FIGURE 6.

Amino acids in the CC′ loop region of the N-terminal domain of PSG1 are essential for induction of TGF-β₁ from RAW264.7 macrophages. A, recombinant proteins used for these studies were separated on a 4–20% NuPAGE gel and stained with GelCode Blue. Lane 1, PSG1-A2-B2-Fc; lane 2, PSG1-N-Fc; lane 3, PSG1-N-CC′-Fc; lane 4, PSG1-N-C′C″-Fc; lane 5, PSG1-N-YHY-Fc; lane 6, PSG1-N-LYHY-Fc. B, TGF-β₁ secretion following incubation of RAW264.7 murine macrophages with 3 μg/ml control Fc protein (cntrl-Fc), PSG1-A2-Fc, and PSG1-N-Fc or 6 μg/ml PSG1-A2-B2-Fc to achieve concentrations equimolar to the single-domain proteins. C, sequence alignment of the CC′ and C′C″ loop regions of CEACAM5, PSG1, and PSG1 N-terminal domain mutants. Residues conserved in CEACAM5 and PSG1 are highlighted in gray. The C, C′, and C″ β-strands are indicated with green arrows above the corresponding sequence. Residues in the PSG1 N-terminal domain mutated to the residues corresponding to the same position in CEACAM5 are colored red. D, TGF-β₁ secretion of RAW264.7 macrophages following incubation with increasing concentrations of PSG1-N-CC′-Fc, PSG1-N-C′C″-Fc, or control Fc protein. E, RAW264.7 macrophages were incubated with 12 μg/ml wild-type PSG1 N-terminal domain (WT), PSG1-N-Y42N/H43R/Y44Q mutant (YHY→NRQ), PSG1-N-L41G/Y42N/H43R/Y44Q mutant (LYHY→GNRQ), or control Fc protein. For B–E, TGF-β₁ following acid activation was measured in the supernatants by ELISA as described under “Experimental Procedures.” *, p ≤ 0.05 by Student's t test; ns, not significant. F, ribbon representation of the structural model of the PSG1 N-terminal domain. β-Sheets are shown in green and labeled with uppercase white letters. The α-helix between strands E and F is shown in orange, and loops are shown in gray. Residues ⁴¹LYHY⁴⁴ of the CC′ loop are labeled and highlighted in red, and their side chains are shown in stick representation. The CC′ and C′C″ loops and the N and C termini of the protein domain are indicated.