Orange Essential Oils Antimicrobial Activities against Salmonella spp

JFS M: Food Microbiology and Safety

Orange Essential Oils Antimicrobial Activities against Salmonella spp. C.A. O’BRYAN, P.G. CRANDALL, V.I. CHALOVA, AND S.C. RICKE

ABSTRACT: Seven citrus essential oils (EOs) were screened by disc diffusion assay for their antibacterial activity against 11 serotypes/strains of Salmonella. The 3 most active oils were selected to determine the minimal inhibitory concentration (MIC) against the same Salmonella. Orange terpenes, single-folded d-limonene, and orange essence terpenes all exhibited inhibitory activity against the Salmonella spp. on the disc diffusion assay. EOs were stabilized in broth by the addition of 0.15% (w/v) agar for performance of the MIC tests. Orange terpenes and d-limonene both had MICs of 1%. The most active compound, terpenes from orange essence, produced an MIC that ranged from 0.125% to 0.5% against the 11 Salmonella tested. Gas chromatography-mass spectrometry (GC/MS) analysis revealed that this orange essence oil was composed principally of d-limonene, 94%, and myrcene at about 3%. EOs from citrus offer the potential for all natural antimicrobials for use in improving the safety of organic or all natural foods. Keywords: citrus essential oils, natural antimicrobials, Salmonella

M: Food Microbiology & Safety

M

Introduction

icrobial pathogens are responsible for an estimated 323000 hospitalizations each year at a cost of $7 to 10 billion annually (Mead and others 1999). The Centers for Disease Control and Prevention (CDC) estimates that financial losses from foodborne illnesses, including medical costs and losses in productivity, range from $500 million to $2.3 billion annually (Frenzen and others 1999; CDC 2001). In 2006, 7 serotypes accounted for 66% of the laboratory confirmed Salmonella infections in humans as follows: Typhimurium 19%, Enteritidis 19%, Newport 9%, Heidelberg 6%, Javiana 5%, Montevideo 4%, and Heidelberg 4% (CDC 2007). Overall, the percentage of the population acquiring Salmonella infections is not decreasing, although the percentage of infections caused by S. Typhimurium decreased significantly (41%) in 2006 as compared to the baseline of 1996 to 1998 (CDC 2007). Other serotypes had significant increases compared with the 1996 to 1998 baseline, including S. Enteritidis (28%), S. Newport (42%), and S. Javiana (92%) (CDC 2007). In recognition of the importance of reducing Salmonella infections and mortalities, the Healthy People 2010 Initiative has established a goal of no more than 6.8 cases of salmonellosis/100000 persons, 1/2 of the initial baseline rate of 13.6/100000 persons. Unfortunately, in 2006 the rate was 14.8 cases/100000 (CDC 2006). Thus, Salmonella-mediated foodborne illness continues to be a serious problem and there may be an increase in the number of contaminated foods (Anonymous 2007) and in the number of persons becoming ill. Novel intervention strategies to reduce or eliminate Salmonella in foods are a priority for food processors and researchers. The effects of essential oils and their components on bacterial growth and survival have been studied for many years. Deans and

MS 20070724 Submitted 3/21/2007, Accepted 4/6/2008. Authors O’Bryan, Crandall, Chalova, and Ricke are with Center for Food Safety-IFSE and Food Science Dept., 2650 Young Ave., Univ. of Arkansas, Fayetteville, AR 72704, U.S.A. Author Chalova and Ricke are also with Poultry Science Dept., Univ. of Arkansas, Fayetteville, AR 72701, U.S.A. Direct inquiries to author Crandall (E-mail: [email protected]).

M264

JOURNAL OF FOOD SCIENCE—Vol. 73, Nr. 6, 2008

Richie (1987) concluded that Gram-positive and Gram-negative bacteria were equally sensitive to citrus essential oils and components. Fisher and Phillips (2006), on the other hand, reported that Gram-positive bacteria were more sensitive than Gram-negative in vitro. Kim and others (1995a) found that components of citrus essential oils carvacrol, citral, and geraniol had strong activity against S. Typhimurium and its rifampicin resistant mutant in vitro. In this same set of experiments, they inoculated fish cubes with S. Typhimurium and found that 3% carvacrol eliminated 100% of the inoculated bacteria. Kim and others (1995b) did not observe any activity of nerolidol, limonene, or β-ionone against E. coli, E. coli O157:H7, or S. Typhimurium. Dabbah and others (1970) used orange, lemon, grapefruit, and mandarine citrus oils and their derivatives lime terpeneless oil, orange terpeneless oil, lemon terpeneless oil, d-limonene, terpineol, and geraniol in vitro against S. Senftenberg, E. coli, Staphylococcus aureus, and Pseudomonas spp. The fractions were more active against all bacteria than were the oils themselves. The fractions in general had more activity against St. aureus than against Gram-negative bacteria. These citrus oils did not reduce populations of S. Senftenberg, but lime terpeneless, terpineol, and geraniol reduced it 100%. Lime terpeneless oil, orange terpeneless oil, lemon terpeneless oil, and d-limonene reduced growth from 100% to 50% ranked from the most to the least. Orange oil and terpineol extended the shelf life of pasteurized milk to more than 56 d when stored at 4 ◦ C. Cold pressed oils are recovered from orange peel during the process of extracting juice from the oranges. The essence oils are collected while the single strength orange juice is being evaporated into juice concentrate (Crandall and Hendrix 2001). Although the cold pressed oil and the essence oil are very similar in composition, the essence oil does not contain the nonvolatile portion consisting of hydrocarbon waxes and some colored compounds found in the peel. The oils can be “folded” or concentrated by vacuum distillation to reduce the d-limonene content; thus a 5-fold oil has about 80% less d-limonene. Terpeneless oils are produced by further folding of cold pressed oils by removing the majority of the d-limonene, R Institute of Food Technologists doi: 10.1111/j.1750-3841.2008.00790.x

 C 2008

Further reproduction without permission is prohibited

then concentrating the volatile components by high-vacuum dis- Modified tube dilution assay tillation (Braddock 1999). This study reports the results of testing Citrus essential oils are not water soluble so it was necessary to various citrus essential oils against several Salmonella serotypes emulsify the oil and keep it stable for the duration of the test. Mann and strains using diffusion and tube dilution assays. and Markham (1998) used 0.15% agar as a stabilizer in a microtiter assay of tea tree oil using resazurin as a growth indicator. DonaldMaterials and Methods son and others (2005) found the microtiter assay to be unsuitable because of the carryover of the volatile oils from one well to anOrange oils other so they used a tube dilution technique using 0.2% agar. In our All of the citrus essential oils were obtained as commercially study, to maintain the oils in a homogeneous mixture in the test available products from Firmenich Citrus Center, Safety Harbor, media, 0.15% agar (BBL, Sparks, Md., U.S.A.) was added to nutrient Fla., U.S.A. and were stored per manufacturer’s recommendations broth. The mixture was allowed to boil for 1 min before dispensing at 4 ◦ C prior to use. At the time of testing, the oils had been in stor- into tubes and autoclaving. Triphenyl tetrazolium chloride (TTC) age for 1 mo. Oils tested included cold pressed Valencia orange oil was prepared as a growth indicator (Matilla 1987) by adding 1 g of terpeneless (C1), Valencia orange oil (C2), cold pressed orange ter- TTC (Amresco, Solon, Ohio, U.S.A.) to 100 mL of distilled water and penes (C3), high purity orange terpenes (C4), d-limonene (C5), ter- mixing without heat until in solution. The TTC solution was filter penes from orange essence (C6), and 5-fold concentrated Valencia sterilized and added to tempered medium at a rate of 1 mL/100 mL. orange oil (7). Serial dilutions were made by placing 10 mL of the NB with 0.15% agar (NBA) in the 1st test tube of a series and 5 mL in the remaining Cultures and media tubes. Oil (100 μL) was added to the 1st (10 mL) tube for an iniEleven strains/serotypes of Salmonella were obtained from the tial concentration of 10 μL/mL. The tubes were vortexed for 10 s culture collection of the Center for Food Safety—IFSE. Strains used to establish a stable emulsion. Serial dilutions were made by transwere S. Enteritidis 1773-92, S. Senftenberg ATCC 42845, S. Sen- ferring 5 mL of the emulsion to the next tube; this was repeated ftenberg 1402-94, S. Tennessee, S. Kentucky, S. Heidelberg ATCC for a total of 4 dilutions. Each dilution was vortexed for 10 s before 8326, S. enteritidis 13076, S. Montevideo G4639, S. Michigan, S. transferring the next aliquot. Five milliliters were removed from the Typhimurium var Copenhagen, and S. Stanley H1256. Cultures last tube and discarded so that all tubes had a constant volume. were propagated in nutrient broth (NB; Oxoid Ltd., Basingstoke, All tubes were inoculated with 50 μL of an overnight culture of the Hampshire, U.K.). A loop of bacteria from a TSA slant (PML Micro- serotype or strain of Salmonella being tested. All tubes were vorbiologicals, Wilsonville, Oreg., U.S.A.) was placed into a 10-mL tube texed for 10 s and incubated for 24 h at 37 ◦ C. Tubes were observed of sterile NB and incubated for 18 h at 37 ◦ C, after which a 100 μL for pink/red color indicating growth. The MIC was determined to aliquot was transferred into a fresh sterile 10 mL of NB, which was be the lowest concentration of oil that resulted in no growth (no subsequently incubated for an additional 18 h. color in the medium).

Disc diffusion assay

GC/MS analysis

Overnight cultures of the Salmonella prepared as described in the previous section were streaked on sterile Standard Methods Agar (EM Science, Gibbstown, N.J., U.S.A.) by dipping a sterile cotton swab into the culture and gently squeezing the swab against the inside of the tube to remove excess fluid. The swab was used to streak the agar plate to produce a lawn of growth by streaking the plate in 1 direction, then streaking at right angles to the first streaking, and finally streaking diagonally, and ending by using the swab to streak the outside diameter of the agar. Sterile 6-mm paper discs (Becton Dickson, Sparks, Md., U.S.A.) were aseptically placed on the agar, 4 discs per plate. The essential oil (EO) treatments were aseptically pipetted onto the discs, 10 μL per disc. A time period of 30 min was allowed for the oils to absorb and diffuse, and then the plates were inverted and incubated at 37 ◦ C for 24 h. Diameters of zones of inhibition were measured in mm and recorded. Analyses were performed a total of 3 times.

An analysis of the principal components in the most active EO was carried out using an Agilent 6890 gas chromotragraph (GC) connected to an Agilent 5973N mass spectrometer (MS) (Agilent Technologies Inc., Santa Clara, Calif., U.S.A.). The GC was equipped with fused silica capillary column, DB-5 with a 30 m × 0.25 mm × 0.25 μm film thickness. The GC/MS was operated at an initial temperature of 85 ◦ C, held for 6 min, ramp of 4 ◦ C/min up to 210 ◦ C, and held for 10 min. The carrier gas was helium at a flow rate of 2 mL/min. One microliter of sample was injected into a split injector with a 200:1 split with a temperature of 250 ◦ C. Identification of the compounds was by comparison of their mass spectra with those of the Wiley and 2 internal libraries.

T

Results and Discussion he disc diffusion assay results are presented in Table 1. Valencia orange oil terpeneless (C1), Valencia orange oil (C2), cold

Table 1 --- Zones of inhibition (in millimeters, including the diameter of 6-mm disk) of orange oils against Salmonella spp. (mean ± standard deviation).

S . Enteritidis 1773-92 S . Senftenberg 43845 S . Senftenberg 1402-94 S . Tennessee 825-94 S . Kentucky 1271-94 S . Heidelberg 8326 S . Enteritidis 13076 S . Montevideo G4639 S . Michigan S . Typhimurium(Copenhagen) S . Stanley H1256

C1

C2

C3

C4

C5

C6

C7

12.7 ± 2.3 8±2 7.3 ± 1.2 7.3 ± 1.2 6.7 ± 1.2 8±0 8.7 ± 2.3 6.7 ± 1.2 7.3 ± 1.2 7.3 ± 1.2 7.3 ± 1.2

6 6 6 6 6 6 6 6 6 6 6

6 6 6 6 6 6 6 6 6 6 6

23 ± 5.8 22 ± 3.5 21 ± 5 22.7 ± 4.2 22.7 ± 3.1 22 ± 5.3 29.3 ± 3.1 20 ± 2 22 ± 5.3 18.7 ± 5.8 23.3 ± 7

25 ± 2.3 23 ± 5.8 18.7 ± 3 19.3 ± 4.2 24 ± 5.3 21.3 ± 5.8 32 ± 6.9 23.3 ± 4.6 20 ± 5.3 23.3 ± 2.3 15.3 ± 3.1

30 ± 2 29 ± 3 27.3 ± 2.3 28 ± 2 30 ± 4 30 ± 5.3 34 ± 5.3 29.3 ± 3.1 28 ± 2 27.3 ± 6.1 28.7 ± 2.3

6 6 6 6 6 6 6 6 6 6 6

Vol. 73, Nr. 6, 2008—JOURNAL OF FOOD SCIENCE

M265

M: Food Microbiology & Safety

Orange oil against Salmonella spp. . . .

Orange oil against Salmonella spp. . . . pressed orange terpenes (C3), and 5-fold concentrated Valencia orange oil (C7) gave essentially no detectable effect on any of the Salmonellae tested. Terpenes from orange essence (C6) produced an average of 29.2 ± 3.7 mm zone of inhibition against all of the Salmonella tested. Kim and others (1995b) reported zones of inhibition ranging from 8 to 12 mm for several constituents of essential oil against S. Typhimurium. The 3 oils (C4, C5, and C6) that had an appreciable zone of inhibition on the disk diffusion assay were chosen to determine MIC using the tube method against all 11 Salmonella. The results are preTable 2 --- MIC (in percent, v/v) of orange oils against 11 Salmonella spp.

M: Food Microbiology & Safety

S . Enteritidis 1773-92 S . Senftenberg 43845 S . Senftenberg 1402-94 S . Tennessee 825-94 S . Kentucky 1271-94 S . Heidelberg 8326 S , Enteritidis 13076 S . Montevideo G4639 S . Michigan S . Typhimurium (Copenhagen) S . Stanley H1256

M266

C4

C5

C6

1 1 1 1 1 1 1 1 1 1 1

1 1 1 1 1 1 1 1 1 1 1

0.25 0.5 0.5 0.5 0.25 0.25 0.13 0.25 0.25 0.5 0.5

JOURNAL OF FOOD SCIENCE—Vol. 73, Nr. 6, 2008

sented in Table 2. Orange terpenes (C4) and d-limonene (C5) both exhibited MICs of 1%. The terpenes from orange essence (C6) had a range of MIC from 0.125% to 0.5% against the Salmonella tested. All Salmonella tested had an MIC of at least 0.5% with the terpenes from orange essence. Kim and others (1995a) found that the size of the zone of inhibition on the disc diffusion assay did not necessarily predict the MIC found in liquid medium for several citrus essential oil components they studied. For instance, citral produced small zones of inhibition against E. coli O157:H7 but had optimal inhibition against this organism in broth culture. They also noted that some compounds had activity against 1 species of bacteria but not against any other species tested. In our assays, we found that the zone of inhibition on the disc assay was predictive of which essential oils would subsequently have the most effect on the Salmonella. Thus, the terpenes from orange essence (C6) had the largest zone of inhibition on average for the Salmonella and also had the lowest MIC in most cases for the strains tested. There also did not appear to be a large difference in response between strains of a serotype or between serotypes of Salmonella tested. This reinforces the observation by Kim and others (1995a) that the essential oil components they tested were not equally useful against a wide variety of bacterial species, but had similar effects within serotypes or strains of the same species. In contrast to our findings that the compound oil fractions were more effective than the d-limonene alone, Dabbah and others

Orange oil against Salmonella spp. . . .

Conclusions

T

his study further confirms the possibility that essential oils from citrus could be an excellent antimicrobial agent against the foodborne bacteria Salmonella. Subba and others (1967) determined that orange and lemon oil had in vitro antibacterial effects on Salmonella and other foodborne microorganisms. Fisher and Phillips (2006) tested orange and lemon oils against several foodborne pathogens, although not Salmonella. They found that both orange and lemon oils exhibited in vitro effects against Listeria, Campylobacter, and E. coli O157:H7. However, some studies have indicated that it would take a much higher concentration of essential oils in actual food systems to inhibit bacteria. Tassou and Nychas (1996) found that it required 10 times as much mastic gum to achieve the same results in pork sausages as compared to in vitro. Smith-Palmer and others (1998) found that foods with higher lipid content would require more essential oil to inhibit bacterial growth. Fisher and Phillips (2006) found that even though lemon oil was effective against Campylobacter on the disc diffusion method, it had no effect when tested on chicken skin. Further studies will be needed to confirm the effects of these orange oils in vivo as well as using them as part of a multihurdle approach to food safety for meat products (Ricke and others 2005).

Table 3 --- Identification and percentage composition of components in orange juice essence oil. Retention time

Chemical name∗

% area under peak

1

2.98

myrcene

2.12

2

3.95

d-limonene

93.96

3

6.63

---