APPLIED AND ENVIRONMENTAL MICROBIOLOGY,0099-2240/01/$04.0010 DOI: 10.1128/AEM.67.1.420–425.2001

Jan. 2001, p. 420–425 Vol. 67, No. 1

Copyright © 2001, American Society for Microbiology. All Rights Reserved.

Direct In Situ Viability Assessment of Bacteria in Probiotic DairyProducts Using Viability Staining in Conjunction with

Confocal Scanning Laser MicroscopyM. A. E. AUTY,1* G. E. GARDINER,1 S. J. MCBREARTY,1 E. O. O’SULLIVAN,2 D. M. MULVIHILL,3

J. K. COLLINS,2 G. F. FITZGERALD,2 C. STANTON,1 AND R. P. ROSS1

Teagasc, Dairy Products Research Centre, Moorepark, Fermoy,1 and Department of Food Science and Technology3

and Department of Microbiology,2 University College Cork, Cork, County Cork, Ireland

Received 24 April 2000/Accepted 21 July 2000

The viability of the human probiotic strains Lactobacillus paracasei NFBC 338 and Bifidobacterium sp. strainUCC 35612 in reconstituted skim milk was assessed by confocal scanning laser microscopy using the LIVE/DEAD BacLight viability stain. The technique was rapid (<30 min) and clearly differentiated live from heat-killed bacteria. The microscopic enumeration of various proportions of viable to heat-killed bacteria was thencompared with conventional plating on nutrient agar. Direct microscopic enumeration of bacteria indicatedthat plate counting led to an underestimation of bacterial numbers, which was most likely related to clumping.Similarly, LIVE/DEAD BacLight staining yielded bacterial counts that were higher than cell numbers obtainedby plate counting (CFU) in milk and fermented milk. These results indicate the value of the microscopic ap-proach for rapid viability testing of such probiotic products. In contrast, the numbers obtained by directmicroscopic counting for Cheddar cheese and spray-dried probiotic milk powder were lower than thoseobtained by plate counting. These results highlight the limitations of LIVE/DEAD BacLight staining and theneed to optimize the technique for different strain-product combinations. The minimum detection limit for insitu viability staining in conjunction with confocal scanning laser microscopy enumeration was ;108 bacte-ria/ml (equivalent to ;107 CFU/ml), based on Bifidobacterium sp. strain UCC 35612 counts in maximum-recovery diluent.

Probiotics are described as “living micro-organisms, whichupon ingestion in certain numbers exert health benefits beyondinherent basic nutrition” (12). Accumulating clinical evidencesupporting the health-promoting characteristics of Lactobacil-lus and Bifidobacterium intestinal isolates (for reviews see ref-erences 24 and 29) has led to increased commercial interest indeveloping novel probiotic food products. Those productswhich have received the most attention as probiotic carriersinclude fermented milks, unfermented milks with culturesadded, ice cream, frozen yogurt, and various cheeses (for re-views see references 18, 31, 32, and 34). It has been suggestedthat probiotic products should contain at least 107 CFU per mlor g (15).

Bacterial viability is typically assessed by plate counting on asuitable growth medium. However, there are a number ofdisadvantages associated with this approach. For example,plate counting is time-consuming, often requiring 2 to 3 days ofincubation, microorganisms may be unevenly distributed in theproduct, and bacteria may occur in chains and/or clumps, re-sulting in underestimation of the true bacterial count (6). Inaddition, oxidative killing of anaerobic microorganisms such asBifidobacterium during plating may also contribute to an un-derestimation of bacterial numbers. A more direct approachmay be the use of a microscopic technique; however, thisrequires differentiation of live and dead bacteria. Direct epi-

fluorescent counting has been described as a suitable methodfor enumeration of total bacteria in environmental samples”(17). Fluorescence microscopy has the advantage of allowing arapid and direct assessment of cell viability (17, 22), althoughparticular strains cannot be identified. Fluorescent indicatorsof viability may be based on membrane integrity, enzyme ac-tivity, membrane potential, respiration, or pH gradient (9, 20,21, 23, 27, 33). The LIVE/DEAD BacLight viability kit (Mo-lecular Probes Inc., Eugene, Oreg.) was developed to differ-entiate live and dead bacteria based on plasma membranepermeability and has been used to monitor growth of bacterialpopulations (38). This kit comprises two fluorescent nucleicacid stains: SYTO9 and propidium iodide. SYTO9 (excitationand emission maxima, 480 and 500 nm) penetrates both viableand nonviable bacteria (Handbook of Fluorescent Probes andResearch Chemicals, 6th ed., Molecular Probes, Inc.), whilepropidium iodide (excitation and emission maxima, 490 and635 nm) penetrates bacteria with damaged plasma membranesonly (16, 21), quenching the green SYTO9 fluorescence. Thus,bacterial cells with compromised membranes fluoresce red andthose with intact membranes fluoresce green.

Confocal scanning laser microscopy (CSLM) has been usedextensively in cell biology (39) and was used to study viabilityof Escherichia coli and Salmonella where rhodamine 123 andpropidium iodide were employed to differentiate viable fromnonviable bacteria based on membrane potential and integrity(19). Conventional epifluorescence microscopy may be usedfor viability staining of liquid samples such as milk (26). How-ever, the optical sectioning capability of CSLM has the advan-tages of increased sensitivity and reduced out-of-focus blur,

* Corresponding author. Mailing address: Teagasc, Dairy ProductsResearch Centre, Moorepark, Fermoy, Co. Cork, Ireland. Phone: 3532542447. Fax: 353 2542340. E-mail: mauty@moorepark.teagasc.ie.

420

on June 8, 2020 by guesthttp://aem

.asm.org/

Dow

nloaded from

enabling observation of subsurface structures of foods in situ(3, 13). In addition, digital acquisition of images by CSLMenables rapid enumeration of bacteria by image analysis (4).

In this study, in situ LIVE/DEAD BacLight bacterial viabil-ity staining in conjunction with CSLM was compared withstandard plate counting for enumeration and viability assess-ment of bacteria in various probiotic dairy products, includingreconstituted skim milk (RSM), fermented milk, full-fat ched-dar cheese, and spray-dried probiotic milk powder.

MATERIALS AND METHODS

Bacterial strains and media. The potentially probiotic strains Lactobacillusparacasei subsp. paracasei NFBC 338, Bifidobacterium sp. strain UCC 35612, andBifidobacterium sp. strain UCC 401 were isolated from the human gastrointes-tinal tract (5, 25) and were obtained from University College Cork, Cork, Ire-land, under a restricted-materials transfer agreement. The Lactobacillus strainwas cultured as described previously (10), while the Bifidobacterium strain wascultured in MRS broth (Difco Laboratories, Detroit, Mich.) supplemented with0.05% (wt/vol) cysteine HCl (Sigma-Aldrich Ireland, Dublin, Ireland) (7). Forcheddar cheese manufacture, cheesemaking starters Lactococcus lactis subsp.cremoris strains 223 and 227 and an adjunct culture of Bifidobacterium lactisBb-12 were obtained from C. Hansen Laboratories (Little Island, Cork, Ireland)in the form of freeze-dried pellets. L. paracasei NFBC 338 was enumerated inmilk and dairy products by pour plating on MRS agar (Difco Laboratories).Tryptone-phytone-yeast extract agar containing NPNL selective solution (neo-mycin sulfate [20 mg/liter], paromomycin sulfate [40 mg/liter], nalidixic acid [3mg/liter], lithium chloride [600 mg/liter]) (30, 37) was used for selective enumer-ation of bifidobacteria from fermented milk and cheddar cheese, while MRS agarsupplemented with 0.05% (wt/vol) cysteine HCl was used for enumeration fromRSM suspensions. All dilutions were performed using maximum-recovery di-luent (MRD) (Oxoid Ltd., Basingstoke, Hampshire, United Kingdom) andplates were incubated under anaerobic conditions at 37°C for 3 days for bothlactobacilli and bifidobacteria.

Plate count enumeration of probiotic bacteria in milk and dairy products.Cells from 100 ml of stationary-phase cultures of L. paracasei NFBC 338 orBifidobacterium sp. strain UCC 35612 were concentrated by centrifugation at3,640 3 g for 10 and 20 min, respectively. The resultant cells were then resus-pended by vortex mixing for 10 s in 100 ml of RSM (10% wt/vol) that hadpreviously been sterilized at 121°C for 5 min. A 10-ml sample of the RSMcontaining resuspended cells was then incubated at 90°C for 5 min. Unheatedand heat-treated RSM cell suspensions were vortex mixed in various proportionsto give mixtures of live and dead bacteria in which the proportion of live bacteriavaried in 10% increments from 0 to 100% (vol/vol). Bacteria in mixtures con-taining 0, 10, 50, 90, and 100% (vol/vol) live bacteria were enumerated asoutlined above.

Pasteurized whole milk supplemented with skim milk powder (16.5% totalsolids) was heat treated at 90°C for 15 min. After cooling to 37°C, the milk wasinoculated (2% vol/vol) with an overnight broth culture of Bifidobacterium sp.strain UCC 401 and incubated at 37°C for 24 h until a pH of 4.8 was reached.Duplicate samples of fermented milk were emulsified in sterile 2% (wt/vol)trisodium citrate, and serial dilutions in MRD were pour plated as describedabove.

A pilot-scale cheesemaking trial was performed according to the experimentalprotocol described by Gardiner et al. (10). A control vat contained a 1.5%(vol/vol) inoculum of starter cultures only and the experimental vat contained anadditional culture of Bifidobacterium sp. strain Bb12, added as an adjunct to thestarter culture to yield ;108 CFU bifidobacteria per ml of cheesemilk. Cheeseswere sampled at 1 and 3 months and bifidobacteria were enumerated by platecounting as described for probiotic fermented milk above.

To manufacture a probiotic-containing spray-dried powder, cells from anovernight MRS broth culture of L. paracasei NFBC 338 (200 ml) were resus-pended in 1,500 ml of RSM (25% wt/vol), which had been previously heat treatedat 90°C for 30 min. The suspension was spray dried in a Buchi B191 mini-spraydryer (Buchi Labortechnik AG, Flawil, Switzerland) as previously described (11).The inlet air temperature was set at 160°C and outlet air temperatures rangingfrom 71 to 78°C were used, yielding skim milk powders containing viable lacto-bacilli at ;1010 CFU/g. Lactobacilli were enumerated in duplicate following 2months of storage at 4°C.

In situ viability staining and CSLM imaging. All microscopy work was per-formed using an LSM310 confocal scanning laser microscope (Carl Zeiss Ltd.,

Welwyn Garden City, Herts., United Kingdom) using the method involvingLIVE/DEAD BacLight viability staining essentially as previously described (11).Randomly selected areas of each sample were imaged using a 363 magnificationobjective with a numerical aperture of 1.4. Confocal illumination was provided bya Kr/Ar laser (488-nm laser excitation) fitted with a long-pass 514-nm emissionfilter. A 580-nm beam splitter was used together with a long-pass 520-nm filter(green fluorescence signal) and long-pass 590-nm filter (red fluorescence signal).Simultaneous dual-channel imaging using pseudocolor was used to display greenand red fluorescence. The confocal pinhole was set to give an x-y resolution of 0.2mm and an axial resolution of 1.0 mm. Red-green-blue images (24 bit), 512 by 512pixels, were acquired using a zoom factor of 2.0, giving a final pixel resolution of0.2 mm/pixel and representing a volume of 1.05 3 1028 ml per field of view. Thus,for direct enumeration of bacteria per milliliter, a microscopic factor of 1.05 3108 was used. For triple-channel imaging, a transmitted photodetector was usedin conjunction with interference contrast optics and the transmitted image wascolored blue. Image analysis was performed on CSLM images using a KontronKS400 image analysis system (Imaging Associates Ltd., Thame, Oxfordshire,United Kingdom). Images of stained bacteria were segmented using colorthresholding to separate the red and green fluorescence signals. Two parameterswere then measured: (i) green fluorescence as a percentage of total green andred fluorescence and (ii) numbers of individual green fluorescing bacteria. Toseparate clusters of bacteria, erosion-dilation algorithms included in the imageanalysis software were used. Direct microscopic counts were normalized to takeinto account the dilution effect caused by adding the viability stain to the sample.To simultaneously visualize the structure of the spray-dried particles and the red-and green-fluorescing bacteria, triple-channel imaging was used. To confirm thatthe glycerol-based staining mixture did not affect viability staining, live andheat-killed L. paracasei NFBC 338 microorganisms in RSM were prepared asdescribed above. When mixed with the glycerol-based stain at a ratio of 1:1, livecells fluoresced green and heat-killed cells fluoresced red.

(i) Minimum detection limit of the in situ viability staining and CSLMenumeration method using bifidobacteria suspended in MRD. To establish thesensitivity of the in situ viability staining technique, an actively growing brothculture of Bifidobacterium sp. strain UCC 35612 was diluted in MRD to yield anapproximate log dilution series of 105 to 109 CFU/ml. Plate count enumerationof the broth culture and in situ staining with CSLM enumeration of the dilutionseries were performed as described for RSM. Bacteria from 50 microscopic fieldswere counted using image analysis, and results were expressed as numbers ofbacteria per milliliter.

(ii) In situ viability staining and CSLM enumeration of lactobacilli andbifidobacteria in RSM. The LIVE/DEAD BacLight viability stain, preparedaccording to the manufacturer’s instructions, was incubated with equal volumesof milk containing 0 to 100% live bacteria, prior to CSLM imaging. The speci-ficity of the two individual LIVE/DEAD BacLight staining components wasverified in the milk by adding 5 ml of each staining component to separate 100-mlsamples of milk inoculated with either a live culture or a heat-treated culture ofL. paracasei NFBC 338. CSLM imaging confirmed that SYTO9 stained both liveand dead bacteria green, whereas propidium iodide stained only heat-killedbacteria red (data not shown). Results from in situ viability staining were ob-tained within 30 min of sampling the milk.

(iii) In situ viability staining and CSLM enumeration of bacteria in fermentedmilk, cheddar cheese, and spray-dried probiotic milk powder. Equal volumes ofprobiotic fermented milk (pH 5.6) and LIVE/DEAD BacLight viability stainwere vortex mixed for 1 min and green fluorescent bacteria from 20 fields wereenumerated. LIVE/DEAD BacLight viability stain (25 ml) was also added tofreshly cut sections of 2-month-old cheddar cheese, and a coverslip was placed ontop. CSLM images were obtained ;10 mm below the level of the coverslip after20 min of incubation in the dark at room temperature. CSLM imaging data fromdry powders were compared with data from the reconstituted product (10%[wt/vol]). To prevent dissolution of the spray-dried powder particles during insitu viability staining, a glycerol-based staining mixture was prepared from theLIVE/DEAD BacLight staining components, as follows. SYTO9 and propidiumiodide were each dissolved in separate 1-ml samples of distilled water to give finalconcentrations of 60 and 300 mM, respectively. Seventy-five microliters ofSYTO9 solution and 25 ml of propidium iodide solution were then added, withvortex mixing, to 400 ml of glycerol (Sigma-Aldrich Ireland). This ratio of SYTO9to propidium iodide was found to be optimal for the production of an adequategreen fluorescence signal. Spray-dried probiotic milk powder (;10 mg) wasgently mixed with 10 ml of this staining mixture on a microscope slide.

Statistical analysis. The significance of the difference between the meansobtained by direct microscopic counting and plate count enumeration was de-termined by a one-tailed Student t test (20 df).

VOL. 67, 2001 CSLM AND STAINING OF BACTERIA IN DAIRY PRODUCTS 421

on June 8, 2020 by guesthttp://aem

.asm.org/

Dow

nloaded from

RESULTS AND DISCUSSION

Minimum detection limit of the in situ viability staining andCSLM enumeration method using Bifidobacterium sp. strainUCC 35612. In order to relate in situ viability staining andCSLM enumeration to plate count data, it was first necessaryto establish the minimum detection limit of the in situ CSLMtechnique. Results of CSLM enumeration of LIVE/DEADBacLight-stained bifidobacteria in MRD indicated a minimumdetection limit of ;108 bacteria/ml or ;107 CFU/ml fromplating (Table 1). These data suggest that plate counting un-derestimates the actual viable cell population by a factor of atleast 10, confirming reports by other researchers (6, 17, 28).Results further indicate that LIVE/DEAD BacLight viabilitystaining may be a suitable means of assessing in situ the via-bility of bacteria in probiotic foods, given that the recom-mended minimum number of probiotic bacteria in such foodproducts is approximately 107 CFU/ml (15). It should be noted,however, that the sensitivity of the viability staining methodcould be greatly increased by filtration and/or centrifugation toconcentrate the recovered cells (17, 26).

In situ viability staining and CSLM enumeration of probi-otic bacteria in RSM. The specificity of the LIVE/DEADBacLight stain was then assessed using known ratios of live todead (heat-killed) bacteria in RSM. Both L. paracasei NFBC338 and Bifidobacterium sp. strain UCC 35612 produced astrong red or green fluorescence depending on whether thecultures were dead or live, respectively (Fig. 1A and B). Back-ground fluorescence from milk proteins was low, enabling cleardiscrimination of viable and nonviable bacterial cells. CSLMobservations indicated that cells of L. paracasei NFBC 338were often in short chains of four cells in addition to largerclumps of up to 200 microorganisms. In contrast, Bifidobacte-rium sp. strain UCC 35612 appeared as individual cells or smallclumps of ,4 cells. A good correlation was obtained betweenpercentages of live bacteria and green cells (green fluorescenceexpressed as a percentage of red and green fluorescence), asmeasured by analysis of CSLM images for L. paracasei NFBC338 (R2 5 0. 99) and for Bifidobacterium sp. strain UCC 35612(R2 5 0.98). This indicates that in situ viability staining andCSLM imaging constitute a valid quantitative technique forestimating the proportion of viable to dead bacterial cells inmilk.

Data obtained by direct CSLM enumeration of bacteria werethen compared with those obtained by plate counting for L. pa-

racasei NFBC 338 and Bifidobacterium sp. strain UCC 35612 inRSM. The correlations of 0.98 and 0.89 for various ratios oflive to dead L. paracasei NFBC 338 and Bifidobacterium sp.strain UCC 35612, respectively, confirm the quantitative capa-bility of in situ viability staining and CSLM enumeration (Fig.2). Relative to the direct microscopic counts, however, L. pa-racasei NFBC 338 and Bifidobacterium sp. strain UCC 35612plate counts were approximately 20-fold and 10-fold lower,respectively. These results are consistent with the greater de-gree of clumping exhibited by L. paracasei NFBC 338 (Fig. 1)compared with Bifidobacterium sp. strain UCC 35612.

In situ viability staining and CSLM enumeration of bi-fidobacteria in probiotic fermented milk. In situ LIVE/DEADBacLight staining showed red- and green-fluorescing bacteriaoccurring singly or in small clumps of up to 20 cells in theprobiotic fermented milk (Fig. 1C). Some background fluores-cence was present in the green channel. Bacterial cells wereirregularly shaped rods with occasional branching, a morpho-logic characteristic of some Bifidobacterium sp. (30). Enumer-ation by CSLM indicated a viable count equivalent to 3.2 3 108

bacteria/ml, comparing favorably to the plate count of 2.3 3108 CFU/ml (Table 2). The higher count (P , 0.001) obtainedby direct microscopic enumeration was most likely due toclumping of bacteria and killing on media selective for bi-fidobacteria.

In situ viability staining and CSLM enumeration in probi-otic cheddar cheese. The in situ viability staining and CSLMimaging technique was used to enumerate total viable bacteriadirectly from cheddar cheese ripened for 2 months and com-pared with plate counts for enumeration of viable bifidobac-teria (Table 2). Enumeration by CSLM indicated a total viablecount equivalent to 1.5 3 108 bacteria/g, which was lower thanthe bifidobacterial plate count of 3.6 3 108 CFU/g. The pro-biotic cheddar cheese contained approximately twice as manyviable bacteria as the control cheese, as determined by the insitu CSLM method. If higher bacterial counts in the probioticcheese were exclusively due to bifidobacteria, in situ viabilitystaining indicated a count of approximately 7.5 3 107 bi-fidobacteria/g. Positive identification of bifidobacteria in situwould require an alternative approach such as fluorescent insitu hybridization or immunofluorescent labeling (14). There-fore, it was not possible to distinguish bifidobacteria in thecheese from nonstarter lactic acid bacteria; rather, it is thecomparison with total numbers in the control cheese that isgiven. However, the probiotic cheddar cheese contained sev-eral star-shaped clusters of rod-shaped bacteria (Fig. 1E) typ-ical of some Bifidobacterium strains (30), including Bifidobac-terium lactis Bb-12. These clusters were not present in thecontrol Cheddar cheese (Fig. 1D). Cell morphology was con-firmed by adjusting the focal plane of the CSLM. Bacteria werenot homogeneously distributed but frequently occurred inclumps of up to 20 cells. Some background fluorescence of theprotein matrix was seen in the green channel, although this wasat a lower intensity than bacterial fluorescence. Fat globulesappeared as dark rounded regions by negative contrast as ob-served in a previous study (8). The homogeneous staining ofthe protein matrix with SYTO9 was most likely due to non-specific binding of the stain to milk proteins. Small (,2-mm)patches of diffuse red fluorescence, possibly due to exogenous

TABLE 1. Direct microscopic counts by in situ viability stainingand CSLM enumeration and plate counts of serially

diluted Bifidobacterium sp. strain UCC 35612 inmaximum-recovery diluent

Plate count(CFU/ml)a

Direct microscopic count of green (viable)fluorescent bacteria

Mean of 50 fields No. of bacteria/ml, normalized

1.25 3 105 0 01.25 3 106 0.1 2.1 3 107

1.25 3 107 2.3 4.8 3 108

1.25 3 108 14.6 3.1 3 109

1.25 3 109b 172.8 3.6 3 1010

a Estimated from serial dilutions of broth culture.b Mean of triplicate analyses.

422 AUTY ET AL. APPL. ENVIRON. MICROBIOL.

on June 8, 2020 by guesthttp://aem

.asm.org/

Dow

nloaded from

FIG. 1. CSLM images of dairy products stained with LIVE/DEAD BacLight viability stain. (A and B) L. paracasei NFBC 338 and Bifidobac-terium sp. strain UCC 35612, respectively, suspended in reconstituted skim milk. Dual-channel CSLM images represent a 50:50 mixture of live andheat-treated bacteria. Live bacteria are green; dead bacteria are red. Note the clumping of lactobacilli in panel A (arrow). Bar 5 10 mm. (C)Dual-channel CSLM projection (z depth, 15 mm) of probiotic milk fermented with Bifidobacterium sp. strain UCC 401. Note extensive clumpingof bacteria. Bar 5 25 mm. (D and E) Dual-channel CSLM images of control (D) and probiotic (E) cheddar cheese. Note the star-shaped clusterof rod-shaped cells characteristic of some Bifidobacterium strains (E, large arrow), short rods and cocci of presumptive nonstarter lactic acidbacteria (small arrows), and background green fluorescence of protein matrix with dark spaces containing fat. Bar 5 25 mm. (F) Probiotic skimmilk powder stained in situ with glycerol-based LIVE/DEAD BacLight viability stain. Triple-channel CSLM image shows live and dead lactobacilli(green and red, respectively) and transmitted image (blue). The arrow indicates a powder particle. Bar 5 5 mm.

VOL. 67, 2001 CSLM AND STAINING OF BACTERIA IN DAIRY PRODUCTS 423

on June 8, 2020 by guesthttp://aem

.asm.org/

Dow

nloaded from

microbial nucleic acids, were also observed in both probioticand control cheddar cheese samples (data not shown).

In situ viability staining and CSLM enumeration of spray-dried L. paracasei NFBC 338 in skim milk powder. The countsof live bacteria in spray-dried form, as determined by imageanalysis of CSLM images and plate counts, are shown in Table2. CSLM enumeration indicated a viable count of 6.3 3 108

bacteria/g in the rehydrated powder, which was significantlylower (P , 0.05) than that obtained by plate counting (1.1 3109 CFU/g). Triple channel imaging using the glycerol-basedmixture of propidium iodide and SYTO9 enabled in situobservation of both red- and green-fluorescing L. paracaseiNFBC 338 cells within the powder particles (Fig. 1F). A lowlevel of background fluorescence from the milk powder wasobserved in the green channel. Serial CSLM optical sectionsindicated that bacteria were encapsulated within the spray-dried powder particles, confirming earlier work (11). Highernumbers of bacteria fluoresced green in the rehydrated than inthe dry powder. The low number of green-fluorescing bacteria

(,1 bacterium/field) in the dry powder compared with that inthe rehydrated powder suggested that the bacterial plasmamembrane was compromised in the dehydrated state, as ex-pected (1, 35), but recovered somewhat when rehydrated.Spray drying has been shown to result in cell membrane dam-age, as indicated by the increased sensitivity of L. paracaseiNFBC 338 to NaCl following drying (11). It is possible thatreversible melting of membrane lipids at temperatures of;50°C (36) and/or removal of bound water from cell wallproteins during the drying process (1) may be responsible. Ithas been reported that slow rehydration procedures can in-crease the viability of spray-dried L. bulgaricus (35). For moredetailed study of the effect of sublethal stress on bacterialviability, other fluorescent viability indicators, such as esteraseactivity, membrane potential, or respiratory activity, may bemore suitable than techniques based on membrane permeabil-ity (2, 19).

Conclusions. The results of this study indicate that in situLIVE/DEAD BacLight viability staining and CSLM enumer-

FIG. 2. Correlation between direct microscopic counts (mean of six replicates) of green-fluorescing bacteria stained with LIVE/DEADBacLight viability stain and plate counts (mean of duplicates) of L. paracasei NFBC 338 (R2 5 0.98) (A) and Bifidobacterium sp. strain UCC 35612(R2 5 0.89) (B) in reconstituted skim milk. The error bars represent 95% confidence intervals.

TABLE 2. Comparison of LIVE/DEAD BacLight viability staining and direct CSLM microscopic enumeration withplate counting for probiotic bacteria in dairy products

Product Sample Microscopic count(viable bacteria/ml or g)a

Plate count(CFU/ml or g)b P

Fermented milk Bifidobacterium sp. strain UCC 401 3.2 3 108 (1.6 3 107) 2.3 3 108 (1.1 3 108) ,0.001Cheddar cheese Control 7.0 3 107 (1.0 3 107) 3.3 3 105 (4.2 3 105) ,0.001

Bifidobacterium sp. strain Bb12 1.5 3 108 (5.6 3 107) 3.6 3 108 (2.8 3 107) ,0.001Spray-dried powder L. paracasei NFBC 338 6.3 3 108 (3.6 3 108) 1.1 3 109 (6.4 3 107) ,0.05

a Mean of 20 fields. Values in parentheses are standard deviations.b Mean of duplicate analyses. Values in parentheses are standard deviations.

424 AUTY ET AL. APPL. ENVIRON. MICROBIOL.

on June 8, 2020 by guesthttp://aem

.asm.org/

Dow

nloaded from

ation may be of value for the rapid estimation of viable bac-teria in some dairy products, which could take over 3 days toachieve by plate counting. The data demonstrate that micro-scopic viability counting of probiotic milk and fermented milkyield consistently higher counts (up to 20-fold for milk) thanplate counting. This may be expected given the high degree ofclumping observed with some of the strains and the possiblekilling of cells by selective media. Microscopic counts werelower than plate counts for cheese products and spray-driedcultures, highlighting the need for further work to establish theeffect of environmental factors such as pH, ionic profile, andwater activity on viability staining.

ACKNOWLEDGMENTS

This work was supported by the European Research and Develop-ment Fund and by the European Union (SM&T-CT98–2235). G.E.G.and S.J.M. were supported by Teagasc Walsh Fellowships.

REFERENCES

1. Brennan, M., B. Wanismail, M. C. Johnson, and B. Ray. 1986. Cellulardamage in dried Lactobacillus acidophilus. J. Food Prot. 49:47–53.

2. Breuwer, P. 1996. Assessment of viability of micro-organisms employingfluorescence techniques. Ph.D. thesis. Aula vab de Landbouwuniversiteit teWageningen, Wageningen, The Netherlands.

3. Brooker, B. E. 1995. Imaging food systems by confocal scanning laser mi-croscopy, p. 53–68. In E. Dickenson (ed.), New physico-chemical techniquesfor the characterisation of complex food systems. Blackie Academic andProfessional, London, United Kingdom.

4. Caldwell, D. E., D. R. Korber, and J. R. Lawrence. 1992. Confocal lasermicroscopy and digital image analysis in microbial ecology. Adv. Microb.Ecol. 12:1–67.

5. Collins, J. K., G. Thornton, and G. O’Sullivan. 1998. Selection of probioticstrains for human applications. Int. Dairy J. 8:487–490.

6. Daley, R. J. 1979. Direct epifluorescence enumeration of native aquaticbacteria: uses, limitations and comparative accuracy, p. 29–45. In J. W.Costerton (ed.), Native aquatic bacteria: enumeration, activity and ecology.ASTM STP 605. American Society for Testing and Materials, Philadelphia,Pa.

7. de Man, J. C., M. Rogosa, and M. E. Sharpe. 1960. A medium for thecultivation of lactobacilli. J. Appl. Bacteriol. 23:130–135.

8. Everett, D. W., K. Ding, N. F. Olson, and S. Gunasekaran. 1995. Applica-tions of confocal microscopy to fat globule structure in cheese. In E. L. Malinand M. H. Tunick (ed.), Chemistry of structure-function relationships incheese. Plenum Press, New York, N.Y.

9. Gant, V. A., G. Warnes, I. Phillips, and G. F. Savidge. 1993. The applicationof flow cytometry to the study of bacterial responses to antibiotics. J. Med.Microbiol. 39:147–154.

10. Gardiner, G. E., R. P. Ross, J. K. Collins, G. F. Fitzgerald, and C. Stanton.1998. Development of a probiotic cheddar cheese containing human-derivedLactobacillus paracasei strains. Appl. Environ. Microbiol. 64:2192–2199.

11. Gardiner, G. E., E. O’Sullivan, J. Kelly, M. A. E. Auty, G. F. Fitzgerald, J. K.Collins, R. P. Ross, and C. Stanton. 2000. Comparative survival of human-derived Lactobacillus paracasei and L. salivarius strains during heat treat-ment and spray drying. Appl. Environ. Microbiol. 66:2605–2612.

12. Guarner, F., and G. J. Schaafsma. 1998. Probiotics. Int. J. Food Microbiol.39:237–238.

13. Heertje, I., P. van der Vlist, J. C. G. Blonk, H. A. C. Hendrickx, and G. J.Brackenhof. 1987. Confocal scanning laser microscopy in food research:some observations. Food Microstruct. 6:115–120.

14. Hugenholtz, J., H. Veldkamp, and W. N. Konings. 1987. Detection of specificstrains and variants of Streptococcus cremoris in mixed cultures by immuno-fluorescence. Appl. Environ. Microbiol. 53:149–155.

15. Ishibashi, N., and S. Shimamura. 1993. Bifidobacteria: research and devel-

opment in Japan. Food Technol. 47:126–135.16. Jepras, T. I., J. Carter, S. C. Pearson, F. E. Paul, and M. J. Wilkinson. 1995.

Development of a robust flow cytometry assay for determining numbers ofviable bacteria. Appl. Environ. Microbiol. 61:2696–2701.

17. Kepner, R. J., and Pratt, J. R. 1994. Use of fluorochromes for direct enu-meration of total bacteria in environmental samples: past and present. Mi-crobiol. Rev. 58:603–615.

18. Lee, Y. K., and S. Salminen. 1995. The coming of age of probiotics. TrendsFood Sci. Technol. 6:241–245.

19. Lopez-Amoros, R., S. Castel, J. Comas-Riu, and J. Vives-Rego. 1997. As-sessment of E. coli and Salmonella viability and starvation by confocal lasermicroscopy and flow cytometry using rhodamine 123, DiBAC4(3), pro-pidium iodide and CTC. Cytometry 29:298–305.

20. Mason, D. J., R. Allman, and D. Lloyd. 1993. Uses of membrane potentialdyes with bacteria, p. 67–81. In D. Lloyd (ed.), Flow cytometry in microbi-ology. Springer-Verlag, London, United Kingdom.

21. Mason, D. J., R. Allman, J. M. Sark, and D. Lloyd. 1994. Rapid estimationof antibiotic susceptibility with flow cytometry. J. Microsc. 176:8–16.

22. McFeters, G. A., P. Y. Feiping, B. H. Pyle, and P. S. Stewart. 1995. Physio-logical assessment of bacteria using fluorochromes. J. Microbiol. Methods21:1–13.

23. Molenaar, D., T. Abee, and W. N. Konings. 1991. Continuous measurementof the cytoplasmic pH in Lactococcus lactis with a fluorescent pH indicator.Biochim. Biophys. Acta 1115:75–83.

24. Naidu, A. S., W. R. Bidlack, and R. A. Clemens. 1999. Probiotic spectra oflactic acid bacteria (LAB). Crit. Rev. Food Sci. Nutr. 38:13–126.

25. O’Riordan, K., and G. F. Fitzgerald. 1998. Evaluation of bifidobacteria forthe production of antimicrobial compounds and assessment of performancein cottage cheese at refrigeration temperature. J. Appl. Microbiol. 85:104–114.

26. Pettipher, G. L., R. Mansell, C. H. McKinnon, and C. Cousins. 1980. Rapidmembrane filtration epifluorescent microscopy technique for direct enumer-ation of bacteria in raw milk. Appl. Environ. Microbiol. 39:423–429.

27. Rodriguez, G. G., D. Phipps, K. Ishiguro, and H. F. Ridgway. 1992. Use ofa fluorescent redox probe for visualization of actively respiring bacteria.Appl. Environ. Microbiol. 58:1801–1808.

28. Roszak, D. B., and R. R. Colwell. 1987. Survival strategies of bacteria in thenatural environment. Microbiol. Rev. 51:365–379.

29. Salminen, S., A. G. Ouwehand, and E. Isolauri. 1998. Clinical applications ofprobiotic bacteria. Int. Dairy J. 8:563–572.

30. Scardovi, V. 1986. Genus Bifidobacterium Orla-Jensen 1924, 472AL, p. 1418–1434. In P. H. Sneath, N. S. Nair, M. E. Sharpe, and J. G. Holt (ed.), Bergey’smanual of systematic bacteriology, vol. 2. Williams and Wilkins, Baltimore,Md.

31. Shah, N. P. 1997. Bifidobacteria: characteristics and potential for applicationin fermented milk products. Milchwissenschaft 52:16–20.

32. Stanton, C., G. E. Gardiner, P. B. Lynch, J. K. Collins, G. F. Fitzgerald, andR. P. Ross. 1998. Probiotic cheese. Int. Dairy J. 8:491–496.

33. Stubberfield, L. C. F., and P. J. A. Shaw. 1990. A comparison of tetrazoliumreduction and FDA hydrolysis with other methods of microbial activity. J.Microbiol. Methods 12:151–162.

34. Tamime, A. Y., V. M. Marshall, and R. K. Robinson. 1995. Microbiologicaland technological aspects of milks fermented by bifidobacteria. J. Dairy Res.62:151–187.

35. Teixeira, P., H. Castro, and R. Kirby. 1995. Spray drying as a method forpreparing concentrated cultures of Lactobacillus bulgaricus. J. Appl. Bacte-riol. 78:456–462.

36. Teixeira, P., H. Castro, C. Mohacsi-Farkas, and R. Kirby. 1997. Identifica-tion of sites of injury in Lactobacillus bulgaricus during heat stress. J. Appl.Microbiol. 83:219–226.

37. Teraguchi, S., M. Uehara, K. Ogasa, and T. Mitsuoka. 1978. Enumeration ofbifidobacteria in dairy products. Jpn. J. Bacteriol. 33:753–761.

38. Virta, M., S. Lineri, P. Kankaanpaa, M. Karp, K. Peltonen, J. Nuutila, andE.-M. Lilius. 1998. Determination of complement-mediated killing of bac-teria by viability staining and bioluminescence. Appl. Environ. Microbiol.64:515–519.

39. Wright, S. J., V. E. Centonze, S. A. Stricker, P. J. DeVries, S. W. Paddock,and G. Schatten. 1993. Introduction to confocal microscopy and three-dimensional reconstruction. Methods Cell Biol. 38:1–45.

VOL. 67, 2001 CSLM AND STAINING OF BACTERIA IN DAIRY PRODUCTS 425

on June 8, 2020 by guesthttp://aem

.asm.org/

Dow

nloaded from