Restriction Landmark Genome Scanning
Restriction landmark genomic scanning (RLGS) is a method that provides both a quantitative genetic and epigenetic (cytosine methylation) assessment of thousands of CpG islands in a single gel without prior knowledge of gene sequence. The method is a two-dimensional separation of radiolabeled genomic DNA into nearly 2,000 discrete fragments that have a high probability of containing gene sequences and are ideal in length for cloning and sequence analysis. Genomic DNA is digested with an infrequently cutting restriction enzyme such as NotI, radiolabeled at the cleaved ends, digested with a second restriction enzyme, and then electrophoresed through a narrow, 60 cm-long agarose tube-shaped gel.
The DNA in the tube gel is then digested by a third, more frequently cutting restriction enzyme and electrophoresed, in a direction perpendicular to the fi rst separation, through a 5% nondenaturing polyacrylamide gel, and the gel is autoradiographed. Radiolabeled NotI sites are frequently used as “landmarks” because NotI can not cleave methylated sites and since an estimated 89% of NotI sites are within CpG islands. Using a methylation-sensitive enzyme, the technique has been termed RLGS-M. The resulting RLGS profi le displays both the copy number and methylation status of the CpG islands. These profi les are highly reproducible and are therefore amenable to inter- and intra-individual DNA sample comparisons. To increase the number of fragments analyzed by RLGS, the DNA samples can be processed with a different series of enzymes. The choice of a “landmark” enzyme is critical since this site determines the bias of the displayed fragments. To maintain a strong bias for CpG islands, landmark enzymes such as NotI, BssHII, or EagI are generally used. Alternatively, a different second and/or third restriction enzyme may be used along with same landmark enzyme to display a different subset of fragments. Differences between RLGS profi les have been used to identify important genes involved in normal cellular processes and in disease states. Two novel imprinted genes, one encoding a ribonucleoprotein auxiliary factor and the second encoding Cdc25Mm, were isolated using this approach.
By determining the proportion of DNA fragments on RLGS profi les that display a potentially imprinted pattern, it has been possible to obtain an estimate of the total number of imprinted genes in the genome (4). These genomic loci were identifi ed as having a parent-of-origin-specifi c methylation pattern and indicated that a 50% change in the intensity of a single-copy DNA fragment was readily detectable in an RLGS profi le. Such a reduction is also apparent in X-chromosome specifi c fragments derived from either males or females, since there is methylation-related inactivation of one X chromosome in the latter. Similarly, comparison of profi les from normal individuals to those from Down’s Syndrome patients has revealed a proportional increase in the intensity of many chromosome 21 specifi c loci. Several chromosome 21 CpG islands were methylated on one copy of chromosome 21 and potentially represented an attenuation mechanism allowing for viability of a trisomy chromosome 21 fetus. RLGS and standard positional cloning have also been combined to identify the gene responsible for cardiomyopathy in Syrian hamsters and to identify the mouse reeler gene. Normal genetic variation among related individuals has also been detected by RLGS.
Thus, RLGS can be used for widespread methylation analysis of CpG islands and also to defi ne a level of genetic or epigenetic change in cells. This approach has been used to identify novel tumor-specifi c targets of DNA amplifi cation, aberrant CpG island methylation, and repetitive sequences that are demethylated in human cancer and in experimentally induced rodent tumors. For example, the gene encoding human cyclin-dependent kinase-6 (CDK6) was identifi ed as a novel target of DNA amplifi cation in human brain tumors. Similarly, the tumor-suppressor gene, P16/INK4, as well as a variety of other CpG islands were identifi ed as frequent targets of aberrant methylation in mouse liver tumors induced by tissue-specific expression of an SV-40 transgene. The chromosomal origin of the majority of DNA fragments displayed on RLGS profi les has been mapped by chromosome-assigned RLGS (CA-RLGS). CA-RLGS profi les were generated from fl ow-sorted human chromosomes and then each individual chromosome-specifi c profi le was integrated into a total genomic DNA profi le. In mouse, a more detailed linkage between 1045 DNA fragments displayed on RLGS profi les and specifi c loci within each chromosome has been generated from profi les derived from a panel of recombinant inbred mouse strains.
RLGS-M has signifi cant advantages when compared to polymerase chain reaction (PCR)-based global methylation-analysis techniques. First, there is a greater than 90% bias toward display of CpG island-containing DNA fragments when using NotI, which is critical since CpG islands are tightly linked with genes (15). CpG island bias is especially important in examining cancer-cell genomes for altered methylation as the majority of such changes occur in noncoding, potentially nonfunctional regions of the genome. Certain repetitive elements in the genome are also rich in CpG dinucleotides and can be subjected to methylation changes in cancer cells. These repetitive elements are displayed as high copy-number fragments on RLGS profi les, distinguishing them from single-copy CpG islands. Second, several thousand CpG island fragments can be analyzed simultaneously, whereas other methylation scanning methods visualize 10-fold fewer methylation sites. Third, quantitative information for each fragment can be derived directly from the profi les and compared to that obtained from many fragments that are of similar size and are invariant in intensity between samples. Fourth, since RLGS does not involve PCR amplifi cation, the particular CpG islands that are displayed are not restricted by the limited ability of PCR enzymes to amplify GC-rich templates.
Tags: Forensic DNA
