Enforced SDF-1 expression by BMSSCs enhances their osteogenic potential. (A) The retroviral packaging line PT67 was used to transduce secondary BMSSC cultures, derived from 3 different bone marrow aspirates, with a pLNCX2 construct containing either human SDF-1 cDNA or vector alone. Stable multicolony-derived high SDF-1–expressing BMSSCs and corresponding control lines were generated after G418 selection. Triplicate samples of tissue culture supernatant were assessed for SDF-1α levels using a commercially available ELISA kit. The data represent the mean values ± standard errors generated from 3 different high SDF-1–expressing BMSSC cell lines versus the corresponding controls. (B) Single-cell suspensions of each of the transduced BMSSC lines were mixed with hydroxyapatite (HA/TCP) particles and then implanted subcutaneously into NOD/SCID mice. The images represent cross-sections of 8-week-old harvested transplants of new bone (b) formed by high SDF-1–expressing BMSSCs (SDF-1) and control cell lines (LNCX2) stained with hematoxylin & eosin (× 200). Images were captured with an Olympus BX50 light microscope (Olympus, Tokyo, Japan) equipped with an Olympus D11 digital camera. Magnification ×200. (C) Each graph represents a different high SDF-1–expressing BMSSC line and corresponding control cell line derived from 3 different bone marrow donors. The level of new bone formation is expressed as a percentage of the total tissue surface area analyzed from 12 representative tissue sections, using Scion Imaging software. The data represent the mean values ± standard errors from duplicate transplants. Statistical differences (^*) of P < .05 between the SDF-1 high-expressing BMSSC lines and corresponding controls were determined using the unpaired t test.