Fig. 4: In vitro cytocompatibility, barrier function, and pro-osteogenic ability.

a Absorbances of the CCK-8 assay exhibiting the proliferation of L929s cultured on the dense layer and BMSCs cultured on the loose layer have no significant difference between CM and BC-g-PNCl/CS-HAP after 1, 3, and 7 days (n = 3 independent samples; Student’s t test; two-tailed P = 0.088, 0.613, 0.056 in dense layer group and 0.938, 0.081, 0.294 in loose layer group for 1, 3 and 7 days compared with CM, respectively; error bars = SD; data are presented as mean values ± SD). b Schematic diagram showing the barrier function evaluation via transwell assay. c Fluorescent images of penetrated cells on the bottom surfaces and penetration depths after 7 days for CM and BC-g-PNCl/CS-HAP, exhibiting their barrier functions against L929s before and after simulated clinical immersion for 4 weeks (blue for cellular nucleus, scale bars = 200 μm). d ALP activities of BMSCs cultured on the loose layers of samples for 3, 7, and 14 days. e Alizarin Red S staining of BMSCs cultured on the loose layers of samples for 21 days (red for calcium nodules). f–i RT-qPCR results of expression levels of osteogenic-related genes (ALP, RUNX2, Col1, and OCN) of BMSCs cultured on the loose layers of samples for 3, 7, and 14 days after osteogenic induction. d–i n = 3 independent samples; ANOVA followed by Tukey’s multiple comparisons; *adjusted P < 0.05, **adjusted P < 0.01, ***adjusted P < 0.001, ****adjusted P < 0.0001; error bars = SD; data are presented as mean values ± SD.