• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of aemPermissionsJournals.ASM.orgJournalAEM ArticleJournal InfoAuthorsReviewers
Appl Environ Microbiol. Apr 2004; 70(4): 1895–1906.
PMCID: PMC383152

Relative Ability of Orally Administered Lactobacillus murinus To Predominate and Persist in the Porcine Gastrointestinal Tract

Abstract

Five porcine-derived Lactobacillus or Pediococcus isolates administered to pigs (n = 4), either singly or as a combination at ~1010 CFU per day varied with respect to intestinal survival and persistence. Two Lactobacillus murinus strains survived best and were excreted at ~107 to 108 CFU/g of feces. In contrast, Pediococcus pentosaceus DPC6006 had the lowest fecal count at ~105 CFU/g and was excreted at a significantly lower level than both L. murinus strains. Fecal L. murinus DPC6003 counts were also significantly higher than both Lactobacillus salivarius DPC6005 and Lactobacillus pentosus DPC6004 (~106 CFU/g). The L. murinus strains persisted for at least 9 days postadministration in both the feces and the cecum. Animals fed a combination of all five strains excreted ~107 CFU of the administered strains/g, with L. murinus predominating, as determined by randomly amplified polymorphic DNA PCR. Postadministration, variation was observed between animals fed the strain combination, but in general, L. murinus DPC6002 and DPC6003 and L. pentosus DPC6004 predominated in the feces and the cecum while P. pentosaceus DPC6006 was detected only in the cecum. Fifteen days after the start of culture administration, mean fecal Enterobacteriaceae counts were significantly lower in some of the treatment groups. In addition, when mean preadministration counts were compared with those obtained after 21 days of culture administration, Enterobacteriaceae counts were reduced by ~87 to 98% in pigs fed L. salivarius DPC6005, P. pentosaceus DPC6006, L. pentosus DPC6004, and the culture mix. In conclusion, the porcine intestinal isolates have potential as probiotic feed additives for pigs, with differences in strain performance highlighting the advantages of using culture combinations.

Intestinal carriage of enteropathogens, such as Salmonella, Escherichia coli O157, Yersinia, and Campylobacter, in farm animals is a cause for concern, as carcass contamination at slaughter can lead to pathogen transmission to humans. Antibiotics are generally unsuitable for the treatment of carrier animals (1, 29), and the use of antibiotics in animal production, for growth promotion and treatment or prevention of disease is controversial, as it has been associated with the emergence of antibiotic-resistant pathogens. Multiresistant strains of Salmonella and E. coli, vancomycin-resistant enterococci, and fluoroquinolone-resistant Campylobacter are being isolated from pigs and poultry in particular at increasing frequencies (1). In light of these concerns and the implementation of European legislation banning the addition of certain antibiotic growth promoters to animal feed, alternatives to antibiotics in animal production are highly desirable (1).

Probiotics, aimed at restoration and maintenance of a healthy gut microflora, are live microorganisms which when administered in adequate amounts exert a health benefit on the host (9). As microbial feed additives, they offer potential as an alternative to antibiotics, both as a means of controlling pathogen carriage and improving growth rate and feed conversion. The cultures most commonly used include lactic acid bacteria, Bacillus organisms, and yeasts such as Saccharomyces boulardii (36). Data from animal trials have shown that probiotics can improve growth performance in pigs (4, 39), chickens (19), and lambs (21). Competitive exclusion cultures have been applied successfully to animals for the reduction of enteropathogen carriage, with probably the best example being a commercially available Food and Drug Administration-approved 29-strain mix that significantly reduces Salmonella carriage in chickens (27). Inoculation with probiotic bacteria can also reduce fecal shedding of E. coli O157:H7 in both cattle (44) and lambs (21). Probiotics offer potential for use in pigs as a means of improving performance and health, particularly in light of the fact that pig rearing has become more intensive in recent years (40). Reducing pathogen carriage is also a potential target, considering that pigs are the predominant carriers of Salmonella, with 23% of pig cecal samples in one study identified as Salmonella positive at slaughter compared with only 0.2 and 0.1% of cattle and sheep rectal samples, respectively (6).

Although the exact mode of action of probiotics is unknown, suggested mechanisms include immunomodulation, suppression of pathogens through competitive exclusion, and/or the production of inhibitory compounds such as organic acids, hydrogen peroxide, and bacteriocins (40). The use of high levels of viable microorganisms selected in vitro on the basis of these criteria may improve probiotic efficacy in vivo. Other selection criteria considered important include survival and persistence in the host, safety, and technological suitability (8, 12). Many studies have reported the isolation and selection of potential probiotic strains for use in pigs (4, 17, 25). However, the results of in vivo feeding trials can be variable (36). This is perhaps understandable, given that the complexity of the intestine can lead to variation between individual animals. Furthermore, commercial animal probiotic products may not contain the strains or species listed on the label at an adequate probiotic dose or may have no indication that the strains used possess any of the recommended probiotic properties (2, 41). On the other hand, while undefined cultures used as competitive exclusion products are particularly effective in pigs (10, 15), uncertainty regarding their exact composition has led to concerns that they may result in pathogen transmission. Therefore, there is a need for rational selection and characterization of strains intended for use as probiotic feed additives. Given that the performance of probiotic strains can vary between individual animals, a good case can be made for the use of mixtures, but the most appropriate approach may be to isolate and characterize individual strains prior to their combination in a probiotic product.

Although there is increasing interest in the use of probiotics as prophylactic and therapeutic agents in pig production, relatively few strains have demonstrated efficacy in vivo. The objective of the present study was to investigate the performance in pigs of five cultures when administered orally both individually and as part of a strain combination. These cultures were previously isolated from the porcine cecum and characterized in vitro with regard to their probiotic potential, based on a number of selection criteria, including the ability to inhibit Salmonella enterica serovar Typhimurium (3).

MATERIALS AND METHODS

Bacterial strains and culture conditions.

Lactobacillus murinus DPC6002 and DPC6003, Lactobacillus pentosus DPC6004, Lactobacillus salivarius DPC6005, and Pediococcus pentosaceus DPC6006, previously isolated from pig cecal contents, were selected from a bank of porcine intestinal isolates on the basis of molecular typing and properties such as antimicrobial activity, bile tolerance, and growth in milk (3). Spontaneous rifampin-resistant (Rifr) variants of these porcine isolates required to facilitate subsequent enumeration in the pig intestine were isolated by spread plating ~109 CFU from an overnight culture onto MRS agar containing 100 μg of rifampin (Sigma Chemical Co., Poole, Dorset, United Kingdom)/ml. Following anaerobic incubation at 37°C for 3 days, the colonies that had grown were selected and stocked in MRS broth containing 40% (vol/vol) glycerol. To confirm that the rifampin-resistant variants were identical to the parent strains, molecular fingerprinting by randomly amplified polymorphic DNA (RAPD) analysis was employed, as outlined below. Both parent and variant strains were routinely cultured at 37°C in MRS broth (Difco Laboratories, Detroit, Mich.) in anaerobic jars with CO2-generating kits (Anaerocult A; Merck, Darmstadt, Germany).

Genetic fingerprinting by RAPD PCR.

RAPD PCR analysis was performed on each administered Lactobacillus or Pediococcus strain and on intestinal isolates recovered from pig fecal and cecal samples. Initially, genomic DNA was isolated from 1.5 ml of overnight MRS broth cultures according to the method outlined by Coakley et al. (5). The extracted DNA was then used as a template in PCR amplifications, which were performed with either R1 (5′-ATGTAACGCC-3′) or R2 (5′-GTGATGTGCTGGTGTTATGTTTA-3′) random primers (manufactured by MWG Biotech, Ebersberg, Germany) as outlined by Gardiner et al. (11) with the following modifications. PCR amplifications were performed in a total volume of 50 μl in an Eppendorf DNA thermal cycler (Eppendorf Scientific Inc., Westbury, N.Y.) with 1.25 U of Taq DNA polymerase (Bioline, London, United Kingdom) added to the reaction mix. The PCR products (10 μl of each reaction mixture) were analyzed on a 1.5% (wt/vol) agarose (Sigma) gel, with a 100-bp ladder (New England Biolabs, Hitchin, Hertfordshire, United Kingdom) as a molecular size standard.

Survival of porcine isolates in gastric juice.

Gastric contents collected from 12 porcine stomachs (obtained from a local slaughter plant) and ranging in pH from 3.1 to 4.0 were pooled and filtered through glass wool. Porcine gastric juice was obtained by centrifugation twice at 13,000 × g for 30 min and filtered through a Whatman no. 113 filter. The gastric juice was then checked for sterility by pour plating on brain heart infusion agar (Merck) and incubating the plates aerobically at 30°C for 2 days. The pH was adjusted to 3.0 with 1 N HCl, and the gastric juice was stored at −20°C until use. For each of the porcine cultures, overnight MRS broth cultures were diluted 10-fold in 1 ml of one-fourth strength Ringers solution (Lab M Ltd., Bury, Lancashire, United Kingdom) and cells were harvested from this dilution, washed twice in one-fourth strength Ringers solution, and resuspended in 1 ml of gastric juice to achieve a final cell concentration of ~107 to 108 CFU/ml. Inoculated gastric juice was incubated in a shaking water bath at 37°C. Samples (100 μl) were taken at 0, 5, 15, 30, 45, and 60 min and immediately added to 900 μl of Ringers solution. Culture viability was assessed by further diluting 10-fold in Ringers solution, spread plating 5-μl volumes of appropriate dilutions in quadruplicate on MRS agar, and incubating the plates anaerobically at 37°C for 2 days. The experiment was performed in triplicate with three separate overnight cultures for each porcine isolate.

Preparation of skim milk cultures for pig-feeding trial.

To investigate the optimal skim milk medium for growth and survival of the porcine strains, each of the five rifampin-resistant variants was inoculated at a rate of 1% (vol/vol) from an overnight MRS broth culture into both 100 ml of heat-treated (121°C for 5 min) 10% (wt/vol) reconstituted skim milk (RSM) and 100 ml of heat-treated 10% (wt/vol) RSM supplemented with 0.5% (wt/vol) yeast extract (Merck). Inoculated milks were then incubated at 37°C for 18 h, and the fermentation was terminated by cooling on ice. The pH was measured with an MP220 pH meter (Mettler-Toledo GmbH, Greifensee, Switzerland), and the cultures were stored at 4°C for 5 days. Viable plate counts were performed on both freshly grown and stored cultures by serially diluting samples in maximum recovery diluent (Oxoid Ltd., Basingstoke, Hampshire, United Kingdom) and pour plating on MRS agar which was incubated anaerobically at 37°C for 3 days. Based on the results of these experiments, 10% (wt/vol) RSM supplemented with 0.5% (wt/vol) yeast extract was chosen as the optimal skim milk medium for growth and survival of all of the strains except for L. salivarius DPC6005, which was grown without the addition of yeast extract. For administration to pigs, the isolates were grown in 450- to 900-ml volumes of the relevant skim milk medium. The resultant culture fermentates were then aliquoted into 100-ml volumes, stored at 4°C, and used within 5 days. In addition, a culture mix was prepared by mixing equal volumes of fermentates of each of the five porcine isolates. Throughout the pig-feeding trial, bacterial numbers were routinely checked in each batch of culture prepared by plating as outlined above.

Pig-feeding trial.

The pig-feeding trial complied with European Union Council Directive 91/630/EEC, which lays down minimum standards for the protection of pigs, and European Union Council Directive 98/58/EC, which concerns the protection of animals kept for farming purposes. A total of 28 crossbred (Large White × Landrace) pigs (14 males and 14 females) were weaned at 24 to 28 days and blocked by sex and weight. Pigs within each block were assigned at random to one of seven treatment groups (n = 4) as follows: group A, control; group B, L. salivarius DPC6005; group C, P. pentosaceus DPC6006; group D, L. pentosus DPC6004; group E, L. murinus DPC6002; group F, L. murinus DPC6003; group G, culture mix containing all five strains. Each animal was penned individually, with control animals penned in isolation from culture-fed animals to prevent cross-contamination. In addition to the cultures or skim milk administered throughout the trial as outlined below, all animals had unrestricted access to water and nonmedicated creep feed manufactured in the Moorepark feed mill. The creep feed was formulated to contain 14.5 MJ of digestible energy per kg and 1.29% (wt/wt) total lysine by using the following ingredients: full fat soy (29% wt/wt), dried whey (20% wt/wt), barley (15.5% wt/wt), wheat (15% wt/wt), maize (17.5% wt/wt), and minerals and vitamins (1.8% wt/wt) with synthetic amino acids added.

The feeding trial consisted of three consecutive periods: baseline period (10 days), culture administration period (21 days), and postadministration period (9 days). During the baseline period, each pig received 100 ml of sterile 10% (wt/vol) RSM supplemented with 0.5% (wt/vol) yeast extract per day. During the administration period, 100 ml of skim milk culture fermentate containing on average ~3 × 108 CFU of the relevant porcine strain/ml (prepared as outlined above) was administered daily to each of the pigs in groups B, C, D, E, and F, giving a total average daily intake of ~3 × 1010 CFU. Pigs in group G received 100 ml of skim milk culture fermentate (~3 × 108 CFU/ml) containing a mixture of all five strains in approximately equal proportions daily, providing a total dose of ~3 × 1010 CFU/day. Pigs in the control group (A) each received 100 ml of sterile skim milk adjusted to pH 4.7 with food-grade lactic acid (Purac Biochem, Gorinchem, The Netherlands) per day. Thereafter, during the postadministration period neither culture fermentate nor skim milk was administered to the animals.

Nine days after culture administration had ceased, two pigs per treatment group were sacrificed by electrical stunning followed by bleeding. Immediately after slaughter, the cecum was removed and the cecal contents were collected and stored on ice during transport to the laboratory.

Microbiological analysis of pig fecal and cecal samples.

Fecal samples were obtained from each animal prior to (day −5), during (days 3, 8, 15, and 22), and 5 days after (day 26) culture administration, where day 1 was the first day of culture administration. Fecal samples were stored at 4°C and analyzed within 24 h. We have previously demonstrated no difference in bacterial counts between fresh and stored samples (data not shown). Fecal samples were homogenized in maximum recovery diluent as 10-fold dilutions with a stomacher (Lab-Blender 400; Seward Medical, London, United Kingdom) and further diluted, and the appropriate dilutions were pour plated. The administered strains were enumerated on MRS-RIF agar, i.e., MRS agar containing 100 μg of rifampin/ml as a selective agent and 50 U nystatin (Sigma)/ml to inhibit yeasts and molds, following anaerobic incubation for 2 days at 37°C. Fecal bacteria in the family Enterobacteriaceae were enumerated on violet red bile glucose agar (Merck) incubated at 37°C for 24 h. Total Lactobacillus counts were obtained on Lactobacillus-selective agar (35) (Becton Dickinson, Cockeysville, Md.) following anaerobic incubation at 37°C for 5 days. In the same way, the administered cultures, Enterobacteriaceae, and total lactobacilli were enumerated in the pig cecal content samples collected at slaughter. The pH of the cecal contents was also measured with a Mettler Toledo MP220 pH meter.

In addition, representative colonies randomly selected from MRS-RIF plates from day 15 and 26 fecal samples and cecal content samples were analyzed by RAPD PCR, as outlined above, and fingerprints were compared with those of the relevant administered strain(s).

Statistical analyses.

For all pig fecal and cecal counts, mean values were calculated for each treatment and the results are presented with standard errors of the means or standard deviations (SD). Analysis of variance (split-plot in time design), used to compare treatments and time points, was performed by using Genstat (16), and Tukey's test was used for separation of means.

RESULTS AND DISCUSSION

The pig-derived strains used in the present study have been characterized by a range of in vitro probiotic tests, including assays for acid and bile tolerance, anti-Salmonella activity, and adherence to intestinal epithelial cells (3). However, while laboratory testing can provide information useful for selection of potentially effective probiotic strains, performance in the gut and effects on intestinal microflora can only be accurately determined in vivo. In the present study, following preliminary gastric transit simulation studies, five different strains were administered to pigs both individually and as a culture mix to investigate survival and persistence in the porcine gastrointestinal tract (GIT) and to assess any impact on the resident gut microflora.

DNA fingerprinting of porcine cultures by RAPD PCR.

RAPD, a molecular technique which employs PCR with primers of arbitrary sequence, is important as an adjunct to the use of rifampin-resistant variants to facilitate tracking of the strains in the porcine GIT and has previously been used in human trials (14). It enables differentiation of the administered strains from the indigenous gut microflora and is useful for identification of the predominating strain(s) in animals fed a strain combination. In this study, RAPD was used to discriminate each of the five porcine isolates, with L. salivarius DPC6005, L. pentosus DPC6004, and P. pentosaceus DPC6006 yielding unique reproducible DNA fingerprints with the R1 random primer (data not shown). However, the L. murinus strains DPC6002 and DPC6003 could not be differentiated from each other with this primer. A range of additional primers was investigated, and one (R2) was found to successfully differentiate these closely related strains (data not shown). The RAPD method was also used to demonstrate that the rifampin-resistant variants yielded identical fingerprints to those of the corresponding parent strains (data not shown).

Survival of porcine isolates in porcine gastric juice.

Among the desirable properties recommended for a probiotic microorganism is the ability to tolerate the acidic conditions encountered in the stomach so that bacterial viability is maintained during gastric transit (8, 12). The potential of each of the porcine cultures to resist gastric transit was initially investigated by means of an in vitro simulation which determined the survival of the strains in porcine gastric juice (pH 3) over a 60-min period. While the viability of all isolates was dramatically reduced under these conditions, L. pentosus DPC6004 was the most tolerant, with 4.6 × 103 CFU/ml of viable cells remaining following the treatment (10,000-fold reduction) (Fig. (Fig.1).1). In comparison, the other four strains tested survived for up to 45 min in gastric juice (pH 3.0), with numbers of P. pentosaceus DPC6006 and L. murinus DPC6002 and DPC6003 reduced to between 2.9 × 101 and 7.4 × 104 CFU/ml at this time point (Fig. (Fig.1).1). L. salivarius DPC6005 was the least tolerant to simulated gastric transit conditions, and only 1.7 × 101 CFU/ml could be recovered after 30 min of exposure, although it was also still detectable at low levels after 45 min of exposure (Fig. (Fig.1).1). Porcine lactobacilli previously investigated by in vitro gastric transit studies also demonstrated considerable strain variation, with an Lactobacillus fermentum strain declining only 10-fold after 60 min of exposure to rabbit gastric juice (pH 2.0), whereas Lactobacillus acidophilus declined 100,000-fold under the same conditions (43). While many studies have investigated L. acidophilus strains of human origin (26, 32), Dunne et al. (8) compared the survival of a range of Lactobacillus spp. in human gastric juice (pH 2.5) and found that all but one survived well for 60 min. In comparison, an Enterococcus faecium strain previously investigated in our laboratory was undetectable after only 8 min of exposure to porcine gastric juice (pH 2) (13). However, comparing the different studies is complicated by the different types of gastric juice, different pH values, and various starting cell numbers employed.

FIG. 1.
Survival of L. pentosus DPC6004 Rifr ([filled square]), P. pentosaceus DPC6006 Rifr ([filled triangle]), L. murinus DPC6002 Rifr (•), L. murinus DPC6003 Rifr (X), and L. salivarius DPC6005 Rifr (♦) in porcine gastric juice (pH 3.0) over time. Values ...

Pig-feeding trial.

Having demonstrated that the porcine intestinal isolates survived, albeit to various degrees, in porcine gastric juice in vitro, a pig-feeding trial was performed to monitor the fate of these strains in vivo in the porcine gut. During the 21-day administration period (days 1 to 22, where day 1 represents the first day of culture administration), each group of pigs ingested ~1010 CFU per day of a different Lactobacillus or Pediococcus strain or the culture mix containing all five strains, also at a mean daily intake of ~1010 CFU per day. Selective enumeration of rifampin-resistant variants, in combination with DNA fingerprinting of selected colonies by RAPD PCR, was used to track the fate of the administered strains. A control group received skim milk acidified with lactic acid to pH 4.7, i.e., approximately the same pH as the culture fermentates. Although rifampin-resistant colonies were recovered from the fecal and cecal samples of some control animals during the trial at low counts, RAPD PCR confirmed that these were not any of the administered cultures (data not shown).

Survival and persistence of administered strains.

Prior to culture administration (day −5), none of the test strains was detected in the feces of any of the animals, although low background counts of rifampin-resistant bacteria were detected in some animals. By day 3, all culture-fed animals excreted 106 to 107 CFU of rifampin-resistant cultures/g of feces and continued to do so at between 105 and 108 CFU/g during the remainder of the 21-day administration period (Fig. (Fig.2A).2A). For the treatment groups fed individual cultures, RAPD PCR analysis confirmed all rifampin-resistant colonies selected from the fecal samples at day 15 as the administered culture (Fig. (Fig.2B).2B). However, even with four animals per group, considerable differences could be seen in excretion rates of the administered cultures during the 21-day administration period. Of all of the individually administered strains, the L. murinus strains showed superior intestinal transit, with mean fecal excretion rates of 1.3 × 108 and 4.7 × 107 CFU/g observed for strains DPC6003 and DPC6002, respectively (Fig. (Fig.2A).2A). Both were excreted at significantly higher numbers (P < 0.05) than P. pentosaceus DPC6006 (which had the lowest mean fecal count, 5.9 × 105 CFU/g). Furthermore, throughout the administration period, the L. murinus strains represented on average 23.7 and 20.8% of the total fecal Lactobacillus population, respectively, which is a higher percentage of total fecal lactobacilli than the other Lactobacillus strains administered (Fig. (Fig.3).3). L. murinus DPC6003 was also excreted at a significantly higher (P < 0.05) level than L. salivarius DPC6005 and L. pentosus DPC6004, both of which were intermediate in their survival ability, being detected at 5.3 × 106 to 5.6 × 106 CFU/g of feces and representing on average only 1.1 to 1.3% of total fecal lactobacilli (Fig. (Fig.2A2A and and3).3). Interestingly, these results correlate to some extent with in vitro findings for survival in gastric juice, where L. pentosus DPC6004 demonstrated superior survival, L. murinus DPC6002 and DPC6003 and P. pentosaceus DPC6006 were intermediate in their survival abilities, and L. salivarius DPC6005 was the least tolerant (Fig. (Fig.1).1). Mean total fecal excretion of the administered cultures in the animals fed the 5-strain combination was 8.2 × 107 CFU/g, which on average constituted 13% of the total fecal Lactobacillus population and was significantly higher (P < 0.05) than mean excretion of P. pentosaceus DPC6006 (Fig. (Fig.2A;2A; Fig. Fig.3).3). Interestingly, for those animals fed the culture mix, the predominant culture excreted at day 15 by all animals was L. murinus, albeit that in some animals L. murinus DPC6003 predominated and in others both strains DPC6002 and DPC6003 were detected (Fig. (Fig.2C2C).

FIG. 2.
(A) Excretion of administered strains in pigs fed ~1010 CFU of each of the porcine cultures/day. L. salivarius DPC6005 Rifr (♦), P. pentosaceus DPC6006 Rifr ([filled square]), L. pentosus DPC6004 Rifr ([filled triangle]), L. murinus DPC6002 Rifr (X), ...
FIG. 3.
Administered strains as a percentage of the total fecal Lactobacillus population in pigs fed ~1010 CFU of the Lactobacillus strains/day. L. salivarius DPC6005 Rifr (♦), L. pentosus DPC6004 Rifr ([filled square]), L. murinus DPC6002 Rifr ([filled triangle]), ...

The administered cultures could still be detected at day 5 postadministration in fecal samples of the majority of animals (Fig. (Fig.2A;2A; Table Table1),1), although none were detected in the two animals per group tested at day 66 postadministration (data not shown). However, the presence of the culture in the feces at day 5 postadministration and the level detected varied depending on the culture fed, and this correlated with the performance of the strain during the 21-day administration period (Table (Table1).1). For example, animals fed L. murinus DPC6002 and DPC6003 still excreted a high mean level (2.4 × 105 to 2.8 × 106 CFU/g) of the administered strains (Fig. (Fig.2A),2A), except for animal 423, which was fed strain DPC6003 (Table (Table1).1). Animals fed the culture mix also harbored high levels of rifampin-resistant bacteria 5 days after culture administration had stopped, with a mean fecal count of 2 × 107 CFU/g obtained for this group (Fig. (Fig.2A).2A). Representative colonies isolated at this time point were identified by RAPD as either L. murinus DPC6002 or DPC6003 or L. pentosus DPC6004, depending on the animal (Fig. 4A and B). This suggests that there was a change in the dynamics of intestinal microbial populations when the strains were no longer administered, as in comparison, L. murinus DPC6002 and DPC6003 were identified as the only predominant surviving strains at day 15. L. salivarius DPC6005, L. pentosus DPC6004, and P. pentosaceus DPC6006 were detected at 103 to 105 CFU/g of feces at day 5 postadministration in only 2, 1, and 3 animals per group, respectively (Table (Table1),1), indicating that they had declined to levels that were too low to be detected in some animals. Interestingly, poor fecal persistence postadministration correlated with lower excretion rates during the administration period. Also, even in animals that harbored high levels of the administered strains at 5 days postadministration, the cultures constituted a lower portion of the total Lactobacillus population at this time than during culture administration. Mean values were 0.5, 1.8, 0.02, 0.002, and 10.5% for cultures DPC6002, DPC6003, DPC6005, DPC6004, and the culture mix, respectively, with large variations between individual animals. This observation that fecal excretion rates decline once culture administration has stopped is common and has been noted previously in pigs (30, 31). Nonetheless, the majority of the strains administered in the present study persisted for at least 5 days postadministration in pig feces, which compares well with previous reports of 3- to 10-day persistence for other Lactobacillus strains (30, 34).

FIG. 4.
(A) Predominant strains detected in the feces at day 5 postadministration and in the cecal contents at day 9 postadministration (as determined by RAPD PCR fingerprinting of a representative number of isolates) in pigs fed ~1010 CFU of a combination ...
TABLE 1.
Number of postadministration pig fecal and cecal isolates identified by RAPD PCR as the administered strain(s) presented as a portion of the total isolates examined

We also investigated the ability of the strains to establish themselves in the cecum, as this is one of the principal intestinal sites of Salmonella colonization in pigs (42) and was the original source of the strains (3). Cecal persistence of the administered cultures was investigated in 2 animals per treatment following slaughter 9 days after culture administration had stopped. Some animals still harbored the cultures at high levels in the cecum, but as at day 5 postadministration, this was very much dependent on the particular strain(s) used (Table (Table1;1; Fig. Fig.5).5). Again the L. murinus strains performed well, persisting at the highest level of all the individually administered strains (103 to 106 CFU/g, 0.01 to 1.3% total lactobacilli) (Fig. (Fig.5),5), which was confirmed by RAPD analysis of selected colonies (Table (Table1).1). In the mixture-fed animals, the total counts for the administered strains were also high (104 and 106 CFU/g) (Fig. (Fig.5)5) and RAPD PCR identified the predominant persisting strains as L. murinus DPC6002 and DPC6003 in one of the animals sacrificed and L. pentosus DPC6004 and P. pentosaceus DPC6006 in the other animal (Fig. (Fig.4C).4C). It is noteworthy that the strains detected in the cecum did not always correspond to those predominating in the feces at day 5 postadministration (Table (Table1;1; Fig. Fig.4).4). When administered individually, P. pentosaceus DPC6006, L. pentosus DPC6004, and L. salivarius DPC6005 persisted poorly in the cecum, which correlates with findings for fecal persistence at day 5 postadministration (Table (Table1).1). These strains were detected as part of the predominant microflora in only one of the two animals slaughtered per treatment and then only as a small portion of the predominant colonies analyzed by RAPD PCR (Fig. (Fig.5;5; Table Table11).

FIG. 5.
Counts on MRS-RIF at day 9 postadministration in the cecal contents of individual pigs administered porcine cultures, as indicated. For animals marked with an asterisk, only a portion of the colonies examined were the administered strain. Animals not ...

While further analyses are required to investigate the exact duration of strain retention in the intestine following cessation of administration, L. murinus DPC6002 and DPC6003 and P. pentosaceus DPC6006, when administered individually, were capable of persisting in the gut, at least for a limited period. In addition, all of the strains in the mixture except L. salivarius DPC6005 persisted as members of the predominant gram-positive microflora in either the feces or the cecum of one or other of the animals. This demonstrates that these strains are able to effectively compete with the native microflora and become established in the intestine to some extent, although the mechanism involved is not fully understood. Indeed, the relative abilities of the five strains to adhere to epithelial (HT-29) cells were found to be similar (data not shown), even though gastrointestinal survival exhibited significant variation, highlighting the complexity of probiotic administration and its impact on the intestinal microflora. Detection of administered strains in the feces is often questioned as a method of determining intestinal survival. For example, Murphy et al. (23) found that strains administered to mice that were unrecoverable from the feces had nonetheless persisted in gastrointestinal tissue. However, we have shown that, in general, fecal survival and persistence correlates well with recovery from the cecum in terms of individually administered strains. The discrepancy between the strains detected in the feces of mixture-fed animals and those predominating in the cecum is worth noting. However, direct comparisons are difficult due to the different sampling times (i.e., day 5 versus day 9 postadministration).

In general, survival during intestinal transit is an important criterion for efficacy of orally administered probiotic strains (8). The strains isolated in our laboratory, especially L. murinus, which accounted for a high percentage of total fecal lactobacilli during culture administration (~21 to 24%) performed well in the porcine intestine. In comparison, in previous studies, strains of L. acidophilus or a 4-strain Lactobacillus mix fed at 1010 CFU per day accounted for no higher than 2.5% of the total Lactobacillus population (30) and a L. shirota strain fed at the same rate was recovered at only 104 CFU/g of feces (28). However, it is not always possible to compare intestinal survival of administered culture(s) in feeding trials, as investigators often do not enumerate the strains fed (4, 38, 39). In the present study, it was perhaps surprising that L. murinus, a species usually associated with mice and rats (18), performed so well in the porcine gut. However, L. murinus was recovered relatively frequently from pig cecal samples in an initial screening in our laboratory (3), indicating that it is a common inhabitant of the pig gut and so may possess properties that enable it to compete successfully in that environment. We also observed strain variation in terms of survival and persistence in the porcine gut, as demonstrated by others (20, 30, 31). This indicates that there may be advantages to feeding a culture mix, where if one strain is unsuccessful, others may compensate. Indeed, we found that individual strains within a mix performed differently in individual animals, and in general, the culture combination resulted in high fecal counts and good strain persistence. Interestingly, this is supported by the work of Pedersen et al. (30), who found that when a culture mix of four Lactobacillus strains was fed, fecal counts were maintained at 104 to 105 CFU/g for 5 to 7 days postadministration, whereas an L. acidophilus strain alone persisted for only 3 days.

Effects on intestinal microflora.

Evaluating the effects of probiotic administration on intestinal microflora is limited by the unculturable nature of the majority of intestinal species and the bias introduced due to the unsuitability of conventional culture methods (37). Nonetheless, we measured total fecal Lactobacillus and Enterobacteriaceae in an attempt to assess any major effects of culture administration on these representative intestinal species.

Examination of Enterobacteriaceae in both the feces and cecum was performed, since this group of gram-negative microorganisms represents pathogenic indicator species, such as E. coli and Salmonella. The five porcine isolates administered have previously been shown to display in vitro anti-Salmonella activity (3) and these strains also inhibit representative fecal Enterobacteriaceae of pig origin in agar plate assays (data not shown). In the present study, there was evidence that some of the cultures may have influenced fecal Enterobacteriaceae counts in the pigs (Table (Table2;2; Fig. Fig.6).6). On average, when all treatment groups were analyzed together, fecal Enterobacteriaceae counts were significantly lower (P < 0.05) towards the end of the culture administration period (i.e., on days 15, 22, and 26) than prior to or during the first week of culture administration (i.e., days −5, 3, and 8). However, fecal Enterobacteriaceae counts were highly variable throughout the trial, with large variance of values within individual treatment groups and fluctuations observed in counts at different time points (Fig. (Fig.6).6). As a result, no significant effects were observed when examining the impact of culture administration on fecal Enterobacteriaceae during the entire culture administration period (day 3 to 22). However, at day 15, mean fecal Enterobacteriaceae counts were significantly lower (P < 0.05) in animals fed L. murinus DPC6003 (1.6 × 105 CFU/g) than in either the L. murinus DPC6002 (2 × 107 CFU/g) or the culture mix (8.7 × 106 CFU/g) groups. Apart from these statistically significant effects, some nonsignificant trends were observed when mean preadministration (day −5) counts were compared with those obtained after 21 days of culture administration (day 22). For example, in pigs fed strains DPC6005, DPC6006, DPC6004, and the culture mix, Enterobacteriaceae counts decreased ~10- to 50-fold from 7 × 106 to 1.3 × 108 CFU/g to 5 × 105 to 4 × 106 CFU/g (Table (Table2).2). This represents mean reductions of 98, 87, 97, and 97%, respectively, when the mean preadministration counts within each treatment group are taken as 100% (Table (Table2).2). However, mean fecal counts also decreased by 83% in the control group (Table (Table2).2). Ratcliffe et al. (33) observed similar Enterobacteriaceae-reducing effects in pigs fed acidified milk and concluded that this was due, at least in part, to its low pH and/or lactic acid content. In our study, in all cases, the Enterobacteriaceae reduction persisted through to day 5 postadministration, when numbers declined even further (Table (Table2).2). Mean counts in the L. murinus DPC6003 group also decreased to ~105 CFU/g at this time point (Table (Table2),2), perhaps reflecting the fact that Enterobacteriaceae counts in this treatment group were significantly lower (P < 0.05) than in some other groups at day 15, as outlined above. In addition, mean Enterobacteriaceae in the cecal contents of the two animals per treatment sacrificed did not differ substantially between treatment groups (Fig. (Fig.7).7). However, animals fed L. murinus DPC6003 and L. salivarius DPC6005 had ~10-fold-lower counts than those fed L. pentosus DPC6004, L. murinus DPC6002, or the culture mix (~2 × 106 versus ~1 × 107 CFU/g), although there was no correlation between lower Enterobacteriaceae counts and low cecal pH (Fig. (Fig.77).

FIG. 6.
Fecal excretion of Enterobacteriaceae in pigs fed either control acidified skim milk (♦) or ~1010 CFU of skim milk fermentates of either L. salivarius DPC6005 Rifr ([filled square]), P. pentosaceus DPC6006 Rifr ([filled triangle]), L. pentosus DPC6004 ...
FIG. 7.
Enterobacteriaceae (white bars) and total Lactobacillus (black bars) counts at day 9 postadministration in the cecal contents of pigs administered porcine cultures either individually or as a combination, as indicated. Values are means of the results ...
TABLE 2.
Effect of administration of porcine cultures on mean fecal Enterobacteriaceae counts in pigs

Interestingly, although L. murinus DPC6002 and DPC6003 were excreted at the highest level in the feces, they did not result in Enterobacteriaceae reductions (although L. murinus DPC6003 may have had some inconsistent effects). This correlates with in vitro findings for Salmonella inhibition, with L. salivarius DPC6005, P. pentosaceus DPC6006, and L. pentosus DPC6004 showing greater inhibitory activity than L. murinus DPC6002 and DPC6003 (3). However, due to the variations observed in counts between individual animals and at different time points and the fact that a reduction was also observed in the control group, it is difficult to draw clear conclusions on the effects of culture administration on intestinal Enterobacteriaceae.

Administration of lactic acid bacteria has been shown to reduce intestinal coliform and Enterobacteriaceae counts in the majority of previous studies (4, 22, 24, 28, 38, 39), although some have seen no effects (7, 13). The trend towards positive effects observed in ours and other studies, although not significant, should not be ignored, as individual variations in the responses of different animals are to be expected due to the complexity of the intestine (36). Furthermore, it has been suggested that hygienic conditions in scientific institutes may be too favorable to investigate effects on pathogenic bacteria without deliberate challenge (36). Future experiments with larger treatment groups and deliberate infection with Salmonella should provide further information on the possible pathogen-lowering ability of the potentially probiotic cultures investigated in the present study.

While others have found increases in total lactobacilli as a result of administration of Lactobacillus strains (20, 28, 38), no significant differences in total fecal Lactobacillus numbers were observed between any of the treatment groups in the present study (Fig. (Fig.8).8). However, nonsignificant ~10-fold increases (from ~108 to ~109 CFU/g) were observed in pigs fed P. pentosaceus DPC6006, L. pentosus DPC6004, and L. murinus DPC6002 and DPC6003 when preadministration (day −5) fecal counts were compared with counts after 21 days of culture administration (day 22) (Fig. (Fig.8).8). A similar nonsignificant increase was also seen in the control group, which contradicts the findings of Ratcliffe et al. (33), who observed a decrease in lactobacilli in pigs fed acidified milk. Interestingly, although L. salivarius DPC6005 produces a bacteriocin with anti-Lactobacillus activity in vitro (data not shown), this strain did not reduce total fecal Lactobacillus populations. However, as we have limited knowledge on its spectrum of activity, it is difficult to say whether this bacteriocin inhibits any of the Lactobacillus species present in the porcine gut. Mean Lactobacillus counts in the cecum show little correlation with fecal counts and were 10-fold higher in pigs fed L. murinus DPC6003 and L. salivarius DPC6005 (2.8 × 108 to 3.9 × 108 CFU/g) than in pigs fed control milk or L. pentosus DPC6004 (3.2 × 107 to 4 × 107 CFU/g) (Fig. (Fig.77).

FIG. 8.
Total fecal Lactobacillus counts in pigs fed either control acidified skim milk (♦) or ~1010 CFU of either L. salivarius DPC6005 Rifr ([filled square]), L. pentosus DPC6004 Rifr (X), P. pentosaceus DPC6006 Rifr ([filled triangle]), L. murinus DPC6002 ...

We have demonstrated that porcine intestinal isolates, selected on the basis of in vitro probiotic criteria, can be effectively delivered to the porcine GIT by oral administration either individually or as a strain combination. All five strains investigated survived gastrointestinal transit in pigs, and some may have the ability to reduce undesirable microorganisms in the gut. However, it was evident that certain cultures survived at higher numbers in the porcine gut, persisted for longer in the cecum after culture administration had ceased, and were more efficacious in reducing pathogenic indicator species. This strain variation highlights the advantages of using combination probiotics. Based on the results of this study, the most promising strains or strain combinations will be selected for administration to larger numbers of animals in future trials. We conclude that, although further characterization of their in vivo efficacy is necessary, the present findings provide a strong basis to explore the potential of these porcine-derived isolates as probiotic feed additives for pigs.

Acknowledgments

The technical assistance of Tom Cullinan, Jim O'Reilly, and the staff of the Moorepark pig unit is gratefully acknowledged. We thank Paula Reid (Teagasc Headquarters) for statistical analyses.

This work was supported by the Irish Government under the National Development Plan 2000-2006.

REFERENCES

1. Barton, M. D. 2000. Antibiotic use in animal feed and its impact on human health. Nutr. Res. Rev. 13:279-299. [PubMed]
2. Canganella, F., M. Gasbarri, S. Massa, and L. D. Trovatelli. 1996. A microbiological investigation on probiotic preparations used for animal feeding. Microbiol. Res. 151:167-175.
3. Casey, P. G., G. D. Casey, G. E. Gardiner, M. Tangney, C. Stanton, R. P. Ross, C. Hill, and G. F. Fitzgerald. 2003. Anti-Salmonella lactic acid bacteria from porcine intestinal sources, p. 280-282. In L. Leonides (ed.), Proceedings of the 5th International Symposium on the Epidemiology and Control of Foodborne Pathogens in Pork, Crete, Greece.
4. Chang, Y. H., J. K. Kim, H. J. Kim, W. Y. Kim, Y. B. Kim, and Y. H. Park. 2001. Selection of a potential probiotic Lactobacillus strain and subsequent in vivo studies. Antonie Leeuwenhoek 80:193-199. [PubMed]
5. Coakley, M., R. P. Ross, and D. Donnelly. 1996. Application of the polymerase chain reaction to the rapid analysis of brewery yeast strains. J. Inst. Brew. 102:349-354.
6. Davies, R., G. Paiba, S. Evans, and B. Dalziel. 2000. Surveys for Salmonella in pigs, cattle and sheep at slaughter in Great Britain. Vet. Rec. 147:695. [PubMed]
7. De Cupere, F., P. Deprez, D. Demeulenaere, and E. Muylle. 1992. Evaluation of the effect of 3 probiotics on experimental Escherichia coli enterotoxaemia in weaned piglets. J. Vet. Med. Bull. 39:277-284. [PubMed]
8. Dunne, C., L. O'Mahony, L. Murphy, G. Thornton, D. Morrissey, S. O'Halloran, M. Feeney, S. Flynn, G. Fitzgerald, C. Daly, B. Kiely, G. C. O'Sullivan, F. Shanahan, and J. K. Collins. 2001. In vitro selection criteria for probiotic bacteria of human origin: correlation with in vivo findings. Am. J. Clin. Nutr. 73:386S-392S. [PubMed]
9. FAO/W. H. O. 2001. Report on joint FAO/W. H. O. expert consultation on evaluation of health and nutritional properties of probiotics in food including powder milk with live lactic acid bacteria. [Online.] http://www.who.int/fsf/probiotics.htm.
10. Fedorka-Cray, P. J., J. S. Bailey, N. J. Stern, N. A. Cox, S. R. Ladely, and M. Musgrove. 1999. Mucosal competitive exclusion to reduce Salmonella in swine. J. Food Prot. 62:1376-1380. [PubMed]
11. Gardiner, G., R. P. Ross, J. K. Collins, G. Fitzgerald, and C. Stanton. 1998. Development of a probiotic cheddar cheese containing human-derived Lactobacillus paracasei strains. Appl. Environ. Microbiol. 64:2192-2199. [PMC free article] [PubMed]
12. Gardiner, G., R. P. Ross, P. M. Kelly, J. K. Collins, G. Fitzgerald, and C. Stanton. 2002. Microbiology of therapeutic milks, p. 431-478. In R. K. Robinson (ed.), Dairy microbiology, 3rd ed., vol. 2. John Wiley & Sons, Inc., New York, N.Y.
13. Gardiner, G., C. Stanton, P. B. Lynch, J. K. Collins, G. Fitzgerald, and R. P. Ross. 1999. Evaluation of cheddar cheese as a food carrier for delivery of a probiotic strain to the gastrointestinal tract. J. Dairy Sci. 82:1379-1387. [PubMed]
14. Gardiner, G. E., C. Heinemann, M. L. Baroja, A. W. Bruce, D. Beuerman, J. Madrenas, and G. Reid. 2002b. Oral administration of the probiotic combination Lactobacillus rhamnosus GR-1 and L. fermentum RC-14 for human intestinal applications. Int. Dairy J. 12:191-196.
15. Genovese, K. J., R. C. Anderson, R. B. Harvey, and D. J. Nisbet. 2000. Competitive exclusion treatment reduces the mortality and fecal shedding associated with enterotoxigenic Escherichia coli infection in nursery-raised neonatal pigs. Can. J. Vet. Res. 64:204-207. [PMC free article] [PubMed]
16. Genstat. 1994. Genstat 5, release 3 reference manual. Clarendon Press, Oxford, United Kingdom.
17. Gusils, C., M. Bujazha, and S. Gonzalez. 2002. Preliminary studies to design a probiotic for use in swine feed. Interciencia 27:409-413.
18. Hemme, D., P. Raibaud, R. Ducluzeau, J. V. Galpin, P. Sicard, and J. Van Heijenoort. 1980. Lactobacillus murinus n. sp., a new species of the autochtoneous dominant flora of the digestive tract of rat and mouse. Ann. Microbiol. 131:297-308. [PubMed]
19. Jin, L. Z., Y. W. Ho, N. Abdullah, and S. Jalaludin. 2000. Digestive and bacterial enzyme activities in broilers fed diets supplemented with Lactobacillus cultures. Poult. Sci. 79:886-891. [PubMed]
20. Jonsson, E. 1986. Persistence of a Lactobacillus strain in the gut of sucking piglets and its influence on performance and health. Swed. J. Agric. Res. 16:43-47.
21. Lema, M., L. Williams, and D. R. Rao. 2001. Reduction of fecal shedding of enterohemorrhagic Escherichia coli O157:H7 in lambs by feeding microbial feed supplement. Small Rumin. Res. 39:31-39. [PubMed]
22. Muralidhara, K. S., G. G. Sheggeby, P. R. Elliker, D. C. England, and W. E. Sandine. 1977. Effect of feeding lactobacilli on the coliform and Lactobacillus flora of intestinal tissue and feces from piglets. J. Food Prot. 40:288-295.
23. Murphy, L., C. Dunne, B. Kiely, F. Shanahan, G. C. O'Sullivan, and J. K. Collins. 1999. In vivo assessment of potential probiotic Lactobacillus salivarius strains; evaluation of their establishment, persistence, and localisation in the murine gastrointestinal tract. Microb. Ecol. Health Dis. 11:149-157.
24. Nemcova, R., A. Bomba, S. Gancarcikova, R. Herich, and P. Guba. 1999. Study of the effect of Lactobacillus paracasei and fructooligosaccharides on the faecal microflora in weanling piglets. Berl. Munch. Tierarztl. Wochenschr. 112:225-228. [PubMed]
25. Nemcova, R., A. Laukova, S. Gancarcikova, and R. Kastel. 1997. In vitro studies of porcine lactobacilli for possible probiotic use. Berl. Munch. Tierarztl. Wochenschr. 110:413-417. [PubMed]
26. Neumann, E., and C. L. L. Fortes Ferreira. 1995. Lactobacillus acidophilus as dietary adjunct: "in vitro' susceptibility to gastric juice, bile salts, lysozyme and chemotherapeutic agents. Rev. Microbiol. 26:59-65.
27. Nisbet, D. 2002. Defined competitive exclusion cultures in the prevention of enteropathogen colonisation in poultry and swine. Antonie Leeuwenhoek 81:481-486. [PubMed]
28. Ohashi, Y., R. Inoue, K. Tanaka, T. Matsuki, Y. Umesaki, and K. Ushida. 2001. Lactobacillus casei strain Shirota-fermented milk stimulates indigenous lactobacilli in the pig intestine. J. Nutr. Sci. Vitaminol. 47:172-176. [PubMed]
29. Ohya, T., T. Marubashi, and H. Ito. 2000. Significance of fecal volatile fatty acids in shedding of Escherichia coli O157 from calves: experimental infection and preliminary use of a probiotic product. J. Vet. Med. Sci. 62:1151-1155. [PubMed]
30. Pedersen, K., G. W. Christensen, M. Steffensen, P. Schyum, and A. K. Johansen. 1992. Transfer of lactic acid bacterial strains from the feed to the sow, the environment, and the piglets. Acta Vet. Scand. 33:297-303. [PubMed]
31. Pedersen, K., and G. W. Tannock. 1989. Colonization of the porcine gastrointestinal tract by lactobacilli. Appl. Environ. Microbiol. 55:279-283. [PMC free article] [PubMed]
32. Pettersson, L., W. Graf, L. Alm, S. Lindwall, and A. Stromberg. 1983. Survival of Lactobacillus acidophilus NCDO 1748 in the human gastrointestinal tract. 1. Incubation with gastric juice in vitro. Symp. Swed. Nutr. Found. XV:123-125.
33. Ratcliffe, B., C. B. Cole, R. Fuller, and M. J. Newport. 1986. The effect of yoghurt and milk fermented with a porcine intestinal strain of Lactobacillus reuteri on the performance and gastrointestinal flora of pigs weaned at two days of age. Food Microbiol. 3:203-211.
34. Rogelj, I., B. Bogovic Matijasic, A. Canzek Majhenic, and S. Stojkovic. 2002. The survival and persistence of Lactobacillus acidophilus LF221 in different ecosystems. Int. J. Food Microbiol. 76:83-91. [PubMed]
35. Rogosa, M., J. A. Mitchell, and R. T. Wiseman. 1951. A selective medium for the isolation and enumeration of oral and fecal lactobacilli. J. Bacteriol. 62:132-133. [PMC free article] [PubMed]
36. Simon, O., W. Vahjen, and L. Scharek. 2003. Microorganisms as feed additives—probiotics, p. 295-318. In R. O. Ball (ed.), Proceedings of the 9th International Symposium on Digestive Physiology in Pigs, vol. 1. University of Alberta, Edmonton, Alberta, Canada.
37. Tannock, G. W. 2001. Molecular assessment of intestinal microflora. Am. J. Clin. Nutr. 73:410S-414S. [PubMed]
38. Terada, A., H. Hara, S. T. Li, S. Yagi, H. Ichikawa, J. Nishi, S. Ko, and T. Mitsuoka. 1994. Effect of a microbial preparation on faecal flora and faecal metabolic products of pigs. Anim. Sci. Technol. (Jpn.). 65:806-814.
39. Tortuero, F., J. Rioperez, E. Fernandez, and M. L. Rodriguez. 1995. Response of piglets to oral administration of lactic acid bacteria. J. Food Prot. 58:1369-1374.
40. Vanbelle, M. 2001. Current status and future perspectives in EU for antibiotics, probiotics, enzymes, and organic acids in animal nutrition, p. 231-256. In A. Piva, K. E. Bach Knudsen, and J. E. Lindberg (ed.), Gut environment of pigs. Nottingham University Press, Nottingham, United Kingdom.
41. Weese, J. S. 2002. Microbiologic evaluation of commercial probiotics. J. Am. Vet. Med. Assoc. 220:794-797. [PubMed]
42. Wood, R. L., and R. Rose. 1992. Populations of Salmonella typhimurium in internal organs of experimentally infected carrier swine. Am. J. Vet. Res. 53:653-658. [PubMed]
43. Yu, B., and H. Y. Tsen. 1993. Lactobacillus cells in the rabbit digestive tract and the factors affecting their distribution. J. Appl. Bacteriol. 75:269-275. [PubMed]
44. Zhao, T., M. P. Doyle, B. G. Harmon, C. A. Brown, P. O. Mueller, and A. H. Parks. 1998. Reduction of carriage of enterohemorrhagic Escherichia coli O157:H7 in cattle by inoculation with probiotic bacteria. J. Clin. Microbiol. 36:641-647. [PMC free article] [PubMed]

Articles from Applied and Environmental Microbiology are provided here courtesy of American Society for Microbiology (ASM)
PubReader format: click here to try

Formats:

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...

Links

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...