Roles of st14a and par2b in mmp9, mmp13, and il1b expression and effects of Mmp inhibition in hai1a MO embryos. (A) Controls, hai1a morphants, and mutant embryos with and without St14a or Par2b deficiency were collected at 30 hpf, and RNA was analyzed by quantitative RT-PCR. Expression relative to controls is shown (mean ± SEM, n = 3). (B) Control and hai1a morphant periderm-nuclear-EGFP embryos with or without mmp9 deficiency or mmp9/13 inhibition were imaged as in Fig. 3 (B–D). The percentage of cells without obvious nuclear fragmentation extruding in the field imaged (mean ± SEM; n = 10) is shown. Mmp knockdown and inhibition were without effect on increased extrusion in hai1a morphants by one-way ANOVA and Bonferroni posttest. NS, not significant. (C–E) Control and hai1a morphant basal-layer-LifeActGFP embryos with or without Mmp9/13 inhibition were imaged and analyzed as in Fig. 4; mean ± SEM (n = 10 embryos/condition) is shown. Approximately 300 cell–cell contacts were analyzed per condition. Two-way ANOVA and Tukey posttest were used to analyze C–E. In C, the hai1a MO group was different from all other groups (*, P < 0.001). (D and E) Control and hai1a morphants periderm-nuclear-EGFP embryos with and without Mmp9 or Mmp9/13 inhibition were incubated with BrdU and analyzed as in Fig. 5. The percentage of BrdU-positive periderm cells (D) and basal layer cells (E) is shown (mean ± SEM; n = 3). In D, the hai1a MO groups were different from the control groups (*, P < 0.0001), and the Mmp9/13 inhibitor hai1a MO group was different from all other groups (*, P < 0.0001). In E, the hai1a MO without Mmp inhibitor treatment group was different from all other groups (*, P < 0.0001), and no other groups differed from control.