The structure of
Escherichia coli-derived rat intestinal fatty acid binding protein
(I-FABP) exhibits a beta-clam topology comprised of two, five-stranded
antiparallel beta-sheets surrounding a large solvent-filled cavity into
which the ligand binds. It also contains two alpha-helices that span
residues E15 through A32 and join beta-strands A and B. This helical
domain is conserved in all proteins of this family for which structures
have been determined. In order to assess the structural and functional
role of the helical domain, we engineered a variant of I-FABP by deleting
residues 15-31 and inserting a Ser-Gly linker after residue 14. Circular
dichroism measurements indicated that this I-FABP variant, termed D17-SG,
has a high beta-sheet content si milar to the wild-type protein.
Two-dimensional NMR spectra of D17-SG revealed patterns similar to those
observed for wild-type I-FABP, except for the selective absence of
resonances and through-space interactions assigned to the helical domain.
The D17 -SG variant was less stable to denaturant than wild-type I-FABP,
but the folding-unfolding transition was highly cooperative and
reversible. Taking into account the lower stability, the refolding
kinetics of D17-SG were essentially identical to wild-type . We conclude
that D17-SG is a helix-less, essentially all-beta-sheet variant of I-FABP
and that the helical domain is not a required element of the beta-clam
topology of I-FABP. In addition, the helical domain does not appear to
serve as a nucleation s ite for the refolding process. As shown in the
accompanying manuscript [Cistola, D.P., Kim, K., Rogl, H. & Frieden, C.
(1996)], the helices may function to regulate the kinetics and energetics
of ligand binding.
Intestinal fatty acid-binding protein (I-FABP) binds a single molecule of long-chain fatty acid in an enclosed cavity surrounded by two antiparallel beta-sheets. The structure also contains two short alpha-helices which form a cap over one end of the bin
ding cavity adjacent to the methyl terminus of the fatty acid. In this study, we employed a helix-less variant of I-FABP known as D17-SG [Kim, K., Cistola, D. P., & Frieden, C. (1996)] to investigate the role of the helical region in maintaining the inte
grity of the binding cavity and mediating the aquisition of ligand. Fluorescence and NMR experiments were used to characterize the energetic, structural and kinetic properties of fatty acid binding to this variant, and the results were compared and contr
asted with wild-type I-FABP and a single-site mutant, R106T. Remarkably, oleate bound to D17-SG with a dissociation constant of 4.5 mM, a value comparable to that for R106T and approximately 20- to 100-fold higher than wild-type I-FABP. Heteronuclear 2-
D NMR spectra for [2-13C]-palmitate complexed with D17-SG revealed a pattern nearly identical to that observed for the wild-type protein, but distinct from R106T. In addition, the ionization behavior of bound [1-13C]-palmitate and the nearest neighbor pa
tterns for [2-13C]-palmitate derived from 13C-filtered NOESY experiments were very similar for D17-SG and the wild-type protein. These results implied that the fatty acid-protein interactions characteristic of the carboxyl end of the fatty acid binding
cavity in the wild-type protein were essentially intact in the helix-less variant. In contrast, 13C-filtered NOESY spectra of [16-13C]-palmitate bound to D17-SG indicated that the fatty acid-protein interactions at the methyl end of the binding cavity we
re disrupted. As determined by stopped-flow fluorescence, the observed ligand association rates for both D17-SG and wild-type I-FABP increased with increasing oleate concentration, but only the wild-type protein exhibited a limiting value of 1000 s-1. T
his rate-limiting process was interpreted as a conformational change involving the helical region that allows the ligand access to the internal cavity. Simulation and fitting of the kinetic results yielded ligand association rates for D17-SG and wild-typ
e I-FABP that were comparable. However, the dissociation rate for wild-type protein was 16-fold lower than that for D17-SG. We conclude that the alpha-helices of I-FABP are not required to maintain the integrity of the fatty acid-binding cavity, but may
serve to regulate the affinity of fatty acid binding by selectively altering the dissociation rate constant. In this manner, conformational changes involving the alpha-helical domain may help control the transfer of fatty acids within the cell.
The rat intestinal fatty acid binding protein is an almost all beta-sheet protein that encloses a large interior cavity into which the fatty acid ligand binds. The protein contains neither cysteine nor proline. In a previous report, six site-directed muta nts were obtained, each having a single cysteine residue [Jiang, N., & Frieden, C., (1993) Biochemistry 32, 11015-11021] either in a turn or pointed into the cavity. In this report, each mutant has been unfolded in denaturant and modified with 5-iodoaceta mido-fluorescein to introduce a large, bulky, and fluorescent group into the protein at a known position. In all cases, fluorescence changes indicated that the modified protein refolded, and circular dichroism measurements suggested that the refolded prot ein appeared to be mostly beta-sheet. Denaturation curves suggest that for two mutants intermediate structures exist at denaturant concentrations well below the midpoint of the unfolding curve. For each modified, folded protein, one- and two-dimensional 1 H NMR spectra were accumulated and compared to the unmodified and wild-type proteins. While the spectra for the modified proteins showed a number of changes in chemical shifts, they were also consistent with folded proteins on the basis of the degree of c hemical shift dispersion. Of the six modified mutant proteins, two appear to have the fluorescein group located in the cavity, but only one of these did not bind fatty acid. The remaining modified proteins are capable of ligand binding.(ABSTRACT TRUNCATED AT 250 WORDS)