Optimized culturing and nucleic acid-based methods for the detection of Salmonella enterica in poultry environments

Optimized culturing and nucleic acid-based methods for the detection of Salmonella entérica in poultry environments J, Schultz,* R, Jarquin,*t S, C, Ricke,i and I, Hanningf^ *Research and Development, Cobb-Vantress Incorporated, PO Box 1030, Siloam Springs, AR; fDepartment of Food Science and Technology, University of Tennessee, 2605 River Dr., Knoxville 37996; and f Center for Food Safety and Department of Food Science, University of Arkansas, 2650 North Young Ave., Fayetteville 72704 ples then were reincubated under the same conditions. After secondary enrichment, the samples were streaked onto Chromogenic agar/ XLT4 bi-plates and incubated under the same conditions. After initial inoculation and each 24-h incubation, a portion of the enriched samples was analyzed using a real-time PCR assay. The results of this trial indicate that recovery of Salmonella in a culture-based assay may be enhanced by up to 3 logs by using the TT as the initial enrichment media compared with BPW, The incorporation of MSRV as a secondary cultural selective media after the TT gave the best recovery of Salmonella. These data indicate that considerable time can be saved by using TT as an initial media for Salmonella recovery.

ABSTRACT The main objective of this trial was to set up a series of assays following quantified inoculation of Salmonella samples in 2 initial enrichment methods to ultimately determine the most effective and fastest detection method for recovery of Salmonella in a poultry environment matrix. Samples were randomly split into 2 different containers containing either buffered peptone water (BPW) -f yeast extract, or tetrathionate broth (TT) with added iodine and Brilliant Green solution 0,1%, A frozen stock Salmonella culture was thawed and serially diluted 10-fold to inoculate 100 |i,L of the dilution into the enriched samples. The samples were incubated at 42 and 37°C, respectively, for 24 h and secondarily enriched in modified semi-solid Rappaport Vassiliadis (MSRV) incubated at 42°C, All sam-

Key words: Salmonella, real-time PCR, culturing, broiler, poultry 2012 Poultry Science 91:2761-2766 http://dx,doi,org/10.3382/ps,2012-02387

INTRODUCTION Salmonella entérica is a leading cause of foodborne illness in the United States and throughout the world (Wegener et al,, 2003; Zhao et al, 2008). The CDC estimates that there are about 1.4 million cases of salmonellosis annually in the United States, costing approximately 2.6 billion dollars in terms of loss of productivity, health care costs, and premature death (USDA Economic Research Service, 2007), Poultry and poultry products can be a leading source of foodborne salmonellosis if contaminated meat is handled improperly or consumed undercooked or raw (Braden, 2006; Jain and Chen, 2006; Linam and Cerber, 2007), Poultry may be colonized by 5. entérica without causing any harm or disease to the animal. Therefore, if gastrointestinal rupture occurs during processing, contamination of the raw product may occur (Byrd et al., 2002).

©2012 Poultry Science Association Inc. Received April 15, 2012. Accepted July 8, 2012. ^Corresponding author: [email protected]

Control of Salmonella on the farm is a critical point in the process of poultry production that can possibly reduce foodborne salmonellosis. One program aimed at reduction of Salmonella on US farms is the National Poultry Improvement Program (NPIP; APHIS-USDA, 2001). Broiler breeders certified and operating under this program are required to sell stock that is free of 5a/moneZZa Enteritidis (APHIS-USDA, 2001), For these reasons, a constant and accurate monitoring program for Salmonella by NPIP participants is required. Monitoring assays that give the earliest detection results are ideal to limit vertical transfer of Salmonella and facilitate early corrective actions, Culturing of Salmonella for detection is considered the gold standard. Several media for culturing exist with recovery results varying based on multiple factors (Koyuncu and Haggblom, 2009). However, culturing can be time consuming and take as long as 4 d for results and up to 2 wk using the NPIP delayed secondary enrichment method (APHIS-USDA, 2001), Polymerase chain reaction assays can be used as detection method with results in as few as 24 h, The main objective of these trials was to set up a series of assays following

2761

2762

SCHULTZ ET AL.

quantified inoculation of Salmonella into feed and drag swab samples in 2 initial enrichment methods to determine the most effective and rapid detection of a 5a/monella recovery method from poultry environmental sample matrices.

MATERIALS AND METHODS

Sample Coliection for Inoculation with Salmonella Three separate trials were conducted using soiled foot covers, and an additional trial was conducted using raw feed ingredients. On 3 separate occasions, 80 soiled foot covers were collected from 1 of 3 farms (80 from each farm). Farms were located in Arkansas, United Kingdom (UK), and Brazil. For the feed trial, 11 feed samples consisting of 8 soy meal and 3 corn germ meal samples were collected and pooled into a sterile plastic bag. The feed ingredients were mixed until a uniform distribution was accomplished. From this mix, 3 subsamples of 10 g were used in 3 separate PCR assays and confirmed as negative for Salmonella.

Salmonella Inoculation and Culturing Conditions For the foot cover trials, a frozen stock Salmonella Enteritidis strain was used that was previously recovered from the same farm sampled, with the exception of the UK farm because no Salmonella had previously been recovered from that farm. For the UK trial, a ministry quality control Salmonella Enteritidis strain was used. Feed trials were performed with 5 replications, each replication with a different strain of Salmonella. Of the 5 strains used, 4 Salmonella strains (Kentucky, Berta, Tennessee, and Typhimurium var. Copenhagen) were previously isolated from feed mills. The fifth strain used Salmonella Enteriditis isolated from a broiler farm. For all experiments, a uniform level of background poultry house microbial matrix or feed ingredient matrix was included in each sample to measure the true selectivity of the system. Additionally, all samples were initially analyzed for the presence of Salmonella as a control. Foot Cover Trials. For foot covers, the 80 samples were randomly split into 2 different containers containing either sterile 1-X solution of the BPW -I- yeast extract, or a 1-X solution of TT with added iodine solution and Brilhant Creen solution 0.1%. For both BPW and TT, these samples were further split into aliquots of 100 mL. A frozen stock Salmonella culture was thawed and immediately serially diluted 10-fold to inoculate 100 jxL of the dilution into the aliquots of enrichment samples. All BPW and TT samples were incubated at 42 and 37°C, respectively, for 24 h. Afi;er this incubation period, samples were inoculated into modified semi-solid Rappaport Vassiliadis (MSRV) and again incubated at 42°C for 24 h. Finally, the MSRV samples were streaked onto Chromogenic/XLT4 agar using a

sterile loop and all plates were incubated under the same conditions. For samples initially enriched in TT, the same procedure was used. After initial inoculation and after each 24-h incubation, all samples were used in a real-time PCR assay. One bag of each set of the foot covers was not inoculated with Salmonella to act as a negative control. Feed Trials. For the feed samples, before splitting the sample and enriching in the 2 media, a 1,000-g subsample of the feed mixture was incubated in 5 L of sterile distilled water at 42°C for 3 h with frequent agitation. The liquid then was filtered out through clean gauze foot swabs (nonmilked). The liquid was split evenly into the BPW or TT broth to make a complete 1-X solution of each enrichment media. Four 1-L volumes of the enrichment medias were then dispensed into 40 wire closure 500-mL bags (100 mL volume per bag), and identified for the spiking trial with the various dilutions of the different Salmonella strains. Feed samples then were inoculated in the same manner as foot cover trials. Feed samples were incubated, passed, and sampled in the same manner as described for the foot covers. One bag of each set of the feed was not inoculated with Salmonella to act as negative controls. For samples enriched in BPW, a 1-mL portion of the sample was inoculated into 9 mL of TT and 100 [iL of the same sample was also enriched in MSRV. All samples then were reincubated under the same conditions. Finally, the samples were streaked onto Chromogenic/XLT4 bi-plates using a sterile loop and all plates were incubated under the same conditions. For samples initially enriched in TT, the same procedure was used without the initial BPW preenrichment step. After initial inoculation and after each 24-h incubation, a portion of all the enriched samples was used in a real-time PCR assay.

DNA Extraction and Reai-Time PCR Assays At each enrichment step after incubation, a portion of the enriched sample was taken and used in a realtime PCR assay and a second sample was streaked onto Chromogenic/XLT4 agar bi-plates (Remel, Lenexa, KS). Any suspect colonies were collected from the agar and also evaluated by PCR. A 1-mL portion of the TT enriched sample was placed into a microcentrifuge tube and single colonies were placed in a microcentrifuge tube containing 1 mL of diethyl pyrocarbonate (DEPC) water (Ambion, Foster City, CA). For MSRV samples, a 1 [xL loop was suspended in 1 mL of DEPC water (Ambion). For all samples, the tubes were placed in a heating block and boiled at 101°C to lyse the cells and release the DNA. All samples were subsequently refrigerated at 4°C for 10 min and finally centrifuged at 16,000 X g centrifuged at room temperature (23°C) for 2 min. All samples then were immediately used for the real-time PCR assay. The SYBR green real-time PCR assay was optimized using an Eppendorf Masterplex thermocycler ep (Eppendorf, Westbury, NY). Cradient

OPTIMIZED DETECTION OF SALMONELLA

Technology in the Eppendorf unit was used to optimize annealing and extension temperatures and times. The assay used published primers which targeted all serovars of Salmonella entérica (Hoorfar et al., 2000) or Salmonella entérica serovar Enteriditis (Agron et al., 2001). The temperatures of the melting curves were 80.5 and 85°G, respectively. All primers were synthesized by Sigma Ghemical Gompany (St. Louis, MO). A 20-^i,L total volume reaction mixture consisted of 10 [xL mL~^ of SYBR Green Premix Ex Taq (Takara; Fisher Scientific, Pittsburgh, PA) 0.5 jxmol mL~^ of each primer, 2 |i,L of DNA template and water to volume. The PGR reactions for both primers were optimized to the conditions of 94°G for 1 min, then 35 cycles of 94°G for 10 s, 57°G for 10 s, and 72°G for 20 s. The threshold cycle number was determined to be the cycle number at which fluorescence was greater than 400 fluorescence units. Melting curves were created and analyzed with the Eppendorf realplex software (version 2.0).

RESULTS For foot cover trials using farm soils sampled in Brazil, the overall detection and recovery of Salmonella was more consistent between PGR and culturing when 5a/monella was initially inoculated into TT versus BPW

2763

(Table 1). Furthermore, a 3- to 4-log difference in Salmonella detection by PGR was achieved when comparing samples incubated for 1 d in TT versus samples incubated for 1 d in BPW (Table 1). Similarly, the detection sensitivity with PGR was increased by 4 logs using TT with further enrichment in MSRV (Table 1). Gultures were recovered after enrichment in TT and BPW with subsequent inoculation into MSRV (Table 1), but the initial TT enrichment increased the sensitivity by 1 log. Finally, culture recovery on chromogenic/ XLT4 bi-plates was increased by 4 logs when samples were inoculated into TT and enriched in MSRV versus initial inoculation into BPW. For trials performed using soiled foot covers sampled from a farm in Arkansas, at least a 2-log difference in PGR detection was determined when comparing samples incubated overnight in TT versus BPW (Table 2). The same increase in PGR sensitivity was achieved after subsequent enrichment in MSRV. A 1-log increase in recovery was determined after further enrichment in MSRV when utilizing TT as the initial inoculum as opposed to BPW. The final set of foot cover trials was performed using soiled foot covers sampled from a farm in the UK. A 2-log increase in sensitivity with PGR detection was determined for initial enrichment in TT, but only for

Table 1. The detection of Salmonella entérica serovar Enteritidis from soiled foot swab samples collected from a farm in Brazil initially enriched in buffered peptone water (BPW) or tetrathionate (TT) d 3

d2

Inoculated logio cfu mL~^

d4

PC1R3

PGRi Salmonella Enteritidis Salmonella

Salmonella MSRV^

Enteritidis Salmonella

Ghromogenic/ XLT4'' agar

0 1 2 3 4 5 6 7

8 9 0 1 2

3 4 5 6 7 8 9

— — -

+

-1-1-

-. — + -1-f -1-

_ _ -1-|-

_ — -1-

+ +

-f -1-h -1-1-

-1-

+

-t-

-1-)-

+ _ — -h -1-1-1-

.-

— -

-h -1-1-h -t-

+

_ — -f -1-1-1-1-1-

_ — -t-1-h -1-1-h -1-

+

-

-1-

+

-1-

-f

+ _ — +

PCR was conducted after 24 h of incubation in TT. ^Samples were further enriched in modified semi-solid Rappaport Vassiliadis (MSRV) after initicd enrichment inTT. ^The PGR was conducted after 24 h of secondary enrichment in MSRV. ''Samples were struck on Ghromogenic/XLT4 agar after secondary enrichment in MSRV. ^Used to suspend a pool of 80 soiled foot covers, which was split into 100-mL samples each receiving a serially diluted Salmonella culture and incubated for 24 h.

2764

SCHULTZ ET AL. Table 2. The detection of Salmonella entérica serovar Enteritidis from soiled foot swab samples collected from a farm in Arkansas initially enriched in buffered peptone water (BPW) or tetrathionate (TT) d 2

d3

PC!Ri Inoculated logio cfu mL~l BPW^ 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9

Salmonella Enteritidis

PC!R3

Salmonella

MSRV2

_ _

Salmonella Enteritidis

d4

Salmonella

Chromogenic/ XLT4'' agar

_ _

•

_

_ _ _ — -1-

+ _ _ _ — -

_ _

_

_

+ + + + _

-f

-1-

-1-

-I-

-h -1-

+

+

_ _ _ -

_ _ _

— +

_

-1-

-1-

-1-I-

-1-

-1-

-1-H

-t-

-1-

-1-

-1-

-1-

-1-

-f

-t-1-

-1-

-h 4-1-

-1-1-h -1-

-1-H -1-

-t-1-

-h

-1-h -1-

-f

_ -f -1-

-1-h -f -1-

PCR was conducted after 24 h of incubation in TT. ^Samples were further enriched in modified semi-solid Rappaport Vassiliadis (MSRV) after initial enrichment inTT. PCR was conducted after 24 h of secondary enrichment in MSRV. "Samples were struck on Chromogenic/XLT4 agar after secondary enrichment in MSRV. ^Used to suspend a pool of 80 soiled foot covers, which was split into 100-mL samples each receiving a serially diluted Salmonella culture and incubated for 24 h.

S'a/mone/ia-specific primers (Table 3). There was no difference in PGR detection using the Salmonella Enteritidis-specific primers between samples inoculated into TT versus BPW. Interestingly, subsequent enrichment in MSRV after initial enrichment in BPW enhanced increased sensitivity by 1 log (Table 3) over TT. A 1-log increase in MSRV sensitivity was achieved utilizing TT as the initial inoculum media versus BPW. In the fourth trial, raw feed ingredients (soy meal and corn germ) were used as the background matrices and inoculated with 1 of 5 Salmonella entérica serovars (Kentucky, Typhimurium var. Gopenhagen, Tennessee, Berta, and Enteritidis; Table 4). After initial inoculation and 24 incubation in TT or BPW, the sensitivity of the PGR assay was increased by 1 to 4 logs depending on the serovar for inoculation in TT versus BPW. None of the MSRV enrichments were positive for any serovar after initial inoculation in BPW. Inoculation into TT enhanced the detection sensitivity for some serovars by 1 log.

DISCUSSION Early detection of Salmonella contamination in poultry flocks can be helpful for preventing colonization of other birds and possible cross-contamination among

different flocks housed in the same facility. In this work, it was determined that PGR detection is possible after 24 h of preenrichment within a solution of TT or BPW. However, in many cases it is necessary to obtain cultures for further analysis such as serotyping or genotyping. This information can be useful in determining persistence of specific strains and to develop intervention or corrective measures. This study demonstrated that detection limits by PGR and culturing were different when samples were initially preenriched in BPW as opposed to TT when using the incubation temperatures of this study. This may be due to the enrichment of any bacteria present in the sample and Salmonella possibly being outcompeted for nutrients by other bacteria. As a result, the final titer of Salmonella may have been lower in the BPW than the TT samples. The BPW is used as a preenrichment nonselective medium because it allows for repair of cell damage and purportedly facilitates the recovery of Salmonella (Edel and Kampelmacher, 1973). Because BPW is buffered, the medium maintains a neutral pH, which is important because stressed cells can be very sensitive to acidic conditions. The TT is a selective enrichment medium and suppresses nontarget organisms while allowing Salmonella to be cultured. Some inhibition of stressed cells has been reported when culturing

OPTIMIZED DETECTION OF SALMONELLA

2765

Table 3. The detection of Salmonella entérica serovar Enteritidis from soiled foot swab samples collected from a farm in the United Kingdom initially enriched in buffered peptone water (BPW) or tetrathionate (TT) d2

d3

PCR 1

Inoculated logio cfu-mL~'

d4

PCI

Salmonella Enteritidis Salmonella

MSRV2

Salmonella Enteritidis Salmonella

Chromogenic/ XLT4 agar''

BPW^ 0 1 2 3 4 5 6 7 8 9

— 4-

— 4-

4-

4-

4-

4-

4-

0 1 2 3 4 5 6 7 8 9

_ _ —

_ — -

44-

44-

. 4-

-

4-

— 4-

-h 4-

44-

4-

444-

4-

4-

4-

4-

44-

4-

44-

4-

_ — 4-

_ — — -

— —

4-

_ _ — — 44-

4-

4-

4-

4-

4-

4-

4-

44-

44-

44-

4-

44-

4-

44-

444-

44-

PCR was conducted after 24 h of incubation in TT. ^Samples were further enriched in modified semi-solid Rappaport Vassiliadis (MSRV) after initial enrichment in TT. •'The PCR was conducted after 24 h of secondary enrichment in MSRV. ''Samples were struck on Chromogenic/XLT4 agar after secondaiy enrichment in MSRV. , ^Used to suspend a pool of 80 soiled foot covers, which was split into 100-mL samples each receiving a serially diluted Salmonella culture and incubated for 24 h.

Table 4. The minimum number of cells needed for detection of 5 different serovars of Salmonella entérica from raw feed ingredients initially enriched in buffered peptone water (BPW) or tetrathionate (TT) d2l

Salmonella serovar BPW^ Kentucky Berta Typhimurium Enteritidis Tennessee Kentucky Berta Typhimurium Enteritidis Tennessee

PCR limit of detection (cfu-mL-1) 2,500 2.9 30 180 180 0.25 0.29 3 1.8 0.18

d4

d3 PCR3 limit

MSRV2 limit of detection Not Not Not Not Not

detected detected detected detected detected 2.5 2.9 3 1.8 1.8

of detection (cfu-mL-l)

Plating'' limit of detection

250 2.9 30 180 180

25 2.9 30 180 1.8

2.5 2.9 3 1.8 1.8

2.5 2.9 3 1.8 1.8

PCR was conducted after 24 h of incubation in TT. ^Samples were further enriched in modified semi-solid Rappaport Vassiliadis (MSRV) after initial enrichment in TT. ^The PCR was conducted after 24 h of secondary enrichment in MSRV. ''Samples were struck on Chromogenic/XLT4 agar after secondary enrichment in MSRV. ^Used to suspend 11 feed samples consisting of 8 soy germ meal and 3 corn germ meal, which was split into 100-mL samples each receiving a serially diluted Salmonella culture and incubated for 24 h.

2766

SCHULTZ ET AL.

in TT because TT is selective and Salmonella must overcome inhibitors aimed at suppressing nontarget organisms (Wilson et al., 2006). In these experiments, a frozen culture was thawed and immediately used to simulate stressed cells. However, the stresses encountered in a poultry environment would most likely be associated with desiccation or heat (Shepherd et al,, 2010), Therefore, this may be a hmitation of the study. The differences in detection limits among samples taken from different locations may be due to several factors. First, background microflora would be expected to be different because climates among the farms were varied. As discussed, this background microflora can have an impact on the ability to recover cultures after enriching. Second, variations in the soil chemistry among the farms could have affected the ability to detect Salmonella by PCR, Chemicals present in soil, such as humic acid, are well known to inhibit PCR reactions (Lantz et al,, 1994; Abu Al-Soud and Râdstrom, 1998). In addition, PCR inhibitors present in the samples may have affected the ability to detect Salmonella. In addition to soil chemicals, other inhibitors present such as those in fecal materials collected from the swabs may have affected the detection limits (Wilson, 1997). Finally, the age of the flock, type of litter used, and the number of birds housed together may all have impacts as well. The result of differences in detection limits between primer sets was an interesting finding. In a few samples, the primers targeting SE had a lower detection limit than the 5. enienca-specific primers. Hoorfar et al, (2000) reported inhibition of the primer set specific for S. entérica by Campylobacter jejuni cultures, Campylobacter jejuni is extremely prevalent in the poultry environment (Lee and Newell, 2006), However, C jejuni does not grow in TT, but can grow in BPW, Therefore, inhibition by C jejuni may partially explain the differences in the sensitivity of the S. entérica primers but only in those samples cultured in BPW. Although, many methods exist for isolating Salmonella, establishing a standard method is essential when multiple laboratories will be comparing results. Standard methods allow data to be compared that are collected from independent laboratories and the development of databases against which future data can be compared. However, as demonstrated by our data, environmental factors that differ among laboratory locations can affect the results of an assay and this may be an inherent problem with standardization that needs to be taken into account. These experiments demonstrate that the sensitivity of Salmonella enhanced if samples are initially enriched in TT versus BPW, The detection of Salmonella is essential for control measures to be implemented. Considering the health burden of Salmonella foodborne illness, sensitive detection can be crucial to prevention of salmonellosis.

ACKNOWLEDGMENTS This study was funded by a Research Advisory Committee grant awarded by Cobb-Vantress (Siloam Springs, AR),

REFERENCES Abu Al-Soud, W., and P. Râdstrom. 1998. Capacity of nine DNA polymerases to mediate DNA amplification in the presence of PCR-inhibiting samples. Appl. Environ. Microbiol. 64:37483753. Agron, P., R. Walker, H. Kinde, S. Sawyer, D. Hayes, J. WoUard, and G. Andersen. 2001. Identification by subtractive hybridization of sequences specific for Salmonella entérica serovar Enteritidis. Appl. Environ. Microbiol. 67:4984-4991. APHIS-USDA (Animal and Plant Health Inspection Service-USDA). 2001. The National Poultry Improvement Plan. Accessed Jun. 22, 2012. http://www.aphis.usda.gov/publications/animal_ health/content/printable_version/npip-helping_you.pdf. Braden, C. R. 2006. Salmonella entérica serotype Enteritidis and eggs: A national epidemic in the United States. Clin. Infect. Dis. 43:512-517. Byrd, J. A., B. M. Hargis, D. E. Corrier, R. L. Brewer, D. J. Caldwell, R. H. Bailey, J. L. McReynolds, K. L. Herron, and L. H. Stanker. 2002. Fluorescent marker for the detection of crop and upper gastrointestinal leakage in poultry processing plants. Poult. Sei. 81:70-74. Edel, W., and E. H. Kampelmacher. 1973. Bull. World Hlth. Org 48:167-174. Hoorfar, J., P. Aherns, and P. Radstrom. 2000. Automated 5' nuclease PCR assay for identification of Salmonella entérica. 3. Clin. Microbiol. 38:3429-3435. Jain, S., and J. Chen. 2006. Antibiotic resistance profiles and cell surface components of salmonellae. J. Food Prot. 69:1017-1023. Koyuncu, S., and P. Haggblom. 2009. A comparative study of cultural methods for the detection of Salmonella in feed and feed ingredients. BMC Vet. Res. 5:6. Lantz, P. G., B. Hahn-Hagerdal, and P. Râdstrom. 1994. Sample preparation methods in PCR-based detection of food pathogens. Trends Food Sei. Technol. 5:384-389. Lee, M., and D. Newell. 2006. Campylobacter in poultry: Filling an ecological niche. Avian Dis. 50:1-9. Linam, W. M., and M. A. Gerber. 2007. Changing epidemiology and prevention of Salmonella infections. Pediatr. Infect. Dis. J. 26:747-748. Shepherd, M., R. Singh, J. Kim, and X. Jiang. 2010. Effect of heatshock treatment on the survival of Escherichia coli O157:H7 and Salmonella entérica Typhimurium in dairy manure co-composted with vegetable wastes under field conditions. Bioresour. Biotech. 101:5407-5413. USDA Economic Research Service. 2007. United States Department of Agriculture Economic Research Service, foodborne illness cost calculator for Salmonella. Accessed Aug. 13, 2009. http://www. ers.usda.gov/Data/FoodborneIllness/salm_Intro.asp. Wegener, H. C, T. Hald, D. L. F. Wong, M. Madsen, H. Korsgaard, F. Bajer, P. Gerner-Smidt, and K. Molbak. 2003. Salmonella control programs in Denmark. Emerg. Infect. Dis. 9:774-780. Wilson, C , W. Andrews, and P. Poelma. 2006. Recovery of Salmonella from milk chocolate using chemically defined medium and five nondefined broths. J. Food Sei. 45:310-313. Wilson, I. G. 1997. Inhibition and facilitation of nucleic acid amplification. Appl. Environ. Microbiol. 63:3741-3751. Zhao, S., D. G. White, S. L. Friedman, A. Glenn, K. Blickenstaff, S. L. Ayers, J. W. Abbott, E. Hall-Robinson, and P. F. McDermott. 2008. Antimicrobial resistance in Salmonella entérica serovar Heidelberg isolates from retail meats, including poultry, from 2002 to 2006. Appl. Environ. Microbiol. 74:6656-6662.

Copyright of Poultry Science is the property of Poultry Science Association, Inc. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.