22
Journal of Food Protection, Vol. 75, No. 1, 2012, Pages 22–28
doi:10.4315/0362-028X.JFP-11-153
Copyright G, International Association for Food Protection
Treatment of Raw Poultry with Nonthermal Dielectric Barrier
Discharge Plasma To Reduce Campylobacter jejuni and
Salmonella enterica
BRIAN P. DIRKS,1 DANIL DOBRYNIN,2 GREGORY FRIDMAN,3 YURI MUKHIN,4 ALEXANDER FRIDMAN,4
JENNIFER J. QUINLAN1,5*
1Department
AND
of Biology, 2Electrical and Computer Engineering Department, 3School of Biomedical Engineering, 4Department of Mechanical Engineering
and Mechanics, and 5Department of Nutrition Sciences, Drexel University, Philadelphia, Pennsylvania 19102, USA
MS 11-153: Received 25 March 2011/Accepted 2 August 2011
ABSTRACT
Nonthermal plasma has been shown to be effective in reducing pathogens on the surface of a range of fresh produce
products. The research presented here investigated the effectiveness of nonthermal dielectric barrier discharge plasma on
Salmonella enterica and Campylobacter jejuni inoculated onto the surface of boneless skinless chicken breast and chicken thigh
with skin. Chicken samples were inoculated with antibiotic-resistant strains of S. enterica and C. jejuni at levels of 101 to 104
CFU and exposed to plasma for a range of time points (0 to 180 s in 15-s intervals). Surviving antibiotic-resistant pathogens were
recovered and counted on appropriate agar. In order to determine the effect of plasma on background microflora, noninoculated
skinless chicken breast and thighs with skin were exposed to air plasma at ambient pressure. Treatment with plasma resulted in
elimination of low levels (101 CFU) of both S. enterica and C. jejuni on chicken breasts and C. jejuni from chicken skin, but
viable S. enterica cells remained on chicken skin even after 20 s of exposure to plasma. Inoculum levels of 102, 103, and 104 CFU
of S. enterica on chicken breast and chicken skin resulted in maximum reduction levels of 1.85, 2.61, and 2.54 log, respectively,
on chicken breast and 1.25, 1.08, and 1.31 log, respectively, on chicken skin following 3 min of plasma exposure. Inoculum
levels of 102, 103, and 104 CFU of C. jejuni on chicken breast and chicken skin resulted in maximum reduction levels of 1.65,
2.45, and 2.45 log, respectively, on chicken breast and 1.42, 1.87, and 3.11 log, respectively, on chicken skin following 3 min of
plasma exposure. Plasma exposure for 30 s reduced background microflora on breast and skin by an average of 0.85 and 0.21 log,
respectively. This research demonstrates the feasibility of nonthermal dielectric barrier discharge plasma as an intervention to
help reduce foodborne pathogens on the surface of raw poultry.
Campylobacter and Salmonella remain the two leading
agents causing foodborne illness, and there continues to be a
need to control the transmittance of these pathogens. The
Center for Disease Control and Prevention FoodNet
estimates the occurrence of disease caused by Salmonella
and Campylobacter at 15.19 and 13.02 per 100,000
population, respectively (2). Poultry is known to be
contaminated with both of these organisms. The rate of
occurrence of Campylobacter on chicken has been reported
to range from 71 to 91% (16, 28) and from 88 to 89% for
Salmonella (6, 18). Elimination of these pathogens from raw
poultry products remains a significant challenge. Methods
have been developed to reduce the population of pathogens
throughout the production process. These include interventions such as the use of chlorine rinses and application of
trisodium phosphate, but these methods may fail to
eliminate all of the pathogens on poultry (1, 13, 26).
Additional disinfection methods throughout the production
and processing are needed to further reduce or eliminate
* Author for correspondence. Tel: 215-762-8456; Fax: 215-762-4080;
E-mail: jjq26@drexel.edu.
pathogens present on carcasses. A treatment that could
remove or reduce the level of Salmonella and Campylobacter from the surface of poultry and therefore eliminate
the potential of cross-contamination in the hands of
consumers could greatly reduce the chances of consumers
contracting illness and reduce the overall morbidity rate
associated with these pathogens (21).
A plasma is an ionized gas, an energetic state of matter
also known as the ‘‘fourth state of matter.’’ As energy
increases, matter transforms from solid to liquid, to gas, and
then to ionized gas, termed plasma (7). When a gas is given
enough energy, the molecules in the gas dissociate to form a
gas of atoms, excited species, ions, and electrons. Although
gas is normally an electrical insulator, when a plasma is
formed, it becomes an electrical conductor due to the
presence of the free electrons and positive ions (7, 10).
A number of recent studies have examined the ability of
nonthermal plasmas to reduce pathogens on the surface of
produce (4, 17, 19, 20). It is important to note that there are
many different types of plasma units that work in different
ways. Critzer et al. (4) used an atmospheric glow discharge
plasma that places the samples outside the unit. The unit
J. Food Prot., Vol. 75, No. 1
NONTHERMAL PLASMA REDUCTION OF PATHOGENS ON CHICKEN
23
FIGURE 1. (A) Schematic for application
of DBD plasma to a surface. (B) Actual
DBD probe applying plasma to boneless
skinless chicken breast.
concentrates the reactive species such as ozone and then
applies them onto the surface of the product by using air
flow. By use of this technology, Escherichia coli O157:H7
on apples was reduced by .2 log, Salmonella on cantaloupe
was reduced by .3 log, and Listeria on lettuce was reduced
by .5 log. This type of plasma is effective on larger and
less uniformly shaped objects. Niemira and Sites (17)
applied gliding arc discharge plasma to apples inoculated
with E. coli O157:H7 and Salmonella and achieved a
maximum reduction of 3.6 and 3.7 log, respectively, after
3 min. Gliding arc discharge plasma also uses airflow to
deliver the reactive species created by the plasma to the
surface of the apple. This type of plasma works by having
the electrode and the sample in an enclosed container and
applying filtered air through the electrode. The flow rate of
the air can be adjusted, and the increased flow rate can be
correlated with an increased reduction of pathogens. Perni et
al. (19) tested the effectiveness of a cold atmospheric
plasma pen on reducing the populations of various
microorganisms on honeydew melon and mango skin and
reported a 3-log reduction of E. coli type 1 (W00871). Use
of the same plasma unit on cut surfaces of mangoes and
cantaloupes resulted in a 2.5-log reduction of both E. coli
type 1 and Listeria monocytogenes on mangoes and 1.5- and
2-log reductions, respectively, on melons. This type of
plasma utilizes helium gas, which when charged creates a
different type of plasma with a lower temperature than that
of other plasmas that just use atmospheric air (20).
The atmospheric pressure nonthermal dielectric barrier
discharge (DBD) plasma used in this research is a type of
plasma that has been studied for biological and medical
applications to selectively inactivate unhealthy cells, aid in
blood coagulation, and kill parasites, bacteria, fungi, and
viruses on the surface of living tissue (11). This lowtemperature, low-pressure plasma poses less hazard and is
more time- and cost-effective than other commonly used
nonthermal sterilization techniques such as ionizing radiation and UV radiation (25). DBD plasma was previously
shown effective in killing microorganisms without destruction of living tissue or damage to heat-sensitive materials
(14). The technology has been shown to completely kill 105
CFU of E. coli (non-O157:H7) per ml in over 120 s on filter
paper as well as inactivate a mixture of the bacteria
Streptococcus spp., Staphylococcus spp., and the yeast
Candida spp. in 5 to 10 s (11, 14). The mechanism of
inactivation of plasmas includes the delivery of charged
particles and neutral reactive oxygen species applied to the
surface, making it safe for heat-sensitive materials and
tissues (5).
In this study the effectiveness of DBD plasma to reduce
and eliminate Salmonella enterica and Campylobacter
jejuni from boneless skinless chicken breast and chicken
skin was evaluated. By assessing the reduction of pathogens
on the surface of meat, we can better direct the development
of DBD plasma for use as a nonthermal pasteurization
technique for food.
MATERIALS AND METHODS
DBD nonthermal plasma. Atmospheric pressure DBD
plasma at room temperature in air was used as described previously
(11). The DBD plasma unit used in this research consists of a probe
connected to a power supply. The power supply delivers highvoltage pulses to the insulated probe. The probe is an electrode,
and its surface is covered in a material that is an electric insulator
(dielectric), which prevents current flow and thus limits any
damage to the second electrode. The surface being exposed acts as
the second electrode, and the plasma is bound between this
dielectric surface and the surface being treated (i.e., chicken meat).
Agar and chicken surfaces to be treated were exposed to plasma by
manually placing and holding the electrode near the surface at a
distance of approximately 1 to 2 mm, which resulted in DBD
plasma treatment of the surface. The powered electrode was made
of a 2.5-cm-diameter solid copper disk covered by a 3.5-cmdiameter, 1-mm-thick quartz dielectric, and the grounded treated
surface acted as the second electrode (Fig. 1). The discharge gap
was kept at 1.5 mm. The microsecond-pulsed DBD discharge was
created by applying alternating current pulsed high voltage of
30 kV in magnitude (peak to peak) and 0.5-kHz frequency between
the electrodes. Current peak duration was 1.2 ms, and the
corresponding plasma surface power density was 0.15 W/cm2.
Microorganisms. Frozen stocks of S. enterica subsp.
enterica serovar Typhi ATCC 19214 were cultured in tryptic soy
broth (TSB) (all media were purchased from Difco, BD, Sparks,
MD, unless otherwise noted) with 20 mg/ml each of tetracycline,
streptomycin, and chloramphenicol and incubated for 24 h at 37uC
prior to each plasma experiment. S. enterica subsp. enterica
serovar Enteritidis ATCC 13076 was used as a control strain and
was cultured in TSB and incubated 24 h at 37uC. C. jejuni RM
2002 and C. jejuni RM 1849 (originating from chicken cloaca)
were generously provided by William G. Miller from the U.S.
Department of Agriculture (15). Frozen stocks of C. jejuni RM
2002 were cultured by plating onto Mueller-Hinton II agar (MHA)
24
DIRKS ET AL.
containing 200 mg/ml kanamycin and incubated 48 h at 37uC under
microaerophilic conditions (5% O2, 10% CO2, and 85% N2). C.
jejuni subsp. jejuni ATCC 700819 (originating from human feces)
and C. jejuni RM 1849 were used as control strains and were
cultured by plating onto MHA without antibiotics and incubated
for 48 h at 37uC under microaerophilic conditions.
Preparation of cultures for plasma exposure. Twenty-fourhour S. enterica cultures from frozen stock (100 ml of stock to
100 ml of TSB) were adjusted to obtain approximately 108 CFU/ml
in 10 ml, using a standard curve of absorbance at 600 nm. The
adjusted culture was centrifuged 12 min at 8,000 | g at 4uC and
resuspended in 100 mM phosphate-buffered saline (PBS), pH 7.2.
The culture was serially diluted in PBS, plated onto Antibiotic
Medium 1 (Penassay Seed Agar) plates to confirm counts, and
applied to the chicken. Frozen C. jejuni stock cultures were
streaked onto MHA plates, incubated for 48 h, subcultured onto a
fresh plate for 24 h, and then harvested from MHA plates with a
swab moistened in Mueller-Hinton broth (MHB) and resuspended
in MHB. MHA was used to obtain sufficient levels of C. jejuni for
experiments using higher levels of inoculum, which were not
obtainable in liquid cultures. Cultures were adjusted to obtain
approximately 108 CFU/ml in 10 ml, using a standard curve of
absorbance at 600 nm. The culture was then serially diluted in
MHB, plated onto MHA to confirm counts, and applied to the
chicken.
Comparison of susceptibility of control and antibioticresistant strains to plasma. Inoculum levels of 101, 102, 103, 104,
105, and 106 total CFU were used. The centers of agar plates were
inoculated with 10 ml of inoculum and spread lightly over a 1-in2
area (,6.45 cm2), which was the size of the area exposed to
plasma. Plates were left to dry for 20 min and then exposed to
plasma for the appropriate amount of time. Each inoculum level
was exposed in duplicate to plasma for 0, 5, 10, 15, and 20 s. S.
enterica strains were plated onto Antibiotic Medium 1 agar plates,
and C. jejuni strains were plated onto MHA plates. Antibioticresistant strains were plated onto both media with appropriate
antibiotics and media without antibiotics to determine if the
presence of antibiotics had an influence on survival and recovery.
S. enterica plates were incubated 48 h at 37uC, and C. jejuni plates
were incubated 48 h at 37uC under microaerophilic conditions.
Following incubation, the presence or absence of bacterial growth
was recorded. These experiments were repeated for a total of three
trials each.
Raw chicken. Boneless, skinless chicken breast and chicken
thigh with skin were used for this study and were purchased in
advance from a local supermarket, frozen at 20uC the day of
purchase, and thawed overnight at 4uC. DBD plasma is most
effective on flat surfaces, so care was taken in selecting areas of the
chicken that were as flat as possible. The surface of the chicken
was scored into 2-in. squares (,25.8 cm2) with a sterile knife. The
squares were inoculated in the center with 10 ml of inoculum and
spread lightly with the pipette tip over a 1-in2 (,6.45-cm2) area.
The inoculated chicken was then allowed to sit at room
temperature for approximately 20 min (22), followed by exposure
to plasma. The inoculum level of 101 total CFU/in2 for both
pathogens was exposed to plasma for 0, 5, 10, 15, and 20 s and
inoculum levels of 102, 103, and 104 total CFU/in2 for both
pathogens were exposed for 0 to 180 s at 15-s intervals. Each trial
consisted of two squares of chicken per time exposure, and a total
of three trials per time exposure were performed. Following
exposure to plasma, each square was sampled by wiping the
J. Food Prot., Vol. 75, No. 1
inoculated surface with a swab (VWR 6-in. critical swabs with a
cotton tip and wood handle) moistened in 1 ml of phosphate buffer
for S. enterica or MHB for C. jejuni. Swabbing was performed for
approximately 15 to 30 s; the entire area was swabbed in a
horizontal direction, and then the swab surface was turned over and
the entire area was swabbed a second time in a vertical direction.
The swab was then returned to the 1 ml of buffer or MHB and
mixed manually by turning the swab in the liquid and squeezing
against the sides of the tube for approximately 15 s. S. enterica
ATCC 19214 samples were plated onto Antibiotic Medium 1
plates with antibiotics and incubated 48 h at 37uC. C. jejuni RM
2002 samples were plated onto MHA with 200 mg/ml kanamycin
and incubated 72 h at 37uC under microaerophilic conditions. Total
CFU per square inch was determined from the plate counts.
Recovery of pathogens on untreated areas for each experiment was
used as the baseline to determine inactivation of pathogens by
plasma. Recovery from untreated areas was generally 80 to 90% of
the population applied to chicken breast for both C. jejuni and S.
enterica but showed greater variability for both pathogens
(approximately 40 to 80%) on chicken skin, with greater recovery
at lower inoculum levels. Inactivation by plasma was determined
by subtracting the average population recovered from plasmatreated areas from the average population recovered from untreated
areas in individual experiments.
Sample enrichment. For experiments using only 101 total
CFU/in2, four squares were exposed to plasma for each time point.
Two squares were sampled and plated as described above, and two
squares were used for enrichments to ensure plasma eliminated all
of the pathogens on the surface of chicken when there was no
growth from direct plating due to counts being below the detection
limit or being in a viable but nonculturable state. For enrichment
of C. jejuni, a modified version of the U.S. Food and Drug
Administration’s Bacteriological Analytical Manual protocol was
used (12). Ten milliliters of Bolton broth (EMD, Gibbstown, MD)
was aliquoted aseptically into sterile 16-mm test tubes. One
milliliter of each sample suspension was added to a tube of
Bolton broth. The tube was incubated for 4 h at 37uC under
microaerophilic conditions. Following this preenrichment, the
incubation temperature was increased to 42uC with shaking for
23 to 24 h. Enrichments were diluted 1:100, and both diluted and
undiluted samples were plated in duplicate onto Campylobacter
blood-free selective agar (CCDA, Oxoid, Basingstoke, Hampshire,
England). Plates were incubated at 42uC for 24 to 48 h under
microaerophilic conditions.
For enrichment of Salmonella, a modified version of the U.S.
Department of Agriculture Food Safety and Inspection Service’s
Microbiological Laboratory Guidebook protocol was used (24).
The 1-ml suspension obtained from the swab was incubated at
35uC for 20 to 24 h. Following incubation, 0.5 ml was transferred
into 10 ml of tetrathionate broth and 0.1 ml was transferred into
10 ml of Rappaport-Vassiliadis R10 broth. The broth tubes were
incubated at 42uC for 18 to 24 h, and one loopful was plated in
duplicate onto xylose lysine Tergitol 4 agar and brilliant green
sulfa agar plates. The plates were incubated at 35uC for 18 to 24 h.
Background microflora. Boneless skinless chicken breast
and chicken thigh with skin were scored into 2-in. squares
(,25.8 cm2) with a sterile knife. Each piece of chicken contained
two squares that were exposed to plasma and two squares that were
not exposed to DBD plasma and used as controls. Pieces were
exposed to plasma for 15 and 30 s because these were the treatment
times at which the greatest initial reductions in pathogens were
observed. Following exposure, the squares were sampled by
J. Food Prot., Vol. 75, No. 1
NONTHERMAL PLASMA REDUCTION OF PATHOGENS ON CHICKEN
25
TABLE 1. Recovery of 101 total CFU of S. enterica ATCC 19214
and C. jejuni RM2002 inoculated on the surface of boneless
skinless chicken breast and chicken thigh with skin a
Plasma
exposure
time (s)
0
5
10
15
20
a
b
c
Skinless chicken breast
S. enterica
C. jejuni
Chicken thigh with skin
S. enterica
C. jejuni
7.67 ¡ 0.29 9.56 ¡ 0.54 8.00 ¡ 1.03
8 ¡ 0.34
—b
—
7.33 ¡ 1.13 3.11 ¡ 0.44c
—
—
4.22 ¡ 0.80c
—
—
—
6.00 ¡ 1.32
—
—
—
3.33 ¡ 1.49c
—
Values are total CFU recovered ¡ standard error.
—, no recovery by plating or enrichment.
Value shows significant reduction (P , 0.05) from zero exposure
time.
wiping the surface with a swab moistened in 1 ml of phosphate
buffer, after which the swab was returned back to the buffer and
mixed. Samples were pour plated with plate count agar and
incubated at 30uC for 48 h. To account for variation in background
on different pieces of chicken, the results were compared with
controls from the same piece of chicken.
Data analysis. Microsoft Excel 2007 was used for all data
analysis. t tests were performed comparing the mean total CFU
recovered from positive controls to the mean total CFU recovered
from experimental samples. A P value of ,0.05 was considered
statistically significant.
RESULTS AND DISCUSSION
Susceptibility of antibiotic-resistant strains to plasma. In order to ensure that pathogens recovered from raw
poultry after plasma treatment were not from naturally
occurring contamination of the poultry, the effectiveness of
plasma treatment was determined by using antibioticresistant strains of S. enterica and C. jejuni. The
antibiotic-resistant strains were found to be as susceptible
to plasma exposure as the wild-type strains on agar plates.
All strains of S. enterica and all strains of C. jejuni showed
no growth at inoculum levels of 104 CFU after 5 s of plasma
treatment and no growth at inoculum levels of 105 and 106
CFU after 10 s of plasma treatment (data not shown). The
greater resistance of higher levels of pathogens to DBD
plasma may indicate that cell density plays a role in plasma
effectiveness at killing bacteria. Yu et al. (27) showed a
decrease in the rate of kill of E. coli K-12 with increased
pathogen level when exposed to plasma on a membrane, and
Fernandez et al. (9) found that the rate of inactivation of S.
enterica serovar Typhimurium by plasma was inversely
proportional to the initial bacterial level, adding further
evidence to the theory that cell biomass is a protective factor
against plasma treatment.
Raw chicken breast and skin. Very low levels (101
CFU) of both S. enterica and C. jejuni on chicken breast
were unable to be detected after 5 s of plasma treatment.
Similar inoculum levels of the pathogens on chicken thigh
skin resulted in no detection of C. jejuni after 10 s of plasma
exposure, but S. enterica was still detectable by both plate
FIGURE 2. Reduction of S. enterica and C. jejuni populations on
chicken breast and thigh when inoculated at levels of 102 CFU and
exposed to plasma for increasing time periods.
count and enrichment methods following 20 s of plasma
treatment (Table 1). Pathogen reduction on breast and skin
at higher inoculum levels can be seen in Figures 2 through
4. In general, survival curves showed high levels of initial
kill of pathogens followed by a tailing-off effect similar to
what was reported by Critzer et al. (4). Reduction of
pathogens compared to controls were found to be
statistically significant (P , 0.05) at all time points at all
three inoculum levels (102, 103, and 104 CFU) with the
exception of Salmonella on chicken skin. Reduction of
Salmonella on chicken skin was significant at all time points
with inoculum levels of 102 but was not significant until 45
and 105 s of plasma exposure with inoculum levels of 103
and 104, respectively. The observed tailing-off effect may be
due to a number of factors. One possible explanation could
be that a subpopulation of the organism is resistant to the
treatment. Alternatively, there may be a protective effect
with higher densities of the pathogens, similar to what we
observed with higher levels of the pathogens when tested on
agar plates and the possible protective effect of biomass (9).
C. jejuni and S. enterica generally did not show
significant differences in their susceptibility to DBD plasma,
with both resulting in maximum log reductions of
approximately 2.5 on chicken breast and approximately
1.3 to 1.8 on chicken thigh skin (P . 0.05). The exception
was a maximum log reduction of 3.11 of C. jejuni on
chicken skin when 104 CFU of the pathogen was applied.
Differences were observed when we compared recovery of
the pathogens from breast to recovery from skin. Initially,
recovery of both pathogens was less from thigh skin than
breast even without plasma application (Figs. 2 through 4).
We hypothesize that the fatty nature of the skin resulted in
greater adherence of the pathogens, which resulted in lower
recovery rates. The surface of the chicken skin is not easily
treated by DBD plasma due to the pores and feather follicles
covering the surface. When the surfaces are exposed to the
DBD plasma, it does not apply evenly but instead
concentrates on the higher topographic points. Therefore,
the rough nature of the skin likely provides sites where the
pathogens may attach and be somewhat protected from the
inactivation effects of the DBD plasma. Greater total log
reductions were observed on chicken breast at all three
inoculum levels of S. enterica (1.65, 2.45, and 2.45) when
26
DIRKS ET AL.
J. Food Prot., Vol. 75, No. 1
FIGURE 3. Reduction of S. enterica and C. jejuni populations on
chicken breast and thigh when inoculated at levels of 103 CFU and
exposed to plasma for increasing time periods.
FIGURE 4. Reduction of S. enterica and C. jejuni populations on
chicken breast and thigh when inoculated at levels of 104 CFU and
exposed to plasma for increasing time periods.
compared with total log reductions on chicken skin (1.25,
1.08, and 1.31). Greater total log reductions were generally
observed on chicken breast at all three inoculum levels of C.
jejuni (1.65, 2.45, and 2.45) when compared with total log
reductions on chicken skin (1.42, 1.87, and 3.11) with the
exception, again, of the log reduction of 3.11 observed when
the skin was inoculated with high levels of C. jejuni. There
were no visible changes to the color of the chicken breast or
chicken thigh skin surface after application of DBD plasma
even at the maximum time period of 3 min.
Overall, our maximum reduction levels on chicken after
3 min of DBD plasma treatment (1.3 to 1.8 log on skin and
approximately 2.5 log on breast for both pathogens) were
slightly less than those observed by researchers investigating the use of plasmas on foodborne pathogens on produce.
Critzer et al. (4) reported .2-log reduction of E. coli
O157:H7 after 2 min of exposure and .3-log reduction of
both Salmonella and L. monocytogenes after 3 min of
exposure on produce samples. Niemira and Sites (17)
observed maximum reductions of 2.96 to 3.72 log CFU/ml
for Salmonella and 2.6 to 3.4 log CFU/ml for E. coli
O157:H7 following 3 min of exposure to plasma deposition
on apples. This may provide further evidence for the
concept of biomass protection of bacteria from plasmas,
since both of the studies cited above took steps to eliminate
background microflora on produce samples prior to
application of pathogens and plasmas, whereas the chicken
samples in our study were found to have background in the
104 to 108 total CFU range and were not pretreated to reduce
background. The lower reduction levels may still be
sufficient to reduce pathogen levels on raw poultry and
render it safer for consumers. Future research should be
aimed at treatment of naturally contaminated poultry
samples with plasmas to determine if treatment could
reduce the ability to detect pathogens at naturally occurring
levels on raw poultry.
0.84 and 0.85 log following 15 and 30 s of exposure,
respectively. DBD plasma reduced the background microflora on chicken thigh skin 0.33 and 0.21 log following 15
and 30 s of exposure, respectively. These results are
consistent with our pathogen reductions rates with respect to
the fact that the plasma appears to be less effective on the
skin than on the breast. The results are inconsistent with our
pathogen reduction rates with respect to the levels of
reduction observed. The total plate counts were in the 104 to
108 total CFU ranges. On chicken breasts inoculated with
104 CFU of pathogens, log reduction rates of 1.26 and 1.28
were observed for S. enterica and C. jejuni, respectively,
following 30 s of plasma exposure. These are greater than
the 0.85-log reduction seen for total background microflora.
The results on chicken skin were less predictable, with log
reduction rates of 0.34 and 1.39 observed for S. enterica and
C. jejuni, respectively, following 30 s of plasma exposure,
compared with total CFU log reduction rates in the 0.22 to
0.41 range for background microflora. One possible
explanation for greater resistance of background microflora
would be if the natural microflora on the chicken has greater
resistance to plasmas than S. enterica and C. jejuni. The
background microflora of raw chicken has been found to be
a mixed population of gram-negative and gram-positive
microorganisms including Pseudomonas, lactic acid bacteria, Enterobacteriaceae, and bacilli (3). There is some
evidence that plasmas may be more effective on gramnegative organisms. Ermolaeva et al. (8) found plasmas to
be more effective against gram-negative organisms (Pseudomonas aeruginosa, Burkholderia cenocepacia, and E.
coli) than against gram-positive organisms but also found
differences in resistance among the gram-positive organisms
(Streptococcus pyogenes, Staphylococcus aureus, and
Staphylococcus epidermidis). Critzer et al. (4) found L.
monocytogenes to be more sensitive to plasma than both
Salmonella and E. coli O157:H7 and Salmonella to be more
sensitive to plasma than E. coli O157:H7. Scholtz et al. (23)
found plasmas to be equally effective against gram-positive
and gram-negative organisms on agar but found that in
liquid suspensions the gram-negative organisms were more
sensitive to plasmas than the gram-positive ones. This leads
to a second possible explanation of these results, which may
be that plasmas are more effective on moister surfaces.
Background microflora. Since the greatest pathogen
kill with the least amount of plasma treatment was observed
in the initial 15 to 30 s of exposure, these time points were
chosen to observe the effect plasmas would have on
background microflora on the raw poultry. DBD plasma
reduced the background microflora on chicken breast by
J. Food Prot., Vol. 75, No. 1
NONTHERMAL PLASMA REDUCTION OF PATHOGENS ON CHICKEN
Although chicken itself is moist, the buffer that carried the
pathogens when they were applied may have added to this
despite the allowance of drying time. With no additional
liquid added, the cell density of the endogenous flora may
be what is limiting the plasma’s effectiveness. It may also be
possible to further moisten the surface of the chicken before
exposure to plasma to determine if that has a higher rate of
reduction. While the limited results presented here with
regard to background microflora are not conclusive, the
reduction of background microflora on raw poultry has the
potential to extend the shelf life of the product in addition to
making it safer, potentially making plasmas a more
attractive intervention for poultry processing. As with
reduction of pathogens, optimization of the plasma
application would be required to result in more consistent
bacterial reduction on poultry.
Limitations of DBD plasma. The greatest limitation of
this study is the requirement of a flat surface for the plasma
probe to be most effective. Care was taken to select flat
pieces of chicken, but there was uncontrollable variation of
1 to 2 mm between individual pieces of chicken, and this
inconsistency may have contributed to increased variation in
our data. An additional limitation of the technology as it was
used here is the limited size of the probe applying plasma.
Further development of the nonthermal DBD plasma used in
this study would be necessary to create an application
method that does not require a flat surface in order to expose
full pieces, or ideally, multiple pieces of chicken to plasmas
prior to packaging. Another possible limitation of the DBD
plasma utilized in this research could be the interaction of
the delivered charged particles and neutral reactive oxygen
species with components in the food to result in oxidation of
lipids, in particular. Further research into the effects of the
DBD plasma on the poultry and poultry skin and how it
might impact acceptability and shelf life would be needed.
While a limited number of studies have examined the
ability of nonthermal plasma to reduce pathogens on the
surface of produce (4, 17, 19, 20), none have examined raw
meat as a surface for plasma treatment. Additionally, this is
the first study we are aware of that has examined the
effectiveness of nonthermal plasma on C. jejuni, the second
leading cause of foodborne illness in the United States (2).
This study demonstrates that DBD plasma is effective at
reducing S. enterica and C. jejuni on the surface of chicken
breast and skin. If further developed, nonthermal DBD
plasma has the potential to become an important coldpasteurization intervention for raw food products.
REFERENCES
1. Berrang, M. E., J. S. Bailey, S. F. Altekruse, B. Patel, W. K. Shaw, R.
J. Meinersmann, and P. J. Fedorka-Cray. 2007. Prevalence and
numbers of Campylobacter on broiler carcasses collected at rehang
and postchill in 20 U.S. processing plants. J. Food Prot. 70:1556–
1560.
2. Centers for Disease Control and Prevention. 2010. Preliminary
FoodNet data on the incidence of infection with pathogens
transmitted commonly through food—10 States, 2009. Morb. Mortal.
Wkly. Rep. 59:418–422.
27
3. Chouliara, E., A. Karatapanis, I. N. Savvaidis, and M. G.
Kontominas. 2007. Combined effect of oregano essential oil and
modified atmosphere packaging on shelf-life extension of fresh
chicken breast meat, stored at 4uC. Food Microbiol. 24:607–617.
4. Critzer, F. J., K. Kelly-Wintenberg, S. L. South, and D. A. Golden.
2007. Atmospheric plasma inactivation of foodborne pathogens on
fresh produce surfaces. J. Food Prot. 70:2290–2296.
5. Dobrynin, D., G. Fridman, G. Friedman, and A. Fridman. 2009.
Physical and biological mechanisms of direct plasma interaction with
living tissue. New J. Phys. 11:115020.
6. Dufrenne, J., W. Ritmeester, E. D. V. Asch, F. Van Leusden, and R.
De Jonge. 2001. Quantification of the contamination of chicken and
chicken products in The Netherlands with Salmonella and Campylobacter. J. Food Prot. 64:538–541.
7. Eliezer, S., and Y. Eliezer. 2001. The fourth state of matter: an
introduction to plasma science. Institute of Physics, London.
8. Ermolaeva, S. A., A. F. Varfolomeev, M. Y. Chernukha, D. S. Yurov,
M. M. Vasiliev, A. A. Kaminskaya, M. M. Moisenovich, J. M.
Romanova, A. N. Murashev, I. I. Selezneva, T. Shimizu, E. V.
Sysolyatina, I. A. Shaginyan, O. F. Petrov, E. I. Mayevsky, V. E. Fortov,
G. E. Morfill, B. S. Naroditsky, and A. L. Gintsburg. 2011. Bactericidal
effects of non-thermal argon plasma in vitro, in biofilms and in the
animal model of infected wounds. J. Med. Microbiol. 60:75–83.
9. Fernandez, A., N. Shearer, D. R. Wilson, and A. Thompson. Effect of
microbial loading on the efficiency of cold atmospheric gas plasma
inactivation of Salmonella enterica serovar Typhimurium. Int. J.
Food Microbiol., in press.
10. Fridman, A., and L. A. Kennedy. 2004. Plasma physics and
engineering. Taylor and Francis, New York.
11. Fridman, G., A. D. Brooks, M. Balasubramanian, A. Fridman, A.
Gutsol, V. N. Vasilets, H. Ayan, and G. Friedman. 2007. Comparison
of direct and indirect effects of non-thermal atmospheric-pressure
plasma on bacteria. Plasma Process. Polym. 4:370–375.
12. Hunt, J. M., C. Abeyta, and T. Tran. 2009. Campylobacter, chap. 7.
Bacteriological analytical manual, 8th ed., revision A 1998. U.S.
Food and Drug Administration, Rockville, MD.
13. Kim, T., J. L. Silva, and T. C. Chen. 2002. Effects of UV irradiation
of selected pathogens in peptone water and on stainless steel and
chicken meat. J. Food Prot. 65:1142–1145.
14. Lee, K., K. H. Paek, W. T. Ju, and Y. Lee. 2006. Sterilization of
bacteria, yeast, and bacterial endospores by atmospheric-pressure
cold plasma using helium and oxygen. J. Microbiol. 44:269–275.
15. Miller, W. G., A. H. Bates, S. T. Horn, M. T. Brandl, M. R. Wachtel,
and R. E. Mandrell. 2000. Detection on surfaces and in Caco-2 cells
of Campylobacter jejuni cells transformed with new gfp, yfp, and cfp
marker plasmids. Appl. Environ. Microbiol. 66:5426–5436.
16. Moran, L., P. Scates, and R. H. Madden. 2009. Prevalence of
Campylobacter spp. in raw retail poultry on sale in Northern Ireland.
J. Food Prot. 72:1830–1835.
17. Niemira, B. A., and J. Sites. 2008. Cold plasma inactivates
Salmonella Stanley and Escherichia coli O157:H7 inoculated on
golden delicious apples. J. Food Prot. 71:1357–1365.
18. Parveen, S., M. Taabodi, J. G. Schwarz, T. P. Oscar, J. Harter-Dennis,
and D. G. White. 2007. Prevalence and antimicrobial resistance of
Salmonella recovered from processed poultry. J. Food Prot. 70:
2466–2472.
19. Perni, S., D. W. Liu, G. Shama, and M. G. Kong. 2008. Cold
atmospheric plasma decontamination of the pericarps of fruit. J. Food
Prot. 71:302–308.
20. Perni, S., G. Shama, and M. G. Kong. 2008. Cold atmospheric
plasma disinfection of cut fruit surfaces contaminated with migrating
microorganisms. J. Food Prot. 71:1619–1625.
21. Rowan, N. J., S. J. MacGregor, J. G. Anderson, R. A. Fouracre, L.
McIlvaney, and O. Farish. 1999. Pulsed-light inactivation of foodrelated microorganisms. Appl. Environ. Microbiol. 65:1312–1315.
22. Sarjeant, K. C., S. K. Williams, and A. Hinton, Jr. 2005. The effect of
electron beam irradiation on the survival of Salmonella enterica
serovar Typhimurium and psychrotrophic bacteria on raw chicken
breasts stored at four degrees Celsius for fourteen days. Poult. Sci. 84:
955–958.
28
DIRKS ET AL.
23. Scholtz, V., J. Julak, and V. Kriha. 2010. The microbiocidal effect of
low temperature plasma generated by corona discharge: comparison
of various microorganisms on an agar surface and in aqueous
suspension. Plasma Process. Polym. 7:237–243.
24. U.S. Department of Agriculture. 1998. Microbiological laboratory
guidebook, 3rd ed. Available at: http://www.fsis.usda.gov/ophs/
microlab/mlgbook.pdf. Accessed 12 April 2007.
25. Vleugels, M., G. Shama, X. T. Deng, E. Greenacre, T. Brocklehurst, and
M. G. Kong. 2005. Atmospheric plasma inactivation of biofilm-forming
bacteria for food safety control. IEEE Trans. Plasma Sci. 2:824–828.
26. Whyte, P., J. D. Collins, K. McGill, C. Monahan, and H. Mahony.
2001. Quantitative investigation of the effects of chemical de-
J. Food Prot., Vol. 75, No. 1
contamination procedures on the microbiological status of
broiler carcasses during processing. J. Food Prot. 64:179–
183.
27. Yu, H., S. Perni, J. J. Shi, D. Z. Wang, M. G. Kong, and G. Shama.
2006. Effects of cell surface loading and phase growth in cold
atmospheric gas plasma inactivation of Escherichia coli K12. J. Appl.
Microbiol. 101:1323–1330.
28. Zhao, C., B. Ge, J. De Villena, R. Sudler, E. Yeh, S. Zhao, D. G.
White, D. Wagner, and J. Meng. 2001. Prevalence of Campylobacter
spp., Escherichia coli, and Salmonella serovars in retail chicken,
turkey, pork, and beef from the Greater Washington, D.C., area. Appl.
Environ. Microbiol. 67:5431–5436.