NeuroToxicology 33 (2012) 53–58

Contents lists available at SciVerse ScienceDirect

NeuroToxicology

Annonacin in Asimina triloba fruit: Implication for neurotoxicity

Lisa F. Potts a, Frederick A. Luzzio b, Scott C. Smith a,c, Michal Hetman c, Pierre Champy d, Irene Litvan a,e,f,*
a
Department of Anatomical Sciences and Neurobiology, University of Louisville, United States
b
Department of Chemistry, University of Louisville, United States
c
Department of Neurological Surgery, University of Louisville, United States
d
Laboratory of Pharmacognosy CNRS UMR8706 BioCIS, University Paris-Sud 11, France
e
Department of Neurology, University of Louisville, United States
f
Department of Neurosciences, University of California San Diego, United States

A R T I C L E I N F O A B S T R A C T

Article history: Introduction: The acetogenin, annonacin, from the tropical annonaceous plant Annona muricata, is a
Received 17 June 2011 lipophilic, mitochondrial complex I inhibitor reported to be more toxic than rotenone to mesencephalic
Accepted 26 October 2011 neurons. The temperate annonaceous plant Asimina triloba (pawpaw) is native to the Eastern United
Available online 23 November 2011
States and products are available online. This study determined whether annonacin is in the pawpaw
fruit pulp and whether it or the crude ethyl acetate extract is toxic to cortical neurons.
Keywords: Methods: Pawpaw extract was prepared by pulp extraction with methanol and liquid–liquid partitioning
Asimina triloba
with ethyl acetate (EtOAc). Annonacin was isolated from the crude EtOAc extract via column
Pawpaw
Annonacin
chromatography using a gradient solvent system of increasing polarity. Mass spectroscopy, nuclear
Annonaceous acetogenins magnetic resonance and infrared spectroscopy were used to compare isolated material with synthetic
Neurotoxicity annonacin data and a natural annonacin sample. Toxicity of isolated annonacin and the total EtOAc
extract was determined in primary rat cortical neurons using the MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-
diphenyltetrazolium bromide) assay.
Results: The average concentration of annonacin in the fruit pulp was 0.0701  0.0305 mg/g. Purified
annonacin (30.07 mg/ml) and crude EtOAc extract (47.96 mg/ml) induced 50% death of cortical neurons 48 h
post treatment. Annonacin toxicity was enhanced in the presence of crude extract.
Discussion: Pawpaw fruit contains a high concentration of annonacin, which is toxic to cortical neurons.
Crude fruit extract also induced neurotoxicity, highlighting the need for additional studies to determine
the potential risks of neurodegeneration associated with chronic exposure to pawpaw products.
ß 2011 Elsevier Inc. All rights reserved.

1. Introduction Khondiker et al., 2007; Hollerhage et al., 2009), and neurodegen-

eration of the basal ganglia and brainstem after its chronic,
Annonaceous acetogenins, specifically annonacin, isolated from intravenous administration to rats (Champy et al., 2004).
Annona muricata (soursop, graviola, guanabana, Brazilian pawpaw, Nigrostriatal pathology, mitochondrial impairment and/or tau
prickly custard apple) fruit, bark and leaves are naturally occurring pathology are seen in numerous neurodegenerative disorders
compounds exhibiting a range of biological activities related to including Parkinson’s disease, Alzheimer’s disease (AD) and
inhibition of mitochondrial complex I of the electron transport progressive supranuclear palsy (PSP) (Albers and Beal, 2002;
chain (Degli Esposti et al., 1994; Bermejo et al., 2005). The Robert and Mathuranath, 2007; Di Filippo et al., 2010). Moreover,
annonaceous plants have been investigated for their potential as regular consumption of fruit and tea from annonaceous plants was
cancer treatments (McLaughlin, 2008; Liaw et al., 2010), but also reported to be higher in patients with PSP and with an
for neurotoxicity (Champy et al., 2004; Lannuzel et al., 2006) due to unclassifiable atypical parkinsonian disorder than controls in
their high content of acetogenins (Champy et al., 2005, 2009). the French West Indies (Caparros-Lefebvre and Elbaz, 1999;
Annonacin has been reported to cause cell death and tau pathology Caparros-Lefebvre et al., 2002; Lannuzel et al., 2007).
in mesencephalic cultures (Lannuzel et al., 2003; Escobar- Annonacin has been reported to be present in the bark and
seeds of Asimina triloba (pawpaw, prairie banana, poor man’s
banana, Ozark banana, Banango, also commonly referred to as its
native states ‘‘banana’’, e.g. Indiana/Hoosier banana, Kentucky
* Corresponding author at: Movement Disorders Program, University of
California San Diego, Department of Neurosciences, 8950 Villa La Jolla Drive, Suite
banana, etc.) (McLaughlin, 2008), which grows throughout the
C112La Jolla, CA 92037, United States. Tel.: +1 858 822 5872; fax: +1 858 822 5743. Eastern United States and is touted as a potential alternative cash
E-mail address: ilitvan@ucsd.edu (I. Litvan). crop to tobacco. Furthermore, there is an annonacin-containing

0161-813X/$ – see front matter ß 2011 Elsevier Inc. All rights reserved.
doi:10.1016/j.neuro.2011.10.009
54 L.F. Potts et al. / NeuroToxicology 33 (2012) 53–58

commercial supplement made from pawpaw twig extracts, Paw- 22 mg/ml and stored at 80 8C. Serial dilutions of drugs were
Paw Cell-RegTM, that is marketed as beneficial for overall health prepared in basal medium Eagle (BME, Sigma) immediately before
(www.naturessunshine.com) and as a safe complement to cancer each experiment.
therapy (Wang et al., 2001; McLaughlin, 2008; Cuendet et al., 2008;
Pomper et al., 2009; Coothankandaswamy et al., 2010). We 2.3. Cell cultures
hypothesized that annonacin is also present in the pawpaw fruit
pulp, which is the most commonly consumed tree product. This Cortical neurons were cultured from postnatal day 0 (P0)
study determined the presence and concentration of annonacin in Sprague-Dawley rat pups according to previously published
the pawpaw fruit and its toxicity to cortical neurons, which are methods (Hetman et al., 1999). The handling of animals followed
affected in several neurodegenerative disorders such as AD and Institutional Animal Care and Use Committee approved proce-
fronto-temporal dementia (FTD). We also determined the toxicity dures. For viability assays, cultured neurons were plated in 96-well
of the crude organic extract. culture plates coated with poly-D-lysine (PDL) at a density of 1500–
2000 cells/mm2. 1-(b-D-Arabinofuranosyl) cytosine (2.5 mm) was
2. Materials and methods added to all wells on day 2 in vitro (DIV 2) to inhibit proliferation of
non-neuronal cells. Cells were treated with annonacin, cEX, or
2.1. Isolation vehicle (DMSO or EtOAc) on DIV 5. Rotenone is a model neurotoxin
and was therefore used as a positive control while the ACG-free
Annonacin was isolated from A. triloba using frozen (http:// extract served as a negative control for the extraction process. MTT
www.integrationacres.com/products.html, May 18, 2011) and fresh (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide)
(A Little Piece of Paradise Farm, Lincoln County, Kentucky) fruit pulp is converted to a formazan product by mitochondrial dehydroge-
following standard extraction and chromatography techniques nase in living cells. As the rate of conversion is proportional to the
(Liaw et al., 2005) using an optimized solvent system. Fruit pulp number of surviving cells, it provides a convenient measure of cell
(1000 g) was homogenized with methanol (800 ml) using a viability (Mosmann, 1983). Therefore, the MTT assay was used as
commercial blender and vacuum filtered at room temperature previously described (Hansen et al., 1989) to assess the toxicity of
overnight or until nearly dry. Liquid–liquid partitioning of the each treatment relative to vehicle-treated controls.
filtrate into ethyl acetate (EtOAc, 3 200 ml) yielded the crude
organic extract (cEX), which was dried and concentrated for weight 2.4. Statistical analysis
determination. Separation of extracts and isolation of annonacin
from a known amount of crude extracts was obtained by repeated One-way analysis of variance (ANOVA) followed by Dunnette’s
open column chromatographies (25 mm  250 mm glass column) post hoc test was used to determine significance and non-linear
on silica gel 60 (Merck 9385, 235–400 mesh) or 62 (Mallinckrodt regression analysis to calculate the LD50 values for annonacin and
6551, 60–200 mesh) using a gradient solvent system of hexane with crude extracts.
increasing concentrations of EtOAc and finally 5% methanol in
EtOAc. Annonacin and acetogenin containing fractions were 3. Results
identified by thin layer chromatography (TLC) with chloroform/
methanol (9:1) using standard visualization techniques (i.e. 3.1. Isolation
anisaldehyde/ethanol/sulfuric acid stain) as well as the Kedde
reagent (dipping of TLC plate in 3,5-dinitrobenzoic acid/methanol Data from TLC, FTMS, 1H- and 13C-NMR, FTIR spectroscopic
(2%, w/v) followed by dipping in potassium hydroxide/methanol analyses and optical rotations revealed the presence of annonacin
2 M) and weight determined following solvent removal in vacuo. The (HRMS calculated for C35H64O7 m/z [m+Na]+: 619.4543; Found:
entire extraction and isolation process was repeated on 5 separate 619.4546  0.6 ppm. [a]D20: +20.48 (c = 0.51, CHCl3)) in extracts
batches of frozen fruit and one batch of fresh fruit. High resolution isolated from pawpaw fruit pulp (Table 1, Fig. 1A). Annonacin
mass spectroscopy (HRMS, specifically FTMS using ESI) and nuclear tetraacetate was successfully prepared (HRMS calculated for
magnetic resonance (NMR, 500 MHz 1H and 13C) were used to C43H72O11 m/z [m+Na]+: 787.4966; Found: 787.4963  0.4 ppm.
compare the isolated annonacin to an annonacin sample from A. [a]D20: +7.458 (c = 0.51, CHCl3)) and isolated as determined by TLC,
muricata and synthetic annonacin data (Hu et al., 2001). An NMR, FTMS and FTIR (Table 1, Fig. 1B).
annonacin tetraacetate derivative was prepared by dissolving crude
annonacin (79 mg, 0.13 mmol) in pyridine (200 ml, 2.5 mmol and 3.2. Quantification
acetic anhydride (100 ml, 1.0 mmol). TLC confirmed full consump-
tion of the starting material during the reaction. Chromatographic The amount of annonacin isolated per gram of frozen fruit pulp
purification of the acetylated product (hexane/EtOAc 2:1) yielded was 0.0701  0.0305 mg as determined by average weight from
the tetraacetate derivative and spectral data were compared to that extractions on 5 separate batches (Table 2). On average, annonacin
of isolated annonacin. Optical rotations and infrared spectra (Fourier made up 10% of the total crude EtOAc extract. One additional
transform infrared spectroscopy, FTIR) were also recorded for extraction was done on locally harvested fresh fruit, which gave an
isolated samples of annonacin and annonacin tetraacetate. annonacin yield of 0.1226 mg/g of pulp.

2.2. Chemicals 3.3. Neurotoxicity

Solvents and reagents used for extractions were American Annonacin treatment reduced the number of viable cells
Chemical Society grade and used as commercially supplied. Stock remaining in the culture 48 h after treatment at concentrations
solutions of 1 mM rotenone (Sigma R8875) were prepared in as low as 10 mM (p < 0.001, Fig. 2A) as determined by MTT assay.
dimethyl sulfoxide (DMSO) or EtOAc prior to each experiment. ACG-free treatment did not result in cell death indicating that the
Purified annonacin and cEX were dissolved in DMSO or EtOAc to processing of extract components per se is not sufficient for
make 50 mM and 75 mg/ml stock solutions, respectively, which procuring toxic material (Fig. 2B). Pure annonacin 30.07 mg/ml (i.e.
were stored at 80 8C. Acetogenin-free fractions (ACG-free) 50.45 mM) and cEX (47.96 mg/ml) reduced the number of viable
isolated from cEX alongside annonacin were diluted in DMSO to cells compared to controls by 50% 48 h after treatment. Compari-
 L.F. Potts et al. / NeuroToxicology 33 (2012) 53–58 55

Table 1
Chemical structures and 1H and 13
C NMR data for annonacin (R5
5H) and annonacin tetraacetate (R5
5Ac).

5H, 1H assignment
R5 R5
5H, 13
C assignment 5Ac, 1H assignment
R5 R5
5Ac, 13
C assignment

– C-1: 174.7 – C-1: 173.4

– C-2: 131.2 – C-2: 130.1
H-3a: dd, 2.40 (15.2, 8.0) C-3: 33.3 H-3a: dd, 2.52 (14.4, 4.0) C-3: 30.8
H-3b: dd, 2.52 (15.2, 8.0) H-3b: dd, 2.52 (14.4, 4.0)
H-4: m, 3.85 C-4: 69.9 H-4: m, 4.80–5.02 C-4: 71.8
H-5–H-9: m, 1.25–1.62 C-5–C-9: 22.7–37.3 H-5–H-9: m, 1.25–1.40 C-5–C-9: 22.6–34.0
H-10: m, 3.59 C-10: 71.7 H-10: m, 4.80–5.02 C-10: 74.0
H-11–H-14: m, 1.25–1.62 C-11–C-14: 22.7–37.3 H-11–H-14: m, 1.25–1.40 C-11–C-14: 22.6–34.0
H-15: dt, 3.41 (12.0, 6.8) C-15: 74.0 H-15: m, 4.80–5.02 C-15: 74.7
H-16: m, 3.81 C-16: 82.7 H-16: m, 3.97 C-16: 79.4
H-17: m, 1.65/m, 1.98 C-17: 28.8/29.7 H-17: m, 1.73/m, 1.94 C-17: 28.0
H-18: m, 1.65/m, 1.98 C-18: 28.8/29.7 H-18: m, 1.73/m, 1.94 C-18: 28.0
H-19: m, 3.81 C-19: 82.7 H-19: m, 3.97 C-19: 79.5
H-20: dt, 3.41 (12.0, 6.8) C-20: 74.1 H-20: m, 4.80–5.02 C-20: 74.9
H-21–H-31: m, 1.25–1.62 C-21–C-31: 22.7–37.3 H-21–H-31: m, 1.25–1.40 C-21–C-31: 22.6–34.0
H-32: t, 0.88 (6.8) C-32: 14.1 H-32: t, 0.88 (7.0) C-32: 14.1
H-33: d, 7.18 (1.4) C-33: 151.9 H-33: d, 7.09 (1.5) C-33: 151.9
H-34: qd, 5.06 (6.8, 1.4) C-34: 78.0 H-34: qd, 5.09 (7.0,1.5) C-34: 77.5
H-35: d, 1.40 (6.8) C-35: 19.1 H-35: d, 1.40 (7.0) C-35: 18.9
– C5
5O (Ac): 170.6; 170.8; 170.9
CH3 (Ac): s, 2.03; s, 2.04; s, 2.07 CH3 (Ac): 21.1–21.2

Isolated and purified materials were dried and dissolved in CDCl3 (chemical shifts d in ppm; coupling constants J in Hz; 100 and 400 MHz).

son of LD50 values between pure annonacin and the amount of corresponds to that of annonacin prepared by stereoselective
annonacin present in cEX (LD50 = 4.80 mg/ml) confirmed that cEX synthesis, (Hu et al., 2001) which confirms that it possess the same
has more toxic potential than annonacin by itself (p < 0.0001, stereochemistry as annonacin. Based on repeated extractions we
Fig. 2C). determined that annonacin represents 0.007% of pawpaw pulp by
weight determination. This is higher than what is reported in
4. Discussion Annona cherimola seeds (0.0004%) (Kim et al., 2001) and A. muricata
(soursop) pulp 0.002% (Champy et al., 2005). The fruit pulp of A.
Herein we have: (1) found annonacin in fruit pulp of the triloba is a new, easily accessible source of biomass and our
annonaceous pawpaw tree, A. triloba, (2) determined the amount of extraction and isolation process procures a higher yield of
annonacin per gram of fruit pulp, and (3) shown that annonacin annonacin than the current, multistep synthetic route (Hu et al.,
and crude acetogenin-containing extracts are toxic to primary 2001), making it useful in exploring its potential as a novel model
cortical neurons. These findings are important particularly in view neurotoxin.
of previous reports of neurotoxicity associated with in vivo and in
vitro exposure to annonacin extracted from fruit or teas of A. 4.2. Neurotoxicity
muricata, a tropical annonaceous relative of pawpaw.
Acute annonacin treatment was toxic to cortical neurons (P0) at
4.1. Isolation and quantification low micromolar concentrations while treatment with the ACG-free
extract did not affect neuronal viability, indicating that the
We optimized a method for reliable extraction and isolation of neurotoxicity of annonacin was not an artifact of the extraction
annonacin from pawpaw fruit pulp. The presence of annonacin was process. Furthermore, annonacin was more toxic in the presence of
determined by TLC and mass spectroscopy (FTMS) using high-field cEX than when purified in terms of the amount of annonacin
NMR and infrared spectroscopy (FTIR) to confirm its identity by present in each treatment, suggesting that multiple acetogenins
comparison of spectral data with data from a natural annonacin synergistically affect neuronal viability. The presence of bis-
sample as well as that reported for synthetic annonacin. When tetrahydrofuranic acetogenins such as bullatacin is likely contrib-
examined by TLC in multiple solvent systems, the isolated uting to the neurotoxicity of cEX (Bermejo et al., 2005; Hollerhage
annonacin co-spotted with the A. muricata sample. NMR shifts et al., 2009). We found annonacin to be less potent in cortical
were consistent with those previously reported for synthetic neurons than previously reported in mesencephalic and striatal
annonacin (Hu et al., 2001) and FTMS and NMR spectra neurons (Escobar-Khondiker et al., 2007; Hollerhage et al., 2009),
corresponded with that of the annonacin sample. The successful which are more sensitive to such treatments. As annonacin was
formation of annonacin tetraacetate provided additional confir- less potent than rotenone in our experimental model, one can
mation that our isolated material was annonacin as noted by shifts speculate that the greater toxicity of the latter may be due to
in FTMS, NMR and FTIR absorptions where expected. Specifically reotenone’s inetarctions with additional cellular targets other than
the loss of OH bending in the FTIR tetraacetate spectra indicated the mitochondrial complex I (Klintworth et al., 2007; Choi et al.,
disappearance of the 4 hydroxyl groups that were evidenced by a 2008; Belcastro et al., 2009; Gill and Perez-Polo, 2009). Prior work
broad absorption band at 3416.74 cm 1 in the annonacin spectra. from Dr. Hetman’s lab has demonstrated that 80% or more of cells
Furthermore, the optical rotation of our purified material in the culture are neurons. Indeed in annonacin treated cells, there
56 L.F. Potts et al. / NeuroToxicology 33 (2012) 53–58

Fig. 1. Fourier transform infrared (FTIR) spectra of annonacin isolated from Asimina triloba. Broad peak at 3416.74 cm 1 = OH groups, 2922.2 cm 1 = CH2s,
2852.19 cm 1 = THF ring, 1744.09 cm 1 = butenolide (A) compared to spectra of the tetraacetate derivative 2927.53 cm 1 = CH2s, 2856.29 cm 1 = THF ring,
1737.72 cm 1 = acetates with butenolide shoulder (B). Note, as expected, in (B) there is no longer a peak corresponding to the OH groups confirming complete
consumption of starting material and tetraacetate formation. Purified samples dissolved in CHCl3 and spotted on salt plates for spectral acquisition using Perkin-Elmer
Spectrum 100 instrument.

were no MAP2-positive cells remaining in the culture (data not

Table 2 shown) after 48 h, indicating toxicity to neuronal cells as MAP2 is a
Annonacin content of Asimina triloba vs. Annona muricata. neuron-specific marker. To date, A. triloba toxicity has only been
Average total Amount Amount annonacin documented in cancer cell lines, and in cancerous murine models
fruit weighta annonacin per gram of fruit (McLaughlin, 2008). Our finding of in vitro cortical neurotoxicity
(g) per fruit (mg) pulpb (mg/g) corroborates previous reports of annonacin neurotoxicity to nigral
Asimina triloba 300 neurons (Lannuzel et al., 2002, 2003; Champy et al., 2004; Escobar-
Frozen fruit pulp 7.0 0.070 Khondiker et al., 2007) broadening the significance of annonacin-
Fresh fruit pulpc 12.3 0.123 induced toxicity as it relates to neurodegeneration.
Annona muricata 800 15.0 0.023
Previous reports assumed human exposure levels based on
a
Note these values reflect weight of the whole fruit (i.e. including skin, seeds and phytochemical and epidemiological studies and comparison of in
pulp). vivo findings. These estimations may be insightful for preliminarily
b
The amount of annonacin per gram of fruit pulp is based on the amount of pulp
used in the extraction, not the weight of the whole fruit.
assessing the risk of consuming pawpaw products; however,
c
Quantities in fresh fruit are based on data from one extraction; one pawpaw human exposure levels (i.e. levels of annonacin in blood or brain
fruit yielded 100 g of pulp on average. parenchyma) may not be extrapolated from our results, which
 L.F. Potts et al. / NeuroToxicology 33 (2012) 53–58 57

used an acute treatment method, the effects of chronic, in vivo

exposure to the total extract should be determined before firm
conclusions can be made regarding the risk associated with regular
consumption of pawpaw products.

5. Conclusions

Identification and quantification of annonacin in pawpaw fruit

is of interest considering the neurotoxicity reported here and in
previous studies (Champy et al., 2004; Escobar-Khondiker et al.,
2007; Hollerhage et al., 2009). The high concentration of annonacin
in pawpaw fruit is particularly relevant because consumption of
pawpaw in the United States could be much more widespread than
that of other annonaceous fruits considering (1) it is the only edible
annonaceous fruit that is not confined to tropical areas, (2) it grows
throughout the Eastern United States, (3) is marketed on the
worldwide web therefore available to consumers all year round
(e.g. frozen pulp, jam), and (4) is currently marketed for its
potential use in alternative medicine (Zhao et al., 1992;
McLaughlin, 2008; Cuendet et al., 2008; Coothankandaswamy
et al., 2010). This growing in popularity of pawpaw fruit highlights
the importance of ours and other’s findings of annonacin
neurotoxicity with regard to public health and supports the
inclusion of pawpaw exposure/consumption in studies investigat-
ing potential risk factors for neurodegeneration.

Conflicts of interest

The authors declare that there are no conflicts of interest.

Acknowledgements

The assistance of Ms. Juan Chen and Dr. Neal Stolowich in

recording 1H and 13C NMR spectra and of research students Mr.
Scott Jones and Ms. Sarah Fox are gratefully acknowledged as well
as the advice of Drs. Martha Bickford, Theo Hagg, and David Hein.
Support of the Mass Spectrometry Facility and specifically the
assistance of Dr. Bogdan Bogdanov of the University of Louisville
CREAM Center supported by NSF/EPSCOR grant EPS-0447479 is
also gratefully acknowledged. Support from the CEGIB through the
NIEHS grant P30ES014443 is also gratefully acknowledged. Dr.
Litvan is supported by NIH grant R01 PAS-03–092, National
Parkinson Foundation and Parkinson Support Center of Kentucki-
ana, and is the founder and CEO of the Litvan Neurological Research
Foundation. The generosity of Dr. Samuel Weakly and Ms. Lydia B.
Fig. 2. Cortical neurotoxicity of Asimina triloba fruit extracts 48 h after treatment.
Miller for donating research funds for this project is gratefully
Panel A: MTT assay of cortical neurons 48 h after treatment with annonacin isolated acknowledged.
from Asimina triloba fruit pulp. LD50 for annonacin = 50.45 mM,(i.e.) 1.07. Diagonal
lines = annonacin (mM), checkered line = rotenone (mM) p-value < 0.001 (***). Panel B: References
crude extract (cEX, light checkered bars) relative viability was significantly lower at
each dose than ACG-free (plaid bars) treatment p-value < 0.0001 (****). All doses of cEX Albers DS, Beal MF. Mitochondrial dysfunction in progressive supranuclear palsy.
significantly reduced cell viability compared to controls (p < 0.001). ACG-free Neurochem Int 2002;40:559–64.
treatment only significantly reduced viability at high concentrations (500 mg/ml, Belcastro V, Tozzi A, Tantucci M, Costa C, Di Filippo M, Autuori A, et al. A2A adenosine
p < 0.01). Bars represent relative viability, normalized to controls (mean  SEM). p- receptor antagonists protect the striatum against rotenone-induced neurotoxicity.
Values determined by one-way ANOVA with Dunnett’s post hoc comparison. Panel C: Exp Neurol 2009;217:231–4.
dose–response curves for cEX (LD50 = 47.96 mg/ml) (solid red line) and for pure Bermejo A, Figadere B, Zafra-Polo MC, Barrachina I, Estornell E, Cortes D. Acetogenins
annonacin (mg/ml) (solid blue line) compared to amount of annonacin (mg/ml) in the from Annonaceae: recent progress in isolation, synthesis and mechanisms of
crude extracts (dashed blue line) 48 h post-treatment. LD50 of pure annonacin alone action. Nat Prod Rep 2005;22:269–303.
(30.07 mg/ml, i.e. 50.45 mM) is significantly higher than that of annonacin in the Caparros-Lefebvre D, Elbaz A. Possible relation of atypical parkinsonism in the French
West Indies with consumption of tropical plants: a case-control study. Caribbean
presence of crude extracts (4.80 mg/ml, p < 0.0001 (****)). LD50 values determined by
Parkinsonism Study Group. Lancet 1999;354:281–6.
nonlinear regression of log (dose) vs. relative viability. Gray dashed lines represent 95%
Caparros-Lefebvre D, Sergeant N, Lees A, Camuzat A, Daniel S, Lannuzel A, et al.
confidence intervals for dose–response data. Guadeloupean parkinsonism: a cluster of progressive supranuclear palsy-like
tauopathy. Brain 2002;125:801–11.
Champy P, Guerineau V, Laprevote O. MALDI-TOF MS, profiling of annonaceous
cannot be directly compared to studies done in vivo. Although it acetogenins in Annona muricata products for human consumption. Molecules
has been demonstrated that annonacin crosses the blood brain 2009;14:5235–46.
barrier (Champy et al., 2004), additional studies should determine Champy P, Hoglinger GU, Feger J, Gleye C, Hocquemiller R, Laurens A, et al. Annonacin,
a lipophilic inhibitor of mitochondrial complex I, induces nigral and striatal
the bioavailability of annonacin and further elucidate the neurodegeneration in rats: possible relevance for atypical parkinsonism in Gua-
neurotoxicity of other compounds in the fruit. Likewise, as we deloupe. J Neurochem 2004;88:63–9.
58 L.F. Potts et al. / NeuroToxicology 33 (2012) 53–58

Champy P, Melot A, Guerineau Eng V, Gleye C, Fall D, Hoglinger GU, et al. Quantification Kim DH, Son JK, Woo MH. Annomocherin, annonacin and annomontacin: a novel and
of acetogenins in Annona muricata linked to atypical parkinsonism in guadeloupe. two known bioactive mono-tetrahydrofuran annonaceous acetogenins from
Mov Disord 2005;20:1629–33. Annona cherimolia seeds. Arch Pharm Res 2001;24:300–6.
Choi WS, Kruse SE, Palmiter RD, Xia Z. Mitochondrial complex I inhibition is not Klintworth H, Newhouse K, Li T, Choi WS, Faigle R, Xia Z. Activation of c-Jun N-terminal
required for dopaminergic neuron death induced by rotenone, MPP+, or paraquat. protein kinase is a common mechanism underlying paraquat- and rotenone-
Proc Natl Acad Sci USA 2008;105:15136–41. induced dopaminergic cell apoptosis. Toxicol Sci 2007;97:149–62.
Coothankandaswamy V, Liu Y, Mao SC, Morgan JB, Mahdi F, Jekabsons MB, et al. Lannuzel A, Hoglinger GU, Champy P, Michel PP, Hirsch EC, Ruberg M. Is atypical
The alternative medicine pawpaw and its acetogenin constituents suppress parkinsonism in the Caribbean caused by the consumption of Annonacae? J Neural
tumor angiogenesis via the HIF-1/VEGF pathway. J Nat Prod 2010;73: Transm Suppl 2006;153–7.
956–61. Lannuzel A, Hoglinger GU, Verhaeghe S, Gire L, Belson S, Escobar-Khondiker M, et al.
Cuendet M, Oteham CP, Moon RC, Keller WJ, Peadon PA, Pezzuto JM. Dietary adminis- Atypical parkinsonism in Guadeloupe: a common risk factor for two closely related
tration of Asimina triloba (Paw Paw) extract increases tumor latency in N-methyl- phenotypes? Brain 2007;130:816–27.
N-nitrosourea-treated rats. Pharm Biol 2008;46:3–7. Lannuzel A, Michel PP, Caparros-Lefebvre D, Abaul J, Hocquemiller R, Ruberg M.
Degli Esposti M, Ghelli A, Ratta M, Cortes D, Estornell E. Natural substances (acet- Toxicity of Annonaceae for dopaminergic neurons: potential role in atypical
ogenins) from the family Annonaceae are powerful inhibitors of mitochondrial parkinsonism in Guadeloupe. Mov Disord 2002;17:84–90.
NADH dehydrogenase (complex I). Biochem J 1994;301(Pt 1):161–7. Lannuzel A, Michel PP, Hoglinger GU, Champy P, Jousset A, Medja F, et al. The
Di Filippo M, Chiasserini D, Tozzi A, Picconi B, Calabresi P. Mitochondria and the link mitochondrial complex I inhibitor annonacin is toxic to mesencephalic dopaminer-
between neuroinflammation and neurodegeneration. J Alzheimers Dis 2010;20 gic neurons by impairment of energy metabolism. Neuroscience 2003;121:287–96.
(Suppl 2): S369–79. Liaw CC, Chang FR, Chen SL, Wu CC, Lee KH, Wu YC. Novel cytotoxic monotetrahy-
Escobar-Khondiker M, Hollerhage M, Muriel MP, Champy P, Bach A, Depienne C, et al. drofuranic annonaceous acetogenins from Annona montana. Bioorg Med Chem
Annonacin, a natural mitochondrial complex I inhibitor, causes tau pathology in 2005;13:4767–76.
cultured neurons. J Neurosci 2007;27:7827–37. Liaw CC, Wu TY, Chang FR, Wu YC. Historic perspectives on annonaceous acetogenins
Gill MB, Perez-Polo JR. Bax shuttling after rotenone treatment of neuronal primary from the chemical bench to preclinical trials. Planta Med 2010;76:1390–404.
cultures: effects on cell death phenotypes. J Neurosci Res 2009;87:2047–65. McLaughlin JL. Paw paw and cancer: annonaceous acetogenins from discovery to
Hansen MB, Nielsen SE, Berg K. Re-examination and further development of a precise commercial products. J Nat Prod 2008;71:1311–21.
and rapid dye method for measuring cell growth/cell kill. J Immunol Methods Mosmann T. Rapid colorimetric assay for cellular growth and survival: application to
1989;119:203–10. proliferation and cytotoxicity assays. J Immunol Methods 1983;65:55–63.
Hetman M, Kanning K, Cavanaugh JE, Xia Z. Neuroprotection by brain-derived neuro- Pomper KW, Lowe JD, Crabtree SB, Keller W. Identification of annonaceous acetogenins
trophic factor is mediated by extracellular signal-regulated kinase and phospha- in the ripe fruit of the North American pawpaw (Asimina triloba). J Agric Food Chem
tidylinositol 3-kinase. J Biol Chem 1999;274:22569–80. 2009;57:8339–43.
Hollerhage M, Matusch A, Champy P, Lombes A, Ruberg M, Oertel WH, et al. Natural Robert M, Mathuranath PS. Tau and tauopathies. Neurol India 2007;55:11–6.
lipophilic inhibitors of mitochondrial complex I are candidate toxins for sporadic Wang LQ, Li Y, Min BS, Nakamura N, Qin GW, Li CJ, et al. Cytotoxic mono-tetrahydrofuran
neurodegenerative tau pathologies. Exp Neurol 2009;220:133–42. ring acetogenins from leaves of Annona montana. Planta Med 2001;67:847–52.
Hu TS, Yu Q, Wu YL, Wu Y. Enantioselective syntheses of monotetrahydrofuran Zhao G, Hui Y, Rupprecht JK, McLaughlin JL, Wood KV. Additional bioactive compounds
annonaceous acetogenins tonkinecin and annonacin starting from carbohydrates. and trilobacin, a novel highly cytotoxic acetogenin, from the bark of Asimina triloba.
J Org Chem 2001;66:853–61. J Nat Prod 1992;55:347–56.