Introduction

We recently reported a fluorescent molecular sensor, an oscillating nano-size device, which contained a vaccinia virus encoded protein linked with a dual-labeled DNA part (1) (also featured in ref. 2). The construct exemplified a practical approach to the design of molecular devices and machines, and illustrated our notion that nano-size constructs that use mechanisms developed in the evolution of biological molecules are simpler and uniquely suitable for nanoscale environments. Here, we present this development for practical usage as an oscillating molecular probe with energy transfer capabil-

From: Methods in Molecular Biology, vol. 335: Fluorescent Energy Transfer Nucleic Acid Probes: Designs and Protocols Edited by: V. V. Didenko © Humana Press Inc., Totowa, NJ

ity and describe its application for simultaneous detection of two specific types of DNA damage in situ.

Selective imaging of specific types of DNA breaks directly in tissue sections is critical for the analysis of cellular damage and can be used for the early detection of disease and for the creation of new drugs. Therefore, several years ago we introduced an in situ ligation approach for selective detection of double strand (ds) breaks in cellular DNA (3-5). It relies on attachment of ds DNA probes with blunt ends, or short 3' overhangs, to the ends of ds DNA breaks bearing 3'OH/5'PO4, such as those produced by DNase I. The ligation reaction occurs in tissue section (3-5) or in live cells in culture (6), and is carried out by the enzyme T4 DNA ligase. The probes used in the reaction can be polymerase chain reaction (PCR)-labeled (3) or synthesized as hairpin-shaped oligonucleotides (4,5). This assay detects exclusively 5' phosphorylated ds breaks, because ligase needs a terminal 5'PO4 in cellular DNA to attach the probe. An important example of the nuclease that can produce such breaks is CAD/ DFF40 (7).

However, breaks in cellular DNA can possess different terminal configurations. Another specific type is represented by ds breaks bearing 5'OH, which are generated by the ubiquitous DNase Il-type lysosomal nucleases (8). DNase II-like nucleases play a role in fundamental biological phenomena such as apoptosis, DNA catabolism, and drug-induced DNA cleavage (8-13). Yet, no ligase can attach the 5'OH end of genomic DNA to the 3'OH end of the probe (14). Attempting to expand the ligation assay to detect 5'OH-bearing breaks, by just adding 5'-phosphates to the probe, will not result in labeling of this type of DNA damage, due to probe attachment to 3'OH ends and probe self-ligation.

To resolve this problem, the T4 DNA kinase-based technique was introduced for the detection of 5'OH-bearing DNA breaks (15). However, in this case, the 5'PO4 breaks must be labeled first so that in the subsequent reaction with T4 DNA kinase all remaining 5'OH breaks are converted to the detectable 5' phosphate format. As a result, the approach requires two successive overnight labeling reactions with multiple controls, making it complicated and time-consuming.

We have, therefore, developed a new approach for fast and reliable visualization of 5'OH ds DNA breaks. The approach uses vaccinia DNA topoisomerase I and a double-hairpin oligonucleotide probe. It is applicable for the detection of DNA breaks in solution and in tissue sections.

The combination of vaccinia DNA topoisomerase I with a double-hairpin oligonucleotide, labeled with two fluorophores that form a donor-acceptor pair, creates an oscillating probe capable of fluorescence resonance energy transfer.

The oscillating probe contains CCCTT3' vaccinia topoisomerase I recognition sequence, located adjacent to the nick formed by the probe's folded 3'- and 5'-ends

(Fig. 1). After mixing the oligonucleotides with vaccinia topoisomerase I, the topoisomerase molecules bind to the oligos and cleave them at the 3'-end of the CCCTT3' recognition sequence (16). This results in cutting of the probe into two blunt-ended hairpins, which will dissociate from each other (Fig. 1). The cleaved phosphodiester bond energy is conserved by formation of a covalent link between the 3' phosphate of the CCCTT3'-carrying hairpin and a tyrosyl residue of the enzyme (Tyr-274) (17). The free energy gain for the breakage reaction is small, on the order of+1 kcal/mol, making the reaction freely reversible (18). The enzyme, which remains bound to the CCCTT motif located on the downstream hairpin, will religate it back to the end of the upstream hairpin. This leads to cycles of self-attachment-disattachment of the hairpins, which are repeatedly religated and then recut by topoisomerase when thermal motion randomly separates and brings them together. In the presence of another acceptor, such as a blunt-ended DNA break with 5'OH, the topoisomerase-carrying hairpin will ligate to it instead and will fluorescently label it.

The assay places a single FITC fluorophore at the end of each DNA break. Therefore, the intensity of fluorescence is directly proportional to the number of breaks. Observation of the results of this labeling reaction is well within the limits of the regular fluorescent microscope. Using a dot spot test we verified that a nonconfocal optical system (Olympus IX-70 microscope with MicroMax digital videocamera) can visualize 1.25 fmol of FITC spotted as a 1.3-mm dot, which corresponds to approx 45,000 FITC molecules per the surface area occupied by a nucleus 0.01 mm in diameter (1). This sensitivity is sufficient for visualization of individual apoptotic cells in all stages of programmed cell death because the number of breaks in apoptosis rises from approx 50,000 per genome, at the initial high molecular weight DNA degradation, to 3 x 106 during internucleosomal DNA fragmentation stage (19) (see Note 1).

Although the topoisomerase-based assay can be used on its own, it can also be combined with ligase-mediated labeling. When T4 DNA ligase is added to the reaction mix, the second hairpin, produced by the split of the double-hairpin probe, can be simultaneously ligated by DNA ligase to breaks with DNase I architecture, bearing 5'PO4 groups (Fig. 1). The combined assay, using both topoisomerase and ligase, is especially informative in the tissue section format because cells with different types of DNA damage can be simultaneously visualized by different fluorophores.

When tested in tissue sections of dexamethazone-treated rat thymus, the assay successfully detected both the primary DNase I-like cleavage in apoptotic thymocyte nuclei and DNase II-like breaks in the cytoplasm of cortical macrophages ingesting apoptotic cells (1).

The in situ application of oscillating double-hairpin probe and topoisomerase-ligase labeling, visualizes both DNase I- and II-type breaks even

Fig. 1. Principle of the assay for DNA damage detection in situ using oscillating probe, vaccinia topoisomerase I, and T4 DNA ligase (see Heading 1.). Adapted with permission from ref. 1. (This figure also appears on the Companion CD.)

in small numbers of cells, thus, increasing the sensitivity of DNA damage detection (see Note 2).

In this chapter, we present complete protocols for topoisomerase and combined topoisomerase-ligase-based detection applicable for fixed tissue sections.

0 0

Post a comment