You are on page 1of 7

Research Article

Received: 11 April 2014 Revised: 3 July 2014 Accepted article published: 4 August 2014 Published online in Wiley Online Library: 26 August 2014

( DOI 10.1002/jsfa.6853

Dairy matrix effect on the transference

of rosemary (Rosmarinus officinalis) essential
oil compounds during cheese making
Armando Moro,a† Celia M Librán,b† M Isabel Berruga,b Manuel Carmonaa,c
and Amaya Zalacaina*

BACKGROUND: The use of aromatic plant extracts as ingredients may be compromised owing to low transference and activity
lack in food matrixes compared with in vitro trials. Rosemary essential oil (REO) was added to sheep milk to study the transference
of its compounds during the cheese-making process and to determine how cheese antimicrobial activity is modified.

RESULTS: The volatile characterization of dairy samples was performed using headspace stir bar sorptive extraction coupled to
gas chromatography/mass spectrometry (HS-SBSE/GC/MS) so that fat matrix interferences were reduced. This method detected
a decrease in volatile recovery concentration of 19.33% when REO was added to milk. A total recovery volatile yield of 62.51%
was measured from the initial quantification of milk to cheese, with hydrocarbon volatiles being transferred in a higher ratio
(64.88%) than oxygenated ones (58.74%). No effects were observed for REO in fortified cheese on the counts of native flora
necessary for ripening processes, but the total inhibition of Clostridium spp. was provoked

CONCLUSION: The study of active compound transference during cheese elaboration was achieved. The antimicrobial results
in fortified cheeses with REO showed a preventive effect in the case of clostridial species, which are responsible for late cheese
© 2014 Society of Chemical Industry

Keywords: rosemary oil; sheep milk; transference; HS-SBSE/GC/MS; antimicrobial activity

INTRODUCTION (polymers and surfactants),6,7 active modified atmosphere pack-

Today, consumers are significantly uneasy, demanding more infor- aging and other storage materials8,9 to improve their bioavailabil-
mation and safety alternatives concerning the use of artificial ity in food matrixes without the requirement for much extract.
additives (e.g. preservatives, colourants, flavourings and artificial The use of these methods in combination with other technolog-
sweeteners) in the food industry.1 – 3 As a result, a major number of ical approaches such as pulsed light, high pressure or magnetic
plant-derived food preservatives have been tested in recent years fields could minimize the organoleptic impact, which would avoid
to reduce the use of synthetic additives.3 In this regard, essential excessive changes in the overall scent and taste of food.3
oils (EOs) have been purposed as effective chemo strategies to EOs are most readily accepted in manufactured foods that are
prevent the growth of undesirable pathogens and to improve the traditionally associated with spices and aromatic plants (meat
shelf life and overall quality of food products.1,3 and fish dishes, cheese, soups and sauces, drinks and desserts
However, the current application of these extracts and their con-
stituents presents major limitations that must be addressed, such

as the lack of antimicrobial activity in food matrixes compared Correspondence to: Amaya Zalacain, Cátedra de Química Agrícola, Escuela Téc-
with in vitro trials. High concentrations of EOs are required to nica Superior de Ingenieros Agrónomos, Universidad de Castilla-La Mancha,
Avda de España, E-02071, Albacete, Spain. E-mail:
achieve sufficient antimicrobial activity to compensate for their
interactions with food matrix components (protein, fat, etc.).1,3,4 † These authors contributed equally to this work
Although EOs are generally recognized as safe (GRAS), the effect
of high concentrations of these materials on human health should a Cátedra de Química Agrícola, Escuela Técnica Superior de Ingenieros
Agrónomos, Universidad de Castilla-La Mancha, Avda de España, E-02071
be taken seriously.3 The resulting intense aroma of EOs should Albacete, Spain
be overcome, because even at low concentrations these aromas
can cause negative organoleptic effects when the threshold that b Departamento de Ciencia y Tecnología Agroforestal y Genética, ETSIA-IDR,
is accep to consumers is exceeded. Different strategies can be Universidad de Castilla-La Mancha, Campus Universitario, E-02071 Albacete,
used to solve this problem; for example, EOs and correspondent
antimicrobial components are currently being incorporated as c Albacete Science and Technology Park Foundation, Universidad de Castilla-La

a part of edible biofilms2,5 through microencapsulation systems Mancha, Campus Universitario, 02071 Albacete, Spain

J Sci Food Agric 2015; 95: 1507–1513 © 2014 Society of Chemical Industry A Moro et al.

that contain fruit and dairy products).3 In the case of some dairy
Table 1. Chemical composition of milk, cheese and whey, and pro-
commodities such as cheese, certain plants have been used as duction and yield obtained during cheese making
flavourings and natural preservatives since ancient times.10 – 16
Rosemary is traditionally used as a natural cheese cover in Spain, Parameter Milk Cheese Whey
particularly in the Castilla-La Mancha region, which results in a
Total solids (%) 17.77 68.81 ± 0.37 7.80 ± 0.32
product that is much appreciated by consumers owing to its
Fat (%) 6.82 35.99 ± 0.11 1.10 ± 0.13
rosemary volatile profile.
Protein (%) 5.55 25.71 ± 0.49 1.00 ± 0.27
EC Regulation No. 1334/2008, which concerns flavourings and
Production (kg per vat) 30.9 ± 0.00 5.75 ± 0.02 25.15 ± 0.02
certain food ingredients with flavouring properties for use in foods,
Yield (%) – 18.60 ± 0.08 81.40 ± 0.08
does not report any volatile compound from rosemary essential
oil (REO) among the undesirable substances. Rosemary has been Values are mean ± SE.
used as a flavouring agent in many food matrixes,17 and its in
vitro antimicrobial activity has been widely tested.18,19 In the case
of cheese matrixes, the antimicrobial activity of REO has been starter culture containing Lactococcus lactis ssp. lactis, Lactococcus
successfully tested in vitro against the microorganisms (Escherichia lactis ssp. cremoris, Lactococcus lactis ssp. lactis biovar. diacetylac-
coli and Clostridium tyrobutyricum) that are responsible for early tis and Streptococcus thermophilus (CHOOZIT MA4001 Cheese Cul-
and late cheese blowing,20 as well as some fungi (Penicillium tures, Danisco, Sassenage, France) was added at 5 Danisco culture
verrucosum).21 units (DCU) per 100 mL. The milk was held at 30 ∘ C for 20 min and
The aim of this research is to determine the degree of interfer- then commercial rennet (CARLINA Animal Rennet, Danisco) was
ence between the matrix (milk, whey and cheese) and the con- added as a coagulation agent. After 50 min, the curd was cut into
stituents of Rosmarinus officinalis L. EO when added to milk. In the 8–10 mm cubes and, to separate whey from curd, a new temper-
resulting cheese, the antimicrobial effect of REO against certain ature gradient, which started at 30 ∘ C and finished at 35 ∘ C and
microbial cheese spoilers will be evaluated. was coupled with shaking movements for 25 min, was performed.
Curds were transferred into 3 kg cheese moulds and pressed in
three steps (20 min at 1.5 bar, 45 min at 2 bar and 45 min at 2.5 bar)
EXPERIMENTAL until the cheese pH reached 5.4–5.5. Unripe cheeses were incu-
Plant material and EO extraction bated in brine (200 g L−1 NaCl adjusted to pH 5.4) for 24 h. Cheeses
Rosmarinus officinalis L. (rosemary) plants cultivated in the (control and REO-fortified ones) were ripened for 5 months at 12 ∘ C
Castilla-La Mancha region of Spain in March 2010 were har- and 80% humidity before analysis. Whey samples were frozen at
vested at their optimal flowering time and dried at 40 ∘ C in the −20 ∘ C for later analysis.
dark (92 g kg−1 moisture content). The plant material was packed The chemical compositions of the milk and cheese were assessed
in sealed plastic bags and stored under dark conditions at room using direct determination equipment (MilkoScan Minor and Foss
temperature until analysis. FoodScan Dairy Analyzer respectively, FOSS, Hillerød, Denmark).
REO was obtained by solvent-free microwave extraction (SFME) Whey samples were analyzed by AOAC methodology22 for total
with an NEOS® apparatus (Milestone, Sorisole, Italy). The extrac- solids and protein fractions, and fat was analyzed by Gerber’s
tion power was set at 600 W for 5 min and then at 250 W for 30 min. method as described by Casado.23 The values obtained, as well as
A rotating microwave diffuser ensured homogeneous microwave some cheese-making parameters, are presented in Table 1.
distribution throughout the plasma-coated polytetrafluoroethy-
lene (PTFE) cavity. The temperature was monitored by an exter- Chemical characterization
nal infrared sensor. Constant conditions of temperature and water Colour analysis of dairy samples
were guaranteed by the reflux of condensed water, which was Colour was determined in triplicate using a Minolta CR-400
achieved by a circulating cooling system at −2 ∘ C. A 150 g vege- colorimeter (Osaka, Japan) with a D65 illuminant and a 10∘
tal sample was placed in the reactor with 250 mL of Milli-Q water. observer, with specific liquid or solid adapters for milk and whey
Exhaustive extraction of REO was obtained in 35 min. For the or cheese measurements. The calibration was performed with a
antimicrobial activity test, the REO was filtered through 0.2 μm Minolta white plate (Y = 93.1, No. plate = 11333110, x = 0.313460,
PTFE syringe filters (Millipore, Madrid, Spain) to ensure the absence y = 0.3195). The results were expressed as the mean of three
of microorganisms. measurements in CIE L*a*b* space. The L* coordinate corresponds
to the brightness (0 = black, 100 = white), the a* coordinate is the
Addition of REO to milk and cheese elaboration red index (−a* = green, +a* = red) and the b* coordinate is the
Bulk tank milk from the Manchega breed experimental ewe flock of yellow index (−b* = blue, +b* = yellow).
Castilla-La Mancha University was employed for cheese manufac-
turing. The sheep milk used had a pH of 6.66, a somatic cell count of REO and dairy sample volatile extractions followed by gas
5.7 log cells mL−1 and an aerobic plate count of 5.9 log cells mL−1 . chromatography/mass spectrometry (GC/MS) analyses
The compositional characteristics of the milk are given in Table 1. The REO obtained was directly injected (0.2 μL) into the gas chro-
Each vat of 30 L of sheep milk was fortified with REO at a final con- matograph following the methodology of Moro et al.21
centration of 0.2 g kg−1 (215 mg L−1 ). A commercial food emulsifier, Milk and cheese volatile extraction was performed by the
previously dissolved in Tween®-20 (Panreac Synthesis, Barcelona, headspace stir bar sorptive extraction (HS-SBSE) method. In the
Spain) at 1:1 (v/v), was added to ensure a uniform solution of EOs case of milk volatile extraction, 10 mL liquid dairy samples (milk
in the milk. The same proportion of emulsifying agent/Tween®-20 and whey) were pipetted separately into headspace glass vials,
was added to the control vat. The milk was heated at 20 ∘ C for whereas cheese volatile extraction was performed following the

30 min for better oil solubilization. For cheese manufacturing, a methodology of Licón et al.24 © 2014 Society of Chemical Industry J Sci Food Agric 2015; 95: 1507–1513
Transference of rosemary EO in dairy matrixes

Table 2. Culture medium information

Medium Provider Microorganism Incubation conditions

Plate count agar (PCA) Biokar Diagnosis (Barcelona, Spain) Total aerobic bacteria 32 ∘ C, 48 h
M17 Biokar Diagnosis (Barcelona, Spain) Lactic acid bacteria 37 ∘ C, 48 h
Brilliant green bile (BGB) Pronadisa Conda (Madrid, Spain) Coliforms 37 ∘ C, 24 h
Reinforced clostridial agar (RCA) Oxoid (Basingstoke, UK) Clostridium spp. 37 ∘ C, 48 h, anaerobic
Potato dextrose agar (PDA) Merck (Darmstadt, Germany) Moulds/yeasts 25 ∘ C, 96 h

For all dairy samples, headspace glass vials were affixed with where Xi indicates the presence of each compound in cheese and
inserts for headspace exposition and supplemented with a X indicates the presence of the same compound in milk. Dairy
1 × 10−3 g kg−1 aqueous solution of the internal standard ethyl samples were analyzed in triplicate.
octanoate (Aldrich Chemical Co., Milwaukee, WI, USA). A poly-
dimethylsiloxane (PDMS)-coated stir bar (0.5 mm film thickness,
Microbial analysis
10 mm length in liquid samples and 20 mm length in cheese
The estimation of several microbial groups was performed in
samples; Twister, Gersterl GmbH, Mülheim an der Ruhr, Germany)
control and REO-fortified ripened cheeses. Samples (10 g) of each
was placed into the insert, and headspace vials were sealed with
cheese were aseptically diluted 1:10 (w/v) with buffered peptone
an aluminium crimp cap. Before analysis, the glass insert and
water (Scharlau, Barcelona, Spain), and serial decimal dilutions
vials were thoroughly cleaned and heat conditioned at 110 ∘ C
were seeded with an automatic sow (Eddy Jet v1.23, IUL, Barcelona,
to avoid any odorous contamination. The extraction of volatile
Spain), to determine the counts of total aerobic bacteria, lactic
samples was performed under stirring at 1000 rpm for 120 min
acid bacteria, coliforms, Clostridium spp. and moulds/yeasts at
(milk and whey) or 240 min (cheese) at 45 ∘ C. The PDMS stir bars
their optimal growing conditions. Culture medium information is
were rinsed with distilled water, dried with cellulose tissue and
summarized in Table 2. Microbial growth was estimated using an
finally transferred into thermal desorption tubes for the GC/MS
automatic Countermat Flash 4.2 (UIL). Results were expressed as
log colony-forming units (CFU) g−1 .
The volatiles from dairy samples that were extracted into PDMS
stir bars were desorbed in an automated thermal desorption sys-
tem (Turbo Matrix ATM, PerkinElmer, Norwalk, CT, USA) under Statistical analysis
the following conditions: oven temperature, 280 ∘ C; desorption Descriptive (mean ± standard error of mean (SE)) and variance
time, 5 min; cold trap temperature, −30 ∘ C; helium inlet flow (ANOVA, P < 0.001) analyses coupled with Tukey’s test (P < 0.05)
rate, 45 mL min−1 . The volatiles were transferred into a Varian were performed to determine group differences among all results
CP-800 gas chromatograph (GC) equipped with a Saturn 2200 using SPSS v19 (IBM Statistics, Chicago, IL, USA).
ion trap mass spectrometer (MS) (Palo Alto, CA, USA) that was
provided with an Elite-Volatiles Specialty phase capillary column
(30 m × 0.25 mm i.d., 1.4 μm film thickness; PerkinElmer, Shelton, RESULTS AND DISCUSSION
CT, USA). The column temperature was set at 35 ∘ C for 2 min The REO antimicrobial activities in in vitro trials against the
and then raised at 5 ∘ C min−1 to 240 ∘ C and held for 5 min. The microorganism responsible for some cheese defects have been
detector temperature was 250 ∘ C and the helium carrier gas flow successfully tested.20,21 However, there have not been any studies
rate was 1 mL min−1 . The electron ionization (EI) mode at 70 eV on how the active rosemary compounds interact with the milk and
was used for the MS analysis. The mass range varied from m/z cheese matrixes, as well as whether such well-known antimicrobial
35 to 300. activity remains in the final product.
To avoid matrix interferences between the REO and dairy matrix
volatiles, the MS identification of volatiles was performed in
single-ion-monitoring (SIM) mode using their characteristic m/z REO extraction and its composition
values and by comparison of their mass spectra with those The REO was obtained from rosemary leaves by SFME, which is
reported in the NIST/ADAMS library data of the GC/MS system based on a combination of microwave heating and distillation. The
as described by Adams.25 The identities of the oil components use of this technique ensures that the concerned antimicrobial
were established from the GC retention indices (relative to Kovats activity depends only on the volatile fraction of the REO and not on
indices) and by comparison with those reported by Adams.25 certain artefacts that are generated when solvents are used with
Quantification was carried out in scan mode and expressed as the other extraction procedures.26,27 In fact, the extract obtained with
relative area using the correction factor for the internal standard organic solvents, supercritical CO2 or other agents should not be
(ethyl octanoate) area. The results of each volatile compound that called an EO according to ISO 9237. The extraction yield obtained
was transferred to the dairy matrix were expressed as relative con- was 1.15% and the density of the EO was 0.92 g mL−1 .
centration area (g kg−1 ) using the internal standard correction fac- Approximately 95.7% of the EO volatile fraction was quantified,
with 22 volatile compounds being identified. These compounds
tor in all samples. Then the transference ratio or recovery yield (%)
were classified into different chemical families, i.e. monoterpene
from milk to cheese of each compound that was found was calcu-
hydrocarbons, oxygenated monoterpenes and sesquiterpene
lated by the equation
hydrocarbons. The oxygenated monoterpenes and hydrocarbons
[ ( ) ( )] (monoterpenes and sesquiterpenes) accounted for 42.15 and

recovery yield (%) = Xi g kg−1 ∕X g kg−1 × 100 (1) 55.38% respectively of the total volatile composition (Table 3).27,28

J Sci Food Agric 2015; 95: 1507–1513 © 2014 Society of Chemical Industry A Moro et al.

impact, as well as to maintain the antimicrobial activity and a safe

Table 3. Chemical composition of REO obtained by SFME
dose. Tween®-20 was selected as a polysorbate surfactant whose
Retention index Volatile compound Content (%) stability and relative lack of toxicity allow it to be used as a deter-
gent and emulsifier for culinary, scientific and pharmacological
Monoterpene hydrocarbons purposes.36,37 Specific surfactant actions are required to improve
928 𝛼-Thujene 1.14 ± 0.22 the affinity of the matrix for volatile compounds, particularly ter-
937* 𝜶-Pinene 23.35 ± 0.10 penes.
951 Camphene 11.62 ± 0.01 It is known that the analysis of dairy matrixes is rather
972 Sabinene 0.20 ± 0.05 complicated because of their high fat (6.82%) and protein
977* 𝛽-Pinene 4.62 ± 0.08 (5.55%) contents. The HS-SBSE technique23,38 – 40 was chosen
988 Myrcene 2.83 ± 0.00 here in order to avoid direct interactions between the dairy
1000* 𝛼-Phellandrene 0.48 ± 0.41 and EO volatile compounds adsorbed on the PDMS coat-
1008 𝛿-3-Carene 1.49 ± 0.01 ing device. The REO volatile concentration detected in the
1015* 𝛼-Terpinene 0.70 ± 0.16 fortified milk was 173.43 × 10−3 g kg−1 , representing an ini-
1022* p-Cymene 2.17 ± 0.05 tial decrease in concentration of 19.33%, because the milk
1025* Limonene 2.10 ± 0.20 was fortified with 215 × 10−3 g kg−1 REO. This concentration
1054* 𝛾-Terpinene 0.84 ± 0.01 difference can be explained by the hydrophobic behaviour
1087 Terpinolene 0.58 ± 0.03 of most terpenes, which induces a lower compound parti-
Oxygenated monoterpenes tion in the headspace and decreases their recovery into the
1027* 1,8-Cineole 19.21 ± 0.02 sorbent-coated rod.41
1095* Linalool 0.87 ± 0.00
1144* Camphor 13.59 ± 0.00
1167 Borneol 2.23 ± 0.03 Cheese and whey
1175* 1-Terpinen-4-ol 1.31 ± 0.00 After cheese elaboration, a total recovery volatile yield of 62.51%
1186* 𝛼-Terpineol 2.12 ± 0.01 (108.4 × 10−3 g kg−1 ) was quantified (Table 4). This yield trans-
1204 Verbenone 1.66 ± 0.02 ference could be considered successful, taking into account the
1284 Bornyl acetate 1.16 ± 0.11 methodology limitations (which interfere with the quantitative
Sesquiterpene hydrocarbons analysis) and the lack of a large amount of REO volatile compounds
1418 𝛽-Caryophyllene 3.26 ± 0.62 during the whey draining phase (29.16 × 10−3 g kg−1 ). A total
Total oxygenated 42.15 volatile concentration of 137.57 × 10−3 g kg−1 within consecutive
Total hydrocarbons 55.38 stages of the cheese elaboration process (29.16 × 10−3 g kg−1
Total detected 97.5 in whey and 108.41 × 10−3 g kg−1 in cheese) was detected,
with a yield of 79.32% from the initial REO milk concentration
Values are mean ± SE from GC/MS sample analysis in triplicate. (173.43 × 10−3 g kg−1 ).
* Retention indices (as Kovats indices) of available standards.
The group of compounds that has been transferred from REO
milk into cheese at a higher rate is the hydrocarbon chemical
family, with a mean value of transference of 64.88%, in com-
Among these chemical families, the major compounds identi- parison with 58.74% transference for the oxygenated family.
fied were are 𝛼-pinene, 1,8-cineole, camphor and camphene. Only The higher retention rate for oxygenated compounds in the
𝛽-caryophyllene was identified as a sesquiterpene hydrocarbon food matrix can be explained by possible interactions (hydro-
in the oil extract. Differences in the EO composition of rosemary gen bonding) between their hydroxyl groups and the receptor
according to Gachkar et al.29 could be attributed to climatic effects groups of proteins, such as NH and CO, in cheese.42 In terms
on plants grown in different habitats and also to effects of genet- of individual compounds, the sesquiterpenoid hydrocarbon
ics, plant vegetative stage, soil characteristics, diseases, predator 𝛽-caryophyllene had the highest recovery yield (82.75%) in cheese
attacks and other factors. The volatile profile found here is iden- samples. In contrast, linalool had the lowest value detected, with
tical to that described by Santoyo et al.30 and Angioni et al.31 for a transference of 35.44%. The other volatiles ranged between
Mediterranean samples. 56.47% (1,8-cineole) and 77.95% (p-cymene) transference. In
terms of concentration, four major compounds (𝛼-pinene, cam-
Volatile composition of dairy samples phene, 1,8-cineole and camphor) were detected in all samples
Milk analyzed.
Whey characterization revealed the same four main volatiles
As previously reported by Tajkarimi et al.,3 the normal concentra-
as in cheese samples, with 𝛼-pinene and camphene as the com-
tion range for spices and herbs used in food systems is between
pounds with higher transference values from milk, at 19.37 and
0.05 and 0.1%. EOs may have an undesirable impact on cheese
19.85% respectively. In fact, the bulk oil composition was identified
sensory properties by modifying the dynamics or proper activity
in each sampling, with the exception of myrcene and verbenone,
of the microbial ecosystem during cheese making and ripening.32
which were not found in whey samples.
This hypothesis results from indirect observations in several tri-
als of hard cooked cheeses33 – 35 and experiments performed by
Tornambé et al.,32 where EO concentration levels higher than Colour analysis in dairy samples
10 g kg−1 resulted in a high sensory impact and consequent rejec- The parameters L*, a* and b* were measured in dairy control
tion by consumers. In this study, an REO concentration of 0.2 g kg−1 samples (Table 5). The lightness (L*) of milk and whey samples
was chosen to study the transference of EO volatile compounds decreased with respect to the control but increased in the case

during the cheese-making process to prevent an excessive sensory of cheese samples. This result was expected, considering that the © 2014 Society of Chemical Industry J Sci Food Agric 2015; 95: 1507–1513
Transference of rosemary EO in dairy matrixes

Table 4. Volatile composition of REO-fortified milk, cheese and whey, and volatile recovery yield from milk to cheese

Content (10−3 g kg−1 )

Retention index Volatile compound REO milk REO cheese REO whey Recovery yield (%)

Monoterpene hydrocarbons
928 𝛼-Thujene 1.32 ± 0.16 0.90 ± 0.00 0.23 ± 0.04 67.82
937* 𝜶-Pinene 54.55 ± 5.03 35.82 ± 0.57 10.57 ± 1.37 65.67
951 Camphene 26.54 ± 1.60 15.79 ± 0.49 5.27 ± 0.62 59.51
972 Sabinene 0.07 ± 0.01 0.05 ± 0.01 0.03 ± 0.00 62.50
977* 𝛽-Pinene 7.30 ± 0.14 4.44 ± 0.27 1.24 ± 0.07 60.81
988 Myrcene 1.78 ± 0.21 1.17 ± 0.02 0.00 65.73
1000* 𝛼-Phellandrene 0.79 ± 0.05 0.57 ± 0.04 0.17 ± 0.01 71.34
1008 𝛿-3-Carene 1.43 ± 0.10 0.95 ± 0.01 0.30 ± 0.03 65.89
1015* 𝛼-Terpinene 1.19 ± 0.06 0.80 ± 0.04 0.27 ± 0.01 66.73
1022* p-Cymene 5.88 ± 0.74 4.58 ± 0.77 1.35 ± 0.07 77.95
1025* Limonene 3.70 ± 0.01 2.65 ± 0.00 1.42 ± 0.00 71.62
1054* 𝛾-Terpinene 1.07 ± 0.01 0.71 ± 0.05 0.28 ± 0.00 66.28
1087 Terpinolene 0.89 ± 0.05 0.60 ± 0.00 0.27 ± 0.01 67.12
Oxygenated monoterpenes
1027* 1,8-Cineole 47.49 ± 1.62 26.82 ± 2.10 1.30 ± 0.13 56.47
1095* Linalool 1.12 ± 0.04 0.40 ± 0.03 0.22 ± 0.01 35.44
1144* Camphor 14.56 ± 0.61 9.41 ± 0.38 4.86 ± 0.15 64.64
1167 Borneol 1.46 ± 0.07 1.14 ± 0.15 0.67 ± 0.01 77.81
1175* 1-Terpinen-4-ol 0.39 ± 0.03 0.25 ± 0.02 0.13 ± 0.01 64.39
1186* 𝛼-Terpineol 0.53 ± 0.04 0.41 ± 0.03 0.28 ± 0.03 77.15
1204 Verbenone 0.53 ± 0.08 0.35 ± 0.04 0.00 65.12
1284 Bornyl acetate 0.39 ± 0.06 0.27 ± 0.01 0.14 ± 0.00 69.90
Sesquiterpene hydrocarbons
1418 𝛽-caryophyllene 0.43 ± 0.05 0.36 ± 0.04 0.16 ± 0.02 82.75
Total oxygenated 66.48 39.05 7.60 58.74
Total hydrocarbons 106.94 69.39 21.56 64.88
Total detected 173.43 108.41 29.16 62.51

Values are mean ± SE from HS-SBSE/GC/MS sample analysis in triplicate.

* Retention indices (as Kovats indices) of available standards.

Antimicrobial activity
Table 5. Effect of REO addition on CIE L*a*b* colour
Table 6 shows the results of the enumeration of the microbial con-
Control REO-fortified tent of cheeses after 5 months of ripening. No negative effects
Sample Coordinate sample sample ANOVA of REO were observed on the counts of total aerobic bacteria,
lactic acid bacteria and moulds/yeasts. However, the addition of
Milk L* 85.84 ± 0.24 79.44 ± 0.39 *
REO into milk for cheese elaboration seemed to have a benefi-
a* −1.97 ± 0.29 −2.51 ± 0.80 *
cial preventive effect in the case of clostridial species, which are
b* 1.64 ± 0.30 1.15 ± 0.17 NS responsible for late cheese blowing, with a reduction of more
Whey L* 80.97 ± 0.06 75.41 ± 1.73 NS than 3 log units being observed in comparison with control
a* −0.49 ± 0.17 0.06 ± 0.17 NS
cheeses. One of the main concerns regarding the use of EOs in
b* 2.13 ± 0.02 2.60 ± 0.04 *
fermented foods is the potential negative effect against starters
Cheese L* 79.28 ± 0.59 81.26 ± 0.17 *
or native flora that are necessary for the fermentative and ripen-
a* −3.11 ± 0.13 −3.22 ± 0.03 NS
ing processes. In the case of REO, its active compounds had no
b* 15.74 ± 0.33 16.43 ± 0.14 NS
inhibitory effect on the lactic flora after 5 months of ripening,
Values are mean ± SE. while this EO was simultaneously able to prevent the growth of
* P < 0.01; NS, not statistically significant.
Clostridium spp.
Generally, the EO concentrations in food matrixes must be
greater than under in vitro conditions, which may be related to
REO was darker in colour than the milk and whey43 but not darker the more complex nature of foods.44 In a previous in vitro study,17
than the cheese. In the case of fortified REO milk samples, a* REO assayed at a similar concentration (0.020% v/v) did not show
and b* decreased, whereas these parameters increased in whey any antimicrobial inhibitory effect; however, in the cheese, this REO
samples. Cheese samples from fortified milk showed a decrease was able to inhibit C. tyrobutyricum growth. An opposite behaviour
in a*, whereas b* increased; however, these differences were not was observed in the case of moulds/yeasts, where this concentra-

statistically significant (P < 0.05). tion assayed under in vitro conditions provoked a 36% inhibition

J Sci Food Agric 2015; 95: 1507–1513 © 2014 Society of Chemical Industry A Moro et al.

5 Avila-Sosa R, Palou E, Jiménez Munguía MT, Nevárez-Moorillón GV,

Table 6. Microbiological counts (log CFU g−1 ) of cheese supple- Navarro Cruz AR and López-Malo A, Antifungal activity by vapor
mented with REO after 5 months of ripening contact of essential oils added to amaranth, chitosan, or starch
edible films. Int J Food Microbiol 153:66–72 (2012).
Microorganism Control cheese REO cheese ANOVA 6 Rao J and McClements DJ, Lemon oil solubilization in mixed surfac-
tant solutions: rationalizing microemulsion and nanoemulsion for-
Total aerobic bacteria 6.24 ± 0.21 6.87 ± 0.02 NS mation. Food Hydrocolloids 26:268–276 (2012).
Lactic acid bacteria 6.21 ± 0.43 6.72 ± 0.06 NS 7 Varona S, Rodríguez Rojo S, Martín Á, Cocero MJ, Serra AT, Crespo
Coliforms ND ND NS T, et al., Antimicrobial activity of lavandin essential oil formula-
Clostridium spp. 3.53 ± 0.12 ND *** tions against three pathogenic food-borne bacteria. Ind Crops Prod
42:243–250 (2013).
Moulds/yeasts 1.19 ± 0.67 1.33 ± 0.12 NS 8 Serrano M, Martínez-Romero D, Guillén F, Valverde JM, Zapata PJ,
Values are mean ± SE; ND, not detected. Castillo S, et al., The addition of essential oils to MAP as a tool
*** P < 0.001; NS, not statistically significant. to maintain the overall quality of fruits. Trends Food Sci Technol
19:464–471 (2008).
9 Tyagi AK, Malik A, Gottardi D and Guerzoni ME, Essential oil vapour and
negative air ions: a novel tool for food preservation. Trends Food Sci
of P. verrucosum growth but was not able to inhibit the growth of Technol 26:99–113 (2012).
fungal species in cheese.21 10 Di Cagno R, Buchin S, De Candia S, De Angelis M, Fox PF and Gobbetti M,
Characterization of Italian cheeses ripened under nonconventional
The importance of the food matrix on the final antimicrobial conditions. J Dairy Sci 90:2689–2704 (2007).
effect of EOs has been confirmed in this study, where the different 11 Gammariello D, Di Giulio S, Conte A and Del Nobile MA, Effects of
behaviour of an EO in regard to in vitro assays and the potential natural compounds on microbial safety and sensory quality of Fior
use of REO to prevent late cheese blowing defects have been di Latte cheese, a typical Italian cheese. J Dairy Sci 91:4138–4146
observed. (2008).
12 Gandomi H, Misaghi A, Basti AA, Bokaei S, Khosravi A, Abbasifar A, et al.,
Effect of Zataria multiflora Boiss. essential oil on growth and aflatoxin
formation by Aspergillus flavus in culture media and cheese. Food
CONCLUSIONS Chem Toxicol 47:2397–2400 (2009).
The addition of REO to sheep milk to study its bioactive com- 13 Licón CC, Carmona M, Molina A and Berruga MI, Chemical, microbiolog-
ical, textural, color, and sensory characteristics of pressed ewe milk
pound transference during cheese elaboration was achieved. In cheeses with saffron (Crocus sativus L.) during ripening. J Dairy Sci
terms of volatile compound quantification, the REO compounds 95:4263–4274 (2012).
𝛼-pinene, camphene, 1,8-cineole and camphor were detected as 14 Licón CC, Carmona M, Rubio R, Molina A and Berruga MI, Preliminary
major components in all samples analysed. A mean total recovery study of saffron (Crocus sativus L. stigmas) color extraction in a dairy
matrix. Dyes Pigm 92:1355–1360 (2012).
yield of 62.51% for REO active compounds was obtained in forti-
15 Smith-Palmer A, Stewart J and Fyfe L, The potential application of
fied cheese samples, with the transference of hydrocarbon chem- plant essential oils as natural food preservatives in soft cheese. Food
ical compounds being more effective than that of oxygenated Microbiol 18:463–470 (2001).
compounds. Regarding the antimicrobial effect of REO in for- 16 Vázquez BI, Fente C, Franco CM, Vázquez MJ and Cepeda A, Inhibitory
tified cheeses, its use provoked the total inhibition of Clostrid- effects of eugenol and thymol on Penicillium citrinum strains in
culture media and cheese. Int J Food Microbiol 67:157–163 (2001).
ium spp. without affecting the growth of lactic acid bacteria, 17 Hayaloglu AA and Farkye NY, Cheese with added herbs, spices and
which supports the use of this plant to prevent late cheese condiments, in Encyclopedia of Dairy Sciences, ed. by John WF.
blowing defects. However, further studies are required to eval- Academic Press, San Diego, CA, pp. 783–789 (2011).
uate the economic costs, legislation, organoleptic effects, safety 18 Jordán MJ, Lax V, Rota MC, Lorán S and Sotomayor JA, Effect of
and practical effectiveness of the application of this EO in dairy bioclimatic area on the essential oil composition and antibacterial
activity of Rosmarinus officinalis L. Food Control 30:463–468 (2013).
products. 19 Teixeira B, Marques A, Ramos C, Neng NR, Nogueira JMF, Saraiva
JA, et al., Chemical composition and antibacterial and antioxidant
properties of commercial essential oils. Ind Crops Prod 43:587–595
20 Librán CM, Moro A, Zalacain A, Molina A, Carmona M and Berruga MI,
The research that led to these results received funding Potential application of aromatic plant extracts to prevent cheese
from the European Union’s Seventh Framework Programme, blowing. World J Microbiol Biotechnol 29:1179–1188 (2013).
which is managed by the REA – Research Executive Agency 21 Moro A, Librán CM, Berruga MI, Zalacain A and Carmona M, Mycotoxi- (FP7/2007–2013) under grant cogenic fungal inhibition by innovative cheese cover with aromatic
agreement number 243638. We thank the entire project con- plants. J Sci Food Agric 93:1112–1118 (2013).
22 AOAC (Association of Official Analytical Chemists), Official Methods of
sortium and the project officer Ms Ryniak for their support and Analysis of AOAC International. AOAC International, Gaithersburg,
collaboration. MD (1998).
23 Casado P, Método butirométrico de Gerber, in Guía para el análisis
químico de la leche y los derivados lácteos, ed. by Casado P. Ayala,
REFERENCES Madrid, pp. 79–88 (1991).
1 Burt S, Essential oils: their antibacterial properties and potential appli- 24 Licón CC, Hurtado de Mendoza J, Maggi L, Berruga MI, Martín Aranda
cations in foods – a review. Int J Food Microbiol 94:223–253 (2004). RM and Carmona M, Optimization of headspace sorptive extraction
2 Shim SM, Seo SH, Lee Y, Moon GI, Kim MS and Park JH, Consumers’ for the analysis of volatiles in pressed ewes’ milk cheese. Int Dairy J
knowledge and safety perceptions of food additives: evaluation on 23:53–61 (2012).
the effectiveness of transmitting information on preservatives. Food 25 Adams RP, Identification of Essential Oil Components by Gas Chromatog-
Control 22:1054–1060 (2011). raphy/Mass Spectometry. Allured Publishing, Carol Stream, IL (2009).
3 Tajkarimi MM, Ibrahim SA and Cliver DO, Antimicrobial herb and spice 26 Paré JRJ and Bélanger JMR, Techniques and instrumentation in analyt-
compounds in food. Food Control 21:1199–1218 (2010). ical chemistry, in Instrumental Methods in Food Analysis, ed. by Paré
4 Gutierrez J, Barry-Ryan C and Bourke P, The antimicrobial efficacy of JRJ and Bélanger JMR. Elsevier, Amsterdam, pp. 395–420 (1997).
plant essential oil combinations and interactions with food ingredi- 27 Okoh OO, Sadimenko AP and Afolayan AJ, Comparative evaluation

ents. Int J Food Microbiol 124:91–97 (2008). of the antibacterial activities of the essential oils of Rosmarinus © 2014 Society of Chemical Industry J Sci Food Agric 2015; 95: 1507–1513
Transference of rosemary EO in dairy matrixes

officinalis L. obtained by hydrodistillation and solvent free 36 Ivanković T and Hrenović J, Surfactants in the environment. Arh Hig
microwave extraction methods. Food Chem 120:308–312 (2010). Rada Toksikol 61:95–110 (2010).
28 Ruberto G and Baratta MT, Antioxidant activity of selected essential 37 Jahan K, Balzer S and Mosto P, Toxicity of nonionic surfactants. Proc.
oil components in two lipid model systems. Food Chem 69:167–174 Environmental Toxicology Conf ., Granada, pp. 281–292 (2008).
(2000). 38 Buettner A, A selective and sensitive approach to characterize
29 Gachkar L, Yadegari D, Rezaei MB, Taghizadeh M, Astaneh SA and odour-active and volatile constituents in small-scale human milk
Rasooli I, Chemical and biological characteristics of Cuminum samples. Flav Fragr J 22:465–473 (2007).
cyminum and Rosmarinus officinalis essential oils. Food Chem 39 Huang X, Qiu N and Yuan D, Simple and sensitive monitoring of sul-
102:898–904 (2007). fonamide veterinary residues in milk by stir bar sorptive extraction
30 Santoyo S, Cavero S, Jaime L, Ibañez E, Señoráns FJ and Reglero based on monolithic material and high performance liquid chro-
G, Chemical composition and antimicrobial activity of Rosmarinus matography analysis. J Chromatogr A 1216:8240–8245 (2009).
officinalis L. essential oil obtained via supercritical fluid extraction. 40 Prieto A, Basauri O, Rodil R, Usobiaga A, Fernández LA, Etxebarria N,
J Food Protect 68:790–795 (2005). et al., Stir-bar sorptive extraction: a view on method optimisation,
31 Angioni A, Barra A, Cereti E, Barile D, Coïsson JD, Arlorio M, et al., novel applications, limitations and potential solutions. J Chromatogr
Chemical composition, plant genetic differences, antimicrobial and A 1217:2642–2666 (2010).
antifungal activity investigation of the essential oil of Rosmarinus 41 Paraskevopoulou A, Tsoukala A and Kiosseoglou V, Monitoring
officinalis L. J Agric Food Chem 52:3530–3535 (2004). air/liquid partition of mastic gum oil volatiles in model alcoholic
32 Tornambé G, Cornu A, Verdier-Metz I, Pradel P, Kondjoyan N, Figueredo beverage emulsions: effect of emulsion composition and oil droplet
G, et al., Addition of pasture plant essential oil in milk: influence size. Food Hydrocolloids 23:1139–1148 (2009).
on chemical and sensory properties of milk and cheese. J Dairy Sci 42 Bartolomé B, Estrella I and Hernández MT, Interaction of low molecular
91:58–69 (2008). weight phenolics with proteins (BSA). J Food Sci 65:617–621 (2000).
33 Buchin S, Martin B, Dupont D, Bornard A and Achilleos C, Influence of 43 Boroski M, Giroux HJ, Sabik H, Petit HV, Visentainer JV, Matumoto-Pintro
the composition of Alpine highland pasture on the chemical, rhe- P, et al., Use of oregano extract and oregano essential oil as antiox-
ological and sensory properties of cheese. J Dairy Res 66:579–588 idants in functional dairy beverage formulations. LWT – Food Sci
(1999). Technol 47:167–174 (2012).
34 Bugaud C, Buchin S, Hauwuy A and Coulon JB, Relationships between 44 Holley RA and Patel D, Improvement in shelf-life and safety of perish-
flavour and chemical composition of Abondance cheese derived able foods by plant essential oils and smoke antimicrobials. Food
from different types of pastures. Lait 81:757–773 (2001). Microbiol 22:273–292 (2005).
35 Martin B, Buchin S and Hauwuy A, Influence of the botanical compo-
sition of the highland pastures on the sensory characteristics of the
Beaufort cheeses. Sci Alim 25:67–75 (2005).


J Sci Food Agric 2015; 95: 1507–1513 © 2014 Society of Chemical Industry