Discussion

Based on results to experiment #1 (cited from first semester technical report):

We were unable to analyze the results from the Ki67 proliferation stain. Currently, we are looking into the CellProfiler settings to attempt to utilize a different set of cell-identifying techniques. If CellProfiler is unable to consistently and accurately identify the number of cells, we plan to use a different proliferation stain such as EdU. With the proliferation results, we hope to have a better understanding of the capabilities of F1R1 in inducing regenerative properties for cardiac cells. In particular, the proliferation data for the F1R1-only group could reveal useful information about the protein’s mechanism of eliciting cardiomyocyte proliferation. If the number of Ki67-positive cells is significantly greater in the F1R1 condition when compared to the control, F1R1 could have more effector functions beyond just inhibiting TGF-B and may act on fibroblasts directly. 

The statistically significant changes seen in aSMA percentage of cells agree with our hypothesis. The percentage of cells differentiating decreased with the addition of F1R1 or both F1r1 and TGF-B and increased with the addition of TGF-B (Figure 1). TGF-B exposure was expected to increase differentiation as it stimulates a known cardiac fibroblast pathway that results in activation and differentiation into myofibroblasts⁷. Testing the influence of F1R1 on fibroblast differentiation contributes to the novelty of this study. There was a significant decrease in differentiation observed in the F1R1 group, suggesting that there is some sort of anti-differentiation mechanism invoked by F1R1.  Furthermore, the decreased differentiation from the F1R1/TGF-B group to a similar level to the control suggests that F1R1 impedes the ability of  TGF-B to activate fibroblasts. This could happen in multiple different ways, whether it be through direct inhibitory binding of F1R1 to TGF-B or by F1R1 mediating an inhibitory interaction between TGF-B and another cellular factor. Further experiments will focus on determining in what manner F1R1 interacts with TGF-B to curb fibroblast activation. Nonetheless, the decrease in fibroblast differentiation exhibited between the TGF-B and TGF-B/F1R1 groups indicates that F1R1 could effectively reduce scar tissue formation in vivo by acting on TGF-B. 

Initial analysis of the control group from aSMA data indicates a high level of basal differentiation at 71.7% (Figure 1). This is substantially higher than some previous studies have shown for cardiac fibroblast differentiation. For example, Li et al. found a basal level of differentiation to be around 20%¹⁶ for neonatal rat cardiac fibroblasts. There are a few possible reasons for this discrepancy. Firstly, Li’s study cultured the rat cardiac cells using fetal bovine serum, meanwhile, the cells in this first experiment were cultured in serum-free media. The media used can have a significant effect on the rate of differentiation for a cell culture¹⁷. Thus, the serum starvation stage performed during our culturing process could have affected the basal level of differentiation. Further factors, such as handling and preparation of the cell cultures, can also result in a change in the differentiation rate. The methods used for handling cell cultures were very similar between our experiment and what has been seen in previous studies. However, cell cultures are sensitive to relatively small changes in the environment and/or physical stimuli¹⁸. It is possible that cardiac cells could have been driven to differentiate from accidental stimuli during the preparation process. Another factor that can potentially cause differentiation is intercellular interactions. The proximity of cells to one another can increase or decrease the basal level of differentiation¹⁷. Since this is not a factor that we controlled for, this could have increased the rate of differentiation in our culture. 

Based on results to further experiments (cited from final technical report):

In order to address the potential effect of serum starvation on cardiac fibroblast differentiation, we reran the same experiment but with serum-containing media. Although the magnitude was lesser, there was still a decrease in aSMA-positive cells from the TGFb to the TGFb/F1R1 treatment groups. Despite this, post hoc testing did not find statistical significance when comparing the extent of activation in these groups. Thus, it is yet to be confirmed if F1R1’s dampening of the TGFb-induced cardiac fibroblast differentiation is reproducible in serum-containing media. A more broad takeaway is that there was still a high level of basal differentiation of ~72% in the control even with the media change (Figure 2). These findings suggest that serum starvation did not influence the activation of the fibroblasts in culture and that other factors may be at play. In itself, the introduction of serum comes with some limitations. Serum contains hormones, growth factors, and many other undefined substances that may influence the differentiation or proliferation of cardiac cells¹⁹. There is not expected to be any interference with serum media and the treatments we used. Because of these limitations, serum-free media is the standard in the field once serum starvation is found to not affect the data²⁰. Moving forward, serum-free media would be preferable for these fibroblast experiments to reduce differentiation and proliferation not induced by TGFb or F1R1 treatment. This conclusion is further supported by our attempted F1R1 dose dependency experiment with serum, which saw wells become overly confluent despite seeding density being appropriate (Appendix 5). 

As previously mentioned, cells in close proximity can also influence each other’s phenotype through intercellular signaling. Furthermore, the distribution of cells in each well was irregular in this experimental iteration, with growth only occurring on the perimeter of each well and the center remaining unpopulated. With this in mind, the denser areas of cells at points around the edge could have bolstered the signaling cues exchanged between fibroblasts, resulting in more ensemble differentiation. Taking the unexpectedly high differentiation in the F1R1-only condition into account, aggressive treatment could have put excessive stress on the cells, pushing them to differentiate. Another factor at play here could be the reaction of the cardiac fibroblasts to tissue culture plastic (TCP). TCP can have a stiffness of up to 3 GPa while a healthy cardiac environment is closer to 7 kPa, which could be triggering the mechanosensing pathways of fibroblasts to induce the myofibroblast transition²¹. These results and subsequent considerations show that our colleague Benjamin Hoang’s pursuit of designing silk hydrogels for optimal cardiac cell growth is justified.  

Unlike the first iteration, we were able to extract meaningful data about the effects of F1R1 and TGFb on cardiac fibroblast proliferation with Ki67 staining. Specifically, there was a significant difference in Ki67 positivity between the F1R1 condition and both the TGFb and F1R1/TGFb groups (Table 2). This suggests that F1R1 is able to upregulate proliferation-promoting pathways in cardiac fibroblasts while TGFb has an opposing effect. Although there was no significance between the control and F1R1 conditions, this could have been influenced by the previously mentioned effects of serum media and intercellular signaling. Cardiomyocytes could have a similar response to F1R1 treatment, as they share differentiation and proliferation pathways with cardiac fibroblasts due to their shared progeny²². This cements the potential for F1R1 to be a catalyst for cardiomyocyte proliferation, especially in vivo where the massively reduced stiffness would be more favorable towards a proliferative environment rather than a differentiative one.

Our results so far suggest that F1R1 inhibits the TGF-B differentiation pathway. The next steps for this study will be to further investigate and confirm the interaction. While there was no time to perform it this semester, we have already established and obtained an antibody panel of proteins associated with TGFb-mediated activation for a Western Blot experiment.