Abstract: The prevailing spread of food-poisoning orchestrated by mycotoxin producing Penicillium expansum strains and others on post-harvested crops and food is a preoccupying issue in tropical countries especially Cameroon. The search for antifungal bio-preservatives to curb this health-threatening issue has become adamant. A study was carried out to evaluate the antifungal potential, constituents of Cymbopogon citratus (CC), Ocimum gratissimum (OG) and Thymus vulgaris (TV) Essential oil (EO) fractions, synergism and antagonism between active and non-active fractions against P. expansum predominant strains in Cameroon. The antifungal potency was determined by supplemented broth dilution technique which revealed EO fractions from OG was significantly active against P. expansum strains than those from CC and TV in that sequence at 1000 ppm. Gaseous phase chromatography coupled with mass spectrometry (GC/MS) illustrated fractions rich with oxygenated terpenes exhibited high antifungal potentials than their whole EO counterparts. Conversely, rich hydrocarbon terpenes fractions displayed a far lesser antifungal potency than their whole EO. Inter-blending active subfractions (CC1C/OG1C and CC1C/TV1B) for the three EO at 1000 ppm depicted a stronger antagonistic effect; whereas non active subfractions (CC1A/OG1A and CC1A/TV1A) inter-blends exhibited strong synergistic effects against the two strains of P. expansum. This pathfinder work urges the exploitation of this synergistic effects of active fractions as a probable optimized bio-preservative agent against P. expansum; but recommending the non usage of active fractions blend exhibiting antagonistic effect’s for the control of the latter.
INTRODUCTION
Post-harvest decay of cereals and subsequent food poisoning of recent had become common in Cameroon. One of the catalysing biotic factor to somewhat behind these ills is P. expansum. This phytopathogen is a post-harvest fungus responsible for the deterioration of fruits, cereals and vegetables in general, causing blue-mould-rot disease. This fungus synthesizes roquefortine C, citrinin, chaetoglobosins and the mycotoxin patulin, responsible for Human and animal’s food poisoning (Andersen et al., 2004; Fung and Clark, 2004).
Studies reveals approximately 25% of marketed food commodities are contaminated by mycotoxins and this constitutes a health-threatening issue and economy loss for most of the tropical countries where these contaminated goods are pulled out of the market for security reasons (FAO, 2003). Restrictions imposed by food industries, trade zones and regulatory agencies on the use of some synthetic food additives (Matthew et al., 2005) had led to a retrospection and the eventual quest for alternatives; essentially biodegradable, eco and health-friendly of plant origin. Essential oils have been shown to be active against the above-mention phytopathogenic agents responsible for the deterioration of stored and conserved food crops (Hammer et al., 2003; Nguefack et al., 2007).
Importantly, recent studies shows essential oils of O. gratissimum (Bengyella et al., 2010) C. citratus and T. vulgaris (Nguefack et al., 2009) possess high antifungal potential against an array of phytofungi and are readily available and accessible for local farmers and wholesalers. Moreover, the bottom-line with these plants is the presence of several bioactive compounds including geranial, neral, thymol, terpinen-4-ol, linalool, carvacrol (Nielsen and Rios, 2000). It’s been shown bio-guided fractionation of EO enhances the antimicrobial activity of fractions, rendering them more active than their complete or whole oil (Nguefack et al., 2009).
However, none of these studies had actually determined the constituents of each fraction nor determine the antifungal potential of these fractions or subfractions on P. expansum strains. Interestingly, little is known about the synergism or antagonism between active or non-active fractions of these plants EO. In this work, we evaluated the level of synergism and antagonism between some active and non-active bio-guided fractions of EO obtained from O. gratissimum, C. citratus and T. vulgaris against two strains of P. expansum; then correlates their activities with their chemical composition and their retention indexes.
MATERIALS AND METHODS
Plant materials were harvested at Obala (for O. gratissimum and C. citratus) and Bafoussam (for T. vulgaris) all in Cameroon. All bioassays and chemical analysis were carried out within the time frame of 2007-to-2009. Each specimen was confirmed by the Cameroon National Herbarium in Yaoundé according to the deposited Voucher specimens (N° Dang 18628/SRF/Cam-1968, Letouzey 5817/SRF/Cam-1966 and Westphal 42851/HNC-1978, respectively). Essential oils were obtained from air-dried leaves by hydrodistillation using the Clevenger apparatus as described by Lamaty et al. (1987). Recovered oils were dried over anhydrous sodium sulphate and stored in darkness at 4°C. Fungal strains (MRC 6935 and MRC 6939) of P. expansum were obtained from the Medical Research Council, PROMEC Unit, Cape Town, South Africa and culture on Potato Dextrose Agar (PDA) medium at 25°C, 12/12 h alternate day and night.
Forty milligrams of each essential oil was fractionated in packed 40 g silica gel 60-200 mesh, 30x750 mm chromatographic column under increasing gradient elution of ethylacetate (EA)-hexane (5%: 95%) to 100% Hexane (H) solely. The composition of each fraction was examined on aluminium backed silica gel thin layer chromatography (TLC) plates with hexane/ethyl acetate (3: 1 v/v) as the developing solvent. Spots were observed under UV light and reveal in an iodine vapour chamber and fractions showing similar patterns were pulled together. The antifungal potential were evaluated for complete essential oils and their fractions using the supplemented broth dilution method (Benjilali et al., 1984) accompanied with colony counts. Complete EO or their fractions were blended at different proportions of (100, 75/25, 50/50, 25/75% v/v) making-up 1000 ppm as final concentration; supplementing PDA broth Difco-conidial mixed of 9: 1 v/v in Eppendorfs. Each use conidia suspension was adjusted to approximately 107 conidia/mL/strain using a Bürker-Türk counting chamber. The Eppendorfs mixtures were incubated and vortexed at interval of 1 h for 3 h. Prior to incubation, three 10 fold serial dilution was made and 50 μL aliquots from each dilution were plated on PDA in petri-dishes of 90 mm. After 72 h of incubation under 12/12 h, 24±1°C alternate light and darkness; number of colony counts (N) were performed and expressed as number of Colony Forming Units (CFU) per ml per plate, calculated as follows:
N = NCFUx20xdilution factor |
The fungicidal activity was determined and expressed as number of decimal reduction of colony forming units per ml (NDRCFU) and calculated using the following formula:
NDRCFU = -Log [N+/No] |
N+: No. of colony forming units per ml on supplemented media with EO.
No: No. of colony forming units per ml on non-supplemented media.
Fractions with antifungal potentials superior or equal to the whole EO or which represented a high proportion of the whole EO were pulled for synergism and antagonism assessment. Synergisms between EO fractions were assessed by comparing the NDRCFU obtained with inter-blended fractions to those obtained with individual fractions. Each blend activity was qualified as synergistic or antagonistic when the NDRCFU obtained were respectively superior or inferior to the arithmetic sum of NDRCFU obtained with individualized fraction at the same concentration of 1000 ppm (Pandey et al., 1983).
The samples were characterized using gas chromatography coupled to mass spectrometry (GC/MS) using Hewlett-Packard GC 6890A equipped with a HP-5MS (cross-linked methyl siloxane) fused column (30 m x0.25 mM, film thickness 0.25 μm) and interfaced with a quadrupole detector (Model 5973). The apparatus was programmed as follows: Injector temperature 220°C, transfer line temperature at 280°C, carrier gas helium flow rate at mL min-1, 70 ev ionization voltage and 1400 ev electron multiplier for mass ranging from 33-500. Comparing their retention time indices and their mass spectrum with standard markers identified compounds. Collected data were analyzed for variance using the One Way ANOVA package and the parametric student-t-test-Newman-Keuls at 95% confidence level.
RESULTS
From the three plants, three whole essential oils, twenty-six fractions were obtained in total and their antifungal potential evaluated. Some physical characteristics, extraction parameters, yields, activities of these oils and their individual fractions are equally displayed in Table 1.
Evaluating antifungal potential of C. citratus at 1000 ppm reveals maximum activity was exhibited by subfraction CC1C with NDRCFU values of 3.34 and 5.02 against MRC6935 and MRC6939, respectively. C. citratus strongest intra-blend antifungal potential was observed with CC1A/CC1C (25:75 v/v) with NDRCFU at 0.81 and 1.07 against MRC6935 and MRC6939, respectively.
| Table 1: | Whole and fractions of EO characteristics and activity against P. expansum strains |
| Data are the Mean±SD of 3 repetitions with p<0.05. Fractions of C. citratus (CC) are: C1A, CC1B, CC1C, CC2 to CC8. Those of O. gratissimum (OG) are OG1A, OG1B, OG1C, OG2 to OG8. Fractions of T. vulgaris (TV): TV1A, TV1B, TV2 to TV8. Subscript 1A, 1B, 1C attached to OG, CC and TV represents subfraction OG1, CC1 and TV1 for each plant EO fractionated with 100% hexane which showed different profiles on TLC plates. Fractionating solvent are hexane (H), water and ethylacetate (EA). The remaining fractions were designated in function of the proportional mixed of their fractioning solvent as displayed on the table. Different superscipts within coloumn show significant differeance at p<0.05 | |
However, an impressive activity was observed with a trio intra-blend of CC1B/CC1C/CC2 (1:1:1 v/v/v) with NDRCFU above unity as shown in Table 2.
Profiling antifungal potential of O. gratissimum at 1000 ppm unveiled OG1C was the most active with NDRCFU values above 6.00 for both pathogenic strains. A maximum antifungal effect was noted with intra-blends of OG1B/OG2 (25:75 v/v) and OG1C/OG2 (75:25 v/v) exhibiting an NDRCFU value above 6.00 against both strains as depicted in Table 3.
| Table 2: | Activities of blended fractions of C. citratus (CC) EO at 1000 ppm against P. expansum strains |
| Data are the Mean±SD of 3 repetitions with p<0.05. CC is the whole or non fractionated EO of C. citratus (CC). Subfractions CC1A, CC1B and CC1C represent C. citratus (CC) EO fractionated with 100% hexane which showed different profiles on TLC plates. CC2 fraction is obtained with H/EA 95/5%v/v fractionating solvent. Different superscipts within coloumn show significant differeance at p<0.05 | |
Assaying antifungal potency at 1000 ppm for T. vulgaris revealed best individualized activity was associated with fraction TV1B with NDRCFU valuing 6.00 for both pathogenic strains. Strongest intra-blend antifungal activity was observed with TV1A/TV1B (25:75 v/v) with NDRCFU of 0.6 and 0.74 for strains MRC6935 and MRC6939, respectively as depicted in Table 4.
Inter-blending C. citratus and T. vulgaris fractions revealed a strong synergistic effect with active fractions CC1A/TV1A (50:50 v/v) with NDRCFU values of 0.58 and 0.79 for strain MRC6935 and MRC6939, respectively. Besides this synergism, a very strong antagonism was observed with inter-blends of CC1C/TV1B (75:25 v/v); CC1C/TV1B (50:50 v/v) and CC1C/TV1B (25:75 v/v) as depicted in Table 5.
Best synergistic effect from inter-blending active fractions of C. citratus and O. gratissimum was noted with CC1A/OG1A (50:50 v/v) with NDRCFU values of 0.81 and 0.68 for strains MRC6935 and MRC6939, respectively. Equally, all inter-blending of of CC and OG were generally more active than their individual fractions. Generally, inter-blending C. citratus and O. gratissimum generally expressed impressive synergism than those of C. citratus and T. vulgaris as depicted in Table 6. Antagonism with active fractions was observed inter-blending CC1C/OG1C (75:25 v/v) and CC1C/OG1C (50:50 v/v); hence, enhancing P. expansum viability to somewhat compared to their individual fractions as depicted in Table 6.
| Table 3: | Activities of mixed fractions of O. gratissimum (OG) EO at 1000 ppm against P. expansum strains |
| Data are the Mean±SD of 3 repetitions with p<0.05. OG is the whole EO, while OG2 was obtained with fractionating solvent H/EA 95/5%v/v. Subfractions OG1A, OG1B and OG1C were obtained with 100% hexane and exhibited different profiles on TLC plates. Different superscipts within coloumn show significant differeance at p<0.05 | |
| Table 4: | Activities of mixed fractions of T. vulgaris (TV) EO at 1000 ppm against P. expansum strains |
| Data are the Mean±SD of 3 repetitions with p<0.05. TV is the whole EO, while TV2 was obtained with fractionating solvent H/EA 95/5%v/v. Subfractions TV1A and TV1B were obtained with 100% hexane and exhibited different profiles on TLC plates. Different superscipts within coloumn show significant differeance at p<0.05 | |
| Table 5: | Activities EO blended Fractions of T. vulgaris (TV) and C. citratus (CC) at 1000 ppm against P. expansum strains |
| bData are the Mean±SD of 3 repetitions with p<0.05. Subfractions CC1A, CC1C, TV1A and TV1B were obtained with 100% hexane and exhibited different profiles on TLC plates, while TV2 was obtained with fractionating solvent H/EA 95/5%v/v. Different superscipts within coloumn show significant differeance at p<0.05 | |
| Table 6: | Activities EO blended Fractions of C. citratus (CC) and O. gratissimum (OG) at 1000 ppm against P. expansum strains |
| Data are the Mean±SD of 3 repetitions with p<0.05. Subfractions of C. citratus (CC1A, CC1C) and subfractions of O. gratissimum (OG1A, OG1B, OG1C) were obtained with 100% hexane showing distinctive profiles on TLC plates. OG2 is a fraction of O. gratissimum obtained with 95/5% v/v H/EA. Different superscipts within coloumn show significant differeance at p<0.05 | |
GC-MS showed whole EO from CC was mainly oxygenated monoterpenes (OMT) (64.03 %), among which 58.94% was citrals. Very active subfraction CC1C (100% hexane fractionation) possessed 45.16% citrals.
| Table 7: | Chemical composition (%) of the EO from C. citratus (CC) and some of its fractions (CC1A, CC1B, CC1C, CC2) |
| Monoterpene Hydrocarbons (MTH); Oxygenated Monoterpenes (OMT); Sesquiterpene Hydrocarbons (STH); Oxygenated sesquiterpens (OST); Retention Index (RI); Whole EO of C. citratus (CC); Subfractions CC1A, CC1B and CC1C obtained with 100% hexane exhibiting different profiles on TLC plates; CC2 fraction was obtained at 95/5% v/v H/EA fractionating solvent | |
Subfraction CC1A was remarkable rich with 86.93% STH, while subfraction CC1B and fraction CC2 had 67.16% and 75.21% of OMT, respectively as depicted in Table 7.
GC-MS revealed essential oil from OG contain approximately 41.79% MTH and 44.61% OMT, predominantly thymol (40.61%) and p-Cymene (23.47%) displayed in Table 8. Active subfraction OG1C and fraction OG2 were 96.46 and 79.13% thymol rich, respectively. Complete EO T. vulgaris was predominantly made-up of 30.87 and 28.11% of p-cymene and thymol, respectively as active molecules as depicted in Table 8. Active TV1B subfraction contained approximately 68.2% thymol, while less active sub-fraction TV1A had approximately 71.11% p-Cymene.
| Table 8: | Chemical composition (%) of O. gratissimum (OG)and T. vulgaris (TV) EO’s and some of their fractions (OG1A, OG1B, OG1C, OG2, TV1A, TV1B, TV2) |
| Fractions of Ocimum gratissimum (OG) essential oil are OG1A, OG1B and OG1C obtained with 100% hexane showing distinctive profiles on TLC plates, while OG2 fraction was obtained at 95/5% v/v H/EA fractionating solvent. Fractions of Thymus vulgaris (TV) essential oil are TV1A and TV1B obtained with 100% hexane showing distinctive profile on TLC plates, while TV2 fraction was obtained at 95/5% v/v H/EA fractionating solvent. Monoterpene Hydrocarbons (MTH); Oxygenated Monoterpenes (OMT); Sesquiterpene Hydrocarbons (STH); Oxygenated sesquiterpens (OST); Aromatic Compound (AC), 1-methyl-4(1-methylethenyl) benzene (MMB). Retention Index (RI) | |
DISCUSSION
The fungicidal activity displayed by C. citratus, O. gratissimum and T. vulgaris oils can be assigned to their high content in OMT 64.03, 44.61 and 43.02%, respectively as reported by Nguefack et al. (2007), this correlated with our findings. The low antifungal activities against the two strains of P. expansum observed with subfractions CC1A, OG1A and TV1A rich in hydrocarbon terpenes compared to their complete EO affirmed OMT in these plants largely contributes for their antifungal potency. This observation further verifies the truism that oxygenated terpenes are overwhelmingly antimicrobial than their hydrocarbon counterparts as proposed by Hammer et al. (2003) and Dorman and Deans (2000).
It has been suggested that, borneol and bornyl-acetate lower antimicrobial activities of EO (Chalchat et al., 1987). This fall in line with our observation and can possibly explain the low antifungal activity of fractions and some subfractions such as CC2, OG2, TV2 and CC1B, OG1B, TV1B, respectively containing the latter. Moreover, low antifungal activity of subfraction OG1B (49.36% thymol; 5.08% carvacrol) compared to subfraction OG1C (92.01% thymol and 0.86% carvacrol) could also be explained by the net decrease in thymol content due to its high isomer-carvacrol content and possibility of racemisation.
Overall, O. gratissimum whole EO exhibited the highest antifungal activity in NDRCFU comparatively to complete EO of C. citratus and T. vulgaris; probably due to the high content of thymol (40.61% in OG, 28.11% in TV and absent in CC). The high antimicrobial activity of thymol had been proposed to be associated to the electron delocalisation system and the hydroxyl group in its structure (Ultee et al., 2002); this possibly explains observed high antifungal activity of O. gratissimum. It’s been proposed structural configuration, functional groups, antagonistic and synergistic interactions between components of oils determines the intrinsic antimicrobial potential of EO (Dorman and Deans, 2000; Delaquis et al., 2002) supporting these factors harmonized well in O. gratissimum than in T. vulgaris and C.citratus.
It’s been shown thymol, citrals, carvacrol and p-cymene play an important role in membrane swelling, causing membrane permeability (Dorman and Dean, 2000; Ultee et al., 2002), but with p-Cymene playing a more dominant role. This permeabilizing activity p-cymene probably enables the influx carvacrol into the cell so that a synergistic effect is achieved when the two are present. This supportively, explains the synergism observed with fractions blend of CC1A/CC1B and OG1A/OG1B due to the combined effect of the different components brought into the cocktail. However, synergism varied with the proportion of fractions blended and was important at 50/50% v/v compared to 25/75 and 75/25% v/v. This may imply 50/50% v/v blend contains sufficient quantity of p-cymene required to puncture P. expansum membrane. Hence, facilitating trans-membrane transportation of citrals, thymol and carvacrol into the cytosol. The synergistic effect observed with non-active fractions blend CC1A/OG1A and CC1A/TV1A could be explain by the increase concentration of oxygenated terpenes present in traces in each of the individual fractions.
The antagonistic effect observed with blended fractions CC1B/CC1C, CC1B/CC2, CC1C/CC2, OG1B/OG1C, OG1B/OG2, OG1C/OG2 and TV1B /TV2 could be related to the presence of borneol or bornyl-acetate; since they lower the activity of their individual fraction and the cocktail. Proportionate reduction of fractions containing the latter substantially increases the antifungal activity of the mixtures, affirming they are antagonist enhancers. Furthermore, fraction rich blend of borneol exhibited the highest antagonistic effect. Amazingly, blends of active fractions CC1C/OG1C and CC1C/TV1B revealed strong antagonism, probably due to neutralizing interactive effects between the active components of the fractions. Ultee et al. (2000) showed antimicrobial activity of carvacrol methyl ether was lower than carvacrol, due to blockage of the hydroxyl group with a methyl group. This implies such neutralizing interactive effect may block active groups of fractional components and enhance antagonism as observed with the latter blends. Reduction in oxygenated terpenes content or increases in borneol and it derivatives such as in fractions blend of CC1A/CC1C, OG1A/OG1C, OG1A/OG2, TV1A/TV1B and TV1A/TV2 possibly explain their antagonism.
CONCLUSION
These results pave way for the possible use of essential oils of these plants as food preservatives; but require further assessment of the pH variance on the synergistic and antagonistic effects. Moreover, this study reveals very small and precise amount of essential oils can be used to produce a differentia antimicrobial effect with a probable minimal alteration in food taste. A comparative assessment of the degree of synergy of these essential oils on other phytopathogens and the ultimate evaluation of their toxicity to Humans and animals, can lead to their formulation as bio-preservatives against mycotoxins biotic agents.
ACKNOWLEDGMENTS
We thank the Chinese Academic of Science (CAS), The Third World Academic of Science (TWAS) and DBT (Department of Biotechnology-Government of India) for scholarship support.