<p>An increasing number of chloroplast proteins have been found to interact with plant virus proteins. This is not surprising because these viruses cause various mosaic, mottles, and chlorosis symptoms on host leaves indicating damage to chloroplasts. A chloroplast protein, AtPsbP, was identified in a yeast two-hybrid screen as interacting with <i>Alfalfa mosaic virus</i> (AMV) coat protein (CP). AMV is a ssRNA virus with a wide host range including Arabidopsis. AtPsbP is an extrinsic subunit of photosystem II and with PsbQ is vital for water oxidation. We found that an RNAi knock-down of PsbP in <i>Nicotiana tabacum</i>, allowed increased replication of AMV and the development of quite severe disease symptoms in comparison to a wild-type <i>N. tabacum</i>. This suggested that PsbP plays an important role in plant resistance to AMV. PsbP, in addition to its role in photosynthesis, has been reported to interact with a wall-associated receptor kinase, WAK1, whereby it may affect plant defense signaling. We found that AtPsbP is a link between AtWAK1 and AMV CP at the plasma membrane. The formation of the AtWAK1-AtPsbP-AMV CP complex activated WAK1 kinase activity causing phosphorylation of PsbP and significant inhibition of AMV replication. We also found that the formation of the ternary complex induced the activation of the MAPK signal pathway. Analysis of the susceptibility of an Arabidopsis WAK1 knock-down indicated that WAK1, like PsbP, is critical for inhibiting AMV replication. Overall, we found a unique virus perception strategy, whereby a chloroplast protein (PsbP) interacts with a virus protein and then a Receptor-like kinase protein (WAK1) to transduce signals through the MAPK signaling pathway to activate defense responses.</p>

Penstemon is one of North America's largest endemic genera with over 280 described species. These species are distributed throughout most of North America from the Arctic northern latitudes to tropics of Central America. The genus has historically been divided into six subgenera, but has recently been reorganized into two subgenera following some recent phylogenetic studies. I made a comprehensive assessment of the Penstemon genus' geographic distribution utilizing herbaria databases by ecoregion to discuss the general ecologic adaptations of each historic subgenera. I also assessed the Penstemon genus' bee pollinator diversity utilizing online databases of bee specimen collections associated with Penstemon flowers. I investigated the efficacy of utilizing the plastid genomes (plastomes) of 29 species in the Lamiales order, including five newly sequenced Penstemon plastomes, for analyzing phylogenetic relationships and resolving problematic clades. I compared whole-plastome based phylogenies to phylogenies based on individual gene sequences (matK, ndhF, psaA, psbA, rbcL, rpoC2, and rps2) and concatenated sequences. I found that my whole-plastome based phylogeny had higher nodal support than all other phylogenies, which suggests that it provides greater accuracy in describing the hierarchal relationships among taxa as compared to other methods. I found that the genus Penstemon forms a monophyletic clade sister to, but separate from, the Old World taxa of the Plantaginaceae family included in our study. My whole-plastome based phylogeny also supports the rearrangement of the Scrophulariaceae family and improves resolution of major clades and genera of the Lamiales. I evaluated 16 accessions of P. cyaneus with 14 accessions of closely related Penstemon species in common garden in two distinct environments in Aberdeen, ID and Provo, UT during 2018 and 2019. I evaluated the accessions for key commercial seed production traits including survival, plant height, number of stems, and seed production. Both common gardens received supplemental irrigation during 2018, but I withheld irrigation during 2019. Plant survival in our Aberdeen, ID site was not significantly different between years, but survival was a significantly lower in 2019 than in 2018 at our Provo, UT site. The mean survival for P. cyaneus accessions ranged from 56% to 94%, and the mean seed production ranged from 91.2 kg/ha to 397.6 kg/ha. I recommend developing a commercial seed source derived from pooling germplasm of six accessions (PECY3-367, PECY3-371, PECY3-376, PECY3-443, PECY3-457, and PECY3-458). These accessions had mean survival rates of 82-94%, and seed production of 196.2-397.6 kg/ha.

Chloroplasts of many species of hornworts (Anthocerotae) have a structure that resembles the pyrenoid of green algae but whether these two structures are homologous has not been determined. We utilized immunogold labelling on thin sections to determine the distribution of ribulose 1,5-bisphosphate carboxylase/oxygenase (RuBisCO), the major protein of algal pyrenoids, in sixteen hornwort species with and without pyrenoids. Several species (Phaeoceros laevis, Anthoceros punctatus, A. formosae, A. laminiferus, Folioceros fuciformis, Folioceros sp., Dendroceros tubercularis, D. japonicus, D. validus, Notothylas orbicularis, N. temperata, and Spaerosporoceros adscendens) have uniplastidic (or primarily uniplastidic) cells with large prominent multiple pyrenoids. In all of these species, the labelling is found exclusively in the pyrenoid and, with the exception of the Folioceros, Dendroceros, and Notothylas species, the labelling is randomly distributed throughout the pyrenoid. In the exceptional species, the pyrenoids have prominent pyrenoglobuli or other inclusions that are unlabelled. In Megaceros flagellaris and M. longispirus, the cells are multiplastidic (with the exception of the apical cell and some epidermal cells) and the chloroplasts lack pyrenoids. Anthoceros fusiformis and Phaeoceros coriaceus have primarily uniplastidic cells but the chloroplasts lack pyrenoids; only an area of stroma in the center of the plastid devoid of starch, reminiscent of a pyrenoid, is found. In all of the species lacking pyrenoids, RuBisCo is found throughout the stroma, including the stromal spaces made by the so-called channel thylakoids. No preferential accumulation of RuBisCo is found in the pyrenoid-like region in A. fusiformis and P. coriaceus. These data indicate that 1) the hornwort pyrenoid is homologous to algal pyrenoids in the presence of RuBisCo; 2) that at least some of the RuBisCo in the pyrenoid must represent an active form of the enzyme; and 3) that, in the absence of pyrenoids, the RuBisCo is distributed throughout the stroma, as in higher plants.

Elektrophysiologische und biochemische Charakterisierung von rekombinanten und nativen Kanalproteinen aus der inneren Chloroplasten-Hüllmembran und dem Cyanobakterium Synechocystis In der vorliegenden Arbeit wurden zwei angenommene Kaliumkanäle heterolog exprimiert und untersucht. Außerdem wurde die elektrophysiologische Charakterisierung des am Proteinimport beteiligten Proteins Tic110 durchgeführt. Ergänzt wurden diese Experimente durch eine elektrophysiologische Charakterisierung von Ionenkanälen in der inneren Chloroplasten-Hüllmembran. Sowohl in Synechocystis sp. PCC 6803 als auch in Arabidopsis thaliana konnten zwei Gene identifiziert werden, die wahrscheinlich Kaliumkanäle codieren. Das Genprodukt aus Arabidopsis ist wahrscheinlich im Chloroplasten lokalisiert und besitzt sechs putative transmembrane Durchgänge. Es ähnelt strukturell den Kaliumkanälen AKT1 und KAT1. Der potentielle Synechocystis-Kanal besitzt aufgrund von Sekundärstrukturvorhersagen Ähnlichkeit mit dem Kaliumkanal KcsA aus Streptomyces lividans und eukaryontischen Kanälen der IRK-Familie. Beide Proteine wurden als His-Tag Fusionsproteine in Baculovirus-infizierten Sf21-Insektenzellen überexprimiert und konnten in der Membranfraktion der Zellen nachgewiesen werden. Patch-Clamp-Messungen in der ’Whole-Cell’ und ’Excised-Patch’ Konfiguration an den Insektenzellen zeigten, dass keine funktionellen Kanäle in der Plasmamembran lokalisiert waren. Beide Kanäle konnten mittels nicht-ionischer Detergentien solubilisiert werden, aber nur der Synechocystis-Kanal konnte mittels Ni-Affinitätschromatographie gereinigt werden. Nach Rekonstitution des Proteins in Azolektin-Liposomen zeigte sich bei Messungen im Bilayer-System jedoch keine Kanalaktivität. Ursache hierfür ist wahrscheinlich eine zu geringe Offenwahrscheinlichkeit des Proteins oder eine nicht funktionelle Rekonstitution. Weiterhin wurden bei elektrophysiologischen Messungen mit der Bilayer-Methode Ionenkanäle in der isolierten inneren Hüllmembran von Chloroplasten untersucht. Es konnte die Existenz eines Kaliumkanals bestätigt sowie zwei bisher unbekannte Kanäle (CIMCC1 und CIMCC2) charakterisiert werden. CIMCC1 ist mäßig selektiv für Kationen (PK/PCl = 3.5), besitzt einen Haupt- und einen Unterleitwert von 680 bzw. 330 pS (in 250 mM KCl) und eine spannungsunabhängige Offenwahrscheinlichkeit von 70 %. Der Kanal könnte aufgrund seiner Eigenschaften am Transport von Aminosäuren über die innere Chloroplasten-Hüllmembran beteiligt sein. CIMCC2 ist Kationen-selektiv (PK/PCl = 5.3), besitzt einen Leitwert von 600 pS (in 250 mM KCl) und schließt bei höheren positiven bzw. negativen Membranpotentialen. Der Kanal ist durch wenige 100 nM des Präpeptids Troe33 blockierbar, weshalb er eine Rolle im Import von Proteinen in den Chloroplasten spielen könnte. Ferner konnte im Rahmen dieser Arbeit gezeigt werden, dass das Protein Tic110 nach heterologer Expression in E. coli und Rekonstitution in Liposome eine hydrophile Pore bildet, deren Eigenschaften denen von Proteinimportkanälen in der äußeren Membran von Chloroplasten und Mitochondrien ähnelt. Aufgrund einer Sekundärstrukturvorhersage und des CD-Spektrums ist, im Gegensatz zu vorherigen Annahmen, eine von b-Faltblättern dominierte Sekundärstruktur wahrscheinlich. Der durch Tic110 gebildete Kanal ist Kationen-selektiv, hat einen Leitwert von 446 pS in 250 mM KCl und einen Porendurchmesser zwischen 15 und 34 Angström. Die spannungsabhängige Offenwahrscheinlichkeit ist maximal bei kleinen Membranpotentialen und nimmt zu höheren positiven und negativen Spannungen hin ab. Weiterhin ist der Kanal durch geringe Konzentrationen des Präpeptids Troe33 (ca. 100 nM) spezifisch hemmbar. Tic110 besitzt funktionell eine starke Ähnlichkeit mit CIMCC2 in der inneren Hüllmembran, was nahelegt, dass es sich hierbei um die gleichen Proteine handelt. Ausgehend von diesen Eigenschaften ist es wahrscheinlich, dass Tic110 die Proteinimportpore der inneren Chloroplasten-Hüllmembran darstellt.

PRAT-C2 In elektrophysiologischen Untersuchungen heterolog exprimierter PRAT-C2 Proben wurde eine Kationen-selektive Kanalaktivität mit einer dreifachen Porenstöchiometrie beobachtet. Durch Variation der experimentellen Parameter und eine detaillierte Analyse der Ergebnisse konnten seitenspezifische Eigenschaften des Kanals aufgedeckt werden, die auf eine ungleiche Ladungsverteilung hindeuten. Auf der Grundlage von Strukturvorhersagen wurde ein Topologiemodell mit vier transmembranen alpha-Helices für PRAT-C2.2 aus Arabidopsis thaliana erstellt. Die zum Stroma und Intermembranraum exponierten Sequenzabschnitte weisen eine sehr unterschiedliche Verteilung geladener Aminosäuren auf. Das Schaltverhalten des Kanals wurde sowohl durch das Signalpeptid eines chloroplastidären als auch eines mitochondrialen Präproteins beeinflusst. Trotz einer Reaktion auf den spezifischen Antikörper gegen PRAT-C2.1 wurden Hinweise dafür erbracht, dass es sich bei der Kanalaktivität um eine porenbildende Kontamination aus dem bakteriellen Expressionssystem handelte. Tic110 Durch den Vergleich der Eigenschaften von Tic110, aufgereinigt aus Chloroplasten von Pisum sativum, und einer verkürzten, heterolog exprimierten Form konnte gezeigt werden, dass der porenbildende Sequenzabschnitt jenseits der ersten 96 Aminosäuren liegt. Die Kanalaktivität zeichnete sich durch ein charakteristisches Schaltverhalten aus, wobei die Schaltfrequenz durch eine Wechselwirkung mit Ca2 -Ionen beeinflusst wurde. Tic110 zeigte eine Präferenz gegenüber Kationen, die jedoch von der Konformation des Kanals abhängig war und in Anwesenheit zweiwertiger Kationen abnahm. Cu2 -Ionen führten ab Konzentrationen von 2 mM zu einem kompletten Stromblock. Ob dieser Effekt auf der Bildung einer Disulfidbrücke basiert und damit biochemische Ergebnisse bekräftigt werden, die eine Regulation durch das Thioredoxin-System im Chloroplasten vermuten lassen, konnte anhand der vorliegenden Ergebnisse nicht eindeutig geklärt werden.

京都大学 / 0048 / 新制・課程博士 / 博士(理学) / 甲第20212号 / 理博第4297号 / 新制||理||1617(附属図書館) / 京都大学大学院理学研究科生物科学専攻 / (主査)教授 鹿内 利治, 准教授 小山 時隆, 教授 長谷 あきら / 学位規則第4条第1項該当 / Doctor of Science / Kyoto University / DGAM

Euglena gracilis is a very ancient eukaryote whose chloroplast acquisition and evolution has been independent of higher plants. The organism is unique in possessing two de novo fatty acid synthases, a true multienzyme complex of great size in the cytosol and a plastid-localized type II fatty acid synthase composed of discrete enzymes and acyl carrier protein (ACP). The enzymology of the early steps of fatty acid biosynthesis differed in the Euglena type II fatty acid synthase compared to those of Escherichia coli and plants. The enzymes of Euglena participating in both priming and elongation reactions to form a new carbon-carbon bond were acetyl-CoA-ACP transacylase, malonyl-CoA-ACP transacylase, and β-ketoacyl-ACP synthase I. The effects of inhibitors on the three different enzymes were noted. All carbon-carbon bond formation was inhibited by cerulenin. Although neither fatty acid biosynthesis nor any of the isolated enzymes were sensitive to diisopropylphosphofluoridate, the three Euglena enzymes studied were sensitive to different sulfhydryl-alkylating agents. Acetyl-ACP supported fatty acid biosynthesis as effectively as did comparable amounts of ACPSH and acetyl-CoA. There was no evidence for a β-ketoacyl-ACP synthase III for priming such as has been reported in type II fatty acid synthase of higher plants and bacteria. The roles of the acetyl-CoA-ACP transacylase and β-ketoacyl-ACP synthase I appear to be unique in the type II fatty acid synthase of Euglena. Acetyl-CoA-ACP transacylase, malonyl-CoA-ACP transacylase, and β-ketoacyl-ACP synthase I were separated from one another and shown to have different molecular weights.

For neutral genes, uniparental inheritance is expected to reduce effective population size relative to biparentally inherited genes. In finite populations, the ensuing genetic drift can cause stronger spatial and temporal differentiation. An intrapopulation polymorphism in chloroplast DNA was used to examine relative spatial and temporal population structure of chloroplast and allozyme markers in the annual plant Phacelia dubia. There was significant differentiation among populations at chloroplast markers but not for allozyme loci. A fine-scale analysis showed significant structure among sites within populations for chloroplast markers and local heterozygote deficiencies at allozyme loci. These spatial analyses suggest that gene flow via pollen exceeds that via seed. Temporal variation in chloroplast markers, assessed over a 10-year period, was evident in two of four populations, and allozyme loci were characterized by temporal variation in rare-allele frequencies. Population structure appeared to be related to the intensity and type of human disturbance influencing each population. Habitat destruction promoted isolation and enhanced differentiation, whereas mowing increased seed dispersal and reduced differentiation for chloroplast markers. At this time, genetic drift appears to be the primary force shaping chloroplast gene frequencies.

Plants have two organelles outside the nucleus which carry their own DNA, mitochondria and chloroplasts. These organelles are descendants of bacteria that were engulfed by their host according to the endosymbiotic theory. Over time, DNA has been exchanged between these organelles and the nucleus. Two polymerases, DNA Polymerases Gamma I and II, are encoded in the nucleus and remain under nuclear control, but are transported into the mitochondria and chloroplasts. DNA polymerases gamma I and II are two organelle polymerases which have been studied through sequence analysis and shown to localize to both mitochondria and chloroplasts. Little has been done to characterize the activities of these polymerases. Work in tobacco showed the homology of these polymerases to each other and to DNA Polymerase I in bacteria. They have been characterized as being part of the DNA Polymerase A family of polymerases. In my research I have studied the effect of T-DNA insertions within the DNA Polymerase Gamma I and II genes. Since these DNA Polymerases are targeted to the mitochondria and chloroplasts, I studied the effect of knocking out these genes. A plant heterozygous for an insert in DNA Polymerase Gamma I grows slightly slower than wild type plants with an approximately 20% reduction in mitochondrial and chloroplast DNA copy number. A plant homozygous for an insert in this same gene has a drastic phenotype with stunted plants that grow to around 1 inch tall, with floral stems, and have an approximately 50-55% reduction in mitochondrial and chloroplast DNA copy number. Wild type plants can grow to a height of 12-18 inches with floral stems as a comparison. A plant heterozygous for an insert in the DNA Polymerase Gamma II gene grows slightly slower than wild type plants and has an approximately 15% reduction in mitochondrial DNA copy number and a 50% reduction in chloroplast DNA copy number. These plants also produce much less seed than do other mutants and wild type plants.

The vast majority of valuable agronomic traits are encoded polygenetically. Chloroplast genetic engineering offers an alternate approach to multigene engineering by allowing the insertion of entire pathways in a single transformation event, while being an environmentally friendly approach. Stable integration into the chloroplast genome and transcription of the phaA gene coding for β-ketothiolase was confirmed by Southern and northern blots. Coomassie-stained gel and western blots confirmed hyperexpression of β-ketothiolase in leaves and anthers, with high enzyme activity. The transgenic lines were normal except for the male sterile phenotype, lacking pollen. Scanning electron microscopy revealed a collapsed morphology of the pollen grains. Transgenic lines followed an accelerated anther developmental pattern, affecting their development and maturation, resulting in aberrant tissue patterns. Abnormal thickening of the outer wall, enlarged endothecium and vacuolation, decreased the inner space of the locules, affecting pollen grain and resulted in the irregular shape and collapsed phenotype. Reversibility of the male sterility phenotype was achieved by exposing the plants to continuous illumination, producing viable pollen and copious amounts of seeds. This is the first report of engineered cytoplasmic male sterility and offers a new tool for transgene containment for both nuclear and organelle genomes. Detailed characterization of transcriptional, posttranscriptional and translational processes of heterologous operons expressed via the chloroplast genome is reported here. Northern blot analyses performed on chloroplast transgenic lines harboring seven different heterologous operons, revealed that in most cases, only polycistronic mRNA was produced or polycistrons were the most abundant form and that they were not processed into monocistrons. Despite such lack of processing, abundant foreign protein accumulation was detected in these transgenic lines. Interestingly, a stable secondary structure formed from a heterologous bacterial intergenic sequence was recognized and efficiently processed, indicating that the chloroplast posttranscriptional machinery can indeed recognize sequences that are not of chloroplast origin, retaining its prokaryotic ancestral features. Processed and unprocessed heterologous polycistrons were quite stable even in the absence of 3'UTRs and were efficiently translated. Unlike native 5'UTRs, heterologous secondary structures or 5'UTRs showed efficient translational enhancement independent of any cellular control. Finally, we observed abundant read-through transcription in the presence of chloroplast 3'UTRs. Such read-through transcripts were efficiently processed at introns present within native operons. Addressing questions about polycistrons, as well as the sequences required for their processing and transcript stability are essential for future approaches in metabolic engineering. Finally, we have shown phytoremediation of mercury by engineering the mer operon via the chloroplast genome under the regulation of chloroplast native and heterologous 5'UTRs. These transgenenic plants hyperexpress were able to translate MerA and MerB enzymes to levels detectable by coomassie stained gel. The knowledge acquired from these studies offer guidelines for engineering multigene pathways via the chloroplast genome.