**3. Results**

#### *3.1. Structural Modification of HU-600*

The syntheses 2-OHOA ethanolamide (HU-585) was done by the addition of anandamide side chain to 2-OHOA (HU-600) (Figure 1).

**Figure 1.** Structures of the synthetic compounds tested for anticancer activity. Structures of (**A**) HU−600 (Minerval, <sup>2</sup>−hydroxyoleic acid), (**B**) anandamide (AEA) and (**C**) HU−585 that was obtained by structural modification and the addition of AEA side chain to HU−600.

*3.2. HU-585 Induces Cell Growth Inhibition, Reduced Colony Formation, and Reduced Migration in the Neuroblastoma Cell Line SK-N-SH In Vitro*

In order to evaluate the antitumorigenic effects of HU-600 and its derivative, HU-585, on SK-N-SH cells, we used MTT, CFU, and migration assays (Figure 2). The choice of concentrations used in these assays (12.5–200 μM) is based on previous data showing that the IC50 of HU600 for most cancer cells studied is in the range of 30–250 μM [16,21,22]. A similar and significant dose-dependent decrease in cell viability was demonstrated by both compounds at a concentration of 12.5 μM (16.7%, *p* < 0.05). The effect on cell viability reduction was significantly better for HU-585 treatment in comparison to HU-600 treatment at concentration of 75 μM and above: 33.5% vs. 14.4% at 75 μM; 34.3% vs. 14.7% at 100 μM; 52.2% vs. 23.2% at 200 μM, respectively. \* *p* = 0.03, \*\* *p* = 0.02, \*\*\* *p* = 0.01, respectively (Figure 2A).

**Figure 2.** Anti-tumorigenic effects of HU-585 and HU-600. (**A**) Cell viability, measured with MTT assay in SK-N-SH cell line following treatment with HU-585 and HU-600. The cells were plated in 96-well plates and treated with increasing concentrations of HU-600 or HU-585 for 72 h. A significant decrease in cell viability was detected for both compounds at concentration of 12.5 μM (16.7%, *p* < 0.05). A better effect in cell viability reduction was obtained for HU-585 at concentration 75 μM and above. The cell viability reduction for HU-585 and HU-600 at 75 μM was 33.5% and 14.4%, respectively (\* *p* = 0.03), at 100 μM, 34.3% and 14.7%, respectively (\*\* *p* = 0.02), and at 200 μM, 52.2% and 23.2%, respectively (\*\*\* *p* = 0.01). Data are expressed as percentage of the vehicle control and are the mean of pooled results from several experiments (*n* = 6) performed in triplicate. Statistical significance was determined by GEE test. \* *p* < 0.05 compared to HU-585 (1.15 < SE values < 1.99). (**B**,**C**) The effect of HU-585 and HU-600 on the colony formation ability of SK-N-SH cells showing a better effect of HU-585. The cells were plated in 6-well plates and treated with 12.5 μM, 37.5 μM and 75 μM of HU-585 or HU-600. Cells were fixed and stained with crystal violet 14 days later. Colonies were counted using ImageJ software. (**B**) Scanned image of representative wells showing different levels of colony formation in SK-N-SH treated cells. (**C**) Representation of the quantified number of colonies in increasing concentrations of the treatment used. Statistical significance was determined by GEE test, \* *p* < 0.05 (*n* = 3, performed in duplicates). (**D**,**E**) Migration rate of SK-N-SH cells was

decreased following treatment with HU-585. Using ImageJ software, pre-migration (0 h) and postmigration (24 h) images of untreated and following treatment with 75 μM of HU-585 or 75 μM of HU-600 were taken (**D**). Treatment with HU-585 resulted in a better inhibitory effect of the migration rate in comparison to untreated and to HU-600 treated cells. Migration rate of the cells was quantified as the average percentage of gap closure following treatment (**E**). Data is reported as mean ± SE of triplicates. Statistical significance was determined by GEE test, \* *p* ≤ 0.05 compared to the two other group (*n* = 3, performed in duplicates).

Cologenic assay was used to determine the effect of HU-600 and HU-585 treatment on the cellular cologenic potential of SK-N-SH cells (Figure 2B,C). The results show a significant reduction in the number of colonies formed following HU-585 treatment in comparison to untreated cells (mean 29.2 colonies/field vs. 40.55 colonies/field respectively, *p* < 0.001). The significant difference between HU-585 treatment and untreated cells was observed in all concentrations used (12.5 μM, 37.5 μM & 75.5 μM, *p* < 0.05) (Figure 2C). A similar effect of reduction in colony formation was not obtained following treatment with HU-600.

Finally, to assess the effect of HU-600 and HU-585 treatment on SK-N-SH cell migration ability, a wound healing assay was performed. The results show a significant reduction in migration rate of cells treated with 75.5 μM of HU-585 in comparison to untreated cells. Decrease in migration rate following treatment with HU-600 was not observed (Figure 2D,E).

Taken together, these results provide evidence that both HU-585 and HU-600 have antitumorigenic effects as demonstrated by reduced cells viability following treatment. However, the new compound HU-585 obtained by structural modification of HU-600, has a better anti-tumorigenic effect in the MTT assay at concentration of 75 μM and above and in all the additional in vitro studies that were performed.

#### *3.3. Apoptotic Cell Death and Senescence Following HU-585 Treatment in SK-N-SH Cells*

To verify our hypothesis that HU-585 induced reduction in NB cell viability was indeed due to apoptotic cell death, we first examined the morphological changes following HU-585 treatment. Microscopic analysis showed that treatment with 75 μM and 100 μM of HU-585 affected cell morphology and increased the number of cells that had lost their normal shape and became rounded, swollen and floated in the medium (data not shown). These results confirmed that HU-585 treatment might induce the appearance of typical features of apoptosis. Next, we used caspase assay and Bcl-2 levels following treatment to better evaluate the apoptotic activity of HU-585. Staurosporine-treated cells were used as the positive control to apoptosis.

Treatment of SK-N-SH cells with 100 μM of HU-585 induced apoptosis as demonstrated by cleavage of caspase-3 represented by the appearance of activated 17 kDa and 19 kDa fragments on western blot (Figure 3A). The apoptotic effect was observed at 24 h and peaked at 48 h (Figure 3B). In order to evaluate Bcl-2 levels following treatment, we first determined its baseline levels in SK-N-SH cells in comparison to fibroblasts and human astrocyte cell line. Western blot revealed a high baseline level of Bcl-2 protein in the SK-N-SH cells in comparison to normal controls (Figure 3C). Treatment with 100 μM of HU-585 resulted in a decreased level of Bcl-2 with a better and significant effect at 48 h in comparison to 24 h (Figure 3D,E).

As senescence is generally regarded as a tumor suppressive process that evolves alongside apoptosis to suppress tumorigenesis, we next assessed the effect of treatment with HU-600 and HU-585 on cell senescence. β-Galactosidase activity was measured by blue-colored cell counting (Figure 3F). Results show that cellular senescence occurred at 48 h following treatment with HU-585, but not with HU-600 (Figure 3G).

**Figure 3.** Growth suppression by HU-585 is mediated by apoptosis and senescence. (**A**–**E**) Apoptotic effects of HU-585. (**A**) Caspase-3 assay. Western blot of caspase-3 (35 kDa) cleavage to 19 kDa and 17 kDa represents the apoptotic effect of HU-585. SK-N-SH cells were treated with 100 μM of HU-585 for 24 h and 48 h. (**B**) Quantification of cleaved caspase-3. Quantification was done by Image Lab software and representation for 24 h and 48 h of incubation are shown. Cleaved 19 kDa level increased significantly in a time-dependent manner in comparison to its control following treatment with HU-585. Statistical differences at 24 h and 48 h were determined by two-way ANOVA (\* *p* < 0.01). (**C**) Bcl-2 protein levels. Western blot of Bcl-2 protein levels determination for human astrocyte cell line (HA), SK-N-SH cells and fibroblast cell line (as a "normal control"). High levels of Bcl-2 protein were detected in SK-N-SH cells compared to the other cell lines. (**D**,**E**) Apoptotic effects of HU-585 analyzed by Bcl-2 levels following treatment. (**D**) Western blot of Bcl-2 protein levels in SK-N-SH cells following treatment with 100 μM HU-585 for 24 h and 48 h. (**E**) Quantification of Bcl-2 protein. Quantification was done by Image Lab software following treatment with 100 μM of HU-585 for 24 h and 48 h. Results revealed that the level of Bcl-2 protein decreased in a time-dependent manner, reaching maximal and significant effect at 48 h. Statistical significance was determined by Unpaired *t*-Test (\* *p* < 0.001). (**F**,**G**) Senescence effect of HU-585 analyzed by β-galactosidase staining. (**F**) Activity of β-galactosidase in SK-N-SH cells following HU-600 and HU-585 treatment was measured by β-galactosidase staining. SK-N-SH cells were plated in 6-well plates and treated with 50 μM, 75 μM and 100 μM of HU-600, HU-585 or no treatment as control. Positivity for β-galactosidase following 48 h of treatment represents the cells that are in senescence. (**G**) Representation of the quantified percentages of senescent cells. Results show an increased number of senescent cells following HU-585 treatment compared to HU-600 treatment and untreated cells. Data is reported as mean ± SE of triplicates. Statistical significance was determined by GEE test. \* *p* < 0.05 compared to HU-600 and # *p* < 0.05 compared to untreated cells.

Overall, these results provide evidence that treatment with HU-585 induces apoptosis and cellular senescence in SK-N-SH cells, while no such effect is demonstrated for HU-600 treatment.

#### *3.4. Combined Treatment of HU-585 with Anti Bcl-2 Compounds ABT-263 or ABT-737 Results in Cell Growth Inhibition In Vitro and in Tumor Growth Delay In Vivo*

As we have shown that SK-N-SH cells express high levels of Bcl-2 (Figure 3C), which have been shown to be important for neuroblastoma survival, we wished to study the effect of combined treatment of HU-585 with anti Bcl-2 compounds. First, MTT assay was performed to evaluate the in vitro effect of combined treatment of ABT-263 with HU-585 or HU-600. The reduction in SK-N-SH cell viability treated with the combination of HU-600 with 2.5 μM of ABT-263 was not significantly different in comparison to HU-600 alone (Figure 4A). In contrast, treatment of the cells with combination of HU-585 and 2.5 μM of ABT-263 resulted in a significantly reduced viability when compared to treatment with HU-585 alone. This effect was dose-dependent, with maximal effect observed in doses of 75 μM and 100 μM of HU-585 (Figure 4B).

Given the observed differential sensitivity of SK-N-SH cells to the various tested compounds, we further examined the in vivo antitumor effect using the nude mice Xn model. Nude immunodeficient mice bearing SK-N-SH Xns expressing mCherry were treated once daily for 21 days with HU-585, ABT-737, a combination of HU-585 and ABT-737 or with a vehicle control. For in vivo studies of senolytic compounds, ABT-737 (75 mg/kg) was used as previously described [23]. It was specifically chosen to be given IP in contrast to its closely related compound ABT-263 used in the in vitro tests that is an orally bioavailable agent. Based on the results of the in vitro studies that demonstrated a better efficacy of HU585 in comparison to its parent compound HU600 at lower doses, the doses selected for HU585 for the in vivo studies (120 mg/kg) were lower than the doses reported for HU600 treatment (200 mg/kg) in animal experiments [21]. Body weight change was assessed as an indicator of side effects and treatment toxicities. No significant weight loss was observed in any treatment group, indicating that this dosing strategy and the dosing intervals used were well tolerated and safe (Data not shown). The average total radiant efficiency signal as measured by IVIS was used to evaluate Xn response to treatment and tumor volume. At the end of treatment, the growth rate and volume of the Xns in the mice that were treated with the combination of HU-585 and ABT-737 were significantly lower in comparison to control or HU585 alone as shown in Figure 4C,D (*p* < 0.05). Moreover, the regression in tumor volume that was observed in the combined treatment group contrasted with the other groups in which progression was demonstrated. Together, the results sugges<sup>t</sup> additive in vitro and in vivo anti-tumorigenic effect of combined treatment of HU-585 with the anti Bcl-2 compounds ABT-263 and ABT-737.

**Figure 4.** The combination of HU-585 with anti Bcl-2 compounds results in enhanced tumor growth delay in vitro and in vivo. (**A**,**B**) The effect of combination therapy of ABT-263 with HU-600 or HU-585 on SK-N-SH cells viability. The cells were plated in 96-well plate and treated with increasing concentrations of HU-600 (**A**) or HU-585 (**B**) for 72 h, with or without 2.5 μM of ABT-263. Viability measurements are shown by MTT assay and demonstrate a better effect of ABT-263 treatment combined with HU-585 in comparison to HU-585 alone. A similar effect was not obtained following combination of ABT-263 with HU-600. All values are normalized to control. Data are expressed as percentage of the vehicle control and are the mean of pooled results from several experiments performed in triplicate. Statistical significance was determined by GEE test, \* *p* < 0.05 compared to HU-585 + ABT263 (0.86 < SE values < 4.22). (**C**,**D**) Combined treatment of ABT-737 with HU-585 inhibited tumor growth in mice model. A total of 5 × 10<sup>6</sup> SK-N-SH expressing mCherry cells were subcutaneously inoculated in the right flank of nude mice. The mice were divided into four groups and given vehicle, HU-585 (120 mg/kg), ABT-737 (75 mg/kg) or combination of HU-585 with ABT-737 (120 mg/kg, 75 mg/kg, respectively) once daily by IP injections for 21 days. (**C**) Tumor burden was followed by IVIS system once a week during treatment. (**D**) Average total radiant efficiency signal was significantly lower following combined treatment with HU-585 and ABT-737 compared to control or HU585 alone. The Kruskal-Wallis test and the post hoc Dunn's multiple comparisons test were used to evaluate significant differences in the growth rate of xenografts between treatment groups, \* *p* < 0.05 compared to control and HU-585.
