Safety tests and antinutrient analyses of noni (Morinda citrifolia L.) leaf
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
Journal of the Science of Food and Agriculture J Sci Food Agric 87:2583–2588 (2007) Safety tests and antinutrient analyses of noni (Morinda citrifolia L.) leaf Brett J West,1∗ Hisanori Tani,2 Afa K Palu,1 Charles B Tolson1 and C Jarakae Jensen1 1 Research and Development Department, Tahitian Noni International, American Fork, Utah, USA 2 Akikawa Research Center, Tangle Wood, Saitama, Japan Abstract BACKGROUND: Noni (Morinda citrifolia L.) leaves have a documented history of food use. However, previous safety and antinutrient studies are absent. The current investigation was conducted to assess the utility of noni leaves as food. RESULTS: No evidence of toxicity or differences in weight gain were observed in acute, subacute, and subchronic oral toxicity tests of ethanol–water (1:1 v/v) and hot-water extracts of noni leaves in mice at doses of 2000, 200, and 20 mg kg−1 body weight, respectively. Acute systemic anaphylaxis tests of the ethanol–water (4:1 v/v) and hot-water extracts were negative. Further, leaf proteins were readily digested in simulated gastric fluid. Tannic acid concentrations in frozen and dried leaf were 1.6 and 25.8 g kg−1 , respectively. Phytic acid was not detected in the raw leaf (
BJ West et al.
One volume of ground leaves was extracted with tract, heart, kidneys, liver, lungs, pancreas, spleen, and
five volumes of hot deionized water for 1 h. The other organs were recorded.
aqueous extract was clarified to 10 g kg−1 total solids The subacute toxicity tests consisted of the same
by centrifugation, decanted and then freeze-dried to animal breed, test and control group composition,
obtain M. citrifolia leaf, hot-water extract (MCW). housing, and animal acclimation and preparation as
Noni leaf ethanol extract (MCE) was prepared in those of the acute oral toxicity tests. However, in this
a similar process, but with ethanol–water solution test, administration by gavage was not used. Rather,
(1:1 v/v) used for the extraction in place of the hot MCW or MCE was mixed with the feed of the test
water, and the ethanol evaporated off before freeze groups at 200 mg per animal per day. Controls received
drying. The purpose of preparing these extracts was no extracts in their feed. The same observations and
to concentrate any potential antinutrient or toxic measurements as those of the acute oral toxicity test
compounds, thus allowing more sensitive toxicity tests were employed in this test. However, the observation
of the leaf. period was doubled to 28 days, and body weights were
For the allergenicity tests, milled dry noni leaves recorded on days 3, 7, 14, 21, and 28.
were extracted with ethanol–water (4:1 v/v) for 4 h. The subchronic toxicity test followed the same
The extract was filtered, ethanol evaporated, and then protocol as the subacute test with the following
freeze-dried to comprise the ethanol extract. Milled exceptions. The dose of the extracts was lowered to
dry noni leaves were also boiled in hot water for 20 mg per animal per day. The observation period was
30 min, then cooled and centrifuged. The aqueous extended to 90 days, and body weights were recorded
extract, or infusion, with 79.5 g kg−1 total solids, was on days 3, 7, 14, 21, 28, 42, 56, 70, 84, and 90.
decanted and used immediately. Statistical analyses of body weight gain were
A sample of noni leaves, harvested from the Society performed for all oral toxicity tests. Summary statistics
Islands of French Polynesia, was collected for the were determined, as well as one-way analysis of
pepsin resistance test. This sample, and samples of variance (ANOVA). Where intergroup differences
dried leaves, were used for tannic acid and phytic were determined, Dunnett’s Multiple Comparisons
acid determinations. Additionally, samples of leaves test14,15 was performed to compare the MCW and
were collected from 22 sites on 11 islands throughout MCE group values with those of the control. Bartlett’s
French Polynesia. These were then tested for the test16 was performed to determine homogeneity of
presence, composition, and variability of oxalic acid variance. Where the data were heterogeneous, the
and phytosterols. Kruskal–Wallis test17,18 was performed to determine
intergroup differences.
Oral toxicity tests Allergenicity tests
Animals for the MCW and MCE oral toxicity tests Adult female albino Hartley guinea-pigs were selected
consisted of two separate groups of 50 each, 6- to assess the allergenic potential of noni leaf. The
week-old Jcl:ICR mice (25 males and 25 females). guinea-pigs were acclimated with a 12 h light/dark
The control group contained the same number and cycle and provided feed and water ad libitum. For
sex ratio. All mice were acclimated for one week at the ethanol extract, 21 animals were divided into four
22 ± 3 ◦ C, 40–60% RH, and 12 h light/dark cycle, groups. The first group of six animals was treated
and were provided feed pellets and water ad libitum. with a sodium chloride vehicle (9 g L−1 ) and Freund’s
Eighteen hours prior to the test, they were provided adjuvant. The second group of three animals was
water only. In the acute oral toxicity test, each of the treated with 1 mg ovalbumin, the positive control,
MCW and MCE extracts was dissolved in water to and Freund’s adjuvant. The next two groups, both
produce 200 mg mL−1 solutions. Mice in the test composed of six animals, were treated with the ethanol
groups were administered the solutions by gavage extract (12.5 g kg−1 ). However, only one of these
at 10 mL kg−1 body weight. A final extract dose last groups was concurrently treated with Freund’s
of 2000 mg kg−1 body weight was provided. Control adjuvant. Sensitization in all animals was induced with
animals received an equivalent dose of purified water. a 1 mL subcutaneous injection of the corresponding
The actual dose in milliliters was determined by animal treatment, 1 per week for 4 weeks. The animals were
weight measured immediately before administration. allowed to rest for 2 weeks and then challenged by
Observations were made 1 and 4 h after administra- oral gavage (10 mL per animal) of the corresponding
tion to record deaths and clinical symptoms resulting treatment, except those in the positive control groups,
from the treatment. Thereafter, observations were which were challenged by intravenous injection of
made once daily through day 14. Each animal was 2 mg ovalbumin. Following the challenge, the animals
weighed prior to the initial administration of the were observed continuously for 30 min for symptoms
extract and control doses, then again on days 3, 7, of acute systemic anaphylaxis: licking and rubbing the
and 14 of the tests. Average weight gain was compared nose, rubbing of fur, labored breathing, sneezing or
between the test and control groups. All animals were coughing, retching, and death. The animals were then
sacrificed on day 14, and gross necropsy was per- observed every 10 min during the following hour, and
formed. Any abnormalities in organs of the digestive finally once per hour for the remaining 3 h.
2584 J Sci Food Agric 87:2583–2588 (2007)
DOI: 10.1002/jsfaSafety tests of noni leaf
In the test of the aqueous extract, or infusion, 10 sample was first extracted with acetone–water (7:3
guinea-pigs were induced to sensitization by injection v/v). The extract was next reacted with Folin–Denis
with 6 mL, 3 days per week, for two consecutive reagent and sodium carbonate for 30 min, after
weeks. Animals were then rested for 37 days and then which absorbance of the sample was measured and
challenged by oral gavage with 550 mg of aqueous compared to standard tannic acid curve prepared by
extract, as a suspension in 11 mL distilled water. A the same method. Phytic acid content was determined
group of five naive control guinea-pigs was similarly by high-performance liquid chromatography on a
challenged. The animals were observed for symptoms macroporous polymer column with an acetonitrile-
of systemic anaphylaxis during challenge and for based mobile phase, following sonic extraction with
120 min afterwards. 0.5 mol L−1 HCl.19
Proteins from noni leaves were extracted with Twenty-two noni leaf samples, each from different
the Plant Total Protein Extraction Kit (Sigma, St sites on 11 islands, were analyzed for oxalic acid,
Louis, MO, USA). Noni leaf protein, lipoxidase, campesterol, stigmasterol, and β-sitosterol content.
and trypsin inhibitor solutions were prepared at Oxalic acid was determined according to AOAC
protein concentrations of 1.91, 2.47, and 1.96 mg Official Method 986.13. Campesterol, stigmasterol,
mL−1 Tris-NaCl buffer, respectively. The total protein and β-sitosterol were determined according AOAC
content of each solution was determined by the Official Method 994.10. The data were evaluated
Bradford method. Simulated gastric fluid was prepared for normality with the Shapiro–Wilk test.20 Basic
(3.2 g L−1 pepsin in 0.05 mol L−1 NaCl, pH adjusted summary statistics (mean, range, and standard
to 2.07 with 0.1 mol L−1 HCl) from porcine gastric deviation) were calculated to evaluate the degree of
mucosa (Sigma). Separate 50 µL aliquots of the inter-island variability.
sample solution were incubated in 200 uL of the
simulated gastric fluid at 37 ◦ C for 0, 15, 30, and 60 s.
Immediately thereafter, each sample was neutralized
RESULTS AND DISCUSSION
with 75 µL of 0.16 mol L−1 Na2 CO3 , then heated at
None of the oral toxicity tests resulted in observable
96 ◦ C for 5 min. A non-allergenic protein, soybean
antinutritive or toxic effects. No deaths or abnormal
lipoxidase, and a known allergen, trypsin inhibitor,
clinical symptoms were noted. Furthermore, no vis-
were used as pepsin-sensitive and resistant reference
ible abnormalities were present in any of the organs
controls, respectively.
examined during gross necropsy. Conversely, all three
All samples were evaluated by sodium dode-
groups of the test and control animals showed signif-
cyl sulfate–polyacrylamide gel electrophoresis (SDS-
icant weight gains/measured time period (Figs 1–3).
PAGE) (10–20% Tris–tricine/peptide gel). Polypep-
There were no treatment-related differences in weight
tide molecular weight standards, as well as biotinylated
gain between the control and treatment groups in any
broad-range molecular weight standards, were run
of the tests.
with each gel, as well as the pepsin solution and
The currently accepted methods for investigating
sample solutions alone. Proteins were visualized by
the potential for Types I (immediate) and IV
Coomassie Brilliant Blue staining.
(delayed) hypersensitivity reactions are not designed
for ingested and chemically complex substances,
Chemical analyses such as foods. These assays involve non-physiological
The tannic acid content of noni leaf was determined routes of exposure, in which digestion is not
spectrophotometrically at 760 nm according to a accounted for. There also seems to be no generally
modified Association of Official Analytical Chemists accepted and standardized methods for assessing
(AOAC) method: Official Method 952.03. The Types II (cytotoxic) and III (immune complex)
45 MCW males
MCE males
43 control males
MCW females
41
MCE females
Mean weight (g)
39 control females
37
35
33
31
29
27
0 3 7 14
Days following adminstration
Figure 1. Mouse weight gain by group and gender in acute oral toxicity test.
J Sci Food Agric 87:2583–2588 (2007) 2585
DOI: 10.1002/jsfaBJ West et al.
40 MCW males
MCE males
control males
MCW females
35 MCE females
Mean weight (g)
control females
30
25
20
0 3 7 14 21 28
Day of test
Figure 2. Mouse weight gain by group and gender in subacute oral toxicity test.
47 MCW males
MCE males
control males
42 MCW females
MCE females
control females
Mean weight (g)
37
32
27
22
0 3 7 14 21 28 42 56 70 84 90
Day of test
Figure 3. Mouse weight gain by group and gender in subchronic oral toxicity test.
hypersensitivities. Further, there is a general lack of proteins in rye and maize, etc. However, ethanol
epidemiological data to determine the sensitization extraction may not include some proteins within
rate for new food sources, and often there is no the higher-molecular-weight range of food allergens
previously sensitized subpopulation from which sera (10–70 kDa).22 Therefore, the aqueous extract, or
may be provided to conduct relevant studies on IgE infusion, was also evaluated to include larger proteins
binding components. Alternative methods, such as while excluding insoluble material.
profiling of recombinant proteins, and comparison In the acute systemic anaphylaxis test of the
against databases containing amino acid sequences of ethanol extract, none of the test group or negative
allergens, are not an economically reasonable approach control group animals demonstrated any symptoms,
for new natural foods, where very large numbers of whereas all animals in the positive control groups
individual proteins may be involved. displayed obvious signs of anaphylaxis. No animals
Another obstacle is that injection of the whole leaf in the aqueous extract, or infusion, acute systemic
to sensitize the animals is inappropriate, as many anaphylaxis test displayed symptoms of anaphylaxis.
insoluble components will not be absorbed, or may These results suggest that noni leaves are unlikely
cause inflammatory reactions that are unrelated to to cause sensitization when used as food. Allergenic
oral sensitization. With these limitations in mind, a proteins in food are typically resistant to digestion by
modified active systemic anaphylaxis test of an ethanol the gastrointestinal tract.23 The property of resistance
extract of noni leaf was performed where intravenous to digestion may be evaluated in vitro and is useful
challenge of the test group was replaced with an oral in estimating the allergenicity of proteins in novel
challenge, as previously done to evaluate noni fruit foods.24,25 Therefore, the resistance of noni leaf
juice.21 This allows the extraction and evaluation proteins to digestion by pepsin was evaluated to
of ethanol-soluble proteins, similar to extraction confirm the conclusion drawn from the guinea-pig
of allergenic wheat gliadin peptides and storage tests.
2586 J Sci Food Agric 87:2583–2588 (2007)
DOI: 10.1002/jsfaSafety tests of noni leaf Figure 4 displays SDS-PAGE results for noni leaf greatly influences leaf morphology. As such, leaf shape proteins alone, and in pepsin solution. Proteins from is inadequate to differentiate genotype and estimate this leaf are very sensitive to the effects of pepsin. The diversity. proteins were digested immediately, at 0 s incubation. As possible diversity may have an impact on the Only pepsin from the simulated gastric fluid remained chemical composition of the leaves, including antinu- after any incubation time. The SDS-PAGE results for trient factors and health-promoting phytochemicals, it the pepsin-resistant and sensitive references revealed is important to assess the content variability of oxalic that the test was effective, with trypsin inhibitor acid, an antinutrient often found in green leafy vegeta- remaining intact at all incubation times, and lipoxidase bles, and the lipophilic phytochemicals campesterol, being degraded rapidly. stigmasterol, and β-sitosterol. Summary statistics for The identification of potential antinutrient sub- the analysis of fresh noni leaves collected from 11 stances is an important aspect of the safety evaluation islands are presented in Table 1. The content ranges of any vegetable food. No known antinutrient com- are narrow, and standard deviations suggest that quan- pounds have been previously identified in noni fruit or tities significantly outside of these ranges are not leaves. Further, oral toxicity tests failed to reveal the expected. Based on these results, noni leaves are fairly presence of any antinutritive effects. However, as other uniform throughout much of French Polynesia, with vegetable foods are known to contain tannic acid, a no substantial inter-island difference. Further, within determination of the quantity in noni leaves is impor- the geographical boundaries of these islands, the over- tant. The frozen noni leaf sample was found to contain all environment is not sufficiently diverse to result in a tannic acid concentration of 1.6 g kg−1 , and an aver- any relevant differences in composition. The oxalic age dried leaf content of 25.8 g kg−1 . These levels are acid content of noni leaves is relatively small when not particularly high, as they are well within the range compared to other commonly consumed green leafy of other well-known vegetable foods.26 – 28 Phytic acid vegetables, and is within the range of values reported was not detectable in the raw leaf (
BJ West et al.
ACKNOWLEDGEMENTS 13 Nakanishi K, Sasaki SI, Kiang AK, Goh J, Kakisawa H,
The modified guinea-pig test protocols were carried Ohashi M, et al, Phytochemical survey of Malaysian plants
preliminary chemical and pharmacological screening. Chem
out under the direction of Lise Bollen at Scantox, Lille Pharm Bull 13:882–890 (1965).
Skensved, Denmark, and George Moore at Product 14 Dunnett CW, A multiple comparison procedure for comparing
Safety Labs, Dayton, New Jersey, USA. Some chemi- several treatments with a control. J Am Statist Assoc
cal analyses were performed at Covance Laboratories, 50:1096–1121 (1955).
Madison, Wisconsin, USA. Leland D White provided 15 Dunnett CW, New tables for multiple comparisons with a
control. Biometrics 20:482–491 (1964).
assistance with manuscript preparation.
16 Bartlett MS, Properties of sufficiency and statistical tests. Proc
R Soc Lond A 160:268–282 (1937).
17 Kruskal WH and Wallis WA, Use of ranks in one-criterion
REFERENCES variance analysis. J Am Statist Assoc 47:583–621 (1952).
1 Morton JF, The ocean-going noni, or Indian mulberry (Morinda 18 Kruskal WH and Wallis WA, Errata for Kruskal–Wallis (1952).
citrifolia, Rubiaceae) and some of its colorful relatives. Econ J Am Statist Assoc 48:907–911 (1953).
Bot 46:241–246 (1992). 19 Lehrfeld J, HPLC separation and quantitation of phytic acid and
2 Nelson SC, Species profiles for Pacific Island agroforestry. some inositol phosphates in foods: problems and solutions. J
[Online]. Available: http//www.traditionaltree.org [3 January Agric Food Chem 42:2726–2731 (1994).
2007]. 20 Shapiro SS and Wilk MB, An analysis of variance test
3 Brown FBH, Flora of southeastern Polynesia. III. Dicotyledons. for normality (complete samples). Biometrika 52:591–611
Bishop Mus Bull 130:1–386 (1935).
(1965).
4 Ochse JJ and van den Brink RCB, Vegetables of the Dutch East
21 West BJ, Jensen CJ, Westendorf J and White LD, A safety
Indies (Edible Tubers, Bulbs, Rhizomes and Spices Included):
review of noni fruit juice. J Food Sci 71:R100–R106 (2006).
Survey of Indigenous and Foreign Plants Serving as Pot-Plants and
22 Lehrer SB, Horner WE and Reese G, Why are some proteins
Side-Dishes. Archipel Drukkerij, Bogor, Java, pp. 630–632
allergenic? Implications for biotechnology. Crit Rev Food Sci
(1931).
Nutr 36:553–564 (1996).
5 Suckcharoen S, Mercury contamination of terrestrial vegetation
23 Astwood JD, Leach JN and Fuchs RL, Stability of food allergens
near a caustic soda factory in Thailand. Bull Environ Contam
to digestion in vitro. Nat Biotechnol 14:1269–1273 (1996).
Toxicol 24:463–466 (1982).
24 Bannon G, Fu TJ, Kimber I and Hinton DM, Protein digestibil-
6 Hall HT, Nagy S and Berry RE, Leaves for food: protein and
ity and relevance to allergenicity. Environ Health Perspect
amino acid contents of leaves from twenty-three tropical and
subtropical plants. Fla State Hortic Soc 88:486–490 (1975). 111:1122–1124 (2003).
7 Yeoh HH and Wee YC, Leaf protein contents and nitrogen-to- 25 Bernstein JA, Bernstein IL, Bucchini L, Goldman LR, Hamil-
protein conversion factors for 90 plant species. Food Chem ton RG, Lehrer S, et al, Clinical and laboratory investigation
49:245–250 (1994). of allergy to genetically modified foods. Environ Health Perspect
8 Aalbersberg WGL, Hussein S, Sotheeswaran S and Parkin- 111:1114–1121 (2003).
son S, Carotenoids in the leaves of Morinda citrifolia. J Herbs 26 Savolainen H, Tannin content of tea and coffee. J Appl Toxicol
Spices Med Plants 2:51–54 (1993). 12:191–192 (1992).
9 Sang S, Cheng X, Zhu N, Stark RW, Badmaev V, Ghai G, et al, 27 Mosha TC, Gaga HE, Pace RD, Laswai HS and Mtebe K,
Flavonol glycosides and novel iridoid glycoside from the leaves Effect of blanching on the content of antinutritional factors
of Morinda citrifolia. J Agric Food Chem 49:4478–4481 (2001). in selected vegetables. Plant Foods Hum Nutr 47:361–367
10 Zin ZM, Hamid AA, Osman A and Saari N, Antioxidative (1995).
activities of chromatographic fractions obtained from root, 28 Yadav SK and Sehgal S, Effect of domestic processing and
fruit and leaf of Mengkudu (Morinda citrifolia L.). Food Chem cooking on selected antinutrient contents of some green leafy
94:169–178 (2006). vegetables. Plant Foods Hum Nutr 58:1–11 (2003).
11 Dignan C, Burlingame B, Kumar S and Aalbersberg W, The 29 McClatchey W, Diversity of uses and growth forms in the
Pacific Islands Food Composition Tables (2nd edn). UN FAO, Morinda citrifolia complex, in Proceedings of 2002 Hawaii Noni
Rome (2004). Conference, ed. by Nelson SC. University of Hawaii, College of
12 Leung WTW, Butrum RR, Chang FH, Rao MN and Polac- Tropical Agriculture and Human Resources, pp. 5–9 (2003).
chi W, Food Composition Table for Use in East Asia. US 30 Haytowitz DB and Matthews RH, Composition of foods:
Department of Health, Education, and Welfare publication vegetables and vegetable products, in USDA Handbook no.
(NIH) 73–465, Bethesda, MD (1972). 8–11. USDA, Washington, DC, p. 11 (1984).
2588 J Sci Food Agric 87:2583–2588 (2007)
DOI: 10.1002/jsfaYou can also read
