Polyadenylation-specific complexes undergo a transition early in the polymerization of a poly(A) tail.

We have analyzed several properties of the complex that forms between RNAs that end at the poly(A) site of simian virus 40 late mRNA and factors present in a HeLa cell nuclear extract. Formation of this polyadenylation-specific complex requires the sequence AAUAAA and a proximal 3' end. We have observed three changes in the polyadenylation complex early in the addition of the poly(A) tail. First, the complex becomes heparin sensitive after the addition of approximately 10 adenosines. Second, a 68-kilodalton protein present in the complex, which can be cross-linked by UV light to the RNA before polyadenylation has begun, no longer can be cross-linked after approximately 10 adenosines have been added. Third, after 30 adenosines have been added, the AAUAAA sequence becomes accessible to a complementary oligonucleotide and RNase H. This accessibility gradually increases with longer poly(A) tail lengths until, with the addition of 60 A's, all substrates are accessible at AAUAAA. Sheets and Wickens (Genes Dev. 3:1401-1412, 1989) have recently demonstrated two phases in the addition of a poly(A) tail: the first requires AAUAAA, whereas the second is independent of AAUAAA but requires a short oligo(A) primer. The data reported here further support a biphasic model for poly(A) addition and may indicate disengagement of specific factors from AAUAAA after the initiation phase.

The enzyme that adds poly(A) to mRNAs is a classical poly(A) polymerase.

Virtually all mRNAs in eucaryotes end in a poly(A) tail. This tail is added posttranscriptionally. In this report, we demonstrate that the enzyme that catalyzes this modification is identical with an activity first identified 30 years ago, the function of which was previously unknown. This enzyme, poly(A) polymerase, lacks any intrinsic specificity for its mRNA substrate but gains specificity by interacting with distinct molecules: a poly(A) polymerase from calf thymus, when combined with specificity factor(s) from cultured human cells, specifically and efficiently polyadenylates only appropriate mRNA substrates. Our results thus demonstrate that this polymerase is responsible for the addition of poly(A) to mRNAs and that its interaction with specificity factors is conserved.

ETO, a target of t(8;21) in acute leukemia, interacts with the N-CoR and mSin3 corepressors.

t(8;21) is one of the most frequent translocations associated with acute myeloid leukemia. It produces a chimeric protein, acute myeloid leukemia-1 (AML-1)-eight-twenty-one (ETO), that contains the amino-terminal DNA binding domain of the AML-1 transcriptional regulator fused to nearly all of ETO. Here we demonstrate that ETO interacts with the nuclear receptor corepressor N-CoR, the mSin3 corepressors, and histone deacetylases. Endogenous ETO also cosediments on sucrose gradients with mSin3A, N-CoR, and histone deacetylases, suggesting that it is a component of one or more corepressor complexes. Deletion mutagenesis indicates that ETO interacts with mSin3A independently of its association with N-CoR. Single amino acid mutations that impair the ability of ETO to interact with the central portion of N-CoR affect the ability of the t(8;21) fusion protein to repress transcription. Finally, AML-1/ETO associates with histone deacetylase activity and a histone deacetylase inhibitor impairs the ability of the fusion protein to repress transcription. Thus, t(8;21) fuses a component of a corepressor complex to AML-1 to repress transcription.

BCOR regulates myeloid cell proliferation and differentiation.

BCOR is a component of a variant Polycomb group repressive complex 1 (PRC1). Recently, we and others reported recurrent somatic BCOR loss-of-function mutations in myelodysplastic syndrome and acute myelogenous leukemia (AML). However, the role of BCOR in normal hematopoiesis is largely unknown. Here, we explored the function of BCOR in myeloid cells using myeloid murine models with Bcor conditional loss-of-function or overexpression alleles. Bcor mutant bone marrow cells showed significantly higher proliferation and differentiation rates with upregulated expression of Hox genes. Mutation of Bcor reduced protein levels of RING1B, an H2A ubiquitin ligase subunit of PRC1 family complexes and reduced H2AK119ub upstream of upregulated HoxA genes. Global RNA expression profiling in murine cells and AML patient samples with BCOR loss-of-function mutation suggested that loss of BCOR expression is associated with enhanced cell proliferation and myeloid differentiation. Our results strongly suggest that BCOR plays an indispensable role in hematopoiesis by inhibiting myeloid cell proliferation and differentiation and offer a mechanistic explanation for how BCOR regulates gene expression such as Hox genes.

The BCL-6 POZ domain and other POZ domains interact with the co-repressors N-CoR and SMRT.

Virtually all diffuse large cell lymphomas and a significant fraction of follicular lymphomas contain translocations and/or point mutations in the 5' non-coding region of the putative oncogene BCL-6, that are presumed to deregulate its expression. BCL-6 encodes a Cys2-His2 zinc finger transcriptional repressor with a POZ domain at its amino-terminus. The POZ (or BTB) domain, a 120-amino-acid motif, mediates homomeric and, in some proteins, heteromeric POZ-POZ interactions. In addition, the POZ domain is required for transcriptional repression of several proteins, including BCL-6. Using a yeast two-hybrid screen, we identified N-CoR and SMRT as BCL-6 interacting proteins. Both N-CoR and SMRT, which were originally identified as co-repressors for the unliganded nuclear thyroid hormone and retinoic acid receptors, are components of large complexes containing histone deacetylases. We show that the interaction between BCL-6 and these co-repressors is also detected in the more physiologically relevant mammalian two-hybrid assay. The POZ domain is necessary and sufficient for interaction with these co-repressors. BCL-6 and N-CoR co-localize to punctate regions of the nucleus. Furthermore, when BCL-6 is bound to its consensus recognition sequence in vivo, it can interact with N-CoR and SMRT. We find, in vitro, that POZ domains from a variety of other POZ domain-containing proteins, including the transcriptional repressor PLZF, as well as ZID, GAGA and a vaccinia virus protein, SalF17R, also interact with varying affinities with N-CoR and SMRT. We find that BCL-6 POZ domain mutations that disrupt the interaction with N-CoR and SMRT no longer repress transcription. In addition, these mutations no longer self associate suggesting that self interaction is required for interaction with the co-repressors and for repression. More recently N-CoR has also been implicated in transcriptional repression by the Mad/Mxi proteins. Our demonstration that N-CoR and SMRT interact with the POZ domain containing proteins indicates that these co-repressors are likely involved in the mediation of repression by multiple classes of repressors and may explain, in part, how POZ domain containing repressors mediate transcriptional repression.

mRNA expression analysis and the molecular basis of neonatal testis defects in Dmrt1 mutant mice.

Transcriptional regulators containing the DM domain DNA binding motif have been found to control sexual differentiation in a diverse group of metazoan animals including vertebrates, insects, and nematodes, suggesting that these proteins may comprise a very ancient group of sexual regulators. Dmrt1, 1 of 7 mammalian DM domain genes, is essential for several aspects of testicular differentiation in mice. The Dmrt1 mutant phenotype becomes apparent shortly after birth, and culminates in severe testicular dysgenesis. To better understand the roles of Dmrt1 in testicular development we have performed a more detailed analysis of its mutant phenotypes, and we have used mRNA expression profiling to identify genes misregulated in the neonatal Dmrt1 mutant testis. We find that Dmrt1 mutant germ cells fail to undergo several of the normal postnatal events of germ cell development, including radial movement, mitotic proliferation, differentiation into spermatogonia, and initiation of meiosis, and they die by P14. During this period Dmrt1 mutant Sertoli cells fail to polarize and form tight junctions, and fail to cease proliferation, eventually filling the seminiferous tubules. Expression profiling at P1 and P2 in Dmrt1 mutant testes indicates defects in several important testicular signaling pathways (Gdnf, retinoic acid, TGFbeta, FSH), and detects elevated expression of the pluripotency marker Stella/Dppa3/Pgc7, providing insight into the molecular basis of Dmrt1 testis defects. This work also identifies a number of new candidate testicular regulators for further investigation.

Purification of RNA and RNA-protein complexes by an R17 coat protein affinity method.

We describe an affinity chromatography method to isolate specific RNAs and RNA-protein complexes formed in vivo or in vitro. It exploits the highly selective binding of the coat protein of bacteriophage R17 to a short hairpin in its genomic RNA. RNA containing that hairpin binds to coat protein that has been covalently bound to a solid support. Bound RNA-protein complexes can be eluted with excess R17 recognition sites. Using purified RNA, we demonstrate that binding to immobilized coat protein is highly specific and enables one to separate an RNA of interest from a large excess of other RNAs in a single step. Surprisingly, binding of an RNA containing non-R17 sequences to the support requires two recognition sites in tandem; a single site is insufficient. We determine optimal conditions for purification of specific RNAs by comparing specific binding (retention of RNAs with recognition sites) to non-specific binding (retention of RNAs without recognition sites) over a range of experimental conditions. These results suggest that binding of immobilized coat protein to RNAs containing two sites is cooperative. We illustrate the potential utility of the approach in purifying RNA-protein complexes by demonstrating that a U1 snRNP formed in vivo on an RNA containing tandem recognition sites is selectively retained by the coat protein support.

The POZ domain: a conserved protein-protein interaction motif.

We describe a novel zinc finger protein, ZID (zinc finger protein with interaction domain). At its amino terminus ZID contains a 120-amino-acid conserved motif present in a large family of proteins that includes both the otherwise unrelated zinc finger proteins, such as Ttk, GAGA, and ZF5, and a group of poxvirus proteins: We therefore refer to this domain as the POZ (poxvirus and zinc finger) domain. The POZ domains of ZID, Ttk, and GAGA act to inhibit the interaction of their associated finger regions with DNA. This inhibitory effect is not dependent on interactions with other proteins and does not appear dependent on specific interactions between the POZ domain and the finger region. The POZ domain acts as a specific protein-protein interaction domain: The POZ domains of ZID and Ttk can interact with themselves but not with each other, POZ domains from ZF5, or the viral protein SalF17R. However, the POZ domain of GAGA can interact efficiently with the POZ domain of Ttk. In transfection experiments, the ZID POZ domain inhibits DNA binding in NIH-3T3 cells and appears to localize the protein to discrete regions of the nucleus. We discuss the implications of multimerization for the function of POZ domain proteins.

BCoR, a novel corepressor involved in BCL-6 repression.

BCL-6 encodes a POZ/zinc finger transcriptional repressor that is required for germinal center formation and may influence apoptosis. Aberrant expression of BCL-6 due to chromosomal translocations is implicated in certain subtypes of non-Hodgkin's lymphoma. The POZ domains of BCL-6 and several other POZ proteins interact with corepressors N-CoR and SMRT. Here we identify and characterize a novel corepressor BCoR (BCL-6 interacting corepressor), which is expressed ubiquitously in human tissues. BCoR can function as a corepressor when tethered to DNA and, when overexpressed, can potentiate BCL-6 repression. Specific class I and II histone deacetylases (HDACs) interact in vivo with BCoR, suggesting that BCoR may functionally link these two classes of HDACs. Strikingly, BCoR interacts selectively with the POZ domain of BCL-6 but not with eight other POZ proteins tested, including PLZF. Additionally, interactions between the BCL-6 POZ domain and SMRT, N-CoR, and BCoR are mutually exclusive. The specificity of the BCL-6/BCoR interaction suggests that BCoR may have a role in BCL-6-associated lymphomas.

Expression of Dmrt1 in the genital ridge of mouse and chicken embryos suggests a role in vertebrate sexual development.

Sex-determining mechanisms are highly variable between phyla. Only one example has been found in which structurally and functionally related genes control sex determination in different phyla: the sexual regulators mab-3 of Caenorhabditis elegans and doublesex of Drosophila both encode proteins containing the DM domain, a novel DNA-binding motif. These two genes control similar aspects of sexual development, and the male isoform of DSX can substitute for MAB-3 in vivo, suggesting that the two proteins are functionally related. DM domain proteins may also play a role in sexual development of vertebrates. A human gene encoding a DM domain protein, DMRT1, is expressed only in the testis in adults and maps to distal 9p24.3, a short interval that is required for testis development. Earlier in development we find that murine Dmrt1 mRNA is expressed exclusively in the genital ridge of early XX and XY embryos. Thus Dmrt1 and Sry are the only regulatory genes known to be expressed exclusively in the mammalian genital ridge prior to sexual differentiation. Expression becomes XY-specific and restricted to the seminiferous tubules of the testis as gonadogenesis proceeds, and both Sertoli cells and germ cells express Dmrt1. Dmrt1 may also play a role in avian sexual development. In birds the heterogametic sex is female (ZW), and the homogametic sex is male (ZZ). Dmrt1 is Z-linked in the chicken. We find that chicken Dmrt1 is expressed in the genital ridge and Wolffian duct prior to sexual differentiation and is expressed at higher levels in ZZ than in ZW embryos. Based on sequence, map position, and expression patterns, we suggest that Dmrt1 is likely to play a role in vertebrate sexual development and therefore that DM domain genes may play a role in sexual development in a wide range of phyla.

Dmrt1, a gene related to worm and fly sexual regulators, is required for mammalian testis differentiation.

The only molecular similarity in sex determination found so far among phyla is between the Drosophila doublesex (dsx) and Caenorhabditis elegans mab-3 genes. dsx and mab-3 contain a zinc finger-like DNA-binding motif called the DM domain, perform several related regulatory functions, and are at least partially interchangeable in vivo. A DM domain gene called Dmrt1 has been implicated in male gonad development in a variety of vertebrates, on the basis of embryonic expression and chromosomal location. Such evidence is highly suggestive of a conserved role(s) for Dmrt1 in vertebrate sexual development, but there has been no functional analysis of this gene in any species. Here we show that murine Dmrt1 is essential for postnatal testis differentiation, with mutant phenotypes similar to those caused by human chromosome 9p deletions that remove the gene. As in the case of 9p deletions, Dmrt1 is dispensable for ovary development in the mouse. Thus, as in invertebrates, a DM domain gene regulates vertebrate male development.

Site-directed ribose methylation identifies 2'-OH groups in polyadenylation substrates critical for AAUAAA recognition and poly(A) addition.

The importance of sugar contacts for the sequence-specific recognition that occurs during polyadenylation of mRNAs was investigated with chemically synthesized substrates containing 2'-O-CH3 groups at selected riboses. An RNA (5'-CUGCAAUAAACAAGU-UAA-3') with 2'-O-CH3 ribose at each nucleotide except for the AAUAAA sequence and 3'-terminal adenosine was efficiently polyadenylated in vitro. Methylation of single riboses within AAUAAA inhibited both poly(A) addition and binding of the specificity factor, but the magnitude of inhibition varied greatly at different nucleotides. Nucleotides that showed sensitivity to base substitutions did not necessarily show sensitivity to ribose methylation, and vice versa. The data indicate that the specificity factor interacts with AAUAAA through RNA-protein contacts involving essential recognition of both sugars and bases at different nucleotide positions.

A region of human chromosome 9p required for testis development contains two genes related to known sexual regulators.

Deletion of the distal short arm of chromosome 9 (9p) has been reported in a number of cases to be associated with gonadal dysgenesis and XY sex reversal, suggesting that this region contains one or more genes required in two copies for normal testis development. Recent studies have greatly narrowed the interval containing this putative autosomal testis-determining gene(s) to the distal portion of 9p24.3. We previously identified DMRT1, a human gene with sequence similarity to genes that regulate the sexual development of nematodes and insects. These genes contain a novel DNA-binding domain, which we named the DM domain. DMRT1 maps to 9p24. 3 and in adults is expressed specifically in the testis. We have investigated the possible role of DM domain genes in 9p sex reversal. We identified a second DM domain gene, DMRT2, which also maps to 9p24.3. We found that point mutations in the coding region of DMRT1 and the DM domain of DMRT2 are not frequent in XY females. We showed by fluorescence in situ hybridization analysis that both genes are deleted in the smallest reported sex-reversing 9p deletion, suggesting that gonadal dysgenesis in 9p-deleted individuals might be due to combined hemizygosity of DMRT1 and DMRT2.