Continuous flow photochemistry as an enabling synthetic technology: Synthesis of substituted-6(5: H)-phenanthridinones for use as poly(ADP-ribose) polymerase inhibitors

Article abstract of DOI:10.1039/c5md00552c

Methods utilizing continuous flow photochemistry, an enabling synthetic technology, have been developed for the generation of phenanthridinones via an intramolecular photochemical cyclization of 2-chlorobenzamides for the purposes of generating poly(ADP-ribose) polymerase inhibitors. Herein we report 16 examples of a single-step flow photocyclization which produces substituted phenanthridinones in yields up to 99%, while a two-step method leads directly to phenanthridinones from 2-chlorobenzoyl chlorides and anilines via a novel continuous flow amidation/photocyclization protocol. Overall, the flow photocyclization reactions typically progress in good to excellent yields, and in a superior fashion to analogous batch methods, greatly enabling the drug discovery process.

Full text of DOI:10.1039/c5md00552c

MedChemComm
View Article Online
View Journal
RESEARCH ARTICLE
Continuous flow photochemistry as an enabling
synthetic technology: synthesis of substituted-
Cite this: DOI: 10.1039/c5md00552c
6
IJ5H)-phenanthridinones for use as polyIJADP-
ribose) polymerase inhibitors†‡
a
ab
Y. Fang and G. K. Tranmer*
Methods utilizing continuous flow photochemistry, an enabling synthetic technology, have been developed
for the generation of phenanthridinones via an intramolecular photochemical cyclization of
2-chlorobenzamides for the purposes of generating polyIJADP-ribose) polymerase inhibitors. Herein we report
Received 1st December 2015,
Accepted 31st January 2016
16 examples of a single-step flow photocyclization which produces substituted phenanthridinones in yields
up to 99%, while a two-step method leads directly to phenanthridinones from 2-chlorobenzoyl chlorides
and anilines via a novel continuous flow amidation/photocyclization protocol. Overall, the flow photo-
cyclization reactions typically progress in good to excellent yields, and in a superior fashion to analogous
batch methods, greatly enabling the drug discovery process.
DOI: 10.1039/c5md00552c
www.rsc.org/medchemcomm
8
–11
subject.
We would like to summarize, however, the bene-
Introduction
fits derived from flow photochemistry, namely: flow photo-
chemistry is more efficient due to the narrow width of the
reactor tubing (Beer–Lambert law ) in comparison to batch;
Heterocyclic ring systems are an essential constituent core of
a large number of bioactive compounds and as a result, serve
as important building blocks for modern drug discovery. In
recent years, new efforts have concentrated upon using the
many benefits of flow chemistry to develop improved
methods for the synthesis of various heterocycles and apply
4
1
,2
the constant elution of product from the flow reaction aids
in scalability; control over flow-rates/volume/concentration
allows for increased control over reaction conditions and UV
irradiation duration, which can control competing reactions
or side-products; decreased levels of product decomposition,
as the final product spends relatively little time being exposed
to photochemical conditions; flow photochemistry can benefit
intramolecular reactions, as low concentration reactions avoid
intermolecular dimerizations/polymerizations; while continu-
ous processing enhances the production of product from low
concentration reactions, a distinct limitation of low concen-
tration batch reactions. Overall, the use of flow photo-
chemistry has experienced increased interest in recent years,
due in part to the advent and utility of bespoke, and commer-
cially available flow photochemistry systems.
3
these advantages within medicinal chemistry programs. Ad-
ditionally, considerable interest has focused upon the use of
4
,5
flow photochemistry as a tool in organic synthesis, and as
an enabling synthetic technology in the drug discovery pro-
cess. In particular, the flow photochemical synthesis of
6
Artemisinin stands out as an excellent example of the many
benefits that can be realized when flow chemistry techniques
are applied to the synthesis of medicinally relevant com-
pounds. In a direct comparison of batch versus flow photo-
chemical reactions, increased productivity and a demon-
strated ability to scale-up the reactions were highlighted as
the main benefits of flow photochemistry, and were cited as
key advantages over batch processes. A detailed explanation
on the myriad of benefits of flow chemistry will not be offered
here, as several reviews have been authored on the
PolyIJADP-ribose) polymerase (PARP) is a family of nuclear
proteins involved in DNA repair and programmed cell death
and are responsible for the synthesis of an adenine-
containing RNA-like polymer. In general, PARP binds to DNA
strand breaks, undergoes automodification, and gives rise to
branched polyIJADP-ribose) polymers, enabling the DNA repair
7
a
College of Pharmacy, Faculty of Health Sciences, University of Manitoba,
1
2
Winnipeg, MB, R3E 0T5, Canada. E-mail: geoffrey.tranmer@umanitoba.ca
process. Several inhibitors of PARP are currently in clinical
b
Department of Chemistry, Faculty of Science, University of Manitoba, Winnipeg,
13
trials as potential cancer therapies, as inhibitors of PARP
MB, R3T 2N2, Canada
are proposed to enhance the effects of cytotoxic alkylating
†
‡
The authors declare no competing interest.
1
4
agents. 6IJ5H)-Phenanthridinone and its structural analogs
Electronic supplementary information (ESI) available. See DOI: 10.1039/
c5md00552c
have been known to be inhibitors of PARP family proteins,
This journal is © The Royal Society of Chemistry 2016
Med. Chem. Commun.

View Article Online
Research Article
MedChemComm
however, some lack specificity and can elicit promiscuous in-
hibitory behaviour on any number of the family of PARP pro-
to moderate yield, 23–74%. Additionally, this method was
performed in cyclohexane which limits the substrate scope
due to solubility issues, while the relatively low auto-ignition
temperature of cyclohexane (245 °C) provides for a safety haz-
ard when using hot UV lamps. Our group set out to provide a
significant improvement over existing methods by using the
many benefits of flow photochemistry to provide a superior
route towards substituted-6IJ5H)-phenanthridinones for use in
our medicinal chemistry program (Scheme 1).
1
5
teins. For instance, in terms of percent inhibition of PARP
activation in macrophages at 1 μM concentration, 6IJ5H)-
phenanthridinone (PJ97A, 1) provided 71% inhibition, while
1
6
PJ34 (2) provided 90% inhibition, however, they are both
1
5,17
considered to be unselective.
Interestingly, similar agents
containing a sulphonamide residue (3) have been found to
1
5
be PARP1-4 selective (Fig. 1).
As part of a medicinal chemistry program, our group
needed to develop a robust method for the synthesis of
substituted-6IJ5H)-phenanthridinones. An examination of the
literature revealed only a single communication which has
been published on the synthesis of 6IJ5H)-phenanthridinones
Results and discussion
In order to perform our flow photochemical cyclizations, we
purchased a Vapourtec R2C+/R4 flow chemistry system
equipped with a UV-150 photochemical reactor which pos-
sesses a medium pressure pure mercury lamp and a 10 mL
fluorinated ethylene propylene (FEP) reactor coil. The system
contains two HPLC-style pumps which progress material
through the UV-150 reactor via sample loops and is mounted
to the front of the Vapourtec R4 heating/cooling module. For
our purposes, we set the UV power output to 75% (112 W)
and maintained a constant temperature of 60 °C within the
reactor using the R4's forced air cooling feature. This temper-
ature control is one of the many advantages of the Vapourtec
flow photochemistry system as the UV-150's integrated tem-
perature probe allows precise knowledge of flow photo-
chemical reaction temperature, which is often problematic
(or unknown) for most bespoke flow photochemistry systems.
Additionally, the UV lamp is housed within the UV-150 reac-
tor, increasing safety by eliminating user UV-exposure and
cooling the system via the R4, reducing the risk of fires
caused by the hot UV lamps approaching the auto-ignition
1
8
for the purposes of PARP1 inhibition. In this communica-
tion, three different inefficient methods were explored to con-
struct the characteristic tricyclic framework of phenanthridin-
one and provide various substitution patterns. Method 1: a
Schmidt reaction was used to convert 9-fluorenones to 3,8-
disubstituted-6IJ5H)-phenanthridinones using sodium azide in
concentrated sulfuric acid, generating hydrazoic acid in situ
which creates a significant safety risk. Additionally, if the
substitution on the 9-fluorenones were not symmetrical, a
mixture of regioisomers was observed. Method 2: an Ullmann
coupling was used to afford a biphenyl ester, followed by re-
duction and amidation, however, the key step only progresses
in 42–50% yield. Method 3: a Suzuki reaction, followed by
intramolecular Friedel–Crafts acylation provided limited ac-
cess to desired products as the key cross-coupling step only
occurs in 40–60% yield. Additionally, other methods have
been used to generate phenanthridinone ring systems, such
1
9
as radical mechanisms
or palladium-catalyzed C–H
2
0
functionalization, however, these reactions often lead to
complex mixtures of side-products and moderate yields
2
0
1
9
when aryl chlorides are used. As a result of the literature
providing poor routes for the synthesis of substituted-6IJ5H)-
phenanthridinones, we set out to develop a superior route
using alternate methods. We envisioned a simplified two-step
process wherein an acyl chloride is reacted with an aniline to
provide a 2-halobenzanilide derivative, which in turn, pro-
vides
a substituted-6IJ5H)-phenanthridinone via a photo-
cyclization reaction. A thorough examination of the literature
provided only a single example of the batch photocyclization
2
1
of 2-halobenzanilides. Unfortunately, the batch method was
very limited in scope and synthetic utility, as the communica-
tion contained only 6 examples and provided products in low
Fig. 1 6IJ5H)-Phenanthridinone PARP inhibitors.
Scheme 1 Methods for 6IJ5H)-phenanthridinone synthesis.
Med. Chem. Commun.
This journal is © The Royal Society of Chemistry 2016

View Article Online
MedChemComm
Research Article
−
1
temperature of flammable solvents (cyclohexane, 245 °C), a
hazard that is often ignored.
reactive substrates, an overall flow rate of 200 μL min was
found to give the best results, however, for substrates with
unreacted starting material remaining, slower flow rates did
give higher rates of conversion. In order to perform a library
of reactions, Table 1, we chose to set the flow rate at 0.2 mL
Extensive optimization reactions were carried out on the
photocyclization of 2-chlorobenzamide (entry 1, Table 1) in order
to ascertain the impact of varying solvent, substrate concen-
tration, residence time/flow rate and UV light filter. Protic
solvents, such as methanol, did not yield any desired prod-
uct, however, both acetonitrile and acetone were found to
produce the product in significant amounts. Additionally, the
photocyclization reaction was also found to proceed if a polar
aprotic solvent, such as DMSO, was added to acetone, ratio
−
1
min , as this setting provided for the best balance between
substrate throughput and adequate conversion of starting
−
1
material. However, slower flow rates (∼0.15–0.1 mL min )
did give higher yields in some cases where starting material
remained, such as entry 2, 4, 6, 11 and 12. We also investi-
gated different filters for our photochemical cyclization
which are available commercially from Vapourtec. Filter 1 is
a simple quartz filter which allows irradiation across the full
spectrum of a medium pressure pure Hg lamp, however, fil-
ter 2 cuts off wavelengths above 400 nm. For the compounds
tested, no observable difference in yield was obtained when
filter 2 was used, implying that the photochemical reaction
was taking place a wavelengths less than 400 nm.
1
: 40 respectively. This result will become useful for future re-
actions when solubility of potential substrates becomes an is-
sue. Overall, the reactions performed best in acetone, while
acetonitrile produced more by-products and showed numer-
ous impurity peaks via HPLC. As for the effect of concentra-
tion on yield, for the parent compound 1, the reaction
performed well in concentration ranges from 0.02 M to 5
mM, as all reactions gave yields of 95+%. However, for some
of the other substrates, a concentration of 5 mM (0.1 mmol/
Using our general reaction conditions of 5 mM of the
−1
2-chlorobenzamide, 0.2 mL min (50 minutes flow-hv exposure
time), in acetone using filter 1 (quartz), we set out to produce
a library of compounds which is highlighted in Table 1. The
parent compound, entry 1, produced a product in near quan-
titative yield, while electron-donating (OMe) or electron with-
drawing (Cl) ortho-substituted anilines (R ) gave similarly
high yields (80, 74%, respectively), indicating little electronic
3
effects for this substitution. para-Substituted anilines (R )
also gave high isolated yields, ranging between 80–93% for
electron withdrawing substituents. Substitution was also per-
mitted on the carbonyl component of the cyclization partner,
yielding mono- and di-substituted products, typically in good
yields (67–85%). We also found three problematic reactions,
entries 14–16, and wish to report that these reactions typi-
cally gave multiple unidentified side-products, and some
unreacted starting material which contributed to the low
yields. We are currently exploring the mechanistic aspects of
this reaction, and further exploring the effects of substitution
2
0 mL) appeared to give the highest yielding reactions, in com-
bination with adequate conversion. At higher concentrations,
some substrates gave incomplete conversion, as starting ma-
terial remained. Obviously, flow rate and concentration are
key factors in ensuring that all starting material has been
converted to product. For the parent compound 1, a 5 mM
1
−
1
solution could be run at 400 μL min without any noticeable
effect on conversion, however, at higher flow rates >0.5 mL
−
1
min , unreacted starting material was observed. For less
Table 1 Flow photochemical synthesis of 6IJ5H)-phenanthridinones from
2-chlorobenzamides
(position and electron-withdrawing/donating nature) on the
yield of this photocyclization and will report results in due
course.
a
Entry
R
1
R
2
R
3
R
4
R
5
R
6
Yield
When the flow-hv reaction was performed on a 2-chloro-
2′-bromoanilide, an unexpected result was obtained,
Scheme 2. The reaction of N-(5-acetyl-2-bromophenyl)-2-
chlorobenzamide under our generalized flow photochemical
conditions produced a benzoxazole ring system in 63% iso-
lated yield, rather than the desired phenanthridinone. A
1
2
3
4
5
6
7
8
9
H
H
OMe
Cl
H
H
H
H
H
H
H
H
H
Me
H
H
H
COMe
H
H
H
H
H
H
H
H
H
H
H
H
H
Cl
H
H
H
H
Cl
F
CF
H
H
Cl
Cl
F
H
H
H
H
H
H
H
Cl
H
H
Cl
Cl
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
CF
H
H
H
H
H
H
H
H
H
Cl
Cl
H
H
F
99
76
80
74
93
80
86
67
85
83
74
67
84
21
26
13
3
10
11
12
13
14
15
16
F
H
H
H
H
H
3
Me
a
Isolated yields; 10 mL flourinated ethylene propylene (FEP) coil,
60 °C, Vapourtec UV-150 medium pressure Hg lamp (75%, ∼112 W).
Scheme 2 Flow-hv reaction of 2-chloro-2′-bromoanilides.
This journal is © The Royal Society of Chemistry 2016
Med. Chem. Commun.

View Article Online
Research Article
MedChemComm
thorough examination of the literature did produce a single
would also like to highlight that by combining two synthetic
steps into a continuous flow sequence, there was no need for
purification or handling of intermediates, offering greener re-
action processing and a more rapid synthesis of drug-like mol-
ecules, reducing the cost and cycle-time for development.
2
2
communication which describes this reaction. However, the
report of this batch photochemical reaction contained only
three examples, and occurred in lower yields ranging from
2
5–60%. Interestingly, the reaction is proposed to proceed via
an oxygen radical which participates in the cyclization reac-
tion, however, following the addition of 1.0 equivalent of bu- Conclusions
tylated hydroxytoluene (BHT), a known radical inhibitor, the
Overall, we have developed a highly efficient method for the
photochemical reaction, and subsequent yield, appeared to be
unaffected, 60+%. An investigation of this reaction mecha-
nism and the development of a protocol for the flow photo-
chemical synthesis of benzoxazoles are proposed for future
study.
synthesis of substituted phenanthridinones using a flow
photochemical cyclization reaction. Using a single step flow
protocol, we were able to produce 16 examples of the photoflow
cyclization of 2-chlorobenzamides which occurred in yields up
to 99%. We were also able to combine two separate synthetic
steps into a single flow process which reacted 2-chlorobenzoyl
chlorides with anilines, developing a novel flow method for
the generation of substituted phenanthridinones in good
yields. Currently, mechanistic studies are underway on the
photocyclization reaction, and we will shortly initiate testing of
our molecules for activity against polyIJADP-ribose) polymerase
in a variety of disease-state models. In the near future, we plan
to expand the scope of this reaction to include other substrates
that will be tested for activity and selectively as PARP inhibi-
tors. The research contained herein represents one of the first
examples of the use of continuous flow photochemistry for the
synthesis of drug-like molecules and can serve as an example
of how the many benefits of flow chemistry can enable the
process of drug discovery.
One of the many benefits of flow chemistry is the ability to
combine several separate synthetic steps into a single flow se-
quence. An early example of this approach is the total synthe-
2
3
sis of oxomaritidine, which combined seven separate syn-
thetic steps into a single flow sequence. It was our plan to
develop a photochemical flow synthesis of phenanthridinones
that would take simple commercially available starting mate-
rials and produce a tricyclic heterocycle, combining a two-step
batch sequence into a novel single flow method. To accom-
plish this task, a 2-chlorobenzoyl chloride is loaded into a sin-
gle sample loop (pump A) while a corresponding aniline is
loaded into a second sample loop (pump B). Each of these
flow streams are then connected via a T-piece through a 10
mL reactor coil to generate an amide which is then directed
through the UV-150 photochemical reactor at a total flow rate
−
1
of 0.2 mL min . In each case, we found the reaction
proceeded as expected and a summary of this single (two-step)
flow protocol can be found in Table 2 where a 2-chlorobenzoyl
chloride was combined with an aniline to produce the corre-
sponding phenanthridinone in yields which are slightly lower
than the single photochemical reaction of the corresponding
amide. Traditionally, the amidation reaction is typically
Acknowledgements
The authors would like to offer a debt of gratitude to the Col-
lege of Pharmacy at the University of Manitoba for financial
support and a University start-up grant (GKT).
Notes and references
3
performed in cooperation with a base, such as Et N or pyri-
dine. We found, however, that the two-step reaction gave
higher isolated yields when no additional base was used. We
1 A. Gomtsyan, Chem. Heterocycl. Compd., 2012, 48, 7–10.
2 J. J. Li, Heterocyclic chemistry in drug discovery, John Wiley
and Sons, Hoboken, 2013.
3
4
5
6
7
M. Baumann, I. Baxendale and S. Ley, Mol. Diversity,
011, 15, 613–630.
J. P. Knowles, L. D. Elliott and K. I. Booker-Milburn, Beilstein
J. Org. Chem., 2012, 8, 2025–2052.
Table 2 Multi-step phenanthridinone flow-hv synthesis via amidation of
anilines and 2-chlorobenzoyl chloride
2
K. Gilmore and P. H. Seeberger, Chem. Rec., 2014, 14,
4
10–418.
D. Kopetzki, F. Lévesque and P. H. Seeberger, Chem. – Eur.
J., 2013, 19, 5450–5456.
L. D. Elliott, J. P. Knowles, P. J. Koovits, K. G. Maskill, M. J.
Ralph, G. Lejeune, L. J. Edwards, R. I. Robinson, I. R.
Clemens, B. Cox, D. D. Pascoe, G. Koch, M. Eberle, M. B.
Berry and K. I. Booker-Milburn, Chem. – Eur. J., 2014, 20,
a
Entry
R
1
R
2
R
3
R
4
R
5
R
6
Yield
1
2
3
4
5
H
Cl
H
H
H
H
COMe
H
H
H
H
H
Cl
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
72
47
51
74
77
1
5226–15232.
8 J. Wegner, S. Ceylan and A. Kirschning, Adv. Synth. Catal.,
012, 354, 17–57.
OMe
2
a
9 J. C. Pastre, D. L. Browne and S. V. Ley, Chem. Soc. Rev.,
2013, 42, 8849–8869.
Isolated yields; 10 mL flourinated ethylene propylene (FEP) coil,
0 °C, Vapourtec UV-150 medium pressure Hg lamp (75%, ∼112 W).
6
Med. Chem. Commun.
This journal is © The Royal Society of Chemistry 2016

View Article Online
MedChemComm
Research Article
1
1
0 D. Webb and T. F. Jamison, Chem. Sci., 2010, 1, 675–680.
1 D. T. McQuade and P. H. Seeberger, J. Org. Chem., 2013, 78,
17 T. Ekblad, E. Camaioni, H. Schüler and A. Macchiarulo,
FEBS J., 2013, 280, 3563–3575.
6
384–6389.
18 J.-H. Li, L. Serdyuk, D. V. Ferraris, G. Xiao, K. L. Tays, P. W.
Kletzly, W. Li, S. Lautar, J. Zhang and V. J. Kalish, Bioorg.
Med. Chem. Lett., 2001, 11, 1687–1690.
1
1
1
1
2 Z. Herceg and Z.-Q. Wang, Mutat. Res., Fundam. Mol. Mech.
Mutagen., 2001, 477, 97–110.
3 C. Underhill, M. Toulmonde and H. Bonnefoi, Ann. Oncol.,
19 B. S. Bhakuni, A. Kumar, S. J. Balkrishna, J. A. Sheikh, S.
Konar and S. Kumar, Org. Lett., 2012, 14, 2838–2841.
20 J. V. Suarez-Meneses, E. Bonilla-Reyes, E. A. Ble-Gonzalez,
M. C. Ortega-Alfaro, R. A. Toscano, A. Cordero-Vargas and
J. G. Lopez-Cortes, Tetrahedron, 2014, 70, 1422–1430.
21 J. Grimshaw and A. P. de Silva, J. Chem. Soc., Chem.
Commun., 1980, 302–303.
22 I.-S. Bae, Y.-S. Kim and Y.-T. Park, Bull. Korean Chem. Soc.,
2003, 24, 916–920.
23 I. R. Baxendale, J. Deeley, C. M. Griffiths-Jones, S. V. Ley, S.
Saaby and G. K. Tranmer, Chem. Commun., 2006, 2566–2568.
2
011, 22, 268–279.
4 M. Rouleau, A. Patel, M. J. Hendzel, S. H. Kaufmann and
G. G. Poirier, Nat. Rev. Cancer, 2010, 10, 293–301.
5 E. Wahlberg, T. Karlberg, E. Kouznetsova, N. Markova, A.
Macchiarulo, A.-G. Thorsell, E. Pol, A. Frostell, T. Ekblad, D.
Oncu, B. Kull, G. M. Robertson, R. Pellicciari, H. Schuler and
J. Weigelt, Nat. Biotechnol., 2012, 30, 283–288.
1
6 P. Jagtap, F. G. Soriano, L. Virág, L. Liaudet, J. Mabley, E.
Szabó, G. Haskó, A. Marton, C. B. Lorigados and F. Gallyas
Jr, Crit. Care Med., 2002, 30, 1071–1082.
This journal is © The Royal Society of Chemistry 2016
Med. Chem. Commun.