, for amylose-free potato starch (Heilersig et al. 2006). However, the outcome is a GM-plant which is based on a transgene, which could nowadays be indicated as an intragene if all functional parts are coming from natural gene sequences of the crop plant itself or from crossable species. In case of an RNAi intragene the deregulation could also be made less costly and less time consuming because all functional parts of the intragene are belonging to genes from the plant itself or from crossable species.

25.6.2 Extension of Genetic Variation

In agricultural practice and post harvest technology gain of functions, such as those influencing quality or resistances to biotic stress factors are very important. These traits can be found e.g. in wild material. The most important problems that have to be solved in this way are susceptibility to many diseases. In apple, resistance to apple scab (Venturia inaequalis) is at this moment a hot item and in potato resistance to late blight, nematodes and wart disease. Resistances to these diseases have to come from wild species and in practice several genes resisting different isolates of the same pathogen have to be combined, which is called gene stacking. These two examples of potato and apple will be discussed below in more detail.

25.6.3 PotatoPhytophthora Interaction

In a case like Phytophthora infestans it is very well known from the past that resistance can be overcome by the pathogen relatively easily (Haverkort et al. 2008).

25 Cisgenesis

605

Until the fifties Solanum demissum was used as main source of resistance to Phytophthora without sustainability in success. All 11 R-genes from S. demissum were broken and attention was shifted to horizontal resistance which is more based on quantitative traits. In potato this approach appeared also to be without sufficient amounts of success. Only a few varieties have been obtained with quantitative resistance. The present solution is searching for resistance in many different crossable species. These resistances do segregate well in crosses with susceptible plants and are effective against different complex isolates. In this way, R-genes are selected which in combination with other major resistance genes could be the basis for more durable resistance (Jacobsen and Vossen 2009). Sources of resistance in different species are found and needed, which brings more often introgression problems because of a lower degree of crossability and linkage drag problems because of decreased cross-over events in the homoeologous parts of the chromosomes of backcross plants during meiosis. Introgression breeding in vegetatively propagated crops brings the same problems of linkage drag as shown above for self-fertilizing crops. Fig. 25.5 shows how many years are needed for a new variety in traditional breeding when (double) bridge interspecific crosses are required to transfer major resistance genes into normal breeding material. In case of bridge crosses with far related species it is expected that during backcrossing meiotic cross-over events occur less frequently in the introgressed areas. This is generally increasing the size of introgressed alien chromosome parts and the chance of negative side effects by other linked alien genes. In case of potato, it has taken over 50 years before varieties with the first R-genes of S. bulbocastanum against Phytophthora could be released as recently has been done with the new dutch varieties Toluca and Bionica (personal comm.). This highly resistant variety probably contains only Rpi-blb2 (unpublished observations) which still provides in the Netherlands total resistance however, in trap fields virulent isolates of late blight have already been isolated (Kessel, pers. comm). Therefore, only one R-gene in a variety is dangerous because of potential breakage problems. The philosophy has to be to stack several major R-genes from different sources at the same time in one genotype (Jacobsen and Vossen 2008). In classical breeding this leads to accumulation of linkage drag problems and, therefore, Double-bridge crosses in potato for resistance to Phytophthora infestans
S. acaule 4x S. bulbocastanum 2x (R genes) AB hybrid 3x colchicine doubling AB hybrid 6x S. ph ureja 2x ABP hybrid 4x S. tuberosum 2x ABPT material 4x R -gene + linkage -drag
Fig. 25.5 Double-bridge crosses in potato introgression breeding for resistance to Phytophthora infestans. This multiple step approach with far related species is difficult and always accompanied with a lot of linkage drag around the donor resistance gene. Stacking of R-genes from different sources, without linkage drag problems, is complex

606

E. Jacobsen and H.J. Schouten

decreasing the chance of providing attractive varieties in a reasonable time. The tetraploid and heterozygous nature of potato further complicates the combination of desired traits in a new variety. Alternatives like stacking durable R-genes via cisgenesis are in development in order to solve this problem. At the moment 15 R-genes from several species like S. demissum, S. bulbocastanum, S. stoloniferum, S. papita and S. venturii have been isolated and are tested with different isolates for their spectrum of resistance. The use of Avr-genes is also speeding up the detection of R-genes in other species with the same or nearly the same specificity. It is getting more and more clear that R-genes can be subdivided into different groups with various resistance mechanisms (Vleeshouwers et al. 2008). These isolated R-genes can be stacked more easily by cisgenesis, and the availability of the corresponding Avr-genes enable testing the simultaneous biological expression of these R-genes in one plant. The possibility of using R-genes in GM-plants is also promoting the development of new resistance strategies as has been applied in the past with Bt based insect resistance genes by using refuges (Babu et al. 2003). In potato three resistance strategies have been described which could be used in practice (Jacobsen and Vossen 2008). It is one by one using single R-genes, stacking R-genes in one variety or using different R-genes in different clones from the same variety in mixed varieties.

25.6.3.1 AppleVenturia Inaequalis Interaction In 1946 crosses were made for introduction of resistance to apple scab (Venturia inaequalis) into commercial apple varieties, using as source of resistance the crab apple Malus floribunda 821 (Hough et al. 1953). The progeny of the cross between M. floribunda 821 and susceptible cultivars segregated in a Mendelian fashion for resistance in a 1:1 ratio. The gene putatively underlying this resistance was named Vf-gene. However, the fruits of the resistant parent M. floribunda 821 were very small, approximately 1 cm. The apples of the progeny were also small, and did not have the fruit quality that was required for commercial cultivars. This was caused by linkage drag: not only the desired resistance gene was inherited to part of the progeny, but also many unwanted alleles leading to poor fruit quality and other unwanted traits. In order to get rid of the unwanted alleles, subsequent crosses had to be carried out between resistant progeny and susceptible high quality cultivars. As indicated in Fig. 25.6 about five generations were required to remove enough unwanted alleles from M. floribunda, yet keeping the desired Vf-gene for scab resistance. Approximately 50 years after the first cross, Vf-cultivars with a reasonable fruit quality were introduced onto the market (Anonymous 1999). Therefore, it has taken half a century to introduce the Vf-gene and remove the linkage drag to an acceptable degree. In the mean time, Venturia inaequalis strains have been detected that are able to infect Vf-cultivars (Parisi et al. 1993). Especially in North-western Europe these strains are present and have spread (Parisi et al. 2006). As a result, several orchards that consist of Vf-cultivars have to be sprayed like orchards with susceptible

25 Cisgenesis

607

Fig. 25.6 Breeding scheme between apple and Malus floribunda 821 for introgression of the Vf resistance gene against apple scab (Venturia inaequalis) providing after 50 years the resistant cultivar Santana

cultivars (Trapman 2006). Fifty years of breeding is fading away in 10 years. Obviously, more individual resistance genes need to be accumulated for obtaining durable resistance. Fortunately, many loci that confer resistance to apple scab have been discovered in Malus, both major genes and QTLs (Calenge et al. 2004; Schmidt and Van de Weg 2005; Gessler et al. 2006; Gardiner et al. 2006). Therefore, sufficient genes for resistance are present in the germplasm of apple. Introgression of one resistance gene took approximately 50 years. Introgression of four or more genes for durable resistance will require more time, when the breeding is performed in the classical way. Would this imply that we are left with an additional 50 years of intensive fungicide applications to scab in apple, before the durably resistant cultivars are introduced? We regard it as our challenge to shorten this period significantly, by introducing the resistance genes and preventing the linkage drag. One way to speed up the breeding process is marker-assisted breeding. This method can be of tremendous use and we advocate this approach, but still this will be timeconsuming, because linkage drag has to be removed through crosses and meiotic recombination. Accumulation of resistance genes from four sources of resistance and sufficient removal of unwanted alleles from the same sources, will probably require at least another five cross generations of apple. As long as the juvenile

608

E. Jacobsen and H.J. Schouten

period and additional evaluation time in apple is about eight years, this would require a minimum of 40 years of breeding. An alternative route is introduction of the resistance genes into susceptible elite cultivars without simultaneous introgression of unwanted alleles, so prevention of linkage drag, rather then removal of linkage drag. Through this process of cisgenesis, durable resistance provided by several resistance genes is added to high quality cultivars in one step, preserving the proven fruit quality and other desired traits of these cultivars. Currently, apple is being sequenced, and other fruit crops probably will soon follow. This provides unprecedented opportunities for identification of genes. In addition, numerous loci have been mapped genetically in diverse germplasms, including fruit crops (Kole 2006). The information on genetic positions on the linkage groups, together with the whole genome sequences, and knowledge of genes from model plant species, offer us great opportunities to isolate alleles for desired traits at an increasing efficiency. We expect that, in the coming ten years, a vast number of major alleles for desired traits will be isolated in many crops, including fruit. So, the treasury of isolated alleles for cisgenesis will be filled at an increasing rate. The already mentioned Vf-gene was isolated by means of map-based cloning and subsequently functionally analyzed (Belfanti et al. 2004). A closer analysis revealed that a tandem repeat of two gene copies provides the Vf-resistance (Malnoy et al. 2007). Therefore, the spelling Vf-gene should be updated to the plural form Vf-genes. In the mean time several other resistance genes to apple scab are being isolated. As soon as the apple genome sequence becomes available to the scientific community, many more genes and their alleles will also be isolated and characterized, and will enrich the wealth of available alleles for cisgenesis. After discovery of the Vf-genes, several research groups in Europe and the USA proceeded in inserting these genes with strong, constitutive promoters in susceptible cultivars, resulting in resistance. However, at nearly the same time, the natural cisgenes were inserted with their own promoters. Apparently, the concept of cisgenesis was a logical step following the isolation of the Vf-genes. Additional resistance genes will also be inserted into apple by means of cisgenesis within a few years, using a combination of Vf-genes and other resistance genes. These stacked functional resistance genes will provide more durable resistance in elite cultivars. We regard cisgenesis as a way to apply the increasing knowledge about alleles to plant breeding, to the benefit of growers, consumers and the environment. An extra advantage of cisgenesis in comparison with cross breeding is that susceptible cultivars can be used that already have a proven high fruit quality and safe use. Apple is self-incompatible. Crossing with apple germplasm scrambles the genetic composition of good cultivars, and restoring such a cultivar through crossings is virtually impossible. However, cisgenesis preserves the genetic assembly of the high quality cultivar, and adds some well-defined apple alleles. Subsequently, the enriched cultivar can be propagated vegetatively by means of grafting, which is a common practise in apple propagation.

25 Cisgenesis

609

25.6.4 Cisgenic Approach in Vegetatively Crops

The stacking and linkage drag problems are, in addition to allopolyploids, most prominent in vegetatively propagated crops, specifically when they are polyploid or perennial as has been described in potato and apple, respectively. Short run success can be obtained in this type of crops when cisgenes are available and can be transferred directly into the variety of interest. During stacking of genes it is important to know whether all the genes of the cassette after insertion are expressed and biologically active. For strong resistance genes which have not been broken at the moment it is important to know whether HR reactions with the cross reacting Avr-genes is occurring or not and whether this is a reliable indication for functional gene expression. The cisgenic approach is, as earlier indicated, free of the multiple introgression problems when the traditional breeding approach has to be applied so that domestication of a number of inserted alien genes can be realized in one step. It is also important to realize that stacked cisgenes in existing varieties can easily be used in traditional cross breeding approaches. We believe that the cisgenic approach is simplifying traditional plant breeding, especially when multiple linkage drag problems have to be overcome. If this approach will be exempted from the GM-regulations worldwide, it will help plant breeding in general.

25.7 Concluding Remarks

This chapter describes cisgenesis as an efficient alternative for several multiple step (pre-breeding) approaches in traditional plant breeding by highlighting with examples for important traits that. It is introducing and domesticating agriculturally important genes in improved existing varieties in one step and providing new crossing parents for variety breeding. This approach is, therefore, highly recommended for breeders in developing countries and for SMEs. The process of producing cisgenic GM-plants results per definition into a GM-plant for the same reason as indicated in the GM-Directive 2001/18/EC (Anonymus 2001) for induced mutations and protoplast fusion between crossable species. However, in parallel with the induced mutation and protoplast fusion approaches, the cisgene source belongs per definition to the well known and already for a long time used breeders gene pool and should, because of this, also be exempted from the existing GM-regulations worldwide.

References
Anonymous (1999)Variety list of fruit crops (in Dutch), Dutch Fruit Growers Organization, P.O. Box 906071, 2585 HK The Hague, The Netherlands, ISN 01696750 Anonymous (2001) Directive 2001/18EC of the European Parliament and of the Council of 12 March 2001 on the deliberate release into the environment of genetically modified organisms and repealing Council Directive 90/220/EEC. Off. J Eur Comm 106:138 (http://europa.eu.int/ eur-lex/pri/en/oj/dat/2001/1_10620010417en00010038.pdf)

610

E. Jacobsen and H.J. Schouten

Babu RM, Sajeena A, Seetharaman K, Reddy MS (2003) Advances in genetically engineered (transgenic) plants in pest managementan overview. Crop Prot 22:10711086 Belfanti E, Silfverberg-Dilworth E, Tartarini S, Patocchi A, Barbieri M, Zhu J, Vinatzer BA, Gianfranceschi L, Gessler C, Sansavini S (2004) The HcrVf2 gene from a wild apple confers scab resistance to a transgenic cultivated variety. ProcNatl Acad Sci U S A 101: 886890 Bentolila S, Alfonso A, Hanson MR (2002) A pentatricopeptide repeat-containing gene restores fertility to cytoplasmic male-sterile plants. Proc Natl Acad Sci 99:1088710892 Buckler E, Thornsberry JM, Kresovich S (2001) Molecular diversity, structure and domestication of grasses. Genet Res Cambridge 77:213218 Calenge F, Faure A, Goerre M, Gebhardt C, Van de Weg WE, Parisi L, Durel C-E (2004) Quantitative Trait Loci (QTL) analysis reveals both broad-spectrum and isolate-specific QTL for scab resistance in an apple progeny challenged with eight isolates of Venturia inaequalis. Phytopathology 94:370379 Friebe B, Jiang J, Raupp WP, McIntosh RA, Gill BS (1996) Characterization of wheat-alien translocations conferring resistance to diseases and pests: current status. Euphytica 91:5987 Gardiner SE, Bus VGM, Rusholme RL, Chagn D, Rikkerink EHA (2006) Apple. In: Kole C (ed) Genome mapping and molecular breeding. Vol 4, Fruit and nuts. Springer, Berlin, pp 262 Gessler C, Patocchi A, Sansavini S, Tartarini S, Gianfranceschi L (2006) Venturia inaequalis resistance in apple. Crit Rev Plant Sci 25:473503 Haverkort AJ, Boonekamp PM, Hutten R, Jacobsen E, Lotz LAP, Kessel GJT, Visser RGF, Vossen Van der EAG (2008) Societal costs of late blight in potato and prospects of durable resistance through cisgenic modification. Potato Res 51:4757 Heilersig HJB, Loonen AS, Bergervoet M, Wolters AMA, Visser RGF (2006) Post-transcriptional gene silencing of GBSSI in potato: effects of size and sequence of the inverted repeat. Plant Mol Biol 60:647662 Hough LF, Shay JR, Dayton DF (1953) Apple scab resistance from Malus floribunda Sieb. Proc Am Soc Hort Sci 44:260272 Jacobsen E (2007) Cisgenesis: next step in advanced traditional breeding. In: Haverkort AJ, Anisimov BV (eds) Potato production and innovative technologies.Wageningen Academic, ISBN:9789086860425, pp 348352 Jacobsen E, Schouten HJ (2007) Cisgenesis strongly improves introgression breeding and induced translocation breeding of plants. Trends Biotechnol 25: 219223 Jacobsen E, Vossen, van der EAG (2009) Plant disease resistance. Cisgenesis: a new tool for traditional resistance breeding. In: Encyclopedia of microbiology, 3rd edn. Mesolio Schaechter Editor) pp 597612 Jansen JPA (1997) Aphid resistance in composites. Patent NL1003261 Jarvis DJ, Hodgkin T (1999) Wild relatives and crop cultivars: detecting natural introgression and farmer selection of new genetic combinations in agroecosystems. Mol Ecol 8:159173 Kole, C. (2006) Genome mapping and molecular breeding in plants. Vol 4, Fruits and nuts. Springer, Berlin Lim KY, Leitch IJ, Leitch AR (1998) Genomic characterisation and the detection of raspberry chromatin in polyploid Rubus. Theor Appl Genet 97:10271033 Liu F, Cui X, Horner HT, Weiner H, Schnable PS (2001) Mitochondrial aldehyde dehydrogenase activity is required for male fertility in maize. Plant Cell 13:10631078 Malnoy M, Xu M, Borejsza-Wysocka EE, Korban SS, Aldwinckle HS (2007) The role of Vfa1, Vfa2 and Vfa4 genes in resistance to Venturia inaequalis in transgenic McIntosh apple plants. ISHS Acta Hort 738:323326 McClure B (2006) New views of S-RNase-based self-incompatibility. Curr Opin Plant Biotechnol 9:639646 Parisi L, Lespinasse Y, Guillaumes J, Kruger J (1993) A new race of Venturia inaequalis virulent to apples with resistance due to the Vf gene. Phytopathology 83:533537 Parisi L, Laurens F, Didelot F, Evans K, Fischer C, Fouillet V, Gennari F, Kemp H, Lateur M, Patocchi A, Schouten HJ, Tsipouridis C (2006) Geographical distribution of Venturia inaequalis strains virulent to the Vf gene. Europe. Bulletin-OILB/SROP 29:4952

25 Cisgenesis

611

Pelletier G, Budar F (2007) The molecular biology of cytoplasmically inherited male sterility and prospects for its engineering. Curr Opin Biotechnol 18:121125 Reynaarts A, Wiele van der H, Sutter de G, Janssens J (1993) Engineered genes for fertility control and their application in hybrid seed production. Scientia Horticulturae 55:125139 Rommens CM (2004) All-native DNA transformation: a new approach to plant genetic engineering. Trends Plant 9:457464 Rommens CM, Ye J, Richael C, Swords K (2006) Improving potato storage and processing characteristics through all-native DNA transformation. J Agric Food Chem 54:98829887 Rommens CM, Haring MA, Swords K, Davies HV, Belknap WR (2007) The intragenic approach as a new extension to traditional plant breeding. Trends Plant Sci 12:397403 Ruiz ON, Daniell H (2005) Engineering cytoplasmic male sterility via the chloroplast genome by expression of beta-kethiolase. Plant Physiol 138:12321246 Schaart J, Krens FA, Pelgrom KTB, Mendes O, Rouwendaal GJA (2004) Effective production of marker-free transgenic strawberry plants using inducible site-specific recombination and a bifunctional selectable marker gene. Plant Biotechnol J 2:233240 Schmidt H, Van de Weg WE (2005) Breeding. In: Tromp J., Webster AD, Wertheim SJ (eds) Fundamentals of temperate zone tree fruit production. Backhuys, Leiden, the Netherlands, pp 136155 Schmidt R (2002) Plant genome evolution: lessons from comparative genomics at the DNA level. Plant Mol Biol 48:2137 Schouten HJ, Krens FA, Jacobsen E (2006a). Do cisgenic plants warrant less stringent oversight? Nat Biotechnol 24:753 Schouten HJ, Krens FA, Jacobsen E (2006b). Cisgenic plants are similar to traditionally bred plants. EMBO Rep 7:750753 Sijacic P, Wang X, Skirpan AL, Wang Y, Dowd PE, McCubbin AG, Huang S, Kao TH (2004) Identification of the pollen determinant of S-RNase-mediated self-incompatibility. Nature 429:302305 Trapman, M. (2006) Resistance management in Vf resistant organic apple orchards. Bulletin OILB/SROP 29:253257 Vetten de N, Wolters AMA, Raemakers K, Meer van der I, Stege ter R, Heeres E, Heeres, Visser RGF (2003) A transformation method for obtaining marker-free plants of a cross-pollinating and vegetatively propagated crop. Nat Biotechnol 21:439442 Vleeshouwers GAA, Rietman H, Krenek P, Champouret N, Young C, Oh SK, Wang M, Bouwmeester K, Vosman B, Visser RGF, Jacobsen E, Govers F, Kamoun S. Vossen, van der EAG (2008) Effector genomics accelerates discovery and functional profiling of potato disease resistance and Phytophthora infestans avirulence genes. PLoS ONE e2875 Wang Z, Zou Y, Li X, Zhang Q, Chen L, Wu H, Su D, Chen Y, Guo J, Luo D, Long Y, Liu YG (2006) Cytoplasmic male sterility of rice with Boro II cytoplasm is caused by a cytotoxic peptide and is restored by two related PPR motif genes via distinct modes of mRNA silencing. Plant Cell 18:676687 William HM, Trethowan R, Crosby-Galvan EM (2007) Wheat breeding assisted by markers: Cimmyts experience. Euphytica 157:307319 Wise RP, Pring DR (2002) Nuclear-mediated mitochondrial gene regulation and male fertility in higher plants: light at the end of the tunnel? Proc Natl Acad Sci 99:1024010242 Yu HX, Tang SZ, Liu QQ, Wang L, Zhao ZP, Xu L, Huang BL, Gong ZY, Tang SZ, Gu MH (2006) Breeding of selectable marker-free transgenic rice lines containing AP1 gene with enhanced disease resistance. Agric Sci Chin 5:805811