(A) An intact graft of GFP-hMSCs was evident at 1–3 days post-implantation. Only a few intact cells survived to 5 days and only cell remnants remained for 7–30 days. Squares demarcate location of high magnification inset. Scale bar = 100 µm. (B) Stereological cell counting confirmed only 15.1% of the 100,000 grafted hMSCs survived to 1 day and 17.3% to 3 days. Survival significantly dropped (n = 5−6; p = 0.004) to 4.5 and 3.5% at 5 and 7 days. By 15 days no intact cells were counted. * denotes significance based on ANOVA and Holms-Sidak as the post-hoc test.

hMSC implantation increased endogenous cell proliferation at 1-30 days post-implantation (A-D, I-J), but not in the contralateral, PBS-injected hemisphere (E-H, L-M). There was no increase in cell proliferation after dead hMSCs were implanted into the striatum (K,N). (O) hMSCs increased cell proliferation in the striatum from 1 to 15 days post-transplantation (n = 5−6). There were significantly more dividing, BrdU⁺ cells in the hMSC-implanted hemisphere than the PBS-injected hemisphere at 5–15 days post-implantation. There was no difference in the number of BrdU⁺ cells in either hemisphere when dead hMSCs were implanted. * and a denote p<0.05 by ANOVA with Holms-Sidak post-hoc analysis.

There was a slight increase in Nestin ⁺ NPCs to the site of the graft (A) compared to the PBS injection (D) at 7 days post-implantation. Many of these recruited Nestin⁺ NPCs remained proximal to the graft until 30 days (G compared to J). There was also an influx of NeuN⁺ neurons (B) and, to a lesser extent, βIII tubulin ⁺ neurons (C) to the hMSC graft compared to the PBS injection (E,F) at 7 days. At 30 days post-implantation there were still more NeuN⁺ (H) neurons and an overwhelming increase in the number of βIII tubulin ⁺ (I) neurons at the site of the graft compared to the PBS-injected, contralateral hemisphere (K,L). Some BrdU⁺ cells co-labeled with Nestin at 7 days (inset image A) and 30 days (inset image G) post-implantation. A few BrdU⁺ cells were NeuN+ at 7 days post-implantation (inset image B). This number had increased by 30 days post-implantation (inset image H). No BrdU+ cells had matured into βIII tubulin⁺ neurons at 7 days post-transplantation (inset image C). By 30 days, many BrdU⁺ cells did co-express βIII tubulin (inset image I). Squares demarcate location of high magnification inset. Scale bar = 50 µm. Arrows = co-labeled cells, arrowheads = single labeled cells.

There was a substantial increase in FGF-2 signaling in the hMSC-implanted hemisphere (A, I) compared to the contralateral, PBS-injected hemisphere (E, M) at 7 and 30 days post-implantation. There was no increase in CNTF signaling at 7 days post-implantation (B) compared to the PBS-injected hemisphere (F) although a slight increase was detected by 30 days post-implantation (J compared to N). There was an increase in VEGF signaling in the hMSC-implanted hemisphere at both 7 (C) and 30 (K) days post-implantation compared to the PBS-injected hemisphere (G,O). NGF signaling appeared to be increased similarly in both the hMSC- and the PBS-injected hemispheres 7 days post-implantation (D,H). By 30 days post-implantation, NGF was still slightly increased in the hMSC-implanted hemisphere (L) compared to the contralateral, PBS-injected hemisphere (P). Scale bar = 50 µm.

Serial sections throughout the striatum were used for striatal volume analysis. Serial sections of the striatum are outlined in both HD mice receiving MSCs or PBS (Ai–Av) and control WT mice (Bi–Bv). There was an apparent increase in striatal volume after MSC implantation at 8 weeks compared to the contralateral, PBS injected hemisphere (Ai–Av). There was no apparent difference in the size of the striatum in the opposite hemispheres in WT mice (Bi–Bv). Scale bar = 1 mm. (C) Stereological analysis at 1 month post-implantation revealed a significant increase in striatal volume after hMSC implantation compared to PBS injection (n = 5−6; *). The PBS-injected striatum was significantly smaller than both the MSC-implanted (a) and PBS WT control striatum (b). There was no difference between the two hemispheres in the WT control group. *, a, and b denote p<0.05 by ANOVA with Holms-Sidak post-hoc analysis.