Structure of a TLR4-interacting SPA4 peptide

Abstract

We have recently identified a Toll-like receptor (TLR4)-interacting SPA4 peptide encoding amino acids: GDFRYSDGTPVNYTNWYRGE, a shorter region of human surfactant protein-A (SP-A). The SPA4 peptide suppressed lipopolysaccharide-induced inflammation (JPET 2011, Innate Immun 2013). In this report, we examined the structure of synthetic SPA4 peptide in solution by circular dichroism (CD) and nuclear magnetic resonance (NMR) spectroscopy. The CD analysis revealed that the SPA4 peptide is composed of ∼35% beta sheet and <5% alpha helix. We used solution NMR to solve the structure of the SPA4 peptide. We calculated NMR structures using Nuclear Overhauser Enhancement (NOE) distance restraints. The superposition of the low energy structures indicated that the central 6-14 amino acids "SDGTPVNYT" of the 20mer SPA4 peptide form a turn, and amino acids on either side (GDFRY and NWYRGE) conform to flexible arms. Furthermore, thermal denaturation experiments demonstrated the structural flexibility of the peptide. The NMR structures of the SPA4 peptide align well with the homologous region within the available structure of rat SP-A and a computationally-docked model of SP-A-TLR4-MD2 protein complex. Together, our results support the structural adaptability of SPA4 peptide for binding to TLR4.

Keywords: Surfactant protein-A-derived peptide; Toll-like receptor-4; structure.

Figure 1

Representative CD spectrum of SPA4 peptide suspended in water or 75% methanol in water. The spectra recorded at 20°C are in units of mean residue ellipticity versus wavelength.

Figure 2

Assessment of thermal stability of the SPA4 peptide. Thermal denaturation of the SPA4 peptide in water was determined by changes in the CD signal intensity at a wavelength of 220 nm. Tm refers to the midpoint of the transition between folded and unfolded states, and was calculated as described in Materials and Methods.

Figure 3

All of the spin systems were assigned using 80 ms mixing TOCSY (A). Sequence-specific assignments were made using 200 ms and 400 ms mixing NOESY. 200 ms mixing NOESY is shown in (B).

Figure 4

Superposition of 25 low energy NMR structures of SPA4 peptide. The structures were superposed by aligning the C-alpha carbons of the amino acid residues “SDGTPVNYT” (in blue). The superposition was performed using the PyMOL molecular graphics program.

Figure 5

The distance in Å between the first- G and last- E residues in the individual 25 NMR structures of SPA4 peptide was assessed using the PyMOL molecular graphics program. NMR structures are shown as a ribbon diagram. The structural alignment is similar to those shown in Figure 4.

Figure 6

Hydrophobic domains within the SPA4 peptide. Orientation is rotated 180 degrees relative to the standard rotation along the vertical axis. The color code is: blue, polar regions; white, in between; orange, hydrophobic regions. The Chimera program was used to obtain information about hydrophobic or hydrophilic regions within the NMR structures of the SPA4 peptide.

Figure 7

Alignment of the region homologous to SPA4 peptide within the available structure of rat SP-A (PDB ID: 1R13; shown in black) with individual NMR structures. The alignment of SP-A region homologous to SPA4 peptide with (A) all structures superposed and (B) individual 25 low energy NMR structures of SPA4 peptide in solution. NMR structures are shown as ribbon diagrams.

Figure 8

In silico analysis of SPA4 peptide structure aligned to homologous regions within SP-A in a computationally-docked model of the SP-A-TLR4-MD2 complex. Part of the computationally-docked model of the SP-A-TLR4-MD2 complex is taken from Ref. (7), and is reproduced with permission from the American Society for Pharmacology and Experimental Therapeutics. (A) Alignment of the lowest energy NMR structure of SPA4 peptide (in red) with SP-A (in green) within a computationally-docked model of SP-A-TLR4-MD2 complex. (B) Previously predicted amino acids of the homologous region of rat SP-A (D, S, Y, T, P, G, green sticks) and TLR4 (E, Q, E, K, F, violet sticks). The amino acids (D, N, Y, T, R, G) of the SPA4 peptide are derived from human SP-A, and are shown as red sticks within the computationally-docked model. (C) The distances between the closest amino acids of SPA4 peptide (D, N, Y, T, R, G) and TLR4 (E, Q, E, K, F) were measured using the PyMOL program. The regions surrounding the interface have been made transparent for better viewing. The distances between the intermolecular residues are shown in Å units in yellow.