Translational control of maskin mRNA by its 3' untranslated region

Abstract

Background information: Maskin is a member of the TACC (transforming acidic coiled-coil) domain proteins found in Xenopus laevis oocytes and embryos. It has been implicated in the co-ordination of the spindle and has been reported to mediate translational repression of cyclin B1 mRNA.

Results: In the present study, we report that maskin mRNA is translationally repressed at the level of initiation in stage 4 oocytes and becomes activated in stage 6 oocytes. The translational repression of maskin mRNA correlates with the presence of a short poly(A) tail on this mRNA in stage 4 oocytes. The 3'-UTR (untranslated region) of maskin can confer the translational regulation to a reporter mRNA, and so can the 3'-UTR of human TACC3. A conserved GUCU repeat element was found to repress translation in both stage 4 and stage 6 oocytes, but deletion of this element did not abrogate repression in stage 4 oocytes. UV cross-linking experiments indicated that overlapping sets of proteins bind efficiently to both the maskin and the cyclin B1 3'-UTRs. As reported previously, CPEB [CPE (cytoplasmic polyadenylation element)-binding protein] binds to the cyclin B1 3'-UTR, but its binding to the maskin 3'-UTR is minimal. By RNA affinity chromatography and MS, we identified the EDEN-BP [EDEN (embryonic deadenylation element)-binding protein] as one of the proteins binding to both the maskin and the cyclin B1 3'-UTRs.

Conclusions: Maskin mRNA is translationally regulated by at least two repressor elements and an activation element. One of the repessor elements is the evolutionarily conserved GUCU repeat. EDEN-BP binds to both the maskin and cyclin B1 3'-UTRs, indicating it may be involved in the deadenylation of these mRNAs.

Figure 1

Maskin protein appears in late oogenesis, long after its mRNA Western blots for cyclin B1, maskin and CPEB (protein panels) and Northern blots for cyclin B1 and maskin mRNA (mRNA panels). Equal numbers of oocyte equivalents from the same batch of oocytes were loaded in each lane. Oocyte stages according to Dumont. 4E: early stage 4 oocytes (0.6-0.8 mm diameter), 4L: late stage 4 oocytes (0.8-1 mm). M: mature stage 6 oocytes.

Figure 2

Maskin mRNA in early stage 4 oocytes is not bound to polyribosomes and has a short poly(A) tail. (A) protein: Western for maskin on stage 4-6 oocytes. RNA: RT-PCR for maskin mRNA on total RNA from stage 4-6 oocytes. Pellet: RT-PCR on RNA from the polyribosomal pellet (translated mRNA). Supernatant: RT-PCR on mRNA from the free mRNP (untranslated mRNA). Marker sizes in KDa are indicated on the left of the panel. (B) Stage 6 oocyte extracts were treated with EDTA to release the ribosomes before pelleting (EDTA) or not, the RNA isolated from the fractions was amplified as in A. (C) Poly(A) test on stage 4 and stage 6 oocyte RNA. PCR: PCR using a fixed forward primer and a reverse primer at the end of the 3' UTR, thus amplifying the 3' UTR excluding the poly(A) tail. PAT: PCR using a fixed forward primer and the adaptor primer wich amplifies the same region including the poly(A) tail. Marker sizes are indicated in numbers of base pairs.

Figure 3

The maskin 3' UTR controls translation by multiple elements (A) The maskin and TACC3 3' UTRs are depicted with the GUCU repeats and the poly(A) signal (AAUAAA) in bold, the potential weak CPEB binding site shaded and the UGU trinucleotides underlined. (B) CAT activity in injected oocytes. CAT reporter mRNAs were injected into stage 4 and stage 6 oocytes and incubated for 7 hours. CAT enzyme activity is expressed as percentage of the activity obtained for the CAT control mRNA. Control: control CAT mRNA, MSK: CAT mRNA with the full maskin 3' UTR, GUCU: CAT mRNA with the GUCU repeat in the 3' UTR, MSKdGUCU: CAT mRNA with the maskin 3' UTR lacking the GUCU repeat, TACC3: CAT mRNA with the full TACC3 3' UTR. Error bars indicate 1 standard deviation. (C) mRNA levels at the start and end of the incubation period as determined by RNase protection.

Figure 4

Multiple proteins including EDEN-BP bind to the maskin and cyclin B1 3' UTRs (A) UV crosslinking assay with stage 4 extract on radioactive vector (BS), maskin 3' UTR (MSK) and cyclin B1 (sB1) 3' UTR transcripts. (B) UV crosslinking as above (UV) followed by immunoprecipitation with αCPEB antibody. (C) RNA affinity purification with RNAs as above, separated on SDS-PAGE and stained with silver stain. Triangles indicate CPEB (white) and EDEN-BP (black), as identified by mass spectrometry. The circles are bubbles trapped under the gel during photography. (D) Comparison of EDEN-BP and MGC69034. The peptides identified by mass spectrometry are boxed.

Figure 5

EDEN-BP binds to the full length maskin 3' UTR only (A) Western for CPEB and EDEN-BP on RNA affinity purification eluates from beads bearing control (BS), maskin 3' UTR (MSK) and cyclin B1 3' UTR (sB1) RNAs. (B) UV crosslinking and immunoprecipitation with a control rabbit IgG and the EDEN-BP antibody using radioactive forms of the RNAs listed for (A). (C) RT-PCR for maskin, c-mos and ribosomal protein S6 (rsp6) on RNA isolated from immunoprecipitates obtained with control (IgG) and EDEN-BP antibody. RT+/RT−: reverse transcriptase added or omitted. (D) UV crosslinking followed by immunoprecipitation as in (B) using the control (BS), full maskin 3' UTR (MSK), the GUCU repeat alone (GUCU), the maskin 3' UTR lacking the GUCU repeat (MSKdGUCU) or the EDEN from the c-Mos 3' UTR (Mos EDEN). (E) Western for EDEN-BP on RNA affinity purification eluates from RNAs as described in (D). (F) UV crosslinking and IP with capped and polyadenylated CAT mRNAs as used in Fig. 3.