CD5L as a promising biological therapeutic for treating sepsis

Abstract

Sepsis results from systemic, dysregulated inflammatory responses to infection, culminating in multiple organ failure. Here, we demonstrate the utility of CD5L for treating experimental sepsis caused by cecal ligation and puncture (CLP). We show that CD5L's important features include its ability to enhance neutrophil recruitment and activation by increasing circulating levels of CXCL1, and to promote neutrophil phagocytosis. CD5L-deficient mice exhibit impaired neutrophil recruitment and compromised bacterial control, rendering them susceptible to attenuated CLP. CD5L^-/- peritoneal cells from mice subjected to medium-grade CLP exhibit a heightened pro-inflammatory transcriptional profile, reflecting a loss of control of the immune response to the infection. Intravenous administration of recombinant CD5L (rCD5L) in immunocompetent C57BL/6 wild-type (WT) mice significantly ameliorates measures of disease in the setting of high-grade CLP-induced sepsis. Furthermore, rCD5L lowers endotoxin and damage-associated molecular pattern (DAMP) levels, and protects WT mice from LPS-induced endotoxic shock. These findings warrant the investigation of rCD5L as a possible treatment for sepsis in humans.

Fig. 1. CD5L⁻ mice have a compromised immune response to mid-grade CLP.

WT and CD5L⁻ mice were subjected to cecal ligation and puncture (CLP) surgery to induce moderate disease severity. a Kaplan–Meier survival curves were generated to compare survival between the two groups and significance was determined by log-rank (Mantel-Cox) test. Graphical representation of pooled individuals (n = 15 in each group) from 5 independent experiments. b Quantification of body weight loss in the experimental setting described in a. Data are presented as mean values ± SEM. c–e WT and CD5L⁻ mice were subjected to mid-grade CLP; control mice (sham) underwent the same surgical procedure but without ligation and puncture of the cecum. Mice were sacrificed 6, 24, or 72 h after surgery. Pooled data from at least 2 independent experiments. c CFU counts of the indicated tissues from WT and CD5L⁻ mice after mid-grade CLP. Statistical differences between groups were analyzed by two-tailed Mann-Whitney test. DL, detection limit = 20 CFU. Mice per group at 6, 24, and 72 h: Peritoneum WT: 7, 6, and 7; peritoneum CD5L⁻: 6, 6, and 8. Blood WT: 7, 6, and 7; blood CD5L⁻: 8, 6, and 8. Lung, liver, and kidney WT: 7, 7, and 7; lung, liver and kidney CD5L⁻: 7, 6, and 8. d Absolute number of total CD45⁺ leukocytes, CD45⁺CD11b⁺Ly6G⁺ neutrophils, CD45⁺CD11b⁺CD11c^-F4/80⁺ macrophages and Ly6C⁺ and CD206⁺ subsets within macrophage population, assessed by flow cytometry of cell populations in the peritoneal cavity. Statistical comparisons were drawn after performing two-tailed unpaired t-tests with Welch’s correction. Mice per group at 6, 24, and 72 h: Sham, WT, and CD5L⁻: 4, 4, and 4. CLP WT: 7, 7, and 7; CLP CD5L⁻: 7, 6, and 8. Floating bars show the minimum, average (line), and maximum values within each group (c-d). e Fold change in the indicated cytokines between CD5L⁻ and WT mice, quantified by bead-based multiplex immunoassay in samples from the peritoneal cavity (left panel) and blood serum (right panel).

Fig. 2. RNA-seq analysis reveals important changes in inflammatory pathways in CD5L⁻ mice upon CLP, when compared with WT mice.

a Principal component analysis (PCA) of RNA-seq expression data for the 4 groups of peritoneal cells [WT and CD5L⁻ mice, either naive (0 h) or 6 h after CLP; n = 3]. b Volcano plot depicting differentially expressed genes in CD5L⁻ vs. WT mice, 6 h after CLP. Red dots represent genes expressed at higher levels in CD5L⁻ mice, while black dots represent genes with higher expression levels in WT controls. Grey dots represent genes bellow the cutoff of significant (|log₂ fold change| ≥ 0.5 and adjusted P values ≤ 0.05). Differential expression was evaluated using the Wald’s test, followed by adjustment for multiple testing with the Benjamini-Hochberg correction. Log₂-fold change values were shrunk with the apeglm method to increase the signal-over-noise ratio of the effect size. c Dot plot of gene set enrichment analysis (mouse hallmark gene set collection) for CD5L⁻ vs. WT mice, 6 h after CLP. The diameter of the dot indicates the degree of significance of the ontology term. Red dots represent terms enriched in CD5L⁻ mice, while black dots represent terms enriched in WT mice. d Heatmap of the relative expression values (z-score of each gene across samples) of the top 7 upregulated pathways. Only differentially expressed genes were represented, excluding genes belonging to more than one pathway.

Fig. 3. Administration of rCD5L reduces lethality of WT mice subjected to high-grade CLP.

a WT mice underwent CLP to induce severe disease. Mice were then injected with two doses of 2.5 mg/kg of rCD5L intraperitoneally (IP, left) or intravenously (IV, right) 3 and 6 h post-surgery. Kaplan–Meier curves compared survival between treated and untreated groups, with significance determined by log-rank (Mantel-Cox) test. Pooled data from 4 independent experiments, number of mice per group (n) is indicated. b Protocol for analysis after rCD5L treatment. Top - analysis at 6 h. Mice were injected IP or IV with 2.5 mg/kg rCD5L at 3 h post-CLP, or PBS, and euthanized 3 h later. Bottom - analysis at 24 h. Mice were injected IP or IV with 2 doses of rCD5L, 3 and 6 h after CLP, or PBS, and euthanized at 24 h. Image created with BioRender.com. c Flow cytometry-based absolute counts of cellular populations in the peritoneum of mice injected via IP or IV routes. Leukocytes, CD45⁺; neutrophils, CD45⁺CD11b⁺Ly6G⁺; macrophages, CD45⁺CD11b⁺CD11c^-F4/80⁺; B1 cells, CD45⁺CD19⁺CD5⁺; B2 cells, CD45⁺CD19⁺CD5^-; T cells, CD45⁺CD3⁺. Statistical comparisons were made using two-tailed unpaired t-tests with Welch’s correction. Mice per group: for IP-treatment groups, for Leukocytes, Neutrophils, and Macrophages at 6 h: 6 untreated and 5 IP-treated; at 24 h: 7 untreated and 6 IP-treated. For B1, B2, and T lymphocytes at 6 h: 3 untreated and 3 IP-treated; at 24 h: 3 untreated and 2 IP-treated. In all groups of IV-treatment, there were 6 mice each. d Fold change in indicated cytokines between IP- or IV-treated mice and PBS-injected controls. Cytokines quantified by bead-based multiplex immunoassays on samples from peritoneum and serum. e CFU counts in tissues from IP- or IV-treated mice and untreated groups. Statistical differences between groups were analyzed by two-tailed Mann-Whitney test. DL, detection limit = 20 CFU. Mice per group: for IP-treatment groups, at 6 h: 6 untreated and 6 IP-treated; at 24 h: 7 untreated and 6 IP-treated. For IV-treatment groups, 6 mice were in each group. Pooled data from at least 2 independent experiments (c–e). Floating bars show the minimum, average, and maximum values within each group.

Fig. 4. High-dimensional data analysis distinguishes rCD5L therapeutic groups in high-grade CLP, and compares with responses of WT vs. CD5L⁻ mice in mid-grade CLP.

a Two-dimensional t-distributed stochastic neighbor embedding (t-SNE) visualization of cytokines, CFUs and neutrophil recruitment for a perplexity of 8. Each ellipse describes a phenotypic subtype and are derived from 36 different measurements (seven for WT, five for IP-treated, and six for the remaining groups). Measurements were log₂ transformed and normalized by the mean value of the “untreated” or the CD5L⁻ subgroups for each setup [intravenous (IV); intraperitoneal (IP) or WT/CD5L⁻]. b Linear discriminant analysis (LDA) biplot showing the overall profile of treatment of CLP with rCD5L administered IV or IP, and mid-grade CLP in WT or CD5L⁻ mice. Ellipses show 95% confidence intervals for each treatment and vectors represent the contribution of each variable to the overall variance.

Fig. 5. Biodistribution of CD5L upon mid-grade CLP.

a Normalized counts of Cd5l transcripts obtained from RNA-seq analysis of peritoneal cells of WT mice at indicated times after CLP. Mice per group: 3. b RT-qPCR quantification of Cd5l expression, normalized with Hprt1, in peritoneal cells and liver tissue at indicated times after CLP. Mice per group: 3. c ELISA quantification of CD5L in peritoneum and blood of WT mice at indicated times after mid-grade CLP, or in sham operated animals. Mice per group: at 0 h: 8 in peritoneum, 9 in blood; CLP at 6, 24, and 72 h (peritoneum and blood): 7, 9, and 7; Sham at 6, 24, and 72 h (peritoneum and blood): 6, 6, and 4. d CD5L⁻ mice were subjected to mid-grade CLP and 3 h later injected IV with 2.5 mg/kg rCD5L, or PBS. Fluids from peritoneum and serum were collected 1 h later and total CD5L was quantified by ELISA. DL, detection limit: 6.25 pg/ml. Mice per group: 4 for PBS; 5 for rCD5L. Pooled data from at least 2 independent experiments (c, d). e Scatterplot illustrates the correlation between peritoneal and blood rCD5L levels following IV injection, accompanied by Spearman’s rank correlation coefficient and the two-tailed P value (n = 5). Statistical differences between groups analyzed by one-way ANOVA with Dunnett’s multiple comparisons correction (b), two-tailed unpaired t-tests with Welch’s correction (c), or two-tailed Mann-Whitney test (d). Floating bars show the minimum, average, and maximum values within each group.

Fig. 6. CD5L bioactivity hinges on its free state dynamics.

a Determination of IgM-bound and free endogenous CD5L in peritoneal fluid and serum of WT mice at specified time points after CLP. Values were obtained through densitometric analysis of western blots normalized to baseline IgM-bound CD5L levels (100) in naïve mice. Each group comprised six mice. b ELISA quantification of total IgM in the peritoneum and serum of WT mice at specified times after CLP. Mice per group at 0, 6, 24, and 72 h were as follows: Peritoneum: 7, 9, 7, and 7; Blood: 8, 9, 6, and 7, respectively. c ELISA quantification of total CD5L in the peritoneal cavity of mice subjected to high-grade CLP, and IP- or IV-treated with rCD5L. At 6 h, mice had received one dose of rCD5L (2.5 mg/kg), or PBS, 3 h after CLP. At 24 h, mice had received 2 doses of rCD5L, or PBS, 3 and 6 h after CLP. Mice per group: IP treatment: 6 untreated and 6 treated, at 6 h; 7 untreated and 5 treated, at 24 h. IV treatment: 8 untreated and 6 treated, at 6 h; 8 untreated and 6 treated, at 24 h. d Determination of IgM-bound and free total (endogenous + recombinant) CD5L in the peritoneal cavity of WT mice treated with 2.5 mg/kg rCD5L, or PBS (untreated), via IP or IV routes, 3 h after high-grade CLP, and euthanized 3 h later. Values were obtained by densitometric analysis of western blot bands corresponding to IgM-bound and free CD5L. Mice per group: IP treatment: 6 untreated and 5 treated. IV treatment: 5 untreated and 6 treated. Pooled data from at least 2 independent experiments (a–d). Statistical differences between groups analyzed by two-tailed unpaired t-tests with Welch’s correction (a), one-way ANOVA with Dunnett’s multiple comparisons correction (b), or two-tailed Mann-Whitney test (c, d). Floating bars show the minimum, average, and maximum values within each group.

Fig. 7. CD5L promotes neutrophil phagocytosis.

a, b In vivo analysis of phagocytosis of pHrodo^TM Red E. coli BioParticles^TM. WT mice were injected IP with 15 μl BioParticles and 50 μg of rCD5L, or PBS (0 μg). Peritoneal cells were recovered 3 h later and analyzed by flow cytometry. a Percentage of large peritoneal macrophages (LPMs, CD45⁺CD11b^+/highF4/80^high) or neutrophils (CD45⁺CD11b⁺Ly6G⁺) that internalized BioParticles. Mice per group: 6. b MFI (geometric mean) of pHrodo channel within the pHrodo-positive LPM or neutrophil populations. Statistical comparisons between groups determined by two-tailed unpaired t-tests with Welch’s correction. Mice per group: 6. c, d Thioglycolate-elicited macrophages collected from mouse peritonea 72 h post-broth injection were infected with cecal bacteria pre-coated with 0, 2 or 10 μg of rCD5L. c After coating, a sample of cecal bacteria was boiled and subjected to SDS-PAGE. Detection of rCD5L bound to bacteria was performed via western blotting using primary anti-His tag mAb. d Gentamycin protection assay for phagocytosis analysis after 1-h incubation with pre-coated cecal bacteria at an MOI of 5 and enumeration of CFUs determined by serial dilution plating. Mice per group: 6. a–d Statistical comparisons between groups established by unpaired t-tests with Welch’s correction. Pooled data from two independent experiments. Floating bars show minimum, average, and maximum values within each group. e, f Resident LPMs and thioglycolate-elicited neutrophils were collected from WT mouse peritonea 6 h post-broth injection and incubated with BioParticles pre-coated with 0 or 10 μg rCD5L for the indicated times before phagocytosis analysis by flow cytometry. e Percentage of LPMs (upper left panel) and neutrophils (lower left panel) that internalized BioParticles, and MFI values of pHrodo channel within the pHrodo-positive LPM (upper right panel) and neutrophil (lower right panel) populations. Statistical differences between groups analyzed by two-way ANOVA with Šídák’s multiple comparisons test. Data shown are mean with SEM of n = 3 samples/group from one representative of two independent experiments. f Histogram overlays of representative examples of pHrodo channel fluorescence of LPMs and neutrophils incubated for the indicated times with uncoated (0) or rCD5L-coated (10) BioParticles.

Fig. 8. CD5L reduces endotoxin and DAMP levels, and enhances mouse survival in septic shock.

a ELISA quantification of LPS in sera from WT and CD5L⁻ mice 72 h after mid-grade CLP; and in sera from WT mice subjected to high-grade CLP followed by rCD5L treatment. In these, mice were IP- or IV-injected with PBS (untreated), or with two doses of 2.5 mg/kg rCD5L, 3 and 6 h after surgery. Sera was collected and analyzed at 24 h. DL, detection limit: 0.16 ng/ml. The experimental groups comprised 7 WT and 6 CD5L⁻ mice for mid-grade CLP, and for high-grade CLP: 7 IP-untreated, 5 IP-treated, 8 IV-untreated, and 6 IV-treated mice. b, c HMGB1 protein expression in WT and CD5L⁻ organs, determined by immunofluorescence. b Representative images of HMGB1 expression in paraffin-preserved sections of lung, liver, and kidney of WT and CD5L⁻ mice. c Quantification of HMGB1-stained areas in 3 regions from 2 independent images, from 2 mice per CLP group (72 h), or from 1 naïve control mouse per group (0 h), normalized by the DAPI-stained area in the corresponding regions. d, e HMGB1 expression in organs of WT mice subjected to high-grade CLP, and IP- or IV-treated with two doses of rCD5L. d Representative images of HMGB1 expression in lung, liver, and kidney of treated and untreated mice. e Quantification of HMGB1-stained areas, as in (c). a, c, e Floating bars represent the minimum, average and maximum values within each group. Statistical differences between groups were analyzed using a two-tailed Mann-Whitney test. b, d Scale bar: 20 μm. f WT and CD5L⁻ mice were IP-injected with a sublethal LPS dose (1.5 mg/kg). Survival was monitored for 6 days. g WT mice were injected with a lethal LPS dose (10 mg/kg), and 3 h later with 2.5 or 5 mg/kg rCD5L, or left untreated (0). Survival was monitored for 4 days. f, g Kaplan–Meier curves were generated to compare survival between groups. Significance was determined by log-rank (Mantel-Cox) test. The graphical representation includes pooled data from 3 independent experiments for (f) and 2 independent experiments for g.

Fig. 9. CD5L promotes neutrophil chemotaxis via CXCL1, and neutrophil activation.

a Quantification of inflammatory chemokines in WT and CD5L⁻ mice at indicated time points after mid-grade CLP, using bead-based multiplex immunoassays. b WT mice subjected to high-grade CLP were IV-injected with 2.5 mg/kg of rCD5L, or PBS (untreated), 3 h later. Inflammatory chemokines were quantified 6 h post-CLP. c CXCL1 concentrations in WT and CD5L⁻ mice after mid-grade CLP (extracted from data presented in a). d CXCL1 concentrations in untreated and IV-treated WT mice after high-grade CLP (extracted from data presented in b). a–d Pooled data from at least 2 independent experiments, analyzed by two-tailed Mann-Whitney test. a, c Mice per group: Peritoneum: 4 WT and 4 CD5L⁻ at 0 h, 4 WT, and 3 CD5L⁻ at 3 h, 6 WT, and 6 CD5L⁻ at 6 h; Blood, 9 WT and 9 CD5L⁻ at 0 h, 3 WT, and 3 CD5L⁻ at 3 h, 8 WT and 5 CD5L⁻ at 6 h. b, d Mice per group: 6 in untreated, 5 in rCD5L IV-treated. e CD5L⁻ mice were injected IV with rCD5L (2.5 mg/kg) 3 h after mid-grade CLP and peritoneal cells were recovered 3 h later. Percentage of CD5L-bound cells within CD45^- cells; neutrophils (CD45⁺CD11b⁺Ly6G⁺), macrophages (CD45⁺CD11b⁺F4/80⁺), other CD11b⁺ (CD45⁺CD11b⁺F4/80^-Ly6G^-), B cells (CD45⁺B220⁺), and T cells (CD45⁺CD3⁺). f MFI values of the CXCL1 channel within CXCL1⁺ cells in each subset defined in e. e, f Pooled data from 2 independent experiments, analyzed by two-way ANOVA. Mice per group: 5 untreated and 6 rCD5L-treated. Ly6G and CD11b MFI in neutrophils collected from the peritoneal cavity of rCD5L IP- (left panel) or IV-treated (right panel) WT mice, after high-grade CLP (g), or WT vs. CD5L⁻ mice after mid-grade CLP (h). Mice per group: IP treatment: 3 at 6 h, 4 at 24 h; IV treatment: 6 at 6 h, 6 at 24 h (g); IP and IV treatment: 4 (h). Pooled data from 2 independent experiments, statistical comparisons between groups were established by two-tailed unpaired t-tests with Welch’s correction.

Fig. 10. Mechanistic action of rCD5L.

Intravenous administration of rCD5L leads to elevated levels of free bioactive protein in the peritoneal cavity (1). Upon binding to target cells of non-hematopoietic origin (2), rCD5L induces significant production of the CXCL1 chemokine (3). The production of CXCL1 establishes a chemotactic gradient, resulting in heightened neutrophil activation (4) and recruitment from the bloodstream (5). The exact mechanism behind neutrophil  activation, whether through increased CXCL1 levels or direct action of rCD5L, remains unclear. Increased neutrophil presence in the peritoneal cavity, coupled with the synergistic effect of rCD5L in bacterial binding and enhanced phagocytosis, facilitates efficient bacterial clearance (6). rCD5L aids in the effective removal of DAMPs, likely from both the bloodstream and the peritoneum (7). Endogenous local release of CD5L from resident macrophages contributes to the overall pool of free CD5L at the site of infection. This illustration is for explanatory purposes and is not drawn to scale. Created with Biorender.com.