Index | Search | Home | Table of Contents

Johnson, H.A., J. Gordon, and J.L. McLaughlin. 1996. Monthly variations in biological activity of Asimina triloba. p. 609-614. In: J. Janick (ed.), Progress in new crops. ASHS Press, Arlington, VA.

Monthly Variations in Biological Activity of Asimina triloba

Holly A. Johnson, John Gordon, and Jerry L. McLaughlin*

  4. Table 1
  5. Table 2
  6. Fig. 1
  7. Fig. 2
  8. Fig. 3
  9. Fig. 4
  10. Fig. 5
  11. Fig. 6

The acetogenins are a class of compounds endogenous to the Annonaceae. Uvaricin, the first acetogenin discovered, was isolated from Uvaria accuminata in 1982 (Jolad et al. 1982). Since then, over 230 annonaceous acetogenins have been identified. The genera studied thus far which produce the acetogenins are Annona, Asimina, Goniothalamus, Rollinia, Uvaria, and Xylopia.

Asimina triloba, the North American paw paw tree or Indiana banana, is native to eastern and central North America. The tree bears an exotic flavored fruit while the bark, roots, twigs, and seeds contain the majority of the potent acetogenins (Ratnayake et al. 1992). The tree is usually pest and disease resistant, and acetogenins present inhibit the feeding of insects and herbivores.

Most acetogenins are potent biologically active molecules. They are inhibitors of Complex I (NADH: ubiquinone oxidoreductase) in mitochondria and, thus, cause a decrease in cellular ATP levels (Londerhausen et al. 1991; Ahammadsahib et al. 1993; Lewis et al. 1993). The acetogenins have shown patented antitumor, pesticidal, and antifeedant activities (Mikolajczak et al. 1988, 1989; Ikekawa et al. 1991; McLaughlin and Hui 1993). The chemical structures of the acetogenins consist of a long aliphatic chain, usually a variable number of tetrahydrofuran rings, followed by another aliphatic chain, ending in a gamma lactone ring. Anywhere from two to six hydroxyls can be found in each acetogenin, but molecules containing three hydroxyls have the highest biological activity (Landolt et al. 1995). Acetogenins are polyketide derived molecules and current research is being conducted to demonstrate that fatty acids are their precursors.

Of all the acetogenins, the adjacent bis-THF ring compounds are the most potent biologically (Hui et al. 1989; Zhao et al. 1994a, b; Landolt et al. 1995). Bullatacin (Fig. 1) inhibits mammalian solid tumor cells at concentrations often a billion times lower than Adriamycinreg., as shown in Table 1 (Zhao, et al. 1994b).

Asimicin, another adjacent bis-THF compound is the epimer of bullatacin at position 24 (Rupprecht et al. 1986). Asimicin has been shown to be effective and the whole class of compounds have been patented as natural pesticides as shown in Table 2 (Mikolajczak et al. 1988). Asimicin is a more effective insecticide than pyrethrins in three out of four assays performed. The F005 (crude extract) of Asimina triloba is also active in the pesticide tests. The F005 extract is 30% more effective in the mosquito larvae assay when compared with rotenone, a classic Complex I mitochondrial electron transport inhibitor (Ernster et al. 1963; Singer 1979). For the nematode assay, pyrethrins killed no nematodes at 10 ppm, whereas asimicin (0.1 ppm) and the F005 extract (10 ppm) showed 100% lethality after 72 hours.

The F005 bark extract has been field tested for pesticidal activity. Figure 2A displays a tree with Eastern tent catepillars (Malacosoma americanum) nested in a Prunus serotina tree. Tween 80, a non-ionic surfactant, was used to solubilize 0.5% F005 extract in water. The concentration of Tween 80 used was 1.0%. The aqueous surfactant extract solution was sprayed on the catepillars. After a thirty minute period, the majority of the insects were killed and fell from the tree (Fig. 2B).

In another experiment, the same 0.5% F005 bark extract solubilized in Tween 80 and water was sprayed on Phlox plants infested with powdery mildew fungus. The Phlox plants were sprayed with the extract solution and observed 10 days later. Figure 3 shows the Phlox without F005 (left-hand side) and with F005 sprayed (right-hand side). The F005 extract solution inhibited the powdery mildew fungus from growing on the Phlox.

Asimicin and bullatacin along with over 30 other bioactive acetogenins and other compounds (Rupprecht et al. 1986; Zhao et al. 1992, 1993, 1994a, b, 1995; Woo et al. 1995a, b) are present in the crude extract of Asimina triloba. For pesticidal purposes crude extracts rather than pure compounds can be used. Using the mixture saves much time and is more economical for the preparation of the pesticide. The crude extract produces a potent acetogenin mixture; hence, this mixture presents a host of different compounds to target different insects. The crude extract could also help to decrease insect adaptation and development of resistance to the acetogenins.


A. triloba twig samples from two trees (labelled east and west) were collected by John Gordon at an orchard in Amherst, New York. The twigs were dried and pulverized in a Wiley Mill to a fine powder. Originally, A. triloba was extracted with 95% ethanol followed by two sets of partitions (Fig. 4) before reaching F005 (Ratnayake et al. 1992). The bioactivity of the extracts was monitored by the brine shrimp lethality test (BST) (Meyer et al. 1982).

A new, more rapid, extraction procedure was developed to monitor the biological activity of multiple biomass samples of paw paw. This new method requires fewer steps and uses even less solvent. The twig samples from each month were extracted in triplicate. The residues of the three separate extractions for each tree each month were screened for activity using the BST. For each extraction, 2.5 g of dried twigs was placed in a scintillation vial and extracted with 20 ml of dichloromethane for 24 h. The dichloromethane eluate was filtered, and the solvent was left to evaporate in a scintillation vial in the hood overnight. The next day the dilutions for the brine shrimp assay were made starting from the residue (F003) left in the vial. The new modified assay is summarized in Fig. 5.


The LC50 value for each assay was calculated using a Finney Probit analysis program on an IBM personal computer (McLaughlin et al. 1991). The mean LC50 value was calculated for each tree each month. Graphical results are shown in Fig. 6. The east tree was sampled all 12 months from Sept. 1994 through Aug. 1995. The west tree was not sampled June and July 1995.

The east tree data shows that the twigs were more biologically active during the summer months of 1995. The month of May showed the highest biological activity with a mean BST of 0.7922 ± 0.250 ppm SD. The months of July and June had activities of 1.2644 ± 0.430 and 1.2972 ± 0.754 ppm, respectively. The winter months had a significant decrease in activity compared to the summer months. There was a one to two order of magnitude decrease in activity in late fall to winter.

The west tree was not collected June and July due to the tree under-going the stress of massive sampling and the drought. The tree was stressed by drought in the month of Aug., and these results were discarded. The top three months of biological activity for this tree were Sept. (0.4680 ± 0.181 ppm), May (0.9073 ± 0.341 ppm), and Apr. (1.4447 ± 1.0864 ppm), respectively. Also, the content of acetogenins was higher in the west tree (high in spring and summer; low in fall and winter) suggesting that genotypic variations might be significant. The data gathered from these two trees, from Sept. 1994 to Aug. 1995, indicates that biomass content of paw paw twigs would be optimized if collections were made in the months of May through July. Slight variations might be expected in other geographical regions; but the production of acetogenins by the trees shows definite monthly variations and seems to be maximized when protection from insects would be most beneficial to the plant species.

This new, rapid, assay will be used to monitor the biological activity of twig samples from a plantation of over 670 A. triloba trees at the Western Maryland Resource and Education Center in Keddysville, Maryland. It is hoped that the new extraction method and the brine shrimp lethality assay will then provide data to help find the most biologically active A. triloba genotypes.


*This work was supported by RO1 grant no. CA30909 from the National Cancer Institute, National Institutes of Health. R. Neal Peterson of the Paw Paw Foundation is greatfully acknowledged.
Table 1. The bioactivities of bullatacin in the brine shrimp test and human solid tumor cell lines.

Compound BST LC50 (µg/ml) ED50 (µg/ml) MCF-7 ED50 (µg/ml) HT-29 ED50 (µg/ml)
Bullatacin 1.59 x 10-3 1.25 x 10-13 >10 1.0 x 10-12
Adriamycin 2.57 x 10-1 1.04 x 10-4 1.76 x 10-2 1.53 x 10-4
BST = brine shrimp lethality test A-549 = human lung carcinoma
Adriamycin® = standard anticarcinoma drug MCF-7 = human breast carcinoma
ED50 = effective dose resulting in 50% cell death HT-29 = human colon carcinoma
LC50 = lethal concentration resulting in 50% deaths

Table 2. Comparison of pesticidal activity of asimicin and paw paw extract vs. standard insecticides.

Mortality Rate (%)
Compound Test organism Time of exposure (h) 0.1 ppm 1 ppm 10 ppm 50 ppm 100 ppm 500 ppm 5000 ppm
Asimicin, purified MBB 72 70 100
F005 (extract) MBB 72 60
Pyrethrins, (57% pure) MBB 72 0 100
Asimicin MA 24 20 100
F005 (extract) MA 24 80
Pyrethrins MA 24 20 100
Asimicin ML 24 100
F005 (extract) ML 24 10 80
Pyrethrins ML 24 100
Rotenone, (97% pure) ML 24 50 100
Asimicin NE 72 100
F005 (extract) NE 72 0 100
Pyrethrins NE 72 0
MBB = mexican bean beetle ML = mosquito larvae
MA = melon aphid NE = nematode (Caenorhabditis elegans)

Fig. 1. Structure of bullatacin.

Fig. 2. Left, Malacosoma americanum on Prunus serotina tree. Right, pesticidal activity of paw paw extract: 30 min. after 0.5% F005 bark extract and 1.0% Tween 80 were applied.

Fig. 3. Antifungal activity of paw paw extract. Phlox with powdery mildew. The plants on the right were sprayed with 0.5% F005 paw paw bark extract + 1.0% Tween 80. Photograph taken after treatment (10 d).

Fig. 4. Original extraction procedure for Asimina triloba.

Fig. 5. Modified extraction scheme for Asimina triloba.

Fig. 6. Monthly biological activity of Asimina triloba.

Last update August 25, 1997 aw