Atio)/d (t)) values that correspond to defined concentrations of extracellular oxygen. Generally, novel intramolecular FRET-based biosensors are pre-calibrated in vitro and subsequently transferred into the living system of interest [38-41]. However, in contrast to other FP-based biosensors which allow the quantification of a certain metabolite, in vitro calibration of FluBO is not an easy task. n-Phenylpiperazine-1-carboxamide Accurate pre-calibration requires biosensor protein, which was expressed and subsequently purified under anaerobic conditions. In addition, in vitro calibration has to be conducted using the gaseous substrate O 2 in an otherwise oxygen-free atmosphere. Furthermore, biosensor pre-calibration basically does not take into account that differences in non-specific interactions between a molecule (that is, the respective biosensor) and its immediate surrounding (for example, a buffer in comparison to the interior of a living cell) can greatly influence the equilibrium and rate of the respective reaction [42,43]. For that reason, the use of pre-calibrated reference probes does not inevitably lead to accurate in vivo calibration of a novel biosensor. Finally, as discussed below, the level of biosensor expression and, to a minorMost of the YFP derivatives described in the literature exhibit a distinct pH sensitivity with pKa values ranging from 6.9 (EYFP) to 5.7 (mCitrine) [28]. Using YFP as a FRET receptor domain, therefore, ideally requires a constant fluorescence brightness which is not affected by intracellular changes of pH. To quantitatively interpret oxygen measurements with FluBO, we utilized an EYFP variant distributed by Clontech-Takara (details are described within the Methods section and the corresponding EYFP sequence is shown in Additional file 1). Analysis of pH sensitivity demonstrated that PubMed ID:https://www.ncbi.nlm.nih.gov/pubmed/25386826 this YFP derivative exhibits a remarkable pH resistance (pKa = 5.2; Additional file 3). Thus, FluBO-based oxygen determination is not affected within the physiological pH range. In addition, the early YFP variants were also sensitive towards chloride [28]. To further rule out that chloride ions influenced the FluBO oxygen response in vivo, we also determined Clsensitivity of the used EYFP. We could demonstrate that YFP fluorescence intensity remained constant in the presence of chloride concentrations up to 100 mM (data not shown). This result clearly indicates that FluBO is also insensitive to changes of chloride-ion concentrations in the physiological relevant range.Limitations of FluBOAlthough FluBO is a useful molecular probe, some aspects must be considered to allow online-detection of intracellular oxygen: First, as mentioned above, oxygendependent chromophore formation is an irreversible process. Consequently, this biosensor indeed allows analyzing changes from low to high oxygen concentrations but it cannot directly be used to detect reduction ofPotzkei et al. BMC Biology 2012, 10:28 http://www.biomedcentral.com/1741-7007/10/Page 9 ofoxygen levels via its change of intramolecular FRET efficiency. Therefore, we analyzed alterations from sufficient to deficient oxygen levels by observing the total changes of dual emission ratio over time. By using this method, recurring oxygen PubMed ID:https://www.ncbi.nlm.nih.gov/pubmed/14445666 limitation was reflected by a stagnating alteration of YFP-to-FbFP fluorescence ratio (as for example, visible in Figure 3B between t2 and t3) and 4-((2-Hydroxyethyl)(methyl)amino)benzaldehyde in turn by decreasing (d (ratio)/d (t)) values (Figure 4A). Furthermore, it should be emphasized that due to the irreversibilit.