DNA REPAIR AND MUTATION DETECTION

ANTHONY T. YEUNG, Ph.D., Member
BING YANG, Ph.D., Postdoctoral Associate
CATHERINE A. OLEYKOWSKI, M.S., Scientific Assistant
DAVID A. BESACK, B.S., Scientific Technician (from April 1998)
JOANNE A. KULINSKI, B.S., Scientific Technician (from September 1998)
XIAO WEN, M.S., Scientific Technician (until July 1998)
HONG MEI LI, Student Volunteer, Central High School, Philadelphia, PA
JASON YEUNG, Student Volunteer, Central High School, Philadelphia, PA
NAGENDRA SRINIVAS KODALI, M.D., Volunteer (from September 1998)

Our environment is a constant source of chemical assault on the genetic material in our cells. The cells have to rid themselves of a variety of deleterious lesions in their DNA using enzymes collectively called DNA repair systems. Some of these enzymes are highly specific for distinct DNA modifications, but others are "flexible" and act on a wide range of lesions that appear to have little in common. The discovery of various flexible repair systems and the elucidation of their mechanisms have been the interest of this laboratory. Our current focus is on nucleases that make use of a common property shared by most if not all DNA lesions, that of single-strandedness. In general, the sites of DNA lesions, either bulky adducts or base pair mismatches, exhibit some properties of single-strandedness. We are interested in whether and how cells use lesion-specific single-strandedness in the mechanisms of DNA repair. Nucleases such as S1, P1 and mung bean nuclease are well known to cut DNA at regions of single-strandedness. These nucleases, however, rely on an acidic pH for function. We looked for nucleases that are active at neutral pH and produce similar or better results than at acidic pHs; the rationale is that the nuclear contents are generally neutral in pH. Furthermore, DNA is more double-stranded at neutral pH than at an acidic pH. Therefore, a nuclease that can act on single-strandedness at neutral pH would be expected to be more specific for mismatches and less prone to non-specific DNA degradation.

We designed oligonucleotide duplexes that contain base-pair mismatches and used them as substrates to search for mismatch-specific endonucleases at neutral pH in a variety of organisms. Such an activity was found by us to be prominent in the plant kingdom. For most vegetables, the activity can be detected in crude extracts. Celery was chosen as the model system for our studies because of the abundance of the mismatch endonuclease enzyme, CEL 1, the relatively low S1-like activity, and the low level of chloroplast pigments in the plant. CEL I is similar to mung bean nuclease, and exhibits some activities as an RNase, a single-stranded nuclease, and a double-stranded DNase. CEL I also requires Zn⁺⁺ for activity. Under appropriate conditions, CEL I activity can be distinguished from the activities of the S1-like mung bean nuclease in several ways. Moreover, the Mg⁺⁺ requirement for CEL I in the mismatch cutting is not observed for any of the S1 family of nucleases, including the mung bean nuclease.

CEL I nicks one strand of a mismatch heteroduplex at the phosphodiester bond on the 3' side of the mismatch nucleotide to produce an unpaired nucleotide. This nucleotide can then be removed by a DNA polymerase containing a 3' to 5' exonuclease activity. In the presence of dNTP, the DNA polymerase will perform a nick translation reaction and restore the double-strandedness of the DNA duplex. This mismatch elimination reaction in vitro is very rapid. However, by itself, it does not provide information on which strand should be used as template for mismatch removal. The mechanism may be more compatible with that of gene conversion where the retention of parental information is not the goal, therefore, strand identity is not important.

Spinach leaves contain a highly active nuclease, SP. The purified enzyme incises single-stranded DNA, RNA, and double-stranded DNA that has been destabilized by A-T rich regions and DNA lesions (Strickland et al., Biochemistry 30:9749, 1991). This broad range of activity suggested that SP may be similar to a family of nucleases represented by S1, P1, and the mung bean nuclease. However, unlike these single-stranded nucleases that require acidic pH and low ionic strength conditions, SP has a neutral pH optimum and is active over a wide range of salt concentrations. We have extended these findings and shown that an outstanding substrate for SP is a mismatched DNA duplex. For base-substitution mismatches, SP incises at all mismatches except those containing a guanine residue. SP also cuts at insertion/deletions of one or more nucleotides. Where the extrahelical DNA loop contains only one nucleotide, the preference of the extrahelical nucleotide is A>>T~C (undetectable at G). The inability of SP to cut at guanine residues and the favoring of A--T rich regions distinguishes SP from the CEL I family of neutral pH mismatch described above (1). SP, like CEL I, does not turnover after incision at a mismatched site in vitro. Like CEL I, the presence of a DNA polymerase or a DNA ligase allows SP to turn over and stimulates its activity in vitro by about 20-fold. It is possible that the SP nuclease may be a natural variant of the CEL I family of mismatch endonucleases. The properties of SP fall between those of CEL I and S1 nuclease, and it shares properties of both of these families suggesting that CEL I and S1 are distantly related.

The CEL I nuclease that we discovered (see above) is a useful new reagent for mutation detection; it is specific for DNA distortions and mismatches from pH 6 to 9. CEL I is a stable enzyme during purification, storage, and assay. We used CEL I to develop a simple method of Enzyme Mutation Detection to efficiently identify mutations and polymorphisms. To illustrate the efficacy of this approach, the exons of the BRCA1 gene were amplified by PCR using primers 5'-labeled with fluorescent dyes of two colors. The PCR products were annealed to form heteroduplexes and subjected to CEL I incision. In GeneScan analyses with a PE Applied Biosystems automated DNA sequencer, two independent incision events, one in each strand, produced truncated fragments of two colors that complement each other, confirming the position of the mismatch. CEL I can detect 100% of the sequence variants present, including deletions, insertions, and base-substitutions. Our results indicate that CEL I mutation detection is a highly sensitive method for detecting both polymorphisms and disease-causing mutations in DNA fragments as long as 1120 base pairs. We have received numerous inquiries about CEL I since publication (1). At present, over 20 laboratories worldwide have received CEL I from us and are using it in their research programs to detect mutations.

1.   OLEYKOWSKI, C.A., BRONSON MULLINS, C.R., GODWIN, A.K., YEUNG, A.T. Mutation detection using a novel plant endonuclease. Nucl. Acids Res. 26:4597-4602, 1998.

BELLACOSA, A., CICCHILLITTI, L., SCHEPIS, F., RISSIO, A., YEUNG, A.T., MATSUMOTO, Y., GOLEMIS, E.A., GENUARDI, M., NERI, G. MED1, a novel human methyl-CpG-binding endonuclease, interacts with DNA mismatch repair protein MLH1. Proc. Natl. Acad. Sci. USA 96:3969-3974.

OLEYKOWSKI, C.A., BRONSON MULLINS, C.R., CHANG, D.W., YEUNG, A.T. Incision at nucleotide insertions/deletions and basepair mismatches by the SP nuclease of spinach. Biochemistry 38:2200-2205, 1999.

U.S. Patent number 5869245. "Mismatch endonuclease and its use in identifying mutations in targeted polynucleotide strands," 1999.

^§   Fox Chase researcher

^a   C. Bronson Mullins: Present address--Route 209 and Kunkletown Road, Kresgeville, PA 18333

^b   D.W. Chang: Princeton University, Princeton, NJ 08540

Illustrations or unpublished data in these reports should not be used without permission of the author.

Fox Chase Cancer Center

Scientific Report 1998