Synthesis of [13C]-isotopomers of indole and tryptophan for use in the analysis of indole-3-acetic acid biosynthesis

Article abstract of DOI:10.1002/jlcr.850

The direct conversion of indole to indole-3-acetic acid without tryptophan as an intermediate has previously been shown to occur in vivo, as well as in vitro, with seedlings of plants. In order to facilitate the purification of the enzymes that carry out the enzymatic synthesis of indole-3-acetic acid from labeled indole, it was necessary to develop an assay that had both high sensitivity and analytical precision. To obtain the required analytical resolution and to allow definitive product identification, [13C6]indole was synthesized for use in GC-MS assays of the enzymatic conversion. Plants have been shown to be able to synthesize indole-3-acetic acid either directly from indole as well as by degradation of tryptophan. Thus, in order to allow the biochemical discrimination between these processes, the synthesized [13C6]indole was used as a starting material for a novel enzymatic synthesis of [13C]isotopomers of L-tryptophan labeled at specific positions. Together, these isotope labeled indolic compounds offer powerful new approaches to understanding and differentiating routes of indole-3-acetic acid biosynthesis in vitro and in vivo.

Full text of DOI:10.1002/jlcr.850

JOURNAL OF LABELLED COMPOUNDS AND RADIOPHARMACEUTICALS
J Label Compd Radiopharm 2004; 47: 635–646.
Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/jlcr.850
Research Article
Synthesis of [13C]-isotopomers of indole and tryptophan
for use in the analysis of indole-3-acetic acid
biosynthesis
1
Nebojsa Ilic and Jerry D. Cohen *
2,
$
!
1 Biotech Center, Cook College, Rutgers University, NewBrunswick, New Jersey 08901,
USA
2 Department of Horticultural Science, Center for Microbial and Plant Genomics,
University of Minnesota, Saint Paul, MN 55108, USA
Summary
The direct conversion of indole to indole-3-acetic acid without tryptophan as an
intermediate has previously been shown to occur in vivo, as well as in vitro, with
seedlings of plants. In order to facilitate the purification of the enzymes that carry out
the enzymatic synthesis of indole-3-acetic acid from labeled indole, it was necessary to
develop an assay that had both high sensitivity and analytical precision. To obtain the
required analytical resolution and to allow definitive product identification,
[13C6]indole was synthesized for use in GC-MS assays of the enzymatic conversion.
Plants have been shown to be able to synthesize indole-3-acetic acid either directly
from indole as well as by degradation of tryptophan. Thus, in order to allow the
biochemical discrimination between these processes, the synthesized [13C6]indole was
used as a starting material for a novel enzymatic synthesis of [13C]isotopomers of
l-tryptophan labeled at specific positions. Together, these isotope labeled indolic
compounds offer powerful new approaches to understanding and differentiating
routes of indole-3-acetic acid biosynthesis in vitro and in vivo. Copyright # 2004 John
Wiley & Sons, Ltd.
Key Words: carbon-13 indole; enzymatic synthesis; indole-3-acetic acid metabolism;
mass spectrometry standards; tryptophan isotopomers
*Correspondence to: J.D. Cohen, Department of Horticultural Science, University of Minnesota, 305
Alderman Hall, 1970 Folwell Avenue, Saint Paul, MN 55108, USA. E-mail: cohen047@tc.umn.edu
Contract/grant sponsor: US Department of Energy; contract/grant number: DE-FG02-00ER15079
Copyright # 2004 John Wiley & Sons, Ltd.
Received 23 April 2004
Accepted 27 April 2004

!
N. ILIC AND J. D. COHEN
636
Introduction
The oxidative degradation of the amino acid tryptophan (trp) has long been
considered to be the primary pathway for the production of the plant growth
hormone auxin (indole-3-acetic acid, 1AA) in plants.1 Isotope labeling
experiments with Lemna gibba, carrot embryos and trp biosynthesis mutants
in maize and Arabidopsis thaliana, however, have shown that IAA can be made
by plants via a trp-independent biosynthetic pathway.1–5 For example, the
maize mutant orange pericarp, which lacks a functional tryptophan synthase b
due to knockout mutations in both tryptophan synthase b genes, does not
make trp but does synthesize IAA de novo. Additionally, neither normal nor
mutant seedlings treated with 15N- or deuterium-labeled trp showed significant
conversion of trp into IAA. These results established the fact that IAA
biosynthesis occurs by a pathway not using trp as an intermediate in both
mutant and normal maize.5
In order to define the enzymes and intermediates involved in the trp-
independent pathway to IAA, we conducted a series of experiments6 that
demonstrated the enzymatic synthesis of IAA from indole by an in vitro
preparation from maize (Zea mays L.). These studies with light grown
seedlings of normal and orange pericarp mutant maize showed that extracts
contained the necessary enzymes and cofactors necessary to convert
[14C]indole to [14C]IAA and that tryptophan was not an intermediate in this
process.6
Our initial approaches, which involved the fractionation of the radio-
labeled intermediates and their identification, were made difficult due
to the low percentage of indole conversion to IAA relative to the form-
ation of other indolic byproducts.7 These results suggested that further
progress in enzyme and intermediate isolation would depend on the
availability of methods to clearly identify and distinguish components of
the process from the competing reactions and resulting products in plant
extracts. Thus, we have designed a robust stable isotope method for mea-
suring indole to IAA conversion in vitro that provides a highly defi-
nitive quantitative measurement of IAA production in such extracts. The
method we describe here involves the chemical synthesis of [13C6]indole for
use as a substrate, then the quantitative analysis of the [13C6]IAA formed
with in vitro reactions using [2H4]IAA as the internal standard for GC-MS
measurements.
Chemically synthesized [13C6]indole was also used together with
l-[13C3]serine in a novel enzymatic synthesis of specifically labeled
l-tryptophans. These heavy labeled l-tryptophan isotopomers are thus
suitable and easily produced for studies of alternative IAA biosynthetic
pathways as well as for tracing the metabolism of the amino acid, trp
containing peptides and other indolic signal messengers.
Copyright # 2004 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2004; 47: 635–646

SYNTHESIS OF [13C]-ISOTOPOMERS OF
637
Results and discussion
In order to develop a stable isotope method for analysis of IAA biosynthesis
from indole, several advances were necessary. First, highly enriched, multiple
[13C]-labeled indole needed to be synthesized so that indole to IAA conversion
could be measured in vitro even in crude preparations where unlabeled tryp-
tophan and indole might be present in the extract. Second, in order to measure
the quantity of isotopic labeled IAA produced, an internal standard of heavy
IAA was necessary for use for isotope dilution based analysis.8 Several stable
isotope labeled IAAs are available and [2H4]IAA, as described by Magnus
et al.,9 proved particularly suitable for the assay procedure since it is four mass
units heavier than unlabeled IAA and two mass units lighter than the expected
product from [13C6]indole. Tryptophan-independent IAA biosynthesis had
previously been studied with in vitro experiments on young maize seedlings by
the analysis of the conversion of radiolabeled [14C]indole and [14C]tryptophan
to IAA. It was shown that in extracts from light-grown maize seedlings IAA
was synthesized from indole but not from tryptophan.6 To allow similar
comparisons between the efficacy of IAA labeling by isotopic indole and
tryptophan, several specifically labeled isotopomers of l-trp were required.
The assay for trp-independent IAA biosynthesis we describe in this report is
based on the analysis of the formation of [13C6]IAA from [13C6]indole using an
isotope dilution method with [2H4]IAA as the internal standard for
quantification by GC-MS. It can be expected that with the ability to easily
distinguish IAA from other products formed in the reaction that it will be
possible to follow the specific enrichment in IAA forming activity upon further
fractionation of the proteins that catalyze these reactions.
In any study of non-traditional IAA biosynthetic routes it is also important
to have detailed knowledge of the fate of tryptophan both in vivo and in vitro.
One significant limitation to such studies has been the lack of stable isotope
labeled tryptophan containing multiple labels in specific positions. To advance
such studies, we developed a cell-free enzyme system for the production of
specifically labeled l-trp from [13C6]indole and [13C]serine. Combinations of
unlabeled and specifically labeled indole and serine provides a highly facile
route to l-trp labeled in any of several specific patterns. The labeled l-
tryptophan isotopomers produced in this way are particularly useful for
studies of the biosynthesis of IAA and should find applications in the analysis
of the routes to indolic compounds derived from tryptophan in plants as well
as other organisms.
Synthesis of [13C6]indole
Indole is an important heterocyclic molecule that forms the structural basis for
many natural products of great physiological and economic importance.
Copyright # 2004 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2004; 47: 635–646

!
N. ILIC AND J. D. COHEN
638
Synthesis of the indole ring itself has been the subject of many studies and
numerous methods for its synthesis have been reported. Most methods use, as
a starting compound, substituted benzene ring compounds subjected to a
variety of cyclization reactions to form the required indole ring. It is, however,
much more common to find methods for production of substituted indoles
rather than indole itself. Although a number of methods for synthesis of indole
are available, the synthesis of indole containing labels in specific ring positions
is not straight-forward. As with many syntheses of labeled compounds there
are several limiting factors, including the availability of appropriately labeled
starting compounds and the suitability of the selected method for small scale
synthesis, that determine the approach that can be used. The commercial
availability at a modest price of [13C6]aniline was an important factor for
selecting the synthetic approach. [13C6]Indole was synthesized in three steps
using a modification of the general method of Nordlander et al.10 with
[13C6]aniline as a starting compound (Figure 1). The reaction of [13C6]aniline
Figure 1. Synthetic steps leading from [13C6]aniline to [13C6]indole, as described
in text. The reaction of [13C6]aniline and bromoacetaldehyde diethyl acetal gave
2-anilino acetal that in an acid-catalyzed cyclization-elimination reaction formed
the indole ring
Copyright # 2004 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2004; 47: 635–646

SYNTHESIS OF [13C]-ISOTOPOMERS OF
639
Figure 2. Mass spectrum of [13C6]indole synthesized from [13C6]aniline, from
GC-MS analysis. Diagnostic ions for unlabeled indole,11 m/z 117 and 90 (M+
-
HCN), are both fully labeled to yield the corresponding ions at m/z 223 and 96.
Note the presence of the m/z 95 ion is due, in part, to the rearrangement to a
cyclic structure with the loss of a hydrogen atom11
and the a-halo aldehyde acetal (bromoacetaldehyde diethyl acetal) gave 2-
anilino acetal that in an acid-catalyzed cyclization-elimination reaction formed
the indole ring. Essential for the reaction was the trifluoroacetylation of the
amino function of 2-anilino acetal before the acid-catalyzed cyclization.
Trifluoroacetylation of 2-anilino acetal was done in situ with excess of
trifluoroacetic anhydride in trifluoroacetic acid before the cyclization reflux.
Saponification of the product in methanolic KOH gave [13C6]indole. We were
not able to achieve the higher yields reported in the original method, possibly
because of the reduced scale and differences in purity between the unlabeled
compound used for optimization and the [13C6]aniline provided by the isotope
supplier. The reaction product was characterized by GC-MS without
derivatization (Figure 2) and noting the ions at m/z 123 and 96 for [13C6]-
labeled indole as compared to molecular ion m/z 117 and ion m/z 90 for the
unlabeled compound.
Use of [13C6]indole as a substrate in maize seedlings assays for IAA biosynthesis
The supernatant from extracts of 9–10 day old maize seedlings was
incubated with [13C6]indole for 5 h at 358C, as previously described using
[14C]indole.6 IAA was purified following the enzyme reaction by ethyl acetate
partitioning, C18 HPLC, followed by methylation with diazomethane. The
conversion of the substrate to [13C6]IAA was quantified by GG-MS
monitoring four ions in the selected-ion mode. Ions at m/z 136 (quinolinium
Copyright # 2004 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2004; 47: 635–646

!
N. ILIC AND J. D. COHEN
640
Figure 3. GC-MS selected ion analysis (monitoring m/z 134, 136, 193 and 195)
of the [13C6]indole-3-acetic acid (analyzed as its methyl ester) formed from
[l3C6]indole by maize seedling extracts (upper spectrum). Selected ion spectrum
of the internal standard [2H4]indole-3-acetic acid methyl ester showing the lack
of ions at m/z 136 and 195 is also shown (lower spectrum)
ion) and m/z 195 were monitored for [13C6]IAA produced by the enzyme
activity from [13C6]indole. The internal standard [2H4]IAA was analyzed
simultaneously by monitoring the corresponding ions at m/z 134 and 193
(Figure 3). Using [13C6]indole and [2H4]IAA in this way for a selected ion
monitoring GC-MS-based assay, the measurement of conversion of very small
amounts of indole to IAA using partially purified extracts could be
accomplished with an exact quantitative isotope dilution analysis of the
product formed (Figure 3). For example, in the reaction shown in Figure 3,
only 9 ng of IAA was produced from 1 mg of [13C6]indole added to the enzyme
reaction.
Copyright # 2004 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2004; 47: 635–646

SYNTHESIS OF [13C]-ISOTOPOMERS OF
641
Synthesis of [13C]tryptophan isotopomers
Previously published methods for enzymatic tryptophan synthesis often rely
upon tryptophan synthase b, the enzyme capable of catalyzing the formation
of trp from indole and l-serine with pyridoxal phosphate as the coenzyme.12
The enzyme frequently used is that obtained from E. coli mutants that
overexpress tryptophan synthasel.13–16 Based on the now classic work by
Widholm12 on tryptophan synthase activity in higher plants and the demon-
stration of rapid conversion of indole to trp by maize liquid endosperm
preparations,7,17 a facile enzymatic synthesis was designed. The tryptophan
synthase from maize liquid endosperm was partially purified by dialysis, thus
removing contaminating, naturally occurring, unlabeled indole, serine and
tryptophan as well as cofactors for potentially competing reactions.
Chemically synthesized [13C6]indole was incubated with the dialyzed enzyme
preparation, pyridoxal phosphate and [13C3]-l-serine. The resulting isotopo-
mer with nine [13C] carbons was isolated by a procedure modified from that of
Michalczuk et al.,18 derivatized to its N-acetyl methyl ester and characterized
by the full scan GC-MS (Figure 4; characteristic ions at m/z 269 and 137).
Other isotopomers were produced, derivatized and analyzed by GC-MS
similarly. In the reaction of [13C6]indole and unlabeled l-serine, the product
was [13C6]tryptophan with characteristic ions at m/z 266 and 136. The
combination of unlabeled indole and [13C3]-l-serine produced [13C3]trypto-
phan with characteristic ions at m/z 263 and 131. The characteristic ions for
unlabeled derivatized tryptophan were at m/z 260 and 130, identical to that
previously described.18
Experimental
Synthesis of [13C6]indole
In a modification of Nordlander’s procedure,10 a solution of 0.250 g
(2.68 mmol) [13C6]aniline (Cambridge Isotopes Laboratories, Inc., Andover,
MA), 0.352 g (1.78 mmol) bromoacetaldehyde diethyl acetal (Aldrich Chemi-
cal Co. Inc., Milwaukee, WI) and 0.226 g (2.68 mmol) NaHCO3 (Sigma
Chemical Co., St. Louis, MO) in 5 ml 95% EtOH was brought to boiling
under reflux for 96 h (Figure 1). After reflux, the reaction mixture was
extracted with ether (3 Â 10 ml) and the ether layer was washed with water
(3 Â 10 ml) and the ether was dried over anhydrous Na2SO4. Rotary
evaporation of the solvent and separation of the residue on normal phase
HPLC (Phenomenex Luna 5 m Silica 250 Â 10 mm column; elution solvent of
isopropanol 25: heptane 75) gave 0.250 g of [13C6]2-anilinoacetaldehyde
diethyl acetal as a major fraction. This acetal was added with a syringe
to 3 ml of a 50/50 (v/v) mixture of trifluoroacetic anhydride/acid at 08C
under nitrogen. After 0.5 h, 2 ml of trifluoroacetic acid was added to this
Copyright # 2004 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2004; 47: 635–646

!
N. ILIC AND J. D. COHEN
642
Figure 4. GC-MS characterization of the [13C]tryptophan isotopomers. Note
the characteristic ions for the isotopomer formed from [13C6]indole and
[13C3]serine with 9 [13C] carbons (m/z 269 and 137); with 6 [13C] carbons from
[13C6]indole and unlabeled indole (m/z 266 and 136; and with 3 [13C] carbons
from unlabeled indole and [13C3]serine (m/z 263 and 131). Note that only the
methylene carbon (in the side chain derived from serine) is present in the
quinolinium ion fragment (m/z 130 in the unlabeled compound)11
Copyright # 2004 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2004; 47: 635–646

SYNTHESIS OF [13C]-ISOTOPOMERS OF
643
mixture and then refluxed for 72 h. After reflux, the reaction mixture was
evaporated to dryness and stirred overnight at room temperature in 3 ml
of 5% methanolic KOH. The methanol was evaporated and the product
was dissolved in 20 ml of water and extracted with ether (3 Â 20 ml). The
ether phase was washed with water (2 Â 15 ml) and dried over Na2SO4. The
ether was evaporated and the product was run on a short (4 Â 2.5 cm) silica
column (Kieselgel 60, 70–230 mesh, Merck, Darmstadt, Germany) with
toluene as the elution solvent. The first 50 ml of the eluent was collected, the
solvent was evaporated and the sample purified on normal phase HPLC (as
above). The fractions with [13C6]indole were collected and the solvent was
evaporated, thus yielding 42 mg of product. The product was characterized by
GC-MS and then dissolved in pentane for ease of use as a substrate in
biochemical studies.
Maize seedling extract assays for IAA biosynthesis using [13C6]indole as a
substrate
The kernels of sweet hybrid corn (Silver Queen, Meyer Seed Co., Baltimore,
MD) were imbibed in the tap water overnight, planted in vermiculite and
grown in a chamber at 258C with constant cool white flourescent light
at 25 mmol/m2/s. The homogenization buffer used with the seedlings consisted
of 75 mM Tris, 50 mM (NH2)2SO4, 1.5% PVPP, 10–20% glycerol, 1%
Zwittergent 3:10, 0.25 M ascorbic acid, l mM benzamide, 1 mM PMSF,
1 mM benzamidine, 5 mM e-amino-n-caproic acid, 1 mM PHMB, 10 mM
EGTA, 0.1 mg/ml pepstatin, 0.02 mg/ml antipain and 0.02 mg/ml leupeptin.
The final pH was adjusted to 8.5 using KOH. Ten grams of 9–10 days
old seedlings were homogenized in 50 ml of prechilled buffer at 48C.
The homogenate was filtered through Miracloth (Calbiochem, La Jolla,
CA) and centrifuged at 12 000 g for 20 min at 48C. Following centrifugation,
1 ml of supernatant was incubated with 1 mg of [13C6]indole for 5 h at
358C. The enzymatic reaction was stopped by adding 40 ml of concentrated
H3PO4. The reaction mixture was then extracted with 1 ml of EtOAc.
The sample was evaporated under a stream of nitrogen gas and then
dissolved in 100 ml of 50% MeOH together with 18.5 ng of [2H4]IAA
and the radiotracer [3H]IAA (20 000 dpm). The sample was then injected
onto the reverse phase HPLC column (Ultracarb (30), 4.6 Â 50 mm;
Phenomenex, Torrance, CA) using 27% methanol/water plus 1% acetic
acid as the mobile phase. The major radioactive fraction was methylated
using diazomethane and the methylated sample was then analyzed by
GC-SIM-MS, monitoring four ions: m/z 136 and 195 for enzymatically
synthesized IAA and m/z at 134 and 193 for the methyl ester of the internal
standard [2H4]IAA.
Copyright # 2004 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2004; 47: 635–646

!
N. ILIC AND J. D. COHEN
644
Synthesis of [13C]-labeled tryptophan
The enzymatic conversion of indole to tryptophan by maize endosperm
enzymes was done essentially as we previously reported using [14C]indole.7
Fresh liquid endosperm of sweet corn (Zea mays L. cv Silver Queen obtained
from a local grocery) was collected by cutting the rows of kernels still on the
cob with a razor blade and the kernel contents pressed out against the rim of a
beaker and further squeezed through a layer of cheese cloth. The resulting
liquid (80 ml) and 8 ml of grinding buffer (0.4 M potassium phosphate, pH 8.5,
10 mM mercaptoethanol and 40 mg/ml pyridoxal phosphates)12 were then
homogenized with a glass homogenizer and centrifuged at 10 000 g at 48C for
20 min. After centrifugation and removal of a fatty layer, 20 ml of the
supernatant was dialyzed against 3.5 l of half-strength grinding buffer, at 48C
for 16 h using dialysis membranes with a MWCO of 12–14 000 Da. Following
the dialysis, 0.8 ml of dialyzate was incubated for 1 h at 308C with 0.2 ml of a
solution of 200 mmol potassium phosphate, pH 8.5, 4.32 mg (40 mmol) l-serine
(alternatively, the same amount of [13C3]-l-serine (Cambridge Isotopes Labs)
was used), 40 mg pyridoxal phosphate and 23.4 mg (0.2 mmol) of indole
(unlabeled or [13C6]). The reaction was stopped by boiling for 5 min and
centrifuged at 10 000 g for 5 min. Following the centrifugation step,
approximately 50 000 dpm of l-[5-3H] tryptophan (33.0 Ci/mmol; Amersham
Life Science, Arlington Heights, IL) was added to the supernatant. The sample
was then diluted 3 Â with double-distilled water and applied to a 4 ml bed
volume column of Dowex 50W-X2, 200-400 mesh (Sigma), preconditioned
with 1 M HC1. The sample on the column was then washed with three bed
volumes of double-distilled water and eluted with two bed volumes of 2 M
NH4OH. The eluate was evaporated to dryness, resuspended in dichloro-
methane and evaporated. After the evaporation of the dichloromethane, l ml
of acetic anhydride (Supelco, Bellefonte, PA) and 2 ml of anhydrous methanol
were added, the evaporation flask was closed with a glass stopper and clip,
mixed for 1 min on a vortex mixer and placed in a sandbath at 658C for 1 h
in order to form the N-acetyl methyl ester of tryptophan.18 Following the
methanol-acetic anhydride derivatization procedure the sample was again
evaporated to dryness, resuspended in 100 ml of 50% methanol and purified
using an HPLC equipped with a C18 Nova-Pak 4 mm column (3.9 Â 150 mm;
Waters, Milford, MA). The mobile phase was 30% methanol/water (v/v) at a
flow rate 1 ml/min. Labeled fractions were collected, evaporated to dryness
and resuspended in 20 ml of ethyl acetate. The sample was then injected onto
the GC-MS and data was collected in the full scan mode. The variously labeled
tryptophan isotopomers from the enzymatic syntheses were characterized by
monitoring the ions at m/z 130 (quinolinium ion) and 260 (molecular ion) for
unlabeled tryptophan, and at M++3, M++6 and M++9 in heavy labeled
Copyright # 2004 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2004; 47: 635–646

SYNTHESIS OF [13C]-ISOTOPOMERS OF
645
experiments (for M++3: 131 and 263 ions; for M++6: 136 and 266 ions and
for M++9: 137 and 269 ions).
Conclusion
A facile synthesis of [13C6]indole from readily available [13C6]aniline provides a
useful new approach to measuring the metabolic pathways involved in the
synthesis of tryptophan and indole. [13C6]indole was synthesized in particular
for use in GC-MS based analysis of IAA biosynthesis in order obtain the
required analytical resolution and to facilitate definitive product identification.
Also, in order to allow the biochemical discrimination between pathways using
indole directly or via a tryptophan intermediate, the synthesized [13C6]indole
was used as the starting material for a novel enzymatic synthesis of
[13C]isotopomers of l-tryptophan labeled at specific positions. Together, this
suite of labeled indolic compounds provides critical tools for understanding
and differentiating routes of indole-3-acetic acid biosynthesis in vitro and
in vivo.
Acknowledgements
The authors thank Dr Richard J. Sundberg and Dr J. George Buta for helpful
advice and suggestions.
References
1. Normanly J, Slovin JP, Cohen JD. Plant Physiol 1995; 107: 323–329.
2. Baldi BG, Maher BR, Slovin JP, Cohen JD. Plant Physiol 1991; 95: 1203–1208.
3. Michalczuk L, Ribnicky DM, Cooke TJ, Cohen JD. Plant Physiol 1992; 100:
1346–1353.
4. Normanly J, Cohen JD, Fink GR. Proc Natl Acad Sci USA 1993; 90: 10355–
10359.
5. Wright AD, Sampson MB, Neuffer MG, Michalczuk L, Slovin JP, Cohen JD.
Science 1991; 254: 998–1000.
.
6. Ostin A, Ilic N, Cohen JD. Plant Physiol 1999; 119: 173–178.
!
.
7. Ilic N, Ostin A, Cohen JD. Plant Growth Regul 1999; 27: 57–62.
!
8. Cohen JD, Baldi BG, Slovin JP. Plant Physiol 1986; 80: 14–19.
9. Magnus V, Schulze A, Bandurski RS. Plant Physiol 1980; 66: 775–781.
10. Nordlander JE, Catalane DB, Kotian KD, Stevens RM, Haky JE. J Org Chem
1981; 46: 778–782.
11. Williams CM, Porter AH, Greer M. Mass Spectrometrv of Biologically Important
Aromatic Acids. University of Florida, Gainesville, 1969.
12. Widholm JM. Physiol Plant 1971; 25: 75–79.
13. Van den Berg EMM, Baldew AU, De Goede ATJW, Raap J, Lugtenburg J. Recl
Trav Chim Pays-Bas 1988; 107: 73–81.
14. Van den Berg EMM, Jansen FJHM, De Goede ATJW, Baldew AU, Lugtenburg
J. Recl Trav Chim Pays-Bas 1990; 109: 287–297.
Copyright # 2004 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2004; 47: 635–646

!
N. ILIC AND J. D. COHEN
646
15. Van den Berg EMM, Van Liemt WBS, Heemskerk B, Lugtenburg J. Recl Trav
Chim Pays-Bas 1989; 108: 304–313.
16. Unkefer CJ, Lodwig SN, Silks III LA, Hanners JL, Ehler DS, Gibson R. J Label
Compd Radiopharm 1991; 29: 1247–1256.
17. Rekoslavskaya NI. Fiziol Rast (Moscow) (Russ J Plant Physiol 42: 143–151)
1995; 42: 65–174.
18. Michalczuk L, Bialek K, Cohen JD. J Chromatogr 1992; 596: 294–298.
Copyright # 2004 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2004; 47: 635–646