How many chromosomes does cannabis have

In situ hybridization on hemp (Cannabis sativa L.) chromosomes


1 From the Institute for Plant Breeding and Plant Protection of the Faculty of Agriculture of the Martin Luther University Halle-Wittenberg In situ hybridization on hemp (Cannabis sativa L.) Chromosomes as a dissertation to obtain the academic degree doctor agriculturarum (Dr. agr.) Presented by a graduate agricultural engineer Marko Riedel Halle / Saale 2005 urn: nbn: de: gbv: [

2 From the Institute for Plant Breeding and Plant Protection of the Agricultural Faculty of the Martin Luther University Halle-Wittenberg In situ hybridization on hemp (Cannabis sativa L.) Chromosomes as a dissertation for obtaining the academic degree doctor agriculturarum (Dr. agr.) Presented by a graduate agricultural engineer Marko Riedel born on in Lutherstadt Wittenberg Reviewer: Prof. Dr. habil W. E. Weber Prof. Dr. habil W. Diepenbrock Dr. habil V. Schubert Defense on: Halle / Saale 2005 urn: nbn: de: gbv: [

3 Table of contents Table of contents 1 Introduction 1 2 State of knowledge Basics of sex determination General mechanisms of sex determination Development of sex chromosomes Sex determination in plants Sexually monomorphic plants Sexually polymorphic plants Hemp botany Cytology Molecular genetics 14 2 Material and methods Material plant material Bacteria and vectors Methods DNA isolation Total plant DNA bacteria Plasmid-DNA (plasmid mini-prep.) Hemp pollen isolation Polymerase chain reaction (PCR) based methods Standard PCR RAPD (random amplified polymorphic DNA) DOP-PCR (degenerated oligonucleotide primed-pcr) I-PEP-PCR (improved primer extension preamplification-pcr) PCR Walking AFLP (amplified fragment length polymorphism) Touchdown PCR DNA sequencing Electrophoresis Fragment isolation from electrophoresis gels Cloning of DNA fragments Permanent bacterial cultures 22


5 Table of contents Establishing FISH in hemp FISH with gender-specific probes FISH with PAR-specific probes 84 5 Summary 86 6 Bibliography 88 7 Appendix 103 List of figures 103 List of tables 104 Tables List of abbreviations used

6 1 Introduction 1 1. Introduction Hemp (Cannabis sativa L.) is one of the most important crops in human history due to its wide range of uses. The main uses are the hemp fibers (textile and paper production), the hemp seeds (food and oil production) and the intoxicating flowers of female hemp plants (for pharmaceutical and spiritual purposes and as luxury food). Since important properties of the hemp such as fiber quality and quantity as well as the intoxicant content are strongly influenced by the sex of the originally dioecious (dioecious) plant, attempts were made to breed monoecious (monoecious) varieties of hemp. With such monoecious varieties, however, purely male plants repeatedly split out (Hoffmann, 1947; Bocsa et al., 1997). Comprehensive studies to elucidate the complicated processes of gender inheritance in hemp are necessary. The aim of the present work is to use molecular and cytological methods to expand knowledge about the structure and organization of gender-linked DNA areas in order to gain insight into the gender inheritance of diocial hemp forms. The methods developed through this work are also to be used in the future to clarify the nature of monoic hemp forms. The experimental part of the work is divided into two parts. The first part focuses on the identification and molecular characterization of gender-specific DNA areas. In the second part, these DNA areas will be used as probes to visualize their position on the hemp chromosomes by means of fluorescence in situ hybridization (FISH). The technique of fluorescence in situ hybridization is a combination of cytogenetic and molecular genetic methods with the aim of hybridizing microscopic specimens with labeled DNA probes. The range of applications for this technology includes, among others. the physical mapping of DNA sequences, the identification and characterization of chromosomes or chromosome segments as well as the identification of chromosomal alterations through recent evolutionary processes (Schwarzacher and Heslop-Harrison, 2000). Several strategies have been pursued to identify gender-specific DNA fragments. On the one hand, PCR techniques should be used to develop gender-specific markers whose DNA fragments can be used as probes.

7 1 Introduction 2 On the other hand, it was tested whether sex-specific clones of another diocesan plant species (Silene latifolia) can be used as a probe for hemp. Another strategy was to look for repetitive sex-specific clones in a hemp DNA library. For this purpose, the clones of a hemp DNA library should be hybridized with male or female probes. Clones that hybridize to only one of the probes should be gender specific. Total DNA from male and female hemp plants should be used as probes. This would make it possible to identify male-specific clones. Female clones cannot be identified with these probes, since both male and female hemp DNA contain an X-chromatin, the hybridization signals of which are superimposed. In order to still be able to identify female-specific clones, probes should be made from the haploid genome of individual hemp pollen. Assuming that hemp has X and Y chromosomes, the three nuclei of hemp pollen each have nine autosomes and only one sex chromosome (X or Y). By amplifying the DNA of individual hemp pollen with degenerate primers, enough DNA should be generated for the production of the probe. Such probes can prevent the hybridization signals from being superimposed.

8 1 Introduction 3 2 State of the art 2.1 Basics of sex determination Sex determination mechanisms in general Both in the animal kingdom and in the vegetable kingdom two systems of sex determination have developed. On the one hand, there is cosexuality, in which the sexual organs of both sexes are localized on one organism (e.g. in marine invertebrates and most plants: hermaphrodite and monoecious). On the other hand, the diocese has developed into the dominant system among vertebrates. This system of separated-sex individuals is also developed in some plant species. It is assumed that the diocese developed out of cosexuality through mutations. The driving force behind this evolutionary development could have been resource and energy efficiency and the avoidance of the accumulation of recessive disease genes through inbreeding (Charlesworth, 1991). The processes of sex determination lead to the physical separation of different individuals of a species, which are able to form male or female gametes (Tanurdzig and Banks, 2004). The genetic regulation of sex determination in animals generally contains three basic components: a primary (genetic) signal, a main regulator that reacts to it, and a switch system that chooses between two alternative sexual programs (Nöthiger and Steinmann-Zwicky, 1987). In animals it is known that the determination of sex can be carried out by three different processes (Ayling and Griffin, 2002). For example, sex-determining transcription processes in lizards, crocodiles and turtles are influenced by the incubation temperature of the embryos. Other systems are gene-based. In such processes, the sex is determined by the allele conformation of a gene (e.g. in the case of state-forming Hymenoptera). The third mechanism of sex determination is determined by the presence of whole sex chromosomes. This type of sex determination occurs in Drosophila, fish, birds and mammals. Some diocesan plants with heteromorphic sex chromosomes also show this type of sex determination.

9 1 Introduction 4 There are various systems for determining the sexing of chromosomes. In the ZZ / ZW system of birds and snakes (Graves, 1998), the females are heterogametic (ZW), while the males with two Z chromosomes are homogametic. Mammals have an XX / XY system in which the males are heterogametic (XY) and the females are homogametic (XX). The fruit fly Drosophila also has an XX / XY system. While in mammals the sex is controlled by the Y chromosome, in Drosophila the relationship between the X chromosome (s) and the autosomes is decisive. Depending on this ratio, various gender and intermediate forms are known. The genes that determine female gender are suspected to be on the X chromosome. Genes that determine the male gender are probably located on the autosomes (Winter et al., 1998). There are different systems for fish and amphibians (e.g. XX / XY, ZW / ZZ, XX / X0). Unfortunately, these species have not been investigated very much so that well-founded statements are often not possible. Development of the sex chromosomes The development of systems of heteromorphic sex chromosomes in phylogenetically distant species suggests that the same or similar causes are involved in these systems. It is believed that all sex chromosomes evolved from pairs of autosomes. So-called proto X and Y chromosomes therefore contained a simple diallelic sex chromosome system (Negrutiu, 2001). During the evolution of animal sex chromosomes, it is generally assumed that the suppression of recombination (crossing over) led to functional and structural degeneration of the Y chromosome. For this degeneration z. B. through the accumulation of fixed mutations on the Y chromosome, inter alia. Processes such as Muller's ratchet, (Muller, 1964) and Hitchhiking (Rice, 1987) were blamed. Muller s ratchet describes a mechanism of genetic drift in which deletions prevent recombination, which in turn results in its fixation. As this process progresses, further mutations accumulate. The model of hitchhiking on Y chromosomes describes the occurrence of favorable mutations in a non-recombining area of ​​the Y chromosome, which leads to the fixation of all

10 1 Introduction 5 deletions of this chromosome. The fixation of the deletions results in a reduced genetic activity of the Y chromosomes (Rice, 1987). Despite the greatest possible suppression of recombination of the sex chromosomes, there are still regions on X and Y in some organisms that can pair and recombine during meiosis. Such regions are called pseudo autosomal regions (PAR). These regions may have arisen through translocations of autosomal genes (Graves, 1995; Graves et al., 1998). The PAR of mammals such as mice and humans is considered to be well studied (Ayling and Griffin, 2002). It enables the exchange between the X and Y chromosomes and ensures a correct separation of the sex chromosome pair during meiosis. In mice it was found that errors in this process led to a reduction in fertility (Hassold et al., 1991; Kipling et al., 1996). In humans, too, deletions in the PAR lead to errors in spermatogenesis. However, PARs are absent in all marsupials and some rodents. Homologous mating is therefore not a universal requirement for fertility (Graves et al., 1998). The PARs of humans and mice have sizes of 2600 kb and kb, respectively (Ayling and Griffin, 2002). In addition to the 2600 kb PAR1, people have a second PAR2 of around 500 kb, the PAR2. This PAR2 is of a more recent evolutionary origin and does not occur in the closest relatives of humans, the primates (Ellis et al., 1990). Outside of the animal kingdom, the occurrence of a PAR has only been described in a genus of diocesan plants. In the Y chromosomes of the light carnation Melandrium album (syn. Silene latifolia), Westergaard (1953) describes differentiating, non-pairing regions and pairing regions (PAR). The first molecular detection of PAR in plants was achieved by Di Stilio et al. (1998) for a plant species (Silene dioica L.) of the same genus. It is assumed that the PAR region disappears in the course of evolution in favor of complete differentiation of the sex chromosomes (Graves et al., 1998). Y chromosomes usually have only a few active genes. Many of the genes located on the Y chromosome that have homologous genes on the X chromosome have lost their function in the course of evolution through the fixation of mutations and have become pseudogenes. Others acquired a male-specific function through mutation. The most important gene for the development of the male sex is SRY (sex-determining region Y). This gene initiates a cascade of gene activation and gene suppression that lead to expression of the male sex (Ayling and Griffin, 2002).

11 1 Introduction 6 Both the mechanisms of sex inheritance and the structure of the sex chromosomes in plants are very similar to the known systems of animals. Properties such as B. the degeneration of the Y-chromosomes in plants and animals are interpreted as an indication of general evolutionary processes in the development of sex chromosomes (Negrutiu, 2001; Charlesworth, 2002). Molecular comparisons of the Marchantia Y chromosome with that of humans show clear similarities (Tanurdzig and Banks, 2004). The same applies to the sex chromosomes of the two plant species Silene latifolia and Rumex acetosa. In contrast to the Y chromosomes of mammals, the Y chromosomes of many diocesan plant species (e.g. Silene, Rumex and Cannabis) are considerably larger than the X chromosomes. The accumulation of repetitive sequences is assumed to be the cause. Such repetitive sequences could be detected among others in Silene latifolia, Asparagus officinalis and Rumex acetosa. Sexually polymorphic plant species with systems of sex determination based on sex chromosomes emerged in the course of evolution several times and independently of one another from hermaphrodite ancestors. As a result of phylogenetic studies of the genus Silene, two origins of the diocy are to be assumed. Overall, it is assumed that diocesan forms in the flowering plants have developed more than 100 times. The time of the separation of the sex chromosomes in Silene is assumed to have been around 20 million years ago (Charlesworth, 2002). In comparison, it is estimated that the evolution of human sex chromosomes began around 240 to 320 million years ago (Lahn and Page, 1999). Plant sex chromosomes are therefore considered to be relatively recent developments in which the early stages of the evolution of sex chromosomes can be investigated (Charlesworth, 2002) Sex determination in plants Approximately 90% of flowering plants are hermaphrodites with hermaphroditic flowers that respond to the production of microspores () or macrospores () are specialized (Charlesworth, 2002). Of the remaining species, around 50% are monoecious. These plants have unisexual flowers. Whereby the flowers of both sexes are localized on one and the same individual. The other 50% (sexually polymorphic) include the diocesan plants. In these plants, the male and female flowers are located on separate individuals.

12 1 Introduction Sexually monomorphic plants Sexually monomorphic plants, to which the hermaphrodites and monoecean plant species belong, represent the largest group of flowering plants. In the hermaphrodites, the genetic control of flower formation was investigated by analyzing mutations in the species Arabidopsis and Antirrhinum. Two basic classes of genes were identified. One class of genes is responsible for the identity of the meristems, while the genes belonging to the other class are responsible for the development of the flower organs. The process of flower formation begins with the transformation of vegetative meristems into inflorescence meristems and then into floral meristems. So far, four genes (EMF: Embyonic Flower, TFL: Terminal Flower, AP1: APETALA1 and LFY: LEAFY) have been identified that are involved in these processes. For example, the suppression of EMF leads to the transformation from the vegetative meristem to the inflorescence meristem. The TFL gene is responsible for the development of the inflorescence meristem or for the suppression of the floral meristem. AP1 and LFY are necessary for the downstream expression of genes (flower organ identity). The genetic control of the flower organ identity is described by the so-called ABC model (Coen and Meyerowitz, 1991; Weigel and Meyerowitz, 1994) (Fig. 1.1). Responsible for this are three classes of homeotic genes that overlap in their function (A, B, and C). The protein sequences of many homeotic genes contain conserved DNA and protein binding motifs. These motifs, called K-Box or MADS-Box, have developed functions similar to those of the animals' homeobox genes, which are important for organ identity. These genes code for different classes of transcription factors. Each class influences the expression of two whorls of the flower system. Each of these whorls forms the origin of one of the four different flower organs. Class A genes influence the formation of whorls 1 and 2 (sepals and petals). Class B genes influence whorls 2 and 3 (crown and stamens), while genes of class C influence the formation of whorls 3 and 4 (stamen and carpels) (Dellaporta and Calderon-Urrea, 1993).

13 1 Introduction 8 whorl sepal petal stamen carpel A B E C D Figure 1.1: Model of the control of the flower organ identity left: The numbers 1 to 4 designate the individual whorls, including the flower organs influenced by the gene function. The gene functions are labeled A, B, C, D and E. Newer models assume two additional gene classes (D and E). Class D genes play an important role in specifying egg identity (Colombo et al., 1995). The genes of class E (Theissen, 2001) express proteins which interact with the proteins of the other class. By identifying these two new classes, Theissen (2001) extended the classic ABC model to the so-called quartet model. In this model, the identity of the flower organs is determined by four different combinations of homeotic proteins. In the other group of sexually monomorphic plants, the monocists, the sex of the flowers is determined locally during flower development. Monocial plants can be divided into two groups in terms of their gender characteristics. There are species that produce exclusively unisexual flowers on the same individual and species that also produce hermaphrodite flowers on the same individual in addition to the unisexual flowers. The maize (Zea mays) belongs to the first group. The unisexuality of maize flowers is brought about by the selective elimination of the stamens in the female inflorescences (cobs) or the pistils in the male inflorescences. Based on the investigation of mutants, a decisive influence of gibberellic acids (GA) and other steroid-like hormones in the suppression of stamen development could be demonstrated (Irish, 1999).

14 1 Introduction 9 The cucumber (Cucumis sativus) belongs to the second group of monoecious plants. Here, too, all flowers are originally hermaphroditic. The interruption of the development of male or female organs leads to the formation of unisexual flowers. Three genes (F, A and M) influence the formation and arrangement of unisex flowers. The semidominant F gene causes an increase in the female character in the apical direction. The A gene is epistatic for this and is also required for the expression of femininity. The M gene is necessary for the development of male flowers. The combination of the different alleles of M and F determines the sex expression of the plant. In addition to these three genes, phytohormones such as gibberellic acids and ethylene also have an influence (Tanurdzig and Banks, 2004). In papaya (Carica papaya), gender is controlled by three different allele states of a single gene. The dominant alleles M and M h lead to the expression of male and hermaphroditic flowers, respectively. Expression of the recessive allele m leads to female flowers (Storey, 1953). Males have the constitution Mm, hermaphrodites M h m and females mm. Homozygous states of dominant alleles (MM, M h M h) as well as heterozygotes (MM h) are presumably lethal Sexually polymorphic plants In addition to primitive plants such as the well liver moss Marchantia polymorpha, some naked species (Gymnospermae) and a few types of bedspamers (Angiospermae) are sexually polymorphic. Only about 6% of the flowering plants covered with seeds are diocesan (Renner and Ricklefs, 1995). These include species such as the light carnations Silene dioica and Silene latifolia, the asparagus (Asparagus officinalis L.), the pistachio (Pistacia vera), the mock hemp (Datisca cannabina), species of the genera Rumex (dock) and Actinia (kiwi-like) as well as both Cannabinaceae hemp (Cannabis sativa) and hops (Humulus lupulus). The genetic determination of the sex of diocesan plants takes place at the level of the whole plant. The sexual dimorphism of diocesan plants is determined in the very early phases of flower development (Ainsworth et al., 1998). This definition takes place during or after the processes of flower formation of the ABC model and is completely independent of this. One plant in which the mechanism of sexual inheritance has been well studied is Marchantia polymorpha. In these primitive plants, the haploid gametophyte dominates the life cycle. The gametes formed by the gametangia of the gametophyte

15 1 Introduction 10 unite to form diploid sporophytes, which in turn produce haploid spores. New gametophytes then develop from the spores. The sex of the gametophytes is determined by the presence of cytologically heteromorphic sex chromosomes. In addition to the eight autosomes, male gametophytes also have a Y chromosome. Female gametophytes have an X chromosome instead of the Y chromosome (Lorbeer, 1934). By constructing X- or Y-specific PAC (P1-based artificial chromosome) banks, Okada et al. (2000, 2001) found that a quarter to a third of the Y chromosome consists of variable repetitive elements. Six potentially protein-coding genes were also found. Two of these genes were specific for the Y chromosome. The other four genes were found on Y and in low copy numbers on the X chromosome. The light carnation Silene latifolia L. has been the best studied of the diocesan flowering plants with blooming seeds. There, the formation of unisexual flowers, similar to the flowers of monoecious plants, is achieved by stopping the development of originally created androecias or gynoecias (Grant et al., 1994). In Silene latifolia, too, sex determination is controlled by sex chromosomes. The Y chromosome is the largest chromosome here. Male plants are heterogametic (XY) and female plants are homogametic of type XX. Early studies on deletion mutants with deletions on the Y chromosome (Westergaard, 1958) led to the finding that the Y chromosome has three regions that are important for sex expression. The first region is called the Su F region. This region acts as a suppressor of the development of female organs. In contrast, the other two regions control the development of the anthers. Later work by Farbos et al. (1999), Lardon et al. (1999) and Lebel-Hardenack et al. (2002) were able to confirm these findings. So far it has been possible to identify four genes that are on the Y chromosome. These genes are SLY-1 (Delichere et al., 1999), SLY-4 (Atanassov et al., 2001), MROS3_Y a (Matsunaga et al., 1996; Guttman and Charlesworth, 1998) and DD44Y ( Moore et al., 2003). Each of these genes has a homologous gene on the X chromosome. In addition, MROS3_Y a appears to belong to a family of low copy genes, the members of which are also distributed on the autosomes (Kejnovsky et al., 2001). Male-specific expression only showed the genes SLY-1 and MROS3_Y a. However, these genes do not seem to be the sex-controlling sites, but rather are controlled in a gender-dependent manner (Charlesworth, 2002).

16 1 Introduction 11 Another well-studied group of diocesan plants is the genus Rumex. The sex determination here is based on heterogametic sex chromosomes. The female plants of this genus have two X chromosomes. The males, on the other hand, have two Y chromosomes in addition to the X chromosome. In the species Rumex acetosa, the sex is determined by the ratio of female factors on the X chromosome and male factors on the autosomes (Ainsworth et al., 1998; Stehlik and Blattner, 2004). The highly heterochromatic Y chromosomes only have an impact on male fertility, but not on gender expression. In asparagus (Asparagus officinalis L.) the sex-determining factor is located on the homomorphic chromosome pair L5. In addition to heterozygous males and homozygous females, there are so-called super males who are homozygous. The hops, which are closely related to hemp, also have an XY-based system of gender determination. Shephard et al. (2000) describe the sex chromosomes as homomorphic. By means of differential staining techniques with the dye DAPI (4,6 diamidino-2-phenylindole dihydrochloride) it was Karlov et al. (2003) still possible to find morphological differences in the sex chromosomes of hops. The Y chromosome was identified as the smallest of all chromosomes. The X chromosome is medium in size. 2.2 Hemp botany Hemp, Cannabis sativa L., belongs within the Dicotyledoneae (dicotyledonous plants) to the Urticales (nettle-like) and within these to the family of the Cannabaceae (syn .: Cannabidaceae). The Cannabaceae family consists of two genera (Cannabis and Humulus). The species of both genera are originally dioecious (dioecious). The division within the cannabis genus is controversial. Older studies divide the genus into two (Hoffmann et al., 1970) or three types (Schultes et al., 1974). More recent studies assume a species with several subspecies. Small and Cronquist (1976) differentiate between subspecies as THC-rich forms (C. sativa ssp. Indica) and THC-poor forms (C. sativa

17 1 Introduction 12 ssp. sativa). Also at the subspecies level, a distinction is made between broad-leaved and densely branched drug hemp from Afghanistan and Pakistan (C. sativa ssp. Indica) and narrow-leaved fiber and drug hemp (C. sativa ssp. Sativa) from the rest of the world (de Meijer, 1999). Hemp is an annual, upright, herbaceous plant with a height of up to five meters. The height of growth depends on the type and the growth conditions. Under suboptimal conditions, the plants reach the generative phase at considerably lower heights (Ranalli, 1999). The generative phase of the short-day hemp plant is delayed by long day lengths. The shoot axis, which is square in the youth stage, develops a hexagonal cross-section as it grows. The root system consists of a radial main root with lateral and secondary roots. The fingered leaves have 5 to 13 lanceolate, serrated blade sections. On the basis of the flower structure, a distinction is made between monoecious (one-euced) and dioecious (dioecious) hemp plants. Male inflorescences of diocesan plants are leafless in the form of a loose panicle. The short-stalked flowers are five-fold. The flowers of diocesan female plants form a leafy and unbranched stem end as false spikes (Hoffman, 1947; Dierks and von Sengbusch, 1967). The female flowers are enclosed by bracts which are covered with a multitude of THC-secreting glandular hairs (Stearn, 1970). In the diocesan forms, the female plants have a longer lifespan (approx. 2-4 weeks for the seeds to ripen) than the males, which die after the pollen is ripe. In monocular forms, the male flowers develop from the leaf axils. Female flowers are localized on side shoots. Different growth and gender types can be distinguished in the monocial hemp forms. According to the expression of the secondary sexual characteristics, a distinction is made between plants with female and male growth types (Hoffmann, 1947). Within these growth types, further distinctions are made based on the ratio of male and female flowers. The female growth types include: normal females, feminized monocists (different proportions of male and female flowers) and feminized males (exclusively male flowers). Normal males, masculinized monocists (different proportions of male and female flowers) and masculinized females (exclusively female flowers) belong to the male growth type. In rare cases, the appearance of hermaphrodite flowers could be determined.

18 1 Introduction 13 The fruits of hemp are monochrome to marbled nuts (achenes). Cytology Diploid hemp has 2n = 20 chromosomes (Hirata, 1929). Tetraploid cells occur in the cells of the primary root cortex (Breslavetz, 1928, 1932; Riedel, 2000). As with some other diocesan species, hemp has a heteromorphic sex chromosome pair (Hirata, 1929; von Sengbusch, 1943; Hoffmann, 1947). The chromosomal configuration of monocular forms of hemp is unknown. Yamada (1943) describes after microscopic examinations that diocesan female plants have two X chromosomes and males one X and one Y chromosome. The Y chromosome is described here as the largest chromosome in hemp. Current investigations using flow cytometry (Sakamoto et al., 1998) confirm these findings and characterize the Y chromosome as subtelocentric with a satellite on the short arm. The satellite region and the long arm of the Y chromosome condense considerably more strongly than the X chromosome and the autosomes during the transition from mitotic prophase to metaphase. The accumulation of Y-specific LINE (long interspersed element) like retrotransposons (Sakamoto et al., 2000) is cited as a possible reason for this behavior. Flow cytometric examinations of diploid forms of hemp showed genome sizes of 1636 megabase pairs (Mbp) in female plants and 1683 Mbp in male plants. The difference between the two sexes is attributed to the considerably larger Y chromosome. This also corresponds to the observations made by Herich (1961), who was able to find differences between pollen with X and Y chromosomes when investigating the size of pollen grains. Y-pollen was significantly larger in comparison to X-pollen. In comparison, wheat with Mbp has a considerably larger genome. The species Arabidopsis thaliana, which is known for its very small genome, has a genome of 260 Mbp that is about six times smaller than that of hemp (Kaneko et al., 1998; Sakamoto et al., 1998). Genome size differences between male and female plants could also be found in other diocesan plants. Male plants of Silene latifolia have a 2% to 5% larger genome than female plants (Costich et al., 1991; Vagera et al., 1994; Dolezel and Göhde, 1995)

19 1 Introduction Molecular Genetics Molecular studies on hemp have been carried out since the mid-1990s. This included marker analyzes using RFLP (restriction fragment length polymorphism) and RAPD (random amplified polymorphic DNA) technology as well as microsatellite analysis (STR: short tandem repeat, SSR: simple sequence repeat) for gender determination (Sakamoto et al., 1995, 2000; Mandolino et al., 1999, 2002; Riedel, 2000; Törjék et al., 2002a, b; Moliterni et al., 2004, Rode et al., 2005) and for the analysis of genetic diversity (Faeti et al., 1996; Jagadish et al., 1996; Shirota et al., 1998; El-Ghany, 2001; Forapani et al., 2001, Kojoma et al., 2002, Alghanim and Almirall 2003). The AFLP (amplified fragment length polymorphism) technique used Flachowsky et al. (2001) and Peil et al. (2003) to identify male and PAR-specific markers. Studies on the investigation of conserved intergenic spacer regions of hemp such as Internal Transcribed Spacer (ITS) and chloroplast genes trnl / f have also been published (Linacre and Thorpe, 1998 and Gigliano, 1999, respectively). During the investigations into molecular sex determination, some AFLP and RAPD markers linked to the male gender could be identified. Sakamoto et al. (1995) found two RAPD primers which generated 500 bp and 730 bp DNA fragments only with DNA from male plants. The conversion of the male-specific RAPD marker OPA8 400 was achieved by Mandolino et al., With the help of AFLP technology, Flachowsky (2003) succeeded in identifying a large number of male-specific markers. The successful conversion of two male-specific AFLP markers into SCAR markers was demonstrated. Also in AFLP marker analyzes on hemp (Flachowsky, 2003; Peil et al., 2003), the markers AGA_AAT_330 and AGA_GAA_510 linked to the father's Y chromosome showed a recombination rate of r = 0.25 in the offspring. It is believed that these markers are located on a region of the Y chromosome of the male parent that has homologies with the X chromosome. These markers indicate a PAR (Pseudo Autosomal Region) on the sex chromosomes of the hemp.

20 2 Material and methods 15 2 Material and methods 2.1 Material plant material The starting material for the cytological investigations was hemp plants of the diocese descent CAN 18 (Genbank, Institute for Plant Genetics and Cultivated Plant Research Gatersleben) and the diocesan Hungarian fiber hemp variety `Kompolti and crossings of` Kompolti with dioecious cultivar `Skunk 1 used. DNA from the diocese hemp population A2 was used for the molecular marker analysis as well as for the other molecular investigations. These are 81 F 1 progeny from a cross between two plants of the diocese ancestry CAN 18. This population was created by Flachowsky et al. (2001) Bacteria and vectors Competent One Shot Chemically Competent E. coli cells of the strain TOP10F (Invitrogen, Karlsruhe) were used for transformations. DNA fragments were cloned into the vector pcr 2.1-TOPO (Invitrogen, Karlsruhe). 2.2 Methods DNA isolation Total vegetable DNA For DNA isolation, a protocol according to Saghai Maroof et al. (1984) used. The protocol was modified as in Flachowsky (2003). Bacteria plasmid DNA (plasmid mini-prep.) To isolate plasmid DNA, single colonies were grown overnight (37 ° C., 200 rpm) in 2 ml LB medium with 2 μl Amp 100 . The bacteria in the suspension were in a

21 2 Material and methods 16 Centrifuge pelleted for 5 min at rpm and then resuspended in 100 μl plasmid solution I. After a 5-minute incubation (on ice) to lyse the bacteria, 200 μl of plasmid solution II were added. The batch was carefully mixed, 150 μl of plasmid solution III were added, stored on ice for 5 min and centrifuged at rpm for 5 min.The supernatant was removed and precipitated with 1 ml of ethanol at 20 ° C. and then pelleted. After pelleting, the plasmids were washed twice with 500 μl of ethanol (70%), dried in vacuo, dissolved in 40 μl of 1 × TE / RNAse buffer and incubated at 37 ° C. for 15 min. Hemp pollen isolation With a dissecting needle, dried hemp pollen was placed on a slide given with µl of distilled water. With the aid of a glass capillary, the opening of which was 20 µm, individual pollen were isolated from this pollen suspension under the microscope and transferred to 0.2 µl tubes with 10 µl distilled water. Polymerase chain reaction (PCR) -based methods Standard PCR The standard PCR was carried out in 1x PCR -Buffer Y (with 15 mm MgCl 2, PeqLab, Erlangen) with 0.2 mm dntp`s, 0.25 µm forward primer, 0.25 µm reverse primer and 1 U Taq polymerase (PeqLab, Erlangen) in the thermal cyclers T. -Gradient (Biometra) or TC480 (Perkin Elmer) carried out. Standard PCR conditions included: number of cycles step 1 5 min at 94 C, initial denaturation phase 1 min at 94 C, denaturation 25 to 35 1 min at C, annealing phase 1 5 min at 72 C, synthesis phase 1 10 min at 72 C, final elongation step

22 2 Material and methods 17 The reaction vessels were then cooled to 4 ° C. The annealing temperatures and length of the synthesis phase depend on the PCR method, primer properties and DNA fragment size RAPD (random amplified polymorphic DNA) In the RAPD technique, 0.25 µm oligonucleotide primers (10 base pairs) with random sequences were used for amplification. An annealing temperature of 37 ° C. and a synthesis time of 2 min were achieved. DOP-PCR (degenerated oligonucleotide primed-pcr) DOP-PCR was carried out with the DOP-PCR Master Kit (Roche, Mannheim) according to the instructions. Amplification conditions: number of cycles step 1 1 min at 94 ° C., initial denaturation phase 5 1.5 min at 30 ° C., primer annealing 3 min at 30 ° C., annealing phase with temperature increase from 3.5 ° C./15 sec to 72 ° C. 3 min at 72 ° C., synthesis phase 35 1 min at 94 C, denaturation 1 min at 62 C, annealing phase 2 min at 72 C, synthesis phase (14 sec synthesis time extension for each subsequent cycle) 1 7 min at 72 C final elongation step I-PEP-PCR (improved primer extension preamplification- pcr) In the I-PEP-PCR, the totally degenerate 15N primer with a sequence of random 15 nucleotides was used. Amplification conditions of the I-PEP-PCR:

23 2 Material and methods 18 Number of cycles Step 1 1 min at 92 ° C. Denaturation 50 2 min at 37 ° C. Annealing, with temperature increase from 0.1 ° C./sec to 55 ° C. and subsequent 30 sec. At 68 ° C. 16 μm 15N primer, 0 , 1 mm dntp, 2.5 mm MgCl 2, 1 x high fidelity PCR buffer (Roche, Mannheim) and 3.6 U Expand high fidelity polymerase (Roche, Mannheim) were used. PCR walking During PCR walking, 2.5 µg of DNA were used in each case cut with various restriction enzymes producing blunt ends (80 U each: DraI, EcoRV, PvuII, ScaI, SmaI and SspI). The ends of the resulting fragments were ligated in a mixture of ligation buffer, 10 U T4 ligase and 5 μm adapter with a known sequence (adapter sequence see Table 7.2 in the Appendix) overnight at 15 ° C. A PCR was then carried out with adapter-specific or fragment-specific primer pairs (10 μm each, length: 27 to 30 bp) using 1 μl of cut DNA, 6 U Advantage Genomic Polymerase Mix (Clonetech, Heidelberg), 1 × PCR buffer (Clontech , Heidelberg), 0.2 mm dntp`s and 5.5 mm Mg (OAc) 2. The PCR was designed as a 2-step amplification first with an outside and then with an inside primer pair (nested PCR). 1. Amplification: number of cycles step 7 25 sec at 94 ° C., denaturation 3 min at 72 ° C., annealing / primer extension sec at 94 ° C., denaturation 3 min at 67 ° C., annealing / primer extension 1 7 min at 67 ° C., final extension die Amplification products of the first PCR were diluted 1:50 and used in a second PCR with the inner adapter- or fragment-specific primer pair.

24 2 Material and methods amplification: number of cycles step 5 25 sec at 94 C, denaturation 3 min at 72 C, annealing / primer extension sec at 94 C, denaturation 3 min at 67 C, annealing / primer extension 1 7 min at 67 C, final extension AFLP (amplified fragment length polymorphism) The AFLP technique was based on a modified protocol from Vos et al. (1995) carried out. 2 µg of total DNA were cut with the restriction enzymes MseI and HindIII. Interface-specific adapters were ligated to the ends of the resulting DNA fragments. The ligation mixture consisted of 0.25 U T4 ligase, 1 × RL buffer, 1.2 mm ATP and 2.5 mmol each of MseI and HindIII adapters. The ligation took place overnight at 15 ° C. In the subsequent preamplification, 5 μl of ligation mixture were mixed with 1 × PCR buffer Y (PeqLab, Erlangen), 1 × enhancer solution (PeqLab, Erlangen), 200 μm dntp s, 75 ng preamplification primer (HindIII + A and MseI + A) and 5 U Taq polymerase (PeqLab, Erlangen) amplified in a 50 µl amplification mixture under the following conditions: number of cycles step 20 1 min at 94 ° C, denaturation 1 min at 60 ° C, annealing phase 2 min at 72 ° C., synthesis phase 1 10 min at 72 ° C., final elongation The preamplification products were diluted 1:20 with water and 2.5 μl of them were used in a further amplification step with primers with three selective bases. The 20 μl amplification mixture consisted of 1 × PCR buffer (PeqLab, Erlangen), 1 × enhancer solution (PeqLab, Erlangen), 10 ng HindIII + ANN primer (Cy3-labeled), 60 ng MseI + ANN primer, 200 μm dntp s and 0.1 U Taq polymerase (PeqLab, Erlangen). The amplification conditions were designed as follows:

25 2 Material and methods 20 Number of cycles Step 9 1 min at 94 C, denaturation 1 min at 65 C, annealing phase 1.5 min at 72 C, synthesis phase, after each cycle: decrease the annealing temperature by 1 C 23 1 min at 94 C, Denaturation 1 min at 56 ° C, annealing phase 1 min at 72 ° C, synthesis phase Touchdown-PCR To increase the product specificity, a touchdown protocol was carried out. The composition of the starting reagents corresponded to that of a standard PCR. The amplification conditions were as follows: number of cycles step 1 5 min at 94 ° C., initial denaturation phase 7 1 min at 94 ° C., denaturation 1 min at 64 ° C., annealing phase, reduction of the annealing temperature by 1 ° C. per cycle 1 min at 72 ° C., synthesis phase 25 1 min at 94 C, denaturation 1 min at 54 C, annealing phase 1 min at 72 C, synthesis phase 1 10 min at 72 C, final elongation step DNA sequencing DNA sequencing reactions were carried out with the Thermo Sequenase TM Cy TM 5 Dye Terminator Kit (Amersham Biosciences, Freiburg) according to the instructions and evaluated with an ALFexpress TM DNA sequencer (Amersham Biosciences, Freiburg).

26 2 Material and methods Electrophoresis AFLP and sequencing reactions were carried out on 0.5 mm thick denaturing PAA gels (7 M urea, 6% PAA, 1 x TBE) with 0.5 x TBE as running buffer in an ALFexpress TM DNA sequencer ( Amersham Biosciences, Freiburg). The same amount of AFLP loading buffer was added to 3-6 μl of amplification products, denatured (90 sec at C) and electrophoretically separated for 540 min at 1500 V, 38 mA, 34 W, 50 C and a sampling interval of 2 sec. The gels were evaluated using the software: ALFwin ™ Version 1.00, ALFwin ™ Sequence Analyzer 2.00 or ALFwin ™ Fragment Analyzer (Amersham Biosciences, Freiburg). To carry out the agarose gel electrophoresis, agarose was dissolved in 1 × TAE to a final concentration of 0.7 to 1.5% and 0.5 μg / ml ethidium bromide was added. The DNA was mixed with 1 x loading buffer (B or Y) and separated in 1 x TAE at V for 1-16 h and evaluated under UV light with a gel documentation system. Fragment isolation from electrophoresis gels AFLP fragments were determined according to Flachowsky et al. (2001) isolated from PAA gels. The fragment was eluted from the gel using the QIAEX II Gel Extraction Kit (QIAGEN GmbH, Hilden) according to the instructions. PCR fragments from agarose gels were eluted using the freeze-squeeze method. The fragment was cut out on the transilluminator with a scalpel. The cut-out piece of gel was then transferred to a 0.5 ml tube, the bottom of which had previously been provided with a piece of viscose wool and three small holes. The tube with the gel piece was shock-frozen in liquid nitrogen, then thawed again, placed in a 1.5 ml tube and centrifuged for 2 min at rpm. The flow-through was purified with 1 volume of phenol / chloroform, precipitated with ethanol, pelleted and dissolved in 20 μl of TE buffer. Cloning of DNA fragments PCR fragments were cloned with the TOPO TA cloning kit (Invitrogen, Kalsruhe). For this purpose, 4 µl PCR product with 1 µl saline solution (1.2 M NaCL,

27 2 Material and methods 22 0.06 M MgCL 2) and 1 µl pcr 2.1-TOPO Vector combined. Depending on the size of the fragments to be ligated, after 5 to 30 minutes of incubation at room temperature, 2 μl of this ligation mixture were placed in a 2 ml tube with 50 μl of One Shot Chemically Competent E. coli cells (Invitrogen, Karlsruhe) and carefully mixed. After incubation on ice for 5 minutes, a heat shock of 30 seconds at 42 ° C. and further storage on ice, 250 ml of SOC medium were added to the cells and these were incubated for 1 hour at 37 ° C. and 200 rpm. Subsequently, 50 and 200 μl each were plated out on LB medium (100 μg / ml ampicillin, 0.1 mm IPTG, 40 μg / ml X-GAL) and grown in the incubator at 37 ° C. overnight. DNA fragments without A-overhangs (e.g. PCR walking fragments) were before ligation in a 10 μl mixture with 1 × PCR buffer (PeqLab, Erlangen), 0.1 mm dntp s and 2-5 U Taq polymerase (PeqLab, Erlangen) incubated for 10 min at 72 ° C. Permanent bacterial cultures Individual E. coli colonies were grown overnight in 2 ml LB Amp100 liquid medium at 37 ° C. and 200 rpm. 800 μl of this bacterial suspension were mixed with 800 μl glycerol, snap-frozen with liquid nitrogen and stored at 80 C. Production of probes DNA fragments were converted into probes by labeling with biotin, digoxygenin (DIG) or radioactive isotopes (32 P). The biotin labeling was carried out by means of PCR. A nucleotide mix was used which contained biotin-labeled dutp. The DIG labeling by means of PCR was carried out analogously using the PCR DIG Labeling Mix (Roche, Mannheim). Large fragments (> 1000 bp) and cloned fragments for which no primers were available were marked with the DIG-Nick Translation Mix (Roche, Mannheim). For this purpose, 1 µg DNA was mixed with 4 µl DIG-Nick Translation Mix in a 16 µl batch and incubated at 15 C for 90 min. The reaction was stopped by adding 1 μl of 0.5 M EDTA and heating to 65 ° C. for 10 minutes.

28 2 Material and Methods 23 Radioactively labeled probes were used to screen a DNA library. For this purpose, the probe DNA was cut with the restriction enzymes HindIII and then radioactively (32 P) labeled with the Rediprime ™ II Random Prime Labellin System (Amersham Biosciences, Freiburg). For this purpose, 16 μl (25-45 ng) of probe DNA together with the labeling mix and 50 μci of 32 P dctp were placed in a reaction vessel and incubated at 37 ° C. for 10 min. The reaction was then stopped at 100 ° C. Southern Blot For Southern blotting, 10 mg of total DNA were digested with 40 U of restriction enzyme each time (Bam-HI, EcoRI, HindIII) in 1 × reaction buffer at 37 ° C. overnight. The 200 μl restriction mixture was precipitated with ethanol, pelleted and dissolved in 20 μl TE buffer. The dissolved DNA was applied to a 0.8% agarose gel and separated at 15V. After the separation, the gel was swirled 2 times for 15 minutes in denaturation buffer for denaturing and then for 2 times for 15 minutes in neutralization buffer. The DNA was then blotted (Southern, 1975) onto a positively charged nylon membrane (Hybond-N +, Amersham Biosciences, Freiburg), washed in 2 × SSC and fixed on the membrane under a crosslinker for 10 minutes using UV light Bacterial colonies Bacterial colonies grown in 384-well plates were stamped onto a positively charged nylon membrane (Hybond-N +, Amersham Biosciences, Freiburg) with the aid of a 384-inch stamp. Each colony was stamped a second time on the same membrane diagonally to the first stamp imprint in order to identify artifacts during hybridization. The membranes with the stamped colonies were placed on LB medium overnight at 37 ° C. to grow. The bacteria were lysed by placing them on 3MM Whatmann paper soaked with 10% SDS for 3 min. The membranes were then placed on Whatmann paper soaked with denaturation buffer for 3 minutes. For neutralization, the membranes were placed 2 times for 3 min on Whatmann paper soaked with neutralization buffer. The membranes were then washed in 2 × SSC and baked at 80 ° C. for 1 h. After treatment with 1 ml proteinase K (1 mg / ml) at 37 ° C. for one hour, the filters were placed between 2 layers of H 2 O-soaked 3MM Whatman paper.

29 2 Material and methods 24 To remove the proteins, the top layer of Whatman paper was pressed onto the membrane with pressure. This process was repeated several times with new layers of Whatman paper each time. The membranes purified from the proteins were then sealed in 2 × SSC and stored in the refrigerator at 7 ° C. DNA-DNA hybridization Nylon membranes with blotted DNA were prehybridized in a hybridization tube for one hour in ml hybridization buffer at 65 ° C. in a rotary oven. The DIG-labeled probe was denatured for 5 min at 95 ° C., stored on ice and added to the hybridization tube together with ml of fresh hybridization buffer and 100 mg / ml herring sperm DNA. Hybridization then took place at 65 ° C. overnight in a rotary oven. This was followed by the stringent washes 2 x 15 min in 2 x SSC and 0.1% SDS at room temperature or in 0.1 x SSC and 0.1% SDS at 65 ° C. After washing for five minutes in washing buffer, incubation followed 1 x Blocking buffer at 37 ° C for 30 minutes. The membrane was then treated with anti-DIG AP and blocking buffer (1: 10000) for 30 minutes. After 5 minutes in the washing buffer, the nylon membrane was treated with detection buffer for 5 minutes. The membrane was then placed between two PE films and incubated for 5 min in detection buffer and CSPD (100: 1). The membrane was welded into a new film and stored in the dark at 37 ° C. for 15 minutes. A chemiluminescent film was then exposed together with the membrane in an exposure box for 2 to 24 hours. Radiolabeled probes were hybridized together with the nylon membranes in 32 P hybridization buffer at 65 ° C. overnight. After the stringent washes (analogous to DIG detection), exposure was carried out on X-ray film or K-Screen (BioRad). Chromosome preparation For the production of chromosome preparations, meristematic tissue from hemp root tips or male hemp flowers was used. Root tip preparations were made by germinating hemp seeds in the dark at 24 C on filter paper moistened with distilled water until the radicles reached a length of 1 to 2 centimeters. To synchronize the cells were

The root tips were incubated for 8 to 17 hours on filter paper soaked with 1.25 mm hydroxyurea solution. In order to enrich the meristem cells of the root tips in the stage of the mitotic metaphase, the radicle tips separated from the seeds were subjected to 19 to 21 hours of ice water treatment or treated with 0.05% colchicine solution for four to six hours. A solution of 3 parts 98% ethanol and 1 part glacial acetic acid was used to fix the root tips. This solution has also been used to fix and hold male hemp inflorescences. Before the production of chromosome preparations, the fixed tissue was subjected to an enzymatic digestion. For this purpose, the tissue was washed 3 × 10 min in 1 × enzyme buffer (40 mm citric acid, 60 mm sodium citrate, pH 4.8) in order to remove the alcohol from the tissue. The plant tissue was then incubated in an enzyme mixture which breaks down cell walls and proteins. For this purpose, root tip tissue was incubated for 1 to 2 hours at 37 ° C. in a mixture of 2% cellulase, 20% pectinase and 1% pectolyase 1x enzyme buffer. During the enzymatic digestion of hemp flowers, 1% cytohelicase was added to this enzyme mixture. The foremost part of the root tips, containing the meristematic tissue, was then removed, transferred to a microscope slide, comminuted in 60% acetic acid and then squeezed with a cover slip. After the enzyme treatment, individual hemp flowers were comminuted in 60% acetic acid on the microscope slide. The slide was then placed on a 50 ° C. hot plate. This tissue suspension was distributed for approx. 60 seconds with circular movements of a dissecting needle. The slides were then rinsed with 3: 1 fixing solution and dried. Completed preparations were examined under a phase contrast microscope to determine the stage of cell division and the quality of the preparation. Suitable preparations were dehydrated in an alcohol series (70%, 90% and 98% ethanol) and then air-dried. The preparations were stored at 20 C FISH (fluorescence in situ hybridization). The fluorescence in situ hybridization was carried out as a multicolor FISH; two differently labeled DNA probes (biotin or digoxygenin labeling) could be used for the hybridization at the same time.

31 2 Material and methods 26 The hybridization was evaluated using incident-light fluorescence microscopes (Zeiss Axioskop). The protocol for performing FISH is given in Table 2.1, Table 2.

32 2 Material and methods 27 Continuation of table 2.1 Amplification of the biotin-labeled probes resp.

33 3 Results 28 3 Results 3.1 Development of male-specific SCAR markers Isolation of male-specific PCR fragments Two male-specific RAPD markers OPC (Fig. 3.1 A) and OPE developed as part of a diploma thesis (Riedel, 2000) should be used in sequence-specific SCAR markers are converted.The male-specific bands of the RAPD amplification products were cut out from an agarose gel, eluted and cloned. 96 positive clones (white colonies) were selected from the OPC fragment and grown in LB liquid medium. In parallel, the insert size of the positive colonies was checked in a PCR with plasmid-specific primers (M13). (Fig.3.1 B). Two clones (clones 4 and 87) had the expected fragment size of ca bp (2700 bp fragment bp plasmid sequence). Permanent cultures were established from both clones. The fragment OPE could not be cloned successfully. AB Fig.3.1: Isolation and cloning of the RAPD marker OPC A: RAPD-PCR with the primer OPC-11 on three male (A2 / 9, A2 / 37 and A2 / 39) and three female plants (A2 / 3, A2 / 13 and A2 / 15). The arrow marks the male-specific fragment (approx. 2.7 kb). B: Checking the insert size: Section from an agarose gel with amplified inserts of the clones. The arrow marks clone 4 with the correct insert size of 2900 bp.

34 3 Results Sequencing of the clones and preparation of sequence-specific primers Clones 4 and 87 of the RAPD fragment OPC were sequenced from both sides to check their sequence and to derive specific primers with universal or reverse sequencing primers. Both clones had identical sequences. The priming sites of the RAPD primer OPC-11 could be found at both ends of the fragment. A sequence comparison of the data of the OPC fragment with the NCBI (National Center for Biotechnology Information) sequence database (blastn) did not reveal any significant sequence homologies. A blastx comparison showed homologies (e = 2-79) to a gag / pol precursor protein of the Gret1 retrotransposon in Vitis vinifera. (Fig.3.2). There were also homologies to gag / pol precursor proteins of retro elements in rice and Beta vulgaris. 2: Blastx comparison of the sequence of the RAPD fragment OPC Comparison of the translated nucleic acid sequence of the fragment OPC (red) with the NCBI protein database (blastx). 44% homology to the gag / pol precursor protein (blue) of a retrotransposon in wine.

35 3 Results 30 The sequence contains three open reading frames (ORF). Of these, a 120 amino acid long ORF showed homologies to the integrase core domain of retroviral elements. In order to amplify the fragment in a specific PCR, the SCAR primer pair C11Komp_L + C11Komp_R was derived (Fig. 3.3). Since the RAPD fragment OPC could not be sequenced in one go due to its size, the primer C11Seq_L lying further within the sequence was derived for further sequencing (sequences of the primers: see Table 7.2 Appendix). 1 AAAGCTGCGG CAACGGGCTG CCCGATATGT CATATATGAT GGAAGATTAT CGTGGAAG CTTCAGTCAA CC 51 ATCGTGGAAG CTTCAGTCAA CCGTTACTTA AATGTATCGA CGGGGAAGAT 101 TGCGACTACG TACTCCGTGA GGTGCACGGA GGTATTTGTG GGAATCATAC 151 TGGTGTTAAT TCCCTTGCCC TAAAAATCAT GCGACAACGG TATTACTGGC 201 CTACCTTGCG ACAAGACACT TTCACTTTCG CAAAAAAATG CGACAAATGT 251 CAGCGGATAG CCACATATGC CCACCAACCT CTGAGCCAGT TGCAGTCCAT 301 CACAAGCCCT TGGCCCTTTG CAGTTTGCGG CATCGACTAG ATAGGTAAAT CAAGTATA TTGCAGTCGC GG 351 TACCCAAAGG AAAAGGCGGA GTCAAGTATA TTGCAGTCGC GGTCGACTAC 401 TTCACGAAAT GGACTGAGGC CAAAGCACTA GCAACCATCA CATCGACTAA 451 GTTTCGTGAG TTTGTCTACA ACTCCATAAT CTTTCGATTT GGCGTCCCTT TGTACGCTAT ACAATGGACT ACAACCAGCT CTTCAAACGA ACCTCTCCGC 51 AAACAGATAG GCGATAACAA CAAAGAGTCT ACCGAGGAAC ATGTAACCGA 1 GAGAGACTTC CGAGCGTAAG AGACCTTAGT TGAACACAAG GGCGTCGAAA AGACTTC CGAGCGTAAG Fig. 3.3: sequence of the two ends of the fragment OPC Shown is the sequence of the ends of the fragment OPC from clone 4. The binding sites of the RAPD primer OPC-11 are marked in green ert. The sequences marked in blue represent the primers C11Komp_L and C11Komp_R. The red sequence marks the primer C11Seq_L.

36 3 Results Test of the SCAR primers on male and female hemp plants To test the specificity of the derived SCAR primers, they were first used in PCRs with DNA from 10 male and 10 female hemp plants of population A2 each. The optimal PCR conditions for the corresponding primers were determined using gradient PCR. With the primers C11komp_L and C11komp_R four bands were amplified at annealing temperatures up to 60 C. By increasing the annealing temperature to 68 C, the number of bands could be reduced to just one band with the expected size of approx. This band could be amplified in all male but also in some female plants. With the primer combination C11Seq_L and C11Komp_R, on the other hand, at a temperature of 65 C, in addition to two to three unspecific bands (clearly visible in Fig. 3.5), an approx.2.2 kb fragment could be amplified that only appeared in male plants (Fig. 3.4). 2.0 kb - 2.0 kb - Fig. 3.4: Test of sex-specific primer pairs Test of the primer pair C11Seq_L and C11Komp_R on 10 male (above) and 10 female (below) plants. The arrow marks the male-specific fragment. The specificity of the primer combination C11Seq_L + C11Komp_R was checked on 74 plants of population A2. The sex-specific cleavage behavior of the combination of the primers C11Seq_L + C11Komp_R could be confirmed by the amplification of the approximately 2.2 kb fragment in all 29 male plants (Fig. 3.5). Table 7.1 (Appendix) shows the marker data for the individual plants.

37 3 Results 32 A * 1 * 2-2.0 kb B - 2.0 kb H 2 0 Fig. 3.5: Test of the male-specific SCAR primer C11Seq_L + C11Komp_R on 74 individual plants of population A2. Amplification of DNA from the 74 plants of population A2 with the primer combination C11Seq_R and C11Komp_L. Traces with DNA from male plants are indicated by red bars. The male-specific fragment of approx. 2.2 kb in size is highlighted by an arrow. A: 100 bp ladder (Fermentas), plants A2 / 1 to A2 / 45, 1 kb ladder (Fermentas); B: 100 bp ladder (Fermentas), plants A2 / 46 to A2 / 100, negative control (H2O), 1 kb ladder (Fermentas). Arrangement of the plants: see table 7.1 * 1 A2 / 11 missing fragment detectable in repeated PCR * 2 A2 / 38 no amplification product can be generated (DNA degradation) 3.2 Development of PAR-specific SCAR markers In previous studies (Flachowsky, 2003; Peil et al. , 2003) specific AFLP markers were found for the pseudo autosomal region (PAR) of the sex chromosomes of hemp. The PAR-specific fragments of the AFLP markers AGA_AAT_330 (Fig. 3.6) and AGA_GAA_510 should be isolated from PAA gels and converted into SCAR markers. For this purpose, the DNA of the plant A2 / 11 was

38 3 Results 33 combination MseI + AAT and HindIII + AGA amplified, loaded onto a PAA gel in order to subsequently isolate the PAR-specific fragment AGA_AAT_330. The same procedure was used for the AGA_GAA_510 marker. Here the PAR-specific fragment from DNA from plant A2 / 9 was amplified. After elution, reamplification and cloning (TOPO TA Cloning Kit, Invitrogen) of the isolated fragments, 96 positive clones each were selected. Then 50 positive clones were amplified with the selective AFLP primers MseI + AAT and HindIII + AGA (marker AGA_AAT_330) or MseI + GAA and HindIII + AGA (marker AGA_GAA_510) and loaded onto a PAA gel. To check the correct insert size, the amplification products of the AFLP-PCR of plants A2 / 8 and A2 / 11 for the marker AGA_AAT_330 and A2 / 9 for the marker AGA_GAA_510 were also applied to the gel (Fig. 3.7). Fig.3.6: AFLP gel with the marker AGA_AAT_330. Section from an AFLP gel of the primer combination MseI + AAT and HindIII + AGA. Plants A2 / 14 are shown; A2 / 15; A2 / 16; A2 / 17 and A2 / 19. The PAR-specific peaks at 330 minutes are marked with arrows. Fig. 3.7: Checking the clones of the AGA_AAT_330 fragment. To check the insert size, 20 clones were loaded onto an AFLP gel with the primer combination MseI + AAT and HindIII + AGA. Plants A2 / 8 (without PAR band) and A2 / 11 (with PAR band) were applied as controls in the first two slots. The arrow marks the PAR-specific fragment.

39 3 Results 34 Five clones (clones: 330_1, 330_2, 330_3, 330_4 and 330_5) of the AFLP fragment AGA_AAT_330 and four clones (clones: 510_A10, 510_B3, 510_B4 and 510_B9) of the AFLP fragment AGA_GAA_510 were selected for further processing Clones and preparation of sequence-specific primers The five (marker AGA_AAT_330) or four (marker AGA_GAA_510) positive clones of the PAR-specific AFLP fragments were sequenced from both sides with sequencing primers to check their sequence and to derive specific primers. The sequences created with the universe or reverse primers overlapped so that the complete sequences of the fragments could be created. From the sequence data of the five clones of the AFLP marker AGA_AAT_330, a consensus sequence could be created for the entire fragment (Fig. 3.8). When evaluating the sequences, the primer binding sites of the HindIII and MseI primers could be found in all clones. The consensus sequence of the AGA_AAT_330 fragment had a length of 280 base pairs and a GC content of 39%. Furthermore, a small CA microsatellite with seven repetitions could be found in the sequence analysis of this AFLP fragment. When the four clones of the AGA_GAA_510 fragment were sequenced, three different sequences were obtained. A binding site for the MseI primer was found at both ends of the clone 510_A10. The remaining clones of this marker could be assigned to two different sequences, each with a HindIII or MseI primer binding site, but of different lengths. The clones 510_B3 and 510_B4 had an identical sequence of 479 bp and a GC content of 50% (Fig. 3.9 above). The clone 510_B9 had a length of 464 bp and a GC content of 39% (Fig. 3.9 below). When the sequence data of the PAR-specific fragments were compared with the NCBI sequence database (blastn / blastx), no significant homologies to known sequences could be found. With the help of the sequence data, specific primer pairs were derived for the respective fragments (Figs. 3.8 and 3.9). The primer pair AAT330Komp (AAT330Komp_L + AAT330-Komp_R) should generate a fragment of 258 bp in length. The length of the fragments that can be amplified by the primer pairs GAA510_B3 (GAA510B3_L + GAA510B3_R) or GAA510_B9 (GAA510B9_L + GAA510B9_R) should be 457 and 443 bp, respectively. In addition to the primers named 330Komp for the entire AGA_AAT_ 330 fragment,

40 3 Results 35 crosatellites of the 330 fragment created an additional primer pair (AAT330CA_L + AAT330-CA_R) (amplicon length 117 bp). For the primer pairs 330Komp, GAA510_B3 and GAA510_B9, the restriction cleavage sites responsible for the AFLP polymorphisms were included in the primer sequences. The sequences of the primers are summarized in Table 7.2 (Appendix). 1 GATGAGTCCT GAGTAAAATA CAGATTGTGT CGTGTTATGT CATAAAAAAA TTAAAATA CAGATTGTGT CGTGTTATG 51 TACGATATTT GTATCGTGCA ATAATTTTAC AGGGTATGCC CGACACGGCC 101 CGCAAGTCTA ATAATTTTGT ATCGTGCTCA GATTCATGAT CAGGGCGGCC AAAGTATGC TCGGACCCAG A 151 AAAAGTATGC TCGGACCCAG ACTCTAACTT ATATAAACCC TTTTACCCCT CAACCCG ACCCCG 201 TCACACACAC ACACAAAAGA AAACGAGAAA AAACTTATAA AGAGTTGGGC 251 ATAAAATACT CTCTAAGCTG GTACGCAGTC TATTTTATGA GAGATTCGAA TATTTTATGA GAGATTCG Fig. 3.8: Consensus sequence of the fragment AGA_AAT_330 The consensus sequence created from 4 clones of the AFLP fragment AGA_AAT_330 is shown. The binding sites of the AFLP primers are marked in color (MseI primer: red, HindIII primer: blue). The selective bases of the AFLP primers are shown underlined. The green and purple marked sequences represent the primers AAT330Komp_L and AAT330Komp_R or AAT330CA_L and AAT330CA_R. Bases highlighted in gray represent the completion of the AFLP restriction cleavage sites. The purple shaded area marks the CA7n microsatellite.

413.9 Sequence of clones AGA_GAA_510_B3 and AGA_GAA_510_B9 Sequence of clones B3 (top) and B9 (bottom) of the AFLP fragment AGA_GAA_510. The binding sites of the AFLP primers are marked in color (MseI primer: red, HindIII primer: blue). The selective bases of the AFLP primers are shown underlined. The green marked sequences represent the sequences of the primer pairs GAA510_B3 and GAA510_B9. Bases highlighted in gray represent the completion of the AFLP restriction cleavage sites.

42 3.2.2 Test of the SCAR primers on hemp plants 3 Results 37 The derived SCAR primers were first used in PCR test series with DNA from selected hemp plants of population A2. The optimal PCR conditions for the corresponding primers were determined using gradient PCR or touchdown PCR. When the primer pair AAT330Komp was tested on 20 plants, an approximately 250 bp fragment was generated in 11 plants at an annealing temperature of 66.5 C. A comparison of the cleavage of the AFLP marker AGA_AAT_ 330 with that of the primer pair AAT330Komp showed that both markers completely cosegregated in all 20 plants. Of the plants examined, 10 each were male and 10 female. The fragment was amplified in 9 male and 2 female plants. The primers AAT330CA for the microsatellite CA7N generated two different bands in a touchdown PCR at 58 ° C. A band with the expected size of approx. 120 bp shows the same splitting as the AFLP marker AGA_AAT_330. The other band, about 490 bp in size, cleaves reciprocally to the AFLP marker AGA_AAT_ 330 (Fig. 3.10). Fig. 3.10: Test of gender-specific primer pairs Test of the primer pair AAT330CA on 16 hemp plants. The first 8 plants are positive for the AFLP marker AGA_AAT_330. In these plants, the expected fragment (120 bp) is amplified with the microsatellite. Co-segregating in addition, the other 8 plants show a fragment approx. 490 bp in size. - 0.5 kb 0.1 kb - Sequencing of the approximately 120 bp fragment confirmed its origin from the AFLP fragment AGA_AAT_330. To subsequently check the specificity of the primers 330Komp and 330CA on a larger number of plants, they were applied to 74 plants of population A2 tested. With the AAT330Komp primers, a fragment approx. 250 bp in length could be generated in 33 plants. This fragment was only amplified in the plants which also have the AFLP-

43 3 results showed 38 markers AGA_AAT_330 (Fig. 3.11). When using the primer pair AAT330CA in a touchdown PCR, the approx. 120 bp fragment could be amplified with the microsatellite in 33 plants. The 490 bp fragment was detectable in the remaining 41 plants (Fig. 3.12). When the primer GAA510_B3 was tested on 20 plants of population A2, even at low annealing temperatures, a fragment could not be amplified in any of the plants. In contrast, a 450 bp fragment could be generated with the primer pair GAA510_B9 at an annealing temperature of 59 C. In gradient PCRs to determine specific reaction conditions, four DNAs each from plants that showed or did not show the AFLP marker AGA_GAA_510 were used. The cleavage of the primer GAA510_B9 did not correspond to that of the AFLP marker AGA_GAA_510. Of the four plants which showed the AFLP marker, a fragment could only be amplified in two plants. In contrast, three of four plants negative for the AFLP marker had a fragment. In order to determine possible polymorphisms within the amplificate of the primer GAA510_B9, a CAPS (cleaved amplified polymorphic sequences) analysis was carried out. The presence and number of corresponding cleavage sites for restriction enzymes could be determined by analyzing the sequence of the AFLP fragment AGA_GAA_B9. In the CAPS analysis, amplicons of the primer pair GAA510_B9 of a plant that showed or did not show the AFLP marker AGA_GAA_510 (A2 / 4 or A2 / 1) with the restriction enzymes AluI, BglII, RsaI, Sau3AI, SspI, TaqI or XhoI cut. When the restriction products were analyzed on an agarose gel, no polymorphism could be found for any enzyme. In an SSCP (single strand conformation polymorphism) analysis of the two fragments, no differences could be found either.

44 3 Results 39 A * 1 - 0.5 kb - 0.1 kb B * 2 * 3 - 0.5 kb - 0.1 kb Fig. 3.11: Test of the PAR-specific SCAR primer AAT330Komp on 74 individual plants of the population A2. Male plants are marked by red bars. Plants with different gender cleavage are indicated by green or blue bars (male green, female blue). The arrow marks the approx. 250 bp PAR-specific fragment. A: 100 bp ladder (Fermentas), plants A2 / 1 to A2 / 44, 1 kb ladder (Fermentas) B: 100 bp ladder (Fermentas), plants A2 / 45 to A2 / 100, blank sample (H2O), 1 kb ladder (Fermentas). Arrangement of the plants: see Table 7.1 (Appendix) * 1 A2 / 38 no amplification product can be generated (DNA degradation) * 2 A2 / 49 missing fragment detectable after repeated PCR * 3 A2 / 71 no AFLP data available *