Urea is Necessary for the Culture of Embryos of the Marsupial Frog Gastrotheca riobambae, and is Tolerated by Embryos of the Aquatic Frog Xenopus laevis

Urea was found in the capsular fluid that bathes Gastrotheca riobambae embryos during incubation in the maternal pouch. The urea concentration in this fluid is higher than in blood from the mother, indicating that urea is accumulated by the embryo during the period of maternal incubation. Gastrotheca tadpoles tolerate up to 500 mM urea with 86% survival after 24 hours and die in solutions of 0.5 mM ammonia. These findings suggest that urea plays a role in the adaptation of G. riobambae embryos to the conditions of water stress within the maternal pouch. To improve the in vitro culture conditions of early embryos taken from the maternal pouch, a saline solution that contains urea was designed (GRS). GRS plus 30 mM urea was used for the culture of cleavage to the neurula stage embryos of G. riobambae. During organogenesis, the urea concentration was raised to 60 mM. Early embryos of Xenopus laevis tolerate urea, and in addition, no inducing effects of urea have been detected in animal cap explants of Xenopus .     

Gastrulation of Gastrotheca riobambae in comparison with other frogs

Blastopore formation, the embryonic disk, archenteron and notochord elongation, and Brachyury expression in the marsupial frog Gastrotheca riobambae was compared with embryos of Xenopus laevis and of the dendrobatids Colostethus machalilla and Epipedobates anthonyi. In contrast with X. laevis embryos, the blastopore closes before elongation of the archenteron and notochord in the embryos of G. riobambae and of the dendrobatid frogs. Moreover, the circumblastoporal collar (CBC) thickens due to the accumulation of involuted cells. An embryonic disk, however, is formed only in the G. riobambae gastrula. We differentiate three gastrulation patterns according to the speed of development: In X. laevis, elongation of the archenteron and notochord begin in the early to mid gastrula, whereas in the dendrobatids C. machalilla and E. anthonyi the archenteron elongates at mid gastrula and the notochord elongates after gastrulation. In G. riobambae, only involution takes place during gastrulation. Archenteron and notochord elongation occur in the post gastrula. In the non-aquatic reproducing frogs, the margin of the archenteron expands anisotropically, resulting in an apparent displacement of the CBC from a medial to a posterior location, resembling the displacement of Hensen's node in the chick and mouse. The differences detected indicate that amphibian gastrulation is modular.     

Chromosome banding in Amphibia. XXX. Karyotype aberrations in cultured fibroblast cells

The present study reports for the first time on the numerical and structural chromosome anomalies that spontaneously arise in aging cultured fibroblast cells of Amphibia. The analyses were conducted on kidney fibroblasts of three anuran species with extremely divergent genome sizes (Bufo rubropunctatus, Scaphiopus holbrooki, Gastrotheca riobambae), in the sixth up to the 14th culture passage. The chromosomal rearrangements were identified by means of the 5-bromodeoxyuridine/deoxythymidine (BrdU/dT) replication banding technique. The aberrations can be either confined to a single chromosome, or else involve all chromosomes of the karyotype. The most frequent structural aberrations in the cell cultures of S. holbrooki and G. riobambae are tandem fusions between two or more chromosomes. These tandem fusions originating in vitro in long-termed cell cultures reflect the chromosome mutations which also took place during amphibian phylogenesis.     

The gastrocoel roof plate in embryos of different frogs

The morphology of the gastrocoel roof plate and the presence of cilia in this structure were examined in embryos of four species of frogs. Embryos of Ceratophrys stolzmanni (Ceratophryidae) and Engystomops randi (Leiuperidae) develop rapidly, provide comparison for the analysis of gastrocoel roof plate development in the slow-developing embryos of Epipedobates machalilla (Dendrobatidae) and Gastrotheca riobambae (Hemiphractidae). Embryos of the analyzed frogs develop from eggs of different sizes, and display different reproductive and developmental strategies. In particular, dorsal convergence and extension and archenteron elongation begin during gastrulation in embryos of rapidly developing frogs, as in Xenopus laevis. In contrast, cells that involute during gastrulation are stored in the large circumblastoporal collar that develops around the closed blastopore in embryos of slow-developing frogs. Dorsal convergence and extension only start after blastopore closure in slow-developing frog embryos. However, in the neurulae, a gastrocoel roof plate develops, despite the accumulation of superficial mesodermal cells in the circumblastoporal collar. Embryos of all four species develop a ciliated gastrocoel roof plate at the beginning of neurulation. Accordingly, fluid-flow across the gastrocoel roof plate is likely the mechanism of left-right asymmetry patterning in these frogs, as in X. laevis and other vertebrates. A ciliated gastrocoel roof plate, with a likely origin as superficial mesoderm, is conserved in frogs belonging to four different families and with different modes of gastrulation.     

Role of progesterone on oocyte maturation in the egg-brooding hylid frog Gastrotheca riobambae (Fowler)

In the egg-brooding frog Gastrotheca riobambae (Fowler), oocyte maturation is comparable to the situation of other frog species. In isolated follicles, progesterone induces only germinal vesicle breakdown (GVBD), while human chorionic gonadotropin (hCG) induces GVBD and ovulation. In addition, defolliculated oocytes respond with GVBD to the treatment with progesterone, while hCG has no effect. As in other frogs, oocyte maturation in vitro depends on hormonal action and on the presence of divalent cations. In this frog, progesterone or a similar hormone conditions the brooding pouch for reproduction and induces pouch closure. Follicles from frogs with closed pouches showed GVBD after 15-17 hours of incubation with progesterone, while those from frogs with open pouches took 19-24 hours for GVBD. These findings suggest that follicles become stimulated for maturation when the pouch is closed and that this stimulated condition is maintained for several weeks in advance of the process of oocyte maturation. In G. riobambae, the external appearance of the pouch aperture indicates the reproductive condition of the ovary.     

Development of the dendrobatid frog Colostethus machalilla

To provide a developmental correlate with other frogs, we prepared a normal table of development for the dendrobatid, Colostethus machalilla and analyzed the morphology of its early development. This frog reproduces in captivity and deposits moderately sized eggs (1.6 mm in diameter) in terrestrial nests. The father guards the embryos until tadpole hatching. We divided development until hatching into 25 stages and implemented methods for in vitro culture of the embryos. The external and internal morphology of embryos were evaluated by observations in whole mount and in sections. Neural, notochord and somite specific antibodies were used to analyze gene expression patterns by immunostaining of embryos. Embryonic development of C. machalilla is slow and deviates from Xenopus laevis. In C. machalilla the elongation of the notochord, neural plate and somite formation occur after blastopore closure, possibly due to a delay in the dorsal convergence and extension movements. The gastrula of C. machalilla also deviates from X. laevis. The archenteron remains small until blastopore closure, where small cells accumulate at the blastopore lips. Simultaneously, the blastocoel roof thins until it becomes a monolayer of cells. Although C. machalilla does not form an embryonic disk, its thick blastopore lips resemble the embryonic disk of the marsupial frog Gastrotheca riobambae and represent an interesting deviation from the gastrulation pattern observed in X. laevis.     

Chromosomal mapping of Xenopus 5S genes: somatic-type versus oocyte-type.

Xenopus 5S RNA genes exhibit a pattern of differential expression during development in which some members (oocyte-type) are transcribed only in oocytes, while others (somatic-type) are expressed in both oocytes and somatic cells. Using cloned DNA probes specific for each gene type, we determined the positions of these genes on Xenopus metaphase chromosomes by in situ hybridization. Somatic-type 5S genes in both X. laevis and X. borealis are located at the distal end of the long arm of only one chromosome (number 9). The oocyte-type 5S RNA genes are found at the distal ends of the long arms of most Xenopus chromosomes, including chromosome 9. Thus, large scale differences in chromosomal location cannot explain the selective expression of these genes, as suggested previously.     

The effect of insulin on intracellular ph and ribosomal protein S6 phosphorylation in oocytes of Xenopus laevis

Both insulin and progesterone are capable of stimulating germinal vesicle breakdown (GVBD) of large, Stage VI oocytes of Xenopus laevis. Numerous studies have shown an increase in intracellular pH (pHi) and ribosomal protein S6 phosphorylation prior to GVBD in oocytes treated with progesterone. In this study the effect of insulin and progesterone on pHi and S6 phosphorylation was compared. Both hormones increased pHi and S6 phosphorylation to similar levels and the time course of pHi change was the same for both hormones. Half-maximal effects of insulin were observed at 7 X 10(-8) M concentrations. In the presence of 1 nM cholera toxin, the ability of progesterone to induce these two responses was inhibited while the action of insulin was unaffected. However, GVBD induced by either hormone was blocked by cholera toxin. In small, Stage IV oocytes that do not undergo GVBD in response to either progesterone or insulin, a partial increase in pHi without S6 phosphorylation occurred in response to progesterone but both events occurred in response to insulin. These results suggest that the inability of Stage IV oocytes to undergo GVBD in response to hormone is not due to a failure to increase pHi or phosphorylate S6. The results in this paper also indicate that these events are regulated differently by insulin and progesterone in Xenopus oocytes.     

Identification of the Xenopus 20S proteasome alpha4 subunit which is modified in the meiotic cell cycle.

The proteasomes are large, multi-subunit particles that act as the proteolytic machinery for most of the regulated intracellular protein degradation in eukaryotic cells. To investigate the regulatory mechanism for the 26S proteasome in cell-cycle events, we purified this proteasome from immature and mature oocytes, and compared its subunits. Immunoblot analysis of 26S proteasomes showed a difference in the subunit of the 20S proteasome. A monoclonal antibody, GC3beta, cross-reacted with two bands in the 26S proteasome from immature oocytes (in G2-phase); however, the upper band was absent in the 26S proteasome from mature oocytes (in M-phase). These results suggest that changes in the subunits of 26S proteasomes are involved in the regulation of the meiotic cell cycle. Here we describe the molecular cloning of one of the alpha subunits of the 20S proteasome from a Xenopus ovarian cDNA library using an anti-GC3beta monoclonal antibody. From the screening, two types of cDNA are obtained, one 856bp, the other 984bp long. The deduced amino-acid sequences comprise 247 and 248 residues, respectively. These deduced amino-acid sequences are highly homologous to those of alpha4 subunits of other vertebrates. Phosphatase treatment of 26S proteasome revealed the upper band to be a phosphorylated form of the lower band. These results suggest that a part of the alpha4 subunit of the Xenopus 20S proteasome, alpha4_xl, is phosphorylated in G2-phase and dephosphorylated in M-phase.     

Biochemical research on oogenesis: Oocytes and liver cells of the teleost fish Tinca tinca contain different kinds of 5S RNA

Previtellogenic oocytes of Tinca tinca accumulate very large amounts of 5S RNA. We show here that 5S RNA stored in oocytes differs from liver 5S RNA in 3 out of 120 nucleotides. Liver and oocyte 5S RNAs, therefore, are produced by different genes. Both kinds of 5S genes are active in oocytes. However, only 5S RNA of the oocyte type accumulates in these cells. In Tinca tinca as in Xenopus laevis, oocyte-type and somatic-type 5S RNAs differ by three properties, ie., primary structure, conformation, and metabolic stability. Nucleotide substitutions occur in different positions in oocyte and somatic 5S RNAs of Tinca tinca and Xenopus laevis. We do not understand how different sets of nucleotide substitutions confer to 5S RNAs of both species similar properties in vivo, namely, increased metabolic stability.     

Four-dimensional imaging of cytoskeletal dynamics in Xenopus oocytes and eggs

The Xenopus laevis (African clawed frog) system has long been popular for studies of both developmental and cell biology, based on a variety of its intrinsic features including the large size of Xenopus oocytes, eggs, and embryos, and the relative ease of manipulation. Unfortunately, the large size has also been considered a serious impediment for high-resolution light microscopy, as has the heavy pigmentation. However, the recent development and exploitation of 4D imaging approaches, and the fact that much of what is of most interest to cell and developmental biologists takes place near the cell surface, indicates that such concerns are no longer valid. Consequently, the Xenopus system in many respects is now as good as other model systems considered to be ideal for microscopy-based studies. Here, 4D imaging and its recent applications to cytoskeletal imaging in Xenopus oocytes and eggs are discussed.     

Brix from xenopus laevis and brx1p from yeast define a new family of proteins involved in the biogenesis of large ribosomal subunits

A clone was isolated from a cDNA library from early embryos of Xenopus laevis that codes for a highly charged protein containing 339 amino acids. Two putative nuclear localization signals could be identified in its sequence, but no other known motifs or domains. Closely related ORFs are present in the genomes of man, C. elegans, yeast and Arabidopsis. A fusion protein with GFP expressed in HeLa cells or Xenopus oocytes was found to be localized in the nucleolus and coiled (Cajal) bodies. Moreover, immunoprecipitation experiments demonstrated that the new Xenopus protein interacts with 5S, 5.8S and 28S RNAs of large ribosomal subunits. The name Brix (biogenesis of ribosomes in Xenopus) is proposed for this protein and the corresponding gene. In Saccharomyces cerevisiae, the essential gene YOL077c, now named BRX1, codes for the Brix homolog, which is also localized in the nucleolus. Depletion of Brx1 p in a conditional yeast mutant leads to defects in rRNA processing, and a block in the assembly of large ribosomal subunits.     

Biochemical research on oogenesis: Oocytes of Xenopus laevis synthesize but do not accumulate 5S RNA of somatic type

The oocytes of amphibians and teleosts begin to accumulate 5S RNA several months before other components of the ribosomes become available. Two types of genes coding for 5S RNA are active during oogenesis of these animals. One type of genes is expressed only in oocytes. The other type is expressed in both oocytes and somatic cells. In this paper, we show that the oocytes of Xenopus laevis do not accumulate 5S RNA of somatic type. We conclude that the products of the two types of genes behave differently during oogenesis. One product is stored by the oocytes, whereas the other is not. The heterogeneity of 5S genes in Xenopus laevis might have arisen because oocytes and somatic cells needed different kinds of 5S RNA. These needs are met by molecules having different primary structures, different conformations, and different metabolic stabilities in vivo. We do not understand how these properties are related to one another.     

Cysteine string proteins are associated with cortical granules of Xenopus laevis oocytes

Cysteine string proteins (csps) are associated with secretory organelles in a wide range of eukaryotic cells. Functional studies of these proteins indicate that they subserve one or more vital steps in the pathway of regulated exocytosis. Here, we document the presence of csps in fully grown (stage VI) oocytes of the frog, Xenopus laevis. Both Northern and immunoblot data support the conclusion that csps are expressed in these cells. In addition, immunoreactive csp is seen even at the earliest stage of oocyte development, namely, in stage I oocytes. Finally, immunoblot and immunocytochemical results indicate that csps are associated with cortical granules of stage II-VI oocytes. These observations suggest that csps participate in the cortical reaction that underlies the sustained block to polyspermy in Xenopus eggs. Moreover, because of the relative ease of manipulating cells as large as Xenopus oocytes, this system harbors considerable promise as a model for studying the role of csps and other proteins in exocytotic events.     

Analysis of the p21 ras system during development of meiotic competence in Xenopus laevis oocytes.

The ability of Xenopus oocytes to undergo insulin- or insulin-like growth factor 1-induced meiotic maturation develops during oogenesis, with cells 1.0 mm in diameter or larger responding in a size-dependent manner. Since insulin-induced oocyte maturation was shown previously to be p21 ras-dependent, experiments were performed to test whether a deficiency in the p21 ras system might account for meiotic incompetence in small oocytes (less than or equal to 0.9 mm diameter). Both small and large oocytes were found to contain comparable levels of membrane-associated p21, as determined by protein immunoblotting. Treatment of both small and large oocytes with 2 microM insulin for 2 hr increased endogenous levels of membrane-associated p21 by approximately 70%. Stimulation of microinjected p21-membrane association by insulin was observed to be both time- and concentration-dependent in large oocytes with an EC50 of 50 nM. In addition, comparable levels of GTPase activating protein were measured in extracts prepared from oocytes ranging from 0.8 to 1.3 mm in diameter. Therefore, the p21 system is apparently not limiting during oogenesis, and expression of some other cellular component must account for development of meiotic competence in Xenopus oocytes.