Xinyu LioJio LiYunjie SuoShiguo ChenXingqin YeDonghong LiuTin Ding

aDepartment of Food Science and Nutrition,National Engineering Laboratory of Intelligent Food Technology and Equipment,Zhejiang University,Hangzhou,Zhejiang 310058,China

bKey Laboratory for Agro-Products Postharvest Handling of Ministry of Agriculture,Zhejiang Key Laboratory for Agro-Food Processing,Hangzhou,Zhejiang 310058,China

cFuli Institute of Food Science,Zhejiang University,Hangzhou 310058,China

Abstract

Keywords:Ultrasound;Inactivation mechanism;Reactive oxygen species(ROS);Staphylococcus aureus;Escherichia coli

1.Introduction

Ultrasound,pressure wave with a frequency of 20kHz or more,is known as potential non-thermal technology to ensure food safety without loss of food sensory properties[1,2].As early as 1929,it has been reported that ultrasound had the ability to kill vegetative microbes[3].Generally,high power ultrasound with lower frequency(20 to 100kHz)is employed for microbial inactivation in food industry[2].The mechanism of microbial inactivation by ultrasound is widely thought to be the responsibility of cavitation phenomenon[4–8].Under strong ultrasonic field,the bubbles will implode coincident with production of intensive pressure,high temperature or even active species(e.g.free radicals).So far,the inactivation mechanisms on microbes by ultrasound has been studied and investigated in some studies[9–11].The mechanical effects(e.g.shear force)generated by ultrasonic wave were commonly thought to the primary inactivation mechanism leading to the completed rupture of microbial cell envelopes and the death of microbes[11–18].However,apart from strong physical effect,increasing temperature and pressure during the collapse of bubbles was enough to trigger sonochemical reaction of surrounding medium and produce active compounds[19,20].According to that,few studies proposed another inactivated mechanism that free radicals produced by the explosion of bubbles might be transported into microbial cells through microjet and react with interior components with intact cell exterior structure[13,14,18,20].Sofar,the exact mechanisms underlying ultrasound inactivation still have no consensus and require to fully elucidate.

The aim of this study is to investigate the molecule mechanisms of ultrasound-induced microbial inactivation.Gram-negativeEsherichia coliand Gram-positiveStaphylococcus aureuswere used the model bacterium in this case.The extracelluar H2O2and intracellular reactive oxygen species(ROS),the membrane integrity and permeabilization,the membrane potentials,ATP production,DNA fragment and morphological properties were measured for the estimation of microbial cells physiological activity during ultrasound treatment in molecular level.We try to provide new insight into the action mode of ultrasound on microbial cells for future research.

2.Materials and methods

2.1.Bacterial strains preparations

Gram-negativeEsherichia coli(ATCC 25922,Hope Bio-Technology Co.,Ltd.,Qingdao,Shandong,China)and Gram-positiveStaphylococcus aureus(ATCC 25923,Hope Bio-Technology Co.,Ltd.,Qingdao,Shandong,China)were stored with 50% glycerol at?80°C.The stock culture of each strain was transferred into 100mL of nutrient broth(NB,Base Bio-Tech Co.,Hangzhou,China)and incubated at 37°C in a well-shaken of 150rpm to reach stationary phase,which required 18h forE.coliand 24h forS.aureus.The enriched culture was centrifuged at 5000rpm for 10min at 4°C to harvest bacteria cells.The precipitated cells were then washed for three times with 0.85% sterile saline solution.The final bacteria concentration of each strain determined by plating count method was approximately 9 log CFU/mL.

2.2.Ultrasound treatment

Thirty milliliter of cell suspension(about 8 log CFU/mL)was placed in a cylindrical tube with a volume of 85mL.An ultrasonic probe(diameter=10mm)was immersed 2.0cm into the bacteria solution. In this case, the input power was 198W,power intensity was 252W/cm2,frequency was 20kHz and treatment times were 0,3,5 and 12min.The temperature was maintained at20±2°C with the use of a thermostatic water bath(DC-1006,Safe Corporation,Ningbo,China).The details of the ultrasound equipment(Scientz-IID;Ningbo Scientz,Zhejiang,China)were described in our previous study[21].

2.3.Microbiological analysis

One milliliter of each sample was serial diluted with 0.85% sterile saline solution.One hundred microliter of appropriate dilution was plated on non-selective medium,tryptone soya agar(TSA,Hope Bio-Technology Co.,Ltd.,Qingdao,Shandong,China)and selective medium,supplemented TSA with 2%(w/w)sodium chloride forE.coliand 7%(w/w)sodium chloride forS.aureus.The plates were then incubated at 37°C for 48h under atmospheric conditions. Both viable and sublethally damaged cells appeared on non-selective medium,while microbial cells with compromised membranes failed to grow on the selective mediums.Therefore,the differences between non-selective and selective medium resulted from the occurrence of sublethal injury on microbes[22,23].Each experiment was conducted in triplicate independent cultures.

2.4.Membrane integrity

Propidium iodide(PI,Sigma-Aldrich Co.,USA)was used to indicate compromised cell membrane.One milliliter of bacteria sample was stained with 10μL of PI solution(1.5mM)for 30min at 37°C in the dark.The excess PI was removed by centrifugation of 8,000rpm at 4°C for 10min.The pelleted cells were then resuspended with 0.85% sterile saline solution and stored in the dark for no more than 1h until flow cytometry analysis.

2.5.Membrane permeabilization and esterase activity

Carboxyfluorescein diacetate(cFDA,Sigma-Aldrich Co.,USA)diffused freely through viable cell membrane and was converted by intracellular esterase into membrane-impermeable green fluorescent dye,carboxyfluorescein(cF).One milliliter of sample was incubated with cFDA solution(50μM)for 30min at 37°C.After centrifugation 8,000rpm at 4°C for 10min,the pellet cells were resuspended with 0.85% sterile saline solution to remove excess cFDA. The stained samples were stored in dark for no more than 1h and analyzed by flow cytometry(Beckman Coulter Inc.,Miami,FL,USA).

2.6.Membrane potential assessment

Membrane potential was measured with a BacLightTMBacterial Membrane Potential Kit(B34950,Molecular Probes,Invitrogen,Grand Island,NY).DiOC2(3)(3,3＇-diethyloxacarbocyanine iodide)was a probe, altering from green to red fluorescence as the membrane potential increasing.CCCP(carbonyl cyanide 3-chlorophenylhydrazone)could destroy cell membrane potential through eliminating membrane proton gradient.Specifically,10μL DiOC2(3)(3mM)was added into 1mL sample and mixed thoroughly.As for control,10μLCCCP(500μM)was transferred into sample and mixed before the addition of DiOC2(3).The mixture was then incubated for 30min at room temperature.The samples were then centrifuged and washed with 0.85% sterile saline solution to remove excess DiOC2(3).The stained samples were stored at dark measured by flow cytometer(Beckman Coulter Inc.,Miami,FL,USA).

2.7.Intracellular ROS detection

The intracellular reactive oxygen species(ROS)level was detected by 2,7-dichlorofluorescin diactate(DCFH-DA,Beyotime,Shanghai,China).DCFH-DA was hydrolyzed by intracellular esterase and then oxidized by intracellular ROS into fluorescent compound,2,7-dichlorofluorescin(DCF).Before ultrasound treatment,1mL of bacteria cells was incubated with 1μL of DCFH-DA solution(10mM)for 30min at 37°C and washed by 0.85% sterile saline solution to remove excess dyes.The DCF fluorescence was detected by flow cytometer(Beckman Coulter Inc.,Miami,FL,USA).

2.8.Extracellular H2O2concentration detection

The concentrations of H2O2generated in the medium was measured and quantified with a Hydrogen Peroxide Assay Kit(Beyotime,Shanghai,China)following the manufacturer’s guidelines.After ultrasound treatment,50μL of each sample solution was placed into the test tube.100μL of test solution was added into each tube and incubated for 30min under room temperature.Then,the absorbance at a wavelength of 560nm was measured immediately with a spectrophotometer.A standard curve with known H2O2concentrations was used for calibration of absorbance values.

2.9.ATP measurement

Intracellular ATP levels were measured by the Bac Tiler-Glo microbial viability assay(Promega,Wisconsin,USA).After ultrasound treatment,100μL of sample solution was placed in an opaque-walled multiwell plate(JingAn Biological Technology Co.,Ltd,Shanghai,China).One hundred microliter of BacTiter-GloTMRegent was added into each well contained samples.The mixture was shaken briefly and incubated for 5min.A Centro LB 960 Microplate Luminometer(Berthold Technologies GmbH&Co.KG,Bad Wildbad,Germany)was used to measure the luminescence values.The wells containing medium without bacteria cells were used to obtained the background luminescence value.

2.10.DNA damage analysis

The genomic DNA of bothE.coliandS.aureuswere extracted by a TIANamp Bacteria DNA Kit(TIANGEN BiotechCo.,Ltd.Beijing,China)following the manufacturer’s instructions.The genomic DNA extracted from bacteria was loaded on a 1% agarose gel with 100V for 40min.Then,the DNA bands were visualized with a long wave UV light.

2.11.Scanning electron microscope analysis

The cells were obtained by centrifugation at 4°C,12,000rpm for 10min and washed twice by 0.85% sterile saline solution.The washed samples were fixed with 2.5% glutaraldehyde(TAAB)for no less than 4h.Then,the samples were washed thrice by phosphate buffer(0.1M;pH 7.0)for 15min,followed by 1.5h of fixation with 1% OsO4.The fixed samples were washed tree times with phosphate buffer(0.1M;pH 7.0)for 15min.As for SEM analysis,the samples were dehydrated by a graded series of ethanol(30,50,70,80,90,95 and 100%)and transferred into a mixture consisting of ethanol and iso-amyl acetate(v:v=1:1)for around 30min.After that,samples were transported into pure iso-amyl acetate and incubated overnight.The samples were coated with the use of gold-palladium and observed under a Hitachi Model SU8010 SEM(Hitachi,Ltd.,Tokyo,Japan)with accelerating voltage of 3.0kV and working current of 10μA.

2.12.Laser confocal microscope analysis

The samples were stained with a Live/Dead BacLight bacterial viability kit(Molecular Probe Inc.,Eugene,OR,USA)according to the manufacturer’s protocol.Brie fly,the equal volume of Component A and Component B was mixed thoroughly.3μL of the dye mixture was mixed with 1mL of each sample thoroughly and incubated at room temperature in the dark for 15min.The stained sample was placed between a slide and a coverslip,and observed with a laser scanning confocal microscopy(LSCM)equipped with a Nikon TE 2000U microscope(Nikon Corporation,Tokyo,Japan).

2.13.Statistical analysis

All the experiments were conducted triplicate from independent trails.The data was performed and analyzed by AVOVA with SPSS v.20(SPSS Inc.,Chicago,IL,USA)software and expressed as mean value±standard deviation(SDs).The data was considered statistically significant at a probability value less than 0.05.

3.Results and discussion

3.1.Viability,sublethal injury and death evaluation

As Table 1 shown,the inactivation rate ofE.coliandS.aureuswas 98.1% and 81.3% with 12-min ultrasound exposure,respectively.The higher resistance ofS.aureusagainst ultrasound might be explained by the thicker,more rigid and robust properties of Gram-positive microbial cell envelopes[24–26].Monsen et al.[26]found that 60 min-ultrasound treatment with a frequency of 40kHz and input power of 350W resulted in more than 90% inactivation ofE.coli,but only 40% reduction ofS.aureus.Besides cell wall thickness,cell shape also plays significant role in ultrasonic resistance.Alliger[25]found that cocci cells showed more resistant towards ultrasound than cells with rod appearance.In our case,compared with rob-shapedE.colicells,spherical property ofS.aureuscells might assist them against ultrasonic stressor.

Table 1Surviving of Escherichia coli and Staphylococcus aureus on non-selective and selective medium after different treatments for various times.

Additionally,the differences of colony forming units between non-selective medium and selective medium didn’t appear significant for bothE.coliandS.aureusduring 12min ultrasound treatment(Table 1).The live/dead fluorescent assay for ultrasound-treatedE.coliandS.aureuscells exhibited only green and red fluorescence for live and dead subpopulations,but no yellow fluorescence(Fig.1).Yellow fluorescence represented the subpopulation sublethally damaged.These results revealed that ultrasound sterilization seemed to result in no sublthally injured Gram-negative and Gram-positive microbes in this case,which was described as an “all-or-nothing”phenomenon[27,28].Similarly,our previous study also found that no sublethal injury of bothE.coliandS.aureusinduced by ultrasound[21].What’s more,Wordon et al.[14]applied low-frequency ultrasound(20kHz)to inactivateSaccharomyces cerevisiaeand found only viable and dead but no sublethally injured cells during 20min treatment time.

3.2.Physiological activity estimation

The cell membrane plays essential role in maintaining normal physiological functions of microorganisms,including transportation of important materials and generation of ATP[29].Once the membrane is damaged irreversibly,microbial cells will be dead immediately.The cell membrane integrity was estimated by a fluorescence,PI.The microbial cell stained with PI meant the loss of membrane integrity.As Fig.2 shown,it was only 1.18% addition for Gram-negativeE.colistained PI after 12min ultrasound exposure.As for Gram-positveS.aureus,the subpopulation lacking of cell membrane integrity(PI-positive)increased by 20.49% during 12min ultrasound treatments.Cell membrane potential is indispensable for normal energy transduction and nutrient uptake in microbial cells and regarded as an important indicator for physiological activity[30–33].Membrane potentials inE.coliandS.aureuswere measured by the Bac LightTMBacterial Membrane Potential Kit.The fluorescent probe,DiOC2(3)exhibited green fluorescence in cells,but tended to emit red fluorescence as it accumulated in the cells due to higher membrane potential.The cells treated by a proton ionophore,CCCP,possessed totally depolarized membrane potentials,as a positive control in this case.As Table 2 shown,ultrasound treatment induced a peak on the membrane potential trend for bothE.coliandS.aureus.The maximum value of membrane potential appeared at 5min for Gram-negativeE.coliand 3min for Gram-positveS.aureus.The phenomenon of the cell membrane potential changes might be related to the change of ion channels on the cell membranes.At the initial phase,the mechanical stressor from ultrasound might trigger the response of partial microorganisms through the change of ion channel activity[34,35].When the bubbles collapsed,extremely high temperature was produced,resulting in pyrolysis of solvent into active species[20].In this case,water,the main sonication medium,generated H?and OH?radicals inside the bubble,which might subsequently recombine into stable molecules,such as hydrogen peroxide(H2O2)in the interfacial layer between the gaseous bubble content and the solvent[20].As Table 2 exhibited,the level of H2O2concentration in 0.85% saline solution increased from 0.16 to 7.25μM after 12min ultrasound exposure.Similarly,Stanley et al.[20]found that H2O2concentration in saline solution ascended as the accumulation of ultrasound(20kHz)treatment time.The ROS produced in the medium could be injected through microbial cell membranes by ultrasonic wave without damage of cell envelopes.As the intracellular ROS accumulation to a certain level(Table 2),calcium ion channel was trigged and lost its normal function to open and let the in flux of calcium ions,therefore the cell membrane potentials began to decrease[36,37].

Fig.2.Fluorescence dot plots of Escherichia coli(A–D)and Staphylococcus aureus(E-H)stained with PI after 0(Control),3(U-3),5(U-5),and 12(U-12)min ultrasound treatment.

Table 2Physiological states of Escherichia coli and Staphylococcus aureus induced by ultrasound treatments for various times.

As Table 2 shown,ultrasound exposure also resulted in a reduction of ATP level for bothE.coliandS.aureus.The start of significant ATP reduction for Gram-negativeE.coliappeared earlier than Gram-positiveS.aureus.3-min ultrasound exposure was enough to cause ATP inE.colito decrease, while it required 5-min ultrasound treatment forS.aureus(Fig.3).Intracellular esterase activity and cell permeabilization was estimated by a fluorescent probe,carboxyfluorescein(cF).As Fig.3 shown,cF-stainedE.colipopulations enjoyed share of 10.08,24.77,31.22 and 15.90% after 0,3,5 and 12min ultrasound exposure.The low proportions of cF-stained forE.colibefore treatments resulted from the presence of stable outer membranes in Gram-negative bacteria,which were composed of lipopolysaccharide(LPS)[11,38].The proportion ofS.aureuscF-stained population decreased from 99.88 to 69.51% during 12min ultrasound treatment.Another essential macromolecule,DNA,is detected by gel electrophoresis technique.As exhibited in Fig.4,compared with untreated cells,the fluorescent density of DNA bands decreased significantly for bothE.coliandS.aureusas treatment time increased.Particularly,the ultrasound treatedE.colicells showed nearly no obvious DNA bands after 3min treatment.

Fig.3.Fluorescence dot plots of Escherichia coli(A–D)and Staphylococcus aureus(E–H)stained with cFDA after 0(Control),3(U-3),5(U-5),and 12(U-12)min ultrasound treatment.

Fig.4.Agarose gel electrophoresis of genomic DNA of Escherichia coli(A)and Staphylococcus aureus(B)after 0(control),3(U-3),5(U-5),and 12(U-12)min ultrasound treatment.

The morphological change after ultrasound treatment was observed with scanning electron microscope(SEM).UntreatedE.coliandS.aureuscells showed intact cell walls and membranes(Fig.5A and C).With 12-min ultrasound treatment,a part of bothE.coliandS.aureusmicrobial cells have been disrupted completely and irreversibly into debris(Fig.5B and D).These debris wrapped up the remaining intact cells.

Fig.5.Scanning electron microscope(SEM)images of Escherichia coli(A–B)and Staphylococcus aureus(C–D)after 0(A and C)and 12min(B and D)ultrasound treatment.

3.3.More than simple cell envelop disruption

Interestingly,subpopulation with loss of cell membrane integrity(PI-stained)among bothE.coliandS.aureusshowed much lower than the inactivation rate from colony results,at the same time,the subpopulation without membrane permeabilization(cF-stained)enjoyed much higher share than survival level.It seemed that not all the dead microorganisms induced by ultrasound lost their cell membrane integrity.What’s more,the significantly decreasing of ATP level and the damage of DNA occurred inE.coliandS.aureus.These phenomena revealed that part of microorganisms suffered from intracellular damage by ultrasound but with intact cell exterior structure.Similarly,Wordon et al.[14]employed flow cytometry to demonstrate that ultrasound could induce intracellular injuries inSaccharomyces cerevisiaeprior to the complete disruption of cell membranes.Alliger et al.[25]also found that subcellular particles of sonicatedS.cerevisiaehad been damaged before breakage of cell walls.Additionally,Ashokkumar et al.[13]discovered that the sonicated Cryptosporidium oocysts with thin cell envelopes had no nuclei,and Yusof et al.[39]thought the generated free radicals might be injected through cavitation microjets into cells,causing the disintegration of nuclei without affecting cell membranes and walls.During ultrasound treatment,a portion of microorganisms,failing to enter into the area of acoustic cavitation,might have damage of intracellular components by ROS produced by ultrasound.As for the microbial cells,who situated closely to the valid distance of acoustic cavitation,the physical force would take the main responsible for the completed disruption of cell envelopes and the immediate death of microorganisms would occur(Fig.6).

Fig.6.The inactivation mechanisms of ultrasound on Escherichia coli(A)and Staphylococcus aureus(B).

4.Conclusions

In this study,we tried to demonstrate the action modes of ultrasound treatment on both Gram-negative and Gram-positive microorganisms.The results here indicated that ultrasound treatment might result in that some microorganisms in liquid medium had intracellular DNA broken and enzymes inactivation,but cause no any disruption of cell walls.These microbial cells might fail to undergo the effect of strong mechanical forces,which,however,the injected free radicals into cells would dam-age interior components severely.The findings in this study help to enhance the understanding of the ultrasound physical and chemical effect on microbial cells and provide avenues for the future research to clarify the ultrasonic inactivation mechanisms.

Conflict of interests

The authors declare that no Conflict of interests.

Acknowledgements

This study was supported by the National Major R&D Program of China(grant 2016YFD0400301)and the National Natural Science Foundation of China(grants 31401608).