Antifungal activity of volatile compounds generated by essential oils against fungi commonly causing deterioration of bakery products

Authors


A. J. Ramos, Food Technology Department, Lleida University, UTPV-CeRTA, Rovira Roure 191, 25198 Lleida, Spain (e-mail: ajramos@tecal.udl.es).

Abstract

Aims: To investigate the volatile fractions of 16 essential oils for activity against the more common fungi causing spoilage of bakery products, Eurotium amstelodami, E. herbariorum, E. repens, E. rubrum, Aspergillus flavus, A. niger and Penicillium corylophilum.

Methods and Results: The study applied 50 μl of pure essential oils in a sterilized filter paper, were carried out at pH 6 and at different water activity levels (0·80–0·90). First, a wheat flour based agar medium was used, where cinnamon leaf, clove, bay, lemongrass and thyme essential oils where found to totally inhibit all microorganisms tested. These five essential oils were then tested in sponge cake analogues, but the antifungal activity detected was much more limited.

Conclusion: Five essential oils showed potential antifungal capacity against all species tested, over a wide range of water availability. Their activity, however, seems to be substrate-dependent. More research is needed to make them work in real bakery products, as in the preliminary study limited effectiveness was found.

Significance and Impact of the Study: The potential of the cinnamon leaf, clove, bay, lemongrass and thyme essential oils against species belonging to Eurotium, Aspergillus and Penicillium genus has been demonstrated.

Introduction

In bakery processing, the most common type of microbial spoilage is mould growth and in many cases it is the major factor governing shelf-life (Earle and Putt 1984). As products of intermediate moisture (0·75–0·90aw, aw being the water activity) (Marín et al., 2002a), cakes are particularly sensitive to xerophilic moulds spoilage (Seiler 1988; Beuchat and Hocking 1990; Pitt and Hocking 1997; Fustier et al. 1998).

Historically, many plant oils and extracts have been used as topical antiseptics, or have been reported to have antimicrobial properties (Hoffman 1987; Lawless 1995). Restrictions imposed by the food industry and regulatory agencies on the use of some synthetic food additives have led to renewed interest in searching for alternatives, as natural antimicrobial compounds, particularly those derived from plants (Delaquis and Mazza 1995; Hammer et al. 1999). Essential oils and their constituents have been used extensively as flavour ingredients in a wide variety of foods, beverages, and confectionery products. Many of such compounds are classified as Generally Recognised As Safe (Kim et al. 1995). The antimicrobial activity of spices and essential oils is well recognized. Several studies have reported results on their preservative action (Mishra and Dubey 1994; Paster et al. 1995; Montes-Belmont and Carvajal 1998; Vázquez et al. 2001). For this reason, these substances can be used as safe and effective alternatives to synthetic preservatives (Wilson et al. 1997). Essential oils of thyme, cinnamon, bay and clove had been proved to possess antimicrobial activity (Farag et al. 1989).

The antifungal activity of essential oils upon solution contact as measured by the broth dilution and agar dilution methods has been studied by many researchers. However, the activity by vapour contact has been reported more rarely (Matsuoka et al. 1990; Inouye et al. 2000). Several spices, herbs and fruits contain volatile antimicrobial compounds (Zaika 1988). Some volatile components of citrus fruit essential oils show a high degree of fungal inhibition (Caccioni et al. 1998). Wilson et al. (1987) found that a number of the volatiles that emanate from peaches, as they ripen, are fungicidal. Volatile antimicrobial substances (essential oils and oleoresins) from mustard, cinnamon and garlic have proven to be efficient in the control of spoilage by common bread spoiling fungi (Nielsen and Rios 2000). Earlier, some investigators reported the antifungal activity of black mustard powder (Goi et al. 1985) or mustard extract (Sekiyama et al. 1996) by gas contact. Arras and Usai (2001) found that thyme essential oil vapours had some antifungal activity. The authors also recorded alteration in Penicillium digitatum morphology.

Although different results have been observed depending on testing conditions, target microorganisms and antimicrobial substance origin, some essential oil compounds such as eugenol, geraniol, thymol and cinnamaldehide have consistently been reported to have antifungal activity (Bullerman et al. 1977; Farag et al. 1989; Salmeron and Pozo 1991; Mahmound 1994; Chao and Young 2000; Chang et al. 2001).

The objective of this work was to evaluate the antifungal activity of 16 essential oils by vapour contact against the most common fungal species causing bakery products spoilage, under different water activity conditions.

Materials and methods

Fungal isolates

A total of seven isolates from different bakery products were used. Five of them, Eurotium amstelodami (3·205), Eurotium herbariorum (3·209), Eurotium rubrum (3·228), Aspergillus flavus (3·226) and Aspergillus niger (3·227) were isolated by Abellana et al. (1997) from Spanish bakery products. These isolates belong to the Food Technology Department microorganisms collection of the Lleida University. The other two isolates, Eurotium repens (IBT18000) and Penicillium corylophilum (IBT6978) were kindly provided by the Department of Biotechnology of the Technical University of Denmark and had been isolated from Danish bakery products.

Study in wheat flour agar

Essential oils.

The essential oils used in this study were of lemon, aniseed, mandarin, grapefruit, cinnamon leaf, lemongrass, rosemary, thyme, basil, sweet fennel, peppermint, ginger, bay, clove, sage and orange. They were all supplied by F. D. Copelans & Sons, Ltd (London, UK).

Experimental design.

The factors assayed were different essential oils as a qualitative factor and aw (0·80, 0·85 and 0·90) as a quantitative factor. The aw values include the range of aw variation of bakery products. The response recorded was colony diameter. A full factorial design was used and all treatments were repeated twice.

Media preparation.

The basic medium used was a 2% wheat flour agar. It was adjusted to achieve the desired aw level and to pH 6 (as a mean pH for bakery products) as previously described by Marín et al. (2002a). Sterile media were poured into 9-cm diameter sterile Petri plates. The final aw and pH levels of the media were checked with an AquaLab (Decagon, Pullman, WA, USA), and a Crison micropH2000 pH-meter (Crison, Barcelona, Spain), respectively.

Inoculation, incubation and measurement.

Plates were kept in an inverted position. A sterilized filter paper (square of 1 × 1 cm) was placed in the centre of the lid and 50 μl of pure essential oil were added to the paper. Blank were made by adding 50 μl of water to it. For each fungal isolate, a conidial spore suspension of 106 spores ml−1 was prepared and Petri plates were needle-inoculated in the centre. Plates were tightly sealed with parafilm and incubated for 42 days at 25°C. Diameters of the growing colonies were measured weekly with the aid of a binocular magnifier.

Chemical analysis of the essential oils.

Analysis of essential oils having antifungal activity in the wheat flour agar study was performed by gas chromatography. The Consejo Superior de Investigaciones Científicas (CSIC, Institute of General Organic Chemistry, Madrid, Spain) kindly performed the analysis, using a Hewlett-Packard chromatograph (Hewlett Packard Co. Geneva, Switzerland), model 6890 coupled with mass spectrometer selective detector 5973N.

Studies in cake analogues

Pure essential oils found to be effective in the agar study were checked (at the same levels of aw and pH) in a Spanish cake dough representative, prepared in the laboratory as previously reported by Abellana et al. (1999). The dough after baking had a pH of about 7·5, and its initial aw was ca. 0·75. To reach the required pH (6·0), citric acid was added to the mixed solid ingredients before baking, whereas aw was adjusted to 0·80, 0·85 and 0·90 by placing the cake analogues in Petri plates containing water–glycerol agar (1·5% of agar and different amounts of glycerol depending on the desired aw, as indicated in Table 1). Previously, two calibration curves were plotted, one to determine the amount of citric acid necessary to reach the desired pH (Marín et al. 2002b) and the other to determine the concentration of glycerol in the agar needed to increase and maintain the aw of the dough during the incubation period (data not shown). After baking, the cakes were aseptically cut into 5 × 5-cm squared pieces, placed in 9-cm sterile Petri plates containing water–glycerol agar and after being sealed with parafilm, incubated at 25°C for a 48-h equilibration period. The procedures for adding the essential oils (50 μl) and for making the blanks were the same as described above, but in this case, paper was placed on the water the glycerol agar beside the cake analogue, as Petri plates had to be incubated in an upright position. The inoculation and incubation procedures, and the measurements of fungal growth were the same as described above. The experiment was performed in triplicate.

Table 1.  Composition of the water–glycerol agar used to adjust aw level of the cake analogues
aw of cake analoguesGlycerol (g) per 100 ml of water
0·80105·8
0·8569·0
0·9039·1

Statistical treatment of the results

Analyses of variance of the effect of aw and essential oil volatile fraction with the colony radius measures at the different incubation days as co-variable, were performed for each isolate separately by using SAS (Statistical Analysis System) program version 8·02 (SAS Institute Inc., Cary, NC, USA).

Results

Studies in wheat flour agar

The combination of natural preservatives in the surrounding atmosphere and reduced aw, had a significant impact on the growth of all isolates tested (Tables 2 and 3).

Table 2.  Mean colony diameter (mm) of Eurotium spp. growing on wheat flour agar medium exposed to different essential oil atmospheres andaw levels, after 42 days of incubation at 25°C
Essential oilE. amstelodamiE. herbariorumE. repensE. rubrum
0·80aw0·85aw0·90aw0·80aw0·85aw0·90aw0·80aw0·85aw0·90aw0·80aw0·85aw0·90aw
  1. Data in the column followed by different letters are significantly different in LSMEANS test.

Blank29·5a43·1a69·6a0·2b30·3ab44·3a14·8a28·5ab61·2ab 5·4bc27·2ab60·6a
Lemon18·8abc44·3a65·5a0·0b29·8ab46·1a 0·0b27·7ab59·7ab19·7ab38·7ab48·7a
Mandarin21·9ab44·2a56·3a3·4a37·8a58·0a11·1a26·1ab64·8a26·1a52·3a66·4a
Aniseed16·5abcd44·0a61·6a0·0b19·4abc50·0a 7·8ab33·2ab61·1ab 6·8bc37·2ab63·7a
Sweet fennel13·9bcde35·7a59·3a0·2b18·1abc55·0a 7·4ab19·6abc48·0abc12·8abc39·1ab62·1a
Orange13·0bcde47·1a70·9a1·4ab35·1a54·8a13·1a26·8ab63·9ab12·7abc47·1ab67·1a
Grapefruit 7·7cde43·8a68·0a0·0b23·9ab50·3a 1·6b36·6a58·4ab 9·9abc34·8ab53·7b
Rosemary 3·0de24·2ab59·4a0·0b49·0ab56·5a 0·0b26·4ab55·6ab 9·6abc24·6abc62·3a
Basil 2·5e36·6a59·5a1·1ab42·8ab54·4a 0·0b20·3abc44·2abc16·6abc29·8ab61·8a
Sage 2·3e20·7ab64·3a1·3ab0·0abc47·1a 1·3b21·2abc33·3bc 9·1abc20·8bc69·7a
Ginger 0·0e23·5ab63·4a0·0b21·4abc51·6a 0·0b16·2bcd49·7abc 0·0c40·5ab65·0a
Peppermint 0·0e 0·6b20·2b0·0b12·9bc52·7a 0·0b 3·7cd20·5cd 0·0c 0·0c49·7a
Lemongrass 0·0e 0·0b 0·0b0·0b 0·0c 0·0b 0·0b 0·0d 0·0d 0·0c 0·0c 0·0b
Clove 0·0e 0·0b 0·0b0·0b 0·0c 0·0b 0·0b 0·0d 0·0d 0·0c 0·0c 0·0b
Cinnamon leaf 0·0e 0·0b 0·0b0·0b 0·0c 0·0b 0·0b 0·0d 0·0d 0·0c 0·0c 0·0b
Bay 0·0e 0·0b 0·0b0·0b 0·0c 0·0b 0·0b 0·0d 0·0d 0·0c 0·0c 0·0b
Thyme 0·0e 0·0b 0·0b0·0b 0·0c 0·0b 0·0b 0·0d 0·0d 0·0c 0·0c 0·0b
Table 3.  Mean colony diameter (mm) of Aspergillus spp. and Penicillium corylophilum growing on wheat flour agar medium exposed to different essential oil atmospheres and aw levels, after 42 days of incubation at 25°C
Essential oilA. flavusA. nigerP. corylophilum
0·85aw0·90aw0·85aw0·90aw0·85aw0·90aw
  1. Data in the column followed by different letters are significantly different in LSMEANS test.

Blank31·1a74·9a14·0a55·2ab1·4a 8·6b
Lemon14·2bc62·9a 0·0b69·0ab0·0b29·0a
Mandarin 6·3cd61·3a 0·0b73·2a0·0b 3·2b
Aniseed15·2bc55·6ab 0·0b66·4ab0·0b 7·1b
Sweet fennel19·4ab72·2a 0·0b72·5a0·0b 1·1b
Orange21·4ab75·2a 2·8b75·5a0·0b 7·7b
Grapefruit18·7abc63·8a 0·0b72·3a0·0b 5·8b
Rosemary12·1bcd68·8a 0·0b66·3ab0·0b22·7a
Basil24·4ab72·2a 0·0b72·9a0·0b 5·6b
Sage 0·0d54·0ab 0·5b66·9ab0·0b 5·5b
Ginger18·0ab56·1ab 0·0b69·9ab0·0b 3·5b
Peppermint 0·0d36·0b 0·7b45·2b0·0b 2·9b
Lemongrass 0·0d 0·0c 0·0b 0·0c0·0b 6·4b
Clove 0·0d 0·0c 0·0b 0·0c0·0b 0·0b
Cinnamon leaf 0·0d 0·0c 0·0b 0·0c0·0b 0·0b
Bay 0·0d 0·0c 0·0b 0·0c0·0b 0·0b
Thyme 0·0d 0·0c 0·0b 0·0c0·0b 0·0b

The behaviour of all species was quite similar and, as expected, the growth was favoured by increasing levels of water activity. Aspergillus spp. and P. corylophilum were unable to grow at 0·80aw. At 0·85awA. niger was markedly affected by low water availability, and then fungal growth was most easily prevented (Table 2).

Among the essential oils, the ones that totally inhibited growth of all species, were cinnamon leaf, lemongrass, thyme, bay and clove, regardless of aw level. Peppermint and ginger were also able to inhibit fungal growth but their activity was restricted to the lower aw levels (0·80–0·85). The remaining essential oils were, in general, ineffective or only delayed growth of some of the species. Sage essential oil slowed E. amstelodami and E. repens growth at 0·80aw, although it favoured E. herbariorum and E. rubrum growth at the same aw level (Table 2); A. niger and A. flavus growth were also inhibited by sage essential oil at 0·85aw (Table 3). The most sensitive species to essential oils at low aw were E. repens and A. niger, as they were inhibited by almost all essential oils at 0·80aw and at 0·85aw, respectively. On the other hand, some promoting effect on fungal growth by the addition of some essential oils was observed; the ones that stood out were mandarin favouring E. herbariorum and E. rubrum growth at low aw, orange favouring E. herbariorum also at low aw, lemon favouring P. corylophilum at 0·90aw and to a lesser extent E. repens at 0·80aw. Finally P. corylophilum grew more at 0·90aw when rosemary essential oil was added.

Composition of the essential oils with antifungal activity in agar medium

The main components of cinnamon leaf, clove, bay, thyme and lemongrass essential oils were identified by GC-mass spectrometer analyses. Their percentage compositions, calculated as the ratio of the peak area to the total chromatographic area, are listed in Table 4. Eugenol was the main component of clove (83·9%), cinnamon leaf (78·5%) and bay (57%) essential oils. Geraniol and neral represented the 50·5 and 29·4%, respectively, of lemongrass essential oil. Finally thyme essential oil was mainly composed of thymol (53·9%) and p-cimene (25·2%).

Table 4.  Qualitative and quantitative composition (expressed as percentage) of essential oils having antifungal activity
ComponentBayCinnamon leafCloveLemongrassThyme
Limonene 2·2 3·8 1·1
Cariophyllene 0·9 4·712·6 4·1<1
Neral29·4
Geraniol50·5
Geranial acetate 1·6
Geranial 5·0
Cariophylene oxide<1<1
Humelene 1·4
Eugenol57·078·583·9
α-cubenene<1
Linalool 2·6 2·6 3·5
2-propenil benzodioxol 1·4
3-phenil propenal 1·1
3-phenil propenol acetate 1·1
Eugenil acetate 1·7
Mirceno12·4 1·7
Cineol 1·6 1·0
1-octen-3-ole 1·0
Cadinene 0·5
Metil eugenol 1·9
γ-terpinene 3·4
p-cimene25·2
Bornil-acetate 1·8
α-terpineol<1
Borneol<1
Thymol53·9
Carvacrol 2·9
Non-identified compounds15·7

Studies in cake analogues

Both factors, the presence of essential oil and aw, and their interaction had a significant impact on the growth of all isolates tested. Further separate analyses were performed for each water activity level and the effect of different essential oils was analysed by the LSMEANS test (Table 5, E. repens). In general, a slight trend to delay fungal growth was observed only at 0·80aw. Moreover, in some cases, as aw increased fungal growth was favoured. All isolates had similar responses.

Table 5.  Mean colony diameter (mm) of Eurotium repens growing on sponge cake analogues, exposed to different essential oils volatile fraction, after 42 days of incubation at 25°C
Essential oil0·80aw0·85aw0·90aw
  1. Data in the column followed by different letters are significantly different at P < 0·01 in LSMEANS test.

Control14·6a25·2b39·9a
Thyme10·9b26·9ab39·7a
Lemongrass 8·8bc29·5ab39·3a
Cinnamon leaf 8·1bc29·2ab39·3a
Bay10·6bc25·6b39·8a
Clove 7·3c31·9a41·0a

Discussion

The ability of 16 essential oils to inhibit bakery spoilage fungi was evaluated. Most studies concerning the antimicrobial activity of natural products have been conducted by testing the plant extracts as contact materials, mostly by supplementing growth media or by incorporation of the plant material itself (Paster et al. 1995). However, these methods do not reflect the activity of the volatile fraction alone (Moleyer and Narashimam 1986; Delespaul et al. 2000). To this end, therefore, the present work aims to study the activity of the volatile portion of the essential oils against fungal growth.

Previous experiments were carried out with the same set of essential oils used in this study, but they were added diluted directly to 2% wheat flour agar (0–1000 ppm). Results showed clearly that cinnamon leaf, bay, clove, rosemary and thyme had some antifungal activity; they prevented or delayed mycelium growth of the seven isolates tested (M. E. Guynot, S. Marín, L. Setó, V. Sanchis, A. J. Ramos, unpublished data). The present study confirmed the antifungal activity of cinnamon leaf, bay, clove and thyme essential oils. Although no growth inhibition was observed by applying rosemary essential oil, lemongrass was found to delay fungal growth by vapour contact. Nielsen and Rios (2000) had proven that the volatile substances from mustard, cinnamon, garlic and clove essential oils were efficient in the control of common bread spoilage fungi. Inouye et al. (2000) also found fungicidal activity in lemongrass essential oil followed by cinnamon bark and thyme oils, by vapour contact. However, comparison of the data obtained by different studies is difficult, because of differences in plants extract compositions, in methodologies followed to assess antimicrobial activity and in microorganisms chosen to be tested (Hammer et al., 1999).

Several investigators support the best antifungal activity of volatile compounds by gaseous contact than by aqueous solution or by agar contact (Goi et al. 1985; Sekiyama et al. 1996; Inouye et al. 2000). It is difficult to differentiate the direct effect of vapour on fungi from the indirect effect of vapour acting after absorption by the medium during a long incubation period (Inouye et al. 2000). Due to the hydrophobic nature of compounds that constitute the volatile fraction of essential oil, it could be expected that their dilution in the agar medium during incubation period was not so important. Sekiyama et al. (1996) measured the amount of allyl isothiocyanate (active principle of some species) in a potato dextrose agar exposed to vapour released from a mustard essential oil, and they detected that only the 15–20% of the extract used was absorbed by the culture medium. Due to the high lipophilic nature of mycelia coupled with a large surface area relative to the volume of a fungus, vapours of essential oils may act mainly by accumulation on mycelia than in the agar (Inouye et al. 2000).

In general, the inhibitory effect has been attributed to the most abundant components and not to the other associated substances of the oil (Farag et al. 1989). However, some workers found more antifungal activity when adding the whole extract to the medium than when adding only the principal compound, suggesting a synergistic effect of some minor constituents of the oil (Bullerman et al. 1977; Paster et al. 1995; Chang et al. 2001).

The composition of the extracts obtained from a plant is known to vary according to the species and variety of the plant, the agronomic conditions, the harvest time and the type of processing followed (Mishra and Dubey 1994; Guillén and Cabo 1996; Hammer et al. 1999). However, in general, the results obtained from the GC analysis of the essential oils used in this experiment were as expected. Eugenol reported by different workers to be the main component of clove, cinnamon leaf and bay essential oils and is also responsible for the antifungal effect of these oils (Ross 1976; Bullerman et al. 1977; Farag et al. 1989; Montes-Belmont and Carvajal, 1998). Bullerman et al. (1977) found eugenol to be the active compound responsible for fungal inhibition by clove essential oil, but they raised the possibility that interactive effects of other compounds present in smaller quantities may also contribute. On the other hand, Mahmound (1994) reported that 1000 ppm of different essential oils constituents, namely geraniol and thymol, completely inhibited fungal growth. Later Paster et al. (1995) found thymol, the main component of thyme essential oil, to be active against Aspergillus spp. when it was added to the medium. Vázquez et al. (2001) reported a total inhibition of Penicillium citrinum adding 2000 ppm of eugenol and thymol in a liquid medium.

Several studies have been conducted to understand the mechanism of action of essential oils (Chang et al. 2001), however it is still unclear. Some studies suggest that compounds penetrate inside the cell, where they interfere with cellular metabolism (Marino et al. 2001). Other studies suggest that they disturb the cellular membrane and react with active sites of enzymes or act as a H+ carrier, depleting adenosine triphosphate pool (Farag et al. 1989; Chang et al., 2001; Ultee et al. 2002).

The promising results of the agar culture study encouraged us to conduct experiments in a sponge cake analogue. It was found, however, that all five essential oils had low antifungal activity, as they inhibited fungal growth only if they were combined with low water availability. Actually, bibliography exists reporting cases of low antimicrobial activity by this type of compounds when they are applied in foodstuff (Kim et al. 1995; Del Campo et al. 2000). Cake dough lipids may affect the activity of essential oils because of the hydrophobic properties of their active compounds. Arras and Usai (2001) showed that thyme essential oil was less effective on orange fruits than in vitro against P. digitatum. Earlier, Kim et al. (1995) found differences in the bactericidal effectiveness of three essential oil compounds (geraniol, citral and carvacol) when they were added in a culture medium or on inoculated fish cubes. Authors argue that this minor activity in vivo could be because of interactions of essential oils active compounds (known to be lipophilic) with food components, such as proteins and lipids, decreasing their level.

As a conclusion, volatile substances from cinnamon leaf, clove, bay, lemongrass and thyme essential oils had a potential antifungal activity against common fungi causing spoilage of bakery products. Due to the poor activity observed when they were used in a sponge cake analogue, further studies are needed before their application in bakery products. Combining them with other preservatives or modified atmosphere packaging could be a valuable alternative.

Acknowledgements

The authors are grateful to the European Union (PL98-4075), to the Spanish government (CICYT, ALI 99-0831) and to the Catalonian government (CIRIT) for their financial support.

Ancillary