Design, Synthesis and Anticancer Evaluation of Nitroimidazole Radiosensitisers

The role of hypoxic tumour cells in resistance to radiotherapy, and in suppression of immune response, continues to endorse tumour hypoxia as a bona fide, yet largely untapped, drug target. Radiotherapy innovations such as stereotactic body radiotherapy herald new opportunities for classical oxygen-mimetic radiosensitisers. Only nimorazole is used clinically as a radiosensitiser, and there is a dearth of new radiosensitisers in development. In this report, we augment previous work to present new nitroimidazole alkylsulfonamides and we document their cytotoxicity and ability to radiosensitise anoxic tumour cells in vitro. We compare radiosensitisation with etanidazole and earlier nitroimidazole sulfonamide analogues and we identify 2-nitroimidazole and 5-nitroimidazole analogues with marked tumour radiosensitisation in ex vivo assays of surviving clonogens and with in vivo tumour growth inhibition.


Introduction
Radiotherapy (RT) plays a central role in cancer treatment with~50% of all cancer patients receiving RT, either in a definitive setting (high-dose fractionated RT with curative intent), as adjuvant therapy after surgery, or as palliation to relieve symptoms of disease [1,2]. The combination of RT with chemotherapy (CRT) has emerged as the standard of care for definitive RT in many tumour types (e.g., cisplatin in head and neck squamous cell carcinoma (HNSCC) [3]), but chemoradiotherapy is limited by normal tissue toxicities both within and beyond the radiation field [4]. Key to further advances in RT will be exploitation of the tumour microenvironment to identify drug targets that will provide tumour-selective radiosensitisation [2,[5][6][7].
A key aspect of the tumour microenvironment that limits the therapeutic effectiveness of RT is hypoxia [8,9], a common characteristic of tumours, owing to the disordered and dysfunctional tumour vasculature that supports tumour growth [8,10]. Higher radiation doses (from two-to three-fold) are required to kill hypoxic cells compared to well-oxygenated cells, since radiation requires oxygen to induce DNA damage [11]. Tumour hypoxia is prevalent in HNSCC tumours [12] and there is compelling evidence that patients with hypoxic human papilloma virus (HPV)-ve HNSCC tumours have inferior responses following RT or CRT [13][14][15][16].
We have previously reported our efforts to develop a novel class of nitroimidazole sulfonamide radiosensitiser that incorporated both highly electron-affinic compounds that act as potent sensitisers through reacting with DNA radicals and bioreduction to cytotoxic species, as well as less electron-affinic compounds that are less effective sensitisers but better tolerated [33,34]. We identified four compounds (5)(6)(7)(8) that effectively sensitised HCT116 colon carcinoma cells and tumours to radiation but, due to their relatively low solubility, required formulation as phosphate prodrugs (9)(10)(11)(12) [34] (Figure 1). Here, we extend our investigation of these nitroimidazole sulfonamide radiosensitisers to characterise a wider compound set in additional cell lines, including two HNSCC cell lines, and to identify effective compounds that do not require a prodrug approach in vivo.

Synthesis
In our earlier investigation, we evaluated 33 novel nitroimidazole sulfonamide radiosensitiser compounds for in vitro cytotoxicity and radiosensitisation in HCT116 cells [33,34]. We identified four candidate alcohols 5-8 with promising in vitro activity, but low aqueous solubility required formulation as the corresponding phosphate prodrugs 9-12 for in vivo evaluation ( Figure 1). A further five compounds (13)(14)(15)(16)(17) were synthesized to explore the use of diol (13,15), triol (14) and morpholine groups (16) to increase aqueous solubility, thus, avoiding the use of phosphate prodrugs (Scheme 1). We also explored the effect of reversing the orientation of the sulfonamide linker (17). The use of diols and triols as neutral solubilising groups have ample precedent in the development of doranidazole [35,36].

Synthesis
In our earlier investigation, we evaluated 33 novel nitroimidazole sulfonamide radiosensitiser compounds for in vitro cytotoxicity and radiosensitisation in HCT116 cells [33,34]. We identified four candidate alcohols 5-8 with promising in vitro activity, but low aqueous solubility required formulation as the corresponding phosphate prodrugs 9-12 for in vivo evaluation ( Figure 1). A further five compounds (13)(14)(15)(16)(17) were synthesized to explore the use of diol (13,15), triol (14) and morpholine groups (16) to increase aqueous solubility, thus, avoiding the use of phosphate prodrugs (Scheme 1). We also explored the effect of reversing the orientation of the sulfonamide linker (17). The use of diols and triols as neutral solubilising groups have ample precedent in the development of doranidazole [35,36].

Physicochemical Properties
The use of a diol sidechain for 2-nitroimidazole 13 provided a large fold) in aqueous solubility compared to the corresponding alcohol 5 (18 ever, addition of another alcohol group in triol 14 did not increase solubil the isomeric 5-nitroimidazole 15, little additional solubility was gained co mM). The use of a charged morpholine group in 5-nitroimidazole 15 pro increase in solubility compared to analogous alcohol 7 (23 mM). Curiou sulfonamide 17 provided an increase in solubility compared to the conve mide analogue 6. Overall, the use of polyols did not provide increased aq compared to the phosphate prodrug approach but did provide sufficient ity to allow full evaluation of the analogues. The compounds were all s culture conditions. The electron affinity of the compounds was assessed b measured one-electron reduction potentials measured for analogues of 5 ature values [28]. Typically, 2-nitroimidazoles provide reduction potent from −380 to −400 mV and the proximity of the electron withdrawing s stituent can increase this to ca. from −340 to −350 mV for a C-1 linkage, increase (ca. 10 mV) for a C-2 linkage [34]. Similarly, the one electron redu for 5-nitroimidazoles (e.g., 3) typically lie in the range from −440 to −460 thyl-5-nitroimidazole analogues as low as −500 mV, and the addition of fonamide substituent increases this by ca. 25 and 10 mV, respectively [34

Physicochemical Properties
The use of a diol sidechain for 2-nitroimidazole 13 provided a large fold) in aqueous solubility compared to the corresponding alcohol 5 (18 ever, addition of another alcohol group in triol 14 did not increase solubil the isomeric 5-nitroimidazole 15, little additional solubility was gained co mM). The use of a charged morpholine group in 5-nitroimidazole 15 pro increase in solubility compared to analogous alcohol 7 (23 mM). Curiou sulfonamide 17 provided an increase in solubility compared to the conve mide analogue 6. Overall, the use of polyols did not provide increased aq compared to the phosphate prodrug approach but did provide sufficient ity to allow full evaluation of the analogues. The compounds were all s culture conditions. The electron affinity of the compounds was assessed b measured one-electron reduction potentials measured for analogues of 5 ature values [28]. Typically, 2-nitroimidazoles provide reduction potent from −380 to −400 mV and the proximity of the electron withdrawing s stituent can increase this to ca. from −340 to −350 mV for a C-1 linkage, increase (ca. 10 mV) for a C-2 linkage [34]. Similarly, the one electron redu for 5-nitroimidazoles (e.g., 3) typically lie in the range from −440 to −460 thyl-5-nitroimidazole analogues as low as −500 mV, and the addition of fonamide substituent increases this by ca. 25 and 10 mV, respectively [34

Physicochemical Properties
The use of a diol sidechain for 2-nitroimidazole 13 provided a large fold) in aqueous solubility compared to the corresponding alcohol 5 (18 ever, addition of another alcohol group in triol 14 did not increase solubil the isomeric 5-nitroimidazole 15, little additional solubility was gained co mM). The use of a charged morpholine group in 5-nitroimidazole 15 pro increase in solubility compared to analogous alcohol 7 (23 mM). Curiou sulfonamide 17 provided an increase in solubility compared to the conve mide analogue 6. Overall, the use of polyols did not provide increased aq compared to the phosphate prodrug approach but did provide sufficient ity to allow full evaluation of the analogues. The compounds were all s culture conditions. The electron affinity of the compounds was assessed b measured one-electron reduction potentials measured for analogues of 5 ature values [28]. Typically, 2-nitroimidazoles provide reduction potent from −380 to −400 mV and the proximity of the electron withdrawing s stituent can increase this to ca. from −340 to −350 mV for a C-1 linkage, increase (ca. 10 mV) for a C-2 linkage [34]. Similarly, the one electron redu for 5-nitroimidazoles (e.g., 3) typically lie in the range from −440 to −460 thyl-5-nitroimidazole analogues as low as −500 mV, and the addition of fonamide substituent increases this by ca. 25 and 10 mV, respectively [34

Physicochemical Properties
The use of a diol sidechain for 2-nitroimidazole 13 provided a large fold) in aqueous solubility compared to the corresponding alcohol 5 (18 ever, addition of another alcohol group in triol 14 did not increase solubil the isomeric 5-nitroimidazole 15, little additional solubility was gained co mM). The use of a charged morpholine group in 5-nitroimidazole 15 pro increase in solubility compared to analogous alcohol 7 (23 mM). Curiou sulfonamide 17 provided an increase in solubility compared to the conve mide analogue 6. Overall, the use of polyols did not provide increased aq compared to the phosphate prodrug approach but did provide sufficient ity to allow full evaluation of the analogues. The compounds were all s culture conditions. The electron affinity of the compounds was assessed b measured one-electron reduction potentials measured for analogues of 5 ature values [28]. Typically, 2-nitroimidazoles provide reduction potent from −380 to −400 mV and the proximity of the electron withdrawing su stituent can increase this to ca. from −340 to −350 mV for a C-1 linkage, increase (ca. 10 mV) for a C-2 linkage [34]. Similarly, the one electron redu for 5-nitroimidazoles (e.g., 3) typically lie in the range from −440 to −460 thyl-5-nitroimidazole analogues as low as −500 mV, and the addition of fonamide substituent increases this by ca. 25 and 10 mV, respectively [34

Physicochemical Properties
The use of a diol sidechain for 2-nitroimidazole 13 provided a large fold) in aqueous solubility compared to the corresponding alcohol 5 (18 ever, addition of another alcohol group in triol 14 did not increase solubil the isomeric 5-nitroimidazole 15, little additional solubility was gained co mM). The use of a charged morpholine group in 5-nitroimidazole 15 pro increase in solubility compared to analogous alcohol 7 (23 mM). Curiou sulfonamide 17 provided an increase in solubility compared to the conve mide analogue 6. Overall, the use of polyols did not provide increased aq compared to the phosphate prodrug approach but did provide sufficient ity to allow full evaluation of the analogues. The compounds were all culture conditions. The electron affinity of the compounds was assessed b measured one-electron reduction potentials measured for analogues of 5 ature values [28]. Typically, 2-nitroimidazoles provide reduction potent from −380 to −400 mV and the proximity of the electron withdrawing s stituent can increase this to ca. from −340 to −350 mV for a C-1 linkage, increase (ca. 10 mV) for a C-2 linkage [34]. Similarly, the one electron red for 5-nitroimidazoles (e.g., 3) typically lie in the range from −440 to −460 thyl-5-nitroimidazole analogues as low as −500 mV, and the addition of fonamide substituent increases this by ca. 25 and 10 mV, respectively [34

In Vitro Cytotoxicity
The new compounds 13-17 were compared to the clinical compounds misonidazole (1), etanidazole (2) and nimorazole (3), the recently described alcohols 5-8, as well as a representative set of nitroimidazole sulfonamides 33-46 [34]. Compounds were tested for cytotoxicity (as IC 50 ) after 4 h of drug exposure under both oxic and anoxic conditions with a five-day regrowth period. This allowed derivation of hypoxia cytotoxicity ratios (HCR) as IC 50(oxic) /IC 50(anoxic) in an HCT116 subclone HCT116/54C and two HNSCC cell lines, FaDu and UT-SCC-74B ( Table 2). The clinical radiosensitisers 1-3 and the lead nitroimidazole sulfonamides 5-8 demonstrated increased cytotoxicity under anoxia relative to oxia consistently across the cell line panel, with more electron affinic 2-nitroimidazoles showing greater HCR values than the corresponding 5-nitroimidazole analogues, for example, comparing 1, 2 and 5 with 3 and 7. Similarly, 2-nitroimidazole 6 was more potent under anoxia leading to increased HCR values across cell lines. The new polyol 2-nitroimidazole ana-Molecules 2023, 28, 4457 5 of 23 logues 13 and 14 were ca. 10-fold less potent under anoxia than 5, while 15 was ca. 5-fold less potent than 6 under anoxia, leading to reduced HCR values for these 2-nitroimidazoles. The 5-nitroimidazole 15 displayed low oxic potency and little hypoxic selectivity consistent with its lower electron affinity. Reversal of the sulfonamide group in compound 17 made little impact on oxic potency or hypoxic selectivity compared to the original orientation in sulfonamide 6. Preparation of 16 allowed a comparison of the effect of a weakly basic amine sidechain across the four nitroimidazole analogues (34, 36, 38 and 16) with their corresponding alcohols 5-8 in vitro and their phosphate prodrugs 9-12 in vivo. The increased solubility of the morpholines compared to the corresponding alcohols allowed determination of HCR values across the panel. The HCR values of the morpholines were broadly similar to the corresponding alcohols (e.g., 5 compared to 34, etc.) and tracked with nitroimidazole electron affinity.  anoxia leading to increased HCR values across cell lines. The new polyol 2-nitroimidazole analogues 13 and 14 were ca. 10-fold less potent under anoxia than 5, while 15 was ca. 5fold less potent than 6 under anoxia, leading to reduced HCR values for these 2-nitroimidazoles. The 5-nitroimidazole 15 displayed low oxic potency and little hypoxic selectivity consistent with its lower electron affinity. Reversal of the sulfonamide group in compound 17 made little impact on oxic potency or hypoxic selectivity compared to the original orientation in sulfonamide 6. Preparation of 16 allowed a comparison of the effect of a weakly basic amine sidechain across the four nitroimidazole analogues (34, 36, 38 and 16) with their corresponding alcohols 5-8 in vitro and their phosphate prodrugs 9-12 in vivo. The increased solubility of the morpholines compared to the corresponding alcohols allowed determination of HCR values across the panel. The HCR values of the morpholines were broadly similar to the corresponding alcohols (e.g., 5 compared to 34, etc.) and tracked with nitroimidazole electron affinity.  anoxia leading to increased HCR values across cell lines. The new polyol 2-nitroimidazole analogues 13 and 14 were ca. 10-fold less potent under anoxia than 5, while 15 was ca. 5fold less potent than 6 under anoxia, leading to reduced HCR values for these 2-nitroimidazoles. The 5-nitroimidazole 15 displayed low oxic potency and little hypoxic selectivity consistent with its lower electron affinity. Reversal of the sulfonamide group in compound 17 made little impact on oxic potency or hypoxic selectivity compared to the original orientation in sulfonamide 6. Preparation of 16 allowed a comparison of the effect of a weakly basic amine sidechain across the four nitroimidazole analogues (34, 36, 38 and 16) with their corresponding alcohols 5-8 in vitro and their phosphate prodrugs 9-12 in vivo. The increased solubility of the morpholines compared to the corresponding alcohols allowed determination of HCR values across the panel. The HCR values of the morpholines were broadly similar to the corresponding alcohols (e.g., 5 compared to 34, etc.) and tracked with nitroimidazole electron affinity. anoxia leading to increased HCR values across cell lines. The new polyol 2-nitroimidazole analogues 13 and 14 were ca. 10-fold less potent under anoxia than 5, while 15 was ca. 5fold less potent than 6 under anoxia, leading to reduced HCR values for these 2-nitroimidazoles. The 5-nitroimidazole 15 displayed low oxic potency and little hypoxic selectivity consistent with its lower electron affinity. Reversal of the sulfonamide group in compound 17 made little impact on oxic potency or hypoxic selectivity compared to the original orientation in sulfonamide 6. Preparation of 16 allowed a comparison of the effect of a weakly basic amine sidechain across the four nitroimidazole analogues (34, 36, 38 and 16) with their corresponding alcohols 5-8 in vitro and their phosphate prodrugs 9-12 in vivo. The increased solubility of the morpholines compared to the corresponding alcohols allowed determination of HCR values across the panel. The HCR values of the morpholines were broadly similar to the corresponding alcohols (e.g., 5 compared to 34, etc.) and tracked with nitroimidazole electron affinity. anoxia leading to increased HCR values across cell lines. The new polyol 2-nitroimidazole analogues 13 and 14 were ca. 10-fold less potent under anoxia than 5, while 15 was ca. 5fold less potent than 6 under anoxia, leading to reduced HCR values for these 2-nitroimidazoles. The 5-nitroimidazole 15 displayed low oxic potency and little hypoxic selectivity consistent with its lower electron affinity. Reversal of the sulfonamide group in compound 17 made little impact on oxic potency or hypoxic selectivity compared to the original orientation in sulfonamide 6. Preparation of 16 allowed a comparison of the effect of a weakly basic amine sidechain across the four nitroimidazole analogues (34, 36, 38 and 16) with their corresponding alcohols 5-8 in vitro and their phosphate prodrugs 9-12 in vivo. The increased solubility of the morpholines compared to the corresponding alcohols allowed determination of HCR values across the panel. The HCR values of the morpholines were broadly similar to the corresponding alcohols (e.g., 5 compared to 34, etc.) and tracked with nitroimidazole electron affinity.  anoxia leading to increased HCR values across cell lines. The new polyol 2-nitroimidazole analogues 13 and 14 were ca. 10-fold less potent under anoxia than 5, while 15 was ca. 5fold less potent than 6 under anoxia, leading to reduced HCR values for these 2-nitroimidazoles. The 5-nitroimidazole 15 displayed low oxic potency and little hypoxic selectivity consistent with its lower electron affinity. Reversal of the sulfonamide group in compound 17 made little impact on oxic potency or hypoxic selectivity compared to the original orientation in sulfonamide 6. Preparation of 16 allowed a comparison of the effect of a weakly basic amine sidechain across the four nitroimidazole analogues (34, 36, 38 and 16) with their corresponding alcohols 5-8 in vitro and their phosphate prodrugs 9-12 in vivo. The increased solubility of the morpholines compared to the corresponding alcohols allowed determination of HCR values across the panel. The HCR values of the morpholines were broadly similar to the corresponding alcohols (e.g., 5 compared to 34, etc.) and tracked with nitroimidazole electron affinity.  anoxia leading to increased HCR values across cell lines. The new polyol 2-nitroimidazole analogues 13 and 14 were ca. 10-fold less potent under anoxia than 5, while 15 was ca. 5fold less potent than 6 under anoxia, leading to reduced HCR values for these 2-nitroimidazoles. The 5-nitroimidazole 15 displayed low oxic potency and little hypoxic selectivity consistent with its lower electron affinity. Reversal of the sulfonamide group in compound 17 made little impact on oxic potency or hypoxic selectivity compared to the original orientation in sulfonamide 6. Preparation of 16 allowed a comparison of the effect of a weakly basic amine sidechain across the four nitroimidazole analogues (34, 36, 38 and 16) with their corresponding alcohols 5-8 in vitro and their phosphate prodrugs 9-12 in vivo. The increased solubility of the morpholines compared to the corresponding alcohols allowed determination of HCR values across the panel. The HCR values of the morpholines were broadly similar to the corresponding alcohols (e.g., 5 compared to 34, etc.) and tracked with nitroimidazole electron affinity.  anoxia leading to increased HCR values across cell lines. The new polyol 2-nitroimidazole analogues 13 and 14 were ca. 10-fold less potent under anoxia than 5, while 15 was ca. 5fold less potent than 6 under anoxia, leading to reduced HCR values for these 2-nitroimidazoles. The 5-nitroimidazole 15 displayed low oxic potency and little hypoxic selectivity consistent with its lower electron affinity. Reversal of the sulfonamide group in compound 17 made little impact on oxic potency or hypoxic selectivity compared to the original orientation in sulfonamide 6. Preparation of 16 allowed a comparison of the effect of a weakly basic amine sidechain across the four nitroimidazole analogues (34, 36, 38 and 16) with their corresponding alcohols 5-8 in vitro and their phosphate prodrugs 9-12 in vivo. The increased solubility of the morpholines compared to the corresponding alcohols allowed determination of HCR values across the panel. The HCR values of the morpholines were broadly similar to the corresponding alcohols (e.g., 5 compared to 34, etc.) and tracked with nitroimidazole electron affinity.  anoxia leading to increased HCR values across cell lines. The new polyol 2-nitroimidazole analogues 13 and 14 were ca. 10-fold less potent under anoxia than 5, while 15 was ca. 5fold less potent than 6 under anoxia, leading to reduced HCR values for these 2-nitroimidazoles. The 5-nitroimidazole 15 displayed low oxic potency and little hypoxic selectivity consistent with its lower electron affinity. Reversal of the sulfonamide group in compound 17 made little impact on oxic potency or hypoxic selectivity compared to the original orientation in sulfonamide 6. Preparation of 16 allowed a comparison of the effect of a weakly basic amine sidechain across the four nitroimidazole analogues (34, 36, 38 and 16) with their corresponding alcohols 5-8 in vitro and their phosphate prodrugs 9-12 in vivo. The increased solubility of the morpholines compared to the corresponding alcohols allowed determination of HCR values across the panel. The HCR values of the morpholines were broadly similar to the corresponding alcohols (e.g., 5 compared to 34, etc.) and tracked with nitroimidazole electron affinity.   Evaluation of the broader set of nitroimidazole sulfonamides 33-46 reinforced the previous observations, with anoxic potency being primarily a function of electron affinity with contributions from amine pKa and lipophilicity [34]. Thus, 2-nitroimidazoles 33 and 34 were strongly hypoxia selective with attenuated selectivity observed for the corresponding analogues 35 and 36 because of reduced electron affinity. This change reflects the reduced influence of the electron withdrawing sulfonamide group with a 2-carbon linker. Similarly, in the 5-nitroimidazole series, the C-2 linked analogue 39 displayed lower hypoxic selectivity than the C-1 linked analogues 37 and 38. Likewise, addition of a weakly electron-withdrawing 2-methyl substituent produced low anoxic potency and hypoxic selectivity across the series 40-44, with only the addition of more basic amine groups in 45 and 46 modestly increasing anoxic potency and hypoxic selectivity. The set of compounds recapitulated the observations from previous work that the more electron affinic 2-nitroimidazoles are likely to display both hypoxic selectivity and oxygen-mimetic radiosensitisation, although this may be mitigated by low pKa and lipophilicity as seen for 13-15. As expected, the cytotoxic potency and hypoxic selectivity of the analogues were consistent between cell lines.

In Vitro Radiosensitisation
Next, we evaluated in vitro radiosensitisation by using a representative subset of nitroimidazole compounds in HCT116/54C cells with a clonogenic survival endpoint. Radiosensitisation was determined by treating cells under anoxia with a range of radiation doses alone and with compounds at equimolar doses (1 mM). Sensitiser enhancement ratios (SER) were calculated as the ratio of radiation dose required for 1% clonogenic survival alone or with compound ( Table 3). The SER values were strongly influenced by the electron affinity of the nitroimidazole, and this was modulated by the influence of the side chain. The more electron affinic 2-nitroimidazoles were more effective sensitisers, for example, 5 compared to 7 and 6 compared to 8. Increased chain length reduced the electron 13 Evaluation of the broader set of nitroimidazole sulfonamides 33-46 reinforced the previous observations, with anoxic potency being primarily a function of electron affinity with contributions from amine pKa and lipophilicity [34]. Thus, 2-nitroimidazoles 33 and 34 were strongly hypoxia selective with attenuated selectivity observed for the corresponding analogues 35 and 36 because of reduced electron affinity. This change reflects the reduced influence of the electron withdrawing sulfonamide group with a 2-carbon linker. Similarly, in the 5-nitroimidazole series, the C-2 linked analogue 39 displayed lower hypoxic selectivity than the C-1 linked analogues 37 and 38. Likewise, addition of a weakly electron-withdrawing 2-methyl substituent produced low anoxic potency and hypoxic selectivity across the series 40-44, with only the addition of more basic amine groups in 45 and 46 modestly increasing anoxic potency and hypoxic selectivity. The set of compounds recapitulated the observations from previous work that the more electron affinic 2-nitroimidazoles are likely to display both hypoxic selectivity and oxygen-mimetic radiosensitisation, although this may be mitigated by low pKa and lipophilicity as seen for 13-15. As expected, the cytotoxic potency and hypoxic selectivity of the analogues were consistent between cell lines.

In Vitro Radiosensitisation
Next, we evaluated in vitro radiosensitisation by using a representative subset of nitroimidazole compounds in HCT116/54C cells with a clonogenic survival endpoint. Radiosensitisation was determined by treating cells under anoxia with a range of radiation doses alone and with compounds at equimolar doses (1 mM). Sensitiser enhancement ratios (SER) were calculated as the ratio of radiation dose required for 1% clonogenic survival alone or with compound ( Table 3). The SER values were strongly influenced by the electron affinity of the nitroimidazole, and this was modulated by the influence of the side chain. The more electron affinic 2-nitroimidazoles were more effective sensitisers, for example, 5 compared to 7 and 6 compared to 8. Increased chain length reduced the electron >25 13 Molecules 2023, 28 Evaluation of the broader set of nitroimidazole sulfonamides 33-46 reinforced the previous observations, with anoxic potency being primarily a function of electron affinity with contributions from amine pKa and lipophilicity [34]. Thus, 2-nitroimidazoles 33 and 34 were strongly hypoxia selective with attenuated selectivity observed for the corresponding analogues 35 and 36 because of reduced electron affinity. This change reflects the reduced influence of the electron withdrawing sulfonamide group with a 2-carbon linker. Similarly, in the 5-nitroimidazole series, the C-2 linked analogue 39 displayed lower hypoxic selectivity than the C-1 linked analogues 37 and 38. Likewise, addition of a weakly electron-withdrawing 2-methyl substituent produced low anoxic potency and hypoxic selectivity across the series 40-44, with only the addition of more basic amine groups in 45 and 46 modestly increasing anoxic potency and hypoxic selectivity. The set of compounds recapitulated the observations from previous work that the more electron affinic 2-nitroimidazoles are likely to display both hypoxic selectivity and oxygen-mimetic radiosensitisation, although this may be mitigated by low pKa and lipophilicity as seen for 13-15. As expected, the cytotoxic potency and hypoxic selectivity of the analogues were consistent between cell lines.

In Vitro Radiosensitisation
Next, we evaluated in vitro radiosensitisation by using a representative subset of nitroimidazole compounds in HCT116/54C cells with a clonogenic survival endpoint. Radiosensitisation was determined by treating cells under anoxia with a range of radiation doses alone and with compounds at equimolar doses (1 mM). Sensitiser enhancement ratios (SER) were calculated as the ratio of radiation dose required for 1% clonogenic survival alone or with compound ( Table 3). The SER values were strongly influenced by the electron affinity of the nitroimidazole, and this was modulated by the influence of the side chain. The more electron affinic 2-nitroimidazoles were more effective sensitisers, for example, 5 compared to 7 and 6 compared to 8. Increased chain length reduced the electron >20 11.0 ± 0. Evaluation of the broader set of nitroimidazole sulfonamides 33-46 reinforced the previous observations, with anoxic potency being primarily a function of electron affinity with contributions from amine pKa and lipophilicity [34]. Thus, 2-nitroimidazoles 33 and 34 were strongly hypoxia selective with attenuated selectivity observed for the corresponding analogues 35 and 36 because of reduced electron affinity. This change reflects the reduced influence of the electron withdrawing sulfonamide group with a 2-carbon linker. Similarly, in the 5-nitroimidazole series, the C-2 linked analogue 39 displayed lower hypoxic selectivity than the C-1 linked analogues 37 and 38. Likewise, addition of a weakly electron-withdrawing 2-methyl substituent produced low anoxic potency and hypoxic selectivity across the series 40-44, with only the addition of more basic amine groups in 45 and 46 modestly increasing anoxic potency and hypoxic selectivity. The set of compounds recapitulated the observations from previous work that the more electron affinic 2-nitroimidazoles are likely to display both hypoxic selectivity and oxygen-mimetic radiosensitisation, although this may be mitigated by low pKa and lipophilicity as seen for 13-15. As expected, the cytotoxic potency and hypoxic selectivity of the analogues were consistent between cell lines.

In Vitro Radiosensitisation
Next, we evaluated in vitro radiosensitisation by using a representative subset of nitroimidazole compounds in HCT116/54C cells with a clonogenic survival endpoint. Radiosensitisation was determined by treating cells under anoxia with a range of radiation doses alone and with compounds at equimolar doses (1 mM). Sensitiser enhancement ratios (SER) were calculated as the ratio of radiation dose required for 1% clonogenic survival alone or with compound ( Table 3). The SER values were strongly influenced by the electron affinity of the nitroimidazole, and this was modulated by the influence of the side chain. The more electron affinic 2-nitroimidazoles were more effective sensitisers, for example, 5 compared to 7 and 6 compared to 8. Increased chain length reduced the electron >18 15 Evaluation of the broader set of nitroimidazole sulfonamides 33-46 reinforced the previous observations, with anoxic potency being primarily a function of electron affinity with contributions from amine pKa and lipophilicity [34]. Thus, 2-nitroimidazoles 33 and 34 were strongly hypoxia selective with attenuated selectivity observed for the corresponding analogues 35 and 36 because of reduced electron affinity. This change reflects the reduced influence of the electron withdrawing sulfonamide group with a 2-carbon linker. Similarly, in the 5-nitroimidazole series, the C-2 linked analogue 39 displayed lower hypoxic selectivity than the C-1 linked analogues 37 and 38. Likewise, addition of a weakly electron-withdrawing 2-methyl substituent produced low anoxic potency and hypoxic selectivity across the series 40-44, with only the addition of more basic amine groups in 45 and 46 modestly increasing anoxic potency and hypoxic selectivity. The set of compounds recapitulated the observations from previous work that the more electron affinic 2-nitroimidazoles are likely to display both hypoxic selectivity and oxygen-mimetic radiosensitisation, although this may be mitigated by low pKa and lipophilicity as seen for 13-15. As expected, the cytotoxic potency and hypoxic selectivity of the analogues were consistent between cell lines.

In Vitro Radiosensitisation
Next, we evaluated in vitro radiosensitisation by using a representative subset of nitroimidazole compounds in HCT116/54C cells with a clonogenic survival endpoint. Radiosensitisation was determined by treating cells under anoxia with a range of radiation doses alone and with compounds at equimolar doses (1 mM). Sensitiser enhancement ratios (SER) were calculated as the ratio of radiation dose required for 1% clonogenic survival alone or with compound ( Table 3). The SER values were strongly influenced by the electron affinity of the nitroimidazole, and this was modulated by the influence of the side chain. The more electron affinic 2-nitroimidazoles were more effective sensitisers, for example, 5 compared to 7 and 6 compared to 8. Increased chain length reduced the electron Evaluation of the broader set of nitroimidazole sulfonamides 33-46 reinforced the previous observations, with anoxic potency being primarily a function of electron affinity with contributions from amine pKa and lipophilicity [34]. Thus, 2-nitroimidazoles 33 and 34 were strongly hypoxia selective with attenuated selectivity observed for the corresponding analogues 35 and 36 because of reduced electron affinity. This change reflects the reduced influence of the electron withdrawing sulfonamide group with a 2-carbon linker. Similarly, in the 5-nitroimidazole series, the C-2 linked analogue 39 displayed lower hypoxic selectivity than the C-1 linked analogues 37 and 38. Likewise, addition of a weakly electron-withdrawing 2-methyl substituent produced low anoxic potency and hypoxic selectivity across the series 40-44, with only the addition of more basic amine groups in 45 and 46 modestly increasing anoxic potency and hypoxic selectivity. The set of compounds recapitulated the observations from previous work that the more electron affinic 2-nitroimidazoles are likely to display both hypoxic selectivity and oxygen-mimetic radiosensitisation, although this may be mitigated by low pKa and lipophilicity as seen for 13-15. As expected, the cytotoxic potency and hypoxic selectivity of the analogues were consistent between cell lines.

In Vitro Radiosensitisation
Next, we evaluated in vitro radiosensitisation by using a representative subset of nitroimidazole compounds in HCT116/54C cells with a clonogenic survival endpoint. Radiosensitisation was determined by treating cells under anoxia with a range of radiation doses alone and with compounds at equimolar doses (1 mM). Sensitiser enhancement ratios (SER) were calculated as the ratio of radiation dose required for 1% clonogenic survival alone or with compound ( Table 3). The SER values were strongly influenced by the electron affinity of the nitroimidazole, and this was modulated by the influence of the side chain. The more electron affinic 2-nitroimidazoles were more effective sensitisers, for example, 5 compared to 7 and 6 compared to 8. Increased chain length reduced the electron 11 Evaluation of the broader set of nitroimidazole sulfonamides 33-46 reinforced the previous observations, with anoxic potency being primarily a function of electron affinity with contributions from amine pKa and lipophilicity [34]. Thus, 2-nitroimidazoles 33 and 34 were strongly hypoxia selective with attenuated selectivity observed for the corresponding analogues 35 and 36 because of reduced electron affinity. This change reflects the reduced influence of the electron withdrawing sulfonamide group with a 2-carbon linker. Similarly, in the 5-nitroimidazole series, the C-2 linked analogue 39 displayed lower hypoxic selectivity than the C-1 linked analogues 37 and 38. Likewise, addition of a weakly electron-withdrawing 2-methyl substituent produced low anoxic potency and hypoxic selectivity across the series 40-44, with only the addition of more basic amine groups in 45 and 46 modestly increasing anoxic potency and hypoxic selectivity. The set of compounds recapitulated the observations from previous work that the more electron affinic 2-nitroimidazoles are likely to display both hypoxic selectivity and oxygen-mimetic radiosensitisation, although this may be mitigated by low pKa and lipophilicity as seen for 13-15. As expected, the cytotoxic potency and hypoxic selectivity of the analogues were consistent between cell lines.

In Vitro Radiosensitisation
Next, we evaluated in vitro radiosensitisation by using a representative subset of nitroimidazole compounds in HCT116/54C cells with a clonogenic survival endpoint. Radiosensitisation was determined by treating cells under anoxia with a range of radiation doses alone and with compounds at equimolar doses (1 mM). Sensitiser enhancement ratios (SER) were calculated as the ratio of radiation dose required for 1% clonogenic survival alone or with compound ( Table 3). The SER values were strongly influenced by the electron affinity of the nitroimidazole, and this was modulated by the influence of the side chain. The more electron affinic 2-nitroimidazoles were more effective sensitisers, for example, 5 compared to 7 and 6 compared to 8. Increased chain length reduced the electron  Evaluation of the broader set of nitroimidazole sulfonamides 33-46 reinforced the previous observations, with anoxic potency being primarily a function of electron affinity with contributions from amine pKa and lipophilicity [34]. Thus, 2-nitroimidazoles 33 and 34 were strongly hypoxia selective with attenuated selectivity observed for the corresponding analogues 35 and 36 because of reduced electron affinity. This change reflects the reduced influence of the electron withdrawing sulfonamide group with a 2-carbon linker. Similarly, in the 5-nitroimidazole series, the C-2 linked analogue 39 displayed lower hypoxic selectivity than the C-1 linked analogues 37 and 38. Likewise, addition of a weakly electron-withdrawing 2-methyl substituent produced low anoxic potency and hypoxic selectivity across the series 40-44, with only the addition of more basic amine groups in 45 and 46 modestly increasing anoxic potency and hypoxic selectivity. The set of compounds recapitulated the observations from previous work that the more electron affinic 2-nitroimidazoles are likely to display both hypoxic selectivity and oxygen-mimetic radiosensitisation, although this may be mitigated by low pKa and lipophilicity as seen for 13-15. As expected, the cytotoxic potency and hypoxic selectivity of the analogues were consistent between cell lines.

In Vitro Radiosensitisation
Next, we evaluated in vitro radiosensitisation by using a representative subset of nitroimidazole compounds in HCT116/54C cells with a clonogenic survival endpoint. Radiosensitisation was determined by treating cells under anoxia with a range of radiation doses alone and with compounds at equimolar doses (1 mM). Sensitiser enhancement ratios (SER) were calculated as the ratio of radiation dose required for 1% clonogenic survival alone or with compound ( Table 3). The SER values were strongly influenced by the electron affinity of the nitroimidazole, and this was modulated by the influence of the side chain. The more electron affinic 2-nitroimidazoles were more effective sensitisers, for example, 5 compared to 7 and 6 compared to 8. Increased chain length reduced the electron 11 Evaluation of the broader set of nitroimidazole sulfonamides 33-46 reinforced the previous observations, with anoxic potency being primarily a function of electron affinity with contributions from amine pKa and lipophilicity [34]. Thus, 2-nitroimidazoles 33 and 34 were strongly hypoxia selective with attenuated selectivity observed for the corresponding analogues 35 and 36 because of reduced electron affinity. This change reflects the reduced influence of the electron withdrawing sulfonamide group with a 2-carbon linker. Similarly, in the 5-nitroimidazole series, the C-2 linked analogue 39 displayed lower hypoxic selectivity than the C-1 linked analogues 37 and 38. Likewise, addition of a weakly electron-withdrawing 2-methyl substituent produced low anoxic potency and hypoxic selectivity across the series 40-44, with only the addition of more basic amine groups in 45 and 46 modestly increasing anoxic potency and hypoxic selectivity. The set of compounds recapitulated the observations from previous work that the more electron affinic 2-nitroimidazoles are likely to display both hypoxic selectivity and oxygen-mimetic radiosensitisation, although this may be mitigated by low pKa and lipophilicity as seen for 13-15. As expected, the cytotoxic potency and hypoxic selectivity of the analogues were consistent between cell lines.

In Vitro Radiosensitisation
Next, we evaluated in vitro radiosensitisation by using a representative subset of nitroimidazole compounds in HCT116/54C cells with a clonogenic survival endpoint. Radiosensitisation was determined by treating cells under anoxia with a range of radiation doses alone and with compounds at equimolar doses (1 mM). Sensitiser enhancement ratios (SER) were calculated as the ratio of radiation dose required for 1% clonogenic survival alone or with compound ( Table 3). The SER values were strongly influenced by the electron affinity of the nitroimidazole, and this was modulated by the influence of the side chain. The more electron affinic 2-nitroimidazoles were more effective sensitisers, for example, 5 compared to 7 and 6 compared to 8. Increased chain length reduced the electron 16 Evaluation of the broader set of nitroimidazole sulfonamides 33-46 reinforced the previous observations, with anoxic potency being primarily a function of electron affinity with contributions from amine pKa and lipophilicity [34]. Thus, 2-nitroimidazoles 33 and 34 were strongly hypoxia selective with attenuated selectivity observed for the corresponding analogues 35 and 36 because of reduced electron affinity. This change reflects the reduced influence of the electron withdrawing sulfonamide group with a 2-carbon linker. Similarly, in the 5-nitroimidazole series, the C-2 linked analogue 39 displayed lower hypoxic selectivity than the C-1 linked analogues 37 and 38. Likewise, addition of a weakly electron-withdrawing 2-methyl substituent produced low anoxic potency and hypoxic selectivity across the series 40-44, with only the addition of more basic amine groups in 45 and 46 modestly increasing anoxic potency and hypoxic selectivity. The set of compounds recapitulated the observations from previous work that the more electron affinic 2-nitroimidazoles are likely to display both hypoxic selectivity and oxygen-mimetic radiosensitisation, although this may be mitigated by low pKa and lipophilicity as seen for 13-15. As expected, the cytotoxic potency and hypoxic selectivity of the analogues were consistent between cell lines.

In Vitro Radiosensitisation
Next, we evaluated in vitro radiosensitisation by using a representative subset of nitroimidazole compounds in HCT116/54C cells with a clonogenic survival endpoint. Radiosensitisation was determined by treating cells under anoxia with a range of radiation doses alone and with compounds at equimolar doses (1 mM). Sensitiser enhancement ratios (SER) were calculated as the ratio of radiation dose required for 1% clonogenic survival alone or with compound ( Table 3). The SER values were strongly influenced by the electron affinity of the nitroimidazole, and this was modulated by the influence of the side chain. The more electron affinic 2-nitroimidazoles were more effective sensitisers, for example, 5 compared to 7 and 6 compared to 8. Increased chain length reduced the electron withdrawal of the sulfonamide group and consequently reduced SER values, for example, 5 compared with 6 and 7 compared with 8. The character of the sidechain also influenced radiosensitisation with reduced lipophilicity resulting in reductions in SER values. This was most dramatically seen in the loss of activity observed for diol 13 and triol 14 compared to alcohol 5. Addition of the weakly basic morpholine sidechain provided increased activity, for example, 16 compared with 8, 34 compared with 5, 36 compared with 6; two 5-nitroimidazole examples provided similar sensitisation, for example, 38 and 42 compared to 7 and 41, respectively. Overall, it was clear that the new analogues did not provide enhanced sensitisation compared to existing analogues and that compounds with a morpholine solubilising side chain provided sufficient solubility [34] to explore the most active 2-nitroimidazole 34 and 5-nitroimidazole 38 sensitisers further in vivo. withdrawal of the sulfonamide group and consequently reduced SER values, for example, 5 compared with 6 and 7 compared with 8. The character of the sidechain also influenced radiosensitisation with reduced lipophilicity resulting in reductions in SER values. This was most dramatically seen in the loss of activity observed for diol 13 and triol 14 compared to alcohol 5. Addition of the weakly basic morpholine sidechain provided increased activity, for example, 16 compared with 8, 34 compared with 5, 36 compared with 6; two 5-nitroimidazole examples provided similar sensitisation, for example, 38 and 42 compared to 7 and 41, respectively. Overall, it was clear that the new analogues did not provide enhanced sensitisation compared to existing analogues and that compounds with a morpholine solubilising side chain provided sufficient solubility [34] to explore the most active 2-nitroimidazole 34 and 5-nitroimidazole 38 sensitisers further in vivo. withdrawal of the sulfonamide group and consequently reduced SER values, for example, 5 compared with 6 and 7 compared with 8. The character of the sidechain also influenced radiosensitisation with reduced lipophilicity resulting in reductions in SER values. This was most dramatically seen in the loss of activity observed for diol 13 and triol 14 compared to alcohol 5. Addition of the weakly basic morpholine sidechain provided increased activity, for example, 16 compared with 8, 34 compared with 5, 36 compared with 6; two 5-nitroimidazole examples provided similar sensitisation, for example, 38 and 42 compared to 7 and 41, respectively. Overall, it was clear that the new analogues did not provide enhanced sensitisation compared to existing analogues and that compounds with a morpholine solubilising side chain provided sufficient solubility [34] to explore the most active 2-nitroimidazole 34 and 5-nitroimidazole 38 sensitisers further in vivo. withdrawal of the sulfonamide group and consequently reduced SER values, for example, 5 compared with 6 and 7 compared with 8. The character of the sidechain also influenced radiosensitisation with reduced lipophilicity resulting in reductions in SER values. This was most dramatically seen in the loss of activity observed for diol 13 and triol 14 compared to alcohol 5. Addition of the weakly basic morpholine sidechain provided increased activity, for example, 16 compared with 8, 34 compared with 5, 36 compared with 6; two 5-nitroimidazole examples provided similar sensitisation, for example, 38 and 42 compared to 7 and 41, respectively. Overall, it was clear that the new analogues did not provide enhanced sensitisation compared to existing analogues and that compounds with a morpholine solubilising side chain provided sufficient solubility [34] to explore the most active 2-nitroimidazole 34 and 5-nitroimidazole 38 sensitisers further in vivo. withdrawal of the sulfonamide group and consequently reduced SER values, for example, 5 compared with 6 and 7 compared with 8. The character of the sidechain also influenced radiosensitisation with reduced lipophilicity resulting in reductions in SER values. This was most dramatically seen in the loss of activity observed for diol 13 and triol 14 compared to alcohol 5. Addition of the weakly basic morpholine sidechain provided increased activity, for example, 16 compared with 8, 34 compared with 5, 36 compared with 6; two 5-nitroimidazole examples provided similar sensitisation, for example, 38 and 42 compared to 7 and 41, respectively. Overall, it was clear that the new analogues did not provide enhanced sensitisation compared to existing analogues and that compounds with a morpholine solubilising side chain provided sufficient solubility [34] to explore the most active 2-nitroimidazole 34 and 5-nitroimidazole 38 sensitisers further in vivo. withdrawal of the sulfonamide group and consequently reduced SER values, for example, 5 compared with 6 and 7 compared with 8. The character of the sidechain also influenced radiosensitisation with reduced lipophilicity resulting in reductions in SER values. This was most dramatically seen in the loss of activity observed for diol 13 and triol 14 compared to alcohol 5. Addition of the weakly basic morpholine sidechain provided increased activity, for example, 16 compared with 8, 34 compared with 5, 36 compared with 6; two 5-nitroimidazole examples provided similar sensitisation, for example, 38 and 42 compared to 7 and 41, respectively. Overall, it was clear that the new analogues did not provide enhanced sensitisation compared to existing analogues and that compounds with a morpholine solubilising side chain provided sufficient solubility [34] to explore the most active 2-nitroimidazole 34 and 5-nitroimidazole 38 sensitisers further in vivo. withdrawal of the sulfonamide group and consequently reduced SER values, for example, 5 compared with 6 and 7 compared with 8. The character of the sidechain also influenced radiosensitisation with reduced lipophilicity resulting in reductions in SER values. This was most dramatically seen in the loss of activity observed for diol 13 and triol 14 compared to alcohol 5. Addition of the weakly basic morpholine sidechain provided increased activity, for example, 16 compared with 8, 34 compared with 5, 36 compared with 6; two 5-nitroimidazole examples provided similar sensitisation, for example, 38 and 42 compared to 7 and 41, respectively. Overall, it was clear that the new analogues did not provide enhanced sensitisation compared to existing analogues and that compounds with a morpholine solubilising side chain provided sufficient solubility [34] to explore the most active 2-nitroimidazole 34 and 5-nitroimidazole 38 sensitisers further in vivo. withdrawal of the sulfonamide group and consequently reduced SER values, for example, 5 compared with 6 and 7 compared with 8. The character of the sidechain also influenced radiosensitisation with reduced lipophilicity resulting in reductions in SER values. This was most dramatically seen in the loss of activity observed for diol 13 and triol 14 compared to alcohol 5. Addition of the weakly basic morpholine sidechain provided increased activity, for example, 16 compared with 8, 34 compared with 5, 36 compared with 6; two 5-nitroimidazole examples provided similar sensitisation, for example, 38 and 42 compared to 7 and 41, respectively. Overall, it was clear that the new analogues did not provide enhanced sensitisation compared to existing analogues and that compounds with a morpholine solubilising side chain provided sufficient solubility [34] to explore the most active 2-nitroimidazole 34 and 5-nitroimidazole 38 sensitisers further in vivo. withdrawal of the sulfonamide group and consequently reduced SER values, for example, 5 compared with 6 and 7 compared with 8. The character of the sidechain also influenced radiosensitisation with reduced lipophilicity resulting in reductions in SER values. This was most dramatically seen in the loss of activity observed for diol 13 and triol 14 compared to alcohol 5. Addition of the weakly basic morpholine sidechain provided increased activity, for example, 16 compared with 8, 34 compared with 5, 36 compared with 6; two 5-nitroimidazole examples provided similar sensitisation, for example, 38 and 42 compared to 7 and 41, respectively. Overall, it was clear that the new analogues did not provide enhanced sensitisation compared to existing analogues and that compounds with a morpholine solubilising side chain provided sufficient solubility [34] to explore the most active 2-nitroimidazole 34 and 5-nitroimidazole 38 sensitisers further in vivo. withdrawal of the sulfonamide group and consequently reduced SER values, for example, 5 compared with 6 and 7 compared with 8. The character of the sidechain also influenced radiosensitisation with reduced lipophilicity resulting in reductions in SER values. This was most dramatically seen in the loss of activity observed for diol 13 and triol 14 compared to alcohol 5. Addition of the weakly basic morpholine sidechain provided increased activity, for example, 16 compared with 8, 34 compared with 5, 36 compared with 6; two 5-nitroimidazole examples provided similar sensitisation, for example, 38 and 42 compared to 7 and 41, respectively. Overall, it was clear that the new analogues did not provide enhanced sensitisation compared to existing analogues and that compounds with a morpholine solubilising side chain provided sufficient solubility [34] to explore the most active 2-nitroimidazole 34 and 5-nitroimidazole 38 sensitisers further in vivo. withdrawal of the sulfonamide group and consequently reduced SER values, for example, 5 compared with 6 and 7 compared with 8. The character of the sidechain also influenced radiosensitisation with reduced lipophilicity resulting in reductions in SER values. This was most dramatically seen in the loss of activity observed for diol 13 and triol 14 compared to alcohol 5. Addition of the weakly basic morpholine sidechain provided increased activity, for example, 16 compared with 8, 34 compared with 5, 36 compared with 6; two 5-nitroimidazole examples provided similar sensitisation, for example, 38 and 42 compared to 7 and 41, respectively. Overall, it was clear that the new analogues did not provide enhanced sensitisation compared to existing analogues and that compounds with a morpholine solubilising side chain provided sufficient solubility [34] to explore the most active 2-nitroimidazole 34 and 5-nitroimidazole 38 sensitisers further in vivo. withdrawal of the sulfonamide group and consequently reduced SER values, for example, 5 compared with 6 and 7 compared with 8. The character of the sidechain also influenced radiosensitisation with reduced lipophilicity resulting in reductions in SER values. This was most dramatically seen in the loss of activity observed for diol 13 and triol 14 compared to alcohol 5. Addition of the weakly basic morpholine sidechain provided increased activity, for example, 16 compared with 8, 34 compared with 5, 36 compared with 6; two 5-nitroimidazole examples provided similar sensitisation, for example, 38 and 42 compared to 7 and 41, respectively. Overall, it was clear that the new analogues did not provide enhanced sensitisation compared to existing analogues and that compounds with a morpholine solubilising side chain provided sufficient solubility [34] to explore the most active 2-nitroimidazole 34 and 5-nitroimidazole 38 sensitisers further in vivo. withdrawal of the sulfonamide group and consequently reduced SER values, for example, 5 compared with 6 and 7 compared with 8. The character of the sidechain also influenced radiosensitisation with reduced lipophilicity resulting in reductions in SER values. This was most dramatically seen in the loss of activity observed for diol 13 and triol 14 compared to alcohol 5. Addition of the weakly basic morpholine sidechain provided increased activity, for example, 16 compared with 8, 34 compared with 5, 36 compared with 6; two 5-nitroimidazole examples provided similar sensitisation, for example, 38 and 42 compared to 7 and 41, respectively. Overall, it was clear that the new analogues did not provide enhanced sensitisation compared to existing analogues and that compounds with a morpholine solubilising side chain provided sufficient solubility [34] to explore the most active 2-nitroimidazole 34 and 5-nitroimidazole 38 sensitisers further in vivo. withdrawal of the sulfonamide group and consequently reduced SER values, for example, 5 compared with 6 and 7 compared with 8. The character of the sidechain also influenced radiosensitisation with reduced lipophilicity resulting in reductions in SER values. This was most dramatically seen in the loss of activity observed for diol 13 and triol 14 compared to alcohol 5. Addition of the weakly basic morpholine sidechain provided increased activity, for example, 16 compared with 8, 34 compared with 5, 36 compared with 6; two 5-nitroimidazole examples provided similar sensitisation, for example, 38 and 42 compared to 7 and 41, respectively. Overall, it was clear that the new analogues did not provide enhanced sensitisation compared to existing analogues and that compounds with a morpholine solubilising side chain provided sufficient solubility [34] to explore the most active 2-nitroimidazole 34 and 5-nitroimidazole 38 sensitisers further in vivo.

In Vivo Pharmacokinetics
The most promising 2-nitroimidazole 34 and 5-nitroimidazole 38 compounds were selected for in vivo evaluation on the basis of solubility, in vitro cytotoxicity and radiosensitisation data. These compounds were both well tolerated by i.v. injection in NIH-III mice at their maximum achievable dose (MAD) of 2.2 mmol/kg. This MAD was also achieved for some of our earlier compounds, but only when we employed a phosphate prodrug strategy owing to their poor solubility [34]. Pharmacokinetic evaluation of 2-nitroimidazole 34 achieved a maximum concentration (C max ) of 1.53 mM 5 min after dosing with a terminal half-life of 1.52 h and AUC of 0.96 mM·h ( Table 4). The C max and AUC exceeded that achieved for earlier phosphate prodrugs 9-12, but was less than 2. On this basis we advanced both morpholine analogues into in vivo radiosensitisation studies.

In Vivo Radiosensitisation
Compounds 34 and 38 were evaluated in an ex vivo clonogenic assay using HCT116/54C tumour xenograft model as compared with 2 and our four phosphate prodrugs 9-12 that had previously demonstrated activity in the same assay in HCT116 parental tumours [34]. Compounds 34 and 38 both significantly enhanced the tumour killing of radiotherapy by targeting the hypoxic cells that radiation spares (Figure 2A), as did 2, 9 and 12, but not 10 and 11. The 2-nitroimidazole 34 had an in vivo sensitisation ratio (SR = surviving fraction after radiation/surviving fraction after drug + radiation) of 14.9 in HCT116/54C tumours, which was superior to that of 2 and the phosphate prodrugs 9-12 ( Table 5). The SR of 38 was 3.43, which was comparable to 2, but 1.8-fold lower than 9. A second tumour model, UT-SCC-74B, was used to confirm in vivo radiosensitisation. All four compounds tested (2, 9, 34 and 38) significantly enhanced tumour cell killing by radiotherapy, with slightly higher SR values of 13.8 and 11.5 for 34 and 38, respectively, than those seen with 2 and phosphate prodrug 9 ( Figure 2B, Table 5). Evidence of single agent activity was observed for 34 in the ex vivo clonogenic assays with 62.3 ± 7.9% cell kill observed in HCT116/54C tumours and 68.4 ± 1.2% cell kill in UT-SCC-74B tumours ( Figure 2) relative to untreated tumours (p < 0.005 and p < 0.0001, respectively). Among the other compounds evaluated, only 9 (65.7 ± 8.2% cell kill; p < 0.005) in HCT116/54C tumours and 2 (55.9 ± 1.1% cell kill; p < 0.0005) in UT-SCC-74B tumours demonstrated significant single agent activity.
Since nitroimidazoles may possess dual mechanisms of cell killing (both electronaffinic oxygen mimetic radiosensitisation and hypoxia-selective cytotoxicity), we evaluated the mechanism of action for 34 and 38. Oxygen-mimetic sensitisers must be present at the time of irradiation to work, whereas hypoxia-selective cytotoxins may exert their effect even when given after radiation. Drugs were dosed either 5 min before radiation or 20 min after radiation in mice with HCT116/54C tumours in the ex vivo clonogenic assay. Compounds 34 and 38 both significantly enhanced radiation cell kill when administered 5 min before radiation with SR values of 9.76 and 8.26, while no significant additional killing over radiation was observed when given after radiation (SR values of 1.49 and 1.38, respectively, Figure 3). This confirms their major mode of action as oxygen-mimetic radiosensitisers rather than hypoxia-selective cytotoxins.

Inhibition of Tumour Growth
The ability of 34 and 38 to inhibit tumour growth was evaluated in HCT116/54C and UT-SCC-74B xenograft models and compared to 2 and phosphate prodrug 9. In mice with HCT116/54C and UT-SCC-74B tumours, a single dose of 34 was able to sensitise 10 Gy radiation to a similar level observed with 2, while 38 and 9 had similar activity to radiation alone ( Figure 4A). In a further comparison using a fractionated radiation schedule Molecules 2023, 28, 4457 9 of 23 (qw × 3, 5 Gy) in HCT116/54C tumours, both 34 and 38 delayed tumour growth relative to radiation alone, with similar activity to 2. All treatments were well tolerated in both tumour models with no or minor reductions in bodyweight compared to radiation alone ( Figure 4B). On evaluation of survival analysis, assessed as time for tumours to quadruple in size following treatment initiation (RTV 4 , relative tumour volume 4×), 2, 34 and 38 prior to radiation significantly increased survival compared to radiation alone (p < 0.05), but only in HCT116/54C tumours treated with fractionated radiation ( Figure 4C). There was no significant difference in activity between any of the radiosensitisers in any tumour model.  Since nitroimidazoles may possess dual mechanisms of cell killing (both electronaffinic oxygen mimetic radiosensitisation and hypoxia-selective cytotoxicity), we evaluated the mechanism of action for 34 and 38. Oxygen-mimetic sensitisers must be present at the time of irradiation to work, whereas hypoxia-selective cytotoxins may exert their effect even when given after radiation. Drugs were dosed either 5 min before radiation or 20 min after radiation in mice with HCT116/54C tumours in the ex vivo clonogenic assay.  10 Gy radiation (RAD) alone or 2.2 mmol/kg drug combined with 10 Gy radiation (n = 3-5), and plated to determine the number of surviving cells relative to untreated controls. ****, p < 0.0001; *, p < 0.05 for comparison of RAD vs. drug + RAD by one-way ANOVA with Sidak's multiple comparison analysis. Molecules 2023, 28, x FOR PEER REVIEW 10 of 23 respectively, Figure 3). This confirms their major mode of action as oxygen-mimetic radiosensitisers rather than hypoxia-selective cytotoxins.

Inhibition of Tumour Growth
The ability of 34 and 38 to inhibit tumour growth was evaluated in HCT116/54C and UT-SCC-74B xenograft models and compared to 2 and phosphate prodrug 9. In mice with HCT116/54C and UT-SCC-74B tumours, a single dose of 34 was able to sensitise 10 Gy radiation to a similar level observed with 2, while 38 and 9 had similar activity to radiation alone ( Figure 4A). In a further comparison using a fractionated radiation schedule (qw × 3, 5 Gy) in HCT116/54C tumours, both 34 and 38 delayed tumour growth relative to radiation alone, with similar activity to 2. All treatments were well tolerated in both tumour models with no or minor reductions in bodyweight compared to radiation alone ( Figure  4B). On evaluation of survival analysis, assessed as time for tumours to quadruple in size following treatment initiation (RTV 4 , relative tumour volume 4×), 2, 34 and 38 prior to radiation significantly increased survival compared to radiation alone (p < 0.05), but only in HCT116/54C tumours treated with fractionated radiation ( Figure 4C). There was no significant difference in activity between any of the radiosensitisers in any tumour model.

Discussion
The development of effective radiosensitisers has been of increasing interest in recent years [5][6][7]. One of the key recommendations from these consensus statements has been that the addition of sensitisers to radiation treatment schedules should not exacerbate nor-

Discussion
The development of effective radiosensitisers has been of increasing interest in recent years [5][6][7]. One of the key recommendations from these consensus statements has been that the addition of sensitisers to radiation treatment schedules should not exacerbate normal tissue toxicities because the therapeutic ratio has been titrated to a maximum with little tolerance in normal tissues for further toxicity. Oxygen-mimetic nitroimidazole radiosensitisers are not a new idea, but evolving treatment paradigms present a new opportunity for this class. Nitroimidazole radiosensitisers are designed to oxidise DNA radicals produced by radiation and to cause DNA strand breaks analogously to oxygen [11]. This physicochemical process is largely independent of cell type or genetic background, and it signals the opportunity to expand the use of these radiosensitisers into other tumour types and treatment modalities where RT is limited by hypoxia. Importantly, the rate of reaction with DNA radicals is considerably less than oxygen, limiting their activity to hypoxic tissue and increasing the effective radiation dose to a treatment-resistant compartment without creating additional toxicities.
Having identified a new chemical class of nitroimidazole radiosensitisers where a sulfonamide linking group provides novelty and modulates the nitroimidazole electron affinity, we have taken the opportunity to design new radiosensitisers that build on known structure activity relationships and discover lead compounds with clinical potential. Although we had previously identified effective radiosensitisers in 5-8, these required a phosphate prodrug approach to provide compounds (9)(10)(11)(12) sufficiently soluble for in vivo use [34]. Here, we explored the use of polyol sidechains 13-15 to enhance aqueous solubility as demonstrated for 4 and related compounds [39]. We also explored reversal of the key sulfonamide group in 17 and, with compound 16, completed an analogous set of morpholine analogues (34, 36, 38 and 16) to the effective alcohols 5-8, respectively. The use of polyol or morpholine sidechains provided useful increases in aqueous solubility, but aqueous solubilities were considerably less than the phosphates 9-12 [34]. The polyols 13-15 showed weak hypoxic selectivity (Table 2) and were ineffective as radiosensitisers (Table 3). Reversing the sulfonamide moiety in 17 made little difference in either hypoxic selectivity or radiosensitisation compared to the analogous 6. Comparison of the weakly basic amines 34, 36, 38 and 16 with their corresponding alcohols 5-8 in vitro showed increased solubility of the morpholines compared to the corresponding alcohols, which was sufficient to allow determination of HCR values across the panel. The HCR values of the morpholines were broadly similar to the corresponding alcohols (e.g., 5 and 34, etc.) and tracked with nitroimidazole electron affinity. When evaluated as radiosensitisers in cells at equimolar concentrations, the SER values tracked with electron affinity with 34 and 36 showing strong radiosensitisation. The morpholine analogues showed increased (34, 36 and 16) or similar (38) radiosensitisation compared to the alcohols 5, 6, 8 and 7, respectively.
We elected to advance 34 and 38 to in vivo evaluation as the best examples of 2-and 5-nitroimidazole analogues, respectively. The compounds were both well tolerated by NIH-III mice and could be given at a MAD of 2.2 mmol/kg, allowing comparison with 2 and the phosphate prodrugs 9-12 at equimolar doses. We evaluated 2, 9-12 and 34 and 38 using an ex vivo excision assay to determine how well they sensitised HCT116/54C tumour xenografts to a single 10 Gy dose of radiation. All compounds provided significant additional cell killing in combination with radiation with the 2-nitroimidazoles 2, 9 and 34 giving strong SR values. Most compounds were inactive as a single agent, except for a small amount of cell killing for 2-nitroimidazoles with high electron affinity in the HCT116/54C (9 and 34) and UT-SCC-74B (2 and 34) models, suggesting that hypoxia-selective cytotoxicity might have contributed to the sensitising ability of these compounds. To further explore the potential contribution of hypoxia-selective killing to radiosensitisation, we compared 34 and 38 when given 5 min before radiation and 20 min after radiation. When given after radiation, the compounds cannot participate in the oxygen mimetic sensitisation process and so any additional cell killing reflects the contribution from hypoxia-selective cytotoxicity to tumour cell killing. Compounds 34 and 38 provided ca. one log of additional cell killing when given 5 min prior to radiation, and they provided a small (but nonsignificant) amount of additional cell killing compared to radiation when given post radiation. While 34 did show substantial HCR values in vitro (HCR values of [23][24][25][26][27][28][29][30][31][32][33][34][35][36][37][38] and some single agent activity in ex vivo excision assays, the small effect seen post-irradiation in this experiment suggests that oxygen-mimetic radiosensitisation is the dominant mechanism for the two nitroimidazoles 34 and 38. One of the limitations in the clinical development of previous radiosensitisers was use with fractionated RT, where fractionation of the radiation dose is designed to allow tumour reoxygenation between radiation doses, thus, reducing the potential for radiosensitisation [40][41][42]. Administering a dose of radiosensitiser with each fraction of radiation was unachievable for misonidazole and etanidazole due to cumulative peripheral neurotoxicity [43,44]. The growing use of stereotactic body radiotherapy (SBRT), which delivers hypo-fractionated (1-8 doses), high-dose (25-60 Gy total dose) radiation to treat tumours [45,46], offers new opportunity for radiosensitisers in hypo-fractionated schedules. This new approach leverages recent advances in the accuracy and precision of radiation delivery to allow dose intensification to small tumours while minimising the effects to adjacent normal tissue. Clinical trials using SBRT to treat various solid tumours have demonstrated comparable control, toxicity and efficacy profiles to fractionated RT [46][47][48], and SBRT is being extensively explored in primary and recurrent HNSCC [49][50][51]. The reduced treatment time and number of patient visits, combined with emerging potential to replace surgery with an outpatient procedure, indicates substantial health, social and economic advantages for SBRT and is driving increasing use of SBRT. However, evidence is emerging that SBRT accentuates the role of hypoxia in radioresistance [52][53][54][55] because the opportunity for reoxygenation is reduced [56,57], and thus, the impact of a radiosensitiser is increased [52,58,59].
We wished to compare our best radiosensitisers 9, 34 and 38 with etanidazole (2) in tumour xenograft models using SBRT-like radiation schedules. We compared the compounds in the HCT116/54C and UT-SCC-74B models using a single 10 Gy radiation dose as well as a hypo-fractionated 3 × 5 Gy schedule in the HCT116/54C model. We evaluated drug toxicity as bodyweight loss and observed that the radiosensitisers were well tolerated compared to radiation alone and produced no overt clinical signs of toxicity. Etanidazole (2) and 34 appeared to provide additional tumour control compared to radiation in the HCT116/54C and UT-SCC-74B 10 Gy single dose models, but these did not translate to significantly prolonged survival times. In the fractionated 5 Gy (qw × 3) dose model in HCT116/54C tumours 2, 34 and 38, all significantly increased survival compared to radiation alone, indicating that both 34 and 38 can effectively sensitise radiation to a similar degree to 2 in SBRT-like schedules. The greater activity observed with the fractionated schedule of 34 and 38, combined with their high tolerance, suggests the use of formulation strategies to enable higher doses or combination with repeat radiation schedules could be used to further enhance efficacy.
In summary, we have identified two new nitroimidazole sulfonamide radiosensitisers 34 and 38 with comparable activity to 2, and with potential to be used in combination with SBRT. This discovery complements our earlier identification of 9 as an effective radiosensitiser of HCT116 tumour xenografts. We are mindful that further evaluation, including optimised drug and radiation scheduling, with additional pharmacokinetic, efficacy and toxicity testing in both sexes, will be required to complete our preclinical evaluation of these promising lead compounds.

General Information
The full chemical names of compounds tested in this study are provided in Table S1. DMF, DCM, MeCN and THF were purchased predried and stored over molecular sieves from Acros Organics. All other reaction solvents were analytical grade. Nonaqueous reactions were carried out under a N 2 atmosphere, unless otherwise noted. Commercial reagents were used without purification. Flash column chromatography was carried out on a silica gel solid phase (Merck (Darmstadt, Germany) 230-400 mesh) using distilled laboratory grade solvents. Thin layer chromatography was carried out using Merck 60 F254 aluminium plates, precoated with silica. Compounds were identified using UV fluorescence or iodine on silica gel. Analyses were carried out in the Microchemical Laboratory, University of Otago, Dunedin, New Zealand. Melting points were determined on an Electrothermal 2300 Melting Point Apparatus (ThermoFisher Scientific, Waltham, MA, USA). High-resolution mass spectra (HRMS) were measured on an Agilent Technologies (Santa Clara, CA, USA) 6530 Accurate-Mass Quadrupole Time of Flight (Q-TOF) LC/MS interfaced with an Agilent Jet Stream Electrospray Ionisation (ESI) source, allowing positive or negative ions detection. Low-resolution mass spectra were gathered by direct injection of methanolic solutions into an Agilent 6120 mass spectrometer using atmospheric pressure chemical ionization (APCI) mode with a fragmentor voltage of 50 V and a drying gas temperature of 250 • C. NMR spectra were recorded on a Bruker (Billerica, MA, USA) Avance 400 spectrometer (1H nuclei, 400 MHz; 13C nuclei, 100 MHz) in (CD 3 ) 2 SO unless specified. All chemical shift (δ) values are reported in parts per million (ppm) relative to the residual 1 H resonance from the deuterated solvent; coupling constants are reported in Hertz (Hz). Final products were analysed by reverse-phase HPLC (Agilent Zorbax Eclipse XDB C8 5 µm column, 150 mm × 4.6 mm or Alltech Altima C8 5 µm column, 150 mm × 2.1) using an Agilent HP1100 equipped with a photodiode array detector. Mobile phases were gradients of 80% CH 3 CN/20% H 2 O (v/v) in 45 mM ammonium formate at pH 3.5 and 0.5-1.0 mL/min. Purity was determined by monitoring at 330 ± 50 nm; all key compounds were greater than 95% pure.

Solubility and Stability
The solubility of the final products was determined by HPLC analysis after incubation of saturated solutions of the drug in αMEM containing 5% foetal calf serum (FCS) and 1% DMSO at 37 • C to mimic cell culture conditions. The stability of the same solutions after 24 h incubation at 37 • C was measured by HPLC analysis.

In Vitro Cytotoxicity Testing
HCT116/54C is an HCT116 subline that originated from a mixed culture with UT-SCC-54C cells [60], which was supplied alongside UT-SCC-74B by Prof. Bradly Wouters (University Health Network, Toronto, ON, Canada). HCT116/54C and UT-SCC-74B cells were passaged in minimum essential media (MEM, ThermoFisher Scientific, Waltham, MA, USA) with 10% foetal calf serum (FCS, Moregate Biotech, Hamilton, New Zealand), 4.5 mg/mL D-glucose (Sigma-Aldrich, St. Louis, MO, USA) and 20 mM HEPES (Sigma-Aldrich). FaDu cells were sourced from American Type Culture Collection and cultured in αMEM (ThermoFisher Scientific) with 10% FCS. All cell lines were confirmed mycoplasma-free by PlasmoTest (InvivoGen, San Diego, CA, USA) and were derived from STR-authenticated cryopreserved stocks. Log-phase cells were seeded into 96-well plates at 800 cells/well for HCT116/54C and 1500 cells/well for FaDu. Cells were allowed to attach for 2 h, then, test compounds were added by dilution from DMSO stocks to give a top concentration in <1% DMSO before serial 3-fold dilution in the plates. After 4 h, cultures were washed 3 times with fresh medium and grown for a further 5 days before staining with sulforhodamine B to determine IC 50 values as previously [61]. For hypoxic exposure to compounds, cells were pelleted by centrifugation, transferred to a Pd-catalyst anaerobic chamber (Bactron-II, Shel Lab), resuspended in anoxic medium, and exposed to drugs as above, but using medium and plates that had been equilibrated in the chamber for at least 3 days. After drug washout, cells were grown and stained as for the oxic IC 50 assays. The hypoxic cytotoxicity ratio HCR was determined as HCR = IC 50(oxic) /IC 50(anoxic) .

In Vitro Radiosensitisation Testing
Cells were seeded onto 96-well plates at 10 5 cells per well and allowed to attach for 2 h. Replicates were treated with compound at the respective anoxic IC 50 concentration 30 min prior to irradiation. For anoxic irradiation, plates were transferred to a custom-built, air-tight, stainless steel, portable box (13 × 16.5 × 3 cm) within the anaerobic chamber, then, sealed and transported to the radiation machine. The plates were irradiated (Eldorado 78 60 Co teletherapy unit,~2 Gy/min) with 15 Gy under anoxia at room temperature for SR experiments. A metal wedge placed on the top of the metal chamber was used to achieve a graduated radiation dose across the plate, varying from 7 to 29 Gy, calibrated by Fricke dosimetry with ammonium thiocyanate as previously described [62,63] for SER experiments. The control plate (compound alone, no radiation) was left inside the anaerobic chamber at room temperature during the irradiation period. After treatment, the cells were trypsinized and suspended in usual media + 1% penicillin-streptomycin (ThermoFisher Scientific), and 10-fold serial dilutions were plated for clonogenic survival. After 10 days, plates were stained with methylene blue (2 g/L in EtOH/H 2 O, 1:1 v/v) and colonies with more than 50 cells were counted. The surviving fraction (SF) was determined as: SF = PE (irradiated) /PE (control) where the plating efficiency (PE) = (No. of colonies)/(No. of cells plated). SF was plotted against radiation dose. Survival ratios (SR) were calculated: SR = (cell survival with radiation)/(cell survival with drug + radiation). Sensitizer enhancement ratios (SER) were calculated: SER = (radiation dose for 1% survival without compound)/(radiation dose for 1% survival with compound). Misonidazole (1) at its hypoxic IC 50 (0.5 mM) was used as an intra-experiment control.