Successful synthesis of BBA and FA-PEG-CM-β-CD
Additional file 1: Figure S1 showed the 1H-NMR and 13C-NMR spectra of 2-hydroxy-5-butylamino benzoic acid that confirmed the successful synthesis of it. Characteristic peaks of 2-hydroxy-5-butylamino benzoic acid appeared at (g) and (h). Its NMR data were: 1H NMR (400 MHz, DMSO): δ 10.97 (s, 1H), 9.84 (s, 1H), 8.10 (s, 1H), 7.64 (s, 1H), 6.88 (s, 1H), 2.24 (t, J = 7.3 Hz, 2H), 1.58 (s, 2H), 0.90 (s, 3H). 13C-NMR (101 MHz, DMSO): δ 172.44 (s), 171.39 (s), 157.32 (s), 131.54 (s), 127.82 (s), 120.60 (s), 117.42 (s), 112.93 (s), 39.00 (s), 19.05 (s), 14.25 (s).
Additional file 1: Figure S2 confirmed the successful synthesis of BBA. Characteristic peaks of BBA appeared at (i). Its NMR data were: 1H NMR (400 MHz, Acetone) δ 9.29 (s, 1H), 8.29 (s, 1H), 7.96 (s, 1H), 7.08 (s, 1H), 2.54 (s, 2H), 2.36 (s, 2H), 1.70 (s, 4H), 0.98 (d, J = 17.7 Hz, 6H). 13C NMR (101 MHz, Acetone) δ 172.79 (s), 172.48 (s), 166.18 (s), 147.33 (s), 138.60 (s), 125.35 (s), 125.12 (s), 123.10 (s), 39.90 (s), 36.73 (s), 30.30 (s), 19.90 (s), 19.11 (s), 14.35 (d, J = 8.6 Hz).
Additional file 1: Figure S3A–C showed the 1H NMR spectra of CM-β-CD (A), FA-PEG-NH2 (B), and FA-PEG-CM-β-CD (C). The NMR data of FA-PEG-CM-β-CD were: 1H NMR (400 MHz, D2O) δ 8.67 (s, 1H), 8.28 (m, 1H), 7.59 (s, 1H), 6.69 (s, 1H), 5.24 (s, 10H), 4.23 (s, 8H), 4.03 (s, 2H), 3.79 (s, 31H), 2.79 (d, J = 5.2 Hz, 8H), 1.99 (s, 1H), 1.81 (s, 1H), 1.28 (m, 1H), 1.09 (s, 1H), 0.98 (s, 3H).
The infrared spectrum was shown in Additional file 1: Figure S3D. The characteristic absorption peaks around 3434 cm−1 and 1653 cm−1 responded to N–H and C = O of the amide bonds in FA-PEG-CM-β-CD (a). The characteristic absorption peak of the conjugated N–H around 3499 cm−1 overlapped with the hydroxyl group (O–H) in the spectra of FA-PEG-NH2 (c). Peaks due to vibrations of the conjugated O–H bond and C = O bond was detected around 3416 cm−1 and 1599 cm−1 in the spectra of CM-β-CD (d). These results further proved that FA-PEG-CM-β-CD was successfully synthesized.
Characterization of BBA/FA-PEG-CM-β-CD
The infrared spectra were shown in Fig. 2A. A characteristic peak of the conjugated O–H bond was detected near 3295 cm−1 in the spectra of BBA (a), and a strong absorption peak of benzene ranged near 1699 cm−1. The absorption peaks of BBA and FA-PEG-CM-β-CD (b) did not simply overlap in the infrared spectrum of the BBA/FA-PEG-CM-β-CD (d). Some characteristic peaks of BBA disappeared or weakened, in which the N–H stretching vibration peak near 3295 cm−1 disappeared, and a peak due to vibrations of the conjugated C–O–C bond around 1103 cm−1 was weakened in the spectra of FA-PEG-CM-β-CD (b). It was concluded that BBA and FA-PEG-CM-β-CD successfully formed the inclusion compound of BBA/FA-PEG-CM-β-CD.
The XRD pattern was shown in Fig. 2B. In the X-ray pattern, the obvious crystal diffraction peak of BBA (a) could be observed, while no obvious crystal diffraction peaks were seen in the FA-PEG-CM-β-CD (b). Compared with the diffraction pattern of BBA (a), the peak position of the BBA/FA-PEG-CM-β-CD inclusion compound (d) shifted, the measurement angle changed, and the characteristic peak disappeared, which preliminarily proved the formation of BBA/FA-PEG-CM-β-CD.
DSC results of BBA (a), FA-PEG-CM-β-CD (b), the mixture of BBA with FA-PEG-CM-β-CD (c), and BBA/FA-PEG-CM-β-CD (d) were shown in Fig. 2C. Two different peaks were observed in BBA (a) and FA-PEG-CM-β-CD (b). However, after BBA was incorporated into FA-PEG-CM-β-CD, the peak at 360 °C decreased greatly. It might be attributed to the formation of BBA/FA-PEG-CM-β-CD.
As shown in Fig. 2D, BBA (a) and FA-PEG-CM-β-CD (b) had a two-step weight loss following as one was below 90 °C due to the water loss, another was ranging 270 °C—390 °C which was the main thermal degradation of BBA and FA-PEG-CM-β-CD. BBA/FA-PEG-CM-β-CD (d) exhibited a 3-step weight loss. The first reduction was seen below 90 °C, due to the water loss of BBA or FA-PEG-CM-β-CD. The other two steps were observed at around 200 °C − 400 °C and 400 °C − 600 °C, which might be attributed to the double thermal degradations of BBA/FA-PEG-CM-β-CD.
As shown in Fig. 2E&F, SEM was used to observe the external morphology of FA-PEG-CM-β-CD (E) and BBA/FA-PEG-CM-β-CD (F). FA-PEG-CM-β-CD existed as lumps. But in the morphological image of BBA/FA-PEG-CM-β-CD, the crystal form was dislocated and the boundary deformed under the action of mechanical force and thus forming small spherical particles that were stacked in some blocks.
The in vitro accumulative release (%) of BBA/FA-PEG-CM-β-CD was investigated by dialysis. As shown in Fig. 2G, BBA/CM-β-CD and BBA/FA-PEG-CM-β-CD released only 82% of BBA in vitro during 96 h, whereas 90% of free BBA was released during the first 24 h. The sustained drug release from BBA/CM-β-CD and BBA/FA-PEG-CM-β-CD with no initial burst release suggested that BBA was not adsorbed on the surface but encapsulated completely inside FA-PEG-CM-β-CD.
Through detecting the encapsulation efficiency of BBA/FA-PEG-CM-β-CD under high temperature and high humidity conditions, it could be seen (Fig. 2H) that as the temperature or humidity increased, the encapsulation rate didn’t change, indicating that BBA/FA-PEG-CM-β-CD in high temperature or high humidity conditions still kept stable.
In vitro anticancer activity of BBA/FA-PEG-CM-β-CD
MTT assay was used to investigate the cell-growth inhibition effects of BBA/FA-PEG-CM-β-CD at various concentrations against CaCo-2 and SW620 cells at 24 h, 48 h, and 72 h post-treatment (Fig. 3A).
Compared with NaB, the dual-prodrug BBA showed some cell-proliferation inhibition effect on both CaCo-2 and SW620 at the time points of 24 h, 48 h, and 72 h (P < 0.0001). Meanwhile, as prediction the treatment of 5-FU (20 μg/mL) drastically enhanced the cell proliferation inhibition rate compared with the BBA group ( P < 0.0001). BBA/FA-PEG-CM-β-CD showed much better inhibitory effects than BBA, BBA/CM-β-CD. The proliferation inhibition effect of BBA/FA-PEG-CM-β-CD on SW620 cells was better than that on CaCo-2 cells, so we took SW620 cells as the cell model for subsequent tumor-relative evaluation.
Uptake of BBA/FA-PEG-CM-β-CD by cells in vitro
As shown in Fig. 3B&C, cells treated with BBA/CM-β-CD (30 μM) or BBA/FA-PEG-CM-β-CD (30 μM) showed co-localization of green fluorescence from coumarin 6 and blue fluorescence from DAPI, suggesting that the inclusion complexes had been ingested by the cells. After treatment with BBA/FA-PEG-CM-β-CD, SW620 cells showed much stronger fluorescence signals than CaCo-2, which may be due to that SW620 cells were more sensitive to BBA/FA-PEG-CM-β-CD than CaCo-2 cells. Meanwhile, in BBA/FA-PEG-CM-β-CD group the fluorescence intensity was much greater than that in BBA/CM-β-CD group, demonstrating a targeting ability of BBA/FA-PEG-CM-β-CD mediated by FA ligand.
As shown in Fig. 3D, both CaCo-2 cells and SW620 cells were treated with BBA/FA-PEG-CM-β-CD containing increasing concentration of FA. The uptake of BBA/FA-PEG-CM-β-CD was competitively inhibited by excess FA. This also proved the activity targeting ability of BBA/FA-PEG-CM-β-CD.
Mechanism analysis about the inhibitive effect of BBA/FA-PEG-CM-β-CD
To fully understand the inhibitive effect mechanism of BBA/FA-PEG-CM-β-CD on the SW620 cells, we further performed cell apoptosis and cell cycle assays. As shown in Fig. 4, SW620 cells were treated with different formulations for 48 h. BBA, BBA/CM-β-CD, and BBA/FA-PEG-CM-β-CD obviously induced apoptosis in SW620 cells compared with the blank group. The apoptotic rates including early apoptosis and late apoptosis were respectively 32.21% ± 8.26%, 46.28% ± 9.4%, and 51.56% ± 8.54% in groups of BBA, BBA/CM-β-CD, and BBA/FA-PEG-CM-β-CD, indicating that BBA/FA-PEG-CM-β-CD induced the strongest apoptosis against SW620 cells.
As shown in Fig. 5, the treatments of BBA, BBA/CM-β-CD, and BBA/FA-PEG-CM-β-CD on SW620 cells induced a significant decrease of cells in the S phase (P < 0.0001) and a significant increase of cells in the G0/G1 phase (P < 0.01), which proved that the SW620 cell was mainly arrested at the G0/G1 phase. Both BBA/CM-β-CD and BBA/FA-PEG-CM-β-CD induced higher cell cycle arrest in G0/G1 phase than BBA.
BBA/FA-PEG-CM-β-CD showing no in vivo acute toxicity
We examined the in vivo toxicity of BBA/FA-PEG-CM-β-CD in normal Kunming mice. It was found that the hair color, diet, and activities of all mice remained normal. No abnormal behavior, toxic symptoms, bodyweight loss, and death were observed after all the treatments except the 5-FU group. As shown in Table 1, no significant difference in the average body weight was observed among the treatment groups except 5-FU, in which the tested mice treated with 5-FU significantly decreased by 4.46% (P < 0.05). These results demonstrated that BBA/FA-PEG-CM-β-CD showed no toxicity to normal mice.
As shown in Table 2, the viscera indexes of heart, liver, lung, and kidney exhibited no significant changes after treatment with BBA and BBA/FA-PEG-CM-β-CD compared with the saline group. However, the liver index increased significantly to 8.34% ± 0.43% after 5-FU treatment. Moreover, the kidney index also increased significantly to 3.06% ± 0.27% after 5-FU treatment, which was significantly different from the saline group (2.30% ± 0.34%, P < 0.01). Compared with 5-FU, the viscera indexes of the mice treated with BBA/FA-PEG-CM-β-CD showed no toxic effect.
In addition to the viscera index, we further conducted serum biochemistry analyses of ALT, AST, BUN, and Cr to assess the potential toxicities of BBA/FA-PEG-CM-β-CD to the liver and kidney of mice. As shown in Table 3, except for the 5-FU treatment, the serum biochemical analysis indexes of other groups after treatment were all within the normal range, and no significant difference was observed. After 5-FU treatment, AST and ALT increased significantly compared to the saline group (P < 0.05). For the kidney function markers of BUN and Cr, all data were normal in all treatment groups and showed no statistical difference compared to the saline group (P > 0.05), suggesting that all preparations had no obvious toxic effect on the kidney function of the treated mice.
We also performed histological studies of the heart, liver, kidney, and lung after treatment with the different formulation mentioned above. At least three sections for each mouse from each group (n = 5) were randomly and blindly analyzed by a pathologist who was blinded to the experimental protocol. The experimental results were shown in Fig. 6. First, it was found that no obvious histological changes appeared in the heart, liver, kidney, and lung of mice treated with the different formulation. Secondly, the lungs of nude mice treated by each group showed unequal amounts of inflammatory cell infiltration in the interstitium of lung tissues, which might be related to tumor growth. Lastly, different degree of bleeding was observed in the lung from 5-ASA, NaB, FA-PEG-CM-β-CD, BBA/CM-β-CD treatment groups, which might be related to improper operation during dissection. The heart, liver, or lung, and kidney tissues of all treatment groups showed no pathological injury related to the treatment of this test, which further proved the non-toxicity of BBA/FA-PEG-CM-β-CD to normal mice.
BBA/FA-PEG-CM-β-CD prolonging the in vivo circulation time of BBA
The SW620 tumor-bearing nude mice were given orally with free BBA, BBA/CM-β-CD, and BBA/FA-PEG-CM-β-CD. The blood concentration of butyric acid was determined at different time points. As shown in Fig. 7A, butyric acid showed a rapid release when delivered as free BBA, but sustained release when delivered as BBA/CM-β-CD or BBA/FA-PEG-CM-β-CD, which was consistent with the in vitro release study. The blood-concentration of butyric acid showed no significant difference between BBA/CM-β-CD and BBA/FA-PEG-CM-β-CD, suggesting that the FA-PEG-NH2 modification did not affect the metabolism of BBA/CM-β-CD. Pharmacokinetic parameters (Table 4) indicated that loading BBA into FA-PEG-CM-β-CD could significantly prolong the in vivo circulation time of BBA.
To assess the targeted ability of BBA/FA-PEG-CM-β-CD to tumor tissue, we analyzed the tissue distribution of BBA, BBA/CM-β-CD, or BBA/FA-PEG-CM-β-CD in the SW620 tumor-bearing nude mice after oral administration. As shown in Fig. 7B, in the first 3 h butyric acid concentration in plasma was lower from both BBA/CM-β-CD and BBA/FA-PEG-CM-β-CD than that from BBA (P < 0.0001). But at time points of 8 h (Fig. 7C) and 12 h (Fig. 7D), the blood concentration of butyric acid from BBA/CM-β-CD and BBA/FA-PEG-CM-β-CD was significantly higher than that from BBA (P < 0.0001), which showed that the drug-carrying inclusion compound made by encapsulating BBA with CM-β-CD could significantly prolong the circulation time of BBA. The butyric acid concentration in the heart, spleen, and lung was kept at low levels in all groups. Both BBA/CM-β-CD and BBA/FA-PEG-CM-β-CD were mainly distributed in blood, liver, kidney, and tumor tissues. In tumor tissue, concentrations of both BBA/CM-β-CD and BBA/FA-PEG-CM-β-CD were higher than that of BBA at all time points, in which BBA/FA-PEG-CM-β-CD accumulated significantly more in tumor than BBA/CM-β-CD.
Further, the tissue distribution of BBA/FA-PEG-CM-β-CD was detected by in vivo small animal imaging system (Fig. 8). The fluorescence in tumor tissues was significantly stronger from DiD-BBA/FA-PEG-CM-β-CD than that from DiD or DiD-BBA/CM-β-CD (P < 0.0001). All three treatments showed obvious fluorescence in the liver and kidney in the SW620 tumor-bearing mice. But DiD-BBA/FA-PEG-CM-β-CD group could be observed a lower level of fluorescence in the kidney than the DiD-BBA/CM-β-CD group. The fluorescence intensity in the liver was significantly lower from DiD-BBA/CM-β-CD and DiD-BBA/FA-PEG-CM-β-CD than that from DID.
From these results mentioned above, it could be concluded that the FA-modified CM-β-CD inclusion compound could prolong the half-life of BBA, and this novel drug delivery system showed a certain tumor-targeting effect.
BBA/FA-PEG-CM-β-CD showing tumor suppression to colon cancer
The SW620 tumor-bearing mice were sacrificed by euthanasia on day 25 after tumor implantation. The results presented in Fig. 9 showed the kinetics of the antitumor activities of the different formulation against SW620 xenografts in nude mice. Figure 9A showed that in the BBA/FA-PEG-CM-β-CD-treated group the mean tumor volume increased very slowly, compared to the saline group (P < 0.0001), indicating that BBA/FA-PEG-CM-β-CD had a significant anti-tumor effect. Meanwhile, BBA/FA-PEG-CM-β-CD gave the anti-tumor effect much higher than BBA.
As shown in Fig. 9B, the mean tumor inhibitory rates in nude mice treated with 5-ASA, NaB, BBA, 5-FU, BBA/CM-β-CD and BBA/FA-PEG-CM-β-CD were 31.33% ± 7.17%, 15.30% ± 5.61%, 41.76% ± 7.56%, 85.85% ± 2.89%, 67.20% ± 6.57%, and 78.56% ± 8.48% against SW620 tumors, respectively. Compared with NaB and 5-ASA, BBA enhanced the tumor inhibition effect, which may result from the combination therapy of anti-inflammatory role of 5-ASA and anti-tumor effect of NaB to colon cancer. Moreover, BBA/FA-PEG-CM-β-CD produced a significant decrease in the tumor weight of mice compared to BBA (P < 0.05), which showed great tumor suppression as well as the positive control of 5-FU. These results demonstrated that the delivery system of BBA/FA-PEG-CM-β-CD had a good inhibitory effect on SW620 xenograft tumors while showing no acute toxicity to mice.
BBA/FA-PEG-CM-β-CD inducing tumor cells necrosis and apoptosis
To further confirm the antitumor activity of BBA/FA-PEG-CM-β-CD, pathological analysis of tumor tissues was performed by H&E assay. As shown in Fig. 10, some typical necrosis, such as nuclear fragmentation, shrink, and dissolution, was observed in tumor-bearing mice treated with 5-FU, BBA, BBA/CM-β-CD, and BBA/FA-PEG-CM-β-CD. The average necrosis rates were 20%, 50%, 50%, 30%, 10%, and 15% for BBA, BBA/FA-PEG-CM-β-CD, 5-FU, BBA/CM-β-CD, NaB, and 5-ASA, which was consistent with the antitumor effect, suggesting that BBA/FA-PEG-CM-β-CD could cause tumor cell necrosis, leading to tumor growth inhibition.
For evaluation of the apoptosis rate, we chose multiple fields to calculate the average fluorescence intensity in the TUNEL analysis. Results showed that the apoptosis rates from groups of saline, FA-PEG-CM-β-CD, NaB, 5-ASA, BBA/FA-PEG-CM-β-CD, BBA, BBA/ CM-β-CD and 5-FU were 0.06%, 0.75%, 1.45%, 2.44%, 5.13%, 3.27%, 4.49%, and 8.87%, respectively. From these results, it can be seen that a higher level of apoptosis rate was observed in the tumor of mice treated with BBA/FA-PEG-CM-β-CD which showed obvious anti-tumor activity. It is consistent with the results showed in Fig. 9.
To further determine whether the antitumor effect of BBA/FA-PEG-CM-β-CD was associated with the antiangiogenic effect, we analyzed the endothelial cell marker VEGFR-3 and the cell proliferation marker Ki-67. The VEGFR-3 positive area was significantly reduced in animals treated with NaB, 5-ASA, 5-FU, BBA, BBA/CM-β-CD, and BBA/FA-PEG-CM-β-CD, respectively. A higher degree reduction in vascularity was observed in SW620 xenografts after treatment with 5-FU and BBA/FA-PEG-CM-β-CD. In these sections, the reduced VEGFR-3 staining was accompanied by tumor necrosis. In terms of the Ki-67 detection, a significant decrease of the positive expression was observed in mice treated 5-FU and BBA/FA-PEG-CM-β-CD, indicating that the BBA/FA-PEG-CM-β-CD effectively inhibited tumor cell proliferation in vivo as well as 5-FU.