Introduction

The eukaryotic genome encodes several thousand genes, the expression of which is coordinated by multiple highly complex genetic networks. To understand the function of these networks, it is crucial to accurately determine the temporal and spatial patterns of expression of each gene in the context of the living organism.

In the past two decades, several reporter gene systems have been developed and used to monitor gene expression in eukaryotic cells. The earliest reporter genes encoded enzymes [i.e., bacterial β-galactosidase, chloramphenicol acetyltransferase (CAT), firefly luciferase, or mammalian alkaline phosphatase], and these systems used enzymatic assays to detect and quantify reporter gene activity (1). More recently, reporter gene systems have been based on fluorescent proteins such as green fluorescent protein (GFP) (2) or fluorescent products of bacterial enzymes such as bacterial β-lactamase (bla) (3) and uroporphyrinogen III methyltransferase (cobA) (4). These systems can be used to monitor gene expression in real-time and with high spatial resolution in living cells. In some cases, such as with luciferase and GFP reporter systems, specially designed derivatives of the reporter proteins have been developed for particular research purposes (5,6). The reporter systems based on optical signaling (i.e., lucif-erase or fluorescent-tagged proteins) offer several advantages, including their noninvasive nature and that they can be used in living cells. However, optical signals only travel a limited distance through living tissue (1–2 cm) due to quenching from absorption and/or light scattering (7). Thus, currently available systems are not suitable for studying gene expression in the internal tissues of living organisms.

Recently, noninvasive reporter gene systems have been developed that overcome the distance/depth limitation associated with optical signals. These include radionuclide-based imaging techniques, such as positron emission tomography (PET) and single-photon emission tomography (SPECT), which have been used successfully to visualize gene expression in living animals (1,7). These methods are highly sensitive and are also suitable for tracking cells after organ or tissue transplants in research setting (8). However, radionuclide-based imaging of gene expression in living systems cannot provide detailed anatomic information. Furthermore, the short-lived radioactive tracer, whose synthesis and detection often involves specialized equipment and facilities, needs to be injected into the patient (or animal) prior to initiating an imaging procedure.

Magnetic resonance imaging (MRI), which has also been used to monitor gene expression, has a significantly higher spatial resolution than PET/SPECT (7). To date, genes encoding the transferrin receptor (9), β-galactosidase (10), and ferritin (11,12) have been used as reporter genes in ¹H-MRI-based reporter gene systems. Detection of the first two marker genes requires a specific contrast agent (i.e., Tf-MION and EgadMe) to detect changes in local ¹H relaxation times, T₂ and T₁, of water molecules, respectively. In contrast, in the ferritin system, T₂ is affected directly by sequestration of endogenous iron; thus, no cofactor or contrast agent is required. Simpler systems based on ³¹P- or ¹⁹F-magnetic resonance spectroscopy (MRS) have also been used to measure expression of creatine kinase (13), arginine kinase (14), and cytosine deaminase (15) reporter genes. Although three-dimensional (3-D) images using these techniques have not yet been reported, they are effective systems for quantifying gene expression in living organisms.

Recently, we reported the development of a novel MRS/MRI-based reporter gene system in Saccharomyces cerevisiae (16). This system uses VTC1 and VMA2, which encode components of the vacuolar transporter chaperone (Vtc) complex (17) and the vacuolar H⁺-ATPase (V-ATPase) (18), respectively. These two genes are required to produce and accumulate inorganic polyphosphate (polyP) in the yeast vacuole (19). polyP is a linear polymer of orthophosphate residues linked by high-energy phosphoanhydride bonds that is found in a wide range of organisms, including bacteria, fungi, plants, and animals (20) and which is readily quantified using ³¹P-MRS/MRI. When VTC1 and VMA2 were linked to several heterologous promoters of different strength, the strength of the promoter and the level of expression correlated well with the amount of polyP that accumulated in the cell (16). These results demonstrated that VTC1 and VMA2 can be used as quantitative reporter genes in yeast in the absence of an exogenous cofactor. Thus, this polyP-generating reporter gene system is a powerful, noninvasive tool for quantifying gene expression in yeast.