In order to assemble both our Control and Treatment Plasmid, we used Gibson assembly as described in experiments.

Assembly of Treatment Plasmid:
After the Gibson assembly of our Treatment Plasmid we performed colony PCR, to ensure the selection of our plasmid within the different E. coli colonies. As shown in Figure 1, our insertion piece was 3627 bp long.

Figure 1: PCR product from transformation of E. coli DH5α with our Treatment Plasmid. The outer wells of the gel contains ladders. Well 2 contains a negative control and the following 8 wells contain different colonies from the E. coli DH5αwith the insertion piece of the plasmid amplified by PCR. As our insertion piece was about 3627 bp long, we chose to purify the plasmid of well 3, 4 and 8, marked with the respective numbers, as these were therefore the correct length.

Based on these results of the PCR reaction we decided to continue with the assembled plasmids from well 3, 4 and 8, as these contained the entire insertion piece. We changed the names of the plasmids to T1, T2 and T3. Upon purification, the plasmids were sent for Sanger sequencing analysis. Analyzing the sequencing results, we discovered that none of the plasmids had mutations, thereby allowing us to continue testing with the chosen plasmids.

Assembly of Control Plasmid:
The control plasmids were also assembled using Gibson assembly, where our insertion piece is 3721 bp long. We used the ladder band that was 3000 bp, as seen in Figure 2 and 3, we used as a guideline to choose the colonies. We chose to continue working with the plasmids that have the same length of our insertion piece, which is 3721 bp long.

Figure 2: PCR product from transformation of E. coli DH5α with our Control Plasmid. The Figure shows ladders in both ends of the gel, and a negative control in the second well. The next 12 wells show different insertions from the E. coli DH5α amplified plasmid. As our insertion piece was about 3721 bp long we chose to use the colonies C, F and L, marked with the respective letters, as these had a similar length.

Figure 3: PCR product from transformation of E. coli DH5α with our control plasmid. The four wells in the middle show different insertions from the E. coli DH5αamplified plasmid. As our insertion piece is about 4000 bp long, we chose to use the colony M, which is marked with the letter, as this has the same approximate length.

Based on the results of the PCR reaction we sent the plasmids marked as C, F, L and M for Sanger sequencing analysis. We changed their names to C1, C2, C3 and C4 and after analyzing the sequencing results, we concluded that C3 and C4 have mutations and could not be used for further testing. C1 and C2 have no mutations and can therefore be used for further testing.

We performed electroporation on our 12Z cell lines with our Control Plasmid, as described in experiments. The control plasmid is supposed to emit green fluorescence if there is a high concentration of ERα in the cells and emit red fluorescence if there is a high concentration of ERβ.

Figure 4: Results from the elctroporation was analyzed via flow cytometry. On the x-axis there is green fluorescent intensity and on the y-axis, there is red fluorescent intesity. The green dots represent the negative control, and the blue dots represent the 12Z cells that have gone through the electroporation with the control plasmid

The graph presented in Figure 4 compares 12Z cells that have been transfected with the Control Plasmid to a negative control of non-transfected 12Z cells, with measurements based on green and red fluorescence emitted by the respective cells. As mentioned, the blue dots represent the negative control sample, meaning that the 12Z cells were not subjected to electroporation, while the green dots represent the 12Z cells that were subjected to electroporation and were introduced to the plasmid. We expected to see that the blue dots would be higher on the y-axis than the green dots, because the red fluorescence should be activated by the high concentration of ERβ in the 12Z cells. However, as seen in Figure 4 both blue and green dots are concentrated in the same location. We later discovered mutations in the BRCA1 promoter of the control plasmid, so we are unsure as to if the promoter was functioning properly. If not, this could have caused the lack of fluorescence.

Electroporation optimization of 12Z and HESC cells with pmaxGFP Vector

After the previous failed attempts, we decided to optimize the electroporation process for 12Z cells with the pmaxGFP Vector, a standard GFP plasmid. We conducted an experiment with 48 different reactions with 3 Nucleofector Solutions (SE, SF, and SG) alongside 15 different electroporation programs, plus a control. Our goal was to find the best combination that would maximize cell transfection efficiency while minimizing cell death. Once we identified the optimal conditions, we aimed to apply them to all future experiments, to transfect our plasmids.

Figure 5: Results from electroporation on 12Z cells. On the y-axis is the percentage of GFP+ cells. On the x-axis are the different programs, that were run on the samples. The blue pillars show the live cells. The PI- stand for propidium iodide negative. Propidium iodide is used to dye dead cells. The Orange pillars show the GFP+ cells. (A) The electroporation is done with a SE solution. There was an error due to too low an event rate in the control sample. (B) The electroporation is done with a SF solution. There was an error due to too low an event rate in the sample that got run with the CM-150 program. (C) The electroporation is done with a SG solution.

Based on the results from the electroporation experiments, we observed distinct transfection efficiencies across the different Nucleofector solutions and electroporation programs. For the SE solution (Figure 5A), transfection efficiency varied significantly between programs, but a noticeable issue arose with the control sample, which had too low an event rate, rendering its results inconclusive. With the SF solution (Figure 5B), the transfection profile followed a similar pattern, but the sample treated with the CM-150 program also experienced an error caused by cells absence, affecting the reliability of that specific condition. However, despite these errors, we identified notable trends, including an increased percentage of GFP+ cells in certain programs. Finally, in the SG solution (Figure 5C), we achieved relatively consistent results across programs, though with varying levels of transfection efficiency. Overall, while no singular solution-program combination produced optimal results across all metrics, the data enabled us to narrow down the most effective conditions for maximizing GFP expression while minimizing cell death. These findings guided the selection of the optimal conditions for subsequent plasmid transfections in 12Z cells.Due to the low cell number in the performed experiment we decided to repeat the electroporation optimization experiment as a general trend was observed favoring SF solution in the subsequent experiment. Due to the low cell number in the performed experiment we decided to repeat the electroporation optimization experiment as a general trend was observed favoring SF solution in the subsequent experiment.

Figure 6: Results from electroporation on 12Z cells with SF solution. On the y-axis is the percentage of GFP+ cells. On the x-axis is the different programs, that were run on the different samples. The blue pillars show the live cells. The PI- stand for propidium iodide negative. Propidium iodide was used to dye dead cells. The Orange pillars show the GFP+ cells. The sample that was run with the CA-137 did not receive any electric impulses. The sample that was run with the DS-130 program had too few cells.

In the optimized electroporation experiment using the SF solution (Figure 6), we observed improvements in transfection efficiency and cell viability across the tested programs. Most notably, programs that applied electric impulses generally showed a higher percentage of GFP+ cells, indicating successful transfection. The CA-137 and control program, which did not receive any electric impulses, predictably showed no GFP+ cells, serving as negative controls for the experiment. Meanwhile, the DS-130 program had an insufficient number of cells, leading to unreliable results for that condition. Despite this, the majority of the remaining programs demonstrated a balance between GFP expression and cell viability, with the blue pillars indicating a healthy population of live cells. These results confirmed that the optimized conditions for electroporation with the SF solution significantly improved both transfection efficiency and cell survival, allowing us to select the most favorable parameters for future experiments.

It is important to note that the percentage of live cells, as indicated by the blue pillars, represents only about 5% of the total number of cells that were initially added to each well. This means that a big portion of cells were lost during the electroporation process, possibly due to stress or suboptimal conditions. Despite this, most of the remaining programs demonstrated a balance between GFP expression and cell viability, highlighting their potential for effective transfection. These results confirmed that the optimized conditions for electroporation with the SF solution significantly improved both transfection efficiency and cell survival, allowing us to select the most favorable parameters for future experiments.

We did multiple transfection attempts, as described in experiments, but unfortunately none of the attempts were successful. In our very first transfection attempt we used a 6-well plate, where two wells contained 12Z cells transfected with the Control Plasmid and one well with HESC cells transfected with the control plasmid. Besides this, we also had a negative control sample of the 12Z cells. The Control Plasmid is designed to emit green fluorescence if there is a high concentration of ERα in the cells and emit red fluorescence if there is a high concentration of ERβ. Generally, there is a higher concentration of ERα in endometrium cells, while there is a higher concentration of ERβ in the endometrial-like cells. Therefore, we would expect similar concentration patterns in our HESC and 12Z cell lines. In a later attempt we used a standard GFP plasmid to determine whether the issue was with our designed plasmid or with the transfection process itself, since the transfections kept being unsuccessful.

Figure 7: Results from transfection both with our plasmid and a standard GFP plasmid. The y-axes show the red fluorescent intensity, and the x-axes show the green fluorescent intensity. (A) Show the density of cells from the transfection with our control plasmid in 12Z cells. Blue dots show the lowest density of cells and red dots show the highest density of cells. The black box indicates where the cells that emit red fluorescent should be seen, which in this case is 0.05% of the sample. (B) Show the density of cells from the transfection with our control plasmid in HESC cells. Only 1.38% of the cells that are in the box. (C) Show all cells from the transfection with the standard GFP plasmid. The black box shows where cells that emit green fluorescent should be seen. (D) Shows the isolated cells that were expressing green fluorescence. In both (C) and (D) the dark blue and green dots represent 12Z cells that have gone through transfection with the GFP plasmid. The red dots represent 12Z cells that went through transfection without the plasmid. The purple dots represent the negative control for 12Z cells, so they have not gone through transfection. The yellow dots represent HESC cells that went through transfection with the plasmid, and the light blue dots represent HESC cells that went through transfection without the plasmid.

×

The graphs, seen in Figure 7, visualizes the results from two different transfection experiments. In Figure 7A the density of 12Z cells that underwent transfection with our control plasmid are shown. We expected to observe a large number of cells within the black box, which indicates the area for cells that expressed red fluorescent. Unfortunately, only 0.05% of cells are located within the black box, indicating an unsuccessful transfection. In Figure 7B the density of HESC cells that underwent transfection with our control plasmid is shown. We expected to observe no cells in the black box, and a large number of cells positioned lower on the y-axis and further to the right on the x-axis, as these cells were expected to emit green fluorescence. This expectation is based on the function of the control plasmid, as the BRCA1 promotor in the control plasmid responds to the higher concentration of ERα, which promotes the expression of GFP - thereby emission of green fluorescence would be expected. Unfortunately, this is not the case as only 1.38% of the cells are located in the black box.
In Figure 7C and D the results from the transfection with the standard GFP plasmid are shown. In Figure 7C all the cells transfected with the standard GFP plasmid are shown. Here we expected to see a large number of dark blue, green and yellow dots within the black box, as these cells were transfected with the standard GFP plasmid. However, this was not the case, as there are more cells outside the box than within the area. In Figure 7D the cells within the black box, seen in Figure 7C, are isolated. As mentioned earlier, we expected to see a lot of dark blue, green and yellow dots in this area. Unfortunately, there are lots of every color in Figure 7D, indicating that cells from almost all of the samples are found there. However, there is mostly dark blue dots, which represent the 12Z cells transfected with the standard GFP plasmid, but there are too few to confirm whether the plasmid was successfully transfected into the cells.

One of the challenges that we faced during our project was that our HESC cell line was growing very slowly. The prolonged growth period caused an issue as it delayed our experimental timeline, due to the fact that the cells needed to be confluent enough (70-90%) to actually be used for an experiment. Therefore, we decided to try and optimize the HESC cells conditions, and thereby their growth, through various media combinations.

The average time between splits in our original HESC cell lines was 11.3 days, before we began our growth optimization experiment. Our cell lines had a doubling time of approximately 8.87 days, which was considerably longer than the typical scenario, where cells are split every 3 days upon reaching a confluency of about 70-90%. Prior to our optimization experiment, the cells were cultured in T75 flasks, but as shown in Figure 8A, transferring the cell lines to a T25 results in a slight change in the growth rate compared to cells grown in the standard T75 flasks, as shown in Figure 8B.

Figure 8: (A) shows the growth rate of original HESC cells in T75 flasks. (B) Shows the growth rate of original HESC cells in T25 flasks.

In the experiment we used several different media combinations, as shown in the table below.

Table 1: This table gives an overview of the different media combinations used in the experiment. Each flask was assigned a code (e.g. “O2”), differentiating the type of media that they contain.

Replacing 10% charcoal-stripped FBS with 10% Human AB Serum results in an improved growth rate of the cells in O1 and O5. The doubling time of the HESC cells growing in O1 media, when compared to the original HESC cell line in a T75 flask, is decreased by 6,74 days in comparison to the original doubling time of 8.87 days, which was initially positive for our experiment. However, in the following days it was observed that the O1 media appeared cloudy, and it was therefore checked for contamination. Although the test did not confirm any contamination, the cells growing in the O1 media looked anomalous and were therefore discarded. The media was no longer used throughout the experiment.

In O2 media, the concentration of Pen-strep was reduced to 0.5% from the recommended 1%. This change in the concentration of Pen-strep is also seen to have a positive effect on the growth rate of the O2 cells, reducing the doubling time by 6.32 days.

The O5 media consists mainly of EmbryoMax and 10% Human AB Serum instead of DMEM-F12 and 10% charcoal-stripped FBS. The media combination shows good improvement in growth rate, decreasing the doubling time by 6.63 days,compared to the original 8.87 days. The O3 media results in a decrease of 6.74 days in doubling time, compared to the original 8.87 days. The single change in the O3 media, is the use of normal 10% FBS instead of 10% charcoal-stripped FBS.

The media combination O4 shows the best results out of all. We switched out the recommended DMEM-F12 with Embryomax, which is the media recommended for our 12Z cell line. This shows a very noticeable increase in cell growth, thereby indicating that Embryomax could have a positive effect on the growth of our HESC cells. The doubling time is decreased by 6.4 days in comparison to the original 8.87 days.

Figure 9: (A) shows the growth rate of O4 HESC cells, (B) shows the growth rate of O3 HESC cells and (C) shows the O1 HESC cells.

In all, our HESC growth optimization experiment results in the O4 media being the most successful media combination, resulting in a decrease in doubling time by 7.09 days as opposed to the original 8,87 days. O1 and O3 are closely following it, each decreasing the cells doubling time by 6.71 days, as compared to the original 8.87 days. To learn more about our HESC optimization experiment, take a look under experiments.

To test if our antibodies are specific for only the ERα and Erβ, we used western blot as described in experiments. The Nanodrop in our lab was broken so we could not measure the protein concentration of our lysed cells. Therefore, we decided to make five different samples for each cell line. The cell lines were lysed and diluted with PBS. When loading the gel, we used a protein ladder, a 12Z sample, and a HESC sample. Afterwards we added our antibodies one after the other, so both antibodies were distributed all over the membrane. The fluorescence was measured at the same time.

Figure 10: Western blot results from antibodies that bind to ERα or ERβ. (A) Results for Estrogen Receptor alpha, the fluorescent antibody that binds to ERα. The red squares show the bands from the antibody. (B) results for ESR2 Monoclonal Antibody, the fluorescent antibody that binds to ERβ. The red squares show the bands form the antibody. The yellow square marks two ladder bands. The upper band is ~70 kDa and the lower band is ~55 kDa.

The red squares, seen in Figure 10A, are indicating where the bands of the ERα are located. ERα is 67 kDa (1). Therefore, we expected to see the ERα band near the ladder band with the length of ~70 kDa. This was also the case as shown in Figure 10A, where the ERα band is located next to the ladder band. This could confirm that the antibody binds specifically, and thereby only to ERα. In Figure 10B the red squared is indicating where the bands of ERβ are located. ERβ is 59 kDa (1), and we therefore expected that the ERβ band would appear between the two ladder bands marked in the yellow square in Figure 10B, as the lower band has a length of ~55 kDa and the upper band has a length of ~70 kDa. However, it appears that there was an error with the staining process, as the ERβ band is next to the ladder band with the size of ~70 kDa and not in between the two ladder bands. Thereby we cannot 100% confirm the specificity of this antibody.

In this experiment, 12Z and HESC are stained for ERα and ERβ with the use of ESR2 Monoclonal Antibody and Estrogen Receptor alpha Antibody coupled with fluorescence. The goal was to titrate the antibodies used for staining, in order to optimize detection of ERα and Erβ in these specific cell lines. ERα staining was performed as an extracellular stain without permeabilizing the cells, while ERβ staining was performed intracellularly after permeabilization. The stained cells were analyzed using flow cytometry to assess the percentage of receptor-positive cells, and to evaluate the antibody titrations.

Figure 11: Results from antibody staining for both 12Z- and HESC cells. (A) Results from antibody staining on 12Z cells with the antibody that binds to ERα. The y-axis is ERα antibody fluorescence, and the x-axis shows the antibody dilution, highest (left) to lowest (right). Blue dots show the individual population and red dots show the highly dense population. (B) A graph visualizing the percentage of cells with positive ERα signal after staining with the different antibody concentrations. The y-axis shows the percentage of ERα+ cells present, and the x-axis shows the concentration of antibody at different dilutions. (C) Results from antibody staining on HESC cells showing the amount of ERα signaling in each dilution range. (D) Results from antibody staining on 12Z cells, showing single cells with ERβ signals. (E) A graph visualizing the affect antibody concentration had on the amount of ERβ+ cells. The y-axis shows the percentage of ERβ + cells present, and the x-axis shows the concentration of antibody at different dilutions. (F) Results of the ERβ stain on HESC cells, showing single cells with ERβ signals.

12Z Cells - ERα Staining (Figure 11A)

For ERα staining in 12Z cells, the antibody was titrated at different concentrations between 1:12.5 and 1:200, doubling at each dilution. 5 wells were stained with ERα antibody extracellularly while 1 additional well remained unstained on the surface. The flow cytometry plot, seen in Figure 12A, was obtained by creating live cells channel by using zombie violet dye and the ERα antibody channel to detect the cells with positive cells with ERα antibody.
The results show a gradient from high concentration / low dilution (left) toward low concentration / high dilution (right). For most samples with antibody staining, the majority of the cell population show low fluorescent signal intensity; therefore, only a small fraction was considered as ERα positive. Only with the highest antibody concentration (least diluted), around 85% of cells were Era positive, but still with relatively low signal.

12Z Cells - ERβ Staining (Figure 11D)

For ERβ intracellular staining in 12Z cells, the antibody is titrated at different concentrations between 1:25 and 1:400 into 5 wells and 2 further wells are used as intracellular unstained cells. The flow cytometry data, seen in Figure 11D, show the majority of the cells exhibiting strong fluorescence signal even at highest dilutions (lowest antibody concentrations), resulting in more than 90% of cells being ERβ positive. These findings support the upregulation of ERβ in endometrial-like cells stated in literature.

HESC Cells - ERα Staining (Figure 11E)

In HESC cells, ERα staining was titrated between 1:12.5 and 1:25 in 2 wells and 1 well for surface unstained control. As shown in Figure 11C, the fluorescence signal for ERα is weak and sporadic, with only a small population of cells showing positive staining. The low amount HESC samples tested restraints the accuracy of the titration curve for these dilutions and limits the descending gradient as well. More samples should be tested to refine the graph and ensure the optimal dilution is accurately identified.

HESC Cells - ERβ Staining (Figure 11F)

For HESC cells, ERβ staining is titrated between 1:50 and 1:100 in 2 wells and 1 additional well is used for intracellular unstained control. Similar to the ERα staining, the small number of samples limits the precision of this titration. More HESC samples across a wider range of dilutions would provide a more accurate assessment of the optimal ERβ antibody concentration, and higher dilutions should be tested to avoid the saturation seen in the current data.

Conclusion

Based on the performed antibody titration, it was found that the optimal dilution of the ERα antibody on 12Z cells was 1:50, as shown in Figures 11A and 11B. For the ERβ antibody titration on the 12Z cells, no optimal dilution factor could be identified, as all investigated dilutions resulted in saturated staining, which can be seen in Figures 11C and 11D. For the antibody titration on the HESC cells, no successful results were obtained due to insufficient cell numbers. However, it was observed that the ERα and ERβ antibodies successfully bound to their targets on the HESC cells, as both positive and negative populations were detected, as shown in Figures 11E and 11F.

We have based our whole project around the fact that research shows, that the cells involved in endometriosis express up to 100x more ERβ than ERα (2), so we wanted to see if this was true for the two cell lines which we have used throughout this experiment: 12Z and HESC. We did this by performing an antibody staining with two antibodies: ESR2 Monoclonal Antibody and Estrogen Receptor alpha Antibody. The ESR2 Monoclonal Antibody is conjugated with CoraLite Plus 488, which emits a green-fluorescent light. The Estrogen Receptor alpha Antibody, on the other hand, is conjugated with PerCP, which emits a red-fluorescent light.

Figure 12: This figure shows the results of staining 12Z cells (A-C) and HESC cells (D-F) with both ESR2 Monoclonal Antibody (green) and Estrogen Receptor alpha Antibody (red). Additionally, the cells were stained with DAPI to visualize the nucleus.

In Figure 12 it can be seen that the 12Z cells (picture A-C) show a slightly clearer expression of ERβ, than in the HESC cells (picture D-F), as ER-β can be seen as a kind of green hue around the cells. On the other hand, it looks like the ERα, in the 12Z cells, has a tendency to lay more around the nuclei of the cells, which can be seen as the slightly red hue “covering” the nuclei. In the HESC cells, it looks like there is way less ERβ present in the cytoplasm of the cells, as not much green fluorescent can be seen in pictures D-F. The ERα is only slightly present in the HESC cells, not showing a very strong signal. These findings roughly align with what we expected to find, which is that the endometrial-like cells (here: 12Z) express noticeably more ERβ than endometrium cells (here: HESC), while the endometrium cells should express more ERα than the endometrial-like cells.

From this experiment alone, it is not possible to conclude the precise localizations of the two receptors, or if there is more of one receptor type present than the other in the cells. If our experiment with transfecting our two cell lines with the Control Plasmid had been successful, those results could have been used to confirm and/or disconfirm our theory on which receptor is expressed more in which cell line. The western blot performed on the antibodies (see Figure 10), confirm that the Estrogen Receptor alpha Antibody is specific for the ERα receptor only. This tells us that the signal of red fluorescent is only coming from ERα receptors in both cell lines. The western blot performed on the ESR2 Monoclonal Antibody, did not confirm its specificity (see Figure 10), so that makes it harder for us to trust that all the green fluorescence in pictures A-F comes only from the ERβ receptors.

Figure 13: The excitation and emission spectrums of the conjugates which the two antibodies contain, respectively. The excitation spectrum is only outlined, while the emission spectrum is shown colored in.

As it can be seen in figure 13 the distance between the two antibodies emission spectrums is not that far, which could cause some trouble. While the emission spectrums are shown as the colored-in graphs in figure 13, it is important to note that the filter of the microscope starts to emit 100 nm before and after the spectrum. Due to this, there is going to be a slight overlap between the two antibodies, which could possibly be the reason why the red fluorescence signal is not that clear. The samples were checked for “bleed-through” - a fantom that occurs when multiple lasers are used at once, causing them to leak into the other channels - and this was not the case. Another reason for the weak red fluorescent signal could be, that the antibodies which we have chosen to work with, are just not compatible with the confocal microscopy used to analyze the samples.