Neuroprotection by PlGF gene-modified human mesenchymal stem cells after cerebral ischaemia

Abstract

Intravenous delivery of mesenchymal stem cells (MSCs) prepared from adult bone marrow reduces infarction size and ameliorates functional deficits in rat cerebral ischaemia models. Placental growth factor (PlGF) is angiogenic to impaired non-neural tissue. To test the hypothesis that PlGF contributes to the therapeutic benefits of MSC delivery in cerebral ischaemia, we compared the efficacy of systemic delivery of human MSCs (hMSCs) and hMSCs transfected with a fibre-mutant F/RGD adenovirus vector with a PlGF gene (PlGF-hMSCs). A permanent middle cerebral artery occlusion (MCAO) was induced by intraluminal vascular occlusion with a microfilament. hMSCs and PlGF-hMSCs were intravenously injected into the rats 3 h after MCAO. Lesion size was assessed at 3 and 6 h, and 1, 3, 4 and 7 days using MR imaging and histology. Functional outcome was assessed using the limb placement test and the treadmill stress test. Both hMSCs and PlGF-hMSCs reduced lesion volume, induced angiogenesis and elicited functional improvement compared with the control sham group, but the effect was greater in the PlGF-hMSC group. Enzyme-linked immunosorbent assay of the infarcted hemisphere revealed an increase in PlGF in both hMSC groups, but a greater increase in the PlGF-hMSC group. These data support the hypothesis that PlGF contributes to neuroprotection and angiogenesis in cerebral ischaemia, and cellular delivery of PlGF to the brain can be achieved by intravenous delivery of hMSCs.

Fig. 1

Flow cytometric analysis of surface antigen expression on primary hMSC (A) and PlGF-hMSC (B). The cells were immunolabelled with FITC-conjugated monoclonal antibody specific for the indicated surface antigen. Dead cells were eliminated by forward and side scatter. PlGF production of hMSC, LacZ-transfected hMSC and PlGF-transfected hMSC were summarized (C). Levels of PlGF in the supernatant were shown.

Fig. 2

DWI- and T₂-weighted images (T₂), collected at 3 h (column 1), 3 days (column 2) and 7 days (column 3) post-lesion induction. The 3 h images were obtained just before intravenous cell or vehicle injection. Images from sham control, hMSC and PlGF-hMSC injections are shown in rows 1, 2 and 3, respectively, for data sets of (A) DWI and (B) T₂. The infarction is observed as a high-intensity area of the ischaemic (right) side of the brain. Scale bar = 5 mm.

Fig. 3

A summary of lesion volumes evaluated with MRI (DWI) (A), T₂-WI (B) and TTC staining (C). Both DWIs (A) and T₂-WIs (B) were obtained 3, 6 and 24 h, 3, 4 and 7 days after MCAO in rats treated with medium (control), hMSC or PlGF-hMSC. Images at 3 h were obtained just before treatment. (C) Seven days after MCAO, there was a reduction in the lesion volume assayed using TTC staining for both hMSC and PlGF-hMSC groups, but the reduction was greater in the PlGF-hMSC group. (D) PlGF in vivo levels assayed with ELISA. Levels of PlGF were significantly increased in the ischaemic hemisphere of hMSC and PlGF-hMSC compared with medium-treated rats. In addition, PlGF levels were greater in the ischaemic hemisphere of PlGF-hMSC-transplanted as compared with rats that received hMSC. Assays were obtained 3 and 7 days after MCAO induction. Note that PlGF was increased in the lesion with medium injection alone.

Fig. 4

Brain sections stained with TTC to visualize the ischaemic lesions 7 days after MCAO. Sections from non-treated MCAO (A-1), hMSC- (A-2) and PlGF-hMSC-treated (A-3) groups. Intravenously administrated PlGF-hMSCs accumulated in the ischaemic lesions (B-1, C-1). Confocal images show the differentiation of the transplanted cells (LacZ in red; B-1 and C-1) into neurons (NeuN in green; B-2) or astrocytes (GFAP in green; C-2). NeuN localizes primarily in the nucleus. Panels B-3 and C-3 confirm the co-labelling of LacZ/NeuN or LacZ/GFAP in the cells, respectively. Higher power confocal images are shown in insets. Scale bar = 5mm (A), 50 μm (B and C) and 30 μm (inserts).

Fig. 5

Cells with DNA fragmentation were present in the ischaemic penumbra 7 days after MCAO. Fewer TUNEL-positive cells were detected in rats treated with PlGF-hMSCs (A-4) than those in hMSC-treated animals (A-3) and medium-treated animals (A-2) in the ischaemic boundary zone (yellow, TUNEL-positive: red, nuclear stain). Normal brain showed no TUNEL-positive cell (A-1). These results are summarized in B. Scale bar = 100 μm (A).

Fig. 6

Seven days after MCAO, the angiogenesis in boundary zone was analysed using a three-dimensional analysis system. A-1 shows the three-dimensional capillary image with systemically perfused FITC-dextran in the normal rat brain. The total volume of the microvessels in the sampled lesion site decreased 7 days after MCAO (A-2), but was greater in the hMSC-treated group (A-3), and greater still in the PlGF-hMSC-treated group (A-4). These results are summarized in B. Scale bar = 150 μm (A).

Fig. 7

Intravenously administrated PlGF-hMSCs differentiated into the endothelial lineage. Confocal images show the transplanted cells (LacZ in red; A-1, B-1) and endothelial cells (vWF in green; A-2, B-2). A-3 and B-3 confirm the co-labelling of LacZ/vWF in the cells. B-4 also demonstrates z-axis projections of the image. The cell density of vWF-positive cells is shown in C. D demonstrates the cell density of vWF-positive cells co-localized with LacZ. Scale bar = 50 μm (A and B).

Fig. 8

Assessment of brain ischaemia-induced functional deficits using an LPT (A) and the treadmill stress test (B). Three hours after MCAO, but before intravenous cell delivery, there was no statistical difference in limb placement score among three groups. Twenty-four hours after MCAO, rats that received PlGF-hMSC achieved significantly higher limb placement scores compared with control rats. From 3 days on both hMSCs and PlGF-hMSCs showed greater functional improvement than medium alone, but the PlGF-hMSC group showed more improvement than the hMSC group. (B) The treadmill stress test demonstrates that the maximum speed at which the rats could run on a motor-driven treadmill was faster in the hMSC and the PlGF-hMSC-treated rats than control. Twenty-four hours after MCAO, rats that received hMSC-PlGF showed higher running speed compared with the medium alone, and the hMSC group. From three days on after MCAO, the hMSC and PlGF-hMSC groups attained higher velocities but the PlGF-hMSC group was greater. The highest level of functional improvement was attained at 7 days for both functional tests.