Refine
Has Fulltext
- yes (123) (remove)
Year of publication
- 2014 (123) (remove)
Document Type
- Journal article (75)
- Doctoral Thesis (48)
Keywords
- gene expression (5)
- Epigenetik (4)
- Maus (4)
- Bioinformatik (3)
- Genexpression (3)
- antibodies (3)
- ants (3)
- biodiversity (3)
- cancer (3)
- cytoskeleton (3)
Institute
- Theodor-Boveri-Institut für Biowissenschaften (123) (remove)
Sonstige beteiligte Institutionen
- DNA Analytics Core Facility, Biocenter, University of Würzburg, Würzburg, Germany (1)
- Department of Animal Ecology and Tropical Biology, University of Würzburg, Würzburg, Germany (1)
- Forschungsstation Fabrikschleichach (1)
- Institut für Tierökologie und Tropenbiologie (1)
- Interdisziplinäres Zentrum für Klinische Forschung (ZIKF), Würzburg (1)
- Klinische Mikrobiologie am Universitätsklinikum Erlangen (1)
- Technische Hochschule Wildau (1)
Localization microscopy is a class of super-resolution fluorescence microscopy techniques. Localization microscopy methods are characterized by stochastic temporal isolation of fluorophore emission, i.e., making the fluorophores blink so rapidly that no two are
likely to be photoactive at the same time close to each other. Well-known localization microscopy methods include dSTORM}, STORM, PALM, FPALM, or GSDIM. The biological community has taken great interest in localization microscopy, since it can enhance the resolution of common fluorescence microscopy by an order of magnitude at little experimental cost.
However, localization microscopy has considerable computational cost since millions of individual stochastic emissions must be located with nanometer precision. The computational cost of this evaluation, and the organizational cost of implementing the complex algorithms, has impeded adoption of super-resolution microscopy for a long time.
In this work, I describe my algorithmic framework for evaluating localization microscopy data.
I demonstrate how my novel open-source software achieves real-time data evaluation, i.e., can evaluate data faster than the common experimental setups can capture them.
I show how this speed is attained on standard consumer-grade CPUs, removing the need for computing on expensive clusters or deploying graphics processing units.
The evaluation is performed with the widely accepted Gaussian PSF model and a Poissonian maximum-likelihood noise model.
I extend the computational model to show how robust, optimal two-color evaluation is realized, allowing correlative microscopy between multiple proteins or structures. By employing cubic B-splines, I show how the evaluation of three-dimensional samples can be made simple and robust, taking an important step towards precise imaging of micrometer-thick samples.
I uncover the behavior and limits of localization algorithms in the face of increasing emission densities.
Finally, I show up algorithms to extend localization microscopy to common biological problems.
I investigate cellular movement and motility by considering the in vitro movement of myosin-actin filaments. I show how SNAP-tag fusion proteins enable imaging with bright and stable organic fluorophores in live cells. By analyzing the internal structure of protein clusters, I show how localization microscopy can provide new quantitative approaches beyond pure imaging.
This thesis explores the influence of social and environmental cues on the nest building behavior of leaf-cutting ants. Especially, the investigations are aimed at evaluating the mechanisms of nest building and how the nest environment can spatially guide building responses that lead to an adaptive nest architecture. The emergence of nest chambers in the nest of the leaf-cutting ant Acromyrmex lundi were evaluated. Rather than excavating nest chambers in advance, at places where workers encounter suitable environmental conditions for brood and fungus rearing, these items have to be present at a site. When presented in the laboratory with a choice between two otherwise identical digging sites, offering suitable environmental conditions, but one containing brood, the workers displayed a higher excavation activity at the site where they encountered the putative content of a chamber. The shape of the excavated cavity was also more round and chamber-like. It is concluded that leaf-cutting ants respond to social cues during nest building. Excavation is a costly process and colonies have to spend a part of their energy stores on nest building, so that regulatory responses for the control of nest excavation are expected to occur. Worker density at the beginning of the digging process influenced digging activity while the presence of in-nest stores did not. Stored brood and fungus did however influence the architecture of the excavated nest, leading to the excavation of larger chambers and smaller tunnels. While self-organized mechanisms appear to be involved in the nest building process, the social cues of the ants’ environment during building clearly influence the nest architecture and lead to an adjustment of the nest size to the current space needs of the colony. Workers secondarily regulated nest size by the opportunistic refilling of unused space with excavated soil pellets. As the ants should provide suitable conditions for brood and fungus rearing, they should show a behavioral response to CO2 concentrations, as the gas is known to hinder fungus respiration. Workers of A. lundi did indeed avoid high CO2-levels for fungus rearing but actually preferred CO2-values in the range encountered close to the soil surface, where this species excavates their nests. However, different CO2-levels did not affect their excavation behavior. While fungus chambers make up part of a leaf-cutting ant nest, most leaf-cutting ants of the genus Atta also spent part of the colony’s energy on excavating large, voluminous chambers for waste disposal, rather than scattering the material aboveground. It is expected that leaf-cutting ants also show environmental preferences for waste management. In experiments Atta laevigata workers preferred deposition in a warm and dry environment and showed no preference for specific CO2-levels. The continued accumulation of waste particles in a waste chamber seems to be based on the use of volatiles. These originate from the waste itself, and seem to be used as an orientation cue by workers relocating the material. The ensuing large accumulation of waste at one site should result in the emergence of more voluminous chambers for waste disposal.
The intracellular pathogen Chlamydia is the causative agent of millions of new infections per year transmitting diseases like trachoma, pelvic inflammatory disease or lymphogranuloma venereum. Undetected or recurrent infections caused by chlamydial persistence are especially likely to provoke severe pathologies. To ensure host cell survival and to facilitate long term infections Chlamydia induces anti-apoptotic pathways, mainly at the level of mitochondria, and restrains activity of pro-apoptotic proteins. Additionally, the pathogen seizes host energy, carbohydrates, amino acids, lipids and nucleotides to facilitate propagation of bacterial progeny and growth of the chlamydial inclusion.
At the beginning of this study, Chlamydia-mediated apoptosis resistance to DNA damage induced by the topoisomerase inhibitor etoposide was investigated. In the course of this, a central cellular protein crucial for etoposide-mediated apoptosis, the tumour suppressor p53, was found to be downregulated during Chlamydia infections. Subsequently, different chlamydial strains and serovars were examined and p53 downregulation was ascertained to be a general feature during Chlamydia infections of human cells. Reduction of p53 protein level was established to be mediated by the PI3K-Akt signalling pathway, activation of the E3-ubiquitin ligase HDM2 and final degradation by the proteasome. Additionally, an intriguing discrepancy between infections of human and mouse cells was detected. Both activation of the PI3K-Akt pathway as well as degradation of p53 could not be observed in Chlamydia-infected mouse cells. Recently, production of reactive oxygen species (ROS) and damage to host cell DNA was reported to occur during Chlamydia infection. Thus, degradation of p53 strongly contributes to the anti-apoptotic environment crucial for chlamydial infection.
To verify the importance of p53 degradation for chlamydial growth and development, p53 was stabilised and activated by the HDM2-inhibiting drug nutlin-3 and the DNA damage-inducing compound etoposide. Unexpectedly, chlamydial development was severely impaired and inclusion formation was defective. Completion of the chlamydial developmental cycle was prevented resulting in loss of infectivity. Intriguingly, removal of the p53 activating stimulus allowed formation of the bacterial inclusion and recovery of infectivity. A similar observation of growth recovery was made in infected cell lines deficient for p53.
As bacterial growth and inclusion formation was strongly delayed in the presence of activated p53, p53-mediated inhibitory regulation of cellular metabolism was suspected to contribute to chlamydial growth defects. To verify this, glycolytic and pentose phosphate pathways were analysed revealing the importance of a functioning PPP for chlamydial growth. In addition, increased expression of glucose-6-phosphate dehydrogenase rescued chlamydial growth inhibition induced by activated p53. The rescuing effect was even more pronounced in p53-deficient cells treated with etoposide or nutlin-3 revealing additional p53-independent aspects of Chlamydia inhibition. Removal of ROS by anti-oxidant compounds was not sufficient to rescue chlamydial infectivity. Apparently, not only the anti-oxidant capacities of the PPP but also provision of precursors for nucleotide synthesis as well as contribution to DNA repair are important for successful chlamydial growth.
Modulation of host cell signalling was previously reported for a number of pathogens. As formation of ROS and DNA damage are likely to occur during infections of intracellular bacteria, several strategies to manipulate the host and to inhibit induction of apoptosis were invented. Downregulation of the tumour suppressor p53 is a crucial point during development of Chlamydia, ensuring both host cell survival and metabolic support conducive to chlamydial growth.
Cell growth and cell division are two interconnected yet distinct processes. Initiation of proliferation of central brain progenitor cells (neuroblasts) after the late embryonic quiescence stage requires cell growth, and maintenance of proper cell size is an important prerequisite for continuous larval neuroblast proliferation. Beside extrinsic nutrition signals, cell growth requires constant supply with functional ribosomes to maintain protein synthesis.
Mutations in the mushroom body miniature (mbm) gene were previously identified in a screen for structural brain mutants. This study focused on the function of the Mbm protein as a new nucleolar protein, which is the site of ribosome biogenesis. The comparison of the relative expression levels of Mbm and other nucleolar proteins in different cell types showed a pronounced expression of Mbm in neuroblasts, particularly in the fibrillar component of the nucleolus, suggesting that in addition to nucleolar components generally required for ribosome biogenesis, more neuroblast specific nucleolar factors exist. Mutations in mbm cause neuroblast proliferation defects but do not interfere with cell polarity, spindle orientation or asymmetry of cell division of neuroblasts. Instead a reduction in cell size was observed, which correlates with an impairment of ribosome biogenesis. In particular, loss of Mbm leads to the retention of the small ribosomal subunit in the nucleolus resulting in decreased protein synthesis. Interestingly, the defect in ribosome biogenesis was only observed in neuroblasts. Moreover, Mbm is apparently not required for cell size and proliferation control in wing imaginal disc and S2 cells supporting the idea of a neuroblast-specific function of Mbm.
Furthermore, the transcriptional regulation of the mbm gene and the functional relevance of posttranslational modifications were analyzed. Mbm is a transcriptional target of dMyc. A common feature of dMyc target genes is the presence of a conserved E-box sequence in their promoter regions. Two E-box motifs are found in the vicinity of the transcriptional start site of mbm. Gene reporter assays verified that only one of them mediates dMyc-dependent transcription. Complementary studies in flies showed that removal of dMyc function in neuroblasts resulted in reduced Mbm expression levels.
At the posttranslational level, Mbm becomes phosphorylated by protein kinase CK2. Six serine and threonine residues located in two acidic amino acid rich clusters in the C-terminal half of the Mbm protein were identified as CK2 phosphorylation sites.
Mutational analysis of these sites verified their importance for Mbm function in vivo and indicated that Mbm localization is controlled by CK2-mediated phosphorylation.
Although the molecular function of Mbm in ribosome biogenesis remains to be determined, the results of this study emphasize the specific role of Mbm in neuroblast ribosome biogenesis to control cell growth and proliferation.
Malaria is a challenging infection with increasing and wide-spread treatment failure risk due to resistance. With a estimated death toll of 1-3 Million per year, most cases of Malaria affect children under the age of five years in Sub-Saharan Africa. In this thesis, I analyse the current status of malaria control (focussing on diagnosis and therapy) in Burkina Faso to show how this disease burdens public health in endemic countries and to identify possible approaches to improvement. MB is discussed as a therapeutic option under these circumstances.
Burkina Faso is used as a representative example for a country in Sub-Saharan Africa with high endemicity for malaria and is here portrayed, its health system characterised and discussed under socioeconomic aspects.
More than half of this country’s population live in absolute poverty. The burden that malaria, especially treatment cost, poses on these people cannot be under-estimated.
A retrospective study of case files from the university pediatric hospital in Burkina Faso’s capital, Ouagadougou, shows that the case load is huge, and especially the specific diagnosis of severe malaria is difficult to apply in the hospital’s daily routine. Treatment policy as proposed by WHO is not satisfactorily implemented neither in home treatment nor in health services, as data for pretreatment clearly show.
In the face of growing resistance in malaria parasites, pharmacological combination therapies are important. Artemisinins currently are the last resort of malaria therapy. As I show with homology models, even this golden bullet is not beyond resistance development. Inconsidered mass use has rendered other drugs virtually useless before. Artemisinins should thus be protected similar to reserve antibiotics against multi-resistant bacteria.
There is accumulating evidence that MB is an effective drug against malaria. Here the biological effects of both MB alone and in combination therapy is explored via modeling and experimental data. Several different lines of MB attack on Plasmodium redox defense were identified by analysis of the network effects. Next, CQ resistance based on Pfmdr1 and PfCRT transporters as well as SP resistance were modeled in silico. Further modeling shows that MB has a favorable synergism on antimalarial network effects with these commonly used antimalarial drugs, given their correct application.
Also from the economic point of view MB shows great potential: in terms of production price, it can be compared to CQ, which could help to diminuish the costs of malaria treatment to affordable ranges for those most affected and struk by poverty.
Malaria control is feasible, but suboptimal diagnosis and treatment are often hindering the achievment of this goal. In order to achieve malaria control, more effort has to be made to implement better adjusted and available primary treatment strategies for uncomplicated malaria that are highly standardised. Unfortunately, campaigns against malaria are chronically underfinanced. In order to maximize the effect of available funds, a cheap treatment option is most important, especially as pharmaceuticals represent the biggest single matter of expense in the fight against malaria.
Dynamic interactions and their changes are at the forefront of current research in bioinformatics and systems biology. This thesis focusses on two particular dynamic aspects of cellular adaptation: miRNA and metabolites.
miRNAs have an established role in hematopoiesis and megakaryocytopoiesis, and platelet miRNAs have potential as tools for understanding basic mechanisms of platelet function. The thesis highlights the possible role of miRNAs in regulating protein translation in platelet lifespan with relevance to platelet apoptosis and identifying involved pathways and potential key regulatory molecules. Furthermore, corresponding miRNA/target mRNAs in murine platelets are identified. Moreover, key miRNAs involved in aortic aneurysm are predicted by similar techniques. The clinical relevance of miRNAs as biomarkers, targets, resulting later translational therapeutics, and tissue specific restrictors of genes expression in cardiovascular diseases is also discussed.
In a second part of thesis we highlight the importance of scientific software solution development in metabolic modelling and how it can be helpful in bioinformatics tool development along with software feature analysis such as performed on metabolic flux analysis applications. We proposed the “Butterfly” approach to implement efficiently scientific software programming. Using this approach, software applications were developed for quantitative Metabolic Flux Analysis and efficient Mass Isotopomer Distribution Analysis (MIDA) in metabolic modelling as well as for data management. “LS-MIDA” allows easy and efficient MIDA analysis and, with a more powerful algorithm and database, the software “Isotopo” allows efficient analysis of metabolic flows, for instance in pathogenic bacteria (Salmonella, Listeria). All three approaches have been published (see Appendices).
In contrast to c-Myc, a deregulated expression of the MYCN gene is restricted to human neuroendocrine tumours. In most cases, the excessive activity of N-Myc results from a MYCN amplification. In neuroblastoma, amplification of MYCN is a predictor of poor prognosis and resistance to therapy. The inability to target the N-Myc protein directly necessitates the search for alternative targets. This project aimed at identifying genes specifically required for growth and survival of cells that express high levels of N-Myc using high-throughput shRNA screening combined with next generation sequencing. The identification and analysis of these genes will shed light on functional interaction partners of N-Myc.
We screened a shRNA library containing 18,327 shRNAs and identified 148 shRNAs, which were selectively depleted in the presence of active N-Myc. In addition, shRNAs targeting genes that are involved in p53 and ARF turnover and apoptosis were depleted in the cell population during the screen. These processes are known to affect N-Myc-mediated apoptosis. Consequently, these results biologically validated the screen. The 148 shRNAs that showed a significant synthetic lethal interaction with high levels of N-Myc expression were further analysed using the bioinformatics program DAVID. We found an enrichment of shRNAs that target genes involved in specific biological processes. For example, we validated synthetic lethal interactions for genes such as, THOC1, NUP153 and LARP7, which play an important role in the process of RNA polymerase II-mediated transcription elongation. We also validated genes that are involved in the neddylation pathway.
In the screen we identified Cullin 3, which is a component of the BTB-CUL3-Rbx1 ubiquitin ligase that is involved in the turnover of Cyclin E. Depletion of cullin 3 and activation of N-Myc was found to synergistically increase Cyclin E expression to supraphysiological levels, inducing S-phase arrest and a strong DNA damage response.
Together with results from a proteomics analysis of N-Myc associated proteins, our results lead us to the following hypothesis: In a neuroblastoma cell, the high levels of N-Myc result in a conflict between RNA polymerase II and the replication machinery during S-phase. The newly identified interaction partners of N- Myc are required to solve this conflict. Consequently, loss of the interaction leads to a massive DNA damage and the induction of apoptosis. In addition, inhibition or depletion of the essential components of the neddylation pathway also results in an unresolvable problem during S-phase.
Die Einführung der Fluoreszenzmikroskopie ermöglicht es, Strukturen in Zellen spezifisch und mit hohem Kontrast zu markieren und zu untersuchen. Da die Lichtmikroskopie jedoch in ihrer Auflösung begrenzt ist, bleiben Strukturinformationen auf molekularer Ebene verborgen. Diese als Beugungsgrenze bekannte Limitierung, kann mit modernen Verfahren umgangen werden. Die Lokalisationsmikroskopie nutzt hierfür photoschaltbare Fluorophore, deren Fluoreszenz räumlich und zeitlich separiert wird, um so einzelne Fluorophore mit
Nanometer-Genauigkeit lokalisieren zu können. Aus tausenden Einzelmolekül-Lokalisationen wird ein künstliches, hochaufgelöstes Bild rekonstruiert. Die
hochauflösende Mikroskopie ist grade für die Lebendzell-Beobachtung ein wertvolles Werkzeug, um subzelluläre Strukturen und Proteindynamiken jenseits der Beugungsgrenze unter physiologischen Bedingungen untersuchen zu können.
Als Marker können sowohl photoaktivierbare fluoreszierende Proteine als auch photoschaltbare organische Fluorophore eingesetzt werden. Während die
Markierung mit fluoreszierenden Proteinen einfach zu verwirklichen ist, haben organische Farbstoffe hingegen den Vorteil, dass sie auf Grund der höheren Photonenausbeute eine präzisere Lokalisation erlauben. In lebenden Zellen wird die Markierung von Strukturen mit synthetischen Fluorophoren über sogenannte
chemische Tags ermöglicht. Diese sind olypeptidsequenzen, die genetisch an das Zielprotein fusioniert werden und anschließend mit Farbstoff-gekoppelten Substraten gefärbt werden. An der Modellstruktur des Histonproteins H2B
werden in dieser Arbeit Farbstoffe in Kombination mit chemischen Tags identifiziert, die erfolgreich für die Hochauflösung mit direct stochastic optical
reconstruction microscopy (dSTORM) in lebenden Zellen eingesetzt werden können. Für besonders geeignet erweisen sich die Farbstoffe Tetramethylrhodamin,
505 und Atto 655, womit der gesamte spektrale Bereich vertreten ist. Allerdings können unspezifische Bindung und Farbstoffaggregation ein Problem bei der effizienten Markierung in lebenden Zellen darstellen. Es wird
gezeigt, dass die Beschichtung der Glasoberfläche mit Glycin die unspezifische Adsorption der Fluorophore erfolgreich minimieren kann. Weiterhin wird der
Einfluss des Anregungslichtes auf die lebende Zelle diskutiert. Es werden Wege beschrieben, um die Photoschädigung möglichst gering zu halten, beispielsweise
durch die Wahl eines Farbstoffs im rotem Anregungsbereich.
Die Möglichkeit lebende Zellen mit photoschaltbaren organischen Fluorophoren spezifisch markieren zu können, stellt einen großen Gewinn für die Lokalisationsmikroskopie dar, bei der ursprünglich farbstoffgekoppelte Antikörper zum Einsatz kamen. Diese Markierungsmethode wird in dieser Arbeit eingesetzt, um
das Aggregationsverhalten von Alzheimer verursachenden -Amyloid Peptiden im Rahmen einer Kooperation zu untersuchen. Es werden anhand von HeLa Zellen verschiedene beugungsbegrenzte Morphologien der Aggregate aufgeklärt. Dabei wird gezeigt, dass intrazellulär vorhandene Peptide größere Aggregate formen als die im extrazellulären Bereich. In einer zweiten Kollaboration wird mit Hilfe des photoaktivierbaren Proteins
mEos2 und photoactivated localization microscopy (PALM) die strukturelle Organisation zweier Flotillinproteine in der Membran von Bakterien untersucht.
Diese Proteine bilden zwei Cluster mit unterschiedlichen Durchmessern, die mit Nanometer-Genauigkeit bestimmt werden konnten. Es wurde außerdem festgestellt, dass beide Proteine in unterschiedlichen Anzahlen im Bakterium
vorliegen.
The synaptonemal complex (SC) is a highly conserved structure in sexually reproducing organism. It has a tripartite, ladder-like organization and mediates the stable pairing, called synapsis, of the homologous chromosomes during prophase of meiosis I. Failure in homolog synapsis result in aneuploidy and/or apoptosis of the developing germ cells.
Since 1956, the SC is subject of intense research and its presence was described in various species from yeast to human. Its structure was maintained during millions of years of evolution consist-ing of two parallel lateral elements (LEs), joined by numerous transverse filaments (TFs) which run perpendicular to the LEs and an electron dense central element (CE) in the middle of the SC. Individual protein components, however, were characterized only in few available model organ-isms, as for example Saccharomyces cerevisiae, Arabidopsis thaliana, Drosophila melanogaster, Ceanorhabditis elegans and Mus musculus. Rather unexpectedly, these characterizations failed to detect an evolutionary homology between the protein components of the different SCs. This fact challenged the general idea of a single origin of the SC in the evolution of meiosis and sexual reproduction.
This thesis now addressed itself to the task to unravel the discrepancy between the high conser-vation of the SC structure and its diverse and apparently non-homologous protein composition, focusing on the animal kingdom. It is the first study dealing with the evolution of the SC in Meta-zoa and demonstrates the monophyly of the mammalian SC components in metazoan species. The thesis demonstrates that at least four out of seven murine SC proteins emerged in Eumeta-zoa at the latest and have been likewise part of an ancient SC as it can be found in the present-day cnidarian species Hydra. This SC displays the common organization and already possesses the minimal protein kit corresponding to the three different structural domains: LEs, TFs and the CE. Additionally, the individual phylogenies of the murine SC proteins revealed the dynamic evolu-tionary history of the ancient SC. Further components were added during the diversification of Bilateria and vertebrates while ancestral proteins likely duplicated in the vertebrate lineage and diversified or got lost in the branch leading to ecdysozoan species. It is hypothesized that the apparently non-homologous SC proteins in D. melanogaster and C. elegans actually do derive from the ancient SC proteins but diversified beyond recognition during the fast evolution of Ar-thropoda and Nematoda.
The study proposes Hydra as an alternative invertebrate model system for meiosis and SC re-search to the standard organisms D. melanogaster and C. elegans. Recent results about the cni-darian SC as well as the possible application of standard methods is discussed and summarized in the concluding section.
1. Pollination of sexually reproducing plants requires pollen transfer agents, which can be biotic, abiotic or a combination of biotic and abiotic agents. The dominance of one of pollination system in wild plant communities depends on climatic factors and/or degrees of anthropogenic influences, which have effects on pollinator diversity and pollination function. Anthropogenic activities and climate change are also considered as main causes of ongoing invasion of invasive species into wild and managed habitats which can bring up competition for pollinators with possible negative consequences for the reproduction of co-occurring native plant species.
2. The study aimed to determine pollination systems and pollination limitation of invasive and native plant communities in natural savannah between 870 – 1130 m and semi-natural (managed) grassland between 1300 – 1750 m above sea level; effects of flower density and pollinator abundance on seed production of cross-pollinated and self-pollinated plants; and relationships of bee abundance and the proportion of cross- pollinated plants at the southern slope of Mount Kilimanjaro, Tanzania.
3. Pollinator-exclusion, open pollination and supplemental hand-pollination treatments were applied to 27 plant species in savannah and grassland habitats. Flowers were counted in each clusters based upon their species. Pollinators were sampled by using pan traps. Information-theory-based multi-model averaging and generalized linear mixed effects models were used to identify and analyze the effects of flower density, pollinator abundance, pollination treatments and habitat types on seed production. Regression models were used to determine relationships of altitude with bee abundance, and with proportion of cross-pollinated plants.
4. My results show that mean seed numbers of native plants were significantly lower in pollinator-exclusion treatments than in open-pollination treatments, indicating their reliance on pollinators for reproductive success. In contrast, seed numbers of invasive plants were similar in pollinator-exclusion and open-pollination treatments, demonstrating an ability of reproduction without pollinators. Despite of higher levels of self-pollination in invasive plants, supplemental hand-pollination treatments revealed pollen limitation in grassland and marginally in savannah habitats. There were no significant difference in seed numbers between supplemental hand pollination and open pollination treatments of native plant communities in savannah and grassland, which indicates no pollination limitation in the studied ecological system for native communities. Besides, grassland plants produced comparatively more seeds than savannah plants, however seeds in grasslands were lighter than those of the savannah which may be due to nutrient limitation in grassland.
5. I found 12 cross-pollinated and 15 self-pollinated plants along altitudinal gradient after comparing seeds from pollinator-excluded and open-pollinated experiments. I also found that proportions of cross-pollinated plants and bee abundance simultaneously decreased with increasing altitude. All cross-pollinated plants were native and grew in savannah habitats, with an exception of one species.
6. Neither effects of focal flower density nor a significant interaction between focal flower densities and bee abundance for self-pollinated plants were observed. However, there were effects of focal flower densities and interactions of flower density with bee abundance for cross-pollinated plants. Non-focal flower density has no significant effects on seed production of cross-pollinated and self-pollinated plants.
7. The results show that native plants depend more on cross-pollination than invasive plants, despite of most native plants in managed habitat (grassland) rely on self-pollination for reproduction. The tendency of having more cross-pollinated plants in natural savannah which are in low altitude coincides with other finding that the cross-pollinated plants and bee abundance simultaneously decrease with increasing altitude. Therefore, our findings support the hypotheses that self-fertilization of flowering plants increases with increasing altitude, and pollinator limitation is most pronounced in managed or disturbed habitats. Despite of reduction of pollinators in grassland, only invasive plants experience pollen limitation, which may be due to poor integration with available pollinator networks.
8. I also found bee abundance and flower density are not the main pollination factors required by self-pollinated plants during reproduction. However, focal flower density, which influences pollinator diversity, is more applicable to cross-pollinated plants. Climate change and anthropogenic activities in natural habitats are factors that influence pollinator abundance and functioning, which lead to a shift of mating systems in plant communities so as to assure their reproduction.