Histone acetylation-mediated regulation of genes in leukaemic cells.

Histone deacetylase (HDAC) and histone acetyltransferase (HAT) functions are associated with various cancers, and the inhibition of HDAC has been found to arrest disease progression. Here, we have investigated the gene expression profiles of leukaemic cells in response to the HDAC inhibitor trichostatin A (TSA) using oligonucleotide microarrays. Nucleosomal histone acetylation was monitored in parallel and the expression profiles of selected genes were confirmed by quantitative polymerase chain reaction (PCR). A large number of genes (9% of the genome) were found to be similarly regulated in CCRF-CEM and HL-60 cells in response to TSA, and genes showing primary and secondary responses could be distinguished by temporal analysis of gene expression. A small fraction of genes were highly sensitive to histone hyper-acetylation, including XRCC1, HOXB6, CDK10, MYC, MYB, NMI and CBFA2T3 and many were trans-acting factors relevant to cancer. The most rapidly repressed gene was MKRN3, an imprinted gene involved in the Prader-Willi syndrome.

An oligonucleotide probe specific for Onchocerca volvulus.

A genomic DNA library of a Liberian strain of Onchocerca volvulus was prepared in the vector bacteriophage lambda gt10. The library was differentially screened by hybridisation with radiolabelled total DNA from the homologous parasite, two heterologous Onchocerca parasites (Onchocerca gibsoni and Onchocerca gutturosa) and human liver cells. A clone (C1A1) was isolated whose binding to O. volvulus DNA was at least 50 times stronger than to the other parasite DNA samples. No binding was observed with human DNA. The insert of C1A1 was subcloned into the filamentous phage vector M13 mp18 and sequenced. Two oligonucleotides, each corresponding to a unique region of 60 nucleotides (out of a total of 154) were synthesised and examined for hybridisation with three different geographical isolates of O. volvulus (including forest and savannah strains) and six other Onchocerca spp. One of the oligonucleotides (C1A1-2) was found to hybridise to the three O. volvulus isolates with an intensity in the region of 300 times greater than to any other Onchocerca spp. Since the other species include the two which may be most closely related to O. volvulus, i.e., O. gibsoni and Onchocerca ochengi, it is concluded that C1A1-2 is likely to represent a truly species-specific probe.

Repetitive DNA as a tool for the identification and comparison of nematode variants: application to Trichinella isolates.

DNA prepared from four isolates of Trichinella was compared by genomic DNA cross-hybridisation, by electrophoresis following restriction endonuclease digestion and by hybridisation studies using a cloned repetitive DNA sequence from T. spiralis. The DNA from T. spiralis, T. nelsoni and T. pseudospiralis isolates was distinct and the interrelationships of these isolates were inferred. In contrast to previous work on T. nativa and T. spiralis, our work suggests that these two isolates are very similar.

Predicted structure of a major Schistosoma mansoni eggshell protein.

The complete sequence for a major Schistosome mansoni eggshell protein gene has been determined from a genomic DNA fragment. The use of an open reading frame encoding a glycine-rich polypeptide was confirmed by in vitro translation of schistosome mRNA in the presence of [3H]glycine and comparison with the amino acid composition of purified, schistosome eggshells. Apart from the extraordinary abundance of glycine and tyrosine which are evenly distributed throughout the polypeptide chain, the most striking features of the deduced amino acid sequence are the presence of five well-conserved tandem repeats of 16-18 residues in the N-terminal region and the asymmetrical distribution of charged residues. Acidic residues (Asp) are confined to the N-terminal region, while basic residues (Lys, His), with the exception of a single histidine, are found in the C-terminal region. A model structure composed of short anti-parallel beta-strands is proposed, in which glycines and residues with small side chains lie within the strands and tyrosines and cysteines are arranged at the bends, where they would be available for cross-linking. Four such strands form one of the tandem repeats which are predicted in turn to form a stack of five closely packed beta-sheets, each of three strands and linked by the more variable fourth strand. The C-terminal region may form a similar but less compact structure. The ordered structure demonstrated by birefringence studies of the schistosome eggshell [Kusel, J. (1970) Parasitology 60, 79-88] could be formed by packing of the polypeptides such that the N-terminal domain contributes counter ions or cross-links to the C-terminal domain of adjacent molecules.

Expression of genes encoding the transcription factor SRF during early development of Xenopus laevis: identification of a CArG box-binding activity as SRF.

cDNA clones encoding the sequence-specific DNA binding protein, serum response factor (SRF), have been isolated from a Xenopus laevis neurula library and their nucleotide sequence determined. The Xenopus SRF (SRFX) gene produces multiple-sized transcripts, present at 10(5) copies per unfertilized egg. A similar level is detected in the embryo during early cleavage, but SRFX transcripts accumulate rapidly following gastrulation. The protein they encode is similar in sequence to human SRF in its central and carboxy-terminal regions, but possesses a divergent amino-terminal portion. We have previously described a Xenopus embryo sequence-specific binding activity that recognized the CArG motif of the cardiac actin gene promoter. Here we show that the DNA-binding characteristics of synthetic SRFX are indistinguishable from those of the embryo factor. Moreover, antiserum raised against the synthetic SRFX recognizes this factor. Together, these results establish that the same factor binds to elements required for constitutive transcription in Xenopus oocytes, muscle-specific gene expression in Xenopus embryos and serum-responsive transcription in cultured amphibian cells.

Muscle-specific expression of SRF-related genes in the early embryo of Xenopus laevis.

We have isolated two members of the RSRF protein family, SL-1 and SL-2, in Xenopus laevis. Both proteins contain SRF-type DNA binding domains and are related to the human protein, RSRFC4. SL-1 constitutes a novel member of the RSRF family whilst SL-2 is similar to human RSRFC4 throughout its length. SL-1 protein recognizes the consensus DNA sequence CTA(A/T)4TAR in vitro and can bind to the same regulatory sites as other A/T-rich sequence-specific binding activities, such as the muscle-specific regulatory factor, MEF-2. Transcription of both Xenopus genes is restricted to the somitic mesoderm of early embryos and subsequently to the body muscle (myotomes) of the tadpole. In contrast, both genes are expressed ubiquitously in the adult frog. A binding activity, antigenically related to both human RSRFC4 and the SL-2 gene product, is detected in Xenopus embryos and after gastrulation is localized to embryonic muscle. An indistinguishable binding activity is detected in many adult frog tissues. We conclude that the RSRF genes undergo a dramatic switch in their patterns of expression during development. We suggest that RSRF proteins may regulate muscle-specific transcription in embryos, but acquire other roles during the course of development.

The RSRF/MEF2 protein SL1 regulates cardiac muscle-specific transcription of a myosin light-chain gene in Xenopus embryos.

We have examined the role of two RSRF/MEF2 proteins in the onset of skeletal and cardiac muscle differentiation in early Xenopus embryos. In normal development, zygotic expression of SL1 (MEF2D) precedes that of SL2 (MEF2A) by several hours, but neither gene is expressed prior to the accumulation of MyoD and Myf5 transcripts in the somitic mesoderm. Ectopic expression of the myogenic factors in explants of presumptive ectoderm induces expression of both SL1 and SL2, whereas in reciprocal experiments, neither RSRF protein activates the endogenous myoD or Myf5 genes. We conclude that SL1 and SL2 lie downstream of these myogenic factors in the skeletal myogenic pathway. SL1 is distinguished from SL2 in being expressed in the presumptive heart region of the early tailbud embryo, prior to detection of any markers for cardiac muscle differentiation. Furthermore, ectopic SL1 induces the expression of an endogenous cardiac muscle-specific myosin light-chain (XMLC2) gene in cultured blastula animal pole explants, whereas SL2 has no comparable effect. These results demonstrate that in addition to a possible role in skeletal myogenesis, SL1 also acts in vivo as a regulator of cardiac muscle-specific transcription.