Nucleic Acids Research

Figure 3. A gel mobility shift assay using HeLa cell extracts to determine the protein expression levels of the full-length and C-terminal deletion fusion constructs shown (except F 1-425) in Figure 2. Lane 1, HeLa cell extract from cells transfected with the reporter (pCAT4) and internal control (pGLtkLUC) plasmids; lane 2, GAL 1-147; lane 3, F 1-455; lane 4, F 1-365; lane 5, F 1-380; lane 6, F 1-395; lane 7, F 1-410; lane 8, F 1-440. The position of the gel shifted `fusion complexes' is shown, free probe and the position of a presumptive degradation products comigrating with the GAL4 1-147 (the DNA binding domain of GAL4) are indicated by arrows.

To identify the critical residue(s) in the region between residues 359 and 380, fusion proteins with deletions at three amino acid intervals between residues 359 and 380 were constructed and tested in the transient transfection assay in HeLa cells (Fig. 4). Deleting amino acids 369-380 had no effect on the activity of the C-terminal activation domain (Fig. 4B, compare F 1-380 and F 1-368). Deleting the next three amino acids, 366-368, resulted in a significant decrease in activity to background level (Fig. 4B, F 1-365). Two further deletions, F 1-362 and F 1-359, also had background activity. We conclude that the region between amino acids 365 and 368 is critical for transactivation. Leu (L) 366 is the last amino acid residue in the proposed amphipathic [alpha]-helix formed by the AF2 core (Fig, 1; 18) and thus may be an important residue.

Figure 4. Determining the critical amino acid residues in the AF2 core of HNF4 in HeLa cells. (A) Schematic representation of the GAL4 DBD, the full-length HNF4 fusion and C-terminal deletion fusions from 359 to 380 at three amino acid intervals. Intermediate constructs between 368 and 380 are not shown since the activities of these constructs were not significantly different to fusion F 1-368 (results not shown). (B) Transactivation relative to the full-length HNF4 fusion set at 100% (see legend to Fig. 2B for further details).

To investigate the significance of L366 in more detail, point mutations of L 366 to A, D, Q and I were made both in the context of GAL4-DBD full-length HNF4 fusion proteins (Fig. 5A) and in the context of native full-length HNF4 (Fig. 5C). All mutations at L366 resulted in decreased activity compared to wild type GAL4-DBD-HNF4 fusion protein or native HNF4 (Fig. 5B and D). In the context of native HNF4 (Fig. 5D), the activity of mutant L 366 I (which of all the mutants differed least in activity from the wild-type), was still statistically significantly different (P < 0.01) from wild-type in a z value test (Materials and Methods). When the activities of mutants 366 A and Q were compared to that of mutant 366 I, they were also found to be statistically significantly different(P < 0.01 in both cases). However, the transactivation activities of mutants 366 D and I were not statistically significantly different (P > 0.05) in a student's t-test. Comparing the activities of the same set of point mutants at L 366 in the context of the GAL4-HNF4 fusion (Fig. 5B) shows that mutants 366 Q and D were statistically significantly different from I (P < 0.01 in both cases), although no significant difference was seen between 366 I and A (P > 0.05) or between mutants 366 A, D and Q. These results show that L 366 is critical for full transactivation.

Figure 5. The roles of L366 and E363 in the activity of the C-terminal AF2 domain of HNF4 in HeLa cells. (A) Schematic representation of the GAL4 DBD, the wild-type GAL4-HNF4 full-length fusion construct with arrows showing the point mutations of L366->A, D, Q or I. F E363->A shows the GAL4-HNF4 full-length construct with the bolded 363E->A mutation of the AF2 core. F E363A/L366A shows the double mutant (B) Relative transactivation (see legend to Fig. 2B for further details). (C) Schematic representation of the native full-length HNF4 and mutants studied as in (A) above. WT = wild-type. (D) Transactivation relative to wild-type native full-length HNF4 set at 100%. APF1-HIV-CAT is a transfection control (see Materials and Methods). (See legend to Fig. 2B for further details.)

Position 4 (E 363 in HNF4) of the consensus AF2 core is an acidic residue in both orphan and nuclear receptors with known ligands (Fig. 1). From the crystal structure of the ligand binding domain of RAR[gamma] bound to all-trans retinoic acid, this position is critical for the formation of a salt bridge between helix 4 and 12 (18), essential for AF2 activity. Thus, we investigated the point mutant 363 E->A in HNF4 in the context of both a GAL4-HNF4 fusion protein (Fig. 5B) and in native HNF4 (Fig. 5D). Transactivation was statistically significantly lower than that of the wild-type fusion protein in both cases (P < 0.001 and P < 0.0001, respectively). These results indicate that the E at 363 is also important for full transactivation.

We next determined the effect of a double mutation of E363A/L366A in the context of a GAL4 full-length HNF4 fusion (Fig. 5A, F E363A/L366A). Unexpectedly, the activity of the double mutant was comparable to that of the single mutants at either position (Fig. 5B, F E363A/L366A compared to F L366A or F E363A), suggesting that E363 and L366 may act cooperatively in regulating the activity of the AF2 domain.

The AF2 activity of HNF4 is regulated by a repressor domain near the C-terminus

A repressor domain near the C-terminus of the progesterone receptor (PR) was recently shown to inhibit the activity of PR in the absence of progesterone (19). Alignment of the PR repressor domain sequence (residues 917-928) with the entire HNF4 protein sequence identified a potential 12 amino acid long sequence homology (Fig. 6A, boxed region) near the C-terminus of HNF4 (residues 430-441). This potential repressor region showed 40% identity to the sequence of the repressor domain of PR, which is also near its C-terminus. This homology and our previous results defining a repressor domain between residues 380 and 455 (see above), prompted us to investigate the effect of fusing a short region of HNF4 containing residues 430-441 downstream of a high activity C-terminal deletion mutant. Thus, when a 14-long peptide (residues 428-441) of HNF4 was fused in-frame downstream of GAL4-HNF4 1-372 (Fig. 6B), its activity was reduced from 300 to 22% of the full-length fusion protein [Fig. 6C, compare F 1-372+428-441 to F 1-455 (P < 0.001)]. This suggested the peptide 428-441 may function as a repressor domain in isolation.

Figure 6. A study of the C-terminal repressor of HNF4. (A) Homology (boxed) between the C-terminal repressor domain of PR and a potential repressor domain in the extended C-terminal region of HNF4 (see text). * indicates the C-terminus of PR. Amino acids are numbered in HNF4 above and PR below. (B) Schematic representation of the GAL DBD, the full-length fusion (F 1-455), the 1-371 C-terminal deletion fusion and a 1-372 C-terminal deletion fusion containing an in-frame fusion with the proposed repressor domain (428-441) of HNF4. (C) Transactivation in HeLa cells relative to the full-length fusion set at 100% (see legend to Fig. 2B for further details).

The role of this repressor domain was further investigated by mutagenesis in the context of the full-length HNF4 fusion protein. Two constructs with multiple point mutations (a run of five or six alanine residues) in the proposed repressor domain (Fig. 6A) were constructed (Fig. 7A). The activity of the N-terminal mutant [YKLLPG->AAAAAA, (427-432)] increased 2.3 times when compared to the wild-type full-length fusion (Fig. 7B). The C-terminal mutation [ITTIV->AAAAA, (434-438)] showed an increase in activity which was 3.5 times that of the wild-type full-length fusion protein (Fig. 7B). Both mutants are significantly different from wild-type (P < 0.01). These results confirm the presence of a repressor domain between amino acids 428 and 441.

Figure 7. Mutation of amino acids in the repressor domain of HNF4 enhances activity. (A) Schematic representation of the GAL 4 DBD, the full-length HNF4 fusion (F 1-455) and two full-length HNF4 fusions with multiple mutations to alanine (as shown) in the repressor domain of HNF4. (B) Transactivation in HeLa cells relative to the wild-type full-length HNF4 fusion. (For further details see legend to Fig. 2B.)

DISCUSSION

In this report we have studied the C-terminal activation region of HNF4, concentrating our attention on two short regions. These are firstly, the well-known AF2 core, which is conserved in nuclear receptors (4) and secondly, a novel short repressor domain. These regions were studied either by deletion analysis or by site-directed mutagenesis of HNF4 fused to the GAL4 DNA binding domain. Cotransfection of such fusion constructs with a chloramphenicol acetyl transferase reporter plasmid containing upstream GAL4 binding sites into either HeLa or HepG2 cells transiently, was used to study the activities of the deleted or mutated constructs (Materials and Methods).

Previous studies of the C-terminal activation domain of HNF4 suggested that the AF2 core (residues 360-366) was important for activity (12-14). Furthermore, a mutation of L366->E, in an HNF4 deletion fusion construct (amino acids 128-370) suggested that residue 366 was important for the activity of the AF2 domain (14). However, this result is preliminary because it was studied on a HNF4 deletion construct which lacked three essential domains, the N-terminal AF1 domain, the DNA binding domain and the C-terminal repressor domain (see below). It is arguable that the results might not be typical of intact HNF4. Another potential criticism of a previous study (14) is that the conclusions as to the importance of particular amino acid residues of the AF2 core particular were based on studies of a single point mutant at each position, 363 and 366, which might not be representative had other mutants been studied. Here we set out to study multiple mutations at position 366 of the AF2 core and to study a double mutant involving both positions 363 and 366 to gain further insight into the role of the AF2 core.

The results of successive C-terminal deletions (Fig. 2) and fine deletion mapping (Fig. 4) initially suggested that L366 might be important for the activity of the C-terminal AF2 domain. Therefore, four different point mutations (L->A, D, Q or I) were constructed to definitively assess the role of L 366 in both full-length HNF4 (Fig. 5C and D) and GAL4 full-length HNF4 fusion constructs (Fig. 5A and B). All mutations decreased the transactivation activity, indicating that Leu 366 is required for full activity. However, in contrast to a previous report where the activity of a mutant at 366 was reduced to [sim]5% of wild-type, we detected activities varying from [sim]35 to 75% of wild-type. This suggested, surprisingly, that transactivation activity is not absolutely dependent on Leu 366, even although it is conserved in a number of orphan receptors (Fig. 1). Other residues e.g. A, D, Q and I were clearly functional at least in our transient transactivation assay. This result is not an artefact of the fusion of intact HNF4 to the GAL4 DNA binding domain, since essentially similar quantitative results were obtained irrespective of whether full-length HNF4, or full-length HNF4 fused to GAL4 DNA binding domain, were studied (compare Fig. 5B and D). However, the fact that a conservative amino acid substitution 366 (L->I) showed the highest activity in our assay ([sim]60-70%), whereas non-conservative amino acid substitutions 366 (L->A, D, Q) showed lower activities (35-60%) still argues for the functional importance of Leu at 366 (Fig. 5).

E363 is also highly conserved between HNF4 and all nuclear receptors with known ligands, although it is substituted by a D in some orphan receptors (Fig. 1). E363 was selected for mutation because it is known to be involved in the formation of a salt-bridge and is critical for transactivation in RAR[gamma] (18). Mutation of E363->A in HNF4 here (Fig. 5) indicated that this residue, like residue 366, was required for full activity. Our result at this position differed from a previous report (where activity was reduced to <5% of wild-type; 14) in that significant residual activity was observed (between 50 and 70% of wild-type; Fig. 5C and D) depending on whether with the full-length HNF4 construct or the GAL4 full-length HNF4 fusion construct was used in the transactivation assay. We attribute this difference to the fact that full-length HNF4 was used in our studies, whereas HNF4 deletion constructs were used in the previous report (14).

It is significant that a double mutant of 363 and 366 also retained [sim]50% activity in our assay (Fig. 5). The fact that the double mutant failed to demonstrate an additive effect over and above that of the two point mutants assayed separately is consistent with E363 and L366 cooperating to regulate the activity of the AF2 domain. These experiments illustrates both the importance of E363 and L366 of the AF2 core, yet also demonstrate the core is partly redundant at least in our transactivation assay. It remains possible that the redundancy observed in the mutations studied here results from a compensatory effect of the AF1 domain, which is known to transactivate (14) and is present in our full-length HNF4 constructs.

Our initial C-terminal deletion experiments suggested the presence of a repressor activity within residues 380-455 (Fig. 2B, and previously proposed to lie between residues 371 and 455; 14). By aligning the amino acid sequence of the PR repressor domain (19) with the entire HNF4 sequence, a short region (residues 430-441) showing 40% identity to a previously characterised repressor domain in PR was localised in HNF4 (Fig. 6A). This presumptive repressor domain was first shown to confer repressor activity by itself (Fig. 6C). Thus, when a short peptide from amino acid 428 to 441 containing the HNF4 repressor domain was fused in-frame downstream of a high activity C-terminal deletion fusion construct, transactivation was severely inhibited, decreasing from >300 to 22%. This is unlikely to be the result of decreased protein expression of this particular fusion construct, since a gel shift analysis of a panel of deletion constructs all showed reasonably similar levels of protein expression (Fig. 3) This implied that the peptide containing amino acids 428-441 was sufficient, by itself, to act as a repressor. Further evidence for this repressor domain came from a study of two constructs with multiple point mutations (Fig. 7C).

It is interesting to note that in the experiment where the isolated repressor domain was fused to F 1-372 (Fig. 6B), the AF2 core was intact; hence an activity closer to the that of the full-length fusion (100%) might have been expected, whereas an activity of 22% was observed (Fig. 6C). This suggests the repressor domain had a more potent inhibitory effect when fused closer to the AF2 core, than if it were in its normal location. Perhaps the close proximity of the repressor domain to the AF2 core may restrict the accessibility of potential coactivator proteins to the AF2 core, either by binding a corepressor protein or simply by masking critical residues in the AF2 core. In the case of the PR repressor domain, the repressor domain was shown to mediate its function through an unknown corepressor (19). Experiments in an attempt to determine if the HNF4 repressor domain functioned through a corepressor were unfortunately inconclusive (data not shown).

It is obviously interesting to know how the repressor domain in HNF4 functions and if it is required to regulate the activity of HNF4 in vivo. An attractive, but speculative mechanism might involve the phosphorylation of threonine residues in the sequence ITTI (residues 434-437). A corepressor might bind to the repressor domain in the phosphorylated state and dissociate in the dephosphorylated state. Thus mutations which disrupt these threonine residues would prevent phosphorylation, thus preventing repression and thereby enhancing the activity of HNF4.

In summary, our characterisation of the AF2 domain in the C-terminal region of HNF4 has given new insights into the function of residues 363 and 366. By extensive mutagenesis we have shown that L366 is essential for full AF2 activity. An E363A/L366A double mutant suggested that E363 and L366 may act cooperatively to regulate the activity of the AF2 domain. We have also shown that the activity of the AF2 domain is regulated by a repressor domain located within a proline-rich region. We have localised this repressor domain to within amino acid residues 428-441 of HNF4, and showed it was necessary and sufficient for repressor activity in transient transfection studies in HeLa cells. Furthermore, multiple point mutations within this repressor domain in the context of the full-length HNF4 fusion construct abolished repressor activity. The repressor domain shares homology with a recently reported C-terminal repressor domain in the progesterone receptor (19).

ACKNOWLEDGEMENTS

This work was supported by the MRC (VPI) and the Wellcome Trust (NHD & GGB) grant number 049537/Z/96. We thank Dr S. Murphy for helpful discussions and the provision of pGAL4 1-147, Dr M. Busslinger for pCAT4 and Dr Y. Talianidis for pCB6HNF4.

REFERENCES

1. Sladek, F.M., Zhong, W., Lai, E. and Darnell, J.E., Jr (1990) Genes Dev., 4,2353-2365. MEDLINE Abstract

2. Kastner, P., Mark, M. and Chambon, P. (1995) Cell, 83, 859-869. MEDLINE Abstract

3. Nagpal, S., Friant, S., Nakshatri, H. and Chambon, P. (1993) EMBO J., 6, 2349-2360.

4. Danielian, P.S., White, R., Lees, J.A. and Parker, M.G. (1992) EMBO J., 3, 1025-1033.

5. Drewes, T., Senkel, S., Holewa, B. and Ryffel, G.U. (1996) Mol. Cell. Biol., 3, 925-931.

6. Hata, S., Tsukamoto, S. and Osumi, T. (1992) Biochem. Biophys. Acta, 1131, 211-213.

7. Holewa, B., Zapp, D., Drewes, T., Senkel, S. and Ryffel, G.U. (1997) Mol. Cell. Biol., 2, 687-694.

8. Chen, W.S., Manova, K., Weinstein, D.C., Duncan, S.A., Plump, A.S., Prezioso, V.R., Bachvarova, R.F. and Darnell, J.E. (1994) Genes Dev., 8,2466-2477. MEDLINE Abstract

9. Crossley, M., Ludwig, M., Stowell, K.M., DeVos, P., Olek, K. and Brownlee, G.G. (1992) Science, 257,377-379. MEDLINE Abstract

10. Naka, H. and Brownlee, G.G. (1996) Br. J. Haematol., 92,231-240. MEDLINE Abstract

11. Yamagata, K., Furuta, H., Oda, N., Kaisaki, P.J., Menzel, S., Cox, N.J., Fajans, S.S., Signorini, S., Stolffel, M. and Bell, G.I. (1997) Nature, 384, 458-460.

12. Galson, D.L., Tsuchiya, T., Tendler, D.S., Huang, L.E., Ren, Y., Ogura, T. and Bunn, H.F, (1995) Mol. Cell. Biol., 4, 2135-2144.

13. Malik, S. and Karathanasis, S.K. (1996) Mol. Cell. Biol., 4, 1824-1831.

14. Hadzopoulou-Cladaras, M., Kistanova, E., Evagelopoulou, C., Zeng, S., Cladaras, C. and Ladias, J.A.A. (1997) J. Biol. Chem., 272,539-550. MEDLINE Abstract

15. Braselmann, S., Graninger, P. and Busslinger, M. (1993) Proc. Natl. Acad. Sci. USA, 90,1657-1661. MEDLINE Abstract

16. Morgan, G.E., Rowley, G., Green, P.M., Chisholm, M., Giannelli, F. and Brownlee, G.G. (1997) Br. J. Haematol., 98,79-85. MEDLINE Abstract

17. Murphy, S. (1997) Nucleic Acids Res., 11,2068-2076.

18. Renaud, J.P., Rochel, N., Ruff, M., Vivat, V., Chambon, P., Gronemeyer, H. and Moras, D. (1995) Nature, 378, 681-689. MEDLINE Abstract

19. Xu, J., Nawaz, Z., Tsai, S.Y., Tsai, M.J. and O'Malley, B.W. (1996) Proc. Natl. Acad. Sci. USA, 93,12195-12199. MEDLINE Abstract

---

*To whom correspondence should be addressed. Tel: +44 1865 275559; Fax: +44 1865 275556; Email: george.brownlee@path.ox.ac.uk

---

This page is run by Oxford University Press, Great Clarendon Street, Oxford OX2 6DP, as part of the OUP Journals
Comments and feedback: www-admin{at}oup.co.uk
Last modification: 17 Apr 1998
Copyright©Oxford University Press, 1998.
CiteULike    Connotea    Del.icio.us    What's this?

This article has been cited by other articles:

				L. W. Harries, J. M. Locke, B. Shields, N. A. Hanley, K. P. Hanley, A. Steele, P. R. Njolstad, S. Ellard, and A. T. Hattersley The Diabetic Phenotype in HNF4A Mutation Carriers Is Moderated By the Expression of HNF4A Isoforms From the P1 Promoter During Fetal Development Diabetes, June 1, 2008; 57(6): 1745 - 1752. [Abstract] [Full Text] [PDF]

				A. D. Petrescu, R. Hertz, J. Bar-Tana, F. Schroeder, and A. B. Kier Role of Regulatory F-domain in Hepatocyte Nuclear Factor-4{alpha} Ligand Specificity J. Biol. Chem., April 29, 2005; 280(17): 16714 - 16727. [Abstract] [Full Text] [PDF]

				J. Eeckhoute, P. Formstecher, and B. Laine Hepatocyte Nuclear Factor 4{alpha} enhances the Hepatocyte Nuclear Factor 1{alpha}-mediated activation of transcription Nucleic Acids Res., May 11, 2004; 32(8): 2586 - 2593. [Abstract] [Full Text] [PDF]

				K. S. Childs and S. Goodbourn Identification of novel co-repressor molecules for Interferon Regulatory Factor-2 Nucleic Acids Res., June 15, 2003; 31(12): 3016 - 3026. [Abstract] [Full Text] [PDF]

				M. D. Ruse Jr., M. L. Privalsky, and F. M. Sladek Competitive Cofactor Recruitment by Orphan Receptor Hepatocyte Nuclear Factor 4{alpha}1: Modulation by the F Domain Mol. Cell. Biol., March 15, 2002; 22(6): 1626 - 1638. [Abstract] [Full Text] [PDF]

				H. Schrem, J. Klempnauer, and J. Borlak Liver-Enriched Transcription Factors in Liver Function and Development. Part I: The Hepatocyte Nuclear Factor Network and Liver-Specific Gene Expression Pharmacol. Rev., March 1, 2002; 54(1): 129 - 158. [Abstract] [Full Text] [PDF]

				F. M. Sladek, M. D. Ruse Jr., L. Nepomuceno, S.-M. Huang, and M. R. Stallcup Modulation of Transcriptional Activation and Coactivator Interaction by a Splicing Variation in the F Domain of Nuclear Receptor Hepatocyte Nuclear Factor 4alpha 1 Mol. Cell. Biol., October 1, 1999; 19(10): 6509 - 6522. [Abstract] [Full Text] [PDF]

				T. Hosoda, K. Monzen, Y. Hiroi, T. Oka, E. Takimoto, Y. Yazaki, R. Nagai, and I. Komuro A Novel Myocyte-specific Gene Midori Promotes the Differentiation of P19CL6 Cells into Cardiomyocytes J. Biol. Chem., September 14, 2001; 276(38): 35978 - 35989. [Abstract] [Full Text] [PDF]

This Article Abstract Print PDF (134K) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to citation manager Search for citing articles in: ISI Web of Science (9) Request Permissions Commercial Re-use Guidelines for Open Access NAR Content Google Scholar Articles by Iyemere, V. P. Articles by Brownlee, G. G. Search for Related Content PubMed PubMed Citation Articles by Iyemere, V. P. Articles by Brownlee, G. G. Social Bookmarking          What's this?