Search
Author
Title
Vol.
Issue
Year
1st Page

# Journal of Animal Science - Animal Nutrition Effects of dietary soybean meal concentration on growth and immune response of pigs infected with porcine reproductive and respiratory syndrome virus1

1. Vol. 93 No. 6, p. 2987-2997

Accepted: Apr 08, 2015
Published: May 29, 2015

2 Corresponding author(s):
View
Share

doi:10.2527/jas.2014-8462
1. S. J. Rochell*,
2. L. S. Alexander*,
3. G. C. Rocha,
4. W. G. Van Alstine,
5. R. D. Boyd§#,
6. J. E. Pettigrew* and
7. R. N. Dilger 2*
1. * Department of Animal Sciences, University of Illinois Urbana-Champaign, Urbana 61801
Department of Animal Sciences, Federal University of Viçosa, Viçosa, Minas Gerais, Brazil
Department of Comparative Pathobiology, Purdue University, West Lafayette, IN 47907
§ Hanor Company Inc., Franklin, KY 42134
# Department of Animal Science, North Carolina State University, Raleigh 27695

## Abstract

An experiment was conducted to determine the effects of dietary soybean meal (SBM) concentration on the growth performance and immune response of pigs infected with porcine reproductive and respiratory syndrome virus (PRRSV). Four experimental treatments included a 2 × 2 factorial arrangement of 2 dietary SBM concentrations, 17.5% (LSBM) or 29% (HSBM), and 2 levels of PRRSV infection, uninfected sham or PRRSV infected. Sixty-four weanling pigs of split sex (21 d of age, 7.14 ± 0.54 kg) were individually housed in disease containment chambers. Pigs were provided a common diet for 1 wk postweaning before being equalized for BW and sex and allotted to 4 treatment groups with 16 replicate pigs per group. Pigs were fed experimental diets for 1 wk before receiving either a sham inoculation (sterile PBS) or a 1 × 105 50% tissue culture infective dose of PRRSV at 35 d of age (0 d postinoculation, DPI). Pig BW and feed intake were recorded weekly, and rectal temperatures were measured daily beginning on 0 DPI. Blood was collected on 0, 3, 7, and 14 DPI for determination of serum PRRSV load, differential complete blood cell counts, and haptoglobin and cytokine concentrations. Infection with PRRSV increased (P < 0.01) rectal temperatures of pigs throughout the infection period, with no influence of dietary SBM concentration. Pigs in the PRRSV-infected group had lower (P < 0.01) ADFI and G:F from 0 to 14 DPI compared with uninfected pigs. In the PRRSV-infected group, pigs fed HSBM tended to have improved ADG (P = 0.06) compared with pigs fed LSBM, whereas there was no influence of SBM concentration on growth of pigs in the uninfected group. At 14 DPI, PRRSV-infected pigs fed HSBM had a lower serum PRRSV load (P < 0.05), a higher (P = 0.02) hematocrit value, and a tendency for greater hemoglobin concentration (P = 0.09) compared with pigs fed LSBM. Serum haptoglobin and tumor necrosis factor-α concentrations of PRRSV-infected pigs were lower (P < 0.05) in pigs fed HSBM at 3 and 14 DPI, respectively, than in pigs fed LSBM. Overall, increasing the dietary SBM concentration modulated the immune response and tended to improve the growth of nursery pigs during a PRRSV infection.

## INTRODUCTION

Porcine reproductive and respiratory syndrome (PRRS), caused by the PRRS virus (PRRSV), is the most prevalent disease of swine globally (Lunney et al., 2010). Infection of nursery pigs with PRRSV leads to a complex immune response that results in fever, lethargy, respiratory stress, reduced feed intake, and ultimately decreased growth performance (Rossow, 1998). In 2005, total annual losses due to PRRS were assessed at $560 million for U.S. swine producers (Neumann et al., 2005), and more recent estimates total$664 million (Holtkamp et al., 2012). Therefore, despite much effort by researchers, PRRS continues to pose a substantial financial burden for the U.S. swine industry.

Previous research concerning the interaction between nutrition and PRRS has primarily focused on specific nutrients and specialty additives (Toepfer-Berg et al., 2004; Che et al., 2011, 2012; Liu et al., 2013), but the effects of most feed ingredients on PRRSV-infected pigs are largely unknown. Soybean meal (SBM) is the primary dietary protein source for swine in the United States, and soy-derived feedstuffs contain a number of biologically active compounds, including isoflavones, saponins, proteins, and peptides (Omoni and Aluko, 2005). Several of these compounds, particularly isoflavones, have been demonstrated to exert antiviral, antioxidant, and anti-inflammatory activity against numerous viruses (Hämäläinen et al., 2007; Andres et al., 2009). Johnston et al. (2010) and Rocha et al. (2013) reported improvements in growth of PRRSV-infected pigs when SBM was increased from 21% to 32% and 12.5% to 22.5% for finisher and nursery pigs, respectively. Thus, the current study was conducted to evaluate the influence of dietary SBM concentration on growth performance of pigs infected with PRRSV at 35 d of age. Furthermore, we sought to determine the potential for SBM to affect systemic immunological stress induced by PRRSV, as indicated by proinflammatory cytokine secretion, acute-phase protein production, and differential blood cell measurements.

## MATERIALS AND METHODS

The protocol for this experiment was approved by the Institutional Animal Care and Use Committee and the Institutional Biosafety Committee at the University of Illinois.

## Animals and Experimental Design

Sixty-four weanling pigs (32 barrows, 32 gilts; 21 d of age; 7.14 ± 0.54 kg) were obtained from the University of Illinois Swine Research Center and individually housed in a disease containment facility at the University of Illinois for 4 wk (−14 to 14 d postinoculation, DPI) immediately after weaning at 3 wk of age. Upon arrival, 3 consecutive daily intramuscular prophylactic injections of lincomycin (11 mg/kg BW; Zoetis, Florham Park, NJ) were administered as a precautionary measure against bacterial infections. The disease containment facility consisted of 2 halls with 8 chambers (3.34 m2) in each hall. Each chamber was equipped with a high-efficiency particulate air filtration system and plastic-coated expanded metal flooring and was divided into 4 individual pens (0.84 m2). Lights were provided on a 12-h cycle, and temperature was maintained at approximately 26°C. Pigs had ad libitum access to feed and water throughout the trial. A common diet (Table 1) that met or exceeded NRC (2012) nutrient requirements for weaned pigs was provided for 1 wk (−14 to −7 DPI) after placement in the disease containment chambers. At −7 DPI, pigs were weighed and allotted to 1 of 4 uniform treatment groups on the basis of sex, BW, and litter of origin (9 litters were represented).

View Full Table | Close Full ViewTable 1.

Ingredient and calculated composition of experimental diets (as-fed basis)1

 Item Phase 1 (−14 to −7 DPI) Phase 2 (−7 to 14 DPI) Common LSBM HSBM Ingredient, % Corn 35.81 46.13 35.55 Soybean meal 20.00 17.50 29.00 Dried whey 28.35 14.95 14.95 Distiller’s dried grains with solubles 3.00 10.00 10.00 Poultry by-product meal2 — 7.00 7.00 Menhaden fish meal 4.00 — — Spray-dried plasma 4.00 — — Choice white grease 2.38 1.50 1.50 Ground limestone 0.56 0.68 0.60 Monocalcium phosphate 0.18 0.27 0.20 Sodium chloride 0.35 0.40 0.40 Vitamin and mineral premix3 0.30 0.30 0.30 Zinc oxide 0.42 — — Copper sulfate 0.08 — — Choline chloride 0.07 0.07 0.07 l-Lys HCl 0.23 0.60 0.24 dl-Met 0.22 0.27 0.16 l-Trp 0.05 0.08 0.03 l-Thr — 0.15 — l-Val — 0.10 — Calculated composition ME, kcal/kg 3,395 3,402 3,398 Standardized ileal digestible AA, % Lys 1.44 1.38 1.38 Met + Cys 0.87 0.83 0.83 Trp 0.28 0.26 0.26 Thr 0.91 0.86 0.87 Val 0.98 0.91 1.00 Ca, % 0.80 0.80 0.80 Digestible P, % 0.37 0.28 0.28
1All pigs received the common diet for 1 wk from −14 to −7 d postinoculation (DPI) immediately following weaning at 3 wk of age. Pigs were then provided either the low soybean meal (LSBM) or high soybean meal (HSBM) diet for the remainder of the trial.
2Low ash pet-food-grade poultry by-product meal (American Proteins Inc., Hanceville, AL).
3Vitamin-mineral premix provided the following per kilogram of complete diet: vitamin A (retinyl acetate), 11,128 IU; vitamin D3 (cholecalciferol), 2,204 IU; vitamin E (dl-α tocopheryl acetate), 66 IU; vitamin K (menadione nicotinamide bisulfite), 1.42 mg; thiamine (thiamine mononitrate), 0.24 mg; riboflavin, 6.58 mg; pyridoxine (pyridoxine hydrochloride), 0.24 mg; vitamin B12, 0.03 mg; d-pantothenic acid (d-calcium pantothenate), 23.5 mg; niacin (nicotinamide and nicotinic acid), 44 mg; folic acid, 1.58 mg; biotin, 0.44 mg; Cu (copper sulfate), 10 mg; Fe (iron sulfate), 125 mg; I (potassium iodate), 1.26 mg; Mn (manganese sulfate), 60 mg; Se (sodium selenite), 0.3 mg; and Zn (zinc oxide), 100 mg.

Four experimental treatments were arranged as a 2 × 2 factorial of 2 dietary SBM concentrations and 2 PRRSV infection states (uninfected sham or PRRSV infected). There were 16 replications for each of the 4 treatments. The low SBM diet (LSBM) contained 17.5% SBM, and the high SBM diet (HSBM) contained 29.0% SBM (Table 1). The experimental diets were formulated to be isocaloric and contain equal standardized ileal digestible concentrations of Lys, Met, Trp, and Thr. Pigs were allowed to adapt to the experimental diets for 1 wk (−7 to 0 DPI) before being intranasally inoculated with either 2 mL of Dulbecco’s PBS (Corning Cellgro, Manassas, VA; sham control) or a 1 × 105 50% tissue culture infective dose of PRRSV (P-129 isolate, Purdue University, West Lafayette, IN) diluted in 2 mL of Dulbecco’s PBS. Uninfected and PRRSV-infected pigs were housed in separate halls to avoid cross contamination. All personnel were required to completely change into protective clothing before entering the disease containment facility. Traffic flow was directed such that all animal care and experimental procedures were first conducted for uninfected pigs, and reentry of personnel or equipment into the hall containing the uninfected pigs after exposure to PRRSV-infected pigs was prohibited. All personnel exited the facility through a shower after complete removal of protective clothing.

### Diet Analyses

Gross energy of the experimental diets was determined using an adiabatic bomb calorimeter (Parr Instruments, Moline, IL), and DM was determined according to AOAC (2002) method 934.01 (Table 2). Isoflavone concentrations of the experimental diets were determined at the USDA-ARS National Center for Agricultural Utilization Research (Peoria, IL) using HPLC according to the procedures of Berhow et al. (2006). A complete AA profile of the diets was determined at the University of Missouri Agricultural Experiment Station Chemical Laboratory according to AOAC (2002) official methods [982.30 E(a,b,c)].

View Full Table | Close Full ViewTable 2.

Analyzed chemical composition and isoflavone content of experimental diets (as-fed basis)1

 Item LSBM HSBM DM, % 91.39 91.51 GE, kcal/kg 4,084 4,128 CP (N × 6.25), % 22.75 26.65 Amino acids, % Indispensable Arginine 1.26 1.66 Histidine 0.52 0.64 Isoleucine 0.87 1.10 Leucine 1.93 2.27 Lysine 1.58 1.66 Methionine 0.57 0.57 Phenylalanine 0.97 1.22 Threonine 0.95 1.03 Tryptophan 0.35 0.35 Valine 1.10 1.22 Dispensable Alanine 1.21 1.40 Aspartic acid 1.87 2.48 Cysteine 0.31 0.38 Glutamic acid 3.35 4.19 Glycine 1.06 1.28 Proline 1.53 1.76 Serine 0.89 1.08 Tyrosine 0.70 0.86 Isoflavones, mg/kg Total genistein 369 638 Total daidzein 257 513 Total glycitin 76 96 Total isoflavones 700 1,246
1Abbreviations: LSBM = low soybean meal, HSBM = high soybean meal.

### Growth, Rectal Temperatures, and Blood Collection

Individual pig and feeder weights were recorded weekly to allow for calculation of ADG, ADFI, and feed efficiency (G:F). Growth data are reported in reference to the inoculation schedule (−14 to 14 DPI). Rectal temperatures were measured using a digital thermometer at approximately the same time each day for 14 d beginning immediately before inoculation on 0 DPI.

Blood (8 mL total) was collected from the jugular vein of each pig into Vacutainer tubes (BD, Franklin Lakes, NJ) at 0, 3, 7, and 14 DPI. Blood samples from pigs at 0 DPI were collected immediately before inoculation. Blood (3 mL) was collected into tubes containing EDTA, placed on ice, and submitted to the University of Illinois Veterinary Clinical Pathology Laboratory for analysis as described below. Additional blood (5 mL) was collected into serum collection tubes, allowed to clot and centrifuged for at room temperature for 20 minutes at 1,250 × g.

## Blood and Sera Measurements

A multiparameter, automated hematology analyzer (CELL-DYN 3700, Abbott Laboratories, Abbott Park, IL) was used to determine the total and differential cell counts in blood collected into EDTA tubes at the University of Illinois Veterinary Clinical Pathology Laboratory. Serum PRRSV load of samples collected at 0, 3, 7, and 14 DPI was evaluated with real-time PCR after extraction of total nucleic acids using a BioSprint One-For-All Vet kit (Qiagen, Hilden, Germany) following the manufacturer’s instructions. Real-time PCR was conducted at the University of Illinois Veterinary Diagnostic Laboratory using EZ-PRRSV Multiplex reagents (Tetracore, Rockville, MD) using identical extraction conditions and a single quantity of starting RNA material for all experimental samples. Both positive and “no-template” controls were run along with experimental samples. The assay used a single-tube method based on fluorogenic probe hydrolysis (TaqMan, Applied Biosystems, Foster City, CA) . Viral load was expressed as cycle threshold (Ct) values, where a higher Ct value represents a lower amount of PRRSV mRNA. Serum concentrations of interferon-γ, tumor necrosis factor α (TNF-α), IL-10, IL-1β, and haptoglobin were measured in duplicate using sandwich ELISA kits according to the manufacturer’s instructions (cytokines, R&D Systems, Minneapolis, MN; haptoglobin, GenWay Biotech Inc., San Diego, CA). Plasma concentrations of interferon-γ and IL-10 were below minimum detectable levels for all time periods. Intra-assay CV for TNF-α, IL-1β, and haptoglobin were 3.0, 3.8, and 5.1, respectively.

### Statistical Analyses

A repeated measures analysis was conducted, and each individual pig was considered an experimental unit. There were 16 replicate pigs for each of the 4 experimental treatments. Data were subjected to a 3-way ANOVA using PROC MIXED of SAS 9.3 (SAS Inst. Inc., Cary, NC) with the following statistical model:where μ = the general mean, αi = the fixed effect of SBM concentration, βj = the fixed effect of infection status, αβij = the interaction between SBM concentration and infection status, γk = the fixed effect of DPI, αγik = the interaction between SBM concentration and DPI, βγjk = the interaction between infection status and DPI, αβγijk = the 3-way interaction among SBM concentration, infection status, and DPI, and εijkl = experimental error. Because all uninfected pigs were determined to be free of PRRSV, a separate 2-way ANOVA that included only the main fixed effects and interactions of SBM concentration and DPI was conducted to evaluate diet effects on serum PRRS viral load of infected pigs. For all outcomes, the SLICE option of PROC MIXED was used to evaluate the main effects and interaction of SBM concentration and infection status at each level of DPI. Orthogonal contrasts were used to evaluate the effects of SBM concentration for pigs within the same infection group at a given DPI. Statistical significance was considered at P ≤ 0.05, and trends were discussed if 0.05 < P ≤ 0.10.

## RESULTS

Analyzed compositions of the experimental diets are provided in Table 2. In accordance with the objective, total analyzed concentrations of each of the first 4 limiting AA (Lys, Met, Trp, and Thr) were no more than 8.5% greater in the HSBM diet than in the LSBM diet, whereas concentrations of CP and other AA, as well as isoflavones, varied on the basis of the dietary concentration of SBM. The analyzed total isoflavone concentration was 78% higher in the HSBM diet relative to the LSBM diet.

### Growth Response

Average daily gain (P = 0.07) and G:F (P = 0.07) tended be greater for pigs fed HSBM than for pigs fed LSBM during the week before inoculation (Table 3). There were interactive effects (P < 0.01) of dietary SBM and PRRSV infection status on ADFI, ADG, and G:F during the first week after infection. However, orthogonal contrasts indicated no diet effects on growth or feed efficiency within the uninfected or PRRSV-infected groups during this period.

View Full Table | Close Full ViewTable 3.

Effects of dietary soybean meal concentration and porcine reproductive and respiratory virus (PRRSV) infection on growth performance of weanlings pigs1

 P-value Uninfected PRRSV infected Main effects LSBM vs. HSBM2 Item LSBM HSBM LSBM HSBM SEM Diet PRRSV Diet × PRRSV Uninfected PRRSV infected BW, kg −7 DPI 7.87 7.76 7.76 7.89 0.23 0.94 0.97 0.58 0.74 0.65 0 DPI 9.08 9.03 9.00 9.65 0.36 0.39 0.44 0.52 0.91 0.19 14 DPI 17.59 17.01 13.39 14.73 0.35 0.27 <0.01 <0.01 0.23 <0.01 −7 to 0 DPI ADG, g/d 172 184 171 253 26.5 0.07 0.19 0.09 0.75 0.03 ADFI, g/d 479 391 425 478 31.7 0.57 0.60 0.14 0.05 0.23 G:F, g/kg 412 455 403 515 43.8 0.07 0.54 0.24 0.47 0.06 0 to 7 DPI ADG, g/d 507 521 257 295 25.6 0.31 <0.01 <0.01 0.70 0.29 ADFI, g/d 742 695 615 589 30.6 0.23 <0.01 <0.01 0.27 0.54 G:F, g/kg 696 763 433 498 40.8 0.10 <0.01 <0.01 0.24 0.26 7 to 14 DPI ADG, g/d 709 631 370 451 25.6 0.94 <0.01 < 0.01 0.03 0.03 ADFI, g/d 1028 988 568 647 30.6 0.51 <0.01 < 0.01 0.36 0.07 G:F, g/kg 691 643 637 712 40.8 0.74 0.85 0.49 0.40 0.20 0 to 14 DPI ADG, g/d 608 576 314 374 21.8 0.52 <0.01 0.04 0.30 0.06 ADFI, g/d 885 842 592 618 27.6 0.76 <0.01 0.21 0.27 0.50 G:F, g/kg 693 703 535 605 28.4 0.16 <0.01 0.29 0.81 0.09
1Abbreviations: LSBM = low soybean meal, HSBM = high soybean meal, DPI = days postinoculation. Values represent least squares means of 15 or 16 pigs. Pigs received a common diet from −14 to −7 DPI (1 wk postweaning) and were provided either the LSBM or HSBM diet at −7 DPI. Effects of DPI (P < 0.01) and interactions of DPI and infection status (P < 0.01) were observed for ADG, ADFI, and G:F.
2Orthogonal contrasts of pigs fed LSBM vs. pigs fed HSBM within the uninfected or PRRSV-infected group.

Interactive effects (P < 0.01) of dietary SBM concentration and PRRSV infection were observed for ADFI and ADG but not for G:F from 7 to 14 DPI (Table 3). Average daily feed intake of pigs fed HSBM tended to be higher (P = 0.07) than that of pigs fed LSBM in the PRRSV-infected group, with no effects of SBM concentration in the uninfected group. In the uninfected group, ADG was higher (P = 0.03) for pigs fed LSBM than for those fed HSBM, whereas in the PRRSV-infected group, ADG of pigs fed HSBM was greater (P = 0.03) than that of pigs fed LSBM. There were no effects of SBM inclusion or PRRSV infection on G:F from 7 to 14 DPI.

Overall, PRRSV infection reduced (P < 0.01) ADFI, ADG, and G:F during the 2-wk infection period (0 to 14 DPI). No main effects of SBM concentration or interactions between SBM concentration and PRRSV infection were observed for ADFI or G:F from 0 to 14 DPI; however, an interaction (P = 0.04) was observed for ADG. This interaction was due to the tendency of greater (P = 0.06) ADG in pigs fed HSBM over those fed LSBM in the PRRSV-infected group, with no effects of SBM concentration on ADG in the uninfected group. There was also a tendency for greater (P = 0.09) G:F for pigs fed HSBM over those fed LSBM within the PRRSV-infected group, with no difference in G:F of pigs due to SBM concentration in the uninfected group. In the PRRSV-infected group, pigs fed HSBM had higher (P < 0.01) final BW at 14 DPI than pigs fed LSBM, whereas there was no difference between final BW of LSBM- and HSBM-fed pigs in the uninfected group.

### Rectal Temperatures and Serum Viral Load

Porcine reproductive and respiratory syndrome virus infection increased (P < 0.01) rectal temperatures of pigs throughout most of the 14-d infection period, with no influence of dietary SBM concentration (Fig. 1). Real-time PCR indicated that all pigs were free of PRRSV on 0 DPI and that all sham-inoculated pigs remained PRRSV free throughout the trial (data not shown). All infected pigs were PRRSV positive at 3, 7, and 14 DPI. There was no influence of dietary SBM concentration on serum viral load at 3 or 7 DPI, but Ct values indicated a reduced (P = 0.04) serum PRRSV load in HSBM-fed pigs compared with LSBM-fed pigs at 14 DPI (Fig. 2).

Figure 1.

Main effect of porcine reproductive and respiratory syndrome virus (PRRSV) infection status on rectal temperatures of pigs during the 14-d infection period. Values represent least squares means of 31 or 32 pigs, and error bars indicate SEM. Interactive effects of DPI and infection status (P < 0.01) were observed for rectal temperatures. An asterisk (*) denotes a difference (P ≤ 0.05) due to infection status at a given time point.

Figure 2.

Serum viral load of porcine reproductive and respiratory syndrome virus-infected (PRRSV) pigs fed low (LSBM) or high (HSBM) soybean meal diets as determined by real-time reverse transcription PCR. Serum viral load is presented in cycle threshold (Ct) values, which are inversely related to the amount of viral RNA detected. Values represent least squares means of 15 or 16 pigs, and error bars indicate SEM. Viral load was influenced by SBM concentration (P < 0.05) and DPI (P < 0.01). An asterisk (*) denotes a difference (P ≤ 0.05) due to SBM concentration at a given time point.

### Serum Haptoglobin and Proinflammatory Cytokines

At 3 DPI, interactions (P < 0.01) between dietary SBM concentration and PRRSV infection on serum haptoglobin, TNF-α, and IL-1β concentrations were observed (Table 4). There was no difference in serum haptoglobin concentration between pigs fed LSBM or HSBM in the uninfected group, but greater (P = 0.03) haptoglobin concentrations were observed in pigs fed LSBM than in those fed HSBM within the PRRSV group. Except for a trend for greater (P = 0.06) TNF-α concentration in pigs fed HSBM within the uninfected group, there were no differences in serum cytokine concentrations due to dietary treatment within either infection group.

View Full Table | Close Full ViewTable 4.

Effects of dietary soybean meal concentration and porcine reproductive and respiratory virus (PRRSV) infection on haptoglobin and cytokine production in weanlings pigs1

 P-value Uninfected PRRSV infected Main effects LSBM vs. HSBM2 Item LSBM HSBM LSBM HSBM SEM Diet PRRSV Diet × PRRSV Uninfected PRRSV infected 3 DPI HAP, µg/mL 979 1,311 2,163 1,363 262 0.36 0.02 0.01 0.36 0.03 TNF-α, pg/mL 85.1 105.5 181.7 175.1 7.5 0.36 <0.01 <0.01 0.06 0.53 IL-1β, pg/mL 0.0 0.0 12.9 8.5 3.0 0.45 <0.01 <0.01 1.00 0.28 7 DPI HAP, µg/mL 1,180 864 2,062 1,657 262 0.16 <0.01 <0.01 0.38 0.26 TNF-α, pg/mL 59.8 63.4 161.3 150.1 8.0 0.62 <0.01 <0.01 0.74 0.30 IL-1β, pg/mL 0.0 0.0 14.7 20.8 3.0 0.30 <0.01 <0.01 1.00 0.15 14 DPI HAP, µg/mL 263 315 1,860 1,857 283 0.93 <0.01 <0.01 0.89 0.99 TNF-α, pg/mL 52.9 55.8 273.1 218.4 10.6 <0.01 <0.01 <0.01 0.79 <0.01 IL-1β, pg/mL 0.0 1.5 22.2 26.6 2.9 0.31 <0.01 <0.01 0.72 0.29
1Abbreviations: LSBM = low soybean meal, HSBM = high soybean meal, DPI = days postinoculation, HAP = haptoglobin, TNF-α = tumor necrosis factor-α. Values represent least squares means of 15 or 16 pigs. Effects of DPI (P < 0.01) and interactions of DPI and infection status (P < 0.01) were observed for TNF-α and IL-1β. Time (DPI) tended to influence HAP (P = 0.07), with an interactive effect (P < 0.05) of DPI and infection also observed.
2Orthogonal contrasts of pigs fed LSBM vs. pigs fed HSBM within the uninfected or PRRSV-infected group.

At 7 DPI, interactions (P < 0.01) between dietary SBM concentration and PRRSV infection were detected for haptoglobin, TNF-α, and IL-1β (Table 4). However, orthogonal contrasts between LSBM and HSBM groups within infection groups indicated no differences for any of these analytes. Interactions between dietary SBM concentration and infection status were also observed (P < 0.01) at 14 DPI for haptoglobin, TNF-α, and IL-1β. There were no differences in haptoglobin or IL-1β concentrations between LSBM- and HSBM-fed pigs within either infection group. Within the PRRSV-infected group at 14 DPI, serum TNF-α concentration was lower (P < 0.01) for pigs fed HSBM compared with those fed LSBM.

### Differential Blood Cell Counts

There were interactive effects of dietary SBM concentration and PRRSV infection on total white blood cells (WBC; P = 0.03), lymphocytes (LYM; P < 0.01), monocytes (MONO; P < 0.01), and the neutrophil (NEU) to lymphocyte ratio (NEU:LYM; P < 0.01) at 3 and 7 DPI (Table 5). At 14 DPI, interactive effects (P < 0.01) of PRRSV infection and SBM concentration were observed on WBC and NEU. In addition, LYM of PRRSV-infected pigs were restored to values similar to those of the uninfected pigs, and the NEU:LYM ratio (P = 0.04) was increased in PRRSV-infected pigs. In the uninfected group, MONO were greater (P = 0.05) in LSBM-fed pigs than in HSBM-fed pigs at 14 DPI, but there was no difference in MONO between dietary groups in PRRSV-infected pigs.

View Full Table | Close Full ViewTable 5.

Effects of dietary soybean meal concentration and porcine reproductive and respiratory virus (PRRSV) infection on blood leukocyte counts in weanlings pigs1

 P-value Uninfected PRRSV infected Main effects LSBM vs. HSBM2 Item LSBM HSBM LSBM HSBM SEM Diet PRRSV Diet × PRRSV Uninfected PRRSV infected 3 DPI WBC, 103/µL 17.52 20.55 15.55 15.15 1.54 0.39 0.05 0.03 0.17 0.84 NEU, 103/µL 8.63 10.27 11.73 10.13 0.96 0.99 0.12 0.15 0.23 0.23 LYM, 103/µL 7.72 8.97 3.26 3.49 0.88 0.40 <0.01 <0.01 0.32 0.85 MONO, 103/µL 0.81 0.93 0.29 0.43 0.10 0.21 <0.01 <0.01 0.42 0.32 NEU/LYM 1.20 1.25 3.00 2.92 0.17 0.93 <0.01 <0.01 0.83 0.76 7 DPI WBC, 103/µL 19.99 21.09 12.60 12.44 1.54 0.75 <0.01 <0.01 0.60 0.94 NEU, 103/µL 8.49 8.49 7.22 6.86 1.00 0.84 0.12 0.48 1.00 0.78 LYM, 103/µL 9.96 10.68 4.82 4.78 0.85 0.69 <0.01 <0.01 0.56 0.98 MONO, 103/µL 0.55 0.85 0.23 0.27 0.10 0.10 <0.01 <0.01 0.05 0.78 NEU/LYM 0.91 0.86 1.62 1.49 0.17 0.59 <0.01 <0.01 0.84 0.56 14 DPI WBC, 103/µL 19.04 17.36 23.87 23.18 1.54 0.44 <0.01 <0.01 0.44 0.74 NEU, 103/µL 6.48 5.35 9.51 9.71 0.96 0.62 <0.01 <0.01 0.40 0.88 LYM, 103/µL 11.71 10.50 11.21 12.13 0.91 0.87 0.52 0.61 0.34 0.45 MONO, 103/µL 0.58 0.28 0.63 0.64 0.11 0.16 0.05 0.07 0.05 0.96 NEU/LYM 0.56 0.52 0.79 0.97 0.17 0.69 0.04 0.20 0.87 0.46
1Abbreviations: LSBM = low soybean meal; HSBM = high soybean meal; DPI = days postinoculation; WBC = white blood cells; NEU = neutrophils; LYM = lymphocytes; MONO = monocytes. Values represent least squares means of 15 or 16 pigs. Effects of DPI (P < 0.01) and interactions of DPI and infection status (P < 0.01) were observed for WBC, NEU, LYM, MONO, and NEU/MONO.
2Orthogonal contrasts of pigs fed LSBM vs. pigs fed HSBM within the uninfected or PRRSV-infected group.

There were minimal effects of dietary SBM concentration or PRRSV infection on red blood cell (RBC) measurements at 3 DPI, with the exception of a tendency (P = 0.07) for more total RBC in pigs fed LSBM (Table 6). At 7 DPI, there were interactive (P < 0.01) effects of SBM concentration and PRRSV infection on hemoglobin content, as well as a tendency (P = 0.06) for reduced hematocrit in PRRSV-infected pigs. By 14 DPI, total RBC, hemoglobin, and hematocrit values were reduced (P < 0.01) in PRRSV-infected pigs, but these reductions were dependent on dietary SBM concentration. Within the PRRSV group, pigs fed HSBM had a greater (P = 0.02) percentage of hematocrit and a tendency (P = 0.09) for higher hemoglobin content compared with pigs fed LSBM.

View Full Table | Close Full ViewTable 6.

Effects of dietary soybean meal level and porcine reproductive and respiratory virus (PRRSV) infection on red blood cell measurements in weanlings pigs1

 P-value Uninfected PRRSV infected Main effects LSBM vs. HSBM2 Item LSBM HSBM LSBM HSBM SEM Diet PRRSV Diet × PRRSV Uninfected PRRSV infected 3 DPI RBC, × 106/µL 7.16 7.00 7.32 6.99 0.13 0.07 0.53 0.25 0.40 0.08 HGB, g/dL 11.32 11.67 11.64 11.41 0.21 0.78 0.91 0.58 0.24 0.44 HCT, % 36.44 37.66 38.21 37.36 0.67 0.78 0.27 0.30 0.20 0.36 7 DPI RBC, × 106/µL 7.01 6.85 7.02 6.86 0.13 0.22 0.92 0.68 0.39 0.39 HGB, g/dL 11.71 12.08 11.21 11.31 0.21 0.25 <0.01 <0.01 0.20 0.73 HCT, % 37.21 38.12 36.22 36.67 0.67 0.29 0.06 0.18 0.32 0.62 14 DPI RBC, × 106/µL 7.26 7.12 6.04 6.25 0.13 0.79 <0.01 <0.01 0.47 0.27 HGB, g/dL 12.82 13.16 10.19 10.70 0.21 0.04 <0.01 <0.01 0.25 0.09 HCT, % 42.01 42.67 32.71 34.84 0.72 0.04 <0.01 <0.01 0.49 0.02
1Abbreviations: LSBM = low soybean meal; HSBM = high soybean meal; DPI = days postinfection; RBC = red blood cells; HGB = hemoglobin; HCT = hematocrit. Values represent least squares means of 15 or 16 pigs. Effects of DPI (P < 0.01) and interactions of DPI and infection status (P < 0.01) were observed for RBC, HGB, and HCT.
2Orthogonal contrasts of pigs fed LSBM vs. pigs fed HSBM within the uninfected or PRRSV-infected group.

## DISCUSSION

Soybean meal contains a number of naturally occurring bioactive components. In particular, isoflavones present in SBM have been shown to exert antiviral activity against PRRSV and other viruses, both in vitro and in vivo (Greiner et al., 2001a; Andres et al., 2009). Thus, the feeding of HSBM diets could serve as an immunomodulatory strategy for swine producers in response to a PRRSV infection. In the current experiment, pigs fed a diet containing 29% SBM generally maintained better growth performance over those fed a diet containing 17.5% SBM during an acute (14 d) PRRSV infection. In addition, pigs fed HSBM had reduced serum concentrations of inflammatory biomarkers compared with those fed LSBM at 3 and 14 DPI (haptoglobin and TNF-α, respectively), as well as a reduced serum viral load at 14 DPI. Although the mode of action is unclear, the results of this study suggest that feeding a HSBM diet may help to minimize immune stress and maintain the growth of nursery pigs suffering from a PRRSV infection.

Soybean meal is often limited in nursery diets because of the sensitivity of weanling pigs to its antinutritional components (i.e., phytate and antigenic proteins) that can induce diarrhea and reduce the feed intake and growth of young pigs (Li et al., 1990). However, pigs fed the HSBM diet in the current study tended to have greater ADG and G:F than pigs fed LSBM during the week before inoculation, which suggests that a practical balance must be achieved between the negative and positive elements of feeding a HSBM diet to nursery pigs. Sporadic, moderate diarrhea was noted for some pigs in all treatment groups during the first 2 wk of placement in the disease containment facility. Consequently, several pigs in each treatment group lost weight or continued to have poor weight gain from −7 to 0 DPI. A higher ADG (253 g/d) and numerically greater G:F indicated that this condition may have been minimized for pigs allotted to the PRRSV-infected, HSBM-fed group. Therefore, the superior growth of pigs fed HSBM within the PRRSV group during the preinfection period must be taken into consideration when interpreting growth response data during the postinfection period.

All pigs in the uninfected group tested negative for PRRSV and were free of clinical symptoms associated with PRRS throughout the trial. Conversely, pigs in the PRRSV-infected group had increased rectal temperatures and decreased appetites and were all PRRSV positive, as indicated by real-time quantitative PCR during the first week of infection. The impact of PRRSV infection on the growth of pigs was slightly less severe but comparable to that in previous studies conducted at our facility using similar infection models. In the current experiment, PRRSV infection reduced overall ADG by 42% and G:F by 18%, whereas Che et al. (2011) and Liu et al. (2013) reported reductions of 59% and 47% for ADG and 29% and 33% for G:F, respectively. Total mortality of pigs was approximately 3% in the current study (2 pigs total), with no impact of PRRSV infection (1 uninfected and 1 PRRSV-infected).

The observed improvements in growth of pigs fed HSBM in the current experiment corroborate the limited previous reports of feeding increased dietary SBM concentrations to pigs infected with PRRSV. Johnston et al. (2010) reported that finisher pigs fed a diet containing 32% SBM had greater weight gain and feed efficiency than pigs fed a 21% SBM diet during a naturally occurring coinfection of pigs with PRRSV and porcine circovirus. During an experimental PRRSV infection, Rocha et al. (2013) determined that nursery pigs fed a diet with 22.5% SBM had lower rectal temperatures and improved feed efficiency compared with pigs fed a 12.5% SBM diet during the first week after infection, but effects were limited to only this period. Measurements in addition to growth in the current study provide further insight into potential mechanisms by which a high SBM diet might ameliorate detriments due to PRRS.

The peak serum PRRSV load at 7 DPI in pigs in the current study is supported by previous reports of peak serum PRRSV concentration within the first 10 d after infection (Greiner et al., 2000; Che et al., 2011, 2012; Liu et al., 2013). There was no influence of dietary SBM concentration on serum viral load during the initial or peak phases of infection, but the lower viral load at 14 DPI in pigs fed the HSBM diet indicates that the HSBM diet may have supported more efficient virus elimination during initial stages of recovery. Considering that serum PRRSV concentration is negatively correlated with feed intake (Greiner et al., 2000), the reduced serum PRRSV load at 14 DPI may have contributed to the trends for greater ADFI and improved ADG in pigs fed the HSBM diet from 7 to 14 DPI.

After recognition of pathogenic stimuli, innate immune cells produce cytokines that induce inflammation and coordinate the adaptive immune response to establish host protection. However, prolonged production of proinflammatory cytokines is detrimental to animal growth, largely because of the anorectic nature of these molecules (Johnson, 1997). Previous reports on the characterization of proinflammatory cytokine kinetics in PRRSV-infected pigs have been inconsistent, and the atypical cytokine response induced by PRRSV has been suggested to be an evolutionary mechanism to evade antiviral immune responses (Liu et al., 2010; Sang et al., 2011). In the current study, infection of pigs with PRRSV led to elevated serum concentrations of IL-1β and TNF-α, most notably at 14 DPI. Considering the 4.5-fold increase in TNF-α due to PRRSV infection at 14 DPI, it was interesting that the serum concentration of this cytokine was 20% lower in pigs fed the HSBM diet at this time point.

There was a marked PRRSV-induced increase in the serum concentration of the acute-phase protein, haptoglobin, at 3 DPI for pigs fed LSBM compared with those fed HSBM. Blood haptoglobin concentration has previously been shown to increase in pigs during PRRSV infection, although the timing, magnitude, and duration of its increased synthesis are variable (Asai et al., 1999; Gnanandarajah et al., 2008; Che et al., 2011, 2012; Liu et al., 2013). Haptoglobin is synthesized by hepatocytes in response to proinflammatory cytokines, with the most potent regulator being IL-6 (Wang et al., 2001). Although serum cytokine concentrations of pigs within the PRRSV group were not influenced by diet at 3 DPI, the lower serum haptoglobin concentrations of pigs fed the HSBM diet may indicate reduced inflammation at this time point. The effect of dietary SBM concentration on serum haptoglobin concentration of pigs within the PRRSV group was diminished by 7 DPI.

Cytokine production induced by PRRSV infection may also impact production of myeloid or erythroid progenitor cells, as well as the subsequent maturation of leukocytes and erythrocytes (Halbur et al., 2002). The transient leukopenia caused by PRRS at 7 DPI and subsequent increase in WBC count at 14 DPI for PRRSV-infected pigs in this experiment are in agreement with previous reports (Halbur et al., 2002; Toepfer-Berg et al., 2004; Liu et al., 2013). The PRRS-induced anemia as indicated by reduced total RBC, hemoglobin, and hematocrit values at 14 DPI also supports previous findings (Halbur et al., 2002; Liu et al., 2013). The interaction between dietary SBM concentration and PRRSV infection on RBC measurements was mostly limited to 14 DPI, where the reduction in hematocrit and hemoglobin values elicited by PRRS was less severe for pigs fed HSBM than for those fed LSBM. Because the severity of anemia is likely a function of PRRSV virulence (Halbur et al., 2002), the improved hematocrit and tendency for greater hemoglobin values for pigs fed HSBM at 14 DPI may have been related to reduced viral load or inflammation (i.e., serum TNF-α) observed at this time point.

Soybeans and soybean-derived feedstuffs are the richest sources of the isoflavones genistein, daidzein, and glycitein (Wang and Murphy, 1994), which are reported to have anti-inflammatory and antiviral activity through various mechanisms (Hämäläinen et al., 2007; Andres et al., 2009). It is possible that the immunomodulatory effects and improved growth of PRRSV-infected pigs fed HSBM in the current study can be partly explained by the concomitant increase in dietary isoflavone concentrations. Dietary genistein concentrations of the LSBM and HSBM diets in the current experiment were 369 and 638 mg/kg, respectively. Greiner et al. (2001a) reported a quadratic response in ADFI for PRRSV-infected pigs fed diets ranging from 0 to 800 mg/kg total genistein, with the greatest ADFI and ADG observed in pigs fed 200 mg/kg genistein. Furthermore, a linear decrease in serum PRRSV concentration and a quadratic decrease in interferon-γ were observed with increasing dietary genistein concentration, whereas spleen size increased linearly. In a separate experiment using the same model, daidzein had no influence on the outcomes with the exception of a linear increase in spleen size (Greiner et al., 2001b). Because bioavailability of isoflavones is influenced by the feed matrix (Cassidy et al., 2006), it is likely that the isoflavones contributed by SBM in the current study were less bioavailable than those provided in a supplemental, purified form by Greiner et al. (2001a,b). However, Kuhn et al. (2004) reported that plasma genistein concentrations of pigs increased with dietary SBM concentration, suggesting that SBM isoflavones are, indeed, bioavailable for growing pigs. Additionally, both synergisms and antagonisms exist among bioactive soy components, and these relationships have not yet been characterized in vivo for feed ingredients such as SBM (Dia et al., 2008).

Diets in the current study were balanced to contain similar concentrations of standardized ileal digestible Lys, Met, Trp, and Thr. Beyond these AA, the CP concentration of the HSBM diet resulted in an average increase of 25% for both indispensable and dispensable AA. The greatest increase was for Arg, followed by Ile and Phe, which were 32%, 26%, and 26% higher, respectively, in the HSBM diet than in the LSBM diet. Activation of the immune system may alter AA requirements (Klasing, 1988; Reeds et al., 1994), and the surfeit AA supply of the HSBM diet may have supported the increased demand for specific AA for the production of immune-specific molecules during the PRRSV infection. In addition, inflammatory cytokine secretion following immune activation induces a metabolic shift in which body protein is catabolized and liberated AA are subsequently deaminated and oxidized via gluconeogenesis (Klasing, 1988). Thus, excess AA in the HSBM diet may have provided more gluconeogenic substrates than the LSBM diet, but this speculation does not consider the metabolic cost associated with elimination of greater quantities of AA-associated nitrogen provided by the HSBM diet.

In conclusion, the immune response and growth performance of nursery pigs were differentially influenced by dietary SBM concentration during an acute PRRSV infection. In general, pigs fed a diet containing 29% SBM had improved growth and reduced levels of immune stress biomarkers compared with pigs fed a diet containing 17.5% SBM during the 2-wk infection period. Although previous research suggests that isoflavones have the potential to ameliorate PRRSV infection, the current study was unable to definitively discern whether the beneficial effects of feeding the HSBM diet to PRRSV-infected weanling pigs were due to isoflavones, AA, other bioactive components, or a combination. Therefore, further studies are warranted to investigate potential immunomodulatory components of SBM for pigs during PRRSV and other inflammatory challenges.