Peter J. Ford

Polyadenylic acid-containing RNA in Xenopus laevis oocytes☆

Abstract The quantity of poly(A)-containing RNA is measured in Xenopus laevis oocytes as a function of developmental stage. The amount of poly(A)-containing RNA per oocyte, 0.7 to 1.0% of the total RNA, remains relatively constant from early vitellogenesis until ovulation. It is largely present in the cytoplasm of the oocyte in the form of a ribonucleoprotein complex. The poly(A) sequence is approximately 100 bases in length and is attached to molecules of heterogeneous sedimentation coefficients.

Differential accumulation of two size classes of poly(A) associated with messenger RNA during oogenesis in Xenopus laevis.

Abstract The RNA of full-grown oocytes of Xenopus laevis contains two distinct size classes of poly(A), designated poly(A) S and poly(A) L , which contain 15–30 (mean = 20) and 40–80 (mean = 61) A residues, respectively. Both poly(A) L and poly(A) S are associated with RNA which is heterogeneous in size. The two classes of poly(A) + RNA can be separated by affinity chromatography: Only poly(A) L + RNA binds to oligo(dT)-cellulose under appropriate conditions, but up to 50% of the poly(A) S + RNA can be isolated from the void fraction by binding to poly(U)-Sepharose. Both classes of poly(A) + RNA are active as messenger RNA in an in vitro system and yield identical patterns of in vitro protein products. Previtellogenic oocytes contain almost exclusively poly(A) L , which accumulates up to vitellogenesis but remains almost constant in amount (molecules/oocyte) during vitellogenesis and in the full-grown oocyte. Poly(A) S accumulates (molecules/oocyte) from early vitellogenesis up to the full-grown oocyte. The total number of poly(A) + RNA molecules per oocyte increases throughout oogenesis from 2 × 10 10 /previtellogenic oocyte [80–90% poly(A) L ] to 20 × 10 10 /full-grown oocyte (25–40% poly(A) L ). It is argued that poly(A) S is protected from degradation in the oocyte, thus stabilizing the “maternal” poly(A) + mRNA.

Sequences of 5S ribosomal RNA from Xenopus mulleri and the evolution of 5S gene-coding sequences

Sequences for 5S RNA from somatic cells and oocytes of Xenopus mulleri are presented. Comparison with sequences previously given for Xenopus laevis indicates that the somatic 5S RNA genes of each species are more closely related to each other than either is to its own set of oocyte genes, suggesting that somatic and oocyte genes within each species are evolving independently. However, detailed analysis of sequence variants in each species suggests that there is a mechanism which allows occasional genetic exchanges between somatic and oocyte-specific genes. Possible genetic mechanisms which allows such an exchange are discussed.

Cell-free translation of messenger RNA from oocytes of Xenopus laevis.

Abstract The poly(A)+ RNA which accumulates during oogenesis in the amphibian Xenopus laevis is shown to be functional mRNA; the RNA was active in the mRNA-dependent “shift assay” for initiation sites in the rabbit reticulocyte lysate, and was an efficient template for protein synthesis in the wheat-germ cell-free system. Analysis of the in vitro protein products showed no differences between the coding properties of poly(A)+ RNA extracted from oocytes at all stages of development from previtellogenesis to maturity. In previtellogenic oocytes, the in vitro products of polysomal and of mRNP-associated poly(A)+ RNA were also identical. Neither was there any evidence for changes in the coding properties of the poly(A)+ mRNA of the oocyte. However, the patterns of oocyte in vivo protein synthesis changed markedly during early vitellogenesis. We conclude that the mRNP-associated poly(A)+ RNA present in mature oocytes constitutes the stored maternal mRNA, and that during oogenesis the coding composition of the poly(A)+ mRNA synthesised does not change markedly, while some form of translational control operates to direct the changing pattern of protein synthesis.

Very long-lived messenger RNA in ovaries of Xenopus laevis

It is possible to label with radioactivity newly synthesized ovarian RNA after intraperitoneal injection of [3H]guanosine and [3H]uridine into immature Xenopus laevis, if ovaries in which only previtellogenic stage 1 oocytes are present. Following the amount of radioactivity in the ovarian pool of acid-soluble precursors indicates a complete clearance of acid-soluble radioactivity within 15–20 days after injection. Incorporation of radioactivity into total RNA (which is almost exclusively 4 and 5S RNAs at this stage) and poly(A)+ RNA ceases between 15 and 20 days after injection, but the total amount of radioactivity in these RNA fractions does not decline appreciably over the next 18 months. During this time, the ovary grows and develops since stage 6 oocytes eventually appear and there is a 10- to 20-fold increase in total RNA content, which changes in composition from almost exclusively (95%) 4 and 5S RNAs to mainly (75%) 18 and 28S RNAs. Thus, despite continued growth and development, radioactive RNA molecules synthesized during previtellogenesis survive for lengths of time commensurate with the length of oogenesis (1–2 years). Although very limited (<7%) reincorporation of radioactivity into RNA is detected, it cannot alone account for the stability of the label in poly(A)+ RNA. These results are interpreted as indicative of synthesis during previtellogenesis of tRNA, 5SrRNA, and messenger RNA molecules which are very long-lived.

Turnover and processing of poly(A) in full-grown oocytes and during progesterone-induced oocyte maturation inXenopus laevis

Changes in the amount and the size of poly(A) sequences during progesterone-induced oocyte maturation were determined by [3H]poly(U) hybridization and by labeling with microinjected [α-32P]ATP and [3H]ATP. Following exposure to progesterone total poly(A) length increases by 10–20 nucleotides, largely due to extension of preexisting poly(A) although there is an increase of 5 to 10% in the number of poly(A) sequences. Immediately following germinal vesicle breakdown (GVBD), poly(A) is degraded with first-order kinetics (t1/2 = 5.7 hr) until about 10 hr after GVBD. This degradation results in the loss of 35–50% of the poly(A) sequences but the size distribution of the remaining poly(A) is unchanged. These changes occur only in oocytes which undergo GVBD, but all the changes in poly(A) content also occur in oocytes enucleated before exposure to progesterone. The poly(A) of ovulated eggs is similar in amount and size to that ofin vitro matured oocytes. The incorporation of [3H]ATP into poly(A) confirms that most of the poly(A) synthesis in both maturing and control oocytes is due to chain extension and turnover rather than tode novo synthesis. The amount of newly synthesized poly(A) is sufficient to allow for the net increase in larger poly(A) sequences by chain extension but not byde novo synthesis. Newly synthesized poly(A) has a size distribution similar to that of steady-state poly(A), and no precursor-like32P-poly(A) molecules are detected. The lengths of poly(A) sequences labeled with [3H]ATP, determined by [3H]AMP:[3H]adenosine ratios, are smaller than those indicated by their electrophoretic mobilities in both control and maturing oocytes. It is argued that the size and stability of oocyte poly(A) sequences are regulated by balanced synthetic and degradative activities of cytoplasmic poly(A) polymerase which result in turnover of the 3′-terminal portion of the poly(A). During progesterone-induced maturation there is a relative increase in poly(A) synthesis until the time of GVBD, when a different degradative activity results in destruction of complete poly(A) sequences. These changes must be cytoplasmically programmed because they occur in the absence of the nucleus.

Regulation of protein synthesis and accumulation during oogenesis in Xenopus laevis

Abstract A two-dimensional polyacrylamide gel electrophoresis procedure has been used to study ribonucleoprotein synthesis and accumulation at three stages (Dumont stages 1, 2, and 6) of oogenesis in Xenopus laevis . Ribonucleoprotein particles are separated effectively by sucrose gradient centrifugation of oocyte homogenates. Extraction of protein from different sucrose gradient fractions has enabled identification of four classes of ribonucleoprotein (7 S RNP, 42 S RNP, messenger RNP, and 80 S ribosome RNP). Proteins were extracted from subcellular fractions of defined numbers of defollicled oocytes which had been incubated with [ 35 S]methionine for 20 hr previously. Electrophoresis first in a Triton X-100-acid-urea polyacrylamide gel followed by a sodium dodecyl sulphate-gradient gel system adequately resolved most of these ribonucleoproteins. A rough estimate of the relative amount of a particular protein was obtained by visual inspection of Coomassie blue-stained gels. Estimates of the relative rates of synthesis of particular proteins were obtained by visual inspection of fluorographs prepared from the gels. Each of the protein classes identified follows a distinct and different pattern of synthesis and accumulation during oogenesis, and appears to be coordinately regulated with the RNA molecule to which it associates.

Control of 5S RNA synthesis in Xenopus laevis

PREVIOUS reports1–3 have shown that somatic cells of Xenopus laevis synthesise a 5S RNA which differs in sequence from 5S RNA synthesised by oocytes. In none of these reports is there any indication of ovary-type sequences being synthesised by somatic cells. Moreover, although the oocyte 5S sequences are heterogeneous, the somatic sequence is essentially homogeneous. We report here that X. laevis liver cells labelled in organ culture do not detectably synthesise ovary-type 5S RNA sequences but find that a transformed tissue culture cell line, derived initially from somatic kidney cells, synthesises significant (10–20%) amounts of ovary-type 5S RNA sequences. Taken together with the observation that some 5S genes are translocated to a novel site adjacent to the nucleolar organiser in these cells4 this result is important for an understanding of the nature of differential control of 5S gene transcription in X. laevis.