FGFs regulate Mkp3 transcripts through ERK MAP kinase in the dorsal sclerotome. (A-E) Whole-mount in situ hybridisation with Mkp3 following FGF4 bead grafts. (A) HH13 embryo, (B) transverse section through the trunk of an HH15 embryo, showing ectopic transcripts throughout somite tissue, (C-E) Mkp3 expression after FGF and pharmacological inhibitor bead implants, indicated by asterisks. (C) SU5402 (n=3/4), (D) PD184352 (n=8/8) and (E) LY294002 (n=8/10). (F) Western blot probed with anti-phospho-ERK antibody and tubulin. Protein isolated from untreated somites (-) or somites dissected after exposure to PD184352 beads (+). (G) SU5402 bead implants resulted in reduced endogenous Mkp3 expression (n=7/7). Double arrowheads indicate the `twin-stripe' expression of Mkp3. (H,I) PD184352 bead implants (H) resulted in reduced endogenous Mkp3 expression (red arrowhead n=11/16), whereas a bead soaked in LY294002 (I) did not have any effect (red arrowhead n=19/19). (J,K) In situ hybridisation for Mkp3 after injection of (J) 2 mM PD184352 into somites (n=15/15) or (K) DMSO (n=10/10). (L,M) Whole-mount in situ hybridisation for scleraxis after injection of (L) 2 mM PD184352 into somites (n=12/13) or (M) DMSO (n=10/10). Brackets in J-M show injected somites. Bead grafts and inhibitor injections were analyzed after 5 hours. (N-Q) FGF receptor and ERK MAP kinase activity is required for Mkp3 expression in somites. Whole-mount in situ hybridisation for chick Mkp3 after electroporation of a FGFR1c-EYFP fusion construct (n=5/5) (N), of pCS2⁺ encoding a FREK:Fc fusion protein (n=5/10) (O), or of pCS2⁺ encoding mouse Spry2 (n=10/15) (Q). (O) Expression of sFREK:Fc is detected using a probe against the Fc domain (red). (P) Electroporation of EGFP expression vector alone (n=11/11). GFP transcripts were detected by in situ hybridisation (red). Brackets indicate electroporated somites.

The expression of Mkp3 over time correlates with differential levels of phosphorylated ERK in response to FGF bead implants. (A-C) Whole-mount in situ hybridisation of Mkp3 in chick embryos with FGF bead grafts, harvested after a 1-hour (A), 5-hour (B) or 24-hour exposure (C). (D) Western blot of untreated somites (-) or somites exposed to an FGF4 bead (+), harvested at the times indicated and probed with anti-phospho-ERK andα -tubulin antibodies. (E) scleraxis was induced by an FGF4 bead after 5 hours. (F,G) Sections demonstrating the distribution of ectopic transcripts after bead grafting, (F) Mkp3 and (G) scleraxis. Red asterisks in C,E indicate the bead.

MKP3 regulates its own expression by regulating ERK MAP kinase. (A) Western blot of protein extracted from untreated somites (lane 1), somites electroporated with EGFP vector (lane 2) or electroporated with hMKP3-GFP vector (lane 3); 15 μg was loaded in each lane. Asterisks on the right identify the EGFP protein (lane 2) and the hMKP3-GFP fusion protein (lane 3).α -tubulin served as a loading control. (B) PCR performed on plasmid DNA with specific primers. Human MKP3-GFP plasmid was detected with hMKP3-GFP primers but not with cMkp3 primers (lane 1). Chick Mkp3 plasmid was detected with cMkp3 primers but not with hMKP3-GFP primers (lane 2). (C) RT-PCR on cDNA obtained from somites expressing EGFP vector (lane 1) or hMKP3-GFP fusion protein (lane 2). (B,C) Templates used indicated at the top, primers used indicated on the left. (D) Immunohistochemistry on a frontal section using MF20 antibody detecting myosin heavy chain (green) and dpERK anti-body detecting active phosphorylated ERK MAP kinase (red). White arrowheads indicate rostrocaudal boundaries between somites. Dorsal root ganglia indicated by a blue arrow; dorsal sclerotome region indicated by a yellow arrow. (E) Section of somites electroporated with a GFP vector (in red) and hybridised with a probe detecting MyoD (dark blue). Arrows indicate extent of tissue electroporated, including dermomyotome, myotome and dorsal sclerotome. (F-I) Manipulating dpERK levels in somites results in loss of scleraxis transcripts, the following vectors were electroporated: (F,G) hMKP3-GFP, (H,I) hMKP3^ΔKIM-GFP. Single and double in situ hybridisation of electroporated embryos with the probes indicated on each panel. (F) Mkp3, bracket indicates detection of high levels of human transcripts, which crossreact; (G) scleraxis and hMKP3-GFP detected using a GFP probe (n=19/22); (H) scleraxis and hMKP3^ΔKIM-GFP detected using a GFP probe (n=8/8); (I) GFP fluorescence.

Manipulating dpERK levels leads to loss of scleraxis expression and rib defects. (A,D,F) Whole-mount in situ hybridisation of (A,D) scleraxis (dark blue) and (F) scleraxis with GFP (red). (A) hMKP3-GFP electroporation and (D) caMEK1 electroporation resulted in loss of scleraxis transcripts (n=19/22 and n=10/14). (F) GFP electroporation had no effect (n=16/16). (B,C,E,G) Alcian Blue staining of cartilage. Loss of dpERK resulted in loss of the distal part of rib one (B, n=8/12) or rib seven (C, n=9/16), depending on the somites targeted. (E) High levels of dpERK caused loss of the distal part of rib one (n=9/15). (G) GFP alone had no effect on rib development (n=12/12). (H) Infection of thoracic somites with RCAS-sFREK:Fc, at day 10 of development, multiple ribs failed to form (n=2/16). (I) In situ hybridisation detecting RCAS gag transcripts 48 hours after infection (red). (J) Transverse section through RCAS infected somite demonstrates presence of transcripts predominantly in the dorsal somite. Asterisks indicate missing ribs (B,C,E,H) compared with rib present in control (G). dm, dermomyotome; my, myotome; sc, sclerotome.