To further understand the binding between HSA and
N-BF2, drug competition and molecular docking experiments were next performed. Phenylbutazone and ibuprofen, which have been known to bind HSA/BSA at site Ⅰ and site Ⅱ, respectively, are used to inspect the binding model. Both
N-BF2/HSA and
N-BF2/BSA complexes shown slight decrease in presence of 50–200 µmol/L phenylbutazone. However, the addition of ibuprofen could induce the fluorescence intensity of
N-BF2 sharply decreased to 30% (
Fig. 3a). It indicated that
N-BF2 most likely to bind HSA/BSA at narrow site Ⅱ which considered as the common drug sites. The following docking study between
N-BF2 and HSA/BSA also demonstrated this result (
Fig. 3b and Fig. S3 in Supporting information). It revealed that oxygen in difluoroboron
β-diketonate could formed two stable hydrogen bonds of 2.1 Å and 2.5 Å with the basic amino acid Arg-410 and Lys-414, which made
N-BF2 specifically bind site Ⅱ. On the other side, hydrophobic interactions of aniline unit with surrounding neutral amino acids, Phe-403, Val-433, Leu-453, Ala-449, Cys-438, Cys-392 and Ile-388, were found thus inhibiting the rotation of aniline and enhancing fluorescence intensity of
N-BF2. Inspired by the stable binding model, titration experiment and double logarithmic equation was further conducted to examine the limit of detection (LOD) and binding constant (
Kb). With the addition of HSA, the fluorescence intensity gradually enhanced and the emission wavelength shifted from 583 nm to 568 nm (
Fig. 3c). A linear relationship (
R2 = 0.9845) was found between fluorescence intensity of
N-BF2 and HSA concentration in the range of 0–10 µmol/L (
Fig. 3d). Based on 3
σ/
k, the corresponding limit of detection (LOD) for HSA was calculated to be 1.21 mg/mL (
Fig. 3d). Besides, with the addition of
N-BF2, the fluorescence intensity at 340 nm gradually decreased and the concentration-dependent fluorescence changes were utilized to find the binding constant, the
Kb was 4.61 × 10
5 L/mol (
Figs. 3e and
f). LOD and
Kb of
N-BF2 toward BSA was also inspected to be 0.93 mg/mL and 4.67 × 10
5 L/mol which indicated the strong chelation between HSA/BSAand
N-BF2 (Figs. S4 and S5 in Supporting information). As an endogenous substance
in vivo, HSA not only has high stability but also acts as a drug carrier which can increase the water solubility of drugs and regulate drug release. Therefore, we used
N-BF2/HSA complexes as a trigger to control the entry and exit of
N-BF2 in living cells. As shown in
Figs. 4a–
c,
N-BF2 could specifically light up lipid droplets in HeLa cells. With the addition of 30 µmol/L HSA, the fluorescence of
N-BF2 in lipid droplets disappeared, indicating that
N-BF2 gradually dissociated from lipid droplets and flowed out of cells to bind HSA (
Figs. 4d–
f). Ibuprofen could bind HSA with higher affinity. Then the addition of ibuprofen replaced
N-BF2 from HSA and the released
N-BF2 could re-cross the cell membrane and go back into lipid droplets which had been confirmed by the recovery of fluorescence in lipid droplets (
Figs. 4g–
i). So, it could be concluded that the complex of
N-BF2/HSA indeed realized the regulation of reversible lipid droplet staining in cells.