Objective To analyze the mechanism of acidified silk protein sponge matrices and methanolized silk protein sponge matrices in promoting wound healing. Methods The experimental method was conducted. Acidified silk protein sponge matrices with vascularization ability and methanolized silk protein sponge matrices without vascularization ability were prepared by improved freeze-drying method. General observation was performed. Internal morphology was observed with scanning electron microscope. The secondary structure was observed with X-ray diffractometer (XRD) and infrared spectrometer. Compressive modulus was tested by tensile machine. Two 3-week-old male Sprague-Dawley (SD) rats were used to isolate bone marrow mesenchymal stem cells (BMSCs) cultured in above-mentioned two silk protein sponge matrices, the number of cells was counted under laser scanning confocal microscope after 1, 6 days of culture. Four full-thickness skin defect wounds were made on each one of twelve 8-week-old male SD rats, which were divided into methanolized silk group (24 wounds) and acidified silk group (24 wounds) covered with the corresponding silk protein sponge matrices. On post operation day (POD) 3, 7, 10, and 14, general observation was performed and the remaining wound area was recorded. On POD 3, 7, and 14, the wounds and marginal tissue were collected for hematoxylin-eosin staining (HE) staining and Masson staining to observe growth of new tissue and collagen deposition and CD34 immunohistochemical staining to observe vascularization. Sample number of each index of each group at every time point in animal experiment was 6. Data were statistically analyzed with analysis of variance of factorial design, analysis of variance for repeated measurement, independent-samples

test, and Bonferroni correction. Results Methanolized silk protein sponge matrices and acidified silk protein sponge matrices had the same composition and similar porous structure, with pore size of 300-500 μm. XRD showed that methanolized silk protein sponge matrices showed a significant crystallization peak, while acidified silk protein sponge matrices was mainly composed of amorphous structure. Infrared spectrometer showed that acidified silk protein sponge matrices appeared a strong absorption peak at 1 650 cm^-1, and the methanolized silk protein sponge matrices appeared a strong absorption peak at 1 630 cm^-1. Compressive modulus of methanolized silk protein sponge matrices was (23.8±1.3) kPa, which was significantly higher than (6.1±0.9) kPa of acidified silk protein sponge matrices (

=19.550,

<0.01). After one day of culture, BMSCs successfully adhered to the two kinds of silk protein sponge matrices, and the cells were not spread. After six days of culture, BMSCs were spread on the two kinds of silk protein sponge matrices, and the number of cells on the acidified silk protein sponge matrices increased significantly. On POD 3, the wounds of the 2 groups did not shrink significantly. On POD 7, the wound area in acidified silk group was significantly smaller than that in methanolized silk group, and new epithelium growth occurred at the wound edge. On POD 14, the wounds of acidified silk group basically healed, and the wounds of methanolized silk group were dry and shrinked significantly. Remaining wound area of acidified silk group on POD 3, 7, 10, and 14 were significantly smaller compared with that in methanolized silk group (

=7.782, 10.620, 3.707, 6.830,

<0.05 or

<0.01). HE staining, Masson staining, and CD34 immunohistochemical staining showed on POD 3, new tissue growing into silk protein sponge matrices of wounds of acidified silk group was more than that in methanolized silk group, the former group secreted a small amount of collagen, collagen formation was not observed in the latter group, the number of vascular endothelial cells migrated into the matrices were more in the former group than the latter group; on POD 7, the area of new tissue covering matrices hole of wounds of acidified silk group was larger than that in methanolized silk group, collagen in the former group was more than that in the latter group and was evenly distributed, the number of blood vessels in the former group was more than that on POD 3, and the new blood vessels in the latter group were scattered; on POD 14, the new tissue in acidified silk group was similar in structure to normal skin tissue and formed a certain thickness, the new tissue in methanolized silk group basically grew into the matrices, the former group had rich collagen deposition, the latter group had scattered collagen, and blood vessels in the former group distributed uniformly and density of blood vessels was significantly higher than that in the latter group ((55.7±6.0) and (34.1±1.0) pieces/mm², respectively,

=9.042,

<0.01). Conclusions Angiogenesis-promoting acidified silk protein sponge matrices have good cytocompatibility, which can facilitate the rapid formation of vascular network in wound area, providing sufficient blood supply to accelerate the tissue regeneration and collagen deposition, thereby promoting wound healing and improving healing quality, these effects are better than methanolized silk protein sponge matrices.