Ammodendrine and N-methylammodendrine enantiomers: Isolation, optical rotation, and toxicity

Article abstract of DOI:10.1021/np0580199

Ammodendrine (1) was found to occur as a mixture of enantiomers in two different collections of plants identified as Lupinus formosus. The ammodendrine fraction was reacted in a peptide coupling reaction with 9- fluorenylmethoxycarbonyl-L-alanine (Fmoc-L-Ala-OH) to give diastereomers, which were separated by preparative HPLC. The pure D- and L-ammodendrine enantiomers were then obtained by Edman degradation. Optical rotation measurements revealed that the D- and L-enantiomers had optical rotations of [a]²⁴ _D +5.4° and -5.7°, respectively. D- and L-N- methylammodendrine enantiomers were synthesized from the corresponding ammodendrine enantiomers, and their optical rotations established as [a] ²³_D +62.4° and -59.0°, respectively. A mouse bioassay was used to determine the difference in toxicity between these two pairs of naturally occurring enantiomers. The LD₅₀ of (+)-D-ammodendrine in mice was determined to be 94.1 ± 7 mg/kg and that of (-)-L-ammodendrine as 115.0 ± 7 mg/kg. The LD₅₀ of (+)-D-N-methylammodendrine in mice was estimated to be 56.3 mg/kg, while that of (-)-L-N-methylammodendrine was determined to be 63.4 ± 5 mg/kg. These results establish the rotation values for pure ammodendrine and N-methylammodendrine and indicate that there is little difference in acute murine toxicity between the respective enantiomers.

Full text of DOI:10.1021/np0580199

J. Nat. Prod. 2005, 68, 681-685
681
Ammodendrine and N-Methylammodendrine Enantiomers: Isolation, Optical
Rotation, and Toxicity
Stephen T. Lee,*,† Russell J. Molyneux,‡ Kip E. Panter,† Cheng-Wei Tom Chang,§ Dale R. Gardner,†
James A. Pfister,† and Massoud Garrossian†
Poisonous Plant Research Laboratory, Agricultural Research Service, United States Department of Agriculture,
1150 East, 1400 North, Logan, Utah 84341, Western Regional Research Center, Agricultural Research Service,
United States Department of Agriculture, 800 Buchanan Street, Albany, California 94710, and Department of
Chemistry and Biochemistry, Utah State University, 0300 Old Main Hill, Logan, Utah 84322
Received February 14, 2005
Ammodendrine (1) was found to occur as a mixture of enantiomers in two different collections of plants
identified as Lupinus formosus. The ammodendrine fraction was reacted in a peptide coupling reaction
with 9-fluorenylmethoxycarbonyl-L-alanine (Fmoc-L-Ala-OH) to give diastereomers, which were separated
by preparative HPLC. The pure D- and L-ammodendrine enantiomers were then obtained by Edman
degradation. Optical rotation measurements revealed that the D- and L-enantiomers had optical rotations
of [R]24 +5.4° and -5.7°, respectively. D- and L-N-methylammodendrine enantiomers were synthesized
D
from the corresponding ammodendrine enantiomers, and their optical rotations established as [R]23D +62.4°
and -59.0°, respectively. A mouse bioassay was used to determine the difference in toxicity between
these two pairs of naturally occurring enantiomers. The LD50 of (+)-D-ammodendrine in mice was
determined to be 94.1 ( 7 mg/kg and that of (-)-L-ammodendrine as 115.0 ( 7 mg/kg. The LD50 of (+)-
D-N-methylammodendrine in mice was estimated to be 56.3 mg/kg, while that of (-)-L-N-methylammo-
dendrine was determined to be 63.4 ( 5 mg/kg. These results establish the rotation values for pure
ammodendrine and N-methylammodendrine and indicate that there is little difference in acute murine
toxicity between the respective enantiomers.
Stereochemical integrity is a significant factor in deter-
mining the specificity of biological effects in both natural
products and synthetic compounds.1 Numerous examples
of differential activity have been reported, with one of the
most notorious being the drug thalidomide, since the
R-enantiomer has sedative effects whereas the S-enanti-
omer is teratogenic. Thalidomide was synthesized as a
racemic mixture of both enantiomers, but purification to
provide only the R-enantiomer for therapeutic purposes
was not a solution because a liver enzyme exists that
converts the R- to the S-enantiomer.2 Many naturally
occurring alkaloids have chiral centers and correspondingly
diverse modes of action. A recent example is that of the
antimalarial alkaloids febrifugine and isofebrifugine, for
which the natural compounds were found to have much
higher activity than their synthetic antipodes.3 When
describing plant-associated toxins, it is therefore also
essential to describe the stereochemical structure of any
chiral centers and accurately determine the optical rotation
in order to validate their inclusion in toxicity studies.
Ingestion of Lupinus species by pregnant cows at specific
gestational periods can result in calves with cleft palate
and front limb contractures, commonly known as crooked
calf disease.4-6 Ammodendrine (1) (Figure 1), a piperidine
alkaloid found in Lupinus formosus, has a chiral center and
is a reported teratogen.7,8 The variability of previously
reported optical rotation values for ammodendrine (Table
1) led us to believe that lupine plants could be producing
enantiomeric mixtures of 1.
Figure 1. Chemical structures of ammodendrine (1) and N-methyl-
ammodendrine (2).
San Mateo County, California, were both characterized
taxonomically as Lupinus formosus Greene (Leguminosae).
GC-MS analysis of the alkaloids in the above-ground plant
material from the Rio Vista and SMIP Ranch sites revealed
significantly different alkaloid profiles (Figure 2). The
chromatogram from the Rio Vista plant material shows
that ammodendrine (1) and N-acetylhystrine (3) are the
major piperidine alkaloids. Lower levels of quinolizidine
alkaloids were also found in this plant material.7 N-
Methylammodendrine (2) has also been found in this plant
material in previous years.7-10 The chromatogram from the
SMIP Ranch material shows that ammodendrine (1),
N-acetylhystrine (3), and N-methylammodendrine (2) are
the major piperidine alkaloids and that no quinolizidine
alkaloids were detected in this plant material. Ammoden-
drine (1) was measured as 0.47% dry weight of the Rio
Vista plant material and 0.63% dry weight of the SMIP
Ranch plant material.
Ammodendrine (1) was isolated from L. formosus plant
material collected in Solano County, California. The optical
Results and Discussion
Lupine plant specimens collected near the Rio Vista
Airport, Solano County, California, and at SMIP Ranch,
rotation of this ammodendrine isolate was [R]24 +2.9°, a
D
value that did not compare well with previously reported
measurements for this compound (Table 1). Ammodendrine
was then isolated from L. formosus collected in San Mateo
County, California, 10 days after the previous collection.
The optical rotation of this isolate was [R]24D +1.8°, which
* To whom correspondence should be addressed. Tel: (435) 752-2941.
Fax: (435) 753-5681. E-mail: stlee@cc.usu.edu.
† Poisonous Plant Research Laboratory, USDA.
‡ Western Regional Research Center, USDA.
§ Utah State University.
10.1021/np0580199 CCC: $30.25
© 2005 American Chemical Society and American Society of Pharmacognosy
Published on Web 04/14/2005

682 Journal of Natural Products, 2005, Vol. 68, No. 5
Lee et al.
Table 1. Optical Rotation and Toxicity Data for Ammodendrine (1) and N-Methylammodendrine (2)
compound
observed or reported optical rotations
LD50 ( CIa
plant source
D-ammodendrineb
[R]24D +5.4° (c 1.27, MeOH)
[R]24D -5.7° (c 1.66, MeOH)
[R]24D +6.65° (c 3.9, EtOH)
[R]24D +7.5° (c 0.22, EtOH)
[R]24D +7.1° (c 0.08, MeOH)
[R]23D -20° (c 0.004, CDCl3)
[R]24D +15° (EtOH)
94.1 ( 6.8
115 ( 7.0
Lupinus formosus
L. formosus
L-ammodendrineb
ammodendrine11
L. formosus
ammodendrine12
L. varius
L. varius
Castilleja miniata
ammodendrine13
ammodendrine14
ammodendrine15
ammodendrine16
[R]24D +11.8° (c 0.288, EtOH)
[R]23D +62.4° (c 0.51, MeOH)
[R]23D -59.0° (c 1.17, MeOH)
[R]24D +40.5° (c 2.0, EtOH)11
[R]23D -44° (c 0.02, CDCl3)13
L. hirsutus
D-N-methylammodendrineb
L-N-methylammodendrineb
N-methylammodendrine
N-methylammodendrine
56.3
63.4 ( 4.7
L. formosus
L. formosus
L. formosus
Castilleja miniata
a CI (95% confidence intervals). b This study.
Figure 2. Gas chromatograms and alkaloid profiles of Lupinus formosus collected from (A) near the Rio Vista Airport and (B) SMIP Ranch.
was not in accord with the value reported for ammoden-
drine (+6.6°) from L. formosus collected at the same site
32 years previously.11
HPLC, and because of their earlier elution than their Fmoc
analogues, this method was scaled up for the separation
and isolation of milligram quantities of the respective
diastereomers (Figure S2).
Attempts to resolve the enantiomers by HPLC, using
chiral columns, proved to be unsuccessful. We resorted to
the conversion of the ammodendrine enantiomers into
diastereomers by a peptide coupling reaction utilizing
Fmoc-L-Ala-OH in the presence of N-(3-dimethylaminopro-
pyl)-N′-ethylcarbodiimide hydrochloride and hydroxyben-
zotriazole dissolved in N,N-dimethyl formamide (DMF)
(Figure 3). The formation of Fmoc-L-Ala-ammodendrine (4)
(MH+ 502) was monitored by atmospheric pressure chemi-
cal ionization-mass spectrometry (APCIMS). The conver-
sion of 1 to 4 allowed reversed-phase HPLC analysis of the
diasteriomers (Figure S1) and confirmed that the ammo-
dendrine enantiomers were present in both the Rio Vista
Airport and SMIP Ranch L. formosus plants.
Treatment of the Fmoc-L-Ala-ammodendrine diastere-
iomers (4) with piperidine in methylene chloride was
monitored by electrospray ionization-mass spectrometry
(ESIMS) for the formation of Ala-ammodendrine (5) (MH+
280). The resultant L-Ala-ammodendrine diastereiomers
were separated from each other using reversed-phase
Conversion of the isolated diastereomers to their respec-
tive enantiomeric forms of ammodendrine (1) was ac-
complished by an Edman degradation procedure to remove
the L-alanine. The progress of the reaction was monitored
by ESIMS and indicated that the major portion of the
reaction residue was ammodendrine (1) (MH+ 209). D- and
L-ammodendrine were isolated from their respective reac-
tion mixture using conventional silica gel column chroma-
tography. Reversed-phase HPLC chromatography of the
isolated ammodendrine enantiomers (1) after conversion
to the Fmoc-L-Ala-ammodendrine (4) (Figure S3) estab-
lished the D-ammodendrine enantiomer to be 91% and
L-ammodendrine to be 98% enantiomerically pure.
The optical rotations of D-ammodendrine ([R]24 +5.4°)
D
and L-ammodendrine ([R]24 -5.7°) are consistent with
D
expected optical rotations of isolated paired enantiomers.
However, these optical rotations differ from previously
reported values (Table 1) where no measures were taken
to ensure enantiomeric purity of the isolates. The greater

Ammodendrine Enantiomers
Journal of Natural Products, 2005, Vol. 68, No. 5 683
rotations are possibly due to contamination of the samples
by other alkaloids with higher rotation values, such as
N-methylammodendrine (2).
N-Methylammodendrine (2) was detected in L. formosus
collected from the SMIP Ranch and was present in previous
collections of L. formosus from the Rio Vista site and has
been found in other Lupinus species.7-10,17 We were there-
fore interested in determining the correct optical rotation
values for the enantiomers of N-methylammodendrine (2).
The N-methyl derivatives were obtained by treating D- and
L-ammodendrine (1) with iodomethane and were separated
from their respective reaction mixtures using silica gel
column chromatography.
The optical rotations of D-N-methylammodendrine and
L-N-methylammodendrine were [R]23 +62.4° and [R]23
D
D
-59.0°, respectively. As in the case of the ammodendrine
enantiomers, these measurements are consistent for pairs
of enantiomers and they are also substantially different
from the measurements previously reported (Table 1).
Toxicity of ammodendrine (1) and N-methylammoden-
drine (2) enantiomers was evaluated in a mouse bioassay.
The LD50 of L-ammodendrine in mice was determined to
be 115.0 ( 7.0 mg/kg, n ) 20, and that of D-ammodendrine
was 94.1 ( 7.8 mg/kg, n ) 23. The LD50 of L-N-methylam-
modendrine in mice was determined to be 63.4 ( 4.7 mg/
kg, n ) 15, and that of D-N-methylammodendrine in mice
was estimated to be 56.3 mg/kg (no CI), n ) 7 (Table 1).
The N-methylammodendrine enantiomers were therefore
appreciably more toxic than the ammodendrine isomers.
Although there was a statistical difference (P < 0.05) in
the measured toxicity between the D- and L-ammodendrine,
the observed difference in physiological activity measured
in the mouse bioassay was not as large as is often the case
with enantiomers.1,3 It is possible that the receptor site
responsible for toxicity is not highly chirally discriminatory.
It is also possible that none of these compounds is the
proximate toxin and that they are metabolized into com-
pounds in which the asymmetric center is lost. This could
occur either by isomerization of the existing double bond
to a position between the two piperidine rings or by an
oxidative process in the liver, resulting in the introduction
of an additional double bond in the fully saturated piperi-
dine ring. An analogy for the latter process exists with the
pyrrolizidine alkaloids, which are not toxic per se but are
converted to toxic dehydropyrrolizidines by P450 liver
enzymes.18
It is recognized that the acute mouse toxicity model is
not a surrogate for teratogenic crooked calf disease. Limi-
tations in the amount of isolated enantiomeric material
available only allowed us to investigate initially possible
differences in toxicity based on a mouse model. However,
a true measure of teratogenicity will need to be tested in a
proper system such as a fetal goat model.19
In this study, the similarities in toxicity between the
enantiomeric pairs of ammodendrine (1) and N-methylam-
modendrine (2) indicate that, in the absence of evidence
to the contrary, both enantiomers should be taken into
account in predicting toxicity of plant samples to livestock,
as has been done previously.
Experimental Section
Figure 3. Conversion of ammodendrine (1) into diastereomers for
separation and conversion of diastereomers back to enantiomerically
pure ammodendrine isomers.
General Experimental Procedures. Optical rotations
were recorded using a Perkin-Elmer model 241 polarimeter.
Electrospray (ESI) and atmospheric pressure-chemical ioniza-
tion (APCI) mass spectrometric data were acquired using an
LCQ mass spectrometer (Finnigan, San Jose, CA). Samples
were loop injected into the ESI or APCI source using a
magnitude of the previously reported optical rotations than
those reported in this study indicates that prior optical
rotations for ammondendrine (1) are incorrect. The larger

684 Journal of Natural Products, 2005, Vol. 68, No. 5
Lee et al.
methanol-20 mM ammonium acetate solution, 50:50 v/v, at
The basic aqueous layer was then extracted (3×) with CHCl3.
The CHCl3 extracts from the basic aqueous portion were
combined, dried with anhydrous Na2SO4, filtered, and rotary
evaporated to dryness. The residue was sampled by flow
injection ESIMS. The mass spectrum indicated that the
residue was Ala-ammodendrine (5) (MH+ 280) and that this
reaction had proceeded to completion, as Fmoc-Ala-ammoden-
drine (4) was not present in the mass spectrum: HRESIMS
m/z [M + H]+ found, 280.2022; calcd for C15H26N3O2, 280.2025.
The Ala-ammodendrine diastereomers (5) were separated
using a preparative scale, 250 mm × 21.2 mm i.d., 5 µm,
Betasil C18 HPLC column (Thermo Hypersil-Keystone). The
mobile phase was 20 mM ammonium acetate-methanol (80:
20, v/v) at a flow rate of 25 mL/min. Two major peaks eluted,
and they were collected and combined with the corresponding
peaks in subsequent runs. The mobile phase was evaporated
to <500 mL and made basic to pH 9 with NH4OH. The basic
aqueous layer was then extracted with equal volumes of
chloroform (3×). The CHCl3 extracts were combined, dried with
anhydrous Na2SO4, filtered, and rotary evaporated to dryness.
The Ala-ammodendrine (5) diastereomers were analyzed
using HPLC using an analytical Betasil C18 column with a 20
mM ammonium acetate-methanol mobile phase (80:20, v/v)
at a flow rate of 0.5 mL/min; mass spectrometry in the ESIMS
mode was used for detection.
Edman Degradation. The L-alanine portion of the Ala-
ammodendrine (5) diastereomers was removed via Edman
degradation. Each Ala-ammodendrine diastereomer (134.5 mg,
0.481 mmol) was treated with a methanol-water-triethyl-
amine-phenylisothiocyanate (3.5:0.5:0.5:0.5, v/v/v/v) solution
with stirring at 50 °C. After 1.5 h, trifluoroacetic acid (0.5 mL)
was added and the reaction was stirred again at 50 °C for an
additional 1.5 h. The solvents from the reaction were evapo-
rated under a stream of N2 at 65 °C. The reaction mixture
was partitioned in 1% H2SO4 and CHCl3. The CHCl3 was
discarded. The aqueous portion was made basic to pH 9, with
the addition of NH4OH, and then extracted with CHCl3. The
CHCl3 extracted from the basic aqueous portion were com-
bined, dried with anhydrous Na2SO4, filtered, and rotary
evaporated to dryness. The residue was analyzed using ES-
IMS, and the resultant mass spectrum indicated that the
major portion of the residue was ammodendrine (1) (MH+ 209).
The chiral ammodendrine (1) was isolated by chromatography
on a 40 cm × 2.3 cm silica gel column using a mobile phase of
CHCl3-MeOH-NH4OH (65:35:l, v/v/v). Finally, 1 was cleaned
by acid/base extraction using 1% aqueous H2SO4 and extract-
ing with CHCl3, and then the aqueous portion made basic to
pH 9 with the addition of NH4OH and extracted with CHCl3.
These extracts were combined, dried with anhydrous Na2SO4,
filtered, and rotary evaporated to dryness to give the pure
compound (80.6 mg, 0.388 mmol, yield 80.6%). No isomeriza-
tion of the chiral ammodendrine (1) occurred using these acid
and base conditions.
Synthesis of D- and L-N-Methylammodendrine. D- or
L-ammodendrine (1) (38.9 mg, 0.187 mmol) was weighed into
a 1 mL Reactivial (Pierce, Rockford, IL) with a magnetic
triangular stirbar. Acetone (200 µL) and iodomethane (27 µL)
were added and the reaction was stirred at 57 °C for 1 h. The
acetone was evaporated off under a stream of N2 at 60 °C, the
reaction mixture was dissolved in CHCl3, and the organic layer
was washed with 10% aqueous NaHCO3. The CHCl3 layer was
then dried with anhydrous Na2SO4 and evaporated to dryness
with N2 at 60 °C. N-Methylammondendrine (2) was separated
from the reaction contaminants using silica gel chromatogra-
phy on a 35 cm × 1.25 cm column with a mobile phase of
CHCl3-MeOH-NH4OH (65:35:l, v/v/v) to give 12.1 mg, 0.0544
mmol, yield 29%.
Mouse Bioassay. Known amounts of the individual alka-
loids were dissolved in physiological buffered saline solution.
The solutions were stored in sterile injection vials for toxicity
testing.
Weanling White Swiss-Webster male mice, 15 to 20 g
(Simonsen Labs, Gilroy, CA), were weighed after a 12 h fast
and were dosed intravenously. Injections were performed via
the tail vein in mice restrained in a plastic mouse block. The
a flow rate of 0.5 mL/min.
Chemicals and Reagents. Glacial acetic acid, ammonium
hydroxide, N,N-dimethyl formamide (DMF), and sulfuric acid
were purchased from Fisher Scientific (Pittsburgh, PA). Hy-
droxybenzotriazole, phenyl isothiocyanate (99%), piperidine
(99.5%), and silica gel (70-230 mesh, 60 Å) for column
chromatography were obtained from Aldrich Chemical (Mil-
waukee, WI). (L)-Fmoc-Ala-OH and N-(3-dimethylaminopro-
pyl)-N′-ethylcarbodiimide hydrochloride were obtained from
Fluka Chemical (Ronkonkoma, NY). Trifluoroacetic acid was
from EM Science (Gibbstown, NJ), sodium sulfate from Baker
(Phillipsburg, NJ), chloroform from Mallinckrodt Baker (Paris,
KY), and ammonium acetate from VWR (Bristol, CT).
Plant Material. Lupine plant material was collected near
Rio Vista Airport, Solano County, CA (38°10.50′ N/121°44.38′
W; elevation 7.9 m), on June 25, 2003, and at SMIP Ranch,
San Mateo County, CA (37°21.66′ N/122°17.87′ W; elevation,
398 m), on July 2, 2003. The plant was in the flowering stage
on these dates, and the samples consisted of the whole plant
except for roots. The plant specimen collected rear the Rio
Vista Airport, accession #239432, and the plant specimen
collected at the SMIP Ranch, accession #239414, were both
taxonomically classified as Lupinus formosus by staff at the
Intermountain Herbarium, Utah State University, where the
voucher specimens are retained.
Extraction and Isolation of Lupine Alkaloids. Aerial
plant material was air-dried and ground to pass through a 2
mm screen. The plant material (366.7 g) was extracted by
steeping at room temperature for 16 h in methanol (4 × 4 L),
the methanol extracts were combined, and the methanol was
removed via rotary evaporation, leaving a dark green residue.
The residue was first partitioned between 1% aqueous H2SO4
(2 L) and CHCl3 (2 × 2 L). The CHCl3 was discarded. The
aqueous portion was made basic to pH 9, with the addition of
NH4OH, and then extracted with CHCl3. The CHCl3 extracted
from the basic aqueous portion was combined, dried with
anhydrous Na2SO4, filtered, and rotary evaporated to dryness.
Ammodendrine (1) was isolated by chromatography on a 40
cm × 3.5 cm silica gel column using a mobile phase of CHCl3-
MeOH-NH4OH (65:35:l, v/v/v).
Synthesis of Ammodendrine Diastereomers. Ammo-
dendrine (1) (106.2 mg, 0.510 mmol), Fmoc-L-Ala-OH (182.0
mg, 0.553 mmol), N-(3-dimethylaminopropyl)-N′-ethylcarbo-
diimide hydrochloride (128.5 mg, 0.670 mmol), and hydroxy-
benzotriazole (90.5 mg, 0.670 mmol) were weighed into a
round-bottomed flask (10 mL) with a magnetic stirbar; N,N-
dimethylformamide (DMF) (7 mL) was then added. The
reaction was stirred under N2 at RT for 16 h. DMF was
evaporated with compressed air, the reaction mixture was
dissolved in CHCl3, and the organic layer was washed (2×)
with distilled deionized water. The CHCl3 phase was then
dried over anhydrous Na2SO4, filtered, and rotary evaporated
to an oily residue. The residue was analyzed using APCIMS,
and the mass spectrum indicated that ammodendrine (1) had
reacted to form Fmoc-L-Ala-ammodendrine (4) (MH+ 502).
Based on reversed-phase HPLCMS analysis, approximately
50% of the ammodendrine (1) was consumed to form Fmoc-L-
Ala-ammodendrine: HRESIMS m/z [M + Na]+ found, 524.2506,
calcd for C30H3N3O4Na, 524.2525.
Plant samples and chiral isolates were derivatized by the
procedure above and analyzed by HPLC using a 100 mm × 2
mm i.d., 5 µm, Betasil C18 column (Thermo Hypersil-Keystone,
Bellefonte, PA). The mobile phase was 20 mM ammonium
acetate-methanol (45:55, v/v) at a flow rate of 0.5 mL/min.
The detector was a Finnigan LCQ mass spectrometer operat-
ing in the APCI mode.
The Fmoc portion of L-Fmoc-Ala-ammodendrine (4) was
cleaved by treatment of 4 with a 25% solution of piperidine (1
mL) in methylene chloride (3 mL) for 1 h at RT. The solvent
was evaporated from the reaction mixture with a gentle flow
of N2 and heat at 65 °C. The reaction mixture was dissolved
in 1% H2SO4 and CHCl3, and the aqueous layer was washed
(3×) with CHCl3 and then made basic to pH 9 with NH4OH.

Ammodendrine Enantiomers
Journal of Natural Products, 2005, Vol. 68, No. 5 685
material. This material is available free of charge via the Internet at
http://pubs.acs.org.
mice were maintained under a heat lamp for 15 min to dilate
the tail vein. The tail was cleaned with 70% ethanol, and i.v.
injections were accomplished with a tuberculin syringe equipped
with a 1.27 cm long 27-gauge needle. The volume injected
varied depending on the dosage delivered. Time of injection,
clinical effects, and time of death were noted and recorded.
The LD50 for individual alkaloid toxicity was determined
using a modified up-and-down method20 and was calculated
using the PROC PROBIT procedures of SAS (SAS Institute,
Cary, NC) on a logistic distribution of the survival data.
Confidence (fiducial) intervals (95%) were also calculated using
the same program.
References and Notes
(1) Brossi, A. Med. Res. Rev. 1994, 14, 665-691
(2) Eriksson, T.; Bjorkman, S.; Hoglund, P. Eur. J. Clin. Pharmacol.
2001, 57, 365-376.
(3) Kobayashi, S.; Ueno, M.; Suzuki, R.; Ishitani, H.; Kim, H. S.; Wataya,
Y. J. Org. Chem. 1999, 64, 6833-6841.
(4) Shupe, J. L.; James, L. F.; Binns, W. J. Am. Vet. Med. Assoc. 1967,
151, 191-197.
(5) Shupe, J. L.; Binns, W.; James, L. F.; Keeler, R. F. J. Am. Vet. Med.
Assoc. 1967, 151, 198-203.
(6) Shupe, J. L.; Binns, W.; James, L. F.; Keeler, R. F. Aust. J. Agric.
Res. 1968, 19, 335-340.
Acknowledgment. The protocol for animal use in this
research was reviewed and approved by the Institutional
Animal Care and Use Committee (IACUC), Utah State Uni-
versity, Logan, UT. We wish to thank J. Christopherson and
S. W. Larsen for technical assistance and E. Thacker for
assistance in plant collection. We are grateful to the proprietor
and manager of SMIP Ranch, San Mateo County, CA, for
permitting plant collection on the property. We appreciate the
assistance of M. E. Barkworth, Director, and K. M. Piep,
Assistant Curator, of the Intermountain Herbarium at Utah
State University for taxonomic confirmation of plant speci-
mens. HRMS was provided by the Washington University
Mass Spectrometry Resource, an NIH Research Resource
(Grant No. P41RR0954).
(7) Keeler, R. F.; Panter, K. E. Teratology 1989, 40, 423-432.
(8) Panter, K. E.; Gardner, D. R.; Molyneux, R. J. J. Nat. Toxins 1998,
7, 131-140.
(9) Wink, M.; Witte, L. Z. Naturforsch. 1987, 42C, 197-204.
(10) Gardner, D. R.; Panter, K. E. J. Nat. Toxins 1994, 3, 107-116.
(11) Fitch, W. L.; Dolinger, P. M.; Djerassi, C. J. Org. Chem. 1974, 39,
2974-2979.
(12) Abel-Halim, O. B.; El-Gammal, A. A.; Abdel-Fattah, H.; Takeya, K.
Phytochemistry 1999, 51, 5-9.
(13) Mohamed, M. H.; Hassanean, H. A. Phytochemistry 1997, 46, 365-
369.
(14) McCoy, J. W.; Stermitz, F. R. J. Nat. Prod. 1983, 46, 902-907.
(15) Dictionary of Natural Products on CD-ROM, version 7.2; Chapman
and Hall: London, 1999.
(16) Takamatsu, S.; Saito, K.; Sekine, T.; Ohmiya, S.; Kubo, H.; Otomasu,
H.; Murakoshi, I. Phytochemistry 1990, 12, 3923-3926.
(17) Panter, K. E.; Mayland, H. F.; Gardner, D. R.; Shewmaker, G. Vet.
Human Toxicol. 2001, 43, 279-282.
(18) Mattocks, A. R. Chemistry and Toxicology of Pyrrolizidine Alkaloids;
Academic Press: Orlando, FL, 1986.
(19) Panter, K. E.; Bunch, T. D.; Keeler, R. F.; Sisson, D. V.; Callan R. J.
J. Clin. Toxicol. 1990, 28, 69-83.
(20) Bruce, R. Fundam. Appl. Toxicol. 1987, 8, 97-101.
Supporting Information Available: Figure S1 shows recon-
structed HPLC ion chromatograms comparing ammodendrine-based
diastereomers from L. formosus plant material collected at the Rio
Vista and SMIP Ranch sites. Figures S2 and S3 show reconstructed
ion chromatograms for L-Ala-ammodendrine (5) and Fmoc-L-Ala-
ammodendrine diastereomers (4) isolated from L. formosus plant
NP0580199