This review examines the vast catalytic and therapeutic potential offered by type I (i.e. oxygen-insensitive) nitroreductase enzymes in partnership with nitroaromatic prodrugs, with particular focus on gene-directed enzyme prodrug therapy (GDEPT; a form of cancer gene therapy). Important first indications of this potential were demonstrated over 20 years ago, for the enzyme–prodrug pairing of Escherichia coli NfsB and CB1954 [5-(aziridin-1-yl)-2,4-dinitrobenzamide]. However, it has become apparent that both the enzyme and the prodrug in this prototypical pairing have limitations that have impeded their clinical progression. Recently, substantial advances have been made in the biodiscovery and engineering of superior nitroreductase variants, in particular development of elegant high-throughput screening capabilities to enable optimization of desirable activities via directed evolution. These advances in enzymology have been paralleled by advances in medicinal chemistry, leading to the development of second- and third-generation nitroaromatic prodrugs that offer substantial advantages over CB1954 for nitroreductase GDEPT, including greater dose-potency and enhanced ability of the activated metabolite(s) to exhibit a local bystander effect. In addition to forging substantial progress towards future clinical trials, this research is supporting other fields, most notably the development and improvement of targeted cellular ablation capabilities in small animal models, such as zebrafish, to enable cell-specific physiology or regeneration studies.

INTRODUCTION

One of the most important advances in the treatment of cancer has been the development of cytotoxic chemotherapeutic agents (such as microtubule inhibitors, anti-metabolites, DNA alkylators, DNA intercalators and DNA strand-breakers). Conventional chemotherapy has now become standard of care for the systemic treatment of advanced, recurrent or metastatic disease; and when employed in combination with locoregional treatment modalities, such as surgery and/or radiation, can improve the survival probability for many cancer types [1,2]. The success of cytotoxic agents relies largely upon their differential toxicity for tumour cells which characteristically have a high mitotic rate and increased dependence on a continuous supply of factors required for growth, compared with normal non-cancerous tissue. Typically, however, these agents are not very tumour-specific (i.e. exhibit a low therapeutic index) and their administration generally results in acute damage to non-cancerous normal tissues that also contain a high proportion of rapidly cycling cells [3].

A better understanding of the molecular and genetic alterations involved in oncogenesis [4] has led to a transition towards more targeted chemotherapies. Rationally designed small molecules that inhibit either single or multiple targets have recently been approved for specific cancer types. For example, in July 2013, afatinib was approved by the U.S. Food and Drug Administration (FDA) for first-line treatment of patients with metastatic non-small-cell lung cancer whose tumours have epidermal growth factor receptor (EGFR) exon 19 deletions or exon 21 (L858R) substitution mutations [5]. This success builds on a rich history of discovering specific oncogene driver mutations and the subsequent development of small-molecule inhibitors of their aberrant functions [6]. However, limitations include on-mechanism inhibition of target in normal tissues with attendant dose-limiting side effects [e.g. afatinib causes skin rash/diarrhoea due to inhibition of EGFR in the skin and gastrointestinal (GI) tract] and the relatively rapid emergence of resistance [6].

Indeed, the emergence of drug resistance is a major cause of treatment failure and poor prognosis with both conventional chemotherapy and targeted agents (see [68] for excellent reviews in this area). Resistance to cytotoxic drugs can occur through mechanisms such as enhanced DNA repair, whereas acquired resistance to targeted agents can develop from mutations and/or amplifications in the target molecule or activation of an alternative oncogenic signalling pathway. Moreover, both types of chemotherapy are susceptible to general resistance mechanisms such as drug efflux from the target cell via multidrug-resistance transporters, or the recalcitrance of hypoxic tumour cells that are most distant from blood vessels and are therefore exposed to lower concentrations of drug [68]. In addition, the rate of cell proliferation decreases as a function of distance from the vasculature and most conventional cytotoxic chemotherapies are primarily effective against rapidly dividing cells [9].

In an attempt to address the limitations of current chemotherapies, promising new treatments such as directed enzyme prodrug therapy (DEPT) are being developed. In DEPT strategies, an exogenous prodrug-converting enzyme is selectively delivered to tumour cells, to specifically sensitize them to that prodrug. A key advantage of employing an exogenous enzyme is the potential to use a prodrug substrate that is not recognized by human enzymes and thereby minimize off-target activity while maximizing toxicity within the tumour environment. A primary disadvantage is the difficulty of delivering an exogenous enzyme effectively and selectively to tumours. Two general approaches have been adopted to achieve this. The first of these is direct delivery of the enzyme itself, conjugated to a tumour-targeting agent. This approach has historically employed monoclonal antibodies directed against tumour-specific epitopes [ADEPT (antibody-directed enzyme prodrug therapy)] [10]; however, other direct delivery mechanisms are emerging, such as gold-coated magnetic nanoparticles that might be directed to tumours using an external magnetic field, with localization monitored via MRI [11]. The second approach is indirect delivery of a genetically encoded therapeutic enzyme to the tumour environment [GDEPT (gene-directed enzyme prodrug therapy)], generally mediated by a tumour-tropic bacterial or viral vector [12,13]. Tumour selectivity in GDEPT is generally achieved via a combination of: (1) the selectivity of the delivery of the DNA construct, e.g. to retinoblastoma tumour-suppressor protein-deficient cells via an E1A gene-deleted adenoviral vector [14] or to hypoxic/necrotic tissues via an obligate anaerobic bacterium [15]; and (2) the selectivity of expression of the gene, controlled by the presence of appropriate promoter and/or other control sequences within the construct [13,16]. Although mechanisms of gene delivery are beyond the scope of the present review, it is worth noting that GDEPT is sometimes further categorized according to the choice of vector [VDEPT (viral DEPT) or BDEPT (bacterial DEPT)] to emphasize an important difference that VDEPT results in the tumour cell expressing the delivered gene, whereas gene expression in BDEPT occurs within the bacterium, which may be either internal or adjacent to a tumour cell [17]. For additional information on the tumour-targeting mechanisms of different bacteria and viruses respectively, readers are referred to superb reviews by Forbes [18] and by Cattaneo et al. [19].

Following successful tumour delivery, the therapeutic enzyme is able to convert a systemically delivered non-toxic prodrug into a potent cytotoxin. The localized activation of prodrug in the tumour creates an improved therapeutic index and, in principle, can generate active drug concentrations at levels that are unachievable using non-targeted traditional agents. Of critical importance to most forms of GDEPT, the activated chemotherapeutic should be able to diffuse to and kill closely neighbouring cells, providing a localized ‘bystander effect’ [20,21] (Figure 1). This is obviously essential when using bacterial vectors as they would otherwise be self-sterilizing, but also necessary for viral vectors to achieve significant tumour ablation without having to fulfil the essentially impossible requirement of delivering the prodrug-converting enzyme to every target cell [22]. Indeed, it is generally recognized that the combination of an efficient prodrug-converting enzyme with a prodrug that exerts a powerful bystander effect might go a long way towards mitigating poor gene delivery [20,23].

GDEPT

Figure 1
GDEPT

(A) Gene (purple rectangle) delivery mediated by a tumour-tropic vector leads to (B) tumour-specific expression of the encoded enzyme (purple pacman), which is able to convert a prodrug (red square) into a highly cytotoxic form. (C) Following systemic administration, enzyme-mediated conversion of the prodrug leads to toxic effects in transfected cells. (D) Neighbouring non-transfected cells may also be killed due to a local bystander effect.

Figure 1
GDEPT

(A) Gene (purple rectangle) delivery mediated by a tumour-tropic vector leads to (B) tumour-specific expression of the encoded enzyme (purple pacman), which is able to convert a prodrug (red square) into a highly cytotoxic form. (C) Following systemic administration, enzyme-mediated conversion of the prodrug leads to toxic effects in transfected cells. (D) Neighbouring non-transfected cells may also be killed due to a local bystander effect.

A number of enzyme–prodrug combinations have been investigated for GDEPT (Table 1); however, only the three most clinically advanced combinations are discussed in the present study. Herpes simplex virus thymidine kinase (HSV-tk) in combination with the nucleoside analogue prodrug ganciclovir was the first suicide enzyme–prodrug pair to be suggested [24] and has been the most widely investigated [25]. In this system, ganciclovir is converted by HSV-tk into ganciclovir 5′-monophosphate, which then undergoes further conversion by cellular kinases into the active metabolite ganciclovir triphosphate [26]. Triphosphorylated ganciclovir can cause cell death via the dual mechanism of misincorporation into elongating DNA strands and direct inhibition of DNA polymerase δ. Retroviral delivery of HSV-tk followed by ganciclovir therapy was incorporated in the treatment of glioblastoma multiforme in the first (and to date only) Phase III GDEPT clinical trial [27]. Although patients receiving intra-tumoral gene therapy exhibited no survival benefit over patients receiving standard of care, it was thought that this failure was mainly due to poor gene transfer rather than failure of the enzyme-prodrug combination [27]. However, another contributing factor may have been that the maximum dose of ganciclovir tolerated in humans (10 mg/kg/day) is far below the dose used in most pre-clinical animal experiments (up to 300 mg/kg/day) [28]. Nonetheless, a subsequent study using an alternative gene delivery system observed a significant increase in median survival over standard of care, from 39 to 71 weeks [29]. An additional advantage of the HSV-tk system is that it has been used as a reporter gene for clinical non-invasive imaging of the location of a gene therapy vector in the patient's body [3032] as HSV-tk can phosphorylate (and thereby entrap within a host cell) a number of radiolabelled nucleoside analogues. This then allows nuclear imaging technologies, such as positron emission tomography (PET), to detect the location of active HSV-tk.

Table 1
Other enzyme–prodrug combinations that have been considered for DEPT

Enzyme–prodrug combinations in bold have undergone evaluation in human clinical trials.

Enzyme (origin) Prodrug Cytotoxin Reference(s) 
β-Galactosidase (E. coliDaun02 Daunomycin [219
β-Glucuronidase (human) Glucuronide-linked doxorubicin Doxorubicin [220
Linamarase (plant: cassava) Linamarin Cyanide [221
Carboxylesterase (human) Camptothecin-11 (CPT-11) (Irinotecan) 7-ethyl-10-hydroxy-camptothecin (SN-38) [222
Carboxypeptidase A (human) Methotrexate-α-peptides Methotrexate [223
Carboxypeptidase G2 (Pseudomonas R16) 4-[(2-chloroethyl)(2-mesyloxyethyl)amino]benzoyl-L-glutamic acid (CMDA) (N,N-(2-chloroethyl)-(2-mesyloxyethyl) aminobenzoic acid) (CMBA) [224
 ZD-2767P (N-[4-[Bis(2-iodoethyl)amino]phenoxycarbonyl]-L-glutamic acid) Phenol-bis-iodo nitrogen mustard [225,226
Horseradish peroxidase (plant: horseradish) Acetaminophen NABQI (N-acetyl benzoquinone imine) [227
 Horseradish indole-3-acetic acid Free radicals [228
Purine nucleoside phosphorylase (E. coli6-Methylpurine-2-deoxyriboside 6-Methylpurine [229
Purine nucleoside phosphorylase (human) 5′-deoxy-5-fluorouridine (5′-DFUR) 5-fluorouracil (5-FU) [230
Thymidine phosphorylase (human) Pyrimidine analogues (e.g. 5′-DFUR) 5-Fluoro deoxyuridine monophosphate [231
Cytosine deaminase (E. coli, S. cerevisiae) 5-FC 5-FU [36,232
NADPH–cytochrome P450 reductase (human) Tirapazamine, EO9, (1-(2-nitro-1-imidazolyl)-3-(1-aziridinyl)-2-propanol) (RSU1069), misonidazole Reduced metabolites [233
Cytochrome P450 enzymes    
e.g. CYP2B6 (human) Cyclophosphamide Phosphoramide mustard and acrolein [234,235
e.g. CYP2B1 (rat) Ifosfamide Phosphoramide mustard and acrolein [236
e.g. CYP3A4 (human) AQ4N AQ4 [237
Thymidine kinase (varicella zoster virus) 6-Methoxypurine arabinoside Adenine arabinoside triphosphate [238
Thymidine kinase (herpes simplex virus) Ganciclovir Ganciclovir triphosphate [27,29
Enzyme (origin) Prodrug Cytotoxin Reference(s) 
β-Galactosidase (E. coliDaun02 Daunomycin [219
β-Glucuronidase (human) Glucuronide-linked doxorubicin Doxorubicin [220
Linamarase (plant: cassava) Linamarin Cyanide [221
Carboxylesterase (human) Camptothecin-11 (CPT-11) (Irinotecan) 7-ethyl-10-hydroxy-camptothecin (SN-38) [222
Carboxypeptidase A (human) Methotrexate-α-peptides Methotrexate [223
Carboxypeptidase G2 (Pseudomonas R16) 4-[(2-chloroethyl)(2-mesyloxyethyl)amino]benzoyl-L-glutamic acid (CMDA) (N,N-(2-chloroethyl)-(2-mesyloxyethyl) aminobenzoic acid) (CMBA) [224
 ZD-2767P (N-[4-[Bis(2-iodoethyl)amino]phenoxycarbonyl]-L-glutamic acid) Phenol-bis-iodo nitrogen mustard [225,226
Horseradish peroxidase (plant: horseradish) Acetaminophen NABQI (N-acetyl benzoquinone imine) [227
 Horseradish indole-3-acetic acid Free radicals [228
Purine nucleoside phosphorylase (E. coli6-Methylpurine-2-deoxyriboside 6-Methylpurine [229
Purine nucleoside phosphorylase (human) 5′-deoxy-5-fluorouridine (5′-DFUR) 5-fluorouracil (5-FU) [230
Thymidine phosphorylase (human) Pyrimidine analogues (e.g. 5′-DFUR) 5-Fluoro deoxyuridine monophosphate [231
Cytosine deaminase (E. coli, S. cerevisiae) 5-FC 5-FU [36,232
NADPH–cytochrome P450 reductase (human) Tirapazamine, EO9, (1-(2-nitro-1-imidazolyl)-3-(1-aziridinyl)-2-propanol) (RSU1069), misonidazole Reduced metabolites [233
Cytochrome P450 enzymes    
e.g. CYP2B6 (human) Cyclophosphamide Phosphoramide mustard and acrolein [234,235
e.g. CYP2B1 (rat) Ifosfamide Phosphoramide mustard and acrolein [236
e.g. CYP3A4 (human) AQ4N AQ4 [237
Thymidine kinase (varicella zoster virus) 6-Methoxypurine arabinoside Adenine arabinoside triphosphate [238
Thymidine kinase (herpes simplex virus) Ganciclovir Ganciclovir triphosphate [27,29

The second enzyme–prodrug combination to enter GDEPT clinical trials was cytosine deaminase in partnership with 5-fluorocytosine, the latter able to be converted by the former into the nucleoside analogue 5-fluorouracil [33]. 5-Fluorouracil is subsequently transformed by cellular enzymes into potent pyrimidine anti-metabolites that lead to growth inhibition and apoptosis-mediated cell death [34]. The first clinical trial was performed on breast cancer patients and involved direct intra-tumoral injection of a plasmid construct containing an Escherichia coli cytosine deaminase gene under transcriptional control of a tumour-specific erbB-2 promoter. The approach was shown to be safe and selective to erbB-2-positive tumour cells [35]. A small pilot study in refractory cancer patients subsequently used an attenuated strain of Salmonella enterica serovar Typhimurium as a gene delivery vector and demonstrated feasibility in two patients [36]. Although there appear to have been no further trials of this system in isolation, a yeast cytosine deaminase has advanced to Phase II trial as part of a multi-modal GDEPT therapy to treat prostate cancer. This complex therapy uses a replication-competent adenovirus to deliver an engineered mutant version of HSV-tk, an yeast-derived cytosine deaminase and a viral protein that enhances oncolytic activity, followed by administration of 5-fluorocytosine, valganciclovir (a pre-prodrug of ganciclovir) and radiotherapy [37,38].

Both the HSV-tk and the cytosine deaminase GDEPT systems have the advantage that their primary prodrugs are independently approved clinical agents, smoothing their developmental pathway. These systems are also more clinically advanced than GDEPT using bacterial nitroreductase enzymes, which is the other main GDEPT modality to have been trialled in humans to date. Nevertheless, nitroreductase GDEPT possesses some distinct advantages over the HSV-tk and the cytosine deaminase systems. For example, as the activated metabolites of ganciclovir and 5- fluorocytosine are nucleotide analogues, they are only effective at killing actively dividing cells; in contrast, nitroreductase enzymes can convert nitroaromatic prodrugs either directly or indirectly into DNA-damaging metabolites that are cytotoxic to both quiescent and actively dividing tumour cells [39]. The nucleotide analogue drugs are also substantially more anti viral and hence likely to prove counter-productive to GDEPT that uses a replicating viral vector [40], and, whereas many nitroreductase-activated drugs can freely diffuse between cell layers [41,42], triphosphorylated ganciclovir is not freely cell-permeant, making its bystander effect primarily dependent on cell–cell contacts such as gap junctions [43], which can be down-regulated in the dysfunctional tumour environment [44].

Perhaps most importantly, there is a large, and to date somewhat untapped, versatility in the choice of nitroreductase and nitroaromatic prodrug combination to provide maximal clinical benefit. Although a very large number of bacterial nitroreductase enzymes exist in Nature [45], the majority of GDEPT studies to date have focused on just a single enzyme, NfsB from E. coli, to the extent that it is often referred to simply as NTR (an abbreviation of nitroreductase), as though it is the singular bacterial nitroreductase enzyme. More recent research to identify additional nitroreductases that offer advantageous properties for GDEPT is summarized in the present review, as are efforts to use targeted enzyme engineering and high-throughput-directed evolution to improve these properties. Accompanying the singular focus on E. coli NfsB has been an emphasis on the use of a single nitroaromatic prodrug, 5-(aziridin-1-yl)-2,4-dinitrobenzamide (CB1954). In fact CB1954 is only one of the many possible nitroaromatic prodrugs with potential utility in GDEPT, some of which have undergone clinical development as hypoxia-targeting chemotherapeutics or have even achieved clinical approval as antibiotics. Owing to the characteristically broad substrate promiscuity of nitroreductases, which is discussed in more detail in the following section, there is potential to repurpose these independently evaluated prodrugs for GDEPT. Alternatively, this promiscuity can be exploited to give medicinal chemists the flexibility and scope to engineer completely new prodrugs tailored specifically for the demanding requirements of GDEPT.

NITROREDUCTASE ENZYMES

Nitroreductases are promiscuous flavin-associated oxidoreductase enzymes, defined by their ability to reduce nitro substituents on aromatic rings. These enzymes divide into two functional classes. Type I nitroreductases are oxygen-insensitive, i.e. can generate nitroso, hydroxylamine and/or amine end-products in the presence of molecular oxygen, whereas type II nitroreductases are oxygen-sensitive and can only generate these products when oxygen is absent [46]. The mechanistic basis of this distinction is that the type II enzymes reduce nitro groups via successive single-electron transfers, which yield nitro radical anion intermediates that are rapidly back-oxidized by molecular oxygen to their original forms, with the concomitant production of superoxide anion in a futile redox cycle [47]. In contrast, type I nitroreductases proceed via concerted two-electron transfers, which allow them to achieve their reduced end-product(s) independent of the oxygen status of their environment (Figure 2).

Schematic of type II (oxygen-dependent) and type I (oxygen-independent) reduction of a nitroaromatic substrate, proceeding via either consecutive 1e or concerted 2e steps

Figure 2
Schematic of type II (oxygen-dependent) and type I (oxygen-independent) reduction of a nitroaromatic substrate, proceeding via either consecutive 1e or concerted 2e steps

If the first reduction is carried out by a type II (oxygen-sensitive, one-electron) reductase [e.g. cytochrome P450 reductase (CYPOR)] [250] the hydroxylamine or amine end-products can only be achieved in the absence of oxygen. In the presence of oxygen, the nitro anion radical generated by a type II reductase is back-oxidized, regenerating the parental compound in a futile redox cycle. In contrast, if all reduction is carried out in 2e increments, as catalysed by a type I nitroreductase (e.g. E. coli NfsB), then reduction proceeds irrespective of the oxygenation status of the environment. Reduction to a hydroxylamine or amine end-product results in a profound change in electronic effect as the nitro group is strongly electron-withdrawing, whereas the reduced hydroxylamine and amine are electron-donating.

Figure 2
Schematic of type II (oxygen-dependent) and type I (oxygen-independent) reduction of a nitroaromatic substrate, proceeding via either consecutive 1e or concerted 2e steps

If the first reduction is carried out by a type II (oxygen-sensitive, one-electron) reductase [e.g. cytochrome P450 reductase (CYPOR)] [250] the hydroxylamine or amine end-products can only be achieved in the absence of oxygen. In the presence of oxygen, the nitro anion radical generated by a type II reductase is back-oxidized, regenerating the parental compound in a futile redox cycle. In contrast, if all reduction is carried out in 2e increments, as catalysed by a type I nitroreductase (e.g. E. coli NfsB), then reduction proceeds irrespective of the oxygenation status of the environment. Reduction to a hydroxylamine or amine end-product results in a profound change in electronic effect as the nitro group is strongly electron-withdrawing, whereas the reduced hydroxylamine and amine are electron-donating.

It is the type I nitroreductases, predominantly from bacteria, that are of particular interest to GDEPT owing to their non-human origins and their ability to catalyse the bioreductive activation of nitroaromatic prodrugs in both oxygenated and hypoxic tumour tissues. Addition of one or more strongly electron-withdrawing nitro substituents to an aromatic ring is a strategy that has been exploited widely in prodrug design, as the nitro group(s) can pull electron density away from other highly reactive substituents and thereby render them relatively inert [48]. Four- or six-electron reduction of the nitro group to a hydroxylamine (-NHOH) or an amine (-NH2) is one of the largest possible changes in electronic effect that can be achieved by a single-enzyme-catalysed reaction [39]. The transition effectively flicks an electronic switch that pushes electrons back into the ring, unleashing the reactive substituents (Figure 2).

Bacterial type I nitroreductases typically adopt a homodimeric quaternary structure and associate with flavins at a ratio of one FMN (or, less commonly, FAD) prosthetic group per monomer [4956]. Catalysis then proceeds via a Ping Pong Bi Bi kinetic mechanism [45], i.e. NAD(P)H enters each active site, two electrons are passed to the flavin, the oxidized nicotinic cofactor exits, a second substrate enters the same active site to accept the electron pair and finally the reduced product dissociates. With nitroaromatic substrates, the entire cycle may be repeated once to form the hydroxylamine or twice to produce the amine (Figure 2). Throughout all of this, the semiquinone states of the flavin are destabilized, rendering one-electron chemistry thermodynamically unfavourable [57,58].

The name nitroreductase is somewhat misleading, as it provides a functional definition rather than a likely physiological role. Nitroaromatic compounds are relatively rare in Nature [59] and it is generally accepted that neither these nor the more widespread xenobiotic nitro-compounds generated by human activities [e.g. 2,4,6-trinitrotoluene (TNT)] are likely to be primary substrates for these enzymes [45]. Rather, type I nitroreductases in bioluminescent bacteria have been shown to act as flavin reductases, providing reduced FMNH2 to power the bioluminescence reaction [60], whereas in other bacteria, it has been proposed that some of these enzymes act as quinone reductases, maintaining a pool of reduced quinones that contribute to antioxidant defence [61,62]. Certainly, identification of a primary physiological role for many of the type I nitroreductases have been made more difficult by their high levels of substrate promiscuity. In addition to nitro-heterocyclic molecules, flavins and quinones, enzymes with type I nitroreductase activity have also been found to reduce hexavalent chromium [63,64], iron [65] and azo dyes [66]. In some enzymes this substrate promiscuity may be promoted by highly plastic active sites that can accommodate substrates of various sizes [57] and it has also been noted that nitroreduction is not dependent on any substrate-specific enzyme conformational changes that might otherwise impose constraints [67]. Thus, irrespective of what their primary physiological role may be, the type I nitroreductases offer great promise as versatile enzymatic tools for the bioreductive activation of synthetic prodrug substrates.

CB1954 AND E. COLI NfsB: THE PROTOTYPICAL NITROAROMATIC PRODRUG–ENZYME PAIRING FOR GDEPT

As noted earlier, the most widely-studied nitroaromatic prodrug for GDEPT is CB1954 (Figures 3 and 4A). The letters in its name derive from the Chester Beatty Laboratories in London where CB1954 was first synthesized, during the late 1960s, as part of a series of compounds tested for activity against the transplantable rat Walker 256 carcinoma [68]. CB1954 showed startling single-agent activity in this model, leading to complete cures with only minimal toxicity [69]. Unsurprisingly, this generated a great deal of interest in elucidating the mechanism of action and the efforts to achieve this are detailed in an excellent review by Knox et al. [70]. Ultimately these efforts led to the discovery that CB1954 was acting as a prodrug, being converted into its 4-hydroxylamine derivative by rat NQO1 [NAD(P)H quinone oxidoreductase 1, a type I nitroreductase also known as DT diaphorase] [71]. Although the difference in DNA-alkylating capability of the aziridine group between the parental and the 4-hydroxylamine forms of CB1954 is modest, the 4-hydroxylamine is able to undergo a spontaneous reaction with acetyl-CoA or other cellular thioesters, yielding a bi-functional alkylating agent capable of cross-linking DNA [72] (Figure 4A). Not only is NQO1 generally expressed at high levels in solid tumours [73], Walker 256 tumours are ineffective at repairing DNA cross-links, which made them particularly susceptible to CB1954 [74].

Structures of potential prodrugs for nitroreductase GDEPT

Mechanism of activation and genotoxicity of metabolites of (A) CB1954 and (B) PR-104A

Disappointingly, equivalent effects were not observed in human tumours. Whereas other rat cell lines were also observed to be sensitive to CB1954, albeit generally to a lesser extent than Walker 256 cells, CB1954 as a sole agent was 500–5000-fold less toxic to human cell lines, even those expressing high levels of NQO1 [75]. Later biochemical studies revealed that purified recombinant rat NQO1 was over an order of magnitude more efficient (in terms of kcat/KM) than the human enzyme at reducing CB1954 [76] and that this discrepancy could largely be attributed to a single amino acid difference between the two enzymes (residue 104, tyrosine in rat NQO1 and glutamine in human NQO1 [77]).

Despite this disappointing finding, excitement around CB1954 persisted, with directed enzyme–prodrug strategies seen as a potential means of recapitulating a CB1954-sensitive phenotype in human cells [70]. Fuelling this excitement was a report that a type I nitroreductase, NfsB from E. coli, was nearly 90-fold more efficient at reducing CB1954 (kcat/KM) than even the rat form of NQO1 [78]. In contrast with NQO1, E. coli NfsB generates both the 2- and the 4-hydroxylamine reduction products of CB1954, in an approximately equimolar ratio, although it appears unable to reduce both nitro groups on the same molecule [79]. Unlike the 4-hydroxylamine, the 2-hydroxylamine derivative of CB1954 remains a monofunctional alkylating agent (i.e. is unable to go on to cross-link DNA; Figure 4A) and consequently for a time was deemed an undesirable by-product [80]. However, it was later demonstrated that the 2-hydroxylamine disproportionates to yield the 2-amine (as well as the 2-nitroso; Figure 4A) and that the 2-amine not only has comparable potency to that of the 4-hydroxylamine against most human tumour cell lines, but also exhibits a far superior bystander effect [81]. Moreover, the acetylation of the 4-hydroxyamine of CB1954 is now known to be catalysed by arylamine N-acetyltransferase 2 (NAT2) which, although important for yielding a cytotoxic metabolite, constrains the bystander effect [82]. Thus, generation of the 2-nitro-reduced derivatives of CB1954 is in fact likely to be highly desirable for DEPT strategies, owing to the innate dependence of these strategies on bystander cell killing.

The combination of E. coli NfsB and CB1954 was initially considered for use in ADEPT; however, the intracellular localization of endogenous NAD(P)H cofactors would necessitate that an externally delivered nitroreductase be supplied with exogenous cofactor analogues that are non-labile in serum. Although several potentially suitable cofactor analogues were identified [83], the complexity added by this requirement would probably have rendered human clinical trials non-viable. In contrast, the system showed great promise in a wide range of pre-clinical GDEPT models, with both viral [8488] and bacterial [89] delivery of E. coli nfsB to murine tumour xenografts resulting in increased median survival times and/or significant reductions in tumour volume upon CB1954 administration, relative to prodrug-free controls. Importantly, significant reduction in tumour size was reported upon treatment with CB1954 in mixed xenograft models where only a small percentage (5–20%) of cells were expressing NfsB, proving the in vivo utility of local bystander-mediated cell killing [9092]. In an immunocompetent mouse model, there was also evidence of a distant bystander effect, i.e. tumour cells killed by activated CB1954 were promoting an immune response against the tumour [93].

It is critical to note, however, that all of these murine studies employed doses of CB1954 far in excess of the human equivalent maximum tolerated dose (MTD), which is approximately 20-fold lower when corrected for body surface area scaling [94]. Most telling are the plasma concentration area under the curve (AUC) measurements; after a typical dose of 50 mg/kg CB1954 the plasma AUC in mice was 320 μM·h [95], some 50-fold greater than the AUC recorded in humans [96,97].

The relatively low tolerance of human patients to CB1954 first became apparent in a Phase I clinical trial in patients having various malignancies (predominantly GI), where dose-limiting hepatic toxicity and diarrhoea were observed at 37.5 mg/m2 intravenous (i.v.) during the dose escalation phase of the study [97]. This dose approximates to 2 mg/kg, uncorrected for potential differences in plasma protein binding between species, but nevertheless far below the 50 or 80 mg/kg that were used in the majority of mouse studies. The basis of CB1954 toxicity was subsequently examined in human liver preparations and concluded to be due to the activity of endogenous type I and II nitroreductases, independent of NQO1 [98], with some evidence for the involvement of nitric oxide synthases in this activity [99]. It has also been suggested that high NAT2 levels may be associated with GI tract and hepatotoxicity in humans [82].

The Phase I trial researchers identified 24 mg/m2 CB1954 as an appropriate and well-tolerated i.v. dose for patients; at this dose, the mean peak serum concentration achieved was 6.3 μM and the mean AUC was 5.8 μM·h [97]. Following confirmation of safety and effective transgene delivery by CTL102, an E. coli nfsB expressing replication-defective adenovirus, in patients with resectable liver cancer [100], the combination of CTL102 and 24 mg/m2 CB1954 was evaluated in a Phase I/II trial in patients with localized prostate cancer [96]. Levels of blood CB1954 were achieved comparable with those reported in the previous Phase I trial (mean peak serum concentration of 7.8 μM, mean plasma AUC of 6.3 μM·h) and no heightened toxicities due to the combined treatments were observed. However, indications of response to the therapy, as judged by a fall in the levels of prostate-specific antigen (PSA), were modest; only seven of 19 patients showed >10% reduction in PSA and the greatest reduction was only 72% [96]. Although a follow-up study further identified an increased frequency of T-cells recognizing PSA in three of 11 patients examined, indicating possible stimulation of an immune response against the tumour [101], the overall results do not appear to have warranted further trials of this particular GDEPT combination.

It seems highly likely that a major factor limiting the success of the human trial was the disparity between the achievable serum concentrations of CB1954 and the KM of E. coli NfsB for this substrate, which has been variously reported as ranging from 79 μM to 17 mM (Table 2). The former value is an especially low outlier and the experience of our team and most others has been that the serum concentrations of CB1954 achievable in patients are well under 1% of the KM of E. coli NfsB (with achievable tumour levels, based on measurements in mice, likely to be lower still) [95]. Thus, the NfsB-catalysed conversion of CB1954 into activated metabolites in patients is likely to be very inefficient. This realization has spurred a search for alternative nitroreductase enzymes that are more effective at converting CB1954 at low prodrug concentrations.

Table 2
Reported activities of isolated bacterial type I nitroreductases with CB1954
Nitroreductase (with GenBank accession numbers) Family1 KM23(μM) kcat23(s−1kcat/KM3(M−1·s−1Cofactor used (and concentration thereof) Product4 Reference Reference(s) for solved crystal structure and PDB code(s)5 
NfsB (E. coli) [WP_000351487] NfsB 860 6 7000 NADH (500 μM) 2HX:4HX (1:1) [78[5053] 1DS7, 1IDT, 1YKI, 1YLR, 1YLU, 1ICR, 1ICU, 1ICV 
  880 6800 NADH (60 μM)  [239 
  680 8 8800 NADH (500 μM)  [103 
  900 6.2 7000 NADH (60 μM)  [104 
  17000 140 7000 NADH (0-1000 μM)  [105 
  8000 50 6000 NADH (50 μM)  [109 
  2600 15 5800 NADPH (50 μM)  [109 
  11000 62 5600 NADH (250 μM)  [108 
  3600 26 7300 NADPH (250 μM)  [108 
  79 26000 NADPH (60 μM)  [240 
FRaseI (A. fischeri) [P46072] NfsB 890 11 12000 NADPH (250 μM) – [129[241] 1VFR 
Hso (Haemophilus somnus) [WP_011608931] NfsB 13 22 600000 NADPH (500 μM) 4HX [106 
  3.5 22 160000 NADH (500 μM)  [106 
NfnB (Campylobacter jejuni) [WP_002869471.1] NfsB 220 6.1 28000 NADPH (500 μM) 4HX [103 
NfnB (H. influenzae) [WP_011272690] NfsB 690 56 81000 NADPH (500 μM) 4HX [103 
NfsB (Citrobacter koseri) [WP_047458272] NfsB 12000 73 4900 NADH (250 μM) [110 
NfsB (Klebsiella pneumoniae) [WP_004178896] NfsB 21000 200 9600 NADH (250 μM) 2HX:4HX (1:1) [110 
NfsB (V. vulnificus) [WP_017421465] NfsB 1300 73 58000 NADH (250 μM) 2HX:4HX (1:3) [110 
Nme (N. meningitidis) [Q9K022] NfsB 2.5 15 6.2 x 106 NADPH (500 μM) 4HX [106 
  2.5 4 1.6 x 106 NADH (500 μM)  [106 
RdxA (Helicobacter pylori) [WP_000670110] NfsB 35 10 300000 NADPH (90 μM) – [240[242] 3QDL 
YdgI (B. subtilis) [WP_003225379] NfsB 1800 7.7 4300 NADH (250 μM) 2HX:4HX (3:2) [110 
YfkO (Bacillus licheniformis) [AAU39748.1] NfsB 30 1100 3.5 x 107 NADPH (500 μM) 4HX [243 
YfkO (B. subtilis) [WP_003243096] NfsB 2400 60 25000 NADPH (250 μM) 4HX [110 
NfsA (E. coli) [WP_000189159] NfsA 140 21 150000 NADPH (50 μM) 2HX [109 
  18 2.6 150000 NADH (50 μM)  [109[49] 1F5V 
  220 16 73000 NADPH (250 μM) 2HX >> 4HX [108 
  46 2.9 63000 NADH (250 μM)  [108 
Frp (Vibrio harveyi) [WP_038898257] NfsA 100 22 220000 NADPH (250 μM) 2HX [110[244,245] 1BKJ, 2BKJ 
NfrA (B. subtilis) [WP_003222161] NfsA 150 14 97000 NADPH (250 μM) 2HX [110[246] 3N2S 
NfsA (A. fischeri) [WP_012535301] NfsA 53 14 260000 NADPH (250 μM) – [110 
NfsA (C. koseri) [WP_012133106] NfsA 160 25 150000 NADPH (250 μM) – [110 
NfsA (Cronobacter sakazakii) [WP_007848889] NfsA 260 26 99000 NADPH (250 μM) – [110 
NfsA (K. pneumoniae) [WP_012068493] NfsA 1700 37 22000 NADPH (250 μM) 2HX [110 
NfsA (Salmonella enterica) [NP_455401] NfsA 21 0.4 18000 NADPH (250 μM) 2HX [110 
NfsA (V. vulnificus) [WP_011082270] NfsA 210 45 220000 NADPH (250 μM) 2HX [110 
YcnD (B. subtilis) [CAB12194] NfsA 35 15 430000 NADPH (250 μM) – [110[247] 1ZCH 
YwrO (B. amyloliquefaciens) [AAL66420] YwrO 620 8.2 13000 NADPH (500 μM) 4HX [102,103 
YwrO (Porphyromonas gingivalis) [WP_021677636.1] YwrO 1200 3.2 2700 NADPH (500 μM) 4HX [103 
AzoR (E. coli) [WP_000048950] AzoR 1400 0.15 110 NADPH (250 μM) 4HX [108[54] 2D5I 
  6600 0.15 23 NADH (250 μM)  [108 
NemA (E. coli) [WP_000093589] NemA 56 0.22 3900 NADPH (250 μM) 4HX [108 
  55 0.048 880 NADH (250 μM)  [108 
NbzA (Pseudomonas pseudoalcaligenes) [AAT71308.1] NbzA 12 – – NADPH (250 μM) – [248 
NtrB (Staphylococcus saprophyticus) [BAE17353.1] NtrB 1100 2.3 2100 NADPH (200 μM) 2HX:4HX (1:1) [249 
YdgI (B. subtilis) [AAB72053] YdgI 3900 30 7800 NADPH (500 μM) 2HX < 4HX [103 
YodC (B. subtilis) [BAA19399] YodC 550 58 11000 NADPH (500 μM) 2HX < 4HX [103 
Nitroreductase (with GenBank accession numbers) Family1 KM23(μM) kcat23(s−1kcat/KM3(M−1·s−1Cofactor used (and concentration thereof) Product4 Reference Reference(s) for solved crystal structure and PDB code(s)5 
NfsB (E. coli) [WP_000351487] NfsB 860 6 7000 NADH (500 μM) 2HX:4HX (1:1) [78[5053] 1DS7, 1IDT, 1YKI, 1YLR, 1YLU, 1ICR, 1ICU, 1ICV 
  880 6800 NADH (60 μM)  [239 
  680 8 8800 NADH (500 μM)  [103 
  900 6.2 7000 NADH (60 μM)  [104 
  17000 140 7000 NADH (0-1000 μM)  [105 
  8000 50 6000 NADH (50 μM)  [109 
  2600 15 5800 NADPH (50 μM)  [109 
  11000 62 5600 NADH (250 μM)  [108 
  3600 26 7300 NADPH (250 μM)  [108 
  79 26000 NADPH (60 μM)  [240 
FRaseI (A. fischeri) [P46072] NfsB 890 11 12000 NADPH (250 μM) – [129[241] 1VFR 
Hso (Haemophilus somnus) [WP_011608931] NfsB 13 22 600000 NADPH (500 μM) 4HX [106 
  3.5 22 160000 NADH (500 μM)  [106 
NfnB (Campylobacter jejuni) [WP_002869471.1] NfsB 220 6.1 28000 NADPH (500 μM) 4HX [103 
NfnB (H. influenzae) [WP_011272690] NfsB 690 56 81000 NADPH (500 μM) 4HX [103 
NfsB (Citrobacter koseri) [WP_047458272] NfsB 12000 73 4900 NADH (250 μM) [110 
NfsB (Klebsiella pneumoniae) [WP_004178896] NfsB 21000 200 9600 NADH (250 μM) 2HX:4HX (1:1) [110 
NfsB (V. vulnificus) [WP_017421465] NfsB 1300 73 58000 NADH (250 μM) 2HX:4HX (1:3) [110 
Nme (N. meningitidis) [Q9K022] NfsB 2.5 15 6.2 x 106 NADPH (500 μM) 4HX [106 
  2.5 4 1.6 x 106 NADH (500 μM)  [106 
RdxA (Helicobacter pylori) [WP_000670110] NfsB 35 10 300000 NADPH (90 μM) – [240[242] 3QDL 
YdgI (B. subtilis) [WP_003225379] NfsB 1800 7.7 4300 NADH (250 μM) 2HX:4HX (3:2) [110 
YfkO (Bacillus licheniformis) [AAU39748.1] NfsB 30 1100 3.5 x 107 NADPH (500 μM) 4HX [243 
YfkO (B. subtilis) [WP_003243096] NfsB 2400 60 25000 NADPH (250 μM) 4HX [110 
NfsA (E. coli) [WP_000189159] NfsA 140 21 150000 NADPH (50 μM) 2HX [109 
  18 2.6 150000 NADH (50 μM)  [109[49] 1F5V 
  220 16 73000 NADPH (250 μM) 2HX >> 4HX [108 
  46 2.9 63000 NADH (250 μM)  [108 
Frp (Vibrio harveyi) [WP_038898257] NfsA 100 22 220000 NADPH (250 μM) 2HX [110[244,245] 1BKJ, 2BKJ 
NfrA (B. subtilis) [WP_003222161] NfsA 150 14 97000 NADPH (250 μM) 2HX [110[246] 3N2S 
NfsA (A. fischeri) [WP_012535301] NfsA 53 14 260000 NADPH (250 μM) – [110 
NfsA (C. koseri) [WP_012133106] NfsA 160 25 150000 NADPH (250 μM) – [110 
NfsA (Cronobacter sakazakii) [WP_007848889] NfsA 260 26 99000 NADPH (250 μM) – [110 
NfsA (K. pneumoniae) [WP_012068493] NfsA 1700 37 22000 NADPH (250 μM) 2HX [110 
NfsA (Salmonella enterica) [NP_455401] NfsA 21 0.4 18000 NADPH (250 μM) 2HX [110 
NfsA (V. vulnificus) [WP_011082270] NfsA 210 45 220000 NADPH (250 μM) 2HX [110 
YcnD (B. subtilis) [CAB12194] NfsA 35 15 430000 NADPH (250 μM) – [110[247] 1ZCH 
YwrO (B. amyloliquefaciens) [AAL66420] YwrO 620 8.2 13000 NADPH (500 μM) 4HX [102,103 
YwrO (Porphyromonas gingivalis) [WP_021677636.1] YwrO 1200 3.2 2700 NADPH (500 μM) 4HX [103 
AzoR (E. coli) [WP_000048950] AzoR 1400 0.15 110 NADPH (250 μM) 4HX [108[54] 2D5I 
  6600 0.15 23 NADH (250 μM)  [108 
NemA (E. coli) [WP_000093589] NemA 56 0.22 3900 NADPH (250 μM) 4HX [108 
  55 0.048 880 NADH (250 μM)  [108 
NbzA (Pseudomonas pseudoalcaligenes) [AAT71308.1] NbzA 12 – – NADPH (250 μM) – [248 
NtrB (Staphylococcus saprophyticus) [BAE17353.1] NtrB 1100 2.3 2100 NADPH (200 μM) 2HX:4HX (1:1) [249 
YdgI (B. subtilis) [AAB72053] YdgI 3900 30 7800 NADPH (500 μM) 2HX < 4HX [103 
YodC (B. subtilis) [BAA19399] YodC 550 58 11000 NADPH (500 μM) 2HX < 4HX [103 

*Defined as the closest E. coli orthologue sharing >25% amino acid identity, where one exists [110].

†With the exception of the kinetic parameters reported in [105], these are apparent KM and kcat values, as measured at the NAD(P)H cofactor concentration employed in each study.

‡Where written in italics, kinetic parameters were derived using a discontinuous HPLC methodology; otherwise a continuous spectrophotometry approach was employed.

§Approximate ratio of 2-NO2 and 4-NO2 reduction products, where measured.

║Where available.

DISCOVERY OF ALTERNATIVE CB1954-ACTIVATING NITROREDUCTASES

The first bacterial nitroreductase other than E. coli NfsB to be investigated for GDEPT potential was YwrO from Bacillus amyloliquefaciens [102]. Relative to E. coli NfsB, B. amyloliquefaciens YwrO exhibited slight improvements in both kcat and KM with CB1954 as a substrate, contributing to an approximately 2-fold increase in kinetic efficiency (kcat/KM, as measured by the same team[78]; Table 2). Despite this, when V79 Chinese hamster ovary cells were co-incubated in culture medium with purified recombinant enzyme, CB1954 and NAD(P)H cofactor, the concentration of CB1954 required to achieve 50% cell killing was >20-fold lower for NfsB than YwrO [102]. Other differences noted between the two enzymes were that, whereas NfsB does not exhibit a strong cofactor preference and reduces both the 2- and the 4-nitro groups of CB1954, YwrO was substantially more active with NADPH than NADH and reduced CB1954 exclusively at the 4-nitro position (Table 2). It is tempting to speculate that the latter characteristic may have contributed in some manner to the surprisingly poor relative ability of YwrO to kill V79 cells in vitro, either due to the heightened reactivity of the 4-hydroxylamine causing it to be quenched by other components of the culture medium or other unknown variables such as the possibility of low NAT2 levels in V79 cells.

In a later study that used the same assay to evaluate several homologues of E. coli NfsB and B. amyloliquefaciens YwrO from other bacterial species [103], the exclusively 4-nitro-reducing enzymes again underperformed relative to E. coli NfsB, despite generally being superior in terms of kcat/KM. For example, the only nitroreductase found to achieve 50% cell killing at even a marginally lower CB1954 concentration (4.7 μM, compared with 6.3 μM for E. coli NfsB) was Haemophilus influenzae NfsB, which had nearly a 10-fold greater kcat/KM with the prodrug (Table 2). In contrast, H. influenzae NfsB appeared highly effective in vivo, generating significant and prolonged anti-tumour effects following repeated cycles of CB1954 GDEPT in a mouse xenograft model, using a Clostridium sporogenes vector [a sub-category of BDEPT sometimes referred to as CDEPT (Clostridium DEPT)] [103].

This work was an important first demonstration of the potential for testing homologues of proven CB1954 reductases to recover more active nitroreductase variants from Nature [103]. This approach has since been adopted by several teams including our own, collectively revealing a large number of nitroreductases that possess superior Michaelis–Menten kinetic parameters for CB1954 reduction relative to those of E. coli NfsB (Table 2). When comparing the activity of different nitroreductases evaluated in different studies, however, it is essential to note that not all kinetic measurements have been obtained using the same assay system. We personally favour the continuous spectrophotometric method optimized by Hyde and colleagues [104] at the University of Birmingham, which is user-friendly and robust, monitoring appearance of the 2- and 4-hydroxylamine end-products at 420 nm, a wavelength at which both reduction products have the same molar absorbance. [This is superior to following NAD(P)H consumption at 340 nm, as in our experience many nitroreductases exhibit low-level NAD(P)H oxidase activity; so, unless assays are run under anoxic conditions, there will be some ‘leakage’ of electrons to molecular oxygen.] The other main assay system described in published literature to date has been a discontinuous HPLC method and in our experience this approach leads to substantially (10–100-fold) lower estimates of KM relative to the spectrophotometric methods. A similar observation was made previously by the Hyde team and attributed the difference as being due to substrate depletion at the high enzyme concentrations and long times used in the discontinuous HPLC assays [105].

It is also important to note that because nitroreductases are dual-substrate oxidoreductases that exhibit a Ping Pong kinetic mechanism, the rate of CB1954 reduction will vary at different NAD(P)H concentrations and hence the data listed in Table 2 are in fact apparent kinetic parameters for CB1954 reduction at the specific NAD(P)H concentration that was used in each study (this should not affect the relative kcat/KM ratio, however). Given that the complexities of GDEPT extend far beyond a simple consideration of enzymatic rates and substrate affinities in a well-defined assay mixture, the discrepancies between the different methodologies that have been employed need not be a major concern provided enzymes are compared using the same assay system (ideally by the same team, on the same day) and ultimately validated in an appropriate GDEPT model.

One very interesting NfsB homologue that has undergone just such a validation is Neisseria meningitidis NfsB, which significantly outperformed E. coli NfsB in CB1954 mouse xenograft growth delay assays, as well as overall survival, when each enzyme was delivered to the tumours via C. sporogenes vectors [106]. Two particularly noteworthy achievements in that study were: (1) the successful integration of each nitroreductase gene into the C. sporogenes chromosome, a major engineering achievement that negates the segregational instability of plasmids in this host and paves the way for regulatory acceptance of CDEPT technologies; and (2) the attainment of a complete cure in a proportion of the treated mice (two out of 16 for the E. coli NfsB cohort and four out of 16 for the N. meningitidis NfsB cohort).

Surprisingly, the nitroreductase family that we consider to hold the most promise for GDEPT, E. coli NfsA and homologues thereof, was not studied in this context until recently. The CB1954 reductase activity of the NfsA enzyme from E. coli was first noted in 2006 [107], although not characterized in detail until a few years thereafter [108,109]. In contrast with E. coli NfsB, NfsA reduces CB1954 almost entirely at the 2-nitro position and exhibits a strong preference for NADPH over NADH as cofactor. Searle and colleagues [109] at the University of Birmingham proposed that this latter phenomenon may help explain why the seminal study of Anlezark et al. [78] recovered NfsB rather than NfsA as the most prominent nitroreductase present in E. coli cell extract, post-addition of exogenous NADH to the extract. (Searle and colleagues [109] also sharply observed that the initial assays of Anlezark et al. [78] used menadione rather than a nitroaromatic substrate to monitor enzyme activity and that NfsB reduces menadione approximately twice as efficiently as NfsA.) Certainly, E. coli NfsA appears superior to E. coli NfsB in a direct comparison with purified enzyme activity with CB1954; in the two independent characterizations conducted by Searle and colleagues [109] and ourselves [108], NfsA was found to possess an approximately 50–60-fold lower KM than NfsB, albeit with a 2.5–5-fold lower kcat (Table 2). Searle and colleagues [109] went on to show that the preference for reduction at the 2-nitro position of CB1954 resulted in an elevated bystander effect in NfsA- compared with NfsB-transfected cell lines; that purified recombinant NfsA was up to 20-fold more effective than purified recombinant NfsB in killing SKOV3 ovarian carcinoma cells in a cell culture assay similar to that used by Anlezark et al. [102] (i.e. an assay system that is possibly more sensitive to the 2-nitro reduction products); and that the nfsA gene was 6.3-fold more effective than nfsB in sensitizing cells to CB1954 when each was delivered to cultured SKOV3 cells via a non-replicating adenoviral vector. Our own human cell data were less clear-cut, suggesting that nfsB was in fact slightly more effective than nfsA at sensitizing stably transfected HCT-116 colon carcinoma cells to CB1954, but confounded by poorer expression of nfsA in those cells irrespective of whether the human EF-1α promoter [108] or the cytomegalovirus (CMV) early immediate promoter [110] was used to drive gene expression.

Variable gene expression levels in stably transfected human cell lines was a prominent theme in the second of those studies, in which we had expanded our library of candidate nitroreductases to comprise 47 enzymes from 11 different bacterial oxidoreductase families, including 12 NfsA and 12 NfsB family members [110]. This gave us sufficient data points to evaluate correlations of enzyme activity with cell killing in different models. However, whereas relative gene expression levels were fairly consistent when the candidate nitroreductases were overexpressed in E. coli and there was a reasonable correlation (R2=0.54) between the kcat/KM of our top 20 purified nitroreductases with CB1954 as substrate and their ability to induce CB1954-mediated DNA damage in E. coli host cells, gene expression was unpredictable in stably transfected HCT-116 cells and there was little correlation between the CB1954 sensitivity of those cell lines and the kcat/KM of the purified nitroreductases [110]. We have since tested gene expression levels for many of these and other nitroreductases in both HCT-116 and HEK (human embryonic kidney)-293 cells, with consistent results, which strongly indicate that long-term stable expression of some nitroreductases is just not tolerated in human cell lines (E.M. Williams, A.M. Mowday, C.P. Guise, J.N. Copp, S.L. Condon, J.B. Smaill, D.F. Ackerley and A.V. Patterson, unpublished work). Our working hypothesis is that this is a substrate promiscuity issue, with certain nitroreductases interfering with essential metabolic pathways in human cells to such an extent that their expression is poorly (or not at all) tolerated. If so, this is not incompatible with the potential use of these enzymes in GDEPT, certainly using bacterial vectors, which compartmentalize the nitroreductase activity away from the human cellular environment, but also probably with viral vectors, which in therapy are more akin to a transient than a long-term stable transfection scenario. Nevertheless, an inability to generate stably transfected human cell lines precludes some valuable pre-clinical experiments, such as studying the bystander effect in 3D mixed multicellular layers [42] or tumour growth delays using xenografts comprise a known proportion of nitroreductase-transfected cells at the time of implant [87,111].

In the same study, we also observed that certain nitroreductases (e.g. YcnD from Bacillus subtilis) were not effective in sensitizing human cells to CB1954, despite being effective in sensitizing E. coli cells and having a very high kcat/KM for CB1954 as a purified recombinant enzyme [110]. This may indicate that competing metabolites present in human cells, but not in E. coli, can impair the activity of some nitroreductases in the human cellular environment. If so, these nitroreductases may still be good candidates for GDEPT using a bacterial delivery vector, but not a viral vector, as the latter mandates that the nitroreductase be able to function within human cells. Ultimately, the heterogeneity in expression of different nitroreductases in different organisms that we observed highlights the importance of considering each nitroreductase candidate for GDEPT in the context of its intended vector, as there may prove to be profound differences in enzyme efficacy when using different vectors in different settings.

Difficulties with stable expression in human cell lines aside, our screening of a large nitroreductase library did uncover some promising new candidates for CB1954 GDEPT [110]. Two NfsB family members, NfsB from Vibrio vulnificus and YfkO from B. subtilis, were found to have a similar kcat to E. coli NfsB but at least a 4-fold lower KM. In contrast, nine of the ten NfsA family members that were purified and analysed, had at least a 40-fold lower KM than NfsB, whereas their kcat values were typically only 1.5–5-fold lower than NfsB (Table 2). In our experience, a lower maximum turnover rate, but much higher affinity (i.e. much lower KM) for a given prodrug substrate is virtually a defining characteristic of the NfsA family relative to the NfsB enzymes hence, our rationale that the NfsA enzyme family holds the greatest promise for nitroreductase GDEPT. One potential drawback of the NfsA enzymes, however, is their characteristic preference for NADPH over NADH (Table 2), as there is some evidence that NADPH is usually present at lower concentrations than NADH, in both mammalian and bacterial cells [112114]. Another distinguishing feature between the two families is that all NfsA enzymes characterized to date reduce CB1954 exclusively at the 2-nitro position, whereas the NfsB enzymes either favour 4-nitro reduction or else generate both the 2- and the 4-nitro reduction products (Table 2). Although the increased bystander capabilities and high toxicity of the 2-amino reduction product of CB1954 [81] do suggest that reduction at the 2-nitro position might generally be preferable for GDEPT, it is certainly conceivable that the bi-functional alkylating derivatives of the 4-hydroxylamine reduction product [72] might prove more effective in some contexts.

In a recent work, we have also uncovered members of some minor type I bacterial nitroreductase families that possess CB1954 reductase activity, including Pseudomonas aeruginosa MsuE [115], E. coli NemA [64,108], E. coli AzoR [108] and homologues of the latter two enzymes [110]. However, none of these enzymes have activity approaching the level of the NfsA or NfsB family members (Table 2), so they do not appear to be attractive prospects for CB1954 GDEPT.

ENGINEERING SUPERIOR NITROREDUCTASES FOR CB1954 GDEPT

An alternative to discovering new nitroreductase enzymes is to engineer existing ones to improve their capacity to activate a prodrug substrate. A powerful tool that has been employed to achieve this is directed evolution, which uses mutagenesis at a single gene level, followed by a screen or selection, to identify enzyme variants that are enhanced in a desirable activity. Mutagenesis can either be random or targeted to regions of particular interest, e.g. the codons that specify active-site residues [116]. The resulting mutant library can then either be screened directly or, given that a large proportion of variants may be non-functional as a result of the mutagenesis, an artificial selection pressure can applied to enrich for functional mutants possessing a desired activity [117]. By replacing natural selection with an artificial screen or selection pressure, mutants that are highly efficient at catalysing a non-natural reaction can sometimes be achieved after only a few rounds of evolution. Nitroreductase enzymes should be particularly well-suited for directed evolution experiments, as promiscuous activities can often be rapidly improved through only a few residue substitutions [117119].

Library generation strategies for improving nitroreductase activity with CB1954 have to date primarily been semi-rational, i.e. utilizing information from solved protein crystal structures to identify specific target residues, rather than employing more random mutagenesis approaches such as error-prone PCR. Methods for subsequently screening the mutant gene libraries have utilized both negative and positive selection, as well as colorimetric screens, to identify nitroreductase variants better able to activate CB1954. An important first example of nitroreductase-directed evolution from Searle and colleagues used the previously solved crystal structure of E. coli NfsB [53] to identify the substrate-binding pocket of the enzyme [120]. A site-saturation mutagenesis strategy was then employed to individually alter each of the nine selected codons to the degenerate codon NNN. This resulted in nine distinct mutant libraries, each comprising 64 gene iterations that collectively specified each of the 20 possible amino acids at a particular active-site position [120]. For screening purposes, each library was transformed into an E. coli nfsB gene-deleted host strain, with individual gene variants integrated into the chromosome of their host cell using a bacteriophage λ vector. A negative selection was then applied to screen for superior CB1954 activating mutants, by replica plating individual lysogens on a series of agar plates that contained escalating CB1954 concentrations. Clones that expressed superior NfsB variants were non-viable at CB1954 concentrations that did not inhibit growth of a wild-type control strain. When delivered to SKOV3 cells via a replication-deficient adenoviral vector, the top (F124K) mutant recovered from these screens sensitized SKOV3 cells to approximately 3–6-fold less CB1954 than the wild-type nfsB control [120] and the purified protein was subsequently shown to be 2.4-fold improved in terms of kcat/KM [104]. Moreover, combining F142K with N71S, another beneficial single mutation identified in the initial screens [120], synergistically improved the kcat/KM of the double mutant beyond the additive level of improvement contributed by each mutation in isolation [104]. It should be noted, however, that such positive epistatic effects cannot be relied upon; in a later study by Searle and colleagues [121], it was observed that combining two individually beneficial mutations could in fact be detrimental to activity with CB1954 relative to an unmutated NfsB control.

Searle and colleagues’ early work [121] provided an important proof-of-principle for the ability to substantially improve, via just a few residue substitutions, prodrug activation by nitroreductases. However, from a practical perspective, the labour-intensive and low-throughput nature of the replica plating really limited the original screening protocol to analysis of individual mutations targeting active-site residues. It has been calculated that when using an NNN codon degeneracy, it is necessary to screen 190 variants to achieve a 95% level of confidence that all possible amino acid changes have been examined [122]; and even using reduced codon degeneracies such as NNK (a total of 32 codons that collectively specify all 20 amino acids and only one stop codon) or NDT (12 codons that specify 12 different amino acids, encompassing all of the main classes of side chains) it would be necessary to screen ∼ 100000 or 5000 variants respectively to thoroughly plumb a gene library in which three different codons had been simultaneously randomized.

Given the unpredictable effects of combining individually beneficial mutations, as well as the fact that it may also be desirable to introduce mutations beyond the active site, which can exert powerful stabilizing effects on the enzyme or improve substrate accessibility [123], there is clear benefit to having a high-throughput screening or selection capability that can reliably interrogate large libraries (e.g. comprising many millions of variants) in a short period of time. Searle and colleagues [120] arrived at an ingenious solution to achieve this for directed evolution of nitroreductases, based on their previous work using lysogenized E. coli to detect CB1954-mediated DNA damage and subsequent recognition that cytotoxicity had almost certainly been mediated by the bacteriophage vector transitioning from a lysogenic to a lytic cycle post-activation of the E. coli SOS (DNA damage repair) response, rather than direct bactericidal effects of the activated CB1954 metabolites [124]. The SOS response is a stress/repair pathway induced by genotoxic damage, leading to the release of ssDNA [125]. In addition to inducing a repair network of genes, the SOS response can also induce chromosomally integrated bacteriophages to enter the lytic cycle and exit their host cell. Searle and colleagues [124] recognized that this phenomenon could be exploited to select for nitroreductase variants able to induce the host SOS response at lower prodrug concentrations [124] (Figure 5). Following the optimization of conditions using their previous improved mutants as controls, they created a new NfsB mutant library in which three of the five possible catalytically important codons per gene variant had been randomized using NNN codon mutagenesis (∼1000000 variants in total). A mixed culture of transfected E. coli lysogens was then transiently challenged with a low dose of CB1954, allowing the most actively expressed nitroreductase variants to convert sufficient prodrug to induce the phage lytic cycle. The released phage were then collected and subjected to two more iterations of CB1954 selection. Following recovery and validation of individual clones, the top triple mutant lysogen (T41Q/N71S/F124T) was 48-fold more sensitive to CB1954 than the lysogens expressing wild-type NfsB [124]. This was mirrored by a 45-fold improvement in kcat/KM with CB1954 as a substrate, measured in a follow-up study [105]. As a whole, the engineering work conducted by Searle and colleagues [124] validated two important tenets of directed evolution in a nitroreductase-specific context: (1) simultaneous mutation of multiple sites can often produce catalytic improvements beyond that achievable via single codon changes; and (2) a high throughput method can be invaluable in enabling large libraries generated by multi-site mutagenesis to be thoroughly screened for superior enzyme variants.

Direct positive selection for active nitroreductases from large-scale mutant libraries via the SOS-triggered bacteriophage λ lytic cycle

Figure 5
Direct positive selection for active nitroreductases from large-scale mutant libraries via the SOS-triggered bacteriophage λ lytic cycle

A nitroreductase mutant library is cloned into λ phages that are then used to infect E. coli. The bacteria are briefly exposed to sub-lethal concentrations of CB1954, which in any E. coli bacteria expressing active nitroreductase mutants will be converted into a DNA-damaging form, triggering the SOS response. This causes RecA activation, leading to proteolysis of the λ cI repressor and switching the λ phage into the lytic cycle. Phage replication occurs and subsequent lysis of the host E. coli releases large quantities of phage that carry the most active nitroreductase variants. Successive cycles of infection and CB1954 exposure lead to populations highly enriched for mutants that possess superior CB1954 reductase activity [124].

Figure 5
Direct positive selection for active nitroreductases from large-scale mutant libraries via the SOS-triggered bacteriophage λ lytic cycle

A nitroreductase mutant library is cloned into λ phages that are then used to infect E. coli. The bacteria are briefly exposed to sub-lethal concentrations of CB1954, which in any E. coli bacteria expressing active nitroreductase mutants will be converted into a DNA-damaging form, triggering the SOS response. This causes RecA activation, leading to proteolysis of the λ cI repressor and switching the λ phage into the lytic cycle. Phage replication occurs and subsequent lysis of the host E. coli releases large quantities of phage that carry the most active nitroreductase variants. Successive cycles of infection and CB1954 exposure lead to populations highly enriched for mutants that possess superior CB1954 reductase activity [124].

Independently, we also turned to the E. coli SOS response as a means for monitoring prodrug activation by an expressed nitroreductase candidate, based on work that one of us (D. F. Ackerley) had previously conducted using a tool called the SOS chromotest to study the genotoxic effects of hexavalent chromium on bacteria [126]. The SOS chromotest is an in vivo assay that uses an E. coli reporter strain containing a lacZ (β-galactosidase) reporter gene under transcriptional control of an SOS-inducible promoter [127]. We showed that plasmid-based expression of a candidate nitroreductase in the SOS chromotest strain, followed by prodrug challenge, enabled the relative levels of DNA damage to be quantified in colorimetric β-galactosidase assays [108] (Figure 6A). The basic system was iteratively improved by identifying and deleting all endogenous nitroreductase genes from the chromosome of the E. coli reporter strain [108,110], as well as creation of variants tailored for certain applications (e.g. deletion of the endogenous nucleotide excision repair gene uvrB to increase sensitivity to DNA cross-links, but not mono-adducts [128], or deletion of the tolC gene that encodes an efflux pump capable of exporting certain prodrugs from the cell [110]). These tools have allowed us not only to rapidly assess the prodrug-activating capabilities of collections of nitroreductases sourced from Nature, while conserving prodrug stocks relative to growth inhibition assays [108,110], but also to successfully enhance CB1954 reduction by a leading nitroreductase candidate (Aliivibrio fischeri FRaseI, an NfsB homologue) via site-saturation mutagenesis strategies similar to those employed by Searle and colleagues, resulting in a series of mutants with up to 8.4-fold improvement in kcat/KM and 12.6-fold improvement in KM relative to wild-type FRaseI (which was already approximately twice as efficient as E. coli NfsB with CB1954 as substrate) [129].

E. coli SOS assays used to monitor the activity of nitroreductases with nitroaromatic prodrugs

Figure 6
E. coli SOS assays used to monitor the activity of nitroreductases with nitroaromatic prodrugs

(A) Overexpression of nitroreductases able to activate a prodrug to a genotoxic form results in DNA damage and activation of the E. coli SOS response. The strength of the response can then be quantified in specialized SOS reporter strains [110,131], in this example by β-galactosidase assay. (B) FACS to enrich active nitroreductase variants from large-scale mutant libraries. E. coli SOS reporter cells engineered to both express a bacterial nitroreductase variant and to generate GFP in response to DNA damage are pooled and exposed to the prodrug of interest. Any individual E. coli cells carrying active nitroreductases convert the prodrug into a DNA-damaging form and subsequently begin to make GFP. The pooled E. coli are then analysed by flow cytometry and the most highly fluorescent cells are selected from the pool and cultured on an individual basis for further analysis [131].

Figure 6
E. coli SOS assays used to monitor the activity of nitroreductases with nitroaromatic prodrugs

(A) Overexpression of nitroreductases able to activate a prodrug to a genotoxic form results in DNA damage and activation of the E. coli SOS response. The strength of the response can then be quantified in specialized SOS reporter strains [110,131], in this example by β-galactosidase assay. (B) FACS to enrich active nitroreductase variants from large-scale mutant libraries. E. coli SOS reporter cells engineered to both express a bacterial nitroreductase variant and to generate GFP in response to DNA damage are pooled and exposed to the prodrug of interest. Any individual E. coli cells carrying active nitroreductases convert the prodrug into a DNA-damaging form and subsequently begin to make GFP. The pooled E. coli are then analysed by flow cytometry and the most highly fluorescent cells are selected from the pool and cultured on an individual basis for further analysis [131].

Although having advantages relative to the earlier replica plating screen of Searle and colleagues [120], β-galactosidase assays still do not allow for high throughput screening of an extensively mutated nitroreductase gene library, as per the elegant bacteriophage λ system that the Searle laboratory developed subsequently [124]. To address this, we created an alternative SOS reporter strain that expresses GFP rather than β-galactosidase in response to DNA damage. Not only does the sensitive GFP signal enable miniaturization of our SOS assays to a 384-well plate format, conserving valuable prodrug stocks, it also allows for FACS to recover the most active nitroreductase variants from a gene library (Figure 6B). Owing to the small size and auto-fluorescence of E. coli [130], the signal to noise ratio is not as high as might be desirable when attempting to recover the most strongly GFP-fluorescent cells via FACS; even so, we have shown that our system can provide a powerful enrichment for the most active CB1954-reducing clones in a mixed population [131]. After a single sort of SOS–GFP cells expressing a strongly active nitroreductase (B. subtilis YcnD) mixed with SOS–GFP cells expressing a poor nitroreductase (E. coli YdjA) at a 1:1000 starting ratio and challenged with CB1954, 38% of recovered cells were found to express YcnD. More impressively, two consecutive sorts of YcnD and YdjA expressing cells mixed in a 1:100000 starting ratio resulted in a 90000-fold enrichment of YcnD-expressing cells [131]. In our ongoing work, we are using this system to improve prodrug activation by a range of lead enzymes, with particular focus on generating more active variants of E. coli NfsA. Although the low signal to noise ratio of E. coli FACS means that a proportion of undesirable clones will always be selected, these can quickly be eliminated using secondary fluorescence assays and/or growth inhibition screens in microtitre plates, and a big advantage of the system is its flexibility and ease of use. In contrast, although it can clearly be very effective, there is evidence that the bacteriophage λ selection system of Searle and colleagues [124] can also be technically challenging to work with [132].

The final enzyme for which improved activity with CB1954 has been reported as a consequence of enzyme engineering is the chromate reductase YieF from E. coli [133]. Although in SOS and purified enzyme kinetic assays the native enzyme was not an efficient activator of CB1954 [108], a YieF variant (Y6, later renamed ChrR6) that had previously been evolved for improved chromate reduction was serendipitously found to also exhibit enhanced activity with CB1954 [107]. When a gene encoding ChrR6 was delivered via an attenuated tumour-tropic Salmonella Typhimurium vector it sensitized cultured HeLa cervical cancer cells to an ∼10-fold lower concentration of CB1954 than an equivalent strain delivering E. coli nfsA [107].

The finding that three distinct nitroreductases, E. coli NfsB and YieF and A. fischeri FRaseI, can be engineered to improve their activity with CB1954 clearly demonstrates the mutational plasticity of these enzymes. However, to date no mutant nitroreductases have been evaluated in clinical GDEPT trials. With the recent advent of clearly superior next generation prodrugs, it is no longer clear that CB1954, even in partnership with substantially more active nitroreductases than have previously been tested, offers a promising way forward for clinical GDEPT.

ALTERNATIVE PRODRUGS FOR NITROREDUCTASE GDEPT

The clinical development of GDEPT has been impeded by the regulatory costs and concerns associated with ensuring the safety of each distinct component of the therapy. These can be mitigated by repurposing a prodrug that has already undergone independent clinical development. Of particular relevance to nitroreductase GDEPT are a range of compounds that have previously been evaluated as hypoxia-activated prodrugs [12]. Hypoxia, a state of abnormally low oxygen that arises to at least some extent in the majority of solid tumours larger than 1 mm3 is a well-established negative prognostic feature for treatment [134]. However, as it does not typically arise in healthy tissues, hypoxia provides a unique physiology for targeted cancer treatments to respond to [12]. Hypoxia-activated prodrugs are converted into highly toxic forms in hypoxic tissue because they are substrates for type II (i.e. oxygen-sensitive) human oxidoreductases that can only generate the reduced toxic product in the absence of oxygen [46,135137]. GDEPT using type I nitroreductases offers the potential to expand the therapeutic range of hypoxia-activated prodrugs to aerobic tumour tissues as well.

A class of hypoxia-activated prodrugs that has received a substantial amount of attention as next-generation nitroaromatic prodrugs for GDEPT is the dinitrobenzamide mustards (DNBMs). The transition from an electron-withdrawing nitro group to an electron-donating hydroxylamine or amine group, which can be catalysed either by type I nitroreductases or by type II nitroreductases specifically under hypoxic conditions, activates the latent mustard moiety [48]. Activated nitrogen mustards are extremely potent drugs due to their ability to bind to the N7 position of guanine and subsequently cross-link DNA [138] and were in fact among the first small molecules ever used to treat cancer [139]. Crucially, the active metabolites have an appreciable half-life and are able to diffuse out of the cell of origin, resulting in bystander cell killing of neighbouring cells [42,140]. When DNBMs have been directly compared against CB1954 in GDEPT models they have consistently proved more efficacious. SN23862 {5-[bis(2-chloroethyl)amino]-2,4-dinitrobenzamide, the dichloro-mustard analogue of CB1954; Figure 3] was one of the first dinitrobenzamide compounds to demonstrate hypoxia selectivity; a 60-fold index for hypoxic compared with oxygenated Chinese hamster ovary UV4 cells compared with 3.6-fold index for CB1954 [141]. It was subsequently found that E. coli NfsB was slightly more efficient with SN23862 as a substrate than with CB1954 (kcat/KM of 10.6 mM−1·s−1 compared with 7.0 mM−1·s−1) [78,142] and that the activated DNBM metabolites demonstrated increased dose potency and bystander cell killing [81,111]. A substantial increase in bystander potential relative to CB1954 has generally been observed across a large panel of DNBMs and has been attributed at least in part to their greater lipophilicity [87,111,140]. However, although this lipophilicity is advantageous for the bystander effect, the concomitant decrease in aqueous solubility can limit in vivo delivery and efficacy. This can be alleviated by attaching a phosphate ester moiety to a pendant alcohol-bearing side chain of the DNBM to generate a water-soluble ‘pre-prodrug’ that in vivo undergoes rapid facile systemic conversion into the lipophilic DNBM prodrug [137].

The apparent superiority of DNBMs over CB1954 for nitroreductase GDEPT has also been observed in tumour xenograft models. In one study mice carrying HCT-116 tumour xenografts seeded with 10% of cells stably expressing E. coli NfsB showed an increase in median survival of only 11 days following CB1954 treatment compared with 62 days with the DNBM SN28343 [87] (Figure 3). Another pre-clinical study that compared CB1954 and the DNBM pre-prodrug PR-104 [2-((2-bromoethyl)(2,4-dinitro-6-((2-(phosphonooxy)ethyl)carbamoyl)phenyl)amino)ethyl methanesulfonate; Figure 3] used C. sporogenes to deliver a codon-optimized E. coli nfsB gene to human cervical carcinoma (SiHa) xenografts in mice [89]. Following treatment with either compound a significant (P<0.05) tumour growth delay was observed for >13 days, with the CB1954-treated tumours growing markedly after that time, whereas the PR-104-treated tumours showed no increase in mean size after 25 days, the last time-point presented [89]. Of profound importance, both prodrugs were administered at approximately 30% of the mouse MTD: 50 mg/kg for CB1954 and 250 mg/kg for PR-104. As noted above, a dose of 50 mg/kg CB1954 in mice yields a plasma AUC approximately 50-fold greater than that achievable in humans [96,97]. In contrast, 250 mg/kg PR-104 results in a murine plasma AUC only 1.2-fold greater than that achievable in humans [143]. Thus, we conclude from this work that PR-104 was more efficacious than CB1954 despite being administered at a substantially lower equivalent dose.

To date PR-104 is the only DNBM hypoxia-activated (pre-)prodrug to have advanced to clinical trials. In vivo, PR-104 is rapidly converted into the corresponding lipophilic DNBM alcohol PR-104A [2-((2-bromoethyl)(2-((2-hydroxyethyl)carbamoyl)-4,6-dinitrophenyl)amino)ethyl methanesulfonate] [137] (Figures 3 and 4B). The reduced hydroxylamine and amine products of PR-104A cause inter-strand DNA cross-links (Figure 4B) that result in disruption of the replication fork upon mitosis, the most likely mechanism of cytotoxicity [144,145]. Phase I studies in patients with solid tumours determined that the main dose-limiting toxicity for PR-104 was delayed onset and prolonged myelosuppression (predominantly thrombocytopenia and neutropenia), observed in a number of patients [146,147]. A factor that is likely to make a significant contribution to the myelosuppression is that PR-104A can undergo a surprising off-mechanism aerobic (i.e. type I) nitro-reduction catalysed by human aldo-keto reductase 1C3 (AKR1C3) [148], an enzyme not previously known to possess any nitroreductase activity but that is expressed at high levels in normal CD34+ myeloid progenitor cells [149]. Disappointingly, this dose-limiting myelotoxicity, together with high levels of clearance mediated by glucuronidation of the alcohol side chain [150], has limited the clinical progression of PR-104 as a hypoxia-activated prodrug. Nonetheless, the advanced clinical status of PR-104 and its high dose-potency relative to CB1954 mean it remains an attractive potential prodrug for GDEPT. It is not unreasonable to suppose that if partnered with a sufficiently active nitroreductase (particularly one able to reduce PR-104A at very low concentrations) significant anti-tumour effects may be possible at the tolerated plasma concentrations of PR-104A. Clinical data suggest that 270 mg/m2 PR-104 can be safely administered on repeated weekly cycles [147,151]. Alternatively, the design of second-generation analogues of PR-104A that are resistant to metabolism by AKR1C3 represents a promising approach that is ongoing in our laboratories.

In terms of the most suitable enzyme partners for these prodrugs, work performed by both Searle and colleagues and our own team has indicated that the NfsA nitroreductase family is generally more effective than the NfsB family in catalysing the bioreductive activation of DNBMs. If anything, the difference in efficacy between the two families is even more pronounced with DNBM prodrugs than with CB1954. In SKOV3 cell lines expressing either E. coli NfsA or E. coli NfsB, matched to be equally sensitive to CB1954, the NfsA-expressing cell line was between 1.8- and 2.8-fold more sensitive to a panel of three different DNBM analogues of CB1954 (SN24927, SN26209 and SN27653; Figure 3) [109]. Likewise, PR-104A challenge induced a significantly greater mean SOS response in our 12 E. coli strains expressing NfsA homologues than the 12 strains expressing NfsB homologues (4.4-fold compared with 2.7-fold induction relative to unchallenged control strains; P<0.05, Student's t test) [110]. When we purified and measured the kinetic efficiency with PR-104A of the top ten NfsA homologues and top seven NfsB homologues from that nitroreductase library, the mean kcat/KM of the NfsA enzymes was approximately 5-fold that of the NfsB enzymes (260 and 54 mM−1·s−1 respectively) [110].

When assessing the kinetic efficiency of nitroreductases with PR-104A as a substrate, it is important to also identify the end-product(s) of reduction, as only reduction of the 4-nitro substituent (para to the mustard) yields an effective cytotoxin (Figure 4B). Reduction of the 6-nitro group (ortho to the mustard) is followed by intramolecular alkylation to form a cyclized non-toxic tetrahydro-quinoxaline derivative [137]. Whereas NfsB family enzymes typically reduce either exclusively the nitro group ortho to the aziridine ring in CB1954 or a mixture of the ortho and para nitro groups, all of the NfsB as well as NfsA homologues we have tested to date exclusively reduce the nitro group para to the mustard in PR-104A and hence generate a cytotoxic product [110]. E. coli NemA is the only bacterial nitroreductase we have observed to generate the tetrahydro-quinoxaline derivative of PR-104A and it does so exclusively [110].

A more clinically advanced hypoxia-activated nitroaromatic prodrug is evofosfamide (TH-302; Figure 3), currently in two Phase III trials, one to treat soft tissue carcinoma in combination with doxorubicin (https://www.clinicaltrials.gov/ct2/show/NCT01440088) and a second to treat pancreatic adenocarcinoma in combination with gemcitabine (https://www.clinicaltrials.gov/ct2/show/NCT01746979). Evofosfamide has, as of 12 May 2015, been awarded fast-track status for each trial, based on its potential to address unmet clinical need. Evofosfamide is a 2-nitroimidazole-based prodrug, for which nitro-reduction triggers an intramolecular fragmentation that releases bromo-isophosphoramide mustard [152]. In addition to activation by hypoxia, evofosfamide has also been shown to be a substrate for NfsA from E. coli [153]. In hyperoxic conditions HCT-116 cells stably expressing NfsA were ∼200-fold more sensitive to evofosfamide as compared with the parental cell line, a value very similar to the increased evofosfamide sensitivity (∼250-fold) of parental HCT-116 in hypoxic compared with normoxic conditions. The bystander effect of reduced evofosfamide metabolites [153] appears to be more modest than those reported for the DNBMs [111,140], which may explain why evofosfamide is yet to receive much attention as a potential prodrug for GDEPT. Nonetheless, the advanced clinical status of evofosfamide suggests that further investigation of evofosfamide-bacterial nitroreductase pairs might be worthwhile, particularly if evofosfamide is soon approved for use in patients.

Whereas use of prodrugs that have already undergone safety evaluations in Phase I clinical trials would be advantageous for GDEPT, researchers have also considered additional nitroheterocyclic compounds that have not as yet been trialled in humans. An excellent review by Denny [39] provides substantial detail on early efforts with a range of potential prodrugs for nitroreductase GDEPT, including other aziridinyldinitrobenzamide analogues of CB1954 and 4-nitrobenzyl carbamate prodrugs. This latter class encompasses a wide variety of chemical structures that share a general mechanism where reduction of the 4-nitrobenzyl moiety to a hydroxylamine triggers intramolecular fragmentation and release of a (previously neutralized) cytotoxic effector. Subsequent investigations of other CB1954 analogues [81] and nitroheterocyclic carbamate-triggered prodrugs [154156] have also been reported. Most recently, however, there appear to be two main classes of prodrug that have emerged as potential candidates for future trials. One of these is the nitro-chloromethylbenzindolines (nitro-CBIs) which, once again, were originally envisaged as hypoxia-activated prodrugs [157] or pre-prodrugs [158]. Following nitroreduction, these compounds exert toxicity via alkylation in the minor groove of DNA [157]. These reduced metabolites possess extremely powerful bystander effects; for example, in oxygenated 3D mixed multicellular layers made up of only 5.8% ‘activators’ (HCT-116 cells expressing NfsB from P. aeruginosa), the remaining non-nitroreductase-expressing ‘target’ cells were only 2-fold less sensitive than the activators to the prodrug nitro-CBI-DEI (nitro-CBI-5-[(dimethylamino)ethoxy]indole; Figure 3), indicating a strikingly efficient transfer of cytotoxicity [159]. Another emerging class of prodrug that may offer prospects for GDEPT is the nitrobenzylphosphoramide mustards (Figure 3), related to evofosfamide; in particular a series of 4-nitrobenzylphosphoramide mustards that also appear to possess useful bystander effects and have been identified as being good substrates for E. coli NfsB, as well as nitroreductases sourced from a selection of trypanosome parasites [160]. In vitro assays demonstrated a comparable potency to that of CB1954 for some of the 4-nitrobenzyl phosphoramide mustards, as well as a higher degree of selectivity between nitroreductase-expressing and-non-expressing cells, in both V79 and SKOV3 lines [160,161]. However, to date, no lead compounds from either the nitro-CBI or the nitrobenzylphosphoramide mustard prodrug classes have been tested in combination with their preferred nitroreductase partners in a relevant pre-clinical model of GDEPT.

Another recent prodrug candidate, which was evaluated in murine GDEPT models, is 6-chloro-9-nitro-5-oxo-5H-benzo(a)phenoxazine (CNOB) (Figure 3). An interesting feature of CNOB is that its reduced toxic product 9-amino-6-chloro-5H-benzo(a)phenoxazine-5-one (MCHB) is far-red-fluorescent. Because of this property, CNOB was first studied as a reporter substrate for E. coli NfsB, used to confirm tumour localization and germination of nfsB-labelled C. sporogenes spores in murine SiHa xenografts [89]. However, it was subsequently demonstrated by Matin and colleagues at Stanford University that the reduced product MCHB is cytotoxic in its own right and was able to trigger apoptosis in a range of tumour cell lines that were expressing the engineered variant of E. coli YieF, ChrR6 [162]. Anti-tumour effects were also observed in vivo against transplanted murine 4T1 breast tumours pre-labelled with ChrR6 and, to a lesser extent, against untransfected 4T1 tumours post-delivery of chrR6 via an attenuated Salmonella Typhimurium vector. Moreover, the fluorescent properties of MCHB enabled direct visualization of a promising bystander effect [162]. However, the polyaromatic nature of CNOB may prove an impediment to human clinical evaluation due to limited solubility and tissue penetration, unless a suitable formulation strategy can be devised (e.g. encapsulation in tumour-targeting microvesicles similar to those being developed by Matin and colleagues [163]).

One last nitroaromatic prodrug class worthy of mention, the 5-nitroimidazole antibiotics, offers prospects for an entirely different GDEPT strategy, one seeking to target the tumour vasculature. For this scenario, it has been argued that a local bystander effect is in fact undesirable, as it would lead to washout of the activated metabolite(s) via the bloodstream, increasing systemic toxicity while decreasing microvascular cell killing [20]. It has also been shown that, for effective vascular disruption in in vitro models, cytotoxic rather than cytostatic activity is required. Thus, the enzyme–prodrug combination of HSK-tk and ganciclovir was ineffective, whereas E. coli NfsB in combination with either CB1954 or the clinically approved 5-nitroimidazole prodrug antibiotic metronidazole (Figure 3) was efficient at disryesupting vascular network-like endothelial cell structures [164]. Metronidazole, an anti-protozoal and also one of the most effective pharmaceuticals against anaerobic bacterial infections, is particularly attractive for such applications due to its lack of any discernible bystander effect [164,165]. We have shown that a wide range of bacterial type I nitroreductases are able to convert metronidazole [128] or its structural analogue tinidazole (Figure 3) [166] (also a clinically approved anti-protozoal) into potent cytotoxins. An additional benefit of nitroreductases that exhibit high levels of antibiotic prodrug activation is that they may provide a valuable safety mechanism during trials, by hyper-sensitizing bacterial GDEPT vectors to those antibiotics. For example, C. sporogenes vectors are routinely cleared using metronidazole [167,168], so use of therapeutic genes that concomitantly enhance sensitivity to metronidazole would be favourable from a regulatory perspective. Other clinical nitro-prodrug antibiotics such as nitrofurazone and nitrofurantoin (Figure 3), which can be aerobically activated by both E. coli NfsA and NfsB [109], may also have value in this regard.

Finally, it is worth considering that the majority of type I nitroreductases are also efficient quinone reductases and hence likely to be capable of bioreductive activation of quinone prodrugs as well as nitroaromatic ones. Indeed, Matin and colleagues [163] reported that both E. coli NfsA and their serendipitously evolved chromate reductase ChrR6 were better able to sensitize cultured HeLa cells not only to CB1954, but also to the clinically approved quinone prodrug mitomycin C [107]. Although similar results were not observed with E. coli NfsB [142], the data of Matin and colleagues [163] suggest potential GDEPT applications for certain bacterial nitroreductases in partnership with clinically advanced or approved quinone prodrugs such as mitomycin C, porfiromycin, RH1 or EO9 (apaziquone; Figure 3) [136].

ADDITIONAL DIRECTIONS FOR NITROREDUCTASE GDEPT

Non-invasive imaging

It is widely recognized that the ability to non-invasively image the location and spread of a gene delivery vector would be extremely beneficial for its clinical development [169173]. Gene therapy vectors are novel, complex and, in some cases, replication-competent; thus to support patient safety, non-invasive imaging that is sensitive, encompasses the whole body and is repeatable (allowing monitoring of vector replication over time) is highly desirable. Without this capability, current clinical practice for vector monitoring typically consists of multi-site biopsies [96,174,175] and/or PCR-based determination of levels of viral DNA/RNA in a patients plasma, urine and stool [29,176,177]. These methods are far from ideal: plasma virus levels give no indication of vector location, whereas the invasive nature of biopsy is not only a burden on the patient, but also precludes the possibility of sampling over time, is subject to sampling bias and restricts the locations where gene delivery can be assessed.

Non-invasive whole body imaging of GDEPT vectors can potentially be achieved by engineering the vector to express a reporter protein that can convert an exogenous probe to an imageable form [178]. For this purpose, nuclear imaging technologies, in which vector location is determined by monitoring a systemically delivered radiotracer that accumulates in cells post-interaction with the reporter protein, offer a number of distinct advantages. They are highly sensitive, quantitative, repeatable, can give reasonable spatial resolution (3–4 mm) and have no clinically relevant depth limit [179181].

Ideally for GDEPT, the reporter protein and the therapeutic enzyme would be one and the same, to enable the imaging capability to directly report on the location and extent of the therapeutic potential. This has previously been achieved for HSV1-tk, which can phosphorylate a number of 18F- or 125I-radiolabelled nucleoside analogues and thereby cause them to become entrapped within the host cell, to enable subsequent PET imaging [178]. A range of fluorinated 2-nitroimidazole probes that similarly become cell-entrapped upon nitro-reduction are independently in clinical development for detection of tumour hypoxia, including EF5 [2-(2-nitro-(1)H-imidazol-1-yl)-N-(2,2,3,3,3-pentafluoropropyl)-acetamide] [182], HX4 (flortanidazole) [183] and F-MISO (fluoro-misonidazole) [184]. Although E. coli NfsB has long been known to be fairly ineffective in reducing 2-nitroimidazole compounds (e.g. misonidazole [142]), we have recently shown that certain members of the NfsA family are far more effective than even endogenous human type II reductases under hypoxia in bioreductive activation of 2-nitroimidazole PET-capable probes (S.P. Syddall, J.N. Copp, C.P. Guise, E.M. Williams, A.M. Mowday, J. Theys, L. Dubois, P. Lambin, A. Ashoorzadeh, J.B. Smaill, D.F. Ackerley and A.V. Patterson, unpublished work). This offers prospects for repurposing clinically advanced hypoxia PET probes for non-invasive imaging of vectors armed with nfsA family genes, although this will of course require that the background signal due to hypoxia not be confounding. An alternative approach adopted by Paulmurugan and colleagues [185] at Stanford University was to generate a bi-functional E. coli NfsB-HSV-tk fusion protein, which they evaluated in stably transfected MDA-MB-231 triple-negative breast cancer lung-metastatic lesions in a mouse model. The fusion protein not only enabled combined treatment with CB1954 and ganciclovir, but also microPET/CT imaging via systemic infusion of [18F] FHBG (9-(4-[18F]-Fluoro-3-[hydroxymethyl]butyl)guanine; an HSV-tk responsive PET probe). This non-invasive imaging capability revealed a decrease in uptake of [18F] FHBG in tumours in the treated group compared with an increase in [18F] FHBG uptake in the control group, with tumour volume measurements confirming regression in the treated group and growth in the control group. The fusion protein strategy has the advantage of enabling synergistic therapeutic effects to be pursued via simultaneous HSV-tk and nitroreductase GDEPT and potentially simplifies the non-invasive imaging of nitroreductase-based therapies by providing access to existing imaging modalities and compounds.

Whereas nuclear imaging is likely to be necessary for clinical applications, alternative optical imaging technologies have been established to facilitate the pre-clinical development of nitroreductase GDEPT. These include a large number of bespoke nitro-quenched fluorescent probes that have been used to successfully monitor transgenic nitroreductase expression both in vitro and in small animal studies [186191]. To be suitable for in vivo imaging, fluorescent probes should emit light in the red or more preferably near-infrared (650–900 nm) spectrum, as these light frequencies are better able to penetrate tissue and avoid much of the background signal generated by tissue auto-fluorescence at shorter wavelengths [192]. Two commercially available probes that can achieve this have been used with success to monitor nitroreductase expression in various murine GDEPT models: CytoCy5S (developed by GE Healthcare) [171,185,193] and CNOB (Life Technologies) [89,162], although, as noted in the previous section, CNOB also becomes cytotoxic upon activation to MCHB [162], so may not be ideal for all imaging applications. Finally, luciferin-based bioluminescent probes have been developed that can be activated by endogenous nitroreductase activity in live bacterial strains [194], E. coli NfsB expressed in cultured human cells [195] or, of particular relevance to GDEPT, NfsB-expressing tumour xenografts in mice [196]. Bioluminescence imaging avoids the issues of photobleaching and background autofluorescence that can confound fluorescent imaging methods and can be extremely sensitive, with bioluminescent signals having been used to successfully track bacterial tumour colonization in small animal models [197,198].

Prosthesis stabilization

As of 2008 there were approximately 1 million hip replacements worldwide per year; of those, it was estimated that approximately 10% of patients would experience prosthetic loosening within 10 years [199]. Revision surgery entails removing the inflammatory interface tissue that causes the loosening followed by replacement of the entire prosthetic, a surgery with high mortality and morbidity rates. Nitroreductase GDEPT, using a non-replicating vector, was proposed as an alternative means for removing the interface tissue; to be followed by injections of bone cement to stabilize the existing prosthesis [200]. This strategy was tested in a Phase I clinical trial with 12 co-morbidity prosthesis patients, using the same E. coli nfsB-armed attenuated adenoviral vector (CTL102) as the earlier human liver cancer trial [100], followed by administration of 16 mg/m2 CB1954, both vector and prodrug being injected directly into the joint cavity [199]. Radiography showed that subsequent injections of bone cement could easily spread into the periprosthetic space to stabilize the prosthetic and patients exhibited improvement in hip function, pain levels, walking distance and independence during 6 months of post-treatment observation [199]. There was no dose-limiting toxicity associated with CB1954 administration; however, nine out of 12 patients experienced adverse GI events and eight patients had hepatic adverse events [201]. Given the similarities of the therapeutic approach with the vascular-targeting GDEPT strategy discussed earlier [164], it seems likely that outcomes could be improved by replacing CB1954 with the nil-bystander prodrug metronidazole or by substituting E. coli NfsB with a more efficient nitroreductase that exclusively generates the low-bystander 4-hydroxylamine reduction product of CB1954, such as B. subtilis YfkO [128].

Oncogene silencing

An exciting new direction for nitroreductase gene therapies has potentially been indicated in the development of 4-nitrobenzyl-caged morpholino oligonucleotides that can be activated by E. coli NfsB [202]. Morpholino oligonucleotides are nuclease-resistant single-stranded nucleic acid analogues that have been used to silence gene expression in a wide range of vertebrate and invertebrate models [203207]. Morpholino oligonucleotides knock down gene expression by interfering with specific mRNA sequences, either by preventing RNA splicing or by blocking translation [203]. ‘Caged’ morpholino oligonucleotides have essentially been converted into a prodrug form, through inactivation via a circularization reaction that can be reversed in response to a stimulus, such as light [208,209]. Yamazoe et al. [202] discovered that morpholino oligonucleotides circularized using a 4-nitrobenzyl linker similar to that used in design of hypoxia-activated prodrugs [155] became uncaged following NfsB-catalysed reduction of the linker. Proof-of-principle was demonstrated in transgenic zebrafish expressing E. coli nfsB (fused to a CFP reporter gene) under control of an insulin promoter. Post-addition of caged morpholino oligonucleotides targeting two key transcription factors, precise ablation of the cyan-fluorescent pancreatic β-cells was observed [202].

Caged nitroreductase-activated morpholino oligonucleotides not only offer exciting prospects for the in-depth study of tissue-specific gene expression, but could also potentially replace or complement nitroaromatic prodrugs in GDEPT. For example, linear morpholinos have been created against oncogenes such c-myc [210,211] that inhibits tumour growth in xenograft models. Phase I clinical trials have determined safe doses of the morpholino oligonucleotides in humans, although some toxic side effects have been observed [210,212,213]. Tumour cells expressing a vector-integrated nitroreductase could therefore be targeted for specific gene silencing, potentially enhancing GDEPT efficacy.

Alternative applications of nitroreductase–prodrug pairings

Much as caged morpholino oligonucleotides have been employed for nitroreductase-mediated cellular ablation in zebrafish [202], genotoxic prodrugs have been used to achieve the same end. The two formative studies in this area also targeted pancreatic β-cells, by creating transgenic zebrafish in which an insulin promoter-controlled expression of E. coli nfsB fused to a fluorescent reporter gene [214,215] (indeed one of these teams, Curado et al. [214], generated the same transgenic zebrafish line that was used later by Yamazoe et al. [202], above). In these studies, however, metronidazole was used to achieve cellular ablation and it was confirmed in each case that this enabled precise ablation without discernible damage to neighbouring cell populations.

The transgenic nitroreductase/metronidazole partnership therefore provides a powerful targeted cellular ablation system to enable cell-specific physiology or regeneration studies and has received widespread uptake [216]. However, as with CB1954, E. coli NfsB is relatively inefficient in catalysing the bioreductive activation of metronidazole [128]. This is a serious problem for targeted ablation in zebrafish, in particular for studies that span multiple days, as near-lethal doses of metronidazole are required (typically 10 mM is administered to the medium, with lethality observed at 15–20 mM) [217]. The high doses of metronidazole not only cause systemic toxicity, including off-target apoptosis [217] but also can promote cellular proliferation and disrupt endocrine signalling, effects that could easily prove confounding [218]. Thus, superior enzymes and/or prodrugs would probably provide substantial benefit and the general biodiscovery and engineering strategies covered above could readily be applied to advance this field as well. Indeed, Mumm and colleagues [217] at Johns Hopkins University have already shown that the top CB1954-activating triple mutant of E. coli NfsB (T41Q/N71S/F124T) evolved by Searle and coworkers [124] is also enhanced in its efficiency with metronidazole as a substrate. When expressed from spinal motor neurons, the mutant NfsB enzyme increased sensitivity to both CB1954 and metronidazole and significant levels of apoptosis in the target cell population were now achievable with a much shorter challenge period [217]. Nitroreductase enzymes selected or engineered specifically for superior metronidazole activation will probably provide even greater benefit.

CONCLUDING REMARKS

The last decade has seen substantial advances in the pre-clinical development of nitroreductase GDEPT. The tangible sense of excitement evident in the field today owes much to the early research conducted with the combination of E. coli NfsB and CB1954, which was in turn launched by the tantalizing ability of CB1954 to independently affect a complete cure in Walker 256 rats. However, it could also reasonably be argued that an overly exclusive focus on E. coli NfsB and CB1954 has ultimately restricted progress. It is our opinion that the most exciting recent advances in the field, other than the development of promising next-generation tumour-targeting vectors (a topic that lies beyond the scope of the present review), have arisen from movements away from the NfsB/CB1954 paradigm. These have included: (1) the potential to leverage the independent development of hypoxia-activated nitroaromatic prodrugs; (2) prospects for non-invasive imaging via a similar strategy of repurposing fluorinated nitroimidazole probes originally developed for PET imaging of hypoxia; (3) the discovery of novel nitroreductases that are far superior to E. coli NfsB in converting these next-generation substrates; and (4) the development of enzyme engineering strategies to maximize the desirable activities of lead enzyme candidates for GDEPT. In particular, this latter aspect offers scope for turning the traditional drug development paradigm on its head, by first embarking on a medicinal chemistry campaign to create a ‘perfect’ prodrug that exhibits optimal physicochemical properties and then tailoring enzyme activity to generate an efficient biocatalyst for conversion of that prodrug. Clearly, multiple roadblocks remain to be surmounted before this pre-clinical promise can be translated into an effective clinical therapy. Nonetheless, there is every reason to believe this can be achieved.

FUNDING

This research has been supported by the Cancer Society of New Zealand (contract 13/01 to D.F.A., J.B.S. and A.V.P.) and the Health Research Council of New Zealand (contract 14-289 to A.V.P., J.B.S. and D.F.A.). MHR is supported by a Cancer Society of New Zealand PhD Scholarship.

Abbreviations

     
  • ADEPT

    antibody-directed enzyme prodrug therapy

  •  
  • AKR1C3

    aldo-keto reductase 1C3

  •  
  • AUC

    area under the curve

  •  
  • BDEPT

    bacterial DEPT

  •  
  • CBI

    chloromethylbenzindoline

  •  
  • CDEPT

    Clostridium DEPT

  •  
  • CNOB

    6-chloro-9-nitro-5-oxo-5H-benzo(a)phenoxazine

  •  
  • DEPT

    directed enzyme prodrug therapy

  •  
  • DNBM

    dinitrobenzamide mustard

  •  
  • EGFR

    epidermal growth factor receptor

  •  
  • GDEPT

    gene-directed enzyme prodrug therapy

  •  
  • GI

    gastrointestinal

  •  
  • HSV-tk

    herpes simplex virus thymidine kinase

  •  
  • MCHB

    9-amino-6-chloro-5H-benzo(a)phenoxazine-5-one

  •  
  • MTD

    maximum tolerated dose

  •  
  • NAT2

    N-acetyltransferase 2

  •  
  • NQO1

    NAD(P)H quinone oxidoreductase 1

  •  
  • PET

    positron emission tomography

  •  
  • PSA

    prostate-specific antigen

  •  
  • VDEPT

    viral DEPT

References

References
1
Chabner
B.A.
Roberts
T.G.
Timeline: Chemotherapy and the war on cancer
Nat. Rev. Cancer
2005
, vol. 
5
 (pg. 
65
-
72
)
[PubMed]
2
DeVita
V.T.
Jr
Chu
E.
A history of cancer chemotherapy
Cancer Res.
2008
, vol. 
68
 (pg. 
8643
-
8653
)
[PubMed]
3
Chatelut
E.
Delord
J.P.
Canal
P.
Toxicity patterns of cytotoxic drugs
Invest. New Drugs
2003
, vol. 
21
 (pg. 
141
-
148
)
[PubMed]
4
Hanahan
D.
Weinberg
R.A.
Hallmarks of cancer: the next generation
Cell
2011
, vol. 
144
 (pg. 
646
-
674
)
[PubMed]
5
Burotto
M.
Manasanch
E.E.
Wilkerson
J.
Fojo
T.
Gefitinib and erlotinib in metastatic non-small cell lung cancer: a meta-analysis of toxicity and efficacy of randomized clinical trials
Oncologist
2015
, vol. 
20
 (pg. 
400
-
410
)
[PubMed]
6
Garraway
L.A.
Jänne
P.A.
Circumventing cancer drug resistance in the era of personalized medicine
Cancer Discov.
2012
, vol. 
2
 (pg. 
214
-
226
)
[PubMed]
7
Szakács
G.
Paterson
J.K.
Ludwig
J.A.
Booth-Genthe
C.
Gottesman
M.M.
Targeting multidrug resistance in cancer
Nat. Rev. Drug Discov.
2006
, vol. 
5
 (pg. 
219
-
234
)
[PubMed]
8
Holohan
C.
Van Schaeybroeck
S.
Longley
D.B.
Johnston
P.G.
Cancer drug resistance: an evolving paradigm
Nat. Rev. Cancer
2013
, vol. 
13
 (pg. 
714
-
726
)
[PubMed]
9
Brown
J.M.
Giaccia
A.J.
The unique physiology of solid tumors: opportunities (and problems) for cancer therapy
Cancer Res.
1998
, vol. 
58
 (pg. 
1408
-
1416
)
[PubMed]
10
Bagshawe
K.D.
Antibody-directed enzyme prodrug therapy (ADEPT) for cancer
Expert Rev. Anticancer Ther.
2006
, vol. 
6
 (pg. 
1421
-
1431
)
[PubMed]
11
Gwenin
V.V.
Gwenin
C.D.
Kalaji
M.
Colloidal gold modified with a genetically engineered nitroreductase: toward a novel enzyme delivery system for cancer prodrug therapy
Langmuir
2011
, vol. 
27
 (pg. 
14300
-
14307
)
[PubMed]
12
Denny
W.A.
Tumor-activated prodrugs–a new approach to cancer therapy
Cancer Invest.
2004
, vol. 
22
 (pg. 
604
-
619
)
[PubMed]
13
Karjoo
Z.
Chen
X.
Hatefi
A.
Progress and problems with the use of suicide genes for targeted cancer therapy
Adv. Drug Deliv. Rev.
2015
14
Johnson
L.
Shen
A.
Boyle
L.
Kunich
J.
Pandey
K.
Lemmon
M.
Hermiston
T.
Giedlin
M.
McCormick
F.
Fattaey
A.
Selectively replicating adenoviruses targeting deregulated E2F activity are potent, systemic antitumor agents
Cancer Cell
2002
, vol. 
1
 (pg. 
325
-
337
)
[PubMed]
15
Kubiak
A.M.
Minton
N.P.
The potential of clostridial spores as therapeutic delivery vehicles in tumour therapy
Res. Microbiol.
2015
, vol. 
166
 (pg. 
244
-
254
)
[PubMed]
16
Saukkonen
K.
Hemminki
A.
Tissue-specific promoters for cancer gene therapy
Expert Opin. Biol. Ther.
2004
, vol. 
4
 (pg. 
683
-
696
)
[PubMed]
17
Lehouritis
P.
Springer
C.
Tangney
M.
Bacterial-directed enzyme prodrug therapy
J. Control. Release
2013
, vol. 
170
 (pg. 
120
-
131
)
[PubMed]
18
Forbes
N.S.
Engineering the perfect (bacterial) cancer therapy
Nat. Rev. Cancer
2010
, vol. 
10
 (pg. 
785
-
794
)
[PubMed]
19
Cattaneo
R.
Miest
T.
Shashkova
E.V.
Barry
M.A.
Reprogrammed viruses as cancer therapeutics: targeted, armed and shielded
Nat. Rev. Microbiol.
2008
, vol. 
6
 (pg. 
529
-
540
)
[PubMed]
20
Dachs
G.U.
Hunt
M.A.
Syddall
S.
Singleton
D.C.
Patterson
A.V.
Bystander or no bystander for gene directed enzyme prodrug therapy
Molecules
2009
, vol. 
14
 (pg. 
4517
-
4545
)
[PubMed]
21
Freeman
S.M.
Abboud
C.N.
Whartenby
K.A.
Packman
C.H.
Koeplin
D.S.
Moolten
F.L.
Abraham
G.N.
The “bystander effect”: tumor regression when a fraction of the tumor mass is genetically modified
Cancer Res.
1993
, vol. 
53
 (pg. 
5274
-
5283
)
[PubMed]
22
Portsmouth
D.
Hlavaty
J.
Renner
M.
Suicide genes for cancer therapy
Mol. Aspects Med.
2007
, vol. 
28
 (pg. 
4
-
41
)
[PubMed]
23
Denny
W.A.
Prodrug strategies in cancer therapy
Eur. J. Med. Chem.
2001
, vol. 
36
 pg. 
577
 
[PubMed]
24
Moolten
F.L.
Tumor chemosensitivity conferred by inserted herpes thymidine kinase genes: paradigm for a prospective cancer control strategy
Cancer Res.
1986
, vol. 
46
 (pg. 
5276
-
5281
)
[PubMed]
25
Duarte
S.
Carle
G.
Faneca
H.
de Lima
M.C.
Pierrefite-Carle
V.
Suicide gene therapy in cancer: where do we stand now?
Cancer Lett.
2012
, vol. 
324
 (pg. 
160
-
170
)
[PubMed]
26
Faulds
D.
Heel
R.C.
Ganciclovir. A review of its antiviral activity, pharmacokinetic properties and therapeutic efficacy in cytomegalovirus infections
Drugs
1990
, vol. 
39
 (pg. 
597
-
638
)
[PubMed]
27
Rainov
N.G.
A phase III clinical evaluation of herpes simplex virus type 1 thymidine kinase and ganciclovir gene therapy as an adjuvant to surgical resection and radiation in adults with previously untreated glioblastoma multiforme
Hum. Gene Ther.
2000
, vol. 
11
 (pg. 
2389
-
2401
)
[PubMed]
28
Caruso
M.
Gene therapy against cancer and HIV infection using the gene encoding herpes simplex virus thymidine kinase
Mol. Med. Today
1996
, vol. 
2
 (pg. 
212
-
217
)
[PubMed]
29
Immonen
A.
Vapalahti
M.
Tyynelä
K.
Hurskainen
H.
Sandmair
A.
Vanninen
R.
Langford
G.
Murray
N.
Ylä-Herttuala
S.
AdvHSV-tk gene therapy with intravenous ganciclovir improves survival in human malignant glioma: a randomised, controlled study
Mol. Ther.
2004
, vol. 
10
 (pg. 
967
-
972
)
[PubMed]
30
Jacobs
A.
Voges
J.
Reszka
R.
Lercher
M.
Gossmann
A.
Kracht
L.
Kaestle
C.
Wagner
R.
Wienhard
K.
Heiss
W.
Positron-emission tomography of vector-mediated gene expression in gene therapy for gliomas
Lancet
2001
, vol. 
358
 (pg. 
727
-
729
)
[PubMed]
31
Peñuelas
I.
Mazzolini
G.
Boán
J.F.
Sangro
B.
Martí-Climent
J.
Ruiz
M.
Ruiz
J.
Satyamurthy
N.
Qian
C.
Barrio
J.R.
, et al. 
Positron emission tomography imaging of adenoviral-mediated transgene expression in liver cancer patients
Gastroenterology
2005
, vol. 
128
 (pg. 
1787
-
1795
)
[PubMed]
32
Yaghoubi
S.S.
Jensen
M.C.
Satyamurthy
N.
Budhiraja
S.
Paik
D.
Czernin
J.
Gambhir
S.S.
Noninvasive detection of therapeutic cytolytic T cells with 18F-FHBG PET in a patient with glioma
Nat. Clin. Pract. Oncol.
2009
, vol. 
6
 (pg. 
53
-
58
)
[PubMed]
33
Mullen
C.A.
Kilstrup
M.
Blaese
R.M.
Transfer of the bacterial gene for cytosine deaminase to mammalian cells confers lethal sensitivity to 5-fluorocytosine: a negative selection system
Proc. Natl. Acad. Sci. U.S.A.
1992
, vol. 
89
 (pg. 
33
-
37
)
[PubMed]
34
Longley
D.B.
Harkin
D.P.
Johnston
P.G.
5-fluorouracil: mechanisms of action and clinical strategies
Nat. Rev. Cancer
2003
, vol. 
3
 (pg. 
330
-
338
)
[PubMed]
35
Pandha
H.S.
Martin
L.A.
Rigg
A.
Hurst
H.C.
Stamp
G.W.
Sikora
K.
Lemoine
N.R.
Genetic prodrug activation therapy for breast cancer: a phase I clinical trial of erbB-2-directed suicide gene expression
J. Clin. Oncol.
1999
, vol. 
17
 (pg. 
2180
-
2189
)
[PubMed]
36
Nemunaitis
J.
Cunningham
C.
Senzer
N.
Kuhn
J.
Cramm
J.
Litz
C.
Cavagnolo
R.
Cahill
A.
Clairmont
C.
Sznol
M.
Pilot trial of genetically modified, attenuated Salmonella expressing the E. coli cytosine deaminase gene in refractory cancer patients
Cancer Gene Ther.
2003
, vol. 
10
 (pg. 
737
-
744
)
[PubMed]
37
Freytag
S.O.
Stricker
H.
Lu
M.
Elshaikh
M.
Aref
I.
Pradhan
D.
Levin
K.
Kim
J.H.
Peabody
J.
Siddiqui
F.
, et al. 
Prospective randomized phase 2 trial of intensity modulated radiation therapy with or without oncolytic adenovirus-mediated cytotoxic gene therapy in intermediate-risk prostate cancer
Int. J. Radiat. Oncol. Biol. Phys.
2014
, vol. 
89
 (pg. 
268
-
276
)
[PubMed]
38
Lu
M.
Freytag
S.O.
Stricker
H.
Kim
J.H.
Barton
K.
Movsas
B.
Adaptive seamless design for an efficacy trial of replication-competent adenovirus-mediated suicide gene therapy and radiation in newly-diagnosed prostate cancer (ReCAP Trial)
Contemp. Clin. Trials
2011
, vol. 
32
 (pg. 
453
-
460
)
[PubMed]
39
Denny
W.A.
Nitroreductase-based GDEPT
Curr. Pharm. Des.
2002
, vol. 
8
 (pg. 
1349
-
1361
)
[PubMed]
40
Patterson
A.V.
Saunders
M.P.
Greco
O.
Prodrugs in genetic chemoradiotherapy
Curr. Pharm. Des.
2003
, vol. 
9
 (pg. 
2131
-
2154
)
[PubMed]
41
Denny
W.A.
Wilson
W.R.
Bioreducible mustards: a paradigm for hypoxia-selective prodrugs of diffusible cytotoxins (HPDCs)
Cancer Metastasis Rev.
1993
, vol. 
12
 (pg. 
135
-
151
)
[PubMed]
42
Wilson
W.R.
Hicks
K.O.
Pullen
S.M.
Ferry
D.M.
Helsby
N.A.
Patterson
A.V.
Bystander effects of bioreductive drugs: potential for exploiting pathological tumor hypoxia with dinitrobenzamide mustards
Radiat. Res.
2007
, vol. 
167
 (pg. 
625
-
636
)
[PubMed]
43
Mesnil
M.
Yamasaki
H.
Bystander effect in herpes simplex virus-thymidine kinase/ganciclovir cancer gene therapy: role of gap-junctional intercellular communication
Cancer Res.
2000
, vol. 
60
 (pg. 
3989
-
3999
)
[PubMed]
44
Trosko
J.E.
The role of stem cells and gap junctional intercellular communication in carcinogenesis
J. Biochem. Mol. Biol.
2003
, vol. 
36
 (pg. 
43
-
48
)
[PubMed]
45
Roldán
M.D.
Pérez-Reinado
E.
Castillo
F.
Moreno-Vivián
C.
Reduction of polynitroaromatic compounds: the bacterial nitroreductases
FEMS Microbiol. Rev.
2008
, vol. 
32
 (pg. 
474
-
500
)
[PubMed]
46
Peterson
F.J.
Mason
R.P.
Hovsepian
J.
Holtzman
J.L.
Oxygen-sensitive and -insensitive nitroreduction by Escherichia coli and rat hepatic microsomes
J. Biol. Chem.
1979
, vol. 
254
 (pg. 
4009
-
4014
)
[PubMed]
47
Bryant
C.
DeLuca
M.
Purification and characterization of an oxygen-insensitive NAD(P)H nitroreductase from Enterobacter cloacae
J. Biol. Chem.
1991
, vol. 
266
 (pg. 
4119
-
4125
)
[PubMed]
48
Siim
B.G.
Denny
W.A.
Wilson
W.R.
Nitro reduction as an electronic switch for bioreductive drug activation
Oncol. Res.
1997
, vol. 
9
 (pg. 
357
-
369
)
[PubMed]
49
Kobori
T.
Sasaki
H.
Lee
W.C.
Zenno
S.
Saigo
K.
Murphy
M.E.
Tanokura
M.
Structure and site-directed mutagenesis of a flavoprotein from Escherichia coli that reduces nitrocompounds: alteration of pyridine nucleotide binding by a single amino acid substitution
J. Biol. Chem.
2001
, vol. 
276
 (pg. 
2816
-
2823
)
[PubMed]
50
Parkinson
G.N.
Skelly
J.V.
Neidle
S.
Crystal structure of FMN-dependent nitroreductase from Escherichia coli B: a prodrug-activating enzyme
J. Med. Chem.
2000
, vol. 
43
 (pg. 
3624
-
3631
)
[PubMed]
51
Johansson
E.
Parkinson
G.N.
Denny
W.A.
Neidle
S.
Studies on the nitroreductase prodrug-activating system. Crystal structures of complexes with the inhibitor dicoumarol and dinitrobenzamide prodrugs and of the enzyme active form
J. Med. Chem.
2003
, vol. 
46
 (pg. 
4009
-
4020
)
[PubMed]
52
Race
P.R.
Lovering
A.L.
Green
R.M.
Ossor
A.
White
S.A.
Searle
P.F.
Wrighton
C.J.
Hyde
E.I.
Structural and mechanistic studies of Escherichia coli nitroreductase with the antibiotic nitrofurazone: reversed binding orientations in different redox states of the enzyme
J. Biol. Chem.
2005
, vol. 
280
 (pg. 
13256
-
13264
)
[PubMed]
53
Lovering
A.L.
Hyde
E.I.
Searle
P.F.
White
S.A.
The structure of Escherichia coli nitroreductase complexed with nicotinic acid: three crystal forms at 1.7 Å, 1.8 Å and 2.4 Å resolution
J. Mol. Biol.
2001
, vol. 
309
 (pg. 
203
-
213
)
[PubMed]
54
Ito
K.
Nakanishi
M.
Lee
W.C.
Sasaki
H.
Zenno
S.
Saigo
K.
Kitade
Y.
Tanokura
M.
Three-dimensional structure of AzoR from Escherichia coli: an oxidoreductase conserved in microorganisms
J. Biol. Chem.
2006
, vol. 
281
 (pg. 
20567
-
20576
)
[PubMed]
55
Adams
M.A.
Jia
Z.
Modulator of drug activity B from Escherichia coli: crystal structure of a prokaryotic homologue of DT-diaphorase
J. Mol. Biol.
2006
, vol. 
359
 (pg. 
455
-
465
)
[PubMed]
56
AbuKhader
M.M.
Heap
J.
Matteis
C.I.
Doughty
S.W.
Minton
N.
Paoli
M.
Crystallization and preliminary X-ray characterization of the Bacillus amyloliquefaciens YwrO enzyme
Acta Crystallograph. Sect. F Struct. Biol. Cryst. Commun.
2007
, vol. 
63
 (pg. 
746
-
750
)
[PubMed]
57
Haynes
C.A.
Koder
R.L.
Miller
A.F.
Rodgers
D.W.
Structures of nitroreductase in three states: effects of inhibitor binding and reduction
J. Biol. Chem.
2002
, vol. 
277
 (pg. 
11513
-
11520
)
[PubMed]
58
Koder
R.L.
Haynes
C.A.
Rodgers
M.E.
Rodgers
D.W.
Miller
A.-F.
Flavin thermodynamics explain the oxygen insensitivity of enteric nitroreductases
Biochemistry
2002
, vol. 
41
 (pg. 
14197
-
14205
)
[PubMed]
59
Ju
K.S.
Parales
R.E.
Nitroaromatic compounds, from synthesis to biodegradation
Microbiol. Mol. Biol. Rev.
2010
, vol. 
74
 (pg. 
250
-
272
)
[PubMed]
60
Zenno
S.
Saigo
K.
Kanoh
H.
Inouye
S.
Identification of the gene encoding the major NAD(P)H-flavin oxidoreductase of the bioluminescent bacterium Vibrio fischeri ATCC 7744
J. Bacteriol.
1994
, vol. 
176
 (pg. 
3536
-
3543
)
[PubMed]
61
Green
L.K.
Flamme
A.C.
Ackerley
D.F.
Pseudomonas aeruginosa MdaB and WrbA are water-soluble two-electron quinone oxidoreductases with the potential to defend against oxidative stress
J. Microbiol.
2014
, vol. 
52
 (pg. 
771
-
777
)
[PubMed]
62
Wang
G.
Maier
R.J.
An NADPH quinone reductase of Helicobacter pylori plays an important role in oxidative stress resistance and host colonization
Infect. Immun.
2004
, vol. 
72
 (pg. 
1391
-
1396
)
[PubMed]
63
Ackerley
D.F.
Gonzalez
C.F.
Keyhan
M.
Blake
R.
Matin
A.
Mechanism of chromate reduction by the Escherichia coli protein, NfsA, and the role of different chromate reductases in minimizing oxidative stress during chromate reduction
Environ. Microbiol.
2004
, vol. 
6
 (pg. 
851
-
860
)
[PubMed]
64
Robins
K.J.
Hooks
D.O.
Rehm
B.H.A.
Ackerley
D.F.
Escherichia coli NemA is an efficient chromate reductase that can be biologically immobilized to provide a cell free system for remediation of hexavalent chromium
PLoS One
2013
, vol. 
8
 pg. 
e59200
 
[PubMed]
65
Takeda
K.
Sato
J.
Goto
K.
Fujita
T.
Watanabe
T.
Abo
M.
Yoshimura
E.
Nakagawa
J.
Abe
A.
Kawasaki
S.
Niimura
Y.
Escherichia coli ferredoxin-NADP+ reductase and oxygen-insensitive nitroreductase are capable of functioning as ferric reductase and of driving the Fenton reaction
Biometals
2010
, vol. 
23
 (pg. 
727
-
737
)
[PubMed]
66
Rafii
F.
Cerniglia
C.E.
Reduction of azo dyes and nitroaromatic compounds by bacterial enzymes from the human intestinal tract
Environ. Health Perspect.
1995
, vol. 
103
 
Suppl. 5
(pg. 
17
-
19
)
[PubMed]
67
Pitsawong
W.
Hoben
J. P.
Miller
A. F.
Understanding the broad substrate repertoire of nitroreductase based on its kinetic mechanism
J. Biol. Chem.
2014
, vol. 
289
 (pg. 
15203
-
15214
)
[PubMed]
68
Khan
A.H.
Ross
W.C.
Tumour-growth inhibitory nitrophenylaziridines and related compounds: structure-activity relationships
Chem. Biol. Interact.
1969
, vol. 
1
 (pg. 
27
-
47
)
[PubMed]
69
Cobb
L.M.
Connors
T.A.
Elson
L.A.
Khan
A.H.
Mitchley
B.C.
Ross
W.C.
Whisson
M.E.
2,4-dinitro-5-ethyleneiminobenzamide (CB 1954): a potent and selective inhibitor of the growth of the Walker carcinoma 256
Biochem. Pharmacol.
1969
, vol. 
18
 (pg. 
1519
-
1527
)
[PubMed]
70
Knox
R.J.
Burke
P.J.
Chen
S.
Kerr
D.J.
CB 1954: from the Walker tumor to NQO2 and VDEPT
Curr. Pharm. Des.
2003
, vol. 
9
 (pg. 
2091
-
2104
)
[PubMed]
71
Knox
R.J.
Boland
M.P.
Friedlos
F.
Coles
B.
Southan
C.
Roberts
J.J.
The nitroreductase enzyme in walker cells that activates 5-(aziridin-1-yl)-2,4-dinitrobenzamide (CB 1954) to 5-(aziridin-1-yl)-4-hydroxylamino-2-nitrobenzamide is a form of NAD(P)H dehydrogenase (quinone) (EC 1.6.99.2)
Biochem. Pharmacol.
1988
, vol. 
37
 (pg. 
4671
-
4677
)
[PubMed]
72
Knox
R.J.
Friedlos
F.
Marchbank
T.
Roberts
J.J.
Bioactivation of CB 1954: reaction of the active 4-hydroxylamino derivative with thioesters to form the ultimate DNA-DNA interstrand crosslinking species
Biochem. Pharmacol.
1991
, vol. 
42
 (pg. 
1691
-
1697
)
[PubMed]
73
Siegel
D.
Ross
D.
Immunodetection of NAD(P)H:quinone oxidoreductase 1 (NQO1) in human tissues
Free Radic. Biol. Med.
2000
, vol. 
29
 (pg. 
246
-
253
)
[PubMed]
74
Knox
R.J.
Lydall
D.A.
Friedlos
F.
Basham
C.
Rawlings
C.J.
Roberts
J.J.
The Walker 256 carcinoma: a cell type inherently sensitive only to those difunctional agents that can form DNA interstrand crosslinks
Mutat. Res.
1991
, vol. 
255
 (pg. 
227
-
240
)
[PubMed]
75
Boland
M.P.
Knox
R.J.
Roberts
J.J.
The differences in kinetics of rat and human DT diaphorase result in a differential sensitivity of derived cell lines to CB 1954 (5-(aziridin-1-yl)-2,4-dinitrobenzamide)
Biochem. Pharmacol.
1991
, vol. 
41
 (pg. 
867
-
875
)
[PubMed]
76
Chen
S.
Knox
R.
Lewis
A.D.
Friedlos
F.
Workman
P.
Deng
P.S.
Fung
M.
Ebenstein
D.
Wu
K.
Tsai
T.M.
Catalytic properties of NAD(P)H:quinone acceptor oxidoreductase: study involving mouse, rat, human, and mouse-rat chimeric enzymes
Mol. Pharmacol.
1995
, vol. 
47
 (pg. 
934
-
939
)
[PubMed]
77
Chen
S.
Knox
R.
Wu
K.
Deng
P.S.
Zhou
D.
Bianchet
M.A.
Amzel
L.M.
Molecular basis of the catalytic differences among DT-diaphorase of human, rat, and mouse
J. Biol. Chem.
1997
, vol. 
272
 (pg. 
1437
-
1439
)
[PubMed]
78
Anlezark
G.M.
Melton
R.G.
Sherwood
R.F.
Coles
B.
Friedlos
F.
Knox
R.J.
The bioactivation of 5-(aziridin-1-yl)-2,4-dinitrobenzamide (CB1954)–I. Purification and properties of a nitroreductase enzyme from Escherichia coli–a potential enzyme for antibody-directed enzyme prodrug therapy (ADEPT)
Biochem. Pharmacol.
1992
, vol. 
44
 (pg. 
2289
-
2295
)
[PubMed]
79
Knox
R.J.
Friedlos
F.
Sherwood
R.F.
Melton
R.G.
Anlezark
G.M.
The bioactivation of 5-(aziridin-1-yl)-2,4-dinitrobenzamide (CB1954)—II: a comparison of an Escherichia coli nitroreductase and Walker DT diaphorase
Biochem. Pharmacol.
1992
, vol. 
44
 (pg. 
2297
-
2301
)
[PubMed]
80
Knox
R.J.
Friedlos
F.
Jarman
M.
Roberts
J.J.
A new cytotoxic, DNA interstrand crosslinking agent, 5-(aziridin-1-yl)-4-hydroxylamino-2-nitrobenzamide, is formed from 5-(aziridin-1-yl)-2,4-dinitrobenzamide (CB 1954) by a nitroreductase enzyme in Walker carcinoma cells
Biochem. Pharmacol.
1988
, vol. 
37
 (pg. 
4661
-
4669
)
[PubMed]
81
Helsby
N.A.
Ferry
D.M.
Patterson
A.V.
Pullen
S.M.
Wilson
W.R.
2-Amino metabolites are key mediators of CB 1954 and SN 23862 bystander effects in nitroreductase GDEPT
Br. J. Cancer
2004
, vol. 
90
 (pg. 
1084
-
1092
)
[PubMed]
82
Mitchell
D.J.
Minchin
R.F.
E. coli nitroreductase/CB1954 gene-directed enzyme prodrug therapy: role of arylamine N-acetlytransferase 2
Cancer Gene Ther.
2008
, vol. 
15
 (pg. 
758
-
764
)
[PubMed]
83
Knox
R.J.
Friedlos
F.
Jarman
M.
Davies
L.C.
Goddard
P.
Anlezark
G.M.
Melton
R.G.
Sherwood
R.F.
Virtual cofactors for an Escherichia coli nitroreductase enzyme: relevance to reductively activated prodrugs in antibody directed enzyme prodrug therapy (ADEPT)
Biochem. Pharmacol.
1995
, vol. 
49
 (pg. 
1641
-
1647
)
[PubMed]
84
Braidwood
L.
Dunn
P.D.
Hardy
S.
Evans
T.R.J.
Brown
S.M.
Antitumor activity of a selectively replication competent herpes simplex virus (HSV) with enzyme prodrug therapy
Anticancer Res.
2009
, vol. 
29
 (pg. 
2159
-
2166
)
[PubMed]
85
Chen
M.J.
Green
N.K.
Reynolds
G.M.
Flavell
J.R.
Mautner
V.
Kerr
D.J.
Young
L.S.
Searle
P.F.
Enhanced efficacy of Escherichia coli nitroreductase/CB1954 prodrug activation gene therapy using an E1B-55K-deleted oncolytic adenovirus vector
Gene Ther.
2004
, vol. 
11
 (pg. 
1126
-
1136
)
[PubMed]
86
McNeish
I.A.
Green
N.K.
Gilligan
M.G.
Ford
M.J.
Mautner
V.
Young
L.S.
Kerr
D.J.
Searle
P.F.
Virus directed enzyme prodrug therapy for ovarian and pancreatic cancer using retrovirally delivered E. coli nitroreductase and CB1954
Gene Ther.
1998
, vol. 
5
 (pg. 
1061
-
1069
)
[PubMed]
87
Singleton
D.C.
Li
D.
Bai
S.Y.
Syddall
S.P.
Smaill
J.B.
Shen
Y.
Denny
W.A.
Wilson
W.R.
Patterson
A.V.
The nitroreductase prodrug SN 28343 enhances the potency of systemically administered armed oncolytic adenovirus ONYX-411NTR
Cancer Gene Ther.
2007
, vol. 
14
 (pg. 
953
-
967
)
[PubMed]
88
Weedon
S.J.
Green
N.K.
McNeish
I.A.
Gilligan
M.G.
Mautner
V.
Wrighton
C.J.
Mountain
A.
Young
L.S.
Kerr
D.J.
Searle
P.F.
Sensitisation of human carcinoma cells to the prodrug CB1954 by adenovirus vector-mediated expression of E. coli nitroreductase
Int. J. Cancer
2000
, vol. 
86
 (pg. 
848
-
854
)
[PubMed]
89
Liu
S.C.
Ahn
G.O.
Kioi
M.
Dorie
M.J.
Patterson
A.V.
Brown
J.M.
Optimized Clostridium-directed enzyme prodrug therapy improves the antitumor activity of the novel DNA cross-linking agent PR-104
Cancer Res.
2008
, vol. 
68
 (pg. 
7995
-
8003
)
[PubMed]
90
Benouchan
M.
Nascimento
F.D.
Perret
G.Y.
Colombo
B.M.
Delivery of the bacterial nitroreductase gene into endothelial cells prolongs the survival of tumour-bearing mice by bystander mechanisms
Int. J. Oncol.
2006
, vol. 
28
 (pg. 
457
-
462
)
[PubMed]
91
Djeha
A.H.
Hulme
A.
Dexter
M.T.
Mountain
A.
Young
L.S.
Searle
P.F.
Kerr
D.J.
Wrighton
C.J.
Expression of Escherichia coli B nitroreductase in established human tumor xenografts in mice results in potent antitumoral and bystander effects upon systemic administration of the prodrug CB1954
Cancer Gene Ther.
2000
, vol. 
7
 (pg. 
721
-
731
)
[PubMed]
92
Westphal
E.M.
Ge
J.
Catchpole
J.R.
Ford
M.
Kenney
S.C.
The nitroreductase/CB1954 combination in Epstein-Barr virus-positive B-cell lines: induction of bystander killing in vitro and in vivo
Cancer Gene Ther.
2000
, vol. 
7
 (pg. 
97
-
106
)
[PubMed]
93
Green
N.K.
McNeish
I.A.
Doshi
R.
Searle
P.F.
Kerr
D.J.
Young
L.S.
Immune enhancement of nitroreductase-induced cytotoxicity: studies using a bicistronic adenovirus vector
Int. J. Cancer
2003
, vol. 
104
 (pg. 
104
-
112
)
[PubMed]
94
Tang
M.H.
Helsby
N.A.
Goldthorpe
M.A.
Thompson
K.M.
Al-Ali
S.
Tingle
M.D.
Hepatic nitroreduction, toxicity and toxicokinetics of the anti-tumour prodrug CB 1954 in mouse and rat
Toxicology
2007
, vol. 
240
 (pg. 
70
-
85
)
[PubMed]
95
Kestell
P.
Pruijn
F.B.
Siim
B.G.
Palmer
B.D.
Wilson
W.R.
Pharmacokinetics and metabolism of the nitrogen mustard bioreductive drug 5-[N,N-bis(2-chloroethyl)amino]-2,4-dinitrobenzamide (SN 23862) and the corresponding aziridine (CB 1954) in KHT tumour-bearing mice
Cancer Chemother. Pharmacol.
2000
, vol. 
46
 (pg. 
365
-
374
)
[PubMed]
96
Patel
P.
Young
J.G.
Mautner
V.
Ashdown
D.
Bonney
S.
Pineda
R.G.
Collins
S.I.
Searle
P.F.
Hull
D.
Peers
E.
, et al. 
A phase I/II clinical trial in localized prostate cancer of an adenovirus expressing nitroreductase with CB1954
Mol. Ther.
2009
, vol. 
17
 (pg. 
1292
-
1299
)
[PubMed]
97
Chung-Faye
G.
Palmer
D.
Anderson
D.
Clark
J.
Downes
M.
Baddeley
J.
Hussain
S.
Murray
P.I.
Searle
P.
Seymour
L.
, et al. 
Virus-directed, enzyme prodrug therapy with nitroimidazole reductase: a phase I and pharmacokinetic study of its prodrug, CB1954
Clin. Cancer Res.
2001
, vol. 
7
 (pg. 
2662
-
2668
)
[PubMed]
98
Tang
M.H.
Helsby
N.A.
Wilson
W.R.
Tingle
M.D.
Aerobic 2- and 4-nitroreduction of CB 1954 by human liver
Toxicology
2005
, vol. 
216
 (pg. 
129
-
139
)
[PubMed]
99
Chandor
A.
Dijols
S.
Ramassamy
B.
Frapart
Y.
Mansuy
D.
Stuehr
D.
Helsby
N.
Boucher
J.L.
Metabolic activation of the antitumor drug 5-(aziridin-1-yl)-2,4-dinitrobenzamide (CB1954) by NO synthases
Chem. Res. Toxicol.
2008
, vol. 
21
 (pg. 
836
-
843
)
[PubMed]
100
Palmer
D.H.
Mautner
V.
Mirza
D.
Oliff
S.
Gerritsen
W.
van der Sijp
J.R.M.
Hubscher
S.
Reynolds
G.
Bonney
S.
Rajaratnam
R.
, et al. 
Virus-directed enzyme prodrug therapy: intratumoral administration of a replication-deficient adenovirus encoding nitroreductase to patients with resectable liver cancer
J. Clin. Oncol.
2004
, vol. 
22
 (pg. 
1546
-
1552
)
[PubMed]
101
Onion
D.
Patel
P.
Pineda
R.G.
James
N.
Mautner
V.
Antivector and tumor immune responses following adenovirus-directed enzyme prodrug therapy for the treatment of prostate cancer
Hum. Gene Ther.
2009
, vol. 
20
 (pg. 
1249
-
1258
)
[PubMed]
102
Anlezark
G.M.
Vaughan
T.
Fashola-Stone
E.
Michael
N.P.
Murdoch
H.
Sims
M.A.
Stubbs
S.
Wigley
S.
Minton
N.P.
Bacillus amyloliquefaciens orthologue of Bacillus subtilis ywrO encodes a nitroreductase enzyme which activates the prodrug CB 1954
Microbiology
2002
, vol. 
148
 (pg. 
297
-
306
)
[PubMed]
103
Theys
J.
Pennington
O.
Dubois
L.
Anlezark
G.
Vaughan
T.
Mengesha
A.
Landuyt
W.
Anné
J.
Burke
P.J.
Dûrre
P.
, et al. 
Repeated cycles of Clostridium-directed enzyme prodrug therapy result in sustained antitumour effects in vivo
Br. J. Cancer
2006
, vol. 
95
 (pg. 
1212
-
1219
)
[PubMed]
104
Race
P.R.
Lovering
A.L.
White
S.A.
Grove
J.I.
Searle
P.F.
Wrighton
C.W.
Hyde
E.I.
Kinetic and structural characterisation of Escherichia coli nitroreductase mutants showing improved efficacy for the prodrug substrate CB1954
J. Mol. Biol.
2007
, vol. 
368
 (pg. 
481
-
492
)
[PubMed]
105
Jarrom
D.
Jaberipour
M.
Guise
C.P.
Daff
S.
White
S.A.
Searle
P.F.
Hyde
E.I.
Steady-state and stopped-flow kinetic studies of three Escherichia coli NfsB mutants with enhanced activity for the prodrug CB1954
Biochemistry
2009
, vol. 
48
 (pg. 
7665
-
7672
)
[PubMed]
106
Heap
J.T.
Theys
J.
Ehsaan
M.
Kubiak
A.M.
Dubois
L.
Paesmans
K.
Mellaert
L.V.
Knox
R.
Kuehne
S.A.
Lambin
P.
Minton
N.P.
Spores of Clostridium engineered for clinical efficacy and safety cause regression and cure of tumors in vivo
Oncotarget
2014
, vol. 
5
 (pg. 
1761
-
1769
)
[PubMed]
107
Barak
Y.
Thorne
S.H.
Ackerley
D.F.
Lynch
S.V.
Contag
C.H.
Matin
A.
New enzyme for reductive cancer chemotherapy, YieF, and its improvement by directed evolution
Mol. Cancer Ther.
2006
, vol. 
5
 (pg. 
97
-
103
)
[PubMed]
108
Prosser
G.A.
Copp
J.N.
Syddall
S.P.
Williams
E.M.
Smaill
J.B.
Wilson
W.R.
Patterson
A.V.
Ackerley
D.F.
Discovery and evaluation of Escherichia coli nitroreductases that activate the anti-cancer prodrug CB1954
Biochem. Pharmacol.
2010
, vol. 
79
 (pg. 
678
-
687
)
[PubMed]
109
Vass
S.O.
Jarrom
D.
Wilson
W.R.
Hyde
E.I.
Searle
P.F.
E. coli NfsA: an alternative nitroreductase for prodrug activation gene therapy in combination with CB1954
Br. J. Cancer
2009
, vol. 
100
 (pg. 
1903
-
1911
)
[PubMed]
110
Prosser
G.A.
Copp
J.N.
Mowday
A.M.
Guise
C.P.
Syddall
S.P.
Williams
E.M.
Horvat
C.N.
Swe
P.M.
Ashoorzadeh
A.
Denny
W.A.
, et al. 
Creation and screening of a multi-family bacterial oxidoreductase library to discover novel nitroreductases that efficiently activate the bioreductive prodrugs CB1954 and PR-104A
Biochem. Pharmacol.
2013
, vol. 
85
 (pg. 
1091
-
1103
)
[PubMed]
111
Wilson
W.R.
Pullen
S.M.
Hogg
A.
Helsby
N.A.
Hicks
K.O.
Denny
W.A.
Quantitation of bystander effects in nitroreductase suicide gene therapy using three-dimensional cell cultures
Cancer Res.
2002
, vol. 
62
 (pg. 
1425
-
1432
)
[PubMed]
112
Klaidman
L.K.
Leung
A.C.
Adams
J.D.
High-performance liquid chromatography analysis of oxidized and reduced pyridine dinucleotides in specific brain regions
Anal. Biochem.
1995
, vol. 
228
 (pg. 
312
-
317
)
[PubMed]
113
London
J.
Knight
M.
Concentrations of nicotinamide nucleotide coenzymes in micro-organisms
J. Gen. Microbiol.
1966
, vol. 
44
 (pg. 
241
-
254
)
[PubMed]
114
Ying
W.
NAD+/NADH and NADP+/NADPH in cellular functions and cell death: regulation and biological consequences
Antioxid. Redox Signal.
2008
, vol. 
10
 (pg. 
179
-
206
)
[PubMed]
115
Green
L.K.
Storey
M.A.
Williams
E.M.
Patterson
A.V.
Smaill
J.B.
Copp
J.N.
Ackerley
D.F.
The flavin reductase MsuE is a novel nitroreductase that can efficiently activate two promising next-generation prodrugs for gene-directed enzyme prodrug therapy
Cancers
2013
, vol. 
5
 (pg. 
985
-
997
)
[PubMed]
116
Gillam
E.M.
Copp
J.N.
Ackerley
D.F.
Directed Evolution Library Creation: Methods and Protocols
2014
2nd edn
New York
Humana Press (Springer)
117
Tracewell
C.A.
Arnold
F.H.
Directed enzyme evolution: climbing fitness peaks one amino acid at a time
Curr. Opin. Chem. Biol.
2009
, vol. 
13
 (pg. 
3
-
9
)
[PubMed]
118
Aharoni
A.
Gaidukov
L.
Khersonsky
O.
Gould
S.
Roodveldt
C.
Tawfik
D.S.
The “evolvability” of promiscuous protein functions
Nat. Genet.
2005
, vol. 
37
 (pg. 
73
-
76
)
[PubMed]
119
Gould
S.M.
Tawfik
D.S.
Directed evolution of the promiscuous esterase activity of carbonic anhydrase II
Biochemistry
2005
, vol. 
44
 (pg. 
5444
-
5452
)
[PubMed]
120
Grove
J.I.
Lovering
A.L.
Guise
C.
Race
P.R.
Wrighton
C.J.
White
S.A.
Hyde
E.I.
Searle
P.F.
Generation of Escherichia coli nitroreductase mutants conferring improved cell sensitization to the prodrug CB1954
Cancer Res.
2003
, vol. 
63
 (pg. 
5532
-
5537
)
[PubMed]
121
Jaberipour
M.
Vass
S.O.
Guise
C.P.
Grove
J.I.
Knox
R.J.
Hu
L.
Hyde
E.I.
Searle
P.F.
Testing double mutants of the enzyme nitroreductase for enhanced cell sensitisation to prodrugs: effects of combining beneficial single mutations
Biochem. Pharmacol.
2010
, vol. 
79
 (pg. 
102
-
111
)
[PubMed]
122
Reetz
M.T.
Carballeira
J.D.
Iterative saturation mutagenesis (ISM) for rapid directed evolution of functional enzymes
Nat. Protoc.
2007
, vol. 
2
 (pg. 
891
-
903
)
[PubMed]
123
Romero
P.A.
Arnold
F.H.
Exploring protein fitness landscapes by directed evolution
Nat. Rev. Mol. Cell Biol.
2009
, vol. 
10
 (pg. 
866
-
876
)
[PubMed]
124
Guise
C.P.
Grove
J.I.
Hyde
E.I.
Searle
P.F.
Direct positive selection for improved nitroreductase variants using SOS triggering of bacteriophage lambda lytic cycle
Gene Ther.
2007
, vol. 
14
 (pg. 
690
-
698
)
[PubMed]
125
Baharoglu
Z.
Mazel
D.
SOS, the formidable strategy of bacteria against aggressions
FEMS Microbiol. Rev.
2014
, vol. 
38
 (pg. 
1126
-
1145
)
[PubMed]
126
Ackerley
D.F.
Barak
Y.
Lynch
S.V.
Curtin
J.
Matin
A.
Effect of chromate stress on Escherichia coli K-12
J. Bacteriol.
2006
, vol. 
188
 (pg. 
3371
-
3381
)
[PubMed]
127
Quillardet
P.
Huisman
O.
D'Ari
R.
Hofnung
M.
SOS chromotest, a direct assay of induction of an SOS function in Escherichia coli K-12 to measure genotoxicity
Proc. Natl. Acad. Sci. U.S.A.
1982
, vol. 
79
 (pg. 
5971
-
5975
)
[PubMed]
128
Prosser
G.A.
Patterson
A.V.
Ackerley
D.F.
uvrB gene deletion enhances SOS chromotest sensitivity for nitroreductases that preferentially generate the 4-hydroxylamine metabolite of the anti-cancer prodrug CB1954
J. Biotechnol.
2010
, vol. 
150
 (pg. 
190
-
194
)
[PubMed]
129
Swe
P.M.
Copp
J.N.
Green
L.K.
Guise
C.P.
Mowday
A.M.
Smaill
J.B.
Patterson
A.V.
Ackerley
D.F.
Targeted mutagenesis of the Vibrio fischeri flavin reductase FRaseI to improve activation of the anticancer prodrug CB1954
Biochem. Pharmacol.
2012
, vol. 
84
 (pg. 
775
-
783
)
[PubMed]
130
Renggli
S.
Keck
W.
Jenal
U.
Ritz
D.
Role of autofluorescence in flow cytometric analysis of Escherichia coli treated with bactericidal antibiotics
J. Bacteriol.
2013
, vol. 
195
 (pg. 
4067
-
4073
)
[PubMed]
131
Copp
J.N.
Williams
E.M.
Rich
M.H.
Patterson
A.V.
Smaill
J.B.
Ackerley
D.F.
Toward a high-throughput screening platform for directed evolution of enzymes that activate genotoxic prodrugs
Protein Eng. Des. Sel.
2014
, vol. 
27
 (pg. 
399
-
403
)
[PubMed]
132
Baker
S.L.
Use of a Direct, Positive Selection Strategy to Generate Improved Prodrug-Activating Enzymes for Cancer Gene Therapy
Ph.D. Thesis
2011
Birmingham, U.K.
University of Birmingham
133
Ackerley
D.F.
Gonzalez
C.F.
Park
C.H.
Blake
R.
Keyhan
M.
Matin
A.
Chromate-reducing properties of soluble flavoproteins from Pseudomonas putida and Escherichia coli
Appl. Environ. Microbiol.
2004
, vol. 
70
 (pg. 
873
-
882
)
[PubMed]
134
Dachs
G.U.
Tozer
G.M.
Hypoxia modulated gene expression: angiogenesis, metastasis and therapeutic exploitation
Eur. J. Cancer
2000
, vol. 
36
 (pg. 
1649
-
1660
)
[PubMed]
135
Brown
J.M.
Wilson
W.R.
Exploiting tumour hypoxia in cancer treatment
Nat. Rev. Cancer
2004
, vol. 
4
 (pg. 
437
-
447
)
[PubMed]
136
Guise
C.P.
Mowday
A.M.
Ashoorzadeh
A.
Yuan
R.
Lin
W.H.
Wu
D.H.
Smaill
J.B.
Patterson
A.V.
Ding
K.
Bioreductive prodrugs as cancer therapeutics: targeting tumor hypoxia
Chin. J. Cancer
2014
, vol. 
33
 (pg. 
80
-
86
)
[PubMed]
137
Patterson
A.V.
Ferry
D.M.
Edmunds
S.J.
Gu
Y.
Singleton
R.S.
Patel
K.
Pullen
S.M.
Hicks
K.O.
Syddall
S.P.
Atwell
G.J.
, et al. 
Mechanism of action and preclinical antitumor activity of the novel hypoxia-activated DNA cross-linking agent PR-104
Clin. Cancer Res.
2007
, vol. 
13
 (pg. 
3922
-
3932
)
[PubMed]
138
Mattes
W.B.
Hartley
J.A.
Kohn
K.W.
DNA sequence selectivity of guanine-N7 alkylation by nitrogen mustards
Nucleic Acids Res.
1986
, vol. 
14
 (pg. 
2971
-
2987
)
[PubMed]
139
Gilman
A.
Philips
F.S.
The biological actions and therapeutic applications of the B-chloroethyl amines and sulfides
Science
1946
, vol. 
103
 (pg. 
409
-
436
)
[PubMed]
140
Atwell
G.J.
Yang
S.
Pruijn
F.B.
Pullen
S.M.
Hogg
A.
Patterson
A.V.
Wilson
W.R.
Denny
W.A.
Synthesis and structure-activity relationships for 2,4-dinitrobenzamide-5-mustards as prodrugs for the Escherichia coli nfsB nitroreductase in gene therapy
J. Med. Chem.
2007
, vol. 
50
 (pg. 
1197
-
1212
)
[PubMed]
141
Palmer
B.D.
Wilson
W.R.
Cliffe
S.
Denny
W.A.
Hypoxia-selective antitumor agents. 5. Synthesis of water-soluble nitroaniline mustards with selective cytotoxicity for hypoxic mammalian cells
J. Med. Chem.
1992
, vol. 
35
 (pg. 
3214
-
3222
)
[PubMed]
142
Anlezark
G.M.
Melton
R.G.
Sherwood
R.F.
Wilson
W.R.
Denny
W.A.
Palmer
B.D.
Knox
R.J.
Friedlos
F.
Williams
A.
Bioactivation of dinitrobenzamide mustards by an E. coli B nitroreductase
Biochem. Pharmacol.
1995
, vol. 
50
 (pg. 
609
-
618
)
[PubMed]
143
Patel
K.
Choy
S.S.F.
Hicks
K.O.
Melink
T.J.
Holford
N.H.G.
Wilson
W.R.
A combined pharmacokinetic model for the hypoxia-targeted prodrug PR-104A in humans, dogs, rats and mice predicts species differences in clearance and toxicity
Cancer Chemother. Pharmacol.
2011
, vol. 
67
 (pg. 
1145
-
1155
)
[PubMed]
144
Gu
Y.
Patterson
A.V.
Atwell
G.J.
Chernikova
S.B.
Brown
J.M.
Thompson
L.H.
Wilson
W.R.
Roles of DNA repair and reductase activity in the cytotoxicity of the hypoxia-activated dinitrobenzamide mustard PR-104A
Mol. Cancer Ther.
2009
, vol. 
8
 (pg. 
1714
-
1723
)
[PubMed]
145
Singleton
R.S.
Guise
C.P.
Ferry
D.M.
Pullen
S.M.
Dorie
M.J.
Brown
J.M.
Patterson
A.V.
Wilson
W.R.
DNA cross-links in human tumor cells exposed to the prodrug PR-104A: relationships to hypoxia, bioreductive metabolism, and cytotoxicity
Cancer Res.
2009
, vol. 
69
 (pg. 
3884
-
3891
)
[PubMed]
146
Jameson
M.B.
Rischin
D.
Pegram
M.
Gutheil
J.
Patterson
A.V.
Denny
W.A.
Wilson
W.R.
A phase I trial of PR-104, a nitrogen mustard prodrug activated by both hypoxia and aldo-keto reductase 1C3, in patients with solid tumors
Cancer Chemother. Pharmacol.
2010
, vol. 
65
 (pg. 
791
-
801
)
[PubMed]
147
McKeage
M.J.
Gu
Y.
Wilson
W.R.
Hill
A.
Amies
K.
Melink
T.J.
Jameson
M.B.
A phase I trial of PR-104, a pre-prodrug of the bioreductive prodrug PR-104A, given weekly to solid tumour patients
BMC Cancer
2011
, vol. 
11
 pg. 
432
 
[PubMed]
148
Guise
C.P.
Abbattista
M.R.
Singleton
R.S.
Holford
S.D.
Connolly
J.
Dachs
G.U.
Fox
S.B.
Pollock
R.
Harvey
J.
Guilford
P.
, et al. 
The bioreductive prodrug PR-104A is activated under aerobic conditions by human aldo-keto reductase 1C3
Cancer Res.
2010
, vol. 
70
 (pg. 
1573
-
1584
)
[PubMed]
149
Birtwistle
J.
Hayden
R.E.
Khanim
F.L.
Green
R.M.
Pearce
C.
Davies
N.J.
Wake
N.
Schrewe
H.
Ride
J.P.
Chipman
J.K.
Bunce
C.M.
The aldo-keto reductase AKR1C3 contributes to 7,12-dimethylbenz(a)anthracene-3,4-dihydrodiol mediated oxidative DNA damage in myeloid cells: implications for leukemogenesis
Mutat. Res.
2009
, vol. 
662
 (pg. 
67
-
74
)
[PubMed]
150
Abbattista
M.R.
Jamieson
S.M.F.
Gu
Y.
Nickel
J.E.
Pullen
S.M.
Patterson
A.V.
Wilson
W.R.
Guise
C.P.
Pre-clinical activity of PR-104 as monotherapy and in combination with sorafenib in hepatocellular carcinoma
Cancer Biol. Ther.
2015
, vol. 
16
 (pg. 
610
-
622
)
[PubMed]
151
Smaill
J.B.
Patterson
A.V.
Ashoorzadeh
A.
Guise
C.P.
Mowday
A.M.
Ackerley
D.F.
Williams
E.M.
Copp
J.N.
Preparation of nitrobenzamide mustard prodrugs metabolized by microorganism nitroreductases or an hypoxic environment for treating hyperproliferative disorders
U.S. Pat.
WO 2014031012 A1, 
2014
152
Duan
J.X.
Jiao
H.
Kaizerman
J.
Stanton
T.
Evans
J.W.
Lan
L.
Lorente
G.
Banica
M.
Jung
D.
Wang
J.
, et al. 
Potent and highly selective hypoxia-activated achiral phosphoramidate mustards as anticancer drugs
J. Med. Chem.
2008
, vol. 
51
 (pg. 
2412
-
2420
)
[PubMed]
153
Meng
F.
Evans
J.W.
Bhupathi
D.
Banica
M.
Lan
L.
Lorente
G.
Duan
J.X.
Cai
X.
Mowday
A.M.
Guise
C.P.
, et al. 
Molecular and cellular pharmacology of the hypoxia-activated prodrug TH-302
Mol. Cancer Ther.
2012
, vol. 
11
 (pg. 
740
-
751
)
[PubMed]
154
Asche
C.
Dumy
P.
Carrez
D.
Croisy
A.
Demeunynck
M.
Nitrobenzylcarbamate prodrugs of cytotoxic acridines for potential use with nitroreductase gene-directed enzyme prodrug therapy
Bioorg. Med. Chem. Lett.
2006
, vol. 
16
 (pg. 
1990
-
1994
)
[PubMed]
155
Hay
M.P.
Anderson
R.F.
Ferry
D.M.
Wilson
W.R.
Denny
W.A.
Synthesis and evaluation of nitroheterocyclic carbamate prodrugs for use with nitroreductase-mediated gene-directed enzyme prodrug therapy
J. Med. Chem.
2003
, vol. 
46
 (pg. 
5533
-
5545
)
[PubMed]
156
Hay
M.P.
Wilson
W.R.
Denny
W.A.
Nitroarylmethylcarbamate prodrugs of doxorubicin for use with nitroreductase gene-directed enzyme prodrug therapy
Bioorg. Med. Chem.
2005
, vol. 
13
 (pg. 
4043
-
4055
)
[PubMed]
157
Wilson
W.R.
Stribbling
S.M.
Pruijn
F.B.
Syddall
S.P.
Patterson
A.V.
Liyanage
H.D.
Smith
E.
Botting
K.J.
Tercel
M.
Nitro-chloromethylbenzindolines: hypoxia-activated prodrugs of potent adenine N3 DNA minor groove alkylators
Mol. Cancer Ther.
2009
, vol. 
8
 (pg. 
2903
-
2913
)
[PubMed]
158
Tercel
M.
Atwell
G.J.
Yang
S.
Ashoorzadeh
A.
Stevenson
R.J.
Botting
K.J.
Gu
Y.
Mehta
S.Y.
Denny
W.A.
Wilson
W.R.
Pruijn
F.B.
Selective treatment of hypoxic tumor cells in vivo: phosphate pre-prodrugs of nitro analogues of the duocarmycins
Angew. Chem. Int. Ed. Engl.
2011
, vol. 
50
 (pg. 
2606
-
2609
)
[PubMed]
159
Green
L.K.
Syddall
S.P.
Carlin
K.M.
Bell
G.D.
Guise
C.P.
Mowday
A.M.
Hay
M.P.
Smaill
J.B.
Patterson
A.V.
Ackerley
D.F.
Pseudomonas aeruginosa NfsB and nitro-CBI-DEI - a promising enzyme/prodrug combination for gene directed enzyme prodrug therapy
Mol. Cancer
2013
, vol. 
12
 pg. 
58
 
[PubMed]
160
Hu
L.
Wu
X.
Han
J.
Chen
L.
Vass
S.O.
Browne
P.
Hall
B.S.
Bot
C.
Gobalakrishnapillai
V.
Searle
P.F.
, et al. 
Synthesis and structure–activity relationships of nitrobenzyl phosphoramide mustards as nitroreductase-activated prodrugs
Bioorg. Med. Chem. Lett.
2011
, vol. 
21
 (pg. 
3986
-
3991
)
[PubMed]
161
Jiang
Y.
Han
J.
Yu
C.
Vass
S.O.
Searle
P.F.
Browne
P.
Knox
R.J.
Hu
L.
Design, synthesis, and biological evaluation of cyclic and acyclic nitrobenzylphosphoramide mustards for E. coli nitroreductase activation
J. Med. Chem.
2006
, vol. 
49
 (pg. 
4333
-
4343
)
[PubMed]
162
Thorne
S.H.
Barak
Y.
Liang
W.
Bachmann
M.H.
Rao
J.
Contag
C.H.
Matin
A.
CNOB/ChrR6, a new prodrug enzyme cancer chemotherapy
Mol. Cancer Ther.
2009
, vol. 
8
 (pg. 
333
-
341
)
[PubMed]
163
Kanada
M.
Bachmann
M.H.
Hardy
J.W.
Frimannson
D.O.
Bronsart
L.
Wang
A.
Sylvester
M.D.
Schmidt
T.L.
Kaspar
R.L.
Butte
M.J.
, et al. 
Differential fates of biomolecules delivered to target cells via extracellular vesicles
Proc. Natl. Acad. Sci. U.S.A.
2015
, vol. 
112
 (pg. 
E1433
-
E1442
)
[PubMed]
164
Hunt
M.A.
Li
D.
Hay
M.P.
Currie
M.J.
Robinson
B.A.
Patterson
A.V.
Dachs
G.U.
Characterisation of enzyme prodrug gene therapy combinations in coated spheroids and vascular networks in vitro
J. Gene Med.
2012
, vol. 
14
 (pg. 
62
-
74
)
[PubMed]
165
Bridgewater
J.A.
Knox
R.J.
Pitts
J.D.
Collins
M.K.
Springer
C.J.
The bystander effect of the nitroreductase/CB 1954 enzyme/prodrug system is due to a cell-permeable metabolite
Hum. Gene Ther.
1997
, vol. 
8
 (pg. 
709
-
717
)
[PubMed]
166
Prosser
G.A.
Williams
E.M.
Sissons
J.A.
Walmsley
K.E.
Parker
M.R.
Ackerley
D.F.
A gain-of-function positive-selection expression plasmid that enables high-efficiency cloning
Biotechnol. Lett.
2015
, vol. 
37
 (pg. 
383
-
389
)
[PubMed]
167
Nuyts
S.
Mellaert
L.
Theys
J.
Landuyt
W.
Lambin
P.
Anné
J.
Clostridium spores for tumor-specific drug delivery
Anticancer Drugs
2002
, vol. 
13
 (pg. 
115
-
125
)
[PubMed]
168
Theys
J.
Landuyt
W.
Nuyts
S.
Mellaert
L.
Bosmans
E.
Rijnders
A.
Van Den Bogaert
W.
van Oosterom
A.
Anné
J.
Lambin
P.
Improvement of Clostridium tumour targeting vectors evaluated in rat rhabdomyosarcomas
FEMS Immunol. Med. Microbiol.
2001
, vol. 
30
 (pg. 
37
-
41
)
[PubMed]
169
Baril
P.
Martin-Duque
P.
Vassaux
G.
Visualization of gene expression in the live subject using the Na/I symporter as a reporter gene: applications in biotherapy
Br. J. Pharmacol.
2010
, vol. 
159
 (pg. 
761
-
771
)
[PubMed]
170
Barton
K.N.
Stricker
H.
Brown
S.L.
Elshaikh
M.
Aref
I.
Lu
M.
Pegg
J.
Zhang
Y.
Karvelis
K.C.
Siddiqui
F.
, et al. 
Phase I study of noninvasive imaging of adenovirus-mediated gene expression in the human prostate
Mol. Ther.
2008
, vol. 
16
 (pg. 
1761
-
1769
)
[PubMed]
171
Bhaumik
S.
Sekar
T.V.
Depuy
J.
Klimash
J.
Paulmurugan
R.
Noninvasive optical imaging of nitroreductase gene-directed enzyme prodrug therapy system in living animals
Gene Ther
2012
, vol. 
19
 (pg. 
295
-
302
)
[PubMed]
172
Chiocca
E.A.
Oncolytic viruses
Nat. Rev. Cancer
2002
, vol. 
2
 (pg. 
938
-
950
)
[PubMed]
173
Rojas
J.J.
Thorne
S.H.
Theranostic potential of oncolytic vaccinia virus
Theranostics
2012
, vol. 
2
 (pg. 
363
-
373
)
[PubMed]
174
Adair
R.A.
Roulstone
V.
Scott
K.J.
Morgan
R.
Nuovo
G.J.
Fuller
M.
Beirne
D.
West
E.J.
Jennings
V.A.
Rose
A.
, et al. 
Cell carriage, delivery, and selective replication of an oncolytic virus in tumor in patients
Sci. Transl. Med.
2012
, vol. 
4
 pg. 
138ra77
 
[PubMed]
175
Breitbach
C.J.
Burke
J.
Jonker
D.
Stephenson
J.
Haas
A.R.
Chow
L.Q.
Nieva
J.
Hwang
T.H.
Moon
A.
Patt
R.
, et al. 
Intravenous delivery of a multi-mechanistic cancer-targeted oncolytic poxvirus in humans
Nature
2011
, vol. 
477
 (pg. 
99
-
102
)
[PubMed]
176
Gollamudi
R.
Ghalib
M.H.
Desai
K.K.
Chaudhary
I.
Wong
B.
Einstein
M.
Coffey
M.
Gill
G.M.
Mettinger
K.
Mariadason
J.M.
, et al. 
Intravenous administration of Reolysin®, a live replication competent RNA virus is safe in patients with advanced solid tumors
Invest. New Drugs
2009
, vol. 
28
 (pg. 
641
-
649
)
[PubMed]
177
Nasu
Y.
Saika
T.
Ebara
S.
Kusaka
N.
Kaku
H.
Abarzua
F.
Manabe
D.
Thompson
T.C.
Kumon
H.
Suicide gene therapy with adenoviral delivery of HSV-tK gene for patients with local recurrence of prostate cancer after hormonal therapy
Mol. Ther.
2007
, vol. 
15
 (pg. 
834
-
840
)
[PubMed]
178
Bhaumik
S.
Advances in imaging gene-directed enzyme prodrug therapy
Curr. Pharm. Biotechnol.
2011
, vol. 
12
 (pg. 
497
-
507
)
[PubMed]
179
Brader
P.
Serganova
I.
Blasberg
R.G.
Noninvasive molecular imaging using reporter genes
J. Nucl. Med.
2013
, vol. 
54
 (pg. 
167
-
172
)
[PubMed]
180
Briat
A.
Vassaux
G.
Preclinical applications of imaging for cancer gene therapy
Expert Rev. Mol. Med.
2006
, vol. 
8
 (pg. 
1
-
19
)
[PubMed]
181
Waerzeggers
Y.
Monfared
P.
Viel
T.
Winkeler
A.
Voges
J.
Jacobs
A.H.
Methods to monitor gene therapy with molecular imaging
Methods
2009
, vol. 
48
 (pg. 
146
-
160
)
[PubMed]
182
Komar
G.
Seppänen
M.
Eskola
O.
Lindholm
P.
Grönroos
T.J.
Forsback
S.
Sipilä
H.
Evans
S.M.
Solin
O.
Minn
H.
18F-EF5: a new PET tracer for imaging hypoxia in head and neck cancer
J. Nucl. Med.
2008
, vol. 
49
 (pg. 
1944
-
1951
)
[PubMed]
183
Dubois
L.J.
Lieuwes
N.G.
Janssen
M.H.
Peeters
W.J.
Windhorst
A.D.
Walsh
J.C.
Kolb
H.C.
Öllers
M.C.
Bussink
J.
van Dongen
G.A.
, et al. 
Preclinical evaluation and validation of [18F]HX4, a promising hypoxia marker for PET imaging
Proc. Natl. Acad. Sci. U.S.A.
2011
, vol. 
108
 (pg. 
14620
-
14625
)
[PubMed]
184
Grierson
J.R.
Link
J.M.
Mathis
C.A.
Rasey
J.S.
Krohn
K.A.
A radiosynthesis of fluorine-18 fluoromisonidazole
J. Nucl. Med.
1989
, vol. 
30
 (pg. 
343
-
350
)
[PubMed]
185
Sekar
T.V.
Foygel
K.
Ilovich
O.
Paulmurugan
R.
Noninvasive theranostic imaging of HSV1-sr39TK-NTR/GCV-CB1954 dual-prodrug therapy in metastatic lung lesions of MDA-MB-231 triple negative breast cancer in mice
Theranostics
2014
, vol. 
4
 (pg. 
460
-
474
)
[PubMed]
186
Cui
L.
Zhong
Y.
Zhu
W.
Xu
Y.
Du
Q.
Wang
X.
Qian
X.
Xiao
Y.
A new prodrug-derived ratiometric fluorescent probe for hypoxia: high selectivity of nitroreductase and imaging in tumor cell
Org. Lett.
2011
, vol. 
13
 (pg. 
928
-
931
)
[PubMed]
187
Guo
T.
Cui
L.
Shen
J.
Zhu
W.
Xu
Y.
Qian
X.
A highly sensitive long-wavelength fluorescence probe for nitroreductase and hypoxia: selective detection and quantification
Chem. Commun.
2013
, vol. 
49
 (pg. 
10820
-
10822
)
188
Li
Z.
Li
X.
Gao
X.
Zhang
Y.
Shi
W.
Ma
H.
Nitroreductase detection and hypoxic tumor cell imaging by a designed sensitive and selective fluorescent probe, 7-[(5-Nitrofuran-2-yl)methoxy]-3H-phenoxazin-3-one
Anal. Chem.
2013
, vol. 
85
 (pg. 
3926
-
3932
)
[PubMed]
189
Li
Z.
He
X.
Wang
Z.
Yang
R.
Shi
W.
Ma
H.
In vivo imaging and detection of nitroreductase in zebrafish by a new near-infrared fluorescence off-on probe
Biosens. Bioelectron.
2015
, vol. 
63
 (pg. 
112
-
116
)
[PubMed]
190
Wan
Q.Q.
Gao
X.H.
He
X.Y.
Chen
S.M.
Song
Y.C.
Gong
Q.Y.
Li
X.H.
Ma
H.M.
A cresyl violet-based fluorescent off–on probe for the detection and imaging of hypoxia and nitroreductase in living organisms
Chem. Asian J.
2014
, vol. 
9
 (pg. 
2058
-
2062
)
[PubMed]
191
Xu
J.
Sun
S.
Li
Q.
Yue
Y.
Li
Y.
Shao
S.
A rapid response “turn-on” fluorescent probe for nitroreductase detection and its application in hypoxic tumor cell imaging
Analyst
2015
, vol. 
140
 (pg. 
574
-
581
)
[PubMed]
192
Nolting
D.D.
Gore
J.C.
Pham
W.
Near-infrared dyes: probe development and applications in optical molecular imaging
Curr. Org. Synth.
2011
, vol. 
8
 (pg. 
521
-
534
)
[PubMed]
193
McCormack
E.
Silden
E.
West
R.M.
Pavlin
T.
Micklem
D.R.
Lorens
J.B.
Haug
B.E.
Cooper
M.E.
Gjertsen
B.T.
Nitroreductase, a near-infrared reporter platform for in vivo time-domain optical imaging of metastatic cancer
Cancer Res.
2013
, vol. 
73
 (pg. 
1276
-
1286
)
[PubMed]
194
Wong
R.H.F.
Kwong
T.
Yau
K.-H.
Au-Yeung
H.Y.
Real time detection of live microbes using a highly sensitive bioluminescent nitroreductase probe
Chem. Commun.
2015
, vol. 
51
 (pg. 
4440
-
4442
)
[PubMed]
195
Porterfield
W.B.
Jones
K.A.
McCutcheon
D.C.
Prescher
J.A.
A “caged” luciferin for imaging cell-cell contacts
J. Am. Chem. Soc.
2015
, vol. 
137
 (pg. 
8656
-
8659
)
[PubMed]
196
Vorobyeva
A.G.
Stanton
M.
Godinat
A.
Lund
K.B.
Karateev
G.G.
Francis
K.P.
Allen
E.
Gelovani
J.G.
McCormack
E.
Tangney
M.
Dubikovskaya
E.A.
Development of a bioluminescent nitroreductase probe for preclinical imaging
PLoS One
2015
, vol. 
10
 pg. 
e0131037
 
[PubMed]
197
Cronin
M.
Akin
A.R.
Collins
S.A.
Meganck
J.
Kim
J.-B.
Baban
C.K.
Joyce
S.A.
van Dam
G.M.
Zhang
N.
van Sinderen
D.
, et al. 
High resolution in vivo bioluminescent imaging for the study of bacterial tumour targeting
PLoS One
2012
, vol. 
7
 pg. 
e30940
 
[PubMed]
198
Yu
Y.A.
Shabahang
S.
Timiryasova
T.M.
Zhang
Q.
Beltz
R.
Gentschev
I.
Goebel
W.
Szalay
A.A.
Visualization of tumors and metastases in live animals with bacteria and vaccinia virus encoding light-emitting proteins
Nat. Biotechnol.
2004
, vol. 
22
 (pg. 
313
-
320
)
[PubMed]
199
De Poorter
J.J.
Hoeben
R.C.
Hogendoorn
S.
Mautner
V.
Ellis
J.
Obermann
W.R.
Huizinga
T.W.J.
Nelissen
R.G.
Gene therapy and cement injection for restabilization of loosened hip prostheses
Hum. Gene Ther.
2008
, vol. 
19
 (pg. 
83
-
95
)
[PubMed]
200
De Poorter
J.J.
Tolboom
T.C.
Rabelink
M.J.
Pieterman
E.
Hoeben
R.C.
Nelissen
R.G.
Huizinga
T.W.
Towards gene therapy in prosthesis loosening: efficient killing of interface cells by gene-directed enzyme prodrug therapy with nitroreductase and the prodrug CB1954
J. Gene Med.
2005
, vol. 
7
 (pg. 
1421
-
1428
)
[PubMed]
201
Poorter
J.J.
Hoeben
R.C.
Obermann
W.R.
Huizinga
T.W.
Nelissen
R.G.
Gene therapy for the treatment of hip prosthesis loosening: adverse events in a phase 1 clinical study
Hum. Gene Ther.
2008
, vol. 
19
 (pg. 
1029
-
1038
)
[PubMed]
202
Yamazoe
S.
McQuade
L.E.
Chen
J.K.
Nitroreductase-activatable morpholino oligonucleotides for in vivo gene silencing
ACS Chem. Biol.
2014
, vol. 
9
 (pg. 
1985
-
1990
)
[PubMed]
203
Bill
B.R.
Petzold
A.M.
Clark
K.J.
Schimmenti
L.A.
Ekker
S.C.
A primer for morpholino use in zebrafish
Zebrafish
2009
, vol. 
6
 (pg. 
69
-
77
)
[PubMed]
204
Griffin
L.B.
January
K.E.
Ho
K.W.
Cotter
K.A.
Callard
G.V.
Morpholino-mediated knockdown of ERα, ERβa, and ERβb mRNAs in zebrafish (Danio rerio) embryos reveals differential regulation of estrogen-inducible genes
Endocrinology
2013
, vol. 
154
 (pg. 
4158
-
4169
)
[PubMed]
205
Howard
E.W.
Newman
L.A.
Oleksyn
D.W.
Angerer
R.C.
Angerer
L.M.
SpKrl: a direct target of beta-catenin regulation required for endoderm differentiation in sea urchin embryos
Development
2001
, vol. 
128
 (pg. 
365
-
375
)
[PubMed]
206
Kos
R.
Reedy
M.V.
Johnson
R.L.
Erickson
C.A.
The winged-helix transcription factor FoxD3 is important for establishing the neural crest lineage and repressing melanogenesis in avian embryos
Development
2001
, vol. 
128
 (pg. 
1467
-
1479
)
[PubMed]
207
Nasevicius
A.
Ekker
S.C.
Effective targeted gene “knockdown” in zebrafish
Nat. Genet.
2000
, vol. 
26
 (pg. 
216
-
220
)
[PubMed]
208
Griepenburg
J.C.
Rapp
T.L.
Carroll
P.J.
Eberwine
J.
Dmochowski
I.J.
Ruthenium-caged antisense morpholinos for regulating gene expression in zebrafish embryos
Chem. Sci.
2015
, vol. 
6
 (pg. 
2342
-
2346
)
[PubMed]
209
Wang
Y.
Wu
L.
Wang
P.
Lv
C.
Yang
Z.
Tang
X.
Manipulation of gene expression in zebrafish using caged circular morpholino oligomers
Nucleic Acids Res.
2012
, vol. 
40
 (pg. 
11155
-
11162
)
[PubMed]
210
Iversen
P.L.
Arora
V.
Acker
A.J.
Mason
D.H.
Devi
G.R.
Efficacy of antisense morpholino oligomer targeted to c-myc in prostate cancer xenograft murine model and a phase I safety study in humans
Clin. Cancer Res.
2003
, vol. 
9
 (pg. 
2510
-
2519
)
[PubMed]
211
Sekhon
H.S.
London
C.A.
Sekhon
M.
Iversen
P.L.
Devi
G.R.
c-MYC antisense phosphosphorodiamidate morpholino oligomer inhibits lung metastasis in a murine tumor model
Lung Cancer
2008
, vol. 
60
 (pg. 
347
-
354
)
[PubMed]
212
Devi
G.R.
Beer
T.M.
Corless
C.L.
Arora
V.
Weller
D.L.
Iversen
P.L.
In vivo bioavailability and pharmacokinetics of a c-MYC antisense phosphorodiamidate morpholino oligomer, AVI-4126, in solid tumors
Clin. Cancer Res.
2005
, vol. 
11
 (pg. 
3930
-
3938
)
[PubMed]
213
Templeton
N.S.
Gene and Cell Therapy: Therapeutic Mechanisms and Strategies
2015
4th edn
Boca Raton
CRC Press
214
Curado
S.
Stainier
D.Y.
Anderson
R.M.
Nitroreductase-mediated cell/tissue ablation in zebrafish: a spatially and temporally controlled ablation method with applications in developmental and regeneration studies
Nat. Protoc.
2008
, vol. 
3
 (pg. 
948
-
954
)
[PubMed]
215
Pisharath
H.
Rhee
J.M.
Swanson
M.A.
Leach
S.D.
Parsons
M.J.
Targeted ablation of beta cells in the embryonic zebrafish pancreas using E. coli nitroreductase
Mech. Dev.
2007
, vol. 
124
 (pg. 
218
-
229
)
[PubMed]
216
White
D.T.
Mumm
J.S.
The nitroreductase system of inducible targeted ablation facilitates cell-specific regenerative studies in zebrafish
Methods
2013
, vol. 
62
 (pg. 
232
-
240
)
[PubMed]
217
Mathias
J.R.
Zhang
Z.
Saxena
M.T.
Mumm
J.S.
Enhanced cell-specific ablation in zebrafish using a triple mutant of Escherichia coli nitroreductase
Zebrafish
2014
, vol. 
11
 (pg. 
85
-
97
)
[PubMed]
218
Cheng
X.
Chen
X.
Li
D.
Jin
X.
He
J.
Yin
Z.
Effects of metronidazole on proopiomelanocortin a gene expression in zebrafish
Gen. Comp. Endocrinol.
2015
, vol. 
214
 (pg. 
87
-
94
)
[PubMed]
219
Farquhar
D.
Pan
B.F.
Sakurai
M.
Ghosh
A.
Mullen
C.A.
Nelson
J.A.
Suicide gene therapy using E. coli beta-galactosidase
Cancer Chemother. Pharmacol.
2002
, vol. 
50
 (pg. 
65
-
70
)
[PubMed]
220
Chen
X.
Wu
B.
Wang
P.G.
Glucuronides in anti-cancer therapy
Curr. Med. Chem. Anticancer Agents
2003
, vol. 
3
 (pg. 
139
-
150
)
[PubMed]
221
Li
J.
Li
H.
Zhu
L.
Song
W.
Li
R.
Wang
D.
Dou
K.
The adenovirus-mediated linamarase/linamarin suicide system: a potential strategy for the treatment of hepatocellular carcinoma
Cancer Lett.
2010
, vol. 
289
 (pg. 
217
-
227
)
[PubMed]
222
Oosterhoff
D.
Witlox
M.A.
van Beusechem
V.W.
Haisma
H.J.
Schaap
G.R.
Bras
J.
Kruyt
F.A.
Molenaar
B.
Boven
E.
Wuisman
P.I.
, et al. 
Gene-directed enzyme prodrug therapy for osteosarcoma: sensitization to CPT-11 in vitro and in vivo by adenoviral delivery of a gene encoding secreted carboxylesterase-2
Mol. Cancer Ther.
2003
, vol. 
2
 (pg. 
765
-
771
)
[PubMed]
223
Hao
X.-K.
Liu
J.-Y.
Yue
Q.-H.
Wu
G.-J.
Bai
Y.-J.
Yin
Y.
In vitro and in vivo prodrug therapy of prostate cancer using anti-gamma-Sm-scFv/hCPA fusion protein
Prostate
2006
, vol. 
66
 (pg. 
858
-
866
)
[PubMed]
224
Davies
L.C.
Friedlos
F.
Hedley
D.
Martin
J.
Ogilvie
L.M.
Scanlon
I.J.
Springer
C.J.
Novel fluorinated prodrugs for activation by carboxypeptidase G2 showing good in vivo antitumor activity in gene-directed enzyme prodrug therapy
J. Med. Chem.
2005
, vol. 
48
 (pg. 
5321
-
5328
)
[PubMed]
225
Schepelmann
S.
Ogilvie
L.M.
Hedley
D.
Friedlos
F.
Martin
J.
Scanlon
I.
Chen
P.
Marais
R.
Springer
C.J.
Suicide gene therapy of human colon carcinoma xenografts using an armed oncolytic adenovirus expressing carboxypeptidase G2
Cancer Res.
2007
, vol. 
67
 (pg. 
4949
-
4955
)
[PubMed]
226
Francis
R.J.
Sharma
S.K.
Springer
C.
Green
A.J.
Hope-Stone
L.D.
Sena
L.
Martin
J.
Adamson
K.L.
Robbins
A.
Gumbrell
L.
, et al. 
A phase I trial of antibody directed enzyme prodrug therapy (ADEPT) in patients with advanced colorectal carcinoma or other CEA producing tumours
Br. J. Cancer
2002
, vol. 
87
 (pg. 
600
-
607
)
[PubMed]
227
Tupper
J.
Greco
O.
Tozer
G.M.
Dachs
G.U.
Analysis of the horseradish peroxidase/indole-3-acetic acid combination in a three-dimensional tumor model
Cancer Gene Ther.
2004
, vol. 
11
 (pg. 
508
-
513
)
[PubMed]
228
Dai
M.
Liu
J.
Chen
D.E.
Rao
Y.
Tang
Z.J.
Ho
W.Z.
Dong
C.Y.
Tumor-targeted gene therapy using Adv-AFP-HRPC/IAA prodrug system suppresses growth of hepatoma xenografted in mice
Cancer Gene Ther.
2012
, vol. 
19
 (pg. 
77
-
83
)
[PubMed]
229
Zhou
J.H.
Tang
B.
Liu
X.L.
He
D.W.
Yang
D.T.
hTERT-targeted E. coli purine nucleoside phosphorylase gene/6-methylpurine deoxyribose therapy for pancreatic cancer
Chin. Med. J.
2007
, vol. 
120
 (pg. 
1348
-
1352
)
[PubMed]
230
Nagata
T.
Nakamori
M.
Iwahashi
M.
Yamaue
H.
Overexpression of pyrimidine nucleoside phosphorylase enhances the sensitivity to 5'-deoxy-5-fluorouridine in tumour cells in vitro and in vivo
Eur. J. Cancer
2002
, vol. 
38
 (pg. 
712
-
717
)
[PubMed]
231
Evrard
A.
Cuq
P.
Robert
B.
Vian
L.
Pèlegrin
A.
Cano
J.P.
Enhancement of 5-fluorouracil cytotoxicity by human thymidine-phosphorylase expression in cancer cells: in vitro and in vivo study
Int. J. Cancer
1999
, vol. 
80
 (pg. 
465
-
470
)
[PubMed]
232
Freytag
S.O.
Stricker
H.
Pegg
J.
Paielli
D.
Pradhan
D.G.
Peabody
J.
DePeralta-Venturina
M.
Xia
X.
Brown
S.
Lu
M.
Kim
J.H.
Phase I study of replication-competent adenovirus-mediated double-suicide gene therapy in combination with conventional-dose three-dimensional conformal radiation therapy for the treatment of newly diagnosed, intermediate- to high-risk prostate cancer
Cancer Res.
2003
, vol. 
63
 (pg. 
7497
-
7506
)
[PubMed]
233
Cowen
R.L.
Williams
K.J.
Chinje
E.C.
Jaffar
M.
Sheppard
F.C.
Telfer
B.A.
Wind
N.S.
Stratford
I.J.
Hypoxia targeted gene therapy to increase the efficacy of tirapazamine as an adjuvant to radiotherapy: reversing tumor radioresistance and effecting cure
Cancer Res.
2004
, vol. 
64
 (pg. 
1396
-
1402
)
[PubMed]
234
Tychopoulos
M.
Corcos
L.
Genne
P.
Beaune
P.
de Waziers
I.
A virus-directed enzyme prodrug therapy (VDEPT) strategy for lung cancer using a CYP2B6/NADPH-cytochrome P450 reductase fusion protein
Cancer Gene Ther.
2005
, vol. 
12
 (pg. 
497
-
508
)
[PubMed]
235
Braybrooke
J.P.
Slade
A.
Deplanque
G.
Harrop
R.
Madhusudan
S.
Forster
M.D.
Gibson
R.
Makris
A.
Talbot
D.C.
Steiner
J.
, et al. 
Phase I study of MetXia-P450 gene therapy and oral cyclophosphamide for patients with advanced breast cancer or melanoma
Clin. Cancer Res.
2005
, vol. 
11
 (pg. 
1512
-
1520
)
[PubMed]
236
Salmons
B.
Löhr
M.
Günzburg
W.H.
Treatment of inoperable pancreatic carcinoma using a cell-based local chemotherapy: results of a phase I/II clinical trial
J. Gastroenterol.
2003
, vol. 
38
 
Suppl. 15
(pg. 
78
-
84
)
[PubMed]
237
McCarthy
H.O.
Yakkundi
A.
McErlane
V.
Hughes
C.M.
Keilty
G.
Murray
M.
Patterson
L.H.
Hirst
D.G.
McKeown
S.R.
Robson
T.
Bioreductive GDEPT using cytochrome P450 3A4 in combination with AQ4N
Cancer Gene Ther.
2003
, vol. 
10
 (pg. 
40
-
48
)
[PubMed]
238
Huber
B.E.
Richards
C.A.
Krenitsky
T.A.
Retroviral-mediated gene therapy for the treatment of hepatocellular carcinoma: an innovative approach for cancer therapy
Proc. Natl. Acad. Sci. U.S.A.
1991
, vol. 
88
 (pg. 
8039
-
8043
)
[PubMed]
239
Searle
P.F.
Chen
M.J.
Hu
L.
Race
P.R.
Lovering
A.L.
Grove
J.I.
Guise
C.
Jaberipour
M.
James
N.D.
Mautner
V.
, et al. 
Nitroreductase: a prodrug-activating enzyme for cancer gene therapy
Clin. Exp. Pharmacol. Physiol.
2004
, vol. 
31
 (pg. 
811
-
816
)
[PubMed]
240
Olekhnovich
I.N.
Goodwin
A.
Hoffman
P.S.
Characterization of the NAD(P)H oxidase and metronidazole reductase activities of the RdxA nitroreductase of Helicobacter pylori
FEBS J.
2009
, vol. 
276
 (pg. 
3354
-
3364
)
[PubMed]
241
Koike
H.
Sasaki
H.
Kobori
T.
Zenno
S.
Saigo
K.
Murphy
M.E.
Adman
E.T.
Tanokura
M.
1.8 Å crystal structure of the major NAD(P)H: FMN oxidoreductase of a bioluminescent bacterium, Vibrio fischeri: overall structure, cofactor and substrate-analog binding, and comparison with related flavoproteins
J. Mol. Biol.
1998
, vol. 
280
 (pg. 
259
-
273
)
[PubMed]
242
Martínez-Júlvez
M.
Rojas
A.L.
Olekhnovich
I.
Espinosa Angarica
V.
Hoffman
P.S.
Sancho
J.
Structure of RdxA - an oxygen-insensitive nitroreductase essential for metronidazole activation in Helicobacter pylori
FEBS J.
2012
, vol. 
279
 (pg. 
4306
-
4317
)
[PubMed]
243
Emptage
C.D.
Knox
R.J.
Danson
M.J.
Hough
D.W.
Nitroreductase from Bacillus licheniformis: a stable enzyme for prodrug activation
Biochem. Pharmacol.
2009
, vol. 
77
 (pg. 
21
-
29
)
[PubMed]
244
Tanner
J.J.
Lei
B.
Tu
S.C.
Krause
K.L.
Flavin reductase P:  structure of a dimeric enzyme that reduces flavin
Biochemistry
1996
, vol. 
35
 (pg. 
13531
-
13539
)
[PubMed]
245
Tanner
J.J.
Tu
S.C.
Barbour
L.J.
Barnes
C.L.
Krause
K.L.
Unusual folded conformation of nicotinamide adenine dinucleotide bound to flavin reductase P
Protein Sci.
1999
, vol. 
8
 (pg. 
1725
-
1732
)
[PubMed]
246
Cortial
S.
Chaignon
P.
Iorga
B.I.
Aymerich
S.
Truan
G.
Gueguen-Chaignon
V.
Meyer
P.
Moréra
S.
Ouazzani
J.
NADH oxidase activity of Bacillus subtilis nitroreductase NfrA1: insight into its biological role
FEBS Lett.
2010
, vol. 
584
 (pg. 
3916
-
3922
)
[PubMed]
247
Morokutti
A.
Lyskowski
A.
Sollner
S.
Pointner
E.
Fitzpatrick
T.B.
Kratky
C.
Gruber
K.
Macheroux
P.
Structure and function of YcnD from Bacillus subtilis, a flavin-containing oxidoreductase
Biochemistry
2005
, vol. 
44
 (pg. 
13724
-
13733
)
[PubMed]
248
Berne
C.
Betancor
L.
Luckarift
H.R.
Spain
J.C.
Application of a microfluidic reactor for screening cancer prodrug activation using silica-immobilized nitrobenzene nitroreductase
Biomacromolecules
2006
, vol. 
7
 (pg. 
2631
-
2636
)
[PubMed]
249
Çelik
A.
Yetiş
G.
An unusually cold active nitroreductase for prodrug activations
Bioorg. Med. Chem.
2012
, vol. 
20
 (pg. 
3540
-
3550
)
[PubMed]
250
Guise
C.P.
Wang
A.T.
Theil
A.
Bridewell
D.J.
Wilson
W.R.
Patterson
A.V.
Identification of human reductases that activate the dinitrobenzamide mustard prodrug PR-104A: a role for NADPH:cytochrome P450 oxidoreductase under hypoxia
Biochem. Pharmacol.
2007
, vol. 
74
 (pg. 
810
-
820
)
[PubMed]