Evaluating a Salmonella Typhimurium, Eimeria maxima, and Clostridium perfringens coinfection necrotic enteritis model in broiler chickens: repeatability, dosing, and immune outcomes

Abstract

Coccidiosis and necrotic enteritis negatively impact poultry production, making challenge model repeatability important for evaluating mitigation strategies. Study objectives were: 1) evaluate Salmonella Typhimurium–Eimeria maxima–Clostridium perfringens necrotic enteritis coinfection model repeatability and 2) investigate E. maxima dose and early S. Typhimurium challenge on immune responses. Three trials using Ross 308 chicks assigned to 12 cages/trial (7 chicks/cage) in wire-floor brooders were completed. Trials 1 and 2 determined E. maxima dose for necrotic enteritis challenge in trial 3. Chicks in trials 1 and 2 were inoculated with 0 (control), 5, 15, or 25,000 sporulated E. maxima M6 oocysts on d 14 and jejunal lesion scores evaluated on d 20. In trial 3, chicks were assigned to control or necrotic enteritis challenge (42 chicks/group). Necrotic enteritis challenge chicks were inoculated with 1 × 10⁵ colony forming units (CFU) S. Typhimurium on d 1, 15,000 E. maxima oocysts on d 14, and 1 × 10⁸ CFU C. perfringens on d 19 and 20 with lesion scoring on d 22. Bird and feeder weights were recorded throughout each trial. Peripheral blood mononuclear cells (PBMC) were isolated from 1 chick/cage at baseline (all trials), 4 chicks/dose (trials 1 and 2) or 8 chicks/challenge (trial 3) 24 h post-inoculation (pi) with E. maxima for immunometabolic assays and immune cell profiling. Data were analyzed by mixed procedure (SAS 9.4) with challenge and timepoint fixed effects (P ≤ 0.05, trends 0.05 ≤ P ≤ 0.01). Inoculating chicks with 15,000 E. maxima oocysts increased d 14 to 20 FCR 79 points (trials 1 and 2; P = 0.009) vs. unchallenged chicks and achieved a target lesion score of 2. While C. perfringens challenge reduced trial 3 performance, average lesion scores were <1. Salmonella inoculation on d 1 tended to increase PBMC ATP production 41.6% 24 hpi with E. maxima vs. chicks challenged with E. maxima only (P = 0.06). These results provide insight for future model optimization and considerations regarding S. Typhimurium's effect on E. maxima immune response timelines.

INTRODUCTION

Globally, coccidiosis due to infection with Eimeria spp. contributes to significant economic losses at £10.4 billion in 2016 (approximately $12.5 billion USD) (Blake et al., 2020). Eimeria invades the intestinal epithelium where it cyclically replicates within host cells before rupturing the cell membrane to be released back into the intestinal lumen, contributing to significant intestinal damage (Blake and Tomley, 2014). In turn, release of cellular proteins into the intestinal lumen provides the ideal environment for existing Clostridium perfringens to rapidly proliferate and cause necrotic enteritis, a secondary infection with economic losses estimated around $3 billion USD annually (Shojadoost et al., 2012; Wade and Keyburn, 2015). Both diseases may present as subclinical infections characterized by reduced production performance or severe clinical disease associated with diarrhea, hemorrhagic/necrotic intestinal lesions, and increased mortality (Williams, 2005; Shojadoost et al., 2012). Antibiotic-free poultry production and the increased resistance to available anticoccidial drugs by Eimeria provides a challenge for controlling both coccidiosis and necrotic enteritis in global poultry flocks while simultaneously emphasizing the importance of researching novel mitigation strategies (Arabkhazaeli et al., 2013; Noack et al., 2019).

Repeatable modeling of both coccidiosis and necrotic enteritis is critical for the development and evaluation of new strategies to control these diseases. Coccidiosis models may implement high doses of attenuated vaccine-strain Eimeria or use virulent wild-type strains isolated from clinical cases to produce repeatable disease, but model variation complicates outcome comparisons between published studies (Soutter et al., 2020). In contrast, necrotic enteritis model repeatability is complicated by C. perfringens presence in the environment and gastrointestinal tract of nondiseased poultry (Hibberd et al., 2011). This means inoculation with C. perfringens alone is not sufficient to induce necrotic enteritis and model success may be impacted by multifactorial variables that include concurrent Eimeria challenge, dietary composition, administration route, timelines relative to concurrent Eimeria, and C. perfringens strain (Al-Sheikhly and Al-Saieg, 1980; Kaldhusdal and Hofshagen, 1992; Drew et al., 2004; Keyburn et al., 2008; Shojadoost et al., 2012).

One published model has accomplished repeatable, severe necrotic enteritis challenge associated with high mortality (∼40%) by including inoculation with Salmonella Typhimurium on d 1 before inoculating birds with Eimeria maxima (Guelph strain or M6 strain) at 14 to 18 d of age followed by consecutive C. perfringens inoculation over 2 d around strain-specific timepoints associated with peak E. maxima lesion score (Shivaramaiah et al., 2011). The lack of reliance on nutritional components to increase disease severity makes this model ideal for evaluating dietary mitigation strategies (Hernandez-Patlan et al., 2019a,b; Coles et al., 2021). These aspects make the S. Typhimurium–E. maxima–C. perfringens coinfection model an ideal candidate for research programs at Iowa State University; however, differences in available facilities and equipment at a new institution requires troubleshooting and optimization before implementation.

Model optimization is a relatively straightforward process that relies on previously established benchmarks and involves the use of multiple small replicate studies to confirm successful and repeatable implementation. At the same time, these small replicate studies provide an opportunity to address additional model-related research questions without subjecting additional experimental animal groups to disease challenge. One aspect of utilizing the S. Typhimurium–E. maxima–C. perfringens coinfection model is including a small subset of birds for the sole purpose of identifying a target E. maxima dose. Within the recommendations of the model, the E. maxima challenge dose should produce subclinical disease; however, E. maxima virulence can be impacted by factors like storage solution and duration, and the proper dose needs to be evaluated before each experiment (Millard and Long, 1974; Gong et al., 2021). Our previous research has evaluated immunometabolic responses to various pathogen challenges, including attenuated vaccine-strain Eimeria spp., to identify potential shifts in glycolytic immunometabolism that could indicate inflammatory immune processes (Rambold and Pearce, 2018; Fries-Craft et al., 2021; Meyer et al., 2022). While previous research has found that increasing Eimeria challenge doses contribute to linear reductions in performance, intestinal integrity, and antimicrobial production (Casterlow et al., 2011; Teng et al., 2021), the peripheral immunometabolic shifts in response to E. maxima have not been characterized.

Previous publications using the S. Typhimurium–E. maxima–C. perfringens necrotic enteritis coinfection model have focused on necrotic enteritis outcomes at peak infection defined around 3 dpi with C. perfringens (Hernandez-Patlan et al., 2019a,b). While this is an effective approach in evaluating feed additive potential for reducing general disease severity, it does not provide insight into potential effects on immune response timelines. Similarly, previous work with this model has not specifically evaluated host immunity or included an E. maxima-only group to evaluate potential S. Typhimurium-related changes in the immune response to the primary challenge pathogen before inoculation with secondary C. perfringens. As a result, it is not clear how early S. Typhimurium inoculation, which is implemented to induce localized immune suppression and increase downstream necrotic enteritis severity, may impact coccidiosis development or systemic immunometabolism (Shivaramaiah et al., 2011). To address these research questions, objectives in this study were 2-fold: 1) successfully implement and optimize the S. Typhimurium–E. maxima–C. perfringens necrotic enteritis coinfection model for future use and 2) evaluate factors such as challenge dose and early inoculation with S. Typhimurium on immune responses within the first 24 h postinoculation (pi) with wild-type E. maxima M6.

MATERIALS AND METHODS

Birds and Housing

The Iowa State University Institutional Animal Care and Use Committee approved all animal protocols used in this study. Three replicate trials of 84 one-day-old Ross 308 chicks were placed in twelve 35.6 cm × 99 cm raised wire-floor cages/replicate (7 chicks/cage) within brooders and allowed ad libitum access to Purina Start & Grow nonmedicated crumble diet and water from an attached trough (Purina Mills, LLC, Gray Summit, MO). Placement dates for the 3 trials were offset by 7 d and each group was placed in a separate brooder unit within the same barn (Petersime Model 2SD20RE; Gettysburg, OH). Temperature in the room ranged from 26°C to 37°C with 55 to 68% humidity and fans were used to improve air circulation when temperatures exceeded 32°C. Key timepoints for challenge, sample collection and lesion scoring were based on placement day for each trial (Figure 1). Trials 1 and 2 were designated to determine target E. maxima M6 doses to be used for the necrotic enteritis model evaluated in Trial 3 to account for variability in challenge outcomes due to oocyst storage or batch. In all 3 trials, baseline blood samples were collected prior to E. maxima challenge on d 14 and additional blood samples collected 24 hpi to evaluate peripheral blood mononuclear cell (PBMC) immunometabolism. Lesion scores were collected 6 dpi with E. maxima in trials 1 and 2, and 3 dpi with secondary C. perfringens in trial 3 (Figure 1).

Figure 1.

Performance measurement and sample collection timelines for 3 trials using Ross 308 broilers. Trials 1 and 2 were used to determine optimal Eimeria maxima challenge dose based on body weight gain reductions and average d 20 lesion score. Trial 3 was conducted to evaluate repeatability of a Salmonella Typhimurium–Eimeria maxima–Clostridium perfringens necrotic enteritis coinfection model.

To minimize potential cross-contamination, cages designated for different challenge groups were assigned within the brooder unit such that unchallenged chicks were placed in the uppermost tiers and those receiving the highest challenge dose or assigned to the necrotic enteritis model were in the lowest cages (3 cages/tier). Daily monitoring was completed on unchallenged brooders prior to handling challenged birds and floors were disinfected daily. Bird and feeder weights were recorded at hatch, prior to inoculation, and at the time of blood sampling or lesion scoring to monitor body weight (BW), BW gain (BWG), average daily gain (ADG), and average daily feed intake (ADFI). All pathogens used in this study were provided by the JKS Poultry Health Laboratory at the University of Arkansas (Fayetteville, AR).

E. Maxima Dose Titration Inoculum Preparation

At 14 d of age, birds in trials 1 and 2 were challenged with 5, 15, or 25,000 sporulated E. maxima M6 oocysts originally isolated from a commercial broiler flock in Florida, USA (21 chicks/dose; Martin et al., 1997). To determine oocyst concentration, stock E. maxima stored in 2.5% potassium dichromate was diluted 100× in 1.18 specific gravity zinc sulfate solution (Ricca Chemical Company, Arlington, TX) and 600 μL pipetted into both reservoirs of a McMaster chamber. After letting the McMaster chamber sit for 5 min at room temperature, sporulated oocysts within each chamber grid were enumerated to determine oocysts/mL in the stock solution. The volume of stock solution needed to prepare each dose was transferred into a 50 mL centrifuge tube and washed 3 times in PBS by centrifugation (1,250 × g, 10 min, 20°C) to remove residual potassium dichromate. After the final wash, oocysts were resuspended in PBS to the volume needed to administer 1 mL to each chick at the target oocyst dose and enumerated by McMaster chamber to confirm the final concentration for challenge. Chicks were inoculated by oral gavage (challenged group) or PBS (control, unchallenged group). At 6 dpi, 15 chicks/dose were euthanized and the distal jejunum was lesion scored using gross lesion scoring criteria published by (Johnson and Reid, 1970) where 0 denoted no observable lesions and 4 indicated severe disease. The E. maxima dose associated with a 25% reduction in BWG and average lesion score of 2 was selected for birds in group 3.

Necrotic Enteritis Modeling Birds and Housing

To attempt the necrotic enteritis model, chicks in trial 3 were assigned to unchallenged and challenged conditions. On d 1, chicks designated for challenge were orally gavaged with 1 × 10⁵ colony forming units (CFU) of an S. Typhimurium poultry isolate (Shivaramaiah et al., 2011). The day before inoculation, tryptone soy broth (TSB) was inoculated with frozen S. Typhimurium stock (−80°C), incubated aerobically at 37°C, and passaged every 8 h to achieve log phase growth (Shivaramaiah et al., 2011). After 24 h, the most recently passaged culture was washed 3 times in sterile PBS by centrifugation (1,800 × g, 15 min, 4°C). The cell pellet was resuspended in 1 mL of sterile PBS and the optical density of the stock solution at 600 nm (OD₆₀₀) was measured by spectrophotometer to estimate stock CFU density. Salmonella Typhimurium stock was diluted to 5 × 10⁵ CFU/mL and 0.2 mL was administered to each chick. The final inoculum was serially diluted and plated on tryptone soy agar (TSA) and incubated aerobically at 37°C overnight to confirm S. Typhimurium dosage.

On d 14, birds in trial 3 that had been inoculated with S. Typhimurium at d 1 were challenged with 15,000 sporulated oocysts of E. maxima prepared as previously described. At timepoints corresponding to 5 and 6 dpi with E. maxima (d 19 and 20), birds were challenged with 1 × 10⁸ CFU Clostridium perfringens isolated from a 2004 poultry necrotic enteritis case (Ausland et al., 2020). To prepare the inoculum, TSB supplemented with 0.25 g/L sodium thioglycolate was inoculated with C. perfringens from frozen stock (−80°C) and incubated at 37°C under anaerobic conditions using a BD GasPak EZ chamber with anaerobe sachets (Becton, Dickinson, and Company, Franklin Lake, NJ). The culture was passaged every 24 h for 3 d and washed 3 times in fresh thioglycolate media by centrifugation at 1,800 × g for 15 min at 4°C. Washed culture was resuspended in PBS, enumerated by hemocytometer, and diluted to 1 × 10⁸ CFU/mL. An aliquot of inoculum was diluted and plated on phenylethyl alcohol agar with 5% sheep blood to confirm challenge dose prior to transporting inoculum to the farm. At 3 dpi with C. perfringens, 15 chicks/group were euthanized for lesion scoring using criteria published by Prescott et al. (1978) where 0 indicated no observable lesions and 4 denoted generalized necrosis and hemorrhage typical for field cases.

Real-Time Immunometabolic Assays

Agilent real-time ATP and glycolytic rate assays (Santa Clara, CA) were conducted on PBMC isolated on d 14 (baseline) and 24 hpi with E. maxima to evaluate early immunometabolic responses. Baseline samples were collected from 1 chick/cage in all 3 trials (12 birds total) and 24 hpi samples were collected from 4 chicks/dose (trials 1 and 2) or 8 chicks/challenge (trial 3). Chicks were anesthetized by CO₂ to collect blood by cardiac puncture into heparin-coated blood tubes before euthanasia by cervical dislocation. Blood was diluted 1:1 in sterile PBS and layered onto a Histopaque 1077 and 1119 density gradient (Sigma Aldrich, St. Louis, MO) for separation by centrifugation at 650 × g for 35 min (low acceleration and no brakes). Cells collected at the density gradient interface were harvested, washed twice in sterile PBS, resuspended in assay media, and enumerated by hemocytometer. For immunometabolic assays, PBMC from each chick were plated in triplicate on a 96-well cell culture plate (200,000 cells/well) and the remaining cells were frozen at −80°C in RPMI supplemented with 42.5% heat-inactivated chicken serum and 7.5% DMSO.

Media and reagents for the real-time ATP and glycolytic rate assays were prepared according to manufacturer's instructions and analyzed by Seahorse XFe96 Analyzer at 40°C. To evaluate ATP production profiles in plated PBMC, the real-time ATP rate assay injects 15 μM oligomycin followed by 5 μM rotenone + antimycin A (Rot/AA) into the assay media. The use of these mitochondrial inhibitors allows the assay to back-calculate relative contributions of mitochondrial respiration and glycolysis to overall ATP production rate. The glycolytic rate assay utilizes 5 μM Rot/AA followed by glycolysis-inhibiting 500 mM 2-deoxy-D-glucose (2-DG) to evaluate shifts in glycolytic activity following mitochondrial inhibition (compensatory glycolysis) and residual glycolytic activity following inhibition by 2-DG (post-2DG acidification). Measurements recorded during the ATP rate assay are converted to ATP production rate (pmol ATP/min) while those recorded during the glycolytic rate assay are converted to proton efflux rate (PER; pmol/min) by Wave software (Agilent, version 2.6.1) and exported for data analysis.

Multicolor Flow Cytometry

To evaluate PBMC immune cell profiles for all 3 trials, frozen PBMC isolated at d 14 (baseline) and 24 hpi with E. maxima were thawed, washed in PBS (400 × g, 10 min, 18°C), and enumerated by hemocytometer. Cells from each sample were equally distributed into polystyrene tubes for extracellular staining. All antibodies were sourced from Southern Biotech (Birmingham, AL) and the staining panel included innate and T cell markers: monocyte/macrophage biotin (clone KUL01; mouse IgG₁κ), cluster of differentiation (CD) 1.1 FITC (clone CB3, mouse IgG₁κ), CD3 Alexa Fluor 700 (clone CT-3; mouse IgG₁κ), CD4 PE-Cy7 (clone CT-4; mouse IgG₁κ), CD8α Pacific Blue (clone CT-8; mouse IgG₁κ), and T cell receptor (TCR) γδ PE (clone TCR-1; mouse IgG₁κ). Each sample included its own fluorescence-minus-one with associated isotype controls to account for nonspecific binding. Antibodies were diluted in PBS, added to designated flow tubes, and incubated for 30 min in the dark at 4°C. After initial staining, cells were washed in PBS and biotin-conjugated antibodies were stained with secondary streptavidin-conjugated Brilliant Violet 785 (BioLegend, San Diego, CA) diluted in PBS and incubated in the dark for 30 min at 4°C. Cells were then washed and resuspended in PBS before analysis by BD FACSCanto cytometer (BD Biosciences, San Jose, CA). Cell phenotype data were analyzed by FlowJo software (BD Biosciences, version 10.5.0). Gating strategies for each cell population are presented in Supplemental Figure 1.

Statistical Analysis

When analyzing performance and lesion scores, outcomes from trials 1 and 2 were not statistically different from each other with the exception of initial BW and were pooled for analysis using initial BW as a covariate, whereas trial 3 was analyzed separately. The following model was used to evaluate performance outcomes:

$$y_{ij}=\mu +C_i+iBW+e_{ij}$$

where y_ij is the variable of interest, μ is the overall mean, C_i is the effect of E. maxima challenge dose (trials 1 + 2) or challenge group (unchallenged vs. necrotic enteritis; trial 3) at the ith level, iBW is the initial BW covariate, and e_ij is the associated error.

As immunometabolic and flow cytometry outcomes were measured at baseline and 24 hpi with E. maxima, outcomes from these trials were pooled differently to address 2 different research questions: 1) are these outcomes affected by increasing E. maxima dose, and 2) how does d 0 S. Typhimurium inoculation affect responses to E. maxima? To accomplish this, outcomes from trials 1 and 2 were analyzed together to address question 1. To address question 2, unchallenged chicks and those dosed with 15,000 sporulated oocysts from trials 1 and 2 were assessed alongside measurements from trial 3. No significant differences between trials were observed for immunometabolic data and these outcomes were pooled and evaluated using the following model:

$$y_{ijk}=\mu +C_i+T_j+{\left(C\times T\right)}_{ij}+e_{ijk}$$

In this model, variables y_ijk, μ, C_i, and e_ijk are the same as described above while T_j denotes the main effect of timepoint (baseline or 24 hpi) at the jth level and (C × T)_ij represents the interaction of dose/challenge group and timepoint.

For flow cytometry data, differences between trials were observed and cell population data were analyzed using the following model:

$$y_{ijkl}=\mu +C_i+T_j+G_k+{\left(C\times T\right)}_{ij}+{\left(C\times G\right)}_{ik}+{\left(T\times G\right)}_{jk}+{\left(C\times T\times G\right)}_{ijk}+e_{ijkl}$$

Variables in this model are similar to those used to analyze immunometabolic data with the addition of G_k to represent the main effect of trial at the kth level (1, 2, or 3) and relevant interaction effects including this new variable.

All reported models were analyzed using the mixed procedure in SAS 9.4 with Tukey's honest significant difference post hoc test to account for multiple comparisons. Orthogonal contrasts were also used whenever the effect of increasing E. maxima dose was evaluated (trials 1 and 2). Coefficients for linear, quadratic, and cubic contrasts were determined using proc IML for the dose effect (SAS 9.4). Statistical significance was denoted at P ≤ 0.05 and trends at 0.05 < P ≤ 0.10.

RESULTS

E. Maxima Titration Performance (Trials 1 and 2)

Performance results from trials 1 and 2 are presented in Table 1. While no performance differences were observed at baseline, a significant linear contrast was observed for BW, FI, and BWG (P ≤ 0.03), indicating that these measures increased when evaluating birds in the top vs. bottom tiers of the brooder unit (Table 1). From d 14 to 20, corresponding with the first 6 dpi, chicks administered 15,000 sporulated E. maxima oocysts had a 79-point less efficient FCR compared to their unchallenged counterparts (P = 0.009); however, only numeric efficiency reductions were observed in chicks dosed with 5,000 and 25,000 oocysts (Table 1). During this same timeframe, significant quadratic effects on BWG and FCR were observed (P ≤ 0.03) where the greatest decline in these measures was observed in chicks inoculated with 15,000 oocysts (Table 1). For the entire 20-day period, E. maxima dose did not significantly affect chick performance; however, a significant quadratic effect was observed on FCR with the greatest efficiency loss occurring in chicks challenged with 15,000 E. maxima oocysts (P = 0.01; Table 1). When examining average jejunal lesion scores on d 20, increasing E. maxima dose increased gross lesion scores stepwise by 0.6 to 0.7 points from 1.3 to 2.6 (P < 0.0001; Table 1).

Table 1.

Performance and lesion scoring outcomes in Ross 308 chicks challenged with increasing doses of sporulated Eimeria maxima M6 oocysts on d 14¹.

	E. maxima challenge dose					P values	Orthogonal contrasts2
Measure	Control	5,000	15,000	25,000	SEM	Dose3	Linear	Quadratic	Cubic
d 0 BW4, g	41.24	40.69	41.40	40.33	1.97	0.98	0.83	0.85	0.75
d 14 BW, g	276.40	284.20	296.10	309.20	10.49	0.17	0.03	0.98	0.93
d 0–14 BWG, g	235.50	243.30	255.20	268.30	10.49	0.17	0.03	0.98	0.93
d 0–14 FI, g	373.90	389.10	417.00	415.60	12.49	0.06	0.02	0.27	0.76
d 0–14 FCR	1.59	1.61	1.63	1.56	0.04	0.67	0.63	0.28	0.76
d 20 BW, g	452.90	437.30	435.40	469.90	18.79	0.55	0.47	0.21	0.98
d 14–20 BWG, g	176.53	153.12	139.36	160.64	11.12	0.16	0.33	0.04	0.90
d 14–20 FI, g	397.70	402.10	408.50	409.90	17.83	0.96	0.60	0.87	0.99
d 14–20 FCR5	2.33b	2.64ab	3.12a	2.64ab	0.14	0.009	0.07	0.003	0.42
d 0–20 BWG, g	412.01	396.40	394.51	428.96	18.79	0.55	0.47	0.21	0.98
d 0–20 FI, g	771.63	791.22	825.50	825.57	27.49	0.44	0.14	0.54	0.88
d 0–20 FCR	1.89	2.00	2.11	1.93	0.06	0.06	0.56	0.01	0.63
d 20 Lesion Score6	0.0d	1.3c	2.0b	2.6a	0.13	<0.0001	<0.0001	0.0001	0.002

¹Data represent the mean pooled outcomes from 2 replicate trials (n = 6 cages/E. maxima dose) on a per bird basis. To determine if pooling data were appropriate, an initial statistical model including the trial effect was used (data not shown). Only d 0 BW was significantly different between trials and was used as a covariate in the final statistical model.

²Linear contrast: −0.6 −0.3 0.2 0.7; Quadratic contrast: 0.5 −0.3 −0.7 0.5; Cubic contrast: −0.4 0.8 −0.5 0.2. Coefficients for each dose were determined using the IML procedure and presented here to the nearest tenth (SAS 9.4, SAS Institute, Cary, NC).

³Cages assigned to each dose were placed in descending order within the brooder unit to minimize cross-contamination (e.g., chicks in cages assigned to the control group were placed in the top tiers while those receiving 25,000 oocysts were at the lowest tier). As such, effects associated with dose prior to d 14 inoculation are associated with position within the brooder unit.

⁴Abbreviations: BW, body weight; BWG, body weight gain; FCR, feed conversion rate; FI, feed intake.

⁵Means with different superscript letters (a–d) are significantly different (P ≤ 0.05).

⁶Lesion scores were assigned based on a system published by Johnson and Reid (1970). Scores of 0 indicate no observable lesions.

Necrotic Enteritis Modeling Performance (Trial 3)

No performance differences between chicks assigned to the control or necrotic enteritis groups were observed within the first 14 d of the study, nor within the first 5 dpi with E. maxima (d 14–19; Table 2). Outcomes from d 19 to 20 represent performance changes between the first and second inoculation with C. perfringens. Between inoculations, BWG and FI were reduced 76.6 and 17.9% in control vs. challenged chicks, respectively (P ≤ 0.05; Table 2). No differences in d 19 to 20 FCR were observed; however, FCR calculated for a single day was highly variable (SEM = 1.52; Table 2). This high variability may be explained by some cages in the necrotic enteritis group having negative BWG, resulting in a negative FCR and lowering the overall treatment mean.

Table 2.

Performance and lesion scoring outcomes in Ross 308 chicks to evaluate a necrotic enteritis (NE) model¹.

	Challenge dose
Measure	Control	NE model	SEM	P value
d 0 BW2, g	35.33	34.86	0.4	0.45
d 14 BW, g	233.60	232.20	9.3	0.92
d 0–14 BWG, g	198.50	197.10	9.3	0.92
d 0–14 FI, g	353.70	355.40	10.2	0.91
d 0–14 FCR	1.78	1.83	0.07	0.70
d 19 BW, g	390.40	381.10	15.0	0.68
d 14–19 BWG, g	156.80	148.90	8.20	0.52
d 14–19 FI, g	351.60	370.90	20.0	0.52
d 14–20 FCR	2.25	2.52	0.15	0.22
d 20 BW, g	415.50	386.90	16.51	0.26
d 19–20 BWG3, g	25.04a	5.85b	4.50	0.02
d 19–20 FI, g	56.86a	46.70b	3.15	0.05
d 19–20 FCR	2.09	2.77	1.52	0.77
d 26 BW, g	659.70	605.40	23.70	0.14
d 19–26 BWG, g	269.30a	224.30b	12.78	0.04
d 19–26 FI, g	453.30	47.00	15.35	0.47
d 19–26 FCR	1.71b	2.09a	0.06	0.002
d 0–26 BWG, g	624.61	570.31	23.70	0.14
d 0–26 FI, g	1158.62	1196.30	28.41	0.38
d 0–26 FCR	1.83b	2.11a	0.07	0.04
d 22 Lesion score4	0.00b	0.80a	0.19	0.01

¹Ross 308 chicks were assigned to either challenge group at placement. Those designated for necrotic enteritis challenge were orally gavaged with 1 × 10⁵ colony forming units (CFU) of Salmonella Typhimurium on d 1, 15,000 sporulated oocysts on d 14, and 1 × 10⁸ CFU of Clostridium perfringens on d 19 and 20. Data represent the mean (n = 6 cages/challenge group) on a per bird basis.

²Abbreviations: BW, body weight; BWG, body weight gain; FCR, feed conversion rate; FI, feed intake.

³Means with different superscript letters (a, b) are significantly different (P ≤ 0.05).

⁴Lesion scores were assigned based on a system published by Johnson and Reid (1970). Scores of 0 indicate no observable lesions.

From d 19 to 26, corresponding to the first 7 dpi with secondary C. perfringens, challenged chicks had 16.7% reduction in BWG compared to their unchallenged counterparts (P = 0.04) without significantly reducing FI, culminating in a 38-point less efficient FCR (P = 0.002). When evaluating performance for the entire 26-day period, chicks in the necrotic enteritis challenge group did not have significantly different BWG or FI, but numerical changes in these measures resulted in a 28-point less efficient overall FCR in challenged vs. unchallenged chicks (P = 0.04; Table 2). Average gross lesion scores taken at 3 dpi with C. perfringens (d 22) were significantly different between groups (P = 0.01). This difference was expected since unchallenged birds did not have observable lesions; however, the average 3 dpi lesion score in necrotic enteritis challenged chicks was 0.8 and lower than expected.

Early PBMC ATP Production Responses to E. Maxima

Outcomes in trials 1 and 2 were pooled to identify potential early immunometabolic responses to increasing E. maxima doses. The timepoint main effect resulted in 19.5% greater ATP production from glycolysis (P = 0.01) in PBMC isolated at baseline (d 14) vs. 24 hpi with E. maxima, while baseline ATP from mitochondrial respiration tended to be 20.6% greater (P = 0.06), resulting in 19.8% increased overall baseline ATP production (P = 0.02) compared to measurements taken at 24 hpi (Figure 2A). Neither E. maxima dose nor the interaction with timepoint affected total ATP production or the percent contributions of glycolysis and mitochondrial respiration to overall ATP production profiles, nor were any significant orthogonal contrasts observed (Figure 2B).

Figure 2.

ATP production profiles in peripheral blood mononuclear cells (PBMC) isolated from Ross 308 chicks at baseline (d 14) and 24 h postinoculation (pi) with Eimeria maxima M6 measured using the Agilent real-time ATP rate assay and Seahorse XFe96 analyzer (Santa Clara, CA). In panel (A), data represent the timepoint main effect on the pooled mean from trials 1 and 2 ± SEM evaluating responses 24 h after oral gavage with 5, 15, or 25,000 sporulated E. maxima M6 oocysts, while panel (B) shows the dose × timepoint interaction from the same data (6 chicks/dose). Panel (C) represents the challenge main effect on the pooled mean from all 3 trials used to evaluate changes in early responses to challenge with 15,000 sporulated E. maxima M6 in chicks ± 1 × 10⁵ colony forming units (CFU) of Salmonella Typhimurium at 1 d of age compared to an unchallenged control. Data in panel (D) represent the challenge × timepoint interaction from this dataset as the mean ± SEM of 9 and 12 chicks in the control group at baseline and 24 hpi, respectively. Data for challenged groups in panel (D) represent the mean ± SEM of 6 and 8 chicks/challenge group at baseline and 24 hpi, respectively. Data were pooled when no significant trial main effect was observed (P > 0.05). Different letter labels within the same-colored bar indicate a significant difference between ATP production from glycolytic or oxidative metabolism while brackets with different letter comparisons denote significant differences in total ATP production rate (P ≤ 0.05). Comparisons marked with # denote a statistical tendency (0.05 < P < 0.10).

To evaluate potential effects of d 1 S. Typhimurium on immunometabolic responses within the first 24 hpi with E. maxima, outcomes from unchallenged birds (all trials) and those receiving 15,000 oocysts (trials 1 and 2) were compared to outcomes in challenged chicks from trial 3 (d 1 S. Typhimurium + 15,000 oocysts E. maxima). The main effect of challenge increased ATP production from glycolysis 25.2 to 25.6% in chicks inoculated with S. Typhimurium on d 1 compared to unchallenged chicks and those that received only E. maxima (P = 0.003) and tended to increase ATP from mitochondrial respiration 21.7 to 27.0% (P = 0.08; Figure 2C). As a result, the challenge main effect significantly increased total ATP production in PBMC isolated from chicks that received d 1 S. Typhimurium 24.1 to 26.1% vs. unchallenged and E. maxima-only groups, respectively (P = 0.009; Figure 2C). Chicks inoculated with d 1 S. Typhimurium tended to have 34.6 and 41.6% greater total ATP production 24 hpi with E. maxima than unchallenged chicks and those administered E. maxima only (P = 0.06). Underlying ATP from glycolysis and mitochondrial respiration in chicks that received d 1 S. Typhimurium tended to be 32.9 to 40.6% and 38.1 to 43.5% greater at 24 hpi with E. maxima, respectively, than unchallenged and E. maxima-only challenged chicks at the same time (P ≤ 0.08; Figure 2D). As a result, no differences in the percent contributions of glycolysis or mitochondrial respiration to overall ATP production were observed.

Glycolytic Rate Changes in PBMC After 24 HPI With E. Maxima

Supplemental Figure 2 illustrates the key measures being recorded during the glycolytic rate assay. The timepoint main effect was associated with 20.9 and 22.4% significantly increased basal glycolysis in PBMC isolated at baseline vs. 24 hpi (P = 0.04; Figure 3A). Compensatory glycolysis was also increased 22.4% in PBMC isolated at baseline vs. 24 hpi due to the timepoint main effect (P = 0.04; Figure 3A). In baseline measures, PBMC isolated from chicks designated for challenge with 5,000 E. maxima oocysts showed distinct separation in glycolytic rate measurements compared to other groups; however, the dose and timepoint interactions were not statistically significant (Figure 3B).

Figure 3.

Glycolytic rate outputs in peripheral blood mononuclear cells (PBMC) isolated from Ross 308 chicks at baseline (d 14) and 24 h postinoculation (pi) with Eimeria maxima M6 measured using the Agilent glycolytic rate assay on the Seahorse XFe96 analyzer (Santa Clara, CA). Panel (A) represents the timepoint main effect on the pooled mean ± SEM from 2 replicate studies evaluating responses 24 h after oral gavage with 5, 15, or 25,000 sporulated E. maxima M6 oocysts. Brackets and letters at timepoints <20 min indicate significantly increased basal glycolysis in PBMC isolated at baseline vs. 24 hpi. Outputs are then split out to represent measurements taken at (B) baseline (6 birds/dose) and (C) 24 hpi (8 birds/dose). In panel (B), brackets at timepoints <20 min denote significantly increased basal glycolysis in PBMC isolated from birds inoculated with S. Typhimurium at d 1 vs. control or E. maxima only. Panel (D) represents the main effect of challenge across pooled trial outcomes to evaluate changes in early responses to E. maxima in chicks ± 1 × 10⁵ colony forming units (CFU) of Salmonella Typhimurium at 1 d of age before inoculating with 15,000 oocysts of E. maxima compared to an unchallenged control. Data in panels (E) represent the mean ± SEM of 9 chicks in the control and 6 chicks/challenge group at baseline and (F) 12 chicks in the control and 8 chicks/challenge group at 24 hpi. Data were pooled when no significant trial main effect was observed (P > 0.05). Brackets with different superscript letters denote significant differences between regions in the assay output (P ≤ 0.05).

When evaluating potential S. Typhimurium effects during E. maxima challenge, the challenge group main effect significantly affected glycolytic rates. Chicks inoculated with S. Typhimurium on d 1 had 29.7 to 30.0% greater basal glycolysis (P = 0.002) and 27.2 to 30.3% greater ability to mobilize glycolysis to meet energy demands after mitochondrial inhibition (compensatory glycolysis; P = 0.0009) compared to unchallenged chicks and those that received E. maxima only (Figure 3D). The challenge group main effect also contributed to a 38.8% greater ability for PBMC isolated from chicks inoculated with S. Typhimurium at d 1 to continue glycolytic metabolism after inhibition (post-2DG acidification) compared to their unchallenged counterparts (P = 0.04; Figure 3D). No other significant differences or trends were associated with timepoint or the challenge group and timepoint interaction (Figure 3E and F).

PBMC Immune Populations

Flow cytometry on the remaining PBMC not plated for immunometabolic assays was conducted to identify any potential immune cell shifts that could be associated with altered metabolic responses. The panel in this study included monocytes/macrophages associated with innate immune cell responses and CD1.1⁺ lipid antigen-presenting cells that would be most likely to respond within the first 24 hpi (Qureshi, 1998; Dvir et al., 2010). While not expected to respond to E. maxima challenge within the first 24 hpi, T cells associated with a later adaptive response and various subtypes were analyzed to determine if increasing E. maxima dose or early S. Typhimurium inoculation affected these immune populations. Of the T cell subtypes analyzed, CD3⁺CD4⁺ helper T (T_H) cells activate other adaptive immune cell types such as B cells for antibody production, CD3⁺CD8α⁺ cytotoxic cells (T_C) directly target infected cells, and CD3⁺TCRγδ⁺ (γδ) T cells represent the dominant TCR phenotype in chickens with proposed functions that range from immune regulation to cytotoxic clearance of intracellular bacteria (Mombaerts et al., 1993; Masson and Belz, 2010; Vantourout and Hayday, 2013; Golubovskaya and Wu, 2016).

Compared to performance and immunometabolic outcomes in this study, the trial main effect on immune cell profiles was statistically significant and required a more complex statistical model for analysis. When evaluating the effect of E. maxima dose on monocyte/macrophage⁺, CD1.1⁺, and CD3⁺ T cells, the trial main effect resulted in 44.4 to 48.4% greater populations of these cells detected in PBMC between trials 2 and 1 (P < 0.0001; Figure 4A–C). Within CD1.1⁺ and CD3⁺ cell populations, baseline samples taken from chicks in trial 2 had 60.1 and 57.4% greater CD1.1⁺ and CD3⁺ cells, respectively, compared to 24 hpi samples within the same trial (dose × timepoint, P < 0.0001; Figure 4B and C). Within underlying T_H subpopulations, populations within trial 2 were 28.8% lower at 24 hpi vs. baseline (P < 0.0001; Figure 4D). For other T cell subtypes measured, PBMC isolated from chicks in trial 2 had 30.5 greater T_C cells (P < 0.0001) and 14.2% fewer γδ T cells (P = 0.03) than chicks in trial 1 (Figure 4E and F). Effects relating to challenge dose did not meet criteria for significance or trend designation, nor were any significant orthogonal contrasts observed, indicating that immune cell populations over a 24 h period were not affected by increasing E. maxima dose.

Figure 4.

Flow cytometric analysis of peripheral blood mononuclear cells (PBMC) isolated from Ross 308 chicks at baseline (d 14) and 24 h postinoculation (pi) with 5, 15, or 25,000 sporulated Eimeria maxima M6 oocysts. Data represent the mean ± SEM for 2 replicate trials (n = 3 chicks/group at baseline; n = 4 chicks/group at 24 hpi). Cell populations are presented as cells positive for (A) monocyte/macrophage, (B) CD1.1, and (C) CD3 within the live cell gate and correspond to innate immune, lipid-antigen, and adaptive immune cells, respectively. (D) CD3⁺CD4⁺ helper, (E) CD3⁺CD8α⁺ cytotoxic, and (F) CD3⁺TCRγδ⁺ T cell subtypes as percentages of cells positive for their respective markers within the overall CD3⁺ cell gate. Comparisons denoted with * indicate a significant trial × timepoint interaction, P ≤ 0.05.

Due to the sample pooling strategy to evaluate the effects of d 1 S. Typhimurium inoculation on early E. maxima responses, the group effect was also found to be a significant contributor to observed immune cell variation. When measures from all 3 trials were visualized, patterns in cell marker detectability in trial 3 were visually intermediate to responses observed in groups 1 and 2 (Figure 5). As such, it was determined that the previously described differences between trials 1 and 2 would obfuscate interpretation of trial 3 data. For simplicity, immune cell populations for trial 3 were analyzed independently from trial 1 and 2 and are presented in Figure 6.

Figure 5.

Flow cytometric analysis of peripheral blood mononuclear cells (PBMC) isolated from Ross 308 chicks at baseline (d 14) and 24 h postinoculation (pi) with 15,000 sporulated Eimeria maxima M6 oocysts ± inoculation with 1 × 10⁵ colony forming units (CFU) of Salmonella Typhimurium at 1 d of age and an additional unchallenged control group. Data represent the mean ± SEM of relevant data from 3 replicate trials. In trials 1 and 2, n = 3 chicks/group at baseline and 4 chicks/group at 24 hpi. In trial 3, n = 6 chicks/group at baseline and 8 chicks/group at 24 hpi. Cell populations are presented as cells positive for (A) monocyte/macrophage, (B) CD1.1, and (C) CD3 within the live cell gate and correspond to innate immune, lipid-antigen, and adaptive immune cells, respectively. (D) CD3⁺CD4⁺ helper, (E) CD3⁺CD8α⁺ cytotoxic, and (F) CD3⁺TCRγδ⁺ T cell subtypes as percentages of cells positive for their respective markers within the overall CD3⁺ cell gate.

Figure 6.

Flow cytometric analysis of peripheral blood mononuclear cells (PBMC) isolated from Ross 308 chicks at baseline (d 14) and 24 h postinoculation (pi) with 15,000 sporulated Eimeria maxima M6 oocysts. Chicks within the E. maxima-challenged group were inoculated with 1 × 10⁵ colony forming units (CFU) of Salmonella Typhimurium at 1 d of age as part of testing a necrotic enteritis model published by (Shivaramaiah et al., 2011). Data demonstrate the mean percentage of immune cells positive for their respective marker ± SEM among chicks in trial 3 only (n = 6 birds/group at baseline; n = 8 chicks/group at 24 hpi). Cell populations are presented as cells positive for (A) monocyte/macrophage, (B) CD1.1, and (C) CD3 within the live cell gate and correspond to innate immune, lipid-antigen, and adaptive immune cells, respectively. (D) CD3⁺CD4⁺ helper, (E) CD3⁺CD8α⁺ cytotoxic, and (F) CD3⁺TCRγδ⁺ T cell subtypes as percentages of cells positive for their respective markers within the overall CD3⁺ cell gate. Bars labeled with different letters are significantly different (P ≤ 0.05).

When analyzed this way, the timepoint main effect had the most significant impact on peripheral blood immune cell profiles. Baseline monocyte/macrophages, CD1.1⁺, CD3⁺, and CD3⁺TCRγδ⁺ cells were 42.0, 36.7, 32.5, and 29.4% greater than populations at 24 hpi, respectively (P ≤ 0.01; Figure 6A–C). While overall T cell populations were impacted more by timepoint, underlying T_H cells at 24 hpi in unchallenged chicks were increased 30.7% relative to their corresponding baseline (P = 0.0007) but no differences in baseline and 24 hpi T_H populations were observed in challenged chicks (Figure 6D). In contrast, CD3⁺CD8α⁺ T cells in chicks inoculated with d 1 S. Typhimurium and challenged with E. maxima were 52.5% increased at 24 hpi compared to their corresponding baseline (P = 007) while no differences were observed in this immune cell population among unchallenged chicks (Figure 6E).

DISCUSSION

Objectives in this study were divided into 2 parts: 1) implement and optimize an existing necrotic enteritis challenge model in a different facility, and 2) evaluate immune responses to early E. maxima challenge in 14-day-old chicks due to increasing challenge dose or d 1 inoculation with S. Typhimurium. Within the first objective, trials 1 and 2 were critical for identifying the appropriate E. maxima dose for the necrotic enteritis model attempted in trial 3. Bird numbers and benchmarks to identify target dose were based on published criteria used in conjunction with larger scale studies (Martin et al., 1997; Shivaramaiah et al., 2011). In these cases, the goal was to identify an appropriate dose rather than test for statistical significance and the low number of significant outcomes was likely due to low sample size. Oocyst viability declines with prolonged storage and E. maxima dose titration before each necrotic enteritis challenge study can ensure that oocyst viability is not a significant contributor to model variability (Millard and Long, 1974; Gong et al., 2021). Dose determination benchmarks were characterized as reduced performance without clinical signs such as severe intestinal damage and mortality, more specifically defined as an approximately 25% reduction in BWG leading up to when peak lesions are expected for E. maxima M6 (6 dpi) and an average 6 dpi lesion score of 2 (Latorre et al., 2018). While BWG reductions were not statistically significant, numeric 13.3, 21.1, and 9.0% reductions were observed in chicks challenged with 5, 15, and 25,000 oocysts, respectively. As such, the challenge dose that best met both the BWG reduction and lesion scoring recommendations was 15,000 oocysts and chosen for use in trial 3 (Table 1).

The minimal performance losses combined with highest observable lesion score in birds inoculated with 25,000 oocysts was unexpected, as previous work has demonstrated that increasing E. maxima doses also increased performance losses and lesion scores as markers of disease severity (Casterlow et al., 2011; Teng et al., 2021). This study was conducted from June to July when recorded temperatures within the brooders were as high as 37°C (99°F) with 50 to 60% humidity despite implementing fans. In broilers, heat stress from temperatures around 34°C resulted in significant production losses (Awad et al., 2020) and was likely a confounding variable during these replicate studies. When comparing performance observations in this study to Ross 308 performance objectives (Aviagen, 2022), d 20 BW and d 0 to 20 FI reported herein were approximately 52 and 69% less than expected, further supporting that heat stress negatively impacted chick performance in this study (Table 1).

Placing birds within the brooder based on assigned dose to control E. maxima cross-contamination accomplished the goal of keeping unchallenged birds from being inadvertently inoculated during routine maintenance (water trough filling or manure removal), but also positioned them at the uppermost tier of cages. As a result, birds at the lowest position in the brooder were in a more direct path of airflow from the supplemental fans and may have been less susceptible to the negative effects of heat stress. The significant linear effect of “dose” on increasing d 14 BW, BWG, and FI was likely associated with cage position and opposing heat stress exposure and E. maxima dose spectra may have also contributed to the unexpectedly minimal performance losses in chicks challenged with 25,000 oocysts despite this group having the highest observable lesion score (Table 1).

No differences in d 14 bird performance were observed in trial 3 (Table 2), likely because birds in this trial reached a point in their grow-out period where no other brooders were present, which potentially improved airflow and mitigated location-based performance differences. Chicks were challenged with C. perfringens at 5 and 6 dpi (d 19 and 20, respectively) with E. maxima M6 at timepoints that closely align with expected peak lesion scores (Johnson and Reid, 1970). Significant performance losses were observed between the 2 C. perfringens inoculations and birds in the necrotic enteritis challenge group displayed depressed BWG for the remainder of trial 3 (Table 2). While these performance results were expected, lesion scores recorded at 3 dpi with C. perfringens (d 21) were <1 and much lower than average necrotic enteritis lesion scores of 1.93 to 2.3 reported previously using the same model (Shivaramaiah et al., 2011; Latorre et al., 2018; Hernandez-Patlan et al., 2019a). This could be explained by the 3.4 mi distance from the lab where C. perfringens cultures were prepared and the farm where chicks were housed. Additionally, the protocols herein involved C. perfringens being washed and resuspended in PBS for inoculating chicks (Wilson et al., 2018). Under aerobic conditions, such as prolonged suspension in PBS, C. perfringens sporulates to promote survivability; however, vegetative cells are responsible for toxin production associated with poultry necrotic enteritis (Shen et al., 2019). While washed vs. unwashed C. perfringens did not differentially impact necrotic enteritis performance reductions in published work using a similar model, BW reductions without observable lesions were also previously reported (Wilson et al., 2018; Chasser et al., 2019). Washing overnight C. perfringens removes preformed toxins from the media and previous work has demonstrated that intestinal toxin production caused by inoculating with vegetative C. perfringens alone may be insufficient for reliably producing necrotic enteritis (Al-Sheikhly and Truscott, 1977; Shojadoost et al., 2012). In other facilities where spaces for inoculum preparation are in close proximity to research animals, washing overnight C. perfringens and administering in PBS may produce a successful model; however, in this study, the transit time coupled with washing overnight cultures increased oxygen exposure, likely causing the inoculum to be dominated by sporulated C. perfringens that would require additional time to germinate and start producing toxin which ultimately reduced necrotic enteritis severity. As such, future use of this model should emphasize protecting C. perfringens infectious viability during transport by not washing overnight cultures and inoculating birds with cultures diluted to the appropriate dose in thioglycolate media vs. PBS.

When evaluating immune responses within the second objective of this study, no differences in immunometabolism were linked to increasing E. maxima dose, whereas d 1 S. Typhimurium inoculation may have impacted responses within 24 hpi with E. maxima. Within the E. maxima dose evaluation trials, numerical differences in total ATP production were observed in chicks receiving 15 and 25,000 sporulated oocysts (Figure 2B). Similarly, glycolytic rate outputs in chicks inoculated with 15 and 25,000 oocysts were numerically depressed between baseline and 24 hpi (Figure 3A and B); however, these patterns were also observed in the unchallenged group. This suggests that observed changes were likely the result of physiological fluctuation rather than a specific response to E. maxima challenge and aligns with previous research where layer chicks inoculated with a 10X dose of vaccine-strain Eimeria spp. did not demonstrate PBMC immunometabolic shifts analyzed by the same real-time assays (Fries-Craft et al., 2023).

While E. maxima dose had minimal impacts on PBMC immunometabolism after 24 hpi, some trends were observed. When examining ATP production outcomes, baseline measures between the E. maxima-only and S. Typhimurium + E. maxima groups were numerically similar while 24 hpi measures tended to be increased in chicks inoculated with S. Typhimurium on d 1 compared to their counterparts challenged with a similar E. maxima dose (Figure 2D). While baseline glycolytic rate measures were also similar between the early S. Typhimurium and E. maxima-only groups, the numerically separate glycolytic rate output in chicks in the S. Typhimurium + E. maxima group at 24 hpi was due more to a depression in glycolytic rate from baseline to 24 hpi in chicks that received only E. maxima (Figure 3E and F). Collectively, this suggests that d 1 S. Typhimurium inoculation likely had a minimal effect on d 14 immunometabolic profiles, but potentially increased overall ATP production by PBMC within 24 hpi with E. maxima without impacting glycolytic rate.

Salmonella induces host tolerance in poultry by inducing an interleukin (IL)-10-mediated response within the intestine (Kogut and Arsenault, 2017). As an anti-inflammatory cytokine, IL-10 from S. Typhimurium colonization could be responsible for localized immune suppression that is important for repeatability in the necrotic enteritis model used in this study (Shivaramaiah et al., 2011). Immunometabolic responses within systemic peripheral blood would not be expected within 24 hpi with E. maxima as early innate responses are typically restricted to local tissues. Potential localized immune suppression via IL-10 within the intestinal environment by S. Typhimurium could have contributed to a systemic response detectable as increased ATP production by PBMC earlier than expected with E. maxima challenge (Figure 2B). In turn, this early reliance on systemic immune compartments during E. maxima challenge could negatively impact host defense against secondary C. perfringens; however, dedicated research into this phenomenon with timepoints extending beyond 24 hpi is needed.

In previous studies, flow cytometric immune cell profiling has been important for potentially identifying immune cell populations responsible for immunometabolic shifts (Fries-Craft et al., 2021). Immune profiling outcomes in this study were difficult to assess due to significant variation in outcomes between the different replicate trials. Notably, the 2- to 3-fold increased populations of CD1.1⁺ cells and CD3⁺ T cells in baseline samples collected in trials 2 and 3 (Figures 4 and 5). The increased detectability of these cell populations could be a potential heat stress response; however, populations in all trials were numerically similar at 24 hpi with unexpectedly high baseline readings representing the peak in most cases. More likely, this is due to an artifact from analysis as cell counts were numerically lower in baseline samples from trials 2 and 3 (data not shown). While decreased overall cell counts should not have theoretically impacted relative proportions of immune cell populations, it cannot be discredited at this time. Detected immune cell populations in chicks exposed to varying E. maxima doses or challenged with E. maxima after receiving d 1 S. Typhimurium were numerically within ranges determined by the minimum and maximum population observed in their unchallenged counterparts. As a result, no E. maxima-specific shifts in PBMC immune profiles can be reliably determined at this time and further research is needed.

Collectively, the results of this study demonstrate some issues faced when replicating challenge models developed at outside institutions with different infrastructure and equipment. While necrotic enteritis outcomes did not align with previously reported outcomes, it is important to note that the ability to pool performance outcomes in trials 1 and 2 and immunometabolism data in all replicates suggests that these responses to E. maxima were repeatable within the same research setting. While an E. maxima dose for downstream necrotic enteritis challenge could be selected based on outcomes from trials 1 and 2, the existing infrastructure and difficulty in managing heat stress may have confounded dose-response performance outcomes. Future implementation of this model will need to assess the costs/benefits of organizing unchallenged and challenged birds within a vertical structure to reduce cross-contamination while taking environmental control factors into consideration. While facilities used by previous research groups could expediently inoculate birds with washed C. perfringens inoculum, future implementation of this model by other groups will need to account for transit time when deciding on PBS vs. thioglycolate media as a carrier. In addition to these concerns, outcomes in this study indicate that d 1 S. Typhimurium inoculation may alter early physiological responses to E. maxima. While this requires further investigation, researchers should be mindful of potential confounders when utilizing S. Typhimurium to repeatably induce necrotic enteritis, especially if research goals include evaluating immunological responses at timepoints between E. maxima and secondary C. perfringens inoculations.