Abstract
Poultry is a major reservoir for foodborne Salmonella serovars. Salmonella Typhimurium, Salmonella Enteritidis, Salmonella Heidelberg, Salmonella Kentucky, and Salmonella Senftenberg are the most prevalent serovars in U.S. poultry. Information concerning the interactions between different Salmonella species and host cells in poultry is lacking. In the present study, the above mentioned Salmonella serovars were examined for invasion, intracellular survival, and their ability to modulate oxidative burst and nitric oxide (NO) responses in chicken macrophage HD11 cells. All Salmonella serovars demonstrated similar capacity to invade HD11 cells. At 24 h post-infection, a 36–43% reduction of intracellular bacteria, in log10(CFU), was observed for Salmonella Typhimurium, Salmonella Heidelberg, Salmonella Kentucky, and Salmonella Senftenberg, whereas a significantly lower reduction (16%) was observed for Salmonella Enteritidis, indicating its higher resistance to the killing by HD11 cells. Production of NO was completely diminished in HD11 cells infected with Salmonella Typhimurium and Salmonella Enteritidis, but remained intact when infected with Salmonella Heidelberg, Salmonella Kentucky, and Salmonella Senftenberg. Phorbol myristate acetate-stimulated oxidative burst in HD11 cells was greatly impaired after infection by each of the five serovars. When newly hatched chickens were challenged orally, a high rate (86–98%) of systemic infection (Salmonella positive in liver/spleen) was observed in birds challenged with Salmonella Typhimurium, Salmonella Enteritidis, Salmonella Heidelberg, and Salmonella Kentucky, while only 14% of the birds were Salmonella Senftenberg positive. However, there was no direct correlation between systemic infection and in vitro differential intracellular survival and modulation of NO response among the tested serovars.
Introduction
Chicken macrophages play a critical role in the defense against microbial infection, in which they detect, phagocytize, and produce microbicidal substances, including reactive radical oxygen species (ROS), nitric oxide (NO), lysozyme, and proteolytic enzymes, to kill the infectious agents (Okamura et al., 2005; Withanage et al., 2005; Babu et al., 2006). Robust macrophage functionality is associated with increased resistance to systemic spread (Wigley et al., 2006) and intestinal colonization (Sun et al., 2008) by Salmonella. However, the role of macrophages in controlling Salmonella infection and the interaction between Salmonella and macrophages in chickens are much less studied and remain mostly unclear.
In the present study, cell invasion, intracellular survival, and modulation of antimicrobial activity (NO and oxidative burst response) in chicken macrophage HD11 cells of the above mentioned five serovars were examined. Additionally, systemic infection by these five serovars in newly hatched chickens was also investigated.
Materials and Methods
Bacteria
Primary poultry isolates Salmonella Typhimurium and Salmonella Enteritidis were obtained from the National Veterinary Services Laboratory (Ames, IA) and were resistant to novobiocin–nalidixic acid (Kogut et al., 1995). Salmonella Heidelberg, Salmonella Kentucky, and Salmonella Senftenberg used in this study were field isolates from broilers and were susceptible to novobiocin–nalidixic acid. All strains were susceptible to gentamicin. Salmonella from stocks were cultured overnight at 41°C in a Tryptic Soy Broth (TSB; Becton, Dickinson, and Company, Franklin Lakes, NJ) and the overnight cultures were transferred to a fresh TSB and cultured for 4 h to reach an exponential growth phase, and the bacteria were collected, washed, and resuspended in phosphate-buffered saline (PBS) at a final concentration of ∼2×109 colony-forming unit (CFU)/mL. Heat-killed Salmonella (HKS) were prepared by incubating the bacterial suspension at 75°C water bath for 15 min.
Chickens
Chickens (Hy-Line W36) were obtained from Hy-Line International (Bryan, TX) on the day-of-hatch. Birds were placed in floor pens with pine shavings in a controlled environment (biosafety level 2) and provided ad libitum access to water and a balanced unmedicated corn-soybean based diet with nutrient rations meeting or exceeding the recommendations of the National Research Council (NRC, 1994). The experiments comply with the Animal Care and Use Experimental Animal Protocol (Southern Plains Agricultural Research Center, Agriculture Research Service, U.S. Department of Agriculture).
HD11 Cells
The MC29 virus–transformed chicken macrophage cell line HD11 (Beug et al., 1979) were maintained in complete Dulbecco's Modified Eagles Medium (DMEM; Invitrogen, Grand Island, NY) containing 10% chicken serum, antibiotics (100 U penicillin/mL and 100 μg streptomycin/mL), and 1.5 mM
Cell invasion and intracellular viability assay
Prior to infection, the culture medium was removed and cells were washed once and replaced with 200 μL of plain DMEM (without chicken serum and other additives). Aliquots of 50 μL of Salmonella suspensions (∼2×109 CFU/mL) were added to each well with four replicate wells for each serovar and incubated for 1 h at 39°C in a 5% CO2 humidified incubator. At 1 h post-infection (hpi), the infection medium was removed and the cells were washed once and treated with 100 μg/mL of gentamicin sulfate in complete DMEM for 1 h to kill extracellular bacteria. After gentamicin treatment, infected cells were washed twice with PBS, lysed for 10 min in 300 μL of 1% Triton X-100 (in PBS). After lysis, 700 μL of PBS was added to each well and mixed thoroughly. Serial 1:10 dilutions were spread onto Difco's xylose-lysine tergitol 4 (XLT4) agar (Becton, Dickinson, and Company) plates and incubated at 41°C for 24 h. Colonies were counted to determine the CFU of intracellular bacteria at 2 hpi. Similarly, intracellular viable bacterial CFU was determined at 24 hpi after an additional culture of the infected cells for 22 h in the medium containing 20 μg/mL of gentamicin sulfate.
NO production assay
Nitrite, a stable metabolite of NO, produced by activated macrophages was measured by the Greiss assay (Green et al., 1982). HD11 cells in 24-well plates were treated, in four replicates, with live (prepared in the same way as above) or HKS as described in the killing assay. After 24 h of incubation, nitrite concentrations in the culture media were determined as previously described (He et al., 2009).
Phorbol myristate acetate (PMA)–stimulated oxidative burst
Oxidative burst of HD11 cells was measured as described (He et al., 2005). To evaluate the effect of Salmonella infection on oxidative burst of HD11 cells, PMA (Sigma) stimulated oxidative burst was measured in both Salmonella-infected and HKS-treated HD11 cells. The cells were first treated in plain DMEM with 10 μL of live or HKS suspensions (∼2×109 CFU) in a final volume of 50 μL/well at 39°C for 1 h. Following the treatment, the cells were washed and stimulated in plain DMEM with PMA (0.5 μg/mL) in a final volume of 100 μL/well containing 10 μg/mL of 2′,7′-dichlorfluorescein-diacetate (DCFH-DA; Sigma) and 100 μg/mL of gentamicin sulfate for 1 h at 39°C in 5% CO2 and 95% humidity. The relative fluorescent units (RFU) at the end of incubation were measured (485/530 nm) using Genios Plus Plate Reader (Tecan US Inc., Durham, NC).
In vivo organ invasion
Day-old chickens, 25 per group, were orally challenged with 0.5 mL of each different Salmonella serovar (∼5×108 CFU/bird) and housed in separated rooms. At 4 days post-infection (dpi), chickens were euthanized with CO2,and liver and spleen were aseptically removed from each chicken and cultured as a combined sample in tetrathionate broth overnight (18–24 h) at 41°C according to guidelines of the U.S. Department of Agriculture (USDA, 1989). After incubation, aliquots of 10 μL of broth were streaked on XLT4 plates and incubated for 24 h at 41°C. Two independent experiments were conducted at different dates, and a total of 50 chickens were used for each treatment group.
Statistical analysis
At least three independent experiments for NO, oxidative burst, and invasion, and two separate experiments for organ invasion were conducted. Statistical difference was determined at the level of p<0.05 by Student's t-test using SigmaStat software (Jandel Corp., Richmond, CA).
Results
Differential modulation of NO production in HD11 cells by Salmonella serovars
Salmonella infection–induced NO production in HD11 cells was serovar-dependent (Fig. 1A): Salmonella Typhimurium and Salmonella Enteritidis induced little or no output of NO, while Salmonella Heidelberg, Salmonella Kentucky, and Salmonella Senftenberg stimulated significant amounts of NO production. This inhibition of NO production in HD11 cells was observed only in treatments with live Salmonella Typhimurium and Salmonella Enteritidis; HD11 cells stimulated with HKS showed a strong NO production regardless of the Salmonella serovar (Fig. 1B).

Effect of Salmonella infection on nitric oxide (NO) production in HD11 cells. HD11 cells were infected with Salmonella for 1 h in 24-well plates at 39°C in a 5% CO2 humidified incubator. At 1 h post-infection (hpi), extracellular Salmonella were killed by incubation with media containing 100 μg/mL of gentamicin sulfate for 1 h; the cells were washed and then cultured for an additional 22 h in a medium containing 20 μg/mL of gentamicin sulfate; and nitrite contents in cell culture media were determined. Treatment with heat-killed Salmonella (HKS) was performed identically as with live Salmonella.
Down-regulation of HD11 cell oxidative burst potential by intracellular Salmonella
To examine the effect of intracellular Salmonella on the oxidative burst capacity of HD11 cells, PMA was used to stimulate oxidative burst in HD11 cells infected with Salmonella or treated with HKS. After invasion, all Salmonella serovars tested significantly diminished the oxidative burst potential of HD11 cells and rendered the macrophages irresponsive to PMA stimulation (Fig. 2A). Metabolically inhibition by intracellular Salmonella was the most likely cause for the loss of oxidative response of Salmonella-infected cells to PMA stimulation, since HD11 cells pretreated with HKS in an identical manner displayed no inhibitory effect on oxidative burst response to a subsequent stimulation with PMA (Fig. 2B).

Effect of Salmonella infection on Phorbol myristate acetate (PMA)–stimulated oxidative burst in HD11 cells. HD11 cells were infected with Salmonella or treated with heat-killed Salmonella (HKS) for 1 h in 96-well optical bottom black plates at 39°C in a 5% CO2 humidified incubator. At 1 h post-infection (hpi), the cells were washed and stimulated with PMA (0.5 μg/mL) in a final volume of 100 μL/well containing 10 μg/mL of DCFH-DA and 100 μg/mL of gentamicin sulfate for 1 h. The relative fluorescent units (RFU) at the end of incubation were measured (485/530 nm).
Cell invasion and intracellular survival in HD11 cells by Salmonella
All Salmonella serovars demonstrated similar capacity (p≥0.05) to invade HD11 cells as demonstrated by the CFU at 2 hpi, even though Salmonella Typhimurium and Salmonella Enteritidis invasion was numerically greater than Salmonella Heidelberg, Salmonella Kentucky, and Salmonella Senftenberg (Table 1). The viability of Salmonella Enteritidis at 24 hpi, however, was significantly higher (p<0.05) than that of Salmonella Typhimurium, Salmonella Heidelberg, Salmonella Kentucky, and Salmonella Senftenberg, indicating that Salmonella Enteritidis was the most resistant to intracellular killing by HD11 cells among the five serovars.
HD11 cells were infected with Salmonella at 2×109 CFU/mL for 1 h at 39°C in a 5% CO2 humidified incubator and intracellular viable Salmonella (CFU in log10 scale) at 2 and 24 h post-infection (hpi) were counted. Data are mean±standard deviations of CFU/well. Reduction (%)={[CFU (2 hpi) – CFU (24 hpi)]/CFU (2 hpi)}/100. Different letters indicate that the differences between these groups are statistically significant (p<0.05).
Systemic invasion by Salmonella in neonatal chickens
Systemic infection in young chickens by these five Salmonella serovars was investigated by examining the presence of Salmonella in the liver/spleen of challenged birds at 4 dpi (Table 2). A high percentage (86–98%) of positive birds was observed in groups infected with Salmonella Typhimurium, Salmonella Enteritidis, Salmonella Heidelberg, and Salmonella Kentucky, whereas only 14% of the birds were positive in the Salmonella Senftenberg challenged group. The unchallenged control birds were all Salmonella negative.
Day-old chickens, 25 per group, were orally challenged with different Salmonella serovar (∼5×108 CFU/bird). At 4 days post-infection (dpi), chickens were euthanized, and liver and spleen were removed to test for organ invasion by Salmonella. Data in columns for Trials 1 and 2 are Salmonella-positive birds in each group at 4 dpi. Data in column for mean±SD are means and standard deviations of the Salmonella-positive birds (%) in each group of the two trials.
Discussion
Survival inside the macrophage is essential for Salmonella virulence and systemic infection (Fields et al., 1986; Schwan et al., 2000; Guiney, 2005). Salmonella virulence depends at least partially on the type III secretion system (T3SS), which secrets and delivers nearly 40 different virulence effectors into host cells, to facilitate invading, surviving, and replicating within host cells (Haraga et al., 2008; Ibarra et al., 2009; Malik-Kale et al., 2011). To defend against the host cell antimicrobial defense mechanisms, Salmonella produce effector proteins which manipulate host cells to delay the phagolysosomal maturation and hence avoid exposure to lysosomal contents (Haraga et al., 2008) and secret various metabolic enzymes which neutralize the antimicrobial effect of free radicals oxygen and nitrogen species (ROS and RNS) (Aussel et al., 2011; Henard and Vázquez-Torres, 2011; Slauch, 2011).
Professional phagocytes generate ROS in the process of an oxidative burst during phagocytosis of microbes or in response to stimulation by microbial components (Fang, 2011). ROS production in response to microbe and microbial component stimulation play a critical role in controlling microbial infection (Ogier-Denis et al., 2008; Lam et al., 2010). The exact role of ROS in controlling intracellular Salmonella in macrophages is debatable (Fang, 2011), since Salmonella carry abundant enzymes (catalases, peroxiredoxins, superoxide dismutases) to neutralize the effect of ROS (Aussel et al., 2011; Fang, 2011). PMA is a protein kinase C activator and stimulates a strong oxidative burst in chicken phagocytes (He et al., 2005). The effect of intracellular Salmonella on the oxidative burst response of HD11 cells to PMA stimulation has not been reported and therefore was examined in the present study. Macrophage HD11 cells infected with the five Salmonella serovars showed severely impaired ROS response to PMA stimulation as compared to the non-infected cells, while the cells treated identically with dead Salmonella (HKS) demonstrated a normal ROS response to PMA. The lack of ROS response to PMA stimulation in Salmonella-infected HD11 cells is likely caused by intracellular Salmonella which may inhibit phagocyte NADPH oxidase activity and metabolically neutralize the ROS products. Our results clearly indicate that Salmonella serovars are well adapted to evade the ROS-mediated killing in macrophage. Although the ROS-mediated direct killing of intracellular Salmonella might be limited in chicken macrophages as our results suggest, accumulated evidence suggests that ROS can act as signaling molecules to indirectly assert an antimicrobial role. For example, ROS have been reported to activate MAP kinase and transcription factors NF-κB and AP-1, up-regulate inflammatory cytokine and chemokine expression, and induce the formation of autophagy (Torres and Forman, 2003; Closa and Folch-Puy, 2004; Huang et al., 2011). Activation of these cellular functions plays a critical role in controlling intracellular Salmonella (Rosenberger and Finlay, 2002; Sahlberg et al., 2007; Jones et al., 2008; Deretic, 2011).
NO response to microbial stimulation is an important innate immune function of macrophages and plays a critical role in controlling the proliferation of intracellular bacterial pathogens such as Salmonella Typhimurium (Mastroeni et al., 2000; Alam et al., 2002; 2008). However, virulent factors secreted via Salmonella T3SS can suppress iNOS activity (Das et al., 2009) and prevent iNOS-containing vesicle trafficking to phagosomes, hence limiting exposure of Salmonella to RNS (Chakravortty et al., 2002). Additionally, Salmonella possess three major enzymes (flavohemoglobin Hmp, flavorubredoxin NorV, and cytochrome c nitrite reductase NrfA) that can detoxify NO under different environmental conditions (Bang et al., 2006; Mills et al., 2008). Previously, infection with Salmonella Typhimurium and Salmonella Enteritidis has been shown to induce NO production in chicken macrophages (Okamura et al., 2005; Withanage et al., 2005; Babu et al., 2006). However, our results show that NO production in HD11 cells was completely inhibited by infection with Salmonella Enteritidis and only minor amounts of NO was produced in cells infected with Salmonella Typhimurium. Infection of HD11 cells with Salmonella Heidelberg, Salmonella Kentucky, and Salmonella Senftenberg induced large amounts of NO. The results demonstrate a dramatic difference in their ability to modulate host cell NO response among these wild-type Salmonella serovars. The lack of or diminished NO response in HD11 cells to Salmonella Typhimurium and Salmonella Enteritidis infection is probably due to the inhibition on iNOS or the metabolic detoxification of NO, since heat-killed Salmonella Typhimurium and Salmonella Enteritidis were able to induce large quantities of NO comparable to the levels induced by the other three serovars. The exact mechanism that enables Salmonella Typhimurium and Salmonella Enteritidis to prevent NO production in HD11 cells is not clear and needs to be further investigated. The discrepancy in NO induction by Salmonella Typhimurium and Salmonella Enteritidis infection between this study and previous studies (Okamura et al., 2005; Withanage et al., 2005; Babu et al., 2006) cannot be readily explained and may be due to different conditions under which these experiments were conducted.
There was no difference in the rate of internalization of Salmonella among the serovars used in this study. Within 24 hpi, HD11 cells were able to limit intracellular Salmonella growth and achieved reduction of viable intracellular Salmonella of all serovars tested. However, Salmonella Enteritidis had the highest viability at 24 hpi among the tested serovars, indicating that Salmonella Enteritidis was the most resistant against macrophage-mediated bacterial killing. These results provide supporting evidence to an epidemiological observation that Salmonella Enteritidis is more adapted to cause systemic infections in chickens (Foley et al., 2011). There was no apparent indication that the ability of Salmonella to down-regulate ROS and RNS responses provided an advantage in survival within chicken macrophages, since Salmonella Typhimurium had a similar, if not the lowest, viability among the five tested serovars despite the fact that it almost completely abrogated NO production in HD11 cells. Therefore, our results indicate that direct killing by ROS or RNS may not play a determinant role in intracellular survival of Salmonella in chicken macrophages.
The ability of Salmonella to invade via the intestine and spread systemically plays an important role in Salmonella colonization of reproductive organs, a main cause for internal contamination of eggs (Gast et al., 2004). Systemic infection and reproductive organ colonization by serovars Salmonella Enteritidis, Salmonella Typhimurium, and Salmonella Heidelberg has been previously reported (Gantois et al., 2008). However, the present study was the first to report that Salmonella Kentucky is capable of causing systemic infection in liver and spleen through intestinal invasion in young chickens. It is interesting to note that chickens challenged with Salmonella Senftenberg had an exceptionally low rate (14%) of systemic infection as compared to chickens challenged with other serovars (86–98%). Although Salmonella Senftenberg has been reported to persist in and frequently isolated from poultry hatching houses, farm houses, and raw feed materials (Liu et al., 1969; Bailey et al., 2001; Kim et al., 2007; Pedersena et al., 2008), it remains a less prevalent strain in chicks, hens, and poultry products. Our results demonstrated that Salmonella Senftenberg lacks the ability to attain systemic infection, suggesting this strain is deficient in its ability to invade.
In summary, we have examined the five most prevalent Salmonella serovars in U.S. poultry for intracellular survival and their ability to modulate antimicrobial activity in chicken macrophage HD11 cells. Intracellular Salmonella impaired the oxidative burst response of HD11 to PMA stimulation. Infection with Salmonella Typhimurium and Salmonella Enteritidis, but not Salmonella Heidelberg, Salmonella Kentucky, and Salmonella Senftenberg, abolished NO production in HD11 cells. Serovar Salmonella Enteritidis was best adapted to survive inside HD11 cells among the tested serovars. Newly hatched chickens were vulnerable to systemic infection by Salmonella Typhimurium, Salmonella Enteritidis, Salmonella Heidelberg, and Salmonella Kentucky, but not Salmonella Senftenberg. These results demonstrate the different capabilities to modulate the immune response in chicken macrophages among Salmonella serovars; however, further investigations are needed to identify factors that control intracellular survival and systemic infection.
Footnotes
Disclosure Statement
No competing financial interests exist.
