Evaluation of the anti-inflammatory effects of synthesised tanshinone I and isotanshinone I analogues in zebrafish

Article abstract of DOI:10.1371/journal.pone.0240231

During inflammation, dysregulated neutrophil behaviour can play a major role in a range of chronic inflammatory diseases, for many of which current treatments are generally ineffective. Recently, specific naturally occurring tanshinones have shown promising anti-inflammatory effects by targeting neutrophils in vivo, yet such tanshinones, and moreover, their isomeric isotanshinone counterparts, are still a largely underexplored class of compounds, both in terms of synthesis and biological effects. To explore the anti-inflammatory effects of isotanshinones, and the tanshinones more generally, a series of substituted tanshinone and isotanshinone analogues was synthesised, alongside other structurally similar molecules. Evaluation of these using a transgenic zebrafish model of neutrophilic inflammation revealed differential anti-inflammatory profiles in vivo, with a number of compounds exhibiting promising effects. Several compounds reduce initial neutrophil recruitment and/or promote resolution of neutrophilic inflammation, of which two also result in increased apoptosis of human neutrophils. In particular, the methoxy-substituted tanshinone 39 specifically accelerates resolution of inflammation without affecting the recruitment of neutrophils to inflammatory sites, making this a particularly attractive candidate for potential pro-resolution therapeutics, as well as a possible lead for future development of functionalised tanshinones as molecular tools and/or chemical probes. The structurally related β-lapachones promote neutrophil recruitment but do not affect resolution. We also observed notable differences in toxicity profiles between compound classes. Overall, we provide new insights into the in vivo anti-inflammatory activities of several novel tanshinones, isotanshinones, and structurally related compounds.

Full text of DOI:10.1371/journal.pone.0240231

PLOS ONE
RESEARCH ARTICLE
Evaluation of the anti-inflammatory effects of
synthesised tanshinone I and isotanshinone I
analogues in zebrafish
Matthew J. Foulkes1,2,3, Faith H. Tolliday2,3, Katherine M. Henry2,3, Stephen A. Renshaw2,3
,
1
Simon Jones
*
ID
1 Department of Chemistry, The University of Sheffield, Sheffield, United Kingdom, 2 The Bateson Centre,
The University of Sheffield, Sheffield, United Kingdom, 3 Department of Infection, Immunity & Cardiovascular
Disease, The University of Sheffield, Sheffield, United Kingdom
a1111111111
a1111111111
a1111111111
a1111111111
a1111111111
* simon.jones@sheffield.ac.uk
Abstract
During inflammation, dysregulated neutrophil behaviour can play a major role in a range of
chronic inflammatory diseases, for many of which current treatments are generally ineffec-
tive. Recently, specific naturally occurring tanshinones have shown promising anti-inflam-
matory effects by targeting neutrophils in vivo, yet such tanshinones, and moreover, their
isomeric isotanshinone counterparts, are still a largely underexplored class of compounds,
both in terms of synthesis and biological effects. To explore the anti-inflammatory effects of
isotanshinones, and the tanshinones more generally, a series of substituted tanshinone and
isotanshinone analogues was synthesised, alongside other structurally similar molecules.
Evaluation of these using a transgenic zebrafish model of neutrophilic inflammation
revealed differential anti-inflammatory profiles in vivo, with a number of compounds exhibit-
ing promising effects. Several compounds reduce initial neutrophil recruitment and/or pro-
mote resolution of neutrophilic inflammation, of which two also result in increased apoptosis
of human neutrophils. In particular, the methoxy-substituted tanshinone 39 specifically
accelerates resolution of inflammation without affecting the recruitment of neutrophils to
inflammatory sites, making this a particularly attractive candidate for potential pro-resolution
therapeutics, as well as a possible lead for future development of functionalised tanshinones
as molecular tools and/or chemical probes. The structurally related β-lapachones promote
neutrophil recruitment but do not affect resolution. We also observed notable differences in
toxicity profiles between compound classes. Overall, we provide new insights into the in vivo
anti-inflammatory activities of several novel tanshinones, isotanshinones, and structurally
related compounds.
OPEN ACCESS
Citation: Foulkes MJ, Tolliday FH, Henry KM,
Renshaw SA, Jones S (2020) Evaluation of the
anti-inflammatory effects of synthesised
tanshinone I and isotanshinone I analogues in
zebrafish. PLoS ONE 15(10): e0240231. https://doi.
org/10.1371/journal.pone.0240231
Editor: Aran Singanayagam, Imperial College
London, UNITED KINGDOM
Received: December 18, 2019
Accepted: September 22, 2020
Published: October 6, 2020
Copyright: © 2020 Foulkes et al. This is an open
access article distributed under the terms of the
Creative Commons Attribution License, which
permits unrestricted use, distribution, and
reproduction in any medium, provided the original
author and source are credited.
Data Availability Statement: All relevant data are
within the manuscript and its Supporting
Information files.
Funding: This work was supported by a Medical
Research Council (MRC) Programme Grant to S.A.
R. (MR/M004864/1) and an MRC Centre Grant
G0700091. The authors also thank The University
Of Sheffield for funding (M.J.F.). The funders had
no role in study design, data collection and
analysis, decision to publish, or preparation of the
manuscript.
Introduction
Tanshinones are a group of aromatic diterpenoid compounds comprising four fused rings;
two of these form a naphthalene or tetrahydronaphthalene moiety, the third ring usually bears
PLOS ONE | https://doi.org/10.1371/journal.pone.0240231 October 6, 2020
1 / 25

PLOS ONE
Synthesised anti-inflammatory compounds in zebrafish
Competing interests: The authors have declared
an ortho- (or para-) quinone, and the fourth ring is a furan or dihydrofuran. Tanshinone I
(TI) 1 is one of the main known tanshinones found naturally in the plant Salvia miltiorrhiza
(also known as Chinese red sage or danshen), whilst others include tanshinone IIA (TIIA) 2,
and the dihydro counterparts, dihydrotanshinone I 3 and cryptotanshinone 4 (Fig 1). The
plant has been traditionally used as a remedy in Chinese herbal medicine since as early as the
first century AD, mainly for treatment of cardiovascular disease [1]. Isolation of milligram
quantities of tanshinones from the plant usually requires serial extractions from the dried
roots, followed by chromatographic separation [2–5].
that no competing interests exist.
Isolated tanshinones have been reported to exert various biological activities both in vitro
and in vivo. In particular, for TI 1 these include anti-inflammatory effects [6–9], antimicrobial
activity [10, 11], cytotoxicity towards cancer cells [12–15], neuroprotective effects [16], benefits
for learning and memory enhancement [17], antioxidant activity [18, 19], and the ability to
induce cell apoptosis [12, 20, 21]. In most cases however, the mechanism of action is poorly
understood, and there is often limited or no knowledge of structure-activity relationships
(SARs).
β-Lapachone 5 and nor-β-lapachone 6 are molecules with a high degree of structural simi-
larity to tanshinones (Fig 2). β-Lapachone 5 is a naturally-occurring ortho-quinone compound
obtained from the bark of the lapacho tree Tabebuia avellanedae in Brazil, a plant used for cen-
turies as a traditional medicine for various treatments, including as an analgesic, an anti-
inflammatory agent, an antineoplastic agent, and a diuretic [22, 23]. In more recent studies, β-
lapachone 5 has exhibited anti-inflammatory effects [24–29], as well as various anti-cancer
activities [30–33], anti-angiogenesis activity [34, 35], antimicrobial effects [36–38], and wound
healing properties [39]. In addition, both β-lapachone 5 and TIIA 2 have been found to act as
inhibitors of NAD(P)H quinone oxidoreductase 1 (NQO1), leading to an increase in cellular
NAD(+) [40–42]. This recent finding connecting the two structures, alongside the clear struc-
tural similarities, may suggest that these two classes of compounds could exhibit common bio-
logical activities in vivo. In contrast, reports of previous studies involving nor-β-lapachone 6
are scarce. Although nor-β-lapachone 6 is considered an anti-cancer drug candidate [43, 44],
at this time there appear to be no studies involving nor-β-lapachone 6 in relation to inflamma-
tion, neutrophils or zebrafish, perhaps because β-lapachone 5 is a naturally occurring com-
pound, whereas nor-β-lapachone 6 is not.
Neutrophils, the most abundant white blood cell in humans, play a vital role in the inflam-
matory response. Their recruitment to the site of pathogen invasion or tissue injury is essential
for elimination of invading pathogens by various mechanisms [45–49]. Strict regulation of
neutrophil function is required: whilst these processes are required for effective host defence
and response, they also need to be limited to avoid persistent inflammation [50]. Subsequent
Fig 1. Structures of some naturally occurring tanshinones 1–4.
https://doi.org/10.1371/journal.pone.0240231.g001
PLOS ONE | https://doi.org/10.1371/journal.pone.0240231 October 6, 2020
2 / 25

PLOS ONE
Synthesised anti-inflammatory compounds in zebrafish
Fig 2. Structures of β-lapachone 5 and nor-β-lapachone 6.
https://doi.org/10.1371/journal.pone.0240231.g002
timely neutrophil clearance from this site, as part of the resolution of inflammation, is also nec-
essary for healthy recovery [51, 52]. Failure of this process can result in retention of persistent
inflammatory neutrophils, which can have damaging effects on the body, playing a key role in
increasingly prevalent inflammatory diseases, for example in chronic obstructive pulmonary
disease (COPD) [53, 54], asthma [55, 56], or rheumatoid arthritis [57, 58]. Current treatments
for unresolved neutrophilic inflammation are non-specific, poorly effective, and can exhibit
many undesired side-effects [59, 60]; thus there is an unmet need for new neutrophil-specific
clinical treatments.
The zebrafish (Danio rerio) is an excellent in vivo model for the study of inflammation and
neutrophil biology [61], and has been utilised effectively in a number of recent studies in the
discovery of previously unidentified compounds with anti-inflammatory properties in vivo
[62–65]. Screening in a transgenic zebrafish inflammation model identified the naturally
occurring TIIA 2 as a compound which significantly accelerated resolution of neutrophilic
inflammation in vivo, without affecting initial neutrophil recruitment or total neutrophil num-
ber [63]. The related molecule cryptotanshinone 4 displayed broadly similar anti-inflamma-
tory effects, although this compound also decreased initial neutrophil recruitment. However,
no other tanshinones have been investigated in this manner, meaning that no knowledge of
any analogous effects of tanshinones based on TI 1 currently exists in the literature; this is also
true for the isotanshinones.
Naturally occurring tanshinones are also inherently carbon-rich and planar, meaning they
possess sub-optimal solubilities and physicochemical properties. To this end, incorporation of
an additional heteroatom and a possible functional group handle for development of related
chemical probes and/or molecular tool compounds, provides further rationale for develop-
ment and evaluation of non-natural tanshinone and isotanshinone analogues. As little is
understood about the in vivo anti-inflammatory effects of these compounds, besides the natu-
rally occurring, unfunctionalised tanshinones, an increased understanding of any SARs would
provide increased drug discovery opportunities for future research. Furthermore, SARs both
within and between different tanshinone classes (TI and TIIA) in this model are currently
unexplored. We here report synthesis of a set of tanshinone I and isotanshinone I analogues,
and use of a zebrafish model for evaluation of the effects of such compounds on different stages
of the in vivo inflammatory response utilising a phenotypic screening approach, as well as ini-
tial assessment of any in vivo toxicity.
PLOS ONE | https://doi.org/10.1371/journal.pone.0240231 October 6, 2020
3 / 25

PLOS ONE
Synthesised anti-inflammatory compounds in zebrafish
Fig 3. Synthesis of tanshinones and isotanshinones.
https://doi.org/10.1371/journal.pone.0240231.g003
Results and discussion
Chemical synthesis of TI analogues and isomers
Synthesis of TI analogues utilised a route similar to that of Jiao and co-workers [66], but with
important modifications (Fig 3). Commercially available 5-bromovanillin 7 was first converted
to the hydroquinone 8 via Baeyer-Villiger oxidation and subsequent ester hydrolysis, followed
by Fe(III)-mediated oxidation to the corresponding benzoquinone 9 [67]. Both transforma-
tions were performed on a multi-gram scale (up to 25 grams) in high yield, with a lack of chro-
matographic purification representing a more time- and cost-effective large-scale isolation of
the desired benzoquinone 9 compared to previous reports. This material was treated with vari-
ous substituted carboxylic acids in a radical decarboxylative coupling. Initial use of 3-
(2-methylphenyl)propionic acid 10, with conditions identified from the literature [66],
resulted in low yields of alkylation product 11 of approximately 20% after chromatographic
purification. Reaction optimisation, through systematic variation of numerous reaction
parameters using the unsubstituted 3-phenylpropionic acid 12, in a Design of Experiments
(DoE) approach (further information provided in the S1 File), led to improved yields of
around 40%, including a 42% yield of alkylation product 11. Throughout these studies,
although formation of a small quantity (< 10%) of doubly-alkylated product was observed,
none of the alternative mono-alkylated product (alkylated adjacent to the methoxide group)
was seen, which was consistent with previous reports, and likely a directing effect due to the
bromide substituent [66, 68].
Use of different carboxylic acids in this step was then considered. The choice of several dif-
ferent substituted carboxylic acids, which would eventually provide various final tanshinone
and isotanshinone molecules, was based on several factors. One of these was a desire to include
a mixture of electron-withdrawing, electron-donating, and electron-neutral groups in the final
molecules, thereby perhaps changing the properties of the tanshinones as a whole. The inclu-
sion of fluorine-containing groups was also of interest, as it was considered that this may have
some effect on in vivo metabolism, in comparison to the solely hydrogen-containing
PLOS ONE | https://doi.org/10.1371/journal.pone.0240231 October 6, 2020
4 / 25

PLOS ONE
Synthesised anti-inflammatory compounds in zebrafish
analogues, although detailed studies and predictions of this nature were beyond the scope of
this work. Variation of the substituent on the phenyl ring of the carboxylic acid in this step
allowed for modification of a position in the final tanshinone as far away as possible from the
most chemically and biologically interesting part of the molecule, the ortho-quinone moiety. It
was hypothesised that modifying in such a position would thereby have a minimal effect on
the interaction of such compounds with their molecular target(s) in vivo. Furthermore, car-
boxylic acid choice was limited to an extent by the commercial availability of such substrates.
Thus, the optimised conditions were applied to the reaction for other substituted carboxylic
acids, and products bearing methyl 11, hydrogen 13, trifluoromethyl 14 and fluoro 15 substit-
uents were all formed in consistently moderate yields of 38–54%. For the methoxy substituent,
the alkylation product 16 was formed in an inseparable 60:40 mixture with the starting mate-
rial 9, corresponding to an approximate 33% overall yield of desired product 16 (see S1 File for
further details).
Employing the conditions for the intramolecular Heck reaction used in the literature for
the methyl-substituted bromide 11 [66], only gave the desired cyclisation product in a maxi-
mum of 40% yield after purification, as a mixture of aromatised 17 and non-aromatised 18
compounds. This was unacceptably low given the high quantities of palladium catalyst used
(45 mol%) and use of a much smaller quantity of catalyst was desired. Optimisation was car-
ried out using the unsubstituted variant 13 as a model system (see S1 File for further details);
whilst changing the phosphine ligand or adding a co-oxidant to the mixture had no beneficial
effect on the yield, performing the reaction at a much lower concentration (approximately
0.01 M), led to a greatly improved of 74% when just 10 mol% of catalyst was used, and this was
conserved (73% yield) when the catalyst loading was reduced to 5 mol%. These results thus
suggested that this particular intramolecular reaction was especially sensitive to concentration
effects. In addition, throughout these studies, varying ratios of aromatised 19 and non-aroma-
tised products 20 were obtained; however, any non-aromatised product 20 was converted to
the fully aromatised product in subsequent steps.
The improved Heck reaction conditions were used to prepare the corresponding analogues
17–26. Whilst mass returns were good for the methyl- and hydrogen-substituted variants 17–
18 and 19–20, heteroatom-substituted compounds were lower. For the methoxy-substituted
compound 16, residual benzoquinone 9 was able to be removed, and the desired products 25–
26 were successfully produced although also with a low mass return (see S1 File for further
details).
Demethylation of the mixture of inseparable methyl ethers 17–18 was achieved by heating
at reflux in a 2 M sodium hydroxide solution with ethanol, with concomitant aromatisation
successfully yielding solely the desired alcohol 27. This avoided the need for additional oxygen
gas to be supplied to the reaction, as previously utilised in the literature [66], and worked well
for all analogues 27–30, and for the methoxy compound 31, where only the single desired
demethylation product was observed.
In a change of conditions from the literature procedure, reaction of the alcohol 27 with
chloroacetone as the solvent with 1 equivalent of ammonium acetate resulted in isolation of
the desired TI 1, together with the isomeric isotanshinone I (iso-TI) 32 [66, 69–72]. The two
isomers were each isolated in 12% yield over three steps, presumably formed via tautomeric
intermediates. This reaction worked consistently for each of the different analogues to yield
tanshinones 1, 33, 35, 37, 39 and isotanshinones 32, 34, 36, 38, 40. Yields were higher for the
electron-neutral methyl- and hydrogen-substituted compounds, and were somewhat lower for
the heteroatom-substituted variants. Only a few examples of isotanshinones are known in the
literature, including iso-TI 32 and isotanshinone IIA 41 [69, 70, 72, 73], and related isotanshi-
nones [74, 75]. These all occur naturally in Salvia miltiorrhiza, although isotanshinone I 32
PLOS ONE | https://doi.org/10.1371/journal.pone.0240231 October 6, 2020
5 / 25

PLOS ONE
Synthesised anti-inflammatory compounds in zebrafish
Fig 4. Synthesis of the TI-acetal 43.
https://doi.org/10.1371/journal.pone.0240231.g004
and isotanshinone IIA 41 have been produced synthetically [69, 70]. However as a class of
compounds, isotanshinones are very much underexplored, both in terms of synthesis and bio-
logical studies. Thus, this synthesis provided efficient access to a range of both substituted tan-
shinone I analogues and isotanshinones of interest, most of which are completely novel
compounds.
To investigate the importance of the ortho-quinone moiety of TI 1 for its biological activity
(as a functional group common to all major tanshinones), a ‘masked’ analogue lacking this
functionality was desired. Synthesis of a dimethyl acetal analogue of TI 1 was readily accom-
plished in a two-step procedure (Fig 4) by reduction of the ortho-quinone 1 with sodium boro-
hydride, followed by quickly treating the diol 42 with 2,2-dimethoxypropane and acid giving
the stable acetal 43 in 34% yield over two steps.
Chemical synthesis of lapachols and β-lapachones
Structurally related β-lapachones were synthesised readily from commercially sourced lapa-
chol 44 (Fig 5). β-Lapachone 5 was synthesised in a single step and in high yield from treat-
ment of lapachol 44 with concentrated sulfuric acid.
Norlapachol 45 was also produced from lapachol, via a Hooker oxidation, in an unopti-
mized yield of 52% after recrystallisation. Finally, nor-β-lapachone 6 was synthesised from its
corresponding precursor norlapachol 45, again via acid-mediated cyclisation, in a near-quan-
titative yield of 95% after simple purification by trituration.
Evaluation of the in vivo anti-inflammatory effects of TI and TIIA
To explore the anti-inflammatory effects of the different tanshinone sub-classes, TI 1 and TIIA
2 were first evaluated in a transgenic zebrafish model of inflammation (Fig 6, Table 1), as pre-
viously described [63, 64]. For neutrophil recruitment experiments, the commercially available
compound SP600125 (SP) was used as the positive control, as a selective and cell-permeable
inhibitor of c-Jun N- terminal kinase (JNK), which prevents activation of various inflamma-
tory genes [76, 77]. Investigation of the effects of the tanshinones on initial recruitment of neu-
trophils to the site of tailfin injury (Fig 6a) revealed that TI 1 (25 μM) reduced the number of
neutrophils recruited at 6 hours post injury (hpi), although had no effect at 10 μM. TIIA 2,
however, did not significantly affect neutrophil recruitment at either 10 or 25 μM, in agree-
ment with previous findings for TIIA 2 using this model [63]. For experiments investigating
resolution of neutrophilic inflammation in vivo (Fig 6b), TIIA 2 (25 μM) was used as the posi-
tive control compound, as this has previously been shown to exhibit highly significant pro-res-
olution activity in this zebrafish model of inflammation [63]. Both TI 1 and TIIA 2 accelerated
resolution of neutrophilic inflammation, at 10 and 25 μM.
PLOS ONE | https://doi.org/10.1371/journal.pone.0240231 October 6, 2020
6 / 25

PLOS ONE
Synthesised anti-inflammatory compounds in zebrafish
Fig 5. Synthesis of norlapachol 45 and β-lapachones 5, 6.
https://doi.org/10.1371/journal.pone.0240231.g005
Evaluation of the effects of TI analogues and isomers on neutrophil
recruitment and resolution in vivo
The synthesised TI analogues and isomers 1, 32–40, 43 were evaluated for any effects on neu-
trophil recruitment in the same zebrafish inflammation model (Table 1, Fig 7), tested using
the same experimental conditions across three sets due to practical constraints on experiment
size, and again with DMSO and SP as the negative and positive control treatments, respec-
tively. As expected, TI 1 significantly reduced the number of neutrophils recruited to the site
Fig 6. Effects of TI 1 and TIIA 2 on a) initial neutrophil recruitment and b) resolution of neutrophilic
inflammation in a zebrafish inflammation model. For a), tailfin transection was performed on transgenic zebrafish
larvae, TgBAC(mpx:GFP)i114 (3 days post-fertilisation, dpf), which were subjected to compound treatments
immediately: DMSO (0.5%), SP600125 (30 μM) and TI 1 and TIIA 2 (10, 25 μM as indicated in brackets). Neutrophil
number at the site of injury was assessed at 6 hpi, by counting GFP+ cells. For b), tailfin transection was performed on
zebrafish larvae (3 dpf), and at 4 hpi, good responders were subjected to compound treatments: DMSO (0.5%), TI 1
and TIIA 2 (10 μM, 25 μM as indicated in brackets). Neutrophil number at the site of injury was assessed at 8 hpi by
counting GFP+ cells. For both graphs, data shown as mean SEM; n = 21–32 larvae per group from 3–4 independent
experiments per graph. � P < 0.05, �� P < 0.01, ��� P < 0.001, ���� P < 0.0001, compared to the DMSO vehicle control
(one-way ANOVA with Dunnett’s multiple comparison post-test).
https://doi.org/10.1371/journal.pone.0240231.g006
PLOS ONE | https://doi.org/10.1371/journal.pone.0240231 October 6, 2020
7 / 25

PLOS ONE
Synthesised anti-inflammatory compounds in zebrafish
Table 1. Summary of the effects of various tanshinones and isotanshinones on neutrophil recruitment and resolu-
tion of neutrophilic inflammation in vivo.
Compound
TI 1
Effect on recruitment
Reduced
Effect on resolution
Enhanced
Enhanced
Enhanced
No effect
TIIA 2
No effect
Iso-TI 32
Reduced
6-hydro-TI 33
No effect
6-trifluoromethyl-TI 35
6-fluoro-TI 37
6-methoxy-TI 39
4-hydro-iso-TI 34
4-trifluoromethyl-iso-TI 36
4-fluoro-iso-TI 38
4-methoxy-iso-TI 40
TI-acetal 43
No effect
No effect
No effect
No effect
No effect
Enhanced
No effect
No effect
Reduced
No effect
Reduced
No effect
No effect
No effect
No effect
Enhanced
https://doi.org/10.1371/journal.pone.0240231.t001
of injury in all three datasets (Fig 7b–7d). Generally, treatment with iso-TI 32 also resulted in a
significant decrease in the number of neutrophils recruited (Fig 7b–7d), with one exception.
The tanshinone analogues 33, 35, 37, 39 are derived from the parent TI structure 1
(Table 1, Fig 7b and 7c, data shown in red). Hydrogen substitution (compound 33) did not
change neutrophil recruitment compared with the DMSO control. Similarly, compounds
substituted with trifluoromethyl (35), fluorine (37) or methoxy (39) groups had no effect on
neutrophil recruitment.
Fig 7. Effects of TI analogues and isomers on neutrophil recruitment. a) Structures of evaluated tanshinones (red),
isotanshinones (blue) and the TI-acetal (green). b)-d) Tailfin transection was performed on transgenic zebrafish larvae,
TgBAC(mpx:GFP)i114 (3 dpf) which were subjected to compound treatments immediately: DMSO (0.5%), SP600125
(30 μM) and compounds 1, 32–40, 43 (25 μM). Neutrophil number at the site of injury was assessed at 6 hpi by
counting GFP+ cells. Data shown as mean SEM; n = 30–39 larvae from 4 independent experiments per graph. �
P < 0.05, �� P < 0.01, ��� P < 0.001, ���� P < 0.0001, ns not significant, in comparison to the DMSO vehicle control
(one-way ANOVA with Dunnett’s multiple comparison post-test).
https://doi.org/10.1371/journal.pone.0240231.g007
PLOS ONE | https://doi.org/10.1371/journal.pone.0240231 October 6, 2020
8 / 25

PLOS ONE
Synthesised anti-inflammatory compounds in zebrafish
The isotanshinones 34, 36, 38, 40 are derived from the parent iso-TI structure 32 (Table 1,
Fig 7c and 7d, relevant data depicted in blue). Replacement of the 4-methyl group with a
hydrogen atom (34) again resulted in the loss of any effect on neutrophil recruitment to the
site of injury. However, replacement of this methyl group with either a trifluoromethyl group
(36) or a fluorine atom (38) resulted in significantly fewer recruited neutrophils in each case,
although the methoxy- substituted analogue 40 had no significant effect on recruitment. The
TI-acetal 43 also had no significant effect on neutrophil recruitment to the site of injury (Fig
7d), possibly due to a lack of the active form TI 1 in vivo.
In general (except for iso-TI 32 in a single dataset), compounds with a methyl substituent
have a clear effect in reducing neutrophil recruitment to the site of injury, yet removal of any
functionality at this position leads to a complete loss of biological effect, possibly due to
decreased aqueous solubility, and/or loss of a particular interaction with the molecular target
(s) in vivo. Compounds with different substituents in the modified position show variable
effects on neutrophil recruitment in vivo, with some isotanshinones resulting in decreased
recruitment, unlike the corresponding tanshinones.
The synthesised TI analogues and isomers 1, 32–40, 43 were then evaluated in a similar
zebrafish model of inflammation resolution (Table 1, Fig 8). For resolution experiments, TIIA
2 was again used as the positive control compound (Fig 8b–8d). Firstly considering the ortho-
quinone tanshinone compounds 1, 33, 35, 37, 39 (Fig 8b and 8c, data shown in red), in larvae
treated with TI 1, there was a trend towards lower neutrophil numbers at the site of injury,
compared to the DMSO-treated larvae (P = 0.063). Replacement of the methyl group in the
6-position (1) with a hydrogen atom (33) resulted in no effect on resolution in vivo. Related
compounds containing a trifluoromethyl group (35) or fluorine atom (37) also showed no
effect. Interestingly, for the 6-methoxy substituted tanshinone 39, there was a clear reduction
in neutrophil number at the site of injury at 8 hpi, in comparison to the DMSO control-treated
larvae, indicating that this compound accelerates resolution of inflammation. This compound
Fig 8. Effects of TI analogues and isomers on resolution of neutrophilic inflammation. a) Structures of tanshinones
(red), isotanshinones (blue) and TI-acetal (green). b)-d) Tailfin transection was performed on transgenic zebrafish
larvae, TgBAC(mpx:GFP)i114 (3 dpf), and at 4 hpi, good responders were subjected to compound treatments: DMSO
(0.5%), TIIA 2 (25 μM) and compounds 1, 32–40, 43 (25 μM). Neutrophil number at the site of injury was assessed at 8
hpi by counting GFP+ cells. Data shown as mean SEM; n = 26–30 larvae per group from 4 independent experiments
per graph. � P < 0.05, �� P < 0.01, ��� P < 0.001, ns not significant, in comparison to the DMSO control (one-way
ANOVA with Dunnett’s multiple comparison post-test).
https://doi.org/10.1371/journal.pone.0240231.g008
PLOS ONE | https://doi.org/10.1371/journal.pone.0240231 October 6, 2020
9 / 25

PLOS ONE
Synthesised anti-inflammatory compounds in zebrafish
39 was of particular interest as it promoted resolution of neutrophilic inflammation but did
not affect initial neutrophil recruitment, making this compound a more promising candidate
for further investigation of specific pro-resolution treatments which do not impair host-
defence. Furthermore, the methoxy group also provides a possible functionalisation point,
which could be useful for developing tool compounds and molecular probes. The methoxy
group also likely confers improved aqueous solubility to the molecule.
For the isotanshinones 32, 34, 36, 38, 40, (Fig 8b–8d, data shown in blue), iso-TI 32 signifi-
cantly accelerated neutrophilic inflammation resolution, whilst the unsubstituted compound
34 again had no effect. The remaining isotanshinones 36, 38, 40 had no effect on inflammation
resolution, irrespective of the functional group in the 4-position. Treatment with the TI-acetal
43 resulted in a reduction in neutrophil number at the site of injury (Fig 8d). One possible
explanation is that the acetal 43 was (partially) hydrolysed and metabolised back to the parent
TI 1 in vivo, which then exerted a pro-resolution effect as seen previously (Fig 6b). These
results suggest that a substituent in the 6- or 4- position for tanshinones and isotanshinones
respectively, is required for activity, the nature of which is particularly important for pro-reso-
lution activity.
Overall, both substituted tanshinones and isotanshinones exhibit differential effects on
recruitment and resolution (Table 1). This is not particularly surprising, as molecules which
affect neutrophil recruitment likely interact with a different molecular target to those which
affect resolution of neutrophilic inflammation. Of particular note are: the abilities of both of
the parent compounds TI 1 and iso-TI 32 to decrease neutrophil recruitment to the site of
injury and accelerate inflammation resolution; isotanshinones 36 and 38 both resulting in
decreased neutrophil recruitment to the injury site yet having no effect on resolution of neu-
trophilic inflammation; and the methoxy-substituted tanshinone 39 exhibiting no effect on
neutrophil recruitment, yet significantly promoting inflammation resolution (analogous to the
findings observed for TIIA 2). This suggests that this particular compound may be an interest-
ing candidate to focus on to investigate compounds which specifically target inflammation res-
olution, as well as providing a lead for synthesis of future functionalised tanshinones. In
addition, the methoxy group of this compound may be improving the aqueous solubility of
this molecule, compared to previously studied naturally occurring tanshinones.
Comparison of the fluorine-containing compounds with their non-fluorine-containing
analogues could, on some level, provide insights into how in vivo metabolism may affect bio-
logical activity. Thus, whereas TI 1 affected both neutrophil recruitment and resolution, nei-
ther the unsubstituted tanshinone 33, the trifluoromethyl-substituted variant 35, nor the
fluorine-substituted compound 37 significantly affected either process. This may suggest that
the methyl group is important for biological for activity, since its removal or replacement with
fluorine atoms (in either the methyl group or instead of an aryl hydrogen) both resulted in loss
of in vivo activity. Meanwhile, for the isotanshinones, replacement of the methyl group with
either a trifluoromethyl group (compound 36) or a fluorine atom (compound 38) did not
affect neutrophil recruitment, with both compounds resulting in reduced neutrophil recruit-
ment, like the parent iso-TI 32. However, whereas inflammation resolution was enhanced for
iso-TI 32, both of the fluoro-containing compounds 36 and 38 had no effect on resolution.
This suggests that for isotanshinones, changing the methyl group to a fluoro-containing group
can be tolerated in relation to effects on neutrophil recruitment, but not resolution. This may
also suggest in this case that in vivo C-H bond metabolism (or some other factor relating to
C-F bond substitution) is important for acceleration of inflammation resolution, but not for
neutrophil recruitment. Yet, with the tanshinones, the results may suggest that C-H bond
metabolism (or other factor(s)) may be important for effects on both neutrophil recruitment
and resolution. Clearly, further work would be required to explore this in more detail.
PLOS ONE | https://doi.org/10.1371/journal.pone.0240231 October 6, 2020
10 / 25

PLOS ONE
Synthesised anti-inflammatory compounds in zebrafish
Besides efficacy, another important consideration in drug discovery and development is
that of toxicity; indeed, these are the two main reasons accounting for the vast majority of the
high drug attrition rates in the pharmaceutical industry [78–80]. One of the great advantages
of using this in vivo zebrafish model to evaluate potential anti-inflammatory compounds is
that it also allows an early stage qualitative assessment of any side-effects on a living organism.
Accordingly, the general health of treated zebrafish larvae was also observed for all experi-
ments. At aqueous concentrations of 25 μM, none of the evaluated tanshinones and isotanshi-
nones resulted in any larval toxicity; larvae were generally healthy in all treatments, as
determined by visual inspection of general body shape, tail shape, presence of circulation, and
heartbeat. However, it is important to note that toxicity was only assessed at the end of the
experiment, after 4–6 hours immersion in the compound of interest. A thorough toxicity
experiment would require leaving the fish in the compound solution for longer periods.
Evaluation of the effects of lapachols and β-lapachones on neutrophil
recruitment and resolution in vivo
Norlapachol 45, lapachol 44, β-lapachone 5, and nor-β-lapachone 6 were also evaluated in
neutrophil recruitment experiments (Table 2, Fig 9). To explore possible dose-response effects,
norlapachol 45 was tested at 0.1, 1, and 10 μM; a higher dose of 25 μM appeared to result in lar-
val abnormalities. Similarly, the remaining three compounds were all evaluated at 1 μM. This
stood in contrast to the use of the various tanshinones at 25 μM previously, in which larvae
were all healthy at this concentration, and is a worthwhile consideration when determining
the viability of such molecules further in the drug discovery and development process. These
observations suggest that the structural differences between these compounds have important
consequences on their toxicity profiles in vivo. This could be due to differences in molecular
target interactions, differences in larval penetration, and/or differences in neutrophil penetra-
tion, possibly due to interactions with relevant drug transporter proteins [81]. TI 1 (25 μM)
was used as the positive control, based on the previous findings and to provide a synthetic con-
trol compound. Norlapachol 45 reduced the number of neutrophils recruited to the site of
injury at both 10 μM and 0.1 μM (Fig 9a). Given the structural differences between norlapachol
45 and tanshinones, norlapachol 45 may have worked in a different way to tanshinones such
as TI 1, possibly by acting on a different protein target. Recruitment was not significantly
reduced at 1 μM. The difference between the effects observed at 1 μM and 0.1 μM may reflect
the natural variability in this type of in vivo experiment, despite all efforts to control for this as
far as reasonably practicable.
The remaining three compounds were evaluated in the same way but at 1 μM (Fig 9b). TI 1
was used at both 25 μM (as a known positive control) and 1 μM, to allow comparison of the
effects of this compound to the other compounds tested at the same dosage. β-Lapachone 5 led
Table 2. Summary of the effects of norlapachol 45, lapachol 44, β-lapachone 5, and nor-β-lapachone 6 on neutrophil recruitment and resolution of inflammation in
vivo.
Compound
Concentration / μM
Effect on recruitment
Effect on resolution
Norlapachol 45
25
10
1
Toxic
Enhanced (���
Enhanced (�)
No effect
)
Reduced (���
)
No effect
Reduced (��
No effect
0.1
1
)
No effect
Lapachol 44
β-Lapachone 5
No effect
1
Reduced (�)
No effect
Nor-β-Lapachone 6
1
No effect
No effect
https://doi.org/10.1371/journal.pone.0240231.t002
PLOS ONE | https://doi.org/10.1371/journal.pone.0240231 October 6, 2020
11 / 25

PLOS ONE
Synthesised anti-inflammatory compounds in zebrafish
Fig 9. Effects of a) norlapachol 45 and b) lapachol 44, β-lapachone 5 and nor-β-lapachone 6 on neutrophil recruitment. Tailfin transection was
performed on transgenic zebrafish larvae, TgBAC(mpx:GFP)i114 (3 dpf) which were subjected to compound treatments immediately: DMSO (0.5%), TI
1 (25 μM; also 1 μM for b only, as indicated in brackets), norlapachol 45 (a; 0.1, 1, 10 μM as indicated) and lapachol 44, β-lapachone 5 and nor-β-
lapachone 6 (b; 1 μM). Neutrophil number at the site of injury was assessed at 6 hpi by counting GFP+ cells. Data shown as mean SEM; n = 29–32
larvae from 4 independent experiments per graph. � P < 0.05, �� P < 0.01, ��� P < 0.001, ���� P < 0.0001, ns not significant, in comparison to the
DMSO control (one-way ANOVA with Dunnett’s multiple comparison post-test).
https://doi.org/10.1371/journal.pone.0240231.g009
to a significant reduction in the number of neutrophils recruited to the site of injury. However,
neither nor-β-lapachone 6 nor lapachol 44 reduced neutrophil recruitment, nor did TI 1 at
this concentration. No signs of toxicity were observed amongst the larvae of any of the treat-
ment groups.
These results were particularly interesting, as β-lapachone 5 is an ortho-quinone compound
which is structurally similar to TI 1, and led to reduced neutrophil numbers at the injury site,
as observed with TI 1 at a 25 μM concentration. However, TI 1 showed no effect when used at
a concentration of 1 μM, suggesting that β-lapachone 5 may have been more active than TI 1
in vivo. This would clearly require further investigation, but is of interest given the toxicity
issues observed with the β-lapachones yet not with TI 1 in this study.
The same four compounds were evaluated for any effects on resolution of neutrophilic
inflammation in vivo (Table 2, Fig 10). Norlapachol 45 was tested at a range of different con-
centrations from 0.1 to 25 μM, as the compound was not toxic to larvae at 25 μM in these
experiments. The higher threshold for toxicity in resolution experiments may be attributed to
the shorter treatment time with compound (4 hours instead of 6 hours in recruitment studies)
and the fact that the tissue has been allowed to heal somewhat before application of com-
pounds. Lapachol 44, β-lapachone 5 and nor-β-lapachone 6 were each tested at 1 μM, as higher
doses were toxic. TIIA 2 (25 μM) was used as a positive control, as in previous resolution
experiments. Norlapachol 45 had no effect on the number of neutrophils at 0.1 and 1 μM (Fig
10a). However, a reduced neutrophil number at the site of injury was observed at 10 μM, and a
highly significant reduction was seen at a concentration of 25 μM, although some of the larvae
did show slow or absent circulation, possibly indicative of a toxic effect of norlapachol 45 at
this concentration. Overall, the resolution data for norlapachol 45 show a good dose-response
relationship. Finally, the remaining compounds lapachol 44, β-lapachone 5, and nor-β-lapa-
chone 6 were analysed in similar inflammation resolution experiments, at 1 μM (Fig 10b). In
these experiments, TIIA 2 was used at 25 μM (as the positive control) and 1 μM, again to com-
pare the effects of this compound to the other compounds tested at the same concentration.
However, none of these experimental compounds had any effect on resolution of neutrophilic
inflammation.
PLOS ONE | https://doi.org/10.1371/journal.pone.0240231 October 6, 2020
12 / 25

PLOS ONE
Synthesised anti-inflammatory compounds in zebrafish
Fig 10. Effects of a) norlapachol 45 and b) lapachol 44, β-lapachone 5 and nor-β-lapachone 6 on resolution of neutrophilic inflammation. Tailfin
transection was performed on transgenic zebrafish larvae, TgBAC(mpx:GFP)i114 (3 dpf), and at 4 hpi, good responders were subjected to compound
treatments: DMSO (0.5%), TIIA 2 (25 μM; also 1 μM for b only, as indicated in brackets), norlapachol 45 (a; 0.1, 1, 10, 25 μM as indicated) and lapachol
44, β-lapachone 5 and nor-β-lapachone 6 (b; 1 μM). Neutrophil number at the site of injury was assessed at 8 hpi by counting GFP+ cells. Data shown
as mean SEM; n = 24–29 larvae from 3–4 independent experiments per graph. � P < 0.05, �� P < 0.01, ��� P < 0.001, ���� P < 0.0001, ns not
significant, in comparison to the DMSO control (one-way ANOVA with Dunnett’s multiple comparison post-test).
https://doi.org/10.1371/journal.pone.0240231.g010
Overall, norlapachol 45 affected both neutrophil recruitment to the site of injury and reso-
lution at higher concentrations, yet at lower concentrations, only initial neutrophil recruit-
ment was significantly affected (Table 2). Lapachol 44 exhibited no observable effect on either
recruitment or resolution, and neither did nor-β-lapachone 6. At a low concentration, β-lapa-
chone 5, affected only recruitment of neutrophils, and not resolution of neutrophilic
Fig 11. Effects of active compounds on neutrophil apoptosis. Freshly isolated human neutrophils were incubated for
6 hours at 37 ᵒC with the following compound treatments: DMSO, and compounds 1, 32, 39 and 43 (25 μM), with
DMSO used as a vehicle control. Percentage neutrophil apoptosis for each condition was quantified from duplicate
cytospins. Cells were classified as either apoptotic or non-apoptotic based on the morphology of their nuclei. Data
shown as mean SEM; n = 4000 neutrophils from 4 independent experiments. ��� P < 0.001, ���� P < 0.0001, in
comparison to the DMSO control (one-way ANOVA with Dunnett’s multiple comparison post-test).
https://doi.org/10.1371/journal.pone.0240231.g011
PLOS ONE | https://doi.org/10.1371/journal.pone.0240231 October 6, 2020
13 / 25

PLOS ONE
Synthesised anti-inflammatory compounds in zebrafish
inflammation. This may suggest a specific function of this compound (at this concentration)
in earlier stages of the inflammatory response.
Evaluation of the effects of active compounds on neutrophil apoptosis
Finally, to investigate effects on neutrophil structure and function directly, and begin to
explore translational relevance, extending our results to a human system, the most active com-
pounds in vivo were taken forward to evaluate their effects on neutrophil apoptosis. Freshly
isolated human neutrophils were cultured for 6 hours with TI 1 and iso-TI 32, both of which
significantly affected resolution and recruitment of zebrafish neutrophils; and 6-methoxy-TI
39 and TI-acetal 43, which accelerated inflammation resolution but had no effect on recruit-
ment. The percentage of neutrophil apoptosis was assessed based on nuclear morphology.
Quantification revealed that treatment with TI 1 and 6-methoxy-TI 39 resulted in a significant
increase in apoptosis, whilst no such effect was observed for treatment with either iso-TI 32 or
TI-acetal 43 (Fig 11). These data suggest that, at least for the structurally similar tanshinones 1
and 39, neutrophil apoptosis substantially contributes to the anti-inflammatory activity of
these compounds demonstrated in the zebrafish model. These data also correlate well with pre-
vious neutrophil apoptosis studies with the structurally similar TIIA 2, which indicated that
treatment with TIIA 2 resulted in an increase in neutrophil apoptosis, both in vivo in the zeb-
rafish model, and in vitro with human neutrophils [63]. However, for the compounds lacking
the ortho-quinone moiety, iso-TI 32 and TI-acetal 43, neutrophil apoptosis appears to play a
much smaller role in the anti-inflammatory effects of these compounds, suggesting either a dif-
ferent mechanism (such as reverse migration of neutrophils), or (perhaps more likely) differ-
ing relative contributions of multiple mechanisms, compared to the tanshinones.
Conclusions
A small library of analogues and isomers of TI 1 was synthesised from commercially available
starting materials using an optimised route, which allowed access to novel tanshinone deriva-
tives as well as the much underexplored isotanshinones, about which little is known generally,
including both chemically and biologically. Synthesis of an acetal variant of TI 1, as well as
structurally similar β-lapachones, was also undertaken. Evaluation of these compounds in a
transgenic zebrafish model of inflammation revealed that various related molecules exhibited
anti-inflammatory effects in vivo. Small changes in the molecular structures of these com-
pounds resulted in varying effects on different stages of the inflammatory response, allowing
for some broad SARs to be constructed. Three compounds– β-lapachone 5 and isotanshinones
36 and 38—resulted in decreased neutrophil recruitment to the site of injury, but did not affect
resolution of neutrophilic inflammation, whilst the methoxy-substituted tanshinone 39 accel-
erated inflammation resolution yet did not affect initial neutrophil recruitment. This com-
pound is considered a particularly attractive candidate for potential pro-resolution
therapeutics, as it exhibited an anti-inflammatory effect without affecting the recruitment of
neutrophils to inflammatory sites. This compound also provides a basis for future studies, for
example for future design and synthesis of molecular probes and other tool compounds. The
acetal 43 also affected resolution but not neutrophil recruitment. Three compounds–tanshi-
nones 1 and 35, and isotanshinone 32—reduced initial neutrophil recruitment and accelerated
inflammation resolution. Furthermore, two of these compounds, tanshinones 1 and 39,
resulted in significant increases in apoptosis of human neutrophils. This study also revealed
differences in the toxicity profiles between tanshinones and the closely related β-lapachones,
exemplifying the utility of using zebrafish phenotypic screening in identifying safety issues
with potential drug treatments early in the drug discovery and development process. Overall,
PLOS ONE | https://doi.org/10.1371/journal.pone.0240231 October 6, 2020
14 / 25

PLOS ONE
Synthesised anti-inflammatory compounds in zebrafish
this work has enabled an increased insight into the anti-inflammatory activities of various
substituted tanshinones, isotanshinones, and structurally related molecules, working towards
identification of candidates for clinical treatment of dysregulated inflammation which act spe-
cifically on inflammatory neutrophils.
Experimental section
Chemical synthesis
General reagents, materials and methods. All chemicals used were purchased from com-
mercial suppliers and were used as received without further purification. Melting points were
determined using a Gallenkamp melting point apparatus equipped with a thermometer. IR
spectroscopy was performed on a PerkinElmer FT-IR Spectrum 65 or Spectrum 100 spectrom-
eter, using either NaCl discs or a Universal diamond ATR. 1H, 13C and 19F NMR experiments
were run on either a Bruker Avance 400 or Bruker Avance III HD 500 spectrometer at 298 K.
Chemical shifts (δ) are reported in parts per million (ppm) relative to the deuterated lock
solvent as an internal standard, where s = singlet, d = doublet, t = triplet, q = quartet,
m = multiplet, br s = broad singlet, br d = broad doublet, br t = broad triplet, dd = doublet of
doublets, ddd = double doublet of doublets, dt = doublet of triplets, td = triplet of doublets. All
coupling constants are reported in hertz, Hz. Mass spectrometry was carried out on either an
Agilent Technologies 6530 or 7200 spectrometer, using either electron impact (EI) or electro-
spray ionisation (ESI). TLC was performed on Merck silica gel 60 F254 aluminium-backed
plates and visualised using ultraviolet light followed by staining with potassium permanganate
dip. Column chromatography was carried out using silica gel obtained from VWR Chemicals,
particle size 40–63 μm. Reactions using microwave conditions were carried out on a CEM
Corporation Discover S-class microwave synthesiser, at pressure � 17 bar and power � 200
W. Synthesis of all intermediates for the tested compounds are provided in the S1 File.
General representative procedure A for reactions with chloroacetone to form tanshi-
nones 1, 33, 35, 37, 39 and isotanshinones 32, 34, 36, 38, 40. A mixture of the alcohol 27–
31 (100 mg, 0.42 mmol), ammonium acetate (32 mg, 0.42 mmol, 4 mmol g-1) and chloroace-
tone (10 mL g-1, 12 mmol) was heated in a sealed tube at 110 ˚C for 15 minutes under micro-
wave conditions. The solution was cooled to room temperature, diluted with water (400 mL
g-1) and extracted with DCM (3 x 400 mL g-1). The organic extracts were combined, dried
(MgSO4), filtered and concentrated in vacuo to give the crude mixture which was purified by
flash column chromatography (silica gel, DCM) to give the tanshinones 1, 33, 35, 37, 39 and
isotanshinones 32, 34, 36, 38, 40.
1,6-Dimethylphenanthro[1,2-b]furan-10,11-dione, TI 1 and 4,8-dimethylphenanthro
[3,2-b]furan-7,11-dione, iso-TI 32. General procedure A was followed, using the alcohol 27
(100 mg, 0.42 mmol) to give the dione 1 (21 mg, 12% over 3 steps) as a dark red solid; mp 232–
234 ˚C (lit. [70] 229–230 ˚C); δH(400 MHz; CDCl3) 9.23 (1 H, d, J 8.8, ArCH), 8.27 (1 H, d, J
8.8, ArCH), 7.77 (1 H, d, J 8.8, ArCH), 7.54 (1 H, dd, J 8.8, 7.1, ArCH), 7.34 (1 H, d, J 7.1,
ArCH), 7.30 (1 H, s, OCH), 2.68 (3 H, s, CH3), 2.30 (3 H, d, J 0.8, CH3); δC(100 MHz; CDCl3)
183.4 (C = O), 175.6 (C = O), 161.2 (ArC), 142.0 (ArCH), 135.2 (ArC), 133.6 (ArC), 132.9
(ArCH), 132.7 (ArC), 130.6 (ArCH), 129.6 (ArC), 128.3 (ArCH), 124.8 (ArCH), 123.1 (ArC),
121.8 (ArC), 120.5 (ArC), 118.7 (ArCH), 19.9 (CH3), 8.8 (CH3); m/z (ESI+) 299 (10%, M+Na+),
277 (100, M+H+). All data were in general agreement with the literature [66, 70, 71]. Also
obtained was the dione 32 (22 mg, 12% over 3 steps) as an orange solid; mp 218–220 ˚C (lit.
[69] 219–220 ˚C); νmax(ATR)/cm-1 1655 (C = O), 1588 (C = O), 1533 (C = C); δH(400 MHz;
CDCl3) 9.66 (1 H, d, J 8.9, ArCH), 8.40 (1 H, dd, J 8.9, 0.8, ArCH), 8.32 (1 H, d, J 8.9, ArCH),
7.63 (1 H, dd, J 8.9, 7.0, ArCH), 7.54–7.52 (1 H, m, OCH), 7.48 (1 H, dt, J 7.0, 0.8, ArCH), 2.76
PLOS ONE | https://doi.org/10.1371/journal.pone.0240231 October 6, 2020
15 / 25

PLOS ONE
Synthesised anti-inflammatory compounds in zebrafish
(3 H, s, ArCH3), 2.41 (3 H, d, J 1.2, CH3); δC(101 MHz; CDCl3) 182.2 (C = O), 177.2 (C = O),
153.9 (ArC), 145.2 (ArCH), 136.0 (ArC), 134.7 (ArC), 133.5 (ArC), 131.2 (ArC), 130.9 (ArCH),
129.8 (ArCH), 129.3 (ArCH), 127.2 (ArC), 126.5 (ArC), 126.2 (ArCH), 122.1 (ArCH), 120.8
(ArC), 20.0 (CH3), 8.7 (CH3); m/z (ESI+) 299 (25%, M+Na+), 277 (100, M+H+). No 13C NMR
spectroscopy data were reported in the literature; all other data were in general agreement
with the literature [69, 70, 72].
1-Methylphenanthro[1,2-b]furan-10,11-dione 33 and 8-methylphenanthro[3,2-b]furan-
7,11-dione 34. General procedure A was followed, using the alcohol 28 (100 mg, 0.44 mmol)
to give the dione 33 (24 mg, 15% over 3 steps) as a dark red solid; mp 229–233 ˚C (lit. [66]
226–229 ˚C); νmax(NaCl discs)/cm-1 2922 (C-H), 2852 (C-H), 1669 (C = O), 1593 (C = C);
δH(400 MHz; CDCl3) 9.42 (1 H, d, J 8.7, ArCH), 8.12 (1 H, d, J 8.7, ArCH), 7.83 (2 H, br d, J
8.7, 2 x ArCH), 7.71 (1 H, br t, J 7.7, ArCH), 7.55 (1 H, t, J 7.7, ArCH), 7.33 (1 H, s, ArCH),
2.32 (3 H, s, CH3). No IR spectroscopy data were reported in the literature. 1H NMR data were
in broad agreement with the literature [66], although precise chemical shift values were slightly
shifted due to the different solvent used for analysis. Also obtained was the dione 34 (28 mg,
17% over 3 steps) as an orange solid; mp 199–202 ˚C; νmax(NaCl discs)/cm-1 2921 (C-H), 1766
(C = O), 1664 (C = O), 1536 (C = C); δH(400 MHz; CDCl3) 9.77 (1 H, d, J 8.5, ArCH), 8.28 (1
H, d, J 8.5, ArCH), 8.18 (1 H, d, J 8.5, ArCH), 7.90 (1 H, d, J 8.5, ArCH), 7.78–7.72 (1 H, m,
ArCH), 7.68–7.61 (1 H, m, ArCH), 7.53 (1 H, s, ArCH), 2.40 (3 H, s, CH3); δC(101 MHz;
CDCl3) 182.2 (C = O), 177.2 (C = O), 153.7 (ArC), 145.3 (ArCH), 136.7 (ArC), 135.1 (ArCH),
134.0 (ArC), 130.9 (ArC), 130.1 (ArCH), 128.7 (ArCH), 128.5 (ArCH), 128.0 (ArCH), 127.0
(ArC), 126.5 (ArC), 122.3 (ArCH), 120.9 (ArC), 8.7 (CH3); m/z (EI+) 262.0622 (100%, M+
C17H10O3 requires 262.0624), 234 (42), 206 (40), 176 (36), 151 (15), 126 (8), 87 (7), 76 (8).
1-Methyl-6-(trifluoromethyl)phenanthro[1,2-b]furan-10,11-dione 35 and 8-methyl-4-
(trifluoromethyl)phenanthro[3,2-b]furan-7,11-dione 36. General procedure A was fol-
lowed, using the unpurified alcohol 29 (435 mg) to give the dione 35 (6 mg, 1% over 3 steps) as
a dark red solid; mp 204–208 ˚C; νmax(NaCl discs)/cm-1 1674 (C = O), 1550 (C = C); δH(400
MHz; CDCl3) 9.65 (1 H, d, J 8.9, ArCH), 8.50 (1 H, d, J 8.9, ArCH), 7.98 (1 H, d, J 8.9, ArCH),
7.94 (1 H, d, J 7.2 ArCH), 7.75 (1 H, br t, J 8.1, ArCH), 7.39 (1 H, br d, J 1.1, OCH), 2.33 (3 H,
d, J 1.1, CH3); δC(100 MHz; CDCl3) 183.3(C = O), 175.2 (C = O), 160.2 (ArC), 142.8 (ArCH),
133.0 (ArC), 132.8 (q, JC-F 3.0, ArCH), 131.0 (ArCH), 130.5 (ArC), 129.8 (ArC), 129.0 (ArCH),
127.0 (q, JC-F 30.3, ArC), 126.0 (q, JC-F 5.9, ArCH), 124.2 (q, JC-F 273.9, CF3), 123.1 (ArC), 122.1
(ArC), 121.2 (ArC), 120.8 (ArCH), 8.8 (CH3); δF(377 MHz; CDCl3) -58.8; m/z (ESI+) 353 (18%,
M+Na+), 331.0580 (100, M+H+ C18H10O3F3 requires 331.0577). Also obtained was the dione
36 (20 mg, 2% over 3 steps) as a dark brown solid; mp 183–186 ˚C; νmax(NaCl discs)/cm-1
1661 (C = O), 1601 (C = O), 1536 (C = C); δH(400 MHz; CDCl3) 10.07 (1 H, d, J 9.0, ArCH),
8.58 (1 H, d, J 9.0, ArCH), 8.45 (1 H, d, J 9.0, ArCH), 8.05 (1 H, d, J 7.1 ArCH), 7.79 (1 H, dd, J
9.0, 7.1, ArCH), 7.57 (1 H, q, J 1.1, OCH), 2.42 (3 H, d, J 1.1, CH3); δC(100 MHz; CDCl3) 181.4
(C = O), 176.6 (C = O), 153.5 (ArC), 145.7 (ArCH), 134.0 (ArC), 132.4 (ArCH), 132.2 (ArC),
131.6 (ArC), 130.8 (q, JC-F 2.7, ArCH), 128.2 (ArCH), 127.2 (ArC), 127.0 (q, JC-F 5.9, ArCH),
126.7 (d, JC-F 3.0, ArC), 126.5 (ArC), 124.3 (q, JC-F 274.0, CF3), 124.0 (ArCH), 121.0 (ArC), 8.7
(CH3); δF(377 MHz; CDCl3) -59.0; m/z (EI+) 330.0496 (90%, M+ C18H9O3F3 requires
330.0498), 261 [100, (M-CF3)+].
6-Fluoro-1-methylphenanthro[1,2-b]furan-10,11-dione 37 and 4-fluoro-8-methylphe-
nanthro[3,2-b]furan-7,11-dione 38. General procedure A was followed, using the unpuri-
fied alcohol 30 (140 mg) to give the dione 37 (18 mg, 2% over 3 steps) as a dark red solid; mp
187–191 ˚C; νmax(NaCl discs)/cm-1 1675 (C = O), 1664 (C = O), 1594 (C = C); δH(400 MHz;
CDCl3) 9.20 (1 H, d, J 8.8, ArCH), 8.45 (1 H, d, J 8.8, ArCH), 7.89 (1 H, d, J 8.8, ArCH), 7.67–
7.60 (1 H, m, ArCH), 7.36 (1 H, q, J 1.2, OCH), 7.22 (1 H, ddd, J 10.0, 7.8, 0.7, ArCH), 2.32 (3
PLOS ONE | https://doi.org/10.1371/journal.pone.0240231 October 6, 2020
16 / 25

PLOS ONE
Synthesised anti-inflammatory compounds in zebrafish
H, d, J 1.2, CH3); δC(100 MHz; CDCl3) 183.0 (C = O), 175.0 (C = O), 160.5 (ArC), 158.8 (d,
JC-F 253.6, ArC), 142.5 (ArCH), 133.2 (d, JC-F 2.5, ArC), 131.0 (d, JC-F 8.5, ArCH), 130.9 (ArC),
129.4 (d, JC-F 7.3, ArCH), 124.7 (d, JC-F 15.4, ArC), 122.5 (d, JC-F 4.7, ArCH), 122.3 (d, JC-F 2.3,
ArC), 122.0 (ArC), 121.0 (ArC), 119.3 (d, JC-F 1.6, ArCH), 111.1 (d, JC-F 19.3, ArCH), 8.8
(CH3); δF(377 MHz; CDCl3) -120.7; m/z (ESI+) 326 (34%, M+2Na+), 303 (20, M+Na+),
281.0612 (100, M+H+ C17H10O3F requires 281.0608). Also obtained was the dione 38 (12 mg,
1% over 3 steps) as a pale orange solid; mp 185–189 ˚C; νmax(NaCl discs)/cm-1 1662 (C = O),
1591 (C = O), 1535 (C = C); δH(400 MHz; CDCl3) 9.56 (1 H, d, J 8.9, ArCH), 8.50 (1 H, d, J 8.9,
ArCH), 8.34 (1 H, d, J 8.9, ArCH), 7.71–7.64 (1 H, m, ArCH), 7.55 (1 H, d, J 1.0, OCH), 7.32 (1
H, dd, J 9.9, 7.8, ArCH), 2.41 (3 H, d, J 1.0, CH3); δC(100 MHz; CDCl3) 181.8 (C = O), 176.7 (C
= O), 158.4 (d, JC-F 253.0, ArC), 153.6 (ArC), 145.5 (ArCH), 134.5 (ArC), 132.0 (d, JC-F 2.8,
ArC), 130.0 (d, JC-F 8.1, ArCH), 127.5 (d, JC-F 6.9, ArCH), 127.1 (d, JC-F 15.8, ArC), 126.8 (d,
J
C-F 2.6, ArC), 126.7 (ArC), 124.0 (d, JC-F 4.6, ArCH), 122.6 (d, JC-F 1.6, ArCH), 121.0 (ArC),
112.1 (d, JC-F 19.3, ArCH), 8.7 (CH3); δF(377 MHz; CDCl3) -120.9; m/z (EI+) 280.0532 (100%,
M+ C17H9O3F requires 280.0530).
6-Methoxy-1-methylphenanthro[1,2-b]furan-10,11-dione 39 and 4-methoxy-8-methyl-
phenanthro[3,2-b]furan-7,11-dione 40. General procedure A was followed, using the alco-
hol 31 (100 mg, 0.39 mmol) to give the dione 39 (22 mg, 5% over 3 steps) as a dark brown
solid; mp 249–253 ˚C; νmax(NaCl discs)/cm-1 1670 (C = O), 1660 (C = O), 1587 (C = C), 1548
(C = C); δH(400 MHz; CDCl3) 9.00 (1 H, d, J 8.8, ArCH), 8.66 (1 H, d, J 8.8, ArCH), 7.82 (1 H,
d, J 8.8, ArCH), 7.62 (1 H, br t, J 8.4, ArCH), 7.34 (1 H, s, ArCH), 6.89 (1 H, d, J 7.8, ArCH),
4.04 (3 H, s, OCH3), 2.32 (3 H, s, CH3); δC(126 MHz; CDCl3) 183.4 (C = O), 175.7 (C = O),
161.2 (ArC), 155.7 (ArC), 142.1 (ArCH), 133.5 (ArC), 131.5 (ArCH), 131.2 (ArCH), 130.5
(ArC), 126.8 (ArC), 122.4 (ArC), 121.8 (ArC), 120.7 (ArC), 118.5 (ArCH), 118.2 (ArCH), 105.5
(ArCH), 55.7 (OCH3), 8.8 (CH3); m/z (ESI+) 315 (11%, M+Na+), 293.0813 (100, M+H+
C18H13O4 requires 293.0808). Also obtained was the dione 40 (21 mg, 5% over 3 steps) as a red
solid; mp 231–234 ˚C; νmax(ATR)/cm-1 1662 (C = O), 1582 (C = O), 1535 (C = C); δH(400
MHz; CDCl3) 9.32 (1 H, d, J 8.8, ArCH), 8.68 (1 H, d, J 8.8, ArCH), 8.24 (1 H, d, J 8.8, ArCH),
7.63 (1 H, dd, J 8.8, 7.9, ArCH), 7.51 (1 H, d, J 0.8, ArCH), 6.95 (1 H, d, J 7.9, ArCH), 4.04 (3 H,
s, OCH3), 2.40 (3 H, d, J 0.8, CH3); δC(101 MHz; CDCl3) 182.3 (C = O), 177.2 (C = O), 155.3
(ArC), 153.9 (ArC), 145.1 (ArCH), 134.4 (ArC), 132.0 (ArC), 130.5 (ArCH), 129.3 (ArC), 129.2
(ArCH), 126.51 (ArC), 126.49 (ArC), 121.6 (ArCH), 120.8 (ArC), 119.8 (ArCH), 106.2 (ArCH),
55.7 (OCH3), 8.8 (CH3); m/z (ESI+) 315 (43%, M+Na+), 293.0810 (100, M+H+ C18H13O4
requires 293.0808).
1,6-Dimethylphenanthro[1,2-b]furan-10,11-dimethyldioxole, TI-acetal 43. TI 1 (27
mg, 0.10 mmol) was dissolved in anhydrous THF (7 mL) and stirred under a nitrogen atmo-
sphere at room temperature. Sodium borohydride (4.0 mg, 0.10 mmol) was added in a single
portion, and the solution stirred for 30 minutes. The solution was poured onto ice/water (20
mL), acidified with aqueous HCl solution (1 M, 3 mL), and extracted with DCM (3 x 20 mL).
The organic extracts were combined, dried (MgSO4), filtered and concentrated in vacuo to
give the crude diol 42 as an olive-green solid (29 mg), which was immediately dissolved in
anhydrous toluene (10 mL). 2,2-Dimethoxypropane (0.025 mL, 0.20 mmol) and para-toluene-
sulfonic acid monohydrate (25 mg, 0.13 mmol) were added, and the reaction was stirred
under a nitrogen atmosphere at reflux for 2 h. The mixture was cooled to room temperature,
aqueous saturated NaHCO3 solution (20 mL) added, and extracted with diethyl ether (3 x 20
mL). The organic extracts were combined, dried (MgSO4), filtered and concentrated in vacuo
to give an off-white solid (38 mg), which was purified twice by flash column chromatography
(silica gel, DCM, and then silica gel, 19:1 40/60 petroleum ether/ethyl acetate) to give the acetal
43 as a beige solid (11 mg, 34%); mp 176–180 ˚C; νmax(ATR)/cm-1 2917 (C-H); δH(400 MHz;
PLOS ONE | https://doi.org/10.1371/journal.pone.0240231 October 6, 2020
17 / 25

PLOS ONE
Synthesised anti-inflammatory compounds in zebrafish
CDCl3) 9.13 (1 H, d, J 8.4, ArCH), 8.22 (1 H, d, J 9.2, ArCH), 7.89 (1 H, dd, J 9.2, 0.8, ArCH),
7.57–7.52 (2 H, m, 2 x ArCH), 7.45 (1 H, dt, J 7.1, 0.9, ArCH), 2.79 (3 H, s, ArCH3), 2.46 (3 H,
d, J 1.3, ArCH3), 1.91 (6 H, s, 2 x CH3); δC(126 MHz; CDCl3) 148.3 (ArC), 141.4 (ArCH), 138.2
(ArC), 137.8 (ArC), 134.1 (ArC), 130.7 (ArC), 128.9 (ArC), 127.3 (ArCH), 125.6 (ArCH), 125.4
(ArCH), 120.4 (ArCH), 119.2 (ArCH), 118.5 (ArC), 114.5 (ArC), 114.3 (ArC), 113.0 (ArC),
112.7 (ArC), 26.1 (2 x CH3), 20.2 (ArCH3), 9.2 (ArCH3); m/z (ESI+) 319.1322 (100%, M+H+
C21H19O3 requires 319.1329), 383 (8), 359 (28), 261 (22).
β-Lapachone 5. Concentrated H2SO4 (1.0 mL) was added slowly to lapachol 44 (25 mg,
0.10 mmol), with stirring, until the solid completely dissolved. The solution was stirred for a
further 5 minutes and poured onto ice (20 g). The precipitate was filtered off by vacuum filtra-
tion and washed with cold water (5 mL) to give an orange solid. The crude product was puri-
fied by flash column chromatography (silica gel, DCM– 19:1 DCM/MeOH) to give the dione 5
(21 mg, 85%) as an orange solid; mp 151–153 ˚C (lit. [82] 152–154 ˚C); δH(400 MHz; CDCl3)
8.07 (1 H, dd, J 7.7, 1.0, ArCH), 7.83 (1 H, d, J 7.7, ArCH), 7.66 (1 H, td, J 7.7, 1.0, ArCH), 7.52
(1 H, td, J 7.7, 1.0, ArCH), 2.58 (2 H, t, J 6.7, CH2), 1.87 (2 H, t, J 6.7, CH2), 1.48 (6 H, s, 2 x
CH3); δC(101 MHz; CDCl3) 179.9 (C = O), 178.6 (C = O), 162.1 (ArC), 134.8 (ArCH), 132.6
(ArC), 130.7 (ArCH), 130.1 (ArC), 128.6 (ArCH), 124.1 (ArCH), 112.7 (ArC), 79.3 [C(CH3)2],
31.6 (CH2), 26.8 (2 x CH3), 16.2 (CH2). All data were in general agreement with the literature
[82, 83].
Norlapachol 45. Lapachol 44 (150 mg, 0.62 mmol) was dissolved in THF (5 mL), sodium
carbonate (72 mg, 0.68 mmol) in water (5 mL) was added, and the solution was heated to 60
˚C. Aqueous hydrogen peroxide (30%, 1.0 mL, 13 mmol) was added and the solution was
heated at 60 ˚C for 4 h. The reaction was cooled to rt, acidified with concentrated HCl (0.2
mL), and quenched with sodium sulfite (3.0 g). Aqueous NaOH (25%, 5 mL) and copper(II)
sulfate (625 mg, 3.9 mmol) in water (5 mL) were added and the mixture was stirred at room
temperature for 1.5 h. The solution was filtered through a pad of Celite, acidified with concen-
trated HCl (7 mL) and extracted with diethyl ether (3 x 75 mL). The organic extracts were
combined, washed with brine (200 mL), dried (MgSO4), filtered and concentrated in vacuo to
give an orange solid. The crude product was purified by recrystallisation from n-hexane to
give the dione 45 (73 mg, 52%) as an orange solid; mp 123–125 ˚C (from n-hexane), (lit. [83]
121–122 ˚C); νmax(NaCl discs)/cm-1 3364 (O-H), 2928 (C-H), 2910 (C-H), 1662 (C = O), 1645
(C = O), 1627 (C = C), 1593 (C = C); δH(400 MHz; CDCl3) 8.15 (1 H, d, J 7.6, ArCH), 8.11 (1
H, d, J 7.6, ArCH), 7.78 (1 H, td, J 7.6, 1.1, ArCH), 7.71 (1 H, td, J 7.6, 1.1, ArCH), 7.55 (1 H, br
s, OH), 6.02 (1 H, br s, CH), 2.01 (3 H, s, CH3), 1.70 (3 H, s, CH3); δC(101 MHz; CDCl3) 184.8
(C = O), 181.6 (C = O), 151.2 (C), 143.6 (C), 134.9 (ArCH), 133.0 (ArCH), 132.9 (C), 129.5 (C),
126.9 (ArCH), 126.1 (ArCH), 120.9 (C), 113.7 (CH), 26.6 (CH3), 21.8 (CH3); m/z (ESI+) 251
(5%, M+Na+), 229.0862 (100, M+H+ C14H13O3 requires 229.0859). All data were in general
agreement with the literature [83–85].
Nor-β-lapachone 6. Concentrated H2SO4 (1.5 mL) was added slowly to norlapachol 45
(30 mg, 0.13 mmol), with stirring, until the solid completely dissolved. The solution was stirred
at room temperature for a further 5 minutes, poured onto ice (20 g) and rinsed with water (20
mL). The solution was extracted with DCM (3 x 40 mL), and the organic extracts were com-
bined, washed with brine (100 mL), dried (MgSO4), filtered, and concentrated in vacuo to give
an orange/pink solid. The crude product was purified by trituration with cold hexane (2 x 2
mL) and dried under high vacuum to give the dione 6 (29 mg, 95%) as an orange solid; mp
190–193 ˚C (lit. [86] 188–189 ˚C); δH(400 MHz; CDCl3) 8.10 (1 H, d, J 7.6, ArCH), 7.68–7.64
(2 H, m, 2 x ArCH), 7.63–7.57 (1 H, m, ArCH), 2.97 (2 H, s, CH2), 1.63 (6 H, s, 2 x CH3);
δC(101 MHz; CDCl3) 181.4 (C = O), 175.7 (C = O), 168.8 (ArC), 134.5 (ArCH), 131.9 (ArCH),
PLOS ONE | https://doi.org/10.1371/journal.pone.0240231 October 6, 2020
18 / 25

PLOS ONE
Synthesised anti-inflammatory compounds in zebrafish
130.9 (ArC), 129.3 (ArCH), 127.9 (ArC), 124.6 (ArCH), 115.0 (ArC), 93.8 [C(CH3)2], 39.3
(CH2), 28.4 (2 x CH3). All data were in general agreement with the literature [83, 86].
Biological evaluation
Ethical permissions. Human neutrophil purification for this study was approved by the
National Research Ethics Committee—Yorkshire & The Humber—Sheffield (05/Q2305/4).
Fully informed written consent was obtained, and all clinical investigation was conducted
according to the principles expressed in the Declaration of Helsinki. Zebrafish work in this
study was approved by Sheffield University Animal Welfare and Ethical Review Body, Project
Applications and Amendments Committee as part of the approval of the Home Office Project
Licence PPL70/8178.
Reagents. All reagents were obtained from Sigma Aldrich/Merck (Darmstadt, Germany)
unless stated otherwise. The c-Jun N-terminal kinase inhibitor SP600125 was obtained from
StressMarq Biosciences (Victoria, Canada), and tanshinone IIA (TIIA) 2 was obtained from
Generon (Slough, UK). All synthetic compounds were synthesised using chemistry as outlined
in the previous section. All compounds were dissolved in dimethyl sulfoxide (DMSO).
Zebrafish husbandry. Zebrafish (Danio rerio) were raised and maintained in UK Home
Office-approved aquaria at the Bateson Centre at The University Of Sheffield, in accordance
with standard protocols and procedures [87]. Transgenic zebrafish larvae which express GFP
specifically in neutrophils, TgBAC(mpx:GFP)i114 [88], 3 days post-fertilisation (dpf), were
used for all experiments.
Zebrafish inflammation assays. All compounds were diluted from the stock solutions
and administered by immersion in aqueous solutions. DMSO (0.5% concentration) was used
as the negative control. In neutrophil recruitment experiments, SP600125 (30 μM) or tanshi-
none I (TI) 1 (25 μM) was the positive control, whilst in inflammation resolution experiments,
TIIA 2 (25 μM) was used.
Zebrafish larvae were injured by transection of the tailfin using a micro-scalpel blade, as
posterior as possible in the region where the pigment pattern was disrupted, inducing an
inflammatory response. Neutrophil counting was performed according to standard procedures
used previously [63, 64, 88]. To assess neutrophil recruitment, zebrafish larvae were injured
and treated immediately with test compounds at the stated concentrations. At 4 or 6 hours
post-injury (hpi), larvae were anaesthetised and the number of neutrophils at the site of injury
was counted. To study the resolution of neutrophilic inflammation, zebrafish larvae were
injured and allowed to recover. Larvae that mounted a good inflammatory response (around
20–25 neutrophils at the site of injury; termed ‘good responders’) were selected for compound
treatment at the stated concentrations at 4 hpi. At 8 hpi, larvae were anaesthetised and the
number of neutrophils at the site of injury was counted. All experiments were performed blind
to treatment groups.
Human neutrophil apoptosis experiments. Peripheral blood neutrophils were purified
from healthy volunteers by dextran sedimentation and Percoll gradients as previously
described [89]. Neutrophils were cultured at 37 ᵒC / 5% CO2 for 6 hours with either DMSO
(negative control) or the compounds stated, in duplicate. Cytospins were produced with a
cytocentrifuge at 300 rpm for 3 minutes for each condition. Cytospin slides were fixed with
methanol then stained with Kwik-Diff (Thermo Scientific). Using an upright light microscope
(Nikon) at 100x magnification, neutrophils were classified as either apoptotic or non-apoptotic
based on the morphology of the nuclei. 1000 Neutrophils were counted per condition (500 per
cytospin) to calculate percentage neutrophil apoptosis.
PLOS ONE | https://doi.org/10.1371/journal.pone.0240231 October 6, 2020
19 / 25

PLOS ONE
Synthesised anti-inflammatory compounds in zebrafish
Statistical analysis. All statistical analysis was carried out using Prism 7.0 (GraphPad, San
Diego, USA), version 7.00. For all experiments, at least three biological repeats were per-
formed, from which data were pooled and then analysed. In all figures, data are shown as
mean SEM of all data points from individual larvae or human neutrophils. Statistical analysis
was performed using a one-way analysis of variance (ANOVA) test with Dunnett’s multiple
comparison post-test for comparing treatment neutrophil numbers to those of the DMSO neg-
ative control.
Supporting information
S1 File. Supporting information. Chemical synthesis of compounds 8–9, 11, 13–31; details of
optimisation studies, including information on Design of Experiments (DoE) studies, and
reactions involving methoxy-substituted compounds; and copies of 1H, 13C and 19F NMR
spectra.
(DOCX)
Acknowledgments
The authors thank Dr. Anne L. Robertson for initial assistance with zebrafish techniques,
Catherine A. Loynes for advice, and the Bateson Centre aquarium staff for providing expert
zebrafish care.
Author Contributions
Conceptualization: Stephen A. Renshaw, Simon Jones.
Formal analysis: Matthew J. Foulkes.
Investigation: Matthew J. Foulkes, Faith H. Tolliday.
Methodology: Matthew J. Foulkes, Katherine M. Henry, Stephen A. Renshaw, Simon Jones.
Supervision: Katherine M. Henry, Stephen A. Renshaw, Simon Jones.
Writing – original draft: Matthew J. Foulkes.
Writing – review & editing: Stephen A. Renshaw, Simon Jones.
References
1. Hoizey D, Hoizey M-J. A History of Chinese Medicine. 2nd Ed. Edinburgh: Edinburgh University
Press, Edinburgh; 1993.
2. Chang HM, Cheng KP, Choang TF, Chow HF, Chui KY, Hon PM, et al. Structure elucidation and total
synthesis of new tanshinones isolated from Salvia miltiorrhiza Bunge (Danshen). J Org Chem. 1990;
55: 3537–3543. https://doi.org/10.1021/jo00298a029
3. Pan X, Niu G, Liu H. Microwave-assisted extraction of tanshinones from Salvia miltiorrhiza bunge with
analysis by high-performance liquid chromatography. J Chromatogr A. 2001; 922: 371–375. https://doi.
org/10.1016/s0021-9673(01)00949-9 PMID: 11486886
4. Zeng A, Wang D, Wu N, Yang R, Nan G, Bian X, et al. Extraction of Tanshinone IIA and Cryptotanshi-
none from the Rhizome of Salvia miltiorrhiza Bunge: Kinetics and Modeling. Sep Sci Technol. 2014; 49:
2330–2337. https://doi.org/10.1080/01496395.2014.922578
5. Xie T, Ma S, Lou H, Zhu R, Sun L. Two novel abietane norditerpenoids with anti-inflammatory properties
from the roots of Salvia miltiorrhiza var. alba. Tetrahedron Lett. 2014; 55: 7106–7109. https://doi.org/10.
1016/j.tetlet.2014.10.153
6. Kim SY, Moon TC, Chang HW, Son KH, Kang SS, Kim HP. Effects of tanshinone I isolated from Salvia
miltiorrhiza Bunge on arachidonic acid metabolism and in vivo inflammatory responses. Phyther Res.
2002; 16: 616–620. https://doi.org/10.1002/ptr.941 PMID: 12410540
PLOS ONE | https://doi.org/10.1371/journal.pone.0240231 October 6, 2020
20 / 25

PLOS ONE
Synthesised anti-inflammatory compounds in zebrafish
7. Kang BY, Chung SW, Kim SH, Ryu SY, Kim TS. Inhibition of interleukin-12 and interferon-γ production
in immune cells by tanshinones from Salvia miltiorrhiza. Immunopharmacology. 2000; 49: 355–361.
https://doi.org/10.1016/s0162-3109(00)00256-3 PMID: 10996033
8. Choi M-S, Cho D, Choi H-K, Im SY, Ryu S-Y, Kim K-M. Molecular mechanisms of inhibitory activities of
tanshinones on Lipopolysaccharide-lnduced nitric oxide generation in RAW 264.7 cells. Arch Pharm
Res. 2004; 27: 1233–1237. https://doi.org/10.1007/BF02975887 PMID: 15646797
9. Jeon SJ, Son KH, Kim YS, Choi YH, Kim HP. Inhibition of prostaglandin and nitric oxide production in
lipopolysaccharide-treated RAW 264.7 cells by tanshinones from the roots of Salvia miltiorrhiza bunge.
Arch Pharm Res. 2008; 31: 758–763. https://doi.org/10.1007/s12272-001-1223-4 PMID: 18563358
10. Zhao J, Lou J, Mou Y, Li P, Wu J, Zhou L. Diterpenoid Tanshinones and Phenolic Acids from Cultured
Hairy Roots of Salvia miltiorrhiza Bunge and Their Antimicrobial Activities. Molecules. 2011; 16: 2259–
2267. https://doi.org/10.3390/molecules16032259 PMID: 21383662
11. Ślusarczyk S, Zimmermann S, Kaiser M, Matkowski A, Hamburger M, Adams M. Antiplasmodial and
Antitrypanosomal Activity of Tanshinone-Type Diterpenoids from Salvia miltiorrhiza. Planta Med. 2011;
77: 1594–1596. https://doi.org/10.1055/s-0030-1270933 PMID: 21412700
12. Lu M, Wang C, Wang J. Tanshinone I induces human colorectal cancer cell apoptosis: The potential
roles of Aurora A-p53 and survivin-mediated signaling pathways. Int J Oncol. 2016; 49: 603–610.
https://doi.org/10.3892/ijo.2016.3565 PMID: 27279458
13. Liu J-J, Liu W-D, Yang H-Z, Zhang Y, Fang Z-G, Liu P-Q, et al. Inactivation of PI3k/Akt signaling path-
way and activation of caspase-3 are involved in tanshinone I-induced apoptosis in myeloid leukemia
cells in vitro. Ann Hematol. 2010; 89: 1089–1097. https://doi.org/10.1007/s00277-010-0996-z PMID:
20512574
14. Su C-C, Chen G-W, Lin J-G. Growth inhibition and apoptosis induction by tanshinone I in human colon
cancer Colo 205 cells. Int J Mol Med. 2008; 22: 613–8. https://doi.org/10.3892/ijmm_00000063 PMID:
18949381
15. Wang L, Liu A, Zhang F-L, Yeung JHK, Li X-Q, Cho C-H. Evaluation and SAR analysis of the cytotoxic-
ity of tanshinones in colon cancer cells. Chin J Nat Med. 2014; 12: 167–171. https://doi.org/10.1016/
S1875-5364(14)60028-5 PMID: 24702801
16. Park JH, Park O kyu, Cho J-H, Chen BH, Kim IH, Ahn JH, et al. Anti-inflammatory Effect of Tanshinone
I in Neuroprotection Against Cerebral Ischemia–Reperfusion Injury in the Gerbil Hippocampus. Neuro-
chem Res. 2014; 39: 1300–1312. https://doi.org/10.1007/s11064-014-1312-4 PMID: 24760430
17. Kim DH, Kim S, Jeon SJ, Son KH, Lee S, Yoon BH, et al. Tanshinone I enhances learning and memory,
and ameliorates memory impairment in mice via the extracellular signal-regulated kinase signalling
pathway. Br J Pharmacol. 2009; 158: 1131–1142. https://doi.org/10.1111/j.1476-5381.2009.00378.x
PMID: 19775283
18. Zhang K, Bao Y, Wu P, Station AE. Antioxidative Components of Tanshen (Salvia miltiorrhiza Bung). J
Agric Food Chem. 1990; 38: 1194–1197. https://doi.org/10.1021/jf00095a007
19. Chu Weng X, Gordon MH. Antioxidant Activity of Quinones Extracted from Tanshen (Salvia miltiorrhiza
Bunge). J Agric Food Chem. 1992; 40: 7331–1336. https://doi.org/10.1021/jf00020a007
20. Kim JY, Kim KM, Nan J-X, Zhao YZ, Park P-H, Lee SJ, et al. Induction of Apoptosis by Tanshinone I via
Cytochrome c Release in Activated Hepatic Stellate Cells. Pharmacol Toxicol. 2003; 92: 195–200.
https://doi.org/10.1034/j.1600-0773.2003.920410.x PMID: 12753423
21. Jing X, Xu Y, Cheng W, Guo S, Zou Y, He L. Tanshinone I induces apoptosis and pro-survival autop-
hagy in gastric cancers. Cancer Chemother Pharmacol. 2016; 77: 1171–1181. https://doi.org/10.1007/
s00280-016-3034-6 PMID: 27100736
ˆ
22. de Miranda FGG, Vilar JC, Alves IAN, Cavalcanti SC de H, Antoniolli AR. Antinociceptive and antiede-
matogenic properties and acute toxicity of Tabebuia avellanedae Lor. ex Griseb. inner bark aqueous
extract. BMC Pharmacol. 2001; 1: 6. https://doi.org/10.1186/1471-2210-1-6 PMID: 11574048
23. de Almeida ER. Preclinical and Clinical Studies of Lapachol and Beta-Lapachone. Open Nat Prod J.
2009; 2: 42–47. https://doi.org/10.2174/1874848100902010042
24. Liu S-H, Tzeng H-P, Kuo M-L, Lin-Shiau S-Y. Inhibition of inducible nitric oxide synthase by β -lapa-
chone in rat alveolar macrophages and aorta. Br J Pharmacol. 1999; 126: 746–750. https://doi.org/10.
1038/sj.bjp.0702341 PMID: 10188987
25. Moon D-O, Choi YH, Kim N-D, Park Y-M, Kim G-Y. Anti-inflammatory effects of β-lapachone in lipopoly-
saccharide-stimulated BV2 microglia. Int Immunopharmacol. 2007; 7: 506–514. https://doi.org/10.
1016/j.intimp.2006.12.006 PMID: 17321474
26. Byun SJ, Son Y, Hwan Cho B, Chung H-T, Pae H-O. β-Lapachone, a substrate of NAD(P)H:quinone
oxidoreductase, induces anti-inflammatory heme oxygenase-1 via AMP-activated protein kinase
PLOS ONE | https://doi.org/10.1371/journal.pone.0240231 October 6, 2020
21 / 25

PLOS ONE
Synthesised anti-inflammatory compounds in zebrafish
activation in RAW264.7 macrophages. J Clin Biochem Nutr. 2013; 52: 106–11. https://doi.org/10.3164/
jcbn.12-80 PMID: 23525626
ˆ
´
´
27. Sitonio MM, de Carvalho Junior CHR, Campos I de A, Silva JBNF, de Lima M do CA, Goes AJS, et al.
Anti-inflammatory and anti-arthritic activities of 3,4-dihydro-2,2-dimethyl-2H-naphthol[1,2-b]pyran-5,6-
dione (β-lapachone). Inflamm Res. 2013; 62: 107–113. https://doi.org/10.1007/s00011-012-0557-0
PMID: 23052183
28. Tzeng H-P, Ho F-M, Chao K-F, Kuo M-L, Lin-Shiau S-Y, Liu S-H. β-Lapachone Reduces Endotoxin-
induced Macrophage Activation and Lung Edema and Mortality. Am J Respir Crit Care Med. 2003; 168:
85–91. https://doi.org/10.1164/rccm.200209-1051OC PMID: 12724123
29. Lee E-J, Ko H-M, Jeong Y-H, Park E-M, Kim H-S. β-Lapachone suppresses neuroinflammation by mod-
ulating the expression of cytokines and matrix metalloproteinases in activated microglia. J Neuroinflam-
mation. 2015; 12: 133. https://doi.org/10.1186/s12974-015-0355-z PMID: 26173397
30. Pardee A, Li Y, Li C. Cancer Therapy with ß-Lapachone. Curr Cancer Drug Targets. 2002; 2: 227–242.
https://doi.org/10.2174/1568009023333854 PMID: 12188909
31. Lee JH, Cheong J, Park YM, Choi YH. Down-regulation of cyclooxygenase-2 and telomerase activity by
β-lapachone in human prostate carcinoma cells. Pharmacol Res. 2005; 51: 553–560. https://doi.org/10.
1016/j.phrs.2005.02.004 PMID: 15829436
32. Lee JI, Choi DY, Chung HS, Seo HG, Woo HJ, Choi BT, et al. Beta-lapachone induces growth inhibition
and apoptosis in bladder cancer cells by modulation of Bcl-2 family and activation of caspases. Exp
Oncol. 2006; 28: 30–5. PMID: 16614704
33. Moon D-O, Kang C-H, Kim M-O, Jeon Y-J, Lee J-D, Choi YH, et al. β-Lapachone (LAPA) Decreases
Cell Viability and Telomerase Activity in Leukemia Cells: Suppression of Telomerase Activity by LAPA.
J Med Food. 2010; 13: 481–488. https://doi.org/10.1089/jmf.2008.1219 PMID: 20438329
34. Kung H-N, Chien C-L, Chau G-Y, Don M-J, Lu K-S, Chau Y-P. Involvement of NO/cGMP signaling in
the apoptotic and anti-angiogenic effects of β-lapachone on endothelial cells in vitro. J Cell Physiol.
2007; 211: 522–532. https://doi.org/10.1002/jcp.20963 PMID: 17192848
´
35. Kalen M, Wallgard E, Asker N, Nasevicius A, Athley E, Billgren E, et al. Combination of reverse and
chemical genetic screens reveals angiogenesis inhibitors and targets. Chem Biol. 2009; 16: 432–41.
https://doi.org/10.1016/j.chembiol.2009.02.010 PMID: 19389629
36. Guiraud P, Steiman R, Campos-Takaki G-M, Seigle-Murandi F, de Buochberg M. Comparison of Anti-
bacterial and Antifungal Activities of Lapachol and β-Lapachone. Planta Med. 1994; 60: 373–374.
https://doi.org/10.1055/s-2006-959504 PMID: 7938274
37. Pereira E, Machado T, Leal I, Jesus D, Damaso C, Pinto A, et al. Tabebuia avellanedae naphthoqui-
nones: activity against methicillin-resistant staphylococcal strains, cytotoxic activity and in vivo dermal
irritability analysis. Ann Clin Microbiol Antimicrob. 2006; 5: 5. https://doi.org/10.1186/1476-0711-5-5
PMID: 16553949
38. Breger J, Fuchs BB, Aperis G, Moy TI, Ausubel FM, Mylonakis E. Antifungal Chemical Compounds
Identified Using a C. elegans Pathogenicity Assay. PLoS Pathog. 2007; 3: e18. https://doi.org/10.1371/
journal.ppat.0030018 PMID: 17274686
39. Kung H-N, Yang M-J, Chang C-F, Chau Y-P, Lu K-S. In vitro and in vivo wound healing-promoting activ-
ities of β-lapachone. Am J Physiol Physiol. 2008; 295: C931–C943. https://doi.org/10.1152/ajpcell.
00266.2008 PMID: 18650264
40. Liu H-Y, Li Q-R, Cheng X-F, Wang G-J, Hao H-P. NAMPT inhibition synergizes with NQO1-targeting
agents in inducing apoptotic cell death in non-small cell lung cancer cells. Chin J Nat Med. 2016; 14:
582–589. https://doi.org/10.1016/S1875-5364(16)30068-1 PMID: 27608947
41. Ough M, Lewis A, Bey EA, Gao J, Ritchie JM, Bornmann W, et al. Efficacy of beta-lapachone in pancre-
atic cancer treatment: Exploiting the novel, therapeutic target NQO1. Cancer Biol Ther. 2005; 4: 102–
109. https://doi.org/10.4161/cbt.4.1.1382 PMID: 15662131
42. Zhang L, Sun H, Chen Z, Liu Z, Huang N, Qian F. Intermolecular Interactions between Coencapsulated
Drugs Inhibit Drug Crystallization and Enhance Colloidal Stability of Polymeric Micelles. Mol Pharm.
2017; 14: 3568–3576. https://doi.org/10.1021/acs.molpharmaceut.7b00591 PMID: 28829143
´
43. da Silva Junior EN, de Souza MCBV, Pinto A V., Pinto M do CFR, Goulart MOF, Barros FWA, et al. Syn-
thesis and potent antitumor activity of new arylamino derivatives of nor-β-lapachone and nor-α-lapa-
chone. Bioorg Med Chem. 2007; 15: 7035–7041. https://doi.org/10.1016/j.bmc.2007.07.043 PMID:
17827021
44. Kongkathip N, Kongkathip B, Siripong P, Sangma C, Luangkamin S, Niyomdecha M, et al. Potent anti-
tumor activity of synthetic 1,2-Naphthoquinones and 1,4-Naphthoquinones. Bioorg Med Chem. 2003;
11: 3179–3191. https://doi.org/10.1016/s0968-0896(03)00226-8 PMID: 12818681
PLOS ONE | https://doi.org/10.1371/journal.pone.0240231 October 6, 2020
22 / 25

PLOS ONE
Synthesised anti-inflammatory compounds in zebrafish
45. Medzhitov R. Origin and physiological roles of inflammation. Nature. 2008; 454: 428–435. https://doi.
org/10.1038/nature07201 PMID: 18650913
46. Dahlgren C, Karlsson A. Respiratory burst in human neutrophils. J Immunol Methods. 1999; 232: 3–14.
https://doi.org/10.1016/s0022-1759(99)00146-5 PMID: 10618505
47. Brinkmann V, Reichard U, Goosmann C, Fauler B, Uhlemann Y, Weiss DS, et al. Neutrophil extracellu-
lar traps kill bacteria. Science (80-). 2004; 303: 1532–5. https://doi.org/10.1126/science.1092385
PMID: 15001782
48. Liu H, Pope RM. Phagocytes: mechanisms of inflammation and tissue destruction. Rheum Dis Clin
North Am. 2004; 30: 19–39. https://doi.org/10.1016/S0889-857X(03)00107-8 PMID: 15061567
49. Fox S, Leitch AE, Duffin R, Haslett C, Rossi AG. Neutrophil apoptosis: relevance to the innate immune
response and inflammatory disease. J Innate Immun. 2010; 2: 216–27. https://doi.org/10.1159/
000284367 PMID: 20375550
50. Lehrer RI, Ganz T, Selsted ME, Babior BM, Curnutte JT. Neutrophils and Host Defense. Ann Intern
Med. 1988; 109: 127–142. https://doi.org/10.7326/0003-4819-109-2-127 PMID: 3289431
51. Serhan CN, Savill J. Resolution of inflammation: the beginning programs the end. Nat Immunol. 2005;
6: 1191–1197. https://doi.org/10.1038/ni1276 PMID: 16369558
52. Headland SE, Norling L V. The resolution of inflammation: Principles and challenges. Semin Immunol.
2015; 27: 149–160. https://doi.org/10.1016/j.smim.2015.03.014 PMID: 25911383
53. Stockley RA. Neutrophils and the Pathogenesis of COPD. Chest. 2002; 121: 151S–155S. https://doi.
org/10.1378/chest.121.5_suppl.151s PMID: 12010844
54. Brown V, Elborn JS, Bradley J, Ennis M. Dysregulated apoptosis and NFκB expression in COPD sub-
jects. Respir Res. 2009; 10: 24. https://doi.org/10.1186/1465-9921-10-24 PMID: 19296848
55. Fahy J V., Kim KW, Liu J, Boushey HA. Prominent neutrophilic inflammation in sputum from subjects
with asthma exacerbation. J Allergy Clin Immunol. 1995; 95: 843–852. https://doi.org/10.1016/s0091-
6749(95)70128-1 PMID: 7722165
56. Jatakanon A, Uasuf C, Maziak W, Lim S, Chung KF, Barnes PJ. Neutrophilic Inflammation in Severe
Persistent Asthma. Am J Respir Crit Care Med. 1999; 160: 1532–1539. https://doi.org/10.1164/ajrccm.
160.5.9806170 PMID: 10556116
57. Pillinger MH, Abramson SB. The neutrophil in rheumatoid arthritis. Rheum Dis Clin North Am. 1995; 21:
691–714. PMID: 8619095
58. Wright HL, Moots RJ, Edwards SW. The multifactorial role of neutrophils in rheumatoid arthritis. Nat
Rev Rheumatol. 2014; 10: 593–601. https://doi.org/10.1038/nrrheum.2014.80 PMID: 24914698
59. Stewart A. Mediators and receptors in the resolution of inflammation: drug targeting opportunities. Br J
Pharmacol. 2009; 158: 933–935. https://doi.org/10.1111/j.1476-5381.2009.00484.x PMID: 19845683
60. Morjaria JB, Malerba M, Polosa R. Biologic and pharmacologic therapies in clinical development for the
inflammatory response in COPD. Drug Discov Today. 2010; 15: 396–405. https://doi.org/10.1016/j.
drudis.2010.03.001 PMID: 20223295
61. Henry KM, Loynes CA, Whyte MKB, Renshaw SA. Zebrafish as a model for the study of neutrophil biol-
ogy. J Leukoc Biol. 2013; 94: 633–642. https://doi.org/10.1189/jlb.1112594 PMID: 23463724
62. Wang X, Robertson AL, Li J, Chai RJ, Haishan W, Sadiku P, et al. Inhibitors of neutrophil recruitment
identified using transgenic zebrafish to screen a natural product library. Dis Model Mech. 2014; 7: 163–
9. https://doi.org/10.1242/dmm.012047 PMID: 24291762
63. Robertson AL, Holmes GR, Bojarczuk AN, Burgon J, Loynes CA, Chimen M, et al. A Zebrafish Com-
pound Screen Reveals Modulation of Neutrophil Reverse Migration as an Anti-Inflammatory Mecha-
nism. Sci Transl Med. 2014; 6: 225ra29. https://doi.org/10.1126/scitranslmed.3007672 PMID:
24574340
64. Robertson AL, Ogryzko N V, Henry KM, Loynes CA, Foulkes MJ, Meloni MM, et al. Identification of ben-
zopyrone as a common structural feature in compounds with anti-inflammatory activity in a zebrafish
phenotypic screen. Dis Model Mech. 2016; 9: 621–632. https://doi.org/10.1242/dmm.024935 PMID:
27079522
65. d’Alenc¸on CA, Peña OA, Wittmann C, Gallardo VE, Jones RA, Loosli F, et al. A high-throughput chemi-
cally induced inflammation assay in zebrafish. BMC Biol. 2010; 8: 151. https://doi.org/10.1186/1741-
7007-8-151 PMID: 21176202
66. Jiao M, Ding C, Zhang A. Facile construction of 3-hydroxyphenanthrene-1,4-diones: Key intermediates
to tanshinone I and its A-ring-modified analogue. Tetrahedron. 2014; 70: 2976–2981. https://doi.org/10.
1016/j.tet.2014.03.019
67. Inman M, Moody CJ. Synthesis of indolequinones from bromoquinones and enamines mediated by Cu
(OAc)2.H2O. J Org Chem. 2010; 75: 6023–6. https://doi.org/10.1021/jo101071c PMID: 20690687
PLOS ONE | https://doi.org/10.1371/journal.pone.0240231 October 6, 2020
23 / 25

PLOS ONE
Synthesised anti-inflammatory compounds in zebrafish
68. Ding C, Tu S, Yao Q, Li F, Wang Y, Hu W, et al. One-Pot Three-Step Synthesis of Naphtho[2,3-a]carba-
zole- 5,13-diones using a Tandem Radical Alkylation-Cyclization- Aromatization Reaction Sequence.
Adv Synth Catal. 2010; 352: 847–853. https://doi.org/10.1002/adsc.200900789
69. Kakisawa H, Inouye Y. Total syntheses of tanshinone-I, tanshinone-II, and cryptotanshinone. Chem
Commun. 1968; 0: 1327b–1328. https://doi.org/10.1039/c1968001327b
70. Inouye Y, Kakisawa H. Total Syntheses of Tanshinone-I, Tanshinone-II and Cryptotanshinone. Bull
Chem Soc Jpn. 1969; 42: 3318–3323. https://doi.org/10.1246/bcsj.42.3318
´
´
´
´
71. Senol FS, Ślusarczyk S, Matkowski A, Perez-Garrido A, Giron-Rodrıguez F, Ceron-Carrasco JP, et al.
Selective in vitro and in silico butyrylcholinesterase inhibitory activity of diterpenes and rosmarinic acid
isolated from Perovskia atriplicifolia Benth. and Salvia glutinosa L. Phytochemistry. 2017; 133: 33–44.
https://doi.org/10.1016/j.phytochem.2016.10.012 PMID: 27817931
72. Kakisawa H, Hayashi T, Yamazaki T. Structures of isotanshinones. Tetrahedron Lett. 1969; 10: 301–
304. https://doi.org/10.1016/S0040-4039(01)87676-0
73. Hayashi T, Inouye Y, Ohashi M, Kakisawa H, Tatematsu A, Kinoshita T. Mass spectral studies on tan-
shinones. Org Mass Spectrom. 1970; 3: 1293–1305. https://doi.org/10.1002/oms.1210031009
74. Guo S, Luo W, Liu L, Pang X, Zhu H, Liu A, et al. Isocryptotanshinone, a STAT3 inhibitor, induces apo-
ptosis and pro-death autophagy in A549 lung cancer cells. J Drug Target. 2016; 24: 934–942. https://
doi.org/10.3109/1061186X.2016.1157882 PMID: 26904961
75. Wu C-Y, Yang Y-H, Lin Y-Y, Kuan F-C, Lin Y-S, Lin W-Y, et al. Anti-cancer effect of danshen and dihy-
droisotanshinone I on prostate cancer: targeting the crosstalk between macrophages and cancer cells
via inhibition of the STAT3/CCL2 signaling pathway. Oncotarget. 2017; 8: 40246–40263. https://doi.
org/10.18632/oncotarget.14958 PMID: 28157698
76. Bennett BL, Sasaki DT, Murray BW, O’Leary EC, Sakata ST, Xu W, et al. SP600125, an anthrapyrazo-
lone inhibitor of Jun N-terminal kinase. Proc Natl Acad Sci U S A. 2001; 98: 13681–6. https://doi.org/10.
1073/pnas.251194298 PMID: 11717429
77. Han Z, Boyle DL, Chang L, Bennett B, Karin M, Yang L, et al. c-Jun N-terminal kinase is required for
metalloproteinase expression and joint destruction in inflammatory arthritis. J Clin Invest. 2001; 108:
73–81. https://doi.org/10.1172/JCI12466 PMID: 11435459
78. Kola I, Landis J. Can the pharmaceutical industry reduce attrition rates? Nat Rev Drug Discov. 2004; 3:
711–716. https://doi.org/10.1038/nrd1470 PMID: 15286737
79. Cook D, Brown D, Alexander R, March R, Morgan P, Satterthwaite G, et al. Lessons learned from the
fate of AstraZeneca’s drug pipeline: a five-dimensional framework. Nat Rev Drug Discov. 2014; 13:
419–431. https://doi.org/10.1038/nrd4309 PMID: 24833294
80. Waring MJ, Arrowsmith J, Leach AR, Leeson PD, Mandrell S, Owen RM, et al. An analysis of the attri-
tion of drug candidates from four major pharmaceutical companies. Nat Rev Drug Discov. 2015; 14:
475–486. https://doi.org/10.1038/nrd4609 PMID: 26091267
81. Foulkes MJ, Henry KM, Rougeot J, Hooper-Greenhill E, Loynes CA, Jeffrey P, et al. Expression and
regulation of drug transporters in vertebrate neutrophils. Sci Rep. 2017; 7: 4967. https://doi.org/10.
1038/s41598-017-04785-4 PMID: 28694436
82. Claessens S, Habonimana P, De Kimpe N. Synthesis of naturally occurring naphthoquinone epoxides
and application in the synthesis of β-lapachone. Org Biomol Chem. 2010; 8: 3790–3795. https://doi.org/
10.1039/c004580b PMID: 20577674
83. dos Santos FS, Dias GG, de Freitas RP, Santos LS, de Lima GF, Duarte HA, et al. Redox Center Modi-
fication of Lapachones towards the Synthesis of Nitrogen Heterocycles as Selective Fluorescent Mito-
chondrial Imaging Probes. European J Org Chem. 2017; 2017: 3763–3773. https://doi.org/10.1002/
ejoc.201700227
84. Elwood TA, Dudley KH, Tesarek JM, Rogerson PF, Bursey MM. The mass spectra of some naphthoqui-
nones. Lapachol, isolapachol and related compounds. Org Mass Spectrom. 1970; 3: 841–861. https://
doi.org/10.1002/oms.1210030703
´
´
85. Sacau EP, Estevez-Braun A, Ravelo AG, Ferro EA, Tokuda H, Mukainaka T, et al. Inhibitory effects of
lapachol derivatives on epstein-barr virus activation. Bioorg Med Chem. 2003; 11: 483–488. https://doi.
org/10.1016/s0968-0896(02)00542-4 PMID: 12538012
´
86. Tapia RA, Salas C, Morello A, Maya JD, Toro-Labbe A. Synthesis of dihydronaphthofurandiones and
dihydrofuroquinolinediones with trypanocidal activity and analysis of their stereoelectronic properties.
Bioorg Med Chem. 2004; 12: 2451–2458. https://doi.org/10.1016/j.bmc.2004.01.046 PMID: 15080940
87. Nusslein-Volhard C, Dahm R. Zebrafish: A Practical Approach. 1st ed. Oxford: Oxford University
Press; 2002.
PLOS ONE | https://doi.org/10.1371/journal.pone.0240231 October 6, 2020
24 / 25

PLOS ONE
Synthesised anti-inflammatory compounds in zebrafish
88. Renshaw SA, Loynes CA, Trushell DM, Elworthy S, Ingham PW, Whyte MK. A transgenic zebrafish
model of neutrophilic inflammation. Blood. 2006; 108: 3976–8. https://doi.org/10.1182/blood-2006-05-
024075 PMID: 16926288
89. Haslett C, Guthrie LA, Kopaniak MM, Johnston RB, Henson PM. Modulation of multiple neutrophil func-
tions by preparative methods or trace concentrations of bacterial lipopolysaccharide. Am J Pathol.
1985; 119: 101–110. PMID: 2984939
PLOS ONE | https://doi.org/10.1371/journal.pone.0240231 October 6, 2020
25 / 25