Interaction between krit1 and icap1alpha infers perturbation of integrin beta1-mediated angiogenesis in the pathogenesis of cerebral cavernous malformation.

Cerebral cavernous malformation (CCM) is a common autosomal dominant disorder characterized by venous sinusoids that predispose to intracranial hemorrhage. CCM is genetically heterogeneous, with loci at 7q, 7p and 3q. Mutations in KRIT1 account for all cases linked to 7q (CCM1), but the pathogenesis of CCM is not understood. Krev Interaction Trapped 1 (krit1) was originally identified through its interaction with the Ras-family GTPase krev1/rap1a in a two-hybrid screen, inferring a role in GTPase signaling cascades. We demonstrated additional 5'-coding exons for krit1, extending the N-terminus by 207 amino acids compared to the previously reported protein. Remarkably, by two-hybrid analysis and co-immunoprecipitation, full-length krit1 fails to interact with krev1/rap1a but shows strong interaction with integrin cytoplasmic domain-associated protein-1 (icap1). Icap1 binds to a NPXY motif in the cytoplasmic domain of beta1 integrin and participates in beta1-mediated cell adhesion and migration. The novel N-terminus of krit1 contains a NPXY motif that it is required for icap1 interaction. Like beta1 integrin, krit1 interacts with the 200 amino acid isoform of icap1 (icap1alpha), but not a 150 amino acid form that results from alternative splicing (icap1beta). In a competition assay, induced expression of krit1 diminishes the interaction between icap1alpha and beta1 integrin. Taken together, these data suggest that beta1 integrin and krit1 compete for the same site on icap1alpha, perhaps constituting a regulatory mechanism. Loss-of-function KRIT1 mutations, as observed in CCM1, would shift the balance with predicted consequences for endothelial cell performance during integrin beta1-dependent angiogenesis.

Characterization of mouse epithelial protein lost in neoplasm (EPLIN) and comparison of mammalian and zebrafish EPLIN.

EPLIN is a cytoskeleton-associated protein that was initially identified as the product of a gene that is transcriptionally down-regulated in cancer cells. In human, there are two known isoforms, EPLIN-alpha and -beta, generated by alternative promoter usage from a single gene. With the exception of a single LIM (lin-11, isl-1, and mec-3) domain, the sequence of EPLIN is unique and does not provide any clues to its function. To identify conserved regions of EPLIN that may be important for its function, we have characterized mouse (m) and zebrafish (zf) EPLIN. As in human, two isoforms, the 593 aa mEPLIN-alpha (77% identity; 83% similarity) and 753 aa mEPLIN-beta (75% identity; 83% similarity), were present in mouse. mEPLIN-alpha is highly expressed in embryonic tissue and adult lung and spleen, whereas mEPLIN-beta is preferentially expressed in kidney, testis, lung and liver. The analysis of mEPLIN gene revealed that the overall organization of the exons in mouse and human are conserved. In zebrafish, there was only one form, the 629 aa zfEPLIN, corresponding to the mammalian EPLIN-beta. Like its mammalian counterparts, ectopically expressed zfEPLIN is co-localized to the actin cytoskeleton. While the overall homology between mammalian and zebrafish EPLIN was not striking (37% identity; 50% similarity), there were seven highly conserved regions, which should be useful in structure-function studies of this novel protein.

Differential gene expression between normal and tumor-derived ovarian epithelial cells.

The majority of ovarian tumors arise from the transformation of the ovarian surface epithelial cells, a single layer of cells surrounding the ovary. To identify genes that may contribute to the malignant phenotype of ovarian cancers, cDNA representational difference analysis was used to compare expressed genes in primary cultures of normal human ovarian surface epithelium (HOSE) and ovarian tumor-derived epithelial cells from the Cedars-Sinai Ovarian Cancer (CSOC) repository. A total of 255 differentially expressed genes were identified, of which 160 and 95 were specifically expressed in HOSE and CSOC cells, respectively. Using cDNA array hybridization, the expression profiles of the genes identified by cDNA-representational difference analysis were examined in an additional 5 HOSE and 10 CSOC lines. The comparison of average signal of each gene revealed 44 HOSE-specific and 16 CSOC-specific genes that exhibited at least a 2.5-fold difference in expression. A large number of genes identified in this study encode membrane-associated or secreted proteins and, hence, may be useful as targets in the development of serum-based diagnostic markers for ovarian cancer. Very few genes associated with protein synthesis or metabolism were identified in this study, reflecting the lack of observable differences in phenotypic or growth characteristics between HOSE and CSOC cells. Northern blot analysis on a subset of these genes demonstrated comparable levels of gene expression as observed in the cDNA array hybridization.

Overexpression of wild-type RhoA produces growth arrest by disrupting actin cytoskeleton and microtubules.

We have investigated the role of Rho GTPase in cell growth by generating stable cells that express the wild-type RhoA (RhoA(wt)) under the control of an inducible promoter. Induction of RhoA(wt) had a biphasic effect on the actin cytoskeleton. At low levels of expression, RhoA(wt) stimulated the assembly of actin stress fibers without affecting cell growth. At high levels, there was a paradoxical disruption of the actin cytoskeleton accompanied by a growth arrest. Cell cycle analysis revealed a dual block at the G(1)/S and G(2)/M checkpoints. The G(1)/S arrest correlated with the accumulation of p21(Cip1), resulting in the inhibition of cdk2 activity, whereas the G(2)/M block correlated with the loss of microtubules. The cyclin B level and the cdc2 kinase activity, however, were increased, suggesting that the progression through mitosis rather than entry into the G(2)/M is defective when RhoA(wt) is overexpressed. Similar cell cycle defects and the loss of microtubules were observed after a cytochalasin D treatment, indicating that the ability of RhoA to regulate the integrity of actin cytoskeleton may be critical for the cell cycle transition through both the G(1)/S and M phase checkpoints.

Characterization of the human EPLIN (Epithelial Protein Lost in Neoplasm) gene reveals distinct promoters for the two EPLIN isoforms.

EPLIN is a novel LIM domain protein that co-localizes to the actin stress fibers and focal adhesion plaques. We previously have demonstrated that two isoforms, the 600aa EPLIN-alpha and the 759aa EPLIN-beta, are generated from a single gene. In the majority of human breast and prostate cancer cell lines, the expression of EPLIN-alpha is significantly reduced, while the expression of EPLIN-beta is either up-regulated or unchanged. To understand the basis of this differential regulation, we have determined the organization of the human EPLIN gene. The human EPLIN100kb and consists of 11 exons. The EPLIN-beta mRNA requires all 11 exons, while the EPLIN-alpha mRNA requires Exons 4-11. The transcriptional start sites of EPLIN-alpha were mapped within the third intron by 5' RACE and S1 nuclease protection. Similarly, the 5' ends of EPLIN-beta were mapped upstream of Exon 1. The DNA sequences flanking the EPLIN-alpha or EPLIN-beta transcriptional start sites were capable of stimulating the expression of promoter reporter constructs. Interestingly, the endogenous transcription of EPLIN-alpha, but not EPLIN-beta, could be stimulated by serum, indicating that the expression of two EPLIN isoforms can be independently regulated. A consensus serum response element was present within 100bp upstream of the transcriptional start sites of EPLIN-alpha. The activity of 0.7kb EPLIN-alpha promoter reporter construct could be enhanced by activated RhoA, indicating that this serum response element is functional.

EPLIN, epithelial protein lost in neoplasm.

We have identified a novel cytoskeletal protein, EPLIN (Epithelial Protein Lost In Neoplasm), that is preferentially expressed in human epithelial cells. Two EPLIN isoforms, a 600 amino acid EPLIN-alpha and a 759 amino acid EPLIN-beta, are detected in primary epithelial cells of oral mucosa, prostate and mammary glands. The expression of EPLIN-alpha is either down-regulated or lost in the majority of oral cancer cell lines (8/8), prostate cancer cell lines (4/4) and xenograft tumors (3/3), and breast cancer cell lines (5/6). The amino acid sequence of EPLIN is characterized by the presence of a single centrally located LIM domain. Both EPLIN isoforms localize to filamentous actin and suppress cell proliferation when overexpressed. These findings indicate that the loss of EPLIN seen in cancer cells may play a role in cancer progression.

Inhibition of Stat1-mediated gene activation by PIAS1.

STAT (signal transducer and activator of transcription) proteins are latent cytoplasmic transcription factors that become activated by tyrosine phosphorylation in response to cytokine stimulation. Tyrosine phosphorylated STATs dimerize and translocate into the nucleus to activate specific genes. Different members of the STAT protein family have distinct functions in cytokine signaling. Biochemical and genetic analysis has demonstrated that Stat1 is essential for gene activation in response to interferon stimulation. Although progress has been made toward understanding STAT activation, little is known about how STAT signals are down-regulated. We report here the isolation of a family of PIAS (protein inhibitor of activated STAT) proteins. PIAS1, but not other PIAS proteins, blocked the DNA binding activity of Stat1 and inhibited Stat1-mediated gene activation in response to interferon. Coimmunoprecipitation analysis showed that PIAS1 was associated with Stat1 but not Stat2 or Stat3 after ligand stimulation. The in vivo PIAS1-Stat1 interaction requires phosphorylation of Stat1 on Tyr-701. These results identify PIAS1 as a specific inhibitor of Stat1-mediated gene activation and suggest that there may exist a specific PIAS inhibitor in every STAT signaling pathway.

Characterization of transformation related genes in oral cancer cells.

A cDNA representational difference analysis (cDNA-RDA) and an arrayed filter technique were used to characterize transformation-related genes in oral cancer. From an initial comparison of normal oral epithelial cells and a human papilloma virus (HPV)-immortalized oral epithelial cell line, we obtained 384 differentially expressed gene fragments and arrayed them on a filter. Two hundred and twelve redundant clones were identified by three rounds of back hybridization. Sequence analysis of the remaining clones revealed 99 unique clones corresponding to 69 genes. The expression of these transformation related gene fragments in three nontumorigenic HPV-immortalized oral epithelial cell lines and three oral cancer cell lines were simultaneously monitored using a cDNA array hybridization. Although there was a considerable cell line-to-cell line variability in the expression of these clones, a reliable prediction of their expression could be made from the cDNA array hybridization. Our study demonstrates the utility of combining cDNA-RDA and arrayed filters in high-throughput gene expression difference analysis. The differentially expressed genes identified in this study should be informative in studying oral epithelial cell carcinogenesis.

Rack1, a receptor for activated protein kinase C, interacts with integrin beta subunit.

The integrin beta subunit cytoplasmic domains are important for activation-dependent cell adhesion and adhesion-dependent signaling events. We report an interaction between integrin beta subunit cytoplasmic domain and Rack1, a Trp-Asp (WD) repeat protein that has been shown to bind activated protein kinase C. The Rack1-binding site on integrin beta 2 subunit resides within a conserved, membrane-proximal region. In the yeast two-hybrid assay, WD repeats five to seven of Rack1 (Rack1-WD5/7) interact with integrin beta 1, beta 2, and beta 5 cytoplasmic domain. In eukaryotic cells, Rack1 co-immunoprecipitates with at least two different beta integrins, beta 1 integrins in 293T cells and beta 2 integrins in JY lymphoblastoid cells. Whereas Rack1-WD5/7 binds integrins constitutively, the association of full-length Rack1 to integrins in vivo requires a treatment with phorbol esters, which promotes cell spreading and adhesion. These findings suggest that Rack1 may link protein kinase C directly to integrins and participate in the regulation of integrin functions.

ICAP-1, a novel beta1 integrin cytoplasmic domain-associated protein, binds to a conserved and functionally important NPXY sequence motif of beta1 integrin.

The cytoplasmic domains of integrins are essential for cell adhesion. We report identification of a novel protein, ICAP-1 (integrin cytoplasmic domain- associated protein-1), which binds to the 1 integrin cytoplasmic domain. The interaction between ICAP-1 and beta1 integrins is highly specific, as demonstrated by the lack of interaction between ICAP-1 and the cytoplasmic domains of other beta integrins, and requires a conserved and functionally important NPXY sequence motif found in the COOH-terminal region of the beta1 integrin cytoplasmic domain. Mutational studies reveal that Asn and Tyr of the NPXY motif and a Val residue located NH2-terminal to this motif are critical for the ICAP-1 binding. Two isoforms of ICAP-1, a 200-amino acid protein (ICAP-1alpha) and a shorter 150-amino acid protein (ICAP-1beta), derived from alternatively spliced mRNA, are expressed in most cells. ICAP-1alpha is a phosphoprotein and the extent of its phosphorylation is regulated by the cell-matrix interaction. First, an enhancement of ICAP-1alpha phosphorylation is observed when cells were plated on fibronectin-coated but not on nonspecific poly-L-lysine-coated surface. Second, the expression of a constitutively activated RhoA protein that disrupts the cell-matrix interaction results in dephosphorylation of ICAP-1alpha. The regulation of ICAP-1alpha phosphorylation by the cell-matrix interaction suggests an important role of ICAP-1 during integrin-dependent cell adhesion.

Exon amplification: a strategy to isolate mammalian genes based on RNA splicing.

We have developed a method, exon amplification, for fast and efficient isolation of coding sequences from complex mammalian genomic DNA. This method is based on the selection of RNA sequences, exons, which are flanked by functional 5' and 3' splice sites. Fragments of cloned genomic DNA are inserted into an intron, which is flanked by 5' and 3' splice sites of the human immunodeficiency virus 1 tat gene contained within the plasmid pSPL1. COS-7 cells are transfected with these constructs, and the resulting RNA transcripts are processed in vivo. Splice sites of exons contained within the inserted genomic fragment are paired with splice sites of the flanking tat intron. The resulting mature RNA contains the previously unidentified exons, which can then be amplified via RNA-based PCR and cloned. Using this method, we have isolated exon sequences from cloned genomic fragments of the murine Na,K-ATPase alpha 1-subunit gene. We have also screened randomly selected genomic clones known to be derived from a segment of human chromosome 19 and have isolated exon sequences of the DNA repair gene ERCC1. The sensitivity and ease of the exon amplification method permit screening of 20-40 kilobase pairs of genomic DNA in a single transfection. This approach will be extremely useful for rapid identification of mammalian exons and the genes from which they are derived as well as for the generation of chromosomal transcription maps.

Structural analysis of the interaction between the human immunodeficiency virus Rev protein and the Rev response element.

The specific interaction between a defined structural element of the human immunodeficiency virus mRNA (RRE, the Rev response element) and the virus-encoded protein Rev has been implicated in the regulation of the export of unspliced or singly spliced mRNA from the nucleus to the cytoplasm. Rev protein was expressed and purified from insect cells using the baculovirus expression system. Chemical and RNase probes were used to analyze the structure of the RRE and the regions involved in Rev binding. Increased reactivity to single-strand-specific probes of nucleotides in two helical domains indicates that Rev binding induces conformational changes in the RRE. Binding of Rev to the RRE primarily protects helical segments and adjacent nucleotides in domain II. A Rev unit binding site is proposed that consists of a six-base-pair helical segment and three adjacent nucleotides. The data also suggest that multiple Rev proteins bind to repeated structural elements of the RRE.

The gene for the RNA component of the mitochondrial RNA-processing endoribonuclease is located on human chromosome 9p and on mouse chromosome 4.

Mitochondrial RNA-processing endoribonuclease (RNAase MRP) has the capacity to cleave mitochondrial RNA complementary to the light strand of the displacement loop at a unique site. The enzyme is a ribonucleoprotein whose RNA component is a nuclear gene product. The 5' flanking region of the primary transcript has control elements characteristic of RNA polymerase II transcription, and the coding region has features of RNA polymerase III transcription signals. The RNA associated with RNAase MRP is the first known RNA encoded by a single-copy gene in the nucleus and believed to be imported into mitochondria. The gene (RMRP) for this RNA component of RNAase MRP was assigned to human chromosome 9 and mouse chromosome 4 by Southern blot analyses of 11 human X rodent hybrids and 11 mouse X rodent hybrids with probe pHM1.0 and probe pSP270, respectively. In situ hybridization of probe pHSTU300 to normal human chromosomes revealed 29 of 100 cells with label on 9p and 9.6% of 302 silver grains located at 9p21--p12.

Regulation by HIV Rev depends upon recognition of splice sites.

The ability of the Rev protein of HIV to regulate the cytoplasmic level of unspliced RNA from a beta-globin gene containing the Rev response element was dependent on the integrity of the 5' and 3' splice sites. A beta-globin pre-mRNA containing the Rev response element is not under regulation by Rev but is made Rev responsive by a mutation at either the 5' or 3' splice site. These mutant RNAs accumulated in the nucleus as unspliced precursors owing to recognition by splicing components. Only in the presence of Rev did these unspliced RNAs appear in the cytoplasm. Thus, regulation by Rev probably involves the dissociation of splicing components and pre-mRNA.

Mouse RNAase MRP RNA is encoded by a nuclear gene and contains a decamer sequence complementary to a conserved region of mitochondrial RNA substrate.

Using complementary oligonucleotide probes, we have isolated the nuclear gene for the RNA moiety of RNAase MRP; it is present as a single copy and encodes an uncapped primary transcript of 275 nucleotides. Direct sequence analysis revealed that the 136 nucleotide RNA that copurifies with RNAase MRP represents the 3' half of the 275 nucleotide primary transcript. The 5'-flanking region of the gene has putative transcriptional control elements homologous to the promoters of RNA polymerase II-transcribed U-series snRNA genes; however, the coding region possesses a box A sequence and terminates at four T residues, both features characteristic of polymerase III-transcribed genes. A decamer sequence, 5'-CGA-CCCCUCC-3', complementary to a conserved sequence adjacent to the enzymatic cleavage site on the mitochondrial RNA substrate, is present in the RNAase MRP RNA. Isolation of a nuclear gene for the RNA component of a mitochondrial enzyme implies that nucleic acids can be transported across mitochondrial membranes.

A mammalian mitochondrial RNA processing activity contains nucleus-encoded RNA.

Ribonuclease mitochondrial RNA processing, a site-specific endoribonuclease involved in primer RNA metabolism in mammalian mitochondria, requires an RNA component for its activity. On the basis of copurification and selective inactivation with complementary oligonucleotides, a 135-nucleotide RNA species, not encoded in the mitochondrial genome, is identified as the RNA moiety of the endoribonuclease. This finding implies transport of a nucleus-encoded RNA, essential for organelle DNA replication, to the mitochondrial matrix.

A novel endoribonuclease cleaves at a priming site of mouse mitochondrial DNA replication.

Priming at the mouse mitochondrial origin of heavy-strand DNA replication is effected by transcripts from the light-strand promoter. The transition from RNA synthesis to DNA synthesis occurs at specific locations between 75 and 165 nucleotides downstream from the transcriptional initiation site. We have identified and partially purified an endonucleolytic activity that cleaves RNA accurately near one of these transition sites; this finding implies a role of specific RNA processing in DNA replication. Cleavage products possess 5'-phosphoryl and 3'-hydroxyl termini. Heterologous assays using mouse or human mitochondrial endoribonuclease with substrates containing the sequences of the human or mouse mitochondrial origins of heavy-strand DNA replication suggest that selection of the cleavage site is guided by sequences adjacent to the actual position of cleavage.

Precise assignment of the light-strand promoter of mouse mitochondrial DNA: a functional promoter consists of multiple upstream domains.

Using deletion mutagenesis we localized the promoter for the light strand of mouse mitochondrial DNA to a 97-base-pair region, from -88 to +9 nucleotides of the transcriptional initiation site. Within this region the light-strand promoter could be dissected into at least three different functional domains. The specificity region, a maximum of 19 base pairs between -10 and +9 of the transcriptional initiation site, was essential and sufficient for accurate transcriptional initiation. A second region, extending to -29 nucleotides from the initiation site, facilitated the formation of a preinitiation complex between the template DNA and factor(s) present in the mitochondrial RNA polymerase fraction and was required for efficient transcription. A third, ill-defined upstream region, which extended up to -88 nucleotides from the initiation site, appeared to influence template transcriptional efficiencies in competition assays. Without the specificity domain, the upstream regions were incapable of supporting any transcription. The presence of multiple upstream domains was confirmed by disrupting nucleotide sequences in the upstream region by using linker insertion and linker replacement techniques.