Serologic analysis of the mouse beta chemokine JE/monocyte chemoattractant protein-1.

Mouse monocyte chemoattractant protein-1 (MCP-1), previously termed JE, is a member of the beta chemokine gene family and a homologue of the human monocyte chemoattractant protein, MCP-1. Mouse rMCP-1 was used to immunize hamsters for the production of mAb. Seven mouse MCP-1-specific mAbs were characterized: two of these mAbs cross-reacted with the human MCP-1, as determined by ELISA. A sensitive and specific capture ELISA for MCP-1 quantitation, which allowed measurement of mouse MCP-1 levels in supernatants from cells stimulated with inflammatory agents, was developed. LPS-stimulated astrocytes produce the highest levels of MCP-1 (80 ng/ml); macrophages and mesangial cells produce lower levels of MCP-1 (2 to 14 ng/ml) after LPS stimulation. IL-1 and TNF-alpha stimulation also can induce low levels of MCP-1 production. Western blot analysis demonstrated that the predominant native form of mouse MCP-1 is a 30-kDa glycoprotein. Two mAbs (2H5 and 6C7) demonstrated dose-dependent neutralization of mouse MCP-1 chemotactic activity. To localize the epitope recognized by one of these neutralizing Abs, the mAb was used to bind a series of genetically engineered truncated variants of human MCP-1. The C-terminal residues 62 to 67 on human MCP-1 molecules seem to be critical to express the epitope recognized by the neutralizing 2H5 anti-MCP-1 mAb. However, multiple sites on the MCP-1 molecule seem to be critical for bioactivity. Thus, these Ab reagents provide a useful tool to explore the biology of the mouse MCP-1 beta chemokine.

Biochemical and biologic characterization of murine monocyte chemoattractant protein-1. Identification of two functional domains.

Monocyte chemoattractant protein-1 (MCP-1) is a member of the chemokine-beta (or C-C) family of cytokines. Murine MCP-1, first identified as the JE gene, differs from human MCP-1 in molecular size and extent of glycosylation. We have used Chinese hamster ovary cells to express recombinant murine MCP-1 and find that its predominant form is a microheterogeneous protein of M(r) approximately 25,000. Most of MCP-1's microheterogeneity is due to variable amounts of sialic acid that are terminally attached to a constant number of O-linked oligosaccharide chains per molecule. This carbohydrate, along with a small amount of N-linked carbohydrate, accounts for 50% of the apparent molecular size of murine MCP-1 and is not required for in vitro monocyte chemoattractant activity. Mutational analysis shows that most of the carbohydrate is added to a 49-amino acid C-terminal domain that is not present in human MCP-1 and is not required for in vitro biologic activity, suggesting that murine MCP-1 consists of an N-terminal domain containing monocyte chemoattractant activity and a heavily glycosylated C-terminal domain of as yet unknown function. MCP-1 produced in COS cells contains a small amount of sulfate, but Chinese hamster ovary-produced MCP-1 does not. The absence of sulfate does not alter MCP-1's in vitro chemoattractant properties. In vitro, highly purified murine MCP-1 attracts monocytes, but not neutrophils, with a specific activity similar to human MCP-1 (EC50 approximately 0.5 nM). Equilibrium binding experiments with human monocytes reveal the presence of approximately 3000 binding sites per cell with a Kd of 0.77 nM. In vivo, injection of up to 1 micrograms murine MCP-1 in a variety of murine strains induces the appearance of a sparse mixed inflammatory infiltrate. The disparity between MCP-1's in vitro and in vivo effects suggests that other factors may be required to elicit a full-blown monocyte chemotactic response to MCP-1 in vivo.