**3. Results**

#### *3.1. E*ff*ect of Omega-3 Supplements on Epithelial Integrity*

The distribution of the ZO-1 was similar after treatment with the 10 di fferent n3-PUFA supplements. No di fference was found among any group of study (Figure 1).

**Figure 1.** Effect of omega-3 supplements on epithelial tight junctions. ZO-1 (red) immunofluorescence were not affected by any of the omega-3 treatments. Nuclei are stained with TOPRO-3. Scale bar 20 μm. (**A**): Control group; (**B**): H2O2 treatment group; ( **C**): DHA (Docosahexaenoic acid) group; ( **D**): EPA (Eicosapentaenoic acid) group; (**E**): EPA/DHA 40/20; TG (Triglycerides) group; (**F**): EPA/DHA 20/40 TG group; ( **G**): EPA/DHA 40/20 EE (Ethylesters) group; ( **H**): DHA 97/3V (Vegetable) group; (**I**): DHA 95/5V group; (**J**): DHA 97/3M (Marine) group; ( **K**): DHA 97/3 VM 1.5 group; (**L**): DHA 95/5 VM 2.5 group.

#### *3.2. E*ff*ect of Omega-3 Supplements on Viability and Proliferation in ARPE-19*

First, we wanted to evaluate the effect of Docosahexaenoic acid (DHA) and Eicosapentaenoic acid (EPA) on cell viability and proliferation. Treatment with both n3-PUFA produced a significant increase in viability (Figure 2A) of ARPE-19 cells compared to untreated controls. Proliferation also seemed to increase, but the effect was not statistically significant (Figure 2D). Secondly, we assessed the effect of different combinations of DHA and EPA on these cellular parameters. Interestingly, only the EPA/DHA 40/20 TG combination was able to significantly increase both viability (Figure 2B) and proliferation (Figure 2E), when compared to the untreated cells. Finally, we evaluated five different supplements containing a mixture of DHA in the form of TG and PL (PL from marine, vegetable, or mixed origin). In this last set of experiments, none of the formulations produced a significant change in cell viability (Figure 2C) or proliferation (Figure 2E).

**Figure 2.** Graphs showing relative cell viability (**A**–**C**) and proliferation (**D**–**F**) compared to a control group under normal conditions of groups 1, 2, and 3 of eicosapentaenoic acid/docosahexaenoic acid (DHA/EPA) formulations, respectively (**D**–**F**). For comparisons, one-way ANOVA with the Bonferroni post-hoc test were used. Data are expressed as mean ± SEM. \* *p* < 0.05. EPA: Eicosapentaenoic acid; DHA: Docosahexaenoic acid; TG: Triglycerides; EE: Ethylesters; PL: Phospholipids. V: Vegetable. M: Marine.

#### *3.3. E*ff*ect of Omega-3 Supplements on Cell Viability and Cell Proliferation under Oxidative Stress and Inflammatory Conditions*

In order to replicate the local oxidative environment of the diabetic retina, we subjected some ARPE-19 cells to H2O2 and measured the response after n3-PUFA treatment. As expected, exposure to 800 μM H2O2 produced a significant decrease in cell viability and proliferation. Treatment with EPA and DHA was able counteract the H2O2 induced decrease on cell viability (Figure 3A), but their effect on cell proliferation, although positive, was not statistically significant (Figure 3D). On the contrary, both n3-PUFA were able to significantly counteract the effect of H2O2 on cell proliferation (Figure 3D). In the combined formulations, all supplements were able to significantly reverse the oxidative effect of H2O2 on viability and proliferation, with the exception of the 40/20 EPA/DHA EE supplement that was not able to significantly mitigate the effect of H2O2 on cell viability (Figure 3B,C,E,F).

**Figure 3.** Graphs showing relative cell viability (**A**–**C**) and proliferation (**D**–**F**) results for ARPE-19 cells treated with different DHA and EPA treatments groups under oxidative stress conditions (H2O2 For comparisons, one-way ANOVA with the Bonferroni post-hoc test were used. Data are expressed as mean ± SEM. \* *p* < 0.05 and \*\* *p* < 0.01. EPA: Eicosapentaenoic acid; DHA: Docosahexaenoic acid; TG: Triglycerides; EE: Ethylesters; PL: Phospholipids. V: Vegetable. M: Marine.

#### *3.4. E*ff*ect of Omega-3 Supplements on ROS Production*

Both DHA and EPA, alone or in combined formulations, produced a significant decrease in ROS when compared with the untreated controls. The same effect was observed in cells treated with the different DHA TG+PL formulations. The EPA/DHA 40/20 TG and DHA 95/5 VM 2.5 formulations had a stronger antioxidant effect, but this difference was not statistically significant when compared with the other combinations (Figure 4A–C).

**Figure 4.** Graphs showing relative Dichloride fluoresceine (DCF) detection results for ARPE-19 cells in groups 1, 2, and 3 of DHA/EPA formulations, respectively (**A–C**). For comparisons, one-way ANOVA with the Bonferroni post-hoc test were used. Data are expressed as mean ± SEM. \*\* *p* < 0.01 \*\*\* *p* < 0.001. EPA: Eicosapentaenoic acid; DHA: Docosahexaenoic acid; TG: Triglycerides; EE: Ethylesters; PL: Phospholipids. V: Vegetable. M: Marine.

#### *3.5. E*ff*ect of Omega-3 Supplements on Caspase-3 in ARPE-19*

In order to test the safety of the different supplements, we evaluated their capacity to activate the early apoptotic mediator caspase-3. As expected, none of the omega-3 formulations induced a significant change in caspase-3 activation, when compared to the untreated control cells (Figure 5).

**Figure 5.** Graphs showing relative caspase-3 expression (positive granules per nucleus) (**A–C**) results for ARPE-19 cells treated with different groups 1, 2, and 3 of DHA/EPA formulations, respectively. For comparisons, one-way ANOVA with the Bonferroni post-hoc test were used. (**D**) Scale bar 20 μm Data are expressed as mean ± SEM. EPA: Eicosapentaenoic acid; DHA: Docosahexaenoic acid; TG: Triglycerides; EE: Ethylesters; PL: Phospholipids. V: Vegetable. M: Marine.

#### *3.6. E*ff*ect of Omega-3 Supplements on Wound Healing Cell Migration Assay*

Treatment with EPA or DHA did not produce any significant change in the migration capacity of ARPE-19 cells (Figure 5D). In the EPA/DHA formulations, the EPA/DHA 20/40 TG supplement produced a small, but significant decrease in the speed of wound closure, when compared with untreated cells. The rest of the combined EPA/DHA formulations and those combining DHA in TG and PL forms did not produce any significant changes in this cellular function (Figure 6).

**Figure 6.** Graphs showing relative migration ratio (**A–C**) results for ARPE-19 cells treated with different groups 1, 2, and 3 of DHA/EPA formulations, respectively. For comparisons, one-way ANOVA with the Bonferroni post-hoc test were used. Data are expressed as mean ± SEM. \*\* *p* < 0.01. EPA: Eicosapentaenoic acid; DHA: Docosahexaenoic acid; TG: Triglycerides; EE: Ethylesters; PL: Phospholipids. V: Vegetable. M: Marine.

#### *3.7. E*ff*ect of Omega-3 Supplements on the VEGF*/*PEDF Ratio*

Treatment with the different omega-3 formulations did not produce a significant effect in the protein levels of VEGF or PEDF. However, combined formulations (especially those with EPA+DHA), showed a tendency to decrease the VEGF/PEDF ratio when compared with untreated controls, but these differences were not statistically significant (Figure 7).

**Figure 7.** Graphs showing relative VEGF/PEDF ratio (**A–C**) results for ARPE-19 cells treated with different groups 1, 2, and 3 of DHA/EPA formulations, respectively. For comparisons, one-way ANOVA with the Bonferroni post-hoc test were used. Data are expressed as mean ± SEM. EPA: Eicosapentaenoic acid; DHA: Docosahexaenoic acid; TG: Triglycerides; EE: Ethylesters; PL: Phospholipids. V: Vegetable. M: Marine.
