Fluorescent reporters are a key tool used for the characterization of genetic parts and are an important component of many iGEM projects. Since the focus of our project is to produce a pH-responsive system that detects acidic or alkaline pH and produce molecules that will neutralize the solution, we tested our circuits' performance at different pH values. In the course of evaluating the behavior of our circuits using BFP and RFP as reporters, we observed variations in the signal at different pH’s. This led us to investigate the properties of the proteins outside of the cellular environment so we could better understand the significance of any observed alterations in signal as we tested our devices.
It is well-known that the optical properties of fluorescent proteins such as lambda max, quantum yield, and brightness, rely heavily on the microenvironment of the chromophore (1). In fact, this feature has been exploited to develop mutants that can be used as pH indicators for studies of pH within specific intracellular compartments (2,4). Fluorescent proteins vary widely in their pH-sensitivity, from the pH stable CFP variant mTurquoise2 to the highly sensitive EYFP (5). The Fluorescent Protein Database (FPBase), a database that documents the properties of over one thousand fluorescent proteins, cautions users seeking the best protein for their application that "all measurements reflect the conditions under which they were performed. Conditions such as temperature, pH, fusion protein, expression level, etc... can all dramatically affect the performance and characteristics of a fluorescent protein (7)." Despite the recognized pH sensitivity of fluorescent proteins, it is often difficult to translate this knowledge into information of practical utility, as pH dependencies can exhibit distinct differences with even small sequence changes and repeated measurements yield pKa values that can vary widely between different labs (6). The iGEM community seeks to tackle issues of repeatability through the InterLab studies (8). These studies have often used red and blue fluorescent protein constructs as reporters, as our team did. We identified the following factors that could have an impact on our use of fluorescent reporters at different pH's:
- Solution conditions affect the spectral properties of fluorescent proteins
- Small differences in fluorescent protein sequence can produce distinctly different spectral properties
- Measurement reproducibility is important to the progress of synthetic biology
Producing Proteins for Our Measurement Study
In this context, our goal was to characterize the behavior of the proteins used as reporters in our study, Strongly Enhanced Blue Fluorescent Protein 2 (SBFP2, Part BBa_K156010) and a monomeric Red Fluorescent Protein (RFP, Part BBa_E1010). These parts were cloned into pSB1C3 under the control of a strong constitutive promoter (Part BBa_J23100) and with the addition of a 6x His tag to aid in purification. A complete description of these new composite parts can be found on our Contribution page.
The above circuits were transformed into DH5alpha cells. Cultures for purification were grown in LB media, the cells lysed via lysozyme treatment and sonication, and the proteins were purified via immobilized metal affinity chromatography (IMAC) as described in our protocol manual and shown in Figures 1 and 2. Protein concentration was determined using absorbance measurements at 280nm and a molar absorptivity coefficient calculated from the amino acid sequence using the Expasy ProtParam server (9). Purification was verified via SDS-PAGE (Figure 3).
Figure 1: Chromatogram of SBFP2-6xHis purification
Figure 2: Chromatogram of RFP-6xHis purification
Figure 3: SDS-PAGE gel showing purification of SBFP2 (27.9kD) and RFP (26.5kD)
Conducting the Study
BFP was loaded into a 96-well plate with buffers in a final concentration of 0.566ng/ml. The buffers, at a concentration of 100mM, ranged from pH 4.0 to 9.0 in 0.5 increments (specific buffer compounds can be found in the buffer table in our protocol manual). To normalize the fluorescence signal, 100μL of each buffer alone was added in another row. Samples were run in triplicate when possible. A Perkin Elmer LS 55 fluorescence spectrometer with a plate reader attachment was used to collect the data. We conducted an excitation scan and found only slight variations in the optimal excitation wavelength. We then conducted an emission scan that measured signal intensity from 400nm to 500nm with a fixed excitation wavelength of 374nm. Results show that at higher pH ranges, the maximum emission wavelength shifts to higher values (Figure 4). In addition, we investigated the pH-dependent variation in signal intensity (Figure 5). We observed 51% fluorescence in SBFP2 at pH 5.5, which is in agreement with previous studies that identified the pKa of fluorescence to be 5.3 (6).
Figure 4: pH dependence of SBFP2 emission spectrum
Figure 5: pH dependence of SBFP2 emission using an excitation of 374 nm and emission at 446 nm
Our results suggest that protein concentrations being equal, different pH values can cause an underestimation of the fluorescence signal of the SBFP2 reporter. We also determined the proportionality of the SBFP2 signal at the pH values measured, using the maximum intensity observed at pH 8.0 as a reference point (Figure 6). We recommend scaling by these factors when comparing SBFP2 reporter expression when comparing samples under different pH conditions. These results impact our interpretation of the growth curve studies on our base-inducible system, since the pH values where are circuit begins to turn on are seen in the region of the pH scale where the BFP signal begins to lose intensity. We have used the scaling factors found in Figure 6 to reconsider the results of our base-inducible riboswitch system.
Figure 6: Table of scaling factors for SBFP2 measurements taken at various pH values
We conducted a similar study with the purified RFP, examining the spectral properties from pH 4.0 to 9.0. We saw no real impact of pH on excitation or emission wavelengths. We also observed somewhat less of a dependence of signal intensity as the pH varied (Figure 7). However, the region of the pH scale where the signal drops is in the acidic range. Since we were using RFP as a reporter for our acid-sensitive circuit, this data indicates that we might be underestimating the degree of circuit induction at pH’s below 6, the region in which we observe a reduction in signal. We did not determine scaling factors as we did have a complete data set with multiple replicates prior to this report.
Figure 7: pH dependence of RFP emission intensity
References
1. Kostyuk, Alexander I et al. “Circularly Permuted Fluorescent Protein-Based Indicators: History, Principles, and Classification.” International journal of molecular sciences vol. 20,17 4200. 27 Aug. 2019, doi:10.3390/ijms20174200
2. Li, Shu-Ang et al. “Progress in pH-Sensitive sensors: essential tools for organelle pH detection, spotlighting mitochondrion and diverse applications.” Frontiers in pharmacology vol. 14 1339518. 10 Jan. 2024, doi:10.3389/fphar.2023.1339518
3. Poëa-Guyon, Sandrine et al. “The enhanced cyan fluorescent protein: a sensitive pH sensor for fluorescence lifetime imaging.” Analytical and bioanalytical chemistry vol. 405,12 (2013): 3983-7. doi:10.1007/s00216-013-6860-y
4. Bizzarri, Ranieri et al. “Green fluorescent protein based pH indicators for in vivo use: a review.” Analytical and bioanalytical chemistry vol. 393,4 (2009): 1107-22. doi:10.1007/s00216-008-2515-9
5. Burgstaller, Sandra et al. “pH-Lemon, a Fluorescent Protein-Based pH Reporter for Acidic Compartments.” ACS sensors vol. 4,4 (2019): 883-891. doi:10.1021/acssensors.8b01599
6. Cranfill, P., Sell, B., Baird, M. et al. Quantitative assessment of fluorescent proteins. Nat Methods 13, 557–562 (2016). https://doi.org/10.1038/nmeth.3891
7. Lambert, TJ: (2019) FPbase: a community-editable fluorescent protein database. Nature Methods 16, 277–278. doi: 10.1038/s103900352-8
8. Beal, Jacob et al. “Comparative analysis of three studies measuring fluorescence from engineered bacterial genetic constructs.” PloS one vol. 16,6 e0252263. 7 Jun. 2021, doi:10.1371/journal.pone.0252263
9. Gasteiger E., Hoogland C., Gattiker A., Duvaud S., Wilkins M.R., Appel R.D., Bairoch A.; Protein Identification and Analysis Tools on the Expasy Server; (In) John M. Walker (ed): The Proteomics Protocols Handbook, Humana Press (2005). pp. 571-607