CXCL12 and Wnt5A can increase T-cell migration in an animal model. EL4 cells treated with CXCL12 (100 ng/mL, 12 hours) demonstrate an increase in Wnt5A expression (A). Treatment of EL4 cells with rWnt5A (50 ng/mL, 16 hours) increases migration in response to CXCL12 (B). AMD3100 (1 μM) inhibits Wnt5A stimulation of CXCL12-induced migration; *P < .01 (B). In vivo data demonstrate that in control animals, after the injection of low numbers of EL4 cells, there are very few metastasizing EL4 cells (C), as demonstrated by CD3 staining, compared with the increased number of CD3-positive EL4 cells in the CXCL12-treated (D arrows) and Wnt5A-treated animals (data not shown). The differences in the density of the livers can be clearly seen, where there are clear spaces surrounding the tumor cells in panel D. Sections from control, CXCL12-, and Wnt5A-treated animals were then costained for Wnt5A and CD3, and there is very little infiltration of Wnt5A⁺ CD3⁺ EL4 cells in controls (E) compared with the increased EL4 infiltration observed in the livers from CXCL12 (F)− and Wnt5A (G)−treated mice. (H) The section is from a Wnt5A-treated animal, incubated with no primary antibody, used as a negative control for staining, in which infiltrating cells are still visible (arrows), but do not stain positive for CD3 or Wnt5A. Images were taken at 40× (C-D), and 64× (E-H). No counterstaining with hematoxylin was performed in these sections. Slides were viewed with a Zeiss Axiovert 2000 microscope using 40×/0.60 KorrPh2 with 1.6× amplification and Permount medium (Fisher Scientific). Stains were done as described in “Methods.” Images were acquired using a Zeiss AxioCam color camera (model 412-312) and Axiovision Version 3.1 (Zeiss) image acquisition software. Images were processed using Adobe Photoshop Version 8.0 (Adobe Systems).