High-resolution spatiotemporal imaging of histidine in single living mammalian cells faces technical challenges. Here, we developed a series of ratiometric, highly responsive, and single fluorescent protein-based histidine sensors of wide dynamic range. We used these sensors to quantify subcellular free-histidine concentrations in glucose-deprived cells and glucose-fed cells. Results showed that cytosolic free-histidine concentration was higher and more sensitive to the environment than free histidine in the mitochondria. Moreover, histidine was readily transported across the plasma membrane and mitochondrial inner membrane, which had almost similar transport rates and transport constants, and histidine transport was not influenced by cellular metabolic state. These sensors are potential tools for tracking histidine dynamics inside subcellular organelles, and they will open an avenue to explore complex histidine signaling.
As an essential amino acid and a precursor to histamine, histidine plays a vital role in human growth, metal transmission1, neurotransmission, and neuromodulation2. An abnormal level of histidine or histidine-rich proteins is also often considered as an indicator of many diseases. For example, high levels of histidine/histidine-rich proteins have been associated with phenylketonuria3, malaria4, and thrombotic disorders5. By contrast, low levels of histidine/histidine-rich proteins have been linked with rheumatoid arthritis6, epilepsy7, chronic kidney disease8, and advanced liver cirrhosis9. Therefore, selective and sensitive detection of histidine in living cells has become significant and indispensable, and this process facilitates the understanding of the pathogenesis for clinical therapy. Existing methods for assaying histidine, such as chromatography10, mass spectrometry11, electrochemistry12, and capillary electrophoresis13, are invasive and not suitable for studying histidine dynamics in intact individual cells. Various chemical probes have been developed to monitor histidine in living cells14,15,16. However, these probes are limited by subcellular location, and in vivo application. In addition, use concentration, incubation time and washing times of chemical probes, and even themselves also often introduce some artificial interferences.
Genetically encoded fluorescent sensors with high spatiotemporal resolution have been developed to sense various intracellular metabolites17. As genetically encoded proteins, these sensors can be easily introduced into living cells by DNA transfection or targeted to different subcellular compartments by tagging with organelle-specific signal peptides18. In addition, these sensors could also be used to establish transgenetic organisms and applied in in vivo research. The sensors derived from fluorescent proteins (FP) provide high sensitivity and great versatility while minimally perturbing the cells under investigation, and are leading a revolution for cell biology research19.
It is interesting to note that researchers have developed various sensors based on circularly permuted fluorescent proteins (cpFPs) for monitoring cellular events20,21,22,23. In these cpFPs, the original N- and C-termini are fused by a polypeptide linker, and new termini are introduced close to the fluorophore, making its fluorescence highly sensitive to the protein’s conformation. A variety of bacterial periplasmic binding proteins (PBPs) are known to sense substrates with high affinity and specificity, including amino acids, sugars, metals and inorganic ions24,25,26. Crystallographic studies of PBPs show that substrate binding often induces conformational changes, providing attractive scaffolds to make indicators24,25,26,27.
To overcome the disadvantages (e.g., invasiveness of sample and low spatiotemporal resolution) of existing methods, we developed a series of single fluorescent protein (FP)-based histidine sensors with various affinities and large dynamic range by combining cpYFP with a bacterial periplasmic binding protein, HisJ. These sensors, dubbed FHisJ, allow for specific, sensitive, and quantitative monitoring of histidine metabolism in various subcellular compartments of mammalian cells.
Generation of cpYFP-based Histidine Indicators
Various genetically encoded fluorescent sensors have been created to monitor cellular events and the microenvironment17,28. These sensors could be generally categorized into fluorescence resonance energy transfer (FRET)–based and single FP-based sensors17,29. Compared with single FP-based sensors, the dynamic ranges of FRET-based reporters usually fall into 10–150% with very few exceptions17,21, making determination of subtle differences in biochemical responses challenging30. We have successively developed first-generation and second-generation NADH biosensors, called Frex31 and SoNar32, respectively, by employing cpYFP. These biosensors display 800%31 or 1500%32 dynamic range, respectively, far exceeding those of other metabolite sensors17. To design single FP-based histidine sensors, we inserted cpYFP into a type II periplasmic binding protein HisJ, which manifests large conformational changes upon histidine binding33,34 (Fig. 1a and b). In this study, a total of 36 chimeric proteins were constructed, in which cpYFP was inserted into the flexible linker 185–193 region of HisJ (Fig. 1c). Among these proteins, seven chimeras showed 240–520% increase in the ratio of fluorescence when excited at 420 and 485 nm upon histidine addition (Fig. 1c and d and Table 1), suggesting the feasibility of the proposed engineering strategy for generating a histidine reporter. Fluorescence titration studies showed that these histidine indicators had different affinities with Kd of ~2.4 μM to ~22 μM (Table 1), which are suitable for measuring subcellular histidine levels in the physiological range. Considering the maximum change in fluorescence ratio and affinity constant, we chose the chimera with cpYFP inserted between Leu 190 and Phe 191 of HisJ for detailed characterization and named this chimera as Fluorescent HisJ (FHisJ) (Fig. 1d).
In vitro Characterization of FHisJ
The fluorescence spectrum of FHisJ is similar to those of other cpYFP-based sensors31,32 with two excitation peaks at approximately 420 and 500 nm and one emission peak near 515 nm (Fig. 2a; Supplementary Fig. 1a and b). FHisJ displayed a ~2.0-fold decrease and ~3.0-fold increase in fluorescence with excitation at 420 and 485 nm, respectively, upon histidine saturation (Fig. 2b). The opposing directional changes of this sensor, also observed in SoNar32, led to a 520% ratiometric change in fluorescence (Figs 1d and 2c). FHisJ possessed high selectivity toward histidine, exhibiting no apparent fluorescence changes toward 18 amino acids and histamine (Fig. 2c). FHisJ also did not respond to low concentrations of arginine (0–100 μM), although addition of 10 mM arginine resulted in a ~3.0-fold increase in the ratio of FHisJ fluorescence with excitation at 420 and 485 nm (Fig. 2c; Supplementary Fig. 1c). Quantitative measurements of intracellular arginine were rare, however, there were reports that intracellular arginine level of 90 μM35, and a large fraction of intracellular arginine was stored within vesicles36. The Kd of FHisJ to arginine is ~400 μM (Supplementary Fig. 1c), thus arginine should not signficantly interfere with FHisJ’s histidine sensing function in the cells. Similar to other cpYFP-based sensors31,32, FHisJ fluorescence depended on pH when excited at 485 nm (Fig. 2d), however, FHisJ fluorescence excited at 420 nm is much more pH resistant (Fig. 2d and e). In addition, FHisJ dynamic range and Kd were more resistant to pH changes (Fig. 2f). At modest pH fluctuations, the pH effects of FHisJ could be corrected by measuring the fluorescence of FHisJ and cpYFP in parallel, similar to Frex and SoNar31,32, because of their very similar pH responses (Supplementary Fig. 1d). Alternatively, FHisJ fluorescence could be measured with 420 nm excitation only when pH fluctuation occurred, but at the expense of reduced dynamic range of only 100% (Fig. 2e), versus 520% fluorescence ratio change when excited at 420 nm and 485 nm (Fig. 2c). For dual color ratiometric imaging, FHisJ may be fused with mCherry, a red fluorescent protein that is not sensitive to physiological pH (e.g. pH 7.0–8.0) or to histidine addition. Therefore, the ratio of FHisJ-mCherry green fluorescence excited at 420 nm and red fluorescence excited at 590 nm can be used to report histidine dynamics specifically, unaffected by intracellular pH or the sensor’s concentration. Hence, these data demonstrate that FHisJ with large dynamic range is a highy sensitive, highly selective, and intrinsically ratiometric biosensor for histidine and would be a promising tool for live-cell application.
Subcellular Distribution of Histidine in Mammalian Cells
We subcloned FHisJ into the pcDNA3.1/Hygro(+) mammalian expression vector and transiently expressed it in human cervical cancer HeLa cells to determine its suitability for intracellular histidine detection. Fluorescence imaging showed that FHisJ sensor with non-tagged sequences was localized exclusively in the cytosol (Fig. 3a). A duplicated cytochrome C oxidase subunit VIII signal peptide was used to lead FHisJ sensor into the mitochondrial matrix of HeLa cells to elucidate the subcellular distribution of histidine in mammalian cells (Fig. 3a). The 485 nm/420 nm fluorescence ratio of FHisJ was higher than that of FHisJ–Mit in HeLa cells when measured with a fluorescence plate reader with dual excitation (Fig. 3b). In Hela cells, the estimated resting pH was ~8.0 in the mitochondrial matrix38,39 and ~7.4 in the cytosol38. Fluorescence titration studies showed that the Kd and dynamic range of FHisJ sensor had slight differences in pH 7.4 and pH 8.0 (Fig. 2f). We carefully calibrated the effect of pH on the fluorescence of FHisJ and found that cytosolic and mitochondrial free-histidine concentrations in glucose-fed cells were ~159 and ~77 μM, respectively (Fig. 3c). However, cytosolic and mitochondrial free-histidine concentrations in glucose-deprived cells were ~20 and ~27 μM, respectively (Fig. 3c), implying that cytosolic histidine levels was more sensitive to environmental changes.
Subcellular Transport of Histidine in Mammalian Cells
Addition of exogenous histidine into the culture medium induced a rapid, dose-dependent, and saturable increase in the fluorescence ratio in the cytosol or the mitochondria of glucose-deprived HeLa cells (Fig. 4a–c). Similar results were obtained after measurement by fluorescence microscopy (Fig. 4d and e; Supplementary Fig. 2a and b), suggesting that histidine was readily transported across the plasma membrane and the inner mitochondrial membrane. By contrast, only slight changes in the fluorescence ratio were found in the cells expressing cpYFP instead of FHisJ when histidine was added to the cell culture medium (Fig. 4d and e). Thus, the possibilities of interference of fluorescence emission were excluded because of pH variations of the cpYFP domain. Furthermore, histidine transport was not affected by nutrient deprivation, nutrient feeding, glycolysis inhibition, or mitochondrial inhibition (Fig. 4f,g), which were assessed by SoNar (Fig. 4h), a highly responsive NAD+/NADH sensor with 1500% dynamic range32. This phenomenon suggested that histidine transport did not depend on cellular metabolic state. Interestingly, plasma membrane and inner mitochondrial membrane had almost similar transport rates and transport constants (K0.5: ~55 μM) (Fig. 4a–c).
In summary, we reported a series of ratiometric, and single FP-based histidine sensors with different affinities. FHisJ sensors were highly specific and highly sensitive, exhibited a large dynamic range, and could be targeted to different subcellular compartments, representing substantial improvement for live-cell histidine measurement over existing methods. It should be noted that FHisJ sensor also have its own limitations. The sensor’s affinity to histine is too high, with a dissociation constant of 22 μM, rendering the sensor largely saturated under physiological histidine concentration. Therefore the sensor’s dynamic range of in living cells is limited comparing to its in vitro dynamic range. Further studies are neccessary to decrease the sensor’s affinity and make it more useful to to detect changes in histidine levels in real cell biology/physiology conditions.
To our knowledge, subcellular free-histidine concentrations in living cells had not been rigorously quantified because of the lack of available non-invasive method. Our quantitative data were nearly equal to histidine levels in the plasma/serum (65–108 μM), standard tissue culture medium (97–200 μM), or whole cell (100–300 μM) as reported across several studies40,41,42. Michaelis constant (Km) of histidine uptake by mammalian cells was reported to be 100–120 μM for histidine41, supporting the idea that these potential histidine carriers are poised to be regulated by local histidine fluctuations.
Many transporters for histidine uptake, such as CAT143, PHT144, PHT245, and SLC38 families46, have been reported, but no clear conclusion has been established. For mitochondrial histidine transport, ORC247 and SLC25A2948 are the only known transporters that might be involved in histidine transport across the inner mitochondrial membrane. However, ORC2 transported histidine with low transport affinity (Km = 1.28 ± 0.14 mM) and slow transport rate (1.2 ± 0.2 mmol/min/g protein)47. Thus, ORC2 might not serve as a mitochondrial histidine carrier in HeLa cells. SLC25A29 could also transport histidine into the mitochondria, but it was mainly responsible for mitochondrial arginine and lysine uptake48, with transport affinities (K0.5) of 0.42 ± 0.04 and 0.71 ± 0.10 mM, respectively. The net import of histidine into the cytosol or mitochondria is necessary for the synthesis of histamine or intramitochondrially translated proteins2,49. Therefore, further study is of vital importance to address these issues on subcellular histidine carrier.
The superior properties of FHisJ sensors facilitated the use of these sensors for real-time tracking of cellular histidine fluctuations in individual mammalian cells using fluorescence imaging. FHisJ sensors are also compatible with high-throughput screening of candidate genes or drugs affecting uptake, efflux, and metabolism of histidine using microplate readers and flow cytometers. We believe that FHisJ sensors should faciliate a more complete understanding of the pathophysiological relevance of organellar histidine metabolism because of the significance of histidine in physiological and pathological conditions. These sensors are good alternatives to existing methods for intracellular histidine detection.
Materials and Methods
The gene encoding HisJ (positions 67–780 relative to ATG) was amplified from Escherichia coli genomic DNA by PCR with the primers P1 (CCCGGATCCGATGGCGATTCCGCAAAAC) and P2 (CCCAAGCTTTTAGCCACCATAAACAT) and cloned into the pRSET-B vector (Invitrogen). The cDNA of cpYFP31,32 was amplified by PCR with the primers P3 (TACAACAGCGACAACGTC) and P4 (GTTGTACTCCAGCTTGTG). The cDNA of FHisJ consisting of HisJ fused to cpYFP was generated by an overlapping PCR and cloned into the BamHI/HindIII sites of pRSET-B vector for bacterial expression. The entire coding sequences of FHisJ were subcloned into pcDNA3.1 Hygro(+) (Invitrogen) behind a Kozak sequence for mammalian expression. For mitochondrial matrix targeting, the duplicated mitochondrial targeting signal50,51,52,53 MSVLTPLLLRGLTGSARRLPVPRAKIHSLGDLSVLTPLLLRGLTGSARRLPVPRAKIHSLGD was inserted at the N-terminus of FHisJ.
Protein Expression and Purification
E. coli JM109 (DE3) cells carrying the pRSETb-FHisJ expression plasmid were grown in Luria-Bertani (LB) media containing 100 μg/ml ampicillin at 37 °C until the cultures reached approximately 0.4–0.6 OD. The expression of His6-tagged proteins was induced by the addition of 0.1 mM IPTG, and the cells were cultured at 18 °C overnight. Bacteria were then centrifuged at 4000 × g for 30 min at 4 °C. The cell pellets were suspended in buffer A (20 mM sodium phosphate, 500 mM sodium chloride, and 20 mM imidazole, pH 7.4) and lysed via ultrasonication. Proteins were purified using a Ni–NTA His SpinTrap column. After washing with 2 column volumes of wash buffer (buffer A containing 50 mM imidazole), the proteins were eluted from the resin using buffer B (20 mM sodium phosphate, 500 mM sodium chloride, and 500 mM imidazole, pH 7.4). The protein preparations were then desalted and exchanged into 20 mM sodium phosphate buffer (pH 7.4) before assay.
In vitro Characterization of FHisJ
We stored the purified protein at −20 °C until use. Excitation and emission spectra of recombinant fluorescent sensor proteins was measured as previously described31,32. The purified sensor protein was placed into a cuvette containing 20 mM sodium phosphate buffer (pH 7.4). Fluorescence was measured using a Cary Eclipse fluorescence spectrophotometer (Agilent). Excitation spectra was recorded with an excitation range from 350 nm to 510 nm and emission wavelength of 530 nm. The emission range for the emission spectra was 505–550 nm, while the excitation wavelength was 490 nm. Readings were taken every 1 nm for excitation spectra and every 3 nm for emission spectra.
For substrate titration, the protein was diluted in 20 mM sodium phosphate buffer (pH 7.4) to a final concentration of 1 μM. The fluorescence value of FHisJ sensor, in the absence of a substrate, was measured by a filter-based Synergy 2 Multi-Mode microplate reader using 420 BP 10 nm or 485 BP 20 nm excitation and 528 BP 20 nm emission band-pass filters (BioTek). The stock solution of histidine was also prepared in HEPES buffer (pH 7.4). Each assay was performed with 10 μl amino acids and 90 μl proteins in black 96-well flat bottom plate. Fluorescence intensity was read immediately after addition of substrate.
Cell Culture and Transfection
HeLa cells were grown in high-glucose DMEM (HyClone) with 10% FBS (HyClone) and cultured at 37 °C in a humidified atmosphere of 95% air and 5% CO2. For DNA transfection, we typically used 0.1 μg of plasmids with 0.3 μl of Lipofectamine 2000 for each well of a 96-well plate according to the manufacturer’s protocol.
Fluorescence Microscopy Imaging
Fluorescence microscopy imaging was performed as previously described32. Briefly, cells were plated on a 35 mm glass bottom dish (segmented) and observed after 30–36 h transfection. Glucose-free HBSS (10 mM HEPES, 136.7 mM NaCl, 5.4 mM KCl, 0.35 mM Na2HPO4, 0.44 mM KH2PO4, 4.2 mM NaHCO3, 1.26 mM CaCl2, 0.81 mM MgSO4, pH 7.4) was used to replace the growth medium before imaging. Images were acquired using a high-performance fluorescent microscopy system equipped with Nikon Eclipse Ti-E automatic microscope, monochrome cooled digital camera head DS-Qi1 Mc-U2 and the highly stable Sutter Lambda XL light source. A Plan Apo 20 × 0.75 NA objective was utilized. Cells were maintained at 37 °C under a humidified atmosphere using a CO2 incubator (Tokai Hit). Dual-excitation ratio imaging was performed using 407 BP 17 nm or 482 BP 35 nm band-pass excitation filters (Semrock) and a 535 BP 40 nm emission filter altered by a Lambda 10-XL filter wheel (Shutter Instruments). Images were captured using 1280 × 1024 format, 12 bit depth, and 2 × gain. Raw data were exported to ImageJ software as 12 bit TIF for analysis. The pixel-by-pixel ratio of the 482 nm excitation image by the 407 nm excitation image of the same cell was used to pseudocolor the images in the HSB color space. The RGB value (255, 0, 255) represents the lowest ratio, and red (255, 0, 0) represents the highest ratio. Color brightness is proportional to the fluorescent signals in both channels.
Live-cell Fluorescence Measurement Using Microplate Reader
HeLa cells were seeded in a black 96-well flat-bottom plate and transfected with the plasmid DNA of FHisJ and cpYFP for 30–36 h. Cells were rinsed twice, incubated in HBSS containing 10 mM HEPES (pH 7.4), and then maintained at 37 °C during the measurement. Dual-excitation ratios were obtained by a Synergy 2 Multi-Mode Microplate Reader (BioTek) with excitation filters 420 BP 10 nm and 485 BP 20 nm and emission filter 528 BP 20 nm emission for both excitation wavelengths. Fluorescence values were background corrected by subtracting the intensity of HeLa cell samples not expressing sensors. Unless otherwise indicated, glucose was not maintained in the buffer.
Calibration of Intracellular Free Histidine Levels
Intracellular free histidine levels were measured after calibration of FHisJ fluorescence in live cells with that of recombinant FHisJ protein as described previously31. Ratiometric measurement of FHisJ fluorescence is possible for FHisJ expressed in the cytosol or mitochondria using the microplate reader. In these experiments, samples containing equal numbers of FHisJ-expressing cells or untransfected (control) cells were measured. The background values (from control cells) were subtracted from those of FHisJ-expressing cells. This correction basically eliminated the interference of not only the autofluorescence of the microplate but also the autofluorescence of the cells. The equation is as follows:
VHis is the fraction of FHisJ sensor bound to histidine, and [His] is the free concentration of histidine. Kd represents KHis, which is the ratio of histidine at which the response of the sensor is half-maximal. Rmin and Rmax represent the F485 nm/F420 nm ratio of the recombinant FHisJ protein in the absence or presence of 1 mM Histidine, respectively. R represents the F485 nm/F420 nm ratio of the cells.
Data are presented either as a representative example of a single experiment repeated at least in triplicate or as three or more experiments. Data obtained are represented as mean values ± SD or mean values ± SEM. All p values were obtained using unpaired two-tailed Student’s t test. Values of p < 0.05 were considered statistically significant (*p < 0.05, **p < 0.01, and ***p < 0.001).
How to cite this article: Hu, H. et al. A genetically encoded toolkit for tracking live-cell histidine dynamics in space and time. Sci. Rep. 7, 43479; doi: 10.1038/srep43479 (2017).
Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
This research was supported by the 973 Program (2013CB531200 to Y.Y.), NSFC (91313301, 31225008, and 31470833 to Y.Y., 31671484 and 91649123 to Y. Zhao), the Shanghai Science and Technology Commission (14XD1401400 to Y.Y., and 15YF1402600 to Y. Zhao), the Specialized Research Fund for the Doctoral Program of Higher Education (20100074110010 to Y.Y.), the Lift Engineering for Young Talent of China Association for Science and Technology (to Y. Zhao), the State Key Laboratory of Bioreactor Engineering (to Y.Y.), the 111 Project (B07023 to Y.Y.) and the Fundamental Research Funds for the Central Universities (to Y.Y. and Y. Zhao).
About this article
RSC Advances (2018)