EDEN (embryo deadenylation element)-dependent deadenylation is a regulatory process that was initially identified in Xenopus laevis early embryos and was subsequently shown to exist in Drosophila oocytes. Recent data showed that this regulatory process is required for somitic segmentation in Xenopus. Inactivation of EDEN-BP (EDEN-binding protein) causes severe segmentation defects, and the expression of segmentation markers in the Notch signalling pathway is disrupted. We showed that the mRNA encoding XSu(H) (Xenopus suppressor of hairless), a protein central to the Notch pathway, is regulated by EDEN-BP. Our data also indicate that other segmentation RNAs are targets for EDEN-BP. To identify new EDEN-BP targets, a microarray analysis has been undertaken.

Introduction

The post-transcriptional regulation of gene expression within the cytoplasm of cells has numerous facets such as translational activation or silencing and localization. Control of gene expression at this level is particularly important in oocytes and early embryos. Indeed, in all species studied to date, transcription is essentially silent during the completion of meiosis (maturation) and the first mitotic divisions. This situation has both highlighted the importance of post-transcriptional controls and enabled the study of these mechanisms. In Xenopus, mouse and probably many other species, one of the major mechanisms that activates or represses the translation of maternally inherited mRNAs is via a control of the poly(A) tail length. Two sequence elements required for post-fertilization deadenylation [poly(A) tail shortening] have been identified: the EDEN (embryo deadenylation element) [1] and the AUUUA containing AU-rich elements [2] both localized within the 3′-UTR (3′-untranslated region) of affected mRNAs. It is necessary to mention that contrary to the general situation in somatic cells, in early embryos deadenylated mRNAs are not immediately degraded and the coupling between mRNA deadenylation and degradation is only restored after the resumption of transcriptional activity at the 12th mitotic cell cycle [24]. This may have developmental importance as the ‘protection’ of deadenylated mRNAs implies that maternally inherited poly(A) mRNAs whose translation is activated after the initial mitotic cycles are also protected. We will concentrate here on the EDEN-dependent pathway.

The EDEN sequence

Mapping for regulatory sequences in the 3′-UTR of mRNAs deadenylated in embryos has shown that functional EDENs are situated within U/purine-rich regions (reviewed in [5]). Indeed, a synthetic motif composed of at least six contiguous (UGUA) motifs can confer post-fertilization deadenylation on a reporter mRNA [6]. However, comparison of the EDEN-containing regions has not identified any particular sequence motif, so, to date, EDENs cannot be efficiently identified using in silico-based methods. To complicate the picture, two auxiliary sequence elements that assist the EDEN mechanism have been identified, an AUU repetition [6] and a single AUUUA motif [7].

In oocytes and early embryos, deadenylation is strongly correlated with translational repression. This functional aspect of the EDEN is evolutionarily conserved, as the translation of a reporter mRNA containing a synthetic EDEN is repressed in the ovaries of transgenic Drosophila [8] whereas the stability and localization of the mRNA were normal. This regulatory mechanism is also probably conserved in humans, as the instability element in the 3′-UTR of human c-jun mRNA is U/purine-rich and behaves like an EDEN when present within the 3′-UTR of a reporter mRNA injected into Xenopus embryos [9].

EDEN-BP (EDEN-binding protein), the associated factor

Most regulatory RNA motifs function as target sites for sequence-specific RNA-binding proteins that are, directly or indirectly, the functional actors in the regulatory processes. The protein binding to the EDEN, EDEN-BP, is a 54 kDa protein containing three RNA recognition motifs, two in the N-terminal region and one in the C-terminal part [1]. It is required for the EDEN-dependent deadenylation process. The particular arrangement of the three RNA recognition motifs is characteristic of a large superfamily of proteins including the Drosophila Elav and bruno proteins and the mammalian Hu proteins. At the amino acid level EDEN-BP has 88% identity with the human protein CUG-binding protein 1 and this protein can functionally replace EDEN-BP in a Xenopus extract [10].

Biological roles

With the intention of identifying a biological role for EDEN-dependent deadenylation, we developed a method to inactivate EDEN-BP in Xenopus embryos using either antisense morpholino oligonucleotides or neutralizing antibodies [11]. The most important phenotype observed was a disruption of somitic segmentation (see Figure 1). Somites are prospective muscle and bones. They are made of a series of similar and successive units by a process termed as segmentation. In EDEN-BP-inhibited embryos, loss of segmentation is associated with a modified expression of a number of markers in the PSM (pre-somitic mesoderm), where segmentation takes place and which comprises the TBD (tail bud domain) and presumptive somites (see Figure 1). Using a candidate gene approach, we found one mRNA that is post-transcriptionally regulated by EDEN-BP in the PSM. This mRNA encodes XSu(H) (Xenopus suppressor of hairless), a protein central in the Notch signalling pathway. EDEN-BP knockdown reduced the efficiency of deadenylation targeted by the 3′-UTR of XSu(H) and caused an overexpression of XSu(H) mRNA. Also, overexpressing XSu(H) caused segmentation defects. XSu(H) is required for the transcriptional activation of Notch target genes and our data indicate that its overexpression disrupts the correct regulation of these genes.

In vivo inhibition of EDEN-BP causes segmentation defects

Figure 1
In vivo inhibition of EDEN-BP causes segmentation defects

Top panel: inhibiting the translation of EDEN-BP mRNA by morpholino oligonucleotides (Mo) causes an accumulation of XSu(H) mRNA. The co-injection of an mRNA encoding EDEN-BP (modified to escape down-regulation by the Mo) prevents XSu(H) overexpression. Ni, not-injected. Middle panel: XSu(H) is a central element in Notch signalling and the mRNAs encoding many proteins implicated in this pathway are expressed in a striped pattern (A, C). Inhibiting EDEN-BP expression by Mo causes important modifications in the expression patterns of at least two of the mRNAs encoding Notch signalling proteins, ESR5 and X-Delta 2 (B, D). S1–S4 correspond to four presumptive somites. Right panel: inhibiting EDEN-BP causes segmentation defects. The lower half of the embryos is derived from one cell of a two-cell embryo injected with anti-EDEN-BP antibodies. The upper half, derived from the non-injected cell, serves as a control. n, notochord; s, somite. Scale bar, 50 μm. Modified from [11] with permission.

Figure 1
In vivo inhibition of EDEN-BP causes segmentation defects

Top panel: inhibiting the translation of EDEN-BP mRNA by morpholino oligonucleotides (Mo) causes an accumulation of XSu(H) mRNA. The co-injection of an mRNA encoding EDEN-BP (modified to escape down-regulation by the Mo) prevents XSu(H) overexpression. Ni, not-injected. Middle panel: XSu(H) is a central element in Notch signalling and the mRNAs encoding many proteins implicated in this pathway are expressed in a striped pattern (A, C). Inhibiting EDEN-BP expression by Mo causes important modifications in the expression patterns of at least two of the mRNAs encoding Notch signalling proteins, ESR5 and X-Delta 2 (B, D). S1–S4 correspond to four presumptive somites. Right panel: inhibiting EDEN-BP causes segmentation defects. The lower half of the embryos is derived from one cell of a two-cell embryo injected with anti-EDEN-BP antibodies. The upper half, derived from the non-injected cell, serves as a control. n, notochord; s, somite. Scale bar, 50 μm. Modified from [11] with permission.

New EDEN-BP targets

If XSu(H) was the only mRNA targeted by EDEN-BP in the PSM, then a morpholino-induced down-regulation of this gene product should rescue the phenotype caused by EDEN-BP inactivation. This was not observed, which implies that other mRNAs expressed in the PSM are also targets for this regulatory pathway [11]. As a first step in the identification of such mRNAs, a DNA microarray screen was initiated to analyse specific changes in the adenylation status of Xenopus tropicalis maternal mRNAs. Of the 3000 gene products tested, approx. 500 showed detectable adenylation changes and could be classified into seven characteristic behaviours (A. Graindorge, Y. Audic, R. Thuret, N. Pollet and H.B. Osborne, unpublished work).

Perspectives

Future work on the molecular characterization of the EDEN-dependent deadenylation mechanism will involve the identification of interacting proteins and a characterization of their functional roles coupled with a wide scale identification of the mRNA targets and a description of their expression patterns in the embryos and adult tissues.

Stem Cells and Development: A Focus Topic at BioScience2005, held at SECC Glasgow, U.K., 17–21 July 2005. Edited by T. Kouzarides (Cambridge, U.K.), S. Newbury (Newcastle upon Tyne, U.K.), B. Richardson (University College London, U.K.), R. Sablowski (John Innes Centre, Norwich, U.K.), D. Tosh (Bath, U.K.), M. Welham (Bath, U.K.) and A. Willis (Nottingham, U.K.).

Abbreviations

     
  • EDEN

    embryo deadenylation element

  •  
  • EDEN-BP

    EDEN-binding protein

  •  
  • PSM

    pre-somitic mesoderm

  •  
  • TBD

    tail bud domain

  •  
  • 3′-UTR

    3′-untranslated region

  •  
  • XSu(H)

    Xenopus suppressor of hairless

The work described in this review was supported by grants from ARC (no. 9529 to H.B.O.; no. 4791 to L.P. and postdoctoral fellowship to Y.A.), the French Ministry of Research BCMS ACI no. 314, the CNRS ‘Xénopuces’ project (no. 02N65/0029 to N.P.) and the LRCC (Bretagne).

References

References
1
Paillard
L.
Omilli
F.
Legagneux
V.
Bassez
T.
Maniey
D.
Osborne
H.B.
EMBO J.
1998
, vol. 
17
 (pg. 
278
-
287
)
2
Voeltz
G.K.
Steitz
J.A.
Mol. Cell. Biol.
1998
, vol. 
18
 (pg. 
7537
-
7545
)
3
Duval
C.
Bouvet
P.
Omilli
F.
Roghi
C.
Dorel
C.
LeGuellec
R.
Paris
J.
Osborne
H.B.
Mol. Cell. Biol.
1990
, vol. 
10
 (pg. 
4123
-
4129
)
4
Audic
Y.
Omilli
F.
Osborne
H.B.
Mol. Cell. Biol.
1997
, vol. 
17
 (pg. 
209
-
218
)
5
Paillard
L.
Osborne
H.B.
Biol. Cell
2003
, vol. 
95
 (pg. 
211
-
219
)
6
Audic
Y.
Omilli
F.
Osborne
H.B.
Mol. Cell. Biol.
1998
, vol. 
18
 (pg. 
6879
-
6884
)
7
Ueno
S.
Sagata
N.
Dev. Biol.
2002
, vol. 
250
 (pg. 
156
-
167
)
8
Ezzeddine
N.
Paillard
L.
Capri
M.
Maniey
D.
Bassez
T.
Aït-Ahmed
O.
Osborne
B.
Proc. Natl. Acad. Sci. U.S.A.
2002
, vol. 
99
 (pg. 
257
-
262
)
9
Paillard
L.
Legagneux
V.
Maniey
D.
Osborne
H.B.
J. Biol. Chem.
2002
, vol. 
277
 (pg. 
3232
-
3235
)
10
Paillard
L.
Legagneux
V.
Osborne
H.
Biol. Cell
2003
, vol. 
95
 (pg. 
107
-
113
)
11
Gautier-Courteille
C.
Le Clainche
C.
Barreau
C.
Audic
Y.
Graindorge
A.
Maniey
D.
Osborne
H.B.
Paillard
L.
Development
2004
, vol. 
131
 (pg. 
6107
-
6117
)