Macrophage migration inhibitory factor was originally identified for its ability to inhibit the random migration of macrophages in vitro 1011. Subsequent work defined MIF as a potent cytokine with mitogenic and pro-inflammatory functions 12. Recent efforts to identify a cellular surface receptor for MIF showed that the CD74/CD44 receptor complex mediates binding of extracellular MIF 1314. However, the process by which extracellular MIF may exert its effect on target cells is still poorly understood 15.

MIF is remarkably well conserved, and its homologues are encoded in evolutionarily divergent species, including different vertebrates, worms, insects and plants 16. For a cytokine, MIF is unusual, since it is abundantly expressed by various cell types 12 and stored within the cytoplasm 17. Early evidence suggesting a role for MIF in cell growth and/or differentiation came from the observations of its expression in developing mouse embryos. The MIF gene is expressed at early embryonic stages prior to implantation 18. At mid-gestation, MIF's expression pattern parallels tissue specification and organogenesis 1920. At later developmental stages, MIF expression is broadly associated with cellular differentiation 21. However, MIF appears to be dispensable for normal development, because MIF-null mice reproduce and grow normally 2223.

Traditionally, the major focus of MIF research has been on its role as a pro-inflammatory mediator within the immune system. Recent studies, however, showed that MIF's functional repertoire is not limited to the immune response, but also extends to the regulation of apoptosis and malignant transformation. MIF overexpression has been observed in various human cancer tissues, including colorectal, breast, lung, bladder and prostate cancer 2425262728. Genetic studies demonstrated that MIF promotes B-cell lymphomagenesis and intestinal tumorigenesis in mice 2429. Importantly, MIF overexpression in several cell types confers resistance to apoptosis through interference with the activity of the p53 tumor suppressor 30. Conversely, when MIF is lost, cell survival and functions are compromised in a p53-dependent manner 2331. Inhibition of MIF expression also phenocopies loss of hypoxia inducing factor-1α (HIF-1α), a well established target of p53 regulation, and induces premature senescence 32. It was hypothesized that upregulation of MIF at sites of chronic inflammation might impair p53-dependent cellular responses towards DNA damage and inappropriate proliferation and, thus, promote the accumulation of oncogenic mutations 30.

In agreement with this, we recently showed that in a mouse model of Burkitt's lymphoma (Eμ-Myc transgenic mouse), loss of MIF expression coincides with the induction of a p53-dependent proliferative block, which profoundly affects normal B-cell development 29. Moreover, inhibited S-phase progression and subsequent differentiation block are at the root of the predisposition of MIF-deficient B-cells to undergo spontaneous p53-dependent apoptosis. Accordingly, almost all lymphomas that arise in MIF-deficient Eμ-Myc mice can be accounted for by mutations within the ARF-p53 axis, indicating that the p53 pathway is the main determinant for tumor suppression in this model system 29.

However, several observations suggest that the functional role of MIF in tumorigenesis is more complex than previously appreciated. In contrast to the notion of MIF as tumor promoter, two reports have indicated that aberrantly low levels of MIF in human tumors could also correlate with poor clinical prognosis and that subcellular compartimentalization of MIF may likewise be relevant 3334. Our own evidence from chemical one-stage skin carcinogenesis experiments revealed that deletion of the MIF gene may lead to increased rates of tumor formation in mice (Fig. 1). These findings are also supported by recent experiments using MIF-knockout mice in a p53-null background, which showed that MIF deficiency leads to a shift in the tumor spectrum: while the expected high frequency of T-cell lymphomas and fibrosarcomas was reduced upon MIF deficiency, the frequency of B-cell lymphomas and carcinomas was strongly increased. Moreover, the shift in the tumor spectrum led to decreased survival of MIF-/-p53-/- mice compared to the p53-/- controls 9. Thus, MIF is unlikely to have sufficient prognostic value when used in isolation from other possible mutations, particularly those affecting p53.

Progression through the cell cycle depends on timely activation of cyclin-dependent kinases (CDKs), which act in conjunction with their corresponding cyclins. The protein levels of cyclins, CDK inhibitors and other major regulatory proteins are quantitatively controlled by the ubiquitin-proteasome system (UPS). Which protein is ubiquitylated and subsequently degraded relies on recognition by two principal E3 ubiquitin ligases, the anaphase-promoting/cyclosome complex (APC/C) and the Skp1-Cullin1-F-box (SCF) complex. These protein complexes appear to fulfill related but nonetheless distinct functions in the regulation of the cell cycle, as SCFs are active from late G1 to early M-phase, whereas the APC/C is activated from mid M-phase to the end of G1-phase 35. The SCF complex has recently taken center stage in regulatory biology because it links extra- and intracellular signals to destruction of various proteins by the proteasome 3637. Thus, SCFs target many key proteins involved in the control of normal cell division, such as cyclin E, c-Jun, c-Myc, p21, p27, β-catenin and Notch 38394041424344. Recent work shows that the SCF complex is also a central effector of DNA damage and repair pathways acting in the S- and G2M-phases of the cell cycle 4546. Therefore, it is not surprising that SCF is frequently a target of genetic alteration in cancer 37, and that deregulated SCF activity greatly promotes cancer development 47.

The SCF complex is made of four components, wherein Rbx1, Cul1 (scaffold protein) and Skp1 (adaptor protein) are invariable subunits. The fourth component, the F-box protein, serves as a substrate recognition unit. Given the variety of F-box proteins that can be recruited to the SCF complex (more than 70 F-box proteins have been identified in humans), the functional diversity of the SCF complex appears tantalizing 48. The activity of the catalytic core of SCF, which is formed from the Rbx1 and Cul1 subunits, is stimulated by the attachment of ubiquitin-like protein Nedd8 to conserved lysines in the cullin-homology domain of Cul1 49. This step invokes recruitment of E2 enzymes and thus promotes assembly of an active SCF complex. Conversely, deneddylation of cullins through the CSN/COP9 signalosome, with its Jab1/CSN5 subunit directly cleaving off Nedd8, decreases E2 recruitment 5051. In addition, CSN recruits a deubiquitylase, Ubp12, which counteracts the intrinsic ubiquitin-polymerizing activity of SCF 52 Furthermore, deneddylated cullins are sequestered by inhibitory Cand1 5354. It was proposed that SCF activity is sustained by dynamic cycles of assembly and disassembly, where both Cand1 and CSN play an essential negative role 55.

Remarkably, a search for intracellular MIF-binding partners by the yeast two-hybrid system yielded Jab1/CSN5 as potential candidate 56. CSN5 is a component of the COP9/CSN signalosome, a multiprotein complex that plays essential roles in differentiation and morphogenesis 5758. Not surprisingly, deletion of individual subunits of the CSN complex is embryonically lethal 5960. Also Jab1-null embryos die soon after implantation due to impaired proliferation and accelerated apoptosis, and these defects have been attributed in part to the accumulation and/or impaired degradation of p53, cyclin E, and p27 61. Importantly, Jab1 possesses an intrinsic metalloprotease activity, which as mentioned above, targets SCF complexes and deconjugates Nedd8 from the cullins 5158. Whereas Jab1 recycles neddylated cullins into more stable unneddylated forms 62, the CSN signalosome can further stabilize the SCF by preventing the autoubiquitination of substrate-recruiting F-box proteins 6364. However, deneddylated cullins are open to interaction with the inhibitory Cand1, which has the capacity to displace both Skp1 and F-box proteins 5354. Therefore, unbalanced activity of Jab1/CSN5 can directly or indirectly block ubiquitin-dependent proteolysis 65. Conversely, MIF binds to Jab1/CSN5 and prevents it from interacting with proteins targeted by the CSN signalosome 66 (Fig. 2). Our recent study showed that this negative regulation of Jab1/CSN5 by MIF is a physiologic requirement to sustain optimal composition and activity of SCF ubiquitin ligases 9.

Cells continuously encounter DNA damage caused either by replication errors at replication forks or by extracellular noxae such as ultraviolet and ionizing irradiation. Failure to properly repair DNA can lead to various disorders, including enhanced rates of tumor development 6768. To protect against such insults to the genome, eukaryotic organisms have evolved an elaborate signalling network of p53-dependent and p53-independent checkpoints. Once the integrity of the genome has been compromised, checkpoint proteins either prevent or delay initiation of the next cell cycle phase. Notably, DNA damage checkpoint kinases (ATM, ATR, Chk1, Chk2) inhibit the Cdk machinery, the normal function of which is to coordinate DNA replication and partitioning of the chromosomes 69 (Fig. 2). The DNA damage response also causes induction of DNA repair functions, such that cells with modest damage may survive. However, cells with more severe damage are induced to undergo apoptosis. Mutations affecting DNA damage pathways allow cell proliferation under conditions of replication stress 6870. Therefore, the DNA damage response is subject to tight control and is regulated at the level of gene expression, protein phosphorylation, protein stabilization or degradation. Cancer development frequently selects for loss of p53 function and hence for loss of the G1 checkpoint 71. Mutations compromising DNA damage checkpoints are also oncogenic 6770, although they are expected to activate p53.

One of the key functional targets of DNA-damage checkpoint regulation is the Cdc25 protein phosphatase family, which controls cell cycle progression by regulating the activity of Cdk2 and Cdk1 kinase complexes 68. We found that in MIF-deficient mice, Chk1/Chk2-regulated checkpoints are uncoupled from proteasomal degradation of Cdc25A under conditions of DNA damage 9. The loss of MIF produced similar dysregulation in the proper degradation of several other cell cycle regulators such as Cyclin A2, p27, E2F1 and DP1 9. Because MIF binds to Jab1/CSN5 and prevents it from interacting with proteins targeted by CSN, notably the cullins, it appears that absence of this interaction can lead to inappropriate deneddylating activity of Jab1/CSN5 and therefore to the ensuing dysfunction of the SCF. Importantly, we showed that DNA damage induces coordinate activity of the G2M checkpoint and Cul1-containing SCF complexes. Moreover, Chk1 delivers a signal that not only marks proteins for degradation but also activates the SCF (Fig 2). These data emphasize the importance of downstream effectors of checkpoint pathways that execute the cell division shut-off program, namely the SCF. In essence, our results imply that the G2M checkpoint and the SCF complex form a functional unit to stop cell cycle progression after DNA damage 9. These data also provide an attractive mechanism explaining the MIF-p53 relationship and the role of MIF in tumor development. Given that MIF plays a key role in the regulation of Cdc25A, Cdk2 and E2F1/DP1 complexes, which can all induce a strong p53-dependent antiproliferative response, our data suggest that the involvement of MIF in p53 function is secondary to p53-independent mechanisms controlling protein stability, checkpoint regulation, and the integrity of the genome. While the loss of MIF expression induces a p53-dependent proliferative block 2329, concomitant loss of p53 rescues these growth defects, but it comes at the price of increased tumorigenesis 9. Accordingly, the tumor phenotype of MIF/p53 compound mutant mice entails defects in the checkpoint response and DNA repair process 9.