Tumor Necrosis Factor: Biosynthetic Regulation, Inhibition, and Mechanism of Action on Cellular Targets

Bruce A. Beutler, M.D.—Associate Investigator
Dr. Beutler is also Associate Professor of Internal Medicine at the University of Texas Southwestern Medical Center at Dallas. After receiving his M.D. degree from the University of Chicago (Pritzker School of Medicine), he served as an intern and resident at the Southwestern Medical Center. His postdoctoral fellowship with Anthony Cerami was completed at the Rockefeller University, which he left as an assistant professor to assume a position at the Southwestern Medical Center.

SEVERE infections may lead to hypotension (low blood pressure); disturbances of metabolism; faulty blood clotting; and injury to such vital organs as the kidneys, liver, lungs, and adrenal glands. It might be supposed that these problems are caused by toxins released from the invasive pathogen (usually a bacterium, fungus, virus, or other parasite). In fact, although microbial toxins do initiate the shock syndrome just described, they have an indirect mode of action. Host factors, normally important in defense against infection, seem largely responsible for the damage that occurs.

These host factors, or cytokines, are proteins of variable size and description that exert powerful hormonelike effects on cells of the immune system. In some cases, cytokines also affect cells that do not belong to the immune system. Although they evolved to fulfill a protective response, they also may injure the host if released in excessive quantity. Thus an understanding of cytokine gene regulation is of paramount importance in an approach to many infectious and inflammatory diseases.

Our laboratory focuses on one of the most important of all cytokines: a molecule known as tumor necrosis factor (TNF), or cachectin. TNF was first recognized as a molecule capable of selectively lysing certain cells that had been malignantly transformed. It is produced predominantly by macrophages, under conditions that simulate host invasion. In particular, if macrophages are exposed to endotoxin (lipopolysaccharide [LPS]), a lipid- and sugar-bearing molecular component of the cell membrane of certain bacteria, enormous quantities of TNF are secreted, often causing shock and death.

My colleagues and I were the first to establish that TNF is one of the principal endogenous mediators of endotoxic shock. This achievement emphasized the importance of TNF as a potential pharmacotherapeutic target. In recent years, my laboratory has sought to define the signaling pathways within macrophages that become active when these cells encounter LPS. We have also developed highly specific inhibitors of TNF, which bind and neutralize the protein once it has been released. Finally, we have studied the signaling properties of the TNF receptors themselves. By producing hybrid receptor molecules that mimic TNF, we caused cytolysis of tumor cells in the absence of the hormone itself. Such molecules may prove to be an effective tool in the treatment of malignant diseases, circumventing the toxic effects of TNF while preserving its tumorolytic effect.

In its initial interaction with the macrophage, LPS binds to a molecule known as CD14, a protein that is linked to the plasma membrane by a tether that does not penetrate into the cytoplasm. Thus CD14 has no obvious means of signaling the presence of LPS, and the events that lead to such signaling are unknown.

One important clue to the identity of proteins responsible for initial generation of the signal is the Lps gene. Mice of the C3H/HeJ strain are unresponsive to LPS, as the result of a mutation that was mapped to mouse chromosome 4. Thus a single genetic locus, Lps, is thought to control all responses to LPS. The product of the Lps gene may well be a crucial intermediate in signal transduction, and its isolation is important to understanding how endotoxic reactions come about.

Whatever the nature of the signal that initiates responses to LPS, a protein kinase cascade ensues, in which phosphorylation of one protein begets phosphorylation of others, ultimately leading to conformational changes in proteins that govern gene expression at the levels of transcription and translation. Our laboratory studied this process in mouse macrophages and determined that the mitogen-activated protein (MAP) kinase pathway becomes activated by LPS. Moreover, signaling through the MAP kinase pathway is essential for the enhancement of TNF gene transcription and TNF mRNA translation that follow induction by LPS.

Dominant inhibitors of the ras and raf proteins, which serve as upstream activators in the MAP kinase pathway, can abolish endotoxin signaling, preventing the biosynthesis of TNF. This finding is consistent with work undertaken in other laboratories, which revealed that certain proteins in the MAP kinase family are required for enhanced translation of TNF mRNA, which occurs subsequent to stimulation of macrophages by LPS.

Once TNF is produced and secreted, it binds to two separate receptors (a 55-kDa and a 75-kDa receptor) present on the plasma membrane of most cells throughout the body. We demonstrated that each trimeric TNF molecule is engaged by a dimeric form of plasma membrane receptor. The receptor then undergoes a conformational change conducive to the generation of a signal in the responding cell.

By producing soluble, dimeric versions of the TNF receptor extracellular domain, we are able to block interactions between TNF and its cellular target in a highly effective, specific manner. These receptor analogues are fashioned by linking the TNF receptor extracellular domain to the distal end of an antibody molecule. A Y-shaped chimeric protein results, containing two binding groups that can each engage a site on a TNF trimer. Only TNF and the closely related protein lymphotoxin- are bound and sequestered by these recombinant soluble receptors. Thus such reagents are highly specific. They are also stable in the plasma of experimental animals and, because they are made from two endogenous components, are essentially nonantigenic.

The TNF inhibitors can be manufactured in enormous quantities in vivo when expressed as the product of adenoviral vectors that contain the synthetic inhibitor gene. Mice transduced with such vectors are rendered refractory to all of the toxic effects of TNF and lymphotoxin, becoming highly resistant to LPS as a result. They are also highly susceptible to infection by facultative intracellular pathogens, suggesting that one of the most important functions of TNF is its ability to retard the growth of such invaders. It was presumably this function that assured the retention of TNF throughout vertebrate evolution, despite its apparent liabilities.

TNF inhibitor molecules may be very effective in the management of inflammatory diseases. Rheumatoid arthritis, for example, is now known to respond favorably to antibodies against TNF, and recombinant TNF inhibitors might prove far more useful for long-term application.

It should not be forgotten that although TNF is an extremely potent inflammatory mediator, it was initially named for its remarkable ability to cause hemorrhagic necrosis of tumors in experimental animals and for its ability to cause cytolysis of certain transformed cell lines in vitro. If its toxic effects could somehow be circumvented, it might be possible to use the tumorolytic activity of TNF for therapeutic purposes.

To this end, we manufactured novel TNF receptors endowed with constitutive activity by virtue of the fact that they are targeted to the plasma membrane and conformationally altered so as to remain dimeric at all times. These molecules, created by fusing the erythropoietin receptor extracellular domain with the extracellular stem, transmembrane, and cytoplasmic domains of each TNF receptor, initiate a suicide program in cells that express them. If introduced into tumor cells by transfection, or using viral transduction systems, the recombinant molecules trigger cell death. Studies with these molecules proved that receptor dimerization, rather than trimerization, is crucial for initiation of a signal. Furthermore, it has been shown that both of the two types of TNF receptor are capable of causing cytolysis.

Ultimately, with the development of sophisticated vector systems for gene delivery, constitutively active TNF receptors might find a place in the treatment of various tumors. No conventional chemotherapeutic agent, and in fact no suicide gene currently available, exhibits TNF's selectivity in its effect on cells that have been malignantly transformed. Therefore there is reason to hope that therapy that mimics the action of TNF, yet avoids its toxic side effects, might one day be as useful as therapy aimed at TNF neutralization.

Grants from the National Institutes of Health and the Council for Tobacco Research provided support for this work.

[ Overview of Cell Biology | Investigators in Cell Biology ]
[ Research in Progress | Biomedical Research ]

HHMI Mission | Biomedical Research | Grants | Where is HHMI? | Offices & Organization | HHMI News | Search | Jobs