Embedded Files
Mean: 0.114, Standard Deviation: 0.124
Experimental Data Table
Experimental Data Table
This data table was produced by the computer attached to the microplate reader. The inner pink wells represent the inner wells used. The outer wells did not contain any cells, so these data points were not used.
Mean: 0.049, Standard Deviation: 0.036
Control Data Table
Control Data Table
This data displays the values produced by the control plate. This plate contained regular growth media, not the media containing Roundup.
Interpreting the Data
The values in the data table are absorbance values. An absorbance value is calculated using the logarithm of the ratio of the intensity of light transmitted through the sample. When using a microplate reader, you input the desired wavelength into the computer, which in my case was 490 nanometers because this is the wavelength that the stained cells with oxidative stress emit. This means the microplate reader gives you a number that corresponds to how much of that specified wavelength is being emitted. The absorbance value is simply put, a measurement of the concentration of cells emitting the wavelength that signifies oxidative stress in each well.
According to Vernier Science Education as well as several other papers, "the useful absorbance rate is 0.1 - 1," meaning values below 0.1 are considered extremely low concentrations and values above 1 are considered extremely high (Vernier, 2014). In both data tables, the majority of the absorbance values are below 0.1, so it is likely there was a "low" amount of oxidative stress in both. However, I can't make the claim that the oxidative stress in both groups was low because without a positive control, I am not certain what a high amount of oxidative stress should look like.
Statistical Analysis
To analyze my data, I ran a 2 Sample 1 Tailed T Test for the alternative hypothesis that the experimental group caused more oxidative stress than the control. This test is used to calculate the probability of getting your data by chance. A lower probability corresponds to higher statistical significance. I used the two means and standard deviations above to calculate the P value using the test function on my calculator. The P value = 0.00037. This means assuming there is no difference between the true experimental absorbance value mean and the control value mean, the probability of getting a difference in sample means of 0.065 purely by chance alone is 0.00037. I used the significance level 0.01, which is equivalent to being 99% confident when rejecting the null hypothesis. Because 0.00037 is less than 0.01, I reject the null hypothesis. There is convincing evidence that the experimental group caused more oxidative stress than the control group.
Close-up Graph of Control vs Experimental
Close-up Graph of Control vs Experimental
Using Google Sheets, I created this graph of the control and experimental means.
Graph of Control vs Experimental
Graph of Control vs Experimental
This graph is a more zoomed out version of the one on the left. It shows the values compared to the linear absorbance value scale where 0.1-1.0 are usable values. I created this separate graph to show that although the experimental mean is higher, in comparison to the absorbance value scale, both values are low.
Discussion
My research question was "to what extent does Roundup, a glyphosate-based herbicide, cause oxidative stress in fibroblast cells?" I hypothesized that Roundup would cause significantly more oxidative stress than the control group. Based on my data, I fail to reject this hypothesis. Although the means 0.114 and 0.049 were both low values, the experimental group did have a significantly higher mean than the control. It is highly likely that Roundup caused more oxidative stress in the cells than the control.
Throughout the project, I experienced several limitations. The "Challenges/Process" page of my blog goes more in depth into the various changes I had to make to my project to accommodate some road blocks. While these alterations were inconvenient, I don't feel that they severely impacted the results of my project. While performing experimental trials, a major limitation was timing. Once seeded into a 96 well plate, cells take 24-48 hours to adhere. However, it was already the end of the week when I approached this step, and I didn't have enough time to start it the following week. I left the cells to adhere over the weekend which was 72 hours. This was too long of a period because the cells became overconfluent. I elaborate on this further in on the "Challenges/Process" section. Despite the cells being overconfluent, I decided to expose them to the Roundup anyway because I didn't have time to completely restart, and the cells weren't dead yet. Fortunately, the cells ended up not dying due to overconfluency, so I was able to continue my project. This could still be a limitation because the cells were not totally healthy during the experiment due to being overconfluent and not receiving enough nutrients. However, because both the control and experimental groups were overconfluent, I can still make the conclusion I did, as the experimental group had more oxidative stress regardless. An additional time limitation was the exposure to Roundup timing. I was again running out of time and had to adjust my project to work with the school schedule, so instead of exposing cells to Roundup for a full 24 hours, I was only able to do 21 hours and 30 minutes. I still got significant data, but the results could have changed if I exposed them to Roundup for longer. Ideally, I would not have to account for weekends and class periods, but as a high school student, these were limitations I had to account for.
Furthermore, a limitation was the data in the outer wells of the 96 well plate. The outer wells were left completely empty and only the inner 60 wells contained cells for both plates. This means the absorbance value for the outer wells should be zero. However, as I was unfamiliar with using a microplate reader, I was unaware that you shouldn't touch the bottom of the plate to avoid fingerprints that could create a false reading. I held both plates by the bottom when placing them into the reader, which is what led to the outer wells showing an absorbance value above zero. I cannot say if the fingerprints also impacted the results of the inner wells. Similarly to the time limitation, I can still be relatively certain about my conclusion because the experimental group yielded higher results regardless. Because I touched the bottom of both plates, this limitation likely didn't severely impact my results. I decided not to include the data captured by the outer wells when calculating the mean and standard deviation because this data doesn't correspond to any cells and is irrelevant.
Healthy NIH cells compared to the overconfluent cells
Cells are clumped together
96 well plate, only inner wells used
Conclusion
Overall, this research addresses the gap in the body of knowledge surrounding glyphosate-based herbicides. As no previous studies had analyzed the effects of a GBH as a whole on oxidative stress, this project has added new information to our current body of knowledge. Glyphosate is still an extremely controversial chemical. In fact, the American Academy of Pediatrics recently changed their stance on glyphosate and GMO foods which has resparked controversy in the scientific community (Abrams et al., 2024). It is crucial to keep studying this chemical in hopes of eventually reaching a consensus, so this research helps us get one step closer to that goal. Additionally, many journal articles suggested that the next step is to do more research on GBHs as a whole rather than just glyphosate. Testing Roundup on a crucial cellular pathway has addressed an important gap.
The next step after this research is to continue analyzing the effects of Roundup-induced oxidative stress on cells. Due to limited time, I was unable to perform additional assays that could determine the effect of oxidative stress. A live/dead cell counting assay could determine the number of cells that had oxidative stress and died. Using fluorescent imaging, I could have analyzed changes in cell structure after oxidative stress. Oxidative stress has two major effects which are inducing cell death and damaging cell structure. Testing these two assays could help provide further information on if the oxidative stress actually impacted the cell negatively. Additionally, future steps would be to test this experiment on human cells. Mouse cells, while similar, are not exactly the same as human cells. I could not work with human cells in our high school lab, so future studies could involve performing this experiment at a higher level lab and using human cells. The implication of these studies would be having more knowledge on the possible health effects of glyphosate-based herbicides.
References
Vernier Science Education. (2024). Why is the absorbance reading on my device (spectrometer/colorimeter) unstable or nonlinear, at values above 1.0? Vernier. https://www.vernier.com/til/2589
Abrams SA, Albin JL, Landrigan PJ, et al; American Academy of Pediatrics Committee on Nutrition; Council on Environmental Health and Climate Change. Use of Genetically Modified Organism (GMO)-Containing Food Products in Children. Pediatrics. 2024;153(1):e2023064774
Page updated
Report abuse