Measuring pH in insulin secretory granules using phasor fluorescence lifetime imaging of a genetically encoded sensor

doi:10.21203/rs.3.rs-4968583/v1

Download PDF

Article

Measuring pH in insulin secretory granules using phasor fluorescence lifetime imaging of a genetically encoded sensor

https://doi.org/10.21203/rs.3.rs-4968583/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

It is widely accepted that the pH of insulin granules is acidic, and that its active regulation during granule maturation plays a role in the process of insulin secretion by β-cells. Yet, a calibrated measurement of the absolute granule pH with organelle specificity is still lacking. To tackle this issue, we used the genetically encoded E¹GFP pH reporter inserted into the C-peptide of proinsulin and expressed in Insulinoma 1E cells. Following verification of correct targeting of the E¹GFP reporter in the insulin granules, phasor-based Fluorescence Lifetime Imaging Microscopy (FLIM) was applied to obtain a calibrated and probe-concentration-independent measurement of insulin-granule pH. Our results confirmed the acidic nature of insulin granules under maintenance cell-culturing conditions, with an average luminal pH of ~ 5.8, and showed that acidity is actively maintained, as evidenced by its near-neutralization upon treatment with the vacuolar H⁺-ATPase inhibitor Concanamycin. Additionally, by exploiting the intrinsic spatial resolution of FLIM, we highlighted that granules which are proximal to the plasma membrane are slightly more acidic (~ 0.1 pH units) than those which are distal, a difference preserved even during the early phase of glucose-induced insulin secretion. This study lays the foundations for future investigations of granule pH in physiology and disease.

Biological sciences/Biophysics/Biological fluorescence

Biological sciences/Biological techniques/Microscopy/Multiphoton microscopy

β-cells

fluorescence

FLIM

insulin secretory granules

GFP

Insulin is synthesized in its immature form, pre-proinsulin, at the endoplasmic reticulum level [1]. After its conversion into proinsulin, it is transported to the trans Golgi network (TGN) and then translocated into insulin secretory granules (ISG) [1]. At this level, the acidification of the luminal pH is supposed to be necessary for both the conversion of proinsulin into active insulin (with release of the C-peptide fragment) and for the ISG maturation and secretion at the plasma membrane level [1], [2]. Despite the crucial role played by pH regulation, a direct and accurate measurement of the absolute pH of ISGs in living cells is still missing. The very first measurements of granule intraluminal pH, performed almost 40 years ago, demonstrated that the pH of the ISG is in the 5.0 to 6.0 range, and that maintaining the acidic condition is ATP-dependent. Yet these measurements referred to isolated granules, i.e. outside the realm natural cellular environment [3], [4]. The almost concomitant advent of pH-sensitive fluorescent probes paved the way to direct investigations of granule pH within living cells. In a seminal study, Pace and Sachs used the weak base Acridine Orange (AO) as pH probe [5]. They showed that β-cell granules in islet cultures could accumulate AO with a characteristic red shift, highlighting the presence of a pH gradient across the granule membrane (i.e. between granule lumen and cell cytoplasm) [5]. Unfortunately, due to the intrinsic photophysical characteristics of AO, it was not possible to provide a reliable estimate of the actual pH value of granules in cells. With similar limitations, in 2001 Barg and colleagues monitored granular pH by supplementing the extracellular solution with LysoSensor Green DND-189, a fluorescent probe, whose fluorescence intensity increases as pH decreases [6]. Although not quantitative in terms of absolute pH, the authors were able to confirm that granular acidification (driven by a V-type H+-ATPase in the granular membrane) is a decisive step in granule priming for exocytosis [6]. Following a similar approach, Stiernet and colleagues measured granule pH in islets but using Lysosensor DND-160, a variant allowing, in principle, ratiometric determination of absolute pH values in acidic compartments [7]. Yet, results were presented in terms of pH differences, rather than in terms of absolute pH. Still, the authors were able to show that an increase in glucose concentration induces rapid and reversible decrease in granular pH in a metabolism and chloride dependent manner [7]. By contrast, in a similar experiment using the fluorescent pH indicator Lysosensor Green DND-189, Eto and colleagues observed that, upon glucose stimulation, the pH of ISG in pancreatic β-cells, was alkalinized by approximately 0.016 pH units [8]. At this stage, it is important to note that in the attempts reviewed so far, researchers employed dyes (such as acridine orange and Lysosensor variants) that are not specific to insulin granules but distribute across all acidic compartments of the cell, including endosomes and lysosomes, potentially affecting the final pH measurement. To bypass these limitations, Tompkins and colleagues, in 2002, targeted a genetically encoded pH sensor in the form of the pH-sensitive variant of green fluorescent protein (EGFP F64L/S65T) to insulin secretory vesicles in RIN1046-38 insulinoma cells by fusing the sensing moiety to the N-terminal leader sequence of human growth hormone [9]. The authors observed that glucose stimulation induces a decrease in granule pH, whereas inhibitors of the V-type H-ATPase increase pH and impair glucose effect [9]. Although specifically targeted to the insulin secretory pathway, this GFP-based pH sensor was intrinsically non-ratiometric, i.e. it reported only relative pH changes among different conditions. Worthy of mention, an attempt to overcome current limitations was performed by Neukman and co-workers sending a quantitative pH reporter to the granule (i.e. eCFP fused to the ICA512-RESP18 homology domain in INS-1 cells) but has so far remained in the form of an unpublished contribution [10].

To summarize, none of the reports available in the literature satisfy the two requirements of a reliable measurement of absolute pH in the ISG, i.e. i) specific targeting of the pH reporter in the insulin secretory pathway and ii) calibration of the reporter to obtain absolute pH values within the desired range. To tackle both issues simultaneously, we inserted the ratiometric and genetically encoded E¹GFP pH reporter within the C-peptide (C-pep) of proinsulin (See Fig. 3a). E¹GFP was selected as it is endowed with a pKa close to 6.0 and it demonstrated to be suitable for absolute pH measurements in acidic compartments [11]. Additionally, the insertion of E¹GFP into the C-pep was shown not to alter the sorting of the whole adduct into the ISG [12], [13], [14]. Phasor-based FLIM was used as a fast, robust, and fit-free method to measure ISG luminal pH independently of probe concentration, while also providing a spatial map of pH values. Our results confirmed the acidic nature of insulin granules under maintenance cell-culturing conditions, with an average luminal pH of ~ 5.8, and showed that acidity is actively maintained, as evidenced by its near-neutralization upon treatment with the vacuolar H⁺-ATPase inhibitor Concanamycin. Additionally, by leveraging the intrinsic spatial resolution of FLIM, we highlighted that granules proximal to the plasma membrane are slightly more acidic (~ 0.1 pH units) with respect to distal once, a difference preserved even during the early phase of glucose-induced secretion.

Plasmids construction

E¹GFP protein harbors two mutations (T65S and T203Y) compared to EGFP (11). Therefore, to generate proinsulin-E¹GFP construct, proinsulin-EGFP plasmid [13] was subjected to two rounds of site-directed mutagenesis using QuikChange XL Site-Directed Mutagenesis Kit (Agilent Technologies). The forward sequence for the T65S mutation was 5’-CCCACCCTCGTGACCACCCTGAGCTACGGCGTGCAGTGCTTC-3’, for the T203Y 5′-CAACCACTACCTGAGCTACCAGTCCGCCCTGAG-3′. For the bacterial expression of recombinant His-tagged E¹GFP, the E¹GFP sequence was amplified from proinsulin-E¹GFP construct by PCR and cloned NdeI/BamHI in pET28c + to generate pET-His-E¹GFP plasmid.

Expression and purification of recombinant E¹GFP

pET-His-E¹GFP construct was transformed in BL21 (DE3) competent cells (Invitrogen). Cells were grown at 37°C till OD600 of about 0.6, and protein expression was induced by the addition of 250 µM IPTG for 24h at 28°C. Cells were harvested by centrifugation and frozen at − 20°C. Cell pellet was resuspended in ice cold lysis buffer (40 mM Tris-HCl pH 8.0, 300 mM NaCl, 1 mM MgCl₂ supplemented with EDTA-free protease inhibitor cocktail (Roche) and DNAseI) and lysed by sonication on ice followed by 1 h treatment with 1% Triton-X100 at 4°C. Bacterial lysate was cleared by centrifugation and filtered through a 0.2 µm filter before loading onto a Bio-Scale Mini Profinity IMAC Cartridge (Bio-rad#7324614) in a fast protein liquid chromatography system (AKTAxpress, GE Healthcare). The His-tagged protein was eluted using a 0-500 mM imidazole gradient. Imidazole was then removed by buffer exchange in 20 mM diethanolamine pH = 8.5. Protein purity was evaluated by SDS-PAGE (Fig. S1a), and the concentration determined by UV absorption measurements using the extinction coefficient calculated in [15]. The absorption spectrum (Fig. S1b) was recorded at RT on a Jasco V550 spectrophotometer (JASCO, Easton, MD, USA) with the following collection parameters: band width 2 nm, scanning speed 1000 nm/min, and data resolution 1 nm.

Cell culture, transfection, and treatments

INS-1E cells were cultured in RPMI 1640 medium containing 11 mM glucose and supplemented with 10% heat-inactivated fetal bovine serum, 100 Units/ml penicillin, 100 µg/ml streptomycin, 2 mM glutamine, 10 mM HEPES, 1 mM sodium pyruvate, and 50 µM β-mercaptoethanol at 37°C in a humidified 5% CO₂ atmosphere. For live cell imaging, INS-1E cells were plated onto IbiTreat µ-Dish 35 mm (Ibidi cat #81156) and transfected the day after using Lipofectamine 2000 (Life Technologies), following the manufacturer's instructions. Cells were imaged 48h post-transfection. Drugs were supplemented to the standard culture medium as follows: concanamycin (MedChemExpress HY-N1724) was used at 100 nM for 1h at 37°C and R(+)-IAA-94 (MedChemExpress HY-12693) at 100 µM for 1h at 37°C. For glucose stimulation, proinsulin-E¹GFP transfected INS-1E cells were incubated for 45 minutes in SAB buffer (114 mM NaCl, 4.7 mM KCl, 1.2 mM KH₂PO₄, 2.5 mM CaCl2, 1.16 mM MgSO4, and 20 mM HEPES (pH 7.4)) containing 2.2 mM glucose prior to microscope acquisitions, then glucose was added to reach a final concentration of 16.7 mM.

pH sensor calibration in vitro and in living cells

Titration of E¹GFP lifetime versus pH was performed diluting recombinant E¹GFP to a final concentration of 2.1 µM in a citrate (2 mM)/phosphate (10 mM) buffer adjusted to the desired pH by addition of 1 M NaOH. WillCo-dish® Glass Bottom Dishes (WillCo Wells, cat #HBST-3522) were incubated with 2% BSA for about 40 min at 37°C to avoid protein sticking to the dish surface. For sensor calibration in living cells, proinsulin-E¹GFP transfected INS-1E cells were incubated for 10 min with an enriched K⁺ buffer containing 120 mM potassium gluconate, 40 mM sodium gluconate, 20 mM HEPES, 0.5 mM CaCl₂, 0.5 mM MgSO₄ adjusted to the desired pH with NaOH and supplemented with 10 µM nigericin. The final pH value used for the calibration curve was measured right after the FLIM acquisition.

Spectroscopic and lifetime measurements of pure low-pH and high-pH states in vitro

Absorption spectroscopy of purified recombinant E¹GFP was carried out according to the procedure reported in [15]. Almost pure A’ and A forms (see text for details) were addressed by setting the pH to 4.85 or 8, respectively by a citrate (2 mM)/phosphate (10 mM) buffer adjusted to the desired pH by addition of 1 M NaOH. Emission spectra were obtained by exciting the protein at 405 nm and collecting emission in the 450–650 nm range with 5 nm step. Lifetime decays of the protein were obtained by exciting at 405 nm and collecting emission in the 480–550 nm range. In both cases, measurements were carried out in a Leica TCS SP5 SMD inverted confocal microscope (Leica Microsystems AG, Buffalo Grove, IL) equipped with an external pulsed diode laser for excitation at 405 and 470 nm and a time-correlated single photon counting acquisition card (PicoHarp 300; PicoQuant, Berlin, Germany) connected to internal spectral detectors. Laser repetition rate was set to 40 Hz. Acquisitions lasted until 100–200 photons per pixel were collected, at a photon-counting rate of 100–500 kHz.

Two-photon microscopy, phasor‐FLIM measurements, and data analysis

Phasor-FLIM measurements were carried out on an Olympus FVMPE-RS microscope coupled with a two-photon Ti:sapphire laser with 80-MHz repetition rate (MaiTai HP, SpectraPhysics) and FLIMbox system for lifetime acquisition (ISS, Urbana Champaign). Measurements were performed at 37°C in a humidified 5% CO₂ atmosphere. E¹GFP was excited at 800 nm and the emission was collected by using a 30× planApo silicon immersion objective (NA = 1.0) in the 500–540 nm range. Calibration of the ISS Flimbox system was performed by measuring the known mono-exponential lifetime decay of Fluorescein at pH = 11.0 (i.e. 4.0 ns upon excitation at 800 nm, collection range: 500–540 nm). To prepare the calibration sample, 100 mM Fluorescein solution in EtOH was diluted 1:5000 in 0.1 M NaOH at pH 11.0. For each measurement, a 512×512 pixels image of FLIM data was collected until about 100k counts in the brightest pixel were achieved. The phasor analysis of experimental lifetime acquisitions was performed by using custom dedicated routines implemented in Python 3.6 as described in [16]. Technically: for each pixel in the image, the fluorescence decays measured in time-domain are mapped onto the so-called “phasor” plot, where a phasor has two coordinates: the real and imaginary parts of the Fourier transform of the fluorescence lifetime decay (with area under the curve normalized at 1), calculated at the angular repetition frequency of the measurement. The phasors stay within the half-disk centered at (½,0) with radius ½ and positive x, where the zero lifetime is located at (1,0) and the infinite lifetime at (0,0). This suggests that by taking the Fourier transformation of a measured decay curve, the lifetime can be estimated based on the position of the phasor inside this so-called universal (semi)circle. The distribution of phasor points originating from FLIM measurements appears on the universal circle for mono-exponential decays, or inside the circle for multi-exponential decays.

Colocalization analysis

INS-1E cells were plated on Ibi Treat µ-Dish 35 mm and transfected with proinsulin-E¹GFP plasmid as described above. 48 h post transfection, cells were incubated with 0.2 µM ZIGIR for 15 minutes and imaged on an inverted Zeiss LSM 800 confocal microscope (Jena, 439 Germany). The acquisition was performed using a 63×/NA 1.4 oil-immersion objective, setting the pinhole aperture at 1 Airy. E¹GFP and ZIGIR fluorescence were collected sequentially, illuminating the sample with a 405 and 561 nm laser and collecting the signal between 400–550 nm and 550–700 nm, respectively. Pearson’s correlation coefficient was calculated using the JaCoP plugin for ImageJ software.

Statistical analysis

After checking the normality by Shapiro-Wilk test, the statistical significance was evaluated by one-way ANOVA test with multiple comparisons.

Characterization of E¹GFP as a lifetime intracellular pH probe

Aequorea victoria fluorescent proteins (avFPs) originate from the ancestor Green Fluorescent Protein (avGFP), originally discovered by Osamo Shimomura in 1962 [17]. The chromophore of avFPs is a hyperconjugated imidazolinone moiety and comes from the autocatalytic post-translational modification of three amino acids located in positions 65–67 of the primary sequence [18]. Y66-chrom proteins, such as avGFP, contain a tyrosine residue in position 66. The protonatable phenolic function of tyrosine lateral chain makes their optical properties sensitive to pH, because the deprotonated chromophore (B state) has a much lower excitation energy than the protonated form (A state). The protonation equilibrium of some avFPs is further complicated by the presence of nearby residues. The protonation or deprotonation of these residues is thermodynamically linked with that of the chromophore. In avGFP and E¹GFP this residue is E222, and its presence leads to the splitting of the protonated form of the chromophore into the optically distinguishable states A’ and A, which refer to protonated and deprotonated E222, respectively (Fig. 1a). This pattern is conserved in the pH probe E¹GFP, which adds the F64L/T203Y mutations to the sequence of avGFP. Interestingly, it can be demonstrated that the overall set of three protonation equilibria mathematically behaves like a single-site deprotonation with a unique pK_a (2S-model) [11], [15]. For E¹GFP, pK_a=6.0 [15], and at pH < < pK_a the absorption spectrum is dominated by the single band of A’ state peaked around 410 nm (Fig. 1a). Conversely, at pH > > pK_a both the A and B states become predominant with a fixed mutual stoichiometric ratio. This is visible in the absorption spectrum as two bands, one at 402 nm (A) and the other at 509 nm (B) (Fig. 1b). Of note, the absorption of the B state of E¹GFP is strongly red-shifted compared to avGFP (+ 34 nm), on account of the π-π stack which is established between the aromatic chains of Y66 and Y203 [19], [20]. For this reason, E¹GFP belongs to the class of yellow fluorescent proteins (avYFPs). The different protonation state of E222 has a major photophysical effect on the emission of the A’ and A states, even though they can be excited at almost the same wavelength. Upon excitation the protonated chromophore undergoes Excited State Proton Transfer (ESPT) to a nearby basic residue in about 1–10 ps [21], [22]. From the excited A state, the proton is discharged to the deprotonated E222 residue. This is not feasible for the excited A’ state, which likely transfers the proton to the adjacent His148 [15]. This mechanism leads to significant differences in both the spectral emission and lifetime properties of the two states upon excitation at 405 nm (or 800 nm by two-photon excitation). The excited A state emits at 516 nm (Fig. 1b), with a monoexponential decay characterized by τ = 3.5 ns (Fig. 1c). In contrast, the excited A’ state emits at 507 nm and exhibits biexponential decay characterized by an average lifetime of < τ> = 0.83 ns (Fig. 1c). The spectral difference underlies the reported intracellular ratiometric pH sensing capability of E¹GFP [11]. Despite the even larger difference between the two states, the use of E¹GFP as a lifetime intracellular pH probe has not been reported yet. Nonetheless, Fluorescence Lifetime IMaging (FLIM) provides a valuable alternative to ratiometric sensing at intracellular level, being free of artifacts due to wavelength-dependent scattering and without the intrinsic loss of S/N due to the computed ratio between optical signals [23]. In this context, the pH sensor E²GFP, which differs from E¹GFP only by the S65T substitution, has been successfully applied to intracellular pH measurements by FLIM using the phasor approach [24]. Therefore, we set out to investigate whether E¹GFP could also be reliably used for intracellular pH measurements using FLIM.

First, we expressed and purified recombinant His-tagged E¹GFP (Fig. S1). Phasor-FLIM measurements were conducted on the purified protein diluted in solutions buffered at various pH values. (Fig. 2). Phasor analysis was carried out using cloud-based custom Python routines [16]: for each pixel, the measured fluorescence decay is Fourier transformed and plotted in the phasor plot (see Methods for more details) originating a phasor cluster whose position depends on the average fluorescence lifetime (Fig S2a). The average lifetime was calculated by determining the phasor barycenter (Fig. S2b). As expected for a single-site protonation behavior, the phasor analysis of the pH calibration revealed a linear distribution of the phasor barycenters (R² = 0.9873). We found that the barycenter for the highest pH was located near the universal circle, indicating a nearly mono-exponential decay at high pH, in agreement with the predominant excitation of the A state (Fig. 2b). At the lowest pH tested, the phasor fell inside the universal circle, witnessing the multiexponential nature of the decay with a phasor-calculated lifetime close to 1.0 ns, in keeping with the in vitro measurements of A’ state emission.

Proinsulin-E¹GFP sensor calibration in INS-1E cells

To target the E¹GFP pH reporter to the ISGs, we generated the proinsulin-E¹GFP construct in which the E¹GFP coding sequence is inserted within the C-pep of proinsulin (Fig. 3a) by site-directed mutagenesis of the previously described proinsulin-EGFP plasmid [13]. To check for proper localization at the granular level, INS-1E cells were transiently transfected with proinsulin-E¹GFP plasmid and stained with a granule-specific Zn²⁺ indicator namely ZIGIR [25]. The E¹GFP signal was enriched in dot-like structures, located both in the cytoplasm and at the plasma membrane level, and showed a good co-localization with ZIGIR (Pearson coeff. = 0.53, M1 = 0,7142, M2 = 0,3466; n = 15 cells) confirming the targeting to the ISGs (Fig. 3b). The relatively low value of Mander’s coefficient M2 was attributed to the presence of the genetically encoded protein in the biogenesis structures of the granule and in not fully mature ISGs, which are not stained by ZIGIR. To calibrate the probe in the cellular environment, INS-1E cells expressing proinsulin-E¹GFP were permeabilized with nigericin, incubated in buffers with known pH values and analyzed using FLIM (Fig. 3c). As for the recombinant protein, phasors were linearly distributed along a segment within the universal circle (R² = 0.997) (Fig. 3d).

The fluorescence lifetime as a function of pH was satisfactorily fitted to a sigmoidal model (R² = 0.9985) yielding a pKa of 5.91, this value is very close to the pKa previously measured using fluorescence intensity-based ratiometric imaging [11] (Fig. 3e). Worthy of note, the absolute values of the fluorescence lifetimes measured in living cells were lower than those obtained in bulk solutions, presumably because of the different microenvironment and crowding effects within the ISGs, that can in turn lead to fluorescence self-quenching.

Measurement of ISGs pH in standard culturing conditions

After calibration, the proinsulin-E¹GFP sensor was employed to measure the luminal pH of ISGs in INS-1E cells under standard culturing conditions (see Materials and Methods for medium supplementation details) (Fig. 4a). By calculating the phasor barycenter, an average pH of 5.79 ± 0.06 was determined for the ISGs, a value in keeping with previous indications of the slight acidity of the granular pH (see Introduction). If, instead of computing the barycenter (i.e. to determine average pH), the cluster lifetime distribution is described in its entirety using a lifetime-dependent LUT, different regions of the cell with varying lifetimes can be highlighted. We observed a slight, yet significant, difference in pH between granules located in proximity of the plasma membrane (5.73 ± 0.06) and granules within the cytoplasm (5.83 ± 0.1) (Fig. 4b). Overall, the difference in pH between granules at the membrane level and granules in the cytoplasm aligns with the what is thought to happen during granule biogenesis and trafficking from the TGN to the plasma membrane [1]. This process involves a series of maturation steps, during which acidification and removal of coat proteins coincide with the conversion of proinsulin to insulin [3], [26], [27]. It is worth noting that in this process a key role has been ascribed to vesicular ATP-dependent proton pump (V-ATPase), which increases the activity of the proinsulin converting enzymes PC1/3 and PC2 [3], [28], [29]. Acidification begins shortly after budding from the TGN and progresses over approximately 30 minutes, during this time proinsulin is converted to insulin and the protein coat is lost. The role of V-ATPase was confirmed here by probing granule pH in INS-1E cells in presence of the vacuolar H+-ATPase inhibitor Concanamycin.

As expected, Concanamycin induced a fast and robust pH shift towards neutrality (pH = 7.07 ± 0.27). Interestingly, by contrast, granule pH was not significantly affected by inhibition of chloride channels by the channel blocker R(+)-IAA-94 (pH = 5.82 ± 0.08 in presence of the drug, as compared to pH = 5.74 ± 0.09 in the control condition), suggesting that other and/or additional counterions participate to granule pH regulation (Fig. 4c).

Monitoring ISGs pH upon glucose stimulation

It should be noted that glucose stimulation is another granular pH regulator that is variably evoked in the literature, although with contrasting claims. While two independent studies suggest that glucose stimulation induces acidification of the ISG lumen [7], [9], Eto and coworkers reported that the same stimulus results in acute alkalinization of granule lumen [8]. To probe glucose-stimulation effect on granule pH, proinsulin-E¹GFP transfected INS-1E cells were first incubated at low glucose concentration for 45 min (i.e. SAB buffer supplemented with 2.2 mM glucose) and then switched to high glucose concentration, by glucose addition, to a final concentration of 16.7 mM. FLIM was performed right before glucose addition (0 min) and 5 and 10 minutes after stimulation (Fig. 5). Not surprisingly, visual inspection of the intensity images collected at different times of exposure to glucose suggests that a fraction of granules is lost due to secretion (Fig. 5a).

Despite this, the color-coded map of the cell shows that a difference in lifetime between granules in proximity of the plasma membrane and granules in the cytoplasm persists throughout the first phase of the secretion process (Fig. 5b). This is more quantitatively expressed by lifetime data extracted from single-cell segmentation of n = 11 stimulated cells (Fig. 5c). Speculatively, it might be assumed that the maintenance of a pH difference between membrane and cytoplasm during the secretory process - i.e. in the presence of a loss of membrane granules by secretion and a concomitant arrival of new granules from the cytoplasm - supports the idea that the priming of new granules from the cytoplasm to the membrane under glucose stimulation is accompanied by their rapid acidification. This in turn, would agree with at least one of the prevailing views that can be found in the literature [6], [7], [9]. More in general, this report paves the way for using phasor-FLIM combined with calibrated genetically encoded reporters to monitor granule pH at the temporal (seconds-to-minutes) and spatial (subcellular) resolution of interest for investigating insulin secretion in physiology and disease.

COMPETING INTERESTS

The authors declare no competing interests.

AUTHORS CONTRIBUTIONS

VDL: Conceptualization, Formal analysis, Investigation, Project administration, Visualization, Writing - original draft, writing - review & editing; SG: Formal analysis, Investigation, Visualization; MB: Data curation, Formal analysis, Visualization, writing - review & editing; GM: Formal analysis, Investigation; BS: Investigation; RB: Conceptualization, Writing - original draft; FC: Conceptualization, Funding acquisition, Supervision, Writing - original draft, writing - review & editing.

ACKNOWLEDGMENTS

This work has received funding from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme (grant agreement No 866127, project CAPTUR3D). The PRIN 2022 BIZZARRI_2022RRFJC4 “Novel protein-based Genetically-Encoded Fluorescent Indicators (GEFI) for Functional Super-Resolution Imaging of Biomolecular Activities in Living Cells” – GEFinder. CUP I53D23003880006 from the Italian Ministry of University and Research is gratefully acknowledged.

DATA AVAILABILITY

Data will be made available upon request to the corresponding author, Francesco Cardarelli ([email protected]).

Omar-Hmeadi, M., Idevall-Hagren, O.: Insulin granule biogenesis and exocytosis, (2021). 10.1007/s00018-020-03688-4
Germanos, M., Gao, A., Taper, M., Yau, B., Kebede, M.A.: Inside the insulin secretory granule, (2021). 10.3390/metabo11080515
Orci, L., et al.: Conversion of proinsulin to insulin occurs coordinately with acidification of maturing secretory vesicles. J. Cell Biol. 103(6) (1986). 10.1083/jcb.103.6.2273
Hutton, J.C.: The internal pH and membrane potential of the insulin-secretory granule. Biochem. J. 204(1) (1982). 10.1042/bj2040171
Pace, C.S., Sachs, G.: Glucose-induced proton uptake in secretory granules of β-cells in monolayer culture. Am. J. Physiol. Cell. Physiol. 11(3) (1982). 10.1152/ajpcell.1982.242.5.c382
Barg, S., et al.: Priming of insulin granules for exocytosis by granular CI- uptake and acidification. J. Cell. Sci. 114(11) (2001). 10.1242/jcs.114.11.2145
Stiernet, P., Guiot, Y., Gilon, P., Henquin, J.C.: Glucose acutely decreases pH of secretory granules in mouse pancreatic islets: Mechanisms and influence on insulin secretion. J. Biol. Chem. 281(31) (2006). 10.1074/jbc.M513224200
Eto, K., et al.: Glucose metabolism and glutamate analog acutely alkalinize pH of insulin secretory vesicles of pancreatic β-cells. Am. J. Physiol. Endocrinol. Metab. 285 (2003). 2 48 – 2 10.1152/ajpendo.00542.2002
Tompkins, L.S., Nullmeyer, K.D., Murphy, S.M., Weber, C.S., Lynch, R.M.: Regulation of secretory granule pH in insulin-secreting cells. Am. J. Physiol. Cell. Physiol. 283 (2002). 2 52 – 2 10.1152/ajpcell.01066.2000
Neukam, M., Sönmez, A., Solimena, M.: FLIM-based pH measurements reveal incretin-induced rejuvenation of aged insulin secretory granules. bioRxiv, (2017)
Serresi, M., Bizzarri, R., Cardarelli, F., Beltram, F.: Real-time measurement of endosomal acidification by a novel genetically encoded biosensor. Anal. Bioanal Chem. 393(4) (2009). 10.1007/s00216-008-2489-7
Ferri, G., et al.: Insulin secretory granules labelled with phogrin-fluorescent proteins show alterations in size, mobility and responsiveness to glucose stimulation in living β-cells. Sci. Rep. 9(1) (2019). 10.1038/s41598-019-39329-5
Rizzo, M.A., Magnuson, M.A., Drain, P.F., Piston, D.W.: A functional link between glucokinase binding to insulin granules and conformational alterations in response to glucose and insulin. J. Biol. Chem. 277(37) (2002). 10.1074/jbc.M112478200
Watkins, S., Geng, X., Li, L., Papworth, G., Robbins, P.D., Drain, P.: Imaging secretory vesicles by fluorescent protein insertion in propeptide rather than mature secreted peptide. Traffic. 3(7) (2002). 10.1034/j.1600-0854.2002.30703.x
Bizzarri, R., et al.: Green fluorescent protein ground states: The influence of a second protonation site near the chromophore. Biochemistry. 46(18) (2007). 10.1021/bi602646r
Bernardi, M., Cardarelli, F.: Phasor identifier: A cloud-based analysis of phasor-FLIM data on Python notebooks. Biophys. Rep. 3(4) (2023). 10.1016/j.bpr.2023.100135
Shimomura, O., Johnson, F.H., Saiga, Y.: Extraction, purification and properties of aequorin, a bioluminescent. J. Cell. Comp. Physiol. 59 (1962). 10.1002/jcp.1030590302
Nifosì, R., Storti, B., Bizzarri, R.: Reversibly switchable fluorescent proteins: ‘the fair switch project’. Rivista del. Nuovo Cimento. 47(2), 91–178 (Feb. 2024). 10.1007/s40766-024-00052-1
Jacchetti, E., Gabellieri, E., Cioni, P., Bizzarri, R., Nifosì, R.: Temperature and pressure effects on GFP mutants: Explaining spectral changes by molecular dynamics simulations and TD-DFT calculations. Phys. Chem. Chem. Phys. 18(18) (2016). 10.1039/c6cp01274d
Wachter, R.M., Elsliger, M.A., Kallio, K., Hanson, G.T., Remington, S.J.: Structural basis of spectral shifts in the yellow-emission variants of green fluorescent protein. Structure. 6(10) (1998). 10.1016/S0969-2126(98)00127-0
Storti, B., et al.: An Efficient Aequorea victoria Green Fluorescent Protein for Stimulated Emission Depletion Super-Resolution Microscopy. Int. J. Mol. Sci. 23(5) (2022). 10.3390/ijms23052482
Chattoraj, M., King, B.A., Bublitz, G.U., Boxer, S.G.: Ultra-fast excited state dynamics in green fluorescent protein: Multiple states and proton transfer. Proc. Natl. Acad. Sci. U S A. 93(16) (1996). 10.1073/pnas.93.16.8362
Battisti, A., et al.: Imaging intracellular viscosity by a new molecular rotor suitable for phasor analysis of fluorescence lifetime Optical Nanosensing in Cells. Anal. Bioanal Chem. 405(19) (2013). 10.1007/s00216-013-7084-x
Battisti, A., Digman, M.A., Gratton, E., Storti, B., Beltram, F., Bizzarri, R.: Intracellular pH measurements made simple by fluorescent protein probes and the phasor approach to fluorescence lifetime imaging. Chem. Commun. 48(42) (2012). 10.1039/c2cc30373f
Ghazvini Zadeh, E.H., Huang, Z.J., Xia, J., Li, D., Davidson, H.W., Li, W.: ZIGIR, a Granule-Specific Zn2 + Indicator, Reveals Human Islet α Cell Heterogeneity. Cell. Rep. 32(2) (2020). 10.1016/j.celrep.2020.107904
Orci, L., Ravazzola, M., Amherdt, M., Madsen, O., Vassalli, J.D., Perrelet, A.: Direct identification of prohormone conversion site in insulin-secreting cells. Cell. 42(2) (1985). 10.1016/0092-8674(85)90124-2
Orci, L., Ravazzola, M., Storch, M.J., Anderson, R.G.W., Vassalli, J.D., Perrelet, A.: Proteolytic maturation of insulin is a post-Golgi event which occurs in acidifying clathrin-coated secretory vesicles. Cell. 49(6) (1987). 10.1016/0092-8674(87)90624-6
Rhodes, C.J., Lucas, C.A., Mutkoski, R.L., Orci, L., Halban, P.A.: Stimulation by ATP of proinsulin to insulin conversion in isolated rat pancreatic islet secretory granules. Association with the ATP-dependent proton pump. J. Biol. Chem. 262(22) (1987). 10.1016/s0021-9258(18)61022-1
Davidson, H.W., Rhodes, C.J., Hutton, J.C.: Intraorganellar calcium and pH control proinsulin cleavage in the pancreatic β cell via two distinct site-specific endopeptidases. Nature. 333(6168) (1988). 10.1038/333093a0

There is NO Competing Interest.

supplementaryfigures.docx

Download PDF

Version 1

posted

You are reading this latest preprint version

Measuring pH in insulin secretory granules using phasor fluorescence lifetime imaging of a genetically encoded sensor

Status:

Version 1

Abstract

Figures

INTRODUCTION

MATERIALS AND METHODS

Plasmids construction

Expression and purification of recombinant E¹GFP

Cell culture, transfection, and treatments

Spectroscopic and lifetime measurements of pure low-pH and high-pH states in vitro

Two-photon microscopy, phasor‐FLIM measurements, and data analysis

Colocalization analysis

Statistical analysis

RESULTS AND DISCUSSION

Characterization of E¹GFP as a lifetime intracellular pH probe

Proinsulin-E¹GFP sensor calibration in INS-1E cells

Measurement of ISGs pH in standard culturing conditions

Monitoring ISGs pH upon glucose stimulation

Declarations

COMPETING INTERESTS

AUTHORS CONTRIBUTIONS

ACKNOWLEDGMENTS

DATA AVAILABILITY

References

Additional Declarations

Supplementary Files

Status:

Version 1

Measuring pH in insulin secretory granules using phasor fluorescence lifetime imaging of a genetically encoded sensor

Status:

Version 1

Abstract

Figures

INTRODUCTION

MATERIALS AND METHODS

Plasmids construction

Expression and purification of recombinant E1GFP

Cell culture, transfection, and treatments

Spectroscopic and lifetime measurements of pure low-pH and high-pH states in vitro

Two-photon microscopy, phasor‐FLIM measurements, and data analysis

Colocalization analysis

Statistical analysis

RESULTS AND DISCUSSION

Characterization of E1GFP as a lifetime intracellular pH probe

Proinsulin-E1GFP sensor calibration in INS-1E cells

Measurement of ISGs pH in standard culturing conditions

Monitoring ISGs pH upon glucose stimulation

Declarations

COMPETING INTERESTS

AUTHORS CONTRIBUTIONS

ACKNOWLEDGMENTS

DATA AVAILABILITY

References

Additional Declarations

Supplementary Files

Status:

Version 1

Expression and purification of recombinant E¹GFP

Characterization of E¹GFP as a lifetime intracellular pH probe

Proinsulin-E¹GFP sensor calibration in INS-1E cells