INTRODUCTION
Avian coccidiosis is major parasitic disease of high economic concern in the poultry industry worldwide and is caused by at least seven distinct species of Eimeria apicomplexan protozoa that infect the various areas of intestinal mucosa. The economic loss for avian coccidiosis is estimated to be more than $3 billion worldwide (Williams, 1999) and this cost includes in-feed medication, mortality, impaired growth rate, inefficient feed utilization and reduction in egg production. According to the previous reports, the Eimeria spp. prevalence has been estimated in more than 50% of the flock level around the world (Williams et al., 1996; Al-Natour et al., 2002; Haug et al., 2008; Nematollahi et al., 2009; Sun et al., 2009; Lee et al., 2010a). In general, the majority of Eimeria-infected chickens are afflicted by multiple species of Eimeria with the most prevalent species including E. tenella, E. praecox, E. acervulina, and E. maxima.
It has been a common practice to implement vaccination against coccidiosis or to add anti-coccidial feed additives either in the form of synthetic chemicals or antibiotic ionophores to control coccidiosis by the poultry industry (Williams, 2005; Lee et al., 2009). Although coccidiosis control programs have been widely used in poultry production worldwide, their impact on the developing host immune system and on host-pathogen interaction has not been studied so far. Because there are scientific data that show the influence of gut microbiota, drugs, and dietary feed-additives (e.g., probiotics and phytonutrients) on developing host immune system (Lee et al., 2010b; Yin et al., 2010; Crhanova et al., 2011; Lee and Lillehoj, 2011; Lee et al., 2011a), it is important to gain better understanding of how these factors affect host ability to elicit immune response to field pathogens.
Recent studies document significant immunoregulatory influence of the gut commensal bacteria on host immune status. For example, Bifidobacteria infantis or Faecalibacterium prausnitzii induced Foxp3+ T regulatory cells and increased the expression of the anti-inflammatory cytokine, interleukin (IL)- 10, whereas segmented filamentous bacteria (Candidatus arthromitus) induced T-helper (Th)17 cell development (Lee and Mazmanian, 2010; Atarashi et al., 2011). In addition, potential interaction of various disease control programs with gut microbiota community, populations of Eimeria spp., and host immunity was recently demonstrated (Lee and Lillehoj, 2011; Lillehoj and Trout, 1996). Infection with Eimeria spp. induces antigen-specific humoral and cell-mediated immune responses (Lillehoj and Trout, 1996) and affects gut microbiota communities (Hume et al., 2006) in broiler chickens. Moreover, broiler chickens that are naturally exposed to Eimeria-contaminated litter developed different patterns of Eimeria-specific antibody response depending on the coccidiosis control programs used (Lee et al., 2011b). Finally, gut microbiota plays an important role in health, immunity, and disease prevention of chickens (Dibner et al., 2008). Altered microbiota is linked to obesity, inflammatory bowel disease, atherosclerosis, and cancer (Wlodarska and Finlay, 2010). Collectively, these aforementioned studies raise a possibility that coccidiosis control programs would affect gut microbiome-immune balance in broiler chickens. Unfortunately, very limited information is available on the effect of different kinds of coccidiosis control programs on the host immune status and its responses in broiler chickens. Thus, the present study was aimed to explore whether various coccidiosis control programs involving in-ovo vaccine, synthetic chemicals, and ionophores would affect the immune responses (e.g., innate and adaptive immunity) in commercial meat-type chickens. In this study, we evaluated various immune parameters including serum acute phase protein concentration, subpopulations of spleen lymphocytes, and serum antibody responses following vaccination against infectious bronchitis virus (IBV) and Newcastle disease (ND), two important respiratory viral diseases of poultry.
MATERIALS AND METHODS
1. Birds, Diets and Experimental Design
Two-hundred eighty, one-day-old broiler chickens (Ross) were randomly placed in eight pens (n = 35/pen) with used litter at the University of Delaware Lasher Poultry Diagnostic Laboratory Research Farm. Coccidiosis control programs included in ovo vaccination, six straight programs with either chemicals or ionophores, or both, and one shuttle program (NCB/SAL) with nicarbazin at 0~18 days and salinomycin at 19~35 days. The straight programs contained the following chemicals or ionophores: diclazuril (CLIN), decoquinate (DECX), monensin (COBN), nicarbazin + narasin (MAXI), and salinomycin (SAL), respectively. In addition to the chemicals or ionophores, both straight and shuttle program treatments contained bacitracin methylene disalicylate (BMD) and roxarsone (3-nitro) during the experimental period (0 to 35 days of age). For the vaccination group (CVAC), 18-day old fertilized and developing eggs were in ovo vaccinated with Inovocox® (Embrex, Pfizer, Durham, NC) and the hatched chicks were provided diets with virginiamycin at 0 to 18 days, and BMD and 3-nitro at 19 to 35 days. The experimental design and the antimicrobial concentrations in diets used are shown in Table 1. The control group (NONE) was provided with non-medicated base diet throughout the whole experimental period. The base diet was a mash type consisting of corn, soybean meal, poultry and animal byproducts, and distiller’s dried grains soluble. The experimental diets were formulated by mixing the base diet with the indicated antimicrobials (Table 1). All broiler chicks used in this study had been vaccinated against ND and IBV via coarse spray after hatch. The population density was set at 0.06 m²/bird at 1 week, 0.07 m²/bird at 2 week, 0.09 m²/bird at 3 week, 0.11 m²/bird at 4 week, and 0.15 m²/bird at 5 or more weeks. All protocols were approved by the Beltsville Animal Care and Use Committees of the Beltsville Agricultural Research Center and the University of Delaware.
2. Used Litter
Used litter was obtained from a commercial broiler house with endemic gangrenous dermatitis, transported to the research facility for the current study, and evenly distributed into eight pens after homogenization. Care was taken to ensure that all pens were given homogenized litter to minimize the potential pen effect. Presence of Eimeria spp. oocysts in litters was as low as 107 (SD 1.8) per g of litter without inter-group differences (Lee et al., 2011b). With Eimeria-specific PCR assay, we also confirmed the presence of E. acervulina, E. maxima, E. mitis, E. praecox, and E. tenella, but not E. brunetti or E. necatrix, in the litter samples.
3. Sampling
At 7 days of age, 5 birds per group were randomly selected for blood after euthanasia. Immediately after blood sampling, spleen was also sampled and pooled by treatment group to obtain splenocytes (Lee et al., 2010b) for analysis. At 14, 25, and 34 days of age, five or six birds per group were randomly selected and blood was sampled by heart puncture immediately after euthanasia. Serum was obtained by gentle centrifugation and stored at -20? until use.
4. Chicken α-1-Acid Glycoprotein ELISA
Levels of chicken α-1-acid glycoprotein (α-1-AGP) in serum samples obtained at 34 days of age were measured by ELISA (Life Diagnostics, Inc., West Chester, PA) according to the manufacturer’s instructions.
5. Splenocyte Proliferation
At 7 days of age, pooled splenocytes per group were seeded at 5.0 × 105 /well in 96-well microtiter plates and in- cubated with medium alone (control) or 2.0 μg/mL of conca- navalin A (ConA, Sigma, St. Louis, MO) in a humidified incubator at 41? with 5% CO2 for 48 hrs. Cell proliferation was measured using 2-(2-methoxy-4-nitrophenyl)-3-(4-nitro- phenyl)-5-(2,4-disulfophenyl)-2H-tetrazolium (Cell Counting Kit-8, Dojindo Molecular Technologies, Gaithersburg, MD) by measuring optical density (OD) at 450 nm (OD450) using a microplate spectrophotometer (BioRad, Hercules, CA) as described (Lee et al., 2010b). Proliferation was expressed as the stimulation index (SI), calculated as the ratio of the mean OD450value of ConA-stimulated cells divided by the mean OD450value of medium alone-stimulated cells. Each analysis was performed in quadruplicate.
6. Analysis of Lymphocyte Subpopulations in Spleen
Splenocytes were sampled at 7 days and pooled by the treatment groups. They were > 95% viable by trypan blue exclusion. The cells were stained with monoclonal antibodies (mAb) against chicken CD4 (T helper cells) (Lillehoj et al., 1988), CD8 (cytotoxic T cells) (Lillehoj et al., 1988), TCR1 (γδ T cell receptor) (Chen et al., 1986), or TCR2 (αβ TCR) (Chen et al., 1986) and analyzed with a FACSCaliburTM flow cytometer (BD Biosciences, San Jose, CA). As a positive control, the cells were stained with mAb K55 (chicken pan lymphocyte) (Chung et al., 1991). As a negative control, the cells were stained with mAb HB2 (human T cell; American Type Culture Collection, Manassas, VA). Data were obtained from a total of 104 viable cells. Each analysis was performed in quintuplicate.
7. Measurement of IBV- and ND-reactive Antibody Responses
Viral antibodies against IBV and ND in sera sampled at 7, 14, 25, and 34 days of age were determined using commercial ELISA kits (IDEXX Laboratories, Westbrook, MA) according to the manufacturer's instructions. The results were expressed as antibody titer that was calculated using the formula provided by IDEXX.
8. Statistical Analysis
All data obtained in this study were compared among the different groups by the Duncan’s Multiple Range test following ANOVA using SPSS 15.0 for Windows (SPSS Inc., Chicago, IL). Statistical significance was pre-set at P<0.05.
RESULTS
The levels of α-1-AGP at 34 days of age were significantly changed depending on the coccidiosis control programs (Fig. 1). In general, the concentration of α-1-AGP ranged as low as 3.0 μg/mL in COBN-treated chickens and as high as 282 μg/mL in CLIN-treated chickens. All treatment groups except the CLIN-treated chickens showed significantly low levels of α-1-AGP compared with the non-medicated controls.
At 7 days of age, in ovo-vaccinated broiler chickens showed the highest splenic mitogenic response to ConA compared with the non-medicated control chickens or the groups treated with anticoccidials (Fig. 2).
Composition of splenocyte subpopulations was also altered by the type of coccidiosis control programs at 7 days of age (Fig. 3). In general, the percentage of CD4+ cells were high (P<0.05) in MAXI-treated chickens compared with the nonmedicated control chickens. Most groups, not including NCB/ SAL, SAL, and CLIN groups, showed lower percentages of CD8+ cells compared with the NONE group. Increase in the percentages of TCR1+ cells of COBN- and MAXI-treated chickens, and decrease in TCR2+ cells in DECX-treated chickens were also noted compared with the non-medicated control chickens. In all groups, the percentage of cells stained by the positive control K55 mAb was consistently > 99%, and the percentage of cells stained by the negative control HB2 mAb was consistently < 2% (data not shown).
Serological monitoring of serum antibody titers against IBV and ND is shown in Table 2 and Table 3. In general, IBV-specific antibody titers decreased with age and then increased at 34 days of age (Table 2). There were no statistical differences in IBV antibody titers among the different treatment groups at all ages. In contrast, ND-reactive antibody titers were changed by treatment groups (Table 3). Especially, at 14 days of age, non-medicated control chickens exhibited significantly lower ND-reactive antibody titers compared with all treatment groups except for DECX which remained intermediate. At 25 days of age, ND antibody titers were significantly low in the in-ovo-vaccinated chickens compared with those of the non-medicated controls or the anticoccidial-treated chickens.
DISCUSSION
The experiment was conducted to monitor immune status of broiler chickens raised on used litter and subjected to various coccidiosis control programs including in ovo vaccination or anticoccidial drugs (e.g., chemicals and ionophores) that are commonly used in poultry production. In summary, the levels of α-1-AGP at 34 days of age were significantly lowered (P<0.05) by most of the treatments compared with the non-medicated controls. The in ovo-vaccinated chickens exhibited the highest ConA-stimulated splenocyte mitogenic response at 7 days of age. Different types of coccidiosis control programs changed T-cell subsets in the spleen. Finally, non-medicated control chickens at 14 days of age and in-ovo vaccinated chickens at 25 days of age had the lowest levels of serum antibody titers against ND. It is clear from this study that the coccidiosis control programs commonly used in poultry production influenced various immune parameters of chickens.
α-1-AGP which has been used as an indicator of innate immunity is an acute phase protein (APP) that contributes to restoring homeostasis and restricting microbial growth in an antibody-independent manner following infection, inflammation, or stress (Murata et al., 2004). In broiler chickens treated with Salmonella Typhimurium-derived lipopolysaccharide (Kefali and Toker, 2006), a 4.0-fold increase in the APP levels was observed. In this study, the levels of α-1-AGP were significantly lowered by most of the anti-coccidial treatments. This result supports a notion that different coccidiosis control programs could reduce factors inducing systemic inflammation or infection in broiler chickens. In addition, since the experimental diets contained antibiotic growth promoters (AGPs) such as virginiamycin or BMD, which are commonly accepted drug programs in poultry industry, the coccidiosis control programs or AGPs, working singly or preferentially in combination, may reduce pathogen loads in the intestines, thereby further decreasing inflammation or infection.
As a measure of general systemic lymphocyte immune function, we stimulated splenocytes with T cell mitogen, ConA. It is well known that cell proliferation is an accurate indicator of general cell-mediated immune function. In this study, CVAC-treated chickens exhibited significantly increased ConA-stimulated splenocyte proliferation compared with untreated controls or anticoccidials-treated chickens. This result is not surprising given that vaccination against coccidiosis can induce robust humoral and cell-mediated immunities in broiler chickens (Jang et al., 2011). However, we failed to see anticoccidials-altered splenocyte proliferation, which is contrary to the previous studies (Baba et al., 1998; Munir et al., 2009) in which antibiotic or anticoccidial treatment exhibited increased mitogen-stimulated proliferation compared with the controls. It cannot be excluded the possibility that chickens treated with anticoccidials for 7 days may not be enough to see the effects on splenocyte proliferation. Unfortunately, we did not assay splenocyte proliferation from chickens sampled at 14, 25, and 34 days. Thus, further studies are warranted to address the concern raised in this study.
Splenocyte subpopulations were changed by the coccidiosis control programs, but the effect was dependent on the types of programs used. Specifically, most T-cell subsets (i.e., CD4, CD8, TCR1, and TCR2) were unaffected or decreased by the treatments compared with the controls. In contrast, CD4+ splenocytes in MAXI-treated chickens, and CD8+ splenocytes in COBN- or MAXI-treated chickens were significantly higher compared with those of the non-medicated controls. In agreement with our result, earlier studies reported discordant results on the lymphocyte subpopulations. For example, the absence of an effect of antibiotic treatment on the percentages of B cells, CD4+ , and CD8+ T cells in spleen or peripheral blood has been noted (Baba et al., 1998). In a study by Arias and Koutsos (2006), they reported that antibiotic-fed chickens raised on fresh litter had higher duodenal lymphocytes, but lower ileal lymphocytes compared with non-medicated control chickens. In contrast, antibiotics-fed chickens raised on the used litter had unaltered duodenal or ileal lymphocytes compared with the non-medicated control chickens (Arias and Koutsos, 2006). Our study is the first report showing the effect of various field coccidiosis control programs on host innate and adaptive immunity.
There have been several attempts to modulate the humoral response, especially in viral- or Eimeria-vaccinated chickens by adding natural alternatives to antibiotics such as probiotics or medicinal plants into broilers’ diets (Talebi et al., 2008; Khaligh et al., 2011; Lee et al., 2011a). Although the mechanisms underlying the diet-induced immunomodulation are not well understood, altered gut microbiota which is known to affect host immunity may play an important role (Lee and Lillehoj, 2011). Thus, any measures (e.g., diet or environment) that influence gut microbiota could affect host immunity. Indeed, several reports including our current study showed the influence of antimicrobials on antibody titers in ND-vaccinated chickens. For example, it was found that chlortetracycline- or salinomycin-treated broiler chickens produced high antibody titers against ND compared with the control chickens (Munir et al., 2007; Murwani and Murtini, 2009). In contrast, the depressive effect of antimicrobials on antibody titers against ND was also reported (Khalifeh et al., 2009). Finally, no clear antibiotic effect on the antibody levels in broiler chickens following vaccination against ND and IBV was observed (Kwon et al., 2008). The discordant results on antimicrobial-induced humoral responses between experiments may be due to several factors including the age of birds, types of chickens (fast growing meat-type chicks vs. slow growing layer-type chick), housing, or diet composition, which might influence the gut microbiota or development of immune systems.
In conclusion, the present study showed that the type of coccidiosis control programs commonly used in field poultry production modulate various systemic immune parameters such as acute phase protein concentration, ConA-stimulated splenocyte proliferation, splenocyte subpopulations, and antibody titers against IBV and ND in broiler chickens. The underlying immunomodulatory mechanism(s) seen in this study may be in part related to altered host immune capability and gut microbiota composition by various coccidiosis control programs. Thus, further studies are required to assess the effects of the coccidiosis control program on microbial community profile and host immune response and their association in disease pathogenesis in broiler chickens.
ACKNOWLEDGMENTS
This project was supported by ARS in-house project 1265- 32000-086-00D. We thank Marjorie Nichols and Stacy Torrey- son for their technical assistance. We also acknowledge Dr Daniel A. Bautista in the Lasher Poultry Diagnostic Laboratory, University of Delaware, for his help on the animal study.
This article was originally published in Korean Journal of Poultry Science. Vol.39, No.1, 17~25 (2012) http://dx.doi.org/10.5536/KJPS.2012.39.1.017.
