3.1 Preparation and characterization of Gel conjugation
As shown in Fig. 1a, hydrogel was obtained by a one-step reaction among OD, chitosan and NH2-TPP, in which the OD-to-NH2-TPP ratio was 1:10, and chitosan (W%=10%) was to adjust the elasticity. As shown in Fig. 1b, the disappearance of characteristic IR absorbance peaks from oxidized dextran (at ~ 1712 cm− 1 for CO-H) and porphyrin derivative (~ 1510 cm− 1 for N-H), and presence of characteristic IR peak from Gel (~ 1650 cm− 1 for C = N) confirm the successful reaction between OD and TPP. The gel-forming behavior under physiological conditions was also evaluated by measuring the storage modulus (G') and loss modulus (G''), where the sol-gel conversion condition could be determined. As shown in Fig. 1c and Fig. 1d, the conversation occurred at ~ 36.5°C for Gel, and ~ 36.8°C for DOX-CA4P@Gel, which indicated that loading of CA4P and DOX did not change the gelation process of TPP, chitosan, and OD. Gel could form within 8.8 min, and DOX-CA4P@Gel formed within 8.5 min. Results showed that the hydrogel can be synthesized quickly at a temperature close to physiological temperature, indicating its injectability in vivo.
As shown in Fig. 2a, DOX was loaded into hMSN to form DOX@hMSN, and then DOX@hMSN was enveloped by OD/TPP together with CA4P to form dual drugs loaded DOX-CA4P@Gel. The morphology and size distribution histogram of hMSN was shown in Fig. 2b. The average diameter of hMSN was 120 nm from TEM assay, smaller than that measured from DLS for hydration effects. Figure 2c and Fig. 2d was SEM images of the lyophilized cross-section of the hydrogel. Results showed that the hydrogel possessed a three-dimensional network structure with a uniform pore size of about 5 µm. Figure 2e-g showed the three-dimensional network structure of the freeze-dried cross-section of hydrogel. The results showed that DOX@hMSN did not affect the morphology of the hydrogel network structure, and the bulges in the three-dimensional structure indicated the locations of hMSN. The surface chemical compositions of the hydrogel layers from XPS also confirmed it. Comparing the X-ray photoelectron spectrum of Gel with that of DOX-CA4P@Gel (Figure S1), the peaks seen at 154.1 eV and 102.5 eV indicate that there were additional Si elements on the surface of Gel, which were resourced from hMSN. According to the absorbance of DOX at 510 nm, the drug loading capacity of DOX in hMSN was 633 mg/g, and 52 mg/g for DOX in hydrogel. The drug loading of CA4P in hydrogel was 12.6 mg/g calculating from HPLC measuring.
Drug sequential release from DOX-CA4P@Gel
DOX-CA4P@Gel was dispersed in phosphate buffer solution (PBS) solution with different pH values, and the elastic modulus of hydrogel was measured using a rheometer. As shown in Fig. 3a, Fig. 3b and Fig. 3c, the energy storage modulus (G'=34,000) of hydrogels at pH of 5.0 was significantly higher than that in pH7.2 (G'=17,000), which proved that the hardness of hydrogels under acidic conditions was more than that in neutral condition, and the hydrogels were more prone to fracture under the same shear force. This result was consistent with the corresponding crushing time (tpH5.0=5.0 min, tpH6.4=5.2 min, tpH7.2=6.4 min). DOX-CA4P@Gel was dispersed in PBS with different pH values, and the degradation of hydrogels was evaluated by measuring the content of decomposed TPP in solution. Results in Fig. 3d showed that the degradation rate of hydrogels was very slow in neutral condition (about 20% at 144 h), while that was significantly increased in weakly acid condition, reaching 65% at 144 h in pH of 6.4. Degradation of hydrogel at pH 5.0 was faster and more than 90% TPP was detected at 144 h. This result testified that pH sensitive degradation of DOX-CA4P@Gel, indicating the application poetical as a tumor-selective drug delivery. The images before and after hydrogel degradation were shown in Fig. 3e and Fig. 3f.
CA4P and DOX exhibited pH sensitive and the release profiles were also observed, where faster release rates were shown at acidic condition (pH 5.0) compared with that in neutral condition (Fig. 3h and Fig. 3i). CA4P and DOX have different release behavior. As shown in Fig. 3j, CA4P was rapidly released at the early stage and relatively stable after 48 h, while DOX released slowly firstly and then go fast after 48 h. 59.13% of CA4P was released from Gel at pH of 6.4 for 48 h, and reached to 63.44% at 144 h (only 5.31% higher than that of 48 h). DOX was intrinsically pH-sensitive. As shown in Fig. 3g, 68.32% of DOX was release at pH of 5.0 for 144 h but only 16.02% released in a buffer with pH 7.4. This significant release characteristics of CA4P and DOX form Gel were shown in Fig. 3j. In the early stage, the release of CA4P was rapid, reaching 71.08% at 48 h, and slowed down in the later stage, reaching only 78.20% at 144 h. As a contrast, DOX released only 14.39% at 48 h, and increased to 61.60% at 144 h.
Photodynamic performance of Gel
In order to prove that the photosensitivity of porphyrin was not changed significantly after the incorporation into the Gel, DPBF was used as the reagent to quantitatively detect the production of singlet oxygen (1O2) in solution. The results of the irradiation of the samples in the presence of DPBF (absorption monitored at 410 nm) with time are shown in Fig. 4a, where the decrease in DPBF absorption over time for Gel compared to TPP, and the decrease was positively correlated with the Gel concentrations (as shown in Fig. 4b). With TPP (φ∆(TPP) = 0.52) as the control [31], the 1O2 yield of TPP@Gel (808 nm, 0.5 W/cm2) was 0.91. After Gel coating, the 1O2 yield of TPP was significantly increased, which may be due to the high viscosity of hydrogel limiting the rotation of porphyrin molecules and avoiding the optical quenching induced by the collision between porphyrin molecules. DCFH-DA was an intracellular singlet oxygen capture agent, and its green fluorescence intensity of cells was directly proportional to the production of 1O2. After incubation of Gel with cells, 1O2 detection was conducted. As shown in Fig. 4c, Gel showed significant green fluorescence, and there was no significant difference in fluorescence intensity between TPP and Gel.
The extracts of blank Gel with different concentrations (0.05, 0.1, 0.15 and 0.2 mg/mL) were incubated with L929 and 4T1 cells for 24, 48 and 72 h, respectively. As shown in Fig. 5a and Fig. 5b, the cellular viabilities of L929 and 4T1 cells were above 85%, indicating that the Gel had no obvious toxicity to cells without laser irradiation. However, the cellular viability significantly decreased after laser treatment (NIR = 808 nm, 0.5 W/cm2, t = 5 min). This result shown in Fig. 5c was particularly significant change at high concentration, that is, at 0.2 mg/ml, the cell viability of the Gel group without laser assistance was 92.43%, while that of with laser group was reduced to 65.31%, with a significant difference (p < 0.05), suggesting that hydrogel has an obvious photodynamic killing effect on 4T1 cells. In order to further intuitively demonstrate the PDT effect of Gel on cancer cells, the live/dead cells staining with calcein AM and PI was carried out. The fluorescence imaging results in Fig. 5d indicated that the proportion of dead cells increased significantly with the elevated concentrations of the Gel extract. Moreover, with the increase of irradiation time (showed in Figure S2), the proportion of dead cells also increased significantly, indicating that the photodynamic treatment effect of hydrogel was obvious concentration and time dependent.
In vitro chemo-photodynamic therapy effect of DOX-CA4P@Gel
Gel, DOX@Gel, CA4P@Gel, and DOX-CA4P@Gel (0.1 mg/ml) were respectively incubated with 4T1 cells for 48 h, and the cell viabilities were measured with or without laser irradiation (808 nm, 0.5 W/cm2, 5 min). PBS was used as the control. As showed in Figure S3, the cell viability of both Gel and DOX@Gel groups were above 90%, indicating that no obvious DOX was released during this period. In contrast, the cell viability of both CA4P@Gel and DOX-CA4P@Gel groups were lower than 70%. Furthermore, cell viability in CA4P@Gel significantly decreased with time lasting to 48 h, indicating that CA4P could be continuously released within 48 h. With laser irradiation, all the groups including Gel + NIR, DOX@Gel + NIR, CA4P@Gel + NIR and DOX-CA4P@Gel + NIR, exhibited enhanced cell killing effect. Especially, the cell activity of DOX-CA4P@Gel + NIR group was lower than 40%, which verified that DOX-CA4P@Gel has significant combinational effect of photodynamic/sequential chemotherapy.
This result was also verified in a flow cytometry assay. As shown in Fig. 6, without laser irradiation, the cell viability (95.5%) of DOX@Gel was not significantly different from that of Gel (95.8%), while CA4P@Gel and DOX-CA4P@Gel decreased to 79.5% and 66.6%, respectively. This is due to DOX-CA4P@Gel mainly showed the release of CA4P within 24 h, while DOX did not show obvious tumor killing effect because of its late release. Furthermore, after the same sample treatment, the cell viability with laser irradiation was significantly lower than that of the non-irradiated group, especially the cell viability of DOX-CA4P@Gel + NIR group decreased from 66.6–53.3%, reflecting obvious combined treatment effect.
Comparing the cell viability of groups (longitudinal groups) without laser assistance, results showed that after 24 h of co-incubation with 4T1 cells, only CA4P@Gel (79.5%) and DOX-CA4P@Gel (66.6%) was significantly lower than that of control group (95.3%), while DOX@Gel group was not significantly different from the control group, which proved that early release of CAP4P from DOX-CA4P@Gel was the major reason for cell toxicity before 24 h. Comparing the Gel groups (lateral groups), it was found that the cell viability with laser assistance was significantly lower than that of the without laser assistance group, especially DOX-CA4P@Gel + NIR was only 50%.
Combinational therapy of DOX-CA4P@Gel
The 4T1 tumor-bearing mice were respectively treated by a single intratumor injection administration of PBS, Free DOX/CA4P (mixture of 0.4 mg DOX and 1 mg CA4P), Gel, CA4P@Gel, DOX@Gel, DOX-CA4P@Gel, Gel + NIR, CA4P@Gel + NIR, DOX@Gel + NIR, DOX-CA4P@Gel + NIR, where NIR was illuminated for 5 min every other day. The body weight, tumor volume and weight of the mice were recorded every day, and the tumor inhibition rate was calculated with PBS as the control group. As can be seen from Fig. 7a, 7b, and 7c, the tumor in the CA4P@Gel group grew slowly in the first 6 days and then accelerated, while the tumor in the DOX@Gel group grew rapidly in the first 6 days but slowed down significantly within 6–12 days, which may be explained by the sequential release of CA4P and DOX. With the assistance of laser, the comparison of treatment results between the DOX-CA4P@Gel + NIR group and other groups showed that the tumors in the DOX-CA4P@Gel + NIR group were the smallest and significantly different from those in other groups (p < 0.01), indicating that sequential delivery therapy combined with photodynamic therapy had a significant impact on tumor growth. It should be noted that after 18 days of treatment, tumor growth was accelerated in the DOX/CA4P group, with tumor volume greater than that of CA4P-DOX@Gel, which may be due to the fact that CA4P stimulated the formation of tumor vascular collateral after a single dose, thus accelerating tumor growth. In the DOX-CA4P@Gel system, CA4P plays its role first, followed by DOX and PDT. Sequential drug release combined with photodynamic therapy can effectively inhibit tumor growth to a certain extent. The therapeutic effect of each group was also evaluated by measuring tumor weight, and the results were shown in Fig. 7d. The results showed that the tumor weight of the DOX-CA4P@Gel + NIR group was the lowest, which was consistent with that of tumor volume growth curves. As showed in Fig. 7e, the calculated tumor inhibition rate was about 70% for DOX-CA4P@Gel + NIR group, which was significantly different from other groups (p < 0.01).
The expression of caspase 3 and DOX fluorescence intensity were used to characterize the location of CA4P and DOX. CA4P can specifically bind to vascular growth factor and prevent its interaction with the receptor, thereby causing apoptosis. The activity of caspase 3 reflects the apoptotic status of tumor cells. At different time points, the caspase 3 activity (apoptotic inducing factor) in the tumors of the DOX-CA4P@Gel group was detected by immunohistochemistry. Results in Fig. 8 showed that compared with that in control group, the caspase 3 expression was significantly up-regulated, reaching the peak on the 3rd day, and gradually went down in the subsequent treatment period. DOX release and accumulation in tumor sections of the DOX-CA4P@Gel group at different time points (3, 6, 10, 14 days) were measured by the fluorescence properties of DOX. The results showed that DOX fluorescence in tumor tissues was extremely weak at the initial stage of treatment (within 2 days), suggesting DOX-CA4P@Gel accumulating in the tumor. With the extension of time (at the 6th days), it retained in tumor for more than 14 d.
In vivo safety evaluation
To evaluate the possible systemic toxicity, body weigh variation and H&E staining was carried out. Figure S4 showed that there was no significant statistical difference in the body weight of mice between the control group and each treatment group, indicating that the therapeutic agent had no significant toxicity in vivo. H&E staining of the representative tissue sections, including heart, liver, spleen, lung and kidney was carried out for the mice treated with PBS and different Gel conjugates at day 21. As shown in Figure S5, no observable systemic toxicity was noted in all the organs of the tumor-bearing mice treated with the drug-loaded hydrogel groups.