Immune profiling reveals umbilical cord blood mononuclear cells from South India display an IL-8 dominant, CXCL-10 deficient polyfunctional monocyte response to pathogen-associated molecular patterns that is distinct from adult blood cells

Abstract

Neonate responses to pathogen-associated molecular patterns (PAMPS) differ from adults; such understanding is poor in Indian neonates, despite recognized significant infectious risk. Immune profiling analysis was undertaken of 10 secreted mediators contextualized with cellular source induced by six PAMPs in umbilical cord (CB; n = 21) and adult-blood (PBMC; n = 14) from a tertiary care hospital in South India. Differential cytokine expression analysis (minimum log2-fold difference; adj P-value < 0.05) identified bacterial PAMPs induced higher concentrations of IL-1β, IL-10, TNF-α in adults versus IL-8, GM-CSF, IFN-γ, and IL-2 in CB. CB responded to poly I:C and SARS-CoV-2 lysate with a dominant IL-8 response, whereas in PBMC, CXCL-10 dominated poly I:C, but not SARS-CoV-2, responses, highlighting potential IL-8 importance, in the absence of Type I Interferons, in antiviral CB immunity. Candida albicans was the only PAMP to uniformly induce higher secretion of effectors in CB. The predominant source of IL-8/IL-6/TNF-α/IL-1β in both CB and PBMC was polyfunctional monocytes and IFN-γ/IL-2/IL-17 from innate lymphocytes. Correlation matrix analyses revealed IL-8 to be the most differentially regulated, correlating positively in CB versus negatively in PBMC with IL-6, GM-CSF, IFN-γ, IL-2, consistent with more negatively regulated cytokine modules in adults, potentially linked to higher anti-inflammatory IL-10. Cord and adult blood from India respond robustly to PAMPs with unique effector combinations. These data provide a strong foundation to monitor, explore, mechanisms that regulate such immunity during the life course, an area of significant global health importance given infection-related infant mortality incidence.

Keywords: cytokines; monocytes; neonates; pathogen-associated molecular patterns; toll-like receptors.

Graphical Abstract

Figure 1.

Magnitude of PAMP-induced cytokine responses differ between cord blood (CB) and adult peripheral blood mononuclear cells (PBMC). Mononuclear cells isolated from infant cord (n = 21) and adult blood (n = 14) were stimulated with multiple PAMPs for 24 h after which supernatants were collected and cytokines were measured by a customized cytokine bead array to study expression of 10 key innate and adaptive analytes. No-stimulation cells were used as negative controls and subtracted from all the PAMP-stimulated samples. (A) Principal component analysis (PCA) of adult and cord blood multiplex dataset including 10 cytokines and 6 stimulations. Principal components of the immune parameters measured in adults and cord blood. Each data point on the plot represents an individual observation in the dataset. (B) The absolute concentrations of analytes in adult and cord blood for bacterial (BCG, LPS, and Pam3CSK4), fungal (Candida albicans) PAMPs and viral PAMPs (poly I:C and SARS-CoV-2 lysate). (C) Adult versus cord blood differential cytokine responses to selected PAMPs are shown in volcano plots. Differentially expressed cytokines (adjusted P-value < 0.05 and absolute fold change > 2) are labelled. Over-expressed cytokines in adult are shown in right compartment (right side to the vertical dotted line) while left compartment have over-expressed cytokines in cord blood group. (D) Pie chart showing the contribution of each cytokine for bacterial (BCG, LPS, and Pam3CSK4), fungal (C. albicans), and viral (poly I:C and SARS-CoV-2 lysate) PAMPs in cord blood and adults. Mann–Whitney test was used for comparisons shown in (B) P < 0.05 was considered significant. *P ≤ 0.05, **P < 0.01, ***P ≤ 0.001, ****P ≤ 0.0001.

Figure 2.

Coordinated expression patterns of PAMP-induced cytokine modules differ markedly between CB and PBMC. Correlation matrices were generated from multiplex data shown in the figure for cord and adult peripheral mononuclear cell responses. Spearman’s correlation method is used for the clustering of cytokine-stimulant pairs. (A) Correlation of all cytokines measured in cord blood (CBMC) (n = 21) across all the PAMPs tested. (B) Correlation of all cytokines measured in adult blood (PBMC) (n = 14) across all the PAMPs tested. (C) Correlation of IL-1β, TNF-α, IL-8, and IL-6 in cord blood (CBMC) across all the PAMPs tested. (D) Correlation of IL-1β, TNF-α, IL-8, and IL-6 in adult blood (PBMC) across all the PAMPs tested. Spearman’s rank correlation was used. P-values of correlation significance were corrected for multiple testing using the Benjamini–Hochberg (BH) method. P < 0.05 was considered significant. *P ≤ 0.05, **P < 0.01, ***P ≤ 0.001, ****P ≤ 0.0001.

Figure 2.

Coordinated expression patterns of PAMP-induced cytokine modules differ markedly between CB and PBMC. Correlation matrices were generated from multiplex data shown in the figure for cord and adult peripheral mononuclear cell responses. Spearman’s correlation method is used for the clustering of cytokine-stimulant pairs. (A) Correlation of all cytokines measured in cord blood (CBMC) (n = 21) across all the PAMPs tested. (B) Correlation of all cytokines measured in adult blood (PBMC) (n = 14) across all the PAMPs tested. (C) Correlation of IL-1β, TNF-α, IL-8, and IL-6 in cord blood (CBMC) across all the PAMPs tested. (D) Correlation of IL-1β, TNF-α, IL-8, and IL-6 in adult blood (PBMC) across all the PAMPs tested. Spearman’s rank correlation was used. P-values of correlation significance were corrected for multiple testing using the Benjamini–Hochberg (BH) method. P < 0.05 was considered significant. *P ≤ 0.05, **P < 0.01, ***P ≤ 0.001, ****P ≤ 0.0001.

Figure 3.

Monocytes were the major cellular source of cytokines and chemokines in CBMC and PBMC. CBMC (n = 21) and adult PBMCs (n = 6) were stimulated with 0.25 × 10⁶ cfu/ml BCG or 1 ng/ml LPS or 10⁶ cfu/ml Candida albicans for 6 h after which cells were stained with a panel of antibodies to study the expression of IL-1β, IL-8, IL-6, TNF-α, IL-17, IL-10, IFN-γ, and IL-2. No-stimulation cells were used as negative controls and subtracted from all the PAMP-stimulated samples. (A) Frequency of immune cell subsets producing at least one of IL-6, TNF-α, IL-1β, and IL-8 cytokines in BCG, LPS, and C. albicans in cord blood (CBMC) and adult blood (PBMC) across immune cell subsets. (B) Frequency of immune cell subsets producing at least one of IFN-γ, IL-2, IL-17, and IL-10 cytokines in BCG, LPS, and C. albicans in cord blood (CBMC) and adult blood (PBMC) across immune cell subsets. (C) Frequency of immune cell subsets producing Granzyme B in No-stimulation control. Mann–Whitney test was used for comparisons shown in (B and C) P < 0.05 was considered significant. *P ≤ 0.05, **P < 0.01, ***P ≤ 0.001, ****P ≤ 0.0001.

Figure 4.

Quality of the PAMP-specific monocytes and NK cell response in cord and adult blood. CBMC (n = 21) and adult PBMCs (n = 6) were stimulated with 0.25 × 10⁶ cfu/ml BCG or 1ng/ml LPS or 10⁶ cfu/ml C. albicans for 6 h after which cells were stained with a panel of antibodies to study the expression of IL-1β, IL-8, IL-6, TNF-α, IL-17, IL-10, IFN-γ, and IL-2 in various immune cell subsets. No-stimulation cells were used as negative controls. Boolean gates were created from the individual cytokines (IL-1β, IL-8, IL-6, and TNF-α) in FlowJo to divide responding cells into 15 distinct subsets corresponding to all possible combinations of these functions, and the data were analyzed using SPICE software. PAMP-induced immune responses to BCG, LPS, and Candida were compared with no stimulation control. Data were analyzed for statistical significance using Wilcoxon signed-rank test. Log data analyzed in all cases. P < 0.05 was considered statistically significant.

Figure 5.

Cellular analysis confirms the paucity of CXCL-10 response in CB to viral PAMPs. CBMC (n = 6) and adult PBMCs (n = 6) were stimulated with 100 µg/ml poly I:C or 1 µg/ml SARS-CoV-2 lysate for 6 h after which cells were stained with a panel of antibodies to study the expression of IL-1β, IL-8, IL-6, TNF-α, CXCL-10, and IFN-α. No-stimulation cells were used as negative controls and subtracted from all the PAMP-stimulated samples. (A) Frequency of immune cell subsets producing CXCL-10, IFN-α, and IL-8 after being stimulated with poly I:C and SARS-CoV-2 lysate in cord blood (CBMC in pink) and adult blood (PBMC in blue). Mann–Whitney test was used for the comparisons shown. P < 0.05 was considered significant. *P ≤ 0.05, **P < 0.01, ***P ≤ 0.001, ****P ≤ 0.0001.