Low-field (L-band, 1.2 GHz) electron paramagnetic resonance using soluble nitroxyl and trityl probes is demonstrated for assessment of physiologically important parameters in the tumor microenvironment in mouse models of breast cancer.
This protocol demonstrates the capability of low-field electron paramagnetic resonance (EPR)-based techniques in combination with functional paramagnetic probes to provide quantitative information on the chemical tumor microenvironment (TME), including pO2, pH, redox status, concentrations of interstitial inorganic phosphate (Pi), and intracellular glutathione (GSH). In particular, an application of a recently developed soluble multifunctional trityl probe provides unsurpassed opportunity for in vivo concurrent measurements of pH, pO2 and Pi in Extracellular space (HOPE probe). The measurements of three parameters using a single probe allow for their correlation analyses independent of probe distribution and time of the measurements.
A key role of the TME in cancer progression and therapy is increasingly appreciated1. Among important physiological parameters of the TME in solid tumors, tissue hypoxia2, acidosis3,4, high reducing capacity5, elevated concentrations of intracellular GSH6,7, and interstitial Pi8 are well documented. Noninvasive in vivo pO2, pH, Pi, GSH, and redox assessments provide unique insights into the biological processes in TME, and help advance tools for pre-clinical screening of anti-cancer drugs and TME-targeted therapeutic strategies. A reasonable radiofrequency penetration depth in tissues by magnetic resonance imaging (MRI) and low-field EPR-based techniques makes them the most appropriate approaches for noninvasive assessment of these TME parameters. MRI relies largely on imaging water protons and is widely used in clinical settings to provide anatomical resolution but lacks functional resolution. The phosphorus-31 nuclear magnetic resonance (31P-NMR) measurements of extracellular Pi concentration and pH based on a signal from endogenous phosphate are potentially attractive for TME characterization, but are normally masked by several times higher intracellular Pi concentrations9,10. In contrast to this, EPR measurements rely on spectroscopy and imaging of specially designed paramagnetic probes to provide functional resolution. Note that exogenous EPR probes have an advantage over exogenous NMR probes due to the much higher intrinsic sensitivity of EPR and absence of endogenous background EPR signals. The recent development of a dual function pH and redox nitroxyl probe11 and multifunctional trityl probe12 provides unsurpassed opportunities for in vivo concurrent measurements of several TME parameters and their correlation analyses independent on probe distribution and time of measurement. To our knowledge, there are no other methods available to concurrently assess in vivo physiologically important chemical TME parameters in living subjects, such as pO2, pHe, Pi, redox, and GSH.
Probes for In Vivo Functional Measurements:
Figure 1 shows chemical structures of the paramagnetic probes used to access TME parameters, which include particulate and soluble probes. High functional sensitivity, stability in living tissue, and minimal toxicity are a few benefits that make particulate probes preferred over soluble probes for in vivo EPR oximetry. For example, particulate probes have increased retention times at the site of tissue implant compared to soluble probes allowing for longitudinal measurement of tissue pO2 over several weeks. On the other hand, soluble probes outperform particulate probes by providing spatial-resolved measurements using EPR-based imaging techniques as well as allowing concomitant analyses from multiple functionalities (pO2, pH, Pi, redox, and GSH).
Figure 1. Chemical structures of the paramagnetic probes that assemble TME assessment assay. This includes the particulate pO2 probe, LiNc-BuO (R = -O(CH2)3CH3), and soluble probes: dual function pH and redox probe, NR; GSH-sensitive probe, RSSR; and multifunctional pO2, pH, and Pi probe of the extracellular microenvironment, the HOPE probe. The synthesis of these probes has been described in the provided references 11,12. Please click here to view a larger version of this figure.
All animal work was performed in accordance with WVU IACUC approved protocol.
1. Probe Synthesis and Calibration
2. Mouse Models of Breast Cancer
3. Probe Delivery for In Vivo Functional Measurements
4. In Vivo Functional Measurements
5. Statistical Analysis
Tissue pO2 Assessment Using the LiNc-BuO Probes:
Using the procedure described under step 1.1, we performed the calibration of freshly prepared LiNc-BuO microcrystals suspension. Figure 2 shows the typical oxygen dependence of the linewidth of the LiNc-BuO probe, as well as its exemplified EPR spectra measured in buffer suspension and in breast tumor tissue in female C57Bl/6 mice grown after tumor initiation by injection of cells with internalized particles (step 3.1). Using the linewidth calibration allows for the calculation of the tissue pO2 value (equal to 7.5 mmHg for the spectrum (b) shown in the Figure 2).
Figure 2. EPR spectral sensitivity of the particulate LiNc-BuO probe to oxygen. Representative dependence of the linewidth of the EPR spectrum of the LiNc-BuO probe suspension in DMEM on oxygen partial pressure is shown. The EPR spectrum (insert a) was measured at 0% of pO2, 37 °C, and represents a pure Lorentzian line with the linewidth ΔHpp = 113 mG. The linewidth increases linearly with pO2 with a slope of 11.7 mG/mmHg. The spectrum shown in (insert b) was measured in mammary tumor tissue of anesthetized female C57Bl/6 mice showing ΔHpp = 204 mG, which corresponds to tissue pO2 of 7.5 mmHg. Please click here to view a larger version of this figure.
Assessment of Tissue Extracellular pH and Redox Using the NR Probes:
Using the procedure described under step 1.2, we performed the calibration of the pH and redox probe, NR. Figure 3 shows the observed hyperfine splitting constant dependence, aN, of the NR radical on pH. The dependence is described by a standard titration curve with the pKa = 6.6 at 37 °C. This dependence of aN(pH) serves as the calibration curve for corresponding in vivo measurements, allowing for pH measurements in the range from 5.6 to 7.6, with accuracy of 0.05 pH units.
Figure 3. EPR spectral sensitivity of the soluble NR probe to pH. L-band EPR spectra of the NR probe solution were acquired at 37 °C at various pH, and pH-dependence of the observed hyperfine splitting constant, aN, was plotted. The solid line is the fit of experimental data with the standard titration curve, , yielding pKa = 6.6, aN(NR-H+) = 14.24 G, and aN(R) = 15.27 G. Insert: Exemplified EPR spectrum of 1 mM NR solution (pH 6.4) acquired using the following spectrometer settings: magnetic field sweep, 80 G; modulation amplitude, 2.5 G, acquisition time, 20 s. The measured hyperfine splitting constant is 14.63 G. Reproduced from reference23 with permission of John Wiley & Sons, Inc. Please click here to view a larger version of this figure.
Figure 4 exemplifies the pH and redox measurements in a breast tumor mouse model, performed as described under step 3.3 (i.t. probe injection) and section 4 (spectra acquisition and spectra analysis).
Figure 4. EPR assessment of tissue reducing capacity in vivo. Insert: L-band EPR spectrum of the NR probe measured in vivo after i.t. injection (10 µL, 10 mM) in mammary tumor tissue of anesthetized female FVB/N mice. The hyperfine splitting, aN, was found to be equal to 14.72 G, which corresponds to the value of pHe = 6.52 assuming tumor tissue temperature 34 °C and pKa = 6.6. The NR reduction rate is assessed by following the decay of the central-field spectral component. The exemplified kinetics measured in mammary tumor (■) and mammary glands (o) demonstrate higher reducing capacity of the tumor tissue vs. normal mammary gland. The analysis of the initial part of the kinetics yields the rates of the EPR signal reduction, kred, in the extracellular media of the tissues as 2.5 x 10-3/s in tumor vs. 0.5 x 10-3/s in normal gland. Reproduced from reference11 with permission of John Wiley & Sons, Inc. Please click here to view a larger version of this figure.
Figure 5 summarizes the pHe and redox measurements performed for the group of female FVB/N MMTV-PyMT mice in tumors and normal mammary glands supporting tumor acidic extracellular microenvironment and high reducing capacity.
Figure 5. In vivo EPR assessment of tissue acidity and reducing capacity. Extracellular tissue pH (filled bars) and the reduction rate (empty bars) values of NR nitroxide in normal mammary glands and mammary tumors of female FVB/N mice measured by in vivo EPR. Error bars denote SE. Reproduced from reference11 with permission of John Wiley & Sons, Inc. Please click here to view a larger version of this figure.
Concurrent Tissue pHe, Redox, and pO2 Assessment by Combining NR and LiNc-BuO Probes:
Particulate LiNc-BuO is the probe of choice when repetitive longitudinal measurements of pO2 values are planned in a predetermined tissue of interest. Unfortunately, a major limitation of the particulate probe is that it only allows for the detection of oxygen and no other physiologically-relevant chemical parameters within the TME, and does not allow for tissue imaging. Thus, the use of soluble probes in EPR have the advantage of imaging and the detection of multiple physiologically-relevant TME parameters. In this respect, pO2 measurements using implanted LiNc-BuO particles can be combined with the dual function pH and redox NR probe because low- and high-field spectral components of the NR EPR spectrum do not overlap with the EPR line of the LiNc-BuO probe.
Figure 6 illustrates a typical EPR spectrum observed by the NR probe immediately after i.t. injection in breast tumor tissue of female C57Bl/6 mice. The observed triplet spectrum of the NR probe is overlaid with the single EPR line of LiNc-BuO microcrystals that have been embedded within the tumor (tumor growth was initiated by the injection of the cells with internalized particles as described in step 3.1. The pH value is calculated by measuring aN hyperfine splitting as a distance between low- and high-field components and using the calibration shown in Figure 3 (equal to 6.88) for the spectrum shown in Figure 6. Note that pO2 assessment is performed based on measurements of the linewidth of the LiNc-BuO signal before injection of the NR probe. The decay of the NR signal provides the value of tissue reducing capacity as illustrated in Figure 4.
Figure 6. Multifunctional TME assessment combining LiNc-BuO and NR probes. EPR spectrum measured in vivo in breast tumor tissue of female C57Bl/6 mouse immediately after i.t. injection of the NR probe. The observed triplet spectrum of the NR is superimposed with a single EPR line of LiNc-BuO probes embedded within tumor. The pH value calculated from the splitting between low- and high-field components is 6.88. The decay of the NR signal provides the value of tissue reducing capacity. pO2 assessment is performed based on measurements of the linewidth of the LiNc-BuO before injection of the NR probe. Please click here to view a larger version of this figure.
In Vivo Assessment of Intracellular GSH:
The reaction of thiol-disulfide exchange between the RSSR probe and GSH splits the disulfide bond of the probe resulting in the formation of two monoradicals22 and cancelation of intramolecular spin exchange between the monoradical fragments. Figure 7A illustrates the effect of the reaction of the RSSR probe with GSH on the EPR spectra resulting in the disappearance of the "biradical" spectral components and corresponding increase of the intensity of the monoradical components. The observed rate constant of the reaction between GSH and RSSR is: kobs = 2.8 ± 0.2 M-1s-1 (T = 34 °C, pH 7.2)11. Figure 7B exemplifies the kinetics of the monoradical spectral peak intensity change measured in mammary tumor (●) and normal mammary gland (o) using L-band EPR. The solid lines are the fits of the initial part of the kinetics by the monoexponent using kobs = 2.8 M-1s-1 and yielding [GSH] = 10.7 mM and 3.3 mM for the tumor and normal mammary gland, respectively.
Figure 7. In vivo EPR assessment of intracellular tissue glutathione concentrations. (A) The X-band EPR spectra of 100 µM RSSR measured at various time points after incubation with 2.5 mM GSH in 0.1 M Na-phosphate buffer, pH 7.2, and 1 mM DTPA at 34 °C. The kinetics analysis provides the observed rate constant value of the reaction between GSH and RSSR, kobs (pH 7.2, 34 °C) = (2.8 ± 0.2) M-1 s-1. (B) The kinetics of the monoradical spectral peak intensity change measured by L-band EPR in mammary tumor (●) and normal mammary gland (o) of FVB/N mice immediately after i.t. injection of RSSR probe (see step 3.3.2). The solid lines are the fits of the initial part of the kinetics by the monoexponent, supposing kobs (pH 7.2, 34 °C) = 2.8 M-1s-1 and yielding [GSH] = 10.7 mM and 3.3 mM for the tumor and normal mammary gland, respectively. Reproduced from reference11 with permission of John Wiley & Sons, Inc. Please click here to view a larger version of this figure.
Multifunctional Assessment of Tissue Extracellular pH, pO2, and Pi Using the Multifunctional HOPE Probe:
Figure 8 illustrates the sensitivity of the spectral parameters of the HOPE probe obtained using calibration procedures described under section 1.4.
Figure 8. Multifunctional assessment of the chemical microenvironment using HOPE probe. (a) The scheme of pH-dependent equilibrium between two ionization states of the probe. (b) L-band EPR spectrum of HOPE. (c) The EPR linewidth of the HOPE is a pO2 marker (accuracy, ≈ 1 mmHg; pO2 range, 1-100 mmHg). (d) The fraction of protonated HOPE is a pH marker in the range from 6 to 8.0 (accuracy, ± 0.05). (e) Dependence of proton exchange rate (expressed in mG) of the HOPE with inorganic phosphate (Pi) concentration extracted by spectra simulation (accuracy, ± 0.1 mM, range, 0.1-20 mM) 12,18. Reproduced from reference8 with permission of Nature Publishing Group. Please click here to view a larger version of this figure.
Figure 9 illustrates multifunctional measurements performed in FVB/N wild type mammary glands and in the TME of MMTV-PyMT transgenic mice, which spontaneously develop breast cancer and emulate human tumor staging20. The mean values of pO2 (50 ± 3 mmHg in TME vs. 58 ± 3 mmHg in normal tissue) and pHe (6.99 ± 0.03 in TME vs. 7.1 ± 0.03 in normal tissue) support an appearance of hypoxic and acidic regions in the tumor. The most dramatic changes were observed in the concentration of interstitial Pi (1.8 ± 0.2 mM in TME vs. 0.84 ± 0.07 mM in normal tissue) indicating a potential role of interstitial Pi concentration as a TME marker of tumor progression. The individual measurements show significant variations around the mean values. An important advantage of a multifunctional probe is that all parameters are measured using a single probe, therefore allowing for correlation analyses independent of probe distribution and time of the measurements as illustrated in Figure 9b–e. The observed positive correlation between pO2 and pHe in normal mammary gland versus the absence of correlation in tumors supports tumor reliance on glycolysis independent of oxygen concentration; in our opinion this is an exemplified in vivo demonstration of the Warburg effect. In turn, the observed negative correlation between interstitial [Pi] and pO2 both in normal and tumor tissues is in agreement with the association of high [Pi] (and low ATP/Pi ratio) with changes in bioenergetics status upon lower oxygen supply.
Figure 9. In vivo tissue pO2, pHe, and Pi assessment using the HOPE probe and EPR. (a) Photograph of the set-up for in vivo L-band EPR measurements shows the anesthetized mouse between the magnets of the EPR spectrometer, with the insert on the right showing placement and positioning of the loop resonator on top of the measured tissue. Note the interstitial extracellular localization of the HOPE probe: it does not penetrate into the cells due to bulky charged structures and the HOPE signal from the blood is not detected by EPR due to signal broadening by HOPE complexation with plasma albumin24. (b–e) Correlation between interstitial pO2, pHe, and Pi values measured in normal mammary glands of FVB/N wild type mice and in the TME of breast cancer in MMTV-PyMT transgenic mice (n = 23). To extend the range of oxygen variations, anoxic conditions in interstitial space were established by i.t. injection of the oxygen-consuming enzymatic system of glucose/glucose oxidase (red symbols). Blue lines represent a linear fit for the total data sets. (b) A positive correlation between pO2 and pHe in normal tissue (r = 0.5, p = 0.014 for black symbols; r = 0.64, p = 1.8 × 10-4 for total data set) vs. (c) no significant correlation between pO2 and pHe in TME (r = 0.01, p = 0.97 for black symbols; r = 0.23, p = 0.3 for total data set) were found. (d) A negative correlation between pO2 and Pi both in normal tissue (r = −0.51, p = 0.013 for black symbols; r = −0.7, p = 2.3 × 10-5 for total data set) and (e) in TME (b, bottom: r = −0.4, p = 0.079 for black symbols; r = −0.62, p = 0.001 for total data set) were found. Adapted from reference8 with permission with of Nature Publishing Group. Please click here to view a larger version of this figure.
The presented methods allow for noninvasive in vivo assessment of the critical parameters of the chemical TME, namely pO2, pH, redox status, and concentrations of interstitial Pi and intracellular GSH. Magnetic resonance techniques, such as MRI and low-field EPR, are the methods of choice for noninvasive in vivo profiling of these TME parameters. MRI visualizes anatomical structures but lacks functional sensitivity. In contrast to MRI, EPR techniques provide functional sensitivity to the local parameters of the microenvironment when used in combination with functional spin probes. To our knowledge, there are no other methods available to concurrently assess in vivo physiologically important chemical TME parameters in living subjects, such as pO2, pHe, Pi, redox, and GSH.
The described protocols are based on using low-field L-band EPR and specially designed paramagnetic probes. A set of four probes presented in Figure 1 can be considered an effective in vivo TME assessment assay. Availability of the probes is an important prerequisite for the described in vivo EPR measurements. The synthesis of the probes according to the published procedures11,12,13,15 requires expertise in synthetic organic chemistry and may become a critical step for the method implementation.
Application of the particulate probe enables repeated measurements of tissue pO2 for up to weeks after implantation14,21. Soluble probes demonstrate sensitivity to an extended number of the parameters beyond oxygen concentration25 and provide an opportunity for spatial-resolved measurements26. In particular, an application of the multifunctional HOPE probe provides unsurpassed opportunity for in vivo concurrent measurements of pH, pO2, and Pi in extracellular space. The measurements of the three parameters using a single probe allows for their correlation analyses independent of probe distribution and time of the measurements8.
The probes cannot be used simultaneously, except for the implanted LiNc-BuO particulate probe, which can be used in combination with any of the soluble probes as demonstrated in Figure 6. Therefore, the experimental design should be based on separate measurements of individual injections of specific functional soluble probes. The hydrophilic extracellular pH and redox NR probe can be delivered both via i.t. injection11,14,26 and systemic delivery27, while two other soluble probes, RSSR and HOPE, allow for i.t. delivery only8,11,12,16. Note that the future structural modifications of RSSR and HOPE probes may overcome the latter limitation of probe delivery by the development of more hydrophilic and less toxic probe analogs for systemic delivery25.
Note that spectroscopic EPR modality provides the average values of the parameters measured in the TME, which is often characterized by its high heterogeneity. This partially masks the potential differences between the parameters or decreases significance in correlation analysis. The use of the described probes in imaging modalities allows for obtaining spatially-resolved functional information26. We believe that the future of multi-functional EPR imaging relies on the further development of relatively new methodologies such as rapid scan EPR imaging and Overhauser-enhanced MRI (OMRI, also termed proton-electron double-resonance imaging, PEDRI). The new directions in the development of functional EPR probes and advanced imaging techniques have been recently discussed in the corresponding feature article25.
The authors have nothing to disclose.
This work was partially supported by NIH grants CA194013, CA192064 and U54GM104942. The WVCTSI is acknowledged for start-up to VVK, AB, and TDE. The authors thank Dr. M. Gencheva and K. Steinberger for the assistance with the illustrative experiments. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH.
L-band EPR spectrometer | Magnettech, Germany | L-band (1.2 GHz) electron paramagnetic resonance (EPR) spectrometer for collection in vitro and in vivo spectra of paramagnetic molecules | |
Temperature & Gas Controller | Noxygen, Germany | Temperature & Gas Controller designed to control and adjust the temperature and gas composition | |
Sonicator | Fisher Scientific | ||
GSH (L-Glutathione reduced) | Sigma-Aldrich | G4251 | |
MMTV-PyMT mice | In house | ||
DMEM | Thermo Fisher Scientific | 11995065 | |
Met-1 murine breast cancer cells | In house | ||
C57Bl/6 wild type mice | Jackson Laboratory | ||
Trypsin | Thermo Fisher Scientific | 25200056 | |
Trypan Blue Exclusion Dye | Thermo Fisher Scientific | T10282 | |
Ohmeda Fluotec 3 | |||
Isoflurane (IsoFlo) | Abbott Laboratories | ||
Sodium phosphate dibasic | Sigma-Aldrich | S9763 | |
Sodium phosphate monobasic | sigma-Aldrich | S07051 | |
Sodium Chloride | sigma-Aldrich | S7653 | |
Hydrochloric acid | sigma-Aldrich | 320331 | |
Sodium Hydroxide | sigma-Aldrich | S8045 | |
Glucose | sigma-Aldrich | ||
Glucose oxydase | sigma-Aldrich | ||
Lauda Circulator E100 | Lauda-Brikmann | ||
pH meter Orion | Thermo Scientific | ||
LiNc-BuO probe | In house | The Octa-n-Butoxy-Naphthalocyanine probe was synthesizided according to ref 13 | |
NR probe | In house | The Nitroxide probe was synthesizided according to ref 11 | |
RSSR probe | In house | The di-Nitroxide probe was synthesizided according to ref 15 | |
HOPE probe | In house | The monophoshonated Triarylmethyl probe was synthesizided according to ref 12 |