Therapeutic mechanism of cord blood mononuclear cells via the IL-8-mediated angiogenic pathway in neonatal hypoxic-ischaemic brain injury

In a clinical trial of cerebral palsy, the level of plasma interleukin-8 (IL-8) was increased, correlated with motor improvement, after human umbilical cord blood mononuclear cell (hUCBC) infusion. This study aimed to elucidate the role of IL-8 in the therapeutic effects of hUCBCs in a mouse model of hypoxic-ischaemic brain injury (HI). In P7 HI mouse brains, hUCBC administration at day 7 after HI upregulated the gene expression of Cxcl2, the mouse IL-8 homologue and increased the expression of its receptor, CXCR2. hUCBC administration restored the sequential downstream signalling axis of p-p38/p-MAPKAPK2, NFκB, and angiogenic factors, which were downregulated by HI. An in vitro assay revealed the downregulation of the angiogenic pathway by CXCR2 knockdown and p38 inhibition. In vivo p38 inhibition prior to hUCBC administration in HI mouse brains produced identical results. Behavioural outcomes revealed a therapeutic effect (ps < 0.01) of hUCBC or IL-8 administration, which was correlated with decreases in infarct size and angiogenic findings in the striatum. In conclusion, the response of the host to hUCBC administration in mice upregulated Cxcl2, which led to the activation of the IL-8-mediated p-p38 signalling pathway. The upregulation of the downstream pathway and angiogenic growth factors via NFκB can be inferred to be the potential therapeutic mechanism of hUCBCs.


Supplementary Methods and Results
As potential downstream signals of the IL-8-mediumted angiogenic pathway, mitogen activated protein kinases known to be involved in angiogenic processes, including the serine/threonine-specific protein kinase Akt and extracellular signal regulated kinase (Erk)1/2 were also evaluated. However, phosphorylated (p-) Akt and p-Erk 1/2 did not show significant changes after cord blood administration in the mouse brain following hypoxic ischaemic brain injury (HI) ( Supplementary Fig. S1).
The intranuclear translocation of p65, the NFᴋB subunit, was promoted by IL-8 treatment in mouse brain endothelial bEnd.3 cells under oxygen glucose deprivation. This finding suggests that NFᴋB activation was elicited by the IL-8-mediated pathway ( Supplementary   Fig. S2).
To find out whether the angiogenic effect of hUCBC occurs confined to the affected brain hemisphere, the unaffected contralateral hemispheres of mice were also sectioned for immunohistochemistry analysis. The upregulation of VEGF and CD31 was not observed in the contralateral hemisphere. Moreover, vessel density observed in the sham group was similar to all other groups of HI, hUCBC and IL-8 treatment, indicating that the sound side hemisphere was not affected by HI or systemic infusion of hUCBC ( Supplementary Fig. S3).
To determine whether IL-8 has an effect on angiogenic processes in vitro, zymography was performed in bEnd.3 cell cultures. Endothelial cells were cultured in DMEM without FBS. After 24 h, the culture supernatant was harvested, and the conditioned medium was diluted so that the samples exhibited the same protein concentration. Samples (10 μL protein) in nonreducing tris-glycine sample buffer with 4% SDS were subjected to electrophoresis in a 7.5% acrylamide gel containing gelatine. The gels were washed repeatedly with washing buffer and then rinsed for 10 min in incubation buffer. The enzymatic activity of MMP-9 was significantly increased under treatment with 50 ng/mL of IL-8 ( Supplementary Fig. S4).
To confirm the angiogenic effect of IL-8, actual tube formation was examined in vitro using mouse brain endothelial bEnd.3 cells prior to IL-8 treatment. The wells of 24-well culture cluster dishes were coated with Corning Matrigel Matrix solution (0.289 ml/well) (Corning Incorporated, MA), which was then allowed to solidify for 1 h at 37 °C. Cells (1.2×10 5 cells/well) were subsequently seeded onto Matrigel with or without 50 ng/ml IL-8.
The plates were incubated for 8 h, and the number of tubes formed was then counted from 9 microscopic fields selected and photographed at random. The number of tubes formed was significantly greater in cells treated with IL-8 (50 ng/ml) ( Supplementary Fig. S5).
To determine the initial time point of CXCR2 upregulation, real-time PCR was performed on brain samples from HI mice sacrificed 30 min after human umbilical cord blood administration. The primers for CXCR2 were CXCR2 forward, 5′-GAAATTTCGCCATGGACTTCTC-3′, and CXCR2 reverse, 5′-ACGAGCTAACAAAAGAAGGCCTT-3′. The expression of CXCR2 was consistently increased in the HI mouse brains treated with hUCBC compared to the HI-only mouse brains at identical time points (Supplementary Fig. S6A). This in vivo finding paralleled the observed in vitro CXCR2 expression after IL-8 treatment according to Western blotting. Mouse brain endothelial bEnd.3 cells were exposed to 24 h of oxygen glucose deprivation prior to IL-8 treatment. CXCR2 expression was continuously increased in cells treated with IL-8 compared with that in cells without IL-8 ( Supplementary Fig. S6B).
To determine the dosage of IL-8 to be used in mouse endothelial bEnd.3 cells, cell viability after IL-8 administration was tested. Cells cultured in Dulbecco's modified Eagle's medium with 10% foetal bovine serum were treated with various dilutions of IL-8 (1, 10, 50, and 100 ng/ml). Cell viability at 24 h post-IL-8 treatment was increased significantly under treatment with up to 50 ng/ml of IL-8 ( Supplementary Fig. S7).
Individual full-length blots are included in the Supplementary Data in the end. Data are shown as the mean ± SEM (n = 3). HI, hypoxic-ischaemic brain injury; DAPI, 4′, 6diamidino-2-phenylindole for nuclear staining; +UCB, HI mice subjected to hUCBC treatment; +IL-8, HI mice subjected to IL-8 treatment; hUCBCs, human umbilical cord blood mononuclear cells