Repeated application of shock waves as a possible method for food preservation, CHEMIA I PIROTECHNIKA, Chemia i ...

[ Pobierz całość w formacie PDF ]
Shock Waves (1999) 9: 49{55
Repeated application of shock waves as a possible method
for food preservation
A.M. Loske
, F.E. Prieto
, M.L. Zavala, A.D. Santana, E. Armenta
Instituto de Fsica, UNAM, A.P. 20-364, 01000 Mexico D.F., Mexico. E-mail: loske@fenix.isicacu.unam.mx
Received 8 April 1998 / Accepted 17 September 1998
Abstract.
In order to study the possibility of using underwater shock waves to cause death in non desired
microorganisms found in certain foods,
Escherichia coli
in suspension was exposed to hundreds of shock
waves on an experimental electrohydraulic shock wave generator. Using a parabolic reflector it was possible
to produce a plane shock front and expose many test tubes to the action of the shock waves at the same
time and under the same conditions. The amount of surviving bacteria was determined by plate counting
for dierent numbers of applied shock waves. Pressure measurements using needle hydrophones are also
reported. Experimental results indicate that electrohydraulically generated shock waves are capable of
producing a signicant reduction in an
E. coli
population. An increase in the applied shock wave number
produced a nearly exponential reduction in the
E. coli
population.
Key words:
Food preservation, Eect of shock waves,
Escherichia coli
1 Introduction
The destructive eects of ultrasonic waves on bacterial
cells, known for many years (Davies 1959), and the dam-
ages on living cells observed during ESWL (Delius et al.
1988), lead to the idea of using underwater shock waves
as a possible method for food preservation.
In the food industry, heat treatments are commonly
used to inactivate pathogenic microorganisms. Neverthe-
less, because heat may aect the organoleptic and nu-
tritional characteristics of food, there is great interest in
non thermal processes like ionizing irradiation (i.e.
,
,
and X rays), addition of preservatives, cold storage, pulsed
electric elds, oscillating magnetic elds, high hydrostatic
pressure and intense light pulses (Downing 1989; Mertens
1994; Pothakamury et al. 1993; Barbosa-Canovas et al.
1994; Qin et al. 1995; Russell 1982). Some of these tech-
niques are still being explored as possible alternatives.
It is the purpose of our investigation to evaluate the
possibility of using underwater shock waves in order to
cause death in non desired microorganisms found in cer-
tain foods, preventing them from carrying out the biolog-
ical processes necessary for their existence and prolifera-
tion. This article reports our rst results of the eects on
Escherichia coli
(
E. coli
) ATCC (American Type Culture
Collections) 10536 under the action of weak underwater
shock waves, generated with an experimental electrohy-
draulic shock wave generator.
Although in the device here described an electric dis-
charge in water is used to produce the shock waves, this
does not mean that the method is similar to the use of high
voltage pulsed electric elds to preserve foods. In our case,
microorganisms are not exposed to an electric eld.
Since its introduction in 1980, extracorporeal shock wave
lithotripsy (ESWL) has become the standard treatment
for the majority of patients with renal and ureteral calculi
(Chaussy et al. 1980; Loske and Prieto 1998) and an alter-
native in the treatment of gallbladder stones (Nahrwold
1993), pancreatic concrements (van der Hul et al. 1993),
and stones of the salivary gland (Hessling et al. 1993).
New clinical applications of shock waves are the treat-
ment of non-union fractures (Haupt et al. 1992), as well as
the management of pseudarthrosis (Schleberger and Senge
1992), tendinopathy and other orthopedic diseases (Haupt
1997). The treatment of tumors with shock waves is an-
other experimental approach (Oosterhof et al. 1991). It
has been shown that colony growth of tumor cells de-
creases as shock wave number increases (Berens et al.
1989). Unfortunately the tumor growth suppression ob-
served
in vivo
is temporary.
Because of the successful applications of shock waves
to medicine (Loske and Prieto 1995), low intensity under-
water shock waves and the behavior of materials under the
influence of low pressure shock waves received increased
attention in the last fteen years. For the same reason, the
eects of shock waves on living cells have also been the
subject of many investigations (Delius 1994; Loske and
Prieto 1995).
Present address:
Laboratorio de Choques Debiles, Fsica Apli-
cada y Tecnologa Avanzada, Universidad Nacional Autonoma
de Mexico, A.P. 1-1010, Santiago de Queretaro, Queretaro
76000, Mexico
50
A.M. Loske et al.: Shock waves as a mechanism for food preservation
As often in this kind of experiments,
E. coli
ATCC
10536 was chosen as the rst microorganism to study the
eect of electrohydraulically generated shock waves, be-
cause it is a well known and easy to handle bacteria. Addi-
tionally, the comparison with the results obtained by Ker-
foot et al. (1992) and by Teshima et al. (1995) promised
to be interesting.
2 Material and methods
2.1 The experimental shock wave generator
A new experimental underwater shock wave generator,
named MEXILIT II was designed and constructed. The
MEXILIT II, similar to its former version (Prieto et al.
1991), consists of a pulsed power circuit, operating be-
tween 10 and 1000 Joules, providing multiple pulses to a
spark gap immersed in water. The spark gap electrode as-
sembly is at the focal point of a parabolic stainless steel re-
flector, with a focal distance of 20.0mm, a
latus rectum
of
80.0mm, a maximum internal diameter of 172.0mm and
a depth of 92.5mm (see Fig. 1), mounted on the bottom
of a 1200
800
600mm Fiberglass water tank (see Fig. 2).
Application of high voltage (up to 30 kV) across a pair of
electrodes induces a spark, creating the sudden ionization
of the water. The fast expansion of the gas bubble gener-
ates a shock wave, propagating into the surrounding wa-
ter and reflecting o the reflector, creating a plane shock
front. The MEXILIT II, shown in Fig. 2, has the Fiberglass
water tank mounted on an iron frame. A three dimensional
computer controlled position system is placed on top of the
device, in order to fasten and move any probe, pressure
transducer, test tube or sample to any desirable position
within the tank. Basically the electric circuit consists of a
computer controlled capacitor charging system and a dis-
charge device. In this work, capacitance and voltage were
set to 80 nF and 20
0.1 kV respectively. The spark gap
was set to one millimeter. Electrodes, with the shape of a
truncated cone (Loske and Prieto 1993), were allowed to
burn in for 400 discharges at 18
0.1 kV. Tap water having
a conductivity of 960
5 microsiemens/cm and a temper-
ature of 27
0.1
C was used. Basically, the MEXILIT II is
similar to electrohydraulic shock wave lithotripters used
in ESWL, except for having a parabolic reflector instead
of the ellipsoidal reflector used for clinical applications.
Fig. 1.
Sketch of the parabolic stainless steel reflector with
the spark gap electrode assembly at its focal point and the
rack holding the test tubes
Fig. 2.
Sketch of the MEXILIT II electrohydraulic shock wave
generator. For clarity the front side of the water tank was not
drawn
The rst was only to conrm the culture growth, and the
second to dilute the sample with saline solution until the
reading corresponding to the desired bacteria concentra-
tion was obtained.
A total of 56 disposable test tubes (Elkay Products
Inc., model 127-P507-STR) were lled and heat sealed.
Half of the pipettes were placed inside the water tank of
the MEXILIT II. The other 28 test tubes remained as con-
trol samples in a separate water bath at the same tempera-
ture, for the same time as each of the \treated" test tubes.
As can be seen in Figs. 1 and 2, the pipettes were placed on
a specially designed plane circular Lucite rack, capable of
holding 28 tubes at the same time. The rack was fastened
with an ordinary laboratory clamp to the position con-
trol system of the MEXILIT II at an arbitrary distance of
122.5
0.5mm from the focus of the reflector, which corre-
sponds to a 50mm separation between the upper border of
the reflector and the Lucite rack (Fig. 1). All samples were
positioned so that their center was 107.5
0.5mm from a
2.2 Sample preparation
A24hrat37
C culture of
E. coli
ATCC 10536 in nutritive
broad (Merck V552243-448) was used. After cultivation,
the cells were collected by centrifugation and resuspended
in a 0.9% NaCl solution. After three washes using the same
procedure, a suspension containing 10
5
{10
7
CFU/mL was
prepared.
The concentration of bacteria before cultivation and
after the washes was registered and adjusted with a pho-
tometer (Lakeside Mannheim Boehringer model 4010) at
623 nm, using media and 0.9% NaCl solution as a blank.
A.M. Loske et al.: Shock waves as a mechanism for food preservation
51
horizontal line (
latus rectum
) going trough the focus of
the reflector. Shock waves were generated at a frequency
of 0.4 Hz. The water level was 45mm over the border of
the reflector.
2.3 Experimental procedure
The experiment was repeated ve times. Each time a to-
tal of 2000 shock waves were applied to the rack hold-
ing the test tubes. After every 500 shock waves, four test
tubes were randomly taken out of the shock wave gener-
ator, mixed in a flask and identied. The same procedure
was applied simultaneously to the 28 control tubes, not
exposed to the shock waves.
The following biochemical analysis were performed in
order to detect possible changes in the
E. coli
metabolism:
Kligler, H
2
S, citrate, mobility, indole and urea.
Samples were serial-diluted (1:10) and the amount of
surviving bacteria determined by plate counting (agar
plate count: Merck V877063708). The number of colony-
forming units per milliliter (CFU/mL) obtained to ll the
pipettes before treatment, was used as a sample for zero
discharges.
Fig. 3.
Pressure record obtained using a needle hydrophone at
about 107mm from the
latus rectum
of the parabolic reflector
shown in Fig. 1 and at about 18mm from its axis of symmetry
it crosses the baseline again. This should not be confused
with some reported data, using ellipsoidal reflectors, where
the width is dened as the time over which the pressure is
greater than one half of the peak compressional pressure
pulse. The implications of using these denitions of the
pulse rise time and widths are explained in the Discussion
section.
2.4 Pressure measurements
The pressure applied was recorded using a needle hydro-
phone (Imotec, GmbH, Wurselen, Germany) with a 20 ns
rise time. Signals coming from the gauge were sent to
the input channel of a
Tektronix 2430A
digital oscillo-
scope (Tektronix, Inc., Beaverton, Oregon), placing the
hydrophone at the axis of symmetry of the parabolic re-
flector, at 107
0.5mm from the focus, and also at ten
other positions, moving the transducer horizontally away
from the axis in 9mm steps. Two hundred measurements
were recorded at each position. A new set of electrodes
was used for each position and allowed to burn in for
400 discharges at 18 kV. In order to measure the pres-
sure drop due to the test tubes, the
Imotec
pressure gauge
was placed inside the water tank at 100
0.5mm from the
spark gap. After burning in the electrodes, 50 pressure
proles were recorded without covering the gauge. After
that, the gauge was immersed in an inactivated
E. coli
suspension inside a test tube and placed at the same po-
sition in order to take another set of 50 measurements.
All measurements were done at the voltage, capacitance,
water temperature and conductivity already mentioned in
2.1, using the cursors of the digital oscilloscope, and fed
into a personal computer for carrying out the statistical
analysis.
In order to save time while measuring with the cursors
of the oscilloscope, all rise times were dened as the time
required for the wave to rise from the baseline to the max-
imum amplitude and not in the conventional way, as the
time required to rise from 10% to 90% of the maximum
amplitude. For the same practical reasons, the widths were
measured at the baseline and dened as the time from
the instant where the pulse rises, to the instant where
3 Results
Figure 3 shows a typical pressure record obtained at 107
0.5mm from the focus of the reflector and at 18
0.5mm
from its axis of symmetry. The signal was obtained using
a50
s/div time base. Each vertical division corresponds
to about 10MPa. The electromagnetic signal of the high
voltage discharge can be seen at the beginning of the trace
at the instant (T) when the oscilloscope was triggered. The
direct shock wave arrives after about 84
s and is followed
approximately 22
s later by the reflected pressure wave.
All pressure variations, recorded at the other positions,
showed a similar behavior.
The average peak positive pressure of the reflected
pulse, corresponding to the rst ten transducer positions
was 44
7MPa, having a width and rise time of 4
0.5
s
and 2.8
0.1
s, respectively, followed by a negative pres-
sure pulse of 6
3MPa. A statistical analysis revealed no
signicant dierence between pressure measurements at
the dierent positions, except for the last two at 81 and
90mm from the axis of symmetry of the reflector, were the
positive pressure dropped about 25 and 40%, respectively.
This is probably due to diraction of the pressure wave at
the borders of the reflector. Because of this, the mentioned
average pressure values refer only to the pressure at the
axis of symmetry and the rst nine consecutive positions.
No test tubes were located at more than 65mm from this
axis.
The measured pressure drop due to test tubes lled
with cell suspension was about 20%.
52
A.M. Loske et al.: Shock waves as a mechanism for food preservation
Fig. 4.
Graphs of the logarithm of survival
E. coli
population
vs. the number of applied shock waves for ve experiments.
Least squares linear ts to the experimental results are shown
where
t
stands for time in minutes,
N
0
for initial number
of microorganisms and
N
for number of microorganisms
which survived after
t
minutes. In this study, results are
given in dose or \applied shock waves", instead of time.
This is due to the fact that the shock wave generation
frequency is a parameter which can be set and modied
depending on the selected voltage and capacitance of the
shock wave generator. In this experiments the mean value
of
K
was 0.0018, with a coecient of variation (standard
deviation divided by the average) of only 0.13.
The average dose
D
=
t=
(log
N
0
−
log
N
), needed to re-
duce the initial amount of microorganisms 90% was about
569 shock waves, having a coecient of variation of 0.14.
Figure 5 is a graph of the logarithm of the survival
population vs. the applied shock wave number, showing
the expected behavior continued to 6
D
. The straight line
has a slope
K
=0
:
0018. This means that the reduction
seems to follow an exponential behavior.
Since in this case a frequency of 0.4 Hz was used, it
would take about 24minutes to generate 569 shock waves.
This means that it would be necessary to apply electrohy-
draulic generated shock waves (at the already mentioned
voltage, capacitance and frequency) for about 24 minutes
to reduce the
E.coli
population from 10
6
to 10
5
CFU/mL.
In order to inactivate the initial population, 6
D
or about
143 minutes are needed (Block 1994).
4 Discussion
Fig. 5.
Expected behavior of
E.coli
growth after shock wave
exposure
The cell container and environment around and within
the cell tube are important because they will directly in-
fluence on the transmission of the shock wave to the cells.
Polypropylene was chosen for the test tubes because its
acoustic impedance approximates that of water. Neverthe-
less, pressure measurements revealed that the shock wave
lost about 20% of its value when passing through the test
tube. Pipettes with thinner walls or made out of a dierent
material, could reduce this pressure attenuation.
Spark gaps in water generate broad band pressure pul-
ses with very short rise times and high pressures which
depend on several parameters, some of which can be con-
trolled and some can not. The reported variations in pres-
sure measurements are typical of electrohydraulic shock
wave generators (Coleman and Saunders 1989; Prieto et
al. 1994). These variations did not aect our results be-
cause microorganisms were exposed to hundreds of shock
waves.
The electrode tips of the shock wave generator wear o
due to the high temperatures and forces acting on them
during each electric discharge. As a result of this erosion,
the electrodes have a limited lifetime. In order to reduce
time between voltage application and spark gap genera-
tion, the electrode gap should not exceed 3mm (Loske and
Prieto 1993). Furthermore, as the electrode gap becomes
larger, the pressure of the shock wave increases. Addition-
ally, in general the electric spark gap does not link the
two electrodes by the shortest path. Therefore, the electric
discharge is rarely located at the focus and leads to dis-
persed pressure peaks around the second focus of ESWL
Biochemical analysis did not reveal any change in the
metabolism of the
E. coli
microorganisms.
Results indicate a nearly logarithmic reduction in the
microorganism population after shock wave exposure. In
order to determine the mortality index of the exposed
E. coli
ATCC 10536 bacteria, an initial count between
10
5
and 10
7
CFU/mL was used. This value is compara-
ble to the concentration reported by the Association of
Ocial Analytical Chemists for some contaminated food
products (Analytical Chemists, vol. I, 15th edition, Wash-
ington DC, Association of Ocial Analytical Chemists,
Inc. (1990) pp 435{436, 803{805).
Figure 4 are the graphs of the logarithm of the survival
population vs. the number of applied shock waves for the
ve experiments, showing a similar slope
K
. Generally,
K
,
referred to as velocity constant or mortality index (Block
1994) is obtained using the formula
N
=
N
0
e
−Kt
;
A.M. Loske et al.: Shock waves as a mechanism for food preservation
53
lithotripters. This can be improved by axial positioning
of the electrodes in the reflector, as in the MEXILIT II.
Considering a 400 discharge burn in at 18 kV, the practical
lifetime of the electrodes was estimated to be about 2400
shock waves at 20 kV, using an 80 nF capacitance. Pres-
sure measurements have shown that between 400 and 2400
discharges, the pressure prole is fairly constant. Beyond
2400 shock waves, the pressure amplitude variation, as
well as the number of misres, increase signicantly. Due
to this, the maximum number of applied shock waves was
2000. The extrapolation of our experimental data (Fig. 5)
revealed that a total of about 3420 shock waves are needed
to completely inactivate the bacteria. In order to replace
the worn-o electrode with a new one, it is necessary to
empty the water tank of the MEXILIT II. It takes about
30 minutes to empty the tub, replace the electrode, ll
the tank again, adjust the water temperature and con-
ductivity to the desired values and burn in the new elec-
trode. During this process, the pipettes would have to be
taken out of the shock wave generator and placed in a
separate water bath having the same temperature. After
that, the experiment could be continued until 3420 shock
waves have been administered. Since a 30 minute waiting
time would signicantly alter the results, the electrode
was not changed and the experiment was stopped after
2000 shock waves. In the future, this shortcoming could
be solved using a dierent type of electrodes or using an
ellipsoidal
reflector in order to increase the pressure and
reduce the number of shock waves needed to perform an
experiment with a D6. This could reveal the existence of
bacteria that were originally resistant to shock waves or
became so during shock wave treatment. As already ex-
plained, the disadvantage of using an ellipsoidal, instead
of a parabolic reflector, is that only
one
pipette should be
placed at the second focus and exposed to the shock waves
at a time. This signicantly increases the experimentation
time. If shock wave application reveals to be a convenient
method to be used in the food or pharmaceutical industry,
other shock wave generation mechanisms will have to be
developed.
Shock waves from electrohydraulic generators are con-
sidered weak. Nonlinear eects appear only in the prox-
imity of regions where the energy is concentrated. This is
the case in extracorporeal lithotripters, using ellipsoidal
reflectors, but not in this study, where a parabolic reflec-
tor was used.
It is important to point out that the radiant output of
the underwater spark has a continuum in the ultraviolet
(UV), having a peak at approximately 55 to 150 nm. This
ultraviolet radiation could contribute to microorganism
death. Nevertheless, the intensity of this radiation is re-
duced signicantly during its path through the water and
the test tube. The influence of this UV radiation on the
reduction of microorganism population is currently been
studied. Experiments on human tumor cells, exposed to
electrohydraulically generated shock waves using opaque
polypropylene pipettes, have shown no evidence of cell
death due to UV light (Berens et al. 1989). Obviously this
result could be dierent when using
E. coli
. The fact that
Ohshima et al. (1991) found that the intact cells of
E. coli
JM 109/pKPDH2 are dicult to be destroyed by shock
waves using a shock tube which does not generate UV
light, indicates a possible influence of the spark-generated
electromagnetic radiation.
Even if it is known that
E. coli
can grow at static pres-
sures up to 55MPa, the response to dynamic pressures is
expected to be dierent, since in this case there is not
an even distribution of pressure in the cell suspension.
Furthermore, static pressures do not produce cavitation
in the suspension. Cavitation is generated whenever there
is a rapid transformation of positive pressure into ten-
sile stress. In the MEXILIT II, the pressure wave initially
produces a high positive pressure, which is rapidly trans-
formed into tensile stress within microseconds, resulting
in the formation of vapor-lled cavities. These cavities im-
plode, creating very high energy densities.
In general, microorganisms can be killed by static pres-
sure of about 100MPa, but the complete sterilization is
often dicult because of so called \persisters". These are
many reasons why simple compression and decompression
does not harm microorganisms in the same way as the re-
peated administration of a very short high pressure pulse
followed by a negative pulse. Cavitation depends on the
pressure of the medium, the presence of microbubbles in
the sample and the existence of a liquid-air interface. The
mechanism by which cavitation may cause biological dam-
age are high localized temperature and pressure gradients.
The bactericidal eect of ultrasound has been attribu-
ted to cavitation (Garca et al. 1989). It is interesting to
point out that the increase in human renal cell carcinoma
xenografts loss in tubes containing air was reported to
be 40% higher as compared to sample tubes without air
(Steinbach et al. 1992). This might be explained by an
increased occurrence of transient cavitation, caused by re-
flection of the pressure wave at the liquid-air interface. The
interface results in perturbation in the shock front with re-
sultant surface shear and cavitation within the suspension.
It is for these reasons that the microorganism death is ex-
pected to reduce in the absence of a liquid-air interface. In
our case, the test tubes were only lled up to about 75%.
As far as we know, Kerfoot et al. (1992) did the rst
experiments designed to isolate the eects of shock waves
on bacterial cells (
Pseudomonas aeruginosa, Streptococ-
cus faecalis, Staphylococcus aureus and Escherichia coli
)
and determine whether bactericidal activity exists. In this
study, the suspension received 200 shock waves at 20 kV
and a rate of 100 shocks per min on a HM3 Dornier elec-
trohydraulic lithotripter (Dornier Medizintechnik GmbH,
Germering, Germany). The experiment was repeated de-
livering 4000 shock waves at the same energy and rate.
Aliquots of bacterial suspensions of each of the four bac-
terial strains were also exposed to 4000 shock waves gen-
erated by a Wolf Piezolith 2200 piezoelectric lithotripter,
which does not generate UV radiation, at energy level 4
and a rate of 120 shock waves per min. Contrary to our
results, the authors concluded that shock waves do not
possess signicant bactericidal activity. It is important to
notice that, even if the MEXILIT II shock wave genera-
[ Pobierz całość w formacie PDF ]

  • zanotowane.pl
  • doc.pisz.pl
  • pdf.pisz.pl
  • tejsza.htw.pl

    •