To TWEAK, or not to TWEAK: That is the question


  • David A. Shafritz M.D.

    Corresponding author
    1. Marion Bessin Liver Research Center and Division of Hepatology, Department of Medicine, Albert Einstein College of Medicine of Yeshiva University, Bronx, NY
    • Marion Bessin Liver Research Center and Division of Hepatology, Department of Medicine, Albert Einstein College of Medicine of Yeshiva University, 1300 Morris Park Avenue, Bronx, NY 10461
    Search for more papers by this author
    • fax: 718-430-8975

  • Potential conflict of interest: Nothing to report.

  • See Article on Page 291.

TWEAK (the designated acronym for tumor necrosis factor [TNF]-like weak inducer of apoptosis) is a member of the TNF family (TNF superfamily, member 12 [TNFSF12]) of cytokines that are mediators of host defense and immune regulation. These cytokines are type II transmembrane proteins expressed on the cell surface that interact with effector cells through direct cell-cell contact or are processed and secreted into the surrounding tissue to interact with receptor proteins on the surface of nearby cells. TWEAK was originally identified by Chicheportiche et al. in 1997 and is expressed ubiquitously in many tissues.1 The behavior of TWEAK is multifaceted, multidimensional, and somewhat schizophrenic, because it modulates a diverse range of cellular processes in different cell types and under different experimental conditions, including proliferation, differentiation, cell migration, cell survival, and cell death (Fig. 1).2

Figure 1.

Opposing effects of TWEAK/Fn14 signaling on the behavior (properties) of progenitor cells versus differentiated cells during tissue regeneration following acute or chronic injury.


NF-κB, nuclear factor κB; PH, partial hepatectomy; TNF, tumor necrosis factor; TWEAK, TNF-like weak inducer of apoptosis.

The receptor for TWEAK is Fn14 (TNFRSF12A), a member of the fibroblast growth factor–inducible gene family that codes for a 14-kD protein (hence the acronym Fn14). It was identified by differential display, cloned, and sequenced as a type 1 transmembrane protein, but its function was initially unknown.3 Subsequently, Wiley et al. cloned a gene coding for a TWEAK-binding protein (TWEAK receptor),4 whose sequence was identical to that reported above3 and Winkles et al. and Thorgeirsson et al. co-reported that Fn14 is hyperexpressed in liver cancer cell lines and in both human and mouse hepatocellular carcinoma tissues.5 Furthermore, Fn14 expression, which is very low in the quiescent liver, is induced within 4 hours after two-thirds partial hepatectomy (PH) (a so-called “immediate early response gene”) and is very highly expressed in several models of mouse hepatocarcinogenesis.5 This suggested a strong role for TWEAK/FN14 signaling in proliferation of hepatic epithelial cells and/or their transformation. How this occurs, however, has not been determined. In addition to TNF signaling, the Fn14 receptor is also linked to other cell pathways, such as nuclear factor κB (NF-κB), mitogen-activated protein kinase, extracellular signal-regulated kinase, c-Jun N-terminal kinase, and phosphoinositide 3-kinase/AKT, so that the cellular functions regulated at least in part by TWEAK/Fn14 and the mechanisms involved are highly complex.2

TWEAK and Fn14 are expressed in epithelial, mesenchymal, and endothelial cells in many tissues2 and the pattern of TWEAK/Fn14 expression suggests that this signaling pathway plays an important role in regulating tissue repair and remodeling under both physiologic and pathologic conditions (for an excellent review, see Burkly et al.6). Studies linking TWEAK/Fn14 signaling to progenitor cell proliferation were first performed in mouse liver, where it was demonstrated that overexpression of TWEAK in transgenic mice causes periportal oval cell hyperplasia.7 Oval cell expansion in mice can also be induced by administration of 3,5-diethoxycarboxyl-1,4-dihydrocollidine (DDC) and in mice lacking expression of the TWEAK receptor (Fn14-null mice), oval cell proliferation is markedly reduced.7 In addition, wild-type mice pretreated with a blocking anti-TWEAK monoclonal antibody prior to DDC administration also showed a similar reduction in oval cell expansion.7 However, TWEAK/Fn14 signaling not only regulates progenitor cell proliferation in hepatic epithelial cells, it also regulates proliferation of mesodermal progenitor cells in the hematopoietic lineage6 and myogenic progenitor cells (so called “satellite” cells) isolated from adult striated muscle tissue and in a tissue regeneration model following muscle injury.8

From studies conducted over five decades, it has been determined that there are two distinct cellular pathways through which liver regeneration can be achieved. In the normal adult liver, hepatocytes are in a quiescent state and turn over very slowly (only 2-3 times/year). However, following abrupt reduction of liver mass (two-thirds PH) or acute toxic liver injury, hepatocytes rapidly enter the cell cycle and proliferate to restore liver mass within 5-7 days. Many years ago, it was shown during liver regeneration following two-thirds PH in a normal young adult rat that periportal hepatocytes initially become labeled with [3H]thymidine, but eventually up to 70%-90% of residual mature hepatocytes throughout the parenchyma engage in DNA synthesis and undergo cell division.9, 10 Liver regeneration is a highly organized, complex, multistep process that involves growth factors and cytokines, gene transcription factors, cell signaling pathways, and expression of cell cycle regulatory genes.11-13 From these studies, it has been concluded that the proliferative activity of adult hepatocytes is sufficient to regenerate the liver following two-thirds PH and that there is no evidence that stem cells participate in this process.11-14

A second pathway for liver cellular regeneration has been identified when the proliferation of hepatocytes has been blocked by chemical or physical damage to cellular DNA and a regenerative stimulus is applied. This pathway involves proliferation of hepatic epithelial progenitor cells which subsequently differentiate into mature hepatocytes and replace host hepatic mass (originally called “oval cells” because of their morphologic appearance15). These cells lie dormant in the canals of Hering (the junctional zone between the biliary tree and the hepatic parenchymal plates) and upon stimulation of liver regeneration under conditions in which hepatocytes cannot proliferate, such as administration of 2-acetaminofluorine (AAF), followed by two-thirds PH, oval/progenitor cells are activated to proliferate, move out into the hepatic parenchyma, and become incorporated into the hepatic lobular structure.16, 17

In this issue of HEPATOLOGY, Tirnitz-Parker et al.18 have extended the studies of Jakubowski et al.7 to show that oval/progenitor cell activation in mice treated with a choline-deficient, ethionine-supplemented (CDE) diet is also mediated by TWEAK/Fn14 signaling. They showed that TWEAK is produced in the liver by natural killer cells and macrophages, both of which are of mesodermal origin, and that activated oval cells, which express pan-cytokeratin (a liver bile duct and epithelial progenitor cell marker) and Ki67 (a cell proliferation marker), express Fn14. They showed further that clonal mouse liver progenitor cell lines, which have been isolated and characterized extensively in their laboratory,19 exhibit increased expression of NF-κB after treatment with recombinant human TWEAK and that knockdown of NF-κB expression by small interfering RNA reduces liver progenitor cell proliferation in vitro. These studies demonstrate that TWEAK signaling through a paracrine mechanism leads to the liver progenitor cell response in CDE diet-induced liver injury and regeneration.

Based on the findings that embryos from Fn14 knockout mice appear normal, it has been stated that TWEAK does not play a role in development, presumably including liver development.6 However, this would seem unlikely given the similarity of gene expression programs in fetal liver stem/progenitor cells and oval/progenitor cells in the adult liver under conditions in which proliferation of adult hepatocytes is blocked. Dabeva and colleagues have reported that TWEAK expression is increased modestly in liver nonparenchymal/Thy1+ cells of rats treated with 2-AAF/PH, and Fn14 expression is increased substantially in purified hepatic oval cells from these same animals.20 Fn14 expression is also increased three-fold to four-fold in purified embryonic day 12.5 (ED12.5) fetal mouse liver stem/progenitor cells compared to adult liver, whereas TWEAK expression is very low (see gene expression microarray data at; accession number GSE6340).21 Similar results have been obtained with purified ED14 rat fetal liver stem/progenitor cells (Dabeva and coworkers, unpublished observations).

The most likely explanation for a normal phenotype in embryos from Fn14 null mice is that the functions of the TWEAK/Fn14 pathway can be expressed redundantly by other TNF family members. The most important issues that need to be addressed in future studies are precisely how TWEAK and Fn14 expression are regulated and how can we manipulate expression of this signaling pathway to enhance liver regeneration under conditions in which hepatocyte proliferation is not blocked and/or utilize it to promote liver repopulation by transplanted cells. It is interesting that fetal liver stem/progenitor cells express increased levels of Fn14 and that after transplantation, these cells effectively repopulate the liver following two-thirds PH.22, 23

Many years ago, it was determined that the regenerative capacity of the liver in rodents decreases with aging24 and this is presumably true in all mammalian species, including humans. If we could induce expression of the TWEAK/Fn14 pathway, perhaps this would activate human hepatic progenitor cell proliferation and enhance liver regeneration in the aging liver. Moreover, if we could transduce adult hepatocytes with Fn14 and transplant these cells into recipients with a genetic disorder in which there is no hepatic parenchymal pathology, such as in inherited hyperbilirubinemias, ornithine transcarbomylase deficiency, hypercholesterolemia, phenylketonuria, etc., then perhaps we could achieve effective liver repopulation using this strategy.

TNFα, NF-κB, interleukin-6, and signal transducer and activator of transcription 3 (Stat3) are all involved in initiating hepatocyte proliferation during the regenerative response in the normal liver.11-13 NF-κB also appears to be activated during oval cell activation, and this is mediated by Fn14, the receptor for TWEAK. Because TWEAK is expressed in inflammatory (natural killer) cells and macrophages,18 there is a direct link between the repair process during and following liver injury and activation of the oval cell/progenitor response. Coordinating these events and their manipulation experimentally could lead to new, interesting, and innovative ways to stimulate liver regeneration, as well as liver repopulation by transplanted cells.