Synthesis of l-indospicine, [5,5,6-2H3]-l-indospicine and l-norindospicine

Article abstract of DOI:10.1039/c6ob01187j

Indospicine is a non-proteogenic amino acid that accumulates as the free amino acid in livestock grazing Indigofera plant species and causes both reproductive losses and hepatotoxic effects. An efficient synthetic route to l-indospicine from l-homoserine lactone is described. The methodology is applicable for the synthesis of both deuterium labelled isotopomers and structural analogues for utilisation in biological studies. The key steps are a zinc mediated Barbier reaction with acrylonitrile and a Pinner reaction that together introduce the target amidine moiety.

Full text of DOI:10.1039/c6ob01187j

Organic &
Biomolecular Chemistry
View Article Online
View Journal
PAPER
Synthesis of L-indospicine, [5,5,6-2H3]-L-indospicine
and L-norindospicine†
Cite this: DOI: 10.1039/c6ob01187j
Cheng-Shan Lang,a Siew-Hoon Wong,a Sharon Chow,a Victoria L. Challinor,a
Ken W. L. Yong,b Mary T. Fletcher,b Dionne M. Arthur,c Jack C. Ngc and
James J. De Voss*a
Indospicine is a non-proteogenic amino acid that accumulates as the free amino acid in livestock grazing
Indigofera plant species and causes both reproductive losses and hepatotoxic effects. An efficient syn-
thetic route to L-indospicine from L-homoserine lactone is described. The methodology is applicable for
the synthesis of both deuterium labelled isotopomers and structural analogues for utilisation in biological
studies. The key steps are a zinc mediated Barbier reaction with acrylonitrile and a Pinner reaction that
together introduce the target amidine moiety.
Received 1st June 2016,
Accepted 20th June 2016
DOI: 10.1039/c6ob01187j
www.rsc.org/obc
those found in rats orally dosed with indospicine3 and it was
concluded that indospicine in Indigofera spp. is responsible
Introduction
L-Indospicine [2(S)-2,7-diamino-7-iminoheptanoic acid] [pre- for the hepatotoxic effects of these plants. Although this
viously also described as L-2-amino-6-amidinohexanoic acid]1 amino acid is commonly associated with liver toxicity, the
(1) (Fig. 1) is an unusual amino acid which occurs only in embryotoxicity and teratogenicity of indospicine,4 as well as its
members of the Indigofera genus with about 700 species of ability to inhibit nitric oxide synthase5 have also been exam-
this genus distributed across tropical Africa, Asia, Australia ined. Indospicine was found to accumulate as the free amino
and North and South America. Interestingly hepatotoxicity has acid in various tissues (including muscle) of animals consum-
been reported in both primary (herbivore) consumers of Indi- ing this toxin. Horses6 were reported to have an apparent
gofera spp. that contain this non-proteogenic amino acid and higher tolerance to the hepatotoxic effects of this toxin, but
also in secondary (carnivore) consumers of meat from these the indospicine residues in their tissues was found to be poi-
herbivores. Various experimental studies have demonstrated sonous to dogs as secondary consumers of meat from horses
the toxicity of Indigofera species containing indospicine to that grazed on Indigofera spp. in central Australia.7 Similarly
cattle, sheep, goats, rabbits, mice and rats, with common signs indospicine residues in commercial dog food containing
being decreased feed intake, weight loss and liver damage.2 camel meat from camels grazing similar regions have resulted
The toxicological symptoms were found to be identical to in canine deaths.8,9
Indospicine was first isolated by Hegarty and co-workers
from Indigofera spicata and was later identified as a toxic
amino acid responsible for the observed hepatoxicity.1,10
Indospicine is structurally analogous to arginine (2), posses-
sing an amidine function rather than a guanidine, and as an
arginine antagonist competes with arginine for the active site
of arginase and competitively blocks incorporation of arginine
into protein with consequential inhibition of protein and DNA
synthesis.11–13 The effect of indospicine on humans (if any)
Fig. 1 L-Indospicine (1) and L-arginine (2).
has yet to be clarified but its toxicity in canine secondary con-
aSchool of Chemistry and Molecular Biosciences, The University of Queensland,
Brisbane 4072, Australia. E-mail: j.devoss@uq.edu.au
sumers suggests that it may represent a potential health
hazard to humans at significant concentrations. In order to
facilitate investigations into the toxicity of indospicine and its
levels and metabolic fate in a variety of plant and animal
tissues, we sought an efficient synthesis of indospicine itself,
as well as deuterated isotopomers for LC/MS quantitation
studies and structural analogues for bioactivity investigations.
bQueensland Alliance for Agriculture and Food Innovation (QAAFI), The University of
Queensland, Health and Food Sciences Precinct, Cooper Plains, QLD 4108, Australia
cNational Research Centre for Environmental and Toxicology, The University of
Queensland, Health and Food Sciences Precinct, Cooper Plains, QLD 4108, Australia
†Electronic supplementary information (ESI) available. See DOI:
10.1039/c6ob01187j
This journal is © The Royal Society of Chemistry 2016
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
The use of stable-isotope dilution analysis (SIDA) methodology availability of the later in deuterated form would permit syn-
in combination with LC-MS/MS9 improves accuracy and thesis of multiply labelled isotopomers.
precision of analysis by eliminating variability related to
chromatographic conditions,14 sample preparation15 and matrix N-cbz-L-homoserine lactone (3) in methanol to give desired
effects.8
ester 4 as an equilibrium mixture with the starting material
The synthesis began with the acid-catalysed esterification of
The total synthesis of DL-indospicine was first published by (3). Simple purification by column chromatography gave 4 in
Culvenor and co-workers in 1971.16 The hydroxyl group of pro- 69% yield along with approximately 30% of recovered starting
tected ( )-2-amino-6-hydroxyhexanoic acid was activated and material 3. Next, the hydroxyl moiety of 4 was activated by
displaced with cyanide and the key functional moiety, the transformation into the corresponding mesylate derivative (5)
amidine group, was then installed via a Pinner reaction; hydro- in 98% yield. Nucleophilic displacement with NaI then yielded
lysis provided DL-indospicine in less than 15% yield. An analo- the iodide (6) (88%). Initial investigation of tributyltin
gous route was later adopted by Bence and co-workers in mediated radical coupling of this iodide with acrylonitrile
which L-lysine was converted into enantiopure 2S-amino-6- proved unsatisfactory and we then investigated a zinc mediated
hydroxyhexanoic acid to provide L-indospicine in 8 steps and Barbier reaction. In a typical Barbier reaction, an alkyl moiety
6% overall yield.17 L-Indospicine has also been reported by is added to a carbonyl group facilitated by various types of
Feldman and Chi via a different synthetic pathway using metal such as zinc, tin or indium to produce a substituted
glutamate as the starting material.18 This synthesis involved alcohol.19 Unsaturation in one of the reactants has been
treatment of an intermediate nitrile with hydroxylamine hydro- reported to improve the yield significantly and this included
chloride to give an N-hydroxyamidine. This was selectively con- the conjugate addition of an alkyl halide to an electron
verted into an O-carbonate derivative that facilitated N–O bond deficient alkene.20 Thus, when a zinc–copper couple in water
cleavage by hydrogenation to provide the desired amidine was first activated under sonication and the iodide 6 and acrylo-
group. Removal of an activating sulphonyl moiety with Na/Hg nitrile were added successively the key intermediate nitrile
amalgam yielded L-indospicine in 9 steps and 29% yield. None (7) was formed in good yield (79%). It was further found that
of these routes appeared flexible enough to allow late stage continuous sonication increased the rate of reaction and that
incorporation of multiple isotopic labels into non-exchange- thorough deoxygenation was required to provide a good yield.
able positions and we thus sought an alternative synthetic
route.
At this stage, as 7 was the last non-ionic compound in the
synthesis, we wished to determine its optical purity by enantio-
selective chromatography. As shown in Fig. 2, enantioselective
HPLC (CHIRALPAK®IA, eluent IPA/hexane 30/70) cleanly
resolved the enantiomers of 7 and indicated its enantiomeric
excess when synthesised via the depicted route in Scheme 1
Results and discussion
The synthesis of L-indospicine reported here (Scheme 1) was >99%. In an earlier iteration of this route, homoserine
employed the Pinner reaction methodology of Culvenor16 and lactone 3 was converted to 4 under basic conditions and whilst
Bence17 to give the desired amidine group. The required nitrile the yield was superior it was found that significant racemisa-
was planned to arise from a Barbier like reaction of an tion had occurred such that 7 produced via this route was
iodo amino acid derivative with acrylonitrile, and the ready approximately 36% ee.
Scheme 1 The synthesis of L-indospicine (1).
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2016

View Article Online
Organic & Biomolecular Chemistry
Paper
Fig. 2 Enantioselective HPLC traces of (S)-7 (>99% ee, top solid line) and a mixture of enantiomers (bottom dotted line). CHIRALPAK®IA column,
IPA/hexane 30/70, flow rate 0.5 mL min−1, UV detector 254 nm, retention times: (S)-7, 23.5 min; (R)-7, 25.6 min.
Nitrile 7 (>99% ee) was then subjected to a Pinner reaction yield (48% based on unrecovered starting material) with utility
with anhydrous hydrogen chloride in methanol to give the for both in vivo and in vitro bioassays.
intermediate imidate ester. This was followed by treatment
In preliminary bioassays, the synthesised indospicine was
with anhydrous ammonia to yield the amidine 8 as a white tested for its biological activity in commercially available cryo-
hygroscopic solid which was used without further purification. preserved rat hepatocytes. The IC50 (50% inhibition of cell
Hydrolysis of 8 in 6 N HCl at room temperature for an growth) values were 915.5 112.0 µg mL−1 and 604.6 75.8
extended period removed the ester and the benzyloxy carb- µg mL−1 for exposure time of 48 and 96 h respectively. This
amate (Cbz) protecting groups without epimerisation of test confirmed the toxicity of synthesised indospicine to rat liver
the α-carbon; hydrolysis above room temperature was not cells and has led to more detailed toxicological studies of its
attempted as it has been shown in literature that the amidine potency against hepatocytes of various animals and humans.
moiety is susceptible to hydrolysis at elevated temperatures.1 These data will be published elsewhere.
Purification of the zwitterionic product then followed the route
The devised synthetic route also provides a convenient
developed by Hegarty which involves formation of the flavia- means to synthesise indospicine analogues. In order to explore
nate salt and subsequent conversion to L-indospicine mono- the utility of this route and to produce labelled material for
hydrochoride (1) via ion exchange chromatography.1 This syn- quantitation studies,9
a multiply deuterated indospicine,
thetic route provides enantiopure 1 in 6 steps and 34% overall [2H3]-1 (Scheme 2) was synthesised employing the same
Scheme 2 The synthesis of deuterated indospicine analogue ([2H3]-1).
This journal is © The Royal Society of Chemistry 2016
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
Scheme 3 The synthesis of L-norindospicine (9).
methodology but substituting acrylonitrile-d3 for acrylonitrile. yield from commercial starting materials. The key step was a
The specific incorporation of deuterium into [2H3]-7 was con- Barbier coupling of an amino acid derived alkyl iodide with
firmed by the appearance in the 13C NMR of a quintet for C5 acrylonitrile via the utilisation of a zinc–copper couple in
(δC 24.0, JCD 20.5 Hz) and a triplet for C6 (δC 16.6, JCD 19.9 Hz), aqueous media to conveniently provide the key nitrile in good
both shifted upfield by approximately 0.4 ppm per deuterium yield. This pathway was shown to be applicable to the syn-
as compared to the unlabelled compound (δC 24.8 and 17.0) thesis both of deuterated isotopomers of 1 as well as structural
and as expected for an α-deuterium effect.21 In addition, three analogues. In this study, use of acrylonitrile-d3 yielded [2H3]-1;
1
upfield protons (δH 1.7 and 2.3) were absent from the H NMR preliminary work indicated that conducting the Barbier reac-
spectrum of [2H3]-7 in comparison to unlabelled 7 and tion in [2H2]-water yielded a monodeuterated product (and
observed molecular ion in the GC/MS was m/z 304 in 7 and at thus potentially a tetradeuterated one in combination with
m/z 307 in [2H3]-7 as expected for incorporation of three deu- acrylonitrile-d3) but this has not been pursued. Substantial
terium atoms. Conversion of [2H3]-7 to [2H3]-1 proceeded (gram) amounts of enantiomerically pure L-indospicine can be
smoothly to yield material chromatographically identical to prepared via this pathway and its availability in both labelled
unlabelled 1. HRMS (ESI) analysis indicated that the deute- and unlabelled form will be valuable for the evaluation of the
rated indospicine contained 95% [2H3]-1 and 5% [2H2]-1.
This deuterated indospicine has been demonstrated to be
useful as an internal standard in LC-MS/MS stable isotope
dilution analysis (SIDA) of indospicine in complex tissue
matrixes to compensate for matrix effects and increase accu-
racy of analysis.9 In the reported analysis of residues in camel
biological activity of this toxic amino acid.
Experimental
General method
meat tissues indospicine (1) was quantitated by selected reac- All solvents used in synthetic procedures were distilled prior to
tion monitoring (SRM) transitions of m/z 174.2 → 111.0 (veri- use. Anhydrous solvents were dried according to standard pro-
fied by transition of m/z 174.2 → 157.1) as compared to cedures and distillation was conducted under vacuum or a N2
transition of m/z 177.1 → 114.0 (verified by transition of m/z atmosphere. NMR spectra were recorded on a Bruker Advance
177.1 → 113.0) for [2H3]-1 as internal standard.
300 MHz spectrometer unless otherwise stated. The experi-
Finally, in order to test the utility of this synthetic methodo- ments were performed in CDCl3 (δH 7.26, δC 77.0) and metha-
logy, an indospicine analogue, L-norindospicine (9) was also nol-d4 (δH 3.31, δC 49.0) with residual solvent peaks as internal
synthesised from commercially available N-cbz-L-serine ethyl standards. High resolution mass spectra were recorded on a
ester (10) (Scheme 3). The route was closely analogous to that Bruker micrOTOFQ spectrometer or a Thermo Scientific™ Q
used for the synthesis of 1, proceeding through the mesylate Exactive™ Hybrid Quadrupole-Orbitrap mass spectrometer
11 and the iodide 12. This iodide (12) was then coupled with (positive ion ESI mode). GC/MS analyses were performed on a
acrylonitrile in the presence of activated zinc–copper couple to Shimadzu GC/MS QP2010-plus spectrometer; ZB-5MS column
provide nitrile 13 in good yield. Subsequently, Pinner reaction – 30 m. Standard GC/MS program: column flow 2.5 mL min−1
mediated conversion provided amidine 14 which was then sub- total flow 76.9 mL min−1; injector 250 °C; detector 250 °C;
jected to hydrolysis to give the desired norindospicine mono- oven 100 °C held for 2 min, increased to 250 °C (16 °C min−1
;
)
hydrochloride salt (9) in 35% overall yield.
and held for 13.6 min; total program time 25 min. Optical
rotations were measured on a Jasco P-2000 polarimeter using a
1 mL quartz cell with a 10 cm path length. Enantioselective
HPLC analyses were performed on an Agilent 1200 Series chro-
matograph using chiral Daicel CHIRALPAK®IA (4.6 mm ×
Conclusions
In summary, an efficient synthetic pathway has been develo- 250 mm). HPLC conditions: eluent IPA/hexane 30/70, flow rate
ped to synthesise L-indospicine in 6 steps and 34% overall 0.5 mL min−1, UV detector 254 nm.
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2016

View Article Online
Organic & Biomolecular Chemistry
Paper
Methyl-L-2-[(benzyloxycarbonyl)amino]-4-hydroxybutyrate 52.5, 53.4, 67.1, 119.3, 128.1, 128.2, 128.5, 136.1, 155.8, 172.5.
(4). N-Cbz-L-homoserine lactone 3 (4.0 g, 17.0 mmol) was dis- GC/MS (EI) m/z (%) 304 (0.4, M+•), 245 (2), 201 (10), 169 (2),
solved in a mixture of 32% HCl (1 mL) in methanol (120 mL) 137 (1), 120 (1), 108 (29), 91 (100), 79 (7), 65 (12), 43 (4). HRMS
at room temperature and the resulting solution was allowed to (ESI) m/z: [M + Na]+ Calcd for C16H20N2O4Na 327.1321; Found
stir overnight. Triethylamine (1.8 mL, 12.5 mmol) was slowly 327.1328.
added to the reaction and the resulting mixture was stirred for
[5,5,6-2H3]-Methyl-L-2-[(benzyloxycarbonyl)amino]-6-cyano-
10 minutes. The solvent was removed in vacuo and the residue hexanoate ([2H3]-7). This was synthesised from 6 (240 mg,
was re-suspended in ethyl acetate (50 mL). Water (30 mL) was 0.64 mmol) using the same procedure as for unlabelled 7,
added and the aqueous layer was extracted with ethyl acetate except that acrylonitrile-d3 (1.2 mL, 98% d, Cambridge Isotope
(20 mL × 2). The combined organic layers were washed with Laboratories) was used instead of unlabelled acrylonitrile.
brine, dried over MgSO4, filtered and concentrated to give the Purification by flash column chromatography on silica gel
crude product. 1H NMR indicated that the product/starting (petroleum ether/ethyl acetate, 3 : 1) gave [2H3]-7 as a colour-
1
material ratio was approximately 2.2/1. Purification by flash less oil (136 mg, 69%). H NMR (300 MHz, CDCl3) δ 1.44–1.52
column chromatography on silica gel (petroleum ether/ethyl (2H, m), 1.70 (1H, m), 1.88 (1H, m), 2.30 (1H, br), 3.76 (3 H, s),
acetate, 1 : 1 then 100% ethyl acetate) gave 4 as a colourless oil 4.40 (1H, m), 5.11 (2H, s), 5.32 (1H, br), 7.32–7.38 (5H, m).
(3.12 g, 69%) with spectroscopic data identical to that pre- 13C NMR (75 MHz, CDCl3) δ 16.6 (t, JCD = 20.5 Hz, ε-CHD),
viously reported.22
Methyl-L-2-[(benzyloxycarbonyl)amino]-4-iodobutyrate
23.99, 24.03 (quintet, JCD = 19.9 Hz, δ-CD2), 31.9, 52.5, 53.4,
(6). 67.1, 119.3, 128.1, 128.2, 128.5, 136.1, 155.8, 172.5. GC/MS (EI)
To a solution of 5 (prepared according to literature pro- m/z (%) 307 (0.5, M+•), 248 (2), 204 (12), 172 (2), 140 (2),
cedures23 from 4 in 98% yield) (3.8 g, 11.0 mmol) in acetone 114 (2), 108 (35), 91 (100), 79 (9), 72 (6), 65 (12), 56 (5), 44 (5).
(50 mL) was added sodium iodide (3.3 g, 22.0 mmol) in one
L-Indospicine (1). To a solution of nitrile 7 (2.0 g, 6.6 mmol)
portion and the resulting mixture was heated under reflux for in dry diethyl ether (30 mL) was slowly added acetyl chloride
40 minutes. Upon completion, the reaction mixture (yellowish (11.5 mL, 161.1 mmol) at 0 °C followed by the slow addition of
slurry) was cooled to room temperature and the solvent was dry methanol (9 mL). The reaction mixture was allowed to stir
removed in vacuo. Water (30 mL) and CH2Cl2 (30 mL) were at 0–4 °C overnight. Subsequently, the reaction mixture was
added to the resulting paste and the aqueous layer was blown dry with N2 at 0 °C and dry methanol (5 mL) was added.
extracted with CH2Cl2 (30 mL × 3). The combined organic Then, a solution of anhydrous methanol (23 mL) saturated
layers were washed with 5% aqueous Na2S2O3 solution and with ammonia was added slowly at 0 °C. After addition, the
brine, dried over MgSO4, filtered and concentrated under reaction mixture was allowed to stir at 0–4 °C overnight. Upon
reduced pressure to give the crude product. Purification by completion, the reaction mixture was purged with N2 at 0 °C
flash column chromatography on silica gel (petroleum ether/ for 1 h to remove residual ammonia. The solvent was removed
ethyl acetate, 3 : 1) gave 6 as a white solid (3.67 g, 88%) with in vacuo at below 20 °C to give a viscous mixture. The desired
spectroscopic data identical to that previously reported.24
product was then taken into a solution of ethyl acetate/
Methyl-L-2-[(benzyloxycarbonyl)amino]-6-cyanohexanoate methanol (3 : 1, 60 mL) and the formed precipitate (NH4Cl)
(7). To a mixture of zinc granular (20 mesh, 0.5 g, 7.6 mmol) was removed by filtration. Evaporation of the filtrate gave the
and copper(I) iodide (0.5 g, 2.7 mmol) was added degassed crude amidine 8 as a viscous and hygroscopic solid (2.33 g,
water (1.4 mL) under Ar at room temperature and the resulting 99%). This crude product was used for the next reaction
mixture was sonicated to give a black suspension (approxi- without further purification. Aqueous HCl solution (6 N,
mately 5 min). To this suspension was added acrylonitrile 50 mL) was added to crude 8 (2.3 g, 6.5 mmol) at room temp-
(2.5 mL) and degassed acetonitrile (2.3 mL) followed by the erature and the resulting mixture was allowed to stir for 8 days.
addition of iodide (1.0 g, 1.3 mmol). This final mixture was The solvent was removed in vacuo and the residue was re-
then allowed to stir at room temperature overnight. Upon com- dissolved in Milli-Q water (10 mL). A solution of flavianic acid
pletion, saturated aqueous NH4Cl solution (5 mL) and diethyl (10.4 g) in Milli-Q water (20 mL) was added and the resulting
ether (10 mL) were added to the reaction mixture and the mixture was stirred for 15 minutes before allowing it to stand
aqueous layer was extracted with diethyl ether (10 mL × 3). The for 3 h. The yellow precipitate formed during this time was
combined organic layers were washed with brine, dried over then collected by filtration. The collected solid was washed
MgSO4, filtered and concentrated to give the crude product. with Milli-Q water excessively to afford the indospicine-flavia-
This crude product was re-dissolved in CH2Cl2 (15 mL) which nate salt as an orange solid. This indospicine-flavianate salt
was then filtered to remove an observed white precipitate, and was converted to the corresponding hydrochloride salt by stir-
the filtrate concentrated to afford an oil. Purification by flash ring it with Dowex 1 × 4 resin (chloride form, 30 mL) in Milli-Q
column chromatography on silica gel (petroleum ether/ethyl water (100 mL) at room temperature overnight. Upon com-
acetate, 3 : 1) gave the desired product as a colourless oil pletion, the resin was filtered and a small amount of fresh
1
(0.64 g, 79%). H NMR (300 MHz, CDCl3) δ 1.43–1.55 (2H, m), resin was added to the filtrate and stirred for another
1.63–1.76 (3H, m), 1.87 (1H, m), 2.33 (2H, t, J = 6.9 Hz), 3.76 30 minutes. The mixture was then filtered and the filtrate
(3H, s), 4.40 (1H, m), 5.11 (2H, s), 5.33 (1H, br), 7.29–7.38 was freeze-dried to give the desired indospicine-mono-
(5H, m). 13C NMR (75 MHz, CDCl3) δ 17.0, 24.2, 24.8, 31.9, hydrochloride salt as a white, hygroscopic solid (1.1 g, 72%).
This journal is © The Royal Society of Chemistry 2016
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
Spectroscopic data (including optical rotation) was in excellent final mixture was then sonicated for 6 h before allowing it to
agreement with that reported in literature.17 HRMS (ESI) m/z: stir at room temperature overnight. Upon completion, a satu-
[M + H]+ Calcd for C7H16N3O2 174.1237; Found 174.1236.
rated aqueous NH4Cl solution (3 mL) and diethyl ether (3 mL)
[5,5,6-2H3]-L-Indospicine ([2H3]-1). This was synthesised were added and the aqueous layer was extracted with diethyl
from [2H3]-7 (135 mg, 0.4 mmol) using the same procedure as ether (3 mL × 3). The combined organic layers were dried over
for unlabelled 1. After hydrolysis of crude [2H3]-8 (159 mg, MgSO4, filtered and concentrated under reduced pressure to
0.4 mmol), the indospicine-flavianate salt was formed, purified give the crude product. Purification by flash column chromato-
and converted to the corresponding hydrochloride salt with graphy on silica gel (petroleum ether/ethyl acetate, 2 : 1) pro-
Dowex 1 × 4 resin (chloride form, 2 mL) as described for 1 to vided 13 as a colourless oil (254 mg, 63%). 1H NMR (300 MHz,
give the desired [5,5,6-2H3]-indospicine-monohydrochloride CDCl3) δ 1.29 (3H, t, J = 7.2 Hz), 1.69–1.86 (3H, m), 2.03 (1H,
salt ([2H3]-1) as a white, hygroscopic solid (63.1 mg, 60%). m), 2.39 (2H, m), 4.22 (2H, q, J = 7.2 Hz), 4.38 (1H, m), 5.11
1H NMR (300 MHz, CD3OD) δ 1.52–1.62 (2H, m), 1.89–2.03 (2H, s), 5.38 (1H, br), 7.32–7.38 (5H, m). 13C NMR (75 MHz,
(2H, m), 2.49 (1H, br), 3.99 (1H, t, J = 6.6 Hz). 13C NMR CDCl3) δ 14.1, 16.8, 21.4, 31.9, 53.0, 61.9, 67.2, 119.0, 128.1,
(75 MHz, CD3OD) δ 25.3, 26.4 (br), 31.1, 32.8 (t, JCD = 19.7 Hz), 128.3, 128.6, 136.0, 155.9, 171.6. GC/MS (EI) m/z (%) 304
53.9, 171.8, 173.0. HRMS (ESI) m/z: [M + H]+ Calcd for (0.7, M+•), 231 (3), 187 (15), 169 (1), 123 (1), 108 (26), 91 (100),
C7H13D3N3O2 177.1425; Found 177.1429.
Ethyl-L-2-[(benzyloxycarbonyl)amino]-O-mesylpropanoate C16H20N2O4Na 327.1321; Found 327.1323.
79 (5), 65 (9), 43 (2). HRMS (ESI) m/z: [M + Na]+ Calcd for
(11). To a solution of 10 (2.3 g, 8.6 mmol) and triethylamine
L-2,6-Diamino-6-iminohexanoic acid (9). To a solution of 13
(1.5 mL, 10.3 mmol) in dry CH2Cl2 (25 mL) was added methane- (250 mg, 0.82 mmol) in dry diethyl ether (3 mL) was added
sulfonyl chloride (733 μL, 9.5 mmol) slowly at 0 °C. The reac- acetyl chloride (1.5 mL, 21 mmol) slowly at 0 °C followed by
tion mixture was gradually warmed up to room temperature dropwise addition of dry methanol (1.2 mL, 30 mmol). The
and stirred for 2 h. Upon completion, saturated aqueous reaction mixture was allowed to stir at 4 °C overnight, then
NH4Cl solution (15 mL) was added and the aqueous layer was evaporated under Ar at 0 °C and dry methanol (1.2 mL) was
extracted with CH2Cl2 (10 mL × 3). The combined organic added. Subsequently, a stream of ammonia gas was bubbled
layers were dried over MgSO4, filtered and concentrated under into the reaction mixture at 0 °C for 3 minutes before leaving
reduced pressure to give the crude product. Purification by it to stir at 4 °C for 3 h. Upon completion (as indicated by
flash column chromatography on silica gel (petroleum ether/ GC/MS analysis), the reaction mixture was blown dry with N2
ethyl acetate, 2 : 1 to 1 : 1) gave the desired product as a colour- at 0 °C. Aqueous HCl solution (1 N, 1 mL) was added and the
1
less oil (2.82 g, 95%). H NMR (300 MHz, CDCl3) δ 1.30 (3H, t, aqueous phase was washed with ethyl acetate (1 mL × 3).
J = 7.2 Hz), 2.97 (3H, s), 4.27 (2H, q, J = 7.2 Hz), 4.51–4.65 (3H, Aqueous HCl solution (6 N, 8 mL) was added to this solution
m), 5.14 (2H, s), 5.66 (1H, br), 7.32–7.38 (5H, m). 13C NMR of crude amidine 14 at room temperature and the resulting
(75 MHz, CDCl3) δ 14.1, 37.4, 53.6, 62.6, 67.4, 68.6, 128.2, mixture was allowed to stir for 8 days. The reaction mixture
128.3, 128.6, 135.9, 155.6, 168.2. HRMS (ESI) m/z: [M + Na]+ was then diluted with Milli-Q water (50 mL), washed with
Calcd for C14H19NO7SNa 368.0780; Found 368.0780.
CH2Cl2 (20 mL × 2) and the aqueous portion was freeze-dried
Ethyl-L-2-[(benzyloxycarbonyl)amino]-3-iodopropanoate (12). to give the crude product. A solution of flavianic acid (1.3 g) in
To a solution of 11 (2.6 g, 7.5 mmol) in acetone (25 mL) was Milli-Q water (2.4 mL) was added to the crude product in Milli-
added NaI (2.8 g, 18.8 mmol) in one portion at room temp- Q water (1.5 mL) and then it was stirred for 10 minutes before
erature. The resulting mixture was then heated to reflux for 1 h allowing it to stand for 1 h. A yellow precipitate formed upon
(precipitate formed). Upon completion, the solvent was standing, which was collected by filtration. The collected solid
removed under reduced pressure and the residue was dissolved was washed with Milli-Q water excessively to afford the flavia-
in water (15 mL) and CH2Cl2 (15 mL). The aqueous layer was nate derivative which was subsequently converted to the corres-
extracted with CH2Cl2 (15 mL × 3) and the combined organic ponding hydrochloride salt by stirring it with Dowex 1 × 4
layers were dried over MgSO4, filtered and concentrated resin (chloride form, 4 mL) in Milli-Q water (8 mL) overnight.
under reduced pressure to give the crude product. Purification Upon completion, the resin was filtered and a small amount
by flash column chromatography on silica gel (petroleum of fresh resin was added to the filtrate and stirred for
ether/ethyl acetate, 10 : 1 to 5 : 1) gave 12 as a white solid another 30 minutes. The mixture was filtered and the filtrate
(2.5 g, 88%) with spectroscopic data identical to that previously was freeze-dried to give the desired norindospicine-mono-
reported.25
hydrochloride salt as a white, hygroscopic solid (106 mg, 66%
Ethyl-L-2-[(benzyloxycarbonyl)amino]-5-cyanopentanoate over 2 steps). [α]2D4 +16.8 (c 1.2, 5 N HCl). H NMR (300 MHz,
(13). To a mixture of zinc (500 mg, 7.7 mmol, activated with CD3OD) δ 1.84–2.09 (4H, m), 2.58 (2H, m), 4.05 (1H, m).
2% HCl(aq)) and copper(I) iodide (515 mg, 2.7 mmol) was 13C NMR (75 MHz, CD3OD) δ 24.2, 30.8, 33.0, 53.5, 171.5,
added degassed water (1.5 mL) under Ar and the resulting 172.4. HRMS (ESI) m/z: [M + H]+ Calcd for C16H14N3O2
mixture was sonicated for 3 minutes at room temperature to 160.1081; Found 160.1078.
1
give a black suspension. To this suspension was added acrylo-
L-Indospicine (1) rat hepatocyte bioassay. Rat recoverable
nitrile (2.2 mL) and degassed acetonitrile (2.2 mL) followed by hepatic cells (Technologies Pty Ltd, Australia) were thawed and
the addition of 12 (500 mg, 1.3 mmol) in one portion. This plated according to the manufacturer’s instructions. Cells were
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2016

View Article Online
Organic & Biomolecular Chemistry
Paper
plated at a density of 0.5 × 105 cell per 100 µL of cell culture 10 M. P. Hegarty and A. W. Pound, Nature, 1968, 217, 354–
medium and were maintained at 37 °C in 5% CO2 in a 355.
humidifier incubator until 80% confluency for exposure to 11 T. Leisinger, D. Haas and M. P. Hegarty, Biochim.
indospicine at 0–2000 µg mL−1 for 48 and 96 h. Cell viability
Biophys. Acta, Nucleic Acids Protein Synth., 1972, 262, 214–
was determined by measuring total metabolic activity of cells
219.
using CellTiter 96® AQueous Non-Radioactive Cell Prolifer- 12 N. P. Madsen, G. S. Christie and M. P. Hegarty, Biochem.
ation Assay (MTS) (Promega, Wisconsin, USA) and the dye Pharmacol., 1970, 19, 853–857.
resazurin.26,27 Cell viability is expressed as a percentage of 13 N. P. Madsen and M. P. Hegarty, Biochem. Pharmacol.,
control. Dose response curves and IC50 values were constructed 1970, 19, 2391–2393.
from 4 replicates of each exposure concentration using GraphPad 14 S. Pollitt, M. P. Hegarty and M. A. Pass, Nat. Toxins, 1999, 7,
Prism software (version 5, La Jolla, CA, USA).
233–240.
15 D. R. Gardner and R.-C. Franklin, Int. J. Poisonous Plant
Res., 2011, 1, 20.
16 C. Culvenor, M. Foster and M. Hegarty, Aust. J. Chem.,
1971, 24, 371–375.
17 A. K. Bence and P. A. Crooks, Synth. Commun., 2002, 32,
2075–2082.
Acknowledgements
This work was supported by a Meat and Livestock Australia
project grant B.AHE.0042 to MLTF, JJDV and JCN.
18 P. L. Feldman, S. Chi, N. Sennequier and D. J. Stuehr,
Bioorg. Med. Chem. Lett., 1996, 6, 111–114.
19 C.-J. Li, Tetrahedron, 1996, 52, 5643–5668.
Notes and references
1 M. P. Hegarty and A. W. Pound, Aust. J. Biol. Sci., 1970, 23, 20 J.-L. Luche and L. A. Sarandeses, in Organozinc
831–842.
Reagents: A Practical Approach, ed. P. Knochel and
2 M. T. Fletcher, R. A. M. Al Jassim and A. J. Cawdell-Smith,
Agriculture, 2015, 5, 427–440.
P. Jones, Oxford University Press, USA, 1999, ch. 15,
p. 307.
3 G. S. Christie, N. P. Madsen and M. P. Hegarty, Biochem. 21 Y. K. Booth, PhD Thesis, The University of Queensland,
Pharmacol., 1969, 18, 693–700. 2007.
4 J. H. Pearn and M. P. Hegarty, Br. J. Exp. Pathol., 1970, 51, 22 M. Tsuji, R. W. Franck, G.-L. Yang and G.-W. Chen,
34–36. PCT Pat., WO2003105769A2, 2003.
5 S. D. Luzzi and M. A. Marletta, Bioorg. Med. Chem. Lett., 23 F. Liu, J. Thomas and T. R. Burke Jr., Synthesis, 2008, 2343–
2005, 15, 3934–3941. 2438.
6 S. M. Ossedryver, G. I. Baldwin, B. M. Stone, 24 G. Nisnevich, M. Gandelman and K. Kulbitski, PCT Pat.,
R. A. McKenzie, A. W. van Eps, S. Murray and
M. T. Fletcher, Aust. Vet. J., 2013, 91, 143–149.
7 M. P. Hegarty, W. R. Kelly, D. McEwan, O. J. William and
R. Cameron, Aust. Vet. J., 1988, 65, 337–340.
8 L. M. FitzGerald, M. T. Fletcher, A. E. Paul, C. S. Mansfield
and A. J. O’Hara, Aust. Vet. J., 2011, 89, 95–100.
9 E. T. T. Tan, M. T. Fletcher, K. W. L. Yong, B. R. D’Arcy and
R. Al Jassim, J. Agric. Food Chem., 2014, 62, 1974–1979.
WO2011154953(A1), 2011.
25 J. S. O’Donnell, S. P. Singh, T. A. Metcalf and A. L. Schwan,
Eur. J. Org. Chem., 2009, 547–553.
26 J. O’Brien, I. Wilson, T. Orton and F. Pognan,
Eur. J. Biochem., 2000, 267, 5421–5426.
27 M. Schriks, C. M. Vrabie, A. C. Gutleb, E. J. Faassen,
I. M. C. M. Rietjens and A. J. Murk, Toxicol. in Vitro, 2006,
20, 490–498.
This journal is © The Royal Society of Chemistry 2016
Org. Biomol. Chem.