@prefix vivo: . @prefix edm: . @prefix ns0: . @prefix dcterms: . @prefix dc: . @prefix skos: . vivo:departmentOrSchool "Forestry, Faculty of"@en ; edm:dataProvider "DSpace"@en ; ns0:degreeCampus "UBCV"@en ; dcterms:creator "Sun, Li-Juan"@en ; dcterms:issued "2009-02-06T22:29:16Z"@en, "1996"@en ; vivo:relatedDegree "Master of Science - MSc"@en ; ns0:degreeGrantor "University of British Columbia"@en ; dcterms:description """DNA-level variability in Endocronartium harknessii (Moore) Hiratsuka, the cause of western gall rust On lodgepole pine (Pinus contorta Dougl.), was assessed using RAPD's and RFLP's. The material analyzed consisted of aeciospores from 120 wild isolates collected at 12 geographically separate locations (10 galls each) ranging from the interior (6 locations) to the coast (4 locations) of the province of British Columbia with two collections from Manning Park, representing one of the passes between these regions. Using the RAPD technique, of 180 primers screened, 64 yielded clear polymorphic bands. Of these 13 were chosen for the main study. These 13 primers yielded 96 polymorphic bands. All galls exhibited a unique RAPD-type. Principle Component Analysis (PCA) revealed that coastal and interior collections were quite distinct, with the exception of two adjacent interior collections (Quesnel and Ten Mile Lake). On the other hand, of the two collections from Manning Park, the high elevation one was clearly related to the interior population, while the low elevation one was clearly coastal indicating that there is probably limited gene flow between the two regions. Dendrograms constructed, using unweighted pair group arithmetic mean analysis (UPGMA) based on similarity measure for individual galls separated most local collections into distinct clusters, and revealed relationships between local collections similar to PCA. The RFLP technique was used for a small sample of galls and revealed considerable variation between them. However, since this technique was rather laborious and required larger amounts of DNA than could be extracted from spores produced by small, young, clean galls, it was not suitable for a large population study. Analysis of DNA from different sectors of large old galls using RAPD's showed some polymorphisms between sectors. The origin of these polymorphisms is not clear, A species-specific dsRNA was detected in several galls, as well as in collections of related Cronartium rusts, suggesting the presence of a mycovirus in these rust populations."""@en ; edm:aggregatedCHO "https://circle.library.ubc.ca/rest/handle/2429/4251?expand=metadata"@en ; dcterms:extent "6271375 bytes"@en ; dc:format "application/pdf"@en ; skos:note "DETECTION OF GENETIC VARIATION AMONG ENDOCRONARTIUM HARKNESSHfOVVLATlONS IN BRITISH COLUMBIA USING rDNA RFLPS AND RAPD MARKERS by LI-JUAN SUN B.Sc. Northeast Forestry University, Harbin, PRC 1980 M.Sc. Chinese Academy of Forest Sciences, Beijing, PRC 1983 THESIS SUBMITTED IN THE PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES (Department of Forest Sciences and The Biotechnology Labotatory) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA February 1996 © Li-Juan Sun, 1996 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of rWf>t\" W f . . A/tlll fe l ^ f e . [ The University of British Columbia Vancouver, Canada DE-6 (2/88) ABSTRACT DNA-level variability in Endocronartium harknessii (Moore) Hiratsuka, the cause of western gall rust On lodgepole pine (Pinus contorta Dougl.), was assessed using RAPD's and RFLP's. The material analyzed consisted of aeciospores from 120 wild isolates collected at 12 geographically separate locations (10 galls each) ranging from the interior (6 locations) to the coast (4 locations) of the province of British Columbia with two collections from Manning Park, representing one of the passes between these regions. Using the RAPD technique, of 180 primers screened, 64 yielded clear polymorphic bands. Of these 13 were chosen for the main study. These 13 primers yielded 96 polymorphic bands. All galls exhibited a unique RAPD-type. Principle Component Analysis (PCA) revealed that coastal and interior collections were quite distinct, with the exception of two adjacent interior collections (Quesnel and Ten Mile Lake). On the other hand, of the two collections from Manning Park, the high elevation one was clearly related to the interior population, while the low elevation one was clearly coastal indicating that there is probably limited gene flow between the two regions. Dendrograms constructed, using unweighted pair group arithmetic mean analysis (UPGMA) based on similarity measure for individual galls separated most local collections into distinct clusters, and revealed relationships between local collections similar to PCA. The RFLP technique was used for a small sample of galls and revealed considerable variation between them. However, since this technique was rather laborious and required larger amounts of DNA than could be extracted from spores produced by small, young, clean galls, it was not suitable for a large population study. ii Analysis of DNA from different sectors of large old galls using RAPD's showed some polymorphisms between sectors. The origin of these polymorphisms is not clear, A species-specific dsRNA was detected in several galls, as well as in collections of related Cronartium rusts, suggesting the presence of a mycovirus in these rust populations. iii TABLE OF CONTENTS ABSTRACT....... . ... H TABLE OF CONTENTS . . . J;t1,.,^::,.= iv LIST OF TABLES....: vi LIST OF FIGURES vii ACKNOWLEDGMENTS viii CHAPTER I GENERAL INTRODUCTION 1 Western gall rust distributions and impact 2 Taxonomy and nomenclature 3 Life cycle and biology 5 Genetic variation in pathogenicity 7 Population structure of WGR 8 Methodology for studying population structure in plant pathogens 9 Objectives 13 CHAPTER II RFLP rDNA marker 14 INTRODUCTION 15 MATERIALS AND METHODS 17 Sample collections. 17 DNA extraction 17 DNA digestion and hybridization 19 Screening for polymorphic markers 20 Population variation assay 20 Data analysis 20 RESULTS 21 DNA extractions 21 Screening of enzyme and probe combination 21 Size of ribosomal DNA 21 Polymorphic markers 23 Variation detected by RFLP rDNA marker 23 DISCUSSION 28 CHAPTER HI RAPD markers 31 INTRODUCTION 32 MATERIALS AND METHODS 34 Development of RAPD protocols 34, 1. DNA isolation method 34 2. DNA amplification 36 iv 3. Selection of primers 36 4. R A P D reproducibility....... 36 Sample collections 38 Data analysis... < 40 R E S U L T S : 41 D N A extraction..; ... 41 R A P D reproducibility 41 R A P D profile 44 Variation within and among populations 50 Genetic variability within individual galls. • 53 D I S C U S S I O N 57 CHAPTER IV General conclusion 69 BIBLIOGRAPHY 73 A P P E N D I X I 82 A P P E N D I X II 90 A P P E N D I X III: 91 A P P E N D I X I V 92 v L I S T O F T A B L E S Table I . D N A fragments (kb) generated by 12 restriction enzymes and 6 heterologous probes ; 22 Table 2. Polymorphisms detection from seven populations by single and double enzyme digestions 24 Table 3. Spore sources and viability tested for D N A mini preparation 35 Table 4. Sequences and number of bands yielded by selected primers 47 Table 5. Number of banding patterns derived from individual galls 56 Table 6. Data matrix of RFLP assays 90 Table 7. Average frequencies of polymorphic bands observed in 12 populations 91 Table 8. Data matrix of 120 individual galls and 96 polymorphic RAPD bands 92 vi LIST OF FIGURES Figure 1. Geographic range of Endocronartium harknessii populations for RFLP assay : 18 Figure 2. Autoradiograph of southern hybridization of D N A from seven locations . screened with Avall/pAm2 25 Figure 3. Autoradiograph of southern hybridization of D N A f r o m f i v e locations and 24 individuals screened with AvaII/pAm2 26 Figure 4. Dendrogram generated based on the RFLP polymorphic bands 27 Figure 5. Geographic range of populations for RAPD assays 39 Figure 6. Phenol- Chloroform and Dry ice DNA extraction from rust spores 42 Figure 7. Spore contamination tests 45 Figure 8. RAPD reproducibility under various conditions 46 * Figure 9. Percentage of bands in the range of polymorphic band present frequencies 48 Figure 10. Examples of RAPD banding patterns with Endocronartium harknessii 49 ; Figure 11. Principal Components analysis based on covariance matrix. 11a, Plot of 1st and 2nd principal components for 120 individuals 51 l ib , Plot of 1st and 2nd principal components for 12 populations 52 Figure 12. Dendrogram of 120 individuals galls 54 Figure 13. Dendrogram of 12 populations 55 Figure 14. Dendrograms generated using randomly chosen primers 14 a, Dendrogram generated by five randomly selected primers (34 RAPD bands) from the 13 primers used in population assay 60 14 b, Dendrogram generated by seven randomly selected primers (60 RAPD bands) from the 13 primers used in population assay 61 vii A C K N O W L E D G M E N T S This project was funded by National Science and Engineering Research Council of Canada grants to Dr. B. J. van der Kamp and Dr. J. E. Carlson. The guidance from my senior supervisor, Dr. B. J. van der Kamp and research supervisor Dr. J. E. Carlson throughout my studies is greatfully acknowledged. I also would like to thank Drs. J. E. Carlson and E. E. White for their contribution of molecular expertise and lab support for this project. And thanks to members of Dr. Carlson's lab in the Biotechnology Laboratory University at the British Columbia and Dr. White's lab at the Pacific Forestry Centre for their helpfulness and friendship. I also owe a debt to Dr. Z. K. Punja* for allowing me taking some time off from work to finish writing this thesis. Special thanks to my family for understanding and invaluable support. * Associate Professor, Director of Center of Pest Management, Simon Fraser University, Burnaby, BC, Canada viii CHAPTER I GENERAL INTRODUCTION G E N E R A L I N T R O D U C T I O N Western gall rust disease distributions and impact Western gall rust (WGR) is a transcontinentally distributed hard pine stem rust. It ranges from Baja California on the Pacific coast to the Yukon and eastward across Canada and the Northern United States to the Atlantic coast and is indigenous to North America (Ziller, 1974). The host range, includes most native hard pines, namely, Jack pine (Pinus banksiana Lamb./ lodgepole pine (P. contorta Dougl.), ponderosa pine (P. ponderosa Laws), Bishop's pine (P. muricata D. Don), Monterey pine (P. radiata D. Don.) as well as exotic hard pines, Scots pine (P. sylvestris L.), maritime pine (P. pinaster Ait.), Swiss mountain pine (P. mugo Turra) and Austrian pine (P. nigra Arnold/ In British Columbia, the native hosts are lodgepole pine, jack pine and ponderosa pine and the geographic range of the disease coincides with the range of the above three hosts. The woody, globose, perennial galls grow larger from year to year, producing spores each spring, until the branches on which they are located are killed. In naturally regenerated stands, damage caused by gall rust is often minor and average mortality of young trees resulting from stem infection is only about 5-10%. Mortality in individual stands ranges from 0-50%. Usually, naturally regenerated stands are well-stocked, and a loss of as much as 25% of stems is not damaging. Western gall rust often acts as a natural thinning agent in such stands and is not known to destroy whole stands (van der Kamp, 1988b and 1994). However, damage can be more severe in plantations and Christmas tree farms. Powell & Hiratsuka (1973) described a plantation of Central Alberta in which about 63% of the young lodgepole was infected. In British Columbia, an average of 22% of trees in provenance trials less than 10-year-old in Northern Interior were infected and in some plantations infection exceeded 50% (Martinsson, 1980; Yanchuk et al., 1988; van der Kamp, 1987). In contrast to the naturally regenerated stands, plantations have fewer stems 2 Kamp, 1987). In contrast to the naturally regenerated stands, plantations have fewer stems per hectare and are therefore damaged more easily. Even though rust does not kill trees directly since the obligate organism's survival depends on the host being alive, branch galls on seedling or young trees do cause malformed boles. Stem galls can girdle young trees quickly and eventual breakage at the gall results in reduced stand density. Cones can be infected by western gall rust (Byler and Piatt, 1971) which causes seed reduction and stem infection. Taxonomy and nomenclature The genus Endocronartium, Uredinales, Basidiomycotina was established in 1969 by Dr. Yasu Hiratsuka to contain species with an endocyclic life cycle and Cronartium-like aecioid telia (Hiratsuka, 1969). Peridermium is a form-genus for rusts in which the telial stage is either unknown or does not occur. Both names of Endocronartium and Peridermium have been used for western gall rust disease. The taxonomy and nomenclature of this fungus has been in dispute since 1969. The two names have appeared with roughly equally frequency in the literature. The name of Endocronartium indicates a rust that reproduces sexually, meiosis taking place at spore germination, and the branches of the germ tube being homologous to basidiospores (Hiratsuka, 1969), while Peridermium indicates a rust that reproduces asexually by true aeciospores. In Endocronartium, aeciospores actually function as teliospores (Hiratsuka, 1969) so that it is called an endocyclic, autoecious rust. In contrast, Peridermium, the anamorph of Cronartium is called a microcyclic, autoecious rust. Christenson (1968) described the cytology of western gall rust collected from a wide range of locations in the United States and showed no support for meiosis, but revealed that consistent differences in nuclear behavior and germ tube morphology exist between heteroecious (Cronartium spp.) and autoecious forms (Endocronartium). Laundon (1976) questioned whether there was sufficient evidence for karyogamy and meiosis in Endocronartium harknessii. 3 cytogenetic studies. They concluded that spore production of western gall rust is via mitosis. Tuskan and Walla (1989) studied variation within western gall rust (Endocronartium harknessii) and eastern gall rust (Cronartium quercuum) populations by using the isozyme technique. Their results showed western gall rust populations to be heterogeneous and homozygous while the eastern gall rust population was heterogeneous and heterozygous. Thus, they concluded that western gall rust has a typical asexually reproduced population structure. Vogler et al. (1987; 1991) collected aeciospores from Pinus radiata, P. muricata and P. coulteri in California and employed starch gel electrophoresis to study isozyme patterns. The results of this study revealed that regardless of hosts or regions, enzyme patterns of most spore sources were monomorphic. This again indicated that western gall rust reproduces vegetatively. Therefore it has been suggested that the old name Peridermium harknessii Moore be used (Epstein and Buurlage, 1988; Vogler et al., 1987 &1991; Tuskan and Walla 1989). This name implies that the fungus has an incomplete life cycle reproducing by the asexual true aeciospore. Nevertheless, the origin of the binucleate aeciospore is still unknown. In fact, the number of nuclei in mature spores is in dispute. Hiratsuka et al. (1966) reported that the majority of mature spores to be uninucleate (Hiratsuka, et al., 1966); Christenson (1968) Epstein and Buurlage (1988), and Kojwang (1989) found the majority to be binucleate; while Vogler et al. (1989) reported both types distributed in different geographic locations. Ziller (1970) showed that the autoecious gall rust occurs in Western Canada by inoculation test. He compared germ-tube morphology of autoecious Endocronartium with that of the Peridermium stage of normal Cronartium rusts, and found that the endocyclic autoecious germtube has short branches and cells separated by septa while the germ tube of Peridermium is aseptate (Anderson et al., 1965). Consequently, western gall rust is called Endocronartium harknessii (Moore) Hiratsuka Y. in Western Canada. Since much of the evidence for the lack of a sexual stage in western gall rust originates from USA collections, mycologists and forest pathologists have questioned 4 originates from USA collections, mycologists and forest pathologists have questioned - whether Canadian races might differ from USA races. Kojwang (1989) studied cytology of aeciospores collected in British Columbia and found no evidence for meiosis at spore germination. Hiratsuka et al. (1966) reported that nuclear fusion and meiosis occur within the first hour of spore germination, while the studies described above by Christenson (1968), Epstein and Buurlage (1988) and Kojwang (1989) did not observe the nuclear state in the germtube until at least an hour after germination. Hence there is a time gap among the observations. Hiratsuka (1991) strongly held that 'germ tubes of western gall rust should be considered homologous to basidia rather than germ tubes of an anamorphic fungus. This fungus should be recognized as a perfect fungus having an endocyclic life cycle. The name Endocronartium harknessii is the appropriate name of the pathogen of the western gall rust'. Until now, the taxonomy and nomenclature of the pathogen of western gall rust remains unresolved. Apart from taxonomic and nomenclatural considerations, resolution of this issue bears critically on disease management and resistance breeding, since it will determine whether recombination of virulence genes in spore genotypes is common and thus will affect strategies for deploying host resistance genes. Life cycle and biology Aeciospores attack the elongating shoots of pine in late spring or early summer. Allen and Hiratsuka (1985) showed that uninucleate hyphae penetrate the cuticle and epidermis, and then grow through the cortex and newly formed phloem to the cambium. There the rust stimulates the cambium to produce more cells to form the gall. Hyphae within infected tissues are uninucleate (True 1938; Hiratsuka et al., 1966; Hopkin and Reid 1988). Fungal development can be through to the pith, with longitudinal growth associated with resin canals and vascular tissue, but the most abundant mycelium is located in the phloem. Gall formation after inoculation occurs at different times. Small 5 following year or two years after infection (Allen et al., 1989; Kojwang 1989; Schulting, 1988). Spores are produced once a year in late spring or early summer and coincide with the newly growing shoots of pine. Spores are produced annually unless hyperparasites (Byler and Cobb, 1969) or rodent feedings (van der Kamp, 1987) cause cessation of spore production. Morphologically, two types of spores have been found in WGR's life cycle, namely, spermogonia and aeciospore or aecioid teliospore. Spermogonia are rare in natural conditions in Western Canada (van der Kamp, per. comm.; Dennis, per. comm.; Walla et al., 1991b; Crane et al., 1995). The role of spermogonia in the life cycle is unclear since aeciospore production occurs in the absence of spermogonia (Crane et al., 1995). Reports of spermogonia have increased in recent years ( Walla et al., 1991b, Crane et al., 1995). However it is not understood how spermogonia are related to sexuality in the fungus. Studies on nucleic condition of the mature spores are rather diverse as described above (Hiratsuka et al., 1966; Christenson, 1968; Hiratsuka, 1991; Epstein and Buurlage, 1988; Kojwang, 1989; Vogler et al., 1991). The origin of the two nuclei in the aeciospore (where that occurs), that is whether they are biparental, one being introduced via spermatia, or uniparental, is still in question. Axenic culture of most rusts is routinely done. Various levels of success have been achieved in culturing over 25 species of rusts (Yamaoka and Katsya, 1985). In 1988, Allen and others first cultured Endocronartium harknessii axenically from the infected tissue at the active stage prior to sporulation, although, the cultures grew very slowly and did not have much mycelium. Lundquist et al. (1994) reported similar results for axenic culture of Peridermium harknessii from mature gall tissue. All successful attempts made so far are limited to culturing western gall rust from the infected tissue. Cultures derived from spores or single spore are not available. 6 Genetic variation in virulence of WGR van der Kamp (1988a) reported there was variation in virulence among geographically separated spore sources as shown by inoculation of a set of pine provenances. Coastal sources appeared more virulent than the interior spore sources when inoculated on coastal trees. Some work has also been done using half-sib families and clones of lodgepole pine with single-gall spore sources of western gall rust. Schulting (1988) inoculated a set of open pollinated pine families with five spore sources, all from Prince George, and each derived from a single gall. Spore sources varied in virulence and families in resistance, but there was no evidence of interaction of these two. Kojwang and van der Kamp (1991) used sixteen grafted lodgepole pine clones and four single-gall spore sources from Prince George to test pathogen-host interactions. The results revealed that one spore source had very low infection performance and significant interaction of pathogen-host was detected. This outcome may suggest that genetic variation, in terms of variability in virulence exist among spore sources. However, without replication of spore sources, it was impossible to distinguish this interpretation from variation in spore quality due to contamination, etc. All spores from a single gall are commonly presumed to be genetically uniform (Vogler et al., 1988; Tuskan and Walla, 1989; Schulting, 1988; Kojwang, 1989) because galls presumably arise from infection by a single spore. However, Allen et al. (1989) pointed out that multiple gall formation is common in artificially inoculated seedlings. The difficulty of identifying the variability in virulence among different groups of the host (other than clones) and spore sources might be due to variation in resistance of the host masking variation in the pathogen since lodgepole pine is a most variable species (Hamrick et al., 1981; Yeh et al., 1985). This difficulty is further exacerbated by the difficulty of producing clonal host lines and the inability to cross pathogen lines. 7 Population structure of WGR Population structure refers to the amount of genetic variation among individuals in a population and its distribution in space. Elucidation of population structure of a pathogen will help to provide genetic insight in the pathogen. Population structure is to a great extent determined by the particular reproductive system of the pathogen. Thus, the genetic structures are very different for pathogens with sexual or asexual reproductive systems (Burdon and Roelfs, 1985; McDonald and Martinez, 1991; Brown et al., 1990; Kohli et al., 1992). Vogler et al. (1987 and 1991)) collected gall rust aeciospores from Pinus radiata, P. muricata and P. coulteri in California and conducted starch gel electrophoresis on them. These studies disclosed that isolates of the rust collected along the California coast were monomorphic whereas those from the inland mountains were polymorphic among locations and monomorphic within locations in terms of isozyme patterns regardless of host or stand origin. Tuskan and others (1990) used 201 isolates of Peridermium harknessii from 13 disjunct geographic locations and three host species to study genetic variation using the isozyme method. Geographic location and stand type appeared to be significant factors. Two genetic distance clusters were identified, namely western locations and eastern locations in which the stand types (windbreaks, plantations, Christmas tree farm and natural forest) were confounded with locations associated with cluster I and cluster II. Such results suggested that western gall rust populations are not panmictic. In British Columbia, information about population structure of western gall rust is not available. 8 Methodology for studying population structure in plant pathogens Virulence, morphology and cultural characteristics Variation in virulence is an important aspect of population structure for most plant pathogens. However, little information is available on genetic variation in virulence of western gall rust, van der Kamp (1988a) reported that coastal rust collections were more virulent than interior ones on coastal host populations, but not on interior host populations. Variation in early symptoms have been observed on lodgepole pine inoculated with western gall rust. Schulting (1988) reported that the frequency of red stain induced by five spore sources varied from 6.92% to 20.65%. That study also found that early symptoms ( red stain) were positively correlated with gall formation. Since the five spore sources were derived from single galls, genetically determined variation in virulence was therefore confounded with possible variation in spore quality (age, host effects, contamination). Yanchuck et al. (1988) studied a lodgepole pine progeny trial consisting of 214 open-pollinated families from 24 provenances in British Columbia. Provenances closer to the coast and from higher elevations had the highest infection level. Morphological variation is applicable in studying population structure in some fungi. Most fungi also can be distinguished by their characteristics in culture. Biological species of Armillaria, the well known causal agent of root rot, can be distinguished by compatible mating interaction to determine mating type variation (Ullrich and Anderson, 1978). This is impossible in western gall rust. Attempts have been made to find variation in morphological and cultural characters of western gall rust. For instance, Christenson (1969) and Walla and Lundquist (1991a) reported that albino and orange color spores are present in nature; the former being rare. Other than coloration, cytology and virulence differentiation were not found. Two color type colonies, namely white and orange colonies in axenic culture were reported by Lundquist et al. (1994). No differences in virulence have been found between these two colony types. 9 In summary, the study of variation in WGR using conventional methods is hampered by the nature of this pathogen. Thus, biochemical or molecular markers need to be considered. Allozyme Allozyme markers have been widely employed to study the population structure of fungal organisms ( Micales, 1986; Burdon and Marshall 1983; Soltis et al., 1989). The patterns of population structure produced by vegetative or sexual reproduction can be distinguished by isozymes (Burdon and Roelfs, 1985). This is a versatile and inexpensive technique. However, there are many limitations of isozyme markers, such as insufficient variation in some organisms, especially in intraspecific analysis, that result from the limited number of genes that can be assayed biochemically. Allendorf (1977) stressed that electromorph identity did not mean identity in DNA base sequence: homology is a 'conditional' concept for isozyme phenotypes. Protein secondary structure, timing of sample collection, aging of the samples, quick thawing and freezing of the samples may all affect the protein migration on the gel that could lead to underestimation or overestimation of polymorphism (Murphy et al., 1991). Some of these factors may account for the lack of variation in western gall rust as revealed by isozyme studies (Vogler et al., 1987 and 1991; Tuskan et al., 1990), which suggests that an alternate genetic marker system is required for detailed population differentiation studies of western gall rust. RFLP (Restriction Fragment Length Polymorphisms) The discovery of restriction enzymes in 1970 provided a simple method for cutting duplex DNA molecules into discrete, reproducible fragments. Restriction endonucleases are enzymes that cut DNA at a constant position within a specific recognition sequence, typically 4-6 base pairs long. Different sized fragments can only be due to sequence 10 differences. If two samples which are digested by a particular restriction enzyme, produce different sets of fragments then the two samples do not have the same DNA base sequence. Gene(s) of interest can be visualized on Southern blots (Maniatis et al., 1982). The method of Southern blotting is based upon the availability of cloned sequences that can be utilized to probe for specified genes or for the presence of variation at the DNA level. Specifically, this variation is monitored as changes in the length of homologous fragments produced by digestion of the DNA sample with restriction endonucleases, and is therefore termed Restriction Fragment Length Polymorphisms' (RFLPs). Restriction fragment length polymorphisms have been studied in many organisms including humans (Botstein et al., 1980), crop plants (Rivin et al., 1983) and fungi (Metzenberg et al., 1984; Wu et al., 1983; Specht et al., 1984; Anderson et al., 1987). RFLPs are very useful molecular markers in phylogenetic and parentage studies. Length and restriction polymorphisms are common between different strains within a species and this condition has been shown to exist in several fungi (Cassidy et al., 1984). The polymorphisms can be the result of one base pair differences (point mutation), insertion or deletion, and sequence rearrangement. Numerous studies have used restriction-site diversity to infer plant pathogen population genetic structure (McDonald and Martinez, 1990), to test strain variations (Specht et al., 1984; Hartung and Civerolo, 1989), to differentiate biological species (Anderson et al., 1987), and to study fungal pathogenesis (Michelmore and Hulberts, 1987; Leong and Holden, 1989; Levy et al., 1991). Restriction fragment length polymorphisms (RFLP) are the most frequently used DNA marker in last two decades. Our pilot study with RFLP markers showed high variability within and among western gall rust populations (Sun et al., 1995b and Chapter II, this thesis). However the use of RFLP markers for WGR population studies requires large amounts of clean spores and a large amount of time to get enough pure DNA. The need for a less tedious method is obvious when a large population survey is contemplated. 11 RAPD (Randomly Amplified Polymorphic DNA) The development of the polymerase chain reaction (PCR) (Mullis and Faloona, 1987) opened new opportunities for the analysis of nucleotide sequence variability. The principle behind PCR is quite simple. Short oligonucleotides (20- 40 bp long) are assigned, homologous to flanking regions on either side of the DNA sequence to be amplified. The oligonucleotide primers are added in great excess to the template DNA, in the presence of buffer, DNA polymerase and free nucleotides. The template DNA is then denatured at 94-95°C and cooled to 40-60°C to allow primer annealing. The temperature is then adjusted to allow extension by the aid of thermally stable DNA polymerase and the process is repeated to 25-40 times. The amplification of DNA is exponential. The major limitation of standard PCR is the requirement of DNA sequence information (Innis et al., 1990). A method described by Williams et al. (1990) for identification of polymorphisms based on PCR is not dependent on having prior knowledge of a DNA sequence. This DNA marker system named RAPD (random amplified polymorphic DNA) is based on the amplification of unknown DNA sequences using single, short (8 to 12 bp), random sequence oligonucleotide primers. The ease and speed of RAPD marker system overcomes many technical limitations of RFLPs and has been used in plant pathology for detection of genetic variation between races (Crowhurst et al., 1991; Goodwin and Annis, 1991; Hamelin et al., 1993; Assigbetse et al., 1994;), differentiation of virulent or avirulent isolates (Guthrie et al., 1992), identification of disease-resistance genes (Michelmore et al., 1991) and population structure analysis (Smith and Stanosz, 1995; Smith, et al., 1992, 1994). One difficulty of using RAPD markers in population studies is that RAPD markers are dominant (Williams et al., 1990). This provides only a single amplifiable allele per locus making homozygotes and heterozygotes indistinguishable. This will reduce the accuracy of estimation of allele frequencies for population genetic analysis (Lynch and Mulligan, 1994). However, this limitation can be avoided by use of binary data to assess 12 the amount of genetic variability and relatedness coupled, with similarity and cluster analysis (Kambhampati et al., 1992, Chen et al., 1993). Objectives Lodgepole pine is a very important species in British Columbia, occupying millions of hectares of forest land. The species regenerates naturally following logging on most sites, and it is often planted on sites where it did not grow before harvest. Hence it will increase in abundance in future forests by reforestation. Approximately 35 million lodgepole pine seedlings are planted annually and this is projected to reach 73 million trees annually by the turn of this century in the province (Ying et al., 1985). WGR is one of the common stem rusts on this host which has been studied for a long time. Elucidation of the pathogen population structure might help us to understand the behavior of the native pathogen populations in natural conditions and will certainly guide us to deal with this disease in altered conditions. This study was designed using rDNA RFLP and RAPD markers to study genetic variation of WGR population in British Columbia with the natural host of lodgepole pine. 13 CHAPTER II DETECTION OF GENETIC VARIATION OF WESTERN GALL RUST POPULATIONS USING rDNA RFLP MARKERS 14 INTRODUCTION The genetics of plant pathogenic fungi generally has been difficult to study because of a lack of easily assayed genetic markers. Most agriculturally important fungal species have been examined using virulence characters as the principal genetic markers. Unfortunately, many phytopathogenic fungi do not possess clearly defined gene-for-gene interactions that make them amenable to genetic analysis using virulence markers, and other genetic markers need to be developed for these fungi. Endocronartium harknessii is one of these fungi. Isozyme markers have limitations for genetic variation studies, especially at the population level (Murphy et al., 1991; Liu and Furnier, 1993; Avise,1994). Isozyme markers identify only those mutations in genes that change the electrophoretic mobility or the stability of enzymes. With DNA markers, either restriction fragment length polymorphisms (RFLP) or DNA sequence variants, a much larger fraction of the genome can be surveyed for within- and between-species differences. The methods for studying population variation in western gall rust are extremely limited because of the lack of morphological variation and the inability to cross individuals in vitro. E. harknessii has no known physiological markers such as specific virulence genes or vegetative incompatibility groups. Virulence variation is thought to exist (van der Kamp, 1988a), but is not clearly defined. Isozyme markers showed little to no variation among isolates (Vogler et al., 1991; Tuskan et al., 1990) which may suggest that isozymes are not the most appropriate method for studying genetic variability within this species. The use of ribosomal DNA (rDNA) as a genetic marker for genetic population structure study in plants and fungi has increased in recent years (Rogers and Bendich, 1987; Zimmer et al., 1988; Bobola et al., 1991; Appels and Dvorak, 1982). Rogers and Bendich (1987) reported that the number and intergenic spacer (IGS) length of ribosomal genes can be so variable among individuals that they can be used to test genetic variability within a population. Ribosomal DNA has been used to describe both the genetic structure 15 variability within a population. Ribosomal DNA has been used to describe both the genetic structure within fungal populations and the phylogenetic relationships among species of fungi. Variability in the IGS region between the tandemly repeated ribosomal genes has provided informative genetic markers for fungal population studies, and variation both within and among individual fungi have been detected (Anderson et al., 1987, 1989; Wu et al., 1983). The objective of this study was to determine the usefulness of rDNA RFLP markers for evaluating population variability in western gall rust. 16 MATERIALS AND METHODS Sample collection Most samples of aeciospores were derived from single galls, and unless otherwise noted, each gall came from a different tree. In some instances spores from several galls were bulked. A total of 27 spore samples was collected in 7 locations (Fig. 1). Spores were extracted from galls by brushing and sieving through a nylon cloth into a petri dish and then drying at room temperature for 24 hours. Air-dried spores were transferred to sterile 1 ml Eppendorf centrifuge tubes and stored at -20° C. The geographic origin of galls used in this study is described in Figure 1. DNA extraction The DNA extraction method for RFLP analysis was adapted from White et al. (1994) with minor modifications: For each sample, 4 ml of PTE (50 mM Tris-HCl, 10 mM EDTA, pH7.8 ) buffer containing 120ul of 10% Triton-X-100 and lmg /ml proteinase K was prepared freshly. An open wiggle-bug (Gresent Dental MFG Co. 1.35A 1 IB Canada PAT 399.762) capsule with the ball bearing in the bottom half was placed in liquid nitrogen. When rapid liquid nitrogen boiling had stopped, the capsule was removed from liquid nitrogen, both halves were placed in a polystyrene holder. Liquid nitrogen was added once more to keep the capsule frozen. Fifty to one hundred milligrams of fresh spores or spores stored at -20° C were poured slowly from a weighing boat into the bottom of the wiggle-bug capsule avoiding liquid nitrogen which would bump and blow the spores out. The capsule was closed and placed in the wiggle-bug; grinding was for exactly 1 minute. Then the capsule was opened and placed in a polystyrene holder. Four ml PTE buffer was added into both halves of the capsule. Macerate was transferred into a 5 ml graduated test tube. The tube was inverted gently a few times to mix and then left at room 17 Figure 1. Geographic range* of Western gall rust sampled for RFLP assay \" 1. Nation River (NR) 4 single galls from individual trees 2. Vanderhoof (VH) 5 single galls from individual trees 3. Prince George (PG) 5 samples from bulked galls which derived from single gall inoculation 4. Quesnel (QS) 4 single galls from individual trees 5. Manning Park (MP) 4 single galls from individual trees, 1 sample from two galls of same tree and 1 bulked sample 6. Richmond (RNP) 1 bulked sample 7. Coombs (C) 1 bulked sample 18 temperature for approximately 1 hour. Three-hundred ul of lOmg/ml ethidium bromide, 500ul 10% N-sarcosine N a and 4.0g CsCl were added and the tube inverted gently to dissolve. Refractive index was adjusted to 1.3850-1.3890. The homogenate was transferred to a 5 ml ultracentrifuge quick seal tube and centrifuged overnight at 53,000rpm. The D N A band was collected under U V light and ethidium bromide was removed by extraction with C s C l saturated isopropanal. Dialysis against I X T E ( 50 m M Tris, 50 m M E D T A , pH7.8 ) buffer for 24 hours was performed. After dialysis, aliquots of genomic D N A were run on 0.6% agarose gels and D N A concentration quantified by comparing staining intensity with lambda D N A molecular weight marker (digested with Hind l l l ) (Boehringer Mannheim Biochemica). DNA digestion and hybridization D N A samples were digested with 12 restriction enzymes (Ava i l , B g l II, Bam HI, Eco RI, Hae III, Hind III, Hin f I, Nde I, PvuII, R S V I, Sma I and Xho II) following instructions from the manufacturer. 250ng D N A was used for each digestion. Southern blottings were done by capillary method from gel to Hybond-N membrane, performed overnight (Maniatis et al., 1982) and hybridized with six heterologous probes (provided by Dr. E . E . White) p A m l , pAm2, histone, actin, B A 4 and H C 5 . Labeling of D N A probes with digoxigenin-dUTP (Boehringer Mannheim Biochemica Cat. No . 1277065), Southern hybridization (high stringency), Lumi-phos 530 (Boehringer Mannheim Biochemica Cat. No. 1277470), and A M P P D (Boehringer Mannheim Biochemica Cat. No . 1357328) visualization by non-radioautographic detection were done according to supplier's instructions. For detailed procedures for probe labeling, Southern hybridization, large scale plasmid isolation, recovering probe-DNA from gel, antibody binding and non-radioautography detection refer to the Appendix I. 19 Screening for polymorphic markers A single sample from each of the seven locations, Nation River (NR), Vanderhoof (VH), Prince George (PG), Umiti Pit Road near Quesnel (QS), Manning Park (MP), Richmond (RNP), and Coombs (C) were screened by 12 enzymes and 6 heterologous probes for polymorphic RFLP markers. The same sample set was screened by both single and double combination digestions of Ava II, Hind III, Pvu II and Bgl II with probes pAml and pAm2. Population variation assay Five populations (VH, 5 samples; NR, 4 samples and one replicate (NR3); QS, 4; PG, 5; and MP, 6 samples) were screened using pAm2/AvaII marker to test for variability within and among populations. Hybridization and detection methods were as above. %. Data analysis Specific DNA restriction fragments were identified by Avail digestion and probe pAm2 hybridization. The molecular weight of bands was determined by plotting the relative migration of bands with standard molecular marker on Log ratio paper. The value (molecular weight) of each RFLP band was determined by regression analysis. Molecular weights were measured by two persons independently. For the population structure study, bands were tabulated as present (1) or absent (0) for each individual to generate a data matrix. Cluster analysis was employed using Numerical Taxonomy and Multivariate analysis system (NTSYS-PC, version 1.80, Rohlf, 1993). To generate a dendrogram, simple matching coefficient and unweighted pair group method, arithmetic average (UPGMA) (Sneath and Sokal, 1973), S = m/n, where m= number of matches, n= total sample size (number of matches and unmatches) were employed. 20 r RESULTS DNA extraction DNA extracted by White's method was intact and cutable with restriction enzymes. There was no correlation between the amount of spores used and DNA extracted, beyond the range of 50 to 100 mg spores. Spore samples less than 30 mg did not yield enough DNA to form a band after ultracentrifugation visible under UV light, probably because of spores being blown away by liquid nitrogen when spores were poured into the extraction device. Using more than 100 mg of spores did not yield more DNA since the space in the container was limited and grinding was poor. Screening of enzyme and probe combinations Of the 6 heterologous probes tested, only two, pAml and pAm2 hybridized to western gall rust DNA. Among 12 tested restriction enzymes, six, namely Ava II, Bam HI, Bgl II, Hind III, Pvu II and Xho II showed polymorphisms with pAm2. Three enzymes Bam HI, Bgl II and Hind III showed polymorphisms with pAml (Table 1.). 0.5-6 kb length fragments were revealed by these combinations. Rsa I, Sma I and Eco Rl generated monomorphic fragments when probed both with pAml and pAm2. The other probes tested failed to hybridize with DNA of E. harknessii (Table 1). Size of ribosomal DNA The total length of an Endocronartium rDNA repeat unit was measured as 8.6 kb by six single and double enzyme combinations probed with pAml and pAm2. These two probes covered the entire repeat of Armillaria mellea rDNA of 10.2 kb (Anderson et al., 1989). 21 Table 1. DNA fragments (kb) generated by 12 restriction enzymes (REs) and detected with 6 heterologous probes3 REs pAml pAm2 Histon Actin BA4 HC5 Avail 3.0 2.2,3.0* . . . . Bam HI 0.8,0.5* 2.0,0.7* . . . . Bgl II 9.0,6.0 * 9.0,6.0,2.8* . . . . Hind III 0.8,0.5* 2.8,1.0,0.8* . . . . Pvu II 0.8 0.9,0.8* . . . . Rsal 8.5 8.5 . . . . Smal 6.5 6.5 -Xho II 0.4 0.8,0.7,0.4* - -EcoRI 8.4 8.4 . . . . Hinfl - -Ndel - -Hae III -a All probes tested here were given by Dr. E.E. White. Polymorphic fragments 22 Polymorphic markers Table 2 represents the results of screening of polymorphic markers. Of seven combinations of four enzymes (PvuII/Avail, PvuII/Hindlll, Avall/Hindlll, Ava II/Bgl II, Bgl II/Hind III, Pvu II and Ava II) probed with pAml and pAm2, pAm2 produced more polymorphic bands. Except for PvuII, all other enzymes were monomorphic with pAml (Table 2). Figure 2 shows the polymorphic bands across the seven samples from seven locations digested with Ava II and probed with pAm2. Variation within and among populations Using the selected Avail / pAm2 RFLP marker, preliminary variation assay in the ribosomal DNA region among 24 single gall samples from 5 sites (Figure 3) was conducted. Variation within local populations and among populations in terms of length of fragments were converted to a binary data matrix based on this one enzyme and probe combination (Appendix II). All samples from the 5 locations shared the fragments of 0.9 kb and 1.2 kb. The polymorphic fragments varied from 2.8 kb to 3.6 kb within and among locations (Fig. 2 and Fig. 3). Figure 4 presents the dendrogram generated by NTSYS-PC using SEVIQUAL and SHAN clustering by the UPGMA method. Figure 4 was derived from the information in Figure 3 and was based on the presence and absence of 7 fragments. The dendrogram revealed two clusters with 66% similarity further divided into 4 clusters at about 70% in which the majority of individuals belonged to the same locations. 100% similarity was observed among 3 out of 5 samples in QS, 3 of 4 in VH, 2 of 5 in PG. and 85.7% for 5 of 6 samples in the MP population. PG6, MP5 and QS3 populations contained individuals that departed the most from their geographically related group. 23 Table 2. Single and double restriction digestions used for screening polymorphisms from seven gall rust populations Enzymes \\ Probes pAm2 pAml PvuII/Avall 1.3, 1.5,2.8 0.5*, 0.9* PvuII/Hindlll 2.1,2.3,2.8 2.5* Avall/Bgll 0.9, 1.5,2.9,3.0,3.8 0.5*, 0.9* Bglll/Hindlll 1.0,2.2,2.3,2.5 2.5* Avall/Hindlll 0.5, 1.5,2.3,2.5, 2.8,3.2 — PvuII 6.5 * 6.0, 6.5 Avail 0.9*, 1.2*, 2.8, 3.2,3.4, 3.6 0.5*, 0.9* * monomorphic across samples from 7 locations Seven locations: Prince George #8, Vanderhoof # 3, Nation River #1, Quesnel (QS) #4, Manning Park #3, Coombs, and Richmond (fragments in kpb) 24 b p 6600 4400 2300 2100 500 Figure 2. Autoradiograph of Southern hybridization of D N A from seven locations screened with AvaII/pAm2. Locations are identified above each lane. Size markers are indicated in base pairs on the left. 25 o ^ O ^ o ^ & >& r%> r%> r4? c2> r^> Figure 3. R F L P autoradiographs of five populations screened vs. AvaII/pAm2. Spore samples are identified above each lane. Size markers are indicated in base pairs on the left. 26 0.00 0.25 0.50 0.75 _L00 QS1 — QS2 'QS4 — P66 PG1 NR5 NR2 PG5 UH4. -PG8 MP5 UH5 UH1 UH2 -UH3 NR1 NR3* NR4 QS3 riP4 riP6 PG3 •npi MP2 T1P3 Figure 4. Dendrogram derived by NTSYS-PC using simple matching coefficient and unweighted pair group method, arithmetic average (UPGMA). A total of 5 populations and 24 individuals were assessed by rDNA RFLP markers. Band presence or absence was scored as 1 or 0, respectively. * NR3 is a replicate of NR1 27 DISCUSSION In this study, polymorphisms in the ribosomal DNA region of western gall rust DNA were revealed among seven geographically separate locations and within locations. Western gall rust DNA samples were derived from both bulked aeciospores and single-gall aeciospores. Even with only one marker and limited samples, the results suggest that rDNA RFLP markers can be used to analyze western gall rust population structure. Ribosomal DNA was targeted for several reasons: First, it exists in multiple copies ( Rogers and Bendich 1987), so that the signal from the hybridized probe is easily detectable with a standard hybridization method. Second, the organization of the rDNA repeat unit varies in different fungal species (all fungi contain the 18s, 5.8s and 26s rRNA genes within the repeat unit). These are always coded on one strand of the rDNA and are spatially ordered (Specht et al., 1984). Third, the intergenic spacer (IGS), which is also known as the NTS or nontranscribed spacer and portions of the external transcribed spacer (ETS) evolve rapidly (Appels and Dvorak 1982). Therefore, by using a probe that will hybridize to the gene itself, it is possible to assess not only the highly conserved regions, but the attached rapidly changing region as well. This study also revealed that the Endocronartium harknessii rDNA unit length (8.6 kb) differs from that of the closely related fungus, Cronartium ribicola, (9.6 kb) as determined by the same two probes (White et al., 1994). This indicates that rDNA RFLP markers could be a means for studying evolution of stem rusts and the origin of Endocronartium. Considerable variation in size is observed among different fungi. For instance, the smallest known rDNA unit (7.8 kb) occurs in Aspergillus nidulans and the largest (>24 kb) in A. ambisexualis (Borsuk, et al., 1982). In Armillaria mellea it is 10.2 kb (Anderson et al., 1989) and in Coprinus cinereus 9.8 kb (Wu et al., 1983). Thus, the rDNA unit lengths can be useful characteristics for phylogenetic studies. Of the six enzyme/pAm2 combinations and one enzyme/pAml combination that 28 yielded polymorphic markers, only one marker was used for the population assay because of limited DNA samples. In£\". harknessii, DNA could only be extracted from aeciospores. The size of spore collections was restricted by various conditions, such as spore production occurring only once a year, timing of collections limited by climate, topography, microclimate and so on, and especially hyperparasites which greatly reduced the opportunity for collection of large amounts of clean spores (Byler and Cobb, 1969). Older galls bore more spores, but were not clean. Young galls usually yielded 10 to 30 mg of spores. The CsCl DNA extraction method gave 2-3 ug pure genomic DNA from 100 mg of spores. In this study, most samples consisted of only about 50 mg of spores, which yielded about 1 ug pure DNA. After testing all the conditions and conducting the screening for polymorphic markers, only barely enough DNA was left for the population assay. Furthermore, the procedures used here were rather time consuming for large scale population assays. Variations occurred both within and among populations. Individual populations could not be unambiguously defined on the basis of available evidence, but some populations (e.g. QS, VH, MP) were clustered in the dendrogram ( Fig. 4). Furthermore, the single gall (MP5) that did not fit with the remaining MP galls was collected several kilometers distant and at an elevation about 200 m higher than the remaining MP galls. Saghai Maroof et al. (1990) studied the genetic diversity and ecogeographical differentiation among rDNA alleles in wild and cultivated barley and concluded that natural selection plays a major role in genetic organization of rDNA variability. Although only one marker and limited samples were employed in this study, which cannot provide strong evidence for suggesting that natural selection is one of the forces accounting for rDNA variation, the trend of the results certainly raised the question. It is noteworthy that the PG population, which unlike the other populations consisted of bulked samples of galls, each derived by inoculation from single-gall spore sources collected near Prince George, was the most variable of all the five populations. 29 Overall, this study showed that r D N A R F L P is a feasible method to detect population variation of western gall rust, and that such variation exists both within and among western gall rust populations. 30 CHAPTER III DETECTION OF GENETIC VARIATION IN WESTERN GALL RUST POPULATIONS IN BRITISH COLUMBIA, USING RAPD MARKERS 31 INTRODUCTION Western gall rust (Endocronartium harknessii (Moore) Hiratsuka ) is an indigenous disease of hard pines in North America. It is especially damaging in plantations. Understanding the genetic makeup of pathogen populations will facilitate the selection of resistant stocks. To understand an organism's genetic structure, its reproductive system must be known. However, there continues to be a dispute about whether western gall rust reproduces sexually (Hiratsuka et al., 1966; Epstein and Buurlage, 1988; Hiratsuka, 1991). Conventional biological methods can not solve this dispute easily. Biochemical and molecular techniques are powerful tools for the study of fungal population genetics. The patterns of population structure produced by vegetative and sexual reproduction can be distinguished by both isozyme and RFLP markers (Burdon and Roelfs,1985; McDonald and Martinez, 1990; Brown et al., 1990). Thus one can deduce the mode of the reproductive system from population structure. Earlier isozyme marker studies of western gall rust isolates (Tuskan et al., 1990; Vogler et al., 1991) showed that populations of western gall rust are not panmictic, implying that western gall rust reproduces vegetatively. Isozyme markers provide a convenient method for the detection of genetic changes but are limited in number and only DNA regions coding for proteins can be sampled. RFLP applications in fungi have increased greatly in recent years (Leong and Holden, 1989; Michelmore and Hulberts, 1987; Michelmore et al., 1991). Restriction-site diversity is useful for inferring population genetic structure (McDonald and Martinez, 1990). However, for population studies, this method is tedious, costly and requires large amounts of tissue. A method described by Williams et al. (1990) for identification of polymorphisms based on PCR (Polymerase Chain Reaction) has provided new opportunities for the analysis of nucleotide sequence variability. This DNA marker system named RAPD (Random Amplified Polymorphic DNA) is based on the amplification of unknown DNA sequences using single, short, random sequence oligonucleotide primers 32 oligonucleotide primers (10-12 bp). The RAPD marker system overcomes many technical limitations of RFLPs and has been used in population structure analysis for various organisms such as plants (Chalmers et al., 1992; Huff et al., 1993; Mosseler et al., 1992), insects (Kambhampati et al., 1992 ) and plant pathogens (Chen et al., 1993; Doudrick et al., 1993; Smith et al., 1992, 1994). RAPD polymorphisms may arise in two ways. Length polymorphisms are the result of insertion or deletion from the amplified fragment between the primer-binding sites. Presence/absence polymorphisms are caused by a change in at least one base pair at either of the primer-binding sites. Many areas in the genome contain a nonbinding sequence that would, with a little alteration, serve as a primer-binding site. Mutation of, insertion, or deletion from this may convert the sequence into a valid primer-binding site. In this case, a new band could appear. Conversely, changes of sufficient magnitude to a functioning primer-binding site would lead to the disappearance of a band, because the fragment that has that site as one of its ends would no longer be amplified exponentially. In Chapter II, it was demonstrated that variation within and among populations can be detected by rDNA markers. That study also showed that the large amount of spores required for DNA extraction is a problem for large scale sample assays. The alternative is to use the simpler RAPD (Random Amplified Polymorphic DNA) method. The purposes of this study were to assess western gall rust population genetic variation and to gain insight into the genetic structure of the fungus using RAPD markers. 33 MATERIALS AND METHODS Development of RAPD protocols 1. DNA isolation method: Table 3 describes the spore samples which were viability tested by germination prior to use for DNA extraction tests. A phenol chloroform DNA extraction method was adapted and modified from Elder et al. (1983), and Rogers et al. (1989) as follows: 10-30 mg of spores were placed in a small mortar with an equal volume of dry ice to freeze the spores which were ground three times for 10 seconds, adding more dry ice each time. The resulting macerate was then washed gently in 400 ul fresh lysis buffer ( 0.5M NaCl, 0.2M Tris (pH 8.0), lOOmM EDTA (pH 8.0), 2%SDS, 1% 2- mercaptoethanol) avoiding formation of bubbles. The homogenate was transferred to a 1.5 ml sterile microcentrifuge tube and the mortar washed once with a further 400 ul lysis buffer which was then transferred to the same microcentrifuge tube. After keeping the homogenate at room temperature for 10 to 30 minutes , 400ul Tris-saturated phenol: chloroform: isoamyl alcohol (PCI) (25:24:1) was added and gently mixed until a homogeneous emulsion had formed. The microcentrifuge tube was then centrifuged at 10,000 rpm for 3 minutes at room temperature and the upper aqueous layer collected. The PCI extraction was then repeated once. An equal volume of ethyl ether was then added, mixed and the resulting emulsion centrifuged at 13,000 rpm at room temperature for 3 minutes, and the upper layer discarded. Then 5 ul RNase A (10 mg/ml) was added and incubated for 30 minutes at 37 °C. Two and a half volumes of cold 100% ethanol were added and the tubes were placed at -20 °C for 30 minutes to overnight. The tubes were then centrifuged at 13,000 rpm at 4 °C to precipitate DNA. The pellet was washed with 70% ethanol, recovered by a quick centrifugation at room temperature , and air dried for 10 - 15 minutes. 34 GO c o o cd s o a a S u o o ON A ON O A o x O ON A o x O oo o N O 00 o D. CO 0) Q i— <2 cu ' GO 4) o U es Q ON ON ON ON ON 00 ON ro ON ON ON ON ON ON oo ON cd C cd CO CD O Ui 3 O GO CD O CV, oo to 3 cs H o w a \"•3 cs w o cd o c o o GO 3 C > 3 O o c cd > cd t O c o o cd AH oo c '2 e cd cd O C o o 4> o CU O O C ' C 3 cd o OH cd PH on c - c c cd cd O c o o 4) \"N c o cd cd o c o o 2 £ o o cd O c o o cd 3 a. GO 3 OS to to cu to ft) % -ft o to ^3 1^ o o ft) I ft o ft) «3 ft o faq t»q 99% (Figure 9) across all the individuals . The percentage of each frequency range is presented in Figure 9 and detailed information for each primer band was listed in Table 7 in Appendix III. Each gall had a unique phenotype based on the 96 polymorphic bands. 44 M 1 2 3 4 5 6 7 8 + -M1 2 3 4 5 6 7 8 +-Figure 7. Pure Penicillium, Endocronartium and mixtures amplified with four primers. Arrows show the disappearing bands amplified from Penicillium D N A . Lane 1 represents pure Penicillium. Lane 2, 3, 4, 5, 6 and 7 represent mixtures of Penicillium and Endocronartium as 50%, 25%, 12.5%, 6.25%, 3.125% and 1.5% (v/v). Lane 8, pure E. harknessii. Each primer is identified above the set of lanes. D N A size marker (M) is a 100 bp ladder (Boehringer Mannheim Canada, Ltd). 45 4 6 4 m m-418 -420 -437 - 438 m m- 418 -420 -437 -438 m m- 437 - 438 - 418 m m BM Taq | PE Taq . — : — 46 Figure 8. RAPD reproducible banding patters under various conditions. 8a. Effect of template DNA concentration: DNA concentrations are identified on the bottom of the figure as A, B and C representing the amount DNA of 0.63ng/ul, 0.365ng/ul, and 0.16ng/ul, respectively. DNA samples of Q8, Q9, WV5-1 and WV5-2 are identified as lanes 1, 2, 3, and 4, respectively. Primers are identified above the lanes. DNA size marker (m) on the extreme left lane is 100 bp ladder (Pharmacia). 8b and 8c. Effect of thermocycler model: RAPD reaction with different machines(8b, 9600 thermal cycler and 8c, 480 thermal cycler) and on the different dates (8b, Dec. 17, 1993 and 8c, Oct. 30, 1993 ) with the same DNA samples which are MP gall sectors. -, represents the negative control (no DNA template). Each primer is identified above the lanes. DNA size marker (m) is 100 bp ladder (Boehringer Mannheim Canada, Ltd) 8d and 8e. Effect of amplification with or without a hot start: Samples arrangement on the gel are as negative control, WV5, WV6, WV7 and WV8 and a space lane between each primer. Primers are identified above the lanes. DNA size marker (m) on the extreme left lane is 100 bp ladder (Pharmacia). 8f. Effect of repeated DNA extractions from a single spore sample: DNA was extracted on March, 1993 from RNP populations and amplified with primer 431. Individual galls are identified by number above each lane. 8g, DNA was extracted on May, 1995, spores from the same RNP 1993 collections and amplified with primer 431. Gall identify number appear above the lanes. DNA size marker (m) is 100 bp ladder (Boehringer Mannheim Canada, Ltd) 8h. Effect of source of Taq enzyme: DNA amplified with Taq polymerase from BM (Boehringer Mannheim Canada), and PE (Perkin Elmer) and primer 406. DNA samples are 9 individuals from MPL population. DNA size marker (m) is 100 bp ladder (Boehringer Mannheim Canada, Ltd) 46a Table 4. Sequences of the selected primers and number of bands amplified for the population assay. Pr imers Sequences #of bands #of polymorphic bands 406 5 ' - G C C A C C T C C T - 3 ' 9 5 418 5 ' - G A G G A A G C T T - 3 ' 8 6 431 5 ' - C T G C G G G T C A - 3 * 11 9 437 5 ' - A G T C C G C T G C - 3 ' 9 9 438 5 ' - A G A C G G C C G G - 3 ' 11 11 467 5 ' - A G C A C G G G C A - 3 ' 10 7 482 5 ' - C T A T A G G C C G - 3 ' 12 9 485 5 ' - A G A A T A G G G C - 3 ' 7 6 489 5 ' - C G C A C G C A C A - 3 ' 6 3 490 5 ' - A G T C G A C C T T - 3 ' 9 8 775 5 ' - G G T T T G G T G G - 3 ' 7 6 777 5 ' - G G A G A G G A G A - 3 ' 10 7 798 5' - G A G A G G A A G G - 3 ' 10 10 Total 119 96 420* 5' - G C A G G G T T C G - 3 ' 6 5 427* 5' - G T A A T C G A C G - 3 ' 4 3 432* 5 ' - A G C G T C G A C T - 3 ' 7 2 464* 5 ' - C A C A A G C C T G - 3 ' 6 1 480* 5' - G G A G G G G G G A - 3 ' 7 4 487* 5 ' - G T G G C T A G G T - 3 ' 10 5 796* b 5' - A G A G G G A G G A - 3 ' 8 2 * Primers were used in other tests. Number of polymorphic fragments were counted based on the primer screening using 6 D N A samples from Manning Park location, b Primer was screened using 8 D N A samples from 8 locations in B C . 47 O O O CVJ N ° V ? X> O * \" tf^ o o o o o o o o o o cd cd •sr c\\i o O O C O O O C O ^ o x o x O O O O O O o V) o o 0) o c 0) 1 CT) 3 C O C O c r o o CD k_ LL 1 CT> C M C M O o 1 CT) 1 1— o o , CT) o o o o s p u B q ojqdeoujA|od j o j equun iv j 48 49 Figure 10. Examples of RAPD banding patterns from E. harknessii DNA. Population abbreviations: C = Coombs, VH = Vanderhoof, LB = Long Beach, PGM = Mackenzie, RNP= Richmond Nature Park, TML = Ten Miles Lake. DNA size marker is a 100 bp ladder ( Boehringer Mannheim Canada, Ltd.). 10-1, Example of a primer (482) yielding polymorphisms within and among populations. 10-2, Example of a primer (775) that distinguishes among populations (arrow). 10-3, Example of a primer (490) that distinguishes geographic populations. The first 6 lanes on this gel excluding the DNA standard size (m) marker, are samples from the Transition zone; the folowing 9 samples are from the Coastal region and the remaining 7 are from the Interior region. Arrows indicate the markers that discriminate between the regions. 49a1 Variation within and among populations The complete data matrix is listed in Table 8 in Appendix IV. It consists of 96 bands and 120 individual galls (10 galls from each of 12 locations.) Principal Component Analysis (PCA) revealed that there was no significant correlation between the variables in the data collected on populations (constants excluded). This means that all variables made similar contributions to the analysis. For instance, the 1st principal component only accounts for 12% of the total variance of all the variables. The results of 35 principal components account for 88% of the total. Two scatter plots of the first and second principal components are shown in Figure 11, with different symbols designating populations (Figure 11a) and regions (Figure lib). These figures show that although a small number of principal components could not take the place of the original variables due to weak correlation, nevertheless, the first two principal components separated individual galls into clear regional clusters, and within regions into fairly clear local population clusters. These trends were confirmed by Cluster analysis. By using similarity analysis on qualitative data and a simple matching coefficient, UPGMA cluster method was performed. The similarity levels for all individual galls ranged from 0.699 to 0.979 and average similarities within populations from 0.792 to 0.873. A dendrogram with 120 individuals was obtained (Figure 12). Different coefficients and clustering methods have been explored. However, according to the cophenetic values (test of goodness of fit of dendrograms), only one tree (UPGMA) was chosen (r =0.7546). There are two major clusters formed at a level of 75% similarity among the RAPD bands. Three populations, namely MPM, PGM and 100 MH formed one cluster. The remaining nine populations divided again into three major clusters at a level of 79% band sharing. The coastal populations Coombs, RNP and WV appear to form one cluster with 85% band sharing centered on the MPL, transition zone population. However, the 50 P R I N 2 2.0 a a r r - J . S -2 .0 NOTE: 18 o b s h a d m U s l n g v e l u o a . 0.0 P R I N 1 PRINCIPAL COMPONENTS ANALYSIS PLOT OF PRIN2'PRIN1 Symbol is value of western gall rust populations Figure 11. A Principal Components analysis based on co variance matrix, 1st and 2nd principal components plotted. There are 18 observations hidden. a. PC A plot for 120 individuals, symbol designation refers to the population, as follows: a, Coombs; b, Long Beach; h, 100 Mile House; 1, Manning Park Lake; m, Manning park Mountain; p, Mackenzie; q, Quesnel; r, Richmond of Vancouver, t, Ten Miles lake; v, Vanderhoof; w, West Vancouver; W, Williams Lake 51 PRIN2 2.0 A • • • • •# • # • • w 0 © • • © . I I 1 1 JPr* i 0 © i I I I I I I I I -2 .0 -1.5 -1.0 -0.5 0.0 0.5 1.0 1.5 2.0 P f l l N I NOTE: 18 oos b&a c i s s i n a values. PRINCIPAL COMPONENTS ANALYSIS PLOT OF PRIN2*PRIN1 Symbol is value of western gall rust regions Figure lib. PCA plot for 12 populations, symbol designation refers to the regions, as follows: #>, Coastal,©, Transition Zone, lower elevation,^ Transition Zone, higher elevation, I, Interior 52 TML population joined the coastal cluster. The Quesnel populations formed a single cluster together with a few coastal individuals. Williams Lake and Vanderhoof populations stayed in one cluster at the 85% similarity level. Within populations, most individuals exhibited an 89-95% similarity level. However, a few individuals were scattered among the second level clusters (Figure 12). In general, within the 4 coastal populations, individuals formed more discreet clusters than did individuals in the other two regions. For instance in the MPM, 100MH, PGM, and Q populations all 10 individuals fell into the same cluster. For the MPL, TML and VH populations, 7-9 galls fell into the same cluster, with only 1 or 3 galls left to an outside group. But, the 4 coastal populations of C, LB, WV and RNP formed more than three clusters (with 10 individuals). The methods of NTSYS similarity for genetic data (SIMGEND) program, Nei's coefficient (Nei, 1972) and UPGMA clustering were chosen to obtain dendrograms for the populations. The resulting population tree (Figure 13) resembles the individuals (Figure 12) dendrogram on the major branches. The maximum accumulated number of codon differences (distance) (Nei, 1972) is about 17.7%, at which point two clusters formed, one consisting of MPM, PGM and 100MH, and the other containing the rest of the populations. The second level of clusters formed at a maximum accumulated number of codon difference of 13.8%. The MPL population clustered with all four coastal populations (9.6%). The other four interior populations formed one cluster at 12.5% difference level (Figure 13). The cophenetic value of this tree is r = 0.7462, which shows a fair degree of fit. Furthermore, the two geographic locations are distinct from each other, as well as having correlation with biogeoclimatic zones (Fig. 13). Genetic variability within individual galls Many studies have been based on the assumption that individual galls arise from infection by a single spore and that, in the absence of a pycnial stage, all the aecia on a gall must therefore be genetically uniform. RAPD markers were used to test uniformity at the DNA level of spores isolated from different sections of three individual large galls. 53 o •a 8 CQ o •a «o C^ oo c o N i — i in ^ n ^ in ^ ro O O O O H s 03 i— CO o T 3 c co -o c o 3 o. o a. • a o E 60 C co I c O o » \"3 8 .S3 13 o •Z3 O g ' N CN I * C D C X I I C E O m z D a X I 2 Z C X CJD C D C J i c t 3 r : r > n i — a e x O D C D C D C*N CZ) NO •m—t. m o J3 o - a c CO o c 03 CO c 00 CO 00 \" G O P c o 3 a, o a. CN o E o3 l — 0 0 o I— c CO Q 0) 3 oo 03 3 T3 - a c c o 3 a. o CX t_ CO a. -4—* C n> oo X) o3 00 -TD C 03 X) 100 galls per m of the shoot length, which is about 100 times as high as the infection rate observed on heavily infected trees in the field during years of heavy infection. Random location of infection points does occasionally lead to two galls being located so close together that they eventually fuse. Such multiple-origin galls usually have irregular or pyriform shapes, although it is possible that on very rare occasions the 63 infection points are so close together that a regular-shaped gall results. However, unless one is willing to pose that successful penetration of the host by a germ tube greatly increases the susceptibility of adjacent entry courts, so that many (most) galls are of multiple spore origin (no evidence for this exists), the assumption that regularly shaped, widely separated galls almost always arise from infection by a single spore, must stand. Galls arising from artificial inoculation with masses of aeciospores will of course be suspect in this regard. The observed variation in RAPD profiles among sectors within the three galls tested may have arisen in ways other than mutation. First, although spermogonia are very rare in this rust, it is theoretically possible that they were in fact produced on these galls some years ago, allowing the introduction of foreign DNA. (Spermogonia are ephemeral, but their old, dead remains are visible for about a year.) Second, the galls used were of necessity (i.e. to obtain enough aeciospores) large, old galls. Such galls have a well-developed rhitidome occupied by a large population of micro-organisms. Hence the possibility that contamination of aeciospore samples accounts for (some of) the variation in RAPD profiles cannot be wholly discounted. If either or both of these two mechanisms are in fact responsible, then the implications for the population study are,less disturbing. In that study, the galls were young, producing spores for the first or second time. Hence spermogonia, if present, would likely have been detected, and spore contamination is unlikely to have been severe enough to lead to spurious RAPD profiles. Where did variations arise in western gall rust populations ? What is a population? In this study, collections of isolates from different locations were treated as separate populations. The distance between most populations was about 100 km, though some were much closer e.g. RNP and WV. In the case of MPM and MPL, the vertical separation was about 300 meters. The dispersal ability of an organism plays an important role in the maintenance of genetically distinct geographic populations. 64 Organisms with potential for air-borne, long-distance dispersal will display greater genetic uniformity across local populations than organisms with very limited dispersal ability. Results of this study indicated that gene flow between some geographically distant populations may have united them into homogenous genetic groups that are evolving together. This appears to be the case for Vanderhoof, Williams Lake, and Ten Mile Lake populations, as well as PGM, 100MH and MPM populations described in Figure 5 and 12. High similarity (low genetic distances) between these populations separated by more than 500 km suggests either that extensive gene flow has blended the populations with each other, or that these populations share a common recent ancestor. However, one exception is more difficult to interpret. For instance, the Quesnel population situated geographicallly between TML and WL, is placed in the dendograms between coastal populations (Fig. 12), but alongside neighbouring populations in Fig 5, and 13. Several mechanisms could facilitate gene flow between populations of E. harknessii. The most obvious one is air dispersal of the aeciospores. Carlson (1969), Peterson (1973) and Blenis at al (1993) demonstrated that aeciospores have the potential to move at least several hundred meters, which would be adequate to move them between adjacent fields. Thus, wind-mediated movement of aeciospores is probably feasible over distances of at least tens of kilometers. Related species C. ribicola, aeciospores are known to have moved more than 100 miles to isolated pine populations (pers. comm., Dr. Bart van der Kamp). Other mechanisms may also facilitate gene flow, e.g. man made collections, planting infected seedlings out of range and so on, although these are unlikely to have had a significant impact to date. Adaptation to local conditions must be happening in the different climates of the coast and the interior, on the different host genotypes. Lodgepole pine is a native tree and adapted to local environments, showing extreme genetic variation in nature (Hamrick et al., 1981; Yeh et al., 1985). Western gall rust is an indigenous disease and infects the native lodgepole pine. The Manning Park populations are an interesting case. MPM 65 native lodgepole pine. The Manning Park populations are an interesting case. MPM samples collected from high elevation always tended to group together with interior populations and MPL from lower elevations stayed with coastal populations by both cluster and PCA analyses (Figs, 1 la, 1 lb) Reproductive system vs. population structure The contribution of sexual and clonal reproduction to population structure has been studied for several fungi (Brown et al., 1990; Burdon and Roelfs, 1985). Except for some fungi with prominent sexual cycles, direct observation of genetic recombination in the field is difficult for many plant pathogens. Analysis of gene or genotype frequencies offers an indirect means to assess the amount of genetic recombination in a population usually with known reproductive system of the fungus. For instance, Puccinia graminis f. sp. tritici requires two different hosts to complete its sexual cycle. The asexual cycle on wheat can be perpetuated indefinitely by urediniospores. For the sexual cycle to take place, barberry is needed as alternate host for infection by basidiospores. In the United States, the P. graminis f. sp. tritici population east of Rocky Mountains is considered asexual due. to the eradication of barberry in the 1920's. The fungal population west of the Rocky Mountains is considered capable of sexual recombination due to the presence of barberry, although the sexual stage is rarely detected on barberry. Burdon and Roelfs (1985) using isozyme markers compared the genetic structure of sexually and asexually reproducing populations of this wheat stem rust pathogen occurring in the United States in 1972, 1975 and 1976. The structure of sexual population was determined directly by electrophoretic analysis of individual isolate collections. And the structure of the asexual population was estimated from a known association of particular isozyme and virulence phenotypes. They found that in all three years, the sexual population always was more diverse than the asexual one. Great genotypic diversity was found in Cryphonectria parasitica (causal agent of 6 6 chestnut blight) within a 25 x 25 m plot in a mixed-oak hardwood forest (Milgroom et al., 1992). For this disease it is uncertain whether reproduction is predominantly sexual or asexual. Thirty-three out of 39 isolates were found to have unique DNA fingerprints which suggested a sexually reproduced pattern. Genotypic diversity analysis of the unique haploitypes showed that the observed genotypic frequencies were consistent with random mating. Thus, the majority of the population was considered to be derived directly from sexual reproduction. The dispute about the reproductive system of this endocyclic rust, Endocronartium, has continued for decades. In this study, it was anticipated that it would be possible to deduce the reproductive system of western gall rust from the population structure since population structure can be strongly influenced by the occurrence (or a lack of occurrence) of genetic recombination. However a final conclusion in this respect cannot be made due to the lack of comparison with related species with known reproductive systems, though both rDNA RFLPs and RAPD markers exhibited high variability within and among populations that is characteristic of sexual reproduction. Since population structure is often determined by the partitioning of genetic variation, inference of population structure is also highly dependent upon the tools used to estimate genetic diversity. A recent study of sea oyster (Crasotrea virginica) showed that phylogenetic trees based on RFLPs of mitochondria and nuclear DNA, were dramatically different from trees based on isozymes, leading to contrasting interpretations of the forces determining population structure (Karl and Avise, 1992). Even different types of DNA markers can yield different relationships among isolates from a single population. McDonald and Martinez (1991) found that in Septoria tritici there was a lack of correlation between the similarity relationships based on single-copy and repetitive DNA probes. C. ribicola is known to be reproduced sexually, however with isozyme markers (Berube, 1994) and RAPD markers (Hamelin et al., 1995) little variation was revealed among populations collected in Ontario. Thus, depending on the population 67 structure as revealed by a particular molecular technique to infer the reproductive system of a pathogen is not always appropriate. Therefore, it appears that there is little basis for a direct comparison of diversity estimates except for those derived with the same genetic tools for populations of the same species. In general, we gain more insight with techniques that have higher discriminating power (such as RFLP vs. isozyme). Ideally, the DNA markers used should be randomly distributed in the genome. The population variation results revealed in this study are dramatically different from the results obtained by isozyme markers (Vogler et al., 1991; Tuskan et al., 1990) which further indicates that different markers may lead to different conclusions. The importance of western gall rust population structure to disease management strategies will depend on how variable the rust populations are. Population structure is a major determinant of breeding strategy for disease management. This study shows that western gall rust populations are highly variable in their genetic make-up. If one assumes that RAPD variability extends to traits responsible for virulence, breeding for long-term resistance may prove elusive. The two major clusters of rust populations detected agree with van der Kamp's (1988a) pathogenicity study and confirms that distinct coastal and interior races exist. 68 CHAPTER IV GENERAL CONCLUSION 69 General Conclusions Populations of western gall rust in British Columbia exhibit considerable variability at the DNA level as shown by both RAPD analysis and by rDNA RFLP's. For instance, each of the 120 collections characterized by RAPD analysis in the main study had a unique RAPD profile. As a rough estimate, about 53.6% of 96 bands amplified by randomly selected primers within populations were polymorphic, and such polymorphic bands appeared at frequencies more or less evenly distributed from 0.01 to 0.99. Similar levels of variability were present in the small sample of gall rust analyzed by RFLP profiles. These results stand in sharp contrast to previous studies that showed low variability of the rust in the USA using isozymes. The RAPD technique proved to be much more suitable for population studies of this rust than RFLP analysis. First of all, RAPD's require much less DNA per assay. The total amount of DNA extractable from the spores of a single gall was sufficient to conduct 100 to 300 RAPD analyses, but only a few RFLP analyses. Secondly, RPLP analysis was restricted by the availability of suitable probes, although additional probes could presumably have been developed. Lastly, RAPD analysis provided a great deal more information per unit of effort and funds expended than RFLP analysis. The main difficulty encountered with the RAPD technique was the common appearance of faint bands. Such bands presumably arise from less than perfect matching of primers to DNA. Strictly speaking, therefore, variation in band intensity between individuals probably indicates DNA variability. However, in this study, bands were scored as either present or 70 absent, overlooking that evidence of quantitative variation in DNA. In almost all instances, the scoring could be done without much difficulty. Occasionally, however, bands varied gradually from clearly present to clearly absent, so that for some individuals an arbitrary decision had to be made. The DNA extraction process involved routine RNase digestion. In the few instances that this step was omitted, one or two dsRNA bands were found. Similar bands were present in related Cronartium rusts. It may be that such dsRNA's are common. One possible interpretation of dsRNA's is that they represent fungal viruses. If so, such viruses do not appear to affect the pathogenicity of these rusts. However, further, study of this phenomenon is warranted, particularly to determine whether dsRNA-free isolates occur, and whether such isolates behave differently with respect to their pathogenicity . The main conclusion of the RAPD population study is that coastal and interior populations of the rust have distinct RAPD profiles, and that local populations of the rust can also often be clearly distinguished from each other by their RAPD profile. Both the PCA analysis and Clustering analysis presented in Chapter III show that, with some exceptions^ regional and local populations tend to be grouped together. Furthermore, the existence of a definable local RAPD profile does indicate limited gene flow between local populations. The most startling case is that of the two Manning Park populations which were collected only a few km's apart, with a 300 m elevational difference. PCA separated these populations into two non-overlapping groups, about as different from each other as average coastal vs. interior populations. In summary, this study has revealed considerable variability at the DNA level in 71 western gall rust. It has demonstrated that RAPD's are a useful tool to elucidate certain aspects of that variability. However, while the RAPD technique can be used to 'fingerprint' species, races, pathotypes, etc. that are already defined by other methods, it cannot by itself reveal the nuclear state of E. harknessii, and that is the major limitation for employing the RAPD technique. 72 i BIBLIOGRAPHY Allen, E. A. and Hiratsuka, Y. 1985. Artificial inoculation of young seedlings of lodgepole pine with Endocronartium harknessii. Can., J. Bot. 63: 1168-1170. Allen, E. A., Blenis, P. V., and Hiratsuka, Y. 1988. Axenic culture of Endocronartium harknessii. Mycologia 80:120-123. Allen, E. A , Blenis, P. V., and Hiratsuka, Y. 1989. Early symptom development in lodgepole pine seedlings infected with Endocronartium harknessii. Can. J. Bot. 68:270-277. Allendorfi F. W. 1977. Electromorphs or alleles. Genetics 87: 821-822. Anderson, G. W. and French, D. W. 1965. Western gall rust in the Lake States. For. Sci. 11:139-141. Anderson, J.B., Petsche, D.M., Smith, M.L. 1987. Restriction fragment polymorphisms in biological species ofArmillaria mellea. Mycologia 79:69-76. Anderson, J. B. Bailey, S. S. and Pukkila, P. J.1989. Variation in ribosomal DNA among biological species of Armillaria, a genus of root-infecting fungi. Evolution 43:1652-1662. Appels, R., Dvorak, J. 1982. The wheat ribosomal DNA spacer region: Its structure and variation in populations and among species. Theor. Appl. Genet. 63:337-348. Assigbetse, K. B., Fernandez, D., Dubois, M. P. and Griger, J.-P. 1994. Differentiation of Fusarium oxysporum f. sp. vasinfectum races on cotton by random amplified polymorphic DNA (RAPD) analysis. Phytopathology 84:622-626. Avise, J.C. 1994. Molecular Markers, Natural History and Evolution. Chapman & Hall New York. London Berube, J. 1994. White pine blister rust population genetics. Report of Biotechnology Research Activities within the Canadian Forest Service. April 1994 Patawawa National Forestry Institute Blenis, P.V., Chang, K.-F., and Hiratsuka, Y. 1993. Spore dispersal gradients and disease gradients of western gall rust. Can. J. For. Res. 23:2481-2486. Bobola, M.S., Eckert, R. T., and Klein, AS. 1991. Restriction fragment variation in the nuclear ribosomal DNA repeat unit within and between Picea rubens and Picea mariana. Can. J. For. Res. 22:255-263. Borsuk, P. A , Nagiec, M. M., Stepien, P. O.and Bartnik, E. 1982. Organization of the ribosomal RNA gene cluster in Aspergillus nidulans. Gene 17:147-152. Botstein, D., White, R.I., Skolnick, M., and Davis, R.W. 1980. Construction of a genetic 73 linkage map of man using restriction fragment length polymorphisms. Am. J. Hum. Genet. 32:314-331. Brown, J. K. M., and Wolfe, M. S. 1990. Structure and evolution of a population of Erysiphe graminis f. sp. hordi. Plant Pathology. 39:376-390. Brown, J. K. M., O'dell, M., Simpson, C. G. and Wolfe, M. S.1990. The use of DNA polymorphisms to test hypotheses about a population of Erysiphe graminis f.sp.hordei. Plant Pathology. 39:391-401. Burdon, J.J. and Marshall, D.R. 1983. The use of isozymes in plant disease research. In Isozymes in Plant Genetics and Breeding. Tanksley, S.D. and Orton, T.J.(eds.) Elsevier Science Publ. Amsterdam pp.401-412. Burdon, J.J. and Roelfs, AP. 1985. The effect of sexual and asexual reproduction on the isozyme structure of populations of Puccinia graminis. Phytopathology 75: 1068-1073 Byler, J.W. and Cobb, F.W.Jr. 1969. Fungi associated with galls caused by Peridermium harknessii. Phytopathology 59:1020. Byler, J.W. and Piatt, W.D. 1971. Cone infection by Peridermium harknessii. Can. J. Bot. 50:1429-1430. Carlson, L.W. 1969. Western gall rust on jack pine nursery stock in Manitoba. Plant Dis. Rep. 53:100. Cassidy, J.R., Moore, D., Lu, B.C., and Pukkila, P.J. 1984. Unusual organization and lack of recombination in ribosomal RNA genes of Coprinus cinereus. Curr. Genet. 8:607-613. Chalmers, K.L., Waugh, R., Sprent, J.I., Simons, A.J. and Powell, W. 1992. Detection of genetic variation between and within populations of Gliricidia sepium and G. maculata using RAPD markers. Heredity 69:465-472. Chen, X., Line, R. F. and Leung, H.1993. Relationship between virulence variation and DNA polymorphism in Puccinia striiformis. Phytopathology 83: 1489-1497. Christenson, J.A. 1968. A cytological comparision of germinating aeciospores in the Cronartium coleosporioides complex. Mycologia 60: 1169-1177. Christenson, J.A. 1969. Cytology of albino Peridermium harknessii. Mycology 61:398-400. Crane, P. E., Jakubec, K M . , and Hiratsuka, Y. 1995. Production of spermogonia and aecia of Endoconartium harknessii on artificially inoculated lodgepole pine and their significance in the life cycle, pp. 101-107 in: Rusts of Pine. Proceedings of the IUFRO 74 Rusts of Pine Working party Conference, 1994 Tsukuba, Japan. Crowhurst, R. N , Hawthorne, B. T. Rikkerink, E. H. A. and Templeton, M. D. 1991. Differentiation of Fusarium solani f. sp. cucurbitae races 1 and races 2 by random amplification of polymorphic DNA. Curr. Genet. 20:391-396. Doudrick, R.L., Nance, W.L., Nelson, CD., Snow, GA. and Hamelin, R.C. 1993. Detection of DNA polymorphisms in a single urediniospore-derived culture of Cronartium quercuum f. sp. fusiforme. Phytopathology 83: 388-392. Elder, R. T.,Loh, E. Y .and Davis, R. W. 1983. RNA from the yeast transposable element Tyl has both ends in the direct repeat, a structure similar to retrovirus RNA. Proc. Natl. Acad. Sci. 80:2432-2436. Ellsworth, D. L., Rittenhouse, K. D., and Honeycutt, R. L. 1993. Artifactual variation in randomly amplified polymorphic DNA banding patterns. BioTechniques 14 (2)214-217. Epstein, L., and Buurlage, M. B. 1988. Nuclear division in germinating aeciospores and its taxonomic significance for the western gall rust fungus, Peridermium harknessii. Mycologia 80: 235-240. Ghabrial, S.A. 1980. Effects of fungal viruses on their hosts. Annu. Rev. Phytopathol. 18:441-461. Goodwin,P. H. and Annis, S.L. 1991. Rapid identification of genetic variation and pathotype of Leptosphaeria maculans by random amplified polymorphic DNA assay. Appl. Environ. Microbial. 57:2482-2486. Guthrie, P. A. I., Magill, C. W.,Fredericksen, R. A. and Odvody, G. N.1992. Random amplified polymorphic DNA markers: A system for identifying and differentiating isolates of Colltotrichum graminicola. Phytopathology 82:832-835. Hamelin, R. C , Bernier, L. and Quellette, G. B. 1993. Identification of Gremmeniella abietina races with Random Amplified Polymorphic DNA markers. Applied and Environmental Microbiology 59:1752-1755. Hamelin, R. C. Beaulieu, J. and Plourde, A. 1995. Genetic diversity in populations of Cronartium ribicola in plantations and natural stands of Pinus strobus. Submitted to Theoretical and Applies Genetics Hamrick, J. L., Mitton, J. B., and Linhart, Y. B. 1981. Levels of genetic variation in trees; influence of life history characteristics. In proceedings of the Symposium on Isozymes of North American Forest Trees. USDA For. Serv. Gen. Tech. Rep PSW48. pp35-41. Hartung, J. S. and Civerolo, E. L. 1989. Restriction fragment length polymorphisms 75 distinguish Xanthomonas campestris strains isolated from Florida citrus. Phytopathology 79:793-799. Hiratsuka, Y., Morf, W , and Powell, J.J. 1966. Cytology of the aeciospores and aeciospore germ tubes of Peridermium harknessii and P. stalactiforme of the Cronartium coleosporioides complex. Can. J. Bot. 44: 1639-1643. Hirastuka, Y. 1969. Endocronartium, Anew genus for autoecious stem rusts. Can. J. Bot. 47:1493-1495. Hiratsuka, Y., 1991. Nuclear cycle, and nomenclature of western gall rust. In Hiratsuka, et al. (eds.) Rust of Pine For. Ca. Northwest Reg., North. For. Cent., Edmonton, Alta. Inf. Rep. NOR-X-317 pp. 92-101. Hiratsuka, Y. 1971. Spore surface morphology of pine stem rusts of Canada as observed under a scanning electron microscope. Can. J. Bot. 49: 371-372. Hopkin, A. A. and Reid, J. 1988. Cytological studies of M-haustorium of Endocronartium harknessii: morphology and ontogeny. Can. J. Bot. 66:2511-2517. Huff, D. R., Peakall, R. and Smouse, P. E. 1993. RAPD variation within and among natural population of outcrossing buffalograss (Buchloe dactyloides (Nutt.) Engelm.) Theor. Appl. Genet. 86:927-934. Innis, M.A.,Gelfond, C H . Sninsky, J. J. and White,T. J.1990. PCR protocols. A guide to methods and applications. Academic Press, New York. Kambhampati, S., Black IV, W. C. and Rai, K. S. 1992. Random amplified polymorphic DNA of Mosquito species and populations (Diptera: Culicidae): Techniques, statistical analysis, and applications. J. Med. Entomol. 29(6) 939- 945. Karl, S. and Avise, J.C. 1992. Balancing selection at allozyme loci in oysters: implications from nuclear RFLPs. Science 256:100-102. Kohli, Y., Morrall, R.A.A., Anderson, J.B. and Kohn, L.M. 1992. Local and trans-Canadian clonal distribution of Sclerotinia sclerotiorum on canola. Phytophathology 82: 875-880. Kojwang, H. O., and B.J. van der Kamp 1991. Infection of Lodgepole pine clones by single gall isolates of western gall rust. In Hiratsuka, et al. (eds.) Rust of Pine For. Ca. Northwest Reg., North. For. Cent., Edmonton, Alta. Inf. Rep. Nor-X-317 pp. 194-199. Kojwang, HO. 1989. Western gall rust on lodgepole pine in British Columbia: A study of variation and inheritance of resistance in a natural pathosystem. Ph.D.dissertation University British Columbia 161pp 76 Laundon, G.F. 1976. Peridermium. Taxon 25:186-187. Leong, S. A. and Holden, D.W. 1989. Molecular genetic approaches to the study of fungal pathogenesis. Annu. Rev. Phytopathol. 27: 463-481. Levy, M., Romao, J., Marchtti, M.A., and Hamer, J.E. 1991. DNA fingerprinting with a dispersed repeated sequence resolves pathotype diversity in the rice blast fungi. Plant Cell 3:95-102. Liu, Z. and Furnier, G.R. 1993. Comparison of allozyme, RFLP, and RAPD markers for revealing genetic variation within and between trembling aspen and bigtooth aspen. Theor. Appl. Genet. 87: 97-105. Lundquist, J.E., Walla, J.A., and Tuskan, G.A. 1994. Characteristics of Peridermium harknessii in axenic culture. Can. J. For. Res. 24:1345-1353. Lynch, M. and Milligan, B. G. 1994. Analysis of population genetic structure with RAPD markers. Mol. Ecol. 3: 91-99. Maniatis, T., Fritsch, E. F., and Sambrook, J. 1982. Molecular cloning. Cold spring Harbor Laboratory, Cold Spring Harbor, New York (A laboratory manual). Martinsson, O. 1980. Stem rusts in lodgepole pine provenance. Silvae Genet. 29:23-26. McDonald, B.A. and Martinez, J.P. 1990. DNA restriction fragment length polymorphisms among Mycospharella graminicola (Anamorph Septoria tritici) isolates collected from a single wheat field. Phytopathology. 80:1368-1373. McDonald, B.A., and Martinez, J.P. 1991. DNA fingerprinting of the plant pathogenic fungus Mycosphaerella graminicola (anamorph: Septoria tritici). Exp. Mycol. 15:146-158. Metzenberg, R. L., Stevens, J.N., Selker, E.U. and MorzyckaWroblewska, E. 1984. A method for finding the genetic map position of cloned DNA fragment. Neurospora News L. 31:35-39. Micales, J.A. 1986. The use of isozyme analysis in fungal taxonomy and genetics Mycotaxon 27: 405-449. Michelmore, R. W., Paran I., and Kesseli R. V., 1991. Identification of markers linked to disease-resistance genes by bulked segregant analysis: A rapid method to detect markers in specific genomic regions by using segregating populations. Proc. Natl. Acad. Sci. USA 88:9828-9832. Michelmore, R. W., Hulberts, W. H., 1987. Molecular markers for genetic analysis of 77 phytopathogenic fungi. Annu. Rev. Phytopathol.. 25: 383-404. Milgroom, M. G., Lipari, S. E., and Powell, W. A. 1992. DNA fingerprinting and analysis of population structure in the chestnut blight fungus, Cryphonectria parasitica. Genetics 131:297-306. Mosseler, A.,Egger, K.N. and Hughes,G.A. 1992. Low levels of genetic diversity in red pine confirmed by random amplified polymorphic DNA markers. Can. J. For. Res. 22:1332-1337. Mullis K.B. and Faloona, F.A. 1987. Specific synthesis of DNA in vitro via a polymerase-catalyzed chain reaction. Meths. Enzym. 155:335-350. Mualidharan, K. and Wakeland, E. K. 1993. Concentration of primer and template qualitatively affects products in random-amplified polymorphic DNA PCR. BioTechniques 14:362-364. Murphy, R.W., Sites, Jr. J.W., Buth, D.G., Haufler, C.H., 1991. Proteins I: Isozyme electrophoresis, in Hillis, D.M. and Moritz, C. (eds.) Molecular Systematics Sinauer Associates, Inc. Publishers pp.45-126. Nei, M. 1972. Genetic distance between populations. The American Naturalist Vol. 106 949:283-293. Peterson, G. W. 1973. Dispersal of aeciospores of Peridermium harknessii in Central Nebraska. Phytopathology, 63:170-172. Powell, J. W. and Hirastuka, Y.1973. Serious damage caused by stalactiform blister rust and western gall rust to a lodgepole pine plantation in Central Alberta. Can. Plant Dis. Surv. 53(2):67-71. Rivin, C. J., Zimmer, E. A., Cullis, C. A., Walbot, V., Huynh, T. and Davis, R. W. 1983. Evaluation of genomic variability at the nucleic acid level. Plant Mol. Reporter 1:9-16. Rogers, R. O., and A. J. Bendich. 1987. Ribosomal RNA genes in plants: variability in copy number and in the intergenic spacer. Plant Molec. Biol. 9:509-520. Rogers, S. O., Rehner, S., Bledsoe, C , Mueller, G. J. and Ammirati, J.F 1989. Extraction of DNA from Basidiomycetes for ribosomal DNA hybridizations. Can. J. Bot. 67:1235-1243. Rohlf, F. J. 1993. Numerical Taxonomy and Multivariate Analysis System. Version 1.80. State University of New York, Stony Brook. Ruano, G., Brash, D.E. and Kidd K. K. 1991. PCR: the first cycles. Amplifications 7:1-4. 7 8 SAS User's Guide: Statistics 1982 Edition. SAS Insititute Statistical Analysis System Saghai Maroof, M.A., Allard, R.W., and Zhong,Q. 1990. Genetic diversity and ecogeographical differentiation among ribosomal DNA alleles in wild and cultivated barley. Proc. Natl. Acad.Sci. U.S.A. 87:8486-8490 Schulting, M. J. 1988. The genetics of disease interaction in the lodgepole pine: western gall rust host:parasite system. M.Sc. thesis UBC 55 pp Smith, D.R. and Stanosz, G. R. 1995. Confirmation of two distinct populations of Sphaeropsis sapinea in the North Central United States using RAPDs. Phytopathology 85:699-704. Smith, M. L. Bruhn, J .N. and Anderson, J. B. 1992. The fungus Armillaria bulbosa is among the largest and oldest living organisms. Nature 356:428-431. Smith, M. L. Bruhn, J. N. and Anderson, J. B. 1994. Relatedness and spatial distribution of Armillaria genets infecting red pine seedlings. Phytopathology 84:822-829 Sneath, P. H. A., and Sokal, R. R. 1973. Numerical Taxonomy— The principles and practice of numerical classification. W. H. Freeman and Company, San Francisco Soltis, P.S., Soltis, D E . and Ness, B.D. 1989. Population genetic structure in Cheilanthes gracillima. Amer. J. Bot. 76(8): 1114-1118 Specht, E.A., Novotny,C. and Ullrich, R.L. 1984. Strain specific differences in ribosomal DNA from the fungus Schizophyllum commune. Curr. Genet. 8: 219-222 Sun, L.-J., Carlson, J E. and van der Kamp, B. J. 1995a. A rapid procedure for preparing high molecular weight DNA from rust aeciospores for RAPD PCR analysis, p. 143-148 in: Rusts of Pine. Proceedings of the IUFRO 4th Rusts of Pine Working Party Conference, 1994 Tsukuba, Japan. Sun, L.-J., van der Kamp, B. J. ,Carlson, J. E. and White E. E. 1995b. Preliminary study of genetic variation in western gall rust populations in British Columbia, Canada, using molecular markers. P149-156 in: Rusts of Pine. Proceedings of the IUFRO 4th Rusts of Pine Working Party Conference, 1994 Tsukuba, Japan. True, R. 1938. Gall development on Pinus sylvestris attacked by the woodgate Peridermium and morphology of the parasite. Phytopathology. 28: 24-50. Tulsieram, L. K., Glaubitz, J. C , Kiss, G. and Carlson, J. E. 1992. Single tree genetic linkage mapping in conifers using haploid DNA from megagametophytes. Bio/Technology 10:686-690. Tuskan, G. A., and Walla, J. A. 1989. Isozyme charaterization of Peridermium harknessii 79 and Cronartium quercuum f. sp. banksianae with starch gel electrophoresis. Phytopathology. 79: 444-448. Tuskan, G. A., Walla, J.A. and Lundquist, J.E. 1990. Genetic geographic variation in Peridermium harknessii in the north-central United States. Phytopathology 80: 857-861. Ullrich, R.C, and Anderson, J.B. 1978. Sex and diploidy in Armillaria mellea. Exp. Mycol. 2:119-129. van der Kamp, B. J. 1987. Western gall rust in British Columbia, in Proceedings Western International Forest Disease Work Conference. pp8-13. van der Kamp, B. J. 1988a. Susceptibility of lodgepole pine provenances to geographically separate western gall rust spore sources. Can J. For. Res. 18: 1203-1205. van der Kamp, B. J. 1988b. Temporal and spatial variation in infection of lodgepole pine by western gall rust. Plant Dis. 72: 787-790. van der Kamp, B. J. 1994. Lodgepole pine stem diseases and management of stand density in the British Columbia interior. Forestry Chronicle 70: 773-779. Vogler, DR., Kinloch, B.B., Cobb, Jr. F.W and Popenuck, T. L. 1987. Isozyme patterns in Endocronartium harknessii from several pine hosts and stands in California. Phytopathology. 77: 1242(Abstr.) Vogler, D.R., Cobb, Jr. F.W. Epstein, L. 1989. Nuclei in mature aeciospores from two zymodemes Peridermium harknessii in California. Phytopathology. 79: 1164 (Abstr.) Vogler, D. R., Kinloch, B. B., Cobb, Jr. F. W and Popenuck, T. L. 1991. Isozyme structure of Peridermium harknessii in the Western United States. Can. J. Bot. 69: 2334-2440. Walla, J. A. Tuskan, G. A. Lundquist, J. E. 1991a. Comparison of white- and orange-spore Peridermium harknessii in native North Dakota Pinus ponderosa Mycologia 83:251-256. Walla, J.A. Lundquist, J.E., GA. and Freeman T.P. 1991b. Observation on the spermogonia stage of the western gall rust fungus. In Rusts of Pine, Hiratsuka et al. (eds.) Northwest Region, Info. NOR-X-317, pp. 54-62. White, E.E., Ford, B.M. and Kinloch, B.B. 1994. Variation in the ribosomal gene cluster of Cronartium ribicola. in press Williams, J.G.K., Kubelik, A.R., Livak, K.L., Rafalski, J.A. and Tingey, S.V. 1990. DNA polymorphisms amplified by arbitrary primers are useful as genetic markers. Nucleic Acids 80 Res. 18:6531-6535. Wu, M.J., Cassidy, J.R., and Pukkila, P.J. 1983. Polymorphisms in DNA of Coprinus cinereus. Current Genetics 7:385-392. Yamaoka, Y., and Katsya, K. 1985. Basic-characteristics of axenic culture of rust fungi: possibilities for culture collections. Bull. Jpn. Fed. Cult. Coll. 1:20-25 (in Japanese) Yanchuk, A.D., Yeh, F.C. and Dancik, BP. 1988. Variation of stem rust resistance in a lodgepole pine provenance - family plantation. For. Sci. 34: 1064-1075. Yeh, F.C, Cheliak, W.M. et al., 1985. Population differentiation in lodgepole pine, Pinus contorta sp. latifolia: a discriminate analysis of allozyme variation. Can. J. Gene. Cytol., 27: 210-218. Ying, C. C , Ulingworth, K. and Carlson, M. 1985. Geographic variation in lodgepole pine and its implications for tree improvement in British Columbia. In: lodgepole pine the species and its management D.M. Baumgartner et al., eds. Coop. Serv. Wash. State. Univ. Pullman Ziller, W. G. 1970. Studies of western tree rusts. VII. Inoculation experiments with pine stem rusts (Cronartium and Endocronartium) Can. J. Bot. 48:1313-1319. Ziller, W. G. 1974. The Tree Rusts of western Canada. Can. Dep. Environ., Can. For. Serv., Ottawa, Ont, Publ. 1329. Zimmer, E.A., Jupe, E.R., and Walbot, V. 1988. Ribosomal gene structure, variation and inheritance in Maize and its ancestors. Genetics 120:1125-1136. 81 APPENDIX I Southern Blot Procedure 1. Stain and photograph the 1% agarose gel with the DNA digests. Trim to lanes to be blotted. 2. Place gel in 1.5 NaCl2/0.5 M NaOH at RT for 1/2 hr. to denature DNA with occasional agitation. 3. Neutralize gels in 1.5M NaCl2 /IM Tris-HCl pH7.4 for l/2hr. 4. While gel is in this solution, prepare Hybond-N membranes, cut to the exact size as the trimmed gel. For each gel, prepare 2 sheets of filter paper and several sheets of paper towel to the size of gel. Have a large stack of paper towel ready. 5. Set up the blots: (from the bottom up, step by step) Pour in a tray 500ml 10X SSC; place a support on the tray, covered with saran wrap except in the place that will be covered by a stack of paper, 4 sheets of filter paper soaked by the SSC buffer and placed on the support (take the air bubbles out); place the gel (DNA face up) on the filter paper, and Hybond-N membrane on top of the gel; stack cut filter paper; stack trimmed paper towel; stack paper towels and weight (about 1kg). 6. Leave blots overnight. 7. Remove weight and papers; peel off upper membrane, rub it gently in 2X SSC to remove any agarose, let dry and label it with pencil (date, gel #, lanes, markers, write on the side of membrane with DNA) 8. Wrap blots in Saran Wrap, place DNA side down on UV light table for 5 min. to covalently bind DNA to membrane. 9. Wrap blots in aluminum foil and store in desiccator. (* handle membrane carefully, DNA will not bind to membrane that has been touched with fingers). 82 D N A probe labeling 1. Aliquot probe (pAml and pAm2) (16 ul DNA at lOng/ul) into a sterile 1.5 ml Eppendorf tube and add 2 ul hexanucleotide mix (10 X concentration, Boehringer Mannheim Cat. No. 1277065), 2 ul Dig DNA labeling mixture and 1 ul Klenow (primer) 2. Briefly centrifuge 3. Incubate at 37° C overnight 4. Add 2 ul EDTA (0.2M, pH8.0) to stop the reaction 5. Add 2.5 ul LiCl (4M) 6. Add 75 ul 100% prechilled ethanol and mix well 7. Leave it at -20° C for 2 hr. 8. Centrifuge at 12000g for 15 min. 9. Keep the pellet, add lOOul 85%) prechilled ethanol; wash the pellet. 10. Spin at 12000g for 5 min. 11. Dry the pellet using vacuum dryer. 12. Dissolve the pellet using 50 ul IX TE buffer at 37° C for 1/2 hr. 13. Boil probe at 95° C for 10 min. to denature the DNA. 14. Cool the tube on ice. Solutions: NRL: 50% deionized formamide 5X SSC 0.1% Sodium Laural Sarcosine(Na+) (SLS) 0.02% Sodium dodecyl sulphate (SDS) 2% blotting agent (milk powder diluted 50% with distilled water, not autoclaved) SSC: 20X stock solution(lL) Dissolve 175.3 g NaCl2, 88.2g Tri-sodium citrate in 800 ml dH 20 Adjust the pH to 7.0 with 1 ON NaOH Adjust the volume to IL 83 Autoclave for 15 min. TAE: 5 OX stock solution 242 g Tris base in 800ml dH 20 Add 37.22g EDTA.2H 20 Adjust pH with glacial acetiuc acid to 8.0 and bring to 1L by dH 20 TPE:10X 108 g Tris base, 15.1 ml 85% phosphoric acid, 40ml 0.5 M EDTA, (pH8.0) Prehybridization and hybridization procedure Take membrane from desiccator u Place membrane in plastic bag u Add 30ml NRL solution to each bag u Seal well and double bag Place the bag in 42° C water bath for 2 hrs minimum (Prehybridization). u Cut open and pour out the NRL solution u Add 3 ml NRL to each bag Add ice cooled denatured, labeled probe (30ul) to each bag u Seal well (remove any air bubbles), and double bag 84 Incubate in 42° C water bath overnight (Hybridization) Stripping the blots and non-radioactive detection 1. Wash the hybridized membrane twice in > 100 ml 2X SSC and 0.1% SDS at RT for 2 min. 2. Wash blots twice with 2X SSC/0.1% SDS at 55° C for 10 min. 3. Wash blots twice with 2X SSC at 55° C for 10 min. 4. Wash blots twice with 0. IX SSC at RT for 2 min. Antibody binding 1. After washing the hybridized blots, transfer blots to a tray with buffer A for 1-5 min. 2. Remove the membrane from buffer A and allow excess liquid to drip off for approximately 5 seconds. Do not allow the membrane to dry. 3. Using a freshly washed dish or a new bag, block the membrane by incubating it in buffer B for 3 firs, with gentle shaking. 4. Near the end of the blocking step, dilute the anti-digoxigenin alkaline phosphatase conjugate 1:10000 in buffer B for working concentration of 75ul/ml. (3ul to 30ml buffer). 5. Remove the membrane from step 3. and transfer it to the antibody conjugate solution at RT for 30 min with gentle shaking. 6. After 30 min. incubation, discard the solution and use fresh buffer A and a clean container to wash the membrane for 15 min.; repeat wash. 7. Equilibrate the washed membrane in buffer C for two min. 85 Lumi-phos visualization (Boerhringer Mannheim lumi-phos 530 kit Cat. No. 1275470) Continued from step 7, above: 8. Place the membrane on a clean sheet of acetate film (e.g., photocopier transparency film). Pipette 750 ul of lumi-phos 530 directly onto the membrane. Place a second sheet of acetate film on top of the filter (membrane) to evenly spread the solution by capillary action so that the entire membrane is covered. 9. Expose to X-ray film immediately or incubate at 37° C for 30 min to allow light emission to reach a steady state. 10. In a dark room, expose the sealed membrane to X-ray film (Kodak XAR film) for 1.5 hr. AMPPD visualization AMPPD[3-(2'-Spiroadamantane)-4-(3 '-phosphoryloxy)-phenyl-1,2-dioxetane] Buffer 1: maleic acid, 0.1 mol/1 ;NaCl2, 0.15mol/l; pH7.5(20°C), adjusted with NaOH, autoclaved. Buffer 2: blocking stock solution diluted 1:10 in buffer 1 (final concentration 1%). Blocking stock solution: Blocking agent (milk powder) 10% w/v in buffer 1, autoclaved and store at 4° C. Buffer 3: Tris-HCl, 0.1 mol/1; NaCl2, 0.1 mol/1; MgCl2, 50mol/l; pH9.5 (20 0 C) Hybridization buffer: Formamide, 50%(v/v); 5X SSC; blocking agent, 2% (W/v); N-lauroyl-sarcosin, 0.1% (w/v); SDS, 0.02% (w/v). Washing buffer: buffer 1+Tween 20, 0.3%(v/v). AMPPD: stock solution:10 mg/ml; 23.5 mmol/1. Final solutions freshly diluted 1:100 in buffer 3 = 0.235 mmol/1. Hybridization and washes follow the same steps as Lumi-Phos 530 visualization, above. 86 Plasmid culture and large scale purification: pAml and pAm2 inserted into E.coli plasmid pUC 9 and stored as a glyceroal stock at -70°C. 1. Culture E.coli on agar plate containing 200 ul/lOOml carbencillin. 2. Make streak culture on the plate and seal the plate into a bag (place upside-down), for overnight. 3. Transfer individual colony to liquid media (200 ul/lOOml Carbenicillin), grow overnight at 37° C with shaking. 4. Centrifuge cells at 800 RPM for 3 min, drain and dry pellet 5. Suspend pellet in 6.6 ml pid 1 solution, sit at RT for 1-2 min. 6. Add 6.7 ml of pid 2 to each tube, gently invert tube to blend. Sit on ice for 5 min. 7. Add 5ml pid 3 to each tube, blend gently. 8. Centrifuge at 9000 RPM for 3 min. 9. Pour supernatant though Kim-wipes into clean tube. 10. Add 9 ml cold (-20°C) isopropanol to each tube. Balance the tubes. 11. Seal with parafilm, leave in -20°C freezer for 2-3 hrs. 12. Centrifuge at 10,000 RPM for 5 min. 13. Add 1 ml PTE to each tube; break the pellet with a Pasteur pipette, transfer to 4 ml graduated centrifuge tube; rinse tubes with PTE and combine rinse in centrifuge tube. Bring volume to 4 ml with PTE. 14. Add4gofCsCl. 15. Add 200 ul of 10 mg/ml Ethium Bromide. Check the refraction index. Adjust to a density of 1.3880-1.3900. 16. Seal tubes; centrifuge overnight at 53,000 RPM (250,000xg) 17. Gradient should have two bands, - upper band is chromosome and linearized DNA and lower band is plasmid DNA; Harvest lower band only; extract ethidum bromide from plasmid DNA with isoamyl alcohol; recover plasmid DNA by ethanol centrifiigation. 87 Solutions Pid 1: EDTA 3.72g/l; Glucose, 9 g/1; Tris, 3.028 g/1; pH 8.0. Pid 2: NaOH, 8 g/1; SDS 10 g/1; freshly made each week. Pid 3: NaAc*3H20, 408 g/1, pH 4.8 with HC1 Extraction of pAml and pAm2 from pUC9 plasmid Digestion of plasmid DNA with enzyme Bam HI 1. Mix 50 ul DNA , 50 ul buffer, 2.5 ul enzyme Bam HI (8-12 units / ul), 400 ul dH20. Incubate at 37°C for 3 hrs. 2. Add 2.5 ul Bam HI (8-12 units / ul) and incubate for another 1 hr. 3. Prepare a 0.6% low melting agarose (Fisher Scientific, BP 1360-25) plus 0.15% Agarose with TAE buffer. 4. Make the gel with only two sample wells, a small one for the Lambda DNA marker and a large one for loading 1225 ul digested plasmid. Electrophorese at low voltage overnight. 5. Cut the insert bands from the gel under the UV light. There should be 3 bands on the gel - remove the pAm2 (upper) and pAml (middle) bands and a discard the third band (vector), with care to avoid contamination. 6. To extract pAml and pAm2 from gel slice. Place the sliced band in a pre- weighed microcentrifuge tube. Determine approximate volume of the excised gel. (0.4 g gel in 1.5 ml microcentrifuge tube) 7. Add 3 times volume of prep-a-gene binding buffer to the gel slice; place the tube in a 50° C water bath. After a minute or two, mix the contents of the tube. The agarose gel should be completely dissociated. Carefully observe the solution while mixing contents to be sure all agarose has dissolved. 8. Add Prep-a Gene matrix to bind DNA (0.2 ug DNA per ul of matrix). Use a minimum 88 of 5 ul. Vortex the insoluble matrix stock until all of the contents are in suspension before using. This requires a minute or so of vigorous mixing. Incubate at RT for 5-10 min. to allow DNA binding. 9. Pellet the matrix with bound DNA. Microcentrifuge at 10,000rpm for 30 seconds. 10. Discard the supernatant(may be kept in separate tubes in case DNA not recovered). 11. Rinse the pellet containing the bound DNA by resuspending it gently in 50 pellet volumes of binding buffer. Centrifuge as above and repeat rinse step, discard supernatant each time. 12. Wash pellet twice with washing buffer(from Prep-A-gene kit). And spin 30 sec. 13. Remove the last drop of supernatant using a Kim-wipe. 14. Add 10 ul 1 x TE buffer to dissolve the pellet (50 °C for 3 min.). Run an aliquot on a mini gel to check the DNA concentration. 15. The DNA is now ready for labeling (probe). 89 APPENDIX II Table 6. Data matrix based on AvaII/pAm2 fragment length variation in 24 DNA samples from 5 populations DNA samples/ 2.8 2.9 3.0 3.1 3.2 3.4 3.6 Fragment length (kb) QSl 0 0 0 0 0 1 0 QS2 0 0 0 0 0 1 0 QS3 1 0 0 0 0 0 0 QS4 0 0 0 0 0 1 0 MP1 1 0 0 0 1 0 0 MP2 0 0 0 1 0 0 0 MP3 1 0 0 1 0 0 0 MP4 1 0 0 0 0 0 0 MP5 0 0 0 0 1 0 0 MP6 1 0 0 0 0 0 0 PG1 0 1 0 0 0 0 0 PG3 1 0 0 0 0 0 0 PG5 0 1 0 0 0 0 0 PG6 0 0 1 0 0 0 0 PG8 0 1 0 0 1 0 0 VH1 0 0 0 0 1 0 0 VH2 0 0 0 0 1 0 0 VH3 0 0 0 0 0 0 1 VH4 0 1 0 0 0 0 0 VH5 0 0 0 0 1 0 0 NR1 0 0 0 0 1 0 1 NR2 0 1 0 0 0 0 0 NR4 0 0 0 0 1 0 1 NR5 0 1 0 0 0 0 0 90 A P P E N D I X III Table 7. Average frequencies of polymorphic bands observed in 12 populations Primer-band Frequency Primer-band Frequency Primer-band Frequency p06-2 0.016667 p82-3 0.216417 p38 -7 0.65 p85-3 0.025 p37-9 0.216667 p98-4 0.658333 p37-l 0.025 p90-8 0.219667 p85-4 0.662083 pl8-3 0.025 p89-4 0.258333 p37-6 0.666667 p98-7 0.036083 p89-l 0.266667 p06-l 0.675 p98-3 0.041667 p98-l 0.278667 p75-2 0.725 p38-2 0.05 p38-ll 0.308333 pl8-2 0.735167 p06-3 0.05 p82-2 0.31175 p85-6 0.746333 p37-4 0.066667 p37-5 0.366667 p31-4 0.776583 p75-4 0.066667 p82-9 0.403083 p37-7 0.783333 p67-4 0.067583 p82-10 0.422417 p75-5 0.791667 p37-3 0.075 p31-l 0.425667 p82-l 0.7975 p38-4 0.075 p31-ll 0.427083 p06-5 0.8 p38-8 0.075 p77-l 0.433333 p31-8 0.832417 p85-7 0.083333 p77-4 0.458333 pl8-l 0.87225 p38-6 0.1 p31-7 0.475 p31-6 0.873167 p38-9 0.1 p98-6 0.483333 p77-8 0.875 p90-2 0.10275 p90-4 0.489333 p75-l 0.883333 pl8-5 0.10925 p85-2 0.49525 p37-2 0.9 p82-6 0.11875 p38-l 0.5 p90-6 0.911583 p89-6 0.125 p67-8 0.5 pl8-8 0.914.833 p90-3 0.12775 p31-3 0.506 p85-5 0.923167 p67 -9 0.135167 p77-6 0.541667 p67-10 0.932417 p82-4 0.135167 p98-9 0.558333 p31-10 0.95 p67-l 0.13975 p98-8 0.558333 p77-3 0.95 p82-7 0.15275 p75-7 0.591667 p77-5 0.966667 p38-10 0.158333 p98-2 0.617583 p38-3 0.966667 p77-ll 0.166667 p67-7 0.633333 P18-4 0.975 p90-5 0.179667 p06-8 0.641667 p98-10 0.975 p82-5 0.181667 p98-5 0.646333 p37-8 0.991667 p90-7 0.188917 p38-5 0.65 p90-l 0.991667 p31-2 0.205583 91 A P P E N D I X I V Table 8 . Data matrix of 120 individual galls and 96 polymorphic RAPD bands p31-1 p31-2 p31-3 p31-4 p31-6 p31-7 p31-8 p31-10 p31-11 p85-2 p85-3 p85-4 p85-5 p85-6 p85-7 p37-1 p37-2 p37-3 p37-4 p37-5 p37-6 p37-7 p37-8 p37-9 p38-1 p38-2 p38-3 p38-4 p38-5 p38-6 p38-7 p38-8 p38-9 p38-10 p38-11 p18-1 p18-2 p18-3 p18-4 p18-5 p18-8 p89-1 p89-4 p89-6 p06-1 p06-2 p06-3 p06-5 p06-8 p77-1 p77-3 p77-4 p77-5 p77-6 p77-8 p77-11 p75-1 p75-2 p75-4 p75-5 p75-6 p75-7 p90-1 p90-2 p90-3 p90-4 p90-5 p90-6 p90-7 p90-8 p98-1 p98-2 p98-3 p98-4 p98-5 p98-6 p98-7 p98-8 p98-9 p98-10 p67-1 p67-2 p67-4 p67-7 p67-8 p67 -9 p67-10 p82-1 p82-2 p82-3 p82-4 p82-5 p82-6 p82-7 p82-9 p82-10 C 1 001111110001110010010110011001010001001001011000111010101101111000011101000001011101101100000110 C 2 000110110101110010000110001000100001101010111000101111101101101000010001000001010101101100000010 C 3 100110111101110010011110101010000001101010111000111111101101101001010001011101110111101110010001 C 4 100110110101110010001110101000100001101010111000101110101101101001010001000001010111101100000010 C 5 000110110101110010000110101010100001101110111000111111101100111101010101100001011101001010000111 C 6 100010110000110011000110101010000001101011111000111111101101101001110001011101110101101100000100 C 7 100110110101110010110110101010100001101010111000111111101101111001110001111101110101101100000001 C 9 011111110101110010000110101010100001101010111000101101101111111001110001011101110101101000000000 C 10 100110110000110010000110001011010101101110011000111111101101111000010101011100110101101 C 11 101110110000110011011110001011000101101011011000111111101101111000010101011100110101101100000011 LB 1 000111111001110010011110101010000011001001101001001010101001111001111001000001010101101100100000 LB 2 010110011101110010101110101010000001101010101001101011101001101000010001011101111111011100000001 LB 3 000011111101100010011110101010000001101010101001110010001000101000000000000001010101001100000000 LB 4 100110011101110010011110101010000001101010101001001010101001111001010001011101110101101010000001 LB 5 000011111000100000001100001000100001001001101001001011101101111000000001010001000100101100000000 LB 7 110110011101110010000110101010000001101010101001001111101101111001011001011101110101101100000000 LB 9 000010111101110010001110111110000001101011101001011111101101111000010001010101010100101100000000 LB 6 011110011101110010000110001000000001101010101001001001101111111001010001011101110101101100000100 LB 8 100110011101110010001110001010000011101010011001111111101001111000011001011101110100001100000011 LB 10 00101011110111001000111000110100000 101100100111110100111100101100000000111 100010001 RNP 7 100110011101110010001110101010000001101011000001001110101001111001010001011011010101101100000100 RNP 10 111110010101110010011110101010000001101010001011101011101001111000010001010101010101101100000100 RNP 24 101110010101110010001110101010000001101011001011101101101001111001010001011101010101101110000100 RNP 25 101110010000100010001110101010100001101010001001101010101001111101000001010101010101101011000001 RNP 28 000110010101111010001110101010000001101011000001001110101111111001010001011101010101101100000010 RNP 29 101110010101110010011110101010100001101011001001011111101111111001010001011101010101101100000000 RNP 31 101O10010101110010011110101010000001101 Oil 001001111111101101111001010001010101010101101100000100 RNP 27 011111010101111010011110101010100001101011100001111111101101111001010001010101010101101100000000 RNP 30 00011001000111001001111001100000000110101001000111101110100111 ...01010101010100001110011110 RNP 33 100110010000110010011110001000100001101011011001111011101001111000011001011101010100101111011110 UV 6 010010100101101010000010101110100011101010101001101111101001101001010010001101010101000100011010 WV 7 010110100101100010000110101010000001101010101001101111101001101000010101001101010101100100000010 WV 8 010110110101111010001110101110100001101010101001101111001001101000010011001101010101100100010000 WV 9 010110100101100010000110101010000001101010101001111111101001101000010001001101010101100100000000 WV 10 000110100101101010011110101010100001101011100001101111001001101001110001000101011101100100010000 WV 11 000010100000100010000110101010100001101011100001010111011000101000010101000101010111011100000000 WV 12 011100010101100010101110101010100001101011100001011111101001101001010101011101010101100100000101 WV 3 011100110101110110101110101010000001101010100001011111111111111001010101001101010101100100100000 WV 4 000010110000110010010010101000100101101010100001001011101000101000010101011101011101111100000010 WV 5 000110110000110010000110001100100101101011100001001111101000101000010011000001011101101100000010 MPM 1 001110011001110010011110001000100101001010001001001111100100111000010010011000111101101101000010 MPM 2 110100110000110000011110001000100100001010001001111110000101111000010010011000110100001110000000 MPM 3 010101111101110000011110001000100110001000001001111111100101111101010010011001111100001100101001 MPM 4 100100100001110000011110001000100110001100000001001010100101111001111010011000111100001100000000 MPM 5 100101111001110010011110001010100001101000001001011110100101111000110011010000110101101010000011 MPM 6 100100111001110000011110001010100101111010001001111110000101111000010010111000111101101110010011 MPM 7 000100111001110010011110001010100011101010001011001010100101111000010010010000110100001010000011 MPM 8 001111111 000011110001000100100010100001011001010100101111000001010011000110100010001010001 MPM- 9 10111111100110000001111000100010010110101000101100101010010111100010001011100011 MPM 10 000110111000110010011110001000100111001000001001101010100101111000010011111000110100001 MPL 2 110110111101110000001110101010100101101011001001101011101001101001010011011101010101101100100000 MPL 4 011100111100110010001110101010000011101011001001101011101001101001111010011101101101101 MPL 5 010010111101100000001110101010100111101010001001101011101001101000010101001101010101101100000000 MPL 6 110010111101100010001110101010100011101010001001101011101001101000010010010101010101101101000000 MPL 9 000010111101100010001110101010100101101011001001101011101001101001010110001101011101101100000000 MPL 10 000010111101100010001110101010100101101010001001101011101001101001010110001101010101101100000000 MPL 11 010010111101110010001110101010100001101010001001111011101001101001010100001101010101101100000000 MPL 12 011100011101110010011110001010101011101011001001101011101001101001010101000001010101101100000000 92 Continuation of Appendix IV: MPL 1 10011000000111000100001001110101000100110101110100111100001010100000101 .100000000 MPL 14 100110111100110010011110001000100111101011001001101011101001101000010101010101010101101100000000 VH 1 001111110001110010011110001010000011001010001001111110101100101001110001000000100101001101010011 VH 2 001111110010110010011111001000100011001010001001101010101100101101111001000000111110011100010011 VH 3 111111110001110010011111001010100010001011001001111010101100100101100011011000110100011100010011 VH 4 101111110001110010000010001010000010001010001001101011101100101001110010011100110101001110100001 VH 5 001111110100110010011110001010001001101011001001101010101100101001110010011000110101001100010001 VH 6 111111111000110110011110001010000001101010001001101010101100101001010000011000110101001 VH 7 101111110100010110011111101010100011101010001001101010101100101001010001011000110101011010000001 VH 9 000111010100010010000010101010101010001010001001101011100100101000000000001000110100001010000011 VH 10 10111011010001001001001000110110001 0011001101011111100101010010000011000110101001100000011 VH 11 101110110100010000011011001101100011101110001001101011101100101000011010011100110100011010000001 PGM 2 000100010000111001000110001101010001001011010001101111111011101011000110101011001000011 PGM 3 000111111101110011000111001001100001101010100000001110001101111000111001011000110100001011101011 PGM 8 101101110101100010001111001000110000001011000000001110001101111000011001010000110100011110011001 PGM 11 100101110000110010001111001000110001101111000011011111101101111011110011011000110100001110101001 PGM 12 000101110000110010010110110010100001101011000001011010101101111100110010011000110100001011001001 PGM 14 000101110001110010000111001010100000001010000001011010101101111000110011011000110100011111001011 PGM 17 010010111001010100000001011100001011010101101111001011010011000110100011010011001 PGM 20 000011110101100010111111001010110001101011101001011111101111111000010001011000110100011111001001 PGM 4 00111111000101001011111100100011000100101000000100101010110011 11011000110100011 PGM 5 100111110001010010000111001010110001101010000001001010101101111011111001011000110100011 TML 1 101111110001100010001110101010101001101010101001111111111101111000011000011000110101101100010010 TML 2 001111110000110010001111101000000001000110000001111111111101111001110000011000110101001100000000 TML 3 101011111001110010001111001000100001000100001001101011011101101101011101001000110101001100010000 TML 4 001111111000100010001110101010000001101100001001111111011110101011011101000001010101101100110010 TML 5 001111111001110010001110101010000001101010000001011111111101101001011101000001010101101100010010 TML 6 101111110000110010000011101010000001101010101000101111111101111000010001011101010101001100000010 TML 7 001111110101110010011110101010000001101010000001011111111101111000010001011001010101101100000001 TML 9 101110110101110010001110101010000001101010000001111111111101111000010010011100110101101100000010 TML 8 101110110000110010011111101011000001101010000001011111111100101110000001011100110101001100000011 TML 10 001110110000110010100111101011000001101010001000101111111101101000110001011100110101001110000011 Q 1 000110111000110010011111101010100001001010000001101010101111111010010001000001010111101111000100 Q 2 001110111000100010001110110011100001101010001001111010101101111001000001010101010101101101000000 Q 3 001011111001100010001111001010100001001010001001101010101101111010010001000001010101101111000000 Q 5 001011111000110010001110001010100001001010001001101001101101111010010001000001010110001111000100 Q 6 001111111000100010000011101110100001101010000101111010101101111000010001010101011100001011000000 Q 7 101110111010100010001010101010100101101010000001111010101101111000010101010101010101101101000100 Q 8 101111111110110010010011001100100001101010000001101010101101111011010001000001010110001011100111 Q 9 000011111101110010000010001001100011001010000001101010101101111000000001010101010100001011100111 Q 10 00001111100011001000111000100010000110101000000110101010110111 01010001010100001101100010 Q 11 00111011100011001000011000100010000110101000000110101010110111 01010001010100001101000010 WL 1 001111111101100010000010101010000001101010001101101010101101101000010000001100110101101100000010 WL 2 000111111001100010000010101010000001101010001001000010001101101000010000001100110101001100000000 UL 3 001011110000100010000011000000100000001010001001100010101100101101011000001101110100001101000000 UL 4 001011111001100010010010101010100011011010000001100010001001101010010000001101110101001100100000 UL 5 001011110001100011001010101010000011101010000001101010001000101001010000001100110100001011100001 UL 6 101011110001100010000010101010100011101010000001100011111100101000010000001100111100001010100001 UL 7 100011110001100010000010101010100011101010000001111010001100101001010010001100110100011011000001 UL 8 010011111101110010000010110011000101101010000001101010101101101001010000001100110100001010000101 UL 9 110111111000110010000111101010100001101010000001101010111100101001010100001101010100001010000100 UL 10 100111111000110010101110001010T00011101010000001101011111100101000010000001101011100001100000010 100MH 3 101111110101111010000010001000100010001011001000011110111101111001010110010000110101101100000010 100MH 5 001111110101111010000011001000101011101110001000011110100101111000010111011000110101101100110010 100MH 6 001110110101110011001010001000101011101110001000011110110101111011010010011000110100001100000000 100MH 7 001111110001110010001010001000101011001110001000011111111101111110010100011010110101101100000010 100MH 8 101111110001010010001010001000001011101010001000111111111101111101011100011010110101101100000011 100MH 10 101110110101111011001010001000101011101010001000011110100101111011010010010000110100001100001011 100MH 1 101110110101111010001111001000100010001010000000011110101101111000010000011010110100001100000001 100MH 11 000110010101111011001010001010101011101110001000011110101101111001010000010000110100001100000011 100MH 2 101111110001010011001110001000101010001011001000001010101101111000010000010000110100001 100MH 4 1011111100010100010011100010101010100010100010001110101011011110000100 0100001 93 0.00 0.25 0.50 0.75 1.00 ci PNP30 RNP33 LB I LB5 RNP25 0 '2 0 ? G 6 0 I 0: 3 0' 5 •cE •cE •cE •cE 0 'J o- IC 0 II 0 8 C2 C1 C3 C7 Zi C9 LB2 LBS LSI LB? LBS RNP? RNP28 PNP10 RNP24 RNP29 P.NP3I RNP27 LB9 UUS W? UU9 UU8 UUIO NPL2 I1PL5 j — HPL9 ~ — I1PL10 NPLII HPL« NPLI2 HPLM MPH UUI2 UU3 CIO Cll TtlLI IflL9 TNL7 TI1L2 — THLS ML 10 TML8 UU1I — m UU5 — C5 . IHL3 WL1 IHL5 LB3 UL UL — UL — Ut» — UL — UL — UL — UL — UL — UL — UL — V H I — U. H 2 — U H 3 — U. H5 — U H< — UK? — V H 9 — U H IO — U HII — LBIO — MPL I — npni — «PH9 — tlPMO — np«2 — HPH« — HPH5 — ma — ma — NPNI — NPN8 — I00NH3 — I00NH5 — I00NHS — I00NH10 — I00NHII — I00NHI — 100MH? — looms — I00NH2 — 100HH1 — PGN2 — PGNI7 — PGN3 — PGH8 — PGH20 — PGNII — PGM12 — PGNH — P6M1 — PGN5 Figure 12. Dendrogram of 120 individual galls from 12 populations drawn by NTSYS TREE DISPLAY using similarity coefficient and UPGMA clustering method. 54 ,00 0.25 0.50 0.75 1.00 H I - c 0 1 0 3 0