A role for VICKZ proteins in the progression of colorectal carcinomas: regulating lamellipodia formation.

VICKZ proteins are a highly conserved family of RNA binding proteins, implicated in RNA regulatory processes such as intracellular RNA localization, RNA stability, and translational control. During embryogenesis, VICKZ proteins are required for neural crest migration and in adults, the proteins are overexpressed primarily in different cancers. We hypothesized that VICKZ proteins may play a role in cancer cell migration. In patients, VICKZ expression varies with tumour type, with over 60% of colon, lung, and ovarian tumours showing strong expression. In colorectal carcinomas (CRCs), expression is detected at early stages, and the frequency and intensity of staining increase with progression of the disease to lymph node metastases, of which 97% express the protein at high levels. Indeed, in stage II CRC, the level of VICKZ expression in the primary lesion correlates with the degree of lymph node metastasis. In culture, VICKZ proteins rapidly accumulate in processes at the leading edge of PMA-stimulated SW480 CRC cells, where they co-localize with beta-actin mRNA. Two distinct cocktails of shRNAs, each targeting all three VICKZ paralogues, cause a dramatic drop in lamellipodia and ruffle formation in stimulated cells. Thus, VICKZ proteins help to facilitate the dynamic cell surface morphology required for cell motility. We propose that these proteins play an important role in CRC metastasis by shuttling requisite RNAs to the lamellipodia of migrating cells.

Defining cis-acting elements and trans-acting factors in RNA localization.

Research over the last 10 to 15 years has revealed that intracellular RNA localization is a widespread phenomenon found in a large range of different cell types in an equally impressive number of different organisms (Bashirullah et al., 1998; St. Johnston, 1995). Efforts have focused both on the molecular mechanisms involved in localizing RNAs to particular intracellular targets and on the functional importance (to the cell) of placing certain RNAs at particular cellular sites. In many cases, an understanding of the role of RNA localization seems to be predicated on a careful analysis of how a particular RNA achieves its characteristic distribution. A generalized model of RNA localization usually invokes cellular factors recognizing RNA target sequences. This review will focus on several systems in which cis-acting elements and trans-acting factors recognizing these elements are involved in RNA localization: how they have been defined, how they relate to each other, and how they interact and function to help achieve defined intracellular localization. Conservation of both RNA elements and protein factors across species suggests that RNA localization is probably a fundamental cellular process.

Dynein light chain binding to a 3'-untranslated sequence mediates parathyroid hormone mRNA association with microtubules.

The 3'-untranslated region (UTR) of mRNAs binds proteins that determine mRNA stability and localization. The 3'-UTR of parathyroid hormone (PTH) mRNA specifically binds cytoplasmic proteins. We screened an expression library for proteins that bind the PTH mRNA 3'-UTR, and the sequence of 1 clone was identical to that of the dynein light chain LC8, a component of the dynein complexes that translocate cytoplasmic components along microtubules. Recombinant LC8 binds PTH mRNA 3'-UTR, as shown by RNA electrophoretic mobility shift assay. We showed that PTH mRNA colocalizes with microtubules in the parathyroid gland, as well as with a purified microtubule preparation from calf brain, and that this association was mediated by LC8. To our knowledge, this is the first report of a dynein complex protein binding an mRNA. The dynein complex may be the motor that is responsible for transporting mRNAs to specific locations in the cytoplasm and for the consequent is asymmetric distribution of translated proteins in the cell.

Vg1 RBP intracellular distribution and evolutionarily conserved expression at multiple stages during development.

We have analyzed the expression and intracellular distribution, during oogenesis and embryogenesis, of Vg1 RBP, a protein implicated in the intracellular localization of Vg1 mRNA to the vegetal cortex of Xenopus oocytes. Vg1 RBP (protein) colocalizes with Vg1 RNA at all stages of oogenesis. Vg1 RBP RNA, however, localizes to the animal pole during late oogenesis, and remains in the animal blastomeres and ectodermal precursors until its zygotic transcription is activated, around stage 12. Vg1 RBP mRNA then becomes expressed throughout the neural epithelium. Vg1 RBP mRNA expression is also detected in what appears to be neural crest cells undergoing delamination and lateral migration. By tailbud stages, Vg1 RBP expression is present in the branchial arches, otic vesicle, pronephros, and along the neural tube. To examine the expression pattern in different species, we cloned the zebrafish homolog of Vg1 RBP by using a highly homologous EST clone to screen an embryonic cDNA library. In situ hybridization reveals that Vg1 RBP RNA localizes early in oogenesis to the animal pole. Although Vg1 RBP RNA is detected in all blastomeres of the early embryo, the expression pattern in the one day old zebrafish embryo is almost identical to that of the equivalent stage Xenopus embryo. These results indicate that the zygotic expression pattern is similar in frogs and fish, and that there is a conserved zygotic expression of Vg1 RBP distinct from its expression in the oocyte.

The Xcad-2 gene can provide a ventral signal independent of BMP-4.

Patterning of the marginal zone in the Xenopus embryo has been attributed to interactions between dorsal genes expressed in the organizer and ventral-specific genes. In this antagonistic interplay of activities, BMP-4, a gene that is not expressed in the organizer, provides a strong ventralizing signal. The Xenopus caudal type homeobox gene, Xcad-2, which is expressed around the blastopore with a gap over the dorsal lip, was analyzed as part of the ventral signal. Xcad-2 was shown to efficiently repress during early gastrula stages the dorsal genes gsc, Xnot-2, Otx-2, XFKH1 and Xlim-1, while it positively regulates the ventral genes, Xvent-1 and Xvent-2, with Xpo exhibiting a strong positive response to Xcad-2 overexpression. Xcad-2 was also capable of inducing BMP-4 expression in the organizer region. Support for a ventralizing role for Xcad-2 was obtained from co-injection experiments with the dominant negative BMP receptor which was used to block BMP-4 signaling. Under lack-of-BMP-signaling conditions Xcad-2 could still regulate dorsal and ventral gene expression and restore normal development, suggesting that it can act downstream of BMP-4 signaling or independently of it. Xcad-2 could also inhibit secondary axis formation and dorsalization induced by the dominant negative BMP receptor. Xcad-2 was also shown to efficiently reverse the dorsalizing effects of LiCl. These results place Xcad-2 as part of the ventralizing gene program which acts during early gastrula stages and can execute its ventralizing function in the absence of BMP signaling.

Nested expression and sequential downregulation of the Xenopus caudal genes along the anterior-posterior axis.

Expression of the Xenopus Xcad-1 and Xcad-2 genes initiates during early gastrulation exhibiting a dorsoventral asymmetry in their domains of transcription. At mid-gastrulation the ventral preference becomes stronger and the caudal genes take up a posterior localization in their expression, which they will maintain until their downregulation along the dorsal midline. Comparison of the three Xenopus caudal genes revealed a temporal and spatial nested set of expression patterns. The transcription of the caudal genes is sequentially downregulated with the one expressed most caudally (Xcad-2) being shut down first, this sequence is most evident along the dorsal midline. This pattern of expression suggests a role for the caudal genes as posterior determinants along the anteroposterior axis. In chicken, mouse, man and Xenopus three members of the caudal family have been identified in the genome. Even though in Xenopus the Xcad-3 gene has been previously described, in order to obtain a better insight on the role of the caudal genes a comparative study of all three frog genes was performed.

RNA-binding protein conserved in both microtubule- and microfilament-based RNA localization.

Vg1 mRNA translocation to the vegetal cortex of Xenopus oocytes requires intact microtubules, and a 3' UTR cis-acting element (termed VLE), which also mediates sequence-specific binding of several proteins. One protein, the 69-kD Vg1 RBP, associates Vg1 RNA to microtubules in vitro. Here we show that Vg1 RBP-binding sites correlate with vegetal localization. Purification and cloning of Vg1 RBP revealed five RNA-binding motifs: four KH and one RRM domains. Surprisingly, Vg1 RBP is highly homologous to the zipcode binding protein implicated in the microfilament-mediated localization of beta actin mRNA in fibroblasts. These data support Vg1 RBP's direct role in vegetal localization and suggest the existence of a general, evolutionarily conserved mechanism for mRNA targeting.

Patterning of the embryo along the anterior-posterior axis: the role of the caudal genes.

Patterning along the anterior-posterior axis takes place during gastrulation and early neurulation. Homeobox genes like Otx-2 and members of the Hox family have been implicated in this process. The caudal genes in Drosophila and C. elegans have been shown to determine posterior fates. In vertebrates, the caudal genes begin their expression during gastrulation and they take up a posterior position. By injecting sense and antisense RNA of the Xenopus caudal gene Xcad-2, we have studied a number of regulatory interactions among homeobox genes along the anterior-posterior axis. Initially, the Xcad-2 and Otx-2 genes are mutually repressed and, by late gastrulation, they mark the posterior- or anterior-most domains of the embryo, respectively. During late gastrulation and neurulation, Xcad-2 plays an additional regulatory function in relation to the Hox genes. Hox genes normally expressed anteriorly are repressed by Xcad-2 overexpression while those normally expressed posteriorly exhibit more anterior expression. The results show that the caudal genes are part of a posterior determining network which during early gastrulation functions in the subdivision of the embryo into anterior head and trunk domains. Later in gastrulation and neurulation these genes play a role in the patterning of the trunk region.

Exogenous tau RNA is localized in oocytes: possible evidence for evolutionary conservation of localization mechanisms.

The multistep pathway leading to intracellular RNA localization is known to involve cis-acting signals in targeted mRNAs, which are presumably recognized by specific RNA-binding proteins and interact with a functional cytoskeleton. Tau RNA is localized to the proximal hillock of rat axons, and this movement requires intact microtubules. Because Xenopus oocytes demonstrate a clear polarity involving microtubule-mediated RNA localization, we have studied the distribution of tau RNA injected into oocytes. We find that a fragment from the 3'-untranslated region of tau RNA is localized to the vegetal cortex of stage III/IV oocytes in a distribution indistinguishable from Vg1 RNA, a vegetally localized oocyte mRNA. A fragment from the tau RNA coding region, however, is homogeneously distributed in oocytes. Tau RNA contains a functional binding site for Vg1 RBP, a Xenopus microtubule-associated protein that binds vegetally localized oocyte RNAs with high affinity, and this binding correlates with vegetal localization ability. The present studies demonstrate, for the first time, localization of heterologous RNA in oocytes. Given the role of Vg1 RBP as a mediator of specific RNA-microtubule interactions, these results are strong evidence that Vg1 RBP is involved in the vegetal localization of RNAs in oocytes and raise the intriguing possibility of the existence of proteins with similar function in neurons.

Vg1 RNA binding protein mediates the association of Vg1 RNA with microtubules in Xenopus oocytes.

Localized RNAs are found in a variety of somatic and developing cell types. In many cases, microtubules have been implicated as playing a role in facilitating transport of these RNAs. Here we report that Vg1 RNA, which is localized to the vegetal cortex of Xenopus laevis oocytes, is associated with microtubules in vivo. Because of the ubiquitous nature of tubulin, the association of specific RNAs with microtubules is likely to involve factors that recognize both RNA and microtubules. Vg1 RNA binding protein (Vg1 RBP), previously shown to bind with high affinity to the vegetal localization site in Vg1 RNA, appears to function in this capacity. Vg1 RBP is associated with microtubules: it is enriched in microtubule extracts of oocytes and is also co-precipitated by heterologous, polymerized tubulin. Furthermore, Vg1 RBP binding activity is required for the specific association of Vg1 RNA to microtubules in vitro. These data suggest a general model for how specific RNAs can be localized to particular sites via common cytoskeletal elements.

A Xenopus nonmuscle myosin heavy chain isoform is phosphorylated by cyclin-p34cdc2 kinase during meiosis.

There are two vertebrate nonmuscle myosin heavy chain (MHC) genes that encode two separate isoforms of the heavy chain, MHC-A and MHC-B. Recent work has identified additional, alternatively spliced isoforms of MHC-B cDNA with inserted sequences of 30 nucleotides (chicken and human) or 48 nucleotides (Xenopus) at a site corresponding to the ATP binding region in the MHC protein (Takahashi, M., Kawamoto, S., and Adelstein, R.S. (1992) J. Biol. Chem. 267, 17864-17871) and Bhatia-Dey, N., Adelstein, R.S., and Dawid, I.B. (1993) Proc. Natl. Acad. Sci. U.S.A. 90, 2856-2859). The deduced amino acid sequence of these inserts contains a consensus sequence for phosphorylation by cyclin-p34cdc2 (cdc2) kinase. In cultured Xenopus XTC cells, we have identified two inserted MHC-B isoforms and a non-inserted MHC-A isoform by immunoblotting of cell extracts. When myosin was immunoprecipitated from XTC cells and phosphorylated in vitro with cdc2 kinase, the kinase catalyzed the phosphorylation of both inserted MHC-B isoforms but not MHC-A. Isoelectric focusing of tryptic peptides generated from MHC-B phosphorylated with cdc2 kinase revealed one major phosphopeptide that was purified by reverse-phase high performance liquid chromatography and sequenced. The phosphorylated residue was Ser-214, the cdc2 kinase consensus site within the insert near the ATP binding region. The same site was phosphorylated in intact XTC cells during log phase of growth and in cell-free lysates of Xenopus eggs stabilized in second meiotic metaphase but not interphase. Moreover, Ser-214 phosphorylation was detected during maturation of Xenopus oocytes when the cdc2 kinase-containing maturation-promoting factor was activated, but not in G2 interphase-arrested oocytes. These results demonstrate that MHC-B phosphorylation is tightly regulated by cdc2 kinase during meiotic cell cycles. Furthermore, MHC-A and MHC-B isoforms are differentially phosphorylated at these stages, suggesting that they may serve different functions in these cells.

A 69-kDa RNA-binding protein from Xenopus oocytes recognizes a common motif in two vegetally localized maternal mRNAs.

Vg1 mRNA, a maternal message encoding a member of the transforming growth factor beta superfamily, undergoes localization to the vegetal cortex of Xenopus laevis oocytes during a narrow period of oogenesis. A 340-nucleotide sequence has been identified in Vg1 RNA that directs its vegetal localization [Mowry, K. L. & Melton, D. A. (1992) Science 255, 991-994]. To understand how cis- and trans-acting factors are involved in Vg1 mRNA localization, we have looked for specific interactions in vitro between oocyte proteins and Vg1 mRNA. S100 extracts of late-stage oocytes contain a protein-binding activity that protects specific regions of labeled Vg1 mRNA from degradation by RNase T1. The use of different regions of Vg1 RNA in competition reactions reveals two binding sites, both in the first half of the 3' untranslated region of Vg1 message. UV crosslinking predominantly labels a 69-kDa protein; saturation analysis and competitor studies indicate that this protein binds with a high affinity to the down-stream site, which corresponds to the 340-nucleotide vegetal localization sequence. Binding to this region is inhibited by another vegetally localized message, transforming growth factor beta 5 but is not inhibited by an animally localized RNA, An2. These data indicate that vegetally localized mRNAs share a binding motif that helps them achieve their intracellular distribution through specific RNA-protein interactions.

A two-step model for the localization of maternal mRNA in Xenopus oocytes: involvement of microtubules and microfilaments in the translocation and anchoring of Vg1 mRNA.

In an effort to understand how polarity is established in Xenopus oocytes, we have analyzed the process of localization of the maternal mRNA, Vg1. In fully grown oocytes, Vg1 mRNA is tightly localized at the vegetal cortex. Biochemical fractionation shows that the mRNA is preferentially associated with a detergent-insoluble subcellular fraction. The use of cytoskeletal inhibitors suggests that (1) microtubules are involved in the translocation of the message to the vegetal hemisphere and (2) microfilaments are important for the anchoring of the message at the cortex. Furthermore, immunohistochemistry reveals that a cytoplasmic microtubule array exists during translocation. These results suggest a role for the cytoskeleton in localizing information in the oocyte.

The process of localizing a maternal messenger RNA in Xenopus oocytes.

The maternal mRNA Vg1 is localized to the vegetal pole during oogenesis in Xenopus. We have cultured oocytes in vitro to begin to understand how this localization occurs. Endogenous Vg1 mRNA undergoes localization when oocytes are cultured in vitro, and synthetic Vg1 mRNA injected into such oocytes is localized in the same fashion. Vg1 mRNA is associated with a detergent-insoluble fraction from homogenized oocytes, suggesting a possible cytoskeletal association. The use of cytoskeletal inhibitors reveals a two-step process for localizing Vg1 mRNA. Microtubule inhibitors such as nocodazole and colchicine inhibit the localization of Vg1 mRNA in late stage III/early stage IV oocytes, but have no effect on Vg1 mRNA once it is localized. The microfilament inhibitor cytochalasin B, however, has little effect on the translocation of Vg1 mRNA in middle-stage oocytes but causes a release of the message in late-stage oocytes. We propose a model for the localization of Vg1 mRNA in which translocation of the message to the vegetal cortex is achieved via cytoplasmic microtubules and the anchoring of the message at the cortex involves cortical microfilaments.

The material mRNA Vg1 is correctly localized following injection into Xenopus oocytes.

The animal and vegetal ends of Xenopus oocytes have distinctly different developmental fates. At the molecular level, several maternal mRNAs have been isolated that are localized to either the animal or vegetal hemisphere. One of these mRNAs, Vg1, is distributed homogeneously throughout the cytoplasm of early-stage oocytes and gets localized during oogenesis to a tight shell at the vegetal cortex of middle and late-stage oocytes. We have used an in vitro culture system to demonstrate that exogeneous Vg1 mRNA injected into middle-stage, but not late-stage, oocytes gets localized in a similar fashion to the endogenous message. Furthermore, translation of Vg1 mRNA is not required for the localization of the message itself. These results show that the information necessary to interpret the animal-vegetal polarity in oocytes is present in the naked mRNA transcript.