Thursday, July 9, 2015

Enhanced saccharification of lignocellulosic biomass in Myb61a knock-down mutants and by pretreatment with ammonium hydroxide

Enhanced saccharification of lignocellulosic biomass in Myb61a knock-down mutants and by pretreatment with ammonium hydroxide

JILLIAN C. VAUGHT
Department of Plant Biology, The University of Oklahoma, Norman, OK

30 April 2015

Abstract

Lignocellulosic material is highly abundant in agricultural waste. Plant cell walls are dense networks of lignin, highly cross-linked with microfibrils of sugar polysaccharides. Sugar found bound in these cell walls could be used to produce a sustainable amount of biofuel and reduce the carbon footprint. Unfortunately, the strong network of sugar polysaccharides makes cell walls resistant to degradation. The deconstruction of lignocellulose to its component sugars is energy intensive. However, a combination of new genetic targets for cell wall mutations and methods of biomass pretreatment are promising for creating a more effective protocol for sugar extraction. Previous data from a wheat straw deconstruction indicates that biomass pretreated with 16.8% NH4OH is 3-fold more digestible. The result of my deconstruction experiment supports these findings in rice straw, showing up to a 9-fold increase in glucose sugar release, or saccharification, when pretreated with 15% NH4OH. In addition to pretreatment methods, targeting the cell wall regulating gene, Myb61a, to produce a knock-down mutant results in biomass with up to 3-fold increase in saccharification. A combination of basic pretreatment (pH 11-12) and genetic modifications to rice could be key to producing a source of digestible biomass for biofuels.

Introduction

Plant cell walls contain a strong network of sugar chains bound in lignocellulose. Sugar released from the cell walls of plants can be converted into biofuels by means of yeast or other microbes that undergo fermentation (1). Currently, the industrial burning of fossil fuels explains the buildup of atmospheric carbon dioxide, creating an immense carbon footprint (2). Using plant biomass to harvest sugar for biofuels has the potential to be a sustainable source of energy. With a global total of 23% of agricultural waste in rice straw alone, it is a promising candidate for harvestable biomass (3). Unfortunately, current methods of extracting sugar from plant cell walls are expensive and energy intensive. Barriers to inefficient sugar extraction include both understanding the mechanisms by which plants build up cell walls and determining pretreatment methods, which enhance the digestibility of cell walls by cellulase enzymes.

The composition of plant cell walls varies among and within plant species. Rice and other monocots exhibit distinct cell wall characteristics including remarkably developed secondary cell walls. The main components of secondary cell walls are cellulose, hemicellulose, and lignin (4). Collectively, these components are referred to as lignocellulose. Cellulose is composed of linear chains of glucose and can make up nearly 50% of the cell wall. Lignin is deposited in plant cells during the maturation of the plant to maintain structural support for the growing tissues. Both cellulose and lignin are dense and chemically cross-linked within the cell walls (1). This dense network of lignin and cellulose contributes to the recalcitrance, or the resistance to digestibility, of secondary cell walls. Highly recalcitrant cell walls are difficult to deconstruct and therefore the amount sugar release from harvested biomass is consistently low.

One approach to decreasing the recalcitrance of cell walls is identifying genes involved in cell wall synthesis. One family of MYB proteins has been found to bind to structural genes and other transcription factors involved in cell wall synthesis (4). Of these MYB proteins, MYB61 in Arabidopsis has been found to regulate the closure of the stomatal aperture (5). Additional data finds MYB61 in Arabidopsis to be involved in the deposition of mucilage (6). In unpublished data of a co-expression network, Myb61a is found to be co-expressed with four cell wall biosynthesis enzymes: COMT1, 4CL3, AT5, and AT4 (Zhao and Bartley, unpublished). Data from this co-expression network suggests that Myb61a in rice plants is a transcriptional activator involved in the regulation of cell wall synthesis. Preliminary evidence leads to the hypothesis that if Myb61a is knocked-down in rice plants, then the biomass will be more digestible than wild-type rice straw when deconstructed under the standard conditions described in the Materials and Methods section. If Myb61a is down-regulated then it will not be able to transcriptionally activate other cell wall genes involved in the biosynthesis of the cell wall, leaving the walls of mutant rice plants less cross-linked and less dense compared to wild-type cell walls. Data from the deconstruction of mutant rice straw in this experiment finds that Myb61a knock-down mutants have a 1.2-fold increase in digestibility compared to the wild-type.


In addition to genetically modifying rice plants, modifications to the deconstruction process itself can facilitate increased glucose yield. The process to converting biomass to biofuel can be divided into three main steps. First, biomass is produced and harvested from a plant source such as rice. Next, the destruction of biomass includes pulverization of materials and the breakdown of lignocellulose into its component sugars such as glucose. The breakdown of lignocellulose is catalyzed by expensive enzymes that are specific to the components of lignocellulose being targeted for sugar release. Examples of these enzymes include cellulase, beta-glucosidase, and xylanase. Lastly, the sugar released from lignocellulose by enzymes is fed to microbes that undergo fermentation to produce biofuels such as ethanol. Experimental evidence suggests that modifications to the deconstruction process can enhance sugar release from biomass. One crucial modification to the protocol is pretreating the pulverized biomass with a harsh acid or base before the biomass is subjected to cell wall enzymes.

The chemical pretreatment of the pulverized biomass has the potential to open up the structure of lignocellulose fibers, making them accessible to enzymes (7). Pretreatment chemicals must not only modify the matrix of the lignocellulose complex, but also be affordable. Previous experiments used 16.8% NH4OH to pretreat wheat straw. The results of this experiment show a 3-fold increase in saccharification, or sugar yield, by pretreated biomass (8). Based on data from this experiment, I hypothesized that pretreating rice straw with 15% NH4OH will increase the glucose yield compared to biomass under standard deconstruction with 0.1M Citrate Buffer. If rice straw is subjected to 15% NH4OH (ammonium hydroxide), the strong base will decrease the cross-linking of lignin fibers and decrease cellulose crystallinity, allowing enzymes to access the glucose polysaccharides. In this experiment, ammonium hydroxide pretreatment increased glucose yields 1.9-fold in pretreated wild-type rice straw compared to rice straw deconstructed under standard conditions with 0.1M Citrate Buffer.


Materials and Methods

Preparation of Biomass
Rice straw from Myb61a-1 knock-down mutants and negative segregant (wild-type) was dried at approximately 45° C. Straw was chopped and ground to a uniform size to produced milled rice straw.

Pretreatment of Milled Rice Straw
To prepare the biomass for enzymatic digestion, samples were pre-treated with either a strong base or a weak acid. Milled rice straw samples between approximately 5-7 mg were weighed and assigned to screw cap tubes. The tubes were organized based on pre-treatment and biomass genotype (Table S1). Four technical replicates of both mutant and wild-type biomass were pre-treated with 15% ammonium hydroxide (NH4OH) or citrate buffer. Biomass was gently shaken in 500 ul of 100mM Citrate Buffer or NH4OH for 5 minutes. Samples were then incubated at 99° C for one hour. After one hour, samples were removed and cooled.

Neutralization of Pre-treated Rice Straw
According to the Novoenzyme Biomass Kit protocol, samples must be neutralized to a pH of 4.5-6.5 for the cellulase complex and beta-glucosidase enzymes to perform. Samples pre-treated with NH4OH were neutralized by adding 2-5 ul of 4M NaCl until the pH neutralized to approximately a pH of 5.00.

Saccharification
To digest the lignocellulosic material into glucose, two enzymes supplied by Novoenzyme Biomass Kit are added. Stock solutions of NS50013 (cellulase) at a 5-fold dilution and NS50010 (B-glucosidase) at a 100-fold dilution were mixed at a 1:1 ration. Four ul of stock enzyme solution were added to all neutralized samples.

Time Points
To determine glucose release over time, a 30 ul aliquot was removed from each spun down test tube and added to a thermocycle strip tube. Thermocycle strips were maintained in a freezer. Samples were incubated at 50° C. Time points were taken at 0, 1, 6, 22, 25.5, and 46 hours in the first experiment and at 0, 1, 6, and 19.5 hours in the follow-up experiment. 

DNS Assay
To quantify the amount of sugar released over time from each time point, 90 ul of DNS reagent was added to each thermocycle strip well. Aliquots of dilutions for the standard curve were also added to the strip. Strips were covered to prevent evaporation and incubated at 95° C for 6 minutes in the thermocycler machine. Samples were then transferred to a microtiter plate. Absorbance of each sample was read at 540 nm by a Biotek plate reader.

Standard Curve
The standard curve was prepared by diluting 0.1M glucose with citrate buffer to known concentrations. Thirty ul of each dilution was added to the thermocyler plate and absorbance was determined. Microsoft Excel was used to plot absorbance vs. concentration. A linear range line was fit to the data and the trendline equation as well as R2 values were determined. The standard curve line was used to calculate the amount of sugar released in mg at each absorbance value (Figure S2).

Statistical Analysis
Sugar release in mg/mL was calculated at each absorbance using the linear trendline generated by the standard curve. Sugar release from each sample was divided by the amount of biomass in mg originally added to each test tube. Sugar release in mg/mg was averaged across each of the four technical replicates and graphed in a scatter plot in Excel. Standard deviation for each sample was calculated across the technical replicate at each time point and added to the graph using errors bars. A two-tailed student’s t-tests was used to calculate the significance of the final time points from each treatment sample. A t-test p-value of 0.05 or less was considered significant.

Results

Part I: Myb61a knock-down mutant rice straw shows enhanced saccharification.

Myb61a knock-down mutant rice straw harvested from a single greenhouse rice mutant plant shows enhanced glucose release compared to wild-type, negative segregant rice plants. Biomass from the rice straw of a mutant and wild-type plants was dried and ground to uniform size. A standard deconstruction method and DNS assay was used to measure glucose release over a period of 46 hours. Milled rice straw from a mutant and wild-type plants was incubated in 0.1M Citrate Buffer for one hour. Cellulase (CTec2) and beta-glucosidase (NS50010) enzymes were then allowed to begin digesting the biomass material. Glucose release in mg per mg of biomass were measured with DNS assay over six time points (Figure 1). As expected, glucose release from the Myb61a mutant rice straw displayed greater glucose release than the pool of wild-type rice straw.


















Figure 1. Myb61a knock-down mutant (MT) rice straw pretreated with Citrate Buffer (CT) shows increased glucose release compared to wild-type (WT). Ground biomass from Myb61a mutants and negative segregants was pretreated with 0.1M Citrate Buffer. Cellulase complex (CTec2) and B-glucosidase (NS50010) enzymes were added to the pre-treated biomass. Myb61a mutant biomass was taken from a single rice plant and wild-type biomass was pooled from all negative segregant rice plants. DNS assay measured glucose release in mg over a period of 46 hours. Mutant rice straw shows a 3-fold increase in glucose yield compared to wild-type with a p-value of 0.0013. A positive control of Kitake wild-type rice straw confirms glucose release from deconstructed rice straw. A linear equation from the glucose standard curve was used to determine glucose in mg at each absorbance. Error bars were calculated using standard deviation. Some error bars are within the symbol of the time point. Asterisks (**) indicate significant data.

Results from the deconstruction of a Myb61a knock-down mutant and wild-type rice straw show the mutant biomass has increased sugar release at the final 46-hour-time-point when compared to wild-type rice straw. Each time point represents the average glucose release in mg/mg of biomass over four technical replicates deconstructed from both mutant and wild-type. As indicated by asterisks, a student’s t-test compares mutant (Mut) glucose release to the final glucose release of the wild-type (WT) to give a p-value of 0.0013. The p-value of this t-test is considered significant because it is less than 0.05. It is important emphasize that mutant rice straw was only taken from a single mutant rice plant, while wild-type rice straw was pooled from all wild-type plants. To confirm that the deconstruction process is releasing glucose, Kitake wild-type biomass was used as a positive control. A negative control (data excluded from Figure 1) consisted of only Novoenzymes, CTec2 and NS50010 in 0.1M Citrate Buffer. DNS assay did not indicate any glucose release from the negative control over the 46-hour time period. Absorbance readings from the negative control were considered background and were subtracted from the absorbance readings of mutant and wild-type biomass wells. The low, significant p-value of increased mutant glucose release provides confidence in the hypothesis that Myb61a mutant biomass is more digestible to cell wall enzymes.

The deconstruction of Myb61a knock-down mutants and wild-type rice straw was repeated with a pool of biomass from the Myb61a mutant plants and wild-type rice straw. The results obtained from the initial experiment testing saccharification of mutant and wild-type straw needed to be confirmed using pooled Myb61a mutant plant material to confirm the significance of increased glucose release during the digestion of mutant plants. The standard deconstruction process and DNS assay was replicated with four technical replicates of each mutant and wild-type deconstruction. In this experiment, only four time points were taken over a period of 19.5 hours. Kitake wild-type rice straw was repeated as a positive control and showed constant sugar release. The negative control (excluded from the Figure) included a well of citrate buffer and enzymes and was measured with DNS assay over the four time points. The results in Figure 2 confirm the trend of increased glucose release by mutant rice straw but do not confirm that Myb61a knock-down mutants show a significant p-value from glucose released at the final time point compared to wild-type rice straw (Figure 2).


Figure 2. Biomass pooled from Myb61a knock-down mutants (MT) pretreated with Citrate Buffer (CT) shows increased glucose release compared to wild-type (WT). Ground biomass from Myb61a mutants and negative segregants was pretreated with 0.1M Citrate Buffer. Cellulase complex (CTec2) and B-glucosidase (NS50010) enzymes were added to the pre-treated biomass. Myb61a mutant biomass and wild-type biomass was pooled from all available rice plants. DNS assay measured glucose release in mg over a period of 19.5 hours. Mutant rice straw shows a 1.2-fold increase in glucose yield compared to wild-type with a p-value of 0.4326. Glucose yield from both mutant and wild-type are increased in this experiment compared to the endpoint yield in the previous deconstruction. A positive control of Kitake wild-type rice straw confirms glucose release from deconstructed rice straw. A linear equation from the glucose standard curve was used to determine glucose in mg at each absorbance. Error bars were calculated using standard deviation. Some error bars are within the symbol of the time point.

Data from Figure 2 confirms the trend of increased glucose release from Myb61a mutant rice straw. However, a student’s t-test gives a p-value of 0.4326 indicating that although mutant rice straw has a greater final glucose release than the wild-type, it cannot be explained by the Myb61a knock-down mutation.

Part II: Pretreatment of rice straw with 15% NH4OH shows enhanced saccharification compared to standard pretreatment with 0.1M Citrate Buffer.

Wild-type rice straw pretreated with 15% NH4OH shows significant increase in glucose release compared to rice straw pretreated under standard conditions with 0.1M Citrate Buffer. Wild-type rice straw was pooled across all available rice plants. Uniformly ground straw was divided into four technical replicates for both the pretreatment of ammonium hydroxide (NH4OH) and Citrate Buffer. 15% NH4OH was added to four biomass samples and 0.1M Citrate Buffer was added to the remaining four biomass samples. All samples were incubated for one hour. After this time, NH4OH (pH 11.63) pretreated samples were neutralized. Cell wall enzymes, CTec2 (Cellulase) and NS50010 (beta-glucosidase) were added to all samples. Time points were taken over 46 hours and glucose release was measured by DNS assay (Figure 3). As hypothesized, NH4OH pretreated samples showed a significant amount of glucose release at the final time point compared to samples undergoing only standard deconstruction.


Figure 3. Wild-type (WT) rice straw pretreated with Ammonium Hydroxide (NH4OH) shows increased glucose release compared to pretreatment with Citrate Buffer (Cit). Ground biomass from wild-type rice straw was pretreated with either 0.1M Citrate Buffer or 15% NH4OH. Cellulase complex (CTec2) and B-glucosidase (NS50010) enzymes were added to the pre-treated biomass. Wild-type biomass was pooled from all negative segregant (WT) rice plants. DNS assay measured glucose release in mg over a period of 46 hours. NH4OH pretreated rice straw shows a 9-fold increase in glucose yield compared to Citrate Buffer pretreated with a p-value of 0.0045. A positive control of Kitake wild-type rice straw confirms glucose release from deconstructed rice straw. A linear equation from the glucose standard curve was used to determine glucose in mg at each absorbance. Error bars were calculated using standard deviation. Some error bars are within the symbol of the time point. Asterisks (**) indicate significant data.

Data from the deconstruction of rice straw with a strong basic pretreatment of NH4OH proves to show significant glucose release as hypothesized. A positive control of Kitake wild-type plant confirms that glucose release increased over a period of incubation due to the digestion of the biomass material by the enzymes. The negative control consisted of only enzymes and citrate buffer and did not show any glucose release which confirmed that the enzymes and citrate buffer showed no signs of contamination. Absorbance from the negative control was subtracted from absorbance of the technical replicates. A standard curve was used to determine mg of glucose release at each absorbance and glucose release was averaged across all technical replicates of each pretreatment condition. A student’s t-test was preformed on the end points of the two pretreatment conditions (NH4OH and Cit) to give a p-value of 0.0045 for the glucose release of the NH4OH pretreated biomass compared to the Citrate Buffer only treated biomass. A p-value less than 0.05 was considered significant. Rice straw pretreated with NH4OH has a 9-fold increase in glucose release compared to the standard deconstructed rice straw. Data from the t-test confirms our hypothesis that a strong basic pretreatment with ammonium hydroxide will alter the structure of the biomass leaving it more digestible to the cell wall enzymes.

The pretreatment of wild-type rice straw with NH4OH compared to 0.1M Citrate Buffer standard deconstruction was repeated to confirm the significant increase in glucose release from ammonium hydroxide pretreated biomass. The deconstruction of wild-type rice straw was repeated with four technical replicates of both NH4OH and Citrate Buffer pretreatment conditions. The initial pretreatment experiment was replicated with the same protocol except during this replication, time points for DNS assay were taken over a period of 19.5 hours (Figure 4). The results from the replicated experiment confirm the significance of the initial experiment.


Figure 4. Wild-type (WT) rice straw pretreated with Ammonium Hydroxide (NH4OH) shows increased glucose release compared to pretreatment with Citrate Buffer (Cit). Ground biomass from Myb61a mutants and negative segregants were pretreated with 0.1M Citrate Buffer. Cellulase complex (CTec2) and B-glucosidase (NS50010) enzymes were added to the pre-treated biomass. Myb61a mutant biomass and wild-type biomass was pooled from all negative segregant rice plants. DNS assay measured glucose release in mg over a period of 19.5 hours. NH4OH pretreated rice straw shows a 1.9-fold increase in glucose yield compared to wild-type with a p-value of 0.0042. Glucose yield from both treatments are increased in this experiment compared to the endpoint yield in the previous pretreatment experiment shown in Figure 3. A positive control of Kitake wild-type rice straw confirms glucose release from deconstructed rice straw. A linear equation from the glucose standard curve was used to determine glucose in mg at each absorbance. Error bars were calculated using standard deviation. Some error bars are within the symbol of the time point. Asterisks (**) indicate significant data.

Data from the replicated deconstruction of rice straw with a strong basic pretreatment of NH4OH confirms the significance of the initial results from Figure 3. As in the initial experiment, a positive control of Kitake wild-type plant confirms the glucose release from rice straw and the negative control (not shown in Figure 4) did not show any glucose release which confirmed that the enzymes and citrate buffer showed no signs of glucose or biomass contamination. Absorbance from the negative control was subtracted from the absorbance of the technical replicates to adjust for background noise. A standard curve was used to determine mg of glucose release at each absorbance and glucose release was averaged across all technical replicates of each pretreatment condition. A student’s t-test was preformed on the end points of the two pretreatment conditions (NH4OH and Cit) to give a p-value of 0.0042 for the glucose release of the NH4OH pretreated biomass compared to the Citrate Buffer only treated biomass. A p-value less than 0.05 was considered significant. Rice straw pretreated with NH4OH had a 1.9-fold increase in glucose release compared to the standard deconstructed rice straw. Although the replicated experiment shows less of a difference between the two pretreatments, the t-test confirms our hypothesis that a NH4OH pretreatment can be used to increase saccharification.

Part III: Myb61a mutant rice straw pretreated with 15% NH4OH shows increased saccharification compared to mutant rice straw deconstructed under standard conditions with 0.1M Citrate Buffer.

In addition to measuring the digestibility Myb61a mutant rice straw under standard deconstruction, the combination of mutant rice straw pretreated with 15% NH4OH was measured to confirm that mutant rice would display even more increased saccharification when pretreated under highly basic conditions. Ground Myb61a mutant biomass was incubated for one hour in 15% NH4OH and then neutralized to a pH of around 5.00. Cell wall enzymes (CTec2 and beta-glucosidase) were then added to the biomass slurry. DNS assay was used to measure glucose release over a period of 46 hours (Figure 5). The results of the deconstruction of Myb61a with NH4OH pretreatment and under standard conditions as well as the deconstruction of wild-type rice straw with NH4OH and under standard conditions are displayed in Figure 5.


Figure 5. Myb61a knock-down mutant (MT) rice straw pretreated with Ammonium Hydroxide (NH4OH) shows significant glucose release compared to Citrate Buffer pretreated mutant rice straw. Ground biomass from Myb61a mutants and negative segregants (WT) was pretreated with 0.1M Citrate Buffer or 15% NH4OH. Cellulase complex (CTec2) and B-glucosidase (NS50010) enzymes were added to the pre-treated biomass. DNS assay measured glucose release in mg over a period of 46 hours. NH4OH pretreated mutant rice straw shows a 4.7-fold increase in glucose yield compared to citrate pretreated mutants with a p-value of 0.0485. A positive control of Kitake wild-type rice straw confirms glucose release from deconstructed rice straw. All combinations of wild-type and mutant pretreatments are graphed for the purpose of comparison. A linear equation from the glucose standard curve was used to determine glucose in mg at each absorbance. Error bars were calculated using standard deviation. Some error bars are within the symbol of the time point. Asterisks (**) indicate significant data.

Additional data from the deconstruction of Myb61a mutant rice straw with 15% NH4OH pretreatment confirms that an ammonium hydroxide pretreatment improves the saccharification of both wild-type and mutant rice straw. The positive control and negative control were consistent in the deconstruction of Myb61a mutant straw in 15% NH4OH with the previous controls. When compared to Citrate Buffer treated mutant rice straw, the pretreatment of mutant rice straw with NH4OH increased glucose yield by 4.7-fold with a p-value of 0.0485. A p-value of less than 0.05 confirms that this basic pretreatment significantly improves the deconstruction of Myb61a mutant plants. This deconstruction provides data to support the advantages of combining successful pretreatments with digestible mutant plants to even further improve saccharification.

Discussion

A global emphasis has been laid on decreasing the amount of carbon dioxide (CO2) released into the atmosphere as a result of the large-scale burning of fossil fuels. The main components of fossil fuels are methane and petroleum (8). These chemical compounds consists of hydrocarbons that when burned release water, energy, and carbon dioxide (9). Climate change has become a household name that refers to the flux of carbon dioxide into the atmosphere that outstrips the rate of carbon dioxide removal. The growth rate of CO2 emissions increased on a global scale from 1.1% a year between 1990-1999 to greater than 3% a year between 2000-2004 (10). The Earth’s atmosphere is maintained by gases such as CO2, water vapor, methane, ozone, nitrous oxide, and halocarbons (9). To maintain the global carbon cycle, the rate of addition and removal of these gases must be equal and steady. An excess of CO2 causes the Earth to retain more heat in the form of infrared rays that would otherwise escape the atmosphere.

Currently, 98% of the energy used for transportation comes from the burning of fossil fuels (9). International efforts are focused toward finding an alternative source of hydrocarbons to replace fossil fuels. Plant biomass is an abundant energy reserve of glucose sugars that can be fed to microbes to produce biofuels. It is estimated that 98-99% of accumulated organic plant matter decays; the energy in the sugars of this decaying matter left unutilized (10).

This experiment makes progress in the utilization of energy stored in plant cell walls through the genetic manipulation of Myb61a and the highly basic pretreatment of biomass with 15% NH4OH. Not only does the deconstruction of Myb61a knock-down mutant plants show increase saccharification, but the improved digestibility of these mutant plants confirms the role of Myb61a in the cell wall biosynthetic process. The results of the Myb61a mutant deconstruction are promising for engineering a genetically modified plant that displays increased sugar yield for the process of biofuel production.

In comparison to other experiments that show increased saccharification due to genetic manipulation, my results prove to be competitive. In one experiment from the literature, an acyl-transferase, OsAT10, involved in the acylation of cell wall matrix polysaccharides was overexpressed in rice. The overexpression mutant, OsAT10-D1, showed a significant increase in glucose yield of approximately 0.035 mg of glucose per mg of biomass when compared to the wild-type which showed approximately 0.029 mg of glucose per mg of biomass (3). In my experiment, Myb61a knock-down mutants show approximately 0.039 mg of glucose release per mg of biomass compared to the wild-type showing approximately 0.013 mg of glucose release per mg of biomass (Figure 1). Data from my experiment suggests that the Myb61a knock-down mutant contends well with other published data for genetically modified rice plants.

In addition to rice, genetic mutants exhibiting increased digestibility have also been identified in maize. A phenotypic mutation known as brown-midrib (bmr) mutation can be induced in maize as well as sorghum. This simple recessive mutation is associated with a pigment in lignified tissues. Bmr mutants show reduced lignin content and a 33% increase in the digestibility of cell wall components when digested with a commercial Cellulase similar to Ctec2 (11). In both Bmr and OsAT10-D1 mutants, increased digestibility was found to correlate to a change in lignocellulosic content. Evidence from my experiment confirms that significant glucose release could in fact be related to Myb61a’s role in regulating the transcription of cell wall components.

If Myb61a is a transcriptional activator for cell wall synthesis, it would be logical to infer that the knock-down mutant would display decreased lignocellulosic content. Further experiments should be conducted to confirm this hypothesis and the general trend of increased saccharification in Myb61a mutant biomass.

One important parameter to consider when examining the data of mutant versus wild-type glucose release (Figures 1 and 2) is that in Figure 1 the Myb61a mutant refers to biomass from a single mutant plant. In Figure 2, the experiment was repeated using a pool of milled rice straw across all available Myb61a mutant plants. Taking rice straw from more than one mutant plant makes the data more credible because the mutant plants now have a greater sample size. As expected, the comparison of a pool of mutant plants to a pool of wild-type plant biomass resulted in an insignificant p-value. If this experiment were to be repeated, it would be interesting to see if the general trend of increased saccharification by mutant rice straw persisted even in a larger sample size.

In addition to the effects of mutation on cell wall digestibility by the cellulase and beta-glucosidase enzymes, biomass pretreated with ammonium hydroxide exhibits a significant increase in saccharification. Wild-type rice straw pretreated with NH4OH shows up to a 3–fold increase in glucose release compared to wild-type rice straw incubated in Citrate Buffer. In a repeated experiment, NH4OH pretreated biomass shows a 1.2-fold increase in glucose yield. Despite the decline in glucose yield in the follow-up experiment, a student’s t-test validates that both data sets show significant increase (p-value <0.05) in glucose yield when pretreated with ammonium hydroxide.

The significance of the rice straw’s increased digestibility due to 15% NH4OH pretreatment has special implications. In a deconstruction of wheat straw, a 16.8% NH4OH pretreatment was used to enhance saccharification (11). Results of this deconstruction also show a 3-fold increase in saccharification of pretreated samples compared to untreated samples. Images captured by scanning electronic microscopy (SEM) of wheat straw before and after pretreatment showed obvious structural changes to lignocellulosic material after being pretreated with NH4OH. Further analysis by X-ray diffraction suggests that the crystallinity of lignocelllulose was significantly decreased by NH4OH pretreatment. Decreasing the crystallinity of lignocellulose may open up the cell wall structure to allow the cellulase enzymes more access to cellulosic material. Evidence obtained by the wheat straw deconstruction suggests that similar structural rearrangements due to the ammonium hydroxide pretreatment may be affecting rice straw. Wheat and rice, both monocots, display the same ratio of increased saccharification under similar pretreatment conditions, which would lead us to presume that a pretreatment that works well for one monocot may work just as well for another monocot.

The data obtained from both mutant and pretreatment deconstructions was put to the test to see if a combination of genetic mutation and ammonium hydroxide pretreatment would yield even further glucose release. In a deconstruction of Myb61a mutant biomass pretreated with NH4OH (Figure 5), the data indicates an outstanding 15.7-fold increase in saccharification compared to the untreated wild-type biomass. This data is especially important to the deconstruction of biomass because it shows that the combination of genetic mutation and pretreatment methods can be used to more than double the sugar yield that would otherwise be obtained by using one of these deconstruction targets by itself. Building on Myb61a mutation and highly basic pretreatment strategies to increase sugar yield has been validated as time-worthy research by the data put forth by this experiment. The benefit of increased sugar yield is magnified by the combination of unique targets outlined in this experiments for the deconstruction of lignocellulosic biomass. It is hopeful that one day the deconstruction of agricultural waste will yield adequate amounts of glucose at a low energy cost.




Citations

1. Bartley, L., & Ronald, P. (2009). Plant and microbial research seeks biofuel production       from lignocellulose. California agriculture63(4), 178-184.
2. Rafelski, L. E., Piper, S. C., & Keeling, R. F. (2009). Climate effects on atmospheric carbon dioxide over the last century. Tellus B61(5), 718-731.
3. Bartley, L. E., Peck, M. L., Kim, S. R., Ebert, B., Maniseri, C., Chiniquy, D., ... & Ronald, P. C. (2013). Overexpression of a BAHD acyltransferase, OsAt10, alters rice cell wall hydroxycinnamic acid content and saccharification. Plant physiology, pp-112.
4. Handakumbura, P. P., & Hazen, S. P. (2012). Transcriptional regulation of grass secondary cell wall biosynthesis: playing catch-up with Arabidopsis thaliana. Frontiers in plant science3.
5. Liang, Y. K., Dubos, C., Dodd, I. C., Holroyd, G. H., Hetherington, A. M., & Campbell, M. M. (2005). AtMYB61, an R2R3-MYB transcription factor controlling stomatal aperture in Arabidopsis thaliana. Current Biology15(13), 1201-1206.
6. Penfield, S., Meissner, R. C., Shoue, D. A., Carpita, N. C., & Bevan, M. W. (2001). MYB61 is required for mucilage deposition and extrusion in the Arabidopsis seed coat. The Plant Cell Online13(12), 2777-2791.
7. Elumalai, S., Roa-Espinosa, A., Markley, J. L., & Runge, T. M. (2014). Combined sodium hydroxide and ammonium hydroxide pretreatment of post-biogas digestion dairy manure fiber for cost effective cellulosic bioethanol production. Sustainable Chemical Processes2(1), 12.
8. Schobert, H. (2013). Chemistry of Fossil Fuels and Biofuels. Cambridge University Press.
9. Jeong, S., Millstein, D., & Fischer, M. L. (2014). Spatially Explicit Methane Emissions from Petroleum Production and the Natural Gas System in California. Environmental science & technology, 48(10), 5982-5990.
10. Raupach, M. R., Marland, G., Ciais, P., Le Quéré, C., Canadell, J. G., Klepper, G., & Field, C. B. (2007). Global and regional drivers of accelerating CO2 emissions. Proceedings of the National Academy of Sciences, 104(24), 10288-10293.
11. Zhong, C., Wang, C., Wang, F., Jia, H., Wei, P., & Zhao, Y. (2014). Enhanced saccharification of lignocellulosic biomass by pretreatment with quaternary ammonium hydroxide. Journal of Chemical Technology and Biotechnology.



Supplementary Information

Table S1. Organization of technical replicates by pre-treatment and biomass type. Myb61a-1 mutants biomass in the first set of experiments was taken from a single Myb61a-1 mutant rice plant. This experiment includes four technical replicates of each pre-treatment and biomass combination. Kitake (KIT) wild-type rice straw was used as the positive control (PC). The negative control (NC) contained enzyme only, to adjust for background readings.










Figure S2. Standard curve of absorbance verses a known amount of glucose dilution used to derive linear equation for determining mg of glucose released at each absorbance. This standard curve was obtained by plotting known glucose dilutions on the x-axis against a given absorbance determined by the thermocycler. A linear trendline of y=0.3987x-0.0859 was obtained from the standard curve and used to calculate mg of glucose per mL of all subsequent thermocycler runs. An R2=0.96379 confirms that the linear trendline obtained from the data of this graph does not display a significant amount of error and can be used to determine glucose release. 

Friday, April 3, 2015

Patterns of expression of Glyma09g31910 of the PLAC-8 family show potential target for improving N2 fixation by Glycine max L.

Patterns of expression of Glyma09g31910 of the PLAC-8 family show potential target for improving N2 fixation by Glycine max L.

JILLIAN C. VAUGHT
Department of Plant Biology, The University of Oklahoma, Norman, OK

3 April 2015

ABSTRACT

Plants are limited in growth and yield by nitrogen availability in the soil. Although N fertilization methods are on the rise, they are costly and energy intensive, often resulting in nitrate run off that degrades water quality. Plants such as the soybean associate with N2 fixing bacteria that convert atmospheric nitrogen into usable organic NH3+. During nodulation, the host plant responds to chemical signals from the infecting soil bacteria in order to incorporate the bacteria into root nodules for nitrogen fixation. Chemical signals, called Nod factors are recognized by the soybean plant and induce developmental changes as well as calcium spiking in root hair cells. The soybean gene, Glyma09g31910, has been shown to be highly expressed in root hairs during inoculation with symbiotic bacteria and is therefore a target gene for increasing plant N2 fixation when N is limiting in the soil. A qRT-PCR analysis of the relative abundance of Glyma09g31910 mRNA was conduct in root hairs un-inoculated and inoculated with rhizobium bacteria. Results of our qRT-PCR indicated that root hairs both inoculated and un-inoculated show high levels of Glyma09g31910 expression. The qRT-PCR results from our study strongly contrast the data mined from SoyKB, which indicates that Glyma09g31910 expression is only high in inoculated root hairs. Results from our qRT-PCR were confirmed in our transgenic roots, which used GUS reporter downstream of the Glyma09g31910 promoter to show high levels of Glyma09g31910 expression in both inoculated and un-inoculated roots. However, transgenic roots showed a different pattern of expression of Glyma09g31910 in areas of the roots that had been infected by rhizobium bacteria when compared to expression of Glyma09g31910 in un-inoculated roots. Patterns in the expression of Glyma09g31910 when associated with bacteria, emphasize Glyma09g31910 as a potential target gene for increasing the efficiency of N2 fixation in legumes.


INTRODUCTION

Nitrogen availability is a crucial factor effecting plant growth and yield. Plants alone do not possess the biochemical pathways needed to convert atmospheric N2 into N (NH3) available for plant metabolism. Popular economic crops such as corn use nitrogen from the soil with small amounts of nutrient return, causing the soil quality to become poor and subsequent crop yield to be limited (Tauer, 1989).  Methods of soil remediation are being sought in response to limited crop growth due to low soil nitrogen content. One method of soil remediation is the addition of nitrogen through fertilization, but the direct addition of nitrogen to soil comes at a high economic price (Miransari et. al, 2013). Not only is the production of nitrogen currently an energy intensive process, but increased nitrogen fertilization has led to nitrate run off that has lowered water quality (Tauer). Finding new ways to implement N remediation into soils that avoids the direct application of nitrogen has an obvious benefit both economically and ecologically.

One investigated solution to soil remediation is the leguminous soybean plant that establishes symbiotic relationships with N-fixing bacteria. Under nitrogen limiting conditions, soybean plants will form nodules associated with the infection of Bradyrhizobium japonicum, an atmospheric N2 fixing bacteria (Miransari). Rotating crops with legumes such as the soybean is already common practice to exploit the N2 fixing ability of bacteria in association with legumes (Tauer). The efficiency of soil remediation to re-introduce usable N into the soil could be greatly increased using biotechnological applications to target genes involved in the nodulation of soybean plants. One specific gene of Glycine max L that has been shown by SoyKB to have expression correlated to the infection of N fixing rhizobia bacteria is Glyma09g31910 (Joshi, 2014). Targeting Glyma09g31910 to manipulate nodulation of N2 fixing relationships with rhizobia could be advantageous to soil remediation.  

Investigation into the evolution of Glyma09g31910 shows it belongs to a highly conserved family of proteins called PLAC-8, which can be found widespread in eukaryotes (Abba et. al, 2011). Proteins containing the PLAC-8 domain have been shown to confer cadmium resistance, regulate cell size and number, control cell growth, and promote cell division (Di Vietro, et. al, 2014).  However, despite the numerous works compiled regarding the PLAC-8 family, no definite function is attributed to this domain (Di Vetro). In a Blast protein sequence similarity search, Glyma09g31910 was found to have 99% similarity with a protein for plant cadmium resistance in Glycine soja and a 62% similarity to a predicted occurrence of cell number regulation in Cicer arietinum (NCBI Resource Coordinators). Further investigation into the PLAC-8 family could help reveal a function associated with this domain.

Although no known function is attributed to the PLAC-8 domain, preliminary evidence suggests that Glyma09g31910 is associated with the process by which rhizobacteria infects roots. The mechanism by which B. japonicum, a rhizobacteria, associates with soybean to fix N2 has been outlined in detail (Oldroyd and Downie, 2008). Research shows that upon root hair infection by the rhizobium, the bacterium secretes lipochitosaccharide-based signal molecules called Nod factors. These Nod factors begin a cascade of biochemical signals, including Ca2+ spiking that is essential for the expression of many soybean genes involved in nodulation (Geurts et. al, 2005). Gaining a better understanding of soybean genes involved in nodulation could be a target for gene modification and plant breeding to increase N2 fixation.

Preliminary results obtained from mining SoyKB indicate that Glyma09g31910 is significantly involved in the nodulation and infection of roots and root hair cells (Figure 1). Data from SoyKB shows high levels of expression of the Glyma09g31910 gene in root hairs when inoculated with rhizobacteria after a period of 12, 24, and 48 hours. Additional literature also indicates that the Glyma09g31910 gene of Glycine max L. shows strong response in terms of gene expression when inoculated with the nitrogen-fixing bacterium B. japonicum (Libault et. al, 2010). These preliminary results lead us to hypothesize that roots inoculated with B. japonicum will express Glyma09g31910 in relatively high mRNA abundance and un-inoculated roots will not show expression of Glyma09g31910. We also expect high patterns of activity of the Glyma09g31910 promoter at sites of rhizobacterial infection and nodulation in the roots and root hair cells of Glycine max L.


Figure 1. Glyma09g31910 Gene Expression in Root Hair Tissue Hours After Rhizobium Bacterial Inoculation



















Figure 1. Gene expression obtained from SoyKB is measured in RPKM found in different tissues, some inoculated with rhizobia bacteria (IN) and other un-inoculated (UN) over a period of time. The y-axis represents reads per kilobase pair per million (RPKM), the amount of short sequences produced during a PCR using specific primers to detect the presence of cDNA that has been reverse transcribed from mRNA of Glyma09g31910. The mean density of reads present is related to cDNA present at the time of mRNA extraction. Reads per kilobase pair per million normalizes the samples in terms of abundance of transcripts so it can be compared across all tissues. In the red bars at 12 hours, un-inoculated (UN) root hairs (RH) show zero abundance of transcript, while inoculated (IN) root hairs show a significant relative abundance. Root hairs inoculated and un-inoculated after 24 hours, and 48 also show highest gene expression under inoculated conditions. Tissue from the nodules shows the highest gene expression. Apical meristem, flower, green pods, leaves, and root tips show no gene expression. Roots, however, are seen to express Glyma09g31910.


RESULTS

Glyma09g31910 is highly expressed in root hair tissues when inoculated with N2 fixing bacteria such as B. japonicum according to the data mined from SoyKB. To confirm these results, mRNA was extracted from 6 major soybean Glycine max L. tissues: pods, nodules, leaves, shoot apical meristem (SAM), and stems. Samples of tissues were taken from root hairs and roots inoculated and un-inoculated with bacteria. Gel electrophoresis was used to determine the integrity of the RNA extraction from tissues (Figure 2).



Figure 2. Gel Electrophoresis of mRNA Extractions to Determine Integrity and Quality of RNA












Figure 2. Lanes 1 was RNA extracted from SAM, 2 from Stems, 3 from Leaves, 4 from Nodules, 5 from roots, and lane 6 extracted from PODS. Gel electrophoresis images on the right and left are the same with the right image adjusted for contrast to show brighter banding. RNA extracted from SAM, Stems, Leaves, Roots, and Pods show banding for smaller nucleic acid molecules consistent with RNA, however the Nodules extraction in lane 4 shows a banding of a large nucleic acid molecule more consistent with DNA contamination. Smearing in the bands indicates the presence of some protein contamination.

RNA extractions confirmed by gel electrophoresis were used to determine relative expression levels of our Glyma09g31910 gene of interest across all six tissue samples. Reverse transcription of mRNA to synthesize cDNA provided us with DNA material to run a qRT-PCR in order show the relative abundance of mRNA extracted from our tissues. To indicate the presence of our gene of interest during our qPCR, primers were ordered specific to the 5’ un-translated region of Glyma09g319190 (Fig. S4). Un-translated regions are less likely to be conserved and are therefore more likely to be more specific to our gene of interest (Fig. S5). Primers were used during qRT-PCR with SYBR Green fluorescence to indicate the relative abundance of Glyma09g31910 across all tissues. The relative abundance of mRNA was normalized using cons6, a gene of constant transcription levels across all tissues. qRT-PCR results (Table 1.) indicated highest levels of Glyma09g31910 in un-inoculated root hairs after a 12 hour incubation, and the second most relative abundance in root hairs that were inoculated with bacteria after 12 hours. Glyma09g31910 was also relatively abundant in nodules and shoot apical meristems.


Table 1. Abundance of mRNA in Tissues of Soybean Obtained from qRT-PCR











Table 1. Indicates the number of cycles that are needed to acquire the same amount of cDNA in each plant tissue using qRT-PCR. The lowest cycle value indicates the highest abundance of cDNA present in the tissue samples. The cDNA is made from the Glyma09g31910 mRNA that has been transcribed in the tissues. The lowest cycle value from RH-BAC 12HAI (the root hairs un-inoculated with bacteria after 12 hours) indicates that Glyma09g31910 has the highest expression in this tissue. RH+BAC 12 HAI followed by nodules and shoot apical meristem (SAM) have increasing levels of Glyma09g31910 expression. No Glyma09g31910 expression was found in all tissues indicated with a cycle number of 0, indicated as no amplification in the table, because no cDNA could be amplified during the qRT-PCR of cDNA made from the mRNA extracted from these samples.

Although our qRT-PCR results (Table 1) strongly contrasted the presence of Glyma09g31910 in un-inoculated root hair tissues from the SoyKB data (Figure 1), the pattern of Glyma09g31910 expression during bacterial inoculation and subsequent nodule development analyzed by a reporter gene can still be useful. In order to look at patterns of expression of the gene across the roots and root hairs, we used the Gateway system to clone the Glyma09g31910 promoter upstream of a GUS reporter. This construct was inserted into soybean plants to transform our roots. To do this, soybean plants were transformed using Agrobacteria rhizogene containing a plasmid for our promoter upstream of a GUS reporter. Agrobacteria rhizogene carrying our vector of interested infected wounds in our soybean plants and inserted our vector with the promoter cloned upstream our GUS reporter to create transgenic soybean roots.

To clone the promoter upstream of the GUS reporter sequence, the promoter was isolated from extracted gDNA. Gel electrophoresis was used to determine the integrity and quality of extracted gDNA (Figure 3). The extracted gDNA was used as DNA template for a subsequence PCR reaction using primers specific to the promoter sequence that also included 50% of attB1 and attB2 boxes that added to the promoter sequence during PCR to make the promoter sequence compatible with the Gateway system (Fig. S6). Adaptors of the attB1 and attB2 boxes were added to elongate the boxes to 100% length in a second PCR reaction (Fig. S7).


Figure 3. Gel Electrophoresis of gDNA to Determine Integrity and Quality of Extraction



















Figure 3. Gel electrophoresis of gDNA extractions was used to determine the integrity and quantity of gDNA. Lane 6 is the extraction used for Glyma09g31910 promoter isolation. It shows good quality and integrity having a solid band of large nucleic acid materials consistent with extracted DNA.

After modified primers added full length attB boxes to the Glyma09g31910 promoter, the promoter was compatible to be used with the Gateway cloning system to clone the promoter upstream of the GUS reporter. A BP reaction was used to make a recombinant pDONR/Zeo plasmid, which cloned our promoter into the pDONR plasmid. pDONR/Zeo plasmids were transformed into E. coli and allowed to grow on nutrient broth plates. Two forms of selection were used to select positive clones: ccdB gene and zeocin antibiotic resistance gene. Positive clones contained the promoter in the place of ccdB, a gene that inhibits bacterial growth. Transformed E.coli was grown on a nutrient solution containing the antibiotic, zeocin. Positive transformed clones contained the gene for zeocin resistance and were able to grow on the plate. E. coli colonies were used as PCR template and run by gel electrophoresis with primers for the Glyma09g31910 promoter to identify positive clones (Figure S8). These plasmids were extracted from positive colonies. An LR reaction was used to make a recombination of the promoter sequence into a donor vector, pYXT1/Kanamycin, containing a GUS reporter downstream the promoter. The donor vector was transformed into Agrobacterium by electroporation and grown on a nutrient plate containing kanamycin antibiotic. Two forms of selection were also used to select positive clones: the ccdB gene and antibiotic resistant gene for kanamycin. A PCR reaction followed by gel electrophoresis, using primers for the Glyma09g31910 and the reporter gene, were used to confirm the presence of positive clones for Glyma09g31910 promoter upstream the GUS reporter (Figure S9).

Agrobacterium rhizogenes uses horizontal gene transfer to infect plants. Agrobacteria transformed with our construct of interest was used to infect plants at wound sites of cut stems on our soybean plants. Soybean plants, with stems cut above the beginning of our roots, were placed in Fibrgro®cubes. The cubes were soaked with transformed agrobacterium nutrients, allowing the agrobacterium to infect the plant and transfer our construct of interest into the plant roots. Plants were grown for 5 weeks in an ultrasound aeroponic system. After this time, roots were stained with GUS staining buffer. The GUS reporter gene, when activated by the Glyma09g31910 promoter, produces β-glucuronidase. β-glucuronidase can be cleaved by X-Gluc, a substrate in the GUS staining buffer, to produce a blue stain. The blue staining observed in transformed plant roots indicates the activity of the Glyma09g31910 promoter in plants that were un-inoculated and inoculated with B. japonicum (Figure 4a-b
Figure 4a. Transformed Roots and Root Hairs Un-Inoculated and Stained with GUS Staining Buffer












Figure 4a. Roots and root hairs of un-inoculated transformed roots stained with GUS staining buffer. The image on the left is taken at 1x magnification showing blue staining of the GUS reporter gene downstream the Glyma09g31910 promoter. Blue stain indicates activity of the Glyma09g31910 promoter. The image on the right was taken at 8x magnification and shows the presence of root hairs with Glyma09g31910 activity. Root hairs appear uncurled as we would expect under un-inoculated conditions.

Figure 4b. Transformed Roots and Root Hairs Inoculated with B. japonicum and Stained with GUS Staining Buffer












Figure 4b. Roots and root hairs of inoculated transformed roots stained with GUS staining buffer. The image on the left is taken at 1x magnification showing blue staining of the GUS reporter gene downstream the Glyma09g31910 promoter. Blue stain indicates activity of the Glyma09g31910 promoter. Blue stain appears spotted which would indicate areas of rhizobacteria infection, which we would expect to activate the Glyma09g31910 promoter. The image on the right was taken at 8x magnification and shows the presence of root hairs with Glyma09g31910 activity. Root hairs appear curled, as we would expect under inoculated conditions.

Based on preliminary evidence obtained from SoyKB data mining, we would expect our Glyma09g31910 gene only to be expressed under inoculated conditions. Images of the stained un-inoculated roots and root hairs (Figure 4a) show significant blue staining which indicates the Glyma09g31910 was activated in these roots. In the image, dark tips can be seen in the 1x image that we would expect to find consistently with the activation of Glyma09g31910 during nodulation. The image of root hairs obtained at 8x does not show curling which would indicate that these root hairs have not been contaminated with rhizobacteria although they do show Glyma09g31910 activation. The results of blue staining in these un-inoculated root images dos not support the preliminary evidence obtained from SoyKB.

Although we did not expect Glyma09g31910 expression in un-inoculated root hairs based on SoyKB, the qRT-PCR results we obtained from un-inoculated root hair tissue (Table 1) indicated the highest amounts of mRNA transcripts of Glyma09g31910, and therefore high expression of Glyma09g31910 even in un-inoculated root hairs. Based on our qRT-PCR data, the staining results from the image of un-inoculated root hairs (Figure 4a) are consistent with the data we obtained (Table 1).

Images of root hairs under inoculated conditions (Figure 4b) appear to have spotted blue staining along the roots in the 1x image. The image taken of the root hairs at 8x have solid blue staining on both sides of the root and in the root hairs. The root hairs in the image appear curled as would be expected as a result of B. japonicum infection. If Glyma09g31910 activity is indeed a response exclusive to rhizobacterial infection, then the blue spots along the transgenic roots indicate areas where rhizobium bacteria entered and infected the roots.

Data analysis from SoyKB indicating that Glyma09g31910 is only expressed under inoculated conditions is supported in the images of the inoculated roots and root hairs (Figure 4b). Unfortunately, because both inoculated and un-inoculated roots show blue staining for Glyma09g31910 activity, our experimental results show no evidence that Glyma09g31910 is expressed exclusively in response to rhizobium bacterial infection. Interestingly, our results of heavy blue staining in un-inoculated roots and slightly less blue staining in inoculated roots agrees with our qRT-PCR results (Table 1) showing the highest amounts of mRNA of Glyma09g31910 in un-inoculated root hairs and slightly less Glyma09g31910 mRNA extracted from inoculated root hairs.

It is important to note the difference in the appearance in the staining of the inoculated and un-inoculated root hairs. While un-inoculated roots appear to have continuous staining down the length of each transgenic root, inoculated roots have spotted staining which is more consistent with sites of infection by rhizobacteria. The difference in these staining patterns could help indicate that the staining of the un-inoculated roots was not due to contamination of rhizobacteria and possibly pertains to issues in the protocol for root staining. Due to the contradiction of our experimental results and data provided by SoyKB, a replication of this experiment would be needed to defend our experimental results with confidence. Replication of this experiment is also needed to confirm that images labeled inoculated and un-inoculated are in fact labeled correctly and were not switched when they were taken back to the lab for analysis, as there was some confusion between group members about the separation of inoculated and un-inoculated plants.

Discussion
A growing worldwide population presents long-term challenges for the production of food crops.
It is estimated that every day, the United States loses 3,000 acres of productive farmland to a variety of developments, both urban and industrial (Land Use Overview, 2013). Loss of cultivatable farmland must be met by advancements in farming techniques such as gene manipulation in crops.

Many common food crops tend to use up nitrogen very readily from the soil, leaving the soil nitrogen deficient. These nitrogen-draining crops tend to fruit and be leafy such as tomatoes, lettuce, and cabbage (Indrasumunar, 2012). Crops grown in nitrogen deficient soil tend to display reduced biomass, decreased growth rates, and increased oxidative damage (Rubio-Wilhelmi, 2011). This reduction in crop yield due to nitrogen deficient soil is of deep concern for farmlands that must meet the growing demand for crop yields.

Fortunately there are crops, such as our soybean, that are able to re-establish useable nitrogen concentrations into the soil by their association with nitrogen fixing bacteria. This trait, indicative of legumes, shows potential to be exploited by biotechnological applications. Genes that prove to be involved in the processes of nitrogen fixation through rhizobacterial association are clear targets for gene manipulation.

Data mined from SoyKB shows clear evidence that Glyma09g31910 has some expressive response to rhizobacterial inoculation soon after inoculation. Although our results failed to clearly support these findings, it is still reasonable to conclude that Glyma09g31910 is involved in the root hair association with nitrogen fixing bacteria, specifically nodulation. Previous studies indicate that genes from the PLAC-8 family may be involved in cell division and remodeling, which would be a consistent with the morphological response of nodulation following bacterial infection with N-fixing bacteria.

Poor conformation of SoyKB data for gene expression in our own experiment could be due to mRNA contamination across our plates prior to qPCR. Impurities in our RNA such as gDNA contamination in our cDNA would also produce the expression patterns seen for Glyma09g31910 in un-inoculated root hairs (Table 1). There is much more uncertainty as to why our transformed roots showed consistent Glyma09g31910 promoter expression under un-inoculated conditions and for this reason it would be beneficial to carefully repeat this experiment. 

High relative abundance of Glyma09g31910 mRNA in nodules and inoculated root hairs is however consistent with our expected results because this gene has been shown to be involved in Nod Factor responses to N-fixing bacteria which occurs in nodules (Geurts). It would be interesting to further investigate the relative abundance of Glyma09g31910 mRNA in nodulation across time points of nodulation from the initial infection to the time of non-expanding nodules. Investigation into the development of nodulation would tell us more about the stages of nodulation that Glyma09g31910 is involved in.

It is worthwhile to remark that high relative abundance of Glyma09g31910 gene transcript is also seen in shoot apical meristems (Table 1). Expanding meristematic tissues are constantly involved in cell division and remodeling. Expression of Glyma09g31910 in dividing cells would provide evidence that the expression of Glyma09g31910 is not only specific to accommodating bacterial infection but also to other types of cell division and remodeling that is being exploited by bacterial infection for the purpose of nodulation. Nevertheless targeting Glyma09g31910 for gene manipulation in efforts to increase legume association with nitrogen fixing bacteria or to exploit its possible role in cell division and nodulation could be profitable strategies for improving crop yield and nitrogen fixing efficiency.

Materials and Methods

Soybean Selection of Tissues
Six major tissues of soybean (Glycine max L.) were chosen for the extraction of gDNA and mRNA. Healthy tissues from pods, roots, nodules, leaves, stems, and shoot apical meristem (SAM) were chosen and homogenized using the beadbeating method. Cell disrupted material was frozen and stored at -80 degrees C. Two types of root tissues were used for mRNA extraction: roots inoculated with B. japonicum and roots un-inoculated with B. japonicum.

Primer Design
Primers were designed for qPCR to indicate presence of Glyma09g31910 gene and to isolate our promoter region for the Gateway method. The soybean genomic resource www.phytozome.net was used to acquire our cDNA sequence of Glyma09g31910. SoyKB at www.soykb.org was used to characterize the expression pattern of this gene. The following parameters were used to choose the qRT-PCR primer: a size of 19-23 base pairs, a PCR product size between 80-120 base pairs, an annealing temperature of the primer at an optimal 60 degrees C, and the percentage of GC content of the primer to be from 40-60%. The forward primer was ordered from Integrated DNA Technologies (IDT) and was 22 bases; the reverse primer was 20 bases. The primers were chosen from the 5’ un-translated region.

Primers were also designed for the isolation of the Glyma09g31910 promoter. The sequence of the promoter was also found using www.phytozome.net and then the promoter region determined using SoyKB. The parameters for the promoter primers were as follows: 19-23 base pairs with an optimal of 21 base pairs, a PCR product size around 2000 base pairs upstream the first codon, annealing temperature at an optimal 60 degrees C, and the GC content 40-60%. The forward primer ordered was 40 base pairs with a GC content of 27.5% and the reverse was 35 base pairs with a GC content of 42.9%. The primers for the promoter included 50% of the attB1 and attB2 boxes used for recombination during the Gateway System Protocol.

gDNA Extraction
Genomic DNA was extracted from soybean tissue before mRNA extraction to avoid contact with DNase. gDNA was extracted from frozen cell disrupted homogenized material according to DNeasy Plant Mini Kit (http://learn.ou.edu/d2l/le/content/2166077/viewContent/2934007/View). After extraction and concentration readings, gDNA was stored at 20 degree C until further use.

mRNA Extraction
mRNA was extracted from frozen cell disrupted homogenized tissue of pods, roots, nodules, leaves, stems, and shoot apical meristem (SAM). The extraction protocol was RNA Mini Kit (Plant) Protocol (http://learn.ou.edu/d2l/le/content/2166077/viewContent/2934007/View). mRNA extracted from the six plant tissues was used to synthesize cDNA using reverse transcription.

gDNA and mRNA Extraction Concentrations Measurements
gDNA and mRNA extraction concentrations were measured using NanoDrop 2000 technology. Two microliters of the extraction sample was pipetted onto the machine and measured using light absorption spectrums. The machine yielded concentration and purification measurments.

Gel Electrophoresis to Confirm gDNA and mRNA Extraction
A gel was made from Buffer and agrose. gDNA and mRNA were run on separate gels with loading dye and water. After 20 minutes the gels were removed and analyzed for the presence of gDNA and mRNA bands to confirm our extractions.

cDNA Synthesis using Reverse Transcription
The Reverse Transcription Protocol (http://learn.ou.edu/d2l/le/content/2166077/viewContent/2934035/View) was used for all six mRNA extractions across all selected tissues to synthesize cDNA for qPCR.

Quantitative PCR of cDNA
Quantitative PCR was performed using the previous cDNA extractions as DNA template. qPCR was performed according the Quantitative PCR Protocol  (http://learn.ou.edu/d2l/le/content/2166077/viewContent/2934011/View). The forward and reverse primers ordered from IDT for qPCR were used to indicate the concentrations of Glyma09g31910 present. Primers were originally lyophilized and water was added to re-suspend primers to 100 mM. After 40-45 PCR cycles, SYBR Green fluorescent dye intensity was measured by a standard dissociation curve program.

PCR of gDNA to Isolate Promoter and Attach attB Boxes
PCR was performed to amplify the promoter sequence while adding 50% of the attB1 and attB2 boxes to each end. Primers ordered from IDT were re-suspended using water and were added to the extracted gDNA. Water, Buffer, Taq polymerase, and dNTPs were also components of the reaction. The final product was the promoter sequence containing 50% of attB1 and attB2 boxes on each side, now compatible for the Gateway System.

A second PCR amplification was run with our promoter sequence using attB1 and attB2 box adaptors to extend the B boxes to their full length. To precipitate the promoter DNA, PEG Precipitation Protocol was used (http://learn.ou.edu/d2l/le/content/2166077/viewContent/2952308/View). Once extended and isolated, promoters were compatible for recombination into pDONR/Zeo plasmid.

BP Clonase Reaction
BP Clonase Protocol (http://learn.ou.edu/d2l/le/content/2166077/viewContent/2952308/View) from Invitrogen was used to produce recombinant pDONOR/Zeo vector to introduce the promoter into the plasmid. Heat shock protocol (http://learn.ou.edu/d2l/le/content/2166077/viewContent/2961459/View) was used to transform E. coli with the promoter containing pDONOR/Zeo plasmid. Transformed E. coli was allowed to recover in SOC media (without antibiotic) and grow for 45 minutes.

Positive Clone Selection of Transformed E. coli Colonies
Transformed E. coli was plated onto LB agar plates that contained Zeocin. E. coli that had been transformed with plasmids that were not recombinant contained the selective ccdB gene that inhibits cell growth and did not grow on medium. E. coli that was not transformed did not grow on plates because it did not contain pDONOR/Zeo plasmid that has the gene for Zeocin resistance. E. coli colonies growing on the LB agar plates + Zeocin were confirmed to contain the Glyma09g31910 promoter using PCR with primers from our forward “screen” and our reverse attB2 box primer ordered from IDT (Fig. S1). The PCR product was analyzed using gel electrophoresis and a band around 533 base pairs was expected. The absence of bands in the gel (Fig, 2S) indicates that the “screen” did not anneal to the Glyma09g31910 promoter at 60 degrees C. Two new screen sequences were ordered from IDT to confirm the presence of the promoter and reporter after the LR Clonase reaction.

Although no colonies were confirmed to contain the promoter, two large colonies were transferred to LB culture and grown overnight in order to generate many plasmids within E. coli for extraction.

Plasmid Extraction from E. coli
Plasmids were extracted from the E. coli that had been incubated and grown overnight. Plasmids were extracted according to E.Z.N.A. Plasmid DNA Mini Kit I Protocol (http://learn.ou.edu/d2l/le/content/2166077/viewContent/2928234/View). Plasmids contain a sequence of attBP1 and AttBP2 boxes sequence flanking the promoter that is recognized by the LR Clonase enzyme.

LR Clonase Reaction
Plasmids containing the promoter extracted from E. coli underwent a reaction using LR Clonase protocol (http://learn.ou.edu/d2l/le/content/2166077/viewContent/2957702/View). LR Clonase reaction  produced recombinant pYXT1 donor vector containing the promoter upstream of  the GUS reporter. pYXT1 donor vectors contain the kanamycin resistant gene.

Electroporation of Agrobacterium rhizogenes strain K599 with pYXT1 Donor Vectors
Electroporation was performed according the Blue Electroporation-Competent Cell Protocol (http://learn.ou.edu/d2l/le/content/2166077/viewContent/2961455/View) from Stratagene. Electroporation transformed Agrobacterium to take up pYXT1 donor plasmids containing the kanamycin resistant gene.

Positive Clone Selection of Transformed Agrobacterium rhizogenes Colonies
Transformed Agrobacterium was allowed to recover in SOC medium and then plated onto LB agar plates containing kanamycin. Colonies were allowed to grow overnight incubating at 37 degrees C. Colonies were selected and used as DNA template for PCR to amplify 300 base pairs between two screens, the first from the promoter and the second from the reporter downstream of the promoter insertion. Gel electrophoresis shows positive bands for the promoter and reporter fusion around 300 base pairs (Fig S3).

Soybean Transformation
The soybean plants were transformed with Agrobacteria rhizogene containing the promoter upstream to the Gus reporter. Soybean plants were cut with ethanol sterilized scissors above the true leaves. Soybean cuts were inserted into Fibrogro cubes. Four mL. of A. rhizogenes was pipetted onto the cubes. Soybean cuts in A. rhizogenes absorbed cubes were covered and incubated at room temperature allowing roots to grow from the cubes. After 5 weeks, transgenic soybean plants were transferred to 3 parts vermiculite 1 part perlite mixture. Half of our transgenic plants were inoculated with B. japonicum and the others remained un-inoculated. After five weeks, the roots were scored for nodules.

GUS Staining
GUS staining buffer was made with X-Guc to reveal the presence of the substrate using a GUS Histochemical Assay with protocols located on D2L (http://learn.ou.edu/d2l/le/content/2166077/viewContent/2984222/View). Five week old transformed plant roots were cut from the soybean plant and incubated in the GUS staining buffer for 30 minutes. Roots were incubated overnight at 37 degrees C. Chemical reaction was then stopped and roots were stored at 4 degrees C until they could be observed under a microscope.


Supplemental Figures

Glyma09g31910 Promoter With Highlighted Screen and attB1 Forward and attB2 Reverse Primers





















Fig. S1. The first yellow highlighted region (TTGTGTGCATTAAGTTGTGAGC) is the attB1 Forward Primer ordered from Integrated DNA Technologies (IDT). The second green highlighted region (GTGAGTGTGGCATCATTATTAAC) is the “screen” primer ordered from IDT. The last yellow highlighted region (TTTGTCTCTGATTAGAGCTTCAAA) is the attB2 Reverse Primer region ordered from IDT. The green highlighted “screen” primer did not amplify and the gel electrophoresis after PCR using the screen primer and the attB2 reverse showing no banding is shown in Figure S2.

Gel Electrophoresis for Selection of Positive E. coli Clones Containing Glyma09g31910 Promoter







Fig. S2. Gel Electrophoresis of G1-12 and H1-4 do not show banding for a product isolated from the promoter to be around 533 base pairs. This indicates our primers did not anneal at the temperatures we used for PCR. As a result of this gel, two new screens were ordered to confirm the presence of the promoter in the pYXT1 plasmid.

Gel Electrophoresis for Selection of Positive Agrobacterium rhizogenes Colonies Recombination with Glyma09g31910 Promoter









Fig. S3. Gel electrophoresis of all columns was taken from the PCR of transformed A. rhizogenes colonies containing Glyma09g31910 promoter. Two new primers were ordered, one within the promoter region and the second within the reporter region. Bands were consistent with a PCR product around 300 base pairs, which was expected from the two new ordered primer “screens” from IDT.

Forward and Reverse Primers for qPCR of cDNA




Fig. S4. Forward and Reverse Primer sequences ordered from Integrated DNA Technologies (IDT). Primers were used to amplify the regions of DNA located between these two primers. The specificity of the primers allow us to infer that an abundance of the amplified product is related to the amount of original mRNA present in the tissues before cDNA synthesis.

Glyma09g31910 Gene Sequence









Fig. S5. Glyma09g31910 gene sequence was acquired using www.phytozome.net. The green region is an area of 5’ un-translated region from which the primers for qPCR were ordered. The genomic sequence is 1416 nucleotides.

Primers Used to Isolate Promoter Sequence and add attB1 and attB2 Boxes



Fig. S6. Glyma09g31910 promoter specific primers with 50% of attB1 and attB2 boxes attach for Gateway System compatibility.

attB1 and attB2 Adaptors to add Full Length Boxes to the Promoter



Fig. S7. attB1 and attB2 adaptors were used in subsequence PCR to amplify the complete boxes to be compatible with the Gateway system.

Gel Electrophoresis Primers for Positive E. coli Transformed Clones








Fig. S8. Gel electrophoresis was run with primers for the Glyma09g31910 promoter insertion into E. coli donor vectors. Columns G1-12 and H1-4 were each colonies of E. coli selected as template for PCR. The gel did not indicate definite bands of expected size and ccdB gene exchange and zeocin resistance selection was used to determine that colonies were, in fact, transformed. New primers were ordered with greater specificity to amplify Glyma09g31910 promoter region for future selections.


Gel Electrophoresis with Primers for Positive Transformed Clones




















Fig. S9. Gel electrophoresis was run with primers for Glyma09g31910 promoter and Gus reporter gene. Columns JV_k and JV_m show two positive colonies containing transformed bacteria to be selected for agrobacteria transformation. Two distinct bands of expected size can be seen to determine that the Glyma09g31910 gene of interest was inserted upstream the GUS reporter.


REFERENCES

Abba, S., Vallino, M., Daghino, S., Vietro, Di L., Borriello, R., and Perotto, S. (2011) A PLAC8-containing protein from an endomycorrhizal fungus confers cadmium resistance to yeast cells by interacting with Mlh3p. Nucleic Acid Research 39 (17): 7548-7563.
DI Vietro, Luigi, Daghino, Stefania, Abba, Simona, and Perotto, Silvia. (2014) Gene expression and role in cadmium tolerance of two PLAC8-containing proteins identified in the ericoid mycorrhizal fungus Oidiodendron maius. Fungal Biology 118: 695-703.
Geurts, R., Fedorova, E., and Bisseling, T. (2005) Nod factor signaling genes and their function in the early stages of Rhizobium infection. Elsevier. Current Opinions in Plant Biology 8:346-352.
Indrasumunar, A., Menzies, N. W., & Dart, P. J. (2012). Calcium affects the competitiveness of acid-sensitive and acid-tolerant strains of Bradyrhizobium japonicum in nodulating and fixing nitrogen with two soybean cultivars in acid soil. Soil Biology and Biochemistry, 46, 115-122.
Joshi, T., Fitzpatrick, M. R., Chen, S., Liu, Y., Zhang, H., Endacott, R. Z., ... & Xu, D. (2014). Soybean knowledge base (SoyKB): a web resource for integration of soybean translational genomics and molecular breeding. Nucleic acids research, 42(D1), D1245-D1252
Land Use Overview (2013).  U.S. Environmental Protection Agency. http://www.epa.gov/agriculture/ag101/landuse.html
Libault, M., Zhang, X.C., Govindarajulu, M., Qui, J., Ong, Y.T., Brechenmacher, L., Berg, R. H., Hurley-Sommer, A., Taylor, C.G., and Stacey, G. (2010) A member of the highly conserved FWL (tomatoe FW2.2-like) gene family is essential for soybean nodule organogenesis. The Plant Journal 62, 852-864.
Miransari, M., Riahi, H., Eftekhar, F., Minaie, A., and Smith, D.L. (2013) Improving soybean (Glycine max L.) N2 Fixation under Stress. J Plant Growth REgul 32:909-921.
NCBI Resource Coordinators (2015) "Database resources of the National Center for Biotechnology Information”
Oldroyd, Giles ED, and J. Allan Downie. "Coordinating nodule morphogenesis with rhizobial infection in legumes." Annu. Rev. Plant Biol. 59 (2008): 519-546.
Rubio-Wilhelmi, M. M., Sanchez-Rodriguez, E., Rosales, M. A., Begona, B., Rios, J. J., Romero, L., ... & Ruiz, J. M. (2011). Effect of cytokinins on oxidative stress in tobacco plants under nitrogen deficiency. Environmental and Experimental Botany, 72(2), 167-173.
Tauer, Loren W. (1989) Economic impact of future biological fixation technologies on United States agriculture. Kluwer Academic Publishers. Plant and Soil 119, 261-270.

endron maius. Fungal Biology 118: 695-703.
Geurts, R., Fedorova, E., and Bisseling, T. (2005) Nod factor signaling genes and their function in the early stages of Rhizobium infection. Elsevier. Current Opinions in Plant Biology 8:346-352.
Indrasumunar, A., Menzies, N. W., & Dart, P. J. (2012). Calcium affects the competitiveness of acid-sensitive and acid-tolerant strains of Bradyrhizobium japonicum in nodulating and fixing nitrogen with two soybean cultivars in acid soil. Soil Biology and Biochemistry, 46, 115-122.
Joshi, T., Fitzpatrick, M. R., Chen, S., Liu, Y., Zhang, H., Endacott, R. Z., ... & Xu, D. (2014). Soybean knowledge base (SoyKB): a web resource for integration of soybean translational genomics and molecular breeding. Nucleic acids research, 42(D1), D1245-D1252
Land Use Overview (2013).  U.S. Environmental Protection Agency. http://www.epa.gov/agriculture/ag101/landuse.html
Libault, M., Zhang, X.C., Govindarajulu, M., Qui, J., Ong, Y.T., Brechenmacher, L., Berg, R. H., Hurley-Sommer, A., Taylor, C.G., and Stacey, G. (2010) A member of the highly conserved FWL (tomatoe FW2.2-like) gene family is essential for soybean nodule organogenesis. The Plant Journal 62, 852-864.
Miransari, M., Riahi, H., Eftekhar, F., Minaie, A., and Smith, D.L. (2013) Improving soybean (Glycine max L.) N2 Fixation under Stress. J Plant Growth REgul 32:909-921.
NCBI Resource Coordinators (2015) "Database resources of the National Center for Biotechnology Information”
Oldroyd, Giles ED, and J. Allan Downie. "Coordinating nodule morphogenesis with rhizobial infection in legumes." Annu. Rev. Plant Biol. 59 (2008): 519-546.
Rubio-Wilhelmi, M. M., Sanchez-Rodriguez, E., Rosales, M. A., Begona, B., Rios, J. J., Romero, L., ... & Ruiz, J. M. (2011). Effect of cytokinins on oxidative stress in tobacco plants under nitrogen deficiency. Environmental and Experimental Botany, 72(2), 167-173.

Tauer, Loren W. (1989) Economic impact of future biological fixation technologies on United States agriculture. Kluwer Academic Publishers. Plant and Soil 119, 261-270.