Fig. 1. Relation between VEGF and the number of surviving DA neurons in vitro. Increased number of surviving 6-OHDA-treated DA neurons treated with VEGF was shown. Treatment with 1, 10, and 100 ng/ml of VEGF resulted in a significant increase in the number of surviving neurons although 100 ng/ml of VEGF displayed less neuroprotective effects than 1 ng/ml of VEGF. Data are shown as mean values ± SE expressed as the number of DA neurons. *P < 0.01 vs. the group with only 6-OHDA by ANOVA. **P < 0.05 vs. the group with 6-OHDA and 1 ng/ml of VEGF by ANOVA. F value; 13.9, degrees of freedom; 5.

Fig. 2. Two sizes of BHK-VEGF capsules secreted VEGF with different doses continuously for a long time. VEGF ELISA data revealed that the small BHK-VEGF capsules secreted VEGF at low doses continuously for 24 weeks after encapsulation. On the contrary, the large BHK-VEGF capsules secreted VEGF 3 to 5 times as much as the small one. Post-explant capsules, which were retrieved before sacrifice, secreted VEGF in similar fashion. The BHK-Control capsules did not secrete VEGF at any time. Data are shown as mean values ± SE expressed as ng/day. N.D.; not detected.

Fig. 3. VEGF reduced the number of amphetamine-induced rotation in a rat model of Parkinson's disease. Reduction of amphetamine-induced rotations in number was observed in rats receiving the small BHK-VEGF capsule, compared to those of rats receiving the large BHK-VEGF capsules as well as rats with the BHK-Control capsules at 2, 4, and 8 weeks after transplantation. Data are shown as mean values ± SE expressed as rotation numbers. *P < 0.01 vs. rats with the small/large BHK-Control capsule by ANOVA. **P < 0.05 vs. rats with the large BHK-VEGF capsule by ANOVA. F value; 26.0, degrees of freedom; 3.

Fig. 4. Low-dose administration of VEGF improved TH immunohistochemistry in a rat model of Parkinson's disease. (A, B) TH staining in the striatum (A) and in the SNc (B) of the non-lesioned side. (C, D) Non-specific staining in the striatum (C) and in the SNc (D) of the non-lesioned side. (E–H) TH-positive fibers in the striatum of the lesioned side. (I–L) TH-positive neurons in the SNc of the lesioned side. Much more preservation of TH-positive fibers in the striatum in the small BHK-VEGF group (G), compared to the large BHK-VEGF group (H), and the small/large BHK-Control group (E, F), Meanwhile, TH-positive neurons in the SNc in the small BHK-VEGF group (K) preserved slightly better than those in the large BHK-VEGF group (L), obviously displaying radical reduction in the small/large BHK-Control group (I, J). Bar: 40 μm.

Fig. 5. Low-dose administration of VEGF showed more neuroprotective effects on 6-OHDA-treated DA neurons in vivo than high dose. Much preservation of the density of TH-positive fibers in the striatum and slight preservation of the number of TH-positive neurons in the SNc in the small BHK-VEGF group compared to the large BHK-VEGF group as well as the small/large BHK-Control group. TH-positive fibers in the striatum were analyzed with a computerized image analysis system. Data are shown as mean values ± SE expressed as percentages of the contralateral side. *P < 0.01 by the group with the small/large BHK-Control capsule by ANOVA. **P < 0.05 by the group with the large BHK-VEGF capsule by ANOVA. ***P < 0.05 by the group with the small/large BHK-Control capsule by ANOVA. F value; 28.3, degrees of freedom; 3.

Fig. 6. High-dose administration of VEGF showed much more angiogenesis and glial proliferation. (A–D) Laminin staining, (E–H) GFAP staining. Laminin staining revealed remarkable angiogenesis in the striatum in the large BHK-VEGF group (D), compared to the small/large BHK-Control group (A/B) and the small BHK-VEGF group (C), although apparent angiogenesis was also recognized in the small BHK-VEGF group. Bar: 20 μm. (E–H) GFAP staining revealed that high-dose administration of VEGF induces the appearance of glial proliferation (H), compared to the small/large BHK-Control group (E/F) and the small BHK-VEGF group (G), although to some extent, glial proliferation was also recognized in the small BHK-VEGF group. Bar: 40 μm. (I) The merged image of laminin staining and nuclear staining revealed that the vessel wall was multilayered. Bar: 100 μm. Table: The ratio to the contralateral side of laminin-positive vessels and GFAP-positive cells in the striatum around the capsule was analyzed with a computerized image analysis system. Data are shown as mean values ± SE expressed as percentages of contralateral side. *P < 0.01 by the small/large BHK-Control group by ANOVA. **P < 0.05 by the small BHK-VEGF group by ANOVA. Laminin: F value; 240, degrees of freedom; 3. GFAP: F value; 43.6, degrees of freedom; 3.

Fig. 7. Specific gravity of the striatum in rats with the large BHK-VEGF capsule apparently reduced. SG of the striatum in rats with the large BHK-VEGF capsule reduced compared to all other groups. *P < 0.01 vs. all other groups by ANOVA. F value; 67.2, degrees of freedom; 4.