In vitro antifungal effi cacy of selected essential oils in controlling fungi associated with the stem-end rot disease of mango (cv. Karutha Colomban) fruits and characterisation of antifungal components

This article is published under the Creative Commons CC-BY-ND License (http://creativecommons.org/licenses/by-nd/4.0/). This license permits use, distribution and reproduction, commercial and non-commercial, provided that the original work is properly cited and is not changed in anyway. Abstract: Karutha Colomban is one of the most delightful mango varieties popular among Sri Lankan consumers. A signifi cant postharvest loss of mango takes place every season due to diseases including stem-end rot (SER), which is caused by a group of endophytic fungal pathogens. In this research, in vitro antifungal effi cacy of diff erent concentrations of essential oils of basil, clove, and cinnamon were evaluated for their ability to control SER causing fungal pathogens of mango (cv. Karutha Colomban) as bio-safe alternatives to conventional fungicides, by conducting liquid and disc volatilisation bioassays. Major bioactive compounds of the selected essential oils were identifi ed by gas chromatography-mass spectroscopy (GC-MS). Basil and cinnamon bark oils (0.20 – 0.30 μL/mL) in liquid bioassay showed high effi cacy against Lasiodiplodia theobromae, while basil and cinnamon leaf oils (0.40 – 0.60 μL/mL) successfully inhibited Pestalotiopsis sp. Cinnamon bark oil (0.60 μL/mL) was identifi ed as the most eff ective oil against Phomopsis sp. According to disc volatilisation bioassay, vapour of cinnamon oils (0.20 – 0.40 μL/mL) was the most eff ective in controlling L. theobromae. Pestalotiopsis sp. was effi ciently controlled by clove and cinnamon bark oil (0.20 – 0.60 μL/mL) vapour. In vapour phase, clove and cinnamon oils (0.40 μL/mL) were the most eff ective against Phomopsis sp. According to GC-MS characterisation, methyl chavicol was the most abundant antifungal component in basil oil while it was (E)-cinnamaldehyde in cinnamon bark oil. Moreover, eugenol displayed the highest abundance in clove and cinnamon leaf oils. Based on in vitro studies, it could be concluded that cinnamon bark oil in liquid and vapour phases demonstrated a higher antifungal effi cacy among the tested essential oils in controlling fungal pathogens causing SER of mango.


INTRODUCTION
Mango (Mangifera indica L.) is a popular fruit worldwide and it has become one of the most desirable fruits in international trade because of its delightful taste and high caloric value (Diedhiou et al., 2007). It bears important nutrients such as vitamins, minerals, polyphenolic and fl avonoid antioxidant compounds as well as prebiotic dietary fi ber (Kothalawala & Jayasinghe, 2017). Asia is recognised as the highest mango producing region, with a record 74.4 % of global production. Present extent under mango cultivation in Sri Lanka is about 3.1 million hectares (FAOSTAT, 2016). 'Karutha Colomban' is one of the most popular cultivars of mango among growers and consumers in Sri Lanka due to its orange fl esh and delightful taste (Kothalawala & Jayasinghe, 2017).
However, the availability of high quality mango for the local and international consumers has been limited by its highly perishable nature and susceptibility to postharvest diseases, which lead to severe postharvest lossess (Bally et al., 2009). Stem-end rot (SER) is considered as one of the most serious and frequently occurring postharvest and post-packaging diseases of mango (Tripathi & Shukla, 2009)
At present, the disease control strategies include prophylactic use of synthetic fungicides to treat fruit crops including mango. Although fungicide usage is a successful chemical control strategy for postharvest diseases of many fruits, it has given rise to many controversial issues owing to their carcinogenicity, acute toxicity, long degradation periods, environmental pollution and contribution to the development of resistant pathogens. Due to these environmentally unsound eff ects, researchers are focusing on fi nding eff ective alternatives to synthetic fungicides in controlling postharvest diseases of fruits (Tripathi & Shukla, 2009). Essential oils extracted from plants are generally assumed to be more acceptable and less hazardous than synthetic compounds as they possess bioactive properties, due to the presence of many chemical components such as eugenol, camphor, camphene, caryophyllene, phellandrene, etc. (Bakkali et al., 2008;Zaker, 2016). In several previous studies, it has been recorded that essential oils of cinnamon (Cinnamomum zeylanicum) leaf and bark, clove (Syzygium aromaticum), and basil (Ocimum basilicum) as very eff ective sources of antifungal compounds (Ranasinghe, 2003;Sukatta et al., 2008;Abeywickrama et al., 2009;Costa et al., 2015).
An in vitro liquid bioassay conducted by Anthony et al. (2004) showed that the essential oils of Cymbopogon nardus and O. basilicum have inhibited L. theobromae isolated from banana with crown rot disease at the minimum concentrations of 1.0 and 0.8 µL/mL, respectively. According to Abeywickrama et al. (2003), vapour of basil oil at 0.05 % -0.2 % has found to be fungicidal against banana crown rot pathogens including L. theobromae. Dip treatment with cardamom (Elettaria cardamomum) oil at 0.70 µL/mL has signifi cantly reduced SER of Karutha Colomban mango (Karunanayake et al., 2018). Furthermore, fumigation with the essential oil of Amomum subulatum leaf has successfully controlled SER of mango (cv. Dasheri) (Dubey et al., 2008).
The objectives of the present study were (i) to investigate in vitro bioactive effi cacy of essential oils of cinnamon (i.e. leaf and bark), basil, and clove as eco-friendly antifungal agents against SER-associated fungi (L. theobromae, Pestalotiopsis sp., and Phomopsis sp.) isolated from Karutha Colomban mango, and (ii) to characterise these oils to identify eff ective antifungal components using gas chromatography-mass spectroscopy (GC-MS).

Test fungi
SER-associated fungi, isolated and characterised in a previous study (Ekanayake, 2017)  In vitro bioassays to determine the antifungal eff ects of essential oils

Liquid bioassay
According to the method described by Anthony et al. (2004), conical fl asks (100 mL) containing 50 mL of SMKY semi synthetic liquid medium (7.0 g yeast extract, 1.5 g KNO 3 , 20.0 g sucrose and 0.50 g MgSO 4 .7H 2 O dissolved in 1 L of distilled water) was autoclaved for 20 min at 1.03 kg cm -2 and 121 ºC. A concentration series of oils; 0.10 -1.00 μL/mL was prepared aseptically adding appropriate volumes of test oils to the fl asks, and Tween 80 (Koch-Light Laboratories, England) was used as a surfactant to disperse oil. Each assay fl ask was then separately inoculated with a 5 mm diameter fungal disc cut from the periphery of a 7-day old pure culture of the respective test pathogen. Contents were mixed thoroughly by placing the fl asks on reciprocal water bath shaker (New Brunswick Scientifi c, Model: R76) at room temperature (28 ± 2 ºC). A control (sterilised Journal of the National Science Foundation of Sri Lanka 48 (2) June 2020 distilled water and Tween 80 without oil) was included for comparison. Eugenol (2.00 µL/mL) prepared in a similar manner was used as an authentic standard to observe the antifungal effi cacy. Carbendazim fungicide (0.1 % w/v) was used as the positive control. Six replicates of each treatment and controls were arranged according to completely randomised design (CRD) in the laboratory. After 7 d of incubation, the mycelia (if any) were recovered on pre-weighed fi lter paper (Whatman No: 1, 5.5 cm diameter), washed three times with sterile distilled water, and placed in a hot air oven (Memmert, UM 600) at 105 ºC overnight until a constant weight was reached to determine the mycelial dry weight (Anthony et al., 2004). The inhibition percentage was determined using the following equation.
(1) C = mean dry weight of mycelium in control fl asks T = mean dry weight of mycelium in test fl asks

Disc volatilisation bioassay
Petri plates (90 × 15 mm) containing potato dextrose agar (PDA) medium (15 mL per plate) were inoculated with a 5 mm diameter mycelial plug from the periphery of a 7-day old pure culture of the respective test pathogen (Perumal et al., 2016). A sterilised fi lter paper disc with a diameter of 90 mm (Whatman No: 1) was placed on the inner surface of the Petri dish lid. Aliquots of the essential oil ranging from 2.0 -20.0 µL were added onto the fi lter paper discs placed in separate Petri dishes. A fi lter paper soaked with sterile distilled water served as the control. Similar treatments with eugenol (20.0 µL) and ethanol (100.0 µL) were used as authentic standard and positive control, respectively. The Petri plates were sealed with para-fi lm immediately after adding essential oil, placed in an inverted position and incubated at room temperature for 7 d (Feng et al., 2011). The experimental design was a CRD with 6 replicates per treatment and controls. The radial mycelial growth of each test pathogen was evaluated after 7 d of incubation by measuring the colony diameter along two axes at right angles to each other. These values were compared with those of untreated controls. The antifungal effi cacy was expressed as the percentage inhibition of radial mycelial growth (% IRMG) using the following formula (Abdollahi et al., 2011).
dc = mean radial mycelial growth in control Petri plates dt = mean radial mycelial growth in treatment Petri plates

Determination of minimum inhibitory and minimum lethal values of test oils
The mycelial plugs that did not show any growth in each in vitro bioassay were transferred to freshly prepared PDA plates and incubated for 7 d at room temperature (28 ± 2 ºC) to observe the recovery of the mycelial growth, in order to diff erentiate between the fungicidal and fungistatic activity of the selected essential oil treatments. Any revival of radial mycelial growth was categorised as the fungistatic eff ect and the corresponding concentration of the test oil was considered as the minimum inhibitory concentration (MIC). No growth of fungal mycelia (no revival) was considered as fungicidal eff ect and the corresponding concentration of the test oil was considered as minimum lethal concentration (MLC) (Sellamuthu et al., 2013;Bill et al., 2015;Perumal et al., 2016).

Gas chromatography-mass spectroscopy analysis of test oils
Basil, clove, cinnamon leaf, and bark oils were subjected to GC-MS analysis at the Industrial Technology Institute, Malabe. The analysis was carried out using a GC apparatus (Thermoscientifi c Trace 1300 model) equipped with an auto-injector AI 1310 (Thermoscientifi c) and a fused silica capillary column (Rtx-wax) of 30 m × 0.25 mm i.d., 0.25 μm fi lm (Thermoscientifi c), using helium as the carrier gas with a fl ow rate of 1.0 mL min −1 . The temperature was programmed to start at 60 ºC, followed by an increase at a rate of 5 ºC min −1 until 220 ºC was reached, then maintaining at 220 ºC for 10 min. The temperatures of the injector and the ion source were 240 ºC and 250 ºC, respectively. A volume of 0.3 μL essential oil diluted with 1 mL of hexane was injected at a partitioning rate of injected volume of 1:50 and a column pressure of 64.20 kPa. The MS was carried out with an ion capture detector operating in electronic impact mode with impact energy of 70 eV, a scan interval of 0.50 fragments, and fragments detected in the range of 50 -450 Da. The essential oil components were identifi ed by comparing the mass spectrum with spectra from the equipment database (NIST11). Additionally, the retention rates were compared with those in the literature (Adams, 2007).

Data analysis
Data with respect to in vitro bioassays were analysed using two-way analysis of variance (ANOVA), using Minitab 17 statistical software. The mean separation was done using Tukey's multiple comparison test (Anthony et al., 2004).
June 2020 Journal of the National Science Foundation of Sri Lanka 48(2)

Antifungal effi cacy of essential oils in liquid phase
In vitro liquid bioassay technique allows determination of the antifungal activity of essential oils which are in direct contact with the mycelia of fungal pathogens growing in an aqueous medium (Anthony et al., 2004). Out of the tested oils, basil oil was the most eff ective against L. theobromae which displayed 100 % mycelial inhibition at 0.20 µL/mL. Cinnamon bark oil was equally eff ective and achieved complete inhibition of mycelia of L. theobromae at 0.25 µL/mL. Clove and cinnamon leaf oils were slightly less eff ective against L. theobromae and displayed 100 % mycelial inhibition only at 0.40 µL/mL ( Figure 1). Accordingly, each of these concentrations were considered as MIC values for basil, cinnamon bark, clove and cinnamon leaf oils, respectively (Table 1). With respect to the fungicidal eff ect, basil oil was more eff ective against L. theobromae than other oils, with an MLC of 0.20 µL/mL. According to two-way ANOVA, percentage inhibition values obtained for diff erent oils were signifi cantly diff erent from the control when the test oil type, concentration and the interaction between test oil type × concentration were considered (p < 0.05). According to a similar liquid bioassay carried out by Anthony et al. (2004), basil oil was fungistatic as well as fungicidal at a minimum concentration of 0.80 μL/mL against L. theobromae isolated from banana, which is comparatively greater than MIC and MLC values determined by the present study. MLC values of cinnamon bark and clove oils against L. theobromae were slightly higher being 0.30 and 1.00 µL/mL, respectively (Table 1). According to a previous record (Ranasinghe et al., 2002) on a liquid bioassay using the same liquid medium, cinnamon bark oil was found to be fungicidal against L. theobromae isolated from banana at 0.45 μL/mL, which is slightly greater than the present value. MLC of clove oil against L. theobromae determined by Ranasinghe et al. (2002) was 0.60 μL/mL. However, present MLC value of clove oil against the same pathogen was somewhat lower compared to the previous result. The present MLC of cinnamon leaf oil against L. theobromae was 1.00 μL/mL and it is slightly greater than the value reported by Ranasinghe et al.  Growth of Pestalotiopsis sp. was completely inhibited by most of the test oils at very low concentrations (i.e. 0.40 -0.60 μL/mL; Figure 2). Cinnamon leaf and bark oils were eff ective against Pestalotiopsis sp., as they displayed 100 % growth inhibition (MIC) at a lower concentration (i.e. 0.40 μL/mL; Figure 3) and fungicidal eff ect (MLC) between 0.60 -0.80 μL/mL (Table 1). Basil and clove oils exhibited fungistatic eff ect against Pestalotiopsis sp. at a slightly higher concentration (i.e. 0.60 μL/mL). Further, MLC of basil oil was also recorded at 0.60 μL/mL indicating its fungicidal eff ect. The twoway ANOVA revealed that, results obtained for test oils were signifi cantly diff erent from the control when the test oil type, concentration and the interaction between test oil type × concentration were considered as factors (p < 0.05).
Further, clove, cinnamon leaf and cinnamon bark oils were more eff ective than basil oil in terms of fungistatic activity against Phomopsis sp. (Table 1; Figure 3). MIC of basil oil was 0.80 μL/mL, which was slightly greater than the MICs of clove and two cinnamon oils (i.e. 0.60 μL/mL). However, when considering the fungicidal activity, cinnamon bark oil was the most eff ective against Phomopsis sp. than other oils. Percentage inhibition values were signifi cantly diff erent when concentration and interaction of test oil type × concentration were considered as factors (p < 0.05). However, there was no signifi cant diff erence in percentage inhibition when only test oil type was considered as a factor (p > 0.05). Carbendazim and eugenol at fi xed concentrations (0.1 % w/v and 2.00 μL/mL, respectively) completely inhibited the growth of all test fungi revealing high antifungal effi cacy (data not presented). Eugenol is a bioactive compound found in the essential oils of many higher plants and its antifungal action has been confi rmed by many studies (Herath & Abeywickrama, 2008;Wang et al., 2010).

Antifungal activity of essential oils in vapour phase
In vitro disc volatilisation bioassay was carried out to identify the vapour phase eff ect of each essential oil against the growth of test fungi on a solid medium. The volatile nature of the bioactive compounds present in essential oils allows them to act against many fungal pathogens (Bill et al., 2015). Vapour of clove, cinnamon Each data point represents the mean of six replicates.  June 2020 Journal of the National Science Foundation of Sri Lanka 48 (2) leaf, and cinnamon bark oils showed a complete inhibition of radial mycelial growth at 2.0 µL/plate concentration against L. theobromae (Table 2). However, according to Sukatta et al. (2008), higher concentration of cinnamon bark oil (i.e. 10 μL/plate) was needed for 100 % inhibition of mycelial growth of L. theobromae, while clove oil at the same concentration controlled the growth of L. theobromae only up to 74.07 %. Further, in the present study, cinnamon leaf and cinnamon bark oils showed a fungicidal eff ect against L. theobromae at concentrations of 2.0 µL/plate and 4.0 µL/plate respectively, whereas, clove oil failed to inhibit the pathogen growth completely. Basil oil vapour achieved a 100 % IRMG at a concentration of 10.0 µL/plate and was the least eff ective oil against L. theobromae compared to other test oils, which could not display fungicidal eff ect either (Table 2). Abeywickrama et al. (2003) reported that basil oil showed a high antifungal eff ect against banana fruit pathogens, Colletotrichum musae, Fusarium proliferatum and L. theobromae at very low concentrations (i.e. MLC = 0.05 -0.20 %) comparable with the results of the present research.   Journal of the National Science Foundation of Sri Lanka 48(2) June 2020 According to Table 2, clove and cinnamon bark oils achieved a 100 % IRMG of Pestalotiopsis sp. at a MIC of 2.0 μL/plate. Therefore, eff ectiveness of those two oils in the vapour form was high, when compared to basil and cinnamon leaf oils, having fungistatic eff ects at 16.0 and 4.0 μL/plate concentrations, respectively. Fungicidal activity of clove and cinnamon bark oils (i.e. MLC = 6.0 and 4.0 μL/plate, respectively) was more pronounced against Pestalotiopsis sp. than that of basil and cinnamon leaf oils (i.e. MLC = 16.0 and 8.0 μL/plate, respectively). Since clove oil demonstrated relatively lower MIC and MLC values than basil oil, it could be considered as a moderately eff ective fumigant against Pestalotiopsis sp. Clove, cinnamon leaf, and cinnamon bark oils displayed fungistatic as well as fungicidal eff ects against Phomopsis sp. at a concentration of 4.0 μL/plate. This value was notably less than the MIC and MLC of basil oil (i.e. 12.0 μL/plate) against Phomopsis sp. Therefore, effi cacy of basil oil vapour against Phomopsis sp. was relatively less than that of clove and cinnamon oils, which could be recognised as the best oils in controlling the target pathogen (Table 2). According to Sukatta et al. (2008), a diff erent strain of Phomopsis (P. viticola) was completely inhibited by cinnamon bark oil at 10 μL/plate concentration, while clove oil at the same concentration controlled the growth of P. viticola up to 91 %. However, the present MIC and MLC values of clove and cinnamon bark oils against mango SER-associated Phomopsis sp. were notably lower when compared to the results of the former study. According to the present study eugenol (20 μL/plate) at vapour phase was capable of inhibiting all test pathogens, completely. Although ethanol (100 μL/ plate) was used as the positive control in this fumigation assay, it was unable to inhibit the growth of test fungi, completely. However, % IRMG values for ethanol were 21.60 and 50.62 % for Pestalotiopsis sp. and Phomopsis sp., respectively. According to Bacílková (2006), vapour of ethanol at 30 -90 % (v/v) has totally controlled Aspergillus niger and Penicillium notatum, which are commonly considered as saprophytes. Since absolute ethanol was used against the two pathogenic fungi tested in the present study, a complete inhibition could not be achieved.

GC-MS characterisation of test oils
GC-MS analysis of the selected essential oils revealed the major chemical constituents of each tested essential oil along with their relative abundance. In basil oil, methyl chavicol was the major compound with a relative abundance of 74.44 % with relatively low amounts of eugenol and (E)-cinnamaldehyde (Table 3). However, eugenol was the most abundant constituent in both clove and cinnamon leaf oils. (E)-cinnamaldehyde in cinnamon leaf was relatively low. The composition of cinnamon bark oil consisted of 72.18 % (E)-cinnamaldehyde with a low abundance of eugenol. Linalool, which is one of the antifungal constituents present in many essential oils, was detected in basil, cinnamon leaf, and cinnamon bark oils with percentages of 15.01, 1.27 and 3.69, respectively (Table 3). All these chemical constituents have previously been reported as bioactive antifungal compounds against many postharvest pathogenic fungi. Methyl chavicol, which is the most abundant component present in basil oil was found to be antifungal against two postharvest pathogenic fungi viz, C. gloeosporioides and Alternaria alternata (Costa et al., 2015). Eugenol, which was recognised as one of the most eff ective bioactive components present in the test oils has previously been reported for its strong antifungal activity against C. musae, F. proliferatum, and Botrytis cinerea (Herath & Abeywickrama, 2008;Wang et al., 2010). According to Hong et al. (2015) and Marei and Abdelgalei (2018), cinnamaldehyde was found to be antifungal against C. gloeosporioides, B. cinerea, Aspergillus niger, and Penicillium digitatum. Further, Soković and van Griensven (2006) have identifi ed linalool as an antifungal agent of controlling Verticillium fungicola and Trichoderma harzianum. Although several antifungal components are present in a single oil in diff erent proportions, they appear to act synergistically to inhibit the growth of target fungi (Anthony et al., 2004).
Antimicrobial constituents of essential oils aff ect cell structure by causing disruption of cell membrane, alteration of cell membrane integrity, and inhibition of cell wall formation. Integrity and functionality of cell membranes are governed by ergosterols, a structural component of cell membranes. Bioactive compounds of essential oils can bind with ergosterols or act as specifi c inhibitors that inhibit the biosynthesis of ergosterols, causing cell membrane destruction. Essential oils can also inhibit the cell wall formation by blocking the formation of β-glucans in fungi, which is one of the important constituents of the cell wall (Nazzaro et al., 2017). Damage to cell wall and cell membrane by essential oils can cause leakage of macromolecules and electrolytes and cell lysis of pathogens (Bakkali et al., 2008). Leakage of contents of C. musae and F. proliferatum conidia treated with essential oils of Ocimum basilicum, Cymbopogon citratus, eugenol and citral was reported by Herath and Abeywickrama (2008). Essential oils are also known to cause the dysfunction of mitochondria and inhibition of effl ux pumps causing serious damages to pathogens. Mitochondrial electron transport chain can be inhibited by the antimicrobial constituents of essential oils, which reduce the mitochondrial membrane potential of fungal pathogens. Activity of proton pumping ATPases can be  Table 3: Chemical components of tested essential oils, their retention time (RT) and relative abundance (RA) as determined by GC-MS inhibited by the essential oils, which diminishes ATP production in microorganisms. Inhibition of effl ux pumps by essential oil constituents will disturb the removal of toxic substances out of the cell (Nazzaro et al., 2017). Essential oils are also capable of damaging cytoplasmic lipids and proteins leading to coagulation of cytoplasm (Bakkali et al., 2008). Many other mechanisms of action for various oil components have been proposed and tested against several fungal pathogens. As essential oils tested in this study contain many bioactive components in diff erent proportions, it could be assumed that they act synergistically in inhibiting the mango SER pathogens with the aid of one or more mechanisms of action as highlighted in the literature.
Journal of the National Science Foundation of Sri Lanka 48(2) June 2020

CONCLUSION
According to the in vitro bioassays carried out during the present study, liquid and vapour forms of basil, clove, and cinnamon oils successfully controlled SER-associated fungi of mango, viz. L. theobromae, Pestalotiopsis sp., and Phomopsis sp. However, their eff ectiveness depends on the type of oil and the fungal species tested. In general, cinnamon bark oil at liquid and vapour phases displayed a higher in vitro antifungal effi cacy against the above SER associated fungi isolated from Karutha Colomban mango. The results highlight the importance of formulating suitable spray or fumigation treatments using these essential oils as eco-friendly alternatives to conventional fungicides, to control SER of mango in order to lengthen the postharvest storage life. Since, L. theobromae was the main pathogen frequently isolated from mango with SER, novel essential oil treatments could be focused towards controlling L. theobromae along with other SER-associated pathogens. Liquid spray and fumigation treatment systems are being currently developed in vivo to control SER in Karutha Colomban mango in combination with passive modifi ed atmosphere packaging and cold storage.