The recent development of caged oligonucleotides that are efficiently activated by ultraviolet (UV) light creates opportunities for regulating gene expression with very high spatial and temporal resolution. By selectively modulating gene activity, these photochemical tools will facilitate efforts to elucidate gene function and may eventually serve therapeutic aims. We demonstrate how the incorporation of a photocleavable blocking group within a DNA duplex can transiently arrest DNA polymerase activity. Indeed, caged oligonucleotides make it possible to control many different protein-oligonucleotide interactions. In related experiments, hybridization of a reverse complementary (antisense) oligodeoxynucleotide to target mRNA can inhibit translation by recruiting endogenous RNases or sterically blocking the ribosome. Our laboratory recently synthesized caged antisense oligonucleotides composed of phosphorothioated DNA or peptide nucleic acid (PNA). The antisense oligonucleotide, which was attached to a complementary blocking oligonucleotide strand by a photocleavable linker, was blocked from binding target mRNA. This provided a useful method for photomodulating hybridization of the antisense strand to target mRNA. Caged DNA and PNA oligonucleotides have proven effective at photoregulating gene expression in cells and zebrafish embryos.
A typical vertebrate animal contains approximately twenty thousand unique protein-coding genes (1). This is a small number when one considers the enormity of the tasks required for creating, growing, and sustaining life. To visualize this complexity, imagine a town with 20,000 human inhabitants, each person representing a different gene. Communication occurs between inhabitants and with others in surrounding communities. These signals help to coordinate activities at different locations and times. Through proper organization, the town is able to thrive and create remarkable structures. The existence today of giant pyramids and cathedrals attests to the power of synchronous, well-coordinated human activity.
Returning to gene expression in the vertebrate animal—it is frustratingly difficult to visualize the complex network of signals that is orchestrated at the molecular level, both intra- and extracellularly. The spatial and temporal role of each protein is important, and protein functions can vary by concentration, cell localization, and activation state (i.e., posttranslational modification). Based on the large number of interacting molecules and their complex environment, it is virtually impossible at the present time to identify all of the interacting partners (i.e., protein-protein interactions) for a given cellular process. We still know relatively little at the molecular genetic level about how cells divide, how cell types are specified and are sometimes re-specified at different stages of life, and how animals achieve and maintain a predetermined size. And, it appears likely that the focus in biochemistry on protein structure and function is somewhat misplaced, considering that of the three billion DNA base pairs in the human genome, most of which are conserved in vertebrates, only 1.5% of the total length serves as protein-coding exons (2). Thus, oligonucleotide-based tools that can probe the function of regulatory sequences, repeat elements, transposons, pseudogenes, junk DNA and RNA, as well as protein-coding sequences, will provide myriad opportunities for pure and applied discovery. In particular, new methods for controlling gene activity quantitatively and with high spatial and temporal resolution will be critical to improving our molecular understanding of many life processes. In this article, we highlight the advent of photochemical methods that will provide unique opportunities for investigating complex biological systems (Figure 1).
The past decade has seen the development of many techniques for controlling genes in cells and animals (3). Gain-of-function is achieved by introducing an exogenous gene into the animal, typically as the fertilized zygote. In contrast, alteration of endogenous genes leading to loss-of-function arises from gene deletion, for example, via recombination (4), mutagenesis, enzymatic degradation of mRNA, as via RNases (5) or Slicer (6), and blockage of ribosomal translation using reverse complementary (anti-sense) oligonucleotides (7,8,9). Transgenic animals, thus produced, often show new phenotypes and patterns of gene expression and can identify in vivo functions of proteins. These data can also be assembled to dissect gene regulatory networks. Unfortunately, many genes cannot be probed effectively in this fashion, as gene deletions are often lethal or produce a compromised transgenic animal. A final challenge is that many genes are co-opted to perform multiple functions: thus, complete gene knockdown throughout an entire organism and for long periods of time (for hours, or extending throughout the lifetime of the organism) will mask layers of genomic complexity.
To address these issues, methods for conditionally perturbing gene activity have gained increasing attention. Several approaches, sometimes used in tandem, deserve mention. These include diverse methods for focally delivering genetic material, such as by electro- or laser poration (10); the use of promoters with known temporal and tissue-specific expression profiles; diffusion of bioactive small molecules, which are typically obtained by screening large compound libraries (11); tetracycline-inducible constructs (12); Cre-loxP recombinase, which can be made inducible by heat-shock (13) or small molecules such as doxycycline, aldo-sterone, and tamoxifen (14); and light-activated (caged) compounds (3). We address the last method here, of which there are many examples: caged neurotransmitters (15,16), hormones (17,18), amino acids (19,20,21,22), peptides (23,24), proteins (23,25,26,27,28,29,30,31,32), calcium (33) and other second messengers such as cAMP (34,35,36,37), and nitric oxide (NO) (38,39). In the caged state, these molecules are transiently blocked from performing their biological functions. The advent of caged cAMP and caged adenosine triphosphate (ATP) 30 years ago propelled neurobiology in exciting new directions. Unfortunately, it proved much more difficult to modulate the activity of oligonucleotides using a single photocleavable blocking agent. Our laboratory has focused on extending sensitive caging strategies to DNA, RNA, and peptide nucleic acids (PNAs) in order to photoregulate gene expression in living cells and embryos.