BDNF-Overexpressing Engineered Mesenchymal Stem Cells Enhances Their Therapeutic Efficacy against Severe Neonatal Hypoxic Ischemic Brain Injury

Abstract

We investigated whether irradiated brain-derived neurotropic factor (BDNF)-overexpressing engineered human mesenchymal stem cells (BDNF-eMSCs) improve paracrine efficiency and, thus, the beneficial potency of naïve MSCs against severe hypoxic ischemic (HI) brain injury in newborn rats. Irradiated BDNF-eMSCs hyper-secreted BDNF > 10 fold and were >5 fold more effective than naïve MSCs in attenuating the oxygen-glucose deprivation-induced increase in cytotoxicity, oxidative stress, and cell death in vitro. Only the irradiated BDNF-eMSCs, but not naïve MSCs, showed significant attenuating effects on severe neonatal HI-induced short-term brain injury scores, long-term progress of brain infarct, increased apoptotic cell death, astrogliosis and inflammatory responses, and impaired negative geotaxis and rotarod tests in vivo. Our data, showing better paracrine potency and the resultant better therapeutic efficacy of the irradiated BDNF-eMSCs, compared to naïve MSCs, suggest that MSCs transfected with the BDNF gene might represent a better, new therapeutic strategy against severe neonatal HI brain injury.

Keywords: brain; brain derived neurotropic factor; cell transplantation; hypoxia-ischemia; infant; mesenchymal stem cell transplantation; newborn; stem cells.

Figure 1

Characterization of BDNF- overexpressing engineered mesenchymal stem cells (eMSCs). (A) Morphology of BDNF-eMSCs. (B) Flow cytometry analyses show that BDNF-eMSCs express specific markers for MSCs such as CD44, CD105, and CD73. N = three per group (C) Cell proliferation rate of BDNF-eMSCs. This result is representative of three independent experiments. (D) BDNF secretion of distinct passages of BDNF-eMSCs measured by ELISA.

Figure 2

Neuroprotective efficacy of BDNF-eMSCs in vitro oxygen–glucose deprivation (OGD)-induced primary cultured rat cortical neurons. Cell viability, which is expressed as relative proliferation rate (%) to normal control group, was assessed with the CCK-8 assay in OGD-induced cultured neurons after co-treatment with BDNF-eMSCs, at doses of approximately 1 × 10³, 1 × 10⁴, 5 × 10⁴, 1 × 10⁵, 1 × 10⁶ cells per 1 mL, (A) and with naïve MSCs at doses of approximately 1 × 10³, 1 × 10⁴, 5 × 10⁴, 1 × 10⁵, 1 × 10⁶ cells per 1 mL (B). (C) Western blotting and ELISA assay for BDNF expression from the culture media of naïve MSCs and BDNF-eMSCs. (D) Cytotoxicity was evaluated with cell viability, expressed as a relative proliferation rate (%) to the normal control group, relative lactate dehydrogenase (LDH) release (%) to the positive control (100% fully killed cells), malondialdehyde (MDA) level in neuronal cells after OGD induction with/without co-treatment of naïve MSCs or BDNF-eMSCs at a dose of 1 × 10⁵ cells per 1 mL. (E) The number of terminal deoxynucleotidyl transferase UTP nick end labeling (TUNEL)-positive cells evaluated in OGD-induced rat cortical neurons after co-culture with naïve MSCs or BDNF-eMSCs at dose of 1 × 10⁵ cells per 1 mL. Data are expressed as the mean ± SD.

Figure 3

Morphological changes in the brains following hypoxic-ischemic brain injury and MSC transplantation. (A) Hematoxylin and eosin (H&E)-stained brain sections showing representative gross morphology of brains at 3 days after modeling. (B) The ratio of intact ipsilateral hemisphere to contralateral hemisphere was evaluated in the histologic brain section at 3 days after modeling in each group. (C) H&E-stained brain sections showing representative brain histology in cortex, striatum, thalamus, and hippocampus at 3 days after modeling. (D) Total sum of histologic scores in each area of brain; cortex, striatum, thalamus, and hippocampus. Data are expressed as mean ± standard error of the mean. * p < 0.05 compared to the HI + vehicle control, # p < 0.05 compared to the HI + naïve MSCs, $ p < 0.05 compared to HI-+ BDNF-eMSCs at dose 1 (approximately 1 × 10⁴ cells). Abbreviations: vehicle, HI + vehicle control, naïve MSCs; HI + naïve MSCs, BDNF MSC_dose1; HI + BDNF-eMSCs at 1 × 10⁴ cells dose; BDNF MSC_dose2; HI + BDNF-MSCs at 5 × 10⁴ cells dose; BDNF MSC_dose3; HI + BDNF-MSCs at 1 × 10⁵ cells dose.

Figure 4

Brain inflammatory cytokines after hypoxic ischemic injury and MSC transplantation. Levels of inflammatory cytokines including interleukin (IL)-1α, IL-β, and IL-6, and tumor necrosis factor (TNF)-α in total brain tissue homogenates at 3 days after modeling. Data are expressed as mean ± standard error of the mean. * p < 0.05 compared to the HI + vehicle control, # p < 0.05 compared to the HI + naïve MSCs, $ p < 0.05 compared to HI + BDNF-eMSCs at dose 1 (approximately 1 × 10⁴ cells), @ p < 0.05 compared to HI + BDNF-eMSCs at dose 2 (approximately 5 × 10⁴ cells). Abbreviations: vehicle, HI + vehicle control, naïve MSCs; HI + naïve MSCs, BDNF MSC_dose1; HI + BDNF-eMSCs at 1 × 10⁴ cells dose; BDNF MSC_dose2; HI + BDNF-eMSCs at 5 × 10⁴ cells dose; BDNF MSC_dose3; HI + BDNF-eMSCs at 1 × 10⁵ cells dose.

Figure 5

Brain damage area after hypoxic ischemic injury and MSC transplantation. (A) Representative serial brain magnetic resonance imaging (MRIs) from each group at right after HI brain injury modeling and 5 weeks after modeling. (B) The volumetric ratio of intact ipsilateral hemisphere to contralateral hemisphere was volumetrically evaluated in the serial images of brain MRI in each group right after HI brain injury modeling and 5 weeks after modeling. Data are expressed as mean ± standard error of the mean. * p < 0.05 compared to the HI + vehicle control. Abbreviations: vehicle, HI + vehicle control, naïve MSCs; HI + naïve MSCs, BDNF MSC_dose1; HI + BDNF-eMSCs at 1 × 10⁴ cells dose; BDNF MSC_dose2; HI + BDNF-eMSCs at 5 × 10⁴ cells dose; BDNF MSC_dose3; HI + BDNF-eMSCs at 1 × 10⁵ cells dose.

Figure 6

Brain tissue injury including inflammation, cell death, and reactive gliosis after hypoxic ischemic injury and MSC transplantation. (A) Representative immunofluorescence photomicrographs of the penumbra area with staining for ED1 (red), glial fibrillary acidic protein (GFAP) (red), TUNEL (green), and 4′,6-diamidino-2-pheylindole (DAPI) (blue) in the brains at 5 weeks after modeling. (B) The average number of ED1-positive cells, TUNEL-positive cells, and mean light intensity of GFAP, DCX, and MBP immunofluorescence per high-power field (HPF) in each group. (C) Levels of inflammatory cytokines including IL-1α, IL-β, and IL-6 and TNF-α in total brain tissue homogenates at 5 weeks after modeling. Data are expressed as mean ± standard error of the mean. * p < 0.05 compared to the HI + vehicle control. Abbreviations: vehicle, HI + vehicle control, naïve MSCs; HI + naïve MSCs, BDNF MSC_dose1; HI + BDNF-eMSCs at 1 × 10⁴ cells dose; BDNF MSC_dose2; HI + BDNF-eMSCs at 5 × 10⁴ cells dose; BDNF MSC_dose3; HI + BDNF-eMSCs at 1 × 10⁵ cells dose.

Figure 7

Sensorimotor behavior function after hypoxic ischemic injury and MSC transplantation. Sensorimotor functional outcomes on the negative geotaxis test at P35 (4 weeks after modeling) and at P42 (5 weeks after modeling) (A) and rotarod test (B) on P40–42 in each group. Data are expressed as mean ± standard error of the mean. * p < 0.05 compared to the HI + vehicle control, # p < 0.05 compared to the HI + naïve MSCs. Abbreviations: vehicle, HI + vehicle control, naïve MSCs; HI + naïve MSCs, BDNF MSC_dose1; HI + BDNF-eMSCs at 1 × 10⁴ cells dose; BDNF MSC_dose2; HI + BDNF-eMSCs at 5 × 10⁴ cells dose; BDNF MSC_dose3.