Please download to get full document.

View again

of 7

Image Analysis Based Quantification of Bacterial Volume Change with High Hydrostatic Pressure

Image Analysis Based Quantification of Bacterial Volume Change with High Hydrostatic Pressure
0 views7 pages
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
         M      :       F      o      o       d       M       i      c      r      o       b       i      o       l      o      g      y       &       S      a       f      e      t      y  JFS  M: Food Microbiology and Safety ImageAnalysisBasedQuantificationofBacterialVolumeChangewithHighHydrostaticPressure M.P ILAVTEPE -C¸ ELIK  ,M.O.B  ALABAN ,H.A  LPAS ,  AND  A.E.Y  OUSEF  ABSTRACT: Scanning electron microscopy (SEM) images of   Staphylococcus aureus   485 and  Escherichia coli  O157:H7933weretakenafterpressuretreatmentsat200to400MPa.Softwaredevelopedforthispurposewasusedto analyze SEM images and to calculate the change in view area and volume of cells. Significant increase in averagecell view area and volume for  S. aureus   485 was observed in response to pressure treatment at 400 MPa. Cell view areafor E.coli  O157:H7933significantlyincreasedat325MPa,themaximumpressuretreatmenttestedagainstthispathogen. In contrast to  S. aureus  , cells of   E. coli   O157:H7 exhibited significant increase in average view area and volumeat200MPa.Thepressure-inducedincreaseintheseparametersmaybeattributedtomodificationsinmem-brane properties, for example, denaturation of membrane-bound proteins and pressure-induced phase transitionofmembranelipidbilayer.Keywords: highhydrostaticpressure,imageanalysis,microbialcell,scanningelectronmicroscopy,volume Introduction I nactivation of microorganisms by high hydrostatic pressure(HHP) has received great attention from investigators in recent years. Hydrostatic pressures below 200 MPa may cause bacterialcell injury whereas greater pressures are lethal to these cells (Ladoand Yousef 2002). Several mechanisms have been proposed forcauses of microbial inactivation by HHP, and currently this areais actively investigated. Recent findings show that pressure affectsprimary functions of the cell membrane through denaturation of membrane-bound proteins and pressure-induced phase transi-tions of the lipid bilayer (Ritz and others 2000; Ulmer and others2002).Electron microscopy has been employed to characterizepressure-induced morphological changes in microorganisms tounderstand the events leading to cell inactivation (Kaletunc¸ andothers 2004). Using scanning electron microscopy (SEM), mor-phological changes in  Leuconostoc mesenteroides   after pressuretreatmentat345MPawereevaluated.Itwasreportedthatwhilecellsize, shape, and surface structure of inactivated cells immediately after pressure treatment were not different from those of living cells, cell lysis was observed after 2 h storage at 4 ◦ C (Kalchayanandand others 2002).SEM examinations of   Listeria monocytogenes   cells revealed thatpressure treatment (400 MPa in citrate buffer, pH 5.6, for 10 minat 20  ◦ C) caused bud scars, pimple-like lesions, and swellings onthe surface of cells (Ritz and others 2002). These observations sug-gest that the cellular wall or membrane could be the target of high-pressure treatment.Limited studies addressed the changes in cell volume and view area induced by HHP, particularly by using SEM micrographs.Tholozan and others (2000) measured the intracellular volume of  MS 20080319 Submitted 4/29/2008, Accepted 8/12/2008. Author Pilavtepe-C¸elik is with Vocational School of Ihsaniye, Kocaeli Univ., 41040, ˙Izmit,Kocaeli,Turkey.AuthorBalabaniswithFisheryIndustrialTechnologyCen-ter, Univ. Alaska Fairbanks, 118 Trident Way, Kodiak AK 99615, U.S.A. Authors Pilavtepe-C¸elik and Alpas are with Food Engineering Dept., Mid-dle East Technical Univ., 06531 Ankara, Turkey. Author Yousef is with Dept.of Food Science and Technology, Ohio State Univ., 2015 Fyffe Rd., Colum-bus, OH 43210, U.S.A. Direct inquiries to author Pilavtepe-C¸elik (E-mail: L. monocytogenes   and  Salmonella typhimurium   with radioactive-labeled probes. Intracellular volume of   L. monocytogenes   in phos-phate buffer (pH 7.0) was 2.48  µ L/mg protein for reference and10.99  µ L/mg protein for the 600 MPa, 10 min, 20  ◦ C treatment.Reference volume of   S. typhimurium   (in phosphate buffer, pH 7.0) was 1.24 µ L/mg protein and its volume was 1.41 µ L/mg protein af-ter pressure treatment at 400 MPa, 10 min, 20  ◦ C. Perrier-Cornetand others (1995) developed an optical device to observe changesin  Saccharomyces cerevisiae   cell volume under high pressure treat-ment. An average decrease of 35% of the initial cell volume was ob-served for a pressure treatment of 250 MPa, 15 min.The objective of this study was to observe pressure-induced vol-ume changes in selected bacteria as determined by a combinationof SEM observations and mathematical modeling.  Staphylococcus aureus  485and Escherichiacoli   O157:H7933weretreatedwithhighpressure and examined by SEM. The resulting images were ana-lyzed, cell view areas were measured and a mathematical model was developed to calculate cell volume. Materials and Methods Highhydrostaticpressureequipment Bacterial cells were treated with high pressure at The OhioStateUniv.high-pressurefoodprocessingfacility(Columbus,Ohio,U.S.A.). A food processor with a 2-L capacity (Quintus QFP6; Flow Pressure Systems, Wash., U.S.A.) was used. A water/propylene gly-col (Houghton-Safe 620-TY, Houghton Intl., Inc., Valley Forge, Pa.,U.S.A.) mixture (1:1, v/v) was used as the working fluid, and its ini-tialtemperaturewascontrolledtoaccountforcompressionheating (3 to 4  ◦ C/100 MPa). The water jacket temperature was also main-tained at treatment temperature to meet the targeted final tem-perature during pressurization. The rate of pressure increase wasapproximately 400 MPa/min and pressure release time was lessthan 20 s. The pressure level, time, and temperature of pressuriza-tion were set manually and controlled during the treatment. Culturesandmedia  The pathogens used in this study were  S. aureus   485 (FDA FoodMicrobiology Laboratory, Wash., U.S.A.) and  E. coli   O157:H7 933 C  2008 Institute of Food Technologists  R  Vol. 73, Nr. 9, 2008 —  JOURNAL OF FOOD SCIENCE  M 423 doi: 10.1111/j.1750-3841.2008.00947.x Further reproduction without permission is prohibited  M  :    F     o    o    d    M  i      c   r    o    b    i      o   l      o     g     y    &    S     a   f      e   t       y    Pressure-induced cell volume change... (M. Doyle, Univ. of Georgia, Griffin, Ga., U.S.A.). Previous study had shown these strains to be relatively pressure resistant (Alpasand others 1999). The strains were cultivated in tryptic soy broth(Merck, Darmstadt, Germany) supplemented with 0.6% yeast ex-tract (Merck) at 37  ◦ C for 16 to 18 h and transferred to fresh brothevery 48 h. Preparationofbacteriaforpressuretreatment  Afterbacterialculturesreachedearlystationaryphase,0.1%pep-tone water (Merck) was inoculated with these cultures to obtainabout 10 7 CFU/mL. The resulting cell suspensions were aseptically transferred to sterile stomacher bags (Fisher Scientific, Pittsburgh,Pa., U.S.A.) in 2 mL portions, vacuum packaged, and the bags wereheat-sealed. The bags were placed inside a 2nd sterile stomacherbagandheat-sealedundervacuumtopreventcontaminationofthehigh-pressure unit if the primary package were to fail. The samples were kept on ice until pressurization. Pressuretreatment S. aureus   485 was subjected to HHP at 200, 250, 300, 350, and400 MPa for 5 min at 40  ◦ C. For  E. coli   O157:H7 933, HHP condi-tions were at 200, 250, 275, 300, and 325 MPa for 1 min 40  ◦ C. Pres-surizationtimesreporteddidnotincludethepressureincreaseandrelease times. A single pressure run was performed for each treat-ment. Immediately after pressurization, the bags were cooled in anice bath in preparation for SEM analysis. A control bag was heldin the ice bath at atmospheric pressure (0.1 MPa). These pressure–time parameters (Table 1) were chosen based on previously col-lectedinactivationkineticdataforeachbacterium(Pilavtepe2007).Differing pressure treatments allowed for optimal comparison be-tween the 2 pathogens. Inactivation level of   E. coli   O157:H7 933,corresponding to 1 min pressurization time, was estimated fromthe kinetic data that were fitted by a Weibul distribution model. Scanningelectronmicroscopeanalysis Immediatelyafterpressurization,cellpelletsof  S.aureus  485and E. coli   O157:H7 933 were prepared from untreated and pressure-treated cell suspensions by centrifugation at 10000 × g   for 10 minand washed twice with 0.1 M phosphate buffer, pH 7.4. Cell pel-lets were then resuspended in 1 mL 0.1 M phosphate buffer. Sus-pended bacteria were filtered (0.22  µ m MF-Millipore, GSWP, Mil-lipore Corp., Billerica, Mass., U.S.A.) and fixed on the membrane with 10 mL 3% glutaraldehyde in 0.1 M phosphate buffer (pH 7.4).Fixatives were left in contact with the cells overnight at 4  ◦ C. Mem-branes were transferred into glass vials and washed 3 times withbuffer, 10 min each, and postfixed for 1 h in 1% osmium tetroxide. Table 1---Selected pressure treatments for  Staphylococ- cus aureus   485 and  Escherichia coli   O157:H7 933 at40 ◦ C, and the corresponding lethalities. Pressure Time Log 10  reductionMicroorganism (MPa) (min) (CFU/mL) S. aureus   485 200 5 0.07250 5 0.17300 5 0.9350 5 2.91400 5 3.24 E. coli   O157:H7 933 200 1 0.04 a 250 1 0.15 a 275 1 0.86 a 300 1 1.84 a 325 1 1.48 a a Log 10  reduction values are estimated from the experimental data by using theWeibull distribution model (Pilavtepe 2007). Membranes were then rinsed twice with buffer, 10min each. Mem-branes were dehydrated through a series of ethanol (EtOH) solu-tions (50%, 70%, 80%, 95%, and 100% ethanol), rinsing with 10 mLofeachsolution,andholdingfor10minforeachrinse.Preparations were chemically dried using hexamethyldisilazane (HMDS). Mem-branes were washed with a series of HMDS solutions as follows: 3:1EtOH:HMDS, 1:1 EtOH:HMDS, 1:3 EtOH:HMDS, and 100% HMDS;all washings were carried out once for 15 min, except the 100%HMDS washing which was carried out twice, 15 min each. Samples were dried in a chemical hood overnight. Dried specimens weremounted on stubs and sputter-coated with a thin film of a heavy metal(gold/palladium)forelectricalconductivity.Thecoatedspec-imens were observed by the SEM (FEI Nova Nanosem 400, FEI Co.,Oreg., U.S.A.) at The Ohio State Univ., Campus Microscopy andImaging Facility, Columbus, Ohio, U.S.A. Depending on the spec-imen, 8 to 12 SEM images were taken from different regions of thesame dried specimen. Machinevisionanalysis DifferentSEMimagesofthesamepressurerunwereusedforcal-culating view area and volume. The SEM images (Figure 1) were“cleaned” using an image-editing software (Corel PhotoPaint 11,Corel Corp., Ottawa, Canada). This involved manually erasing thebackground, leaving bacterial cells visible in full form (that is, notpartially blocked by other cells), and excluding dividing cells, andthose touching the image edges. In addition, based on visual ob-servation, all those cells that were not oriented with their long axisparallel to the picture plane were cropped out and not included inanalysis. Number of cells that were cleaned and analyzed from dif-ferent SEM images is given in Table 2. The scale line of the SEMimage was used to make a size reference square using the rect-angle tool in the image editing software (assuming square pixels,Figure 2). Commercial software (LenseEye version 9.3.1, Engineer-ing and Cyber Solutions, Gainesville, Fla., U.S.A.) was used to cal-culate the view area of the cells, taking the square shape as a sizereference. Volume calculation was based on the assumption that thereexists a “curve of symmetry” (CS) of the 2-dimensional image(Figure 3A). The organism was assumed to have a volume of rev-olution around the CS. The curve of symmetry was found fromthe “distance transform (DT)” function (Alsuwaiyel and Gavrilova2000). First, the perimeter points of the image were identified, thenthe distances of every point within the image to all the perimeterpoints were calculated, and the minimum distance was taken foreach point. DT resulted in a CS (Figure 3B). Once the CS was de-fined, then perpendiculars from each CS node with respect to linesconnecting the previous node and extending them to perimeter weredrawnatregularintervals(Figure4A).Eachslicewasassumedto be a circular section. If the slices are not perpendicular to theCS, then the cross sections of the slices will not necessarily be cir-cles,butellipses;thiswouldhavemadethecalculationsmuchmorecomplex. The perpendicular sections are shown in Figure 4A. Cir-cularportionsofthisimagewererotatedaroundthesymmetryaxisto result in a 3-dimensional solid of rotation. The volume of eachslice was then calculated by assuming cylindrical cuts. Addition-ally, the ends were considered as spherical cuts and their volumes wereaddedtothetotal.InFigure4B,thecentersoftheleftandrightsphericalcapsareshown,aswellastheCS,theperpendicularlines,and the circular approximations. The 1st line on the left (denotedby 1) represents the spherical cut on the left, and the rightmostline (represented by 19) shows the spherical cut on the right. Thecylindricalcutsaretruncatedcylindersorcylindricalwedges—with1 cutting plane perpendicular and the other inclined to the axis. M 424  JOURNAL OF FOOD SCIENCE — Vol. 73, Nr. 9, 2008         M      :       F      o      o       d       M       i      c      r      o       b       i      o       l      o      g      y       &       S      a       f      e      t      y Pressure-induced cell volume change... Statisticalanalysis One-way analysis of variance (ANOVA) and Tukey’s pairwisecomparisons tests were performed to determine the effect of pres-sure on volume and view area of microorganisms. Statistical soft- ware (MINITAB 13 for Windows, Minitab Inc., State College, Pa.,U.S.A.) was used for this purpose. Calculated view area and volumevalues were presented as  ±  standard deviation (SD). Comparisontests were performed within the confidence interval of 95% ( P   < 0.05).The accuracy of the volume calculations depends on the num-ber of perimeter points, the number of slices, and the angle of the CS relative to a horizontal line. The effect of these variableson the calculated volume was investigated by using a test image(a prolate spheroid) developed with a known geometry (radii  = 8 and 4) and therefore a known volume of rotation (536.16)(Figure 5). The typical average error in calculating the volume was Figure 1---Scanning electronmicroscope micrograph of Escherichia coli   O157:H7 933 thatwas pressure-treated at 325 MPa and40  ◦ C for 1 min.Table 2---Average view area and volume of  Staphylococcus aureus   485 and  Escherichia coli   O157:H7 933 for eachpressure treatment at 40  ◦ C. Pressure–time Number of Average Average viewMicroorganism combination cells ( n  ) volume ( µ m 3 ) area ( µ m 2 ) S. aureus   485 Control 47 0.138 ± 0.028 0.356 ± 0.044200 MPa–5 min 48 0.138 ± 0.022 0.364 ± 0.038250 MPa–5 min 57 0.142 ± 0.018 0.366 ± 0.034300 MPa–5 min 53 0.144 ± 0.024 0.368 ± 0.043350 MPa–5 min 50 0.148 ± 0.025 0.374 ± 0.041400 MPa–5 min 70 0.154 ± 0.028 a 0.383 ± 0.042 a E. coli   O157:H7 933 Control 25 0.469 ± 0.110 0.938 ± 0.177200 MPa–1 min 30 0.676 ± 0.136 a 1.267 ± 0.202 a 250 MPa–1 min 39 0.544 ± 0.152 1.052 ± 0.214275 MPa–1 min 33 0.416 ± 0.105 0.913 ± 0.186300 MPa–1 min 31 0.430 ± 0.075 0.904 ± 0.123325 MPa–1 min 34 0.557 ± 0.125 1.097 ± 0.202 a ± Standard deviation (SD). a The increase in these values is significant compared to control cells ( P   <  0.05). 0.56% (range from –1% to 1%, data not shown). This method of cal-culatingvolumeseemsquitestable,independent oftheangleofCSfrom the horizontal, number of perimeters taken, and number of slicesallowed.Thissuggestsarobustmethodofvolumeestimation.Figure 6 shows details of volume calculation of the cells using theLensEye Software.TheeffectofSEMmagnificationontheviewareaofcellswasalsostudied. Average error was 2.5% for 4 different magnification levels(25000 × , 20000 × , 15000 × , and 10000 × ) on the same microorgan-isms (data not shown). Results and Discussion B iological specimens must generally first be converted to adry state before they can be studied by SEM to prevent ex-tensive changes to their 3-dimensional shape or chemical consti-tution (Sunner and others 2003). This involves extensive sample Vol. 73, Nr. 9, 2008 —  JOURNAL OF FOOD SCIENCE  M 425  M  :    F     o    o    d    M  i      c   r    o    b    i      o   l      o     g     y    &    S     a   f      e   t       y    Pressure-induced cell volume change... Figure 2---Isolated organisms fromFigure 1, with size reference square.Figure 3---(A) An isolated organism, (B) its perimeter andits curve of symmetry obtained by image analysis. preparations (fixation and dehydration). There are several tech-niques that have been successfully used to study biological spec-imens by SEM. However, there are also considerable problems. Withrespecttothemaingoalofpreservingmorphologicalfeatures,present methods are only moderately successful. In particular, ex-tensive shrinking almost always occurs. Fratesi and others (2004)testedsometechniquesontheappearanceofbacteriaandbiofilmsin the carter sandstone and observations showed that the srcinal Figure 4---(A) Volume calculation detail for the organismin Figure 3, (B) with different number of volume slicestaken. morphology of individual bacteria was best preserved by ethanoldehydration with hexamethyldisilazane (HMDS) drying.Gusnard and Kirschner (1977) found that the critical point dry-ing process itself caused most of the shrinkage in mouse hepato-cytenucleiandinhumanerythrocytes.Glutaraldehydefixationandethanol dehydration caused only minimal size reduction, prior tocriticalpointdrying.Substitutionofaninertdehydrationtechniquedid not alter the final result. M 426  JOURNAL OF FOOD SCIENCE — Vol. 73, Nr. 9, 2008         M      :       F      o      o       d       M       i      c      r      o       b       i      o       l      o      g      y       &       S      a       f      e      t      y Pressure-induced cell volume change... The SEM procedure applied in our study includes ethanol de-hydration with HMDS drying and prior fixation with glutaralde-hyde and osmium tetroxide. According to literature we can say thatshrinkage problems were minimized in terms of volume measure-ments by the applied SEM procedure. It seems logical that if boththe control and the treated cells are similarly prepared, then thechange in volume would be the same. Average view area and volume of   S. aureus   485 and  E. coli  O157:H7 933 cells, calculated for each pressure treatment, are pre-sented in Table 2. View area and volume distributions of   S. aureus  485 and  E. coli   O157:H7 933 cells are given in Table 3 to 6. Aver-age view area and volume of   S. aureus   485 cells did not change forpressure levels of 200, 250, 300, and 350 MPa compared to con-trol cells, whereas the increase in these values was significant at400 MPa ( P   <  0.05). Average volume for  E. coli   O157:H7 933 cellsdid not change for pressure levels of 250, 275, 300, and 325 MPacompared to control cells ( P   <  0.05). However, the increase in av-erage view area was significant at 325 MPa for  E. coli   O157:H7933 cells ( P   <  0.05). Interestingly, the increase in average view area and volume was significant at 200 MPa for  E. coli   O157:H7933 cells.Tholozan and others (2000) measured the intracellular volumeof  L.monocytogenes   and S.typhimurium  cellswithradioactivelyla-beled probes.  L. monocytogenes   cell volume was unchanged up toa pressure treatment of 325 MPa in citrate buffer (pH 5.6), and upto425MPainphosphatebuffer(pH7.0).Thehigher-pressuretreat-ments causing more than 3 log  10  reduction of the cell populationsin both buffers led to an increase in the calculated cell volume upto fourfold. Cell volume of   S. typhimurium   suspensions was un-changed regardless the pressure treatment or the buffer used tosuspend the cells. An insitu  observationof  Saccharomycopsisfibuligera  underhighhydrostatic pressure was reported by Perrier-Cornet and others(1995). Cell volume variations were measured and shrinkage inaverage cell volume was observed after a pressure treatment of 250 MPa. The observed compression rate (25%) under pressureand the partial irreversibility of cell compression (10%) after re-turntoatmosphericpressureledtotheconclusionthatmasstrans-fer between cell and cultivation medium occurred. In the study by Perrier-Cornet and others (1995), the histogram of the initial yeastcell volume had a normal distribution. During the pressure appli-cation, cell volume kept the same normal distribution form butclearly shifted to a smaller volume range. This displacement cor-responded to an average decrease of 35% in volume. When yeasts were observed after pressurization, population histogram did notdiffer from that of the population under high pressure. It was con- Figure 5---Test image to confirm volume calculationaccuracy. cluded that yeast cells did not recover their initial volume after ahydrostatic treatment of 250 MPa for 15 min. This observation hasconfirmed the hypothesis of an irreversible mass transfer that oc-curred during pressure holding time.In the current study, the view area and volume of   S. aureus   485and the view area of   E. coli   O157:H7 933 increased in responseto the highest-pressure treatment tested. However, a lower pres-sure (200 MPa) increased the view area and volume of   E. coli  O157:H7 933 cells. On the distribution representations of Table3 to 6, view area and volume distributions of pressurized cellsshifted to the higher levels (increase in size) compared to thecontrol distribution. These results are similar to those reportedby Tholozan and others (2000). Behavior of   S. aureus   485 underpressure, in this study, is comparable to that observed by Perrier-Cornet and others (1995) on yeast cells. Yeasts and  S. aureus   cellsshare spherical morphology; this may explain their similar volumechange in response to pressure. The increase in volume at high-pressurelevelscanbeexplainedbyamasstransferbetweencelland Figure 6---Volume calculation detail for different organ-isms from Figure 2. Vol. 73, Nr. 9, 2008 —  JOURNAL OF FOOD SCIENCE  M 427
View more
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!