Species in Rhododendron subsection Fortunea are known for fragrant flowers, tolerance to severe climate, and for having an attractive form, all of great horticultural value. Many Fortunea species have been crossed to produce outstanding hybrids. R. fortunei is most widely hybridized, as it is the hardiest species with scented flowers. R. orbiculare and R. platypodum have interesting leaves, while R. sutchuense is a beautifully symmetrical large bush. To date, 26 species have been identified in Rhododendron subsection Fortunea, 25 of which occur in southwestern China. In fact, Fortunea occurs in more provinces of China than any other Rhododendron subsection. Taxonomically, species in Fortunea are defined by morphological characters such as size of the plant, leaf shape, and characteristics of the flower. Such a classification, however, gives no information about the relation of subsection Fortunea species to other subsections or the relationships between Fortunea species. The goal of this project is to determine the phylogenetic relationships within subsection Fortunea through cladistic analysis of DNA molecular data and to further investigate whether subsection Fortunea is a monophyletic group by comparing sequences with species in other subsections that are either sympatric or morphologically similar. To accomplish this goal, 38 samples representing 14 species and 9 subspecies of Fortunea Rhododendrons were collected at various locations in southwestern China. Selected DNA regions of nuclear genes RPC-1, E19 and RPB2-i were amplified and sequenced. Using these data and information from similar and sympatric Rhododendron subsections, I will infer phylogenies using parsimony, maximum likelihood and Bayesian algorithms. Preliminary results from 16 of the 31 samples indicate a closer relation between different species growing in the same location than between the same species in different locations. Whether this indicates a serious difficulty with Fortunea classification should be made clear with additional data.

A considerable amount of genetic variation exists among organisms of the same species, yet the variability of their phenotype is minimal. The protein Hsp90 has been shown to play a role in maintaining this phenotypic robustness by buffering genetic variation and keeping it phenotypically silent. When Hsp90 function is decreased by mutation or environmental stress, previously cryptic genetic variation is expressed, thereby significantly affecting phenotype. The existence of frequently occurring cryptic genetic variation together with a plausible natural release mechanism - environmental stress - likely facilitates evolutionary processes. Hsp90 allows phenotypes to remain robust over 1) the course of a lifetime, e.g. in the face of environmental changes such as weather or disease, and 2) over many generations, e.g. in the face of inherited genetic variants or mutations. Based on the properties of Hsp90, I hypothesize there are other proteins that similarly maintain phenotypic robustness and hope to identify the genes encoding them. The Queitsch lab has identified several candidate genes in the small RNA pathway that maintain robustness in seedling traits of the plant Arabidopsis thaliana. Currently, I am creating conditional knock-out mutants for these candidate genes, inducible in the presence of estradiol. I will test these genes for their ability to maintain phenotypic robustness by 1) determining the level of variance for a single quantitative trait amongst many individuals of each mutant genotype, 2) qualitatively identifying morphological abnormalities, and 3) testing the environmental responsiveness of these mutants across a temperature gradient. My work will contribute to a better understanding of the molecular mechanisms contributing to phenotypic robustness and will yield insights into the molecular underpinning of evolutionary processes.

Temporal and seasonal awareness in plants is an area of great interest in plant biology. An ability of plants to recognize and respond to the subtle changes in day length associated with seasonal change is dependent upon the function of specific “circadian clock regulated” proteins. To accurately measure the changes in day length, the regulation of the expression of these proteins is a crucial aspect. Here, we in the Imaizumi lab are interested in understanding day length sensing mechanisms. We have recently identified that the transcription factors belonging to the beta-helix-loop-helix (bHLH) superfamily play a role in the activation of the key regulator CONSTANS (CO) gene expression in Arabidopsis thaliana. CO in turn is responsible for the regulation of several downstream clock regulated genes that induce flowering. We are aware that other plant species have CO orthologs that also play a role in flowering time regulation, but the involvement of the bHLH transcription factor superfamily in its regulation is unknown. We have identified structural homologs to the Arabidopsis bHLH transcription factors within the genome of several different plant species, such as rice, maize, barley, wheat, poplar, and grape. Our experiment aims to identify the function of these bHLH homologs in CO regulation within Arabidopsis. Once isolated, we will transform the bHLH homologs into Arabidopsis and analyze the effects that these genes have on flowering regulation.

The maladaptive feature- altruism, sacrificing personal benefit for the advancement of the group or another individual, is not widely understood. Within altruistic species two prerequisites must be met. First, sacrificial organisms must contribute a resource benefiting their genetic relatives. Secondly, a population of organisms carrying self-destructive genes must maintain heterogeneity in expression (if all self-destruct the traits will be lost.) We are investigating the evolution of altruism with toxin-producing E. coli. The plasmid of this bacterium has three main features: toxin production (colicin), a colicin-specific immunity gene, and a lysis gene. Under conditions of stress, both the lysis and toxin gene are expressed, resulting in the manufacture of colicin and distribution through lysis. Bacteria lacking the colicin-specific immunity will die if exposed to the toxin. To fulfill the second prerequisite of heterogeneity there are three main strategies for genetically homogenous populations: stochasticly (environmentally independent), conditionally (environmentally dependent) or a combination of both. We induced conditional bistablity within the population through UV cues-- some cells are “coerced” into self-destructing at high rates while others benefit from their sacrifice. Evolving a conditionally self-destructive system, we (1) expose the experimental population to different enviornments (UV cue for suicidal response) and (2) “reward” genes responding to the UV cue through a public good (production of toxic colicins benefiting close relatives with immunity). We plan to evolve producer cells for approximately 300 generations to evaluate toxin production throughout by assaying the death rate of sensitive cells (DAP auxotroph). Comparing the proportion of surviving producer cells (proxy of fitness) within the evolved experimental treatment to those obtained within the evolved control treatments (greater death of sensisitve cells and survival of producer cells) than within our control treatments. We will then further explore the conditions favoring this strategy by altering various environmental factors.

There are 275,000 species of flowering plants, representing the most dramatic adaptive radiations in recent evolutionary history. This diversity is thought to be the result of co-evolution between the plants and their animal pollinators. However, we have almost no knowledge of the genes responsible for variation in traits that pollinators find attractive, such as flower color, size, shape, and nectar reward. We wish to find the genes that control floral traits by mutagenizing Mimulus lewisii, a species in a genus with tremendous variation in floral form. We have discovered and begun to characterize these genes by chemically mutagenizing Mimulus lewisii with ethylmethanesulfate, and screening for individuals that display relevant floral mutations. Mutants have been grouped by phenotype and tested for complementation to learn more about the nature of each gene as well as the number of genes that affect each trait. Traits we have found include suppressed petal expansion and petal fusion. In addition, flowers with either no pigment or nectar guides that were either enhanced or reduced were common. In nature, evolution in floral morphology by changes that arise by mutation is driven by specificity in pollinator-flower interaction. The traits we have observed which alter specific floral patterns known to guide pollinators have potential implications for pollinator attraction, and may elucidate the basis for evolutionary change in floral morphology.

Methylobacterium extorquens AM1 is a methylotrophic bacterium ubiquitous in the environment and in particular on the undersides of leaves. This organism is able to metabolize both single carbon compounds like methanol and multi carbon compounds such as succinate and pyruvate. M. extorquens has been studied for decades because of its potential for creating value added compounds from methanol, yet genes required for methylotrophic growth are still being discovered. The first step in methylotrophic growth is the oxidation of methanol by methanol dehydrogenase encoded by the mxaF genes. The M. extorquens genome contains two homologs of the mxaF gene, xoxF1 and xoxF2. Here we show that single mutations in the xoxF1 gene lead to both a decrease in growth on methanol and decreased methanol dehydrogenase activity. Surprisingly, when both xoxF1 and xoxF2 are mutated, AM1 is unable to grow on methanol and has no methanol dehydrogenase activity, identical to the phenotype of the mxaF mutant. Further studies revealed that the level of MxaF protein in the cell is greatly reduced in this xoxF1 xoxF2 mutant suggesting that XoxF1 and XoxF2 are required for either expression of mxaF or the stability of the MxaF protein.

Costa Rican soils are vulnerable to erosion and landslides due to heavy precipitation, compounded by agricultural replacement of natural forests, particularly in coffee fields, which are often on steep slopes. Erosion is combated by well-developed root systems and strong soil aggregates, held together by organic matter. The soil protein glomalin, dubbed the “soil glue,” is particularly effective at holding aggregates together, and persists in soils for decades due to high heat tolerance. This glycoprotein is produced exclusively by Arbuscular Mycorrhizal fungi (AM). We hypothesized that over-fertilization could decrease glomalin production in Costa Rican coffee farms, due to a reduction of AM fungi. We examined the effect of reduced fertilization on standing soil glomalin levels and soil glomalin production over a 30 day period. Within ten coffee farms in the Tarrazu region of Costa Rica, experimental plots were established with reduced fertilizer applications. Glomalin was extracted in 50 mM sodium citrate at 121° C from soil cores to assess standing levels (2 pooled cores/transect, 3 transects/treatment, 10 farms), and from fungal in-growth tubes that were buried to capture production over a 30 day period (1 core/transect, 3 transects/treatment, 10 farms). Glomalin was assessed using two methods: the Bradford method measures all heat-stable protein, while the ELISA assay uses a glomalin-specific antibody as a more specific measure. Both the Bradford and the ELISA assays showed that standing glomalin levels in soils were significantly higher with reduced fertilizer; however, there were no significant differences in glomalin production over 30 days. The rate of glomalin production was quite low relative to other studies, and standing levels of glomalin were negatively correlated with the age of coffee farms, leading to the conclusion that glomalin levels may be slowly declining each year, after conversion from tropical forest to coffee fields.