Intracellular pH imaging in cancer cells in vitro and tumors in vivo using the new genetically encoded sensor SypHer2

    Intracellular pH imaging in cancer cells in vitro and tumors in vivo using the new genetically encoded sensor SypHer2 Marina V. Shirmanova, Irina N. Druzhkova, Maria M. Lukina, Mikhail E. Matlashov, Vsevolod V. Belousov, Ludmila B. Snopova, Natalia N. Prodanetz, Varvara V. Dudenkova, Sergey A. Lukyanov, Elena V. Zagaynova PII: DOI: Reference:

S0304-4165(15)00126-9 doi: 10.1016/j.bbagen.2015.05.001 BBAGEN 28202

To appear in:

BBA - General Subjects

Received date: Revised date: Accepted date:

11 November 2014 30 April 2015 4 May 2015

Please cite this article as: Marina V. Shirmanova, Irina N. Druzhkova, Maria M. Lukina, Mikhail E. Matlashov, Vsevolod V. Belousov, Ludmila B. Snopova, Natalia N. Prodanetz, Varvara V. Dudenkova, Sergey A. Lukyanov, Elena V. Zagaynova, Intracellular pH imaging in cancer cells in vitro and tumors in vivo using the new genetically encoded sensor SypHer2, BBA - General Subjects (2015), doi: 10.1016/j.bbagen.2015.05.001

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT Intracellular pH imaging in cancer cells in vitro and tumors in vivo using the new genetically encoded sensor SypHer2 Marina V. Shirmanova1,2, Irina N. Druzhkova1, Maria M. Lukina1,2, Mikhail E. Matlashov3,

Nizhny Novgorod State Medical Academy, 10/1 Minin and Pozharsky Sq., 603005 Nizhny

Novgorod, Russia 2

Lobachevsky State University of Nizhny Novgorod, 23 Gagarin Ave., 603950 Nizhny

NU

Novgorod, Russia 3

IP

1

SC R

Sergey A. Lukyanov1,3, and Elena V. Zagaynova1,2

T

Vsevolod V. Belousov1,3, Ludmila B. Snopova1, Natalia N. Prodanetz1, Varvara V. Dudenkova2,1,

Shemyakin–Ovchinnikov Institute of Bioorganic Chemistry RAS, 16/10 Miklukho-Maklaya St.,

MA

117997 Moscow, Russia

Key words: intracellular pH, genetically encoded sensor, SypHer2, cancer cell, tumor,

TE

D

ratiometric imaging

Corresponding author: e-mail [email protected], Phone: +7 831 465 5672, Fax: +7 831 465

CE P

4113

Background: Measuring intracellular pH (pHi) in tumors is essential for the monitoring of cancer

AC

progression and the response of cancer cells to various treatments. The purpose of the study was to develop a method for pHi mapping in living cancer cells in vitro and in tumors in vivo, using the novel genetically encoded indicator, SypHer2. Methods: A HeLa Kyoto cell line stably expressing SypHer2 in the cytoplasm was used, to perform ratiometric (dual excitation) imaging of the probe in cell culture, in 3D tumor spheroids and in tumor xenografts in living mice. Results: Using SypHer2, pHi was demonstrated to be 7.34 ± 0.11 in monolayer HeLa cells in vitro under standard cultivation conditions. An increasing pHi gradient from the center to the periphery of the spheroids was displayed. We obtained fluorescence ratio maps for HeLa tumors in vivo and ex vivo. Comparison of the map with the pathomorphology and with hypoxia staining of the tumors revealed a correspondence of the zones with higher pHi to the necrotic and hypoxic areas.

ACCEPTED MANUSCRIPT Conclusions: Our results demonstrate that pHi imaging with the genetically encoded pHi indicator, SypHer2, can be a valuable tool for evaluating tumor progression in xenograft models.

General significance: We have demonstrated, for the first time, the possibility of using the

T

genetically encoded sensor SypHer2 for ratiometric pH imaging in cancer cells in vitro and in

SC R

provide high accuracy and spatiotemporal resolution.

IP

tumors in vivo. SypHer2 shows great promise as an instrument for pHi monitoring able to

1. Introduction

Intracellular pH (pHi) is known to be an important regulator of many cell functions. In normal

NU

cells, intracellular pH is lower than extracellular pH (pHe), with the pHi and pHe values lying mostly in the range 7.0–7.2 and 7.3–7.4 respectively. Cancer cells are generally associated with

MA

higher values of pHi 7.12 -7.65 and lower pHe 6.2–6.9. Such a reversed intra-extracellular pH gradient is considered to be a hallmark of neoplastic tissue, assisting the progression of the cancer [1, 2]. An elevated pHi allows cell proliferation and the evasion of apoptosis, provokes a

D

metabolic switch from oxidative phosphorylation to aerobic glycolysis (Warburg effect) and

TE

possibly promotes genetic instability and the multidrug resistance (MDR) of cancer cells. An increased pHi and a decreased pHe coordinately enhance invasion and metastasis. Therefore it

CE P

appears that the measurement of pHi in tumors could represent an important method for monitoring the progression of cancers and the responses of cancer cells to various treatments. Given the significant role of pHi in tumor development, it is crucially important to be able

AC

to measure it with high accuracy and spatiotemporal resolution. Fluorescence imaging based on pH-sensitive fluorescent probes offers excellent opportunities as a highly-sensitive, low cost technique for real-time, non-invasive pH determination in cells and tissues. The majority of commercially available probes are small organic molecules that have to be introduced into the cells or tissues exogenously. The most widely used of these are derivatives of fluorescein (BCECF, BCPCF, fluorescein, fluorescein sulfonic acid, carboxyfluorescein) and benzoxanthene (SNAFL, SNAFR, SNARF) [3]. Despite the variety of pH sensitive fluorescent dyes available, measuring intracellular pH remains problematic, especially in living tissues. These shortcomings include problems with the intracellular delivery; with self-redistribution of the dyes and their leakage from the cells; their interactions with other molecules in the cell, and their own cytotoxicity [4]. As a result, the current applications for such synthetic probes are limited to the assessment of pHi in cell cultures, in dissociated spheroids and tissues and in tissue slices following incubation with the probe [5-9].

ACCEPTED MANUSCRIPT Although some of these dyes are able to enter cells in vitro, the mapping of pHi in solid tumors in vivo remains impracticable and, so far, only pHe measurements in tumors have been carried out with the use of this type of synthetic probe. For example, Mordon et al. ratiometrically measured pHe in vivo in subcutaneously grafted lymphoid leukemia P388 tumors

T

in mice, using 5,6-carboxyfluorescein, 5,6-CF [10]. Robey et al. determined pHe in MDA-MB-

IP

231 tumor xenografts following the injection of SNARF-1 free acid [11]. In the paper by Hight

SC R

et al. pHe was quantified with SNARF-4F in dye-perfused surgically-resected tumor specimens [12].

Therefore, it is desirable both to engineer new indicators for intracellular pH measurement and to develop appropriate methods for their use in vivo. In this context, pH-

NU

reporters based on green fluorescent protein (GFP) represent promising instruments for overcoming the limitations of the synthetic dyes [13, 14]. Being genetically encoded, they can be

MA

directed to any particular compartment within a cell, or allowed to distribute themselves by diffusion within the cytosol. Consequently, they enable the subcellular measurement of pH with unrivalled specificity. The stable expression of GFP-based probes opens up possibilities for

D

continuous pH monitoring in living cells and tissues.

TE

In this issue, Matlashov et al. [15] report on an improved pH-sensitive ratiometric indicator SypHer2 based on the cpYFP fluorophore. It was generated from the H2O2 indicator

CE P

HyPer-2 [16] by mutation of the H2O2-sensing cysteine residue to serine (supplemental Fig. S1), causing a total loss of sensitivity to H2O2. SypHer2 has two excitation peaks, at 420 nm and 500 nm, and one emission peak at 516 nm, similar to HyPer [17], however, in mammalian cells,

AC

compared to SypHer [18], it has a 2 to 3 fold brighter fluorescence signal. The purpose of the study was to develop a method for pHi sensing in living cancer cells in vitro and tumor xenografts in vivo, using this new genetically encoded indicator SypHer2. Ratiometric (dual excitation) imaging of the probe was performed in cell culture, in 3D tumor spheroids and in tumor xenografts in living mice. Quantitative assessment of the in vitro pHi was made possible by using a calibration curve. Additionally, histopathology and hypoxia in the tumors were characterized, to assist in interpreting the ratiometric imaging data.

2. Materials and Methods 2.1. SypHer2 spectral characterization The SypHer2 protein was extracted and purified from competent E.coli XL1-Blue cells transfected with pQE30-SypHer2 plasmids. Individual bacterial clones were picked and grown overnight on solid LB-agar plates at 37oC followed by 24 hrs at room temperature. Cells were lysed using B-PER reagent (Thermo Fisher Scientific), then equal aliquotes of the cell lysates

ACCEPTED MANUSCRIPT were added to buffers (25 mM Tris-HCl и 150 mM NaCl) with pH values 7.1, 7.6, 8.1, 8.6, 9.0. Fluorescence excitation spectra were recorded using Cary Eclipse (Varian) fluorimeter in 350510 nm range (530 nm emission). Fluorescence emission spectra in 440-620 nm range were recorded with 420 nm excitation. The obtained excitation and emission spectra of SypHer2 are

T

presented in Fig. 1A.

IP

Measuring of the excitation spectra at different pH showed that the excitation peak at 420

SC R

nm decreases with pH proportionally to the increase in the peak at 500 nm, allowing ratiometric measurement of pH (Fig. 1B).

2.2. Monolayer and 3D cell cultures

NU

A HeLa Kyoto cell line, stably expressing SypHer2 (HeLa-SypHer2) in the cytosol, was used. The cells were cultured in DMEM supplemented with 10% FBS (Hyclone), 2 mM glutamine

MA

(PanEco), 10 U/ml penicillin and 10 mg/ml streptomycin. For fluorescence microscopy the cells were seeded (1x105 in 2 mL) into glass-bottomed 35 mm FluoroDishes, and incubated overnight at 37°C, 5% CO2 and 80% relative humidity. Then the cells were washed with PBS and placed

D

into Henk's solution for imaging.

TE

The protocol for 3D cultures (tumor spheroids) was modified from the plate manufacture's protocol. A cell suspension of HeLa-SypHer2 cells in complete growth medium

CE P

was seeded into the wells of a round bottom 96-well Ultra-Low Attachment plate (Corning, USA) at a density of 100-150 cells/200 µL/well. Complete culture medium was added on day 3. Once spheroid formation had been completed, on the day 7 they were carefully washed with PBS

AC

and transferred into glass-bottom FluoroDishes with PBS for investigation.

2.3. Animals and tumors All animal protocols were approved by the Ethics Committee of Nizhny Novgorod State Medical Academy. Experiments were performed on female athymic nude mice purchased from the Pushchino animal nursery (Pushchino, Russia). Mice of 20-22 g body weight were inoculated subcutaneously in the left flank with HeLa cells expressing cytosolic SypHer2 (2x106 in 200 µL PBS). Imaging started 7 days after the cell injection, when the tumors had reached 3–4 mm in diameter. The animals were monitored on days 7, 10, 14 and 18. Before fluorescence imaging the mice were anesthetized intramuscularly with a mixture of Zoletil (40 mg/kg, 50 µL, Virbac SA, Carros, France) and 2% Rometar (10 µL and 10 mg/kg, Spofa, Czech Republic). To reduce signal attenuation, a skin flap over the tumor was repeatedly opened for image acquisition and closed with a 6-0 suture immediately afterwards. On day 18 the animals were sacrificed by

ACCEPTED MANUSCRIPT cervical dislocation and the tumors were excised. The total number of animals in the study was 10.

2.4. Fluorescence microscopy

T

The fluorescence intensity of the monolayer cell cultures was measured on an inverted Nikon

IP

Eclipse TI N-STORM microscope (Nikon, Japan). Epi-fluorescence mode with a BV-2A filter

SC R

(Ex: 400-440 nm, Em: 470 nm) and a YFP filter (Ex: 490-500 nm, Em: 520-560 nm) was used for the detection of SypHer2 fluorescence.

For the tumor spheroids, those measuring ~200 µm were analyzed on day 7 after seeding, using an inverted Carl Zeiss LSM 710 laser scanning confocal microscope (Carl Zeiss,

NU

Germany). An argon laser with a diffraction grating was used for fluorescence excitation at wavelengths of 405 nm and 488 nm. Emissions were recorded in the range 435 - 689 nm for

MA

excitation at 405 nm, and at 509 – 689 nm for excitation at 488 nm. Twelve randomly selected spheroids were examined. Starting from one pole of each spheroid, confocal z-stacks were obtained through them with confocal slice thicknesses of 1.2 μm and a step size of 2.47 μm

D

between adjacent optical planes.

TE

After animal sacrifice, the tumors were excised, cut in two parts and immediately placed into liquid nitrogen. Each piece of tumor was then embedded into O.C.T. compound (Optical

CE P

Cutting Temperature, Tissue-Tek, Sakura Finetek) on a cryotome holder. Prior to cryosectioning the tumor pieces were brought to -20°C and allowed to equilibrate for 30 minutes. The frozen samples were cut into 20 µm thick slices, placed on glass slides and immediately imaged on a

AC

Leica DMIL (Leica, Germany) microscope using both CFP ET (Ex: BP 436/20, Em: BP 480/40) and YFP ET (Ex: BP 500/20, Em: BP 535/30) filtration. The fluorescence images were processed with ImageJ 1.39p software (NIH, USA). The background signal, taken from an empty region, was subtracted from the measurements, and the ratio of emission intensity resulting from excitation at the two wavelengths was calculated (I500/I420). 2.5. pH calibration in vitro The protocol for pH calibration was modified from Ref. [19]. A set of buffer solutions containing 130 mM KGluconate, 2 mM CaCl2, 1mM MgCl2, 10 µM nigericin and 30 mM MOPS was prepared with pH values of 6.7, 6.9, 7.1, 7.3, 7.5, 7.7, 8.0. The pH was adjusted with 1M KOH or 1M HCl at room temperature (the registration of fluorescence was also performed at this same temperature). The cells were incubated with buffer for 5 minutes, and then the fluorescence was registered as described above. For each pH value the fluorescence signal from 5 fields of view

ACCEPTED MANUSCRIPT was acquired, with the fluorescence ratio being calculated for 5-10 cells for each field of view in order to obtain an averaged value of signal. Since two different microscope types were used for monolayers and for spheroids, separate calibration curves were constructed for them.

T

2.6. Fluorescence imaging of tumors in vivo

IP

The tumor-bearing mice were imaged in vivo using an IVIS-Spectrum imaging system (Caliper

SC R

Life Sciences, USA). Fluorescence was excited at two wavelengths 430±15 nm (I430) and 500±15 nm (I500), and detected using a 530–550 nm band filter. The images were processed with ImageJ 1.39p software as described above.

NU

2.7. Histopathology

20 µm tumor tissue cryosections were stained with hematoxylin and eosin (H&E) and examined

MA

with a Leica DM2500 microscope. The histological structures were matched with the fluorescence signals recorded for the corresponding sections.

D

2.8. Hypoxia assessment

TE

For the investigation of tumor tissue hypoxia, mice were intravenously injected with Hypoxyprobe™-1 (pimonidazole HCl) solution at a dosage of 60 mg/kg body weight, 40 min

CE P

before tumor excision. 20 µm cryosections were stained with FITC-conjugated IgG1 mouse monoclonal antibodies diluted 1:100, according the manufacturer’s protocol.

AC

2.9. Statistical analysis

The mean values (M) and standard deviations (SD) were calculated for the results of pH measurements for the cells and spheroids, and for the results of the SypHer2 fluorescence ratio measurements of tumors. The Student’s t-test and the one-way ANOVA with Fisher’s post-hoc test were used to compare the data (p ≤ 0.05 was considered statistically significant). To construct the curve fit for the pH calibration data the least-squares method was used in Microsoft Excel.

3. Results 3.1. pH registration in monolayer cell culture To test the sensitivity of SypHer2 to pH changes and to calibrate the signal, their fluorescence at two wavelengths, 500 nm and 420 nm, and the corresponding I500/I420 ratios were registered in monolayered HeLa cells in buffer solutions with known pH values. In vitro calibration was performed over the pH range 6.7-8.0 using MOPS buffers and nigericin. As expected, with

ACCEPTED MANUSCRIPT increasing pH, the fluorescence excited at 420 nm decreased while that at 500 nm increased, resulting in an increase of the ratio of intensities (Fig. 2A). The relationship between the measured fluorescence ratio and the pH was well approximated by an exponential fit (r2 = 0.99, the solid line in Fig. 2B).

T

The resulting calibration curve was used to determine the pHi values in HeLa cells in

IP

vitro (Fig. 2C). It was found to be 7.34 ± 0.11 (mean±SD, n=40) under standard cultivation

SC R

conditions.

3.2. pH mapping in tumor spheroids

It is widely recognized that 3D cellular systems, and spheroids, are biologically more relevant to

NU

solid tumors in vivo than are conventional 2D cell cultures, as they exhibit some of the morphological and physiological properties of the tumors and mimic the in vivo

MA

microenvironment. For example, spheroids show heterogeneous expression of the membrane receptors that regulate cell adhesion and metabolism, produce an intercellular matrix, and have oxygen and nutrient gradients that lead to the formation of a necrotic core and to different

D

proliferative activity [20, 21]. Due to proliferation and to the pO2 gradients, the pH in spheroids

TE

may also vary in different areas [8, 9]. To examine these pHi gradients in HeLa spheroids, we used the genetically encoded pH indicator SypHer2.

CE P

The spheroids were generated from HeLa cells stably expressing SypHer2. To convert the ratio signals to pH values in the spheroids we used the appropriate calibration curve obtained for monolayered cells. In a preliminary study using a confocal microscope we measured the

AC

attenuation of the probing light as it passed through the spheroid and found that violet and blue light intensities attenuate equally, at least, up to the depth of ~90 µm (data not shown). However, at the depths greater than 50 µm signal-to-background ratio did not exceed 10 for both wavelengths. Therefore, to avoid any distortion of the ratio owing to the poor permeability to light, the confocal scanning of the spheroids and, consequently, the measurement of pH, was limited to this 50 µm depth. Fig. 2D shows a pHi gradient from the core of the living spheroid to its periphery. Quantitative analysis of the pHi values revealed that the pHi in peripheral parts of the spheroid was 7.43±0.06, while in the central areas it was 7.23±0.04 (p=0.00000, n=12). Lower pHi in the cores of the spheroids may be a result of the hypoxia-induced accumulation in the cell cytoplasm of metabolic by-products, mostly lactate.

3.3. pH mapping of tumors in vivo

ACCEPTED MANUSCRIPT Noninvasive pH measurement of tumors in vivo by optical methods is a complicated and challenging task due to the extent of light absorption and scattering by the tissues. Another problem is the delivery of the probe to the tumor cells. Therefore ratiometric genetically encoded indicator stably expressed in the cancer cells has the potential to be a valuable tool for in vivo pH

T

monitoring.

IP

We induced tumors in mice using HeLa cells stably expressing SypHer2. In vivo

SC R

fluorescence imaging was performed to visualize the SypHer2 emissions at 520 nm, sequentially excited at 430 and 500 nm.

Initial experiments showed that the skin above the tumor significantly reduces the level of detected SypHer2 fluorescence in both channels. Furthermore, as blue light (500 nm) penetrates

NU

through the skin better than violet (430 nm), this means that this effect will bias the measured I500/I430 ratio upwards. Therefore, the resulting SypHer2 ratio measured from the intact

MA

subcutaneous tumor does not reflect the actual SypHer2 signal. To overcome this obstacle, a skin flap over the tumor was surgically opened for image acquisition, and the fluorescence was detected directly from the tumor surface.

D

Calculation of the I500/I430 ratio showed that the SypHer2 signal in the HeLa tumor was

TE

highly heterogeneous, indicating differences in pHi values across the tumor (Fig. 3A). The spatial distribution of the signal patterns was specific to each particular tumor. Our data support

CE P

the notion that tumors are phenotypically and functionally heterogeneous, and this heterogeneity can be observed both between different tumors and within individual tumors [22]. Heterogeneous pH distribution in solid tumors has been demonstrated previously using, for

AC

example, fluorescence imaging with 5,6-CF [10] or MRI [23]. Macroscopically, the tumors had a multinodular, fleshy appearance with blood vessels, plus yellowish and red-colored areas (Fig. 3A). Analysis of the I500/I430 ratio map revealed a spatial gradient of pH within each separate tumor nodule, with a lower ratio (more acidic pH) in the center, similar to that in the spheroid model. It is probable that the acidic pHi in the nodule center is a result of the increased metabolism of glucose and poor blood perfusion. Comparison of the mapped ratios with the conventional histopathology showed a correspondence of the areas with highest ratio (more alkaline pH) to those showing necrosis (Fig. 3A). Quantitative estimation of the ratio signals in 10 tumors, each containing 3-6 nodules, displayed statistically significant differences between the central and peripheral parts of the nodules and the necrotic zones. As the tumors grew, the heterogeneity of the signals increased, but without any notable changes in their spatial distribution (Fig. 3B).

ACCEPTED MANUSCRIPT 3.4. Histopathology and hypoxia in tumors Since pH could influence the viability of tumor cells, and changes in pH values are associated with both the oxygen consumption and the metabolic state of the cells, histopathological investigation and hypoxia analysis of SypHer2 expressing tumors were performed. Fresh

T

cryosections of the tumors were examined for SypHer2 fluorescence and stained with H&E,

IP

while additional subsequent cryosections were stained with pimonidazole. In a preliminary study

SC R

on tumor spheroids we showed that cryofixation does not change spatial distribution of SypHer2 ratio (supplemental Fig. S2).

We observed higher I500/I420 ratios within necrotic areas and hypoxic zones, and lower

NU

ratios – within vital tumor tissue (Fig. 4).

4. Discussion

MA

This work was the first application of the genetically encoded pH probe SypHer2 for assessing pHi distribution in cultured cancer cells and tumors. Various genetically encoded pHi indicators have been developed so far [13, 14], however

D

in cancer studies, their use has previously only been recorded in cultured cells. For instance,

TE

Bizzarry et al. developed the ratiometric excitation and emission pH indicator E2GFP and applied it for pHi probing in U-2 OS cells [24]. Similarly, EYFP [25] and cytoSypHer [18] have

CE P

been used to measure cytosolic pH in HeLa cells. Owing to the possibility of the subcellular targeting of FPs and FP-based sensors by fusing them to any protein of interest, the analysis of pH within cellular organelles became

AC

possible with the introduction of GFP-based pH indicators. A number of GFP mutants, such as EYFP-mito [25], MitoSypHer [18], mtAlpHi [26] and GFP-pH [27] have been engineered to monitor the pH of the mitochondrial matrix and their use in HeLa cells has also been demonstrated. There are also examples of pH-sensitive FPs targeted to the Golgi apparatus. Using a Golgi-targeted green fluorescent protein GT-EGFP as a probe, Rivinoja et al. demonstrated the presence of a more alkaline pH in the medial/trans-Golgi in cancer cell lines, MCF-7, HT-29 and SW-48 [28]. Golgi-associated GT-EGFP/GT-ECFP or GT-EYFP/GT-ECFP have been used in HeLa cells to examine the contribution of actin dynamics to Golgi pH homeostasis [29]. Serresi et al. have described an application of E1GFP fused to the HIV-Tat protein in the measurement of pH changes along the endo-lysosomal pathway in HeLa cells [30]. Since animal tissues have poor permeability to light, pHi sensing with genetically encoded indicators in tissues in vivo remains challenging, and to our knowledge, it has not previously been implemented in the more complicated models of cancer, such as tumor spheroids and animal tumors. For this purpose, the new optimized GFP-based ratiometric pH

ACCEPTED MANUSCRIPT indicators are required, with absorption and fluorescence in the optical window, and with enhanced brightness and broad dynamic ranges. It should be noted that ratiometric indicators represent the most useful class of fluorescent sensors because they are relatively independent of such factors as optical path length, focusing, excitation intensities, emission collection

T

efficiencies, probe concentration, and photobleaching, in contrast to the intensiometric indicators.

IP

SypHer2, as used in this study, is a new ratiometric pH sensor with increased brightness

SC R

compared to its parental analogue, SypHer (HyPer-C199S), engineered by Poburko et al. [18]. Using SypHer2, we determined the pHi values in HeLa cells grown in monolayered culture. The pHi value of 7.34 ± 0.11 that we obtained using SypHer2 agreed well with cytosolic pH values reported for HeLa cells: 7.33 ± 0.13 [25], ~7.4 [18, 31], but was slightly higher than

NU

the 7.11±0.12 demonstrated in another report [32].

In our study, a core-periphery difference in the fluorescence ratio of SypHer2 was

MA

recorded in living HeLa spheroids, indicating a spatial pHi gradient, with the lower pHi in the core. Similar results had previously been demonstrated in Ref. 7-9, where synthetic pH probes were used. Using carboxy-SNARF-1, Swietach et al. found that spheroids of RT112 bladder

D

carcinoma cells have an acidic core with a pHi 0.25 units lower than at the surface [7]. With

TE

living OvCa nodules, pH imaging using SNARF-4F has demonstrated a pH gradient along their radii, with more acidic cores being observed consistently for both small and large spheroids [8].

CE P

Hulikova et al. also studied pHi in spheroids loaded with carboxy-SNARF-1 and showed 0.1-0.2 units difference in pHi between the center and the periphery for both HT29 and HCT116 spheroids, with the more acidic pH in the core [9]. However, in our experiments, the very limited

AC

penetration of the probing light used for excitation of SypHer2 did not allow the ratiometric measurement of pHi from deeper than 50 µm in living spheroids. Using SypHer2 we obtained a fluorescence ratio map for pHi in HeLa tumors in vivo and ex vivo. In solid tumors in vivo, translation of the fluorescence ratios into pH units requires careful correction, given the optical properties of the tumor tissue, in particular, the different extent of penetration and attenuation of the excitation illumination and of the fluorescence emissions of different wavelengths. The apparent heterogeneity of the SypHer2 ratios in the in vivo images therefore could theoretically be explained by differential light absorption and/or scattering in different areas of the object. However, the correlation of the in vivo ratio mapping with that of the ex vivo tissue sections, where these factors do not impact on the ratio, allows us to conclude that the observed heterogeneity results from differences in pHi. The development of ratiometric pH indicators working in the red range or fluorescence lifetime based pH sensors represents possible solutions for measuring pH in living tissues. The

ACCEPTED MANUSCRIPT pH-dependence of the SypHer2 fluorescence lifetime and the development of a FLIM-based method of pH analysis in cells in vitro and in vivo will be the subject of further research. Comparison of the fluorescence ratio map with pathomorphology and hypoxia staining in the tumors revealed a correspondence of the zones with higher pHi to the necrotic and hypoxic

T

areas. Hypoxia, a common feature of solid tumors, is known to be responsible for inducing

IP

acidosis in tumors through a shift from oxidative phosphorylation to glycolytic metabolism,

SC R

which generates a high acid load in the tumor microenvironment [33]. However, prolonged deprivation of oxygen leads to chronic hypoxic stress and to necrotic cell death, so necrosis is also a common characteristic of tumors. Using microelectrode techniques, it was shown in the 1980s, that in those areas with gross, presumably long-established necrosis, the pH shifts

NU

towards more alkaline values [34, 35]. This happens probably due to a lack of formation of acidic metabolites once the glycogen stores are exhausted. As a consequence, necrotic cells

MA

demonstrate acidophilic (eosinophilic) staining.

In summary, we have shown, for the first time, the possibility of using the genetically encoded sensor SypHer2 for ratiometric pHi imaging in cancer cells in vitro and for tumors in

D

vivo. Our results demonstrate that pH imaging can be a valuable tool for evaluating metabolic

CE P

Acknowledgements

TE

state of tumor and progression in xenograft models.

This work was supported by the Russian Science Foundation (project # 14-15-00646). The authors are grateful to Natalia Klementieva for her kind assistance with the Nikon Eclipse TI N-

AC

STORM microscope.

References

[1] M. Damaghi, J.W. Wojtkowiak, R.J. Gillies, pH sensing and regulation in cancer, Front Physiol. 4 (2013) 370. [2] B.A. Webb, M. Chimenti, M.P. Jacobson, D.L. Barber, Dysregulated pH: a perfect storm for cancer progression, Nat Rev Cancer. 11 (2011) 671-677. [3] pH Indicators, chapter 20 in The molecular probes® handbook: A Guide to Fluorescent Probes and Labeling Technologies, 11th Edition (2010). [4] J. Han, K. Burgess, Fluorescent indicators for intracellular pH, Chem Rev. 110 (2010) 27092728. [5] C. Martin, S. H. F. Pedersen, A. Schwab, C. Stock, Intracellular pH gradients in migrating cells, American Journal of Physiology: Cell Physiology, 300 (2011) C490-C495.

ACCEPTED MANUSCRIPT [6] M.J. Boyer, M. Barnard, D.W. Hedley, I.F. Tannock, Regulation of intracellular pH in subpopulations of cells derived from spheroids and solid tumours // Br. J. Cancer 68 (1993) 890-897. [7] P. Swietach, S. Wigfield, P. Cobden, C.T. Supuran, A.L. Harris, R.D. Vaughan-Jones,

T

Tumor-associated carbonic anhydrase 9 spatially coordinates intracellular pH in three-

IP

dimensional multicellular growths // J. Biol. Chem. 283 (2008) 20473-20483.

SC R

[8] C.L. Evans, A.O. Abu-Yousif, Y.J. Park, O.J. Klein, J.P. Celli, I. Rizvi, X. Zheng, T. Hasan, Killing hypoxic cell populations in a 3D tumor model with EtNBS-PDT, PLoS ONE 6 (2011) e23434.

[9] A. Hulikova, R.D. Vaughan-Jones, P. Swietach, Dual role of CO2/HCO3(-) buffer in the

NU

regulation of intracellular pH of three-dimensional tumor growths, J Biol Chem. 286 (2011) 13815-13826.

MA

[10] S. Mordon, J.M. Devoisselle, V. Maunoury, In vivo pH measurement and imaging of tumor tissue using a pH-sensitive fluorescent probe (5,6-carboxyfluorescein): instrumental and experimental studies, Photochem Photobiol. 60 (1994) 274-279.

D

[11] I.F. Robey, B.K. Baggett, N.D. Kirkpatrick, D.J. Roe, J. Dosescu, B.F. Sloane, A.I.

TE

Hashim, D.L. Morse, N. Raghunand, R.A. Gatenby, R.J. Gillies, Bicarbonate increases tumor pH and inhibits spontaneous metastases, Cancer Res 69 (2009) 2260-2268.

CE P

[12] M.R. Hight, D.D. Nolting, E.T. McKinley, A.D. Lander, S.K. Wyatt, M. Gonyea, P. Zhao, H.C. Manning, Multispectral fluorescence imaging to assess pH in biological specimens, Journal of Biomedical Optics 16 (2011) 016007.

AC

[13] R. Bizzarri, M. Serresi, S. Luin, F. Beltram, Green fluorescent protein based pH indicators for in vivo use: a review, Anal Bioanal Chem 393 (2009) 1107–1122. [14] M. Benčina, Illumination of the spatial order of intracellular pH by genetically encoded pHsensitive sensors, Sensors 13 (2013) 16736-16758. [15] M. Matlashov et al. Fluorescent ratiometric pH indicator SypHer2: applications in neuroscience and regenerative biology, BBA General Subjects, 2015 [16] K.N. Markvicheva, D.S. Bilan, N.M. Mishina, A.Yu. Gorokhovatsky, L.M. Vinokurov, S. Lukyanov, V.V. Belousov, A genetically encoded sensor for H2O2 with expanded dynamic range, Bioorganic & Medicinal Chemistry 19 (2011) 1079–1084. [17] V.V. Belousov, A.F. Fradkov, K.A. Lukyanov, D.B. Staroverov, K.S. Shakhbazov, A.V. Terskikh, S. Lukyanov, Genetically encoded fluorescent indicator for intracellular hydrogen peroxide, Nat Methods. 3 (2006) 281-286. [18] D. Poburko, J. Santo-Domingo, N. Demaurex, Dynamic regulation of the mitochondrial proton gradient during cytosolic calcium elevations, J. Biol. Chem. 286 (2011) 11672-11684.

ACCEPTED MANUSCRIPT [19] J.A. Thomas, R.N. Buchsbaum, A. Zimniak, E. Racker, Intracellular pH measurements in Ehrlich ascites tumor cells utilizing spectroscopic probes generated in situ, Biochemistry, 10 (1979) 2210-2218. [20] F. Hirschhaeuser, H. Menne, C. Dittfeld, J. West, W. Mueller-Klieser, L.A. Kunz-Schughart,

T

Multicellular tumor spheroids: An underestimated tool is catching up again, J. Biotechnol.

IP

148 (2010) 3-15.

SC R

[21] R. Chignola, A. Del Fabbro, M. Farina, E. Milotti, Computational challenges of tumor spheroid modeling, J. Bioinform. Comput. Biol. 09 (2011) 559-77. [22] C.E. Meacham, S.J. Morrison, Tumour heterogeneity and cancer cell plasticity, Nature 501 (2013) 328–337

NU

[23] G. Liu, Y. Li, V.R. Sheth, M.D. Pagel, Imaging in vivo extracellular pH with a single paramagnetic chemical exchange saturation transfer magnetic resonance imaging contrast

MA

agent, Molecular Imaging 11 (2012) 47–57.

[24] R. Bizzarri, C. Arcangeli, D. Arosio, F. Ricci, P. Faraci, F. Cardarelli, F. Beltram, Development of a novel GFP-based ratiometric excitation and emission pH indicator for

D

intracellular studies, Biophysical Journal 90 (2006) 3300–3314.

TE

[25] J. Llopis, J. M. McCaffery, A. Miyawaki, M.G. Farquhar, R.Y. Tsien, Measurement of cytosolic, mitochondrial, and Golgi pH in single living cells with green fluorescent proteins,

CE P

Proc. Natl. Acad. Sci. USA. 95 (1998) 6803–6808. [26] M.F.C. Abad, G. Di Benedetto, P.J. Magalhães, L. Filippin, T. Pozzan, Mitochondrial pH monitored by a new engineered green fluorescent protein mutant, J. Biol. Chem. 279 (2004)

AC

11521–11529.

[27] R. Rossignol, R. Gilkerson, R. Aggeler, K. Yamagata, S.J. Remington, R.A. Capaldi, Energy substrate modulates mitochondrial structure and oxidative capacity in cancer cells, Canc Res, 64 (2004) 985–993. [28] A. Rivinoja, N. Kokkonen, I. Kellokumpu, S. Kellokumpu, Elevated Golgi pH in breast and colorectal cancer cells correlates with the expression of oncofetal carbohydrate T-antigen, J. Cell. Physiol. 208 (2006) 167–174. [29] F. Laґzaro-Dieґguez, N. Jimeґnez, H. Barth, A.J. Koster, J. Renau-Piqueras, J.L. Llopis, K.N.J. Burger, G. Egea, Actin filaments are involved in the maintenance of Golgi cisternae morphology and intra-Golgi pH, Cell Motility and the Cytoskeleton 63 (2006) 778–791. [30] M. Serresi, R. Bizzarri, F. Cardarelli, F. Beltram, Real-time measurement of endosomal acidification by a novel genetically encoded biosensor // Anal Bioanal Chem 393 (2009) 1123–1133.

ACCEPTED MANUSCRIPT [31] F. Cianchi, M.C. Vinci, C.T. Supuran, B. Peruzzi, P. De Giuli, G. Fasolis, G. Perigli, S. Pastorekova, L. Papucci, A. Pini, E. Masini, L. Puccetti, Selective inhibition of carbonic anhydrase IX decreases cell proliferation and induces ceramide-mediated apoptosis in human cancer cells, J Pharmacol Exp Ther. 334 (2010) 710-719.

T

[32] X. He, Y. Wang, K. Wang, J. Peng, F. Liu, W. Tan, Research of the relationship of

IP

intracellular acidification and apoptosis in Hela cells based on pH nanosensors, Sci China Ser

SC R

B-Chem 50 (2007) 258-265.

[33] J. Chiche, M.C. Brahimi-Horn, J. Pouysségur, Tumour hypoxia induces a metabolic shift causing acidosis: a common feature in cancer, J. Cell. Mol. Med. 14 (2010) 771-794. [34] P.W. Vaupel, S. Frinak, H.I. Bicher, Heterogeneous oxygen partial pressure and pH

NU

distribution in C3H mouse mammary adenocarcinoma, Cancer Res 41 (1981) 2008-2013. [35] F. Kallinowski, P. Vaupel, pH distributions in spontaneous and isotransplanted rat tumours,

MA

Br. J. Cancer 58 (1988) 314-321.

Fig. 1. A) Fluorescence excitation (em. 530 nm) and emission (ex. 420 nm) spectra for SypHer2

TE

D

protein from crude lyzates registered at pH 7.1. B) Excitation spectra of SypHer2 at different pH.

Fig. 2. SypHer2 ratiometric imaging in HeLa cells in vitro. A) fluorescence images using

CE P

excitation at 420 nm and 500 nm (detection at 520 nm) and I500/I420 ratio images at different pH values; B) SypHer2 calibration curve (exponential fit, r2 = 0.99), mean±SD, n=30; C) pH imaging in resting HeLa cells transfected with SypHer2 and maintained under physiological

AC

medium; D) Confocal z-stack of living tumor spheroid shows more acidic pHi in the spheroid core. Depth of scanning is indicated on the images. Excitation at 405 nm and 488 nm, detection at 435-689 nm and 509-689 nm respectively. Scale bars 40 µm.

Fig. 3. A) in vivo pHi mapping of HeLa tumor expressing the genetically encoded pH-indicator SypHer2. Upper row: fluorescence images with excitation at 430 nm and 500 nm (detection at 540 nm) and the calculated I500/I430 ratio; quantitative analysis of the I500/I430 ratio in the center and at the periphery of tumor nodules (n=42) and in necrotic areas (n=16), M±SD. Lower row: enlarged I500/I430 ratio map of the area highlighted by the dotted line, corresponding to the photograph of the tumor from the histological slide (H&E, sagittal plane). N - necrosis. The blue to red colors of the I500/I430 lookup table indicate pHi values ranging from more acidic to more alkaline conditions. The data was recorded on day 18 of tumor growth. A separate tumor nodule with a radial pH gradient is highlighted by the dashed circle. Cutting line is shown by the white

ACCEPTED MANUSCRIPT line; B) in vivo monitoring of the fluorescence I500/I430 ratio during the course of tumor growth. The HeLa tumor expressing SypHer2 was imaged with the skin flap opened.

Fig. 4. Correspondence between the pHi map (A, I500/I430 ratio from a freshly frozen tissue

T

section), its histopathology (B, H&E staining) and the distribution of hypoxia (C, pimonidazole)

SC R

IP

in a SypHer2 expressing tumor. N – necrosis, T – vital tumor tissue. Scale bar 100 µm.

AC

CE P

TE

D

MA

NU

Figure 1.

ACCEPTED MANUSCRIPT

AC

CE P

TE

D

MA

NU

SC R

IP

T

Figure 2.

ACCEPTED MANUSCRIPT

AC

CE P

TE

D

MA

NU

SC R

IP

T

Figure 3.

ACCEPTED MANUSCRIPT

AC

CE P

TE

D

MA

NU

SC R

IP

T

Figure 4.

ACCEPTED MANUSCRIPT Highlights

T

IP

SC R NU MA D TE



CE P

  

We developed a method for pHi mapping in living cancer cells in vitro and in tumors in vivo The novel genetically encoded indicator, SypHer2, was used Intracellular pH was measured in HeLa cells in monolayer and tumor spheroids We obtained fluorescence ratio maps, representing the pHi distribution, for HeLa tumors in vivo and ex vivo A correspondence of the zones with higher pHi to the necrotic and hypoxic areas was demonstrated.

AC

