Discussion
There were several reasons for choosing the Sugar Maple (Acer saccharum Marsh) and the Green Ash (Fraxinus pennsylvanica) as sample sources in this experiment. Both species are deciduous, and therefore share many physiological traits. Convenience was also a determining factor when deciding which leaves to use as samples. Both trees are very common in the Eastern United States, and are located in various places around JMU’s campus.
The hypothesis of the experiment, which stated that Rubisco levels would be higher in green leaves than in yellow leaves, was not supported by the experiment. Although it was determined that the green leaves contained more general protein than the yellow leaves, the relative levels of the protein of interest, Rubisco, were not found with any level of certainty.
In the first portion of the experiment, protein was isolated from each sample and the stock concentrations were determined. Then the sample protein was analyzed, using the Western Blot Protocol. As shown in Figure 4, the acrylamide gel did not show any protein bands, indicating that no protein was present in the samples. This result was consistent with the absence of antibody detection on the western transfer membrane Figure 5, which also suggests that there was no Rubisco in the samples.
These results are surprising because the original stock samples clearly contained protein. The presence of protein in the samples was determined by the original absorbance readings, shown in table 1, which were taken with a spectrophotometer. Knowing that the samples did have protein, and that the most abundant protein in plants is Rubisco, it is safe to say that the reason for the blank gels was not caused by the samples lacking Rubisco.
If the Western blot would have run as expected, visible bands would have been present in each lane around 47,600Da (between the green and violet markers). This is the area on the gel where the large subunit of Rubisco would have migrated. After the proteins from this gel were transferred onto a membrane, they were treated with primary and secondary antibodies, linked to horseradish peroxidase, which would have resulted in antibiotic detection at all the places where the Rubisco protein had bound.
There are many possible reasons why a sample that is known to contain protein could produce contradictory Western Blot data. One possibility is that the proteins were significantly degraded between the time their absorbencies were recorded and the time they were run in the western blot. Such denaturation could occur via sudden changes in pH, temperature, or from the addition of an unknown contaminant in the environment of the samples. However, it is very unlikely that the degradation would be significant enough to prevent all of the Rubisco from being visible in the gel. What is even less likely is that such considerable degradation occurred in all four sample vials. Another situation when inaccuracies in protein data can arise is when the expression of another protein in the leaf is associated with an inhibitor, whose presence somehow prevents Rubisco’s expression. However, this scenario also does not seem applicable in this case. When an unwanted inhibitor is associated with the protein of interest, there is an inverse relationship between the sample’s protein concentration and the signal observed. This is because, as the total protein concentration increases, so does the inhibitor, which would in turn decrease the expression of Rubisco. If this were the case, the lanes with the highest amount of protein would have the weakest signals. On the experimental western blot gel and transfer membrane, there were not any visible signals.
A more plausible explanation of the unexpected Western Blot results could be that some kind of miscalculation was made when deciding how much of the protein samples to load into the wells of the acrylamide gel. The amounts loaded into each gel were chosen so that 30ug of protein would be run in each lane. It is possible that the necessary amount was calculated based on the concentration of protein in the stock solution, and then actually taken from one of the diluted solutions, which were used to record absorbencies. If this is the case, it certainly accounts for the lack of signal visible on the gel and transfer membrane.
Whatever the reason for the absence of bands and signaling in the acrylamide gel and transfer membrane, the same conclusion must ultimately be drawn; the data obtained from the protein analysis does not refute the null hypothesis. While table 1 seems to suggest that the green leaf samples (S1 and S3) have a higher protein content than the yellow leaf samples (S2 and S4), without reliable data from the western blot gel and transfer membrane, there is no way of telling if the stock protein levels reflect Rubisco levels.
Unfortunately, the DNA analysis also did not provide any additional insight for comparing the levels of Rubisco between yellow and green leaves. The DNA region that codes for the Rubisco in the chloroplast was amplified using Real Time PCR. In accordance with the hypothesis, the expected results of the RT-PCR were to get lower Ct values in the green samples, than in the yellow samples. During real time PCR, quantitative data is obtained by a computer program that measures the replication of a particular gene. The program quantifies replication by monitoring the appearance of syber-green dye. The number of cycles it takes for the appearance of syber green to cross the appearance threshold is the Ct value. As a result, genes that are more abundant will cross the threshold faster before genes that are less abundent. The PCR results also include melting curves for each of the samples. The melting curve tells the experimenter what temperatures the C(t) values were measured at and is necessary for determining the Tm.
The observed results of the PCR reaction showed that no C(t) values were could be established from the samples run. As was the case with the protein analysis, it seems that there was not enough DNA loaded into the samples. Each of the four samples were run at two different dilutions, with two different primers, for a total of 16 PCR reactions. The concentrations of the DNA obtained prior to running the reaction imply that each sample's DNA concentration was more then sufficient for the reaction to run properly. However, a comparison of the absorbencies at 260nm with the absorbencies at 280 is indicative of the samples being impure. The 260nm level is significant because the nitrogenous bases of DNA absorb light at this frequency. The aeromatic portion of amino acids absorb light around 280nm. The average of the ABS260/ASB280 ratio was .958. A ratio around 1.8 is regarded as a good sample, while a ratio around 1.2 is regarded as bad (impure). At this point in the experiment, the DNA should have been re-purified, and then its concentration re-recorded. A sample being this impure would prevent the PCR reaction from running as intended. This is because the amount of sample that was alloquoted into each PCR tube was based on the assumption that the samples were relatively pure. Such a mistake would certainly account for the melting curve obtained from the PCR. In addition to increasing the dilution factor, an impure sample could have also introduced potentially harmful contaminates to the reaction. This could have also added to the faulty PCR results.
An agarose gel run by electrophoresis confirmed the results of the PCR reaction. The only staining seen was at the very bottom of the gel, which was probably caused by the excessive amount of primer used in the reaction. If the gel had run as expected, bright bands would be present wherever the amplified gene had migrated. The absence of bands on the gel support the idea that the samples weren't pure enough to be properly amplified.
The results for the protein and DNA analysis did not provide enough evidence to accept the hypothesis. However, this is not a result of the experimental layout being problematic. The main reason this experiment was not successful seems to be a result of miscalculations made by the experimenter. If the experiment were repeated, any calculations involving estimation would be much more conservative, so that if anything, the concentrations would be slightly greater than intended. Additionally, in the repeated experiment more care would be taken to isolate the protein and DNA under sterile conditions, to avoid any unknown contaminants that could produce inaccurate results. It would be interesting to repeat the experiment, using various species of leaves, to see a correlation between leaf color and Rubisco level is apparent on a broad scale. It would also be useful to examine the effect chlorophyll degradation has on functional plant mechanisms other than the expression of Rubisco.