A NIR fluorescent smart probe for imaging tumor hypoxia

Abstract Background Tumor hypoxia is a characteristic of paramount importance due to low oxygenation levels in tissue negatively correlating with resistance to traditional therapies. The ability to noninvasively identify such could provide for personalized treatment(s) and enhance survival rates. Accordingly, we recently developed an NIR fluorescent hypoxia‐sensitive smart probe (NO 2 ‐Rosol) for identifying hypoxia via selectively imaging nitroreductase (NTR) activity, which could correlate to oxygen deprivation levels in cells, thereby serving as a proxy. We demonstrated proof of concept by subjecting a glioblastoma (GBM) cell line to extreme stress by evaluating such under radiobiological hypoxic (pO2 ≤ ~0.5%) conditions, which is a far cry from representative levels for hypoxia for brain glioma (pO2 = ~1.7%) which fluctuate little from physiological hypoxic (pO2 = 1.0‐3.0%) conditions. Aim We aimed to evaluate the robustness, suitability, and feasibility of NO 2 ‐Rosol for imaging hypoxia in vitro and in vivo via assessing NTR activity in diverse GBM models under relevant oxygenation levels (pO2 = 2.0%) within physiological hypoxic conditions that mimic oxygenation levels in GBM tumor tissue in the brain. Methods We evaluated multiple GBM cell lines to determine their relative sensitivity to oxygenation levels via measuring carbonic anhydrase IX (CAIX) levels, which is a surrogate marker for indirectly identifying hypoxia by reporting on oxygen deprivation levels and upregulated NTR activity. We evaluated for hypoxia via measuring NTR activity when employing NO 2 ‐Rosol in in vitro and tumor hypoxia imaging studies in vivo. Results The GBM39 cell line demonstrated the highest CAIX expression under hypoxic conditions representing that of GBM in the brain. NO 2 ‐Rosol displayed an 8‐fold fluorescence enhancement when evaluated in GBM39 cells (pO2 = 2.0%), thereby establishing its robustness and suitability for imaging hypoxia under relevant physiological conditions. We demonstrated the feasibility of NO 2 ‐Rosol to afford tumor hypoxia imaging in vivo via it demonstrating a tumor‐to‐background of 5 upon (i) diffusion throughout, (ii) bioreductive activation by NTR activity in, and (iii) retention within, GBM39 tumor tissue. Conclusion We established the robustness, suitability, and feasibility of NO 2 ‐Rosol for imaging hypoxia under relevant oxygenation levels in vitro and in vivo via assessing NTR activity in GBM39 models.


| INTRODUCTION
Oxygen-deprived tumor tissue can transition toward aggressive phenotypes that promote malignancy and increase resistance to standard treatment regimens such as cytotoxic chemotherapy (CCT) and radiotherapy (RT). [1][2][3][4][5][6] The ability to noninvasively identify hypoxia in tumor tissue could (i) improve patient prognosis; (ii) afford patient stratification, which would advance efforts toward providing personalized treatment; and (iii) consequently increase the patient survival rates. [7][8][9] When utilized for the purpose of imaging tumor hypoxia, the nonoptical imaging techniques of magnetic resonance imaging (MRI) and positron emission tomography (PET) suffer from the drawback of reporting on the blood oxygenation levels and providing poor contrast levels (ie, tumor-to-background ratios, TBRs, of ≤~2.0), respectively, due to, in part, employing non-smart probes (ie, radiotracers) such as 18 F-fluoromisonidazole ( 18 F-FMISO). [10][11][12][13][14] On the other hand, optical imaging techniques, such as fluorescence imaging, is a noninvasive imaging technique that is well-suited for imaging tumor hypoxia particularly because such modality affords high sensitivity, high spatiotemporal resolution, and multiplexing capabilities. [15][16][17][18][19] Other fluorescence optical imaging constructs designed for hypoxia imaging suffer from the drawbacks of photophysical and physicochemical deficiencies that limit their efficacy, such as inherently bearing nonneutral net charges that could prevent cell membrane translocation, and therefore are ill-suited for such purpose. 15,[20][21][22][23][24] To overcome such drawbacks, we recently implemented a rational design strategy to develop a hypoxia-sensitive near-infrared (NIR) fluorescent smart probe (NO 2 -Rosol) that is activatable via bioreductive activation afforded by the nitroreductase (NTR) class of enzymes, whose reductive activity toward the nitro group of nitroaromatic moieties could negatively correlate to oxygenation levels. 25 NO 2 -Rosol is a nitrated congener of THQ-Rosol, which itself derives from the rosol molecular platform. 26 As such, NO 2 -Rosol contains a xanthene core-based scaffold that is synthesized from the methyoxybenzene-based tetrahydroquinoxaline (THQ) moiety beginning with a Friedel-Crafts acylation using the acyl chloride of 3-nitrobenzene followed by an intermolecular condensation with resorcinol. NO 2 -Rosol demonstrates excellent solubility in water as determined from its (linearly) increasing absorbance value (at 550 nm) and varied increasing concentration between 1 to 30 μM.
In our previous findings, when compared to a GBM tumor cell line that was propagated under normoxic conditions (pO 2 = 20%) in the same study, our activatable NIR fluorescent smart probe afforded a remarkable 12-fold OFF-ON NIR fluorescence enhancement from NTR activity in such cells that experienced pathological hypoxic (pO 2 ≤ 1.0%) and radiobiological hypoxic conditions (pO 2 ≤ 0.5%) due to such cells being propagated under extreme hypoxic conditions (pO 2 = 0.5%). Pathological hypoxia is defined as the oxygenation level in tumor tissue in which such elicits nonuniform aberrances in its physiological (pO 2 = 1.0-3.0%) hypoxic response that promotes a homeostatic hypoxic environment, whereas radiobiological hypoxia is defined as the oxygenation level in tumor tissue whereby the cytotoxic effect of RT toward such is half maximal (ie, radiation treatment is half as effective due to the hypoxic tumor tissue transitioning to a more radioresistant phenotype) ( Figure 1). 4,5,[27][28][29][30][31] As such an extreme state of oxygen deprivation (pO 2 = 0.5%) unrepresentative of typical conditions for hypoxia in glioma tissue in the brain was ideal for efficiently determining the basic functional viability of NO 2 -Rosol to selectively undergo bioreductive activation via NTR activity, we found it imperative to extend our earlier work by assessing its ability to afford hypoxia imaging in vitro and in vivo models of patient-derived GBM tumor cell lines propagated under applicable physiological hypoxic (pO 2 = 2.0%) conditions that (i) reflect the typical oxygenation levels in glioma tumor tissue of low-and highgrade in the brain (pO 2 =~1.7%) and (ii) afford the GBM tumor cell lines to provide homeostatic hypoxic responses. In doing so, we F I G U R E 1 Generalized representation of tumor tissue classification as a function of oxygen tension. Low oxygen tension (hypoxia) criteria are in accordance with measured glioma tumor tissue oxygenation levels (of low-and high-grade including GBM). 27,28 The unit for a measured oxygenation level is represented by the partial pressure of oxygen (pO 2 ) expressed as a percentage of total environmental pressure. Figure not drawn to scale sought to evaluate the robustness and suitability of NO 2 -Rosol for imaging hypoxia by better mimicking such critical environmental aspects, especially as NTRs can exhibit plasticity in their activity based on oxygenation levels. Importantly, we also sought to establish the feasibility of NO 2 -Rosol to afford the imaging of tumor hypoxia in vivo via targeting any upregulated NTR activity. In particular, we set out to ascertain if our NIR fluorescent smart probe was capable of promptly diffusing throughout, activating within, becoming sufficiently retained in, and consequently providing effective contrast levels (ie, high TBRs) in xenograft murine GBM tumor models postadministration such that it would successfully afford in vivo imaging of tumor hypoxia via providing effective contrast levels analogous to that which the initial rosol framework also involving a NIR emissioninducing THQ moiety had allowed for in lymphatic mapping applications (ie, identifying the sentinel lymph node) via NIR fluorescence imaging. 26,32 Accordingly, herein, we present the evaluation of the robust-  which are grossly atypical hypoxic conditions for glioma tissue in the brain unless under acute additional stress or necrosis. 35 The environmental stress responses from such cells were elicited by nonhomeostatic hypoxic conditions whose cellular machinery could have been inconsistently operating or simply irretrievably dysregulated, and thus such cells may have not been affording NTR activity that was to be representative of typical hypoxic glioma tumor tissue. As such, any observed NTR activity could have been artificially low or high due to any such dysregulation, and thereby potentially under-qualifying or over-qualifying, respectively, the capability for our NIR fluorescent smart probe to render accurate identification (imaging) of hypoxia in glioma tumor tissue (via of detecting NTR activity) that would typically be under mild but relevant physiological hypoxic conditions (pO 2 = 2.0%) when hypoxic. Therefore, it was of paramount importance to undertake the studies performed herein and accordingly evaluate such results.  Figure 2C). Accordingly, the measured extinction coefficient of NO 2 -Rosol and NH 2 -Rosol were determined to be 11 000 and 10 300 M −1 cm −1 , whereby their measured quantum yield was determined to be 0.05% and 2.05%, respectively. indicates that the results are unlikely due to chance.

| Fluorescence microscopy
We used the GBM39 cell line to directly assess for any correlative NTR activity via employing our NIR fluorescent smart probe toward cell imaging studies using confocal fluorescence microscopy ( Figure 5).
As such, we performed hypoxia imaging experiments in vitro using live

| In vivo tumor hypoxia imaging
We assessed the feasibility of NO 2 -Rosol to afford in vivo tumor hypoxia imaging via evaluating its NIR fluorescence response elicited by upregulated NTR activity when applied to murine GBM39 tumor models of glioma tumor under presumably such relevant oxygenation levels. More specifically, we evaluated the capability of our NIR fluorescent smart probe to provide effective contrast levels that are superior to those afforded by noninvasive imaging modalities (eg, PET) that employ non-smart probes (eg, 18 F-FMISO) for imaging hypoxic tumor tissue. 13,14 As the goal of this study was to determine the feasibility of our NIR fluorescent smart probe to afford tumor hypoxia imaging in vivo, we prepared and utilized xenograft murine GBM39 tumor models (as opposed to orthotopic murine GBM39 tumor models) to F I G U R E 4 Cell viability assay using varied concentration levels of NO 2 -Rosol with GBM39 cells. Cells receiving no treatment were used as a negative control and cells treated with 15% DMSO as a positive control. Cells were stained with Calcein-AM, a live cell stain, and subsequently the fluorescence intensity was measured at 516 nm (λ ex = 494 nm, n = 5, *P < .05, ***P < .001, using a one-way analysis of variance followed by Tukey post hoc tests with the intensity from the cells receiving no treatment serving as the control)

| Synthesis of NO 2 -Rosol
The methyoxybenzene-based THQ moiety was prepared according to literature methods. 40,41 Such moiety (366 mg, 1.13 mmol, 1 eq) was added to a small round bottom flask and dissolved in 20 mL CH 2 Cl 2 , wherein a stir bar was added and the flask was sealed with a rubber septum. The solution was cooled to 0 C and bubbled with N 2 gas via introducing a continuous gentle stream of N 2 gas to the solution via a needle alongside a venting needle piercing the septum. The septum was briefly lifted whilst maintaining a strong positive pressure of N 2 gas and aluminum chloride (904 mg, 6.78 mmol, 6 eq) was added quickly followed by the continuous bubbling of N 2 gas. Solid 3-nitrobenzoyl chloride (230 mg, 1.24 mmol, 1.1 eq) was added in a similar fashion followed by the additional bubbling of the mixture with N 2 gas. Once the solvent volume was reduced by half, the venting needle was removed maintaining a positive N 2 pressure on the flask. It was sonicated intermittently and allowed to warm to room temperature over 45 minutes.
The work-up involved pouring the reaction mixture over ice and gradually basifying to~pH 6 with a saturated solution of sodium bicarbonate.
The product was extracted in a separatory funnel with CH 2 Cl 2 (100 mL × 3). The organic layers were combined, dried over magnesium sulfate, and the solvent removed in vacuo. The crude mixture was purified via column chromatography on silica gel. The eluent consisted of 100% methylene chloride which was used to separate out impurities followed by the addition of 1-5% EtOAc to elute the acylated derivative

| Statistical analysis
Unless otherwise noted (and where appropriate), data were expressed as the mean ± SE of the mean (SEM) and analyzed using i) a one-way analysis of variance, followed by Tukey post hoc tests or ii) an unpaired t-test with Welch's correction, from GraphPad Prism 5 (Gra-phPad Software, La Jolla, CA).

| CONCLUSION
Here, we evaluated and validated the robustness, suitability, and feasibility of our NIR fluorescent smart probe for imaging hypoxia in vitro and in vivo by assessing the sensitivity of our smart probe to hypoxic conditions and the reflected NTR activity for a panel of GBM models that were under mild but relevant oxygenation levels (pO 2 = 2.0%) within physiological hypoxic conditions effectively mimicking the typical oxygenation levels of hypoxic glioma tumor tissue in the brain (pO 2 =~1.7%), whereby a homeostatic hypoxic environment is particularly crucial for determining the suitability (ie, practical operational range) of NO 2 -Rosol due to the plasticity in NTR activity. In measuring the total expression level of CAIX in all four patient-derived cell lines, we observed the GBM39 cell line to demonstrate the relatively highest CAIX total expression level under mild but relevant physiological hypoxic conditions (pO 2 = 2.0%) that represent that of glioma tissue (including GBM) in the brain. As such, any correlative NTR activity