« Restriction Landmark Genome Scanning
COBRA - Combined Bisulfite Restriction Analysis »

Analysis of RLGS Profiles

Using direct visual assessment of profi les has proven very reliable. There is a 100% concordance between alterations of DNA fragments detected in RLGS profi les and, following cloning, alteration demonstrated by Southern blotting. Comparisons by computer or by visual assessment are facilitated by the fact that RLGS profi les from different tissues or from different individuals are identical at the majority of loci.

Visual assessment is performed by overlaying two RLGS profi les on a light box and comparing relative intensities of fragments. Overlaying “master” profi les (profi les used as a standard for comparison) with clear acetate sheets allows one to mark differences or similarities between multiple profi les and to generate cumulative data sets. This is also a convenient form in which to retain a usable record of the analysis results, and is a standard that has allowed the sharing of results between labs. Several computerassisted analysis systems, which were originally designed for analysis of two-dimensional (2-D) protein gels, have been developed for the 2-D DNA analysis. One of the disadvantages of using RLGS for global methylation analysis of CpG islands, at least in the analysis of tumor tissue, is that a loss of a fragment from an RLGS profi le could be due to either deletion or methylation. Once the fragment is cloned, these possibilities are distinguished by Southern blotting. It has been our experience that the vast majority of “loss” events detected by RLGS in human tumors are due to methylation.

To tailor the profiles for detection of genetic loss events only, methylation-insensitive landmark enzymes are used. The identifi cation and cloning of DNA fragments that display a tumorspecifi c increase in intensity has led to two very different fi ndings, which are indistinguishable until the DNA fragments are cloned and tested as probes on Southern blots. The increased intensity in some cases corresponded to gene amplifi cation while for others identifi ed DNA fragments from repetitive sequences that had become demethylated in a tumor-specifi c manner. Thus, the interpretation of tumor-specifi c intense fragments on profi les must be refi ned by cloning and Southern-blot analysis. As additional fragments are sequenced, the profi les will become even more informative. The following cloning protocol combines the advantages of the two published cloning procedures: the restriction trapper-based direct cloning of NotI/Hinf I and the PCR-mediated method.

Standard RLGS gels contain a large amount of unlabeled background Hinf I or Hinf I/EcoRV fragments, which could compete with target NotI/Hinf I fragments in the cloning process. A protocol was developed that eliminates more than 90% of the EcoRV/EcoRV fragments, while retaining NotI/EcoRV fragments. RLGS profi les generated with purifi ed NotI/EcoRV fragments are indistinguishable from total genomic DNA profi les. Using the human genome sequence, it is also possible to identify the sequence and chromosomal position of fragments through in silico restriction digests. Since the 5′ overhangs created by NotI digestion are “fi lled in” during the labeling step, standard DNA “analysis” gels are not suitable for cloning with NotI adapters. To generate profi les suitable for such cloning, only 1/5 of the restriction trapper-purifi ed DNA is subjected to labeling and is then mixed with the remaining 4/5 unlabeled DNA prior to loading the fi rst-dimension agarose gel. If suffi cient DNA can be eluted from the restriction trapper gels direct cloning into a vector prepared with NotI/Hinf I ends is possible. For cloning from limited amounts of DNA, the following protocol should be used.


Tags:

This entry was posted on Sunday, January 27th, 2008 at 11:41 am and is filed under Forensic DNA. You can follow any responses to this entry through the RSS 2.0 feed. Responses are currently closed, but you can trackback from your own site.