Edwin M. Southern

Detection of specific sequences among DNA fragments separated by gel electrophoresis

This paper describes a method of transferring fragments of DNA from agarose gels to cellulose nitrate filters. The fragments can then be hybridized to radioactive RNA and hybrids detected by radioautography or fluorography. The method is illustrated by analyses of restriction fragments complementary to ribosomal RNAs from Escherichia coli and Xenopus laevis , and from several mammals.

Molecular interactions on microarrays

The structural features of nucleic acid probes tethered to a solid support and the molecular basis of their interaction with targets in solution have direct implications for the hybridization process. We discuss how arrays of oligonucleotides provide powerful tools to study the molecular basis of these interactions on a scale which is impossible using conventional analysis.

Analyzing and comparing nucleic acid sequences by hybridization to arrays of oligonucleotides: Evaluation using experimental models

An efficient method was developed for making complete sets of oligonucleotides of defined length, covalently attached to the surface of a glass plate, by synthesizing them in situ. A device carrying all octapurine sequences was used to explore factors affecting molecular hybridization of the tethered oligonucleotides, to develop computer-aided methods for analyzing the data, and to test the feasibility of using the method for sequence analysis. Further development is needed before the method can be used routinely, but our work shows that it has a number of potential advantages over gel-based methods: it should be easy to automate; the quality of the sequence results can be evaluated statistically; it provides a powerful way of comparing related sequences and detecting mutation; it can be applied to both DNA and RNA; and specific motifs can be incorporated into all sequences of the array to focus analysis on sequences of biological interest.

Use of restriction enzymes to study eukaryotic DNA methylation: I. The methylation pattern in ribosomal DNA from Xenopus laevis☆

Abstract Restriction enzymes Hpa II, Ava I, Hha I and Hae II have been found to distinguish Xenopus laevis somatic (erythrocyte) rDNA † from amplified rDNA. Amplified rDNA is cleaved at many sites by each enzyme whilst somatic rDNA is relatively resistant to digestion. The difference is attributed to the presence of 5-methylcytosine rendering many restriction sites in somatic rDNA resistant to nuclease attack. The major methylated sequence in eukaryotic DNA is CpG, and recognition sites for all four enzymes contain this sequence. The “ enzymes” thus identified have been used to study the pattern of methylation in erythrocyte rDNA. Unmethylated CpGs were located by mapping vulnerable restriction sites and the distribution and level of CpG methylation was determined. For most detected CpGs the level of methylation is high (about 99%). However there is one site which is unmethylated in 30 to 60% of rDNA repeat units. There is also a region in the non-transcribed “spacer” within which unmethylated sites for Hpa II and Ava I are concentrated, though it is not yet certain whether this is due to undermethylation.

Measurement of DNA length by gel electrophoresis.

Abstract Plotting fragment length against reciprocal of mobility gives a straight line over a wider range than the conventional semilogarithmic plot. Curvature can be removed by a simple correction. A method is also given for determining molecular weights from mobilities by direct calculation.

Long range periodicities in mouse satellite DNA

Abstract Escherichia coli restriction enzyme II breaks mouse satellite DNA into fragments which form a series of bands on gel electrophoresis. The DNA in the strongest band has a length of 220 to 260 nucleotide pairs and the other bands are multiples of this length. It is shown that these fragments are linked together in long arrays in the satellite sequence. The reassociation register of the DNA is about half the length of the 220 to 260 nucleotide pair fragment. In the electrophoresis pattern of the Eco RII † fragments other weaker bands can be seen. The stronger bands of the minor patterns fall half-way between the bands of the main pattern and the smallest is 120 to 130 nucleotide pairs long. The properties of the minor fragments suggest short spacings of the restriction site which have been produced by unequal crossing-over. The extents of divergence and unequal crossing-over are estimated. From this analysis and the sequence analysis described in the accompanying paper (Biro et al., 1975) it is proposed that mouse satellite DNA consists of an hierarchy of four periodicities which reflect stages in the evolution of the sequence. Digestion of mouse satellite DNA with Hae III produces fragments with the same sizes as those produced by Eco RII, but the yields are much lower. It is suggested that Hae III sites have been introduced by divergence and subsequently spread by unequal crossing-over.

Determining the influence of structure on hybridization using oligonucleotide arrays.

We have studied the effects of structure on nucleic acid heteroduplex formation by analyzing hybridization of tRNAphe to a complete set of complementary oligonucleotides, ranging from single nucleotides to dodecanucleotides. The analysis points to features in tRNA that determine heteroduplex yield. All heteroduplexes that give high yield include both double-stranded stems as well as single-stranded regions. Bases in the single-stranded regions are stacked onto the stems, and heteroduplexes terminate at potential interfaces for coaxial stacking. Heteroduplex formation is disfavored by sharp turns or a lack of helical order in single-stranded regions, competition from bases displaced from a stem, and stable tertiary interactions. The study is relevant to duplex formation on oligonucleotide microarrays and to antisense technologies.

Heterogeneity of the ribosomal genes in mice and men

The structures of mouse and human ribosomal DNA were studied using the restriction enzymes Eco R1 and Hind III. Individual mice or humans showed a heterogeneous pattern of restriction fragments resulting from differences in the non-transcribed spacer DNA. Six individual mice from the inbred strain CBA/H-T6 had identical patterns. The same pattern was shown by another CBA strain and by C3H. These strains were originally derived from a BALB X DBA cross made in 1920. Different patterns were found for BALB/c, C57BL and Mus poschiavinus. Cultured cells derived from C3H mice (L cells) showed a pattern quantitatively different from that of the parent strain, but two myeloma cell lines derived from BALB/c showed the same pattern as BALB/c mice. Ribosomal DNA in man is also heterogeneous. Differences were observed between human DNAs in the amounts of the different spacer classes. Studies on mouse-human cell hybrids suggest that some spacer classes are present on more than one of the five human nucleolus organizers.

Partial sequence analysis of mouse satellite DNA: Evidence for short range periodicities

Abstract The sequence of satellite DNA from Mus musculus was investigated both by direct analysis of the pyrimidine tracts derived from each strand and by conventional analysis of complementary RNA transcribed from each DNA strand and labelled with each of the four α- 32 P-labelled nucleotide triphosphates. The sequence is shown to be based on a repeating unit less than 20 nucleotide pairs in length. Four major oligonucleotides are identified, all of which could derive from an original sequence d(GA 5 TGA) (light strand) which probably comprises about half of the basic repeating unit. The sequence is heterogeneous and yields several oligonucleotides in low amounts that are related to the major oligonucleotides and which could have arisen from them by divergence. Other possible explanations for the occurrence of the minor sequences are considered.

Measurement of DNA length by gel electrophoresis II: Comparison of methods for relating mobility to fragment length

Relationships between DNA length and electrophoretic mobility in an agarose gel have been compared by estimating the lengths of known DNA polymer fragments, using other fragments in the series as standards. Global estimates were made using 10 fragments as standards; local estimates were made using the two closest fragments on either side of the unknown as standards. Most relationships were fitted by least squares. All the relationships gave more accurate local than global estimates. The most accurate results were obtained using the reciprocal relationship, where the maximum error in the local estimates was less than 0.1%. The semilog relationship gave the least accurate results, with a maximum error in the local estimates of almost 5%. The polymer fragments were also used as standards to estimate the lengths of lambda DNA restriction fragments. Here the estimates were in error by up to 3%, indicating the influence of base composition and sequence on electrophoretic mobility.