Tricyclic [1,2,4]triazine 1,4-dioxides as hypoxia selective cytotoxins

Article abstract of DOI:10.1021/jm800967h

A series of novel tricyclic triazine-di-N-oxides (TTOs) related to tirapazamine have been designed and prepared. A wide range of structural arrangements with cycloalkyl, oxygen-, and nitrogen-containing saturated rings fused to the triazine core, coupled with various side chains linked to either hemisphere, resulted in TTO analogues that displayed hypoxia-selective cytotoxicity in vitro. Optimal rates of hypoxic metabolism and tissue diffusion coefficients were achieved with fused cycloalkyl rings in combination with both the 3-aminoalkyl or 3-alkyl substituents linked to weakly basic soluble amines. The selection was further refined using pharmacokinetic/pharmacodynamic model predictions of the in vivo hypoxic potency (AUC_req) and selectivity (HCD) with 12 TTO analogues predicted to be active in vivo, subject to the achievement of adequate plasma pharmacokinetics.

Full text of DOI:10.1021/jm800967h

J. Med. Chem. 2008, 51, 6853–6865
6853
Tricyclic [1,2,4]Triazine 1,4-Dioxides As Hypoxia Selective Cytotoxins
Michael P. Hay,* Kevin O. Hicks, Karin Pchalek,† Ho H. Lee, Adrian Blaser, Frederik B. Pruijn, Robert F. Anderson,
Sujata S. Shinde, William R. Wilson, and William A. Denny
Auckland Cancer Society Research Centre, Faculty of Medical and Health Sciences, The UniVersity of Auckland, Auckland, New Zealand
ReceiVed July 30, 2008
A series of novel tricyclic triazine-di-N-oxides (TTOs) related to tirapazamine have been designed and
prepared. A wide range of structural arrangements with cycloalkyl, oxygen-, and nitrogen-containing saturated
rings fused to the triazine core, coupled with various side chains linked to either hemisphere, resulted in
TTO analogues that displayed hypoxia-selective cytotoxicity in vitro. Optimal rates of hypoxic metabolism
and tissue diffusion coefficients were achieved with fused cycloalkyl rings in combination with both the
3-aminoalkyl or 3-alkyl substituents linked to weakly basic soluble amines. The selection was further refined
using pharmacokinetic/pharmacodynamic model predictions of the in vivo hypoxic potency (AUCreq) and
selectivity (HCD) with 12 TTO analogues predicted to be active in vivo, subject to the achievement of
adequate plasma pharmacokinetics.
Introduction
rapid metabolism of TPZ in hypoxic tissues, leading to limited
penetration of the drug into hypoxic tissue.39-45
The increasingly well-defined role of tumor hypoxia in driving
tumor metabolism, progression, invasion, and metastasis,1-7 as
well as resistance to therapy,8-13 emphasizes the cogent need
for clinical agents that selectively target hypoxia.14,15
Overcoming this limited extravascular transport (EVT) re-
quires balancing the rate of bioreductive metabolism relative
to the tissue diffusion coefficient: analogues with low rates of
bioreductive metabolism lack cytotoxic potency, while analogues
with high rates of metabolism suffer limited EVT due to
excessive consumption of the prodrug.40,46-49 Optimisation of
EVT is but one element required in maximizing activity against
hypoxic cells in tumors. We have recently demonstrated a
spatially resolved pharmacokinetic/pharmacodynamic (PK/PD)
model that integrates hypoxic cytotoxicity and selectivity, EVT,
and plasma pharmacokinetics to successfully predict in vivo
activity against hypoxic cells in HT2945 and SiHa tumors.46
We have applied this model to guide drug synthesis and testing
in two studies of BTO analogues and identified several BTOs
with improved activity against hypoxic cells in vivo.48,49
We initially determined structure-activity relationships (SAR)
between A-ring substituents and one-electron reduction potential
[E(1)], hypoxic cytotoxicity, and hypoxic selectivity,50 but issues
of aqueous solubility and EVT were not addressed directly in
this study. Another study of neutral BTOs demonstrated an
increase in tissue diffusion coefficients, measured using mul-
ticellular layer (MCL) cultures of analogues with increased
lipophilicity.47 Recently, we reported48 a systematic evaluation
of 3-aminoalkylamino BTOs, seeking analogues with improved
aqueous solubility and EVT. This study showed that electron-
donating substituents in the 6-position could be used to
counteract increased rates of metabolism caused by the 3-ami-
noalkylamino substituents, e.g., 2, and demonstrated that suf-
ficiently lipophilic analogues provide MCL penetration similar
to TPZ and show activity against hypoxic cells in tumor
xenografts. A subsequent study49 demonstrated that removal of
hydrogen bond donors at the 3-position could further increase
tissue diffusion coefficients but that the less electron-donating
3-alkyl substituents raised reduction potentials and hypoxic
metabolism, which could be counteracted by adding electron-
donating substituents to the A-ring. Several BTO analogues with
improved in vivo activity relative to TPZ were identified in these
studies.
One class of hypoxia-selective cytotoxins under development
is the benzotriazine 1,4-dioxides (BTOs) represented by the
archetype tirapazamine (1, TPZ).a16,17 TPZ acts as a prodrug
that undergoes selective bioreductive activation18 under hypoxic
conditions to ultimately form an oxidizing radical that may react
with DNA,19-22 leading to DNA strand breaks and poisoning
of topoisomerase II.23-27 TPZ shows selective toxicity to
hypoxic cells in vitro and in experimental tumors,17,28-31 and
a range of clinical trials have produced modest therapeutic
results to date.32-34 This approach, using TPZ to selectively
kill hypoxic cells in tumors, is an early example of “physi-
ologically-targeted therapy.” Recent prognostic data35,36 from
a clinical trial in head and neck cancer37 has highlighted the
importance of preselecting patients with hypoxia when using
such a targeted therapy, and this may in part explain the modest
results with TPZ in trials.
TPZ has limitations that blunt its efficacy in vivo and,
potentially, in the clinic. It displays lower hypoxic selectivity
in vivo (2- to 3-fold) compared to in vitro (hypoxic cytotoxicity
ratio, HCR, ca. 50- to100-fold),38 which has been ascribed to
* To whom correspondence should be addressed. Phone: 64-9-9235698,
ext. 86598. Fax: 64-9-3737502. E-mail: m.hay@auckland.ac.nz. Address:
Auckland Cancer Society Research Centre, Faculty of Medical and Health
Sciences, The University of Auckland, Private Bag 92019, Auckland 1142,
New Zealand.
†
Current address: Boehringer Ingelheim Pharma GmbH & Co. KG,
55216 Ingelheim am Rhein, Germany.
a Abbreviations: TPZ, Tirapazamine; BTO, 1,2,4-benzotriazine 1,4-
dioxide; HCR, hypoxic cytotoxicity ratio; EVT, extravascular transport; E(1),
one-electron reduction potential; PK/PD, pharmacokinetic/pharmacody-
namic; D, tissue diffusion coefficient; MCL, multicellular layer; kmet, first-
order rate constant for bioreductive metabolism; SAR, structure-activity
relationships; 9-BBN, 9-borabicyclo[3.3.1]nonane; P7.4, octanol/water coef-
ficient at pH 7.4; CT10, area under concentration-time curve providing 10%
surviving fraction; M10, amount of BTO metabolized for 10% surviving
fraction; X1/2, calculated penetration half-distance into hypoxic tissue; SF,
surviving fraction; LCK, log cell kill; AUCreq, area under the plasma
concentration-time curve required to give 1 log of cell kill of hypoxic
cells in HT29 tumors; HCD, hypoxic cytotoxicity differential.
However, these findings and the limited SAR studies by
others51-55 have all been on analogues based on the 1,2,4-
10.1021/jm800967h CCC: $40.75
2008 American Chemical Society
Published on Web 10/11/2008

6854 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 21
Hay et al.
Scheme 1a
under acidic conditions gave nitroaniline 48, which underwent
Arndt cyclization to give 1-oxide 49, which was oxidized to
TTO 3. The 1-oxide 49 was converted to the 3-chloride 50,
and this underwent displacement with a series of lipophilic
aminoalkylamines to give 1-oxides 51-55, which were oxidized
to the corresponding TTOs 4-8.
A similar sequence of reactions was carried out using the
angular nitroacetamide isomer 46. Deprotection gave nitroaniline
56, which was converted to 1-oxide 57 and then to 3-chloride
58, which was elaborated to a series of 3-aminoalkyl 1-oxides
59-63 and oxidized to TTOs 9-13, respectively.
a Reagents: (i) NH2CN, HCl, ∆, then 30% NaOH, ∆; (ii) CH3CO3H,
CH3CO2H; (iii) NaNO2, CF3CO2H; (iv) DMF, POCl3, ∆; (v) t-BuNO2,
CH2I2, CuI, THF, ∆; (vi) R2CH2CH2NH2, DME, ∆; (vii) Stille or Heck
coupling; (viii) CF3CO3H, CF3CO2H, DCM.
Access to the methylcyclopentane analogues 14 and 15 was
accomplished by elaboration of 2-methylindanone 64 (Scheme
3). Nitration gave 4- and 6-nitroisomers 65 and 66. Reduction
and acetylation of 66 gave acetamide 67, which was nitrated to
give nitroacetamide 68 and deprotected to nitroaniline 69.
Cyclization of 69 gave 1-oxide 70, which was diazotized and
chlorinated to 3-chloride 71. Displacement with amines gave
72 and 73, which were oxidized to TTOs 14 and 15, respectively.
Similar sequences were followed to elaborate R-tetralone (74)
to linear tricyclic 1,4-dioxides 16 and 17 and the angular
analogue 18 (Scheme 4) as well as to convert benzosuberone
(88) to the cycloheptane TTO 19 (Scheme 5).
The use of alkoxy substituents to lower electron affinity has
previously48,49 yielded active analogues, e.g., 2, thus a series
of fused dihydrobenzofuran analogues were synthesized (Scheme
6). Friedel-Crafts acylation of dihydrobenzofuran (96) gave
ketone 97, which was converted to the oxime and underwent
Beckmann rearrangement to acetamide 98. Nitration gave
nitroacetamide 99, which was deprotected to give nitroaniline
100. Conversion of 100 to the “[3,2-g]” 3-amino 1-oxide isomer
101 and subsequent chlorination gave 3-chloride 102, which
underwent displacement with amines to give 1-oxides 103 and
104, which were oxidized to the TTOs 20 and 21, respectively.
Figure 1
benzotriazine 1,4-dioxide core. With the exception of several
studies on the parent 1,2,4-triazine dioxides56,57 and isosteric
cyanoquinoxaline 1,4-dioxides,58-61 there has been little ex-
ploration of related heterocyclic dioxides as hypoxia-selective
cytotoxins. In this study, we have synthesized 40 novel tricyclic
triazine 1,4-dioxides (TTOs) 3-42 (Figure 1) and examined
their in vitro activity as hypoxia-selective cytotoxins. We have
focused on using lipophilic ring systems of an electron-donating
nature in an effort to increase EVT by more rapid diffusion
through tumor tissue and reduced rates of hypoxic metabolism.
We have used in vitro PK/PD modeling to assess the effect of
the structural changes on EVT, and this approach has identified
20 novel TTOs of diverse structure with promising in vitro
profiles as hypoxia-selective cytotoxins.
Access to the “[2,3-g]” isomer pattern was achieved by
diazotization of nitroaniline 100 in the presence of H3PO2,
reduction of the resulting nitrobenzofuran 105 to the intermediate
aniline, and protection as the acetamide 106, and nitration and
deprotection to give nitroaniline 107. Cyclization gave the
isomeric BTO 1-oxide 108, which was directly oxidized to the
1,4-dioxide 22 or elaborated, via the chloride 109, to 3-ami-
noalkylamino 1-oxides 110-112 and subsequently to the
corresponding TTOs 23-25.
The acetamide 114 of 3,4-methylenedioxoaniline (113)
underwent nitration to give nitroacetamide 115, which was
deprotected to give nitroaniline 116 (Scheme 6). Cyclization
of 116 gave the 1-oxide 117, which was converted to the
3-chloride 118, displaced with amine side chain to give 119
and oxidized to TTO 26.
Nitration of 4-chromanone (120) gave predominantly the
6-nitro isomer 122, which was reduced and acetylated to give
acetamide 123 in moderate (56%) yield (Scheme 7). Alterna-
tively, zinc reduction of 120 gave chroman 124, which
underwent Friedel-Crafts acylation to give 125 with subsequent
oxime formation and Beckmann rearrangement to 123. Although
lower yielding (41%), this route was more convenient on a large
scale. Nitration of 123 gave a ca. 1:1 ratio of the two isomeric
nitroacetamides 126 and 127, which were deprotected to give
corresponding nitroanilines 128 and 133. Cyclization of 128
gave 1-oxide 129, which was converted, via chloride 130, to
1-oxides 131 and 132 and subsequently oxidized to 1,4-dioxides
27 and 28. A similar sequence converted the isomeric nitroa-
niline 133 to TTO 29.
Chemistry
Synthesis. Our synthetic strategies build on well established
methodology for the formation of the 3-amino-1,2,4-benzotri-
azine 1-oxide core48,50 and elaboration to 3-alkyl analogues,62,63
and these general methods are described below. Bicyclic
nitroanilines were treated with cyanamide and cHCl to generate
the intermediate guanidines, which were cyclized under strongly
basic conditions to give 3-aminotriazine 1-oxides (Scheme 1).
Oxidation of 3-aminotriazine 1-oxides to 3-amino TTOs was
achieved with peracetic acid. Conversion of 3-aminotriazine
1-oxides to 3-chlorotriazine 1-oxides was achieved by diazo-
tization in trifluoroacetic acid, followed by chlorination of the
intermediate phenol. Displacement of 3-chlorides with a variety
of amines gave the 3-aminoalkyltriazine 1-oxides. 3-Aminot-
riazine 1-oxides were also converted to 3-iodides by diazoti-
zation with tBuNO2 in THF and iodination with CH2I2. Reaction
of 3-halotriazines with various stannanes using Stille conditions
gave 3-alkyltriazine 1-oxides. Alternatively, reaction of 3-iodides
with allyl alcohol under Heck conditions gave 3-alkyltriazine
1-oxides. Oxidation of 3-substituted triazine 1-oxides with
trifluoroperactic acid, using an excess of trifluoroacetic acid to
protect any aliphatic amines present, gave a variety of TTOs.
In particular, our first set of targets included cyclopentane-,
cyclohexane-, and cycloheptane-fused benzotriazine dioxides
with a range of 3-aminoalkyl substituents bearing solubilizing
side chains. Thus, acetylation of 5-aminoindane (43) gave
acetamide 44, which was nitrated to give isomeric nitroaceta-
mides 45-47 (Scheme 2). Deprotection of nitroacetamide 45

Tricyclic [1,2,4]Triazine 1,4-Dioxides
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 21 6855
Scheme 2a
a Reagents: (i) Ac2O, dioxane; (ii) KNO3, H2SO4; (iii) 5 M HCl, ∆; (iv) NH2CN, HCl, ∆, then 30% NaOH, ∆; (v) CH3CO3H, CH3CO2H; (vi) NaNO2,
CF3CO2H; (vii) DMF, POCl3, ∆; (viii) RCH2CH2NH2, DME, ∆; (ix) CF3CO3H, CF3CO2H, DCM.
Scheme 3a
Scheme 4a
a Reagents: (i) cHNO3; (ii) H2, Pd/C, cHCl, EtOH; (iii) Ac2O, dioxane;
(iv) cHNO3, CF3CO2H; (v) cHCl, EtOH, ∆; (vi) NH2CN, HCl, ∆, then
30% NaOH, ∆; (vii) NaNO2, CF3CO2H; (viii) DMF, POCl3, ∆; (ix)
RCH2CH2NH2, DME, ∆; (x) CF3CO3H, CF3CO2H, DCM.
Recent work49 had shown that analogues with a soluble side
chain appended to the benzo ring of BTOs displayed excellent
hypoxic cytotoxicity and selectivity, so we explored analogues
with solubilizing functionality linked to, or part of, the saturated
ring of the TTOs. Thus mesylation of 2-indanol (137) and
displacement with dimethylamine gave amine 138, which was
nitrated to give 5-nitroindanamine 139 (Scheme 8). Reduction
and acetylation gave 140, nitration of which gave an inseparable
mixture of nitroacetamide isomers. Deprotection and frac-
tional crystallization gave the single nitroaniline isomer 141
in 50% yield for the two steps. Cyclization of 141 gave
1-oxide 142, which was converted to 3-chloride 143,
displaced with ethylamine to give 3-amine 144, and selec-
tively oxidized to TTO 30.
a Reagents: (i) fHNO3, H2SO4; (ii) H2, Pd/C, cHCl, EtOH; (iii) Ac2O,
dioxane; (iv) KNO3, H2SO4; (v) 5 M HCl, ∆; (vi) NH2CN, HCl, ∆, then
30% NaOH, ∆; (vii) NaNO2, CF3CO2H; (viii) DMF, POCl3, ∆; (ix)
RCH2CH2NH2, DME, ∆; (x) CF3CO3H, CF3CO2H, DCM.
amination of 7-nitro-tetrahydroisoquinoline 152 with acetic
formic anhydride gave 153. Reduction of the nitro group and
protection gave acetamide 154, nitration of which gave an
inseparable mixture of nitroacetamides. Deprotection allowed
purification of the isomeric nitroanilines 155 and 156. Cycliza-
tion of nitroaniline 156 gave 1-oxide 157, which was converted
to TTO 32 in a similar manner to 142.
3-Alkyl BTOs displayed increased hypoxic potency compared
to the analogous 3-aminoalkyl BTOs but often had reduced
aqueous solubility.49 To avoid this limitation with analogous
Nitration of bis(bromomethyl)benzene (145) gave 4-nitroben-
zene 146, which was cyclized to 2-ethylisoindole 147. Reduction
of 147 and acetylation gave acetamide 148, which was nitrated
and deprotected to give nitroaniline 150. Cyclization of 150 gave
1-oxide 151, which was oxidized to TTO 31. Reductive

6856 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 21
Hay et al.
Scheme 5a
give nitroaniline 195. Cyclization of 195 gave 1-oxide 196,
which was diazotized and converted to the 3-iodide 197.
Protection of the side chain as the THP ether 198 allowed Stille
coupling to form 3-ethyl 1-oxide 199. Deprotection of the THP
ether gave alcohol 200, which was oxidized to TTO 38. Alcohol
200 was further functionalized by reaction with mesyl chloride
and morpholine to give morpholide 201, which was selectively
oxidized to TTO 39.
A series of 3-alkyl TTOs were also explored bearing
heterocyclic saturated rings to potentially balance the effect of
the 3-alkyl group on reduction potential and consequently rates
of hypoxic metabolism. Conversion of dihydrofuran 1-oxide 108
to the iodide 109 followed by Heck coupling with allyl alcohol
gave aldehyde 203 (Scheme 13). This aldehyde underwent
reductive amination to give amine 204, which was oxidized to
TTO 40. Stille reaction of the tetrahydroisoquinoline 3-chloride
158 gave the 3-ethyl 1-oxide 205, which was oxidized to TTO
41. Cyclization of the nitroaniline 155 gave the 1-oxide 206,
which was converted to the chloride 207. Stille coupling to gave
3-ethyl 1-oxide 208, which was oxidized to TTO 42.
One Electron Reduction Potential Measurements, E(1).
Pulse radiolysis experiments were performed on the University
of Auckland Dynaray 4 (4 MeV) linear accelerator (200 ns pulse
length with a custom-built optical radical detection system).64
E(1) values for the TTOs were determined in anaerobic aqueous
solutions containing 2-propanol (0.1 M) buffered at pH 7.0 (10
mM phosphate) by measuring the equilibrium constant for the
electron transfer between the radical anions of the compounds
and the appropriate viologen or quinone reference standard.65
Data were obtained at three concentration ratios at room
temperature (22 ( 2 °C) and used to calculate the ∆E between
the compounds and references, allowing for ionic strength effects
on the equilibria.
a Reagents: (i) fHNO3, cH2SO4; (ii) H2, Pd/C, cHCl, EtOH; (iii) Ac2O,
dioxane; (iv) KNO3, H2SO4; (v) 5 M HCl, ∆; (vi) NH2CN, HCl, ∆, then
30% NaOH, ∆; (vii) NaNO2, CF3CO2H; (viii) DMF, POCl3, ∆; (ix)
Me2NCH2CH2NH2, DME, ∆; (x) CF3CO3H, CF3CO2H, DCM.
TTOs, we sought to maintain aqueous solubility by attaching a
solubilizing moiety via the 3-position or attached to the third
(saturated) ring of the TTO. Diazotization of the 3-amino
1-oxide 70 and reaction with CH2I2 gave the iodide 160, which
underwent a Heck reaction with allyl alcohol using Pd(OAc)2
as the catalyst to give aldehyde 161 (Scheme 9). Reductive
amination with morpholine gave both the terminal alcohol 162
as well as the tertiary morpholide 163. Oxidation of alcohol
162 and amine 163 gave TTOs 33 and 34, respectively.
Two routes were explored to synthesize nitroaniline 168
(Scheme 10). Reaction of 1,2-bis(bromomethyl)-4-nitrobenzene
(146) with diethyl malonate, with subsequent hydrolysis and
decarboxylation giving indane-2-carboxylic acid 164, which was
reduced to the corresponding alcohol 165, protected as the N,O-
diacetyl compound 166 and nitrated to give nitroacetamide 167.
Low yields in the first steps of this sequence prompted us to
explore an alternative. Thus, hydrolysis of nitrile 174 gave the
carboxylic acid 175, nitration of which gave an inseparable
mixture of isomeric nitroindane 2-carboxylic acids 164 and 176.
Reduction of this mixture gave a mixture of alcohols 165 and
177, which underwent further reduction, protection as the N,O-
diacetyl compounds, and subsequent nitration to give nitroac-
etamide 167 in 37% overall yield for the six steps. Deprotection
of 167 gave nitroaniline 168, which was cyclized to 3-amino
1-oxide 169. Protection of 1-oxide 169 as the TBDMS ether
170 and iodination gave 3-iodide 171, which underwent Stille
coupling with SnEt4 and Pd(PPh3)4 to give the 3-ethyl derivative
173. Oxidation and simultaneous deprotection of 173 gave TTO
35. The 3-ethyl silyl ether 173 was also deprotected to alcohol
178, and reaction with mesyl chloride and morpholine gave the
tertiary amine 179. Selective aromatic N-oxidation of 179 gave
TTO 36.
Similarly, Stille reaction of 171 with allyltributyltin gave
alkene 180, which underwent hydroboration with 9-borabi-
cyclo[3.3.1]nonane (9-BBN) to give the primary alcohol 181
(Scheme 11). Mesylation of 181 and displacement by morpho-
line gave amine 182, which was deprotected to alcohol 183 and
oxidized to TTO 37.
Condensation of indanone 184 with glyoxylic acid gave the
enone acid 185, which was reduced to acid 186 (Scheme 12).
Esterification of 186, reduction to the alcohol 188 with LiAlH4,
and acetylation gave 189. Nitration of 189 gave a mixture of
nitro isomers 190 and 191, which were reduced to the corre-
sponding acetamides 192 and 193. Further nitration of the
mixture gave the single isomer 194, which was deprotected to
Physicochemical Measurements. Solubilities of the TTOs
3-42 were determined in culture medium containing 5% fetal
calf serum at 22 °C. Octanol/water partition coefficients were
measured at pH 7.4 (P7.4) for TPZ, 2, and a subset of four TTOs
by a low volume shake flask method, with TTO concentrations
in both the octanol and buffer phases analyzed by HPLC as
previously described.48,66 These values, in conjunction with
previously determined values for related BTOs,47-50 were used
to “train” ACD LogP/LogD prediction software (v. 8.0,
Advanced Chemistry Development Inc., Toronto, Canada) using
a combination of ACD/LogP System Training and Accuracy
Extender. Apparent (macroscopic) pKa values for the side chain
were calculated using ACD pKa prediction software (v. 8.0,
Advanced Chemistry Development Inc., Toronto, Canada).
Biological Assays. Cytotoxic potency was determined by IC50
assays, using 4 h drug exposure of HT29 and SiHa cells under
aerobic and anoxic (H2/Pd anaerobic chamber) conditions in
96-well plates as described previously.50 The hypoxic cytotox-
icity ratio (HCR) was calculated as the intraexperiment ratio
aerobic IC50/anoxic IC50 (Tables 1-3).
The relationship between cell killing, drug exposure, and drug
metabolism (i.e., the in vitro PK/PD model) was measured as
previously described43,44 by following the clonogenicity of
stirred single cell suspensions of HT29 cells for 3 h, at a drug
concentration giving approximately one log kill by 1 h, with
monitoring of drug concentrations in extracellular medium by
HPLC. These concentration-time data were fitted to determine
the apparent first-order rate constant for metabolic consumption,
kmet. The measured parameter values are given in Tables 1-3,
and the associated error estimates are tabulated in the Supporting
Information. The best fit PK/PD model,44,48,49 the CT10 (area

Tricyclic [1,2,4]Triazine 1,4-Dioxides
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 21 6857
Scheme 6a
a Reagents: (i) AlCl3, AcCl, DCM; (ii) NH2OH·HCl, pyridine, then HCl, Ac2O, HOAc; (iii) cHNO3, HOAc; (iv) cHCl, EtOH, ∆; (v) NH2CN, HCl, ∆,
then 30% NaOH, ∆; (vi) NaNO2, CF3CO2H; (vii) DMF, POCl3, ∆; (viii) R2CH2CH2NH2, DME, ∆; (ix) CF3CO3H, CF3CO2H, DCM; (x) NaNO2, cH2SO4,
then H3PO2; (xi) H2, PtO2, THF, EtOH; (xii) Ac2O, dioxane; (xiii) CH3CO3H, HOAc; (xiv) NaOMe, MeOH, ∆.
Scheme 7a
a Reagents: (i) KNO3, cH2SO4; (ii) H2, Pd/C, aq HCl, EtOAc/EtOH; (iii) Ac2O, dioxane; (iv) Zn, HOAc, ∆; (v) AlCl3, AcCl, DCM; (vi) NH2OH·HCl,
pyridine, then HCl, Ac2O, HOAc; (vii) fHNO3, HOAc; (viii) cHCl, EtOH, ∆; (ix) NH2CN, HCl, ∆, then 30% NaOH, ∆; (x) NaNO2, CF3CO2H; (xi) DMF,
POCl3, ∆; (xii) R2CH2CH2NH2, DME, ∆; (xiii) CF3CO3H, CF3CO2H, DCM.
under the concentration-time curve providing a 10% surviving
fraction) and M10 (amount of BTO metabolized for a 10%
surviving fraction), were estimated by interpolation and are
tabulated in the Supporting Information along with the associ-
ated error estimates.
Tissue diffusion coefficients, D, of six TTOs were measured
using HT29 MCLs as previously described under 95% O2 to
suppress bioreductive metabolism.43 These values, along with
previously reported measurements for BTO analogues44,47 and
data for other BTOs (a total of 73 compounds), were used to
develop a multiple regression model to calculate D for the other
analogues as described previously.48 This model extends the
reported47 dependence, for neutral BTOs, of D on logP and
molecular weight (MW) by replacing logP with the octanol/
water distribution coefficient at pH 7.4 (logP7.4) and by adding
terms for the numbers of hydrogen bond donors (HD) and
acceptors (HA) (eq 1).67 The calculated and measured values
and the associated error estimates are tabulated in the Supporting
Information. The calculated values for all TTOs are shown in
Tables 1-3.
log(DMCL) ) a + b log MW +
c
(1)
log P7.4 - x + y · HD + z · HA
1 + exp
(
)
w

6858 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 21
Hay et al.
Scheme 8a
a Reagents: (i) MsCl, iPr2Net, DCM; (ii) aq NHMe2, DMF, ∆; (iii) cHNO3, CF3CO2H; (iv) H2, Pd/C, aq HCl, EtOAc/EtOH; (v) Ac2O, Et3N, DCM; (vi)
fHNO3, HOAc; (vii) HCl, EtOH, ∆; (viii) NH2CN, HCl, ∆, then 30% NaOH, ∆; (ix) NaNO2, CF3CO2H; (x) DMF, POCl3, ∆; (xi) EtNH2, DME, ∆; (xii)
CF3CO3H, CF3CO2H, DCM; (xiii) KNO3, cH2SO4; (xiv) EtNH2 ·HCl, Et3N, DMF, ∆; (xv) HCO2H, Ac2O, THF, then BH3 ·DMS.
Scheme 9a
Scheme 10a
a Reagents: (i) t-BuNO2, CH2I2, CuI, THF, ∆; (ii) AllylOH, Pd(OAc)2,
nBu4NBr, NaHCO3, DMF, ∆; (iii) morpholine, MeOH, then NaCNBH3,
HOAc; (iv) CF3CO2H, CF3CO3H, DCM.
A one-dimensional EVT parameter, the penetration distance
into hypoxic tissue assuming planar geometry (X1/2, eq 2), was
calculated from the opposing effects of D and kmet on tissue
transport as previously reported,48 providing a ready comparison
between analogues; values are given in Tables 1-3, and the
associated error estimates are tabulated in the Supporting
Information.
D
X1/2 ) ln(2)
(2)
k
ꢀ
met
a Reagents: (i) NaH, (EtO2C)2CH2, Et2O; (ii) NaOH, EtOH; (iii) xylene,
∆; (iv) BH3 ·DMS, THF; (v) H2, Pd/C, MeOH; (vi) Ac2O, Et3N, DCM;
(vii) cHNO3, CF3CO2H; (viii) 5 M HCl, MeOH, ∆; (ix) NH2CN, HCl, ∆,
then 30% NaOH, ∆; (x) TBDMSCl, iPr2NEt, DMF; (xi) t-BuNO2, CH2I2,
CuI, THF, ∆; (xii) Et4Sn, Pd(PPh3)4, DME, ∆; (xiii) 1 M HCl, MeOH, ∆;
(xiv) CF3CO3H, DCM; (xv) MsCl, iPr2NEt, DCM, then morpholine, DMF,
∆; (xvi) cHCl, dioxane, ∆.
PK/PD Modeling. The in vivo 3D PK/PD model has been
described in detail recently.45,48 Briefly, transport is modeled
in the extravascular compartment of a representative tumor
microvascular network by solving the Fick’s Second Law
diffusion-reaction equations using a Green’s function method,
providing a description of the PK at each point in the tissue
region. The transport parameters used in the model are the tissue
diffusion coefficient D (estimated in MCLs) and kmet for
bioreductive drug metabolism under anoxia (scaled to MCL cell
density) determined in vitro as described above. Using the
homogeneous PK/PD model established in vitro for each
compound, the log cell kill (LCK) was predicted at each position
in the tumor microregion. The overall cell kill through the whole
region was then calculated for drug only and for drug plus 20
Gy radiation (using the reported radiosensitivity parameters for
aerobic and hypoxic HT29 cells).44 The difference between these
[LCK(drug+RAD) - LCK(drug alone)] gives the model-predicted logs
of hypoxic cell kill (LCKpred). The in vitro PK/PD relationship
was used to predict the area under the plasma concentration-time
curve (AUCreq) that would be required to give 1 log of cell kill
in addition to that produced by a single 20 Gy dose of
γ-radiation. In addition, the hypoxic cytotoxicity differential
(HCD) is calculated as a measure of hypoxic selectivity in the
tumor:
LCKpred in the hypoxic region (<4µM O2)
HCD )
LCKpred in the well oxygenated region (>30µM O2)
(3)
where LCK is predicted for the drug alone.

Tricyclic [1,2,4]Triazine 1,4-Dioxides
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 21 6859
Compounds with high potency (low plasma AUCreq) and high
in vivo hypoxic selectivity (HCD) have potential to demonstrate
improved in vivo hypoxic cell kill compared to TPZ. Compari-
son using these criteria was made to evaluate the potential of
the analogues as improved analogues of TPZ.
Results and Discussion. Targeting the cycloalkane ring
systems 3-19 (Table 1) using the established benzotriazine ring
formation was readily achieved. The cycloalkyl rings conferred
increased lipophilicity relative to TPZ. The low solubility of 3
was remedied by addition of an aminoalkylamine at the
3-position, e.g., 4 (Table 1). This effect was general, with only
the piperidine 7 failing to show increased solubility.
We have previously demonstrated48 that replacing the strongly
electron-donating 3-NH2 (σp ) -0.66) with the dimethylami-
noethylamine side chain raises the one electron reduction
potential, E(1), of TPZ by 60 mV. This can be offset by electron-
donating substituents at the 6-position, with a 6-Me substituent
lowering the E(1) by 40 mV and a 6-OMe substituent, e.g., 2,
lowering the E(1) by 104 mV to a value of -500 mV. Thus,
the cycloalkane rings (4, 9, and 19) lowered the E(1) by ca. 90
mV and the inclusion of a lower pKa amine (13) resulted in a
further reduction in E(1).
The fusion of a cycloalkane ring at the 6,7- or 7,8-positions
of the BTO nucleus provided TTOs with similar hypoxic
cytotoxicity and selectivity to TPZ with the main variation in
hypoxic potency resulting from variation in amine side chain
pKa. Thus the morpholides 8, 13, and 15 were considerably less
potent than more basic analogues. This drop in potency was
also reflected in lowered hypoxic metabolism, with the cyclo-
hexyl analogue 17 showing unusually high hypoxic metabolism
and potency relative to the other morpholides. The increased
lipophilicity from the cycloalkyl rings resulted in increased
diffusion relative to TPZ. Consequently, most of these TTOs
showed increased EVT as defined by the 1-D transport pa-
rameter X1/2
.
Figure 2. (a) Predicted hypoxic selectivity (HCD) and potency
(AUCreq) for BTOs. (black circles) Compounds predicted inactive; (red
solid upward-pointing triangles) active in vivo; (red open upward-
pointing triangles) not tested due to structural similarity to actives; (blue
solid downward-pointing triangles) not active in vivo. Data from refs
48 and 49. (b) Predicted hypoxic selectivity (HCD) and potency
(AUCreq) for TTOs. (black circles) TTOs from Table 1; (blue solid
upward-pointing triangles) TTOs from Table 2; (red solid downward-
pointing triangles) TTOs from Table 3.
Calculation of the predicted AUC required to achieve 1 log
of hypoxic cell kill in tumors (AUCreq) and the in vivo hypoxic
selectivity (HCD) allowed comparison with previously studied
BTO analogues.48,49 In these studies, analogues with high
predicted hypoxic selectivity (HCD) and high predicted in vivo
potency (AUCreq) were most likely to be active in vivo (upper-
left quadrant, Figure 2a). A number of compounds predicted to
be active but were not tested because they were structurally
similar to other actives (red open upward-pointing triangles).
The predicted activity is subject to the in vivo toxicity and
plasma pharmacokinetics and in a few examples very low
plasma AUC values precluded activity (blue solid downward-
pointing triangles). Using these data as a guide (HCD > 6,
AUCreq < 14000 µM·min), six cycloalkyl-TTOs (4, 6, 8, 11,
15, 17) demonstrated the potential for in vivo activity (black
circles, Figure 2b).
Scheme 11a
The inclusion of a heteroatom in the fused ring resulted in
TTOs with considerably lower lipophilicity than the correspond-
ing cycloalkyl analogues (e.g., compare 20 or 23 with 4; 27
with 16) (Table 2). Again the inclusion of a basic side chain
conferred excellent aqueous solubility, with the neutral TTO
22 displaying very low solubility. The indanamine 30 was not
stable in medium and was not evaluated in vitro.
The oxygen atom of the [3,2-g]dihydrofuran 20 had little
effect on E(1) compared to 4, whereas the [2,3-g] isomer 23
shows a stronger influence and a similar effect is seen with the
dioxole 26 having a similar E(1) to 23 (Table 2). This reflects
the stronger electronic influence of substituents at the “6”-
position compared to the “7”-position.50
a Reagents: (i) AllylSnBu3, Pd(PPh3)4, DME, ∆; (ii) 9-BBN, THF, then
NaOAc, H2O2; (iii) MsCl, iPr2NEt, DCM, then morpholine, DMF; (iv) 1
M HCl, MeOH; (v) CF3CO3H, CF3CO2H,DCM.
The heterocyclic TTOs showed similar hypoxic potency to
TPZ and slightly lower selectivity, with the weakly basic
morpholides again displaying reduced potency. The TTO 22
showed low hypoxic potency and selectivity and failed to pass

6860 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 21
Hay et al.
Scheme 12a
a Reagents: (i) 50% aq HCOCO2H, cH2SO4, dioxane, ∆; (ii) H2, Pd/C, MeOH, dioxane; (iii) cH2SO4, EtOH; (iv) LiAlH4, THF; (v) Ac2O, pyridine,
DMAP, DCM; (vi) Cu(NO3)2 ·3H2O, Ac2O; (vii) Ac2O, dioxane; (viii) cHNO3, CF3CO2H; (ix) 5 M HCl, MeOH, ∆; (x) NH2CN, HCl, ∆, then 30% NaOH,
∆; (xi) t-BuNO2, I2, CuI, THF, ∆; (xii) dihydropyran, PPTS, DCM; (xiii) Et4Sn, Pd(PPh3)4, DME, ∆; (xiv) MeSO3H, MeOH; (xv) CF3CO2H, CF3CO3H,
DCM; (xvi) MsCl, iPr2NEt, DCM, then morpholine, DMF, ∆.
Scheme 13a
the criterion for hypoxic selectivity (HCR > 20) in the PK/PD
model48,49 and was not evaluated further. The decreased
lipophilicity of the heterocyclic TTOs 20-32 resulted in lowered
diffusion, with only the weakly basic amines 28, 30, and 32
having D values greater than TPZ. The rates of hypoxic
metabolism were influenced by the electronic nature of the fused
ring and the side chain amine pKa, with the weakly basic
morpholides generally having lower kmet values compared to
more basic analogues. For TTOs with a 7-oxa atom in the ring,
e.g., 20 and 27, kmet values were relatively high, reflecting the
weaker dependence of E(1) on σp for 7-substituents.50 Only
heterocyclic TTOs with very low kmet values achieved a balance
with the low D values, and thus only 21, 24, and 25 were
predicted to display improved EVT based on their X1/2 values.
Low hypoxic potency and kmet contributed to high AUCreq
values, which when combined with modest predicted in vivo
hypoxic selectivity (HCD) as a consequence of modest D values,
resulted in only heteroalkyl-TTO 24, satisfying the criteria for
predicted in vivo activity (AUCreq < 14000; HCD > 6, blue
solid upward-pointing triangle, Figure 2b).
A series of 3-alkyl TTOs 33-40 with solubilizing groups
attached via the 3-alkyl substituent or the 7-position of the
indane ring was prepared. Two examples where the solubilizing
moiety was included within the isoquinoline ring (41, 42) were
a Reagents: (i) t-BuNO2, CH2I2, CuI, THF, ∆; (ii) allyl alcohol, Pd(OAc)2,
also prepared (Table 3). The strategy was to increase EVT
nBu4NBr, NaHCO3, DMF; (iii) morpholine, NaCNBH3, MeOH, DMF; (iv)
through increased lipophilicity and to optimize the hypoxic
metabolism by balancing the electron-donating effects of the
fused rings. Replacement of the 3-aminoalkyl substituent with
a 3-ethyl group (e.g., 35 compared with 4) raised the reduction
potential by ca. 30 mV, and the inclusion of a morpholine group
on the side chain further raised the E(1) by ca. 40 mV. A similar
elevation of E(1) was seen with the morpholide 40 when
compared to 21. Placement of the morpholino group in a remote
position off the indane ring (39) had a negligible effect on E(1).
TTOs 33-42 were generally more lipophilic than similar
3-alkylamino analogues, with only the morpholide 40 showing
significantly reduced lipophilicity relative to 25 due to the
polarity of the dihydrofuran ring. Morpholide side chains
provided increased aqueous solubility relative to TPZ, whereas
the neutral analogues had reduced solubility. TTO 42 was not
evaluated because of instability in culture medium. The neutral
TTOs were generally less potent than related analogues bearing
morpholide side chains, which had similar hypoxic cytotoxicity
to TPZ. TTOs in this series were hypoxia-selective with the
CF3CO3H, CF3CO2H, DCM; (v) Et4Sn, Pd(PPh3)4, DME, ∆; (vi) NH2CN,
HCl, ∆, then 30% NaOH, ∆; (vii) NaNO2, CF3CO2H; (viii) DMF,
POCl3, ∆.
exception of 41, which was not evaluated further. The removal
of the 3-NH group resulted in large increases in diffusion
coefficient with the addition of a hydrogen bond donor (OH)
having a similar negative effect to that of a morpholide group
(e.g., compare 33 with 34, 35 with 36, and 38 with 39).
Hydrogen bond donors such as hydroxyl groups, although only
modestly lowering the logP7.4, have a strong negative influence
on diffusion rates.47 The strongly polar nature of the [2,3-g]-
dihydrofuran moiety dominated in TTO 40, leading to a low
D value. The cycloalkyl TTOs 33-39 displayed similar rates
of hypoxic metabolism to TPZ with the presence of a
morpholide side chain, producing a small elevation in kmet
.
The increased diffusion coefficients combined with modest
rates of hypoxic metabolism gave significantly increased EVT
as defined by X1/2. These factors combined to give relatively

Tricyclic [1,2,4]Triazine 1,4-Dioxides
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 21 6861
Table 1. Physicochemical, in Vitro, and Modelling Parameters for TPZ, BTO 2, and TTOs 3-19
a Calculated using ACD pKa. b Calculated using ACD logD. c Solubility of HCl salts in culture medium. d Hypoxia cytotoxicity ratio ) oxic IC50/hypoxic
IC50. e Diffusion coefficient in HT29 MCLs × 10-7 cm2s-1
.
f Error estimates are provided in the Supporting Information. g First-order rate constant for
metabolism in anoxic HT29 cell suspensions, scaled to the cell density in MCLs. h Penetration half-distance in anoxic HT29 tumor tissue (see text). i Predicted
area under the plasma concentration-time curve required to give 1 log of cell kill in addition to that produced by a single 20 Gy dose of γ radiation. j In
vivo hypoxic cytotoxicity differential ) LCKhypoxic/LCKoxic.
k Data from ref 48. l Not applicable.
modest AUCreq values and high HCD across the series with
TTOs (33, 35, 36, 38, 39) identified as likely to be active in
vivo (AUCreq < 14000; HCD > 6.0) (red solid downward-
pointing triangles, Figure 2b).
The addition of bulky fused rings, either in a linear or angular
manner, or the placement of a polar hydroxyl or charged amine
solubilizing group in either hemisphere of the molecule resulted
in TTOs that were hypoxia-selective. This wide substrate
tolerance is consistent with the description of cytochrome P450
reductase (CYP450R) as the major reductase responsible for
bioactivation of TPZ.18 CYP450R’s prime role is to reduce the
cofactor in cytochrome P450 and consequently it has a wide cleft
allowing access of the substrate enzyme to the active site.68,69
For this enzyme interaction, electron transfer occurs with
minimal overlap of cofactors and is driven by the difference in
reduction potential between the substrate and the cofactor.

6862 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 21
Hay et al.
Table 2. Physicochemical, in Vitro, and Modelling Parameters for TPZ and TTOs 20-32
a Calculated using ACD pKa. b Calculated using ACD logD. c Solubility of HCl salts in culture medium. d Hypoxia cytotoxicity ratio ) oxic IC50/hypoxic
IC50. e Diffusion coefficient in HT29 MCLs × 10-7 cm2s-1
.
f Error estimates are provided in the Supporting Information. g First-order rate constant for
metabolism in anoxic HT29 cell suspensions, scaled to the cell density in MCLs. h Penetration half-distance in anoxic HT29 tumor tissue (see text). i Predicted
area under the plasma concentration-time curve required to give 1 log of cell kill in addition to that produced by a single 20 Gy dose of γ radiation. j In
vivo hypoxic cytotoxicity differential ) LCKhypoxic/LCKoxic.
k Data from ref 48. l Not applicable.
Optimization of rates of hypoxic metabolism and tissue
diffusion coefficients has been possible for at least 12
different TTO analogues, and these compounds are predicted
to be active in vivo. Key to the ongoing development of this
class of compounds is the determination of SAR for animal
toxicity (maximum tolerated dose) and in vivo pharmacoki-
netics. The attainment of a sufficiently high Cmax and AUC
has been demonstrated as necessary for activity against
hypoxic tumor cells in vivo.45,48,49 Studies are currently
underway with the 12 candidates predicted to be active in
vivo to identify SAR for in vivo toxicity and pharmacokinetic
parameters.
of structural arrangements with cycloalkyl, oxygen-, and
nitrogen-containing rings fused in a linear or angular fashion
to the benzotriazine core, coupled with neutral, polar, and
charged side chains linked to either hemisphere, resulted in
TTO analogues that displayed hypoxia-selective cytotoxicity
in vitro.
The fused cycloalkyl rings are sufficiently electron-
donating in combination with both the 3-aminoalkyl or
3-alkyl substituents to position the one electron reduction
potential in an appropriate range for optimal rates of hypoxic
metabolism. The stronger electronic and polar influences of
the “oxa” rings were more difficult to balance and mostly
led to poor EVT properties. The use of amine containing
rings or substituents, while providing increased aqueous
solubility, provided either unstable TTOs or only modest
activity. The lipophilic nature of the cycloalkyl rings also
increased lipophilicity and led to increased diffusion coef-
Conclusions
We have been able to build on previously described SAR
to design a wide range of novel tricyclic TTO analogues and
explore the effect of these structural modifications on their
in vitro activity as hypoxia-selective cytotoxins. A wide range

Tricyclic [1,2,4]Triazine 1,4-Dioxides
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 21 6863
Table 3. Physicochemical, in Vitro, and Modelling Parameters for TPZ and TTOs 33-42
a Calculated using ACD pKa. b Calculated using ACD logD. c Solubility of HCl salts in culture medium. d Hypoxia cytotoxicity ratio ) oxic IC50/hypoxic
IC50. e Diffusion coefficient in HT29 MCLs × 10-7 cm2s-1
.
f Error estimates are provided in the Supporting Information. g First-order rate constant for
metabolism in anoxic HT29 cell suspensions, scaled to the cell density in MCLs. h Penetration half-distance in anoxic HT29 tumor tissue (see text). i Predicted
area under the plasma concentration-time curve required to give 1 log of cell kill in addition to that produced by a single 20 Gy dose of γ radiation. j In
vivo hypoxic cytotoxicity differential ) LCKhypoxic/LCKoxic.
k Data from ref 48. l Not applicable.
ficients, which when combined with weakly basic morpholine
side chains gave the best balance of solubility and increased
chromatography, eluting with a gradient (0-10%) of MeOH/
DCM, to give the 1-oxide.
Preparation of 3-Chlorotriazine 1-Oxides: Methods (iii,
iv), Scheme 1. Sodium nitrite (10 mmol) was added in small
portions to a stirred solution of 1-oxide (5 mmol) in trifluoro-
acetic acid (20 mL) at 0 °C, and the solution was stirred at 20
°C for 3 h. The solution was poured into ice/water, stirred for
30 min, filtered, washed with water (3 × 10 mL), and dried.
The solid was suspended in POCl3 (20 mL) and DMF (0.2 mL)
and stirred at 100 °C for 1 h. The solution was cooled, poured
into ice/water, stirred for 30 min, filtered, washed with water
(3 × 30 mL), and dried. The solid was suspended in DCM (150
mL), dried, and the solvent evaporated. The residue was purified
by chromatography, eluting with 5% EtOAc/DCM, to give the
chloride.
Preparation of 3-alkylamino 1-Oxides: Method (vi), Scheme
1. Amine (3.0 mmol) was added to a stirred solution of chloride
(1.0 mmol) in DME (50 mL) and the solution stirred at reflux
temperature for 8 h. The solution was cooled to 20 °C, the
solvent evaporated, and the residue partitioned between aqueous
NH4OH solution (100 mL) and EtOAc (100 mL). The organic
fraction was dried and the solvent evaporated. The residue was
purified by chromatography, eluting with a gradient (0-10%)
of MeOH/DCM, to give the 1-oxide.
diffusion.
The selection was further refined using PK/PD model
predictions of the AUCreq and HCD and 12 TTOs were predicted
to be active in vivo subject to adequate plasma pharmacoki-
netics.
Experimental Section
Chemistry. General experimental details are described in the
Supporting Information. TPZ and BTO 2 were synthesized as
previously described.48
Example of Synthetic Methods. See Supporting Informa-
tion for full experimental details.
Preparation of 3-Aminotriazine 1-Oxides: Method (i),
Scheme 1. A mixture of nitroaniline (20 mmol) and cyanamide
(80 mmol) were mixed together at 100 °C, cooled to 50 °C,
cHCl (10 mL) added dropwise (CAUTION: exotherm), and the
mixture heated at 100 °C for 4 h. The mixture was cooled to
50 °C, 7.5 M NaOH solution added until the mixture was
strongly basic, and the mixture stirred at 100 °C for 3 h. The
mixture was cooled, diluted with water (100 mL), filtered,
washed with water (3 × 30 mL), washed with ether (2 × 5
mL), and dried. If necessary, the residue was purified by

6864 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 21
Hay et al.
(10) Brown, J. M.; Wilson, W. R. Exploiting tumor hypoxia in cancer
treatment. Nature ReV. Cancer 2004, 4, 437–447.
Preparation of 1-Oxides: Method (vii), Scheme 1. Pd-
(PPh3)4 (0.1 mmol) was added to a stirred, degassed solution
of halide (2.0 mmol) and stannane (2.4 mmol) in DME (20 mL),
and the solution was stirred under N2 at reflux temperature for
16 h. The solvent was evaporated, and the residue was dissolved
in DCM (10 mL) and stirred with saturated aqueous KF solution
(10 mL) for 30 min. The mixture was filtered through celite,
the celite was washed with DCM, and the combined organic
filtrate was washed with water. The organic fraction was dried,
the solvent evaporated, and the residue purified by chromatog-
raphy, eluting with DCM to give product, which was, if
necessary, further purified by chromatography, eluting with 20%
EtOAc/petroleum ether, to give the 3-alkyl 1-oxide.
Preparation of 1,4-Dioxides 3-42: Method (vii),
Scheme 1. Hydrogen peroxide (70%, 10 mmol) was added
dropwise to a stirred solution of trifluoroacetic anhydride (10
mmol) in DCM (20 mL) at 0 °C. The mixture was stirred at 0
°C for 5 min, warmed to 20 °C, stirred for 10 min, and cooled
to 5 °C. The mixture was added to a stirred solution of 1-oxide
(1.0 mmol) [and where aliphatic amine side chains are present,
TFA (5.0 mmol)] in DCM (15 mL) at 0 °C, and the mixture
was stirred at 20 °C for 4-16 h. The solution was carefully
diluted with water (20 mL) and the mixture made basic with
aqueous NH4OH solution, and the mixture was stirred for 15
min and then extracted with CHCl3 (5 × 50 mL). The organic
fraction was dried and the solvent evaporated. The residue was
purified by chromatography, eluting with a gradient (0-15%)
of MeOH/DCM, to give 1,4-dioxides.
(11) Nordsmark, M.; Overgaard, M.; Overgaard, J. Pretreatment oxygen-
ation predicts radiation response in advanced squamous cell carcinoma
of the head and neck. Radiother. Oncol. 1996, 41, 31–39.
(12) Fyles, A. W.; Milosevic, M.; Wong, R.; Kavanagh, M. C.; Pintilie,
M.; Sun, A.; Chapman, W.; Levin, W.; Manchul, L.; Keane, T. J.;
Hill, R. P. Oxygenation predicts radiation response and survival in
patients with cervix cancer. Radiother. Oncol. 1998, 48, 149–156.
(13) Koukourakis, M. I.; Bentzen, S. M.; Giatromanolaki, A.; Wilson, G. D.;
Daley, F. M.; Saunders, M. I.; Dische, S.; Sivridis, E.; Harris, A. L.
Endogenous markers of two separate hypoxia response pathways
(hypoxia inducible factor 2 alpha and carbonic anhydrase 9) are
associated with radiotherapy failure in head and neck cancer patients
recruited in the CHART randomized trial. J. Clin. Oncol. 2006, 24,
727–735.
(14) Denny, W. A.; Wilson, W. R.; Hay, M. P. Recent developments in
the design of bioreductive drugs. Br. J. Cancer 1996, 74, S32–S38.
(15) Brown, J. M.; Giaccia, A. J. The unique physiology of solid tumors:
opportunities (and problems) for cancer therapy. Cancer Res. 1998,
58, 1408–1416.
(16) Brown, J. M. SR 4233 (tirapazamine): a new anticancer drug exploiting
hypoxia in solid tumours. Br. J. Cancer 1993, 67, 1163–1170.
(17) Denny, W. A.; Wilson, W. R. Tirapazamine: a bioreductive anticancer
drug that exploits tumour hypoxia. Expert Opin. InVest. Drugs 2000,
9, 2889–2901.
(18) Patterson, A. V.; Saunders, M. P.; Chinje, E. C.; Patterson, L. H.;
Stratford, I. J. Enzymology of tirapazamine metabolism: a review. Anti-
Cancer Drug Des. 1998, 13, 541–573.
(19) Anderson, R. F.; Shinde, S. S.; Hay, M. P.; Gamage, S. A.; Denny,
W. A. Activation of 3-amino-1,2,4-benzotriazine 1,4-dioxide antitumor
agents to oxidizing species following their one-electron reduction.
J. Am. Chem. Soc. 2003, 125, 748–756.
(20) Shinde, S. S.; Anderson, R. F.; Hay, M. P.; Gamage, S. A.; Denny,
W. A. Oxidation of 2-deoxyribose by benzotriazinyl radicals of
antitumor 3-amino-1,2,4-benzotriazine 1,4-dioxides. J. Am. Chem. Soc.
2004, 126, 7853–7864.
Acknowledgment. We thank Jane Botting, Dr. Maruta Boyd,
Alison Hogg, Sisira Kumara, and Sarath Liyanage, for technical
assistance. We thank Degussa Peroxide Ltd, Morrinsville, New
Zealand for the generous gift of 70% hydrogen peroxide. This
work was supported by the U.S. National Cancer Institute under
grant CA82566 (M.P.H., K.P., K.O.H., F.B.P., W.R.W., W.AD.),
the Health Research Council of New Zealand (W.R.W., R.F.A.,
S.S.S.), and the Auckland Division of the Cancer Society of
New Zealand (W.A.D.). Further support from Proacta Thera-
peutics Ltd (H.H.L.) and the Australian Institute of Nuclear
Sciences and Engineering is acknowledged.
(21) Daniels, J. S.; Gates, K. S. DNA cleavage by the antitumor agent
3-amino-1,2,4-benzotriazine 1,4-dioxide (SR4233). Evidence for
involvement of hydroxyl radical. J. Am. Chem. Soc. 1996, 118, 3380–
3385.
(22) Chowdhury, G.; Junnotula, V.; Daniels, J. S.; Greenberg, M. M.; Gates,
K. S. DNA strand damage product analysis provides evidence that
the tumor cell-specific cytotoxin tirapazamine produces hydroxyl
radical and acts as a surrogate for O2. J. Am. Chem. Soc. 2007, 129,
12870–12877.
(23) Wang, J.; Biedermann, K. A.; Brown, J. M. Repair of DNA and
chromosome breaks in cells exposed to SR 4233 under hypoxia or to
ionizing radiation. Cancer Res. 1992, 52, 4473–4477.
(24) Siim, B. G.; van Zijl, P. L.; Brown, J. M. Tirapazamine-induced DNA
damage measured using the comet assay correlates with cytotoxicity
towards hypoxic tumour cells in vitro. Br. J. Cancer 1996, 73, 952–
960.
(25) Siim, B. G.; Menke, D. R.; Dorie, M. J.; Brown, J. M. Tirapazamine-
induced cytotoxicity and DNA damage in transplanted tumors:
relationship to tumor hypoxia. Cancer Res. 1997, 57, 2922–2928.
(26) Peters, K. B.; Brown, J. M. Tirapazamine: A Hypoxia-Activated
Topoisomerase II Poison. Cancer Res. 2002, 62, 5248–5253.
(27) Evans, J. W.; Chernikova, S. B.; Kachnic, L. A.; Banath, J. P.; Sordet,
O.; Delahoussaye, Y. M.; Treszezamsky, A.; Chon, B. H.; Feng, Z.;
Gu, Y.; Wilson, W. R.; Pommier, Y.; Olive, P. L.; Powell, S. N.;
Brown, J. M. Homologous recombination is the principal pathway
for the repair of DNA damage induced by tirapazamine in mammalian
cells. Cancer Res. 2008, 68, 257–265.
(28) Brown, J. M.; Lemmon, M. J. Potentiation by the hypoxic cytotoxin
SR 4233 of cell killing produced by fractionated irradiation of mouse
tumors. Cancer Res. 1990, 50, 7745–7459.
(29) Brown, J. M.; Lemmon, M. J. Tumor hypoxia can be exploited to
preferentially sensitize tumors to fractionated irradiation. Int. J. Radiat.
Oncol., Biol., Phys. 1991, 20, 457–461.
Supporting Information Available: Experimental details and
characterization data for synthetic intermediates 43-208 and TTOs
3-42; experimental details for the physicochemical and biological
methods; tables of physicochemical and in vitro data with estimates
of errors. This material is available free of charge via the Internet
at http://pubs.acs.org.
References
(1) Semenza, G. L. HIF-1 mediates the Warburg effect in clear cell renal
carcinoma. J. Bioenerg. Biomembr. 2007, 39, 231–234.
(2) Gatenby, R. A.; Gillies, R. J. Glycolysis in cancer: a potential target
for therapy. Int. J. Biochem. Cell Biol. 2007, 39, 1358–1366.
(3) Harris, A. L. Hypoxiasa key regulatory factor in tumour growth.
Nature ReV. Cancer 2002, 2, 38–47.
(4) Semenza, G. L. HIF-1 and tumor progression: pathophysiology and
therapeutics. Trends Mol. Med. 2002, 8, S62–S67.
(5) Pennachietti, S.; Michieli, P.; Galluzzo, M.; Mazzone, M.; Giordano,
S.; Comoglio, P. M. Hypoxia promotes invasive growth by transcrip-
tional activation of the met protooncogene. Cancer Cell 2003, 3, 347–
361.
(6) Cairns, R. A.; Hill, R. P. Acute hypoxia enhances spontaneous lymph
node metastasis in an orthotopic murine model of human cervical
carcinoma. Cancer Res. 2004, 64, 2054–2061.
(7) Subarsky, P.; Hill, R. P. The hypoxic tumour microenvironment and
metastatic progression. Clin. Exp. Metastases 2003, 20, 237–250.
(8) Durand, R. E. The influence of microenvironmental factors during
cancer therapy. In ViVo 1994, 8, 691–702.
(9) Tannock, I. F. Conventional cancer therapy: promise broken or promise
delayed? Lancet 1998, 351, 9–16.
(30) Dorie, M. J.; Brown, J. M. Tumor-specific, schedule-dependent
interaction between tirapazamine (SR 4233) and cisplatin. Cancer Res.
1993, 53, 4633–4636.
(31) Dorie, M. J.; Brown, J. M. Modification of the antitumor activity of
chemotherapeutic drugs by the hypoxic cytotoxic agent tirapazamine.
Cancer Chemother. Pharmacol. 1997, 39, 361–366.
(32) Marcu, L.; Olver, I. Tirapazamine: From Bench to Clinical Trials.
Curr. Clin. Pharmacol. 2006, 1, 71–79.
(33) Rischin, D.; Fisher, R.; Peters, L.; Corry, J.; Hicks, R. Hypoxia in
head and neck cancer: studies with hypoxic positron emission
tomography and hypoxic cytotoxins. Int. J. Radiat. Oncol., Biol., Phys.
2007, 69, S61–S63.

Tricyclic [1,2,4]Triazine 1,4-Dioxides
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 21 6865
(34) Rischin, D.; Peters, L.; O’Sullivan, B.; Giralt, J.; Yuen, K.; Trotti,
A.; Bernier, J.; Bourhis, J.; Henke, M.; Fisher, R. Phase III study of
tirapazamine, cisplatin and radiation versus cisplatin and radiation for
advanced squamous cell carcinoma of the head and neck. J. Clin.
Oncol. 2008, 26, abstract LBA6008.
(35) Rischin, D.; Hicks, R. J.; Fisher, R.; Binns, D.; Corry, J.; Porceddu,
S.; Peters, L. J. Prognostic significance of [18F]-misonidazole
positron emission tomography-detected tumor hypoxia in patients
with advanced head and neck cancer randomly assigned to
chemoradiation with or without tirapazamine: a substudy of Trans-
Tasman Radiation Oncology Group Study 98.02. J. Clin. Oncol.
2006, 24, 2098–2104.
relationships for benzotriazine di-N-oxides. Int. J. Radiat. Oncol., Biol.,
Phys. 1989, 16, 977–981.
(52) Minchinton, A. I.; Lemmon, M. J.; Tracy, M.; Pollart, D. J.; Martinez,
A. P.; Tosto, L. M.; Brown, J. M. Second-generation 1,2,4-benzotri-
azine 1,4-di-N-oxide bioreductive antitumor agents: pharmacology and
activity in vitro and in vivo. Int. J. Radiat. Oncol., Biol., Phys 1992,
22, 701–705.
(53) Kelson, A. B.; McNamara, J. P.; Pandey, A.; Ryan, K. J.; Dorie, M. J.;
McAfee, P. A.; Menke, D. R.; Brown, J. M.; Tracy, M. 1,2,4-
Benzotriazine 1,4-dioxides. An important class of hypoxic cytotoxins
with antitumor activity. Anti-Cancer Drug Des. 1998, 13, 575–592.
(54) Jiang, F.; Yang, B.; Fan, L.; Heb, Q.; Hu, Y. Synthesis and hypoxic-
cytotoxic activity of some 3-amino-1,2,4-benzotriazine-1,4-dioxide
derivatives. Bioorg. Med. Chem. Lett. 2006, 16, 4209–4213.
(55) Jiang, F.; Weng, Q.; Sheng, R.; Xia, Q.; He, Q.; Yang, B.; Hu, Y.
Synthesis, structure and hypoxic cytotoxicity of 3-amino-1,2,4-
benzotriazine-1,4-dioxide derivatives. Arch. Pharm. (Weinheim, Ger.)
2007, 340, 258–263.
(56) Cerecetto, H.; Gonzalez, M.; Onetto, S.; Saenz, P.; Ezpeleta, O.; De
Cerain, A. L.; Monge, A. 1,2,4-Triazine N-oxide derivatives: studies
as potential hypoxic cytotoxins. Part II. Arch. Pharm. (Weinheim, Ger.)
2004, 337, 247–258.
(57) Cerecetto, H.; Gonzalez, M.; Risso, M.; Saenz, P.; Olea-Azar, C.;
Bruno, A. M.; Azqueta, A.; De Cerain, A. L.; Monge, A. 1,2,4-Triazine
N-oxide derivatives: Studies as potential hypoxic cytotoxins. Part III.
Arch. Pharm. (Weinheim, Ger.) 2004, 337, 271–280.
(58) Monge, A.; Palop, J. A.; Gonzalez, M.; Martinez-Crespo, F. J.; De
Cerain, A. L.; Sainz, Y.; Narro, S.; Barker, A. J.; Hamilton, E. New
hypoxia-selective cytotoxins derived from quinoxaline 1,4-dioxides.
J. Hetercycl. Chem. 1995, 32, 1213–1217.
(59) Monge, A.; Palop, J. A.; De Cerain, A. L.; Senador, V.; Martinez-
Crespo, F. J.; Sainz, Y.; Narro, S.; Garcia, E.; de Miguel, C.; Gonzalez,
M. Hypoxia-selective agents derived from quinoxaline 1,4-di-N-oxides.
J. Med. Chem. 1995, 38, 1786–1792.
(60) Monge, A.; Martinez-Crespo, F. J.; De Cerain, A. L.; Palop, J. A.;
Narro, S.; Senador, V.; Marin, A.; Sainz, Y.; Gonzalez, M.; Hamilton,
E. Hypoxia-selective agents derived from 2-quinoxalinecarbonitrile
1,4-di-N-oxides. 2. J. Med. Chem. 1995, 38, 4488–4494.
(61) Amin, K. M.; Ismail, M. M. F.; Noaman, E.; Soliman, D. H.; Ammard,
Y. A. New quinoxaline 1,4-di-N-oxides, Part 1: Hypoxia-selective
cytotoxins and anticancer agents derived from quinoxaline1,4-di-N-
oxides. Bioorg. Med. Chem. 2006, 14, 6917-6923.
(62) Hay, M. P.; Denny, W. A. A new and versatile synthesis of 3-alkyl-
1,2,4-benzotriazine-1,4-dioxides: preparation of the bioreductive cy-
totoxins SR4895 and SR4941. Tetrahedron Lett. 2002, 43, 9569–9571.
(63) Pchalek, K.; Hay, M. P. Stille Coupling Reactions in the Synthesis of
Hypoxia-Selective 3-Alkyl-1,2,4-Benzotriazine 1,4-Dioxide Anticancer
Agents. J. Org. Chem. 2006, 71, 6530–6535.
(64) Anderson, R. F.; Denny, W. A.; Li, W.; Packer, J. E.; Tercel, M.;
Wilson, W. R. Pulse radiolysis studies on the fragmentation of
arylmethyl quaternary nitrogen mustards upon their one-electron
reduction in aqueous solution. J. Phys. Chem. A 1997, 101, 9704–
9709.
(65) Wardman, P. Reduction potentials of one-electron couples involving
free radicals in aqueous solution. J. Phys. Chem. Ref. Data 1989, 18,
1637–1755.
(66) Siim, B. G.; Hicks, K. O.; Pullen, S. M.; van Zijl, P. L.; Denny, W. A.;
Wilson, W. R. Comparison of aromatic and tertiary amine N-oxides
of acridine DNA intercalators as bioreductive drugsscytotoxicity,
DNA binding, cellular uptake, and metabolism. Biochem. Pharmacol.
2000, 60, 969–978.
(67) Pruijn, F. B.; Patel, K.; Hay, M. P.; Wilson, W. R.; Hicks, K. O.
Prediction of Tumour Tissue Diffusion Coefficients of Hypoxia-
Activated Prodrugs from Physicochemical Parameters. Aust. J. Chem.
2008, 61, 687–693.
(68) Wang, M.; Roberts, D. L.; Paschke, R.; Shea, T. M.; Masters, B. S. S.;
Kim, J-J. P. Three-dimensional structure of NADPH-cytochrome
P450reductase: prototype for FMN- and FAD-containing enzymes.
Proc. Natl. Acad. Sci. U.S.A. 1997, 94, 8411–8416.
(69) Zhao, Q.; Modi, S.; Smith, G.; Paine, M.; Mcdonagh, P. D.; Wolf,
C. R.; Tew, D.; Lian, L.-Y.; Roberts, G. C. K.; Driessen, H. P. C.
Crystal structure of the FMN-binding domain of human cytochrome
P450 reductase at 1.93 Å resolution. Protein Sci. 1999, 8, 298–306.
(36) Peters, L.; Rischin, D.; Fisher, R.; Corry, J.; Hicks, R. Identification
and therapeutic targeting of hypoxia in H&N cancer. Int. Congr.
Radiat. Res. 2007,
(37) Rischin, D.; Peters, L.; Fisher, R.; Macann, A.; Denham, J.; Poulsen,
M.; Jackson, M.; Kenny, L.; Penniment, M.; Corry, J.; Lamb, D.;
McClure, B. Tirapazamine, cisplatin, and radiation versus fluorouracil,
cisplatin, and radiation in patients with locally advanced head and
neck cancer: a randomized phase II trial of the Trans-Tasman Radiation
Oncology Group (TROG 98.02). J. Clin. Oncol. 2005, 23, 79–87.
(38) Durand, R. E.; Olive, P. L. Physiologic and cytotoxic effects of
tirapazamine in tumor-bearing mice. Radiat. Oncol. InVest. 1997, 5,
213–219.
(39) Durand, R. E.; Olive, P. L. Evaluation of bioreductive drugs in multicell
spheroids. Int. J. Radiat. Oncol, Biol., Phys. 1992, 22, 689–692.
(40) Hicks, K. O.; Fleming, Y.; Siim, B. G.; Koch, C. J.; Wilson, W. R.
Extravascular diffusion of tirapazamine: effect of metabolic consump-
tion assessed using the multicellular layer model. Int. J. Radiat. Oncol.,
Biol., Phys. 1998, 42, 641–649.
(41) Kyle, A. H.; Minchinton, A. I. Measurement of delivery and
metabolism of tirapazamine to tumour tissue using the multilayered
cell culture model. Cancer Chemother. Pharmacol. 1999, 43, 213–
220.
(42) Baguley, B. C.; Hicks, K. O.; Wilson, W. R. Tumour cell cultures in
drug development. In Anticancer Drug DeVelopment; Baguley, B. C.;
Kerr, D. J., Eds.; Academic Press: San Diego, 2002; pp 269-284.
(43) Hicks, K. O.; Pruijn, F. B.; Sturman, J. R.; Denny, W. A.; Wilson,
W. R. Multicellular resistance to tirapazamine is due to restricted
extravascular transport: a pharmacokinetic/pharmacodynamic study in
multicellular layers. Cancer Res. 2003, 63, 5970–5977.
(44) Hicks, K. O.; Siim, B. G.; Pruijn, F. B.; Wilson, W. R. Oxygen
dependence of the metabolic activation and cytotoxicity of tira-
pazamine: implications for extravascular transport and activity in
tumors. Radiat. Res. 2004, 161, 656–666.
(45) Hicks, K. O.; Pruijn, F. B.; Secomb, T. W.; Hay, M. P.; Hsu, R.;
Brown, J. M.; Denny, W. A.; Dewhirst, M. W.; Wilson, W. R. Use of
three-dimensional tissue cultures to model extravascular transport and
predict in vivo activity of hypoxia-targeted anticancer drugs. J. Natl.
Cancer Inst. 2006, 98, 1118–1128.
(46) Hicks, K. O.; Myint, H.; Patterson, A. V.; Pruijn, F. B.; Siim, B. G.;
Patel, K.; Wilson, W. R. Oxygen dependence and extravascular
transport of hypoxia-activated prodrugs: comparison of the dinitroben-
zamide mustard PR-104A and tirapazamine. Int. J. Radiat. Oncol.,
Biol., Phys. 2007, 69, 560–571.
(47) Pruijn, F. B.; Sturman, J.; Liyanage, S.; Hicks, K. O.; Hay, M. P.;
Wilson, W. R. Extravascular transport of drugs in tumor tissue: effect
of lipophilicity on diffusion of tirapazamine analogues in multicellular
layer cultures. J. Med. Chem. 2005, 48, 1079–1087.
(48) Hay, M. P.; Hicks, K. O.; Pruijn, F. B.; Pchalek, K.; Siim, B. G.;
Wilson, W. R.; Denny, W. A. Pharmacokinetic/pharmacodynamic
model-guided identification of hypoxia-selective 1,2,4-benzotriazine
1,4-dioxides with antitumor activity: the role of extravascular transport.
J. Med. Chem. 2007, 50, 6392–6404.
(49) Hay, M. P.; Pchalek, K.; Pruijn, F. B.; Hicks, K. O.; Siim, B. G.;
Anderson, R. R.; Shinde, S. S.; Phillips, V.; Denny, W. A.; Wilson,
W. R. Hypoxia-selective 3-alkyl 1,2,4-benzotriazine 1,4-dioxides: the
influence of hydrogen bond donors on extravascular transport and
antitumor activity. J. Med. Chem. 2007, 50, 6654–6664.
(50) Hay, M. P.; Gamage, S. A.; Kovacs, M. S.; Pruijn, F. B.; Anderson,
R. F.; Patterson, A. V.; Wilson, W. R.; Brown, J. M.; Denny, W. A.
Structure-activity relationships of 1,2,4-benzotriazine 1,4-dioxides as
hypoxia-selective analogues of tirapazamine. J. Med. Chem. 2003, 46,
169–182.
(51) Zeman, E. M.; Baker, M. A.; Lemmon, M. J.; Pearson, C. I.; Adams,
J. A.; Brown, J. M.; Lee, W. W.; Tracy, M. Structure-activity
JM800967H