The plant Dieffenbachia Camille has been selectively bred over generations to display variegated leaves that are pleasing to the eye. The distinct white and green splash pattern on the leaves is a result of a mutation in the pigment in chlorophyll of the cells on the leaves. This unique patterning of the leaves attracted this research into whether rubisco is expressed in greater quantities in different areas of the leaf. We suspect that the areas of the leaf that have the mutation for the non functional chlorophyll, the white sections, will have a reduced amount of rubisco protein. While the mutation itself does not affect the gene for rubisco or its light independent photosynthetic pathway, the nonfunctioning chlorophyll may result in the down regulation of photosynthetic proteins such as rubisco in an effort to save energy.

To begin the experiment raw DNA and protein were extracted from white and green plant tissue. These procedures went well, obtaining usable amounts of DNA and protein from each type of tissue. The ratio of A260/A280 was very near 1.8 so we suspect that our DNA was mostly free of protein (table 1). Using this DNA preparation, real time polymerase chain reaction was used to quantify the level of the rubisco gene present in the chloroplast of the cells. The protein samples were used in western blots that non specifically targeted protein and in another case targeted rubisco specifically with antibodies.

Western Blot
            Electrophoresing two SDS PAGE gels containing identical materials and Dieffenbachia protein for the green leaf and white leaf were then marked in two separate ways. The first western blot gel, which was non-specifically stained with comassie blue was intended to confirm the presence of protein in the sample. After imaging the gel under visible light (figure 8), it was clear that neither protein sample was stained. It can be concluded that either the proteins extracted from dieffenbachia were at to low a concentration to be stained on the gel, or that no protein was present. Further results from the antibody tagging of the second western blot will help confirm the presence of protein. In a non specific stain such as this, we expected to see large amounts of protein collected from the leaves present in both lanes. In the future, increasing the nanograms of protein put into each well might give better results. Due to the low concentrations extracted of white leaf protein (.59mg/ml), only 18 micrograms of protein was able to be placed into each well due to volume restraints. 18mg is 12mg below the recommended value for this experiment and perhaps increasing the concentration of protein might improve the staining results. Also, including a positive control on the gel along with the molecular weight marker will help determine if there is an error in staining or thin the stain itself.

            Using the second SDS PAGE gel containing Dieffenbachia protein, it was electro blotted onto a membrane. The membrane was then washed and exposed to primary chicken antibody and then secondary bovine antichicken antibodies. This technique vastly increasing the resolution of detection of proteins and can bind specifically to target proteins. In this case, using a primary antibody specific to rubisco allowed for detection of the only the rubisco protein in each type of leaf. After bathing the gel in luminal and exposing it in the BIORAD Chemiluminescent imager, distinct bands were present in both lanes of Dieffenbachia protein (Figure 9). These results confirm the integrity of the green and white leaf protein samples, but go against the hypothesis. The two western blots in this experiment were intended to confirm the integrity of the extracted protein, and then target rubisco to check for its presence or absence in the tissue. According to the hypothesis, higher levels of rubisco were expected in the green leaf samples and the white leaf samples were expected to have little to no rubisco. The data from the results indicate that both white leaf area and green leaf area express rubisco protein in the cell. The two luminescent bands present on the membrane show that the antibodies were successfully able to target and bind the protein in both tissue types. The expected results would have been to see a luminescent band only in the green tissue lane; however both lanes showed bands of equal strength. It can be concluded that white leaf tissue from Dieffenbachia expresses rubisco protein similar to the green tissue. Future experiments could go further and attempt to quantify the amounts of rubisco protein present in each tissue as the antibody tagging preformed cannot accurately quantify protein amounts. While the green and white tissues both express the protein, they may express it in different amounts. The white leaf tissue, which lacks functional chlorophyll may attempt down regulate proteins associated with photosynthesis in an attempt to conserve energy. Using another technique such as an immuno-assay to accurately measure the amounts of rubisco protein in each tissue might elucidate the amount of expression seen in each type of cell.

RT-PCR
Real Time Polymerase Chain Reaction is a technique used to give accurate data on the rate at which DNA is being amplified. By measuring the fluorescence of each sample at set intervals during the reaction, graphs of the amplification as well as the melting curves can be created. These can be then analyzed to determine sample integrity and the rate at which they replicate. Using two sets of primers specific to the rubisco subunit gene in chloroplast DNA, two concentrations of each Dieffenbachia tissue type were run. Since the primer sets were universal primers and not made specifically for Dieffenbachia rubisco, two sets of primers were used, primers rbcl2f/ RBCL-fonfana (primer A) and primers rbcl2f/ RBCL-Savolainen (primer C). These two primer sets were chosen because they amplify small fragments (200-500bp) which are recommended for RT-PCR. Using two sets of primers also gives added flexibility when analyzing the data in the even that one does not bind well or fails to amplify correctly. For the tissue samples, two concentrations of extracted DNA were run of each sample (50ng and 100ng) in an effort to estimate an optimal concentration for best results in the RT-PCR.

Figure 2. - A graph of an RT-PCR melting curve containing four tissue samples from Dieffenbachia Camille as well as a control containing no DNA. Identified by color the major peaks signify melting points of the sample. This graph illustrates the abnormal position of the 100ng and 50ng green tissue samples using primer set A.

To begin analyzing the graphs from the RT-PCR, the melting curve of the entire reaction is shown and an optimal temperature is chosen to take plate reads from. Figure 2 shows a melting curve of products of primer set A for both white and green tissue at two concentrations and a control containing primer set A but no DNA template. The peaks associated with the green tissue (red and dark blue lines) are cause for concern. It was expected to see the template containing samples to peak around 83 degrees; however these peaked below the primer dimer peaks of the control. It can be concluded that primer set A did not properly bind in the green tissue samples either due to operator error or an inability to attach to the template. Due to this, the results for both tissues for primer set A cannot be used.

This abnormal results from primer set A may be due to a variety of reasons. Primer set A only differed from primer set C in one region. It's likely that the primer set A samples, problably the 100ng green and the 50ng green samples, had some sort of contamination. This would alter the ability of the primers to bind and give skewed results on the RT-PCR.

Figure 3. – A graph of an RT-PCR melting curve containing four tissue samples from Dieffenbachia Camille as well as a control containing no DNA, all using primer set C.  Identified by color the major peaks signify melting points of the sample. Specifically this graph illustrates the optimal plate read temperature of 81 degrees.

The data collected from primer set C came out more favorably. Figure 3 illustrates the melting curve of white and green tissue at 50ng and 100ng using primer set C and controls containing primers C and primers A. All four samples peaked at approximately 83 degrees. The controls peaked lower at approximately 79 degrees. By looking at the peaks on Figure 3, the cycle for a clean plate read was determine to be 81 degrees. At this temperature, the primer dimers of the control seen melting of at 79 degrees are not present but the DNA product has yet to be melted.

Figure 4. – A plate read of green and white tissue samples at 100ng concentration and one control using primer set C at step 11 (81 degrees).  The dotted line sets the threshold at which the Ct values are determined (table 2).

Using step 11 as the plate read (81 degrees), figure 4 illustrates the rate at which each DNA sample amplified. The highlighted red line, the 100ng green tissue, started and ended with more product and displayed a Ct value of 16. The green line, the white 100ng sample, amplified slower and had a Ct value of 17.71. The control for comparison had a Ct value of 32.9. Based on these results for the 100ng sample, it can be said that the green tissue has a higher concentration of rubisco DNA.

Figure 5. – A plate read of green and white tissue samples at 50ng concentration and one control using primer set C at cycle 11 (81 degrees). The dotted line sets the threshold at which the Ct values are determined (table 2).

            Using cycle 11 again, this graph displays the results for the 50ng tissue samples along with a control for primer set C. Unlike the first set of tissue, these show the white and green tissue samples at nearly the same Ct value (Table 3). In fact, the white sample had more product at cycle 11. This calls into question the initial conclusion that green tissue had more rubisco DNA. These results indicate that the rubisco chloroplast DNA present in tissue is more likely to be equal.

Figure 6. - A graph of an RT-PCR melting curve containing four control samples and one tissue sample from Dieffenbachia Camille, using two primer sets as indicated in the legend. The highlighted brown control using primer set C has an irregular peak far downfield from the other three controls. The other controls, two from primer set A and one from primer set C, in addition to one DNA sample were added for reference to the abnormal peaks of the brown line.

            This graph of the melting curve of all 4 controls and an arbitrarily determine tissue sample illustrations the contamination of one control. The brown highlighted line was a control for primer set C. As compared to the peaks of the other 3 controls, two for primer set A and one for primer set C, it was not what would be expected of a non template containing tube.  Fortunately, due to redundancy in the controls, these results were determined to be an isolated contamination and another primer C control created in the same conditions was used.

            Using the RT-PCR product, an agarose gel was run. By performing this procedure, purity of the PCR products can be confirmed and the presence of rubisco specific product can be verified by comparing the size to a molecular weight ladder. After staining with ethidium bromide, the resulting gel was not what was expected. As seen in figure 1, the gel presented a series of smears instead of bands. The molecular weight marker smeared as well which suggests that there might be a flaw in the gel or an error in staining. Unfortunately these results are inconclusive. The RT-PCR graphs though show what is believed to be DNA product so it is still possible that the RT-PCR amplified the correct gene. However the lack of banding on the gel is disturbing and makes it so it cannot be said with 100% certainty that this is in fact the rubsico gene being amplified.

            Based on the results collected through these series of experiments we can reject our hypothesis that white tissue would have less rubisco than green tissue in Dieffenbachia Camille. Both the western blot and RT-PCR results indicate that each tissue type expresses rubisco protein and has a measurable amounts of rubisco chloroplast DNA. One reason why the results turned out in this way is due to the nature of rubisco and the mutation that makes these planted variegated. The non-functional chlorophyll and the rubisco protein exist in different parts of the photosynthetic system. Chlorophyll an integral part of the light dependant pathway does not interact with the light independent rubisco. The mutations in the pigment of the leaves may have not been far reaching enough to have an effect on the rubisco gene or protein. There might be slightly different levels of DNA but express the same amount of protein as visible on the western blot. It could be a possibility that rubsico is regulated at the transcriptional level or another form of control besides the levels of DNA.