The use of real-time in vivo systems to detect the presence of chemical contaminants or agriculturally important biotic/abiotic stressors provides an imperative first line of protection for environmental- and plant health [1–5]. Sentinels utilizing plants genetically engineered to produce a specific signal upon induction by an agent is envisioned as one phytosensor platform [6–8]. Engineered phytosensors could contain contaminant- or biotic/abiotic-inducible promoters driving the expression of a reporter gene, whose presence is, in turn, detected by visual observation or some device [9]. When such transgenic plants encounter the agent, the specific inducible promoters are triggered and subsequently drive the expression of the reporter gene producing a signal for detection.

Plant sentinels, or ‘phytosensors’, potentially have tremendous utility as wide-area detectors. The use of an inducible promoter is a desirable feature when contemplating its use for the design of a phytosensing system. However, the complexity of the expression of many plant promoters is inherent to the fact that they contain several cis-regulatory elements, each of which can be activated or repressed by more than one agent. To avoid this limitation, synthetic promoters can be designed based on selected cis-regulatory elements [9]. Yet, from our previous results [10,11], one major problem associated with employing either native or synthetic inducible promoters directly fused to a reporter gene is the lack of sufficient expression of the reporter, i.e., the inducible promoter is not strong enough to produce robust signals upon full induction—the most fundamental feature that is expected to result when induced by the agent in a phytosensing system. Although increasing basal level of the reporter expression by addition of enhancer elements in the synthetic inducible promoter constructs was achievable [11], detection of a false reporter signal from high-level background of reporter expression prior to the real-time induction in response to the agent would be considered as a potential drawback.

To address the problem associated with the inducible promoters directly fused to a reporter in phytosensing systems, we expanded the phytosensing platforms to include a site-specific recombination strategy. Site-specific recombination involves enzyme mediated rearrangement of DNA fragments that do not possess a high degree of homology [12,13]. Conservative site-specific recombination (CSSR) and transposition are two classes of site-specific recombination [13,14]. CSSR involves exchange or recombination at highly specific regions within short stretches (recombination sites) of identical sequences in the participating DNA fragments, while transposition does not require any homology between the recombination sites [13]. CSSR can result in different DNA rearrangements depending on the relative orientation of the recombination sites, in cis and oriented in the same direction results in deletion of the DNA fragment contained within the two sites, while recombination sites in cis and in the opposite orientation results in inversion of the DNA fragment [12,13]. On the other hand, having recombination sites in trans on two linear DNA molecules results in exchange of DNA fragments. If one of the DNA molecules involved in the reaction is circular, recombination results in a cointegration event, but this event is kinetically less favorable and less likely to occur [12]. There are several CSSR systems identified and shown to be functional in higher eukaryotes. These simple and efficient CSSR systems can have a wide variety of applications in plant biotechnology. Some of these applications include excision of selectable marker genes from transgenic plants, excision of redundant copies of transgene in crop plants to reduce the extensive screening required to obtain single-copy transgenic lines, and site-specific integration of transgenes [15–30].

Site-specific recombination systems have been used primarily for the deletion of transgenes resulting in antibiotic-free transgenic plants and/or for direct integration of transgenes into plant genomes. In this study, we propose possible application of this system for phytosensing purposes. We employed one of the well-characterized site-specific recombination systems—the FLP/FRT from Saccharomyces cerevisiae, where FLP (flipping DNA) is the recombinase which recognizes FRT (FLP recombination target) sites [12,21]. We used an inducible heat-shock promoter to produce a system in phytosensing using both tobacco and Arabidopsis as phytosensor models. We used a two gene cassette system in which the inducible heat-shock promoter, instead of controlling the expression of the GUS reporter gene directly, regulates the expression of the FLP recombinase gene, so that once induced by heat, the recombination would result in the excision of the DNA fragment between the FRT recognition sites, then placing the CaMV 35S promoter in close proximity of the reporter gene. This (in case of a weak inducible promoter driving the recombination) would lead to efficient detection of the signal (Figure 1). Our results provide additional insights into the possible application of site-specific recombination system as a tool for the reporter signal in phytosensing.

A two gene cassette scheme of an inducible FLP/FRT recombination system was examined in tobacco and Arabidopsis plants. A plant expression vector containing the inducible FLP/FRT recombination system was constructed and the utility of the system was examined by employing a heat-shock promoter for induction of the system. In this two gene cassette system, the FLP recombinase gene was driven by soybean heat-shock inducible promoter where the expression of the FLP gene upon heat-induction would lead to excision of the DNA fragment between the FRT sites. This subsequent recombination event would place the CaMV 35S promoter within close proximity with the GUS gene thereby resulting in an active expression of the GUS reporter (Figure 2). Histochemical analysis of GUS activity for tobacco (Figure 3A) and Arabidopsis (Figure 3B) showed that the heat-shock treatment successfully excised the DNA fragment between the FRT recognition sites by driving the expression of the FLP recombinase gene further resulting in the CaMV 35S promoter driven expression of the GUS reporter gene in a predictable fashion. The expression of GUS in the transgenic plants that contained the recombination system was comparable to the positive control where the GUS expression was driven by CaMV 35S in a direct fusion system. A total of six lines out of 11 independent tobacco lines and six lines out of 14 independent Arabidopsis lines carrying the recombination system exhibited GUS activity upon heat induction. These plants displayed a uniform pattern of GUS expression with a range of medium to high GUS activity relative to the control (data not shown). We also observed GUS activity in a few of the uninduced counterparts in both tobacco and Arabidopsis plants. The leakiness of heat-shock promoters has been also observed in other studies [34,35].

For further verification, the recombination event upon heat induction was analyzed by PCR. A successful recombination event in a heat-treated plant would result in a 931 bp PCR fragment whereas in an untreated plant would result in a 4,165 bp PCR fragment. As shown in Figure 4, the 931 bp PCR fragment was amplified for the heat-induced plants confirming the excision of the DNA fragment between the two FRT sites whereas the 4,165 bp PCR fragment was amplified for the untreated plants. Induction was analyzed by PCR in all the independent tobacco and Arabidopsis lines exhibited GUS activity upon heat induction and similar results were obtained (data not shown).

Taken together, our results demonstrate that this two gene cassette organization of the inducible FLP/FRT recombination system is functional in tobacco and Arabidopsis plants, suggesting its possible application for reporter signal manipulation in phytosensing.

Phytosensors could serve as a valuable and a cost-effective method for sensing environmental contaminants in real time. An inherent problem associated with this application is the inadequate level of reporter signal for detection, which depends, at least in large part, on the strength of the inducible promoters involved in phytosensing. In this study, we demonstrated that a two gene cassette scheme of the inducible FLP/FRT recombination system is functional in tobacco and Arabidopsis. This system can be used as a potential tool for the reporter gene signal from phytosensor plants when the inducible promoters may provide sensing capabilities but are not sufficiently strong to produce robust signals needed for phytosensing.