Figure 4.

Kinetics of velocity response. Velocity measurements were averaged for KM101 cells (A) and MG-63 cells (B) at coating concentrations of 10 ug/ml (Left Panels) and 100 ug/ml (Right Panels). Each point represents the average for 8 images (acquired every 30 minutes over four-hour time intervals) from triplicate wells in each of 3 experiments. Surface coatings are indicated as follows: laminin (triangles), collagen I (open squares), collagen IV (closed squares), and mock-coated plastic (open circles). Linear trend lines were added, and statistical analysis was applied as described (Methods). For KM101 cells at dose level 10 ug/ml (upper left panel), the estimated velocity model is: Collagen I: v = 0.3327+0.0015*time; Collagen IV: v = 0.2641-0.00002*time; Laminin: v = 0.4532+0.0018*time; Plastic: v = 0.2938-0.0007*time. The p-values for the intercepts and slopes are all significantly positive (p < 0.0001) except that the slope for Collagen4 is nonsignificant (p = 0.9217). The overall test for the equivalence of the starting velocity is significant (p < 0.0001), which means not all the starting velocities are equal, and the Bonferroni adjusted pair wise tests show that all the initial velocities are significantly different. The overall test for the equivalence of the slope is significant (p < 0.0001), which means that not all the slopes are equal, and the Bonferroni adjusted pair wise tests show that only Collagen I and Laminin have equal slope (p = 0.5100). For KM101 cells at coating concentrations of 100 ug/ml (upper right panel), the p-values for the intercepts and slopes are all significant. The overall test for the equivalence of the initial velocity is significant (p < 0.0001), which means not all the initial velocities are equal, and the Bonferroni adjusted pairwise tests show that only Collagen I and Collagen IV have the same initial velocity (p = 0.0202). The overall test for the equivalence of the slope is significant (p < 0.0001), which means not all the slopes are equal, and the Bonferroni adjusted pairwise tests show that only Collagen IV and Plastic have the same slope (p = 0.8705). For MG63 cells at dose level 10 ug/ml (lower left panel), the p-values for the intercepts and slopes are all significant (all p < 0.0001). The overall test for the equivalence of the initial velocity is significant (p < 0.0001), which means not all the initial velocities are equal, and all the Bonferroni adjusted pair wise tests show significant difference. The overall test for the equivalence of the slope is significant (p < 0.0001), which means not all the slopes are equal. The Bonferroni adjusted pair wise tests show that Collagen I and Laminin have the same slopes (p = 0.5176), Collagen IV and Plastic have the same slope (p = 0.6623), and other pairs have significantly different slopes. For MG63 cells at 100 ug/ml (lower right panel), the p-values for the intercepts and slopes are all significantly positive (p < 0.0001). The overall test for the equivalence of the starting velocity is significant (p < 0.0001), which means not all the starting velocities are equal, and the Bonferroni adjusted pair wise tests show that only Collagen IV and Plastic have the same initial velocity (p = 0.0428).). The overall test for the equivalence of the slope is significant (p = 0.0084), which means that not all the slopes are equal, and the Bonferroni adjusted pair wise tests show that only Collagen IV and Plastic have significantly different slope (p = 0.0042), others are not significant.

Bahnson et al. BMC Cell Biology 2005 6:19   doi:10.1186/1471-2121-6-19
Download authors' original image