The occurrence of toxigenic mycoflora and mycotoxins in medicinal and herbal
plants has been confirmed in several studies (Aziz, 1998;
Halt, 1998; Abou-Arab, 1999;
Efuntoye, 1999; Costa and Scussel,
2002; Elshafie et al., 2002). Potential contamination
of these plants with mycotoxins poses concerns over their use. Aspergillus
and Penicillium are two major reported genera that produce mycotoxins
(Rodriguez-Amaya and Sabino, 2002). Several environmental
factors are reported to influence mycotoxin production, but temperature and
relative humidity are considered to be the most critical (Simsek
et al., 2002). Since, the mycotoxins are secreted and can contaminate
products of vegetal origin consumed by livestock such as cattle, they therefore
contaminate human secondarily.
The effects of these mycotoxins especially aflatoxins and ochratoxin on various
organs has been reported in some studies. Ochratoxin A induced liver injury
involves a reduction in the ability to counterbalance oxidative stress leading
to altered gap junction intercellular communication and loss of cell adhesion
and polarity. The mild oxidative damage in combination with other cytotoxic
effects trigger the promotion of liver tumors after exposure to ochratoxin A
(Gagliano et al., 2006). The cancer-producing
capability of the aflatoxins has been demonstrated in laboratory animals and
is suggested from epidemiologic data to occur in man. Aflatoxin B1
decreases serum proteins, which are sensitive early indicators of hepatotoxicity
(Quezada et al., 2000). The most important mycotoxins
are aflatoxins which are hepatotoxic, hepatocarcinogenic and ochratoxins, which
are nephrotoxic and nephrocarcinogenic (Rastogi et al.,
2001; Orsi et al., 2007). This study was
conducted to evaluate the predominant mycoflora, the extent of fungal contamination
in herbal plants, the ability of fungi isolated to produce mycotoxins and the
effects of these mycotoxins on the liver and kidneys of test mice. Female albino
mice were used since most herbs are used by women after delivery in Saudi Arabia.
MATERIALS AND METHODS
Sampling: Twenty five samples of well-known plant herbs from random
herb markets in the city of Riyadh, Saudi Arabia were collected between February
and March 2008 to assess the predominant mycoflora and the extent of fungal
contamination. The herbal products were chosen on the basis of their commercial
availability and popularity of use (Table 1). Every sample
weighed 3 kg and was preserved in clean plastic bags at a temperature ranging
from 4 to 5°C.
Isolation and evaluation of fungal contamination: The fungal flora of the samples was detected by using the dilution plate method. Two types of media were used, (1) Glucose-Czapeck’s agar medium in which glucose (10 gm L-1) replaced sucrose and potato dextrose agar medium, chloramphenicol (20 μg mL-1) and Rose Bengal (30 ppm) were added to suppress bacterial growth, (2) Sabouraud’s agar and (3) Rose Bengal. Every sample was examined for Total Fungi Count (TFC). Five grams of each sample were mixed with 45 mL of distilled water from which tenfold serial dilution were made. Three milliliter
from each dilution was inoculated in sterile petri dishes and mixed well. Plates were incubated upside down at 26±2°C for 7 days. After incubation, the fungal colonies were counted, recorded and the colony-forming units (cfu) per gram were calculated. The fungal isolates were stored at 25°C. Identification was performed by cultural and morphological characteristics and followed taxonomic schemes of Raper and Fennel for the genus Aspergillus and Pitt for the genus Penicillium.
Extraction and evaluation of toxigenic potential of fungal isolates:
Aflatoxin was extracted using SMKY (Sucrose: Magnesium sulphate: Potassium nitrate:
Yeast extract) medium (Bugno et al., 2006). Twenty
five grams of each raw sample was shaken with 50 mL of chloroform in 250 mL
flask for 24 h. The defatted residue was re-extracted for another 24 h in a
shaker with 50 mL chloroform. Chloroform extracts were combined, washed with
an equal volume of distilled water, dried over anhydrous sodium sulfate, filtered,
then concentrated and left to dry. The dried materials were transferred to vials
with small amount of chloroform which was evaporated to near dryness. The analysis
of extract for the detection of different aflatoxins was performed using the
HELICA total aflatoxin assay (Hellica Biosystems, Fullerton, California, USA).
An aflatoxin specific antibody optimized to cross react with all four subtypes
of aflatoxin is coated to a from microwell.
Toxins are extracted from a ground sample with 70% methanol. The extracted
sample and HRP-conjugated aflatoxin B1 are mixed and added to the
antibody-coated microwell. Aflatoxin from the extracted sample and HRP-conjugated
aflatoxin B1 compete to bind with the antibody coated to the microwell.
Microwell contents are decanted and non-specific reactants are removed by washing.
An enzyme substrate (TMB) is added and color (blue) develops. The intensity
of the color is directly proportional to the amount of bound conjugate and inversely
proportional to the concentration of aflatoxin in the sample or standard. Therefore,
as the concentration of aflatoxin in the sample or standard increases, the intensity
of the blue color will decrease. An acidic stop solution is added which changes
the chromogen color from blue to yellow. The microwells are measured optically
by a microplate reader with an absorbance filter of 450 nm (OD450).
The optical densities of the samples are compared to the OD’s of the kit standards
and an interpretative result is determined. Ochratoxin A was extracted by acetonitrile
4% Kclaq (9-1), separated by migration in toluene-ethyl acetate- formic acid
(5:4:1, v/v/v) and quantified by fluorimetry at 333 nm (Boudra
et al., 1995). Mycotoxins were identified by comparison with appropriate
standards for aflatoxin and ochratoxin A. (Supelco, USA).
Evaluation of mycotoxins on test mice: One hundred and thirty adult female albino mice weighing between 130-160 g were procured from the Pharmacy College of King Saud University in Riyadh, Saudi Arabia. The animals were housed in well ventilated aluminum cages with a temperature of 18 to 25°C at the animal lab of King’s Khalid University Hospital, Riyadh, Saudi Arabia. All mice were maintained on a standard pellet diet (mouse breeding diet, Pillsbury Ltd., Birmingham, UK) and tap water. Overall nutrient composition of the diet was 36.2% carbohydrate, 20.9% protein, 4.4% fat, 38.5% fiber with added vitamins and minerals and with a metabolizable energy content of 1.18 MJ/100 g. A water extract was made using 1.5 per 1 mL of the herbs as the solute and sterile water as the solvent. The solution was mixed well and was placed in sterile glass jars with a firmly closed lid.
The extract was left for a couple of hours to cool down and settle before it
was filtered. Fifty milliliters were taken from every filtered water extract
and was given orally to female albino mice instead of water (Bokhari
et al., 2007).
The liquid cultures of fungi that were isolated and found to produce mycotoxins
were filtered and were placed in sterile glass jars with a firmly closed lid.
Portions of 50 mL were given to the test mice instead of water (Waggas,
Test mice were segregated into three groups, the herbal extract group, the fungal extract group and the control group. The herbal extract group was given aqueous herb extracts for five weeks through drink bottles. After five weeks, mice were fasted for 12 h then placed under anesthesia using diethyl ether before blood was extracted. Blood was with drawn from the heart by direct cardiac heart puncture and collected in sterile blood tubes (Terumo Europe N.V.3001 Leuven, Belgium) for biochemical analysis including alanine aminotransferase (ALT), aspartate aminotransferase (AST), gamma glutamyltransferase (GGT), serum creatinine and urea. Biochemical analysis of blood was performed by Dimension ® Clinical Chemistry System Flex ®reagent Cartridge (Dade Behring Limited, United Kingdom).
The second group, fungal extract group was given a liquid medium in which fungi has been cultured and filtered. The same protocol used for the herbal extract group was administered for the test mice and the control group. The control group of mice was given normal drinking water for the entire 5 weeks and was also maintained on standard pellet diet.
Statistical analysis: Laboratory values for creatinine, urea, ALT, AST and GGT were expressed as Mean±SD. Student t-test was used to compare differences between treated and control group and in between the two variables. A p-value<0.05 was considered statistically significant.
Fungal contaminants and their frequency of distribution in herbal plants:
Table 2 presents the frequency distribution of total fungal
counts in the 25 herbal plants analyzed. The results shown in Table
3 revealed that 39.1% of the samples exceeded the limit determined by the
US Pharmacopoeia. Calligonum comosum (2x105 cfu g-1),
grained mixed herbs (24x103 cfu g-1) and Salvia officinalis
(23x103 cfu g-1) were the most contaminated samples. The
predominant mycoflora obtained was distributed in 4 genera. The genus Aspergillus
was the most dominant (142 isolates) followed by Penicillium (14 isolates)
and these two genera were found in 85.0 and 11.0% of the samples analyzed. Aspergillus
fumigatus, Aspergillus flavus and Aspergillus ochraceus were
the most dominant and frequently isolated (47.3, 46.5 and 18.1%, respectively),
followed by Aspergillus citrinum (11.0%).
Effects of the fungal contaminants in herbal plants on test mice: In
the biochemical assessment on test mice fed with specific herbal plant, the
level of urea was significantly high with Ducrosia ismaelis (9.6±1.05
mmol L-1 versus the control group, 6.7±1.03 mmol L-1,
p = 0.0041).
|| Frequency distribution of total fungal counts in 25 herbal
||Distribution of herbal plant samples according to the counts
of fungi (n = 23)*
|*No fungi were isolated from 2 herbal plants.
Several other herbs increased the urea levels on test mice including Carum
carvi (8.4±1.27 mmol L-1), Pimpinella anisum (8.02±1.6
mmol L-1), Cuminum cyminum (7.62±.0 mmol L-1),
Thymus serpyllum (7.53±1.2 mmol L-1), Foeniculum vulgare
(7.5±1.2 mmol L-1), Astralagus sarcocolla (7.44±1.8
mmol L-1), Aloe vera (7.4±0.9 mmol L-1),
Lepidium sativum (7.36±1.5 mmol L-1), Ammi visnaga
(7.34±1.0 mmol L-1), Zingiber officinale (7.2±1.5
mmol L-1), Salvia officinalis (7.13±1.2 mmol L-1),
Nigella sativa (6.98±1.1 mmol L-1) and Achilea fragrantissima
(6.94±1.3 mmol L-1). Mice fed with the other test herbs had
urea levels below the control group with significantly lower levels in Trigonella
foenum (5.26±0.5 mmol L-1, p = 0.0270) and Anethum
graveolens (4.7±0.9 mmol L-1, p = 0.0163).
The mean creatinine level in the control group was 30±11.7 μmol L-1. Twenty one of the tested herbs caused the elevation of the creatinine levels, the extract of D. ismaelis posted the highest significant mean creatinine level of 51.6±4.5 μmol L-1, p = 0.0072, followed by A. vera extract (47.2±10.8 μmol L-1, p = 0.0452) and C. cyminum (46.4±7.0 μmol L-1, p = 0.0307). All extracts from other tested herbs induced higher creatinine levels compared to the control group except extracts of Z. officinale (29.4±11.4 μmol L-1, p = 0.9366), N. sativa (26.2±3.3 mmol L-1, p = 0.5051) and Cinnamum zeylanicum (25.2±10.5 mmol L-1) which resulted in creatinine levels lower than those of the control group.
The mean AST level of the control group was 536±263 U L-1. Test mice fed with grained mixed herbs posted the highest AST level with 1238.8±835 U L-1, followed by mixed herbs (889.6±357 U L-1, p = 0.1152), A. sarcocolla (834.6±525 L-1, p = 0.2895), Arabica coffea (756.6±573 U L-1, p = 0.4575), Vigna adiata (654.4±406 U L-1, p = 0.5997), T. serpyllum (634.5±424 U L-1, p = 0.6711), Z. officinale (643±227 U L-1, p = 0.5243), C. zeylanicum (567.8±244 U L-1, p = 0.8481), Commiphora momol (538.6±225 U L-1, p = 0.9870) and Artemisia herba alba (462.4±202 U L-1, p = 0.6340). All other test herbs resulted in insignificantly lower levels of AST as compared to the control group.
The mean ALT level of the control group was 306.2±345 U L-1. Test mice fed with grained mixed herbs posted the highest ALT level with 740.8±987 U L-1, followed by L. sativum (505.6±542 U L-1, p = 0.5078), T. serpyllum (399.8±309 U L-1, p = 0.7532), A. sarcocolla
(380.4±577 U L-1, p = 0.8116) and C. arabica (338.6±535
U L-1, p = 0.9123). All other test herbs resulted in ALT levels
lower than those of the control group with C. comosum, the lowest value
recorded was 70.2±15 U L-1 (p = 0.1668).
||The four isolated fungal species and their mycotoxin content
||Biochemical analysis of mycotoxin effect in 130 albino mice*
|Values are expressed as Mean±SD (range)
The mean GGT level of the control group was 13.6±2.4 mmol L-1.
Test mice fed with Z. officinale posted the highest GGT level with 15.8±6.2
mmol L-1 (p = 0.4821) followed by mixed herbs with 13.8±3.6
mmol L-1 (p = 0.9203). All other test herbs posted GGT levels insignificantly
lower than that of the control group.
Mycotoxin production by isolated fungal strains and their effects on the
liver and kidneys of test mice: The four strains of Aspergillus and
Penicillium isolated were evaluated for their ability to produce aflatoxins
B1 and B2 and ochratoxin A. A. flavus revealed
production of 3.5 μg kg-1 of aflatoxin B2 and 3.8
μg kg-1 of ochratoxin A. A. flavus produced 7.45 μg
kg-1 of aflatoxin B1 whereas A. ochraceus produced
21.7 μg kg-1 of aflatoxin B2 and 7.25 μg kg-1
of ochratoxin A (Table 4).
Biochemical analysis of mycotoxin effects in 130 albino mice showed that mean
creatinine was significantly higher in the mice group that was fed with herbal
extract than the control group (38.2±7.8 μmol L-1 vs.
30.0±11.7 μmol L-1, p = 0.0264). Urea was also higher
in the herbal extract fed group than the control group (7.8±5.3 mmol
L-1 vs. 6.7±1.0 mmol L-1, p = 0.6154). Similarly,
the AST, ALT and the GGT values in the group fed with herbal extracts was higher
than that of the control group (p = 0.9676, p = 0.9627 and p = 0.2011, respectively).
In the group of mice fed with fungal extract, creatinine, AST, ALT and GGT values
were also higher than the control group of mice fed with water only (p = 0.7802,
p = 0.8032, p = 0.9424 and p = 0.3418, respectively). Among the three groups,
the group of mice fed with fungal extract had the highest values for AST and
ALT whereas, the group that was fed with herbal extract posted the highest urea,
creatinine and GGT values (Table 5).
In the study, Aspergillus sp. were the predominant fungi recovered and as such the major toxigenic species.
Present findings support the findings of Bugno et al.
(2006). Furthermore, in this attempt to determine the mycotoxin content
of fungal isolates, A. fumigatus, A. flavus and A. ochraceus
were aflatoxigenic. Interestingly, present study showed the presence of fungi
in 23 of 25 herbs analyzed, of which 31.9% of samples were above the enumeration
limits set by the US Pharmacopoeia. A. fumigatus predominated in present
study. Approximately, 39.1% of the samples showed enumeration limits of more
than 2x102, the enumeration limit for total fungal count as set by
the US Pharmacopoeia. Considerably though, the high percentage of mycotoxin
production as revealed by present results indicated the inherent capacity of
these moulds to instigate deleterious effects on humans when consumed. Aflatoxin
B1 was high at 7.45 μg kg-1 extracted from A. flavus.
Moreover, Aflatoxin B2 was even higher at 21.7 μg kg-1
from A. ochraceus and 3.5 μg kg-1 from A. fumigatus.
Ochratoxin A was also extracted from A. fumigatus (3.8 μg kg-1)
and A. ochraceus (7.25 μg kg-1). These values are higher
than levels reported by Elshafie et al. (2002)
and Halt (1998).
When the extracts of these herbs and even the mycotoxin extracts were tested on test mice, their hepatotoxicity and nephrotoxicity was evident in the results. Fourteen of this tested herbs showed nephrotoxicity with elevations of urea beyond the control group. Furthermore, the creatinine levels were consistently high in almost all of our tested herbs except for Z. officinalis (ginger), C. ceylanicum (cinnamon) and V. radiata (Mung beans), which are understandably being used in most of oriental and Middle Eastern cuisine.
These high urea and creatinine levels explain the nephrotoxicity of the mycotoxins
present in these herbs which can be potentially dangerous since they alter kidney
function and/or liver function and are teratogenic (Berndt
et al., 1980). The differences in the biochemical test results between
test mice and control groups showed that aflatoxin caused hepatic and renal
tissue degeneration, although present results did not reach significant levels.
This may indicate that there is an interference with normal development and
arrest of differentiation in the various internal organs. Furthermore, the high
percentage of aflatoxins (B1 and B2) and ochratoxin A
in this study indicate the degree of contamination of these herbs. Use and consumption
of these potentially harmful herbs especially in Saudi Arabia where most of
these herbs are being used by women after delivery, can be deleterious and hazardous
to health. Information campaigns on the potential harm of the herbs and their
safe use can benefit the majority.
This study confirms earlier studies demonstrating the predominance of Aspergillus species as contaminants in herbal and medicinal plants. Furthermore, the capability of this fungal species in the production of aflatoxin, promoted by appropriate relative humidity and temperature induces nephrotoxicity and hepatoxicity in animals and even in humans.