Early-phase administration of human amnion-derived stem cells ameliorates neurobehavioral deficits of intracerebral hemorrhage by suppressing local inflammation and apoptosis

Abstract

Background: Intracerebral hemorrhage (ICH) is a significant cause of death and disabilities. Recently, cell therapies using mesenchymal stem cells have been shown to improve ICH-induced neurobehavioral deficits. Based on these findings, we designed this study to evaluate the therapeutic efficacy and underlying mechanisms by which human amnion-derived stem cells (hAMSCs) would ameliorate neurobehavioral deficits of ICH-bearing hosts.

Methods: hAMSCs were induced from amnia obtained by cesarean section and administered intravenously to ICH-bearing mice during the acute phase. The mice were then subject to multitask neurobehavioral tests at the subacute phase. We attempted to optimize the dosage and timing of the hAMSC administrations. In parallel with the hAMSCs, a tenfold higher dose of human adipose-derived stem cells (ADSCs) were used as an experimental control. Specimens were obtained from the ICH lesions to conduct immunostaining, flow cytometry, and Western blotting to elucidate the underlying mechanisms of the hAMSC treatment.

Results: The intravenous administration of hAMSCs to the ICH-bearing mice effectively improved their neurobehavioral deficits, particularly when the treatment was initiated at Day 1 after the ICH induction. Of note, the hAMSCs promoted clinical efficacy equivalent to or better than that of hADSCs at 1/10 the cell number. The systemically administered hAMSCs were found in the ICH lesions along with the local accumulation of macrophages/microglia. In detail, the hAMSC treatment decreased the number of CD11b⁺CD45⁺ and Ly6G⁺ cells in the ICH lesions, while splenocytes were not affected. Moreover, the hAMSC treatment decreased the number of apoptotic cells in the ICH lesions. These results were associated with suppression of the protein expression levels of macrophage-related factors iNOS and TNFα.

Conclusions: Intravenous hAMSC administration during the acute phase would improve ICH-induced neurobehavioral disorders. The underlying mechanism was suggested to be the suppression of subacute inflammation and apoptosis by suppressing macrophage/microglia cell numbers and macrophage functions (such as TNFα and iNOS). From a clinical point of view, hAMSC-based treatment may be a novel strategy for the treatment of ICH.

Keywords: Apoptosis; Human amnion-derived stem cell; Inflammation; Intracerebral hemorrhage; Macrophage; Microglia.

Fig. 1

Intravenous administration of hAMSCs dose-dependently improves ICH-induced neurobehavioral deficits during the subacute phase. A A protocol of neurobehavioral tests to optimize the administration dose of hAMSCs for mice bearing intracranial hemorrhage (ICH). The mice were divided into five groups: high-dose hAMSC group (n = 24), low-dose hAMSC group (n = 11), hADSC group (n = 11), ICH group (n = 32), and sham group (n = 32). The mice were then subject to the following neurobehavioral tests at Day 29 and later: open space swimming test (B), water maze learning test (C), and passive avoidance learning test (D). Data are plotted in mean ± SEM. P values are based on Tukey–Kramer test. *P < 0.05 and **P < 0.01 compared with the ICH group

Fig. 2

Early intravenous administration of hAMSCs might improve ICH-induced neurobehavioral deficits during the subacute phase. A A protocol of neurobehavioral tests to optimize the timing of the hAMSC administration. The ICH-bearing mice were divided into four groups: D1-hAMSC group (n = 12), D3-hAMSC groups (n = 12), untreated ICH group (n = 11), and sham group (n = 12). The mice were then subject to the following neurobehavioral tests at Day 29 and later: open space swimming test (B), water maze learning test (C), and passive avoidance learning test (D). Data are plotted in mean ± SEM. P values are based on Tukey–Kramer test. *P < 0.05 and **P < 0.01 compared with the ICH group. ^#P < 0.05 and ^##P < 0.01 compared with the D3-hAMSC group

Fig. 3

A small number of hAMSCs pass through the blood–brain barrier and interact with macrophage or microglial cell. Representative images of Stem101 staining of the ICH lesions at Day 4. A The hAMSC group, magnification: ×10. B The hAMSC group, magnification: ×40. C The ICH group, magnification: ×10. D The number of STEM101-reactive cells around the ICH lesions were enumerated. Double staining of Stem101 and Iba1 in the ICH lesion in the hAMSC group (E, F). Scale bars indicate 100 μm

Fig. 4

A hAMSC administration decreases the number of CD11b⁺CD45⁺ cells, Ly6C⁺ and Ly6G⁺ cells in the ICH lesions. A Representative gating strategy of the flow cytometry for CD11b CD45 Ly6C Ly6G cells. Data were obtained from the ICH lesions at Days 2, 4 and, 8. B The total cell number. C The number of CD11b⁺CD45⁺ cells. D The number of CD11b⁺CD45⁺Ly6C⁺Ly6G⁻ cells. E The number of CD11b⁺CD45⁺Ly6C⁻Ly6G⁺ cells. Data are plotted in mean ± SEM. P values are based on the Wilcoxon test. *P < 0.05 compared with the ICH group

Fig. 5

hAMSC administration does not affect CD11b⁺CD45⁺ cells in the spleen. A Representative gating strategy for the flow cytometry of CD11b CD45 Ly6C Ly6G cells. Data were obtained from spleen at Days 2, 4 and, 8. B The total cell number. C The number of CD11b⁺CD45⁺ cells. D The number of CD11b⁺CD45⁺Ly6C⁺Ly6G⁻ cells. E The number of CD11b⁺CD45⁺Ly6C⁻Ly6G⁺ cells. Data are plotted in mean ± SEM. P values are based on the Wilcoxon test. *P < 0.05 compared with the ICH group

Fig. 6

hAMSCs administration decreases the TUNEL-positive cells in the ICH lesions. Representative images of TUNEL staining of the ICH lesions at Days 4 and 8. Magnifications: ×40. A The ICH group at Day 4. B The ICH group at Day 8. C The hAMSC group at Day 4. D The hAMSC group at Day8. E The numbers of TUNEL-reactive cells (arrows) around the ICH lesions were enumerated. Data are plotted in mean ± SEM. P values are based on Wilcoxon test. *P < 0.05 compared with the same Day of ICH group. Scale bars indicate 100 μm

Fig. 7

Intravenous administration of hAMSCs suppress protein expression of iNOS and TNFa. The ICH tissues were extracted and subject to Western blotting for macrophage-related factors iNOS, TNFα, and arginase 1. A Representative image of protein levels of iNOS, TNFα, and arginase 1. B Data were plotted in mean ± SEM. P values are based on Tukey–Kramer test. *P < 0.05 compared with the ICH group at the same day

Fig. 8

Intravenous administration of TNFα counteracts the inhibitory effect of hAMSCs on TUNEL-positive cells. A A protocol of TNFα rescue experiment. The ICH-bearing mice were divided into three groups: B the ICH group at Day 4, C the hAMSC group at Day 4, and D the hAMSC+TNFα group at Day 8. E The numbers of TUNEL-reactive cells around the ICH lesions were enumerated. Data are plotted in mean ± SEM. P values are based on Tukey–Kramer test. *P < 0.05 compared with the Day 4 of ICH group. Scale bars indicate 100 µm