Rapid synthesis and zebrafish evaluation of a phenanthridine-based small molecule library

Article abstract of DOI:10.1039/c0ob00449a

A Heck cyclisation approach is described for the rapid synthesis of a library of natural product-like small molecules, based on the phenanthridine core. The synthesis of a range of substituted benzylamine building blocks and their incorporation into the library is reported, together with a highly selective cis-dihydroxylation protocol that enables access to the target compounds in an efficient manner. Biological evaluation of the library using zebrafish phenotyping has led to the discovery of compound 20c, a novel inhibitor of early-stage zebrafish embryo development.

Full text of DOI:10.1039/c0ob00449a

View Article Online / Journal Homepage / Table of Contents for this issue
Organic &
Biomolecular
Chemistry
Dynamic Article Links
Cite this: Org. Biomol. Chem., 2011, 9, 2233
www.rsc.org/obc
PAPER
Rapid synthesis and zebrafish evaluation of a phenanthridine-based small
molecule library†‡
Lauren R. Donaldson,a Stephen Wallace,a David Haigh,§b E. Elizabeth Pattonc and Alison N. Hulme*a
Received 19th July 2010, Accepted 20th December 2010
DOI: 10.1039/c0ob00449a
A Heck cyclisation approach is described for the rapid synthesis of a library of natural product-like
small molecules, based on the phenanthridine core. The synthesis of a range of substituted benzylamine
building blocks and their incorporation into the library is reported, together with a highly selective
cis-dihydroxylation protocol that enables access to the target compounds in an efficient manner.
Biological evaluation of the library using zebrafish phenotyping has led to the discovery of compound
20c, a novel inhibitor of early-stage zebrafish embryo development.
The phenanthridine core 1 (Fig. 1) lies at the heart of many
Introduction
bioactive natural products including the antiviral lycorine 2,6 the
tubulin polymerisation inhibitor chelidonine 3,7 and the insect
Forward chemical genetics techniques seek to emulate the success
of classical genetics by using small molecules to probe the function
of gene-products and thus reveal information about biological
processes.1 Typically, a small molecule library is screened against
anti-feedant 3-O-acetyl-narcissidine 4.8 We recently disclosed
methodology for the efficient Heck cyclisation of carbamate
precursor 5a (Scheme 1), to afford cis-ring fused phenanthridine
double bond isomers 6a (D1,2 alkene), 7a (D2,3 alkene) and 8a
a cellular or embryonic system and the induction of any abnormal
(D3,4 alkene) in excellent yield.9 By combining this methodology
with a range of benzylamine building blocks 9 (Scheme 2), rapid
phenotype is identified. The protein target that has been modu-
lated may then be deduced;2 thus providing both information on
its function and a small molecule modulator from one process.
The ongoing search for modulators of all “druggable” targets
puts heavy demands on the sourcing of small molecule libraries
which may be obtained from in-house compound collections,3
commercially acquired combinatorial libraries,1 or even virtual
libraries.4 However, many such compound collections suffer from
a somewhat reduced scope, which has led to renewed interest in
libraries based on the diverse core-structures of natural products.5
These complex architectures have evolved to fulfil specific biologi-
cal functions and, as a result of the close evolutionary relationship
shared by the genomes of all organisms, bind to proteins similar
to human proteins.5 Natural products are therefore unmatched in
terms of their inherent bioactivity and their ability to probe chem-
ical space; thus they represent excellent, biologically-validated
scaffolds upon which to base chemical library design.
Fig. 1 The phenanthridine core 1 and biologically active phenanthridine
natural products 2–4.
aSchool of Chemistry, The University of Edinburgh, West Mains Road,
Edinburgh, UK, EH9 3JJ. E-mail: Alison.Hulme@ed.ac.uk
bGlaxoSmithKline, Medicines Research Centre, Gunnels Wood Road, Steve-
nage, Hertfordshire, UK, SG1 2NY
cEdinburgh Cancer Research Centre & MRC Human Genetics Unit, The
University of Edinburgh, Crewe Road South, Edinburgh, UK, EH4 2XR
† Electronic supplementary information (ESI) available: Experimental
details for the preparation and characterisation of all the other library
members. See DOI: 10.1039/c0ob00449a
‡ This paper is part of an Organic & Biomolecular Chemistry web theme
issue on chemical biology.
Scheme 1 Heck cyclisation approach to the phenanthridine core.9 6a =
D1,2; 7a = D2,3; 8a = D3,4. Reagents and conditions: Cationic: Herrmann–
Beller palladacycle (5 mol%), Ag2CO3, DMF, 140 ◦C, (99%, 6a : 7a : 8a
83 : 15 : 2); Neutral: Herrmann–Beller palladacycle (5 mol%), MeNCy2,
DMF, 140 ◦C (95%, 6a : 7a : 8a 36 : 38 : 26); or Pd2(dba)3, tBu3P·HBF4
(10 mol%), MeNCy2, MeCN, r.t.-50 ◦C, (95%, 6a : 7a : 8a 34 : 37 : 29).
§ Current address: Centre for Cancer Research & Cell Biology, Queens
University Belfast, 97 Lisburn Road, Belfast, UK BT9 7BL
This journal is
The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 2233–2239 | 2233
©

View Article Online
Table 2 Synthesis of Heck cyclisation precursors 5b–g
Scheme 2 Proposed library synthesis.
Reagents and conditions: (a) Boc2O, Et3N, CH2Cl2, 16 h; (b) NaH,
3-bromocyclohexene, DMF, 0 ◦C to rt, 16 h.
access to a collection of A-ring modified phenanthridine analogues
should be achieved, each with an alkene handle suitable for further
functionalisation. Due to the prevalence of hydroxylated C-ring
components in bioactive phenanthridine natural products,6–8,10
subsequent functionalisation of the alkene isomers 6–8 was en-
visaged using a dihydroxylation protocol. In order to evaluate the
resultant novel natural product-inspired small molecules, library
synthesis was coupled with phenotypic evaluation in embryonic
zebrafish.11
Entry Substrate R =
Yield 16 (%) Yield 5 (%)
1
2
3
4
5
6
b
c
d
e
f
4-Me
4-F
99
88
85
99
70
71
72
84
86
75
5-OMe
4,5-di-OMe
3,4-CH CH-CH CH- 90
4,5-O(CH2)O-
g
—
Table 3 Heck cyclisation of carbamates 5a–g
Library synthesis
Initial efforts focused on the synthesis of Heck cyclisation precur-
sors incorporating a range of benzylamine building blocks. To this
end, standard transformations were employed for the synthesis of
benzylamines 9b–f from their commercially available benzoic acid
counterparts 10b–f (Table 1).
Reagents and conditions: aryl halide (1 eq), MeNCy2 (4 eq),
Herrmann–Beller palladacycle (5 mol%), DMF, 140 ◦C.
Table 1 Synthesis of amine building blocks 9b–f
Entry 5
R =
t (h) Yield 6–8 (%) Ratioa 6 : 7 : 8
1
2
3
4
5
6
8
9
a
b
c
d
e
f
g
g
g
H
4-Me
4-F
5-OMe
4,5-di-OMe
3
12
5
6
12
95
76
72
73
75
74
36 : 38 : 26
26 : 57 : 17
27 : 44 : 29
36 : 44 : 20
41 : 32 : 27
42 : 42 : 16
41 : 26 : 33
18 : 46 : 36
37 : 39 : 24
Reagents and conditions: (a) (i) SOCl2, 60 ◦C, 3 h; (ii) NH4OH, rt, 16 h;
3,4-CH CH-CH CH- 5
59 (32)b
80 (11)c,d
99e
(b) SOCl2, 60 ◦C, 3 h; (c) (i) LiAlH4, AlCl3, Et2O, D, 16 h; (ii) HCl in Et2O.
4,5-O(CH2)O-
4,5-O(CH2)O-
4,5-O(CH2)O-
24
18
18
Yield 11 Yield 12 Yield 9
10
Entry 10 R =
(%)
(%)
(%)
a Ratio identified from 1H NMR of isolated product mixture. b 32%
debrominated material recovered. c Pd2(dba)3 (5 mol%), tBu3P·HBF4
(10 mol%), MeNCy2 (4 eq), MeCN, rt. d Starting material recovered. e As
entry 9 but at 50 ◦C.
1
2
3
4
5
b
c
d
e
f
4-Me
4-F
5-OMe
4,5-di-OMe
95
94
73
99
99
73
97
99
88
95
80
70
65
74
55
3,4-CH CH-CH CH-
N-Boc-protection of benzylamine hydrochlorides 9b–f, followed
by alkylation of the corresponding carbamates 16b–g gave the
desired cyclisation precursors 5b–g in good to excellent yield
(Table 2).
To access the corresponding piperonyl building block (R =
4,5-OCH2O-), a high yielding, two-pot Curtius rearrangement
approach was used. This allowed direct access to the N-Boc pro-
tected derivative of this analogue 16g via isocyanate intermediate
15 (Scheme 3).12
With the cyclisation precursors 5b–g in hand we studied
their reaction under the neutral Herrmann–Beller palladacycle
catalysed Heck cyclisation conditions we had previously developed
(Table 3).9 We found that for the majority of substrates, these
conditions promoted cyclisation to the expected mixture of double
bond regioisomers 6–8. However, conversions were lower in all
cases than that exhibited by the parent Boc substrate 5a (95% in
3 h),9 and reaction times were longer, especially with the electron-
rich aromatics (Table 3, entries 2, 4, 5, 8). For piperonyl substrate
5g (Table 3, entry 8) a significant amount of the debrominated
starting material was recovered, suggesting this substrate was
particularly sluggish to react under the palladacycle-catalysed
conditions. Fortunately, application of our previously developed
low temperature (tBu3P)2Pd conditions (Table 3, entry 9),9 enabled
the recovery of the desired product mixture 6–8g in 80% yield at
Scheme 3 Synthesis of piperonyl cyclisation precursor 16g. Reagents and
conditions: (a) DBDMH, 4 M NaOH, H2O, rt, 16 h, 91%; (b) dppa, Et3N,
CH2Cl2, 0 ◦C → rt, 0.5 h; (c) (i) silica plug filtration (partial conversion);
(ii) PhMe, 110 ◦C, 1 h (complete conversion); (d) tBuOH, 80 ◦C, 16 h, 50%
over 3 steps.
2234 | Org. Biomol. Chem., 2011, 9, 2233–2239
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
rt, and 99% yield at 50 ◦C (Table 3, entry 10). The regioisomeric
mixtures of phenanthridine double bond isomers were used in the
subsequent investigation of the dihydroxylation of the C-ring.13
Molecular models suggested a cupped conformation for the cis-
ring junction phenanthridines 6–8; thus it was anticipated that
dihydroxylation would be more favoured on the convex face. In
order to assess the likely diastereoselectivity of the dihydroxylation
reaction across the library, parent phenanthridines 6–8a were
separated by HPLC14 and subjected individually to standard
dihydroxylation conditions (Scheme 4).15 Quantitative conversions
to the corresponding diols were achieved for all three isomers,
with diastereoselectivities from 83 : 17–85 : 15 as determined by
1H NMR. The relative stereochemistry of the major product was
confirmed as the exo-diol using 2D NMR studies.16
Table 4 cis-Dihydroxylation of phenanthridines 6–8a–g
Reagents and conditions: OsO4, NMO, THF–H2O, rt, 18 h.
Entry
6–8
R =
Yield 17–19 (%)a
1
2
3
4
5
6
7
a
b
c
d
e
f
H
4-Me
4-F
5-OMe
4,5-di-OMe
3,4-CH CH-CH CH-
4,5-O(CH2)O-
70
78
82
95
70
87
87b
g
a Isolated yield of mixture following flash chromatography. b includes 2%
D2,3 endo-syn diol product as characterised by 2D NMR studies.15
Scheme 4 cis-Dihydroxylation of phananthridines 6a–8a. Reagents and
conditions: (a) OsO4, NMO, THF–H2O, rt, 18 h (17a 99%; 18a 99%; 19a
99%).
For the last step in our library synthesis, we took the diol
mixtures and isolated compounds and converted each into the
hydrochloride salts, with yields as summarised in Table 5. This
afforded 18 isolated compounds and 5 mixtures for biological
testing.
When this dihydroxylation protocol was applied to phenanthri-
dine double bond isomer mixtures 6–8a–g the corresponding diol
mixtures 17–19a–g were afforded in good to excellent yield (Table
4). In keeping with the high diastereomeric ratio we obtained
for the cis-dihydroxylation of individual carbamates 6–8a, we
observed the exo-diols as the major products in all cases.18
Following dihydroxylation, samples of the three major products
exo-syn diols 17, 18 and 19 were successfully isolated by silica-gel
chromatography or HPLC separation. However, we were unable
to separate the 4-methyl diol products 17b–19b under a range
of conditions. Similarly, the HPLC fractions for the D1,2 and
D2,3 dimethoxy analogues 17e and 18e overlapped significantly,
and thus the D2,3 sample had significant D1,2 contamination. For
these two substrates, mixtures of the appropriate compounds were
carried through to the biological testing stage.
Zebrafish Phenotyping
Zebrafish (Danio rerio) embryos are transparent for the first five
days post fertilisation (dpf), and develop outwith the mother’s
body, permitting visual inspection of their anatomical develop-
ment to take place.11 Fertilised embryos were collected and arrayed
in 96 well plates (3 eggs per well; from 6 h post fertilization, hpf)
containing embryo buffer (200 mL), and then library members
in their buffer/0.5% DMSO stock solutions (10–100 mM for
the mixtures, and 1–50 mM for the isolated compounds), were
exchanged for the buffer, one compound or mixture per well.19
The embryos were incubated at 28.5 ◦C and examined visually for
Table 5 Boc-deprotection to phenanthridinium hydrochloride library
Reagents and conditions: (i) TFA, CH2Cl2, 1 h, rt; (ii) NaOH, adjust to pH 9; (iii) HCl in Et2O, 0 ◦C.
Entry
17–19
R =
Yield 20–22 (%)a
20 (%)
21 (%)
22 (%)
1b
2
3
a
b
c
d
e
f
H
4-Me
4-F
5-OMe
4,5-di-OMe
3,4-CH CH-CH CH-
4,5-O(CH2)O-
—
69
99
89
87
—
69
86
—
99
82
88
67
61
60
—
86
92
60c
61
81
53
—
74
84
60
69
73
4
5
6b
7
g
a Performed on mixture. b No diol mixture was deprotected for these analogues as significant isolated material was available as compared to the other
substrates. c Approximately 1 : 1 mixture D1,2:D2,3
.
This journal is
The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 2233–2239 | 2235
©

View Article Online
developmental defects11 using a light-dissecting microscope at 1,
2, 3 and 4 dpf.
in conjunction with other methodologies that enable access to
individual library members. Through the development of various
aryl cyclisation precursors and a highly selective dihydroxylation
protocol, we were able to rapidly synthesise a small library of
novel phenanthridine analogues suitable for biological testing. The
preliminary evaluation of this library using zebrafish phenotyping
has led to the discovery of a novel inhibitor of zebrafish embryonic
development. Further biological investigations including cell-
based assays and affinity chromatography followed by protein
micro-sequencing,21 should enable identification of the target
protein and provide an understanding of the mode of action of
compound 20c at a molecular level.
Our preliminary screening was performed using the mixtures of
diols where available, so that precious isolated material was not
wasted. Fluorinated compound mixture 20c–22c was identified
as causing reproducible embryo death by 1 dpf, and so we
examined each component of the mixture individually. Analogue
20c (D1,2 isomer) was found to interfere reproducibly with zebrafish
development (as assessed by the head–trunk angle, which may be
used as a measure of developmental hours20) at concentrations
of 100, 50 and 20 mM. By almost 70 hpf all fish were delayed
developmentally in a dose-dependent manner, and exhibited lower
than normal pigmentation (Fig. 2). At 5 dpf, the yolk sac of the
fish treated with 100 mM 20c was significantly larger than that of
the control, providing further evidence for delayed development
(Fig. 3). No notable difference from the control was observed in
the fish exposed to lower drug concentrations at 5 dpf. None of the
other compound mixtures or isolated compounds exhibited any
specific phenotype when tested on the zebrafish embryos.
Experimental
General methods
All non-aqueous reactions were carried out under an atmosphere
of nitrogen using flame- or oven-dried glassware that was cooled in
a desiccator prior to use. Unless otherwise noted, starting materials
and reagents were obtained from commercial suppliers and were
used without further purification. Dichloromethane (CH2Cl2) and
triethylamine (Et3N) were distilled from and stored over calcium
hydride under a nitrogen atmosphere. Tetrahydrofuran (THF)
was distilled from sodium–benzophenone and stored under a
nitrogen atmosphere. Diethyl ether (Et2O) was dried and purified
by passage through activated alumina columns using a solvent pu-
rification system from www.glasscontour.com. Anhydrous DMF
and acetonitrile (MeCN) were used as supplied by BakerDRY.
Saturated aqueous solutions of inorganic salts are represented
Fig. 2 Zebrafish embryos treated with compound 20c are delayed during
embryonic development. Zebrafish embryos were treated with 20c for
three days, and screened for toxicity or developmental phenotypes. One
compound, 20c, caused a dose–dependent developmental delay. Control
treated embryos appeared normal (A), and developed pigmentation (dark
colouring). The head–trunk angle, used as a measure of developmental
hours,20 indicates that the control animals are aged at approximately
61 hpf (148◦ angle from middle of the ear and eye to the notochord).
By comparison, zebrafish treated with 20 mM concentrations of 20c (B)
appear correctly formed but slightly delayed in development by about 9 h
(head–trunk angle 138◦). Embryos treated with 100 mM concentrations
of 20c (C) are strongly delayed for development, as indicated by a low
head–trunk angle (62◦; equivalent to approximately 24 hpf).
1
as (volume, sat. aq.). H and 13C NMR spectra were obtained
on Bruker instruments at the stated frequency; J values are in
Hz. Infrared spectra were recorded using a Biorad FTS-7 or
Perkin-Elmer Paragon 1000 FT-IR spectrometer as thin films.
Electron impact (EI) mass spectra were obtained using a Finnigan
4500 mass spectrometer, and electrospray (ESI) and fast atom
bombardment (FAB) mass spectra were obtained on a Kratos
MS50TC mass spectrometer. Melting points were determined on
a Gallenkamp Electrothermal Melting Point apparatus and are
uncorrected. Flash chromatography was carried out using Merck
Kieselgel 60 (Merck 9385) under positive pressure by means of
an airline or hand pump. Eluent compositions are quoted as
v/v ratios. High performance liquid chromatography (HPLC) was
carried out using a Gilson instrument fitted with a refractive index
detector, using a Spherisorb silica column (length 250 mm, i.d.
20.0 mm, particle size 5 mm). Flow rates are a quoted for each
separation in mL min-1.
Fig. 3 Zebrafish embryos treated with 100 mM compound 20c at 5 dpf.
Some of the embryos treated with 20c underwent development at a similar
rate to the control fish, but did not consume their yolk (B, red dotted line),
as compared to the control treated fish (A, red arrow). This can indicate
developmental defects in intestinal development, or a more general toxicity
to embryonic development.
2-Bromo-4-fluorobenzamide 11c
A mixture of 2-bromo-4-fluorobenzoic acid (1.00 g, 4.57 mmol)
and thionyl chloride (15 mL) was heated at 60 ◦C for 3 h. The
thionyl chloride was removed under reduced pressure and the
residue was dissolved in NH4OH (15 mL, conc.) and stirred for
16 h at rt. The reaction mixture was filtered and the precipitate
dried on the high vacuum line for several hours to afford amide
11c as a colourless solid (940 mg, 94%).
Conclusions
We have demonstrated that libraries based upon the core structure
of natural products can be fruitful in the chemical genetics quest
for the discovery of novel small molecule probes for biological pro-
cesses. The diversity of such libraries can be enhanced by exploiting
multi-product-generating reactions such as the Heck cyclisation,
2236 | Org. Biomol. Chem., 2011, 9, 2233–2239
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
Rf [CH2Cl2–MeOH, 95 : 5] = 0.4; MP 155 ◦C (H2O); nmax
2-Bromo-4-fluorobenzyl tert-butylcarboxamide 16c
1
(CHCl3)/cm-1 3400 (NH), 1639 (C O); H NMR d (360 MHz,
A solution of amine hydrochloride 9c (167 mg, 0.69 mmol) and
Et3N (195 ml, 1.39 mmol) in CH2Cl2 (8 mL) was stirred for 10 min
then cooled to 0 ◦C. Boc2O (167 mg, 0.76 mmol) was added and
the reaction mixture was allowed to warm to rt and stirred for 4 h.
The reaction was diluted with CH2Cl2 (15 mL) and washed with
NaCl (3 ¥ 15 mL, sat. aq.). The organics were combined, dried
(MgSO4), concentrated under reduced pressure and purified by
flash chromatography (hexane–EtOAc, 10 : 1) to afford carbamate
16c as a colourless oil (184 mg, 88%).
CD3OD) 7.50 (1H, dd, J 8.4, 5.8, ArH), 7.46 (1H, dd, J 8.6, 2.6,
ArH), 7.22 (1H, td, J 8.4, 2.6, ArH); 13C NMR d (90.6 MHz,
CD3OD) 172.6 (C), 164.4 (d, J 252.1, C), 136.5 (C), 131.7 (d,
J 9.0, CH), 121.7 (d, J 25.1, CH), 121.2 (d, J 9.9, C), 115.9
(d, J 21.7, CH); m/z (FAB, 3-NOBA) 220 ([81BrM + H]+, 64%),
218 ([79BrM + H]+, 66), 167 (12), 165 (11), 136 (100); HRMS
(EI) Found: [81BrM]+, 218.9511. C7H5ON81BrF requires 218.9513.
Found: [79BrM]+, 216.9528. C7H5ON79BrF requires 216.9533.
Rf [hexane–EtOAc, 3 : 1] = 0.85; nmax (CHCl3)/cm-1 3343 (NH),
1696 (C O); 1H NMR d (250 MHz, CDCl3) 7.35 (1H, dd, J 8.3,
6.0, ArH), 7.27 (1H, dd, J 8.3, 2.5, ArH), 6.99 (1H, dt, J 8.3, 2.5,
ArH), 5.11 (1H, br s, CHN), 4.33 (2H, d, J 6.3, CH2Ar), 1.44 (9H,
s, 3¥CH3); 13C NMR d (62.9 MHz, CDCl3) 163.4 (C), 157.5 (d,
J 245.8, C), 133.8 (C), 130.5 (d, J 7.8, CH), 123.2 (d, J 9.6, C),
119.8 (d, J 24.5, CH), 114.4 (d, J 20.9, C), 79.6 (C), 44.0 (CH2),
28.2 (3¥CH3); m/z (EI) 306 ([81BrM + H]+, 1%), 304 ([79BrM + H]+,
1), 248 (14), 246 (13), 189 (17), 187 (18) 168 (100); HRMS (EI)
Found: [79BrM]+, 303.0279. C12H1579BrFNO2 requires 303.0265.
2-Bromo-4-fluorobenzonitrile 12c
A mixture of amide 11c (940 mg, 4.30 mmol) and thionyl chloride
(6 mL) was refluxed at 60 ◦C for 3 h. The reaction was concentrated
under reduced pressure to afford nitrile 12c as a colourless solid
(840 mg, 97%).
Rf [hexane–EtOAc, 3 : 1] = 0.78; MP 76 ◦C (EtOH); nmax
1
(CHCl3)/cm-1 2228 (CN); H NMR d (360 MHz, CD3OD) 7.88
(1H, dd, J 8.6, 5.4, ArH), 7.69 (1H, dd, J 8.3, 2.5, ArH), 7.35 (1H,
td, J 8.6, 2.5, ArH); 13C NMR d (90.6 MHz, CD3OD) 166.4 (d, J
259.3, C), 138.0 (d, J 10.1, CH), 127.6 (d, J 10.3, C), 122.5 (d, J
26.1, CH), 117.6 (C), 117.3 (d, J 22.9, CH), 113.5 (C); m/z (FAB,
3-NOBA) 202 ([81BrM+H]+, 3%), 200 ([79BrM+H]+, 2), 154 (64),
121 (11); HRMS (EI) Found: [81BrM]+, 200.9407. C7H381BrFN re-
quires 200.9407. Found: [79BrM]+, 198.9422. C7H379BrFN requires
198.9427.
2-Bromo-4-fluorobenzyl cyclohex-2-enyl-carbamic acid tert-butyl
ester 5c
A solution of Boc carba◦mate 16c (175 mg, 0.56 mmol) in DMF
(4 mL) was cooled to 0 C and NaH (46 mg, 60% dispersion in
mineral oil, 1.15 mmol) was added. The reaction was warmed
to rt for 40 min, then cooled to 0 ◦C and 3-bromocyclohexene
(133 ml, 1.15 mmol) was added dropwise. The reaction allowed
to warm to rt and stirred for 16 h. Et2O (10 mL) was added
and the organics washed with NaCl (3 ¥ 15 mL, sat. aq.).
The organics were dried (MgSO4), concentrated under reduced
pressure and purified by flash chromatography (hexane–EtOAc,
100 : 1) to afford cyclohexenyl amine 5c as a colourless oil (157 mg,
71%).
Rf [hexane–EtOAc, 3 : 1] = 0.75; nmax (CHCl3)/cm-11694 (C O);
1H NMR d (360 MHz, 323 K, CDCl3) 7.29–7.22 (2H, m, 2¥ArH),
7.01 (1H, dt, J 8.4, 2.6, ArH), 5.86–5.84 (1H, m, CH = CH),
5.46 (1H, d, J 10.2, CH CH), 4.83 (1H, br s, CHN), 4.30–4.26
(2H, m, CH2Ar), 2.06–1.85 (3H, m, CH2+CHAHB), 1.79–1.73
(1H, m, CHAHB), 1.70–1.55 (1H, m, CHCHD), 1.55–1.20 (10H, m,
CHCHD+3¥CH3); 13C NMR d (90.6 MHz, 323 K, CDCl3) 161.0
(d, J 249.0, C), 155.7 (C), 135.0 (C), 131.4 (CH), 128.5 (CH),
128.0 (CH), 121.7 (d, J 9.5, C), 119.4 (d, J 24.5, CH), 114.1 (d,
J 20.9, CH), 79.9 (C), 53.1 (CH), 47.0 (CH2), 28.2 (3¥CH3), 28.1
(CH2), 24.5 (CH2), 21.3 (CH2); m/z (FAB, 3-NOBA) 386 ([81BrM +
H]+, 86%), 384 ([79BrM + H]+, 95), 330 (98), 328 (98), 301 (47),
299 (49), 284 (96), 282 (97), 250 (96), 248 (97), 246 (94), 204
(77), 202 (87); HRMS (FAB, 3-NOBA) Found: [79BrM]+, 384.0969.
C18H2479BrFNO2 requires 384.0969.
2-Bromo-4-fluorobenzylamine hydrochloride 9c
To a suspension of LiAlH4 (95 mg, 2.50 mmol) in Et2O (5 mL) was
added AlCl3 (334 mg, 2.50 mmol) and the mixture was cooled to
0 ◦C and stirred for 10 mins. The reaction was warmed to rt, nitrile
12c (249 mg, 1.25 mmol) was added portionwise, the reaction was
stirred for 30 min and then heated at 40 ◦C for 18 h. The reaction
was quenched by the addition of Na2SO4·5H2O portionwise,
filtered and the filtrate stirred vigorously with potassium sodium
tartrate (100 mL, sat. aq.) for 1 h. The Et2O layer was separated
and the aqueous phase extracted with Et2O (3 ¥ 60 mL). The
combined organic phases were dried (MgSO4) and concentrated
under reduced pressure. The residue was taken up in CH2Cl2
(1 mL), and HCl in Et2O (20.0 mL, 1 M in Et2O) was added. The
precipitate was removed by filtration and dried to afford amine
hydrochloride 9c as a colourless solid (210 mg, 70%).
MP 251 ◦C (Et2O); 1H NMR d (250 MHz, CD3OD) 7.63
(1H, dd, J 8.8, 6.0, ArH), 7.58 (1H, dd, J 8.3, 2.8, ArH),
7.28 (1H, td, J 8.5, 2.8, ArH), 4.29 (2H, s, CH2); 13C NMR d
(62.9 MHz, CD3OD) 163.5 (d, J 252.4, C), 133.1 (d, J 9.0, CH),
129.6 (C), 125.3 (d, J 10.3, C), 120.9 (d, J 25.1, CH), 115.9 (d,
21.6, CH), 42.8 (CH2); m/z (FAB, 3-NOBA) 206 ([81BrM + H]+,
14%), 204 ([79BrM + H]+, 17), 189 (37), 187 (40), 149 (47); HRMS
(ESI+, CH2Cl2–MeOH/NH4OAc) Found: [M + H]+, 203.9820.
C7H8NF79Br requires 203.9819. Free Amine: Rf [CH2Cl2–MeOH,
95 : 5] = 0.46; nmax (CHCl3)/cm-1 3285 (NH), 2926, 1597, 1484; 1H
NMR d (250 MHz, CDCl3) 7.33 (1H, dd, J 8.5, 6.0, ArH), 7.26
(1H, dd, J 8.3, 2.5, ArH), 6.98 (1H, td, J 8.3, 2.5, ArH), 3.85 (2H,
s, CH2); 13C NMR d (62.9 MHz, CDCl3) 160.6 (d, J 249.3, C),
137.4 (C), 129.1 (d, J 7.9, CH), 122.6 (d, J 9.6, C), 119.1 (d, J 24.8,
CH), 113.9 (d, J 20.8, CH), 45.4 (CH2).
Fluoro phenanthridines 6–8c
To a degassed solution of cyclohexene 5c (80 mg, 0.21 mmol)
in DMF was added the Herrmann–Beller palladacycle (10 mg,
10 mmol) and MeNCy2 (175 ml, 0.84 mmol), and the reaction
was heated at 140 ◦C for 5 h. The mixture was allowed to cool
and then diluted with Et2O (20 mL) and washed with NaCl (3 ¥
20 mL, sat. aq.). The organics were combined, dried (MgSO4) and
This journal is
The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 2233–2239 | 2237
©

View Article Online
concentrated under reduced pressure to give the crude product.
Flash chromatography (hexane–EtOAc, 100 : 1–100 : 6) afforded
phenanthridines 6–8c as a colourless oil (46 mg, 72%). 1H NMR
of this oil showed it to be a 27: 44 : 29 mixture of double bond
isomers (6c:7c:8c).
m, 2¥ArH), 6.92 (1H, td, J 10.9, 2.6, ArH), 4.70–4.62 (3H, m,
2¥CH+CHX HYAr), 4.31 (1H, d, J 16.9, CHXHYAr), 3.73–3.68
(1H, m, CH), 3.37 (1H, br s, CH), 2.51 (1H, br s, OH), 1.92–1.84
(1H, m, CHAHB), 1.70–1.52 (2H, m, CHAHB+CHCHD), 1.51 (9H,
s, 3¥CH3), 1.41–1.32 (1H, m, CHCHD); 13C NMR d (90.6 MHz,
323 K, CDCl3) 161.9 (d, J 245.2, C), 154.8 (C), 135.3 (d, J 6.9, C),
129.1 (C), 128.1 (d, J 8.2, CH), 113.5 (d, J 21.7, CH), 112.5 (d, J
22.4, CH), 80.1 (C), 71.1 (CH), 67.4 (CH), 46.9 (CH), 43.0 (CH2),
42.9 (CH), 28.4 (3¥CH3), 27.3 (CH2), 24.2 (CH2); m/z (EI) 337
([M]+, 1%), 280 ([M-tBu]+, 100), 236 ([M - Boc]+, 61), 192 (19),
162 (26).
(4aSR,10bSR)-9-Fluoro-4,4a,6,10b-tetrahydro-3H-
phenanthridine-5-carboxylic acid tert-butyl ester 6c (D1,2 isomer)
Rf [hexane–EtOAc, 3 : 1] = 0.57; nmax (CHCl3)/cm-1 1692 (C O);
1H NMR d (360 MHz, 323 K, CDCl3) 7.07 (1H, dd, J 8.4,
5.7, ArH), 7.00 (1H, dd, J 9.9, 2.6, ArH), 6.88 (1H, td, J 8.4,
2.6, ArH), 6.12–6.06 (1H, m, CHCH = CH), 5.89–5.87 (1H,
m, CH CHCH2), 4.65 (1H, d, J 16.3, CHXHYAr), 4.40 (1H,
br s, NCHCH), 4.36 (1H, d, J 16.3, CHXHYAr), 3.53 (1H,
br s, NCHCH), 2.31–2.18 (1H, m, CHAHB), 2.17–2.08 (1H,
m, CHAHB), 1.77–1.68 (1H, m, CHCHD), 1.53–1.46 (10H, m,
3¥CH3+CHCHD); 13C NMR d (90.0 MHz, 323 K, CDCl3) 161.9
(d, J 244.1, C), 154.6 (C), 140.0 (d, J 6.8, C), 129.0 (CH), 127.5
(d, J 8.0, CH), 126.5 (CH), 123.4 (C), 114.4 (d, J 22.1, CH),
113.0 (d, J 21.8, CH), 79.8 (C), 50.2 (CH), 43.0 (CH2), 37.3 (CH),
28.5 (3¥CH3), 25.4 (CH2), 24.0 (CH2); m/z (EI) 303 ([M]+, 8%),
247 (95), 246 (90), 202 (27), 193 (100), 148 (36); HRMS (EI)
Found: [M]+ 303.1626. C18H22FNO2 requires 303.1629. Diagnostic
1H NMR data for 7c (D2,3 isomer) 1H NMR d (360 MHz, 323 K,
CDCl3) 7.12 (1H, dd, J 8.3, 5.7, ArH), 6.94 (1H, d, J 10.5, ArH),
6.92–6.87 (1H, m, ArH), 5.69–5.66 (1H, m, CH = CH), 5.46–5.41
(1H, m, CH CH), 4.57 (1H, d, J 16.2, CHXHY), 4.46 (1H, br s,
NCH), 4.45 (1H, d, J 16.1, CHXHY), 3.18 (1H, s, NCHCH), 2.79–
2.73 (1H, m, CHAHB), 2.67–2.60 (1H, m, CHAHB), 2.27–2.20 (1H,
m, CHCHD), 1.53–1.45 (10H, m, CHCHD+3¥CH3). Diagnostic 1H
(2RS,3SR,4aSR,10bSR)-9-Fluoro-2,3-dihydroxy-2,3,4,4a,6,10b-
hexahydro-1H-phenanthridine-5-carboxylic acid tert-butyl ester
18c (D2,3 isomer)
Rf [CH2Cl2–MeOH, 9 : 1] = 0.53; Rt (EtOAc–hexane, 3 : 1, flow
rate: 10 mL min-1) = 20 min; nmax (CHCl3)/cm-1 3421 (OH), 1670
1
(C O); H NMR d (360 MHz, 323 K, CDCl3) 7.17 (1H, d, J
10.1, ArH), 7.11–7.07 (1H, m, ArH), 6.91 (1H, td, J 8.3, 1.8,
ArH), 4.80–4.72 (1H, m, CH), 4.66 (1H, d, J 17.0, CHX HYAr),
4.31 (1H, d, J 17.0, CHXHYAr), 3.93 (1H, br s, CH), 3.64–3.60
(1H, m, CH), 3.23 (1H, br s, CH), 2.40–2.23 (2H, m, CH2), 1.92–
1.89 (1H, m, CHAHB), 1.52 (9H, s, 3¥CH3), 1.50–1.42 (1H, m,
CHAHB); 13C NMR d (62.9 MHz, CDCl3) 164.9 (d, J 133.4, C),
154.7 (C), 136.8 (C), 128.6 (C), 127.9 (d, J 8.3, CH), 113.4 (d, J
21.6, CH), 112.5 (d, J 21.9, CH), 80.2 (C), 69.1 (CH), 66.6 (CH),
46.3 (CH), 42.8 (CH2), 36.4 (CH), 31.3 (CH2), 29.1 (CH2), 28.4
t
(3¥CH3); m/z (EI) 337 ([M]+, 1%), 280 ([M - Bu])+, 100), 236 ([M
- Boc]+, 27), 218 (19), 192 (35), 162 (34), 148 (24).
1
NMR data for 8c (D3,4 isomer) H NMR d (360 MHz, 323 K,
(3RS,4SR,4aRS,10bSR)-9-Fluoro-3,4-dihydroxy-2,3,4,4a,6,10b-
hexahydro-1H-phenanthridine-5-carboxylic acid tert-butyl ester
19c (D3,4 isomer)
CDCl3); 5.69–5.66 (1H, m, CH = CH), 5.52 (1H, dd, J 10.2, 1.0,
CH CH), 5.08 (1H, br s, NCH), 4.84 (1H, d, J 16.4, CHXHY),
4.19 (1H, d, J 16.4, CHXHY), 3.28 (1H, br s, NCHCH), 2.40–2.30
(1H, m, CHAHB), 2.12–2.00 (1H, m, CHAHB).
Rf [CH2Cl2–MeOH, 9 : 1] = 0.47; Rt (EtOAc–hexane, 3 : 1, flow
rate: 8 mL min-1) = 17 min; nmax (CHCl3)/cm-1 3404 (OH), 1670
9-Fluoro-3,4-dihydroxy-2,3,4,4a,6,10b-hexahydro-1H-
phenanthridine-5-carboxylic acid tert-butyl ester 17–19c
1
(C O); H NMR d (360 MHz, 323 K, CDCl3) 7.11 (1H, dd,
J 8.5, 5.7, ArH), 7.03 (1H, dd, J 10.0, 1.7, ArH), 6.92 (1H, m,
ArH), 4.73–4.61 (2H, m, CH+CHX HYAr), 4.41 (1H, d, J 16.4,
CHXHYAr), 3.96 (1H, d, J 2.7, CH), 3.30–3.25 (2H, m, 2¥CH),
2.60 (1H, br s, OH), 2.35–2.22 (1H, m, CHAHB), 2.17–2.12 (1H,
m, CHAHB), 1.87–1.82 (1H, m, CHCHD), 1.60–1.50 (10H, m,
CHCHD+3¥CH3); 13C NMR d (90.6 MHz, 323 K, CDCl3) 162.1
(d, J 245.1, C), 155.2 (C), 137.1 (d, J 2.8, C), 128.9 (C), 128.0 (d,
J 8.2, CH), 113.4 (d, J 21.7, CH), 112.3 (d, J 22.1, CH), 81.0 (C),
69.5 (2¥CH), 52.7 (CH), 42.1 (CH2), 37.1 (CH), 28.4 (3¥CH3),
25.4 (CH2), 20.1 (CH2); m/z (EI) 337 ([M]+, 1%), 281 (36), 280
To a solution of phenanthridines 6–8c (50 mg, 0.165 mmol) in
THF (920 ml) and H2O (184 ml) was added OsO4 (144 ml, 2.5%
w/w in tBuOH, 11.5 mmol) and NMO (58 mg, 0.495 mmol) and the
reaction was stirred for 16 h at rt. The reaction mixture was poured
onto Na2SO3 (30 mL, sat. aq.) and extracted with EtOAc (3 ¥
30 mL). The combined organics were dried (MgSO4), concentrated
under reduced pressure and purified by flash chromatography
(CH2Cl2–MeOH, 100 : 2) to afford a mixture of diols 17–19c
(46 mg, 82%). HPLC (EtOAc–hexane, 3 : 1) of this mixture
afforded D1,2 diol 17c (10 mg, 18%), D2,3 diol 18c (15.3 mg, 2%),
D3,4 diol 19c (11.4 mg, 20%), and mixed diol fractions (9 mg, 16%)
giving an overall yield (45.7 mg, 82%), all colourless oils.
t
([M - Bu])+, 9), 236 ([M - Boc]+, 100), 218 (11), 206 (12), 192 (13),
162 (40).
(1RS,2SR,4aSR,10bSR)-9-Fluoro-1,2,3,4,4a,5,6,10b-octahydro-
phenanthridine-1,2-diol 20c (D1,2 isomer)
(1RS,2SR,4aSR,10bSR)-9-Fluoro-1,2-dihydroxy-2,3,4,4a,6,10b-
hexahydro-1H-phenanthridine-5-carboxylic acid tert-butyl ester
17c (D1,2 isomer)
To a solution of diol 17c (10 mg, 30 mmol) in CH2Cl2 (2 mL) was
added TFA (5 mL) and the reaction was stirred at rt for 2 h. The
reaction was diluted with H2O (15 mL), adjusted to pH 8–9 by
the addition of NaOH pellets, and then extracted with CH2Cl2
(3 ¥ 15 mL). The combined organics were dried (MgSO4) and
Rf [CH2Cl2–MeOH, 9 : 1] = 0.43; Rt (EtOAc–hexane, 3 : 1, flow
rate: 10 mL min-1) = 15 min; nmax (CHCl3)/cm-13425 (OH), 1664
1
(C O); H NMR d (360 MHz, 323 K, CDCl3) 7.13–7.04 (2H,
2238 | Org. Biomol. Chem., 2011, 9, 2233–2239
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
concentrated under reduced pressure. The resultant oil was taken
up in CH2Cl2 (1 mL), cooled to 0 ◦C and HCl (excess, 1 M in
Et2O) added. The resultant oil was washed with Et2O and dried
under vacuum to afford amine hydrochloride 20c as a colourless
oil (9 mg, 99%).
1H NMR d (360 MHz, D2O) 7.09 (1H, dd, J 8.6, 5.7, ArH),
7.03 (1H, dd, J 10.2, 2.6, ArH), 6.94 (1H, td, J 8.6, 2.6, ArH), 4.28
(1H, d, J 16.2, CHXHYAr), 4.22 (1H, d, J 16.2, CHXHYAr), 3.88
(1H, d, J 7.6, CH), 3.78–3.74 (2H, m, 2¥CH), 3.19 (1H, dd, J 8.2,
4.4, CH), 2.01–1.91 (1H, m, CHAHB), 1.81–1.72 (1H, m,CHAHB),
1.68–1.60 (2H, m, CH2); 13C NMR d (90.6 MHz, D2O) 162.3 (d, J
243.8, C), 135.0 (d, J 7.7, C), 129.2 (d, J 8.5, CH), 123.7 (C), 116.4
(d, J 23.7, CH), 115.5 (d, J 22.1, CH), 70.9 (CH), 68.2 (CH), 52.0
(CH), 43.4 (CH2), 39.5 (CH), 25.6 (CH2), 21.9 (CH2); m/z (ESI+)
238 ([M+H]+, 100%), 236 (86); HRMS (ESI+) Found [M +H]+,
238.1237. C13H17O2NF requires 238.1238.
Jennifer Richardson for their assistance during the zebrafish
screening process.
Notes and references
1 (a) D. R. Spring, Chem. Soc. Rev., 2005, 34, 472–482; (b) N. J. Westwood,
Philos. Trans. R. Soc. London, Ser. A, 2004, 362, 2761–2774.
2 (a) F. Landi, C. M. Johansson, D. J. Campopiano and A. N. Hulme,
Org. Biomol. Chem., 2010, 8, 56–59; (b) S.-E. Ong, M. Schenone, A. A.
Margolin, X. Li, K. Do, M. K. Doud, D. R. Mani, L. Kuai, X. Wang, J.
L. Wood, N. J. Tolliday, A. N. Koehler, L. A. Marcaurelle, T. R. Golub,
R. J. Gould, S. L. Schreiber and S. A. Carr, Proc. Natl. Acad. Sci. U. S.
A., 2009, 106, 4617–4622; (c) B. J. Leslie and P. J. Hergenrother, Chem.
Soc. Rev., 2008, 37, 1347–1360.
3 R. J. Spandl, A. Bender and D. R. Spring, Org. Biomol. Chem., 2008,
6, 1149–1158.
4 S. Ghosh, A. Nie, J. An and Z. Huang, Curr. Opin. Chem. Biol., 2006,
10, 194–202.
5 K. S. Lam, Trends Microbiol., 2007, 15, 279–289.
6 (a) N. Unver, Phytochem. Rev., 2007, 6, 125–135; (b) J. R. Lewis, Nat.
Prod. Rep., 1994, 11, 329–332.
7 A. Panzer, A. M. Joubert, P. C. Bianchi, E. Hamel and J. C. Seegers,
Eur. J. Cell Biol., 2001, 80, 111–118.
Zebrafish screening
Small breeding tanks were set up in the evening, each containing a
male and a female wildtype zebrafish separated by a divider. The
tanks were kept in darkness until the next morning when the lights
were switched on and the dividers removed, causing the fish to
breed. The embryos were collected, pooled and washed with E3
medium. Dead, delayed or unfertilised eggs were discarded, and
the adult fish were returned to their main tank. The embryos were
stored in an incubator at 28.5 ◦C until required they reached the
‘high stage’ of development (approximately 3.3 hpf). Chemicals
were diluted into the E3 screening medium.19 Aliquots of 200 mL
were prepared at concentrations ranging from 100 mM to 1 mM,
all with 0.5% v/v DMSO. Once the embryos had reached the
desired age they were distributed in 96-well plates as appropriate
to the screen, with two or three embryos per well. The surrounding
medium was removed from the embryos using a wide-tipped
Pasteur pipette and then the appropriate 200 mL chemical aliquot
was added to the well. Four control wells were used per plate, each
8 O. Santana, M. Reina, A. L. Anaya, F. Hernandez, M. E. Izquierdo and
A. Gonzalez-Coloma, Z. Naturforsch., C: J. Biosci., 2008, 63, 639–643.
9 L. R. Donaldson, D. Haigh and A. N. Hulme, Tetrahedron, 2008, 64,
4468–4477.
10 Z. Jin, Nat. Prod. Rep., 2003, 20, 606–614.
11 (a) For small molecule screening in zebrafish, see: K. L. Taylor, N. J.
Grant, N. D. Temperley and E. E. Patton, Cell Commun. Signaling,
2010, 8, 11; (b) J. Berger and P. Currie, Curr. Med. Chem., 2007, 14,
2413–2420; (c) L. I. Zon and R. T. Peterson, Nat. Rev. Drug Discovery,
2005, 4, 35–44.
12 V. Namboodiri, B. Campbell, G. A. Flynn, T. Fujimoto, F. P. Hollinger,
M. Bukhtiyarova, E. B. Springman and M. Karpusas, Bioorg. Med.
Chem. Lett., 2005, 15, 5274–5279.
13 In order to obtain characterisation details for the D1,2 product,
we also performed each cyclisation using our cationic conditions.9
1
Diagnostic H NMR data were then determined for the D2,3 and D3,4
phenanthridine isomers by comparison with the data for isolated parent
Boc-phenanthridines 6–8a14.
14 L. R. Donaldson, D. Haigh and A. N. Hulme, Synlett, 2009, 1587–
1590.
15 I. Kadas and L. Toke, J. Heterocycl. Chem., 1998, 35, 343–348.
16 Confirmation that the major product was the exo-diol was obtained
later in this study when 2D NMR was performed on both the major and
minor products from the dihydroxylation of piperonyl phenanthridine
7g. The major product did not show an NOE correlation between either
of the ring junction protons and either of the CHOH protons, whereas
the minor product (the endo–syn diol) did17.
v
containing E3 medium with 0.5% /v DMSO. The plates were
incubated at 28.5 ◦C and examined periodically over the next five
days by optical microscope.
17 L. R. Donaldson, PhD Thesis, The University of Edinburgh, 2009.
18 Determination of the exact ratio was not possible due to the complex
mixture of isomeric products.
Acknowledgements
We thank the EPSRC (DTA award to LRD), the Society for
Chemical Industries (Messel Scholarship to LRD), Medical
Research Scotland, the MRC and GlaxoSmithKline for financial
support of this work. LRD also thanks Corina Anastasaki and
19 R. D. Murphey and L. I. Zon, Methods, 2006, 39, 255–261.
20 C. B. Kimmel, W. W. Ballard, S. R. Kimmel, B. Ullmann and T. F.
Schilling, Dev. Dyn., 1995, 203, 253–310.
21 R. W. King, Chem. Biol., 1999, 6, R327–R333.
This journal is
The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 2233–2239 | 2239
©