**3. Results**

### *3.1. Crocetin Reduced Intracellular Fat in 3T3-L1 Adipocytes*

Oil Red O (OR) staining at the FD time-point allowed visualizing the e ffect of CCT on the storage of intracellular lipid in di fferentiated 3T3-L1 adipocytes. OR staining allows estimating the amount of intracellular fat. The maximum intracellular fat detected in the control di fferentiated cells (CCT 0 μM) at the FD time-point was set to 100%, and the relative intracellular fat levels at the two CCT concentrations at the same time-point are depicted in Figure 2. As shown in this figure, CCT-di fferentiated cells at 5 μM showed a significant decrease in the content of intracellular fat (77.4 ± 11.2%, *p* < 0.01) compared

to control differentiated cells. No significant effect was observed on the content of intracellular fat in cells differentiated with 1 μM CCT (97.4 ± 5.4%) or with DMSO (101 ± 10%), with respect to control differentiated cells. As expected, rosiglitazone significantly enhanced intracellular fat (219 ± 42%) as compared to control and CCT-differentiated cells (*p* < 0.001), whereas genistein significantly reduced intracellular fat (78 ± 7%) as compared to the control (*p* < 0.05). The effect of 5 μM CCT did not present a significant difference compared to genistein.

**Figure 2.** Effect of CCT on lipid accumulation. The maximum intracellular fat detected in control differentiated cells (CCT 0 μM) was set to 100%; the relative intracellular fat levels of CCT-differentiated cells are shown. Values were obtained at the FD time-point. Data from at least 10 replicates in each condition from three independent experiments are expressed as the mean ± SD. \*\* indicates a significant difference compared with the control differentiated cells (*p* < 0.01); aa indicates a significant difference compared with 1 μM CCT (*p* < 0.01).

### *3.2. Crocetine Did Not A*ff*ect the Total Number of Lipid Drops or Their Size*

Additionally, we evaluated whether the presence of CCT during the induction of differentiation resulted in a reduction in the number or size of lipid droplets, which could explain the lower lipid load seen in Figure 2. This was analyzed at the FD time-point using the aforementioned WinLipid software. It was observed that the induction of differentiation in the presence of the two CCT concentrations did not affect the total number of lipid droplets (Figure 3A). Regarding the size of the lipid droplets, there was no statistical difference after the induction of differentiation in the presence of CCT; however, a slight shift to the left was observed in the size distribution frequency curve. Thus, after the induction of differentiation in the presence of CCT, the highest droplet percentage was found within a smaller interval size (60–109 Px) than that of the control differentiated cells (110–159 Px) (Figure 3B).

**Figure 3.** Effect of CCT on the number and size of lipid droplets. (**A**) Final lipid droplet number. Box-and-whisker plot depicting the minimum to maximum values, with the middle line representing the mean ± SD. (**B**) Intervals of frequency–area distribution. The vertical dotted line indicates the size

range in which most lipid droplets were found. Droplets were analyzed at the FD time-point using WinLipid software with 1–2 photomicrographs per well from three independent experiments; 0 μM CCT (control; *n* = 15), 1 μM CCT (*n* = 11), 5 μM CCT (*n* = 10), where *n* represents the number of analyzed photomicrographs. Px, pixels.

Thus, this result indicates that the reduction in intracellular fat produced by CCT was neither due to a lower generation of lipid droplets nor a reduction in their size.

### *3.3. Crocetin Did Not A*ff*ect 3T3-L1 Cell Viability*

In order to assess the safety of CCT, two cell viability assays were conducted at the FD time-point. Figure 4A displays the percentage cellular metabolic activity of the 3T3-L1 adipocytes obtained using the MTT assay. This assay uses metabolic activity as an indicator of cell viability by evaluating the efficiency of mitochondrial enzymes. Mitochondrial activity generates a change in color measured by spectrophotometry. The amount of color produced is proportional to the metabolic activity, thus estimating cell viability. Our results indicate no statistically significant di fference in cellular viability between the control di fferentiated cells (CCT 0 μM) and CCT-di fferentiated adipocytes. Therefore, the viability of 3T3-L1 adipocytes was not a ffected by the concentration of CCT (1 or 5 μM: 107 ± 8% or 97 ± 8%, respectively) or DMSO (106 ± 17%). Additionally, the percentage of viable cells was determined using TB dye. Figure 4B displays the percentage of live cells with respect to total cells. Similarly, to the results obtained with MTT, no statistically significant di fference in the percentage of living cells was found between the control di fferentiated cells (CCT 0 μM) and CCT-di fferentiated adipocytes. Thus, the viability of 3T3-L1 adipocytes was not a ffected by the concentration of CCT (1 or 5 μM: 90 ± 13% or 103 ± 13%, respectively) or DMSO (95 ± 12%); furthermore, their viability was neither a ffected by rosiglitazone (111 ± 8%) nor genistein (102 ± 8%).

**Figure 4.** Effect of CCT on cell viability. The maximum viability detected for control di fferentiated cells (CCT 0 μM) was set to 100%; the relative viability of CCT-di fferentiated cells is shown. Values were obtained at the FD time-point. Data are expressed as the mean ± SD. (**A**) 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay: data were obtained from at least 10 replicates in each condition from three independent experiments. (**B**) Trypan blue (TB) assay: two readings were performed per well using an automated cell counter. The percentage of viable cells (alive cells with respect to total cells) was averaged. Data were obtained from at least two replicates in each condition from three independent experiments.

In summary, our results indicate that neither CCT nor DMSO decreased the adipocyte mass due to cell death; thus, the decrease in intracellular fat observed in CCT-di fferentiated adipocytes was independent of nonspecific cell toxicity.

### *3.4. Crocetin Altered the Expression of Early and Late Genes during the Adipogenic Process*

To determine the effect of CCT on the induction of the adipogenic process at a molecular level, the expression of early (*C*/*EBP*β and *C*/*EBP*δ) and late (*PPAR*γ and *C*/*EBP*α) genes of adipogenesis was quantified using qRT-PCR at the ID and FD time-points, respectively.

As shown in Figure 5 (panel A), 5 μM CCT significantly decreased *C*/*EBP*β mRNA levels by 48.2% compared to control differentiated cells; however, this concentration of CCT did not alter *C*/*EBP*δ mRNA levels. On the other hand, 1 μM CCT modified the mRNA levels of both genes. As for the late genes, both doses of CCT (1 and 5 μM) significantly decreased *C*/*EBP*α mRNA levels (by 36.4% and 35.7%, respectively), without affecting the levels of *PPAR*γ (Figure 5, panel B). The expression of *aP2-1* as a specific adipocyte marker was also evaluated at the FD time-point, showing no statistical difference in mRNA level with respect to control differentiated cells (Figure 5, panel B).

**Figure 5.** Effect of CCT on messenger RNA (mRNA) expression levels of genes regulating the adipogenic process. The relative quantitative real-time PCR (qRT-PCR) values were corrected to actin expression levels and normalized to control differentiation (0 μM CCT). Data were obtained from three independent experiments and are expressed as the mean ± SD. The maximum mRNA expression level for 0 μM CCT was set to 1 (dashed line); the relative mRNA expression levels obtained with the two CCT concentrations (1 and 5 μM) at the same time-point are depicted. (**A**) Cells were collected at the ID-time-point for the expression of early genes, CCAAT/enhancer-binding protein (*C*/EBP)β and *C*/*EBP*δ, evaluated using qRT-PCR with specific primer pairs. (**B**) Cells were collected at the FD time-point for the expression of late genes, peroxisome proliferator-activated receptor γ (*PPAR*γ) and C/*EBP*<sup>α</sup>, evaluated using qRT-PCR with specific primer pairs; the expression of *aP2-1* (as a specific adipocyte marker) was also evaluated. \*\* indicates a significant difference compared with control differentiated cells (*p* < 0.01).
