Hormonal stimulation of starfish oocytes induces partial degradation of the 3' termini of cyclin B mRNAs with oligo(U) tails, followed by poly(A) elongation

Abstract

In yeast, plant, and mammalian somatic cells, short poly(A) tails on mRNAs are subject to uridylation, which mediates mRNA decay. Although mRNA uridylation has never been reported in animal oocytes, maternal mRNAs with short poly(A) tails are believed to be translationally repressed. In this study, we found that 96% of cyclin B mRNAs with short poly(A) tails were uridylated in starfish oocytes. Hormonal stimulation induced poly(A) elongation of cyclin B mRNA, and 62% of long adenine repeats did not contain uridine residues. To determine whether uridylated short poly(A) tails destabilize cyclin B mRNA, we developed a method for producing RNAs with the strict 3' terminal sequences of cyclin B, with or without oligo(U) tails. When we injected these synthetic RNAs into starfish oocytes prior to hormonal stimulation, we found that uridylated RNAs were as stable as nonuridylated RNAs. Following hormonal stimulation, the 3' termini of short poly(A) tails of synthesized RNAs containing oligo(U) tails were trimmed, and their poly(A) tails were subsequently elongated. These results indicate that uridylation of short poly(A) tails in cyclin B mRNA of starfish oocytes does not mediate mRNA decay; instead, hormonal stimulation induces partial degradation of uridylated short poly(A) tails in the 3'-5' direction, followed by poly(A) elongation. Oligo(U) tails may be involved in translational inactivation of mRNAs.

Keywords: mRNA uridylation; oligo(U); oocyte; poly(A); polyadenylation; starfish.

FIGURE 1.

Uridylated short poly(A) tail of cyclin B mRNA in starfish oocytes and nonuridylated long poly(A) tail of cyclin B mRNA in oocytes stimulated with the hormone 1-MA. (A) Elongation of the 3′ region of mRNA after hormonal stimulation. Adaptor-ligated cyclin B mRNAs from oocytes treated with (+) or without (−) 1-MA were RT-PCR amplified. The products were separated by agarose gel electrophoresis and visualized by ethidium bromide staining. (B) Sequencing results of the 3′ terminal region of cyclin B mRNA from oocytes treated with [(+) 1-MA treatment] or without 1-MA [(−) 1-MA treatment]. Gray shaded sequences show the 3′ terminal portion of the 3′ UTR, and the following sequences are the tail regions of cyclin B mRNA. The number of clones obtained is indicated. (C) Histogram of the number of uridine residues at the 3′ ends of cyclin B mRNAs in starfish oocytes not subjected to hormonal stimulation. Inset shows box-plot analysis of oligo(U) tail length. The 25th and 75th percentiles are shown by box edges, the median value is indicated by the thick line, and whiskers show the maximum and minimum values. (D) Box plot analysis of poly(A) tail length of cyclin B mRNAs in oocytes. Median of poly(A) size increased from 12 to 50 nt after hormonal stimulation. (E) Trimming activity at the 3′ region of cyclin B mRNA after hormonal stimulation. GVBD occurred 18 min after 1-MA treatment. At the indicated time after hormonal stimulation, total RNA was purified from oocytes, and adaptor-ligated cyclin B mRNAs were RT-PCR amplified using 3′ adaptor Rev primer 1 and sfcycB short Fwd primer. The PCR products were subjected to high-resolution acrylamide gel electrophoresis (15%) and visualized by SYBR-Green I staining.

FIGURE 2.

The 3′ UTR of cyclin B mRNAs is sufficient to recapitulate endogenous poly(A) elongation. (A) Band shift of cycB-oligo(A) + U in injected oocytes after hormonal stimulation (left and middle lanes), and the control RT-PCR product from oocytes not subjected to RNA injection (right lane). (B) Sequencing results of the 3′ terminal region of cycB-oligo(A) + U before injection (Control), after injection without hormonal stimulation ([−] 1-MA treatment), and after injection with hormonal stimulation ([+] 1-MA treatment). (C) Quantitation of poly(A) tail lengths. Layout and labeling are the same as in B.

FIGURE 3.

Uridylated and nonuridylated RNAs were similarly stable in starfish oocytes. Asterisks in the schematics denote locations of radiolabels. (A) Similar stability of uniformly labeled uridylated and nonuridylated RNAs in oocytes without hormonal stimulation. Upper panel shows representative autoradiography results. The intensity of each band was calculated using ImageQuant. Lower graphs show the remaining amounts of injected RNA after 20 h incubation. Error bars represent standard deviation from more than three experiments. (B) An experiment similar to the one shown in A was performed using 3′-end radiolabeled cycB-oligo(A) + U or cycB-oligo(A). Upper panel shows representative autoradiography results, and lower graphs show the remaining amounts of injected RNA after 20 h incubation. Error bars represent standard deviation (n = 3). (C) Similar band shifts of uniformly labeled uridylated and nonuridylated RNAs in oocytes after hormonal stimulation. Bands for both cycB-oligo(A) + U and cycB-oligo(A) were shifted up after 45–60 min.