Effects of Different Eimeria Inoculation Doses on Growth Performance, Daily Feed Intake, Gut Health, Gut Microbiota, Foot Pad Dermatitis, and Eimeria Gene Expression in Broilers Raised in Floor Pens for 35 Days

Simple Summary Coccidiosis, which is induced by Eimeria spp., is one of the most predominant diseases and causes tremendous economic losses in the world. The effects of Eimeria infection on broilers at slaughter ages in floor pen conditions should be elucidated to conduct further studies investigating the effects of feed additives and bioactive compounds as alternatives for anti-coccidial drugs in broilers. The study was aimed to investigate the effects of different inoculation doses of E. acervulina, E. maxima, and E. tenella with different doses on growth performance, gut ecosystem, and oocyst shedding in broilers raised in floor pens for 35 days. Eimeria infection decreased body weight (BW) in the acute phase (D 21), and this effect was prolonged to the final day (D 35). Eimeria oocysts were observed in the litter until D 35, which may indicate that Eimeria spp. reinfected broilers. Eimeria infection dramatically reduced crude fat (CF) digestibility in the acute phase, which may be associated with reduced fat content in the broilers on D 35. Gut microbiota was negatively affected by Eimeria infection in both acute phase and on D 35. In conclusion, Eimeria infection negatively affected the growth performance and gut ecosystem in broilers, and the negative effects were prolonged to D 35 in floor pen conditions. Abstract The study was conducted to investigate the effects of different Eimeria inoculation doses on the growth performance, gut ecosystem, and body composition of broilers in floor pens for 35 days. A total of 750 15-day-old broilers were allocated to five experimental groups with six replicate pens. The five experimental groups included unchallenged control (CON); Eimeria dose 1 (ED1): E. acervulina: 31,250/E. maxima: 6250/E. tenella: 6250; Eimeria dose 2 (ED2): E. acervulina: 62,500/E. maxima: 12,500/E. tenella: 12,500; Eimeria dose 3 (ED3): E. acervulina: 125,000/E. maxima: 25,000/E. tenella: 25,000; and Eimeria dose 4 (ED4): E. acervulina: 250,000/E. maxima: 50,000/E. tenella: 50,000. On D 21, BW were linearly reduced by increased Eimeria inoculation doses (p < 0.01). On D 35, the Eimeria challenge groups had significantly lower BW compared to the CON group. Increased Eimeria inoculation doses linearly decreased crude fat (CF) (p < 0.01) on D 21. Increased Eimeria inoculation doses tended to increase the relative abundance of the phylum Proteobacteria (p = 0.098) on D 21. On D 35, lean:fat was linearly reduced by increased Eimeria inoculation doses (p < 0.05). Eimeria infection negatively influenced growth performance and gut health in broilers in the acute phase, and the negative effects were prolonged to D 35 in floor pen conditions.


Lesion Score, Oocyst Shedding, Gut Permeability, and Digesta and Litter Moisture Content
On D 21 (6 dpi), D 28, and D 35, duodenum, jejunum-ileum section, and ceca were collected, and lesion scores for each section were evaluated according to the 4-score scale [15]. On 22 (7 dpi), D 28, and D 35, cloaca content from one sacrificed bird per pen and approximately 30 g of litter samples were collected for oocyst shedding. Litter samples were collected in the different areas of the pen except areas near water nipples, and the litter samples were thoroughly mixed. Tap water (40 mL) were added to the samples (5 g), and Animals 2023, 13, 2237 4 of 28 the samples were placed at room temperature overnight to dissolve hard fecal particles. Afterwards, the samples were vortexed and diluted 10 times with saturated salt solution. E. acervulina, E. maxima, and E. tenella in the solution were counted using a hemocytometer (Hausser Scientific Company, Horsham, PA, USA). Ileal content and litter samples were oven-dried (75 • C) until a constant weight was achieved to determine their moisture content [16]. Gut permeability by using fluorescein isothiocyanate-dextran (molecular weight: 4 kDa; FITC-D4; Sigma-Aldrich Co., St. Louis, MO, USA) was determined on D 20 (5 dpi) and D 27 according to Choi et al. [12]. The concentration of FITD-F4 in the serum was determined using a prepared standard curve.

Foot Pad Dermatitis (FPD) and Body Composition Analysis
On D 35, severity and incidence of FPD were recorded according to Sorin et al. [17], as shown in Table 2. Table 2. Footpad dermatitis (FPD) scoring system according to Sorin et al. [17].

0
No lesion 1 FPD covers less than 50% of the food pad 2 FPD covers more than 50% of the food pad One bird per pen was euthanized via cervical dislocation and scanned using a dualenergy X-ray absorptiometry (DEXA, GE Healthcare, Madison, WI, USA) as shown in Figure 1. approximately 30 g of litter samples were collected for oocyst shedding. Litter sam were collected in the different areas of the pen except areas near water nipples, and litter samples were thoroughly mixed. Tap water (40 mL) were added to the samples ( and the samples were placed at room temperature overnight to dissolve hard fecal p cles. Afterwards, the samples were vortexed and diluted 10 times with saturated sal lution. E. acervulina, E. maxima, and E. tenella in the solution were counted using a he cytometer (Hausser Scientific Company, Horsham, PA, USA). Ileal content and litter s ples were oven-dried (75 °C) until a constant weight was achieved to determine their m ture content [16]. Gut permeability by using fluorescein isothiocyanate-dextran (mol lar weight: 4 kDa; FITC-D4; Sigma-Aldrich Co., St. Louis, MO, USA) was determined D 20 (5 dpi) and D 27 according to Choi et al. [12]. The concentration of FITD-F4 in serum was determined using a prepared standard curve.

Foot Pad Dermatitis (FPD) and Body Composition Analysis
On D 35, severity and incidence of FPD were recorded according to Sorin et al. as shown in Table 2. Table 2. Footpad dermatitis (FPD) scoring system according to Sorin et al. [17].

FPD Score
Description 0 No lesion 1 FPD covers less than 50% of the food pad 2 FPD covers more than 50% of the food pad One bird per pen was euthanized via cervical dislocation and scanned using a d energy X-ray absorptiometry (DEXA, GE Healthcare, Madison, WI, USA) as shown in ure 1.

Apparent Ileal Digestibility (AID) of Nutrients
On D 21 (6 dpi) and D 35, ileal content was collected between 10 cm below of Meckel's diverticulum and 10 cm upper of the ileo-ceca-colonic junction. Feed (0.5 g) ileal samples (0.3 g) were ashed at 600 °C overnight, and concentrations of titanium d ide were analyzed according to Short et al. [18]. The concentrations of CP and crud (CF) were determined using nitrogen combustion analyses according to AOAC inte tional (2000) analytical method 990.03 and analytical method 942.05, respectively. Ap ent ileal digestibility (AID) of dry matter (DM), organic matter (OM), ash, CP, and CF w calculated according to Lin and Olukosi [19].

Apparent Ileal Digestibility (AID) of Nutrients
On D 21 (6 dpi) and D 35, ileal content was collected between 10 cm below of the Meckel's diverticulum and 10 cm upper of the ileo-ceca-colonic junction. Feed (0.5 g) and ileal samples (0.3 g) were ashed at 600 • C overnight, and concentrations of titanium dioxide were analyzed according to Short et al. [18]. The concentrations of CP and crude fat (CF) were determined using nitrogen combustion analyses according to AOAC international (2000) analytical method 990.03 and analytical method 942.05, respectively. Apparent ileal digestibility (AID) of dry matter (DM), organic matter (OM), ash, CP, and CF were calculated according to Lin and Olukosi [19].

Intestinal Morphology
On D 21 (6 dpi) and D 35, the duodenum (mid-part of the duodenal loop), jejunum (10 cm upper of the Meckel's diverticulum), and mid-ceca were collected, and remaining digesta was rinsed with PBS and placed into 10% neutral buffered formalin solution. After 72 h fixation, samples were embedded in paraffin and sliced into 4 µm sections, and hematoxylin and eosin (H&E) staining was performed. The stained slides were pictured using a microscope (BZ-X810; Keyence, Osaka, Japan). The villus height (VH) and crypt depth (CD) were measured for the duodenum and jejunum samples, and CD was measured for the ceca samples using ImageJ (National Institutes of Health, Bethesda, MD, USA).

Activities of Jejunal Brush Border Digestive Enzymes and Alkaline Phosphatase in the Serum
On D 21 (6 dpi) and D 35, the jejunum (located 10 cm above the Meckel's diverticulum) was collected and immediately snap-frozen in liquid nitrogen. It was then stored at −80 • C for further analysis. Around 100 mg of the jejunum samples (10 cm upper of the Meckel's diverticulum) were homogenized in 1.8 mL PBS by using a beads beater (Biospec Products, Bartlesville, OK, USA). The homogenized samples were centrifuged at 12,000× g and 4 • C for 15 min, and the protein concentration of the supernatant was determined using Pierce BCA Protein Assay Kits (Thermo Fisher Scientific, Waltham, MA, USA). Activities of maltase and sucrase were evaluated according to Fan et al. [20]. Activities of alkaline phosphatase in the intestine and serum were determined according to Lackeyram et al. [21]. Activities of aminopeptidase N (APN) were measured according to the method by Maroux et al. [22]. Lipase activities were analyzed according to the method of Elgharbawy et al. [23]. The activities of digestive enzymes were expressed as values per mg protein.

RNA Extraction and Quantitative Real-Time Reverse Transcription PCR (qRT-PCR)
On D 21 (6 dpi), whole-tissue samples of the duodenum (mid-part of the duodenal loop), jejunum (10 cm upper of the Meckel's diverticulum), and mid-ceca were collected and immediately snap-frozen in liquid nitrogen. They were then stored at −80 • C for further analysis. Around 100 mg of the whole-tissue of the duodenum (mid-part of the duodenal loop), jejunum (10 cm upper of the Meckel's diverticulum), and mid-ceca was homogenized in QIAzol lysis reagents (Qiagen, Valencia, CA, USA) using a beads beater (Biospec Products, Bartlesville, OK, USA). RNA was extracted according to the manufacturer's procedure, and RNA quantity was measured using a NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA). One microgram of RNA was utilized to synthesize the first-strand cDNA using high-capacity cDNA synthesis kits (Applied Biosystems, Foster City, CA, USA). Primers used in the study are presented in Table 3.
Quantitative real-time reverse transcription PCR (qRT-PCR) was conducted using SYBR Green Master Mix with a Step One thermocycler (Applied Biosystems, Foster City, CA, USA). The final PCR volume (10 µL) contained 5 µL of SYBR Green Master Mix, 1.5 µL of cDNA, 0.5 µL of forward and reverse primers (10 µM), and 2.5 µL of water. Thermal cycle conditions for all reactions were as follows: 95 • C denature for 10 min, 40 cycles at 95 • C for 15 s and 60 • C for 1 min, 95 • C for 15 s, 60 • C for 1 min and 95 • C for 15 s. The melting curve of each gene was checked to confirm the specificity of each PCR product. Several PCR products from each gene were stained with 6 × DNA loading dye (Thermo Fisher Scientific, Waltham, MA, USA), electrophoresed on a 3% agarose gel in a Tris-acetate-EDTA buffer, and visualized by adding ethidium bromide to confirm the specificity of each PCR product. Relative abundance of Eimeria 18s genes was normalized by using host reference genes (geometric mean of beta actin and glyceraldehyde 3-phosphate dehydrogenase) to quantify Eimeria spp. in the host tissue, and relative abundance of Eimeria genes was normalized with a housekeeping gene (18s) for each Eimeria spp. [24]. Relative mRNA abundance was determined by using the 2 −∆∆Ct method [25]. The negative control, containing no cDNA, was included in each run, and each sample was run in duplicate.

DNA Extraction and Microbiome Analysis
On D 21 (6 dpi) and D 35, the cecal content was collected and immediately snap-frozen in liquid nitrogen. It was then stored at −80 • C for further analysis. DNA was extracted from the cecal content by employing QIAamp ® DNA stool mini kit (Qiagen GmbH, Hilden, Germany) according to manufacturer's protocol. Quality and quantity of extracted DNA were checked using a NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific), and samples were shipped to LC sciences (Houston, TX, USA) for 16s rRNA gene sequencing.
Qimme2 (version 2022.02) was used to process and analyze 16s rRNA gene sequences [26]. Using Qiime2's built-in functions, alpha diversity, beta diversity, and the phylum and family level composition were analyzed and presented.

Statistical Analyses
Statistical analyses and graph construction were performed using SAS (version 9.4; SAS Inst. Inc., Cary, NC, USA) and GraphPad Prism (Version 9.1.0; GraphPad Software, San Diego, CA, USA), respectively. All experimental groups were compared using PROC MIXED in a completely randomized design followed by the Tukey's HSD (honestly significant difference) test. For quantitative beta diversity measurement, each experimental group was placed as the control group, and experimental groups were compared using PROC MIXED with Dunnett's post hoc test. Eimeria lesion score data were compared using Kruskal-Wallis test followed by the Dwass-Steel-Critchlow-Fligner test. Orthogonal polynomial contrasts were performed to analyze the significance of linear or quadratic effects of different Eimeria doses, and the inoculation doses of E. acervulina, E. maxima, and E. tenella were normalized by using the base 2 logarithm of the number of sporulated Eimeria oocyst number for orthogonal polynomial contrasts. Statistical significance was set at p < 0.05, and trends (0.05 ≤ p ≤ 0.1) were also shown.

Growth Performance and Daily Feed Intake (DFI)
The results of the growth performance are shown in Table 4. On D 21, BW was linearly (p < 0.01) and quadratically (p < 0.05) decreased due to increased Eimeria inoculation doses, the ED1 and ED2 groups had significantly lower BW compared to the CON group, and the ED4 group had significantly lower BW compared to the ED1 and ED2 groups. Increased Eimeria inoculation doses linearly and quadratically reduced ADFI (p < 0.01), and the CON group and the ED4 group had the highest and lowest ADFI among the experimental groups, respectively (p < 0.05). The FCR was linearly increased by increased Eimeria inoculation doses (p < 0.01), and the ED4 group and the CON group had the highest and lowest FCR, respectively, among the experimental groups (p < 0.01).
On D 28, increased Eimeria inoculation doses linearly (p < 0.01) and quadratically (p < 0.05) decreased BW, ADG, and FCR, and the Eimeria challenge groups had lower BW, ADG, and ADFI compared to the CON group. On D 35, Eimeria challenged groups had significantly lower BW compared to the CON group, and increased Eimeria doses linearly (p < 0.05) and quadratically (p < 0.05) reduced BW. In the whole phase, increased Eimeria inoculation doses linearly (p < 0.01) and quadratically (p < 0.01) reduced ADG, and the Eimeria infected groups had lower ADG compared to the CON group (p < 0.01). The ADFI were linearly (p < 0.01) and quadratically (tendency; p = 0.076) decreased by Eimeria infection, and the Eimeria challenged groups had lower ADFI compared to the CON group in the whole phase (p < 0.01). The CON group and the ED4 group had the lowest and highest FCR, respectively, among the experimental groups (p < 0.01), and increased Eimeria inoculation doses resulted in a linear increase in FCR in the whole phase (p < 0.01).
The DFI during the entire experimental period is presented in Figure 2. On D 19, DFI was linearly reduced by increased Eimeria inoculation doses (p < 0.01). From D 20 to 23, DFI was linearly (p < 0.01) and quadratically (p < 0.05) decreased by increased Eimeria inoculation doses. Increased Eimeria inoculation doses linearly decreased DFI on D 26 (p < 0.01) and D 27 (p < 0.05). On D 28, increased Eimeria inoculation doses tended to linearly (p = 0.055) and quadratically (p = 0.075) reduce DFI. On D 29, the Eimeria challenge tended to linearly reduce DFI (p = 0.057). No differences were observed in the DFI from D 30 to 35 (p > 0.1).

Lesion Score and Gut Permeability
On D 21, the CON group had significantly lower lesion scores for E. acervulina, E. maxima, and E. tenella compared to the Eimeria infected groups as shown in Figure 3. The ED4 group had significantly higher E. tenella lesion scores compared to the ED1 group. On D 28, although no statistical differences in E. acervulina, E. maxima, and E. tenella lesion scores were observed among the experimental groups (p > 0.1), lesion scores of E. acervulina and E. maxima were observed in all experimental groups, and E. tenella lesion score was observed in the ED2, ED3, and ED4 groups. On D 35, E. acervulina lesion scores were observed only in the ED3 group, and E. maxima lesion scores were observed in all experimental groups. E. tenella lesion scores were observed in all experimental groups except the ED2 group.
As shown in Table 5, on D 20, the ED4 group had significantly higher gut permeability compared to the CON and ED1 groups, and the ED3 group had significantly higher gut permeability compared to the ED2 group. However, no significant differences were observed in gut permeability on D 27 (p > 0.1).

Digesta and Litter Moisture Content and Foot Pad Dermatitis (FPD)
As shown in Table 6, on D 21 and D 35, increased Eimeria inoculation doses resulted in a linear increase in ileal moisture content (p < 0.01), and the ED4 group had significantly higher ileal moisture content compared to the CON group. On D 21, litter moisture content tended to be increased in a linear trend by increased Eimeria inoculation doses (p = 0.098). Increased Eimeria inoculation doses caused a quadratic increase in the litter moisture content on D 35 (p < 0.05).

Lesion Score and Gut Permeability
On D 21, the CON group had significantly lower lesion scores for E. acervulina, E. maxima, and E. tenella compared to the Eimeria infected groups as shown in Figure 3. The ED4 group had significantly higher E. tenella lesion scores compared to the ED1 group. On D 28, although no statistical differences in E. acervulina, E. maxima, and E. tenella lesion scores were observed among the experimental groups (p > 0.1), lesion scores of E. acervul- Experimental groups were compared using PROC MIXED followed by the Tukey's HSD (honestly significant difference) test, and different letters mean significant differences (p < 0.05) among the experimental groups on the same day. Orthogonal polynomial contrasts were conducted to see linear and quadratic patterns among the experimental groups on the same day.  [15]. Experimental groups were compared using Kruskal-Wallis test followed by the Dwass-Steel-Critchlow-Fligner test. Different letters mean significant differences (p < 0.05) among the experimental groups.
As shown in Table 5, on D 20, the ED4 group had significantly higher gut permeability compared to the CON and ED1 groups, and the ED3 group had significantly higher gut permeability compared to the ED2 group. However, no significant differences were observed in gut permeability on D 27 (p > 0.1).   [15]. Experimental groups were compared using Kruskal-Wallis test followed by the Dwass-Steel-Critchlow-Fligner test. Different letters mean significant differences (p < 0.05) among the experimental groups. The severity and incidence of FPD on D 35 were quadratically increased by increased Eimeria inoculation doses (p < 0.05; Figure 4).
Tukey's HSD (honestly significant difference) test. Different letters in the same row means si cant differences (p < 0.05) among the experimental groups. Orthogonal polynomial contrasts conducted to see linear and quadratic patterns among the experimental groups. 2 Standard err the means.
The severity and incidence of FPD on D 35 were quadratically increased by incre Eimeria inoculation doses (p < 0.05; Figure 4).  Severity (0 to 2), and incidence of FPD was recorded according to Sorin et al. [17]. Experimental groups were compared using Kruskal-Wallis test followed by the Dwass-Steel-Critchlow-Fligner test. Orthogonal polynomial contrasts were conducted to see linear and quadratic patterns among the experimental groups. Table 7 shows that increased inoculation doses of E. acervulina, E. maxima, and E. tenella linearly increased the number of oocysts of E. acervulina and E. tenella (p < 0.05) and tended to linearly increase E. maxima oocyst in the cloaca content (p = 0.081) on D 22. The oocyst number of E. maxima in the cloaca were quadratically modulated by increased doses of E. maxima (p < 0.05). Linearly increased number of oocysts of E. acervulina and E. tenella in the litter were observed by increased infection doses of Eimeria spp. (p < 0.01), and the ED3 and ED4 groups had significantly higher E. acervulina oocysts in the litter compared to the ED1 group, and the ED2 and ED3 groups had significantly higher E. tenella oocysts compared to the ED1 group on D 22. However, no statistical differences were observed in the number of Eimeria oocysts among the Eimeria challenged groups (p > 0.1), Eimeria oocysts were observed either in the cloaca content and litter in the Eimeria infected groups on D 28 and 35.

Apparent Ileal Digestibility (AID) of Nutrients
As shown in Table 8, increased Eimeria inoculation doses linearly decreased the AID of CP and EE (p < 0.01) on D 21, and the AID of EE tended to be quadratically reduced by increased Eimeria inoculation doses (p = 0.056). The ED4 group had a significantly lower AID of CP compared to the CON group. The ED3 and ED4 groups had a significantly lower AID of EE compared to the CON group, and the ED2, ED3, and ED4 groups had minus values for the AID of CF. On D 35, the AID of ash was linearly decreased by increased Eimeria inoculation doses (p < 0.05).

Intestinal Morphology
As shown in Table 9, duodenal VH were linearly reduced by increased Eimeria inoculation doses (p < 0.01), and the ED3 and ED4 groups had significantly lower duodenal VH compared to the CON and ED1 groups. Duodenal CD was linearly deepened by increased Eimeria inoculation doses (p < 0.01), and the CON group had lower duodenal CD compared to the ED2, ED3, and ED4 groups (p < 0.01). Duodenal VH:CD were linearly reduced by increased Eimeria inoculation doses (p < 0.01), and the Eimeria challenged groups has significantly lower duodenal VH:CD compared to the CON group (p < 0.01). Jejunal VH was linearly decreased by increased Eimeria inoculation doses (p < 0.01), and the ED3 and ED4 groups had significantly lower jejunal VH compared to the CON and ED1 groups. Jejunal CD was linearly deepened by increased Eimeria inoculation doses (p < 0.01), and the CON group had the lowest jejunal CD among the experimental groups (p < 0.01). Jejunal VH:CD was linearly reduced by increased Eimeria inoculation doses (p < 0.01), and the ED3 and ED4 groups had the lowest jejunal VH:CD among the experimental groups (p < 0.01), and the CON group had the highest jejunal VH:CD among the experimental groups (p < 0.01). Cecal CD was linearly deepened by increased Eimeria inoculation doses (p < 0.01), and the ED3 and ED4 groups had significantly deeper cecal CD compared to the CON group (p < 0.05). On D 35, duodenal CD was quadratically deepened by increased Eimeria inoculation doses (p < 0.01), and the ED2 group had significantly higher duodenal CD compared to the CON group. Increased Eimeria inoculation doses resulted in a quadratic increase in duodenal VH:CD (p < 0.05).

Activities of Jejunal Brush Border Digestive Enzymes
As shown in Table 10, activities of APN were lower in the ED1 group compared to the CON group on D 21 (p < 0.05). Activities of serum alkaline phosphatase (SAP) were linearly reduced by increased Eimeria inoculation doses (p < 0.01), and the ED3 and ED4 groups had significantly lower activities of SAP compared to the CON group. The ED2 group had significantly higher sucrase activities compared to the ED3 group. Increased Eimeria inoculation doses linearly decreased maltase activities (p < 0.01), and the ED1 and ED4 groups had significantly lower maltase activities compared to the CON group. On D 35, activities of APN were linearly decreased by increased Eimeria infection doses (p < 0.05).

Alpha Diversity of the Cecal Microbiome Communities
No differences were observed in the alpha diversity parameters (biodiversity of the samples) on D 21 (p > 0.1; Table 11). On D 35, alpha diversity parameters including pielou evenness (evenness; p < 0.05), faith phylogenetic diversity (biodiversity based on phylogeny; p < 0.05), shannon entropy (richness and evenness; p < 0.05), and observed features (richness; p < 0.01) were linearly reduced by increased Eimeria inoculation doses.

Beta Diversity of the Cecal Microbiome Communities
As shown in Figure 5, no differences were observed in unweighted unifrac distance (dissimilarity among samples without considering abundance information) to each experimental group both on D 21 and D 35 (p > 0.1).

Beta Diversity of the Cecal Microbiome Communities
As shown in Figure 5, no differences were observed in unweighted unifrac distance (dissimilarity among samples without considering abundance information) to each experimental group both on D 21 and D 35 (p > 0.1).  Figure 6 shows that on D 21, the ED3 group had significantly higher weighted unifrac distance (dissimilarity among samples with considering abundance information) to compared to the CON and ED1 groups (p < 0.05). On D 35, the ED2 (p < 0.01) and ED4 (p < 0.05) groups had significantly higher weighted unifrac distance compared to the CON and ED1 groups. The ED2 group had significantly higher weighted unifrac distance compared to the ED3 group (p < 0.01).  Figure 6 shows that on D 21, the ED3 group had significantly higher weighted unifrac distance (dissimilarity among samples with considering abundance information) to compared to the CON and ED1 groups (p < 0.05). On D 35, the ED2 (p < 0.01) and ED4 (p < 0.05) groups had significantly higher weighted unifrac distance compared to the CON and ED1 groups. The ED2 group had significantly higher weighted unifrac distance compared to the ED3 group (p < 0.01).   Each experimental group was placed as the control group, and experimental groups were compared by using one-way PROC MIXED with Dunnett's post hoc test. ** represents when p < 0.05, and *** represents when p < 0.01. Figure 7, no visual differences were observed in beta diversity parameters including unweighted unifrac and weighted unifrac on D 21 and D 35. As shown in Figure 7, no visual differences were observed in beta diversity parameters including unweighted unifrac and weighted unifrac on D 21 and D 35.

Taxa Abundance of the Cecal Microbiome Communities
As shown in Figure 8, increased Eimeria inoculation doses tended to increase the relative abundance of the phylum Proteobacteria (p = 0.098). Increased Eimeria inoculation doses tended to linearly enlarge the relative abundance of the family Enterobacteriaceae (Figure 9; p = 0.091). The relative abundance of the family Bacillaceae was enlarged in a linear trend by increased Eimeria inoculation doses (p < 0.05). Increased Eimeria inoculation doses linearly (p < 0.05) and quadratically (tendency; p = 0.074) reduced the relative abundance of the family Christensenellaceae, and the ED2 group had significantly lower the relative abundance of the family Christensenellaceae compared to the CON group. Increased Eimeria inoculation doses linearly reduced the relative abundance of the family Peptostreptococcaceae, and the ED4 group had a significantly lower relative abundance of the family Peptostreptococcaceae compared to the CON group.

Taxa Abundance of the Cecal Microbiome Communities
As shown in Figure 8, increased Eimeria inoculation doses tended to increase the relative abundance of the phylum Proteobacteria (p = 0.098). Increased Eimeria inoculation doses tended to linearly enlarge the relative abundance of the family Enterobacteriaceae (Figure 9; p = 0.091). The relative abundance of the family Bacillaceae was enlarged in a linear trend by increased Eimeria inoculation doses (p < 0.05). Increased Eimeria inoculation doses linearly (p < 0.05) and quadratically (tendency; p = 0.074) reduced the relative abundance of the family Christensenellaceae, and the ED2 group had significantly lower the relative abundance of the family Christensenellaceae compared to the CON group. Increased Eimeria inoculation doses linearly reduced the relative abundance of the family Peptostreptococcaceae, and the ED4 group had a significantly lower relative abundance of the family Peptostreptococcaceae compared to the CON group. The phylum with statistical differences were presented. Experimental groups were compared using PROC MIXED followed by the Tukey's HSD (honestly significant difference) test, and different letters mean significant differences (p < 0.05) among the experimental groups. Orthogonal polynomial contrasts were conducted to see linear and quadratic patterns among the experimental groups.
On D 35, the relative abundance of the phylum Actinobacteria were quadratically enlarged by increased Eimeria inoculation doses (p < 0.05), and the ED4 group had a significantly lower relative abundance of the phylum Actinobacteria compared to the ED2 group (Figure 8). Increased Eimeria inoculation doses tended to increase the relative abundance of the family Enterobacteriaceae with a linear trend (p = 0.061; Figure 9). The relative The phylum with statistical differences were presented. Experimental groups were compared using PROC MIXED followed by the Tukey's HSD (honestly significant difference) test, and different letters mean significant differences (p < 0.05) among the experimental groups. Orthogonal polynomial contrasts were conducted to see linear and quadratic patterns among the experimental groups.
ically increased the relative abundance of the family Streptosporangiceae (p < 0.01 the ED2 group had significantly higher relative abundance of the family Streptosp giceae compared to the ED4 group. Increased Eimeria inoculation doses quadratica creased the relative abundance of the family Erysipelotrichaceae (p < 0.01), and th group had significantly higher relative abundance of the family Streptosporangiceae pared to the CON, ED3, and ED4 groups.

Eimeria Gene Expression
As shown in Table 12, different Eimeria doses did not modulate mRNA express E. acervulina and E. tenella genes including APN, flagella-related protein, elongatio tors (EF), and gametocyte proteins (GAM) in broilers on D 21 (p > 0.1). However, h Eimeria inoculation doses linearly upregulated the gene expression of E. maxima AP 0.05), and the ED3 group had significantly higher E. maxima APN compared to th and ED2 groups. Higher E. maxima doses linearly increased the gene expression of E ima EF2 (p < 0.05). Higher E. maxima doses linearly upregulated the gene expression maxima GAM56 (p < 0.01), and the ED3 group had significantly higher gene express E. maxima GAM56 compared to the ED1 and ED2 groups. Gene expression of E. m The family with statistical differences were presented. Experimental groups were compared using PROC MIXED followed by the Tukey's HSD (honestly significant difference) test, and different letters mean significant differences (p < 0.05) among the experimental groups. Orthogonal polynomial contrasts were conducted to see linear and quadratic patterns among the experimental groups.
On D 35, the relative abundance of the phylum Actinobacteria were quadratically enlarged by increased Eimeria inoculation doses (p < 0.05), and the ED4 group had a significantly lower relative abundance of the phylum Actinobacteria compared to the ED2 group ( Figure 8). Increased Eimeria inoculation doses tended to increase the relative abundance of the family Enterobacteriaceae with a linear trend (p = 0.061; Figure 9). The relative abundance of the family Ruminococcaceae was linearly reduced by Eimeria infection (p < 0.05), and the ED3 group had a significantly lower relative abundance of the family Ruminococcaceae compared to the ED1 group. Increased Eimeria inoculation doses quadratically increased the relative abundance of the family Streptosporangiceae (p < 0.01), and the ED2 group had significantly higher relative abundance of the family Streptosporangiceae compared to the ED4 group. Increased Eimeria inoculation doses quadratically increased the relative abundance of the family Erysipelotrichaceae (p < 0.01), and the ED2 group had significantly higher relative abundance of the family Streptosporangiceae compared to the CON, ED3, and ED4 groups.

Eimeria Gene Expression
As shown in Table 12, different Eimeria doses did not modulate mRNA expression of E. acervulina and E. tenella genes including APN, flagella-related protein, elongation factors (EF), and gametocyte proteins (GAM) in broilers on D 21 (p > 0.1). However, higher Eimeria inoculation doses linearly upregulated the gene expression of E. maxima APN (p < 0.05), and the ED3 group had significantly higher E. maxima APN compared to the ED1 and ED2 groups. Higher E. maxima doses linearly increased the gene expression of E. maxima EF2 (p < 0.05). Higher E. maxima doses linearly upregulated the gene expression of E. maxima GAM56 (p < 0.01), and the ED3 group had significantly higher gene expression of E. maxima GAM56 compared to the ED1 and ED2 groups. Gene expression of E. maxima GAM82 was linearly increased by higher doses of Eimeria (p < 0.01), and the ED3 group had a significantly higher gene expression of E. maxima GAM82 compared to the ED1 group. . Experimental groups were compared using PROC MIXED followed by the Tukey's HSD (honestly significant difference) test. Different letters in the same row means significant differences (p < 0.05) among the experimental groups. Orthogonal polynomial contrasts were conducted to see linear and quadratic patterns among the experimental groups. 2 Relative abundance of Eimeria 18s genes was normalized by using host reference genes (beta actin and glyceraldehyde 3-phosphate dehydrogenase), and relative abundance of Eimeria genes including aminopeptidase N (APN), elongation factor (EF), and gametocyte protein (GAM) were normalized by using Eimeria 18s genes. 3 Standard errors of the means.

Body Composition
The body composition of broilers infected with Eimeria spp. on D 21, 28, and 35 is shown in Table 13. On D 21, total weight was linearly reduced by increased Eimeria inoculation doses (p < 0.05), and the ED4 group had a significantly lower total weight compared to the CON group. Increased Eimeria infection doses linearly reduced fat weight (p < 0.05) and lean weight (p < 0.01). No differences were observed on D 28 (p > 0.1). On D 35, fat weight was linearly reduced by increased Eimeria inoculation doses (p < 0.05), and the ED4 group tended to have lower fat weight compared to the CON group (p = 0.055). The lean weight tended to be linearly decreased by increased Eimeria inoculation doses (p = 0.068). Fat percentage tended to be reduced with a linear trend by increased Eimeria inoculation doses (p = 0.060), and lean percentage tended to be increased with a linear trend by increased Eimeria inoculation doses (p = 0.059). Lean:fat ratio was linearly reduced by increased Eimeria inoculation doses (p < 0.05). Table 13. Effects of different Eimeria infection doses on body composition parameters including total weight (kg), fat (kg), lean weight (kg), fat percentage (%), lean percentage (%), and lean:fat in broilers 1 .

Discussion
The purpose of the study was to investigate the effects of different Eimeria inoculation doses on growth performance, litter moisture content, nutrient digestion and absorption, incidence and severity of FPD, gut microbiota, oocyst shedding, Eimeria gene expression, and body composition in broilers raised in floor pens for 35 days. Cross contamination among pens in the Eimeria infection studies in floor pens would be problematic [27]. To minimize cross contamination between the unchallenged group and Eimeria challenged groups, an anti-coccidial drug (Coban 90) was supplemented in the unchallenged group (CON group) in the current study. The dose of Coban 90 (500 mg/kg) was determined based on our previous study (unpublished). Broilers challenged with Eimeria spp. and fed 500 mg/kg of Coban 90 showed improved body weight compared to the challenged group without Coban 90, and their BW was similar to that of the non-challenged group. Furthermore, the supplementation of 500 mg/kg of Coban 90 did not adversely affect the growth performance of broiler chickens without an Eimeria challenge. However, although workers were extra careful during the entire experimental period, there was still Eimeria cross contamination (e.g., oocyst shedding and lesion score) to the unchallenged group in the current study. These results indicate that supplementation of Coban 90 at 500 mg/kg did not completely inhibit the colonization of Eimeria spp. in broilers even with indirect infection [28]. However, significant statistical differences in body weight (BW) were observed between the control group and the Eimeria challenged groups on Day 35. Additionally, throughout the current study, the control group consistently had lesion scores for Eimeria spp. lower than 0.5 and a minimum gut permeability. These findings suggest that the level of Eimeria infection in the CON group did not have a significant impact on serving as the negative control for the CON group. The DFI was measured during the entire experiment period to check the severity of Eimeria infection and whether Eimeria re-infection occurred in floor pen conditions. The DFI could be one of the powerful and non-invasive parameters to indicate incidence and severity of Eimeria infection in broilers. Measuring DFI has benefits in easiness and time over other non-invasive methods including measuring core body temperature or fecal moisture content. Our previous study [29] showed that Eimeria spp. Infection reduced ADFI in broilers in the acute phase (0 to 6 days post infection (dpi)). The time points 5 dpi and 6 dpi were considered as the peak time points for Eimeria infection according to our previous studies based on the results of gut permeability and daily feed intake [14,29]. In the current study, DFI was dramatically decreased on 5 and 6 dpi and continued to be linearly reduced by increased Eimeria inoculation doses from 11 to 14 dpi (D 26 to 29), which potentially indicates that re-infection of Eimeria has occurred in the current study. This was supported by the results of Eimeria lesion and oocyst shedding of Eimeria spp. on D 28 and 35, while the average values for Eimeria lesion scores on D 35 were below 1 out of 4 in the current study. Once exposed to Eimeria infection, chickens develop strong humoral and cellular immunity against re-infection of Eimeria [30], which may demonstrate no statistical differences in gut permeability and intestinal morphology in broilers after the acute phase in the current study. Still, our current study showed that reinfection of Eimeria was able to decrease feed intake of broilers in the floor pen conditions. Reduced feed intake in broilers infected with Eimeria spp. might be due to alternation in immune response and endocrine system in broilers [31,32].
In the acute phase (0 to 6 dpi; D 15 to 21), reduced feed intake with impaired feed efficiency dramatically decreased BW and ADG in broilers infected with Eimeria spp. in the current study, which was in agreement with Teng et al. [14]. On D 21 to 28, BW and ADG were reduced, along with decreased ADFI, without affecting feed efficiency in Eimeria-infected broilers. These results indicate that severe Eimeria infection reduced growth rate of broilers by damaging the capacity of nutrient digestion and absorption and reducing the feed intake in broilers, but re-infection of Eimeria decreased growth rate via decreasing only feed intake in broilers. These indicate that decreasing feed intake could be a sensitive sign in Eimeria-infected broilers. On D 35, Eimeria infection reduced only BW without affecting ADG, ADFI, and FCR in the Eimeria challenged groups compared to the CON group on D 35. These results were in consistent with a study reported that Eimeria inoculation on D 15 decreased BW in broilers on D 42 in floor pens [33]. In previous studies, compensatory growth happened in broilers infected with E. maxima in the recovery phase (6 to 13 dpi) [12] and in broilers challenged with E. acervulina and E. maxima (14 to 21 dpi) [34] by improving feed intake or feed efficiency after the acute phase of Eimeria infection. However, potentially, infection of E. acervulina, E. maxima, and E. tenella would not induce compensatory growth in the current study because they induced severe damage in the gastrointestinal tract of broilers. More studies are required to explain the potential mode of actions of Eimeria infection on reduced feed intake and to elucidate compensatory growth after Eimeria infection in broilers.
Higher Eimeria inoculation doses above the threshold where the maximal reproductive potential reached may result in reduced oocyst shedding and downregulated Eimeria genes relating to viability and sexual reproduction due to crowding effects in the gastrointestinal tract [35]. In our previous study, while higher doses of E. tenella increased oocyst shedding 5 to 6 dpi, oocyst shedding 6 to 8 dpi was not affected by challenging doses [36]. This potentially designates that Eimeria can control themselves to determine maximal oocyst production in broilers, which is called crowding effect [35,37]. In the current study, the ED4 group (the highest dose group) had numerically similar oocyst shedding of E. maxima and E. tenella compared to the ED2 and ED3 groups in the cloaca content on D 22. Eimeria oocysts were counted in the cloaca content and litter instead of feces because it is not feasible to collect fresh fecal samples in the litter condition. To elucidate crowding effects of Eimeria spp., whole duodenal, jejunal, and cecal tissue samples were collected on 6 dpi, one day prior to peak date (7 dpi) for oocyst production of E. maxima and E. tenella [38], and gene expression of APN, EF2, GAM56, and GAM82 of E. maxima were modulated by increased Eimeria inoculation doses along with quadratically modulated oocyst production of E. maxima in the current study. The APN, EF, and GAM of Eimeria spp. play important roles in Eimeria viability and sexual reproduction to produce oocysts [24,39,40]. The inconsistency between linearly increased gene expression of APN, EF, and GAM and quadratically modulated oocyst shedding of E. maxima is still in question because only transcriptional level was analyzed in the current study. However, our current study showed that different inoculation doses of Eimeria can alter Eimeria gene expression. Increased inoculation doses only affected E. maxima mRNA expression in the current study potentially because E. maxima is more sensitive to crowding effects due to their largest size among chicken Eimeria spp. [41].
In the current study, ileal digesta and litter moisture contents were increased due to Eimeria infection in the acute phase. In our previous study, E. tenella infection did not increase ileal digesta moisture content on 5 to 7 dpi and even decreased ileal moisture content on 6 dpi [36]. Increased ileal moisture content is mainly due to infection of E. acervulina and E. maxima in broilers. Increased digesta moisture content may imply shortened digesta transit time, which can reduce the capacity of nutrient digestion and absorption [42] and is associated with reduced nutrient digestibility in the current study. Litter moisture content was linearly increased in the acute phase and was quadratically increased by Eimeria infection on D 35 in the current study. Many studies demonstrated that litter moisture is closely associated with FPD in broiler chickens [6,43,44]. FPD is a type of skin inflammation that induces necrotic lesions on the plantar surface of foot pad in broilers [45]. FPD can reduce the marketability of chicken feet, be an entry route for pathogenic bacteria, cause lameness, and reduce the growth performance of broiler chickens [46]. Increased litter moisture content due to Eimeria infection can increase the incidence and severity of FPD in broilers by increasing the litter ammonia concentration [47,48]. The litter moisture content on D 35 and severity/incidence of FPD on D 35 showed similar trends among the experimental groups in the current study, while the severity and incidence of FPD was mild potentially because of the dry conditions of the room and early slaughter age (D 35) in the current study. A previous study by El-Wahab et al. [43] showed that litter moisture should be above 35% to induce FPD in poultry. Eimeria infection still has the potential to increase the incidence and severity of FPD by increasing litter moisture content in broilers.
Gut permeability measured by FITC-D4 is an important indicator to represent functionality and integrity of gut in broilers [49]. Increased gut permeability indicates that more pathogens and toxins can permeate into the blood stream across the epithelial layer, which can cause systemic infection in broilers [50]. In the current study, Eimeria infection increased gut permeability on 5 dpi in broilers, which is consistent with our previous study [14]. Our previous study demonstrated that E. maxima infection had a significant impact on increasing gut permeability [51], whereas E. tenella infection did not affect gut permeability in broilers [36]. Potentially, the infection of E. acervulina and E. maxima disrupts tight junction proteins and the mucus layer, which play an important role in maintaining gut barrier integrity, and this would make the intestinal wall thinner (more permeable) in broilers [52,53]. In contrast, Vicuña et al. [49] showed that FITC-D4 can be deposited in the cecal tissue, and E. tenella infection did not increase gut permeability potentially because E. tenella infection thickened the intestinal wall in the ceca [36]. No differences were observed in the gut permeability on D 27 in the current study, and this suggests that the re-infection of Eimeria did not severely damage gut functionality and integrity, while it reduced feed intake in broilers in the floor pen conditions.
Reduced growth performance and feed efficiency might be mainly attributed to the reduced capacity of nutrient digestion and absorption in Eimeria-infected broilers. In the current study, the AID of CP was linearly reduced by Eimeria infection in broilers in the acute phase (0 to 6 dpi), which is consistent with our previous study by Teng et al. [14], which reported that Eimeria infection reduced the AID of CP in Eimeria-infected broilers. E. acervulina and E. maxima are the main Eimeria spp. that directly reduce the AID of CP because they inhabit in the duodenum and jejunum, respectively [54,55]. Our previous study by Choi et al. [36] reported that E. tenella infection in the ceca did not directly affect AID in the acute (6 dpi) and recovery phase (9 dpi) in broilers. The AID method does not account for the endogenous loss of nutrients in the gastrointestinal tract of broilers [54]. Both nutrient disappearance and increased endogenous loss may have affected the AID of CP values in the Eimeria-infected broilers because Eimeria infection impaired the intestinal morphology and activities of jejunal brush border digestive enzymes in the current study. Eimeria infection is known to increase endogenous losses of proteins (e.g., plasma proteins, mucin, and cell debris) in the gastrointestinal tract during Eimeria colonization and reproduction activities [56,57]. The AID of CF was dramatically reduced by Eimeria infection, and AID values of CF were negative in the ED2, ED3, and ED4 groups in the current study. Consistently, Ghareeb et al. [58] reported that E. maxima infection dramatically reduced the AID of CF in broilers. The negative values for AID of CF indicate that there were high endogenous losses in Eimeria-infected broilers. Endogenous fat from the gastrointestinal tract includes bile, cell debris, intestinal secretions, and microbial lipids [59]. Adams et al. [60] demonstrated that infection of E. acervulina decreased the secretion of bile, which has an essential role in fat digestion by emulsifying fat in the gastrointestinal tract. Potentially, reduced bile secretion and increased cell debris from the gastrointestinal tract may have decreased fat digestibility and increased endogenous loss of fat, which resulted in negative values for the AID of CP in Eimeria-infected broilers in the current study. However, more studies are needed to specify the factors such as bile, cell debris, intestinal secretions, and microbial lipids that increased endogenous losses of fat in Eimeria-infected broilers.
Intestinal morphology is an important indicator to represent capacity of nutrient digestion and absorption in the gastrointestinal tract of chickens [61]. In the current study, infection of E. acervulina and E. maxima reduced VH and increased CD in both duodenum and jejunum on D 21. These results agree with several previous studies [62,63]. The reduced VH and increased CD implies there was high tissue turnover in the intestine by Eimeria infection because bigger crypts, which are reservoirs for enterocytes, indicates a high demand for new tissue in the villus [62,64]. Impaired jejunal morphology is highly associated with reduced activities of jejunal maltase in the current study. This is because mature enterocytes in the villus, which mainly express brush border digestive enzymes, would be quickly deceased due to a higher turnover rate in Eimeria infection conditions [65]. Furthermore, negatively modulated duodenal and jejunal morphology would explain decreased AID of CP and EE in the current study. While no differences were observed in the jejunal morphology on D 35, duodenal VH:CD were quadratically reduced with increased duodenal CD in Eimeria-infected broilers in the current study. Reduced VH:CD due to increased CD indicates more energy and nutrients are required for gut maintenance, which can result in growth retardation in broilers [62]. However, while statistical differences were not observed, the ED3 and ED4 groups exhibited numerically higher duodenal VH and numerically lower VH:CD compared to the control group on D 35. This numerical trend suggests that gut maintenance and new tissue generation may have occurred on D 35. This result also indicates that negative effects of E. acervulina infection lasted through to D 35 in broilers in the current study.
Chicken ceca have only crypts without villus, similarly to the colon of mammals (humans and pigs) [66]. In the current study, E. tenella infection deepened the cecal crypts as we observed in our previous study [36]. However, it is still uncertain whether thickened cecal wall due to E. tenella infection decreases the permeation of microbial metabolites (e.g., endotoxins and short chain fatty acids) across epithelium. In the current study, Eimeria infection negatively affected cecal microbiota in the acute and chronic phases. In the current study, Eimeria infection increased the phylum Proteobacteria and the family Enterobacteriaceae in the acute phase, which includes diverse pathogenic bacteria such as Escherichia coli, Salmonella spp., Vibrio spp., and Pseudomonas spp. [67]. Moreover, in the acute phase, Eimeria infection reduced the relative abundance of the family Christensenellaceae and Peptostreptococcaceae, which play important roles in fiber degradation and short chain fatty acid production in broilers [68]. Potentially, negatively altered cecal microbiota may have reduced volatile fatty acids (VFA) production in the ceca, and this would explain the reduced activities of SAP, which needs VFA production [36,69]. Whereas alpha diversity indices in the cecal microbial communities on D 21 were not affected by Eimeria infection, all alpha diversity indices including pielou evenness (evenness), faith phylogenetic diversity (biodiversity based on phylogeny), shannon entropy (richness and evenness), and observed features (richness) were linearly decreased on D 35 in the current study. Reduced alpha diversity indices suggest the presence of unhealthier and immature gut microbiota in chickens [70]. Furthermore, the relative abundance of the family Enterobacteriaceae was quadratically increased by Eimeria infection, and the relative abundance of the family Ruminococcaceae, which play an important role in fiber degradation and short-chain fatty acid production in the ceca of chickens [71], was linearly reduced by Eimeria infection on D 35 in the current study. These results imply that Eimeria infection on D 15 still negatively influenced cecal microbiota in broilers on D 35. Our previous study showed that E. tenella infection negatively influenced cecal microbiota toward increasing the abundance of pathogenic bacteria and reducing microbial VFA (e.g., an important energy source for chickens) production mainly by impairing the mucosal immune system of broilers and increasing the protein concentration in the cecal content [72]. While E. tenella infection would be the main factor to alter cecal microbiota, the infection of E. acervulina and E. maxima potentially can affect cecal microbiota by increasing endogenous loss of proteins and undigested dietary proteins in the gastrointestinal tract because the phylum Proteobacteria mainly ferment protein sources for their growth and reproduction [73][74][75]. More studies are required to investigate whether the infection of E. acervulina and E. maxima itself can alter cecal microbiota in broilers.
Body composition is a crucial parameter in broiler production because body composition is closely associated with meat yield and quality [76]. In the current study, body composition of broilers was altered by Eimeria infection. On D 35, the lean:fat ratio was linearly enhanced, and fat accumulation was reduced by increased Eimeria infection doses in the current study. Potentially, Eimeria infection stimulated the immune system, resulting in the excessive usage of energy sources (e.g., fat and glycogen) in broilers [77,78]. Furthermore, reduced fat digestibility and decreased cecal VFA production due to negatively altered microbiota on D 21 could result in decreased fat accumulation as a chronic effect on D 35. These results suggest that Eimeria infection can influence the body composition and meat quality of broilers.

Conclusions
In conclusion, Eimeria infection negatively affected the growth performance, gut health, gut barrier integrity, nutrient digestion and absorption, gut microbiota, and body composition of broilers in the acute phase. Increased Eimeria inoculation doses modulated the relative mRNA expression of Eimeria genes relating to viability and sexual reproduction. Eimeria infection negatively affected the growth performance, gut microbiota, FPD, and body composition in broilers, and the negative effects were prolonged to D 35 in the floor pen conditions.