Reduced Ubiquitin-Dependent Degradation of c-Jun After Phosphorylation by MAP Kinases

See allHide authors and affiliations

Science  17 Jan 1997:
Vol. 275, Issue 5298, pp. 400-402
DOI: 10.1126/science.275.5298.400


The proto-oncogene-encoded transcription factor c-Jun activates genes in response to a number of inducers that act through mitogen-activated protein kinase (MAPK) signal transduction pathways. The activation of c-Jun after phosphorylation by MAPK is accompanied by a reduction in c-Jun ubiquitination and consequent stabilization of the protein. These results illustrate the relevance of regulated protein degradation in the signal-dependent control of gene expression.

The ubiquitin-dependent protein degradation system is used in the cell not just to eliminate proteins that are either damaged or no longer needed. Instead, it fulfills important functions in cell regulation and signal transduction such as the cell cycle-specific degradation of cyclins and the cytokine-induced breakdown of the transcription factor inhibitor IkB (1, 2, 3, 4, 5).

The transcription factor c-Jun is an in vivo substrate for multi-ubiquitination (6). We investigated whether the ubiquitin-dependent breakdown of c-Jun is a constitutive process or is regulated and whether it might contribute to signal transduction through c-Jun.

One mechanism by which intracellular information is transduced to c-Jun is the phosphorylation of the protein by MAPK-type enzymes, such as the JNKs and the ERKs (7, 8, 9). Phosphorylation of c-Jun by MAPK on Ser63 and Ser73, as well as on Thr91 or Thr93, or both, increases its trans-activating potential and DNA-binding activity (10, 11, 12). The location of these sites in the vicinity of the δ domain, which mediates c-Jun multi-ubiquitination and degradation (6), raised the possibility of a functional connection between phosphorylation and ubiquitination. To investigate this idea, we examined the effect of phosphorylation on the ubiquitination of c-Jun in vivo. Histidine-tagged c-Jun was expressed with epitope-tagged ubiquitin in NIH 3T3 cells. This experimental design permits the purification and detection of multi-ubiquitinated forms of c-Jun, which can be visualized as ladders of immunoreactivity on a protein immunoblot (Fig. 1).

Fig. 1.

Inhibition of c-Jun ubiquitination after phosphorylation by JNK. NIH 3T3 cells were transfected with expression vectors (1 μg) for c-Jun-His6, c-JunAla-His6, ubiquitin-HA (Ubi-HA), and 2 μg of each expression plasmid containing cDNA for human JNK1 (JNK) and the activated form of Cdc42 (Cdc42L61). His6-tagged c-Jun-ubiquitin conjugates were purified from lysates and analyzed by SDS-PAGE (10% gel) and protein immunoblotting with a monoclonal antibody to HA (anti-HA) and a chemoluminescence detection system (upper panels) (18). The apparent molecular sizes (in kilodaltons) of protein standards are shown on the left. The position of c-Jun-ubiquitin conjugates is indicated on the right. The lower panels show the same blots analyzed with polyclonal antibody to Jun. The arrowhead indicates the position of dephosphorylated c-Jun. The phosphorylated Jun protein migrates more slowly. This immunoblot was developed with a phosphatase-conjugated second antibody; under these conditions only the majority of c-Jun (which is in the nonubiquitinated form) is visualized.

To examine a potential effect of c-Jun phosphorylation, we carried out the same experiment in the presence of vectors directing the expression of JNK1 (13) and an activator of JNK1, a gain-of-function mutant of the small guanine nucleotide-binding protein Cdc42 (14). Under these conditions, most of the histidine-tagged c-Jun became phosphorylated, as shown by the slower migration of the protein in the SDS-polyacrylamide gel (Fig. 1). Concomitant with the increase of phosphorylation, the multi-ubiquitination of c-Jun was reduced (Fig. 1). To exclude the possibility that increased JNK1 or Cdc42 activity has a general effect on the ubiquitination machinery in the transfected cells, we analyzed a mutant of c-Jun in which the MAPK phosphorylation sites were replaced by alanine residues (c-JunAla). The ubiquitination of this nonphosphorylatable version of c-Jun was not decreased by cotransfection of JNK and Cdc42, indicating that the decrease in ubiquitination is a direct consequence of phosphorylation of c-Jun.

To investigate whether phosphorylation of the MAPK sites in c-Jun might be sufficient to decrease ubiquitination, we examined the ubiquitination of a mutant of c-Jun (c-JunAsp) in which the phosphorylation sites were replaced by phosphate-mimetic aspartic acid residues. This mutant, c-JunAsp, acts as a gain-of-function form of c-Jun (11, 15). Indeed, c-JunAsp was inefficiently multi-ubiquitinated, whereas the corresponding alanine-replacement mutant was ubiquitinated with at least the same efficiency as the wild-type protein (Fig. 2).

Fig. 2.

In vivo ubiquitination of c-Jun substitution mutants c-JunAla and c-JunAsp. Expression vectors for hexahistidine (His6)-tagged c-Jun (1 μg) or His6-tagged c-Jun substitution mutants, c-JunAla-His6 and c-JunAsp-His6, were transfected into Hela TK cells along with vectors for HA-tagged ubiquitin (1 μg) as indicated (19). Ubiquitin-conjugates of His6-tagged c-Jun were purified by nickel-chelate affinity chromatography and analyzed by SDS-PAGE (10% gel) and protein immunoblotting with anti-HA (upper panel) (19). The molecular sizes of protein standards (in kilodaltons) are shown on the left. The position of conjugates of His6-tagged c-Jun proteins and HA-tagged ubiquitin is indicated on the right. The lower panel shows the same blot developed with a polyclonal antibody raised in rabbits against bacterially expressed full-length c-Jun. The arrowhead indicates the position of the various His6-tagged c-Jun proteins. The lower electrophoretic mobility of the c-JunAsp mutant reflects the “pseudo-phosphorylated” properties of the protein (11).

Phosphorylation by JNK not only suppressed multi-ubiquitination but also stabilized c-Jun in vivo (Fig. 3). Epitope-tagged wild-type c-Jun and the corresponding JunAla and JunAsp mutants were transiently transfected in 3T3 cells and metabolically labeled with 35S. The decay of c-Jun after the radioactivity was removed from the culture medium was consistent with the previously reported half-life of ∼90 min. When phosphorylated by cotransfection of JNK, however, the half-life of phosphorylated c-Jun was two- to threefold longer than that of dephosphorylated c-Jun. JunAla was refractory to JNK phosphorylation and consistently showed fast degradation kinetics regardless of whether or not it was cotransfected with the kinase. In contrast, JunAsp, even in the absence of JNK, had a half-life that was 3.5 times as long as that of the wild-type and JunAla proteins. These results indicate that phosphorylation by JNK regulates the half-life of c-Jun.

Fig. 3.

Stabilization of c-Jun by phosphorylation. The stability of HA-tagged c-Jun and JunAla before and after phosphorylation and of JunAsp was compared by measuring the decay of radiolabeled proteins (arrowheads) in 3T3 cells (20). c-Jun phosphorylation was induced by cotransfection of JNK1 and anisomycin treatment during and after labeling and is apparent by the additional slower migrating bands (multiple arrowheads). The top two panels show that the half-life of phosphorylated c-Jun is longer than that of dephosphorylated c-Jun (two- to threefold, as determined by phosphoimaging analysis). The stability of nonphosphorylatable JunAla is not affected by JNK (middle panels) and is similar to that of dephosphorylated c-Jun. c-JunAsp shows increased stability even without JNK induction (bottom panel).

The regulatory process described here likely contributes to the efficient activation of target genes after exposure of cells to growth factors, stress, or other inducers of c-Jun activity. Such an effect might be compounded by a similar regulation of the Jun partner molecule, c-Fos, which also exhibits phosphorylation-dependent changes of its half-life (16).


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
  17. 17.
  18. 18.
  19. 19.
  20. 20.
  21. 21.
View Abstract

Navigate This Article