Role of CD44 in increasing the potency of mesenchymal stem cell extracellular vesicles by hyaluronic acid in severe pneumonia

Abstract

Background: Although promising, clinical translation of human mesenchymal stem or stromal cell-derived extracellular vesicles (MSC EV) for acute lung injury is potentially limited by significant production costs. The current study was performed to determine whether pretreatment of MSC EV with high molecular weight hyaluronic acid (HMW HA) would increase the therapeutic potency of MSC EV in severe bacterial pneumonia.

Methods: In vitro experiments were performed to determine the binding affinity of HMW HA to MSC EV and its uptake by human monocytes, and whether HMW HA primed MSC EV would increase bacterial phagocytosis by the monocytes. In addition, the role of CD44 receptor on MSC EV in the therapeutic effects of HMW HA primed MSC EV were investigated. In Pseudomonas aeruginosa (PA) pneumonia in mice, MSC EV primed with or without HMW HA were instilled intravenously 4 h after injury. After 24 h, the bronchoalveolar lavage fluid, blood, and lungs were analyzed for levels of bacteria, inflammation, MSC EV trafficking, and lung pathology.

Results: MSC EV bound preferentially to HMW HA at a molecular weight of 1.0 MDa compared with HA with a molecular weight of 40 KDa or 1.5 MDa. HMW HA primed MSC EV further increased MSC EV uptake and bacterial phagocytosis by monocytes compared to treatment with MSC EV alone. In PA pneumonia in mice, instillation of HMW HA primed MSC EV further reduced inflammation and decreased the bacterial load by enhancing the trafficking of MSC EV to the injured alveolus. CD44 siRNA pretreatment of MSC EV prior to incubation with HMW HA eliminated its trafficking to the alveolus and therapeutic effects.

Conclusions: HMW HA primed MSC EV significantly increased the potency of MSC EV in PA pneumonia in part by enhancing the trafficking of MSC EV to the sites of inflammation via the CD44 receptor on MSC EV which was associated with increased antimicrobial activity.

Keywords: CD44; Extracellular vesicles; Hyaluronic acid; Mesenchymal stem cell; Pseudomonas aeruginosa pneumonia.

Fig. 1

MSC-derived Extracellular Vesicles and Influence of Molecular Weight of Hyaluronic Acid on EV Binding. a Mean size and size distribution of MSC EV by NanoSight analyses. b Binding ability of different molecular weights of HA with MSC EV. PKH26-labeled MSC EV preferentially bound to HMW HA (1.0 MDa) compared with PBS, LMW HA (40 KDa), or HMW HA (1.5 MDa). Data are mean ± SD, ****P < 0.0001 by ANOVA (Bonferroni), N = 19–28. Dose of MSC EV (50 μl) used = 1.1 × 10¹⁰ particles

Fig. 2

Effect of Different Molecular Weights of Hyaluronic Acid on MSC EV Uptake by Human Monocytes. a Effect of pre-incubation of MSC EV with different doses of HMW HA (1.0 MDa) on its uptake by LPS stimulated human monocytes. Compared with MSC EV alone, MSC EV pre-incubated with HA doses of 0.2, 1, or 5 μg/ml HMW HA (1.0 MDa) showed increased uptake by LPS stimulated human monocytes. There were no significant differences between the three doses of HMW HA. Data are mean ± SD for each condition, ****P < 0.0001 by ANOVA (Bonferroni), N = 9–18. Bar = 1 mM. b Effect of MSC EV pre-incubated with different molecular weights of HA on MSC EV uptake by LPS stimulated monocytes. LMW HA (40 KDa) primed MSC EV, HMW HA (1.0 MDa) primed MSC EV, and HMW HA (1.5 MDa) primed MSC EV all showed increased uptake by LPS stimulated monocytes when compared with MSC EV alone. HMW HA (1.0 MDa) primed MSC EV demonstrated the highest uptake by the monocytes. Data are mean ± SD, ****P < 0.0001 by ANOVA (Bonferroni), N = 9. Bar = 1 mM. c HMW HA (1.0 MDa) primed MSC EV significantly decreased the secretion of TNFα by LPS stimulated human monocytes when compared with MSC EV alone. Data are mean ± SD, **P < 0.01, ****P < 0.0001 by ANOVA (Bonferroni), N = 16. Dose of MSC EV (50 μl) used = 1.1 × 10¹⁰ particles

Fig. 3

Effect of Different Molecular Weights of Hyaluronic Acid Primed MSC EV on PA103 Bacterial Phagocytosis. a, b Extracellular bacterial CFU levels in the culture medium and Intracellular bacterial CFU levels as measured in LPS stimulated monocytes when treated with different MW HA primed MSC EV. Administration of HMW HA (1.0 MDa) primed MSC EV reduced extracellular PA103 CFU and increased intracellular PA103 CFU levels when compared with MSC EV or LMW HA (40 KDa) primed MSC EV or HMW HA (1.0 MDa) alone. c, d Intracellular bacterial CFU levels of LPS stimulated monocytes when treated with CD44 siRNA pretreated MSC EV with or without HMW HA. CD44 siRNA pretreatment of MSC EV prior to HMW HA incubation decreased bacterial phagocytosis by human monocytes compared to negative control (NC) siRNA pretreated MSC EV. Data are mean ± SD, **P < 0.01, ***P < 0.001, ****P < 0.0001 by ANOVA (Bonferroni), N = 3. Individual P values describe the statistical comparison between the two groups by Student’s t test. A dose of MSC EV (90 μl) was used = 2.0 × 10¹⁰ particles

Fig. 4

Therapeutic Effects of MSC EV Primed HMW HA in Mice Injured with PA103 Bacterial Pneumonia. a Intravenous administration of MSC EV significantly decreased PA103 CFU levels in the BALF and blood when compared with injured mice. Treatment of HMW HA (1.0 MDa) primed MSC EV further reduced the PA103 CFU levels in the BALF and blood compared with MSC EV treated mice. Data are median with IQR, *P < 0.01, **P < 0.01, ****P < 0.0001 by Kruskal-Wallis test with Dunn’s correction, N = 6–20. Individual P values describe the statistical comparison between the two groups by Mann-Whitney U test. b Administration of MSC EV primed with or without HMW HA (1.0 MDa) both significantly reduced TNFα and IL-6 levels in the BALF and plasma in mice injured with severe PA pneumonia compared with injured mice. Compared to MSC EV, HMW HA (1.0 MDa) primed MSC EV further decreased TNFα and IL-6 levels in the BALF and plasma. Data are as mean ± SD, *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 by ANOVA (Bonferroni), N = 5–15. Individual P values describe the statistical comparison between the two groups by Student’s t test. A dose of MSC EV (90 μl) was used = 2.0 × 10¹⁰ particles

Fig. 5

Effect of HMW HA Primed MSC EV on Inflammatory Cell Infiltration into the Injured Alveolus. a There were no significant differences in the total white blood cell (WBC) nor absolute neutrophil (ANC) counts in either the BALF or plasma. Data are mean ± SD, N = 6–20. b By histology, administration of MSC EV reduced alveolar inflammatory cells infiltration, interstitial wall thickening, and blood/edema, which was further reduced with HMW HA primed MSC EV. Lung injury assessed by semiquantitative scoring was decreased by MSC EV and further reduced with HMW HA primed MSC EV. Magnification × 4, Bar = 1 mM. Data are as mean ± SD, ***P < 0.001, ****P < 0.0001 by ANOVA (Bonferroni), N = 20. A dose of MSC EV (90 μl) was used = 2.0 × 10¹⁰ particles

Fig. 6

Trafficking of MSC EV With or Without HMW HA in Mice with PA103 Bacterial Pneumonia. Intravenous administration of HMW HA (1.0 MDa) primed MSC EV increased trafficking of MSC EV to the lung, liver, and spleen. a Representative immunofluorescence from each organ with treatment of HMW HA (1.0 MDa) primed MSC EV or MSC EV alone. Magnification 4X, Bar = 1 mM. b Quantification of the fluorescent intensity. Data is mean ± SD, ***P < 0.001, ****P < 0.0001 by Student’s t test, N = 12–16. A dose of MSC EV (90 μl) was used = 2.0 × 10¹⁰ particles

Fig. 7

Schematic of Potential Mechanisms Underlying the Therapeutic Effects of HMW HA Primed MSC EV. By binding to CD44 on MSC EV, HMW HA further increased the trafficking, adhesion, and internalization of MSC EV into target cells in the injured alveolus, increasing the potency of the EV. We speculate that HMW HA may be a linker between MSC EV and target immune cells which leads to the internalization of the EV