Proof of Concept
Introduction
Fungilyzer is intended as a biological fertilizer for agriculture to store and provide nutrients to plants in a sustainable way. To simulate this environment and the relationship of the fungi with living plants as well as its efficiency and effect on plants, we tracked plant growth and phosphate levels in soil. This idea works as the blueprint for a proof of concept we tested with nonmodified yeast.
Figure 1-2: The zucchini plants in our laboratory
General Methods
The main goal for our plant experiments is to show the effect of phosphate, our target nutrient, in plants, in combination with the model organism S. cerevisiae and the modified Fungilyzer. To test and prove how Fungilyzer can improve plant growth, we ran two testing series using Cucurbita pepo subsp. pepo convar. giromontiina as a model organism. Our original plan was to work with Arabidopsis thaliana, however, we switched to zucchini after talking to Prof. Bauer from the botanical institute of the HHU. She recommended this plant species to us, as the effects of nutrient deficiencies would be more noticeable in it. Since showing how small differences in nutrient concentrations, in our case phosphate, can drastically change the growing speed of plants is a vital aspect to our proof of concept, we decided to take her advice.
Our first testing series started in the beginning of September and consisted of two groups, one with and one without Saccharomyces cerevisiae BY4741. This was to test if nonmodified yeast could by itself influence the growth of plants. These plants are part of our control groups. We had originally planned to repeat the experiment containing another group of zucchini that would be fertilized with our modified yeast, however, our modified organism was not readily transformed in time to carry out such a test.
We planted in “Einheitserde”, a type of soil containing nearly no nutrients and watered all of our plants with Hoagland solution of different compositions, so we could carefully control and compare different nutrient conditions. The composition of the ‘Einheitserde’ is as follows: Nitrogen (N) 200-500 mg/l, phosphate (P2O5) 200-500 mg/l, potassium oxide (K2O) 300-1000 mg/l.
Starting off with the first series, two zucchini seeds per pot were planted in the same amount of soil. The seeds were planted about three centimeters deep as described on the packaging instructions and with a thin coat of sand on top to prevent insects or other organisms from utilizing the soil for their benefit and thus interfering with results. Afterwards we watered the plants with Hoagland solutions of different compositions. The first series consisted of two subgroups: the first control group and the second control group. The first control group (Table 1) consists of seeds planted in plain soil regularly watered with a Hoagland solution of the same composition apart from two variables, copper concentration of 0,5 μM of CuSO₄ on the one hand and 100 μM of CuSO₄ on the other hand; as well as concentrations of either 0 mM or 1 mM of KH₂PO₄. The total composition of the Hoagland nutrient solution is to be explained below in further detail. The planting process for the second control group of series 1 (Table 1) was the same. The sole difference was that the second control group is aiming to show how S. cerevisiae affects plant growth and nutrient uptake in the soil. S. cerevisiae was added to the soil with a separate pipette while watering the individual pots with Hoagland solutions in order for the water to spread the cells evenly throughout. We added 1 mL of 1 OD₆₀₀ yeast which contains approximately 15,3 * 10⁶ yeast cells to each of the pots.
Table 1: Overview of Control Group 1 and 2 of Series 1
The plants were grown for a period of 21 days for control group 2 and 25 days in case of control group 1 which was planted a bit earlier. The planting pots were placed under a table with minimal natural lighting, instead they received artificial sunlight from lamps for 16 hours a day. We measured the height of the plant from just above the soil up to the highest standing leaf every few days. However, we found some issues in the comparability of this data as our plants grew and leaf sizes showed increasing variation. Therefore, we planned on using a different method in our second planting series.
The second series was planted with the same methods and under the same conditions also with two control groups, plain soil and yeast in the soil with one small change being that during planting, we did not immediately water the plants with Hoagland solution but instead let the soil soak up some demineralized water. After three days, when the initial moisture had dried down, we continued watering with Hoagland solution each time, as we had done in our first testing series. Originally, we had planned a third group with our modified yeast to compare if it has a better impact on growth than nonmodified yeast, however, this wasn’t possible anymore due to time running out. We had plans for improving our method of measuring plant height to be more unified in this testing series, measuring only the length of the shoot axis and not up to the highest leaf. Also, we hoped to measure the plants with methods that would prove more insightful, such as measuring the surface of the leaves. So far, we didn’t get to utilize these methods, as our second planting series only started sprouting and the plants were not of measurable height at the time of writing this Wiki.
Hoagland
Within these groups, there were two different concentrations of CuSO₄: 0,5 μM and 100 μM. This was necessary as our construct containing the death circuit is inhibited by copper and we needed to check if higher concentrations of copper would significantly impact growth. After fertilizing, the copper concentration would slowly drop in correlation with the phosphate concentration until reaching a critical level at which the death gene would be expressed. We also included four different levels of phosphate to study how much the phosphate available through the soil or through fertilizer influenced the growth of plants. We chose the levels of 1 mM, 0,5 mM, 0,1 mM and 0 mM of KH₂PO₄, with 1 mM being the recommended concentration of phosphate in Hoagland solution. This was because we wanted to get an overview that was as complete as possible about the different types of phosphate conditions one could find in soil out on fields and how, in theory, our Fungilyzer would interact with plants under these conditions. The 1 mm of phosphate represent ideal, nutrient rich soil, whilst the others show conditions under which fertilizing and the use of Fungilyzer might be more needed in practice.
Table 2: Hoagland Solution Composition1
Results
During the growing phase the plant height was observed and documented. At the end of the series we observed and documented the following variables: final height, biomass, amount of leaves and their special features, the amount of flower buds that had already formed on some of the plants and the levels of phosphate in the soil. The goal is to compare and draw conclusions for the effect the modified yeast (Fungilyzer) would have on the plants in comparison to the control groups.
Only plants that grew at all were put into these statistics as we counted not sprouted seeds as outliers that are not likely to have any correlation to the conditions set by the nutrients or the yeast.
Table 3: Overview on the Percentages of Sprouted Plants From Each Group in Series 1
Plant Growth
Figure 3: The growth diagram shows us the effects of yeast over the overall height of the plants. In the first week of growth there is no statistically significant difference between the two groups. This trend continues until the 17th day, after that point the overall height difference starts to be visible. By the end of the third week overall height difference reaches 0.3 cm
Yeast Height Correlation
Figure 4-8: The samples were compared using a Mann-Whitney U-Test.2
- n.s. not statistically significant to a level of 0,05 or lower
- *statistically significant to a level of 0,05
- **statistically significant to a level of 0,01
These groups compare the impact of nonmodified yeast on the growth of zucchini over 21 days, side to side with zucchini grown by themselves without any added other organisms. The group with yeast contains the growth data of 28 plants, the group without yeast includes the data of 30 plants that grew. The x-axis is the height of the plants in cm, the diagrams are seperated by the age of the plants as of measuring the height.
Looking at the results of this comparison, it’s possible to hypothesize that the existence of nonmodified S. cerevisiae together with the plants acts as a statistically significant boost to plant growth in the first stages of development. However, this influence seems to cease after a certain period. On the other hand, we did not carry out any microbiological analyses of the soil so we could not confirm the survivability of yeast in soil. Literature also seems to confirm that yeast can have growth-boosting properties for plants. It can improve phosphate solubilization and improve the plant’s resistance to stress and plant pathogens3 and be able to boost crop plants' height5. These observations would support the results of our testing series.
Copper Height Correlation
Figure 9-12: The samples were compared using a Mann-Whitney U-Test. 2
- n.s. not statistically significant to a level of 0,05 or lower
Figure 13: The samples were compared using a two sample t-Test with a significance level of 0,05.
- n.s. not statistically significant to a level of 0,05 or lower
The diagrams depict the growth of zucchini plants over the course of 21 days separated into those who were watered with low copper Hoagland solution and those watered with high copper Hoagland solution. Both groups are made up of 29 plants each that were all grown under different phosphate and yeast conditions which will be compared separately in the diagrams below. The x-axis is the height of the plants in cm, the diagrams are seperated by the age of the plants as of measuring the height.
Visible above are two box plots with different Cu concentrations. Box plot one ( low Cu, high P) shows strong growth in height (cm) after the seventh day. Between days 13 to 17 the height has no statistically significant change, but another noticeable change comes after the 17th day. Maximum average height after 25 days has been measured as 21cm for low Cu and high Phosphate conditions.
Box plot two (high Cu, high P) shows strong growth (cm) after the seventh day. A similar growth rate can also be seen after the 13th day for the zucchinis. Maximum average height after 13 days has been measured as 11 cm for high Cu and high Phosphate Conditions.
Box plot two has height (cm) data until the 13th day which is why the comparison of these graphs will include the data until day 13. When the two box plots are compared, we can see noticeable differences and also similarities in growth rate and time. As a clear similarity, we can say that the germination process takes approximately seven days for a Zucchini plant as in both groups we see small sprouts on day seven. The strong growth after the seventh day can also be observed in both low Cu and High Cu conditions. In order to see the effect of the two Cu conditions the height of the Zucchini plants on days seven, ten, and 13 can be compared. The zucchini plants with low Cu conditions always have an approximate positive Δ 1.0 cm in comparison to the high Cu conditions. For example, on day ten the zucchini plants have an average height of 9.6 cm in high Cu conditions, on the other hand, the high Cu group has an average height of 10.5 cm which shows an Δ 0.9 cm between the two conditions.
Copper is normally an essential plant micronutrient and lack of it can cause yield loss. However, in our experiment, we used Cu for internal communication which is why our “low Cu condition” is actually between the “normal” intervals and High Cu conditions are actually over the maximum limits. We initially hypothesized that the high Cu level has a toxic effect on Zucchini plants which is what we disproved with this experiment and the significance tests.
According to the data from our experiment, the difference between the growth of plants with 0,5 μM of CuSO₄ available and that with 100 μM of CuSO₄ available is statistically insignificant.
Copper Biomass Correlation
We also analyzed the correlation between copper concentration and biomass.
Figure 14: We compare the biomass of copper in different concentrations of low copper [0,5 μM] dark blue and high copper [100 μM] light blue. On the x-axis is the copper concentration and on the y-axis the biomass in gramm.
The significance of the data was tested with the Mann-Whitney-U test 2 and the results came up as statistically non-significant, which means that the different copper concentrations did not influence the growth of the plants.
Phosphate Height Correlation
Figure 15-19: The samples were compared using a Mann-Whitney U-Test. 2
- n.s. not statistically significant to a level of 0,05 or lower
- *statistically significant to a level of 0,05
- **statistically significant to a level of 0,01
- ***statistically significant to a level of 0,001
The diagrams above compare the growth of zucchini plants that were watered with three different amount of phosphate inside the Hoagland solutions and one control group with no phosphate added over the course of 21 days. In the no phosphate, low phosphate and medium phosphate group, 15 plants grew for each group. The high phosphate group only contains 13 grown plants. Each of the groups also contains different copper and yeast conditions. The x-axis is the height of the plants in cm, the diagrams are seperated by the age of the plants as of measuring the height.
The analysis of our experiment data concludes that the statistical significance of phosphate influence on plant growth shifts over the course of the three weeks of growth. However, at any stage in the plants development, there were groups that showed statistically significant differences in height, most often between the groups with no phosphate and with a moderate amount of 0,5 mM of KH₂PO₄. It also seems to be that in the beginning of growth, there is a large difference between the high phosphate and the no phosphate group, which disappears as the plants grow older.
The following sets of data are not completely comparable, as the plants were not the same age when we wrote down the biomass, number of leaves and buds, so the results of these comparisons may differ from the actual growth of plants. The plants that were grown with yeast are only 21 days old while the plants grown with no yeast are 25 days old.
Phosphate Biomass Correlation
The biomass of all plants in series one was documented and analyzed based on the phosphate concentration and yeast content.
Figure 20: The effect of 4 different phosphate concentrations on biomass is shown. The test group was grown in soil without any yeast, in high phosphate concentrations [1 mM] shown in dark green, medium Pi concentrations [0,5 mM]shown in green, low Pi concentrations [0,1 mM] shown in light green and in soil with no phosphate [0 mM] shown in a whitish green. On the x-axis is the phosphate concentration in M(Molar) and on the y-axis the biomass is measured in gramm.
Plants grown in high phosphate concentrations stay in a range around 6 gramm with the outliers lying at 8,6 and 2,8. Plants grown in medium phosphate concentration stay in a broader range between 6 and 3,5 gramm with no outliers. In low phosphate conditions the biomass of plants hits a maximum amongst all others at 7,2-5,8 gramm with an outlier at 3,5 gramm. The largest range has plants grown in soil with no phosphate going from 6,5 gramm to 3,8 gramm.
Based on the statistical test one-way Anova that was conducted, the significance of phosphate in biomass differences in the plants is insignificant for our data.
The next group was documented with focus on the same variable, phosphate concentration, the only difference being that S. cerevisiae was added in the soil of this group.
Figure 21: The effect of phosphate concentrations in plants grown in soil with S. cerevisiae. The deep yellow shows data on high Pi concentrations [1 mM], the yellow medium Pi concentration [0,5 mM], the light yellow low Pi concentration [0,1 mM] and the cream color stands for no Pi in soil [0 mM]. On the x-axis are the 4 phosphate concentrations in M (Molar) and on the y-axis the biomass is measured in gramm. Between the group medium Pi and no Pi is a statistical significance [*] on Pi effect.
Plants grown in high Pi concentrations have a shorter range in comparison to the rest of the groups, starting from around 7 gramm to just under 6 gramm. Plants grown in medium phosphate concentration stay in a range between 7,8 and 6 gramm and hit a maximum amongst all others, with an outlier just under 6 and one reaching the highest data point of all the groups at 11 gramm. In low phosphate conditions the biomass of plants ranges from around 7 to 5 gramm. Last but not least the group of no Pi has the lowest biomass of all the groups as the range goes from just above 6 gramm to 5 gramm.
Based on the statistical test one-way Anova conducted, the significance of phosphate in biomass differences in the plants is statistically non-significant for our data apart between the group with medium Pi concentration and the groups with no Pi in soil. There is a significance of * as the p- value lies just under 0.05.
To conclude the analysis of the biomass we observe the difference between plants grown with yeast and without.
Figure 22: The comparison of biomass is conducted between plants with and without yeast based on their phosphate concentration. The color coordination, as seen in the legend, for plants grown without yeast is: high phosphate concentrations [1 mM] shown in dark green, medium Pi concentrations [0,5 mM] shown in green, low Pi concentrations [0,1 mM] shown in light green and in soil with no phosphate [0 mM] shown in a whitish green and for plants grown with yeast is: high phosphate concentrations [1 mM], the yellow medium Pi concentration [0,5 mM], the light yellow low Pi concentration [0,1 mM] and the cream color stands for no Pi in soil [0 mM]. The concentrations of phosphate are on the x-axis and on the y-axis is the mass in gramm.
There is no statistical significance in the role of phosphate in the data apart from three pairs. Medium Pi concentration without yeast and medium Pi concentration with yeast is statistically significant [*] concerning the change in biomass dependent on phosphate. One can observe that plants grown with yeast under the same phosphate concentration [0,5 mM] have greater mass than those grown in plain soil. The next pair where the difference in biomass is statistically significant, is the one also mentioned above, medium Pi concentration with yeast and no Pi in soil with yeast meaning that a higher phosphate concentration accounts for greater biomass. The last pair with statistically significant difference [*] is medium Pi concentration with yeast and no Pi in soil without yeast, showing that the range of biomass in plant growth with medium Pi concentration in soil and yeast lies around 7,8 and 6 gramm while the biomass range of plants grown without any Pi or yeast lies around 6,8 and 3,8 gramm. From that it seems that plants with yeast as well as a 0,5 mM Pi concentration have greater mass than those grown without yeast and without any Pi. Based on literature 4 the biomass of plants grown under Pi stress is lower than those grown in high Pi conditions.
Phosphate Leaf Correlation
The leaf number of all plants in series one was documented and analyzed based on the phosphate concentration and yeast content.
Figure 23: The effect of phosphate concentration on leaf number is shown.The test group was grown in soil without any yeast, in high phosphate concentrations [1 mM] shown in dark green, medium Pi concentrations [0,5 mM] shown in green, low Pi concentrations [0,1 mM] shown in light green and in soil with no phosphate [0 mM] shown in a whitish green. On the x-axis is the phosphate concentration in M (Molar) and on the y-axis the number of leaves.
The range in number of leaves in high Pi concentration lies around 7 and 6 with two outliers at 11 and 4. For medium Phosphate concentration the range becomes even broader from 7 to 5 leaves, however for low Pi concentration the leaf number goes as high as 12 . At last plants grown with no phosphate in soil have a range of 9 to 5 leaves. It is important to mention that based on the One-Way Anova test conducted with the data there is no statistical significance in the difference of leaf count based on the phosphate concentration.
Next we documented the leaves in plants grown with yeast and different Pi concentrations.
Figure 24: The effect of phosphate concentrations on leaf formation in plants grown in soil with S. cerevisiae. The deep yellow shows data on high phosphate concentrations [1 mM], the yellow medium Pi concentration [0,5 mM], the light yellow low Pi concentration [0,1 mM] and the cream color stands for no Pi in soil [0 mM]. On the x-axis are the four phosphate concentrations in M (Molar) and on the y-axis the number of leaves.
Plants grown with yeast and a high Pi concentration have a leaf range of 7 to 3,8 leaves, those grown with medium Pi concentration 6,2 to 4 leaves, those in low Pi concentration have a range of 7 to 3,9 leaves and those grown without any Pi have 7 to 5 leaves. Based on the analysis of the significance of the data using One-Way Anova there is no statistical significance of phosphate concentration in the number of leaves in plants.
Last but not least we compare the difference in leaf number between the groups of plants grown with and without yeast.
Figure 25: The comparison of leaf number implants grown with and without S. cerevisiae is shown. The color coordination for plants grown without yeast is: high phosphate concentrations [1 mM] shown in dark green, medium Pi concentrations [0,5 mM] shown in green, low Pi concentrations [0,1 mM] shown in light green and in soil with no phosphate [0 mM] shown in a whitish green and for plants grown with yeast: high phosphate concentrations [1 mM], the yellow medium Pi concentration [0,5 mM], the light yellow low Pi concentration [0,1 mM] and the cream color stands for no Pi in soil [0 mM]. On the x-axis are the 4 phosphate concentrations in M (Molar) and on the y-axis the number of leaves.
There is no statistical significance on the difference in leaf number based on the One-Way Anova test. One thing to observe however is that plants grown without yeast seem to have a greater leaf number than those grown with yeast.
Malachite Green Assay
Figure 26: Malachite Green Assay 1
During our final evaluation of Growth Series 1, we also conducted a Malachite Green Assay to quantify the amounts of phosphate left inside the pots. 6
Firstly, we washed the nutrients out of the soil using 75 ml of distilled water for each pot. We then had 32 samples of solved nutrients which we placed inside the wells B1 to E8 in the following order:
Table 4: Run-off Water From Planting Pots
The cells that are marked in light yellow are pots in which only one of the seeds grew, the orange cell is a pot in which no plant sprouted.
Inside the wells A1 to A8 we placed the comparison samples recommended by the manufacturer of the assay.
Table 5: Standard Phosphate Concentrations
Wells A9 to A12 are our personal comparison samples, containing the concentrations of phosphate that we put into our Hoagland solution with which we watered the plants.
Table 6: Phosphate Concentrations From Hoagland Solution
After reading the results with a plate reader, we got the following table of Optical Densities, measured at a wavelength of 620 nm.
Table 7: O.D.₆₂₀ Malachite Green Assay
From the phosphate standards inside wells A1 to A8, we created a phosphate standard curve, the formula of which we then used to create the correlating concentrations of phosphate to the O.D.s that we measured in our samples.
Figure 27: Malachite Green Assay Phosphate Standard Curve
The results of calculating all of the phosphate concentrations with help of the standard curve are listed in the table below in the unit of µM. Positive results are marked in light green, concentrations over 40µM are marked in a deeper green.
Table 8: Phosphate Concentrations Inside the Wells in µM, According to the Standard Curve
The large sum of negative results might be due to an unclear standard curve, as the ideal graph through the data starts only at 0,2887 O.D. However, it can be seen in the diagram that the real O.D. measurements of phosphate concentration standards around zero to four µM are actually lower than that.
In addition, it is noteworthy that the calculated concentrations of our Hoagland solution for comparison, inside the wells A9 to A12, are also much lower than expected, this may be due to the recommended range of usage for the Malachite Green Assay lying between zero and 40 µM. Overall, as expected we can observe that the Phosphate concentrations in run-off water from pots that were watered with high phosphate Hoagland solutions are much higher than those who were watered with Hoagland Solutions containing no or low concentrations of phosphate.
There seems to be a tendency for more phosphate to be detected in the run-off water in the pots that also included yeast. However, due to the small amount and accuracy of our data, this is also possible to be a coincidence.
We did another Malachite Green Assay later on, also including several dilutions of our samples to get more exact results within the assay’s intended range. Due to experimenting errors and a lack of time to repeat the experiment, this data is however unusable for analysis.
Planting Series 2
At the end of September, we started a second planting series, including the same groups as the previous one and another group that contained no yeast. We were hoping to add our modified yeast to these pots, to properly compare the growth and if our construct would stand the test as a biological fertilizer. Since our Funglilyzer didn’t get finished as we had hoped, this step could not be completed. In the remaining time, the zucchini plants of series two also didn’t grow enough to gather comparable data. Most plants didn’t sprout at all and the ones that did stayed, with one exception, under 2cm.
Therefore, we decided to abandon this testing series and focus on thoroughly analyzing our existing data, though with more time we would have continued this test to further improve our method of proving the functionality of Fungilyzer and establish an engineering cycle.
Future Steps
If the project Fungilyzer was to be continued, we would run further tests with larger sample sizes and would also include a group of plants grown together with the modified yeast to gauge the effects of Fungilyzer on crops.
It would also be beneficial to use several kinds of crops as testing organisms and to include pots in which the concentration of phosphate in the soil would vary throughout the test, for example, a testing group that would be “fertilized” with large amounts of phosphate in the beginning but wouldn’t receive any further phosphate throughout the test. These kinds of groups would be helpful in proving the practicality of Fungilyzer in the situation that it was meant for: as a co-fertilizer to support the quick uptake of conventional fertilizers and maximize their efficiency. It would also be helpful using different watering methods in order to simulate the real life situations like strong rain storms during the rain seasons. This would help to see in such an occasion how much phosphate will be washed out from the soil and how well we can prevent it with our method. In addition to those steps it would also be very helpful to build a completely sealed farming room in the institute to improve the realism of the experiment and get more reliable results.
In testing series two, we were planning to make several improvements to our planting and measuring methods.
We changed the method of measuring the plant height to be more unified and were planning to collect more data throughout, like a constant leaf count and measuring the leaves surface. A microbiological analysis near the end would also be helpful, in order to see if our nonmodified yeast, and eventually our modified yeast would even survive to have a positive effect on the crop plants. Another improvement would be changing the method of measuring the phosphate level in the water. We claimed the water under our pots but it would be more accurate when it is done with proper equipment for gardening. A better place for plants to grow would be beneficial. Our plants share the same laboratory space with other ongoing project steps, it would be better for them to have a sealed place in order to grow properly. More time for growth would also be helpful to see the effect of Phosphate levels on every cycle of the plant's life.
References
Full Plant data sadly we were not able to upload the excel sheets so here are they in PDF format:
- van Delden SH, Nazarideljou MJ, Marcelis LFM. Nutrient solutions for Arabidopsis thaliana: a study on nutrient solution composition in hydroponics systems. Plant Methods. 2020 May 18;16:72. doi: 10.1186/s13007-020-00606-4. PMID: 32612669; PMCID: PMC7324969.↩
- H. B. Mann. D. R. Whitney. "On a Test of Whether one of Two Random Variables is Stochastically Larger than the Other." Ann. Math. Statist. 18 (1) 50 - 60, March, 1947. https://doi.org/10.1214/aoms/1177730491↩
- Mukherjee, A., Verma, J.P., Gaurav, A.K. et al. Yeast a potential bio-agent: future for plant growth and postharvest disease management for sustainable agriculture. Appl Microbiol Biotechnol 104, 1497–1510 (2020). https://doi.org/10.1007/s00253-019-10321-3↩
- Lyudmila V. Ignatova, Yelena V. Brazhnikova, Ramza Z. Berzhanova, Togzhan D. Mukasheva,Plant growth-promoting and antifungal activity of yeasts from dark chestnut soil, Microbiological Research, Volume 175, 2015, Pages 78-83, ISSN 0944-5013, https://doi.org/10.1016/j.micres.2015.03.008.↩
- Bechtaoui N, Rabiu MK, Raklami A, Oufdou K, Hafidi M, Jemo M. Phosphate-Dependent Regulation of Growth and Stresses Management in Plants. Front Plant Sci. 2021 Oct 28;12:679916. doi: 10.3389/fpls.2021.679916. PMID: 34777404; PMCID: PMC8581177.↩
- Malachite Green Assay: https://b2b.sigmaaldrich.com/deepweb/assets/sigmaaldrich/product/documents/171/212/mak307bul.pdf↩