• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of aemPermissionsJournals.ASM.orgJournalAEM ArticleJournal InfoAuthorsReviewers
Appl Environ Microbiol. Sep 2010; 76(17): 5990–5994.
Published online Jul 2, 2010. doi:  10.1128/AEM.00701-10
PMCID: PMC2935039

Flavin Mononucleotide-Based Fluorescent Reporter Proteins Outperform Green Fluorescent Protein-Like Proteins as Quantitative In Vivo Real-Time Reporters[down-pointing small open triangle]


Fluorescent proteins of the green fluorescent protein (GFP) family are commonly used as reporter proteins for quantitative analysis of complex biological processes in living microorganisms. Here we demonstrate that the fluorescence signal intensity of GFP-like proteins is affected under oxygen limitation and therefore does not reflect the amount of reporter protein in Escherichia coli batch cultures. Instead, flavin mononucleotide (FMN)-binding fluorescent proteins (FbFPs) are suitable for quantitative real-time in vivo assays under these conditions.

In the last decade, quantitative analysis of cellular processes has enabled detailed insights into the complexity of metabolic and regulatory networks in various microorganisms, especially in the fields of systems and synthetic biology. Within the respective research areas, fluorescent proteins (FPs) are used as versatile in vivo reporters to study gene regulation and protein synthesis, folding, localization, and activity (1, 5, 7) in bacteria and yeast. The most widely used FPs are the green fluorescent protein (GFP) and its derivatives. However, their use as in vivo reporter proteins is restricted by various environmental factors affecting their signal intensity, including the availability of oxygen during maturation (18, 20). In addition, time-consuming protein folding and chromophore formation cause a time lag between the detectable fluorescence signal and the amount of reporter protein produced at a certain time point. Recently, we have developed an alternative family of fluorescent proteins that binds flavin mononucleotide (FMN) as a chromophore (3, 4). In contrast to all members of the GFP family, these novel FMN-binding fluorescent proteins (FbFPs) exhibit bright cyan-green fluorescence under both aerobic and anaerobic conditions and can be used to label facultative and strict anaerobic bacteria (3, 16) and yeast (19).

Here, we comparatively analyzed whether FbFP and a GFP-derived enhanced yellow fluorescent protein (YFP) can be used as reporters for quantitative real-time analysis of gene expression in living Escherichia coli cells. Furthermore, we studied whether the detectable FP fluorescence intensities correlate with the amounts of reporter proteins continuously during cell growth. Therefore, both FPs were expressed in E. coli using vector pRhokHi-2 (12), which harbors the aphII promoter of the kanamycin resistance gene and allows the constitutive expression of both fluorescence proteins at moderate levels. The pRhokHi-2-YFP expression vector, encoding yellow fluorescent GFP-10C derivative YFP (15) (available from Clontech-Takara Bio Europe, Saint-Germain-en-Laye, France), was constructed as described by Katzke et al. (12). Expression vector pRhokHi-2-FbFP, encoding E. coli FbFP (EcFbFP) (GenBank accession number ABN71355) (evoglow-Bs2; evocatal GmbH, Düsseldorf, Germany), was generated by PCR amplification of the fluorescence reporter gene, while NdeI and XhoI restriction sites were introduced into its 5′ and 3′ ends, respectively. Subsequently, the reporter gene was cloned into the corresponding sites of the expression vector pRhokHi-2. PaphII-dependent FP expression basically avoids protein misfolding caused by high-level gene expression and alterations of FP expression levels due to changes in the inducer-to-cell ratio during cultivation.

In order to continuously monitor FP expression and in vivo fluorescence, E. coli DH5α (8) strains carrying the FP-encoding expression vector pRhokHi-2-YFP or pRhokHi-2-FbFP were generated. The respective empty vector was used as a negative control. Subsequently, the recombinant E. coli strains were cultivated in TB medium (13) under defined conditions (37°C, 700 rpm shaking frequency, 3 mm shaking diameter) in a BioLector microbioreactor system (m2p-labs GmbH, Aachen, Germany) over 16 h. To allow aeration of the cultures, a 48-well microtiter plate (MTP) (Greiner bio-one GmbH, Frickenhausen, Germany) was sealed with air-permeable sealing tape (AB-0718; ABgene, Epsom, United Kingdom). During cultivation, the BioLector provides quantitative online data of (i) biomass via measuring of scattered light at an excitation wavelength of 620 nm, (ii) dissolved oxygen tension (DOT) using an immobilized oxygen-sensitive fluorescent dye (m2p-labs GmbH, Aachen, Germany), and (iii) FP-mediated fluorescence in a continuously shaken MTP, as described previously (11, 13, 17). YFP fluorescence was measured at an excitation wavelength of 510 nm and an emission wavelength of 540 nm, whereas FbFP fluorescence was monitored at 492 nm upon excitation at 460 nm. Background fluorescence of the control strain E. coli DH5α (pRhokHi-2) was used to normalize the fluorescence signals during cultivation of the YFP and FbFP expression strains (Fig. 1A and B).

FIG. 1.
Comparative analysis of YFP and FbFP expression and in vivo fluorescence in E. coli. E. coli DH5α carrying expression plasmid pRhokHi-2 encoding YFP (A) or FbFP (B) was batch cultured using a BioLector microbioreactor system. The development of ...

Within the first 4 h of cultivation, where cells switched from the log phase to the early exponential growth phase, YFP and FbFP fluorescence as well as cell density increased simultaneously. Remarkably, development of YFP in vivo fluorescence (Fig. (Fig.1A)1A) stagnated after 5 h of cultivation, although exponential growth of the expression strain was not affected. More surprisingly, after 11.5 h of cell growth, YFP fluorescence rapidly increased about 2.5-fold within the next 2 h. At this point, the culture had clearly reached the stationary growth phase. In contrast, fluorescence of FbFP increased continuously until the E. coli culture reached the stationary growth phase (Fig. (Fig.1B1B).

Analysis of dissolved oxygen tension during E. coli cultivation demonstrated that the high respiratory activity of the bacterial cells during the logarithmic and the early stationary growth phases led to a limitation of oxygen in both expression cultures (Fig. 1A and B). Obviously, the delayed YFP fluorescence development corresponded to the phase of low oxygen tension. Hence, we analyzed whether O2 deprivation during logarithmic cell growth leads to either low YFP expression levels or inefficient chromophore maturation. To this end, we determined the amounts of accumulated transcript and protein at various representative time points during cellular growth for both YFP and FbFP (t1 to t7) (Fig. 1C to F). For the analysis of transcript and protein levels, eight 500-μl TB medium cultures were inoculated in parallel with aliquots of precultures to yield an initial cell density of 0.1 (optical density at 600 nm [OD600]). Sampling was conducted at the given time points (t1 to t7) without interruption of the shaking movement. Subsequently, the OD600 of the samples was determined using a Genesys 20 spectrophotometer (Thermo Scientific, Waltham, MA), and cells were harvested by centrifugation (14,000 rpm, 4°C, 10 min). Total RNA was isolated from aliquots of cell cultures by using an RNeasy mini kit (Qiagen, Hilden, Germany). One hundred fifty nanograms of total RNA was used for reverse transcription and subsequent real-time PCR (RT-PCR) with a QuantiTect SYBR green one-step RT-PCR kit (Qiagen, Hilden, Germany). Reverse transcription and real-time PCR were performed using a Mastercycler epgradient S with a realplex4 module (Eppendorf, Hamburg, Germany). For specific detection of FP transcripts, the following primer pairs were used: YFP-f primer, 5′-AGAAGAACGGCATCAAGGTGAACT-3′, and YFP-r primer, 5′-GGACTGGTAGCTCAGGTAGTGGTTG-3′; fbfp-f primer, 5′-TGCAGATCCAGAACTACAAGAAGGAC-3′, and fbfp-r primer, 5′-TATTCCTTCTGCTTGGTGATGTCG-3′; and aphII-f primer, 5′-CTATCAGGACATAGCGTTGGCTACC-3′, and aphII-r primer, 5′-GAACTCGTCAAGAAGGCGATAGAAG-3′. The transcript obtained from the constitutively expressed aphII gene was used as an internal standard and reference. For data analysis, threshold cycle (ΔCT) values of YFP and aphII as well as FbFP and aphII transcripts were calculated from a minimum of triplicate measurements. Levels of YFP and FbFP mRNA were determined by calculating ΔΔCT and 2−ΔΔCT values for the respective transcripts. The ΔCT value calculated for transcripts accumulated at t1 was referred to as the reference value for each time point (t2 to t7). Protein extracts corresponding to a final cell density of 1 (OD600) were separated by SDS-PAGE (NuPAGE, 4 to 12% Bis-Tris gel, 1 mm; Invitrogen, Paisley, United Kingdom) and transferred to a polyvinylidene difluoride (PVDF) membrane (Bio-Rad Laboratory, München, Germany) by Western blotting (NuPAGE; Invitrogen) using standard methods. YFP and FbFP proteins were visualized using the respective antibodies (rabbit anti-GFP [BD Biosciences, Erembodegem, Belgium] and YtvA-specific antiserum). The overall changes in FP protein levels were quantified by charge-coupled-device (CCD) camera-based analysis of chemiluminescence signals obtained from YFP and FbFP immunodetection (Stella imaging system analyzed with AIDA advanced image data analyzer; Raytest, Straubenhardt, Germany). As expected, quantitative RT-PCR and Western blot analyses clearly demonstrated that both reporter genes were constitutively expressed in E. coli (Fig. 1C to F). The change in FP transcript and protein levels never exceeded a factor of 2 over the complete time of cultivation. These results indicated that the delay in YFP fluorescence signal development that occurred in vivo during the early logarithmic growth phase as well as its rapid increase during the stationary growth phase was caused predominantly by a reversible inhibition of chromophore maturation under oxygen-limited conditions.

The applicability of an FP as an in vivo reporter is dictated by its individual robustness within a respective test system. Thus, an ideal in vivo fluorescence reporter must exhibit a constant fluorescence-to-protein ratio that is independent from growth parameters and growth phases. In this regard, our comparative in vivo fluorescence measurements in E. coli revealed that FbFP but not YFP can be used as a reporter protein for accurate quantitative analyses in living cells. These findings became even more obvious when fluorescence intensities of FbFP and YFP were normalized against cell densities. Figure Figure22 shows that normalized FbFP fluorescence was almost constant over the entire cultivation time. In addition, the increased fluorescence intensity observed within the first 7 h of cell growth correlated with higher FbFP accumulation (Fig. 1D and F, t1 and t2). In contrast, under equivalent experimental conditions, normalized YFP fluorescence intensity showed dramatic changes over the complete period of cell growth. After a decrease of more than 6-fold during the first 8 h of cell growth, YFP fluorescence subsequently increased 2.5-fold during the late stationary growth phase. Thus, YFP accumulated at constant levels in all phases of bacterial cell growth (Fig. 1C and E), but these levels were not reflected by the fluorescence signals determined during growth. The results demonstrate that changes in normalized YFP fluorescence are caused by an impaired development of YFP signals under the respective environmental or metabolic conditions. Due to this inconsistency in YFP fluorescence intensity, quantification of the fluorescence signal is not sufficient to reliably analyze the expression of the respective reporter gene in vivo.

FIG. 2.
Fluorescence characteristics of YFP and FbFP during cell growth. Fluorescence intensities derived from online measurements during E. coli cultivation (Fig. 1A and B) were normalized against the respective cell densities (scattered light intensity I-Io). ...

We further analyzed YFP and FbFP fluorescence in the respective E. coli DH5α expression strains batch cultured in Luria-Bertani (LB) medium (13) at 37°C in a 48-well MTP by using the BioLector to determine potential changes in YFP fluorescence during cell growth caused by oxygen limitation. As described previously, the cultivation of E. coli in unbuffered complex LB medium leads to a fast exhaustion of carbon sources and a rapid basification of the culture medium, resulting in an early inhibition of cell growth (14). As shown in Fig. Fig.33 A and B, these limiting growth conditions also caused lower respiratory activity of the E. coli cells, as reflected by a minimal dissolved oxygen tension of approximately 50% at the end of the logarithmic growth phase. Presumably, the development of YFP fluorescence (Fig. (Fig.3A)3A) is not affected by oxygen depletion, as observed in TB medium (Fig. (Fig.1A).1A). In contrast to FbFP-mediated fluorescence, which increased parallel to the cell density, analysis of YFP-mediated fluorescence revealed a remarkably delayed development of the reporter signal (Fig. 3A and B). As a direct consequence, the ratio between YFP fluorescence and cell density was, again, not constant during cell cultivation but showed a 2-fold increase in fluorescence after the bacteria reached the stationary growth phase (Fig. (Fig.3C).3C). In contrast, FbFP-dependent fluorescence normalized against cell density exhibited an almost linear behavior over 16 h of cultivation (Fig. (Fig.3C).3C). The maturation of GFP-like proteins is a process that includes time-consuming protein folding (2, 6) and chromophore formation (9, 10) and may thus significantly delay the development of fluorescence signals. Calculation of the YFP fluorescence signals while assuming an average YFP maturation time of 1 h, as indicated by the respective online fluorescence measurements (Fig. (Fig.3A),3A), revealed that fluctuations of normalized YFP fluorescence observed in the logarithmic growth phase were partially reduced (Fig. (Fig.3C,3C, YFPrec). Therefore, these results further support our initial assumption that the use of GFP derivatives as in vivo real-time reporters is also limited by time-demanding maturation.

FIG. 3.
Comparative expression and in vivo fluorescence of YFP (A) and FbFP (B) in E. coli DH5α cultured in LB medium. Development of biomass (cell density), dissolved oxygen tension (DOT), and FP-mediated autofluorescence were online monitored in each ...

In this article, we describe the performances of two different fluorescent proteins when used as real-time in vivo reporters in a microbioreactor system. We demonstrated that FbFP, in contrast to YFP, is insensitive toward variable levels of dissolved oxygen during cell cultivation and that its fluorescence signal develops immediately after gene expression. Fluorescence signals generated by FbFP rather than GFP-related FPs thus reflect in vivo gene expression and protein production levels. Their unique properties make FbFPs superior tools for systems and synthetic biology as well as for various biotechnological approaches.


We thank Frank Kensy (m2p-labs GmbH, Aachen, Germany) and Michael Puls (evocatal GmbH, Düsseldorf, Germany) for kind support, as well as Ulrich Krauss (Heinrich Heine University, Düsseldorf, Germany) for helpful discussion.


[down-pointing small open triangle]Published ahead of print on 2 July 2010.


1. Chudakov, D. M., S. Lukyanov, and K. A. Lukyanov. 2005. Fluorescent proteins as a toolkit for in vivo imaging. Trends Biotechnol. 23:605-613. [PubMed]
2. Crameri, A., E. A. Whitehorn, E. Tate, and W. P. C. Stemmer. 1996. Improved green fluorescent protein by molecular evolution using DNA shuffling. Nat. Biotechnol. 14:315-319. [PubMed]
3. Drepper, T., T. Eggert, F. Circolone, A. Heck, U. Krauss, J.-K. Guterl, M. Wendorff, A. Losi, W. Gärtner, and K.-E. Jaeger. 2007. Reporter proteins for in vivo fluorescence without oxygen. Nat. Biotechnol. 25:443-445. [PubMed]
4. Eggert, T., T. Drepper, J. Guterl, A. Heck, U. Krauss, and K.-E. Jaeger. April 2007. New light, oxygen, voltage domains, and proteins containing them, useful as fluorescent markers, e.g. in gene therapy, cloning and cell sorting, are altered to prevent covalent bonding to flavine mononucleotide. Patent DE102005048828-A1.
5. Frommer, W. B., M. W. Davidson, and R. E. Campbell. 2009. Genetically encoded biosensors based on engineered fluorescent proteins. Chem. Soc. Rev. 38:2833-2841. [PMC free article] [PubMed]
6. Fukuda, H., M. Arai, and K. Kuwajima. 2000. Folding of green fluorescent protein and the cycle3 mutant. Biochemistry 39:12025-12032. [PubMed]
7. Giepmans, B. N. G., S. R. Adams, M. H. Ellisman, and R. Y. Tsien. 2006. The fluorescent toolbox for assessing protein location and function. Science 312:217-224. [PubMed]
8. Hanahan, D. 1983. Studies on transformation of Escherichia coli with plasmids. J. Mol. Biol. 166:557-580. [PubMed]
9. Heim, R., A. B. Cubitt, and R. Y. Tsien. 1995. Improved green fluorescence. Nature 373:663-664. [PubMed]
10. Heim, R., D. C. Prasher, and R. Y. Tsien. 1994. Wavelength mutations and posttranslational autoxidation of green fluorescent protein. Proc. Natl. Acad. Sci. U. S. A. 91:12501-12504. [PMC free article] [PubMed]
11. Huber, R., D. Ritter, T. Hering, A. K. Hillmer, F. Kensy, C. Müller, L. Wang, and J. Büchs. 2009. Robo-Lector—a novel platform for automated high-throughput cultivations in microtiter plates with high information content. Microb. Cell Fact. 8:42. [PMC free article] [PubMed]
12. Katzke, N., S. Arvani, R. Bergmann, F. Circolone, A. Markert, V. Svensson, K.-E. Jaeger, A. Heck, and T. Drepper. 2010. A novel T7 RNA polymerase dependent expression system for high-level protein production in the phototrophic bacterium Rhodobacter capsulatus. Protein Expr. Purif. 69:137-146. [PubMed]
13. Kensy, F., E. Zang, C. Faulhammer, R. K. Tan, and J. Büchs. 2009. Validation of a high-throughput fermentation system based on online monitoring of biomass and fluorescence in continuously shaken microtiter plates. Microb. Cell Fact. 8:31. [PMC free article] [PubMed]
14. Losen, M., B. Frölich, M. Pohl, and J. Büchs. 2004. Effect of oxygen limitation and medium composition on Escherichia coli fermentation in shake-flask cultures. Biotechnol. Prog. 20:1062-1068. [PubMed]
15. Ormö, M., A. B. Cubitt, K. Kallio, L. A. Gross, R. Y. Tsien, and S. J. Remington. 1996. Crystal structure of the Aequorea victoria green fluorescent protein. Science 273:1392-1395. [PubMed]
16. Piekarski, T., I. Buchholz, T. Drepper, M. Schobert, I. Wagner-Doebler, P. Tielen, and D. Jahn. 2009. Genetic tools for the investigation of Roseobacter clade bacteria. BMC Microbiol. 9:265. [PMC free article] [PubMed]
17. Samorski, M., G. Müller-Newen, and J. Büchs. 2005. Quasi-continuous combined scattered light and fluorescence measurements: a novel measurement technique for shaken microtiter plates. Biotechnol. Bioeng. 92:61-68. [PubMed]
18. Shaner, N. C., P. A. Steinbach, and R. Y. Tsien. 2005. A guide to choosing fluorescent proteins. Nat. Methods 2:905-909. [PubMed]
19. Tielker, D., I. Eichhof, K.-E. Jaeger, and J. F. Ernst. 2009. Flavin mononucleotide-based fluorescent protein as an oxygen-independent reporter in Candida albicans and Saccharomyces cerevisiae. Eukaryot. Cell 8:913-915. [PMC free article] [PubMed]
20. Tsien, R. Y. 1998. The green fluorescent protein. Annu. Rev. Biochem. 67:509-544. [PubMed]

Articles from Applied and Environmental Microbiology are provided here courtesy of American Society for Microbiology (ASM)
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...