K14.GFP+ cells have greater metastatic potential that K14.GFP− cells <i>in vivo</i>.

<p>(A) Primary tumor diameters from mice injected with 4T1 control, or K14.GFP reporter cell lines (K14.GFP+ or K14.GFP−), measured over the course of the experiment. <i>n</i> = 8 (control), 7 (K14.GFP+), and 8 (K14.GFP−); *<i>p</i> < 0.05, **<i>p</i> < 0.001 by one-way ANOVA followed by Newman-Keuls multiple comparisons posttest. (B) Final tumor masses measured after mice were euthanized. (C) Lung metastases were quantified by measuring the average percent area of lung tissue occupied by the tumor in 5 sections (5-μm thick) cut at 200-μm intervals. (D) Representative images of H&E-stained lung sections containing metastases from mice injected with K14.GFP+ or K14.GFP− cell lines. Arrows indicate metastasis; scale bar 1 mm. (E) Correlation analysis of primary tumor mass and lung metastasis. Scatter plot of the percent tumor area in lung tissue compared to the primary tumor mass for each mouse analyzed (<i>n</i> = 23). The slope does not significantly differ from zero by linear regression analysis (<i>p</i> = 0.1097). Quantification of percent of positive cells for Ki67 (F) and CC3 (G) in tumors from K14.GFP+ and K14.GFP− cells. Shown are means ± SD of quantifications of whole tumor sections. (H) Final tumor masses measured after mice were euthanized. For DT treatment, the mice were injected i.p. with DT (25 mg/kg) on days 7, 9, 11, and 13. <i>n</i> = 7 (K14.GFP+; no DT), 4 (K14.GFP+; with DT), 8 (K14.GFP−; no DT), and 4 (K14.GFP−; with DT). <i>p</i> = 0.0425 by unpaired <i>t</i> test. (I) Lung metastases were quantified for the mice described in H. Statistical analysis for (B), (C), and (I) was calculated by one-way ANOVA followed by Tukey’s multiple comparisons posttest; *<i>p</i> < 0.05 or n.s. (J) Fluorescent IHC was performed for K14 and GFP on primary tumors generated from K14.GFP+ cell lines either DT− or DT treated (“DT+”) as described in (H); scale bar 40 μm. (K) Same staining as described in J was carried out on metastatic lung of K14.GFP+-injected mice; scale bar 20 μm. (L) Immunoblots of lysates from primary tumors were analyzed for GFP expression. Every lane represents a different tumor. Blots were also probed with antibodies for ERK1/2 as loading control. If not otherwise indicated, all graphs show mean ± SEM. DT, diphtheria toxin; ERK1/2, extracellular signal-regulated kinase 1/2; GFP, green fluorescent protein; H&E, hematoxylin and eosin; IHC, immunohistochemistry; i.p., intraperitoneally; K, cytokeratin; n.s., not significant