IR and 1H-NMR
PA-PSH was synthesized as reported (Fig. S1) 19,20. The crosslink of PA and PSH was confirmed by FT-IR, which the C = O stretching vibration peak of the acrylate group was appeared at 1732 cm− 1 (Fig. S2). Furthermore, the chemical structures of PA and PA-PSH conjugates were confirmed by 1H-NMR. 1H-NMR spectrum of PA in Fig. 2a,b clearly showed increased signals at 1.2 ppm and 6.0-6.5 ppm comparing with the spectrum of Poloxamer 407 (P407), which represented the protons from acryl and propylene oxide units.
The synthesized PSH was further confirmed by FT-IR (Fig. S2), the C = O stretching vibration of terminal thio-propionyl groups in PSH was appeared at 1736 cm− 1. For disulfide crosslinked P407 (PSS), three peaks appeared on the GPC curve at 21.19, 22.23 and 23.84 min. (Fig. S3). For PSH, the shoulder peak at 21.29 min disappeared, and for PA-PSH mixture, the peak at 24.01 min broadened largely and a remarkable shoulder peak at 25.34 min appeared, which resulted from products with larger molecular weights had formed based on the thiol-ene reaction 20. All these results suggested the successful synthesis of PA-PSH.
Sol–gel transition
Figure 2c,d presented the sol–gel transition of PSH in a simulated physiological environment. At 25 °C, pH 3.5, PA-PSH solution was a transparent solution (Fig. 2c), while at 37 °C, pH 7.4, the copolymer solution turned into a transparent gel (Fig. 2d).
Submucosal injection of in-situ gel in resected porcine stomachs
The feasibility of in-situ gel was assessed in fresh resected porcine stomachs in different concentrations, with PBS and glycerol as controls. The results were shown in Fig. 3a. Although adequate mucosal elevations occurred in each group, the in-situ gel created SFCs were more durable than that of PBS or glycerol. Moreover, no significant change in size, shape or consistency of SFCs under the submucosa elevated by the in-situ gel was observed over 1 h. In contrast, the elevation created with PBS and glycerol collapsed quickly in 15 min. In short, the ability of mucosal elevation and maintenance exhibited a sequence in-situ gel > glycerol > PBS.
Submucosal injection of in-situ gel in living minipigs
The in vivo efficacy in living minipigs were further evaluated, as shown in Fig. 3b. The mucosa was lifted by different agents immediately after injection. However, the SFCs created by glycerol collapsed obviously after 15 min injection and disappeared at 30 min. In contrast, the SFCs of the in-situ gel remained unchanged in 30 min. Even after 9 days, the elevation of the in-situ gel still presented an apparently visible edge at the injection site without obvious ischemia or ulceration, while the injection sites of the controls was hard to find (data not shown).
PA-PSH-assisted ESD in living minipigs
As shown in Fig. 4a,b, the procedure of ESD was performed with the assistance of the methylene blue labled in-situ gel, it could be accomplished in early stage with the guidance of the mucosal elevation created by the in-situ gel, with no significant side effects like major bleeding and perforation. All of the procedures were successful and all the minipigs were sacrificed on 9th day after the ESD procedure, and its stomach was dissected with an ulcer displayed at early healing stage.
Biosafety of PA-PSH in-situ gel
The in vivo biodegradation of PA-PSH in-situ gel was examined in minipigs (Fig. 4c). An obvious elliptic protrusion appeared at the injection site after injection and the integrity of the mucosal elevation remained at least 9 days. The height and size of the protrusion decreased due to the degradation of the matrix in vivo. The degradation of the in-situ gel was attributed the hydrolysis of ester bonds in polyoxypropylene segments, and the final degradation products were biocompatible 21. The gel protrusion was clearly observed at 0th and 9th day, separately (Fig. 4c). Additionally, after injection of in-situ gel solution, neither apparent epithelial damage nor inflammation cells were observed in the injection site (Fig. 4d). After 9 days of injection, granulation tissues were appeared on the postoperative ulcer bed and the edge of the ulcer was re-epithelialized (Fig. 4e). All these results confirmed the considerable biocompatibility of the PA-PSH in-situ gel as a submucosal injection agent.