Secretory pathway function, but not cytoskeletal integrity, is required in poliovirus infection.

We examined the importance of two interactions between poliovirus and its host cell: the putative association between viral proteins and a rearranged intermediate filament (IF) network and the apparent requirement for functional vesicle budding machinery within the host-cell secretory pathway. Poliovirus capsid proteins appeared to associate with reorganized IF proteins during infection. Treatment of cells with cytochalasin D and nocodazole in combination disrupted normal cytoskeletal organization and prevented the poliovirus-induced redistribution of IF proteins to a juxtanuclear location. However, this treatment had no effect on viral yields from single-cycle infections, indicating that neither cytoskeletal integrity nor a specific poliovirus-induced rearrangement of IF proteins is required in the poliovirus life cycle. In contrast, we report that the inhibition of poliovirus replication by brefeldin A (BFA), an inhibitor of secretory membrane traffic, is specific to the host cell. Polioviral yields were not affected by BFA in two BFA-resistant cell lines, demonstrating that BFA targets a host protein or process required by poliovirus. No BFA-resistant virus was detected in these experiments, further supporting the hypothesis that poliovirus replication requires secretory pathway function, perhaps for the generation of vesicles on which viral RNA replication complexes are assembled.

Evidence that the deep keratin filament systems of the Xenopus embryo act to ensure normal gastrulation.

To study the role of keratin filaments in Xenopus development, fertilized eggs were injected with anti-keratin monoclonal antibodies. The anti-keratin monoclonal antibodies AE1 and AE3 induce abnormal gastrulation; in the most severely affected embryos gastrulation fails completely. In contrast, embryos injected with the anti-keratin antibody 1h5 develop normally. Immunocytochemical data indicate that injected 1h5 binds to the dense superficial keratin filament system of the embryo but not to the deeper keratin filament networks of ectodermal and subectodermal cells. Injected AE1 and AE3 do not bind to the superficial keratin system but appear to interact preferentially with the deep keratin filament systems of the embryo. We conclude that the superficial keratin filament system is not involved in the process of gastrulation per se but may protect the embryo from mechanical damage. On the other hand, our results suggest that the integrity of the deeper keratin filament systems is required for the mechanical integration of the morphogenetic movements that underlie gastrulation in Xenopus.

Inhibition of poliovirus RNA synthesis by brefeldin A.

Brefeldin A (BFA), a fungal metabolite that blocks transport of newly synthesized proteins from the endoplasmic reticulum, was found to inhibit poliovirus replication 10(5)- to 10(6)-fold. BFA does not inhibit entry of poliovirus into the cell or translation of viral RNA. Poliovirus RNA synthesis, however, is completely inhibited by BFA. A specific class of membranous vesicles, with which the poliovirus replication complex is physically associated, is known to proliferate in poliovirus-infected cells. BFA may inhibit poliovirus replication by preventing the formation of these vesicles.

Cytokeratin phosphorylation, cytokeratin filament severing and the solubilization of the maternal mRNA Vg1.

During meiotic maturation, the cortical cytokeratin filament system of the Xenopus oocyte disappears (Klymkowsky, M. W., and L. A. Maynell. 1989. Dev. Biol. 134:479). Here we demonstrate that this disappearance results from the severing of cytokeratin filaments into a heterogenous population of oligomers, with S- values ranging from 12S and greater. Cytokeratin filament severing correlates with the hyperphosphorylation of the type II cytokeratin of the oocyte. Both the severing of cytokeratin filaments and cytokeratin hyperphosphorylation are reversed by treatment with cycloheximide. These data suggest that fragmentation of cytokeratin filaments is controlled, at least in part, by the phosphorylation of the type II cytokeratin, and that the cytokeratin kinase activity responsible is biosynthetically labile. Cytokeratin filaments have been suggested to anchor the maternal mRNA Vg1 to the vegetal cortex of the oocyte (Pondel, M., and M. L. King. 1988. Proc. Natl. Acad. Sci. USA. 85:7216). By injecting fractions containing active maturation promoting factor or a purified, mutant cyclin protein, we find that the bulk of the Vg1 mRNA in the oocyte can be solubilized under conditions that block the fragmentation of cytokeratin filaments, and that the fragmentation of cytokeratin filaments itself leads to the solubilization of only a minor fraction of the Vg1 mRNA. Thus, at best, cytokeratin filaments directly anchor only a minor fraction of the Vg1 mRNA in the oocyte. Moreover, factors distinct from maturation promoting factor appear to be required for the complete solubilization of Vg1 mRNA during oocyte maturation.

MPF-induced breakdown of cytokeratin filament organization in the maturing Xenopus oocyte depends upon the translation of maternal mRNAs.

In Xenopus, one of the most dramatic events during oocyte maturation is the breakdown of the oocyte's asymmetrically organized system of cytokeratin-type intermediate filaments. Following oocyte maturation in vitro, we found that (1) the breakdown of cytokeratin filament organization proceeds in an animal to vegetal direction, (2) cytokeratin filament breakdown occurs normally in enucleated oocytes and so is independent of nuclear components, and (3) the injection of maturation-promoting factor (MPF) induces the breakdown of cytokeratin filaments. While the MPF-induced breakdown of the nuclear envelope is independent of new protein synthesis, the MPF-induced breakdown of cytokeratin filament organization requires the translation of maternal mRNAs. These results strongly suggest that the factors regulating cytokeratin reorganization in the oocyte are distinct from those involved in the breakdown of the nuclear envelope.

Polar asymmetry in the organization of the cortical cytokeratin system of Xenopus laevis oocytes and embryos.

We have used whole-mount immunofluorescence microscopy of late-stage Xenopus laevis oocytes and early embryos to examine the organization of their cortical cytokeratin systems. In both mature oocytes and early embryos, there is a distinct animal-vegetal polarity in cytokeratin organization. In mature (stage-VI) oocytes, the cytokeratin filaments of the vegetal region form a unique, almost geodesic network; in the animal region, cytokeratin organization appears much more variable and irregular. In unfertilized, postgerminal vesicle breakdown eggs, the cortical cytokeratin system is disorganized throughout both animal and vegetal hemispheres. After fertilization, cytokeratin organization reappears first in a punctate pattern that is transformed into an array of oriented filaments. These cytokeratin filaments appear first in the vegetal hemisphere and are initially thin. Subsequently, they form bundles that grow thicker through the period of first to second cleavage, at which point large cytokeratin filament bundles form a loose, fishnet-like system that encompasses the vegetal portion of each blastomere. In the animal region, cytokeratin filaments do not appear to form large fibre networks, but rather appear to be organized into a system of fine filaments. The animal-vegetal polarity in cytokeratin organization persists until early blastula (stage 5); in later-stage embryos, both animal and vegetal blastomeres possess qualitatively similar cytokeratin filament systems. The entire process of cytokeratin reorganization in the egg is initiated by prick activation. These observations indicate that the cortical cytoskeleton of Xenopus oocytes and early embryos is both dynamic and asymmetric.