Research Note: Contamination of eggs by Salmonella Enteritidis and

Salmonella Typhimurium in experimentally infected laying hens in indoor

cage-free housing

Richard K. Gast ,*,1 Deana R. Jones ,* Rupa Guraya,* Kenneth E. Anderson ,y and
Darrin M. Karcher z
*
U. S. National Poultry Research Center, USDA Agricultural Research Service, Athens, GA 30605, USA;
y
Department of Poultry Science, North Carolina State University, Raleigh, NC 29765, USA; and zDepartment of
Animal Sciences, Purdue University, West Lafayette, IN 47907, USA

ABSTRACT Contaminated eggs are a leading source the hens in each room were orally inoculated with
of human Salmonella infections and this problem contin- 8.0 £ 107 cfu of 2-strain mixtures of either S. Enteritidis
ues to challenge public health authorities and egg indus- (2 rooms) or S. Typhimurium (2 rooms), and the entire
tries around the world. Salmonella invasion of the internal contents of all eggs laid 5 to 30 d postinocula-
ovaries and oviducts of infected laying hens can result in tion in nest boxes or on the flooring substrate were cul-
bacterial deposition inside the edible portions of devel- tured to detect Salmonella. Contaminated eggs were
oping eggs. The introduction, persistence, and transmis- laid between 8 and 28 d postinoculation. The overall
sion of salmonellae in commercial egg-laying flocks are incidence of S. Enteritidis isolation from eggs (3.41%)
influenced by flock management practices, but the food was significantly (P = 0.0005) greater than S. Typhimu-
safety ramifications of different types of laying hen hous- rium (1.19%). The contamination frequencies associated
ing remain unresolved. The present study assessed the with the 2 egg collection locations were not significantly
frequency of internal contamination of eggs after experi- different (P > 0.05). These results demonstrate that oral
mental Salmonella Enteritidis and S. Typhimurium infection of a relatively small proportion of laying hens
infection of laying hens in indoor cage-free housing. in indoor cage-free housing with invasive Salmonella
Groups of 72 hens were housed on wood shavings in iso- serovars can result in the production of internally con-
lation rooms simulating commercial cage-free barns with taminated eggs at low frequencies over a period of nearly
community kick-out nest boxes and perches and 1/3 of a month postinoculation.
Key words: Salmonella Enteritidis, Salmonella Typhimurium, laying hens, cage-free housing, eggs
2021 Poultry Science 100:101438
https://doi.org/10.1016/j.psj.2021.101438

INTRODUCTION strains of this pathogen are often genetically identical to

laying flock isolates. Contaminated eggs have also been
Contaminated eggs can transmit Salmonella infec- implicated as the sources of human S. Typhimurium
tions to humans and this problem has challenged public infections, particularly in Australia. Meaningful reduc-
health authorities and egg industries around the world tions in the occurrence of human Salmonella infections
for more than 3 decades (Chousalkar et al., 2018). Eggs have been attributed to the widespread implementation
and egg products have been identified as the most com- of comprehensive risk reduction and flock testing pro-
mon food vehicles for human salmonellosis in Europe, grams for egg-producing flocks. Nevertheless, as the inci-
principally caused by Salmonella enterica serovar Enter- dence of human S. Enteritidis infections in the United
itidis. The prevalence of S. Enteritidis in commercial States has not declined significantly for more than a
egg-laying chickens has been directly correlated with the decade, refinements or improvements in pathogen con-
incidence of human infections, and human outbreak trol strategies for laying flocks continue to be topics of
considerable interest.
Published by Elsevier Inc. on behalf of Poultry Science Association Infections with invasive Salmonella serovars are typi-
Inc. This is an open access article under the CC BY-NC-ND license cally initiated by oral ingestion of the pathogen, leading
(http://creativecommons.org/licenses/by-nc-nd/4.0/). to intestinal colonization and rapid dissemination to
Received June 17, 2021.
Accepted August 15, 2021.
internal organs such as the liver and spleen. In mature
1
Corresponding author: Richard.Gast@usda.gov laying hens, subsequent involvement of the ovaries and

1
2 RESEARCH NOTE

oviducts can result in bacterial deposition inside the isolation room simulated a commercial cage-free barn
developing yolk and albumen prior to shell formation with community kick-out nest boxes and perches.
and oviposition (Gantois et al., 2009). Higher frequen- Hens were housed on a wood shavings flooring sub-
cies of egg contents contamination have often been strate at a density of 1,710 cm2 of horizontal space
observed with S. Enteritidis than with other serovars, per bird. Water was supplied ad libitum via auto-
associated with greater adherence to reproductive tract matic nipple-type drinkers and feed (antibiotic-free,
mucosa and invasion of ovarian granulosa cells. How- ground mash layer ration) was provided according to
ever, even after infection of hens with high oral doses, S. standards for commercial egg production. All experi-
Enteritidis is generally found inside eggs at low frequen- mental protocols were approved by the Institutional
cies and in low concentrations. Because hens in commer- Animal Care and Use Committee of the US National
cial laying flocks are most likely exposed to salmonellae Poultry Research Laboratory.
from environmental reservoirs in relatively small doses,
egg contamination occurs very infrequently.
Opportunities for the introduction, persistence, and Experimental Infection of Laying Hens With
transmission of salmonellae in commercial egg-laying Salmonella
flocks are strongly influenced by flock management
In each of 2 trials, 24 of the 72 hens in one isolation
strategies and practices as well as by some intrinsic
room were orally inoculated with a measured dose of a 2-
design features of poultry housing facilities. In recent
strain mixture of S. Enteritidis and 24 of the 72 hens in
years, concerns about animal welfare have spurred the
the other room were similar infected with a 2-strain mix-
commercial egg industry to consider transitioning from
ture of S. Typhimurium. One room in each trial was
conventional cage-based hen housing to less space-inten-
infected at 23 wk of age and the other room at 24 wk,
sive alternatives such as indoor cage-free aviaries or out-
with the order of administration of the 2 serovars
door free-range systems. However, the public health
reversed between trials. All Salmonella strains were orig-
ramifications of laying hen housing are not entirely clear,
inally isolated from internal organs of naturally infected
as previous research regarding the effects of different
chickens in commercial settings. Each inoculum strain
poultry management systems on Salmonella and other
was resuscitated by transfer into tryptic soy (TS) broth
food safety pathogens has yielded diverse and sometimes
(Acumedia, Neogen Corp., Lansing, MI) for 2 successive
contradictory results. A number of confounding issues,
cycles of 24-h incubation at 37°C. After cell numbers in
including differences in bird stocking densities, levels of
each incubated culture were estimated by determining
exposure to contaminated dust and feces, and popula-
its optical density at 600 nm, equal numbers of the 2
tions of biological vectors, have often obscured meaning-
inoculum component strains were combined, and further
ful comparisons between systems in these studies
serial dilutions in 0.85% saline were performed to
(Holt et al., 2011). Nevertheless, prior data has estab-
achieve the desired final cell concentration. Plate counts
lished that the unique inherent characteristics of each
on brilliant green (BG) agar (Acumedia) confirmed
housing system present correspondingly unique chal-
that the final cell concentration in each 1.0-mL oral inoc-
lenges for pathogen control efforts (Jones et al., 2015).
ulum dose was 8.0 £ 107 cfu. Orally infected hens were
Understanding how housing and management factors
identified by colored leg bands.
affect the outcomes of Salmonella infections in poultry is
a key issue for ensuring safe transition into cage-free sys-
tems. The objective of the present study was to assess Pre-inoculation Cloacal Swab Samples
the frequency of internal contamination of eggs after
experimental Salmonella Enteritidis and Salmonella Immediately before inoculation, sterile cotton swabs
Typhimurium infection of laying hens in indoor cage- were used to collect cloacal swab samples from 56 ran-
free housing. domly selected hens in each room. Each sample was
transferred into 10 mL of buffered peptone water (Acu-
media) and incubated for 24 h at 37°C. A 0.1 mL portion
MATERIALS AND METHODS of each culture was then transferred into 10 mL of Rap-
paport-Vassiliadis broth (Acumedia) and incubated for
Experimental Housing of Laying Hens 24 h at 41.5°C. A 10-mL portion from each of these broth
In each of 2 similar trials, 144 female Tetra Brown cultures was then streaked onto BG agar supplemented
chicks (a strain used by the commercial egg industry) with 0.02 mg/mL of novobiocin (Sigma Chemical Co.,
were obtained from a breeding company and reared St. Louis, MO). These plates were incubated for 24 h at
in cage-free barns at Purdue University (West Lafay- 37°C and then examined for the presence of typical Sal-
ette, IN) without vaccination against Salmonella monella colonies.
infection. At 20 wk of age, the young hens were
transferred to the US National Poultry Research Cen- Egg Contents Samples
ter (Athens, GA), distributed evenly between two
separate rooms of a disease-containment facility, and In each trial, 60 eggs were collected (30 per room)
allowed to acclimate to experimental housing condi- prior to Salmonella inoculation. Beginning at 5 d postin-
tions for 3 wk prior to Salmonella infection. Each oculation and continuing through 30 d postinoculation
 RESEARCH NOTE 3
in each trial, all eggs were collected daily from each contamination was 1.19% (1.74% from eggs laid in nest
room. Eggs were stored at room temperature for up to 3 boxes and 1.13% from eggs laid on the flooring sub-
d until transport to the laboratory for culturing to strate). Internal contamination with S. Enteritidis was
detect internal contamination with Salmonella. Eggs detected from 8 to 26 d postinoculation (Table 1), at an
collected from nest boxes and from the flooring substrate overall frequency of 3.41% (3.37% for nest box eggs and
were separately identified. Eggshell surfaces were disin- 3.74% for flooring substrate eggs). For both S. Typhimu-
fected by dipping for 5 s in 70% ethanol and the shells rium and S. Enteritidis, no significant differences in egg
were then broken against a sharp edge covered by sterile contamination frequencies (P > 0.05) were found
foil strips. The entire liquid contents of each egg were between the 2 experimental replicates or between the 2
transferred to 50 mL of TS broth, mixed by vigorous egg collection locations. The frequencies of recovery of S.
shaking for 15 s, and incubated for 24 h at 37°C. A 0.1- Enteritidis were greater than those of S. Typhimurium
mL portion of each incubated TS broth culture was from eggs laid on the flooring substrate (P = 0.0498)
transferred to 10 mL of Rappaport Vassiliadis broth and from all eggs (P = 0.0005).
(Acumedia) and incubated for 24 h at 41.5°C. A 10-mL Hens infected with both Salmonella challenge serovars
aliquot from each of these broth cultures was then in the present study laid internally contaminated eggs,
streaked onto BG agar plus novobiocin and incubated but S. Enteritidis was recovered from the contents of sig-
for 24 h at 37°C. After incubation of these plates for 24 h nificantly more total eggs (and also from more eggs laid
at 37°C, typical S. Enteritidis colonies were subjected to on the flooring substrate) than was S. Typhimurium.
biochemical and serological confirmation. Salmonella serovars Enteritidis and Typhimurium are
both highly prevalent in commercial poultry. Both sero-
vars are highly invasive and can colonize reproductive
Statistical Analysis tissues (Gantois et al., 2008), although this has been
shown to lead to deposition inside developing eggs more
For each replicate within the 2 serovar-specific trials
often for S. Enteritidis than S. Typhimurium. More
(and for both replicates combined), significant differen-
intense tissue pathology and immune responses associ-
ces (P < 0.05) between the 2 egg collection locations in
ated with S. Typhimurium infection have been postu-
the mean frequencies of Salmonella isolation from inter-
lated to diminish the likelihood of deposition inside eggs.
nal contents of eggs were determined by Fisher’s exact
Fecal contamination of shells appears to be the most
test. Egg contamination frequencies associated with the
likely source of Salmonella Typhimurium contamination
2 Salmonella serovars were similarly compared using the
of eggs in Australia (Pande et al., 2016). The ability of
chi square statistic with Yates correction (because of the
S. Enteritidis to persist in avian reproductive tracts and
larger number of values involved). Data were analyzed
survive inside eggs has been attributed to genes which
with Instat biostatistics software (GraphPad Software,
regulate cell surface lipopolysaccharides and flagella or
San Diego, CA).
are responsible for stress responses (Raspoet et al.,
2014), often manifested as small changes in multiple
RESULTS AND DISCUSSION genes.
Until relatively recently, nearly all egg-laying hens
None of the preinfection fecal samples were Salmo- were housed in conventional cage systems. Considerable
nella-positive in either trial. Internal contamination data are available that characterize the progress, out-
with S. Typhimurium was detected in eggs laid from 10 comes, and control of Salmonella infections in caged
to 28 d after oral inoculation of one-third of the hens poultry. High stocking densities in conventional cages
(Table 1). The overall frequency of S. Typhimurium egg have been associated with an increased frequency of

Table 1. Recovery of Salmonella Typhimurium and Salmonella Enteritidis from the internal contents of eggs laid by experimentally
infected laying hens in indoor cage-free housing.1

First and last

contaminated eggs
Replicate 1 Replicate 2 All replicates (d postinfection)
Salmonella-positive/total (%)
S. Typhimurium
Eggs laid in nest boxes 1/27 (3.70)a 1/88 (1.14)a 2/115 (1.74)abcd 13−22
Eggs laid on flooring substrate 7/572 (1.22)a 6/577 (1.04)a 13/1149 (1.13)ac 10−28
All eggs 8/599 (1.34) 7/675 (1.08) 15/1264 (1.19)ab 10−28
S. Enteritidis
Eggs laid in nest boxes 16/493 (3.25)a 15/426 (3.52)a 31/919 (3.37)d 8−26
Eggs laid on flooring substrate 3/71 (4.23)a 1/36 (2.78)a 4/107 (3.74)bd 10−21
All eggs 19/564 (3.37) 16/462 (3.46) 35/1026 (3.41)cd 8−26
1
Eggs were collected for sampling between the 5th and 30th d after 24 of 72 hens in each cage-free housing room were orally inoculated with approxi-
mately 8.0 £ 107 cfu of two-strain mixtures of either S. Typhimurium or S. Enteritidis. The remaining hens were exposed to infection by horizontal
contact.
a,b,c,d
Values in columns within individual replicate trials or in the column for all replicates combined that share no common superscripts are signifi-
cantly (P < 0.05) different.
4 RESEARCH NOTE

internal organ invasion by S. Enteritidis in experimen- opportunities for recently infected individuals to lay
tally infected hens (Gast et al., 2016), perhaps due to contaminated eggs.
immunosuppressive stress. Less information is available
about Salmonella in laying flocks housed in noncage sys-
tems. Comparative studies have produced variable ACKNOWLEDGMENTS
results, as some studies have found higher frequencies of
Salmonella in environmental, tissue, and egg samples We gratefully express appreciation for excellent tech-
from conventional cage systems and others have nical assistance from Stephen Norris, and Robin Wood-
reported greater Salmonella prevalence and horizontal roof (U. S. National Poultry Research Center, Athens,
transmission in cage-free systems. In an experiment that GA). This study was supported by appropriated funds
examined conventional cage, enriched colony, and avi- of the USDA Agricultural Research Service.
ary housing systems managed under commercial condi-
tions at the same location, no differences were observed
in Salmonella contamination of environmental and egg DISCLOSURES
shell samples (Jones et al., 2015). Dust, feces, rodents, The authors declare that they have no known compet-
and insects contribute to perpetuating Salmonella con- ing financial interests or personal relationships that
tamination in poultry facilities and supporting horizon- could have appeared to influence the work reported in
tal dissemination of infection among hens. Higher risks this paper.
of Salmonella infection are reported among laying flocks
that are larger, consist of older birds, or are housed in
older facilities.
REFERENCES
Despite the fact that all hens in the current experi-
ment were of identical genetic stock and had been Chousalkar, K., R. Gast, F. Martelli, and V. Pande. 2018. Review of
raised in the same location under identical conditions egg-related salmonellosis and reduction strategies in United States,
Australia, United Kingdom and New Zealand. Crit. Rev. Micro-
as chicks and pullets, the groups infected with S. biol. 44:290–303.
Typhimurium laid the majority of their eggs on the Gantois, I., R. Ducatelle, F. Pasmans, F. Haesebrouck, R. Gast,
flooring substrate and the groups infected with S. T. J. Humphrey, and F. Van Immerseel. 2009. Mechanisms of egg
Enteritidis laid the majority of their eggs in the nest contamination by Salmonella Enteritidis. FEMS Microbiol. Rev.
33:718–738.
boxes. This illustrates some of the difficulties pro- Gantois, I., R. Ducatelle, F. Pasmans, F. Haesebrouck, and
ducers face in effectively managing consistent egg F. Van Immerseel. 2008. Salmonella enterica serovar Enteritidis
production with these types of housing systems. How- genes induced during oviduct colonization and egg contamination
ever, although exposure to Salmonella from feces in laying hens. Appl. Environ. Microbiol. 74:6616–6622.
Gast, R. K., R. Guraya, D. R. Jones, K. E. Anderson, and
would presumably increase the likelihood of shell con- D. M. Karcher. 2016. Colonization of internal organs by Salmo-
tamination among eggs laid on the flooring substrate, nella Enteritidis in experimentally infected laying hens housed in
no differences in egg contents contamination were enriched colony cages at different stocking densities. Poult. Sci.
95:1363–1369.
observed between the 2 egg collection sites for either Gast, R. K., P. Regmi, R. Guraya, D. R. Jones, K. E. Anderson, and
serovar. In a prior study using the same cage-free sys- D. M. Karcher. 2019. Contamination of eggs by Salmonella Enteri-
tem and a similar experimental infection model as the tidis in experimentally infected laying hens of four commercial
current investigation, extensive horizontal spread genetic lines in conventional cages and enriched colony housing.
Poult. Sci. 98:5023–5027.
among co-housed hens occurred within two weeks Gast, R. K., D. R. Jones, R. Guraya, K. E. Anderson, and
after the introduction of S. Enteritidis infection D. M. Karcher. 2020. Research note: Horizontal transmission and
(Gast et al., 2020). In the present study, internal egg internal organ colonization by Salmonella Enteritidis and Salmo-
contamination was detected at relatively low overall nella Kentucky in experimentally infected laying hens in cage-free
housing. Poult. Sci. 99:6071–6074.
frequencies between 8 and 28 d after one-third of the Holt, P. S., R. H. Davies, J. Dewulf, R. K. Gast, J. K. Huwe,
hens were orally inoculated with S. Enteritdis or S. D. R. Jones, D. Waltman, and K. R. Willian. 2011. The impact of
Typhimurium. Earlier data have suggested that indi- different housing systems on egg safety and quality. Poult. Sci.
vidual hens seldom lay contaminated eggs for longer 90:251–262.
Jones, D. R., N. A. Cox, J. Guard, P. J. Fedorka-Cray, R. J. Buhr,
than 3 wk after oral infection (Gast et al., 2019). The R. K. Gast, Z. Abdo, L. L. Rigsby, J. R. Plumblee, D. M. Karcher,
present study demonstrated that infection of a rela- C. I. Robison, R. A. Blatchford, and M. M. Makagon. 2015. Micro-
tively small proportion of laying hens in indoor cage- biological impact of three commercial laying hen housing systems.
free housing with invasive Salmonella serovars could Poult. Sci. 94:544–551.
Pande, V. V., R. L. Devon, P. Sharma, A. R. McWhorter, and
result in the production of internally contaminated K. K. Chousalkar. 2016. Study of Salmonella Typhimurium infec-
eggs at low frequencies over a period of nearly a tion in laying hens. Front. Microbiol. 7:203.
month postinoculation. Horizontal transmission of Raspoet, R., C. Appia-Ayme, N. Shearer, A. Martel, F. Pasmans,
F. Haesebrouck, R. Ducatelle, A. Thompson, and
salmonellae in cage-free systems, perhaps enhanced F. Van Immerseel. 2014. Microarray-based detection of Salmonella
by extensive access of hens to each other and to enterica serovar Enteritidis genes involved in chicken reproductive
feces-contaminated flooring substrate, may prolong tract colonization. Appl Environ. Microbiol. 80:7710–7716.