Rational design of novel, potent small molecule pan-selectin antagonists.

This report describes the first results of a rational hit-finding strategy to design novel small molecule antiinflammatory drugs targeting selectins, a family of three cellular adhesion molecules. Based on recent progress in understanding of molecular interaction between selectins and their natural ligands as well as progress in clinical development of synthetic antagonists like 1 (bimosiamose, TBC1269), this study was initiated to discover small molecule selectin antagonists with improved pharmacological properties. Considering 1 as template structure, a ligand-based approach followed by focused chemical synthesis has been applied to yield novel synthetic small molecules (MWr < 500) with a trihydroxybenzene motif, bearing neither peptidic nor glycosidic components, with nanomolar in vitro activity. Biological evaluation involves two kinds of in vitro assays, a static molecular binding assay, and a dynamic HL-60 cell attachment assay. As compared to controls, the novel compounds showed improved biological in vitro activity both under static and dynamic conditions.

The functional interaction of the beta 2 integrin lymphocyte function-associated antigen-1 with junctional adhesion molecule-A is mediated by the I domain.

Binding of the beta(2) integrin LFA-1 (alpha(L)beta(2)) to junctional adhesion molecule-A (JAM-A) has been shown to enhance leukocyte adhesion and transendothelial migration. This is mediated by the membrane-proximal Ig-like domain 2 of JAM-A; however, the location of the JAM-A binding site in LFA-1 has not been identified. We have deleted the I domain in the alpha(L) subunit of LFA-1 and expressed this alpha(L) mutant in alpha(l)-deficient Jurkat J-beta(2).7 cells to demonstrate that the I domain of LFA-1 is crucial for their adhesion to immobilized JAM-A. This was substantiated by blocking the stimulated adhesion of wild-type Jurkat T cells or monocytic Mono Mac 6 cells to JAM-A using the I domain-directed mAb TS1/22 or the small molecule antagonist BIRT 377, which stabilizes the low-affinity conformation of the I domain. The immobilized LFA-1 I domain locked in the open high-affinity conformation was sufficient to support binding of transfected Chinese hamster ovary cells expressing JAM-A. Solid-phase binding assays confirmed a direct interaction of recombinant JAM-A with the immobilized locked-open I domain. These data provide the first evidence that the I domain of LFA-1 contains a functional binding site for JAM-A.

Single amino acids in the lumenal loop domain influence the stability of the major light-harvesting chlorophyll a/b complex.

The major light-harvesting complex of photosystem II (LHCIIb) is one of the most abundant integral membrane proteins. It greatly enhances the efficiency of photosynthesis in green plants by binding a large number of accessory pigments that absorb light energy and conduct it toward the photosynthetic reaction centers. Most of these pigments are associated with the three transmembrane and one amphiphilic alpha helices of the protein. Less is known about the significance of the loop domains connecting the alpha helices for pigment binding. Therefore, we randomly exchanged single amino acids in the lumenal loop domain of the bacterially expressed apoprotein Lhcb1 and then reconstituted the mutant protein with pigments in vitro. The resulting collection of mutated recombinant LHCIIb versions was screened by using a 96-well-format plate-based procedure described previously [Heinemann, B., and Paulsen, H. (1999) Biochemistry 38, 14088-14093], enabling us to test several thousand mutants for their ability to form stable pigment-protein complexes in vitro. At least one-third of the positions in the loop domain turned out to be sensitive targets; i.e., their exchange abolished formation of LHCIIb in vitro. This confirms our earlier notion that the LHCIIb loop domains contribute more specifically to complex formation and/or stabilization than by merely connecting the alpha helices. Among the target sites, glycines and hydrophilic amino acids are more prominently represented than hydrophobic ones. Specifically, the exchange of any of the three acidic amino acids in the lumenal loop abolishes reconstitution of stable pigment-protein complexes, suggesting that ionic interactions with other protein domains are important for correct protein folding or complex stabilization. One hydrophobic amino acid, tryptophan in position 97, has been hit repeatedly in independent mutation experiments. From the LHCIIb structure and previous mutational analyses, we propose a stabilizing interaction between this amino acid and F195 near the C-proximal end of the third transmembrane helix.