Analysis of Swainsonine and Swainsonine N-Oxide as Trimethylsilyl Derivatives by Liquid Chromatography-Mass Spectrometry and Their Relative Occurrence in Plants Toxic to Livestock

Article abstract of DOI:10.1021/acs.jafc.6b02390

There are limited data concerning the occurrence of swainsonine N-oxide in plants known to contain swainsonine and its relative impact on toxicity of the plant material. A liquid chromatography-mass spectrometry method based on a solvent partitioning extraction procedure followed by trimethylsilylation and analysis using reversed phase high-pressure liquid chromatography-mass spectrometry was developed for the analysis of swainsonine and its N-oxide. The concentrations of each were measured in several swainsonine-containing taxa as well as two endophytic isolates that produce swainsonine. In vegetative samples the relative percent of N-oxide to free base ranged from 0.9 to 18%. In seed samples the N-oxide to free base ratio ranged from 0 to 10%. The measured concentrations of swainsonine N-oxide relative to swainsonine only slightly increases the actual toxicity of the various plant samples in a combined assay of both compounds.

Full text of DOI:10.1021/acs.jafc.6b02390

Article
pubs.acs.org/JAFC
Analysis of Swainsonine and Swainsonine N‑Oxide as Trimethylsilyl
Derivatives by Liquid Chromatography−Mass Spectrometry and
Their Relative Occurrence in Plants Toxic to Livestock
Dale R. Gardner* and Daniel Cook
Poisonous Plant Research Laboratory, Agricultural Research Service, U.S. Department of Agriculture, 1150 E 1400 N, Logan, Utah
84341, United States
ABSTRACT: There are limited data concerning the occurrence of swainsonine N-oxide in plants known to contain swainsonine
and its relative impact on toxicity of the plant material. A liquid chromatography−mass spectrometry method based on a solvent
partitioning extraction procedure followed by trimethylsilylation and analysis using reversed phase high-pressure liquid
chromatography−mass spectrometry was developed for the analysis of swainsonine and its N-oxide. The concentrations of each
were measured in several swainsonine-containing taxa as well as two endophytic isolates that produce swainsonine. In vegetative
samples the relative percent of N-oxide to free base ranged from 0.9 to 18%. In seed samples the N-oxide to free base ratio ranged
from 0 to 10%. The measured concentrations of swainsonine N-oxide relative to swainsonine only slightly increases the actual
toxicity of the various plant samples in a combined assay of both compounds.
KEYWORDS: swainsonine, swainsonine N-oxide, locoweeds, HPLC-MS, trimethylsilyl derivative
There are several reported methods of analysis for swainsonine
that include TLC,5 HPLC,18−21 and MS22 methods for the direct
analysis of the free base alkaloid as well as gas chromatography
methods for the analysis of the trimethylsilyl ether or the triacetyl
derivatives of swainsonine.23−25 In addition to the occurrence of
the free base swainsonine, swainsonine N-oxide, 2 (Figure 1), has
been reported to occur in several locoweed plants based on
isolation and a qualitative TLC method.4,5,26 Qualitative TLC
estimates suggested that some locoweeds had similar concen-
trations of swainsonine and its N-oxide, whereas the N-oxide was
not detected by gas chromatography in several Swainsona species
tested.23 Additionally, it has been shown that swainsonine N-
oxide was as effective as swainsonine in inhibiting α-
mannosidase.4 To better define the relative contribution of
swainsonine N-oxide to the hydroxylated indolizidine alkaloid
content in plants, a method for the simultaneous detection of
swainsonine and swainsonine N-oxide was developed and used
to determine the relative concentrations of each in a variety of
swainsonine-containing taxa.
INTRODUCTION
■
Swainsonine, 1 (Figure 1), a trihydroxy indolizidine alkaloid,1 has
been reported in species of Ipomoea2 and Turbina3 (Con-
Figure 1. Chemical structures of swainsonine, 1, and swainsonine N-
oxide, 2.
volvulaceae) and Astragalus,4 Oxytropis,5 and Swainsona1
(Fabaceae) and in Sida carpinifolia (Malvaceae).6 Ingestion of
these swainsonine-containing plants causes a toxicosis in
livestock leading to a chronic wasting disease, depression, altered
behavior, decreased libido, infertility, and death.7,8 Other species
within these genera do not contain swainsonine, and many are
important forages that pose no swainsonine-related risk to
livestock.
Recent studies provide evidence that swainsonine is not a
plant-derived secondary metabolite but is produced by fungal
endophytes associated with all swainsonine-containing taxa
investigated to date.8 Astragalus, Oxytropis, and Swainsona spp.
are associated with fungal endophytes belonging to Alternaria
spp. section Undifilum (Pleosporales),9−14 whereas Ipomoea
carnea has been shown to be associated with a fungal endophyte
belonging to order Chaetothyriales.15 Thus far, plants that lack
the association with the respective endophyte do not contain
swainsonine.15−17
MATERIALS AND METHODS
■
Materials. Swainsonine was obtained by extraction and purification
from Astragalus lentiginosus as previously described.27 Solvents used for
HPLC were purified water using a Water Pro PS, purified to 18.2 ΩM
(Labconco, Kansas City, MO, USA), and HPLC grade acetonitrile
(Sigma-Aldrich, St. Louis, MO, USA). All other solvents and reagents
were of analytical reagent grade. Ammoniated methanol used in the solid
phase extraction procedure was prepared by dilution of 50 mL of stock
ammonia-saturated methanol (stored at 4 °C) into 450 mL of additional
methanol. NMR spectroscopic data were acquired using a Bruker
Received: June 1, 2016
Revised: July 19, 2016
Accepted: July 20, 2016
This article not subject to U.S. Copyright.
Published XXXX by the American Chemical
Society
A
DOI: 10.1021/acs.jafc.6b02390
J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Journal of Agricultural and Food Chemistry
Article
Table 1. Concentrations of Swainsonine and Swainsonine N-Oxide in Vegetative Material from a Collection of Several
Swainsonine-Containing Taxa
species
variety
location
swainsonine N-oxide (%)
swainsonine (%)
Astragalus lentiginosus
araneosus
diphysus
Milford, UT, USA
St. Johns, AZ, USA
Juntura, OR, USA
Notom, UT, USA
Alpine, TX, USA
0.02
0.13
0.26
0.19
0.17
0.33
0.18
0.18
0.002
0.35
0.38
0.06
0.05
0.09
0.11
0.05
0.1
0.02
lentiginosus
wahweapensis
earlei
0.01
0.01
Astragalus mollissimus
0.02
mollissimus
Farley, NM, USA
0.007
0.008
0.0002
0.02
Kenton, OK, USA
Sanders, AZ, USA
Green River, WY, USA
Fort Davis, TX, USA
Des Moines, NM, USA
Park Valley, UT, USA
Virgindale, CO, USA
thompsonae
Astragalus pubentissimus
Astragalus wootoni
Oxytropis sericea
0.02
0.003
0.003
0.007
0.02
a
Swainsona canescens
́
Ipomoea carnea
Patos, Paraiba, Brazil
Marajo, Para, Brazil
0.0008
0.0009
0.008
0.003
0.0003
0.0006
́
́
Ipomoea riedelii
Zebele, Bahia, Brazil
0.21
0.2
Ipomoea sericophylla
Ipomoea verbascoidea
Turbina cordata
Zebele, Bahia, Brazil
Sertan
̂
ia, Pernambuco, Brazil
0.005
0.05
Juazeiro, Bahia, Brazil
a
The plant material was from greenhouse-grown plants derived from seeds of S. canescens collected in Western Australia.
Table 2. Concentrations of Swainsonine and Swainsonine N-Oxide in Seeds from a Collection of Several Swainsonine-Containing
Taxa
species
variety
location
swainsonine N-oxide (%)
swainsonine (%)
Astragalus lentiginosus
araneosus
diphysus
Milford, UT, USA
St. Johns, AZ, USA
Notom, UT, USA
Alpine, TX, USA
Farley, NM, USA
Kenton, OK, USA
Sanders, AZ, USA
Green River, WY, USA
Fort Davis, TX, USA
Des Moines, NM, USA
Park Valley, UT, USA
a
0.001
nd
0.12
0.04
0.46
0.41
0.13
0.59
nd
wahweapensis
earlei
0.001
0.0007
0.0007
0.01
Astragalus mollissimus
mollissimus
thompsonae
nd
Astragalus pubentissimus
Astragalus wootoni
Oxytropis sericea
0.002
0.0001
nd
0.04
0.11
0.009
0.04
0.02
0.07
0.42
0.05
0.05
0.0005
0.002
nd
Swainsona canescens
Ipomoea carnea
́
Patos, Paraiba, Brazil
Ipomoea riedelii
Zebele, Bahia, Brazil
0.001
0.0003
nd
Ipomoea verbascoidea
Turbina cordata
Sertan
̂
ia, Pernambuco, Brazil
Juazeiro, Bahia, Brazil
a
Seeds of S. canescens were purchased from B&T World Seeds.
Avance III HD spectrometer (500 MHz for H, 125 MHz for 13C)
(Bruker Biospin, Billerica, MA, USA) using solutions in deuterium oxide
(Sigma-Aldrich) (δH 4.70) and with acetonitrile (δC 1.47) as the
chemical shift reference.
swainsonine and its N-oxide.10,15,17 All plant and fungal materials were
sampled from known locations and populations with previously
identified voucher specimens known to contain swainsonine.
1
Preparation of Swainsonine N-Oxide. Swainsonine (10 mg;
0.058 mmol) was dissolved in absolute ethanol (1 mL) in a 7 mL screw-
cap vial. To this solution was added hydrogen peroxide (30%) (0.2 mL)
and the vial sealed with a Teflon-lined cap. After 16 h at room
temperature, the reaction was checked for completeness by flow
injection esi(+)MS (see high-resolution MS instrumentation below).
The presence of unreacted swainsonine (relative abundance = 7%) was
evident by the observation of the protonated molecule (MH+) at m/z
174, so the reaction mixture was heated to 60 °C for 4 h, after which the
absence of an ion at m/z 174 in the esi(+)MS spectrum indicated
completeness of the reaction. The reaction mixture was diluted with 1%
acetic acid in water (1 mL) and applied to a strong cation exchange
(SCX) solid phase extraction (SPE) column, Strata SCX (500 mg)
(Phenomenex, Torrance, CA, USA) conditioned as per the
Plant and Endophyte Materials. Field samples, vegetative and
floral material if available, of the following swainsonine-containing taxa
were collected: Astragalus lentiginosus var. araneosus, diphysus,
lentiginosus, and wahweapensis; Astragalus mollissimus var. earlei,
mollissimus, and thompsonae; Astragalus pubentissimus; Astragalus
wootoni; Oxytropis sericea; Swainsona canescens; Ipomoea carnea; Ipomoea
riedelii; Ipomoea sericophylla; Ipomoea verbascoidea; and Turbina cordata
(Table 1). Seeds were also collected of select taxa when available (Table
2). Plant material was dried and ground and subsequently analyzed for
swainsonine and its N-oxide. Mycelia from fungal cultures grown on
potato dextrose agar of the swainsonine-producing isolates of Alternaria
section Undifilum oxytropis cultured from O. sericea and the
Chaetothryiales isolate cultured from I. carnea were analyzed for
B
DOI: 10.1021/acs.jafc.6b02390
J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Journal of Agricultural and Food Chemistry
Article
manufacturer’s guidelines. The loaded column was rinsed with water (10
mL) and methanol (10 mL) before the swainsonine N-oxide was eluted
with ammoniated methanol (10 mL). The ammoniated methanol was
evaporated under a flow of nitrogen at 60 °C to give swainsonine N-
oxide (10.8 mg; 0.057 mmol; 98% yield): high-resolution esi(+)MS, m/
HPLC and autosampler (Agilent Instruments, Santa Clara, CA, USA)
and heated electrospray ion source. The MS instrument was calibrated
via the manufacturer’s standard procedures. Operational parameters
included the sheath gas flow (40), auxiliary gas flow (5), spray voltage
(3.5 kV), capillary temp (275 °C), and auxiliary heater temperature (300
°C). The mass spectrometer scanned a mass range of m/z 200−800.
1
z 190.10671 (C8H16NO4 requires 190.10738); H NMR (D2O, 500
MHz; reference to HDO at δ 4.70) δH 4.53 (ddd, J = 8.5, 6.8, 2.2 Hz, H-
2), 4.41 (dd, J = 6.7, 4.3 Hz, H-1), 4.28 (td, J = 10.5, 4.7 Hz, H-8), 3.74
(dd, J = 12.7, 8.2 Hz, H-3α), 3.49 (dd, J = 12.8, 2.1 Hz, H-3β), 3.32 (dd, J
= 12.1, 4.2 Hz, H-5α), 3.08 (td, J = 12.8, 3.6 Hz, H-5β), 2.94 (dd, J =
10.1, 4.2 Hz, H-8a), 2.1 (m, H-7α), 2.02 (m, H-6β), 1.77 (br d, J = 14.8
Hz, H-6α), 1.34 (qd, J = 13.1, 4.6 Hz, H-7β); 13C NMR (D2O, 125
MHz) (spectrum referenced to added acetonitrile at δ 1.47) δC 77.0
(CH, C-8a), 76.2 (CH2, C-3), 70.7 (CH, C-1), 69.2 (CH, C-2), 63.2
(CH2, C-5), 63.1 (CH, C-8), 31.0 (CH2, C-7), 19.4 (CH2, C-6),
consistent with prior literature.4
RESULTS AND DISCUSSION
■
Preparation of Swainsonine N-Oxide. Swainsonine N-
oxide was prepared by oxidation of swainsonine with hydrogen
peroxide. The complete 1H NMR data have not been previously
reported, but with regard to the similarity to the free base,
significant downfield shift of H-3, H-5, and H-8a protons was
noted.4,29
Analytical Method Development for Swainsonine and
Swainsonine N-Oxide. The only method that previously
reported the detection of swainsonine N-oxide in plant material
was a qualitative TLC method used to screen some of the
reported locoweed plant materials at the time.5 A method based
on GC-FID, using a 2 m × 2 mm glass column, 3% OV-1 on
Chromosorb W, was reportedly used for swainsonine N-oxide
detection (but failed to detect any N-oxide in any Swainsona
species);23 however, because of the highly polar nature of
swainsonine N-oxide, even derivatized, it is not expected to be a
good candidate for modern capillary GC type analysis. Thus,
despite observing the TMS derivative of swainsonine, the TMS
derivative of swainsonine N-oxide was not observed using
capillary GC conditions, J&W DB-5 MS, 30 m × 0.25 mm, and
MS detector. As for HPLC type analyses, swainsonine N-oxide is
not retained under reversed phase liquid chromatography
conditions.30 Attempts to replicate a reported hydrophilic
interaction liquid chromatography (HILIC) method for
swainsonine30 were unsuccessful. Upon the investigation of
several other HPLC columns, a 100 mm × 2 mm, 5 μm, Porous
Graphitic Carbon (Thermo Scientific) produced the best
retention and separation, but the peak shape was less than
ideal (Figure 2) for swainsonine N-oxide.
Extraction and Derivatization. Powdered dry plant material (100
mg) was treated, as previously described,28 with 2% acetic acid (5.0 mL)
and chloroform (4 mL) in a Teflon-lined screw-capped 15 mL
centrifuge tube for 16 h by mechanical rotation. The samples were
centrifuged, and the upper aqueous layer was removed to a screw-cap
vial. An aliquot (100 μL) of this aqueous acid extract was placed into an 8
mL screw-cap vial, and ammoniated methanol (1.5 mL) was added to
ensure no swainsonine salts were present; then the solvent was removed
by evaporation under a flow of nitrogen at 60 °C. To the vial was then
added dry pyridine (200 μL) and N,O-bis(trimethylsilyl)-
trifluoroacetamide (BSTFA) silylation reagent (50 μL) (Supelco,
Bellefonte, PA, USA), and the vial was capped and heated for 30 min
at 60 °C. After heating, the samples were diluted with dimethylforma-
mide (DMF) (1.0 mL), and the samples were heated again for 30 min at
60 °C. For HPLC-esi(+)MS analysis an aliquot (25 μL) of the solution
was further diluted with DMF (975 μL) in an LC autosampler vial.
Caibration Standards. A swainsonine-negative matrix solution was
prepared from alfalfa (100 mg) extracted with 2% acetic acid and
chloroform as described above. The aqueous fraction was then used to
prepare standard stock solutions of swainsonine (50 ppm) and
swainsonine N-oxide (5 ppm). An aliquot (100 μL) of each of the
standard stock solutions was mixed with ammoniated methanol (1.5
mL) in an 8 mL vial, evaporated to dryness under a flow of nitrogen at 60
°C, and derivatized as described above. An aliquot (100 μL) of the final
DMF solution was then further diluted with DMF (1.9 mL) and serially
diluted (1:1) to provide a series of standards at 200, 100, 50, 25, 12.5,
6.25, and 3.12 ng/mL for swainsonine and at 20, 10, 5, 2.5, 1.25, 0.625,
and 0.312 ng/mL for swainsonine N-oxide.
Spike and Recovery. For extraction recovery analysis, samples of
powdered, dry plant material (100 mg) were spiked with swainsonine
solution (1.0 mg/mL) (100 μL) and swainsonine N-oxide solution
(0.100 mg/mL) (100 μL). The samples (n = 4) were then extracted as
outlined above and analyzed by HPLC-esi(+)MS.
HPLC-esi(+)MS Analysis. The HPLC column used was a 50 × 2.1
mm, 2.6 μm, Kinetex EVO C18 column (Phenomenex) used in a
gradient elution mode with a mixture of acetonitrile (A) and 20 mM
ammonium acetate (B) at a flow rate of 0.400 mL/min. The linear
gradient was as follows: 50% A (0−1 min); 50−98% A (1−10 min); 98−
50% A (10−11 min); 50% A (11−16 min), with a sample injection
volume of 5 μL.
For high-resolution MS, a Thermo Scientific (Thermo Scientific, San
Jose, CA, USA) Exactive Plus Orbitrap mass spectrometer was coupled
with a Dionex Ultimate 3000 HPLC and autosampler and heated
electrospray ion source. The MS instrument was calibrated via the
manufacturer’s standard procedures. Operational parameters included
the sheath gas flow (35), auxiliary gas flow (10), spray voltage (4 kV),
capillary temp (320 °C), S-lens Rf (35), and auxiliary heater temperature
(300 °C). The mass spectrometer scanned a mass range of m/z 200−
800 at a resolution of 35,000. Peak areas were measured from
reconstructed ion chromatograms for protonated swainsonine-(TMS)3
(m/z 390.22813) and protonated swainsonine-(TMS)3 N-oxide (m/z
406.22308) with a mass range of 10 ppm.
Figure 2. HPLC-esi(+)MS chromatograms for swainsonine and
swainsonine N-oxide (nonderivatized compounds) separated on a
Hypercarb column.
Analyte derivatization is often useful to improve chromato-
graphic properties. The phenyl boronate derivative of
swainsonine and its N-oxide, conjugating with the cis-diol
groups,27 were not stable in aqueous solutions within the
chromatographic analysis time. The trimethylsilyl derivatives of
swainsonine have been previously used in GC analysis, but the
use of TMS derivatives in reversed phase HPLC is very limited
For unit mass resolution and MS/MS data a Thermo Scientific Velos
Pro LTQ mass spectrometer was coupled to a model 1260 Agilent
C
DOI: 10.1021/acs.jafc.6b02390
J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Journal of Agricultural and Food Chemistry
Article
due to the instability of the derivatives in the presence of active
hydrogens and thus aqueous environments. However, the
trimethylsilyl derivatives of swainsonine and its N-oxide
displayed good chromatographic retention and separation as
well as good esi(+)MS detector response using reversed phase
HPLC-MS column and conditions. The relatively short Kinetex
EVO C18 base column gave good separation of swainsonine and
its N-oxide tri-TMS derivatives, with a reasonably rapid elution
time of <6 min (Figure 3).
Figure 4. Measurement of swainsonine-(TMS)3 and swainsonine-
(TMS)3 N-oxide stability in four different analytical solutions:
acetonitrile (MeCN), dimethylformamide (DMF), methanol
(MeOH), and 50% acetonitrile in water (50% MeCN). Repeat
measurements were made over a 16 h period.
96, 114, 120, 143, 184, 210) were more pronounced under higher
energy collision-induced dissociation (HCD) conditions (Figure
5B).
The limit of detection (LOD) was found to be similar between
using the reconstructed ion chromatograms from the high-
resolution MS and the MS/MS ion chromatograms (data not
reported). Under the reported extraction protocol the LODs in
dry plant material were estimated to be 0.13 μg of swainsonine/
100 mg and 0.03 μg of swainsonine N-oxide/100 mg (Table 3).
The resulting analytical method was found to be repeatable
(RSD = 2.6−7.8%), and recovery was acceptable (recovery =
109−122%).
Swainsonine and Swainsonine N-Oxide in Various
Plant Samples. Swainsonine and its N-oxide were detected in
all vegetative plant material of the Fabaceae and Convolvulaceae
taxa surveyed. Concentrations of swainsonine measured ranged
from 0.002% (dw, dry weight basis) in A. mollissimus var.
thompsonae to 0.38% in A. wootoni, consistent with previous
reports for these respective taxa.30 Concentrations of swainso-
nine were consistent with reports in regard to other taxa
surveyed.8,13,30 Concentrations of swainsonine N-oxide ranged
from 0.0002 to 0.02% (dw) with a swainsonine N-oxide to
swainsonine ratio of 0.9−18% among the taxa surveyed. The
ratio of swainsonine N-oxide to swainsonine was greatest in
A. lentiginosus var. araneosus and S. canescens, whereas they were
generally smallest in the Ipomoea and Turbina spp. surveyed. In
general, taxa with greater swainsonine concentrations had greater
concentrations of swainsonine N-oxide. Swainsonine N-oxide
was detected in Astragalus and Oxytropis spp., consistent with
that previously reported;5,26 however, the semiquantitive
estimates of swainsonine and swainsonine N-oxide reported
previously differed from those reported here. Swainsonine N-
Figure 3. (A) HPLC-esi(+)HRMS reconstructed ion chromatogram for
the trimethylsilyl (TMS) derivatives of swainsonine (MH+ m/z 390)
and swainsonine N-oxide (MH+ m/z 406); (B) corresponding HRMS
spectra.
Some initial problems with the stability of the trimethylsilyl
derivatives were observed. The derivatized compounds were
stable within a single chromatographic run but significant loss of
signal was observed within 2 h if samples were prepared in
solvents containing any water or protic solvents. The aprotic
solvent dimethylformamide was used for sample dilution and the
derivatized samples were stable for at least 16 h (Figure 4) which
is sufficient for running a number of samples in an overnight
analysis.
Consequently, the quantitative analysis of the trimethylsilyl-
ated derivatives of swainsonine and its N-oxide was achieved
using reconstructed ion chromatograms, displaying the respec-
tive protonated molecules with a mass tolerance of 10 ppm,
from the high-resolution mass spectrum (Figure 3). For
comparative purposes of different MS detection modes, the
esi(+)MS/MS spectrum of swainsonine-(TMS)3, under CID
conditions, was dominated by loss of one O-silyl ether group (m/
z 300; MH+ − 90) (Figure 5A). The N-oxide likewise gave a
major ion at m/z 316 with the loss of the O-silyl ether along with
possible loss of HOCH3 at m/z 374. Smaller fragment ions (m/z
D
DOI: 10.1021/acs.jafc.6b02390
J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Journal of Agricultural and Food Chemistry
Article
Figure 5. HPLC-esi(+)MS/MS spectra of swainsonine-(TMS)3 and swainsonine-(TMS)3 N-oxide under (A) CID- and (B) HCD-induced
fragmentation.
Table 3. Measurements of Recovery and Limits of Detection (LOD) for Swainsonine and Swainsonine N-Oxide in Dry Plant
Material
a
b
analyte
swainsonine
unfortified, n = 4 (μg/100 mg)
fortified, n = 4 (μg/100 mg)
spike amount (μg)
% recovery
109
LOD, ng/mL (μg/100 mg)
105.4 5.0
4.8
214.2 5.6
2.6
100
0.05 (0.13)
RSD (%)
swainsonine N-oxide
RSD (%)
5.88 4.45
7.8
18.13 0.48
2.7
10
122
0.01 (0.03)
a
Recovery (%) = [(CF − CU)/CA] × 100, where CF = concentration of analyte measured in fortified sample, CU = concentration of analyte measure
b
in unfortified sample, and CA = concentration of analyte added. LOD = ng/mL of analyte injected on-column and then converted to concentration
(μg/100 mg) in original sample material given the dilution factors of the extraction procedure.
oxide was detected in S. canescens, in contrast to a previous study
in which no swainsonine N-oxide was detected in S. canescens.23
We suspect that differences between the data reported here and
in previous studies is likely due to improved analytical
instrumentation.
Swainsonine was detected in all seed samples (0.009−0.59%,
dw) with the exception of A. mollissimus var. thompsonae sample,
whereas swainsonine N-oxide ranged from “not detected” to
0.002%. The ratio of swainsonine N-oxide to swainsonine was
0.09−0.1% among the taxa with detectable amounts. In general,
E
DOI: 10.1021/acs.jafc.6b02390
J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Journal of Agricultural and Food Chemistry
Article
lysosomal storage disease in goats caused by the ingestion of Turbina
cordata in northeastern Brazil. Toxicon 2007, 49, 111−116.
(4) Molyneux, R. J.; James, L. F. Loco intoxication: indolizidine
alkaloids of spotted locoweed (Astragalus lentiginosus). Science 1982,
216, 190−191.
(5) Molyneux, R. J.; James, L. F.; Panter, K. E.; Ralphs, M. R. Analysis
and distribution of swainsonine and related polyhydroxyindolizidine
alkaloids by thin layer chromatography. Phytochem. Anal. 1991, 2, 125−
129.
(6) Colodel, E. M.; Gardner, D. R.; Zlotowski, P.; Driemeier, D.
Identification of swainsonine as a glycoside inhibitor responsible for Sida
carpinifolia poisoning. Vet. Hum. Toxicol. 2002, 44, 177−178.
(7) Panter, K. E.; James, L. F.; Stegelmeier, B. L.; Ralphs, M. H.; Pfister,
J. A. Locoweeds: effects on reproduction in livestock. J. Nat. Toxins
1999, 8, 53−62.
(8) Cook, D.; Gardner, D. R.; Pfister, J. A. Swainsonine-containing
plants and their relationship to endophytic fungi. J. Agric. Food Chem.
2014, 62, 7326−7334.
(9) Braun, K.; Romero, J.; Liddell, C.; Creamer, R. Production of
swainsonine by fungal endophytes of locoweed. Mycol. Res. 2003, 107,
980−988.
(10) Pryor, B. M.; Creamer, R.; Shoemaker, R. A.; McClain-Romero, J.;
Hambleton, S. Undifilum, a new genus for endophytic Embellisia
oxytropis and parasitic Helminthosporium bornmuelleri on legumes.
Botany 2009, 87, 178−194.
(11) Yu, Y.; Zhao, Q.; Wang, J.; Wang, J.; Wang, Y.; Song, Y.; Geng, G.;
Li, Q. Swainsonine-producing fungal endophyte from major locoweed
species in China. Toxicon 2010, 56, 330−338.
(12) Baucom, D.; Romero, M.; Belfon, R.; Creamer, R. Two new
species of Undifilum, the swainsonine producing fungal endophyte from
Astragalus species of locoweed in the United States. Botany 2012, 90,
866−875.
(13) Grum, D. S.; Cook, D.; Baucom, D.; Mott, I. W.; Gardner, D. R.;
Creamer, R.; Allen, J. G. Production of the alkaloid swainsonine by a
fungal endophyte in the host Swainsona canescens. J. Nat. Prod. 2013, 76,
1984−1988.
(14) Woudenberg, J. H. C.; Groenewald, J. Z.; Binder, M.; Crous, P. W.
Alternaria redefined. Stud. Mycol. 2013, 75, 171−212.
(15) Cook, D.; Beaulieu, W. T.; Mott, I. W.; Riet-Correa, F.; Gardner,
D. R.; Grum, D.; Pfister, J. A.; Clay, K.; Marcolongo-Pereira, C.
Production of the alkaloid swainsonine by a fungal endosymbiont of the
Ascomycete order Chaetothyriales in the host Ipomoea carnea. J. Agric.
Food Chem. 2013, 61, 3797−3803.
(16) Oldrup, E.; McLain-Romero, J.; Padilla, A.; Moya, A.; Gardner, D.
R.; Creamer, R. Localization of endophytic Undifilum fungi in locoweed
seed and influence of environmental parameters on a locoweed in vitro
culture system. Botany 2010, 88, 512−521.
(17) Grum, D. S.; Cook, D.; Gardner, D. R.; Roper, J. M.; Pfister, J. A.;
Ralphs, M. H. Influence of seed endophyte amounts on swainsonine
concentrations in Astragalus and Oxytropis locoweeds. J. Agric. Food
Chem. 2012, 60, 8083−8089.
(18) Yang, G.-D.; Gao, R.; Wang, Y.; Li, J.-C.; Hu, Y.-C.; Kang, D.-J; Li,
Y.-H.; Li, H.-L.; Geng, G.-X.; Wang, J.-H. Determination of swainsonine
in the endophytic Undifilum fungi by high-performance liquid
chromatography with evaporative light-scattering detector. Toxicon
2012, 60, 44−49.
(19) Li, D.-L.; Dong, C.-Y.; Gao, Y. A simple and sensitive HPLC-UV
method for the determiniation of swainsonine in rat plasma and its
application in a pharmacokinetic study. Arzneim. Forsch. 2012, 62, 209−
212.
(20) Li, J.; Yang, G.; Kang, D.; Li, Y.; Wang, Y.; Geng, G.; Wang, J.
Optimization and validation of HPLC-ELSD method for determination
of swainsonine in Chinese locoweeds. Asian J. Chem. 2013, 25, 9653−
9657.
there was less swainsonine N-oxide detected in seed samples,
resulting in a lower ratio of swainsonine N-oxide to the free base
swainsonine compared to the samples of vegetative material from
the same taxon.
Swainsonine and its N-oxide were detected in Undifilum
oxytropis cultured from O. sericea and the Chaetothryiales isolate
cultured from I. carnea in ratios similar to those found in plant
material of the respective host taxa (data not shown). These data
would suggest that swainsonine and its N-oxide are produced by
the endophyte associated with the respective plant hosts and that
the amounts found in planta are reflective of the concentrations
produced by the respective endophytes. Endophytes are
associated with seed material of each respective species and are
the means of transmission to new plant material. We have no
known explanation as to the apparent lower ratios of swainsonine
N-oxide to swainsonine that were measured in seed material
versus plant material other than seed samples in some cases were
not necessarily sampled directly from their associated vegetative
material (i.e., seed and leaf samples from the same plant) or
further metabolism and/or possible selective transportation of
swainsonine to the reproductive parts of the plant.
In summary, a method was developed to detect swainsonine
and its N-oxide. The relative concentrations of each were
determined in several swainsonine-containing taxa as well as two
endophytic isolates that produce swainsonine. All swainsonine-
containing taxa analyzed in the Fabaceae and Convolvulaceae
contained swainsonine N-oxide. Among the samples analyzed,
the concentrations of swainsonine N-oxide relative to
swainsonine were sufficiently low that the addition of the N-
oxide compound would only slightly increase the overall effective
toxicity of the various plant samples. Therefore, the continued
practice of plant sample analyses in which only swainsonine is the
target analyte is sufficient to determine potential toxicity of
plants.
AUTHOR INFORMATION
Corresponding Author
*(D.R.G.) Phone: (435) 752-2941. Fax: (435) 797-5681. E-mail:
dale.gardner@ars.usda.gov.
■
Funding
We thank Utah State University and the National Science
Foundation (CHE-1429195) for use of and funding for the
Bruker 500 MHz NMR. Mention of trade names or commercial
products in this publication is solely for the purpose of providing
specific information and does not imply recommendation or
endorsement by the U.S. Department of Agriculture.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
We thank Scott Larsen and Jessie Roper for their technical
assistance in sample preparation and analysis.
■
REFERENCES
■
(1) Colegate, S. M.; Dorling, P. R.; Huxtable, C. R. A spectroscopic
investigation of swainsonine: an α-mannosidase inhibitor isolated from
Swainsona canescens. Aust. J. Chem. 1979, 32, 2257−2264.
(2) Molyneux, R. J.; McKenzie, R. A.; O’Sullivan, B. M.; Elbein, A. D.
Identification of the glycosidase inhibitors swainsonine and calystegine
B2 in weir vine (Ipomoea sp. Q6{aff. calobra}) and correlation with
toxicity. J. Nat. Prod. 1995, 58, 878−886.
(3) Dantas, A. F. M.; Riet-Correa, F.; Gardner, D. R.; Medeiros, R. M.
T.; Barros, S. S.; Anjos, B. L.; Lucena, R. B. Swainsonine-induced
(21) Singh, D.; Kaur, G. Production, HPLC analysis, and in situ
apoptotic activities of swainsonine toward lepidopteran, SF-21 cell line.
Biotechnol. Prog. 2014, 30, 1196−1205.
(22) Gardner, D. R.; Molyneux, R. J.; Ralphs, M. H. Analysis of
swainsonine: extraction methods, detection, and measurement in
F
DOI: 10.1021/acs.jafc.6b02390
J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Journal of Agricultural and Food Chemistry
Article
populations of locoweeds (Oxytropis spp.). J. Agric. Food Chem. 2001,
49, 4573−4580.
(23) Dorling, P. R.; Colegate, S. M.; Huxtable, C. R. Toxic species of
the plant genus Swainsona. In Swainsonine and Related Glycosidase
Inhibitors; James, L. F., Elbein, A. D., Molyneux, R. J., Warren, C. D.,
Eds.; Iowa State University Press: Ames, IA, USA, 1989; pp 14−22.
(24) Stahr, H. M.; Martin, P. J. Analytical methods for swainsonine and
slaframine. In Swainsonine and Related Glycosidase Inhibitors; James, L. F.,
Elbein, A. D., Molyneux, R. J., Warren, C. D., Eds.; Iowa State University
Press: Ames, IA, USA, 1989; pp 118−137.
(25) Molyneux, R. J.; James, L. F. Swainsonine, the locoweed toxin:
analysis and distribution. In Handbook of Natural Toxins, Vol. 6,
Toxicology of Plant and Fungal Compounds; Keeler, R. F., Tu, A. T., Eds.;
Marcel Dekker: New York, 1991; pp 191−214.
(26) Molyneux, R. J.; James, L. F.; Panter, K. E.; Ralphs, M. H. The
occurrence and detection of swainsonine in locoweeds. In Swainsonine
and Related Glycosidase Inhibitors; James, L. F., Elbein, A. D., Molyneux,
R. J., Warren, C. D., Eds.; Iowa State University Press: Ames, IA, USA,
1989; pp 100−117.
(27) Gardner, D. R.; Lee, S. T.; Molyneux, R. J.; Edgar, J. A. Preparative
isolation of swainsonine from locoweed: extraction and purification
procedures. Phytochem. Anal. 2003, 14, 259−266.
(28) Gardner, D. R.; Cook, D. A comparison of alternative sample
preparation procedures for the analysis of swainsonine using LC-MS/
MS. Phytochem. Anal. 2011, 22, 124−127.
(29) Colegate, S. M.; Dorling, P. R.; Huxtable, C. R. Structural
elucidation of swainsonine: relationship of structure to activity. In
Swainsonine and Related Glycosidase Inhibitors; James, L. F., Elbein, A. D.,
Molyneux, R. J., Warren, C. D., Eds.; Iowa State University Press: Ames,
IA, USA, 1989; pp 91−99.
(30) Ralphs, M. H.; Creamer, R.; Baucom, D.; Gardner, D. R.; Welsh,
S. L.; Graham, J. D.; Hart, C.; Cook, D.; Stegelmeier, B. L. Relationship
between the endophyte Embellisia spp. and the toxic alkaloid
swainsonine in major locoweed species (Astragalus and Oxytropis). J.
Chem. Ecol. 2008, 34, 32−38.
G
DOI: 10.1021/acs.jafc.6b02390
J. Agric. Food Chem. XXXX, XXX, XXX−XXX