Rapid radiosynthesis of [11C] and [14C]azelaic, suberic, and sebacic acids for in vivo mechanistic studies of systemic acquired resistance in plants

Article abstract of DOI:10.1002/jlcr.1951

A recent report that the aliphatic dicarboxylic acid, azelaic acid (1,9-nonanedioic acid) but not related acids, suberic acid (1,8-octanedioic acid) or sebacic (1,10-decanedioic acid) acid induces systemic acquired resistance to invading pathogens in plants stimulated the development of a rapid method for labeling these dicarboxylic acids with ¹¹C and ¹⁴C for in vivo mechanistic studies in whole plants. ¹¹C-labeling was performed by reaction of ammonium [ ¹¹C]cyanide with the corresponding bromonitrile precursor followed by hydrolysis with aqueous sodium hydroxide solution. Total synthesis time was 60min. Median decay-corrected radiochemical yield for [¹¹C]azelaic acid was 40% relative to trapped [¹¹C]cyanide, and specific activity was 15GBq/μmol. Yields for [¹¹C]suberic and sebacic acids were similar. The ¹⁴C-labeled version of azelaic acid was prepared from potassium [¹⁴C]cyanide in 45% overall radiochemical yield. Radiolabeling procedures were verified using ¹³C-labeling coupled with ¹³C-NMR and liquid chromatography-mass spectrometry analysis. The ¹¹C and ¹⁴C-labeled azelaic acid and related dicarboxylic acids are expected to be of value in understanding the mode-of-action, transport, and fate of this putative signaling molecule in plants.

Full text of DOI:10.1002/jlcr.1951

Research Article
Received 17 August 2011,
Revised 11 October 2011,
Accepted 31 October 2011
Published online 25 November 2011 in Wiley Online Library
(wileyonlinelibrary.com) DOI: 10.1002/jlcr.1951
Rapid radiosynthesis of [11C] and [14C]azelaic,
suberic, and sebacic acids for in vivo
mechanistic studies of systemic acquired
resistance in plants
a
c
a
Marcel Best,a,b Andrew N. Gifford, Sung Won Kim, Ben Babst,
*
Markus Piel,b Frank Rösch,b and Joanna S. Fowlera
A recent report that the aliphatic dicarboxylic acid, azelaic acid (1,9-nonanedioic acid) but not related acids, suberic acid (1,8-
octanedioic acid) or sebacic (1,10-decanedioic acid) acid induces systemic acquired resistance to invading pathogens in
plants stimulated the development of a rapid method for labeling these dicarboxylic acids with 11C and 14C for in vivo
mechanistic studies in whole plants. 11C-labeling was performed by reaction of ammonium [11C]cyanide with the corre-
sponding bromonitrile precursor followed by hydrolysis with aqueous sodium hydroxide solution. Total synthesis time
was 60 min. Median decay-corrected radiochemical yield for [11C]azelaic acid was 40% relative to trapped [11C]cyanide,
and specific activity was 15 GBq/mmol. Yields for [11C]suberic and sebacic acids were similar. The 14C-labeled version of aze-
laic acid was prepared from potassium [14C]cyanide in 45% overall radiochemical yield. Radiolabeling procedures were ver-
ified using 13C-labeling coupled with 13C-NMR and liquid chromatography–mass spectrometry analysis. The 11C and 14C-
labeled azelaic acid and related dicarboxylic acids are expected to be of value in understanding the mode-of-action, trans-
port, and fate of this putative signaling molecule in plants.
Keywords: azelaic acid; 1,9-nonanedioic acid; systemic acquired resistance; plant signaling; plant hormone
detection of b- decay using autoradiography or liquid scintilla-
Introduction
tion counting in dried plant material.
Labeling with 11C (half-life, 20.4min) requires chemistry that can
Localized foliar infections in plants subsequently result in a
higher resistance of the entire plant to secondary infections, a
be conducted rapidly and with a minimal number of postlabeling
phenomenon termed systemic acquired resistance (SAR).1,2 SAR
seems to require salicylic acid and possibly methyl salicylate.3,4
A recent study reported that azelaic acid (1,9-nonanedioic acid)
accumulates in the leaves after primary infections and also that
exogenous application of azelaic acid primes the plant to gener-
ate salicylic acid upon secondary infections in other parts of the
plant, a process involving activation of the AZI1 (AZELAIC ACID
INDUCED 1, At4g12470) gene.5 Furthermore, the same study also
reported that whereas application of azelaic acid can induce
SAR, the related dicarboxylic, suberic (1,8-octanedioic acid), and
sebacic acids (1,10-decanedioic acid) were inactive, indicating
that the induction of SAR is specific to azelaic acid over chemi-
cally similar acids differing by only one methylene group in chain
length.
steps. In prior studies involving 11C-labeling of aliphatic dicar-
6
boxylic acids, Thorell et al. employed a scheme in which the
two-carbon dicarboxylic acid, oxalic acid, was labeled in a three-
step procedure from [11C]cyanide via methyl [11C]cyanoformate.
7
De Spielgeer et al. synthesized the three-carbon dicarboxylic
acid, [11C]malonic acid, via a shorter two-step procedure from
[11C]cyanide using either chloro-acetate, bromo-acetate, or
iodo-acetate as the precursor. To our knowledge, longer chain
dicarboxylic acids have not yet been labeled with 11C.
In the current study, we developed a two-step 11C-labeling of
suberic, azelaic, and sebacic acids via reaction of ammonium
[11C]cyanide with the corresponding bromonitrile precursors
In order to investigate the transport and fate of azelaic acid in
living plants in vivo, we investigated approaches to label azelaic
acid, as well as the chemically related but inactive suberic
and sebacic acids, with either 11C or 14C. 11C-labeled version of
the acids were synthesized to enable short-term tracking of
movement of labeled acids away from the site of application
via external detection of gamma or b+ emissions in the living
plant, whereas the corresponding 14C-labeled versions of the
acids were synthesized to enable longer duration studies via
aMedical Department, Brookhaven National Laboratory, Upton, NY 11973, USA
bInstitut für Kernchemie, Johannes-Gutenberg Universitaet, 55128 Mainz,
Germany
cNational Institute on Alcohol and Alcoholism, NIH, Bethesda, MD 20892, USA
*Correspondence to: A. N. Gifford, Medical Department, Brookhaven National
Laboratory, Upton, NY 11973, USA.
E-mail: gifforda@bnl.gov
J. Label Compd. Radiopharm 2012, 55 39–43
Copyright © 2011 John Wiley & Sons, Ltd.

M. Best et al.
followed by hydrolysis with aqueous sodium hydroxide solution extracted with ethyl ether and purified by chromatography (silica
(scheme 1). 14C-labeled versions of these acids were subsequently gel, 1:5 ethyl acetate:hexane) to give 9-bromononanenitrile (6.9 g,
prepared using a similar procedure using potassium [14C]cyanide. 34%) as a pale yellow oil. 1H-NMR (400 MHz, CDCl3), d
(ppm): 1.30–1.44 (m, 8H), 1.65 (m, 2H), 1.81 (m, 2H), 2.29 (t, 2H,
J = 7.0 Hz), 3.4 (t, 2H, J = 6.6 Hz); 13 C-NMR (100 MHz, CDCl3):
d119.8, 33.8, 32.5, 29.1, 28.5, 27.9, 26.1, 25.3, 17.8).
Experimental
General
Labeling of azelaic and related acids with 11C
Commercial reagents and solvents were purchased from Sigma–
Aldrich. Organic solvents were distilled before use. Anhydrous
DMSO was stored over molecular sieves. Flash chromatography
was performed using silica gel 60 (Acros Organics). Analytical
thin layer chromatography (TLC) measurements were conducted
with 0.25 mm silica Gel 60-F254 and radioTLC with Macherey–
Nagel polygramsil G/UV254 plastic-back TLC plates. Radioactivity
11C was generated by proton bombardment of nitrogen gas
(14N(p,a)11C) with a trace of oxygen to produce [11C]CO2. For
[11C]cyanide production, the [11C]CO2 was first reduced to
[11C]CH4 via a nickel catalyzed hydrogenation at 390 ꢀC for
3 min. [11C]CH4 was then mixed with ammonia gas and passed
through a platinum furnace at 950 ꢀC to produce [11C]NH4CN.
distribution of TLC plates was determined using a Bioscan system
200 imaging scanner (Bioscan Inc., Washington DC). RadioHPLC
was performed using a HPLC system (Knauer, Berlin, Germany). 1H
and 13C-NMR spectra were measured with a Bruker spectrometer
(Bruker Biospin Corp., Billerica, MA) at 400 and 100 MHz, respec-
tively. Mass spectrometry data was obtained using an Agilent
6300 ion trap LC-MS (Agilent Technologies Inc., Santa Clara, CA).
The latter was trapped in DMSO (200 mL) to which aqueous
potassium hydroxide solution (6 N 1 mL) had been added. To this
solution, the bromonitrile precursor (1 mg, 4.7 mmol) in DMSO
(100 mL) was then added, and the mixture was heated to
140 ꢀC for 5 min. After cooling, a sample of the reaction mixture
was in some cases analyzed by radioTLC to determine the extent
of incorporation of [11C]cyanide into the dintrile. Mean incor-
poration was 90% for both the C-8 and C-9 dintrile and 83%
for the C-10 dintrile. Water (1 mL) was then added to the re-
maining reaction mixture, and the mixture passed through a
solid-phase extraction cartridge (Waters, Sep-PakW, C18 Plus).
The cartridge was washed with water and then with diethyl
ether (2 mL) to obtain the [11 C]dinitrile intermediate. Following
evaporation under the stream of argon, NaOH solution (6 M,
300 mL) was added to the [11 C]dinitrile, and the reaction mixture
was heated at 140 ꢀC for 5 min. The mixture was neutralized with
6 M HCl and purified HPLC using a Gemini seimipreparative C-18
column (250 mm  10 mm) at a flow rate of 5 mL/min. Retention
times and mobile phases were as follows: [11C]suberic acid RT
13 min (0.1% formic acid/acetonitrile, 80/20), [11C]azelaic acid
RT 12 min (0.1% formic acid/acetonitrile, 75/25), [11C]sebacic acid
RT 13 min (0.1% formic acid/acetonitrile, 70/30). Total synthesis time
was approximately 60 min relative to end-of-bombarbment (EOB).
All three carboxylic acids were analyzed by Liquid chromatography–
mass spectrometry; for suberic acid (C8H14O4 [M–H]- calcd 173.1
found 173.1), for azelaic acid (C9H16O4 [M–H]- calcd 187.1 found
187.1), and sebacic acid (C10H18O4 [M–H]- calcd 201.1 found 201.1).
Typical procedure for the preparation of 7-
bromoheptanenitrile (2a)
To a solution of sodium cyanide (6 g, 120 mmol) in water
(13.5 mL) was slowly added 1,6-dibromohexane (25 g, 100 mmol)
in isopropyl alcohol (66 mL). The reaction mixture was refluxed
for 15 h and then extracted with toluene. The toluene layer was
washed with aqueous NaOH (1 M) followed by distilled water.
The toluene was evaporated under vacuum to obtain a yellow
liquid, which was purified by chromatography (silica gel, ethyl
acetate/n-hexane, 1/4). Following evaporation of the solvent,
6.4 g (33%) of 7-bromoheptanenitrile was obtained as colorless
oil. NMR (400 MHz, CDCl3), d (ppm) = 1.3 (m, 4H), 1.7 (m, 2H),
1.79 (m, 2H), 2.4 (t, 2H), 3.2 (t, 2H). 13 C-NMR (100 MHz, CDCl3):
119.6, 33.6, 32.3, 27.8, 27.3, 25.5, 17.06.
8-bromooctanenitrile (2b)
Synthesis of 8-bromooctanenitrile (2b) was performed using 1,7-
dibromoheptane (10 g, 39 mmol) and an equimolar amount of
NaCN. After addition of 1 M NaOH (8 mL), the product was
extracted with ethyl ether and purified by chromatography
(silica gel, 1:5 ethyl acetate:hexane) to give 8-bromooctanenitrile
(2.8 g, 35%) as a pale yellow oil. 1H-NMR (400 MHz, CDCl3),
d (ppm): 1.30–1.40 (m, 6H), 1.63 (m, 2H), 1.80 (m, 2H), 2.28
(t, J = 7.0 Hz), 3.45 (t, J = 6.6 Hz). 3 C-NMR (100 MHz, CDCl3):
119.5, 37.8, 34.1, 33.7, 32.5, 28.4, 27.9, 17.9.
Identification of labeled carboxylic acids using [13C]KCN
Radiosynthesis was performed as described earlier except that
[13C]KCN (0.5 mg, 7.6 mmol) was added as a carrier to the DMSO
solvent prior to trapping [11C]NH4CN. The labeled dicarboxylic
acids were analyzed by 13C-NMR after decay of the 11C radioac-
tivity. Only one 13C signal for each carboxylic acid was observed,
with a chemical shift consistent with the expected acid.
9-bromononanenitrile (2c)
Labeling of azelaic acid with 14C
Synthesis of 9-bromononanenitrile (2c) was performed using 1,8-
dibromooctane (25 g, 90 mmol) and an equimolar amount of To a solution of [14C]KCN (390 mCi, 14 MBq) and 6 M KOH (1 mL) in
NaCN. After addition of 1 M NaOH (25 mL), the product was DMSO (400 mL), the bromonitrile precursor (2 mg, 9.4 mmol) was
Scheme 1. Synthesis of 11C-labeled dicarboxylic acids; a: n = 4, b: n = 5, c: n = 6.
www.jlcr.org
Copyright © 2011 John Wiley & Sons, Ltd.
J. Label Compd. Radiopharm 2012, 55 39–43

M. Best et al.
added, and the mixture was heated at 140o for 15 min. After
cooling, subsequent hydrolysis of the dinitrile and purification
of the dicarboxylic acid by semipreparative HPLC were per-
formed as described for the [11C] radiosynthesis. A slightly mod-
ified procedure involving extraction of the labeled dintrile into
ether rather than sep-pack purification was employed for the
other acids,but with lower yields.
Table 1. Effect of solvent conditions and temperature on
reaction of 7-bromooctanenitrile with KCN
Temperature
Entrya
Solvent
DMSO only
DMSO only
DMSO only
DMSO, NaOH (6 M, 1 mL)
DMSO, KOH (6 M, 1 mL)
DMSO, KOAc (6 M, 1 mL)
DMSO, Triethylamine
(1 mg)
(ꢀC)
Yieldb
1
2
3
4
5
6
7
90
120
140
140
140
140
140
10%
15%
20%
70%
90%
50%
40%
Autoradiography
For 11C autoradiography, a solution of the labeled acid (1–2 mCi;
37–74 MBq) in water (0.5 mL) was applied to the tip of a leaf.
Following uptake, the radioactivity was imaged by placing a
phosphor plate over the fresh leaf. The high activity of the 11C
coupled with the ease of detection of b+ particles and gamma
rays emanating from the thin leaf tissue allowed the image to
be collected using just a brief 15 min exposure to the plate, after
which it was scanned using a Fuji BAS2500 phosphoimager
(Fujifilm Corporation, Tokyo, Japan). For 14C autoradiography, a
solution of the labeled acid (1 mCi; 37 KBq) in 10 mm phosphate
buffer (50 mL, pH 7) containing 0.1% Triton X-100 was applied to
aReaction conditions: 7-bromooctanenitrile (1 mg), KCN
(1 mg), DMSO (300 mL), 5 min at indicated temperature.
bMeasured by GC-MS (area integration of total ion counts)
Because of the poor UV absorbance of the aliphatic dicar-
the tip of a leaf. Following uptake, the whole plant was placed boxylic acids and their trace concentration due to the use of
over a SaranW-wrapped phosphor screen. The plant and phosphor no-carrier-added [11C]cyanide, neither ultraviolet absorption
plate were then placed in a vacuum bag and covered with several nor refractive index detection could be used to detect the acids
layers of paper and a few grams of drying agent (DrieriteW) to dry eluting from the HPLC column. To address this problem during
the plant in situ. The bag was evacuated and sealed, and the plate the development of the labeling procedure, we spiked the
was exposed to the radioactive plant for 2–3 days. Following reaction solvent with [13C]KCN prior to trapping the [11C]NH4CN.
exposure, the phosphor plate was scanned using a Cyclone When the 11C had decayed, the formation of the 13C-dicarboxylic
phosphoimager (Packard Instrument Co. Inc., Downers Grove, IL). acids in the reaction mixture could be confirmed by 13C-NMR. A
chemical shift at 177 ppm, specific for the carboxylic group, was
clearly observed. This procedure was also used to confirm the
Results and Discussion
formation of labeled suberic and sebacic acids. A similar strategy
Bromonitriles were chosen for the cyanide displacement rather of using [13C]cyanide to monitor the [11C]cyanide reaction was
than the corresponding chloronitriles because of their higher used in the development of a novel synthesis of [11C]formalde-
reactivity. Gas chromatography–mass spectrometry analysis of hyde from [11C]methyl iodide.8
the bromonitrile precursors showed only one peak; and thus,
they could be used directly after column chromatography with- chemistry for the dicarboxylic acids. By generating a standard
out additional purification. curve from the total ion chromatogram (negative mode) and
ESI-MS analysis was used to further confirm the labeling
A series of optimization experiments with unlabeled KCN comparing it to that from the labeled product fraction, the spe-
determined that the maximum yields of the dinitrile from the cific activity of the acids could be determined. Using this
bromonitrile were obtained using DMSO at 140 ꢀC for 5 min with approach, the specific activity for [11C]azelaic acid following
added 6 M NaOH or KOH (Table 1). Yields were significantly HPLC purification was calculated as 0.4 Ci/mmol (15 GBq/mmol).
reduced by using lower temperatures or by eliminating the base. Median decay-corrected radiochemical yield for [11C]azelaic acid
Nitriles can be hydrolyzed to carboxylic acids using either relative to the amount of trapped [11C]cyanide was 40% (range
strong acids or strong bases. Initial optimization of the hydrolysis 35%–50%; n = 6). For suberic and sebacic acids, the median
conditions for conversion of aliphatic nitriles to carboxylic acids radiochemical yields relative to trapped [11C]cyanide were 30%
were conducted using 4-phenylbutyronitrile as a model compound (n = 4) and 35% (n = 1), respectively.
due to its ease of detection in the HPLC UV chromatogram. In
[14C]Azelaic acid, as well as related acids, was prepared using a
a direct comparison of hydrolysis of 4-phenylbutyronitrile similar procedure as with the 11C compounds. Because of the
using either aqueous HCl, H2SO4, KOH, or NaOH (3, 6, and lower specific activity of 14C over 11C and thus higher mass of
9 M; 5 min at 140 ꢀC), NaOH gave the highest yields of the compound eluting from the HPLC, a small UV absorbance peak
carboxylic acid. In our hands, acidic hydrolysis conditions (wavelength set at 200 nm) from the radiolabeled compound
resulted in the formation of a side product in addition to the could be observed above the background signal (Figure 1). The
free carboxylic acid. This side product was not further charac- specific activity of [14C]azelaic acid, calculated from the UV
terized but likely represented the amide. On the basis of these absorbance, was 16 mCi/mmol (0.6 GBq/mmol). [14C]-labeled
results, 6 M NaOH was selected as the preferred conditions for suberic and sebacic acids gave similar specific activities. Final
the labeled dinitrile hydrolysis. In subsequent optimization radiochemical yield for [14C]azelaic acid following purification
experiments using the [11C]dinitriles, direct one-pot hydrolysis was 45% (n = 1).
without removal of DMSO from the previous labeling step was
Autoradiographic studies were performed with the 11C and
not successful. An intermediate extraction of the [11C]dinitriles 14C-labeled azelaic acids. An example of an autoradiogram taken
using a Sep-Pak cartridge prior to hydrolysis with NaOH was on a tobacco leaf after application of [11C]azelaic acid to the tip is
thus implemented.
shown in Figure 2. The site of application is clearly visible and
J. Label Compd. Radiopharm 2012, 55 39–43
Copyright © 2011 John Wiley & Sons, Ltd.
www.jlcr.org

M. Best et al.
Figure 1. HPLC purification of [14C]azelaic acid. HPLC conditions: C-18 column
250 Â 10 mm, flow rate 5 mL/min, gradient from 10% acetonitrile, 0.1% formic acid
to 80% acetonitrile (no formic acid) over 25 min.
Figure 3. Phosphor-screen autoradiogram of a dried Arabidopsis plant at 24 h
after topical application of [14C]azelaic acid to a leaf. Arrow indicates the location
of the application leaf. The application site was removed immediately prior to
imaging to reduce ‘fogging’ of the phosphor screen around the site.
they have been administered. In the case of azelaic acid, endo-
genous levels in the plant sap range from 5 mm under
basal conditions to 30 mm following pathogen attack.5 For the
[11C]azelaic acid, specific activity was high and unlikely to
perturb the physiology of the plant. However, for the 14C-
experiments, specific activity is several orders of magnitude
lower. For the 14C autoradiography experiments, approximately
0.06 mmol of labeled compound was applied to the surface of
the leaf and 6%–10% of this was found to be exported from
the application site after 24 h. Assuming the latter distributes
evenly throughout the plant, it can be calculated that the final
average concentration of the 14C-labeled azelaic acid in a 1 g
plant will be approximately 4–6 mm.
Figure 2. Phosphor-screen autoradiogram of a fresh tobacco leaf at 50 min after
topical application of [11C]azelaic acid to the tip of the leaf.
Conclusion
some early movement of the label up the vascular tissue of the In summary, we have developed a rapid two-step radiolabeling
leaf is apparent. Figure 3 shows an example of an autoradiogram scheme for labeling the putative plant signaling molecule,
of a whole Arabidopsis plant after application of 14C-labeled aze- azelaic acid, as well as related inactive dicarboxylic, suberic,
laic acid to one of the leaves. Movement of 14C-label throughout and sebacic acids, with 11C or 14C. These radiolabeled molecules
the plant is apparent.
should facilitate further investigations into the movement and
Labeled compounds should preferably be at tracer levels to metabolism of these compounds in plants and the role of azelaic
avoid interfering with the physiology of the organism to which acid in priming plants for systemic acquired resistance.
www.jlcr.org
Copyright © 2011 John Wiley & Sons, Ltd.
J. Label Compd. Radiopharm 2012, 55 39–43

M. Best et al.
Acknowledgements
References
[1] J. A. Ryals, U. H. Neuenschwander, M. G. Willits, A. Molina, H. Y. Steiner,
M. D. Hunt, Plant Cell 1996, 8, 1809–1819.
[2] J. Shah, Curr. Opin. Plant Biol. 2009, 12, 459–464.
[3] J. Malamy, J. P. Carr, D. F. Klessig, I. Raskin, Science 1990, 250,
1002–1004.
This work was supported by the US Department of Energy, Office
of Biological and Environmental Research under contract DE-
AC02-98CH10886. Additional support was from the National
Institute on Alcohol Abuse and Alcoholism (S.W.K.). We thank
Michael Schueller for the cyclotron operations, DohyunKim and
David Alexoff for the help with the optimization of [11C]cyanide
production, Lisa Muench for the assistance with 11C-labeling,
Laura Reffert for measuring the 13C-NMR spectra, and Richard
Ferrieri for the advice on the plant studies.
[4] S. W. Park, E. Kaimoyo, D. Kumar, S. Mosher, D. F. Klessig, Science 2007,
318, 113–116.
[5] H. W. Jung, T. J. Tschaplinski, L. Wang, J. Glazebrook, J. T. Greenberg,
Science 2009, 324, 89–91.
[6] J. O. Thorell, S. Stone-Elander, N. Elander, J. Label. Compd. Radiopharm.
1993, 33, 995–1005.
[7] B. Despiegeleer, P. Goethals, G. Slegers, E. Gillis, W. Vandenbossche,
P. Demoerloose, J. Nucl. Med. 1988, 29, 1107–1113.
[8] J. M. Hooker, M. Schoenberger, H. Schieferstein, J. S. Fowler, Angew.
Chem. Int. Ed. 2008, 47, 5989–5992.
Conflict of Interest
The authors did not report any conflict of interest.
J. Label Compd. Radiopharm 2012, 55 39–43
Copyright © 2011 John Wiley & Sons, Ltd.
www.jlcr.org