Evidence that cytosine residues within 5'-CCTGG-3' pentanucleotides can be methylated in human DNA independently of the methylating system that modifies 5'-CG-3' dinucleotides

In contrast to the complex sequence specificities of the prokaryotic DNA methylating systems, the mammalian machinery identified thus far methylates cytosine residues within the context of a 5'-CG-3' dinucleotide. To explore the possibility that cytosine residues that do not precede guanine may be independently methylated in mammalian DNA, we have examined a region of the human myogenic gene, Myf-3, which is not targeted by the methylating system that methylates 5'-CG-3' dinucleotides. Our investigations have revealed cytosine methylation within the 5'-CCTGG-3' pentanucleotides specified by the 0.8-kb Myf-3 probe. We have also found that in DNA from neoplastic cells, in which 5'-CG-3' dinucleotides within Myf-3 become abnormally hypermethylated, cytosine residues within 5'-CCTGG-3' pentanucleotides are not methylated. Moreover, methylation of 5'-CCTGG-3' pentanucleotides was not detected within the closely related Myf-4 gene, which is normally 5'-CG-3' hypermethylated. These findings indicate the existence of a system that methylates 5'-CCTGG-3' pentanucleotides independently of the system that methylates cytosine residues within 5'-CG-3' dinucleotides. It is possible that the 5'-CCTGG-3' methylating system influences the fate of foreign integrated DNA.     

Sequence homologies in the protamine gene family of rainbow trout

We have sequenced five different rainbow trout protamine genes plus their flanking regions. The genes are not clustered and do not contain intervening sequences. There is an extremely high degree of sequence conservation in the coding and 3' untranslated regions of the gene. Downstream sequences exhibit little homology though conserved regions are found 250 base pairs 3' to the gene. There are four regions upstream of the gene that are highly conserved in the six clones, including the canonical Goldberg - Hogness box which is 45 base pairs 5' to the coding region. A second homologous region is found 90 bases upstream. Although in the same approximate location as the CAAT box found upstream of other genes, it does not contain the canonical CAAT sequence. Further upstream of the protamine genes at -115 there is an A-T rich sequence while a 25 base pair conserved sequence is located 150 bases upstream. In addition we report the presence of a potential Z-DNA region of predominantly A-C repeats approximately one kilobase downstream of one of the genes.     

Nucleotide sequence of the 5'- and 3'- domains for rabbit 18S ribosomal RNA.

By direct RNA sequence analysis we have determined the primary structures of both the 5' and 3' domains for rabbit 18S ribosomal RNA. Purified 18S rRNA was labeled in vitro at either its 5' or 3' terminus with 32P, base-specifically fragmented enzymatically and chemically, and the resulting fragments electrophoretically fractionated by size in adjacent lanes of 140 cm long polyacrylamide sequencing gels run in 90% formamide. A phylogenetic comparison of both the mammalian 5' proximal 400 residues and the 3' distal 301 nucleotides with the previously determined yeast and Xenopus laevis 18S rRNA sequence shows extensive conservation interspersed with tracts having little homology. Clusters of G + C rich sequences are present within the mammalian 5' domain which are entirely absent in both the Xenopus laevis and yeast 18S rRNAs. Most base differences and insertions within the mammalian 18S rRNA when compared with yeast or Xenopus rRNA result in an increase in the G + C content of these regions. We have found nucleotide sequence analysis of the ribosomal RNA directly permits detection of both cistron heterogeneities and mapping of many of the modified bases.     

Ordered distribution of modified bases in the DNA of a dinoflagellate.

In DNA of the dinoflagellate Crypthecodinium cohnii, 38% of the thymine is replaced by the modified base 5-hydroxymethyluracil, and approximately 3% of the cytosine is replaced by 5-methylcytosine. Both of the modified bases are non-randomly distributed in the DNA. Determinations of 3' nearest neighbors show that HOMeU is preferentially located in the dinucleotides HOMeUpA and HOMeUpC. Pyrimidine tract analysis shows that HOMeU is also greatly enriched in the trinucleotide purine-HOMeU-purine. As in other eukaryotes, methylcytosine in C. cohnii DNA occurs predominantly in the dinucleotide MeCpG. By analysis of restriction endonuclease digestion patterns of C. cohnii total DNA and ribosomal DNA, we have found that the central CpG dinucleotides in the sites for the enzymes Hpa II (CCGG) and Hha I (GCGC) are extensively methylated in both total DNA and ribosomal DNA. Results of digestion with Ava I, however, indicated that not all CpG dinucleotides in the sequence CCTCGGAG are methylated in C. cohnii DNA.     

Anomalous dinucleotide frequencies in both coding and non-coding regions from the genome of the human malaria parasite Plasmodium falciparum

We have statistically analysed the distribution of nucleotides and dinucleotides in 21 genes of the 81% A + T-rich human malaria parasite Plasmodium falciparum. The mRNA-synonymous strands of this protozoan show in general a marked excess of purines over pyrimidines, correlated with abnormally high levels of Lys and Glu. We have used the large differences in base composition between coding and non-coding regions to estimate that the parasite possesses in the range of 2700-5400 genes. The dinucleotide preference patterns are compared with consensus patterns derived from other organisms [Nussinov, Nucl. Acids Res. 12 (1984) 1749-1763]. Patterns in the coding regions surprisingly resemble those of higher, rather than lower eukaryotes, particularly with respect to TG elevation and CG suppression. The latter is correlated with an abnormally low level of Arg in these parasites. In the non-coding regions, the four dinucleotides made up of C and/or G are found with significantly higher frequencies than expected (approx. 50-150%), specifically to the 5' side of the coding regions. The possible role of these dinucleotides in control sequences is discussed.     

A marsupial phosphoglycerate kinase (PGK) processed pseudogene

A clone that cross-hybridized with a full-length human cDNA PGK probe was isolated from a hill kangaroo (Macropus robustus: Marsupialia) lambda EMBL4 EcoRI genomic library. The clone was sequenced and demonstrated to be a pseudogene, with two deletions (one of 3 bases, the other 24 bases long), one single base insertion, and a nonsense mutation with respect to the functional human X-linked gene. It is flanked by terminal repeats in the 5' and 3' noncoding regions, but it has no 3' poly(A) remnant. The 3' untranslated region has a 34-bp sequence, with 29 bp homologous to the human 3' untranslated region. The overall percentage homology with the mouse and human X-linked PGK indicates that this pseudogene is probably more closely related to eutherian X-linked PGK genes than to the autosomal form. The results also suggest that pseudogenes are of considerable antiquity (greater than 100 MYr) in the mammalian lineage.     

Structure of the sea urchin U1 RNA repeat.

The genes coding for U1 RNA in the sea urchin L. variegatus are present in a 1400 base pair tandem repeat. One member of the repeat has been cloned and its sequence determined. The repeat unit contains a single copy of the gene for L. variegatus U1 RNA. This gene encodes an RNA which is 75% homologous to mammalian U1 RNA. The L. variegatus U1 RNA could assume a secondary structure similar to that proposed for other U1 RNAs. In addition the L. variegatus U1 RNA is precipitated by anti-SM and anti-RNP antisera. Analysis of the L. variegatus genomic DNA using the cloned U1 gene as a probe reveals a major and a minor type of repeat unit. The two repeated units are the same length but differ in a number of restriction enzyme sites clustered 200-500 bases down-stream from the gene. The monomer we have cloned and sequenced is a representative of the minor repeat. A sequence (GATAA) which is -41 to -37 bases 5' to the gene has homology to the putative RNA polymerase II promoter. Fifteen bases 3' of the gene is a sequence (CAAAGAAAGAAAA) which is very similar to the sequence found 3' of the sea urchin histone genes. The two Hha I, Hpa II and Ava I sites in the repeat are all unmethylated in sperm DNA.     

Isolation and characterization of two linked mouse U1b small nuclear RNA genes.

A 6.9 kilobase Eco R1 fragment containing genes for two U1 RNAs has been isolated from a library of mouse DNA. The two genes code for an RNA which is very similar, if not identical, to mouse U1b RNA as judged by S1 nuclease mapping. This RNA is one base longer than the mouse U1a RNA, human U1 RNA, and rat U1 RNA and differs in six nucleotide substitutions from rat U1 RNA. The two genes are five kilobases apart and the U1 RNAs are coded for on opposite strands of the DNA with the 5' ends juxtaposed. The sequences flanking the genes are identical for 700 bases 5' to the gene and at least 80 bases 3' to the gene.     

Interactions between the terminal bases of mammalian introns are retained in inosine-containing pre-mRNAs.

Nuclear pre-mRNA splicing has a fundamentally similar two-step mechanism to that employed by group II self-splicing introns. It is believed that nuclear pre-mRNA splicing involves a network of RNA-RNA interactions which form the catalytic core of the active spliceosome. We show here a non-Watson-Crick interaction between the first and last guanosine residues of a mammalian intron. As in Saccharomyces cerevisiae, substitution of the conserved guanosines at the 5' and 3' splice sites by A and C respectively, specifically suppresses step 2 splicing defects resulting from the individual mutations. No other combination of terminal nucleotides was able to restore splicing. We additionally provide independent evidence for an indirect interaction between other nucleotides of the consensus splice sites during step 2 of splicing. Substitution of the nucleotide in the +3 position of the 5' splice site affects competition between closely spaced AG dinucleotides at the 3' splice site, although the interaction is not via direct differential base pairing. Finally, we show that complete substitution of guanosine residues by inosine in a pre-mRNA has only a modest effect upon step 2 of splicing, although earlier spliceosome assembly steps are impaired. Predictions can thus be made about the precise configuration of the non-Watson-Crick interaction between the terminal residues.     

Asymmetrical distribution of CpG in an 'average' mammalian gene.

The frequency and distribution of the rare dinucleotide CpG was examined in 15 mammalian genes. CpG is highly methylated at cytosine in mammalian DNA (1,2) and 5-methylcytosine (5mC) is thought to undergo a transition mutation via deamination to produce thymine (3). This would result in the accumulation of TpG and CpA and depletion of CpG during evolution (4). Consistent with this hypothesis, the gene sample of 26,541 dinucleotides contained CpG at 40% the frequency expected by base composition and the CpG transition products, TpG+CpA, were significantly elevated at 124% of expected random frequency. However, because CpG occurs at only 25% of expected random frequency in the genome, the sampled genes were considerably enriched in this dinucleotide. CpGs were asymmetrically distributed in sequences flanking the genes. 5'-flanking sequences were enriched in CpG at 135% of the frequency expected assuming a symmetrical distribution of all the CpGs in the sampled genes (p less than 0.01), while 3'-flanking regions were depleted in CpG at 40% of expected values (p less than 0.0001). This asymmetry may reflect the role of 5-methylcytosine in gene expression. In contrast the frequencies of GpC and GpT+ ApC did not differ significantly from that predicted by base composition and these dinucleotides were not asymmetrically distributed.     

Comparison of the sequence specificity of bleomycin cleavage in two slightly different DNA sequences.

The sequence specificity of bleomycin damage was investigated utilising 340 bp alpha-DNA (a middle repetitive sequence in the human genome) as a target sequence. The following significant facts were found:- i) The dinucleotides GT and GC were cleaved on all occasions, GA most of the time, and AT, AC, GG and AA cleaved some of the time; ii) The base immediately 5' to the purine-pyrimidine dinucleotides was found to be statistically highly significant in determining the degree of damage caused by bleomycin, while other nearest neighbour bases had no significant effect; iii) The sequence specificity of bleomycin damage was determined on both strands and it was found that damage on either strand follows the above dinucleotide preference and is independent of the extent of damage on the opposite strand; iv) Bleomycin damage was compared between genomic 340 bp alpha-DNA and a cloned alpha-DNA with eleven base substitutions relative to the "consensus" sequence. There were forty-nine detectable differences in intensity of damage between these two DNA molecules. Although four of the differences can be directly attributed to changes in base sequence, the remaining differences were not at the base substitution sites. Some of the differences were over fifty base pairs from the nearest base substitution. We propose that the majority of these differences are due to microvariation in the structure of DNA with a slightly different DNA sequence.     

The majority of methylated deoxycytidines in human DNA are not in the CpG dinucleotide

The only natural postsynthetic modification known to occur in mammalian DNA is the methylation in the 5 position of deoxycytidines. Of the four 5'-CpN-3' dinucleotides (ie. CpG, CpC, CpA, and CpT), the dinucleotide which contains the highest proportion of deoxycytidines methylated is CpG, with 40 to 80% methylation in different mammalian genomes. It has also been shown that CpA, CpT, and CpC are methylated as well but to a much lower extent. Here we report the result of a full nearest neighbour analysis (together with quantitation of methylation levels in the 4 CpN dinucleotides) for DNA from human spleen. Using the values we have calculated the overall frequencies for all the methylated dinucleotides in the human genome. Because of the relative underrepresentation (by 7 to 10 fold) of the CpG dinucleotide, only 45.5% of total mC was present in mCpG, with 54.5% in mCpA, mCpT plus mCpC. These calculations have implications for studies into the function and significance of DNA methylation in mammalian cells.     

Guanine-06 methylation reduces the reactivity of d(GpG) towards platinum complexes.

6-methylated guanine dinucleotides were used to study the influence of hydrogen bonding on the specific binding of the antitumor drug cDDP, cis-PtCl2(NH3)2, to DNA. In this interaction, the guanine-06 site appears to be important in explaining the preference for a pGpG-N7(1),N7(2) chelate, which results from H-bridge formation with the ammine ligand of cDDP. Guanine-06 methylated dinucleotides and the nonmodified dinucleotides were reacted with [Pt(dien)Cl]+, cis-PtCl2(NH3)2, and cis-[Pt(NH3)2(H2O)2]2+ and the reaction products were characterized by 1H NMR using pH titrations. Methylation at guanine-06 clearly reduces the preference for the guanine. In competition experiments monitored by NMR and experiments using UV spectrophotometry a decreasing reactivity towards [Pt(dien)(H2O)]2+ and cis-[Pt(NH3)2(H2O)2]2+ was found, in the order of d(GpG) greater than d(GomepG) greater than d(GpGome) greater than d(GomepGome). The difference in reactivity between 5' guanine methylation and 3' guanine methylation is ascribed to differences in the H-bond formation with the backbone phosphate. The resulting reduced stacking of the bases in both modified dinucleotides, compared to the bases in d(GpG), results in a preference for the 3' guanine over 5'.