Anti-Tumor Necrosis Factor With a Glyco-Engineered Fc-Region Has Increased Efficacy in Mice With Colitis.

BACKGROUND & AIMS: Although tumor necrosis factor (TNF) antagonists reduce many clinical features of inflammatory bowel disease, complete mucosal healing occurs in fewer than 50% of patients. The Fc-region of monoclonal antibodies against TNF has immunosuppressive properties via effects on macrophage polarization. We examined the interaction between the anti-TNF Fc-region and Fcγ receptors (FcγR), and whether the absence of the Fc core fucose (which increases binding to FcγRIIIa) increases the efficacy of anti-TNF in mice with colitis. METHODS: We generated Rag1-/- mice that lack all activating FcγRs (FcγRI, FcγRIII, and FcγRIV; called FcγR-/-Rag1-/- mice). We produced hypo-fucosylated antibodies against mouse and human TNF (adalimumab). Colitis was induced in mice by transfer of CD4+CD45RBhi to FcγR-/-Rag1-/- or Rag1-/- littermates; mice were given different antibodies against TNF or isotype (control) antibodies and disease activity index scores were determined. Colon tissues were collected and analyzed by histology. Human peripheral blood mononuclear cells (PBMCs) were isolated from blood of healthy donors. T-cell proliferation and proportions of CD206+ (immune regulatory) macrophages were measured in mixed lymphocyte reactions. Human PBMCs were genotyped for FCGR3A158 (the FcγRIIIa-158F allotype displays a lower Fc binding affinity) using the TaqMan single nucleotide polymorphism genotype assay. RESULTS: Rag1-/- mice with colitis given anti-TNF had near complete mucosal healing and Rag1-/- mice given an isotype control antibody developed severe colitis. In contrast, FcγR-/-Rag1-/- mice were refractory to the effects of anti-TNF: their histological colitis scores were as severe as those from FcγR-/-Rag1-/- mice given a control antibody. Colons from Rag1-/- mice that received anti-TNF had an increased number of CD206+ macrophages compared with Rag1-/- mice given control antibody; in FcγR-/-Rag1-/- mice given anti-TNF these numbers were as low as FcγR-/-Rag1-/- given the control antibody. In human PBMCs, anti-TNF increased the number of CD206+ macrophages: this required expression of FcγRIIIa; numbers of these cells were reduced in PBMCs with the low-affinity FcγRIIIa-158F genotype. A hypo-fucosylated form of adalimumab bound human FcγRIIIa with a higher affinity than control adalimumab. When hypo-fucosylated adalimumab was added to PBMCs, a larger number of CD206+ macrophages formed and T-cell proliferation was reduced, compared with addition of a control adalimumab. Hypo-fucosylated adalimumab increased the number of CD206+ macrophages in PMBCs that expressed the low-affinity FcγRIIIa. In mice with colitis, hypo-fucosylated anti-TNF significantly increased the number of CD206+ macrophages in the colon compared with control anti-TNF and was more effective in reducing colitis severity as measured by histology. CONCLUSIONS: In a study of the in vitro and in vivo mechanisms of anti-TNF, we found FcγR engagement by anti-TNF to be required for reduction of colitis in mice and development of CD206+ macrophages. A hypo-fucosylated form of anti-TNF binds FcγRIIIa with higher affinity and induces development of CD206+ macrophages in human PBMCs, especially PBMCs that express low-affinity FcγRIIIa. Hypo-fucosylated anti-TNF might be more effective in patients with inflammatory bowel disease.

Long-range modulation of chain motions within the intrinsically disordered transactivation domain of tumor suppressor p53.

The tumor suppressor p53 is a hub protein with a multitude of binding partners, many of which target its intrinsically disordered N-terminal domain, p53-TAD. Partners, such as the N-terminal domain of MDM2, induce formation of local structure and leave the remainder of the domain apparently disordered. We investigated segmental chain motions in p53-TAD using fluorescence quenching of an extrinsic label by tryptophan in combination with fluorescence correlation spectroscopy (PET-FCS). We studied the loop closure kinetics of four consecutive segments within p53-TAD and their response to protein binding and phosphorylation. The kinetics was multiexponential, showing that the conformational ensemble of the domain deviates from random coil, in agreement with previous findings from NMR spectroscopy. Phosphorylations or binding of MDM2 changed the pattern of intrachain kinetics. Unexpectedly, we found that upon binding and phosphorylation chain motions were altered not only within the targeted segments but also in remote regions. Long-range interactions can be induced in an intrinsically disordered domain by partner proteins that induce apparently only local structure or by post-translational modification.

Backbone-driven collapse in unfolded protein chains.

Collapse of unfolded protein chains is an early event in folding. It affects structural properties of intrinsically disordered proteins, which take a considerable fraction of the human proteome. Collapse is generally believed to be driven by hydrophobic forces imposed by the presence of nonpolar amino acid side chains. Contributions from backbone hydrogen bonds to protein folding and stability, however, are controversial. To date, the experimental dissection of side-chain and backbone contributions has not yet been achieved because both types of interactions are integral parts of protein structure. Here, we realized this goal by applying mutagenesis and chemical modification on a set of disordered peptides and proteins. We measured the protein dimensions and kinetics of intra-chain diffusion of modified polypeptides at the level of individual molecules using fluorescence correlation spectroscopy, thereby avoiding artifacts commonly caused by aggregation of unfolded protein material in bulk. We found no contributions from side chains to collapse but, instead, identified backbone interactions as a source sufficient to form globules of native-like dimensions. The presence of backbone hydrogen bonds decreased polypeptide water solubility dramatically and accelerated the nanosecond kinetics of loop closure, in agreement with recent predictions from computer simulation. The presence of side chains, instead, slowed loop closure and modulated the dimensions of intrinsically disordered domains. It appeared that the transient formation of backbone interactions facilitates the diffusive search for productive conformations at the early stage of folding and within intrinsically disordered proteins.

Transient-state kinetic analysis of transcriptional activator·DNA complexes interacting with a key coactivator.

Several lines of evidence suggest that the prototypical amphipathic transcriptional activators Gal4, Gcn4, and VP16 interact with the key coactivator Med15 (Gal11) during transcription initiation despite little sequence homology. Recent cross-linking data further reveal that at least two of the activators utilize the same binding surface within Med15 for transcriptional activation. To determine whether these three activators use a shared binding mechanism for Med15 recruitment, we characterized the thermodynamics and kinetics of Med15·activator·DNA complex formation by fluorescence titration and stopped-flow techniques. Combination of each activator·DNA complex with Med15 produced biphasic time courses. This is consistent with a minimum two-step binding mechanism composed of a bimolecular association step limited by diffusion, followed by a conformational change in the Med15·activator·DNA complex. Furthermore, the equilibrium constant for the conformational change (K(2)) correlates with the ability of an activator to stimulate transcription. VP16, the most potent of the activators, has the largest K(2) value, whereas Gcn4, the least potent, has the smallest value. This correlation is consistent with a model in which transcriptional activation is regulated at least in part by the rearrangement of the Med15·activator·DNA ternary complex. These results are the first detailed kinetic characterization of the transcriptional activation machinery and provide a framework for the future design of potent transcriptional activators.

S100 proteins interact with the N-terminal domain of MDM2.

S100 proteins interact with the transactivation domain and the C-terminus of p53. Further, S100B has been shown to interact with MDM2, a central negative regulator of p53. Here, we show that S100B bound directly to the folded N-terminal domain of MDM2 (residues 2-125) by size exclusion chromatography and surface plasmon resonance experiments. This interaction with MDM2 (2-125) is a general feature of S100 proteins; S100A1, S100A2, S100A4 and S100A6 also interact with MDM2 (2-125). These interactions with S100 proteins do not result in a ternary complex with MDM2 (2-125) and p53. Instead, we observe the ability of a subset of S100 proteins to disrupt the extent of MDM2-mediated p53 ubiquitylation in vitro.

Converting inactive peptides into potent transcriptional activators.

Significant efforts have been devoted to the development of artificial transcriptional activators for use as mechanistic tools, as therapeutic agents, and for biomanufacturing applications. One of the primary challenges has been the development of artificial activators that exhibit potency in cells comparable to that of endogenous activators; the vast majority function only moderately in the cellular context. Here we demonstrate that the superimposition of two distinct binding modes, a masking interaction and an interaction with the transcriptional machinery, has a profoundly positive effect on the cellular activity of artificial activators, with up to 600-fold enhancement observed. Incorporation of this feature into future generations of small molecule transcriptional activators should increase their nuclear uptake and facilitate their accessibility to their target proteins, thus significantly augmenting both their activity and utility.

Functional specificity of artificial transcriptional activators.

Misregulated transcription is linked to many human diseases, and thus artificial transcriptional activators are highly desirable as mechanistic tools and as replacements for their malfunctioning natural counterparts. We previously reported two artificial transcriptional activation domains obtained from synthetic peptide libraries screened for binding to the yeast transcription protein Med15(Gal11). Here we demonstrate that the transcriptional potency of the Med15 ligands is increased through straightforward structural alterations. These artificial activation domains upregulate transcription via specific Med15 binding interactions and do not function in mammalian cells, which lack Med15. This functional specificity stands in contrast to most natural or artificial activation domains that function across all eukaryotic cell types. The results indicate that the screening strategy holds excellent promise for identifying peptide and small molecule transcriptional activators that function by unique mechanisms with advantageous specificity properties.

Targeting the transcriptional machinery with unique artificial transcriptional activators.

The link between a growing number of human diseases and misregulation of gene expression has spurred intense interest in artificial transcriptional activators that could be used to restore controlled expression of affected genes. To expand the repertoire of activation domains available for the construction of artificial transcriptional regulators, a selection strategy was used to identify two unique activation domain motifs. These activation domains bear little sequence homology to endogenous counterparts and bind to unique sites within the transcriptional machinery. A comparison with two well-characterized activation domains, VP2 and P201, demonstrated for the first time that functional potency is not solely dictated by binding affinity. Finally, the selection strategy described is readily applicable to the identification of small molecule activation domains.