Fig 2: CDNF mediated Nrf2/HO-1 axis activation is implicated in decreasing UPR marker expression and tempering GSK3β activation.A Representative Western blot images of nuclear Nrf2, Histone 3, Ho-1, GAPDH in the striatum and the area of hemorrhagic lesion at different time points. B, C The protein expression levels of Ho-1 and Nrf2 in the nucleus were detected in the striatum of indicated groups of rats receiving sham + saline, ICH + saline, or ICH + rhCDNF. Increased levels of nuclear Nrf2 and Ho-1 were shown in the rhCDNF-treated group at 6 h and 24 h post-ICH. D Representative Western blot from tissue lysates of intact or hemorrhagic striatum probed with antibodies to UPR markers, p-GSK-3β (S9), p-GSK-3β (Y216), and GSK-3β. Β-actin was used as an internal marker. E–J Data points show protein expression levels of GRP78, Atf4, Atf6, and Chop in the naïve or hemorrhagic striatum of indicated groups of rats at 6 h and 24 h post-ICH. Expression of p-GSK-3β (S9) (I), and p-GSK-3β (Y216) (J) were analyzed by immunoblotting quantified by densitometry and normalized to expression of GSK-3β. Results are presented as fold change increase compared to sham plus saline control. Mean± SEM, n = 3–4/sham group, n = 4–8/ group of ICH plus saline or ICH plus rhCDNF. Two-way ANOVA + post-hoc Bonferroni test, *p < 0.05; **p < 0.01; ***p < 0.0001 in comparison to the ICH + saline group.

Fig 3: CDNF treatment enhanced expression of scavenger receptors on activated microglia/macrophages at the early stage of ICH.A–D Figures show immunofluorescent staining of cells in hemorrhagic striatum with anti-CD11b (green), anti-CD163 (red) and counterstained with DAPI (blue). The insets show the hematoma edge in higher-magnification photomicrographs immunostained with CD11b and CD163 antibodies. Immunofluorescence pictures showing co-localization of CD163 and CD11b in the saline-treated perihematomal area at 1 d (A) and 3 d (B), respectively. C–E rhCDNF injections increased CD163 immunostaining in the peri-hematoma area at 1 d, but not at 3 d post-ICH, compared with saline group. **p < 0.01 Bonferroni’s multiple comparisons test, following two-way ANOVA [effect of treatment: F(1,32) = 5.026, P = 0.03]. F–J WB bands of CD163, CD36, CD91, and GAPDH at different time points post-ICH. G Increased the level of CD163 in the rhCDNF-treated hemorrhagic striatum at 1 d but not at 3 d post-ICH. **P < 0.001, Bonferroni’s multiple comparisons test, following two-way ANOVA [effect of treatment: F(1,12) = 44.52, P < 0.001]. H, I Increased levels of cell surface CD36 and intracellular CD36 were detected in rhCDNF-treated hemorrhagic striatum at 1 d post-ICH. *p < 0.05, **p < 0.01, Bonferroni’s multiple comparisons test, following two-way ANOVA [effect of treatment: F(1,18) = 30.2, P < 0.001. J rhCDNF treatment did not increase the level of CD91 in the hemorrhagic striatum, compared to a saline-treated group post-ICH. The mean ± SEM of three independent experiments is shown. Scale bars: A, B: 1000 µl, C, D: 50 µl.

Fig 4: Endogenous CDNF affects the hemorrhagic lesion after ICH.A Photograph of representative films demonstrating temporal changes in CDNF protein in the naive striatum and ICH-affected striatum, at 3 h to 7 days post-ICH in SD rats, which were assessed using Western blotting. B Bar graph showing the relative levels of CDNF protein in the striatum of naïve rats and rats at 3 h, 6 h, 24 h, 72 h, and 7 days after ICH. Data were analyzed as repeated measures by one-way ANOVA followed by Bonferroni corrections (n = 4/time point). C Bar graph showing time course of CDNF mRNA levels in hemorrhagic striatum at 3 h to 7 days after ICH in SD rats. Data were analyzed as repeated measures by one-way ANOVA followed by Bonferroni corrections (n = 3/time point). D Representative coronal sections (1 mm thickness) showing brain hemorrhagic areas of WT and Cdnf−/− mice killed 3 days after ICH. E Lesion volume on days 3 (n = 7–8, each group) post-ICH was determined by morphometric measurement. Data were analyzed as two-tailed Student’s t-test. F Volcano plot of gene expression profiles in hemorrhagic striatum collected after collagenase-induced ICH in WT and Cdnf-/- mice, showing distribution of significance [−log10(adjusted P value)] vs. fold change [log2(fold change)] for all genes. The blue dots indicate downregulated genes (fold change < −1.5, adjusted P val <0.05), the red dots indicate upregulated genes (fold change >1.5, adjusted P val <0.05), and the black dots indicate genes with no significant change post-ICH. *P < 0.05 by multiple comparisons using the Holm-Šidák method. Mean ± SEM is shown. Scale bars: 5 mm.

Fig 5: CDNF directly increases RBC engulfment by microglia and upregulates scavenger receptor expression on BV2 cell lines.A Representative immunofluorescence images show engulfment of PKH-26-labeled erythrocytes (red) in microglia (Iba-1; green) after erythrocyte incubation for 6 h. BV2 cells were incubated with saline or rhCDNF for 1 h before treatment with erythrocytes. Scale bar: 100 μm. B WB bands of CD163, CD36, and HO-1 in BV2 cell lines after incubation with erythrocytes and different concentrations of rhCDNF treatment for 6 h. C Quantification of erythrophagocytosis. n = 3 independent experiments/group; each experiment includes 3 technical replicates. **p < 0.01 vs. RBC + vehicle. Data were analyzed as repeated measures by one way ANOVA followed by Bonferroni correction. D Quantitative analysis of CD163, CD36, and HO-1 expression in BV2 cell lines treated with rhCDNF (0.5 or 1 μg/ml) for 1 h before incubation with erythrocytes. *P < 0.05, **P < 0.01 by one-way ANOVA followed by Bonferroni corrections. Mean ± SEM is shown. E A photomicrograph of CD11b-immunolabled microglia in cultures at 6 h after exposure to CFDA-labeled RBCs in the presence of 1 μg/ml rhCDNF or PBS. F Phagocytosis was measured by measuring the fluorescence intensity in the microglial lysate at 6 h after exposure to RBCs. RhCDNF at 1 μg/ml significantly augmented phagocytosis, when compared to PBS treatment. Data were analyzed as two-tailed Student’s t-test. *P < 0.05 (n = 5).