Introduction

In situ hybridization techniques, and in particular its fluorescence-based variants, are routinely applied in a wide range of disciplines including genetics, developmental biology, pathology, and cell biology to study genome composition or gene expression in cells. This is primarily possible due to the many improvements in probes, labels, hybridization protocols, and microscope systems that have been realized since the introduction of the in situ hybridization technique in 1969. Since then, in situ hybridization studies have led to important advances in our understanding of the organization and composition of the genome and its aberrations as well as of the expression and localization of gene transcripts. One should realize, however, that in situ hybridization approaches have been developed mainly for the detection of nucleic acids in fixed, morphologically preserved specimens and consequently provide static rather than dynamic views on nucleic acid localization. Furthermore, specimen preparation and nucleic acid denaturation may cause redistributions or loss of target nucleic acid sequences, and these artifacts may hamper the interpretation of in situ hybridization data (1,2,3). The most important reason, however, for implementing live-cell imaging techniques in current research was to obtain a better understanding of complex cellular processes, including chromatin organization and transcription regulation. Some of these techniques have been developed to enable the overall in vivo labeling of chromatin or RNAs to obtain a more general view on nucleic acid behavior within a living cell. Other techniques, however, were designed to specifically label defined chromatin regions or particular RNA species.

To detect and track specific endogenous RNAs in a living cell, in vivo hybridization-based techniques have been developed that make use of a variety of different nucleic acid probe types and fluorescent detection methods. In essence, all these techniques are developed to pursue optimal detection sensitivity and specificity. Particularly over the past few years, we have witnessed the development of various probe types showing improved affinity and specificity for target sequences and resistance to cellular nucleases. At the same time, it became clear that an efficient delivery of probes into living cells is also an essential step in the visualization procedure. Most methods that pursue this have been previously described (4) and are not discussed here in detail.

Parallel to ongoing developments in nucleic acid-based probe technologies, approaches have been developed that take advantage of fluorescent proteins that specifically interact with DNA or RNA sequences. These approaches have the advantage that the fluorescent (fusion) proteins are made by the cell's own transcription and translation machinery, precluding the need for invasive techniques that may have an impact on the physiology of the cell. This review focuses on the various methods that are used to image the localization and mobility of nucleic acids in living cells, addresses technical limitations, and provides an outlook for future developments.

In Vivo Hybridization

DNA or RNA molecules can be visualized in living cells by incorporating fluorescent nucleotides using the cell's own replication or transcription machinery (5,6,7) or by binding DNA-or RNA-associating fluorescent dyes such as the membrane-permeable dyes Cyto 14 (8), dihydroethidium (9), and DRAQ5™ (10). These methods result in an overall labeling of cellular DNA or RNA in vivo, but do not allow for the detection of defined DNA sequences or specific RNA molecules in living cells. Fluorescence in vivo hybridization has long been the method of choice to detect specific endogenous RNA species in living cells (4,11). This is technically possible because RNAs are, at least partly, single-stranded molecules. Detection of DNA sequences by fluorescence in vivo hybridization requires denaturation of the double-stranded DNA molecule, which is not compatible with live-cell studies. To date, a plethora of different probe types and detection concepts have been used to visualize RNAs in living cells.

Linear Phosphodiester Oligodeoxynucleotide Probes

Phosphodiester oligodeoxynucleo-tides (ODNs) can be easily synthesized and fluorescently labeled, are inexpensive, and hybridize specifically to complementary RNA target sequences in vitro. For these reasons, ODNs were the first to be employed for the detection and tracking of RNA species in living cells. Indeed, it was shown that fluorescein-labeled ODNs are efficiently taken up by cells and delivered to the cell nucleus (12). These observations were followed by pioneering studies by Pedersons’ group with regard to the visualization of endogenous poly(A) messenger RNA (mRNA) movement within living cells using ODNs (13). The question to be answered was how the poly(A) RNAs would travel through the cell nucleus before they were released into the cytoplasm of cells. By applying fluorescence correlation spectroscopy and fluorescence recovery after photobleaching (FRAP) measurements, it was suggested that a vast amount of poly(A) RNA molecules moves randomly throughout the nucleus without the need for energy consumption (13). In subsequent studies, these findings were confirmed using similar probes, labeled with a nonfluorescent caged fluorescein that fluoresced upon irradiation (14,15). Furthermore, these studies suggested that poly(A) RNA moves through interchromatin channels, avoiding chromatin dense regions. Together, these observations indicated that mRNAs do not travel along a directed pathway from their site of synthesis toward the cytoplasm. Instead, it was suggested that it is only by the process of diffusion, thus a matter of change, that an mRNA molecule reaches a nuclear pore, after which it can pass through the nuclear membrane and enter the cytoplasm of a cell.