LIBRARY Michigan State University PLACE N RETURN BOXtonmmfihdnckMflomywrrocord. TO AVOID FINES Mum on Of Mon dd. duo. ‘ DATE DUE DATE DUE DATE DUE MSU inAnNflmdlchciion/Ewommimwon mm: Chemical and Plastome-Mutator Mutagenesis of Oenothera By Susan Baldwin A Thesis Submitted to Mighigan State University in partial fulfillment of the requirement for the degree of MASTER OF SCIENCE Department of Botany and Plant Pathology 1995 ABSTRACT CHEMICAL AND PLASTObm-MUTATOR MUTAGENESIS OF OENTHERA By Susan E. Baldwin In Oenothera, an elevated spontaneous mutation rate occurs when a plastome mutator allele (pm) is homozygous. Seeds of the plastome mutator line were treated with nitrosomethyl urea in an attempt to isolate antibiotic resistance mutations. The chemicals 9-aminoacridine hydrochloride, nalidixic acid, and novobiocin were used to test for synergism in the induction of chlorotic mutations. According to chi-squared analyses, acridine and nalidixic acid increased the pmamutation rate in an additive fashion. In contrast, novobiocin had a synergistic effect on the mutation frequency of the pmlpm plant lines, while having little impact on the wild-type control. These results implicate subunit B of gyrase as the possible product of the pm gene. ACKNOWLEDGEMENTS I wish to express my sincerest gratitude to Dr. Barbara Sears for her support, patience, and guidance. To my parents for their love and support, and a very special thanks to my family, Rick, Kristen and Scott for being there. TABLE OF CONTENTS LIST OF TABLES O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O 0 LIST OF FIGURES O O O O O O O O O O C O O O O O O O O O O O O O O O O O O O O I O O O O O O O O IMODUCTION. O O O O O O O O O O O O O O O C O O O O O 0-. O O O O O O O O O O O O O O O O O O 0 CHAPTER 1. NMU MUTAGENESIS AND SELECTION OF ANTIBIOTIC RESISTANCE SECTORS IN OENOTHERA IntrOduCtionoooooooocoo-0000000000000...0.00.0000... materials and MethMSOOOOOOOOOOOOOOOOOOOOOOOO0.... ResultSCOOO0.00...0......OOOOOOOOOOOOOOOOOOO00.... DiscuSSiODOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO CHAPTER 2. CHEMICAL MUTAGENESIS USING ACRIDINE, NALIDIXIC ACID AND NOVOBIOCIN IntIOduCtionoooooooo0000000000000.000.000.000.000... Materials and Methods 9-Aminoacridine hYdIOChloride O O O O O O O O O O O O O O O O O O O “alidiXic ACid OOOOOOOOOOOOOOOOOOOOOOOOOOO..0... NOVObiOCin .0...0......OOOOOOOOOOOOOOOOOOOO0.... Results 9-Aminoacridine hydrochloride Dosage Trials ............................... Phenotypic Alterations ...................... Inheritance .0...OCOOOOOOOOOOOOOOOOOO00...... iv iii vi 11 12 18 20 29 30 3O 31 4O 41 Nalidixic Acid Dosage Trials ............................... Phenotypic Alterations ...................... Inheritance ................................. Novobiocin Dosage Trials ..................2............ Phenotypic Alterations ...................... Inheritance ................................. Discussion.......................................... CHAPTER 3. CHEMICAL MUTAGENESIS OF THE PLASTOME.MUTATOR GENOTYPE Intrwuction 00.0.0.0....00...OOOOOOOOOOOOOOOOOOOOO Materials and Methods ............................. Results 9-aminoacridine hydrochloride ‘pmzpm IV genotype ............................ Acridine treatment of the mixed pm/pm, pm/+ line 00.0.0.0...OOOOOOOOOOOOOO0.0.0.0.... Nalidixic Acid Preliminary tests with inbred pm/pm line ..... Nalidixic acid treatment of the mixed pm/pm, pm/+ line OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO Novobiocin Preliminary tests with inbred pm/pm line ..... Novobiocin treatment of the mixed pm/pm, pm/+ line 00.0.0.0...0......OOOOOOOOOOOOOOOOOO V 49 49 50 59 59 60 64 68 72 72 74 82 82 88 88 DiscuSSionOOOO00....OOOOOOOOOOOOOOOOOOOO0....O..0... 92 CONCLUSIONOOOOOOO0.000000000000000...0.00.00.00.00.0... 96 LIST OF REFERENCESCOOOOOOOOOOOOOOOOOOCOO...0.......0... 101 vi LIST OF TABLES Chapter 1 Table 1. Sector frequency in seedlings treated with 5mM nitroso-mthYlureaOOOOOOOOOOOOOOOOOOOOIOOOOO0.0...0.0... 15 Table 2. Sector frequency in seedlings treated with 5mM nitroso—methylurea for 0.5 hours........................ 16 Table 3. Sector frequency in the control seedlings treated With ethanol for 0.5 hours.............................. 17 Chapter 2 Table 4. Sector frequencies for wt seedlings containing plastome II treated with varying doses of 9-aminoacridine hYdrwhlorideOOOOOOOOOOOOOOOOOOOOOOO0.00...0......0.... 34 Table 5. Sector frequencies for wt seedlings containing plastome II treated with varying doses of 9-aminoacridine hYdIOChlorideoooooooooooooooooooooooocoo...coo-coocoo-o 36 Table 6. Sector frequency for wt seedlings containing plastome IV treated with 2 ug/ml of 9-aminoacridine HYdrxhlorideOOOOO...OOOOOOOOOOOOOOOOOOCOOO0......0.... 37 Table 7. Sector frequency for wt seedlings containing plastome II treated with 9-aminoacridine hydrochloride. 37 Table 8. Sector frequency for wt seedlings containing plastome II treated with 9-aminoacridine hydrochloride. 38 Table 9. Sector frequency for wt seedlings containing plastome IV treated with 9-aminoacridine hydrochloride. 39 Table 10. Sector frequency for wt seedlings containing plastome I treated with 9-aminoacridine hydrochloride.. 39 Table 11. Crosses of 9-aminoacridine hydrochloride-induced mutants...0.0....OOOOOOOOOOIOOOOOOOOOOO0.0.0.0.....0... 44 vii Table 12. Self—pollination of 9-aminoacridine hydrochloride-induced mutants.......................... 48 Table 13. Sector frequency for wt seedlings containing plastome II treated with varying doses of nalidixic aCidOOOOOO0......OOOOOOOCOOOOOOOOOOOO0......00000000.... 52 Table 14. Sector frequency for wt seedlings containing plastome II treated with varying doses of nalidixic aCidOOOCOO0......OOOOCOOOOOOOOOOOOOCOOOO0..OOOOOOOCOOOOO 53 Table 15. Sector frequency for wt seedlings containing plastome II treated with varing doses of nalidixic acid. 54 Table 16. Sector frequency for wt seedlings containing plastome IV treated with 10 ug/ml nalidixic acid........ 55 Table 17. Sector frequency for wt seedlings containing plastome IV treated with nalidixic acid................. 56 Table 18. Sector frequency for wt seedlings containing plastom IVOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOCOOO 56 Table 19. Sector frequency for wt seedlings containing plastome I treated with nalidixic acid.................. 57 Table 20. Crosses of nalidixic acid-induced mutants...... 58 Table 21. Self-pollinations of nalidixic acid-induced mutantSOOOOOCIOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOCOOOOO 58 Table 22. Sector frequency for wt seedlings containing plastome II treated with varying doses of novobiocin.... 61 Table 23. Sector frequency for wt seedlings containing plastome IV treated with varying doses of novobiocin.... 62 Table 24. Sector frequency for wt seedlings containing PlaStom IVOOOOOOOOOOOOO0.0...OOOOOOOOOOOOOOOOOOOO0.0... 62 Table 25. Self-pollination of novobiocin-induced mutants. 63 Chapter 3 Table 26. Sector frequencies in pm/pm seedlings containing plastome II treated with varying doses of 9-aminoacridine hyarxhlorideOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO. 78 viii Table 27. Sector frequency for pm/pm seedlings containing plastome II treated with varying doses of 9-aminoacridine hyarocnlorideOOOOOOOOOOOOOOOOOOOIIOOOOOOOOOOOOOOOOOOOOOO 79 Table 28. Sector frequency for pm/pm seedlings containing plastome IV treated for 16 hours with 9-aminoacridine hyarxhloridQOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO 81 Table 29. Sector frequency for seedlings containing plastome I treated with 9-aminoacridine hydrochloride............ 81 Table 30. Sector frequency for pm/pm seedlings containing plastome II treated with varying doses of nalidixic aCidOOOOOO0.00.00.00.00.00000000000000000000.00.00.00.00 84 Table 31. Sector frequency for pm/pm seedlings containing plastome II treated with varying doses of nalidixic aCidCOOOOO0......0......OOOOOOOOOOOOOOOOCCO0.......00... 85 Table 32. Sector frequency for pm/pm seedlings containing plastome II treated with varying doses of nalicixic aCidOOOOOOOOOOOIO0.0.0.0000...0.0.0.0000...0......00.... 86 Table 33. Sector frequency for pm/pm,ypm/+ seedlings containing plastome I treated with nalidixic acid....... 87 Table 34. Sector frequency for pmzpm seedlings containing plastome II treated with varying doses of novobiocin.... 90 Table 35. Sector frequency for a mixed population of pm/pm and pm/+ seedlings containing plastome I treated with nov0bixin00000000000000OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO 91 LIST OF FIGURE Figure 1. Transmission electron micrograph of mottled tissue from a 9-aminoacridine hydrochloride-induced mutant.......... 41 INTRODUCTION Chloroplasts have a genome that codes for approximately 123 genes (Reviewed by Mullet 1988). The choroplast genome provides genetic information for components of the protein complexes present in the plastid and must rely on the nuclear genome to code for many of its components (Reviewed by Birky 1978 and Mullet 1988). The chloroplast depends on the nucleus to encode many gene products that are transported into the plastid. Should the nuclear DNA that codes for one of these gene products become mutated so that the gene product produced is damaged or lost it may severely disrupt chloroplast function. One such critical gene for the chloroplast genetic system is probably defined by the pm- locus of Oenothera. In mutagenesis experiments performed by Mel Epp, Oenothera hookeri seeds were treated with ethylmethyl-sulfonate (EMS). He observed a plant with several different chlorotic sectors in the M2 generation (Epp 1973). Through reciprocal crosses and self-pollinations with this plant, Epp demonstrated that the mutations were initially induced by a recessive nuclear allele. However, the non-Mendelian inheritance of the variegated tissue indicated that the mutations occur in the chloroplast genome (plastome). Once the chlorotic sector is induced, its expression is independent of the nuclear gene. This p1astome.mutator (pm) gene increases the frequency of spontaneous mutation in the plastome 200 - lOOO-fold (Epp 1973; Sears and Sokalski, 1991). The plastome mutator of oenothera is known to induce mutations that appear as sectors having pigment deficiences. Sectors range from.white to light green and may be mottled or solidly pigmented. These differences in pigmentation suggest that a variety of mutations are occurring, and are affecting a number of different loci. Studies by Blasko et al. (1988) and Chiu et al. (1990) have shown large deletions in chloroplast DNA (chNA) specific to the plastome mutator line. Two separate deletions studied by Blasko et al. (1988) occurred in a large open reading frame. Both pm—induced deletions were characterized from green, photosynthetically- competent plants. Chiu et a1. (1990) reported that pm causes deletions at five hotspots on the chNA molecule including the site characterized by Blasko et al. (1988). Deletions at these sites occurred in green isolates as well as mutant lines. Thus, these separate studies concluded that the large deletions did not directly cause the mutant phenotypes and postulated that other small deletions or point mutations were responsible for the chlorotic sectors. 3 Sears and Sokalski (1991) investigated the Oenothera plastome (mutator line by treating the seeds with nitroso methylurea (NMU), which has been shown to cause base substitutions in DNA (e.g., Richardson et al. 1987). They observed a synergistically high rate of mutation in the young seedlings. The interaction of the chemical mutagen and the plastome (mutator line in the production of chNA mutations points to the likelihood that there is a defect in the chNA repair system. However, these data are not entirely consistent with the previous observations that the plastome mutator elevates spontaneous deletion frequencies in the chNA, since such deletions would likely result from a defect in a different chNA repair pathway. 0n the other hand, perhaps several types of mutation would result from erroneous replication or repair mechanisms from a defect in: replication bypass, fidelity of the replication system, specificity of the damage detection and removal systems, and DNA breaking and joining reactions. If a mutator allele codes for a disfunctional protein (or no protein at all) that is a component of any of these systems, it is conceivable that an increase in several types of mutations would be seen. My goal was to further test the plastome mutator line by using the chemical mutagen, acridine, which induces DNA lesions different from those caused by NMU. If a synergistic response is again elicited, it would indicate that a defect 4 in a general chNA repair pathway is the most likely cause for the increased frequency of mutation induced by the ‘plastome mutator. Similar experiments were performed with the gyrase inhibitors, nalidixic acid and novobiocin. If a change in the pubinduced mutation frequencies is observed with those chemicals, it would indicate that supercoiling of the chNA helix was involved in the generation of mutations. Data collected from these investigations may allow determination of a role for the plastome mutator gene product in replication or repair of chNA. CHAPTER 1 NMU MUTAGENESIS AND SELECTION OF ANTIBIOTIC RESISTANCE SECTORS IN OENOTHERA INTRODUCTION Many of the plastome mutants of higher plants originated spontaneously or were induced by nuclear 'mutator' genes (Boerner and Sears 1986). The induction of mutations by chemical agents was originally investigated in studies of prokaryotic and eukaryotic systems, but was subsequently adapted to organelle genetic systems. Chemical mutagenesis with alkylating agents, particularly nitroso-compounds, has been the most effective means for inducing chloropast mutations in a wide range of plants, including Antirrhinum (Hagemann 1982), Nicotiana (Fluhr et al. 1985), Brassica (walters et al.1990) and Cenothera (Sears and Sokalski 1991). 6 One of the common characteristics of chemical mutagens is their ability to form.cova1ent products with DNA and thus interfere with the process of DNA replication and repair. Misreplication or misrepair of DNA can lead to incorporation of incorrect bases and the fixing of mutations in the DNA. The error produced by the chemical is either corrected or results in a permanent alteration in the DNA. The damage by alkylating agents consists of the addition of alkyl groups (usually methyl or ethyl groups) to reactive sites on the bases or phosphates of the DNA. The methylation causes atypical base pairing during replication or repair. Nitroso-compounds are well studied mutagens in many other genetic systems. Humans exposed to these types of chemicals exhibited liver damage and with prolonged dosage, hepatocellular carcinoma (Magee and Barnes 1956). Rats have shown chromosomal damage (carcinogenesis, chromosone aberration in somatic cells, polyploidy) after treatment (Monakhou et al. 1990). Genetic studies with bacteria and bacteriophages have shown that many point mutation induced by alkylating agents involve guanine to adenine transitions (Mutagenesis; Drake, Koch 1976; Loveless 1969; Richardson et al. 1987). One experiment showed that all of the NMU-induced mutations in Escherichia coli were GC:AT transitions and that of these transitions, 82% of the mutations occurred at the 7 middle guanine of the sequence 5'-GG(A or T)-3' (Richardson et a1. 1987). The transitions are a consequence of alterations in the normal base pairing during replication due to the methylation of guanine at the N-7 position which is the most reactive (Lawley and Brooks 1961; Nagata, Imamura, Saito, Fukui 1963) or the O-6 position (Castellani 1987). Other oxygens and nitrogens can be methylated on the remaining bases, but they are not as highly reactive as the guanine base. Prokaryotic and eukaryotic cells have repair systems to deal with damage to DNA. .All of the systems use enzymes to repair the damaged bases, which are normally eliminated, bypassed or repaired by the organism. In addition, to deal with alkylation damage, prokaryotes have the enzyme 06- methylguanine methyltransferase which recognizes the alkylated base in the DNA and removes the methyl group (Lawley 1961 and Castellani 1987). This changes the base to its original form. NMU has been used in Oenothera to study mutation frequency (Sears and Sokalski 1991) and in Brassica to recover a cytoplasmic marker mutation, maternally inherited variegation (walter et al. 1990). These mutational lesions were visualized as chlorotic sectors in the plant tissue. In Nicotiana (Fluhr et al. 1985) and other Solanaceae (McCabe et 8 al. 1989), the mutagen was used to induce antibiotic resistance. Antibiotic levels were used for selection that caused wild-type seedlings to exhibit chlorosis; the presence of antibiotic-resistant ribosomes in the chloroplast could be visualized as green sectors in the chlorotic background. Chloroplast ribosomes are similar to prokaryotic ribosomes, in terms of size and sensitivity to various drugs that inhibit protein synthesis. Many of the inhibitors of protein synthesis are familiar antibiotics, such as streptomycin, spectinomycin, and erythromycin. These antibiotics bind to sites on the ribosome that are defined by both the rRNAs and the ribosomal proteins. Thus, resistance to an antibiotic may occur by mutation in a gene for one of the rRNAs or for one of the polypeptides. All of the chloroplast rRNAs are encoded by the plastome; many of the ribosomal proteins are encoded in the nuclear genome while the remaining are encoded in the plastome (Shinozaki et al. 1986). Antibiotics, such as streptomycin, chloramphenicol, tetracycline, and erythromycin inhibit protein synthesis in prokaryotes by different mechanisms. Streptomycin interacts with the site containing the protein 812 in the 308 ribosomal subunit. This antibiotic interferes with the initiation of protein synthesis and also causes the mRNA to be misread by allowing different amino acids to be incorporated into the polypeptide (Galili et a1. 1989). Mutants in E. coli have 9 been obtained that show streptomycin resistance conferred by a C to U transition at position 912 of 16S rRNA (Montandon et al. 1986, Moazed and Noller 1987, Gauthier et al. 1988) and in Nicotiana there is a mutation in the 168 rRNA gene of the SPC23 line at position 860 (C to A) that confers streptomycin resistance (Btzold et al. 1987). The antibiotic spectinomycin inhibits translocation of peptidyl-tRNA from the A site to the P site during early rounds of peptide bond formation. Resistance to spectinomycin has been shown to be conferred by mutations in ribosomal protein SS and by a C to 0 transition at position 1,192 in the 3' major domain of 168 rRNA (Moazed and Noller 1987). In Nicotiana, the SPCl line had an A to U transition at position 1138 and the SPC2 line had a C to U transition at position 1139 both of which caused spectinomycin resistance and disrupted a conserved 16S rRNA stem.structure (Svab and Maliga 1991). The SPC23 line had a mutation in the 16S rRNA at position 1333 (G to A) that caused spectinomycin resistance (Svab and Maliga 1991). The investigation described in this chapter was designed to determine whether mutations caused by NMU could confer antibiotic resistance to Genothera seedlings via mutations in the chloroplast genome. This experiment was modeled after that of Fluhr et al. (1985), in which antibiotic-resistant 10 mutants were induced and selected in Nicotiana. NMU has been used to successfully induce chlorotic plastome mutations in Oenothera (Sears and Sokalski 1991). My goal was to determine if NMU could also induce mutations that would result in antibiotic-resistance. Such mutants in Oenothera could be useful for inheritance studies and could be utilized for investigating aspects of the plastome mutator gene which are under intense study at the present time in our laboratory. In my investigation, seeds of Oenothera were treated with the mutagen and placed on media containing an antibiotic. Because the chloroplasts are prokaryotic and their ribosomes are sensitive to the antibiotics, the plant tissue bleaches out or becomes chlorophyll-deficient under these circumstances due to the loss of the chloroplast's ability to translate RNA messages. If NMU induced mutations that caused the chloroplast ribosomes to become resistant to the antibiotic, green islands of plant tissue would appear on the bleached plant. This experiment has been successful in Nicotiana, where NMU mutagenesis of seeds results in a very high incidence of mutations to photosynthetic incompetence, and a definite (but low) frequency of streptomycin-resistant sectors on the seedlings' true leaves (Fluhr et al. 1985). 11 MATERIALS AND METHODS Plant material. Homozygous plants of Oenothera hookeri strain Johansen containing plastome type II were constructed by Professor W. Stubbe (University of Duesseldorf). These lines were maintained by self-pollination. Nitrosomethyl urea mutagenesis. Fungal contamination had to be controlled before proceeding with the experiment. Several modifications were incorporated: 1) placing the seeds in a sucrose solution to encourage fungal spore germination followed by daily surface sterilization with 50% bleach of the seeds, 2) using a 10% bleach solution on the germinated seedlings and 3) up to three daily surface sterilizations ‘with 50% bleach solution before seed germination. The final method for the reduction of the fungus included an initial alcohol and flame sterilization of the seed capsule, followed by the three surface sterilizations. After a four hour imbibition, seeds were surface-sterilized by incubation in a solution of 50% bleach and 0.1% SDS for 30 minutes, followed by one rinse with 0.01 N HCl and several rinses with sterile water. Nitroso methylurea (Sigma) was added at concentrations of 5-20 mM for 30 minutes. Since the NMU stock was dissolved in ethanol, the control seeds were soaked in 6.25% ethanol for a corresponding time period. This was followed by several rinses with sterile water before placing the seeds on seed media where they were allowed to 12 germinate. There were 50 seeds placed on each petri plate. When the second true leaves started to appear, approximately 2 weeks after germination, the seedlings were transferred to seed media containing streptomycin (400ug/ml), spectinomycin (10ug/ml) or lacking antibiotics. This should have allowed adequate time for mutations to sort out, since chlorotic secters can begin to be recognized 2-weeks post-NMU mutagenesis. Seed media. Seed media was according to Nagata and Takebe (1971) with the following modifications: 0.5 x concentration of the major elements, 2% sucrose, and buffered with Hcl. .Maintenance and scoring of plants. The petri plates containing the seed media were sealed with parafilm, placed under broad spectrum lights(GE F40/PL/AQ), and checked periodically for seedlings with chlorotic sectors. RESULTS Optimization of seed sterilization procedures. In the experiments performed with NMU, one-half of the seeds received a mock treatment with ethanol and were placed on three types of seed media: two types contained different antibiotics and one lacked antibiotics. The other half of the seeds were treated with the chemical, NMU, and also placed on the three media. In the first experiments the 13 fungal contamination was so pervasive that all seedlings were lost. Contamination rates ranged from 23.3% to 98%, comparisons between the control and the treated seedlings were not possible due to the extensive loss of some trials. Therefore, these experimental data are not included here. As indicated in the materials and methods, three protocols were evaluated for control of fungal contamination. It was determined that the number of surface sterilizations was negatively associated with contamination. The final method for the reduction of the fungus included an initial alcohol and flame sterilization of the seed capsule, followed by the three surface sterilizations of the seeds. NMU.mutagenesis and selection f0r antibiotic resistance. Oenothera seeds were treated with NMU as described in the Materials and Methods section. After treatment, seedlings were placed on media containing the antibiotics streptomycin (400 ug/mfl) or spectinomycin (10 ug/ml) and were allowed to grow. TA dosage of 5mM NMU and the exposure time of 30 minutes was taken from.previous trials, to optimize germination and seedling viability. The seedlings were scored by examining the true leaves for green islands amidst bleached tissue. The plants on antibiotic media were bleached after approximately 25 days on 14 the media. Therefore, the scoring started at 25 days and continued for 60 days or until the seedlings were bleached completely and/or dead. It was observed that seedlings on streptomycin plates bleach slower than those on the spectinomycin plates. Data on the effect of the mutagen on seed germination, viability, and green sectors are presented in Tables 1, 2 and 3. .Although the treatment resulted in a high frequency of chlorotic sectors in the M1 seedlings on soil (data not shown), I recovered no bleached plants with persistent green sectors on the antibiotics. 15 Table 1. Sector frequency in seedlings treated with 5mM nitroso-methyl urea for 0.5 hours. 50 wild-type seeds. Each trial contained Seedling Leaf Coloration Germination Viability Green Bleached Bleached N % N % N % with - green sectors 50(100%) 49(98.0%) 49(100%) 0 0 50(100%) 46(92.0%) 0 0 46(100%) 50(100%) 29(58.0%) 0 0 29(100%) 1 400 ug/ml Streptomycin 2 10 ug/ml Spectinomycin Table 2. 16 Sector frequency in seedings treated with 5mM nitroso-methyl urea for 0.5 hours. Each trial contained 550 wild-type seeds, except for the control which contained 549 seeds. Germination N % Viability N % Green N % Bleached 'with green sectors Bleached N % 549(100%) 200 200 o 0 (36.4%) (100%) w/strep1 550(100%) 175 0 0 175 (31.8%) (100%) 550(100%) 129 0 o 129 (23.5%) (100%) 1 400 ug/ml Streptomycin 2 10 ug/ml Spectinomycin 17 Table 3. Sector frequency in the control seedlings treated with ethanol for 0.5 hours. Each trial contained 50 wild- type seeds. The plates containing the control and the streptomycin seeds were lost to contamination. —-- —= I: Seedling Leaf Coloration Seed Germination Viability Green Bleached Bleached Media N % N % with N % green sectors . control 50(100%) 0 0 0 0 lw/ strep1 50(100%) 0 0 0 0 ' 50(100%) 1 400 ug/ml Streptomycin 2 10 ug/ml Spectinomycin 18 DISCUSSION Experiments with Nicotiana have indicated that NMU induces plastome-encoded antibiotic resistant mutations, as well as chlorophyll-deficient mutations (Fluhr et al. 1985). In a previous study of NMU mutagenesis with Oenothera, sectors were recovered in up to 66% of the seedlings. My experiments were designed to test if a subset of mutations induced by NMU could cause the ribosomes of the Oenothera chloroplast to become antibiotic resistant. It is known that NMU causes point mutations (Richardson et al. 1987), specifically transitions. Such a base substitution could change the nucleotide sequence of the rRNA or an amino acid in the sequence of a protein. Changes due to missense mutations are often minor; the protein with this damage will often fold up in its normal configuration or very close to it and retain its basic function. The significant effect concerning this change in amino acid sequence may be in the binding site for the antibiotics in the chloroplast ribosome, such that the antibiotic will not bind. Although I was able to reproduce a high mutation rate from the application of NMU, I did not recover any antibiotic- resistant mutants. Four possibilities could explain this failure: 1) an inadequate sample size; 2) NMU does not target the right base to confer streptomycin- or 19 spectinomyCin-resistance in Oenothera; 3) NMU does not cause base substitutions in Oenothera but causes other types of mutations, such as deletions; and 4) antibiotic binding sites of the Oenothera chloroplast ribosome are defined by many gene products, with an alteration in a single one not affecting antibiotic sensitivity. To overcome these possible problems, I would suggest the following modifications to the experimental procedure: 1) a larger sample size, 2) trying a different mutagenic chemical that targets a different base in the DNA strand, 3) selection with different antibiotics. CHAPTER 2 CHEMICAL MUTAGENESIS USING ACRIDINE, NALIDIXIC ACID AND NOVOBIOCIN INTRODUCTION Three chemicals were chosen to test as mutagens of the chloroplast genetic system: acridine, nalidixic acid, and novobiocin. These agents have been used by others to induce mutations and to perturb gyrase of higher plants. Studies on the mutagenic effects of 9-aminoacridine hydrochloride have used barley (D'Amato 1950; Eherenberg 1956), Allium cepa (D'Amato 1952, 1954; Nuti-Ronchi and D'Amato 1961), Lycopersicum esculentum (Buitti and Ragazzini 1966), and Vicia faba (Michaelis and Rieger 1963). Nalidixic acid and novobiocin experiments with Solanum nigrum (Ye and Sayre 1990), pea (Lam and Chua 1987), and Nicotiania tabacum 20 21 (Heinhorst et al. 1985) were done to inhibit gyrase and to observe its effects on chloroplast transcription. Since the discovery of the plastome.mutator (pm) in Oenothera by Mel Epp in 1973, researchers have been trying to unravel its mode of operation. Currently it is suspected that the lplastome.mutator interferes with repair or replication of chloroplast DNA. In keeping with this line of thought, several chemicals that target DNA metabolism in other organisms were used, to test whether the chloroplast genetic system of Oenothera would be susceptible, resulting in the appearance of chlorotic sectors in the seedlings. The chemical mutagen, 9-aminoacridine hydrochloride, was first used on plants in 1950 by D'Amato, who treated barley seeds with it and saw a retardation of growth along with chlorosis. It is now known that this chemical intercalates between the stacked nitrogen bases at the core of the double helix (Nasim.and Brychcy 1979). The binding of 9-aminoacridine hydrochloride to DNA is non- covalent and its mutagenic action is vastly different than other acridines harboring bulky side-chains. 9-aminoacridine hydrochloride is a DNA-intercalating molecule that interacts preferentially with B-form DNA (Reviewed in Ferguson and Denny 1991; reviewed in Nasim and Brychcy 1979). Using x-ray 22 crystallography, Sakore et al. (1977; 1979) showed that 9- aminoacridine hydrochloride binds to the dinucleotide 5- iodocytidyl (3'-5') guanosine. The planar acridine molecules lie parallel to the plane of the base pairs. This intercalation of the molecule increases the length of the DNA helix and consequently reduces DNA twist at that site. This twist reduction unwinds the double-helix and could affect the processes of replication, repair and/or transcription. 9-aminoacridine hydrochloride mimics base pairs, causing the deletion or addition of a base upon replication. Previous experiments with acridine in Salmonella (Hoffman et al. 1989; McCoy et al. 1981) and E. coli (Thomas and McPhee 1985; Pons and Muller 1989; Gordon et al. 1991) indicate that acridines induce single base-pair duplications and deletions that result in frameshifts. 9-aminoacridine hydrochloride has also been shown to be a frameshift mutagen in the lambda bacteriophage (Pons 1984). However, Levin et al. (1984) saw that 9-aminoacridine hydrochloride did not induce deletion revertants for hisG428 by basepair substitution or small deletions of three to six basepairs. When 9-aminoacridine hydrochloride intercalates into the bacteriophage DNA it causes frameshifts, producing predominantly -1 frameshifts in runs of 3 or more identical G:C basepairs with 90% occurring at hot spots (Skopak and Hutchinson 1984). 23 Newton et al. (1972) synchronized chromosomal replication in E. coli and found that acridine-induced mutagenesis is associated with the DNA replication fork. Newton et al. also observed the same frequency of reversion in wt, recAr, and recBr strains. Similarly, Ferguson and MacPhee (1983) saw that in both recA+ and recli- strains of Salmonella typhimurium, 9- mminoacridine hydrochloride was an effective frameshift mutagen. Since the recA product is involved in both recombination and inducible error-prone repair these findings suggest that those processes are not involved in creating the mutations in response to the presence of 9-aminoacridine hydrochloride. Thomas and MacPhee (1985) included the lexA gene in their study, and concluded that it, too, does not impact the induction of mutations by 9-aminoacridine. Regarding DNA repair, Ferguson and Denny (1991, pg. 144) state that the DNA damage caused by 9-aminoacridine hydrochloride does not appear to trigger common DNA repair systems (Podger and Hall 1984). Thomas and MacPhee (1985) examined the effects of uvrbdependent repair. They found that in uvrB- E. coli, 9-aminoacridine hydrochloride induced reversion of one lacz frameshift mutation, but reduced the reversion frequency of a second lacz frameshift mutation. They proposed that 9-aminoacridine hydrochloride stimulates a uvr-dependent excision, which then plays a role in the introduction of the frameshift mutations. Isamoto et al. 24 (1985) found that 9-aminoacridine hydrochloride did not induce mutations in mtDNA nor in nuclear DNA in Saccharomyces cerevisiae. Skopek and Hutchinson (1984) showed that a prophage is more susceptible to 9-aminoacridine hydrochloride mutagenesis in a mutL host that was deficient in mismatch repair than in a wild-type host. Using Salmonella typhimurium Hoffman et al. (1989) demonstrated by the presence of His+ colonies and the lack of Trp+ colonies on selective media that 9-aminoacridine hydrochloride was ineffective as an inducer of genetic duplications when a hisC3076 uvrB- strain was reverted. In addition to determining if 9-aminoacridine hydrochloride could cause mutations in chNA, it was my goal to determine if distortions due to positive and negative supercoiling could be mutagenic. The rationale for this inquiry is that superhelicity influences many processes in repair and replication. Pommier et al. (1987) found that the 9- aminoacridine-induced DNA unwinding does not appear to inhibit mouse leukemia DNA topoisomerase II because 9- aminoacridine hydrochloride lacks bulky side chains. All bacterial and eukaryotic cells possess topoisomerases that alter the topology of DNA by introducing supercoils 25 (topoisomerase II) and/or removing supercoils (topoisomerase I). Type II topoisomerase is also called DNA gyrase, and its activity is ATP-dependent. Since chloroplast DNA (chNA) is a supercoiled molecule in vivo, for its replication to proceed, the DNA must be unwound at the replication fork by a helicase. The additional tension or supercoiling imposed on the chNA is released by the type I topoisomerase. To restore supercoiling to the closed, circular DNA molecule, the phosphodiester bond must be cleaved in both of the strands, the helix must be rewound, and closed again by repairing the broken bonds. This is the activity of the type II topoisomerases, represented by the bacterial gyrase (Reviewed by Cozzeralli 1977 and 1980). The chloroplast DNA gyrase resembles that of prokaryotes, which is composed of two subunits, A and B (Thompson and Mosig 1985; Lam and Chua 1987). Both subunits are required for the enzyme's activity. The A subunit contains a DNA binding site and mediates the ability to cut and rejoin double-stranded DNA. The B subunit binds ATP and is probably responsible for the introduction of negative supercoils into the double- stranded DNA (Lui and wang 1978; Cozzarelli 1990). DNA gyrase cleavage is highly site-specific (Sugino et al. 1977; Gellert et al. 1977). Analysis of sites showed that cutting between a TG doublet is common to nearly all gyrase 26 cleavages (Morrison and Cozzarelli 1979). When a gyrase cleaves a DNA strand it becomes covalently linked to a DNA phosphoryl group (weintraub 1985; Wang 1986). Gyrase is thought to remain bound to DNA at discrete locations with ATP bringing about a conformational change in the enzyme. Gyrase inhibitors interfere with both DNA replication and the initiation of transcription (Kubo et al. 1979). Because promoters are generally associated with palindromic DNA sequences it has been hypothesized that they loop out to form hairpin—like structures in supercoiled DNA (Gellert et al. 1976). In the absence of gyrase activity, supercoils cannot be generated, and hence the promoters would not be distinguishable. In contrast to these expectations, Thompson and Mosig (1987) found a C. reinhardtii chloroplast promoter activity stimulated by the DNA gyrase inhibitor novobiocin. Since novobiocin is known to reduce torsional stress in E. coli DNA (Thompson and Mosig 1985), they interpret their results to mean that their particular promoter is regulated by torsional stress in chNA. Since Gellert's discovery of gyrase in 1976, the enzyme has been manipulated in a series of experiments using both novobiocin and nalidixic acid in various organisms, including E. coli (Gellert et al. 1976 a 1977; Sugino et al. 1977 a 1978), Daucus carota (Ciarrocchi et al. 1985), C. reinhardtii 27 (Thompson and Mosig 1985), and Solanum nigrum (Ye and Sayre 1990). With nalidixic acid treatment, a reduction in chNA synthesis was observed in Euglena gracilis (Lyman 1967), N. tabacum (Heinhorst et al. 1985), and Solanum nigrum (Ye and Sayre 1990). Novobiocin and nalidixic acid have not been used as mutagens in other organisms, but since gyrase plays such a critical role in DNA metabolism, I have attempted to determine if gyrase inhibitors may affect the mutation rate. Gellert et al. (1976) found that novobiocin, an inhibitor of nucleic acid synthesis in vivo, is an inhibitor of DNA gyrase in vitro. Novobiocin interrupts ATP binding and may block ATP access without sharing binding sites (Cozzarelli 1990). When Chlamydomonas reinhardtii was treated with novobiocin (Thompson and Mosig 1984; Thompson and Mosig 1990) the superhelical tension in chloroplast DNA was reduced. Nalidixic acid targets the A subunit of DNA gyrase in E. coli. This is accomplished through interference with the breakage-and-reunion component by stabilizing the enzyme-DNA cleavage complex (Gellert 1981). It is also likely that nalidixic acid does not just inactivate its target protein but corrupts it, converting it into a poison (Hsiang et al. 1985). The physiological effect of the inactivation of gyrase is the inhibition of replicative DNA synthesis, as observed after treatment with nalidixic acid in Nicotiana 28 tabacum (Heinhorst et al. 1985) and E. coli (Gellert et al. 1976; Cozzarelli 1977; Radl 1990). Both chemicals, novobiocin and nalidixic acid (Gellert et al. 1977, 1976) have been shown to inhibit DNA gyrase in vitro in E.coli, and in vivo in Daucas carota (Ciarrocchi et al. 1985) and Chlamydomonas reinhardtii (Thompson and Mosig 1985, 1987). Thompson and Mosig (1985, 1987) saw changes in transcription with both agents. Novobiocin differentially affected transcripts; some increased and some decreased. Nalidixic acid caused many transcripts to become more 'abundant. Ye and Sayre (1990) found that both novobiocin and nalidixic acid reduced chNA content of Solanum nigrum suspension culture cells. Neither chemical inhibited or reduced nuclear DNA content in vivo, indicating that a chloroplast localized gyrase is targeted. .The objective of the following experiments was to determine whether novobiocin or nalidixic acid would increase the frequency of spontaneous plastome mutations seen in wild-type Oenothera. In addition, the use of acridine, novobiocin, and nalidixic acid as mutagens might provide information relevant to the processes that affect the chloroplast genetic system. 29 Specifically, I hoped to determine if supercoiling of the ch A was essential for replication fidelity in vivo, and whether acridine could induce mutations in the chloroplast genetic system. CHEMICAL TREATMENT Materials and methods Plant material. Homozygous plants of Oenothera hookeri strain Johansen containing plastome types I, II, and IV were constructed by Professor W. Stubbe (University of Duesseldorf). These lines were maintained by self- pollination. 9-aminoacridine hydrochloride mutagenesis. After a four hour imbibition, seeds were surface-sterilized by incubation in a solution of 50% bleach and 0.1% SDS for 30 minutes, followed by one rinse with 0.01 N HCl and several rinses with sterile water. To establish a dose curve, 9-aminoacridine hydrochloride (Sigma) was added at the specified final concentrations (see Results) for 30 minutes, 8 hours, 16 hours, and 32 hours. Since the 9-aminoacridine hydrochloride was dissolved in ddnzo, the control seeds were soaked in ddeo for a corresponding time period. This was followed by several rinses with sterile water before placing the seeds in a beaker of ddeo where they were allowed to germinate. 3O Nalidixic acid mutagenesis. Seeds were surface-sterilized and germinated as described above. Since the nalidixic acid was dissolved in ddHZO the control seeds were soaked in ddHZO for a corresponding time period. Novobiocin mutagenesis. Seeds were surface-sterilized and germinated as described under 9-aminoacridine hydrochloride mutagenesis. Since the novobiocin stock was dissolved in ddHZO the control seeds were soaked in ddHZO for a corresponding time period. Maintenance and scoring of plants. Following mutagenesis, seeds were surface-sterilized daily until they germinated. Seedlings from the chemical treatments were planted in multipots (Hummert Seed Co.), and the trays were placed under broad spectrum lights(GE F40/PL/AQ). Seedlings from crosses and self-pollinations were surface-sterilized for three days then placed on supplemental seed media. All seedlings were scored approximately every two weeks for chlorotic sectors until final scoring was completed at forty days after germination. Scoring consisted of checking the seedlings for chlorotic sectors in their true leaves. Seed media. The supplemental seed media is identical to the seed media mentioned in Chapter 1. 31 Crossing Strategies. Wild-type plants with chlorotic sectors were transplanted to pots and placed in the greenhouse. Reciprocal crosses were performed with control plants to examine the inheritance patterns of the mutations. Self— pollinations were performed to provide an M2 population to test for nuclear recessive mutation. To induce bolting, plants ranging from 2 1/2 months to 6 months in age were sprayed for 15 out of 20 days with a solution containing 10-4 M Gibberillic acid in 0.005% Tween. RESULTS 9-aminoacridine hydrochloride Dosage trials A dose curve was completed to determine the concentration and time interval that would result in the highest frequency of mutations. Six experiments were done to establish a dose curve (Tables 4 - 9). Wild-type (wt) and seeds homozygous for the plastome mutator allele (pm/pm ) were treated concurrently, up until Table 5, where it was feared that samples of the wt and pm/pm seed may have been switched in the 2 ug/md dosage at 0.5 hour treatment. At this treatment, the wt gave a high number of mutants while the treatment of pmlpm seeds (Chapter 3, Table 26) yeilded no plants with chlorotic sectors. After this time, experiments to determine the dosage curve for the wt seeds were undertaken separately from.the experiments with the pm/pm seeds. 2 ug/ml acridine «as and is s a St expc ug/n Sect cons as 1' trar CORE Iti DOt Plan 32 was chosen for further testing because the trials of Tables 1 and 2 suggested it might be the most mutagenic concentration. As shown in Table 6, increasing exposure times failed to show a steady increase in the mutation rate but the 16 hour exposure produced several mutations. As shown in Tables 4 - 10 , not every experiment produced mutants. Subsequently, 10 ug/ml acridine was tested on larger lots of seeds (Table 7). Sectors were observed in seeds treated for 16 hours. No consistent negative effect on seed germination was noted, but as indicated in the footnotes of these tables various transient developmental abnormalities were observed as a consequence of the acridine treatments. In the concentration chosen, (2 ug/ml), the total number of plants containing chlorotic sectors was 22 out of a total of 3402 viable seedlings, giving a mutation frequency of 0.65%. Mutations were not recovered from seeds exposed to greater concentrations. At the 16 hour time trial, a total of 651 seeds produced 12 plants with chlorotic sectors, with a 1.84% mutation rate. It is unknown why the increased exposure to the chemical did not appear to cause an increased level of lethality in the plants, nor a higher rate of mutagenesis. Tables 4 - 10. Dosage trials for 9-aminoacridine hydrochloride. 34 Table 4. Sector frequencies for wt seedlings containing plastome II treated with varying doses of 9-aminoacridine hydrochloride. Each trial contained 50 seeds. ug/mg Exposure Germination Viability :eedlings time with (hours) sectors 0 0.5 25(50.0%) l9(76.0%) o 0 a 35(70.0%) 25(71.4%)1 0 0 16 17(34.0%) 13(76.5%) o 0 32 32 64.0% 18 56.3% 0 1 0.5 37(74.0%) 31(83.8%) 0 1 a 39(78.0%) 31(79.5%)2 0 1 16 21(42.0%) 15(71.4%) 0 fl 1 32 1 2.0% 0 o 2 0.5 40(80.0%) 31(77.5%) 0 fl 2 a 43(86.0%) 39(90.7%) 2(5.1%) H 2 16 22(44.0%) 16(72.7%) 0 2 17 34.0% 11 64.7% 0 4 36(72.0%)7 31(86.1%)3 0 4 a 26(52.0%) 21(80.8%) 0 I 4 16 11(22.0%) 6(54.5%) 0 I 4 32 28(54.0%) 15(53.5%) 0 a 0.5 23(46.0%) 15(65.2%) 0 a a 36(72.0%) 26(72.2%) o a 16 28(56.0%) 21(75.0%) 0 8 32 19 38.0% 13 68.4% 4 0 i 16 0.5 41(82.0%) 40(97.6%) o a I 16 a 38(76.0%) 36(94.7%)5 0 n 16 16 34(68.0%) 26(76.5%) 0 n 16 32 46 92.0% 22 47.3% 0 32 0.5 46(92.0%) 33(71.7%)6 0 32 e 40(80.0%) 38(95.0%)7 0 32 16 20(40.0%) 15(75.0%) o 32 32 28(56.0%) 23(82.1%)8 0 35 Table 4 (cont'd). The number of plants seen with developmental abnormalities (leaves growing together) : 1.2.4.7 one plant 3 two plants 3:5:6 three plants 36 Table 5. Sector frequencies for wt seedlings containing plastome II treated with varying doses of 9-aminoacridine hydrochloride. Each test involved 50 seeds. ug/ml Exposure Germination Viability Seedlings time with (hours) sectors 0 0.5 29(58.0%) 22(75.9%) 0 0 7.5 6(12.0%) 4(66.7%)1 0 0 16 16(32.0%) 7(43.8%) 0 0 42__ 24 48.0% 19 79.2% 0 1 0.5 18(36.0%) l4(77.8%) 0 l 7.5 28(56.0%) 25(89.3%) 0 l 16 14(28.0%) 6(42.9%) 0 42 31 62.0% 31 100% ‘ 0 2 0.5 17(34.0%) 17(100%) 7(41.2%) 2 7.5 11(22.0%) 11(100%) 0 2 16 7(14.0%) 1(14.3%) 0 2 42 28g56.0%g 24g85.7%! 0 4 0.5 32(64.0%) 32(100%) 0 4 8 18(36.0%) 18(100%) 0 4 16 24(48.0%) 13(54.2%) 0 4 11 22.0% 7 63.6% 0 8 0.5 32(64.0%) 32(100%) 0 8 8 23(46.0%) 20(87%) 0 8 16 2(4.0%) 0 0 8 42 6(12.0%) 3g50.0g 0 16 0.5 32(64.0%) 32(100%) 0 16 8 18(36.0%) l4(77.8%) 0 16 16 5(10.0%) 0 0 16 42 23 46.0% 20 87.0% 0 32 0.5 20(40.0%) 17(85.0%) 0 32 8 16(32.0%) 14(87.5%) 0 32 16 8(16.0%) 0 0 1 32 42 =29(58.0%) 4216(89.7%) 0 1 one plant with abnormal leaf development. 37 Table 6. Sector frequency for wt seedlings containing plastome IV treated with 2 ug/ml of 9-aminoacridine hydrochloride. Each trial contained 100 seeds, except for the 16 hour trial which contained 200 seeds. =l= Germination Viability Seedlings with sectors 60(60.0%) 60(100%) 1(1.7%) “ 74(74.0%) 69(93.2%) 0 fl 128(64.0%) 122(95.3%) 4(3.3%) “ 53(53.0%) 50(94.3%) 0 | Table 7. Sector frequency for wt seedlings containing plastome II treated with 9-aminoacridine hydrochloride. Each trial contained 500 seeds, except the 1 hour trial, which contained 512 seeds. Germination Viability Seedlings - with sectors 209(41.8%) 84(40.2%) 10 1 512(100%) 286(55.9%)1 0 10 6.5 363(72.6%) 194(53.4%)2 0 10 16.5 466(93.2%) 318(68.2%)3 3(0.9%) 10 32 403(80.6%) 239(59%.3)4 0 baa—um“ 1 ten stunted plants and one plant with abnormal leaf development, . 2 three plants with abnormal leaf development, 3 three stunted plants, one plant with abnormal leaf development, one leaf fusion, 4 seven stunted plants, two plants with abnormal leaf development, one plant with leaves fused midrib, five plants with two apical meristems. Table 8. 38 Sector frequency for wt seedlings containing plastome II treated with 9-aminoacridine hydrochloride. Each trial contained 500 seeds. 6 ug/ml Exposure Germination Viability Seedlings time vdth (hours) sectors 0 32 460(92.0%) 1294(63.9%) 0 2 1 452(90.4%) 2379(83.8%) 0 2 6 451(90.2%) 376(83.3%) 0 2 16.5 376(75.2%) 3352(93.6%) 3(0.9%) 2 32 463(92.6%) 399(86.2%) O = = = L 1 Lost 26 plants to tray drying out, one plant with abnormal leaf, 2 3 three plants with abnormal leaf development, one plant with double meristem, one stunted plant, one plant with abnormal leaf development and same plant as two apical meristems, one plant without trichomes on one-half of leaf margin. 39 Table 9. Sector frequency for wt seedlings containing plastome IV treated with 9-aminoacridine hydrochloride. Each trial contained 500 seeds with an exposure time of 16 hours. (ug/ml) Gsrminatia: Viability Seedlings with sectors 0 224(44.8%) 1120(53.6%) O 2 298(59.6%) 2160(53.7%) 0 $ 1 two plants with abnormal leaf development, 2 two plants with holes in the leaf. Table 10. Sector frequency for wt seedlings containing plastome I treated with 9-aminoacridine hydrochloride. Each trial contained 500 seeds with an exposure time of 16 hours. ug/ml Germination Viability Seedlings with sectors 0 245(49.0%) l34(54.6%) 0 2 236(47.2%) 1157(66.5%) _ -- r~—# 77~— , —~——~#— 1 three plants with abnormal shaped leaf margins. 40 Mutant Phenotypes resulting from acridine treatment This chemical induced a wide spectum of colors and patterns in Oenothera, in contrast to NMU, which causes predominantly solid white sectors (Chapter 1 in this thesis). In the acridine treatments, I observed white, yellow, and light green sectors in the plant tissue, both as solid sectors and mottled. Developmental anomalies were also observed. These ranged from abnormal leaf shapes/margins, leaves with fused midribs, leaves fused side-to-side, leaves with holes, and one plant without trichomes on half of its leaf margin. All appeared to be transitory and were never seen beyond the second set of true leaves. Two other aberrations present that were not transitory were stunted plants and plants with two apical meristems. Using transmission electron microscopy, mottled mutant tissue was sampled from a 9-aminoacridine hydrochloride-induced sector from.mutant APE. The micrographs provided by the Center for Electron Optics indicated that the plastids in the white tissue were very large with few grana (Figure 1). The green tissue within the mottled sector contained chloroplasts with an excessive amount of densely stacked grana. In both pictures the mitochondria appear normal. Progeny from crosses of mutants were germinated on supplemented media, but 41 64:: ...~... ‘4 J '5 Figure 1. Transmission electron micrograph of mottled tissue from a 9-aminoacridine hydrochloride-induced ' mutant sector. 42 died within 45 days of their transfer to soil, where they were maintained under continuous light. This suggests that the primary mutational lesion affects the function of the chloroplast, not the mitochondrion. Inheritance All wt plants with acridine-induced chlorotic sectors were placed in the greenhouse and sprayed with gibberellic acid. These plants were then self-pollinated and crossed to determine the inheritance pattern of the mutation. In Oenothera, chloroplasts are transmitted from both parents to the progeny (Kirk and Tilney-Bassett 1978), while mitochondria are transmitted only from the female parent (Brennicke and Schwemmle 1984). To follow patterns of non-Mendelian inheritance, it is important to know that in Oenothera, when a full periclinal chimera produces a flower, the germ line contains entirely mutant plastids. If this plant is chosen as the female parent, mutant plastids will predominate in this plant's progeny (Kirk and Tilney-Bassett 1978; Stubbe and Herrmann 1982). Unfortunately, the plants containing the new mutant sectors in these experiments never developed as complete periclinal chimeras. At some point in time, sectors did extend into the leaf margins of a few of the mutant plants so that a subset of the offspring received some mutant plastids. 43 This led to difficulty in establishing the mode of inheritance of the trait. Thus some uncertainty remains about whether the mutations were induced in the mitochondrial or chloroplast genome. However, non-Mendeliam inheritance and vegetative segregation could be demonstrated unequivocally. The variegated plants obtained from the experiment shown in Table 3 were not used for crosses. The progeny from these plants had a bleached phenotype, reminiscent of a particular type of plastome-genome incompatibility. PCR assays kindly performed by Lara Steben indicated that the bleached leaves contained plastome III, indicating that the seed lot provided had been contaminated with seeds containing this somewhat incompatible plastome. For analysis of the crosses, the entire seed capsule was germinated and resulting progeny scored. One plant, designated A-E, had one half of the leaf margin containing chlorotic tissue, and hence, the mutation. Since only half of the germ line would contain the mutation, when used as the female in a cross, approximately half of the progeny should inherit mutant plastids. In fact, when the mutation was carried by the maternal parent, (A-E), 103 out of 205 progeny contained sectors (See Table 11). There were 9 entirely white plants in which the mutation was inherited strictly Table 11 . 44 induced mutants. Crosses of 9-aminoacridine hydrochloride mm Parents1 # of seeds F1 F1 viable seedlings germinated with sectors White 1: Green 385 365(94.8%) 355(97.3) 12(3.38%) A-B(Il)3xIV4 M- 1 M—mutation-not in v-1 1 J leaf margin P- 6.0 green I White x Green 396 l49(37.6%) 149(10096) 70(46.S%) A-E(Il)xl M - 2 (I M-l/Z leaf V-68 margin (3-0 P-green Green :1 White 390 342(87.7%) 342(10096) 0 le-E(ll) M-green P-partial White 7: Green 100 3Z(32.0%) 32(100%) 19(59.4%) A-E(Il)le M - 2 M-1/2leaf margin v-17 P-green G-O White x Green 24 24(10096) 24(10096) l4( 96) A-E(ll)le M - 5 M-1 /Zleaf margin v-9 P-green 5-0 Green 7: White 109 100(91.7%) 100 (100%) O leA-EUI) M-green P-l /2Ieaf margin White )1 Green 33 30(90.9%) 30(10096) 15(50.0%) A-G(ll)xl M-4 M-partial V-l 1 li P-green G-O Green 7: White 358 49(1 3.69%) 49(10096) O le-G(ll) M-green P-1/2leaf margin 45 Table 11 (cont.) Parents1 : of seeds F1 F1 viable seedlings germinated With sectors White )1 Green 10 2(20%) 2(100%) 0 A-F(Il)xl M-poss. partial P-green White )1 Green 272 262(96.3%) 192(73.3%) 23(11.98%) A-F(Il)x| M-S M-partial , V-18 P-green G-O Green x White 200 176(88.0%) 176(10096) 20.14%) leA-F(ll) M-O M-green V-Z P-poss. partial G-O The flowers used as the maternal parent were deduced to contain varying amounts of mutant tissue, based on the abundance of the mutant tissue in the LII tissue layer of the subtending leaf. 1 M = maternal parent P = paternal parent. Notations about the leaf margin describe the prevalence of the mutant tissue as follows: 1/2 Leaf margin - chlorotic sector is contained in one—half of the leaf margin. Partial - chlorotic sector is contained in less than half of the leaf margin. 2 Seedlings with sectors: M - seedling with solely mutant plastids G -green progeny with no sectors V -variegated seedlings 3 Plants designation and roman numerals are the plastome types. '1 When only a Roman numeral is listed, it refers to the wild- type, green plastome type in a nuclear background of 0. johansen. 46 from the female. The 94 variegated plants could have inherited white plastids from the female and green plastids from the male or both types could have been contributed by the female. NO variegation was seen in any offspring from A- E descendants when the male carried the mutation. I observed similar results with plants A-B, A-F and A—G. When the mutation was carried by the maternal parent, variegation was seen in the progeny. The F1 generation from A-B gave a total of 11 plants with sectors and 1 entirely white plant; A-G gave 11 sectored plants with 4 completely white plants; and A-F gave 18 variegated plants with 5 totally white plants. Self-pollinations were done to determine whether recessive nuclear mutations had been induced by acridine. An attempt was made to self-pollinate flowers that did not contain the mutant tissue that had been noted previously, so that other traits could be observed among the progeny. However, plants AeA, A—C, A-D, and AFE had variegated offspring in the M2 generation. This ranged from 9.5% to 37.8% as shown in Table 12. There were no other apparent nuclear anomalies observed. If the mutation was carried in the Oenothera mitochondrial genome, which has been shown to be inherited solely from the 47 female (Brennicke and Schwemmle 1984), variegated plants should be observed only when the female carries the mutation and never when the male carries the mutation. In one instance, pollen contributed mutant plastids to the progeny. When the variegated plant A-F, was used as the pollen donor, I observed two of its progeny with sectors. 48 Table 12. Self-pollination of 9-aminoacridine hydrochloride induced mutants. Self- # of seeds M2 M2 viable seedlings pollinations germinated with sectors A-A 99 93(93.9%) 64(68.8%) 24(37.S%) 14-8 100 0 0 0 A-C 100 90(90.096) 70(77.78%) 16(22.86%) A-D 102 102(10096) 42(41.18%) 4(9.52%) A-E 317 262(82.6596) 37(14.1296) 14(37.84%) A-K 100 97(97.0%) 48(49.48%) 0 A-M 100 87(87.0%) 87(10096) O A-N 100 98(98.0%) 63(64.29%) 0 A-O 100 20(20.0%) 1400.096) 0 A-T 100 1 9(1 9.096) 1 8(94.7%) 0 49 Nalidixic acid Results Dosage trials Initial experiments tested concentrations of 1 - 100 ug/ml nalidixic acid for 0.5 - 24 hours exposures (Tables 13 - 16). As shown in Tables 13 and 16, nalidixic acid did not induce mutations in every experiment. The mutants that did appear occurred almost entirely in the 10 ug/ml concentration at various lengths of exposure. Subsequent experiments utilized more seeds to further test the exposure time (Tables 14 - 15). Based on these experiments, the dosage and exposure time chosen was 10 ug/ml for 16 hours. Mutant phenotypes resulting from nalidixic acid treatment The phenotypes of the M1 generation varied from white to pale green sectors with some mottling; in other plants aberrant leaf shapes were also seen. Many of the plants had leaves with developmental holes in them. These anomalies were seen in the first true leaves but never developed in any of the subsequent leaves, thus, they may have indicated transitory physiological damage, rather than mutations. Other developmental anomalies observed included aberrently shaped leaves, stunting, presence of two apical meristems and presence of thin, bleached leaves. 50 Inheritance In the concentration chosen, 10 ug/ml, a total of 1897 viable seedlings were examined, and 8 plants containing chlorotic sectors were observed. Wild-type sectored plants were transferred to the greenhouse and treated with gibberellic acid to induce bolting. The plants did not produce periclinal chimeric leaves but some of the mutant tissue did get to the leaf margin. Therefore, some mutant plastids could possibly be inherited by the progeny. Due to limitations on the flower viability, number of control plants, and the placement of the sector, no crosses were performed in which the pollen carried the mutation. When a mutation was visible in one leaf's margin, the next leaf up did not always contain the chlorotic tissue in its leaf margin. Out of 444 viable F3 plants from crosses of N-C mutant as the female parent, 27 plants inherited solely mutant plastids, with no variegated individuals noted. An additional 26 plants displayed small clustered true leaves. Two variegated M1 plants were selfed, N-C and N-E, and only one plant from the offspring exhibited variegation. 51 Tables 13 - 19. Dosage trials for nalidixic acid. 52 Table 13. Sector frequency for wt seedlings containing plastome II treated with varying doses of nalidixic acid. ug/ml Exposure # of seeds Germination1 Viability Seedlings time vdth (hours) sectors 0 0.5 50 25(50.0%) 22(88%) 0 o 6 47 10(21.3%) 3(30%) 0 _wfi__g__fing__k24 __“_;gL__#m_ 8 16.0%____ 2 25% o , 1 0.5 42 9(21.4%) 7(78%) 0 I 1 I 6 50 1(2.0%) I 0 I 0 I 1 I 24 50 1(2.0%) I o I o 10 0.5 so 15(30.0%) 14(93%)2 0 10 6 50 8(16.0%) 7(88%) 1(14%) 1o 24 50 32(64.0%) 27(84%) 0 100 0.5 50 5(10.0%) 5(100%) o 100 6 35 10(28.6%) 6(60%) 0 100 24 50 4(s.0%) 3(75%) 0 1 Unknown as to why germination was so poor. 2 two plants with leaf abnormalities. 53 Table 14. Sector frequency for wt seedlings containing plastome II treated with varying doses of nalidixic acid. ug/ml Exposure Seedlings time # of seeds Germination Viability with ( hours) sectors 0 0.5 50 22(44.0%) 21(95.5%) o 0 6 100 70(70%.0) 62(88.6%) 0 o 24 50 1 2.0% 1 100% 0 1 0.5 50 30(60.0%) 29(96.7%) 0 I 1 I 6 100 I 40(40.0%) 31(77.5%) 0 I i 7 14.0% 3 42.9% 0 18(36.0%) 18(100%) 5 I 10 6 100 I 52(52.0%) 43(82.7%) 0 I I 10 ________24_____”____5g1_____ 25 50.0% 18 72.0% 0 I 100 0.5 50 20(40.0%) 20(100%) 0 I 100 I 6 100 I 32(32.0%) I25(7e.1%)1I 0 I I 100 I 24 100 I 12(12.0%) I 12(100%) I 0 I 1 three stunted seedlings. 54 Table 15. Sector frequency for wt seedlings containing plastome II treated with varying doses of nalidixic acid. Each trial contained 50 seeds. with sectors 16(32.0%) 5(31.3%) 21(48.0%) 20(95.2%) 14(28.0%) 7(50.0%) ,_—___ ——_—__——__——____.—_—._—— 9(13.0%) 3(33.5;;. 16(32.0%) 12(75.0%) 6(12.0%) 3(50.0%) 11(22.0%) 24(48.0%) 27(54.0%) 9(91.s%) 13(75.0%) 25(92.6%) 5(10.0%) 2(40.0%) 100 6 16(32.0%) 14(37.5%) 100 24 12(24.0%) 3(25.0%) 55 Sector frequency for wt seedlings containing Table 16. plastome IV treated with 10 ug/ml nalidixic acid. Exposure # of seeds Germination Viability Seedlings time with (hours) sectors 0.5 100 57(57.0%) 55(96.5%) 0 6 100 55(55.0%) 50(90.9%) 2(4.0%) 25 95 58(61.0%) 56(96.6%) l(l.8%) Table 17. plastome IV treated with nalidixic acid. contained 500 seeds. 56 Sector frequency for wt seedlings containing Each trial _ ug/ml Exposure Germination Viability Seedlings time with (hours) sectors 0 32 209(41.8%) 84(40.2%) 0 2 1 287(57.4%) 179(62.4%) 0 2 6.5 373(74.6%) 157(42.1%)1 0 II 2 16.5 352(70.4%) 167(47.4%)2 0 2 345(69.0%) 122(35.4%)3 0 1 seven stunted plants, 2 four plants with longitudally cracked stems (all survived), two plants with abnormal leaves, six very stunted plants, 3 six plants with abnormal leaf development, eight stunted plants, three plants have double meristems. Table 18. Sector frequency for wt seedlings containing plastome IV. Each trial contained 500 seeds. Nalidixic Exposure Ghrmination ‘Viability Seedlings Acid time with (hours) sectors 0 32 451(90.2%) 413(91.6%) 0 10 1 449(89.8%) 214(47.7%)1 0 10 6.5 456(9l.2%) 331(72.6%)2 0 10 16.5 475(95.0%) 379(79.8%)3 1(0.3%) 10 32 449(89.8%) 374(83.3%)‘ 0 u=|==== = = 1 one abnormal leaf, two plants with aberrantly shaped leaves, 2 lost twelve seedlings due to tray drying, one abnormal leaf two stunted plants, one plant with aberrantly . shaped leaves, 3 three stunted plants, one plant aberrantly shaped leaves, 1 one plant with aberrantly shaped leaves. 57 Table 19. Sector frequency for wt seedlings containing plastome I treated with nalidixic acid. Each trial contains 500 seeds with an exposure time of 16 hours. with sectors 0 245(49.0%) 134(54.7%) 330(66.0%) 259(7a.5%)1 1 three plants with thin bleached leaves. Table 20 . 58 Crosses of nalidixic acid-induced mutants. seedfings germinated with sectors : White )1 Green 200 178(89.096) 140(78.796) 0 1 N-C(IV)le M-1/2 leaf margin P-green II White 1: Green 200 183(91.S96) 18300096) 27041596) N-C(|V)le M-Z7(1 0096) M-1/2 leaf V-O margin P-O P-green (26 green plants with small clustered true leaves) White 3: Green 293 159(54.396) 121(76.196) 0 N-C(IV)le M-partial P—green L— m Table 2 1 . mutants . Parents M2 M2 viable Self-pollinations of nalidixic acid-induced seedfings germinated with sectors N-CxN-C 16S 160(96.9796) 94(58.7596) 1(1.06 96) N-ExN-E 100 S4( 54.096) 45(83396) 0 59 Novobiocin Results Dosage trials For novobiocin, two experiments were done to test for an effective dose, using increasing numbers of seeds (Tables 22 and 23). Three variegated plants were obtained from these experiments, two of which came from the 1,000 ug/md concentration. For these the largest time period was 32 hours, therefore, 1,000 ug/ml at 32 hours was chosen as the dose for a larger scale experiment (Table 24). Unfortunately, in the last experiment, seed germination was only about 50%, so the sample size was not as large as I would have liked. Phenotypic alterations resulting from novobiocin treatment Novobiocin-induced chlorotic sectors were quite small and pale green with some mottling in two plants while the third plant had white sectors. Double apical meristems, a swirled plant, ruffled leaves,and thin bleached leaves were some of the anomalies observed in the M1 seedlings. Using transmission electron microscopy a sample was removed from a small novobiocin-induced sector of the third plant. The micrographs provided by the Center for Electron Optics indicated that the cells were under physiological stress, since they were necrotic, had little cytoplasm, and had 60 vacuoles filled with debris. Many cells had cytoplasm that was very vacuolated. The chloroplasts had normal ultrastructure. However, the cells were not fixed well and thus the micrographs are of marginal quality. Therefore, the pictures are not included in this chapter. Inheritance Two of the three plants containing novobiocin-induced sectors were transferred to the greenhouse and treated with gibberellic acid. Crossing plants which contained sectors proved impossible because the sectors eventually disappeared. One plant, No-B, was sterile: when the No-B plant was used as the female, capsules didn't develop and pollen was not produced by the anthers. The other plant, No-A, never produced sectors in the margin of leaves and, thus, mutations could not be transferred to the progeny through the germ layer. Therefore, only self-pollinations of No-A were available to test for the presence of recessive nuclear mutations, and no variegated progeny resulted (Table 25). 61 Table 22. Sector frequency for wt seedlings containing plastome II treated with varying doses of novobiocin. Each trial contained 50 seeds. ug/ml Exposure # of seeds Germina- Viability Seedlings time tion with (hours) sectors 0 0.5 50 35(70.0%) 26(74.3%) 0 o 1 so 44(88.0%) 44(100%) o o 24 50 45(90.0%) 40(88.9%) 0 o 32 50 40 80.0% 35 87.5% o 10 0.5 50 29(58.0%) 28(96.6%) o 10 1 50 45(90.0%) 36(80.0%) o 10 24 51 51(100%) 40(78.4%) o 10 32 50 13 26.0% 9 69.2% 0 100 0.5 50 41(82.0%) 38(92.7%) 0 100 1 50 27(54.0%) 127(100%) 1(3.7%) 100 24 50 15(30.0%) 212(80.0%) 0 100 32 50 39 78.0% 30 76.9% 0 1000 0.5 50 28(56.0%) 324(85.7%) 0 1000 1 50 11(22.0%) ‘10(90.9%) 0 1000 24 50 47(94.0%) 43(91.5%) o 1000 32 50 39(78.0%) 529(74.4%) 1(3.4%) ==================================================================== WOUND-0 two one two one one plants with two apical meristems, plant with two apical meristem, plants with two apical meristem, plant appears swirled/deformed, plant with leaves ruffled. 62 Table 23. Sector frequency for wt seedlings containing plastome IV treated with varying doses of novobiocin. Each trial contains 100 seeds with an exposure time of 32 hours. .UNH __ ug/ml Germination Viability Seedlings with sectors 0 80(80.0%) 76(95.0%)1 0 10 98(98.0%) 89(90.8%)2 0 100 92(92.0%) 89(96.7%)3 1(1.1%) 1000 93(93.0%) 78(83.9%)‘ 0 m m two plants with thin bleached leaves, two plants with thin bleached leaves, nine plants with thin bleached leaves, one plant with thin bleached leaves. Table 24. Sector frequency for wt seedlings containing plastome IV. time of 32 hours. Novobiocin ug/md Germination Viability Each trial contains 500 seeds with an exposure Seedlings with sectors 245(49.0%) 134(54.70%) 0 242(48.4%) 1 204 (84.30%)1 0 seven plants with thin bleached leaves. *The control, 0 ug/ml, is the same as the acridine experiment dated 5/18/93. 63 Table 25. Self-pollinations of novoviocin induced mutants. : Parents # of seeds M2 germinated M2 viable seedlings \ thh i sectors No-AxNo—A 100 99(99.0%) 95(95.996) 0 64 Discussion Several chemicals were used in an attempt to induce plastome mutations in Oenothera. The chemicals chosen were acridine (because it intercalates into DNA), novobiocin and nalidixic acid (because both chemicals inhibit gyrase). The seedlings were screened for mutations that would cause a visible phenotype, chlorosis, appearing in sectors of the M1 plants. Treatment of Oenothera seeds with these chemicals resulted in some temporary morphological and physiological abnormalities in some seedlings, such as fused leaves, holes in leaves, missing trichomes on leaf margin, etc. This points to the chemical being toxic to the plant cell. Mutagenic compounds damage DNA, but some also damage other macromolecules (RNA or proteins) that are present in the cell during the treatment. This damage may cause aberrations but most of these would be transient (Ferguson and Denny 1991). Conceivably, the mottled phenotype could be attributed to a leaky mutation or could represent a series of back mutations. Epp (1973), Stubbe and Herrmann (1982), and Sears and Herrmann (1985) attempted to select green progeny from mottled plastome mutants and were not successful, suggesting that mottling is not produced by back mutations. 65 Few pigment-deficient sectors were observed as a consequence of nalidixic acid and novobiocin. This may be attributed to any of several possibilities: 1) the chemicals not entering the plastid, 2) the gyrase binding the chemicals, but not becoming disfunctional, 3) gyrase being inhibited by the chemicals, but this not being mutagenic, as inferred in Solanum nigrum, where cell growth recovers after removal of the inhibitors (Ye and Sayre 1990), or 4) the existence of a novobiocin-insensitive bypass replication in the Oenothera chloroplast as seen in Chlamydomonas reinhardtii (woelfle et al. 1993). Our studies show that acridine can induce mutations that affect chlorophyll accumulation. Most sectors induced by acridine are mottled pale green and appear approximately 40 days after treatment. In contrast, NMU sectors are white and the majority can be recognized between 25 and 30 days after treatment. This may indicate that a different type of DNA lesion has been produced by acridine, than nitroso methylurea (NMU), which targets the plastome (Sears and Sokolski 1991). One possibility is that acridine may target mtDNA rather than CpDNA . Precedence exists for mitochondrial mutations affecting chloroplast function. A mitochondrial mutation that indirectly cause chloroplasts to become disfunctional has 66 been characterized in Zea mays. The nonchromosomal striped 2 (NCSZ) mutant of maize has a DNA rearrangement in the mitochondrial genome, whose abundance correlates with the severity of the albinism (Roussell et al. 1991; Hunt and Newton 1991), but no chloroplast DNA alterations were detected (Roussell et al. 1991). It appears that abnormalities in the mitochondrial genome can result in pleiotropic effects upon the chloroplast. In another investigation, severe stunting and striping in the nonchromosomal stripe 3 (NCS3) mutant was seen, but it was not determined if the chloroplast genome was altered (Hunt and Newton 1991). In Oenothera, the crossing data should allow us to distinguish between chloroplast and mitochondrial mutations. The sectors are certainly non-Mendelian because vegetative segregation is observed in the progeny. In Oenothera, chloroplasts are inherited from.both parents in most crosses (reviewed by Chiu and Sears 1993). In contrast, mitochondria are thought to be inherited solely from the maternal parent (Brennicke and Schwemmle 1984), based on the observation that restriction fragments of mitochondrial DNA in a number of hybrids have only the maternal mtDNA type. Since this investigation was not very rigorous, the occurrence of strictly maternal inheritance of mitochondria in Oenothera should be viewed with caution. However, if it is true that 67 the mitochondria are only maternally transmitted then my crosses point to a plastome-location of the acridine mutations because mutant A—F produced two variegated progeny when the male carried the mutation (Table 8). In conclusion, evidence obtained from the crossing data and from chemical treatment data, suggests that acridine can act as a plastome mutagen. CHAPTER 3 CHEMICAL MUTAGENESIS OF THE PLASTOME MUTATOR GENOTYPE INTRODUCTION Plastome mutator alleles that result in mutations in the chNA are present in a variety of plants. These mutator genes appear to differ in the types of mutations induced and the mode of induction. For example, in maize and barley they cause the same non-Mendelian albino defect to be expressed over and over again, while the plastome mutator of Oenothera (Epp 1973; Epp and Parthasarthy 1987), Arabidopsis (Redei and Plurad 1973), and petunia (Potrykus 1970) cause a variety of plastid mutations. Crossing experiments verified that the plastome mutator trait found in Oenothera was a nuclear recessive allele (Epp 1973). Since the nucleus probably encodes most of the components of 68 69 the replication and repair processes necessary for chloroplast DNA (chNA), it is highly probable that one ofthese products is absent in the pm lines. Studies of Sears and Sokalski (1990) using the Oenothera {plastome mutator line treated with nitroso methylurea (NMU), reported numerous chlorophyll-deficient sectors in the young seedlings. The dramatically high mutation rate that they observed was interpreted to indicate that the two components, the chemical mutagen and the plastome mutator gene, interact synergistically. It appeared that the guanosine alkylation incurred by the NMU is not being repaired and suggests the the pm line has a defect in the alkylation-damage repair pathway. This interpretation seems to be in conflict with the finding that the plastome mutator causes deletions of short direct repeats in the chNA (Blasko et al. 1988; Chiu et al. 1990). Such a consequence would not be expected to result from the defective repair of alkylated bases. Rather, the occurrence of high frequency deletions involving short direct repeats would point to a process involving recombination or replication slippage. Conceivably, the pma defect lies in a general aspect of chNA replication or repair that could increase the frequency of many sorts of mutational lesions. For example, if the exonuclease subunit of DNA polymerase was defective or absent, the proof-reading function would be missing. Alternatively, an elevated level 70 of error-prone repair could result in the fixation of many types of mutation. Another possibility is that an accessory enzyme for replication and repair, such as a helicase or topoisomerase is missing, causing stalling, replication slippage, and difficulties in making the DNA accessible for repair enzymes. In order to elucidate which aspect of chNA metabolism.is affected by the plastome mutator, it would be helpful to test the integrity of both the replication and repair processes in Oenothera. Thus, I chose to test for synergism of the plastome mutator genotype and a mutagen that causes a type of mutation that is quite different from those caused by NMU. Furthermore, I attempted to use inhibitors of gyrase to alter the extent of supercoiling of the chNA, to determine if that would affect the ability of the plastome mutator to induce mutations. The mutagen 9-aminoacridine hydrochloride was chosen because it is known to cause frame-shifts in Salmonella (Hoffman et al. 1989) and E. coli (Thomas and McPhee 1985; Gordon et al. 1991). Acridine intercalates between the stacked nitrogen bases at the core of the double helix (Nasim and Brychcy 1979). It mimics base pairs causing the deletion or addition of a base upon replication, with hotspots occurring in runs of G-C basepairs (Skopek and Hutchinson 1984). 71 An increase in torsional stress of the chNA could be responsible for a higher frequency of replication slippage in the pmbline compared to wild-type. Such a situation could result if a topoisomerase I is absent, or if the supercoiling activity of topoisomerase II was elevated. In bacteria, gyrase activity seems to be intimately involved in creating deletions between direct repeats, since such deletions are gyrA-dependent, but recA independent (Saing et al. 1988; Maiura-Masuda and Ikeda 1990). To test whether an imbalance between relaxation and supercoiling results in pmbinduced mutation, I attempted to inhibit gyrase activity. Novobiocin and nalidixic acid have been shown to inhibit DNA gyrase in vitro in E.coli (Gellert et al. 1977, 1976), and in vivo in Daucas carota (Ciarrocchi et al. 1985) and Chlamydomonas reinhardtii (Thompson and Mosig 1985, 1987) and these experiments have also shown a concurrent decrease in DNA replication. Ye and Sayre (1990) found that both novobiocin and nalidixic acid reduced Solanum nigrum chNA content and that neither chemical inhibited or reduced nuclear DNA content in vivo. These findings suggest that an inhibitor-sensitive DNA gyrase participates specifically in chNA replication. In vivo experiments using nalidixic acid and novobiocin have shown that these chemicals suppress DNA gyrase in the 72 chloroplast. If excessive supercoiling is responsible for pm mutations, then it is conceivable that the gyrase inhibitors would help relax the molecule and reduce the pm mutation frequency. My objective was to first use wildftype Oenothera plants and determine if certain mutagenic chemicals (acridine, novobiocin, or nalidixic acid) would affect the frequency of spontaneous mutations in the plastome (Chapter 2). Then the goal was to apply these chemicals to Oenothera plants homozygous for the pm gene and determine if a synergistic response is observed. Chemical Treatment Materials and Methods The Materials and Methods for the seeds homozygous for the pm gene are identical to the materials and methods used for the respective chemicals and the wt seeds (Chapter 2). Results 9-aminoacridine hydrochloride effects on plastome mutator lines. pm/pm IV genotype. Since Sears and Sokalski (1991) had found that the combination of the pm/pm genotype and plastome IV gave a 73 spontaneous mutation rate that was 10% of the other plastome types, this was the initial plastome-genome combination of choice. As a control, wt seeds containing the same plastome type as the pm seeds were also treated with the chemical for comparison (Chapter 2 and 3, Tables 4 a 26; 5 a 27; 9 a 28; 10 a 29). The first experiment with the pm/pm seeds [containing plastome IV (Table 9 a 28)] produced an enormous number of sectors even in the pm/pm control with no chemical mutagen (Table 28). With these high numbers it would be extremely difficult to distinguish between an additive or a synergistic induction of mutation. Lacking a homozygous {pm/pm line with a low mutation rate, I pursued inbred pm/pm lines with plastome II, while Dr. Sears performed the crosses necessary to produce an alternative pm/pm line with a low mutation rate. To obtain a pm/pm line with a low background mutation rate, seeds with a reestablished pm/pm nucleus were used. These seeds were produced by crossing a +[pm line as the female parent with a pm/pm line as the pollen parent, resulting in an equal mixture of pm/+ and pm/pm seeds. As Epp demonstrated (1973), such a newly restored pm/pm line has an initially low mutation rate, and maximally can be expected to reach 18.6%. 74 {pm/pm II genotype. The initial 9-aminoacridine hydrochloride dosage trials utilized stocks with plastome I (Chapter 2, Table 10), plastome II (Chapter 2, Tables 4, 5, 7, and 8), and plastome IV (Chapter 2, Tables 6 and 9) in the wild-type nuclear background. It was determined that an exposure to 2 ug/ml for 16 hours was optimal for the induction of mutations. Wt ' and pm/pm seeds were treated concurrently, up until Table 5 (Chapter 2), where it was feared that samples of the wt and {pm/pm seed may have been switched in the 2 ug/ml dosage at 0.5 hour treatment. At this treatment, the wt gave a high number of mutants while the pm/pm sample did not give any plants with chlorotic sectors (Table 26). After this time, experiments to determine the dosage curve for the wt seeds were undertaken separately from the experiments with the {pm/pm seeds. Hence the data in Tables 26 and 27 on pm/pm seedlings were generated concurrently with Tables 4 and 5 in Chapter 2, and were considered only in the establishment of the optimal dosage. .Mutagenesis of a mixed pmlpm,{pm/+ line. After the dosage tests, an experiment was done with wt seeds (Chapter 2, Table 10) for comparison with the new seeds (Table 29). .Almost a two-fold increase in mutations occurred when the pm line was treated with 9-aminoacridine hydrochloride. The data obtained from the wt plants (Table 75 10) and the pm plants can be compared to determine if 9- aminocacridine hydrochloride caused a synergistic increase in the pm mutation frequency. The data contained in Table 10 and 29 were analyzed by applying the chi-square goodness-of-fit test to the numbers of observed and expected mutations. The expected number of mutations is approximated by the formula (if the chemical mutagenesis and the genetic condition independently produce mutations): (cu) + 1/2(NP) where N is the population size of the pm/pm seeds, C is the frequency of mutations in the chemically treated wild-type plants, and P is the observed frequency of mutation of pmApm alone (without chemical treatment). Because each pm/pm plant will be scored for a single mutation, when in fact, mutations would be caused by both acridine and the plastome mutator, this value must be halved. Thus, the expected number of mutations in the acridine-treated seeds of +/pm and pm/pm genotype is calculated as: [(CN + 1/2(NP)] - 1/2[CP (1/2N)] = Expected. A correction factor, l/2[CP(1/2N)], is necessary because in half the population, (1/2N), of mixed seeds (those of the pm/pm genotype), both the chemical and the pm gene may cause mutation in the same plant. The likelihood that two 76 mutations occur in the same plant is the frequency of mutation seen in the chemically treated wild-type plant multiplied by the frequency of mutation in the pm/pm line (without chemical treatment). Scoring after 40 days showed an increase in mutation frequency in pm/pm plants as a result of 9-aminoacridine hydrochloride treatment, but the differences in the final mutation frequency of treated and control pm/pm plants was not statistically significant (0.05 < p < 0.10). 77 Tables 26 - 29. 9-aminoacridine hydrochloride treatment of pulp. seeds. 78 Table 26. Sector frequencies in pm/pm seedlings containing plastome II treated with varying doses of 9-aminoacridine hydrochloride. Each trial contained 50 seeds. a... ug/ml Exposure Germination Viability Seedlings time with (hours) sectors 0 0.5 31(62.0%) 16(51.6%) 12(75.0%) 0 8 37(74.0%) 21(56.8%) 11(52.4%) 0 16 30(60.0%) 14(46.7%) 6(42.9%) 0 32 19 38.0% 3 15.8% 1 33.3% 1 0.5 41(82.0%) 13(31.7%) 3(23.1%) 1 8 35(70.0%) 12(34.3%) 6(50.0%) 1 16 32(64.0%) 15(46.9%) 6(40.0%) 1 32 6 12.0% 0 o 2 0.5 34(68.0%) 11(32.4%) 6(54.5%) 2 8 35(70.0%) 9(25.7%) 3(33.3%) I 2 16 28(56.0%) 13(46.4%) 6(46.2%) I 2 32 10 20.0% 6 60.0% 4 66.7% 4 0.5 28(56.0%) 26(92.9%) l4(53.8%) 4 8 38(76.0%) 26(68.4%) l6(61.5%) 4 16 16(32.0%) 3(18.8%) 2(66.7%) g__~fljL_____ 32 11 22.0% 4 36.4%____ 3 75.0% 8 0.5 29(58.0%) 28(96.6%) 10(35.7%) 8 8 33(66.0%) 30(90.9%) 17(56.7%) ' 8 16 34(68.0%) 32(94.1%) 18(56.3%)I _‘ 8 ___~__ 32 __u_ 31 62.0% 24 77.4% 14 58.3% 16 0.5 28(56.0%) 22(78.6%) 13(59.1%) 16 8 33(66.0%) 29(87.9%) 9(31.0%) 16 16 28(56.0%) 23(82.1%) 15(65.2%) __fl~_1§______ 32_____ 27 54.0% 25 92.6% 16 64.0% i 32 0.5 31(62.0%) 13(41.9%) 6(46.2%) 32 8 32(64.0%) 23(71.9%) 12(52.2%) I I 32 16 39(78.0%) 28(71.8%) 17(60.7%) I 32 32 39(78.0%) 32(82.1%) 15(46.9%) .......== .14....== 79 Table 27. Sector frequency for pm/pm seedlings containing plastome II treated with varying doses of 9-aminoacridine hydrochloride. Each trial contained 50 seeds. ug/ml Exposure Germination Viability Seedlings time with (hours) sectors 0 0.5 11(22.0%) 10(90.9%) 5(50.0%) 0 7.5 27(54.0%) 8(29.6%)1 7(87.5%) 0 l6 20(40.0%) 18(90.0%) l(5.6%) 0 42 25 50.0% * 5 1 0.5 15(30.0%) 12(80.0%) 6(50.0%) I 1 7.5 26(52.0%) 26(100%) 18(69.2%) l 16 14(28.0%) 14(100%) 2(14.3) I 1 42 20 40.0% * 4 2 0.5 31(62.0%) 26(83.9%) 0 II 2 7.5 24(48.0%) 5(20.8%) 2(40.0%) I 2 16 9(18.0%) 7(77.8%) o I 0.5 11(22.0%) 11(100%) 4 4(36.4%) 4 8 21(42.0%) 21(100%) 7(33.3%) 4 16 9(18.0%) 8(88.9%)1 1(12.5%) 25 50.0% 8 23(46.0%) 16(69.6%) 4(25.0%) 8 8 17(34.0%) 17(100%) 6(35.3%) 8 16 0 0 0 II 8 21 42.0% 16 8 17(34.0%) 17(100%) 7(41.2%) 16 16 8(16.0%) 7(87.5%) 5(71.4%) __17 34.0% 12(24.0%) 10(83.3%) 32 8 20(40.0%) 19(95.0%) 4(21.1%) 32 16 1(2.0%) 1(100%) 0 I 32 42 18‘36.0%I * 0 u 80 Table 27 (cont'd). *These plant trays were not counted for the total number of viable plants, but were counted for the total number of sectors. 1 Severe bug damage to plants was responsible for low viability. 81 Table 28. Sector frequency for pm/pm seedlings containing plastome IV treated for 16 hours with 9-aminoacridine hydrochloride. = r ug/ml # of seeds Germination Viability Seedlings with sectors 558(100%) 337(60.4%) 175(51.9%) 476(95.2%) 342(71.8%) 193(56.4%) I ; _—___a Table 29. Sector frequency for seedlings containing plastome I treated with 9-aminoacridine hydrochloride. Seeds are mixed +[pm & pm/pm. Each trial contained 500 seeds that were treated for 16 hours. Germination Viability Seedlings with sectors 0 468(93.6%) 332(70.9%)1 l3(3.92%) 2 467(93.4%) 292(62.5%)a 22(7.53%) 1 40 plants with fused leaves, 5 stunted plants, 2 41 plants with fused leaves, 10 stunted plants. 82 Nalidixic acid effects on plastome mutator lines. Preliminary tests with inbred pm/pm lines. The initial nalidixic acid dosage trials utilized stocks with plastome I (Chapter 2, Table 19), plastome II (Chapter 2, Tables 13, l4,and 15), and plastome IV (Chapter 2, Tables 16, 17,and 18) in the wild-type nuclear background. Tables 30, 31, and 32 provide data on pm/pm seedlings, which were generated concurrently with Tables 13, 14, and 15 in Chapter 2. These data could not be used for the chi-square test due to size deficiencies in the chemical trials, and they were considered only in the establishment of the optimal dosage. It was determined that 10 ug/ml at 16 hours was optimal, in that occasional mutant sectors were observed at that concentration. Nalidixic acid treatment of the mixed pm/pm,‘pm/+ line. The data contained in Table 33 and Table 19 (Chapter 2) were analyzed by applying the chi-square goodness-of-fit test to determine if the mutation frequencies differed between the control and treated samples. Table 19 contained the data from the chemically treated wild-type seedlings that is necessary for the correction factor. Also, additional considerations, such as halving the population size, had to be made since the seed lot contained a mixture of seeds that were pm/+ and pm/pm (see above). 83 Scoring after 40 days showed a slight increase in mutation frequency in pm/pm plants as a result of the nalidixic acid treatment, but the differences in the final mutation frequency of treated and control pm/pm plants were not statistically significant (p > 0.5). Table 30. 84 Sector frequency for pm/pm seedlings containing plastome II treated with varying doses of nalidixic acid. Each trial contained 50 seeds. Seedlings with (hours) sectors 0 0.5 26(52.0%) 21(80.0%) 6(28.6%) 0 6 3(6.0%) 3(100%) 2(66.7%) 0 24 8(16.0%) 2(25.0%) 0(<50%) 1 0.5 7(14.0%) 7(100%) 3(42.9%) 1 6 2(4.0%) 0 0(<0%) l 24 3(6.0%) 1(33.3%) 0(<100%) '1_,____1 0.5M1(38. 177.54) (47.1 I 10 6 2(4.0%) 1(50.0%) 0(<100%) 37(74.0%) 26(70.3%) 13(50.0%) 16(32.0%) ll(68.8%) 6(54.5%) 100 6 30(60.0%) 21(70.0%) 10(47.6%) 100 45(90.0%) 40(88.9%) 14(35.0%) I _—— —--- -—--—-—— Table 31. 85 Sector frequency for pm/pm seedlings containing plastome II treated with varying doses of nalidixic acid. I E ug/ml Exposure # of seeds Germina— Viability Seedlings time tion with (hours) sectors 0 0.5 50 21(42.0%) 14(66.7%) 6(42.9%) 0 6 100 33(33.0%) 20(60.6%) 10(50.0%) 0 24 50 25(50.0%) 19(76.0%) 6(31.6%) 1 0.5 50 11(22.0%) 10(90.9%) 8(80.0%) l 6 100 34(34.0%) 27(79.4%)1 17(63.0%) l 24 50 9(18.0%) 9(100%) 6(66.7%) 10(20.0%) 9(18.0%) 6(60.0%) 4(66.7%) 10 6 50 32(44.0%) 31(96.9%) 18(58.1%) 10 24 50 12(24.0%) 8(66.7%) 5(62.5%) 8(8%8.9) 6(75.0%) I 100 6 100 24(48.0%) 21(87.5%) 11(52.4%) 100 24 100 51(51.0%) 47(92.2%) l7(36.2%) 1 two stunted plants 86 Table 32. Sector frequency for pm/pm seedlings containing plastome II treated with varying doses of nalidixic acid. Each trial contained 50 seeds. Exposure Germination Viability Seedlingsfi time with (hours) sectors 25(50.0%) 17(68.0%) 2(1l.8%) I 29(58.0%) 18(62.1%) 4(22.2%)1 15(30.0%) 12(80.0%) 4(33.3%) 1 0.5 15(30.0%) 15(100%) 5(33.3%) i 1 6 13(26.0%) 5(38.5%) 1(20.0%)1 1 24 15(30.0%) 13(86.7%) 5(38.5%) I _ _I __ii i_ __—:T_——_____n___'___; __ _._.___*_+i_______————._:——."__"_____—__—jJ 26(52.0%) 22(84.6%) 10(45.5%) 18(36.0%) 9(50.0%) 2(22.2%)1 11(22.0.%) 10(90.9%) 6(60.0%) i i I_ ____ __ _ ___ _ ____ _ _____ ________ 24(48.0%) 24(100%)_ 24(48.0%) 12(50.0%) 3(25.0%)1 19(38.0%) 13(68.4%) 2(15.4%) 1 .All seedlings had aberrantly shaped leaves. 2 .All seeds from the six hour treatment came from the same seed lot, which was different then the rest of the trials. 87 Table 33. Sector frequency for pm/pm, pm/+ seedlings containing plastome I treated with nalidixic acid. Each trial contained 500 seeds that were treated for 16 hours. ug/ml Gemination Viability Seedlings with sectors 468(93.60%) 332(70.94%) 13(3.92%)1 483(96.60%) 361(74.74%) 20(5.54%)2 1 40 plants with fused leaves and 5 stunted plants, 2 24 plants with fused leaves and 6 bleached plants with thin leafs. 88 Novobiocin effects on plastome mutator lines. Preliminary tests with inbred pm/pm lines. The initial novobiocin dosage trials utilized stocks with plastome II (Chapter 2, Table 22), and plastome IV (Chapter 2, Table 23 and 24) in the wild-type nuclear background. The data on pm/pm seedlings in Table 34 was generated concurrently with Table 22 in Chapter 2, and was considered only in the establishment of the optimal dosage. It was determined that 1000 ug/ml at 32 hours was necessary to obtain an impact on chloroplast genetic system, since transient sectors were occasionally observed after such a treatment . Novobiocin treatment of the mixed pm/pm, pm/+ line. The data contained in Table 35 and Table 24 (Chapter 2) were analyzed by applying the chi-square goodness-of-fit test to determine if the mutation frequencies differed between the control and the treated samples. Table 24 contains data from the wild-type seedlings necessary for the correction factor. Additional considerations, such as the mutation rate, had to be made since the seed lot contained a mixture of seeds that were pm/+ and pm/pm (see above). Scoring after 40 days showed an increase in mutation frequency in pm/pm plants when compared to wt plants as a result of novobiocin treatment and the differences in the 89 final mutation frequency of treated and control pm/pm plants were statistically significant (0.025 > p > 0.01). 90 Table 34. Sector frequency for pm/pm seedlings containing plastome II treated with varying doses of novobiocin. Each trial contained 50 seeds except for trial 0 ug/ml at 32 hours which contained seeds. ug/ml Exposure Germination Viability Seedlings ; time with I (hours) sectors 0 0.5 25(50.0%) 8(32.0%) 4(50.0%) ! 0 1 18(36.0%) 11(61.l%) 7(63.6%) i ' 0 24 22(44.0%) 9(40.9%) 2(22.2%) I 0 32 20(52.6%) 20(100%) 13(65.0%) I ,"716 77"".5" “44.01) “—734“ " I 10 1 25(50.0%) 8(32.0%) 2(25.0%) i 10 24 25(50.0%) 5(20.0%) 2(40.0%) I I 10 32 22(44.0%) 17(77.3%) 8(47.1%) ‘ 100 0.5 14(28.0%) 3(21.4%) 0(<33%) 100 1 27(54.0%) 12(44.4%) 6(50.0%) I 100 24 12(24.0%) 6(50.0%) 1(16.7%) ' . 100 32 12(24.0%) 1(8.3%) 0(<100%) I f ‘ 1000 ""15“” W‘I 1000 1 8(16.0%) 4(50.0%) 2(50.0%) I 1000 24 4 10(20.0%) 5(50.0%) 1(20.0%) ; 1000 32 28(56.0%) 3(10.7%) 2(66.7%) -——— E w- J Table 35. 91 Sector frequency for a mixed population of pm/pm and pm/+ seedlings containing plastome I treated with novobiocin. Each trial contained 500 seeds treated for 32 hours. ug/ml Germination Viability Seedlings win: sectors 0 468(93.60%) 332(70.94%) 13(3.92%)1 1000 465(93.00%) 222(47.74%) 15(6.76%)2 E 1 40 plants with fused leaves, 5 stunted plants. 2 9 plants with abnormal leaf development. 92 DISCUSSION Previous experiments with the plastome mutator gene of Oenothera have shown that the plastome.mutator activity is determined by a nuclear recessive allele. Thus, it is likely that one of the elements for the chloroplast DNA repair and/or replication processes, which are encoded in the nucleus, is lost due to a null mutation. Mutagenesis was used to test the plastome's repair and replication pathways, with chlorotic sectors identifying a newly mutated chloroplast. Acridine was used because it intercalates into the double helix causing frameshifts and addition/deletion events, and would verify the efficiency of the general repair processes in the plastome mutator system. The usage of novobiocin and nalidixic acid should interrupt the chloroplast DNA gyrase, and would thus test another element of the replication pathway. Even though the number of visible sectors in wild-type Oenothera was low after treatment with the chemicals nalidixic acid and novobiocin, I still proceeded with the treatment of Oenothera plants homozygous for the plastome mutator gene for the following reason: if DNA damage was caused by these two chemicals, it may have been efficiently repaired in the wild-type plant lines, thus, giving a low mutation frequency. However, the plastome mutator defect 93 might render susceptible the chloroplasts of pmeplants. Other simultaneous investigations performed by Lara Steben (personal communication) using the restored plastome mutator (pm/pm and pm/+) line found late developing sectors when the plants were put into the field, consistent with the observations of Epp (1973). Conceivably, plastome mutator activity requires a great deal of dilution of the Emu product present in the egg cell. If this was the case, then the embryos exposed to chemical mutagenesis in the seed may not yet have possessed plastome mutator activity. However, one would expect that with the number of cell divisions that occur to produce the embryo, the pm+ product would be adequately diluted such that the chloroplast genetic system would be susceptible to mutation. In any case, this is the only reasonable recourse to eliminate the tremendous starting mutation rate seen in seeds produced by pm/pm self- pollinations. Scoring the seedlings for visible sectors showed an increase in the mutation frequency in the pm/pm plant line as a result of 9-aminoacridine hydrochloride treatment, but the differences in the final mutation frequencies of control and treated pm/pm seedlings were not statistically significant (0.05 < p < 0.10). In fact, neither acridine nor nalidixic acid could be demonstrated to affect the mutation frequency in the pm plant lines. 94 The difference between the control and acridine-treated pm/pm seedlings are not statistically significant, yet there were many chlorotic sectors observed in the wt seedlings. This suggests, along with the crossing data, that 9-aminoacridine hydrochloride is a potent plastome mutagen. The lack of a synergistic effect in inducing chlorosis in conjunction with the pm/pm plant line implies that the DNA lesion caused by 9- aminoacridine hydrochloride does not interfere with the pm defect. Hopefully, more trials with larger sample sizes will allow a clearer statistical determination. From the experiments with nalidixic acid, as in the wild-type Oenothera experiments (Chapter 1) it appears that a functional gyrase subunit A is regenerated upon removal of the inhibitor, as inferred in Solanum nigrum, where cell growth recovers after the removal of the chemical (Ye and Sayre 1990) and the gyrase subunit A portion of the chloroplast replication system is not involved with the pm gene, either directly or indirectly. Conversely, the novobiocin treatment of pm/pm seeds resulted in about twice the number of visible sectors as obtained ‘without the chemical treatment. This also contrasted with the novobiocin treatment of wt seedlings which produced few chlorotic sectors. These differences in the final mutation frequencies of control and treated mepm seedlings were 95 statistically significant. This suggests that damage resulting from novobiocin is corrected in the wild-type plants but not in pm/pm plants. Many experiments show that novobiocin perturbs gyrase (Gellert et al. 1977, 1978; Sugino et al. 1977, 1978; Ciarrocchi et al. 1985; Thompson and Mosig 1985; Ye and Sayre 1990). If this is also true in Oenothera, it suggests that the wild-type Oenothera chloroplast may have a novobiocin-insensitive bypass replication as seen in Chlamydomonas reinhardtii (WOelfle et al. 1993). In that case, the wt seedlings would not show any effects or few effects from novobiocin while the pm/pm plants would show a synergistic increase in the induction of chlorotic sectors, if they lack the novobiocin-insensitive gyrase. Further testing is needed to determine if topoisomerase II/gyrase is indeed the target for novobiocin. If the target is gyrase, it then becomes necessary to find out how transcription is affected in the pmlpm plant line and why no sectors were seen in wt seedlings when they were treated with novobiocin. CONCLUSION Previous investigations have shown that in the plastome .mutator line of Oenothera hookeri, chloroplast DNA is the target of the increased spontaneous mutation rate. Significant chNA alterations and effects include: 1) pm induces a wide variety of mutations indicating that a number of loci can be mutated (Epp et al. 1987), 2) deletions of variable sizes occur at hot spots on the chNA molecule (Chui et a1. 1990), 3) direct repeats are involved in the high frequency deletions (Blasko et al. 1988; Chiu et al. 1990), 4) the chNA deletions are separable from the mutant phenotypes (Chui et al. 1990) and 5) NMU mutagenesis interacts synergistically with the plastome mutator activity (Sears and Sokowski 1991). These findings and the inability to obtain antibiotic-resistance from pm/pm callus indicate that the mutations causing the chlorosis are likely to be very small deletions/insertions which eliminate gene function. Summarized, it appears that the pm gene mutates the chNA at many sites, causing small deletions/insertions of direct repeats and favoring certain sites. Nalidixic acid and novobiocin have both been shown to inhibit gyrase in Escherichia coli (Gellert et al. 1976 and 1777; Sugino et al. 1977 and 1978; Drica and Snyder 1977; Lockshon 96 and Morris 1983), cultured carrot cells ( Ciarrocchi et al.1985), cultured Nicotiana tabacum cells (Heinhorst et al. 1985), Chlamydomonas reinhardii ( Thompson and Mosig 1984 and 1987), pea chloroplasts (Lam and Chua 1987), and cultured Solanum nigrum cells (Ye and Sayre 1990)). The synergism observed between novobiocin and the pm line implicates subunit B of gyrase as being involved in the plastome mutator defect or it may point to the possibility that a novobiocin bypass system similar to that found in C. reinhardtii (Woelfe et al. 1993) may be present in Oenothera but defective in the pm line. Both nalidixic acid and novobiocin have been used on many organisms as mentioned above, to perturb gyrase. These chemical agents have been shown to bind gyrase, resulting in a secondary effect on transcription, with a decrease of some transcripts and a increase in other transcripts. Thus, gyrase seems to be utilized in transcription control, as well as in replication and repair. One hypothesis is that there are two different gyrases present in the chloroplast. One of these gyrases may be preferential and perform most or all of the supercoiling of the DNA in the chloroplast. In this hypothesis the plastome mutator gene may encode a faulty B subunit for the preferred gyrase resulting in loss of its function. It now becomes necessary for the second 97 gyrase to perform all supercoiling in the chloroplast. This second gyrase may not be as well adapted to supercoiling as the perferred gyrase. This could result in an underwound or a highly supercoiled chNA molecule resulting in aberrant transcription, and in others, a high rate of mutation. With an underwound chNA, the replicating machinery might tend to slip ahead, deleting direct repeats or even a single base pair as suggested by Skopek and Hutchinson (1984). An explanation as to why an increase in chlorosis is observed with the novobiocin treatment but not with the nalidixic acid treatment is due to the chemicals affecting different subunits of gyrase. When nalidixic acid is used to treat Oenothera seeds, the mutagen targets the A subunit of the preferred gyrase which in the plastome mutator line has lost function due to the mutator activity. Since the wild-type plants have a functional preferred gyrase, there would be little impact on their genetic system and a dramatic increase in mutation rate would not be observed. Conversely, when novobiocin is used to treat Oenothera seeds, the mutagen targets the B subunit of the second gyrase. The chloroplast of the pm line lack a functional preferred gyrase due to the Ipm~mutation, and contain a novobiocin-perturbed second gyrase leaving them with no functioning gyrase for the duration of the novobiocin-treatment. The consequence would be the synergistic increase in chlorosis seen in pm/pm plants. 98 The hypothesis presented could follow the premise introduced by Volff et al. (1993), who found that the occurrence of genetic instability, deletions and amplifications of DNA sequences in streptomycetes is specific to chemical treatments interfering with DNA replication, including gyrase inhibitors. Because the inhibition of DNA replication induces the SOS response causing deletions in E. coli and Salmonella typhimurium (Ishii and Kondo, 1975), Volff et al. (1993) suggested that either DNA gyrase is directly involved in genetic instability or that it may be part of an SOS—like system. In addition to the above hypothesis, a gyrase-mediated recombination system was proposed by Ikeda et al. in 1981. The plastome mutator system.may be similar in mode of action. If the gyrase cleaves DNA within each direct repeat in a molecule, single strands protruding at each end of the repeats would be complementary to each other and could pair. If the gyrase ligated both ends to join the two double strands, this would result in the formation of a deletion by eliminating the segment between the repeated sequences and one copy of the repeats. This would require that the gyrase cleavage site be within a repeat and may explain why one direct repeat suffers deletions over other direct repeats within the chNA. This gyrase-mediated recombination system, if faulty in the plastome mutator plant line, does not 99 account for the high rate of mutation frequency that is observed in Oenothera. In summary, the data produced here lends supports to Epp's findings in 1973, that the plastome.mutator is a null mutation. This null mutation may be a gyrase subunit and result in an underwound chNA molecule. The distortion of the chNA may cause the slippage of the replicating machinery and produce permanent alterations in the DNA of the plastids. This in turn, could affect a variety of loci and cause chlorosis as is observed in the pm plant line. 100 LIST 01" REFERENCES Birky, C. (1978) Transmission genetics of mitochondria and chloroplasts. Ann. Rev. Genet. 12:471-512. Blasko, K., Kaplan, S., Higgins, K., WOlfson, R. and Sears, B. (1988) Variation in copy number of a 24-base pair tandem repeat in the chloroplast DNA of Oenothera hookeri strain Johnsen. Curr. Genet. 14:287-292. Boerner, T. and Sears, B. (1986) Plastome Mutants. Plant Mol. Bio. Reporter 4:69-92. Brennicke, A. and Schwemmle, B. (1984) Inheritance of mitochondrial DNA in Oenothera berteriana and Oenothera odorata hybrids. z. Naturforsch 39:191-192. Buiatti, M. and Ragazzini, R. (1966) Mutagenic effect of acridine orange in tomato (Lycopersicum esculentum). Mutation Res. 3:360-361. Castellani, A. (1987) DNA Damage and Repair, Plenum Press, New York and London, pp. ' Chiu, W., Stubbe, W. and Sears, B. (1988) Plastid inheritance in Oenothera: organelle genome modifies the extent of biparental plastid transmission. Curr. Genet. 13:181-189. Chiu, W., Johnson, E., Kaplan, S., Blasko, K., Sokalski, M., WOlfson, R. and Sears, B. (1990) Oenothera chloroplast DNA polymorphisms associated with plastome mutator activity. Mol. Gen. Genet. 221:59-64. Chiu, W. and Sears, B. (1993) Plastome-genome interactions affect plastid transmission in Oenothera. Genetics 133:989- 997. Ciarrocchi, G., Nielson, E. and Cella, R. (1985) Effect of nalidixic acid and novobiocin on the metabolism of suspension cultured carrot cells. Physiol. Plant 64:513-518. Cooper, C. and Grover, P.(1990) Chemical Carcinogenesis and Mutagenesis I, Springer-verlag Berlin Heidelberg, pp.231 101 102 Cozzarelli, N. (1977) The mechanism of action of inhibitors of DNA synthesis. Annu. Rev. Biochem. 46:641-668. Cozzarelli, N.R. (1980) DNA gyrase and the supercoiling of DNA. Science 207:953-960. Cozzarelli, N. and Wang J. (1990) DNA topology and its biological effects. Cold Spring Harbor, NY, USA, Cold SpringHarbor Laboratory Press. 20:1-489. D'Amato, F. (1950) Mutazioni clorofilliane nell'oro indotte da derivati acridinici. Caryologia 3:211-220. D'Amato, F. (1952) Mutagenic activity of acridines. Caryologia 4:388-413. D'Amato, F. (1954) A survey of acridines for chromosome- breaking (mutagenic) action in Allium. Proc. 9th Int. Congr. Drake, J. and Kock, R. (1976) Mutagenesis. Benchmark Papers in Genetics, VOl.4, Dowden, Hutchinson and Ross Stroudsburg, PAe ppe 220-254e Eherenberg, L., Gustafson, A. and Lundquist, U. (1956) chemically induced mutation and sterility in barley,. Acta Chem. Scand. 10:492-494. Epp, M. and Parthasarthy, M. (1987) Nuclear gene-induced plastome mutations in Oenothera hookeri. II. phenotypic description with electron microscopy. Amer. J. Bot. 2:143- 151. Epp, M. (1973) Nuclear gene-induced plastome mutations in Oenothera HOokeri. I. Genetic analysis. Genetics 75:465-483. Etzold, T., Fritz, C., Schell, J. and Schreier, P. (1987) A point mutation in the chloroplast 16S rRNA gene of a streptomycin-resistant Nicotiana tabacum. FEBS Lett. 219:343- 346. Ferguson, L. and Denny, W. (1991) The genetic toxicology of acridines. Mutation Research 258:123-160. Ferguson, L. and MacPhee, D. (1983) Frameshift mutagenesis by 9-aminoacridine, ICR 191, AMSA and related experimental antitumour acridines in recA+and recAl strains of Salmonella typhimurium. Mutation Res. 116:289-296. 103 Fluhr R, Aviv D, Galun E, and Edelman M (1985) Efficient induction and selection of chloroplast-encoded antibiotic- resistant mutants in Nicotiana. Proc Natl Acad Sci USA; 82:1485-1489. Galili, S. Fromm, H., Aviv, D. Edelman, M. and Galun, E. (1989) Ribosomal protein 812 as a site for streptomycin resistance in Nicotiana chloroplasts. Mol. Gen. Genet. 218:289-292. Gauthier, A., Turmel, M., and Lemieux, C.(1988) Mapping of chloroplast mutations conferring resistance to antibiotics in Chlamydomonas: Evidence for a novel site of streptomycin resistance in the small subunit rRNA. Mol.Gen.Genet. 214:192- 197 Gellert, M., Mizuuchi, K., O'Dea, M.H., and Nash, H. (1976) DNA gyrase: An enzyme that introduces superhelical turns into DNA. Proc. Natl. Acad. Sci USA 73:3872-3876. Gellert, M. O'Dea, M., Itoh, T., and Tomizawa, J. (1976) Novobiocin and coumermycin inhibit DNA supercoiling catalyzed by DNA gyrase. Proc. Natl. Acad. Sci. USA 73:4474-4478. Gellert, M., Mizuuchi, K., O'Dea, M., Itoh, T. and Tomizawa, J. (1977) Nalidixic acid resistance: a second genetic character involved in DNA gyrase activity. Proc. Natl. Acad. Sci. USA 74:4772-4776. Gellert, M., Mizuichi, Kiyoshi, Mizuichi, M., and O'Dea, M. (1978) Cloning and simplified purification of Escherichia coli DNA gyrase A and B proteins. J. Bio. Chem. 259:9199- 9201. Gellert, M.(1981) DNA topoisomerases. Annu. Rev. Biochem. 50:879-910. Gordon, A., Halliday, J., Horsfad, M. and Glickman, B. (1991) Spontaneous and 9-aminoacridine-induced frameshift mutagenesis: second-site frameshift mutation within the N- terminal region of the lacI gene of Escherichia coli. Mol. Gen. Genet. 227:160-164. Harris, EH (1988) The Chlamydomonas Sourcebook, Academic Press, Inc. California, pp.384. Hagemann R (1982) Induction of plastome mutations by nitroso- urea compounds. In: Edelman M, Hallick RB, Chua N-H, (eds) Methods in Chloroplast Molecular Biology. Elsevier Biomedical Press. pp. 119-127. 104 Hammond, G., Cassidy, P., and Overbye, K. (1991) Novobiocin- dependent topA deletion mutants of Escherichia coli. Journal of Bacteriology 173:5564-5567. Heinhorst, S., Cannon, G. and weissbach, A. (1985) Chloroplast DNA synthesis during the cell cycle in cultured cells of Nicotiana tabacum: inhibition by nalidixic acid and hydroxyurea. Archives of Biochemistry and Biophysics 239:475- 479. Hoffmann, G., Freemer, C., and Parente, L. (1989) Induction of genetic duplications and frameshift mutations in Salmonella typhimurium by acridines and acridine mustards: Dependence on covalent binding of the mutagen to DNA. Mol. Gen. Genet. 218:377-383. Horowitz, D. and wang, J. (1987) Mapping the active site tyrosine of Escherchia coli gyr A gene coding for the A subunit of DNA gyrase. J. Mol. Bio. 262:5339-5344. Hosticka L., Hanson M. (1984) Induction of plastid mutations in tomatoes by nitroso-methylurea. J. Heredity 75:242-246. Hsiang,Y., Hertzgerg, R. and Hecht, S. (1985) Camptothecin induces protein-linked DNA breaks via mammalian DNA topoisomerase I. J. Bio. Chem. 260:14873-14878. Hunt, M. and Newton, K. (1991) The NCS3 mutation: genetic evidence for the expression of ribosomal protein genes in Zea .mays mitochondria. EMBO journal 10:1045-1052. Iwamoto, Y., Mifuchi, I. and Yielding, L. (1985) Photodynamic mutagenic action of acridine compounds on yeast Saccharomyces cerevisiae. Mutation Res. 158:169-175. Kirk, J.T.O. and Tilney-Bassett, R.A.E. (1978) The Plastids. pp.1-l960, Elsevier Amsterdam. Kubo, M., Kano, Y., Nakamura, H., Nagata, A., Imamoto, F. (1979) In vivo enhancement of general and specific transcription in Escherichia coli by DNA gyrase activity. Gene 7:153-157. Lam, E., and Chua, N. (1987) Chloroplast DNA gyrase and in vitro regulation of transcription by template topology and novobiocin. Plant Mol. Bio. 8:415-424. Lawley, P. (1961) Action of alkylating agents in DNA. J. Chem. Phys. 11:1011-1020. 105 Lawley, P. And Brookes, P. (1961) Acidic dissociation of 7:9- dialkylguanines and its possible relation to mutagenic properties of alykating agents. Nature 192:1081-1082. Levin, D. Marnett, L. and Ames, B. (1984) Spontaneous and mutagen-induced deletions: Mechanistic studies in Salmonella tester strain TA102. Genetics 81:4457-4461. Lindahl T, Sedgwick B, Sekiguchi M, Nakabeppu Y (1988) Regulation and expression of the adaptive response to alkylating agents. Ann. Rev. Biochem. 57:133-57. Liu, L.F. and wang, J.C. (1978) Micrococcus luteus DNA gyrase: active components and a model for its supercoiling of Loveless (1969) Possible relevance of O-6 alkylation of deoxyguanosine to the mutagenicity and carcinogenicity of nitrosamines and nitrosamides. Nature 223:206-208. Lyman, H. (1967) Specific inhibition of chloroplast replication in Euglena gracilis by nalidixic acid. Lyman, H. Jupp, A. and Larrinua, I. (1975) Action of nalidixic acid on chloroplast replication in Euglena gracilis. Plant Physiol. 55:390-392. Magee, PN and Barnes, JM (1956) Production of malignant primary hepatic tumours in the rats by feeding dimethylnitrosamine. Br. J. Cancer 10:114-122. McCabe, P. Timmons, A. and Dix, P. (1989) A simple procedure for the isolation of streptomycin resistant plants in Solanaceae. Mol. Gen. Genet. 216:132-137. McCoy, E., Rosenkranz, E., Petrullo, L., and Rosenkranz, H. (1981) Frameshift mutations: Relative roles of simple intercalation and of adduct formation. Mutat. Res. 90:21-30. Michaelis,A. and Rieger, R. (1963) On the time period during which chemically-induced chromatid breaks are available for interaction. Exptl. Cell Res. 31:202-205. Miura-Masuda, A. and Ikeda, H. (1990) The DNA gyrase of Escherichia coli participates in the formation of a spontaneous deletion by recA-independent recombination in vivo. Mol. Gen. Genet. 220:345-352. Moazed, D. and Noller, H. (1987) Interaction of antibiotics ‘with functional sites in 16S ribosomal RNA. Nature 327:389- 394. 106 Monakhov, A., Iakovleva, T., and Anisimov, V. (1990) Chromosomal damage in peripheral blood lymphocytes of rats in nitrosomethylurea-induced carcinogenesis. Eksp-Onkol. 12:40- 43. Montandon, P., wagner,R. and Stutz, E. (1986) Streptomycin- resistance of Euglena gracilis chloroplast: identification of a point mutation in the 16S rRNA gene in an invariant position. Nucleic Acids Res. 13:4299-4309 Morrison, A. and Cozzarelli, N.R. (1979) Site-specific cleavage of DNA by E. coli DNA gyrase. Cell 17:175-184. Mullet, J. (1988) Chloroplast development and gene expression. Ann. Rev. Plant Physiol. Plant Mol. Biol. 39:475- 502. Nagata, C., Imamura, A., Saito, H., and Fukui, K. (1963) Changes of pi-electron distribution of deoxyribonucleic acid after alkylation and their possible relation to the biological effect. Gann. 54:109-117. Nagata, T. and Takebe, I. (1971) Plating of isolated tobacco mesophyll protoplasts on agar medium. Plant 99:12-20. Nasim, A., and Brychcy, T. (1979) Genetic effects of acridine compounds. Mutation Research 65:261-288. Newton, A., Masys, D., Leonardi, E., and wygal, D. (1972) Association of induced frameshift mutagenesis and DNA replication in Escherichia coli. Nature New Biology 236:19- 22. Nuti-Ronchi, V. and D'Amato, F. (1961) New data on chromosome breakage by acridine orange in the Allium test. Caryologia 14:163-192. Pommier, Y., Covey, J., Derrigan, D., Mattes, W., Markovits, J., and Kohn, K. (1987) Role of DNA intercalation in the inhibition of purified mouse leukemia (L1210) DNA topoisomerase II by 9-aminoacridines. Biochemical Pharmacology 36:3477-3486. Pons, F. (1984) Genetic analysis of clear-plaque mutations induced in bacteriophage lambda by 9-aminoacridine. Mutation Res. 129:311-317. Podger, D. and Hall, R. (1984) Induction of SOS functions is not required for rear-dependent mutagenicity of 9- aminoacridine in Salmonella typhimurium trp-E8. Mutation Res. 131:115-121. 107 Potrykus, I. (1970) Mutation und Ruckmutation extrachromosomale vererbter Plastidenmerkmale von Petunia. z. Pflanzenzuecht 63:24-40. Prina, A.R. (1992) A mutator nuclear gene inducing a wide spectrum of cytoplasmically inherited chlorophyll deficiences in barley. Theor. Appl. Genet. 85:245-251. Radl, S. (1990) Structure-activity relationships in DNA gyrase inhibitors. Pharmacol. Ther. 40:1-17. Redei, G. and Plurad, S. (1973) Hereditary structural alterations of plastids induced by a nuclear mutator gene in Arabidopsis. Protoplasma 77:361-380. Richardson, K.K., Richardson, F.C., Crosby, R.M., Swenberg, J.A., and Skopek, T.R. (1987) DNA base changes and alkylation following in vivo exposure of Escherichia coli to N-methyl-NL nitrosourea or N-ethyl-N-nitrosourea. Roussell, D., Thompson, D. Pallardy, 8., Miles, D. and Newton, K. (1991) Chloroplast structure and function is altered in the NCSZ maize mitochondrial mutant. Plant Physiol. 96:232-238. Saing, K., Orii, H., Tanake, Y., Yanagisawa, K., Miura, A. and Ikeda, H. (1988) Formation of deletion in Escherichia coli between direct repeats located in the long inverted repeats of a cellular slime mold plasmid: Participation of DNA gyrase. Mol. Gen. Genet. 214:1-5. Sakore, T., Jain, S., Tsai, C. and Sobell, H. (1977) Mutagen- nucleic acid intercalative binding: structure of a 9- aminoacridine:5-iodocytididylyl (3'-5') guanosine crystalline complex. Proc. Natl. Acad. Sci. USA; 74:188-192. Sakore, T., Riddy, B. and Sobell H. (1979) Visualisation of drug-nucleic acid interactions at atomic resolution 1V structure of an aminoacridine-dinucleoside monophosphate crystalline complex, 9-aminoacridine-5-iodocytidylyl (3'-5') guanosine. J. Mol. Biol. 135:763-785. Sears BB, Sokalski MB (1991) The Oenothera plastome mutator: effect of UV-irradiation and nitroso-methyl urea on mutation frequencies. Mol Gen Genet 229:245-252. Shinozaki, K., Ohme, M., Tanaka, M., Wakasugi, T., Hayashida, N., Matsubayashi, T., Zaita, N., Chunwongse, J., Obokata, J., Yamaguchi-Shinozaki, K., Ohto, C., Torazawa, K., Meng, Y., Sugita, M., Deno, H., Kamogashira, T., Yamada, K., Kusuda, J., Takaiwa, F., Kato, A., Tokdah, N., Shimada, H., and 108 Sugiura, M. (1986) The complete nucleotide sequence of the tobacco chloroplast genome: its gene organization and expression. EMBO J. 5:2043-2049. Sigmund, C., Ettayebi, M., and Morgan, E.A. (1984) Nucleic Acids Res. 12:4653-4663 Singer B, Grunberger D (1983) Molecular Biology of Mutagens and Carcinogens. Plenum Press, N.Y. Skopek, T. and Hutchinson, F. (1984) Frameshift mutagenesis of lambda prophage by 9-aminoacridine, proflavin and ICR-191. Mol. Gen. Genet. 195: 418- 423. Stubbe, W. and Herrmann, R.G. (1982) Selection and maintenance of plastome mutants and interspecific genome/plastome hybrids from Oenothera. Methods in Chloroplast Molecular Biology. Elsevier Biomedical Press, Amsterdam. Sugino, A, Peebles, C.L., Kreuzer, K.N., Cozzarelli, N.R. (1977) Mechanism of action of nalidixic acid: purification of Escherichia coli nalA gene product and its relationship to DNA gyrase and a novel nicking-closing enzyme. Proc. Natl. Acad. Sci. USA 74:4767-4771. Sugino, A. Higgins, P., Brown, P., Peebles, C. and Cozzarelli, N. (1978) Energy coupling in DNA gyrase and the mechanism of action of novobiocin. Proc. Natl. Acad. Sci. USA 75:4838-4842. Svab, z., Maliga, P. (1991) Mutation proximal to the tRNA binding region of the Nicotiana plastid rRNA confers resistance to spectinomycin. Mol. Gen. Genet. 228:316-319. Thomas, S., and MacPhee, D. (1985) Frameshift mutagenesis by 9-aminoacridine and ICR191 in Escherichia coli: effects of uvrB, recA and lexA mutations and of plasmid pKM101. Mutation Research 151:49-56. Thompson R.J. and Mosig, G. (1984) Light and genetic determinants in the control of specific chloroplasts transcripts in Chlamydomonas reinhartii. Plant Physiol. 76:1- 6. Thompson R.J. and Mosig, G. (1985) An ATP-dependent supercoiling topoisomerase of Chlamydomonas reinhardtii affects accumulation of specific chloroplast transcripts. Nucleic Acids Research 13:873-891. 109 Thompson, R.J. and Mosig, G. (1987) Stimulation of a Chlamydomonas chloroplast promoter by novobiocin in situ and in E. coli implies regulation by torsional stress in the chloroplast DNA. Cell 48:281-287. Thompson, R.J. and Mosig, G. (1990) Light affects the structure of Chlamydomonas chloroplast chromosome. Nucleic Acids Res. 18:2625-2631. Welters, T. And Earle, E. (1990) Cytoplasmic mutants from seed mutagenesis of Brassica compestris with NMU. J. Heredity 81:214-216. Wang, J.C. (1985) DNA topoisomerases. Ann. Rev. Biochem. 54:665-697. wang, J.C. (1986) DNA topoisomerases: From a laboratory curiosity to a subject in cancer chemotherapy. NCI Monographs: First Conference on DNA Topoisomerases in Cancer Chemotherapy. weintraub, H. (1985) Assembly and propagation of repressed and derepressed chromosomal states. Cell 42:705-711. WOelfle, M.A., Thompson, R.J., Mosig, G. (1993) Roles of novobiocin-sensitive topoisomerases in chloroplast DNA replication in Chlamydomonas reinhardtii. Nucleic Acids Res. 21:4231-4238 VOlff, J., Vandewiele, D., Simonet, J. and Decaris, B. (1993) Stimulation of genetic instability in Streptomyces ambofaciens ATCC 23877 by antibiotics that interact with DNA gyrase. Journal of General Microbiology 139:2551-2558. WOlfson, R., Higgins, K., Sears, B. (1991) Evidence for Replication Slippage in the Evolution of Oenothera Chloroplast DNA. Mol. Biol. Evol. 8:709-720 Ye, J. and Sayre, R. (1990) Reduction of chloroplast DNA content in Solanum nigrum suspension cells by treatment with chloroplast DNA synthesis inhibitors. Plant Physiol. 94:1477- 1483. NICHIGAN STATE UNIV. LIBRARIES iiIIIIIIIIIIIIIWIII”IIIHIIIWI 31293014096568