The regulation of cellular gene expression via activation and inactivation of a network of signal transduction pathways enables cells to grow, differentiate, and communicate with each other and their surroundings, as well as to react to other changes in their environment [1Stathopoulos A, Levine M. Genomic regulatory networks and animal development Dev Cell 2005; 9(4): 449-62.].

These crucial events in cell biology are of great interest for many types of cell studies. For example, in the case of stem cell-related research a specific set of transcription factors, Oct4, Sox2, c-myc, and Klf4, has been found to be sufficient to reprogram fibroblasts into pluripotent-like stem cells [2Yu J, Vodyanik MA, Smuga-Otto K, et al. Induced pluripotent stem cell lines derived from human somatic cells Science 2007; 318(5858): 1917-20., 3Takahashi K, Tanabe K, Ohnuki M, et al. Induction of pluripotent stem cells from adult human fibroblasts by defined factors Cell 2007; 131(5): 861-72.]. These results emphasize the powerful effect of promoter regulation via transcription factors.

In order to enable further discoveries in the field of signal transduction, it is crucial for cell biologists to be able to monitor promoter activation via a reliable, easy to measure, and consistent reporter system. To date, a variety of reporter systems have been developed, all of which rely on the expression of a reporter-encoding gene driven by the promoter of interest. Reporter detection can then be correlated with promoter activation.

The reporter itself, and its method of detection, vary between the different assays. Reporter activity is typically measured either by the fluorescence intensity of the reporter itself or by detection of a specific enzymatic activity of the reporter. Examples of fluorescent reporters include Aequorea victoria-based GFP variants [4Patterson GH, Knobel SM, Sharif WD, Kain SR, Piston DW. Use of the green fluorescent protein and its mutants in quantitative fluorescence microscopy Biophys J 1997; 73(5): 2782-90.], Aequorea coerulescens-based AcGFP [5Gurskaya NG, Fradkov AF, Pounkova NI, et al. A colourless green fluorescent protein homologue from the non-fluorescent hydromedusa Aequorea coerulescens and its fluorescent mutants Biochem J 2003; 373(Pt 2): 403-8.], and Anthozoa-based fluorescent reporters including ZsGreen and DsRed variants [6Matz MV, Fradkov AF, Labas YA, et al. Fluorescent proteins from nonbioluminescent Anthozoa species Nat Biotechnol 1999; 17(10): 969-73.]. One advantage of fluorescent protein reporters is that detection does not rely on any cofactors or exogenous substrates. Rather, the fluorescent signal results from the intrinsic chromophore structure. This enables researchers to design completely homogeneous assays to monitor gene expression continuously, on a cell by cell basis.

There are two potential disadvantages to using fluorescent reporters. The first is that unfortunately, most fluorescent proteins have slow turnover rates, which cause the reporter to accumulate over time. This is problematic if the goal is to monitor fast activation and inactivation of a promoter of interest. Destabilized fluorescent proteins, which have a drastically accelerated turnover due to a proteasome targeting sequence fused to the fluorescent reporter protein [7Li X, Zhao X, Fang Y, et al. Generation of destabilized green fluorescent protein as a transcription reporter J Biol Chem 1998; 273(52): 34970-5.], have been developed to address this problem. These destabilized fluorescent proteins have a shorter life span and are indeed able to reflect fast up- and down-regulation of the promoter. A second potential disadvantage of fluorescent reporters is that the signal generated by fluorescent proteins is not based on an enzymatic, amplifying reaction. Therefore, the sensitivity of these assays is lower compared to enzyme based reporter assays.

The most commonly used enzyme based assays are bioluminescence based luciferase assays. Luciferases are oxygenases; that is, they use oxygen together with their respective substrate to produce a bioluminescent signal [8Hastings JW. Chemistries and colors of bioluminescent reactions: a review Gene 1996; 173(1 Spec No): 5-11., 9Wilson T, Hastings JW. Bioluminescence Annu Rev Cell Dev Biol 1998; 14: 197-230.].

Cytosolic luciferase reporters such as Firefly and Renilla luciferase fall into two classes. The first class, which includes Firefly luciferase, requires both luciferin and ATP as substrates in order to produce a luminescent signal. It has been used for a variety of assay applications [10Sala-Newby GB, Campbell AK. Expression of recombinant firefly luciferase in prokaryotic and eukaryotic cells Biochem Soc Trans 1992; 20(2): 143S.], including assays designed to detect cytosolic ATP levels (e.g., apoptosis and cytotoxicity assays) [11Corey MJ, Kinders RJ, Brown LG, Vessella RL. A very sensitive coupled luminescent assay for cytotoxicity and complement-mediated lysis J Immunol Methods 1997; 207(1): 43-51.]. The second class, which includes Renilla luciferase and several other luciferases, use coelenterazine as their substrate and do not require ATP [12Lorenz WW, McCann RO, Longiaru M, Cormier MJ. Isolation and expression of a cDNA encoding Renilla reniformis luciferase Proc Natl Acad Sci USA 1991; 88(10): 4438-2.].

Both classes of luciferase are very sensitive and have broad dynamic ranges. However, the cells must be lysed in order to allow the substrate to access these cytosolic luciferase reporters. Therefore, unlike fluorescent proteins, cytosolic luciferases do not permit the development of live cell assays and the researcher is limited to a single data point per experiment. This is a serious disadvantage, especially if access to specific cells is limited or if long term monitoring of the same cells is the main purpose of the experiment (as in time course, differentiation, or toxicity studies). However, some new membrane permeable luciferase substrates have been developed, which allow the detection of luciferase activity inside the cell without the requirement for cell lysis.

An additional drawback is that the bioluminescent signal detected from luciferases only indicates the total quantity of reporter expressed by the cell population as a whole. It does not reveal if the majority of sampled cells are expressing the reporter, or only a subset.

Previous attempts to solve this problem have included creating secreted variants of Renilla luciferase tagged with a secretion signal peptide sequence. Renilla activity was detectable in the media supernatant; however, the specific activity of the secreted form of Renilla luciferase was approximately 10-fold less than that of cytosolic Renilla luciferase when measured in mammalian cells. This may be due in part to cysteine residues in the Renilla luciferase sequence which can be oxidized in the secretory pathway environment, compromising the enzymatic activity of the secreted protein. The detection limit for secreted Renilla luciferase in cell media was 100 fg [13Liu J, O'Kane DJ, Escher A. Secretion of functional Renilla reniformis luciferase by mammalian cells Gene 1997; 203(2): 141-8.].

Another approach is to avoid this problem by using naturally secreted proteins which are inherently optimized for proper folding and structure in the oxidizing environment of the secretory pathway. One such example is secreted alkaline phosphatase (SEAP) [14Berger J, Hauber J, Hauber R, Geiger R, Cullen BR. Secreted placental alkaline phosphatase: a powerful new quantitative indicator of gene expression in eukaryotic cells Gene 1988; 66(1): 1-10.]. This protein has already been successfully used as a promoter reporter in a variety of applications [15Zhu Y, Yao J, Meng Y, et al. Profiling of functional phosphodiesterase in mesangial cells using a CRE-SEAP-based reporting system Br J Pharmacol 2006; 148(6): 833-44., 16An J, Xu QZ, Sui JL, Bai B, Zhou PK. Downregulation of c-myc protein by siRNA-mediated silencing of DNA-PKcs in HeLa cells Int J Cancer 2005; 117(4): 531-7.] including effective use in transgenic animals for long-term monitoring of ovarian tumor growth and response to therapeutics [17Nilsson EE, Westfall SD, McDonald C, Lison T, Sadler-Riggleman I, Skinner MK. An in vivo mouse reporter gene (human secreted alkaline phosphatase) model to monitor ovarian tumor growth and response to therapeutics Cancer Chemother Pharmacol 2002; 49(2): 93-100.].

SEAP does have disadvantages, in that most cell culture media samples and blood samples contain naturally occurring serum alkaline phosphatase which can interfere with the detection of the reporter molecule SEAP. Fortunately, SEAP is heat resistant, so samples can be heat treated to inactivate serum alkaline phosphatase before the actual reporter SEAP is measured. However, the requirement for a heat inactivation step limits throughput and can be especially problematic for screening assays. The heat inactivation step can be avoided by using serum free medium. This allows immediate measurement of SEAP activity in the sample without the requirement for heat inactivation.

More recently, naturally secreted luciferases have been cloned and characterized. Since there is no luciferase naturally present in serum, these secreted luciferases can be measured immediately after substrate addition, without an intervening heat inactivation step.

Secreted Gaussia luciferase (Gaussia princeps) [18Tannous BA, Kim DE, Fernandez JL, Weissleder R, Breakefield XO. Codon-optimized Gaussia luciferase cDNA for mammalian gene expression in culture and in vivo Mol Ther 2005; 11(3): 435-.] and secreted Metridia luciferase (Metridia longa) [19Markova SV, Golz S, Frank LA, Kalthof B, Vysotski ES. Cloning and expression of cDNA for a luciferase from the marine copepod Metridia longa. A novel secreted bioluminescent reporter enzyme J Biol Chem 2004; 279(5): 3212-7.] have been cloned from members of the same family of Crustacea. Both use coelenterazine as their substrate. Gaussia luciferase has been shown to emit light with a flash-like bioluminescence characteristic [18Tannous BA, Kim DE, Fernandez JL, Weissleder R, Breakefield XO. Codon-optimized Gaussia luciferase cDNA for mammalian gene expression in culture and in vivo Mol Ther 2005; 11(3): 435-.]. Metridia luciferase has a more “glow-like” bioluminescence characteristic. Its signal stability is about 10 times greater than that of Gaussia luciferase (data not shown).

All secreted reporter have the intrinsic advantage of not relying on a fast turnover rate, whereas cytosolic reporters do, in order to guarantee a low background. Any secreted reporter that accumulates in the media before the actual experiment, for example due to promoter leakage, can easily be removed by changing the media just before starting the actual experiment. This can not be done with cytosolic reporters. Secreted reporter actually benefit from their stability, because a larger amount of the reporter can accumulate in the media over the course of the experiment ensuring a larger dynamic range.

Secreted reporters enable applications including time course experiments, long term monitoring during differentiation, repeated induction, and other experiments which cannot be done in an efficient manner using cell lysis-based assays. Multiplexing secreted reporters, like Metridia luciferase and SEAP, offers additional advantages, such as the ability to monitor two signaling pathways simultaneously or to use one of the two reporters as a control for expression and the integrity of the secretory pathway. This multiplex capability is especially important when using secreted reporters to screen chemical compound libraries.

However, even though a secreted reporter system allows for live cell monitoring, it is still limited to use as a cell population assay. Therefore combining a live-cell population assay using a secreted reporter for quick and easy assay sampling, with a live cell by cell assay which uses fluorescent proteins as described above to capture more precise, cell-based data points, can be an even more powerful method.

This study provides data which establishes secreted Metridia luciferase as a very useful reporter for single reporter assays and as a multiplex partner with another secreted reporter, SEAP. In addition, data is presented to support a new strategy for developing multiplex assays which combine the advantages of a bioluminescent secreted reporter system with those of a fluorescent protein based “cell by cell” assay.

This combination helps to close the gap between High throughput screening (HTS) and high content screening (HCS) approaches.

The data presented here demonstrate new opportunities in live-cell assay development, by combining secreted bioluminescent reporter assays with each other as well as by multiplexing them with fluorescent protein reporter assays using multi-mode plate readers.

With a very low limit of detection—as little as 2 fg of recombinant protein per well of a 96-well plate—and a broad linear range of 6 logs, the enzymatic characteristics of Metridia luciferase are comparable with those of other, non-secreted luciferase reporters as well as with another secreted chemiluminescent reporter, SEAP [26Bronstein I, Martin CS, Fortin JJ, Olesen CE, Voyta JC. Chemiluminescence: sensitive detection technology for reporter gene assays Clin Chem 1996; 42(9): 1542-6.].

We have shown that the utility of Metridia luciferase as a secreted reporter is not compromised by the fact that it must pass through the secretory pathway in order to be secreted outside the cell. The protein is secreted efficiently and does not accumulate within the secretory pathway.

The ease of detecting a reporter via direct addition of substrate to the media supernatant of cells, without sacrificing the cells themselves is just one advantage of secreted reporter systems. An even greater advantage lies in the fact that this system can be multiplexed with other live cell assays, including other secreted reporter assays such as SEAP and even fluorescent protein assays.

This is particularly useful in screening applications, where compounds in a chemical library may interfere with the secretory pathway. This could lead to the identification of a false positive compound because the change in the detectable amount of Metridia luciferase in the media is not caused by an effect of the compound on the promoter of interest, but by targeting the secretory pathway itself. Using a cotransfected, constitutively expressed secreted reporter such as SEAP could act as a very important control for the proper function of the secretory pathway.

As we have shown here, it is possible to dissect two signal transduction pathways from each other by using the two secreted bioluminescent reporters Metridia luciferase and SEAP. We have clearly demonstrated that two signal transduction pathways, when induced by TNF-α and Forskolin respectively, are independent from each other. Although this result could also have been obtained using cytosolic bioluminescent reporter systems, the cell lysis required to gain access to the intracellular reporter would have made it impossible to obtain any additional data sets that required live cells or to perform long term monitoring of the identical cell population.

Toxicity assays are another example of a field where there would be a benefit to repeated data acquisition from live cells. Using a secreted reporter would make it possible to monitor the toxic effects of compounds or treatments over a prolonged exposure time. It would also allow the monitoring of potential drug interactions over time, dependent on the time as well as the order in which the drugs were administered.

We have also demonstrated that using a secreted reporter system allows for other, more innovative multiplexing applications by combining bioluminescent assays with fluorescent protein based assays utilizing more advanced, new generation plate readers. These plate readers offer the ability to simultaneously detect absorbance, fluorescence, luminescence, and other readouts.

We were able to combine secreted Metridia luciferase as a reporter for TNF-α induced NFκB activation, with the HEK-293 fluorescent protein based Proteasome Sensor cell line [23Paddison PJ, Silva JM, Conklin DS, et al. A resource for large-scale RNA-interference-based screens in mammals Nature 2004; 428(6981): 427-31., 27Rickardson L, Wickstrom M, Larsson R, Lovborg H. Image-based screening for the identification of novel proteasome inhibitors J Biomol Screen 2007; 12(2): 203-10.]. Combining these assays made it possible to prove the requirement for proteasomal activity (monitored by an increase or decrease in green fluorescence intensity) in TNF-α dependent activation of the NFκB pathway, as described by Haas et al. and Sun et al. [24Haas M, Page S, Page M, et al. Effect of proteasome inhibitors on monocytic IkappaB-alpha and -beta depletion, NF-kappaB activation, and cytokine production J Leukoc Biol 1998; 63(3): 395-404., 25Sun L, Carpenter G. Epidermal growth factor activation of NF-kappaB is mediated through IkappaBalpha degradation and intracellular free calcium Oncogene 1998; 16(16): 2095-102.]. It is important to note that the results from this experiment were obtained from the same well by simply changing between fluorescence and luminescence detection modes on the plate reader, increasing the amount of data that could be obtained simultaneously. In addition, the format of such a multiplexed assay saves a considerable amount of time and reduces the possibility of error due to additional handling.

Multiplexing secreted bioluminscent assays and fluoresecent protein based assays, particularly translocation assays, allows one to combine a “cell population” assay (for example, a secreted luciferase promoter reporter assay) with a “cell by cell” assay (for example, a fluorescent fusion protein translocation assay). In the alternative scenario, where the cells must be lysed in order to access to a cytosolic “cell population” reporter, the information that could have been gained from a fluorescent protein based translocation assay is lost.

Using a secreted reporter molecule as the multiplex partner eliminates this problem and makes it possible to combine high throughput screening assays with high content screening assays, all in the same well. First, a fast screen of all wells for the secreted bioluminescent reporter could identify potential hits. Then only the wells for those potential hits would need to be analyzed for more detailed information based on the high content screen. This would allow more efficient and detailed data mining of these wells, without compromising throughput.

We anticipate that the interest in live cell assays and their multiplexing capabilities will increase, due to the growing scientific interest in primary cells and stem cells. These cells are often only available in limited numbers [28Chase LG, Firpo MT. Development of serum-free culture systems for human embryonic stem cells Curr Opin Chem Biol 2007; 11(4): 367-72., 29Lei T, Jacob S, Ajil-Zaraa I, et al. Xeno-free derivation and culture of human embryonic stem cells: current status, problems and challenges Cell Res 2007; 17(8): 682-8.], due to their elaborate tissue culture requirements. Therefore it is imperative to develop multiplexing assays, in order to gain the maximum amount of data possible without sacrificing these precious cells.

The use of secreted reporter assays in stem cell research would, for example, allow for continuous monitoring of media supernatant from the same culture during subsequent differentiation stages. In the case of primary neuronal cells, live cell assays based on fluorescent proteins and/or secreted bioluminescent reporters could be developed to monitor the effect of repeated stimulation on receptor downregulation and desensitization in the same set of primary cells via continuous sampling. This type of long term monitoring experiment on valuable cells is only possible with live cell assays.

Although the reporters used in this work demonstrated very good performance in this set of applications, further improvements could enable even broader applications. For example, the signal decay rate of secreted Metridia luciferase to half of its original intensity is 10 minutes, which requires improvement. Although other bioluminescent reporters like Gaussia luciferase have considerably faster signal decay rates (in the range of 30–60 seconds) ¹, the signal stability of Metridia luciferase activity will need to be further stabilized. This will be especially important when slower plate readers are used to analyze 384- or even 1536-well plates. If the signals from all wells are acquired simultaneously, fast signal decay is not problematic. However if acquisition occurs on a well-to-well basis, the signal intensity measured in the first and the last well would differ considerably.

Signal stabilization could be accomplished by a variety of approaches: Once the enzymatic site in secreted Metridia luciferase is identified, it may be possible to mutate this domain in order to change the enzymatic turnover rate of the substrate. Another possible approach could involve targeting the substrate itself. Changes in the structure of the substrate could lead to changes in the turnover rate, including association of the substrate with the luciferase and the dissociation of the product after the enzymatic reaction. Structural modifications to the substrate could also affect the intensity of the bioluminescent signal it produces upon interaction with the luciferase.

Further improvements also need to be made to fluorescent proteins for use in live cell reporter applications, in order to guarantee the lowest possible level of cytotoxicity. Minimizing cytotoxicity is especially important for applications in primary cells, stem cells, and animal models. Additionally, brighter, further red shifted fluorescent proteins must be developed to enhance live cell applications, due to their better signal to noise ratio when expressed in vivo.

Fluorescent proteins that act as sensors for specific events in a cell, either by changing their emission spectra or their fluorescence intensity, also need to be improved in order to create options for even more informative multiplexed assays, combining fluorescent, bioluminescent, and other assays such as absorbance based assays, all in the same live cell.