a The effect of E. piscicida infection on the expression of CD44a variants in zebrafish larvae. b, c The antibacterial effects of zebrafish CD44a variants at 6 and 24 hpi. d Larval survival analysis in the WT zebrafish microinfected with FLAG, CD44a_tv1-FLAG or CD44a_tv2-FLAG (n = 60 for each group). e Cartoon showing the position of the target site and its sequence in the zebrafish CD44a locus. f Representative Sanger sequencing results of the PCR amplicons from homozygous mutations with 14 bp deletions (CD44a^−/−-14del), 10 bp insertions (CD44a^−/−-10IS) or WT. g The expression of CD44a variants in the WT and CD44a^−/−-14del zebrafish larvae. h Zebrafish larvae from the WT and CD44a^−/−-14del mutants were collected at the indicated post-infection time points, and homogenates were made for CFU counts. i Zebrafish larvae from the WT and CD44a^−/−-10IS mutants were collected at the indicated post-infection time points, and homogenates were made for CFU counts. j Larval survival analysis in the WT and CD44a^−/−-14del zebrafish larvae infected with E. piscicida (n = 120 for each group). k Larval survival analysis in the WT and CD44a^−/−-10IS zebrafish larvae infected with E. piscicida (n = 120 for each group). For b, c, h and i, each symbol represents the average counts of ten larvae. For a–c and g–i, the data are presented as means ± SD (n = 3). **p < 0.01.

a The quantification analysis of mRNA levels of key genes involved in the NOD1-RIPK2 signaling pathway in the WT, CD44a^−/−-14del and CD44a^−/−-10IS zebrafish larvae, as determined by qRT-PCR. Quantification data represent the expression level of genes compared with those in the WT zebrafish. b Zebrafish larvae from the WT and NOD1^−/− mutants microinjected with p3×FLAG or CD44a_tv1-FLAG were collected at the indicated post-infection time points, and homogenates were made for CFU counts. c Zebrafish larvae from the WT and NOD1^−/− mutants microinjected with p3×FLAG or CD44a_tv2-FLAG were collected at the indicated post-infection time points, and homogenates were made for CFU counts. d Zebrafish larvae from the WT and CD44a^−/−-14del mutants microinjected with p3×FLAG or NOD1-FLAG were collected at the indicated post-infection time points, and homogenates were made for CFU counts. For a–d, the data are presented as means ± SD (n = 3). *p < 0.05, **p < 0.01; ns not significant.

a The numbers of differentially expressed genes and isoforms at 24 hpi, 48 hpi and 7 dpf from the WT and CD44a^−/−-14del zebrafish larvae with or without the infection of E. piscicida. b KEGG enrichment analysis of the down-regulated DEGs at 24 hpi, 48 hpi and 7 dpf from the WT and CD44a^−/−-14del zebrafish larvae with or without the infection of E. piscicida. c KEGG enrichment analysis of the up-regulated DEGs at 24 hpi, 48 hpi and 7 dpf from the WT and CD44a^−/−-14del zebrafish larvae with or without the infection of E. piscicida.

a DEGs involved in autophagy at 24 hpi by RNAseq analysis of mRNA expression levels in the WT and CD44a^−/−-14del zebrafish larvae infected with E. piscicida. b Validation of DEGs involved in autophagy at 24 hpi by qRT-PCR. c Immunoblotting analysis and quantification of p62 and LC3b levels in the WT and CD44a^−/−-14del zebrafish larvae infected with E. piscicida collected at 24 hpi. WB were repeated at least three times, and shown were the representative data. d mRNA levels of CD44a and autophagy-related genes in the WT and CD44a^−/−-14del zebrafish larvae without or with the treatment of the autophagy inhibitor chloroquine (CQ). e Zebrafish larvae from the WT and CD44a^−/−-14del mutants without or with the treatment of CQ were collected at the indicated post-infection time points, and homogenates were made for CFU counts. f Larval survival analysis in the WT and CD44a^−/−-14del zebrafish larvae without or with the treatment of CQ (n = 60 for each group). g Zebrafish larvae microinjected with FLAG, CD44a_tv1 or CD44a_tv2 without or with the treatment of rapamycin or zebrafish larvae from the WT and CD44a^−/−-14del mutants without or with the treatment of rapamycin were collected at the indicated post-infection time points, and homogenates were made for CFU counts. h Larval survival analysis for zebrafish larvae microinjected with FLAG, CD44a_tv1 or CD44a_tv2 without or with the treatment of rapamycin (n = 60 for each group). i Larval survival analysis for zebrafish larvae from the WT and CD44a^−/−-14del mutants without or with the treatment of rapamycin (n = 60 for each group). Data are presented as mean values ± SD (n = 3) and p values by Student’s t test are shown in (b–e, g). *p < 0.05, **p < 0.01; ns not significant.

a DEGs involved in p53 signaling pathway at 24 hpi by RNAseq analysis of mRNA expression levels in the WT and CD44a^−/−-14del zebrafish larvae infected with E. piscicida. b Validation of DEGs involved in p53 signaling pathway at 24 hpi by qRT-PCR. c Cell viability analysis of the WT and CD44a^−/−-14del cells in the absence of infection. The cells were collected at 12, 24, 36 and 48 h. d, e Cell viability analysis of the WT and CD44a^−/−-14del cells with or without the overexpression of p53. The cells were collected at 24 h (d) and 48 h (e). f Cell viability analysis of the WT and CD44a^−/−-14del cells infected with E. piscicida. The cells were collected at 6, 12 and 24 hpi. g–i Cell viability analysis of the WT and CD44a^−/−-14del cells with or without the overexpression of p53 following the E. piscicida infection. The cells were collected at 6 hpi (g), 12 hpi (h) and 24 hpi (i). j mRNA levels of CD44a and many genes involved in p53 signaling pathway in the WT and CD44a^−/−-14del zebrafish with or without the overexpression of p53 following the E. piscicida infection. The cells were collected at 24 hpi. Data are presented as mean values ± SD (n = 3) and p values by Student’s t test are shown in b–j. *p < 0.05, **p < 0.01; ns not significant.

The effects of zebrafish CD44a_tv1 (a) and CD44a_tv2 (b) on the expression of total p53 proteins. The effects of zebrafish CD44a_tv1 (c) and CD44a_tv2 (d) on the expression of cytoplasmic p53 proteins. The effects of zebrafish CD44a_tv1 (e) and CD44a_tv2 (f) on the expression of nuclear p53 proteins. g The interaction between zebrafish CD44a variants and p53. For a–g, WB were repeated at least two times, and shown were the representative data. h The colocalization between zebrafish CD44a variants and p53 in the CD44a^−/−-14del cells following the E. piscicida infection. Scale bar, 10 µm.

a DEGs involved in apoptosis at 24 hpi by RNAseq analysis of mRNA expression levels in the WT and CD44a^−/−-14del zebrafish larvae infected with E. piscicida. b Validation of DEGs involved in apoptosis at 24 hpi by qRT-PCR. c The effect of zebrafish CD44a deficiency on the early and late apoptosis rates at 6 hpi. d mRNA levels of many genes involved in apoptosis in the WT and CD44a^−/−-14del zebrafish with or without the overexpression of p53 following the E. piscicida infection. The larvae were collected at 24 hpi. e, f Zebrafish larvae microinjected with FLAG, CD44a_tv1 or CD44a_tv2 without or with the treatment of Z-IETD-FMK were collected at the indicated post-infection time points, and homogenates were made for CFU counts. g Larval survival analysis in the WT zebrafish larvae microinjected with FLAG, CD44a_tv1 or CD44a_tv2 without or with the treatment of Z-IETD-FMK (n = 60 for each group). h The effect of Z-IETD-FMK on the CD44a_tv1-mediated apoptosis rates at 6 hpi. i The effect of zebrafish CD44a variants on the casp8 activity. j The effect of pifithrin-μ on the early apoptosis regulated by zebrafish CD44a variants at 6 hpi. k The effect of pifithrin-μ on the late apoptosis regulated by zebrafish CD44a variants at 6 hpi. l The effect of pifithrin-μ on the CD44a_tv1-mediated antibacterial activity. m The effect of pifithrin-μ on the CD44a_tv1-mediated larval survival rate (n = 60 for each group). Data are presented as mean values ± SD (n = 3) and p values by Student’s t test are shown in b–f, h–l. *p < 0.05, **p < 0.01; ns not significant.

a The 5′ flanking regulatory sequence of zebrafish p53. The transcription factor binding sites were underlined. b The relative luciferase activity of p53-pGL3 promoter. c The effects of zebrafish CD44a variants on the promoter activity of p53 without or with the infection of E. piscicida. d The effects of zebrafish CD44a variants on the transcription of p53 in the CD44a^−/−-14del cells. e Zebrafish larvae microinjected with FLAG, CD44a_tv1 or CD44a_tv2 without or with the treatment of pifithrin-μ were collected at 48 hpi, and homogenates were made for CFU counts. f Larval survival analysis in the WT zebrafish larvae microinjected with FLAG, CD44a_tv1 or CD44a_tv2 without or with the treatment of pifithrin-μ (n = 60 for each group). g Immunoblotting analysis and quantification of p62 and LC3b levels in the WT zebrafish microinjected with FLAG, CD44a_tv1 or CD44a_tv2 without or with the treatment of pifithrin-μ. WB were repeated at least two times, and shown were the representative data. Data are presented as mean values ± SD (n = 3) and p values by Student’s t test are shown in b–e. *p < 0.05, **p < 0.01; ns not significant.

CD44a deficiency regulates the expression levels of many genes involved in p53 signaling, apoptosis and autophagy, which leads to inhibition of apoptosis and induction of autophagy in the case of E. piscicida infection. In zebrafish, two CD44a variants including CD44a_tv1 and CD44a_tv2 are identified. The zebrafish CD44 variants exert protective effects against E. piscicida infection. Both zebrafish CD44a_tv1 and CD44a_tv2 can promote the nucleoplasmic translocation of p53 and interact with p53 in the cytoplasm. The antibacterial effect of zebrafish CD44a_tv1 is depending on the CASP8-mediated apoptosis, whereas the antibacterial effect of zebrafish CD44a_tv2 is dependent of the cytoplasmic p53-mediated inhibition of autophagy.