Impact of dietary vitamin D₃ supplements in nursery diets on subsequent growth and bone responses of pigs during an immune challenge

Abstract

Limited evidence is available to validate beneficial responses from extra nutrient supplements for mediation of growth suppression that results from immune challenges. Extrarenal roles of vitamin D metabolites in immune function implicate vitamin D₃ supplements as a nutrient for potential beneficial effects. The current objective was to assess growth and bone ash responses to dietary vitamin D₃ (D) supplements for growing pigs undergoing an immune challenge. At weaning, 216 crossbred pigs (4 pigs/pen, 6 pens/treatment) were randomly allotted within sex and weight blocks to 1 of the 9 treatments. Treatments included D supplements (0, 100, or 800 IU/kg) in a factorial arrangement with 3 vaccine (V) protocols; no injection (0 × V), a single 2 mL injection of a Lawsonia intracellularis vaccine at day 14 (1 × V), or 2 mL injections of the same vaccine at days 0 and 7 (2 × V). An adjustment diet with no supplemental D was fed for 1 wk, then assigned D diets for 2 wk (P2). After P2, all pigs were phase-fed standard diets (D = 280 IU/kg) to assess subsequent growth to 115 kg. No differences due to D supplements or vaccination protocol were detected in ADG (0.233 ± 0.021 kg/d) or GF (0.642 ± 0.028 kg/d) over the 21-d nursery trial; however, ADFI was lower (P < 0.10) in pigs fed D levels of 0 vs. 100 and 800 (0.340 vs. 0.375, 0.372 ± 0.027 kg/d). Bone mineral content (g) from whole-body dual energy X-ray absorptiometry scans at 9 wk (n = 4 pigs/treatment) was lower in pigs fed 0 vs. 100 and 800 IU of D (287 vs. 325, 323 ± 34.1 g/pig). Growth from nursery to 115 kg was lower (P < 0.01) in pigs fed D levels of 0 vs.100 and 800 (0.828 vs. 0.876, 0.889 ± 0.021 kg/d). At market, approximately two-thirds of pigs showed positive L. intracellularis serology titers regardless of treatment. Limited evidence for D-mediation of an immune challenge using the vaccination protocols may be a consequence of limited vaccine effects on growth in the nursery and seroconversion of most pigs to L. intracellularis by market.

Introduction

Implications of vitamin D deficiency with autoimmune diseases (Nieves et al., 1994; Talat et al., 2010) and with skeletal deformities (Deluca, 2004; Dusso et al., 2005; Plum and DeLuca, 2010; Rortvedt and Crenshaw, 2012) have generated questions about whether additional supplemental vitamin D is required to meet the needs of an immune challenge beyond the amounts needed for growth and Ca homeostasis. In production animals, questions arise on whether dietary resources are shifted to sustain immune function, thereby compromising growth and bone ash accumulation, or if additional supplemental vitamin D is needed to enhance immune function (Arnold et al., 2015). Continued efforts to supplement mega-doses of vitamin D have not prevented health issues from arising in pigs. Additionally, the prolonged effects of mega-doses of vitamin D supplementation in weaned pigs on bone development are poorly characterized. Even less characterized is the relationship between vitamin D supplementation and interactions on the immune system. Furthermore, the absence of animal studies, which demonstrate the effects of overall herd health on nutrient requirements at the time of preventative treatments, has provided an opportunity for further exploration. Thus, the objective of this experiment was to assess growth and bone ash responses to dietary vitamin D₃ (D) supplements in growing pigs under control or immune challenges.

Materials and Methods

Ethics Statement

All animal procedures were approved (Protocol A005674) by the College of Agricultural and Life Sciences Animal Care and Use Committee, University of Wisconsin–Madison. The experiment was completed at the University of Wisconsin–Madison Swine Research and Teaching Center (SRTC); a facility accredited by the Association for the Assessment and Accreditation of Laboratory Animal Care. Animals selected for tissue collections were humanely euthanized by electrical stunning followed by exsanguination.

Experimental Design and Nursery Diets

The current experiment was designed to determine whether D supplements alter growth responses to Lawsonia intracellularis vaccination protocols over a 3-wk nursery period. Subsequent growth of pigs to market weight (115 kg) was evaluated for potential carry-over effects of the nursery treatments. Pigs were produced and housed at SRTC until shipment at market weight.

At weaning, 216 crossbred F1 (n = 24, Large White × Landrace) and F2 (n = 192, F1 × PIC Line 19) gilts and barrows were selected based on uniformity of weight across multiple litters within a contemporary farrowing group. Breed and sex (n = 72 gilts and 144 barrows) were distributed across treatment groups. Pigs were weaned at 23 ± 3 d of age (3 wk) and averaged 7.8 kg. At weaning, pigs were randomly assigned to 1 of the 3 dietary treatment groups and one of 3 vaccination protocols (Fig. 1). The 3 diets and 3 vaccination protocols were arranged as a 3 × 3 factorial design to provide 9 treatments. Pigs were fed an adjustment diet with no supplemental D (phase 1) for 1 wk, prior to being fed treatment diets (phase 2) for 2 wk (Table 1). Nursery treatment (phase 1) diets were formulated to supply 120% of the Ca and P requirements for 5 to 10 kg pigs (NRC, 1998). Nursery treatment (phase 2) diets were formulated to supply 0, 100, or 800 IU vitamin D₃/kg, each with 100% of the Ca and P requirements for 10 to 20 kg pigs. After the 2 wk dietary treatment period, all pigs were fed standard SRTC diets, which were supplemented with D (280 IU/kg) for pigs at each production phase until market weight at 115 kg. Dietary Ca, P, and other nutrients were supplied in diets for each production phase at concentrations that met or exceeded recommended concentrations. Diets and nutrient formulations were the same as that of standard diets fed to pigs at SRTC during the grower and finisher phases.

Figure 1.

Timeline for applications of procedures from weaning until the end of the nursery phase and until pigs were marketed. Pigs were administered no injection (0 × V), a single injection (1 × V) on day 14, or 2 injections (2 × V) on days 0 and 7 of 2 mL L. intracellularis vaccine. All pigs were fed diets with no supplemental vitamin D₃ for 7 d after weaning, and then were fed diets with 0, 100, or 800 IU/kg vitamin D₃ from days 7 to 21. Vaccination protocols and vitamin D₃ supplemental diets were designed in a 3 × 3 factorial arrangement of treatments. All pigs were fed diets with 280 IU/kg vitamin D₃ from day 21 to market weight. Subsets of barrows from treatment groups were selected for whole-body dual energy X-ray absorptiometry (DXA) scans to determine bone mineral content on day 37.

Table 1.

Ingredients and nutrient composition of phases 1 and 2 nursery diets (as-fed basis)¹

Ingredients, g/kg	Phase 1	Phase 2
Corn grain	384.24	658.42
Soybean meal, 48% crude protein	295.73	298.28
Oat groats	150.00	—
Whey	100.00	—
L-Lysine HCl	2.50	1.00
Methionine	1.00	—
Threonine	0.65	—
Monocalcium phosphate	15.78	9.87
Limestone	10.10	8.43
Corn oil	20.00	10.00
Sodium chloride	5.00	4.00
VTMM-G without D32,3	15.00	10.00
Total	1000.0	1000.0
Calculated analysis 4
Ca, %	0.96	0.70
P, %	0.78	0.60

¹Diets were formulated to supply 120% of the Ca and P requirements (NRC, 1998) for 5 to 10 kg (phase 1) and 100% of the Ca and P requirement for 10- to 20-kg (phase 2) pigs. Supplemental vitamin D₃ was not added to phase 1 diet. The Ca and P requirements for 5 to 10 kg (phase 1) are 0.80% and 0.65%, respectively. For 10 to 20 kg (phase 2), Ca and P requirements are 0.70% and 0.60%, respectively.

²VTMM-G without vitamin D₃ is a custom-mixed vitamin–trace mineral mix for growth without vitamin D₃ which provides the following nutrients per kilogram of diet: vitamin A, 2,800 IU; vitamin D₃, 0 IU; vitamin E, 14 IU; vitamin K, 0.75 IU; niacin, 22 mg; pantothenic acid, 12 mg; riboflavin, 8 mg; vitamin B₁₂, 33 μg; Cu (from copper sulfate), 1.5 mg; I (from ethylene-diamine dihydroiodide), 0.3 mg; Fe (from ferrous sulfate), 38 mg; Se (from sodium selenite), 0.2 mg; and Zn (from zinc oxide), 90 mg. In phase 1 diets, the nutrients provided were 1.5 times the amounts indicated.

³Phase 2 diets included Rovimix D₃ (500,000 IU/g) added to the premix to supply either 100 or 800 IU/kg vitamin D₃ per kilogram of complete diet for the respective dietary treatments.

⁴Calculated concentration of nutrients was based on nutrient compositions of feed ingredients in NRC (1998).

The 3 vaccine injection protocols were randomly assigned to groups at weaning. Vaccination protocols (Fig. 1) consisted of no injection (0 × V), a single 2 mL injection of a L. intracellularis vaccine at 2 wk after weaning (1 × V), or a 2 mL injection of a L. intracellularis vaccine at weaning (d-0) and 1 wk after weaning (2 × V). The L. intracellularis vaccine was a commercial product (Porcilis Ileitis, Merck Animal Health) that had been used for 1 year prior to this experiment in our research herd for prevention of ileitis.

At weaning, pigs (n = 4 pigs/pen) were housed in nursery pens (1.5 × 0.9 m) for the duration of a 3-wk nursery trial. Pigs were allowed ad libitum access to their assigned diets and water throughout the nursery phase. After the nursery phase, pigs were housed in grower-finishing pens in 2 production wings of SRTC to accommodate animal flow. Housing allowed 5 to 6 pigs per pen in grower pens (2.9 × 1.4 m) or 9 to 11 pigs/pen in finisher pens (3.5 × 1.5 m). During the grower-finishing phase, pigs were sorted by sex and locations, which resulted in mixing of pigs across the 9 nursery treatments. Feed consumption was not recorded during the grower-finisher phases, only BW. Pigs were allowed ad libitum access to water and standard SRTC diets for growing pigs at the appropriate stage of production. The SRTC is an environmentally controlled building with no direct sunlight exposure during any stage of the life cycle. Fluorescent lights were controlled to provide cycles of 12-h light, 12-h dark during all production phases.

Pig BW were recorded at weaning, and then at weekly intervals during the 3-wk nursery period. Feed consumption was recorded at weaning and at weekly intervals through the end of the nursery phase (D21). After the nursery phase, feed consumption was not recorded; however, BW were recorded at monthly intervals until pigs were marketed.

Whole Body, Femur, and Front Foot Bone Ash Measurements

At 9 wk of age, 36 barrows were selected for whole-body dual energy X-ray absorptiometry (DXA) scans and femur collections. Pigs were selected from the 2 lightest (1 and 2) and heaviest (5 and 6) weight blocks assigned at allotment. The average weight of 1 male pig per pen was chosen from the 4 weight blocks. The lightest and heaviest weight blocks were intentionally chosen to include pigs that were below or above the average in growth, and thus provide a wide range in weight distribution across the 6 weight blocks. Selected male pigs were euthanized prior to scans and tissue collections. The entire euthanized pig was scanned in a ventral position using DXA (software version 12.20; GE Lunar Prodigy, Waukesha, WI) while in rigor as previously described (Amundson et al., 2016). After whole-body DXA scans, femurs were excised and separately scanned using DXA (small animal scan mode). Analysis of the entire pig and excised femur scans were used to determine bone mineral content (BMC, g bone ash per pig or femur) and bone mineral density (BMD, g bone ash/cm² of skeletal tissue).

At market weight, ~115 kg, pigs were transported to a commercial slaughter plant (JBS Swift, Marshalltown, IA) in 2 separate groups. The 2 groups were chosen based on the final BW of all pigs at 126 d after weaning. The first group included the heaviest 93 pigs at day 126 after weaning. The second group included the remaining 82 pigs, which were shipped 2 wk after the first group. At the facility, the front left feet of the remaining pigs (n = 175) were collected from carcasses for DXA scans. Live animal tattoos, applied at SRTC, were used to tract individual identity of pigs through the slaughter facility. Thus, identity of the collected feet could be maintained for DXA scans. Five pig feet were lost during the slaughter process (2 pig feet were lost on the processing line and 3 feet were not identified as the carcasses were trimmed, which resulted in loss of the unique tattoo identifier for those pigs). Intentions were to cut each foot at the plant to include all the metacarpal bones and the first and second carpal bones. Due to cutting errors when feet were removed, 20 feet were deleted from the DXA analysis, as all carpal bones were missing. The individual pig foot was positioned and scanned as described earlier (Hoffman et al., 2007) to accurately determine the bone ash content of the entire foot, expressed as BMC, gram per entire foot.

Kyphosis Scores

Subjective scores were collected by 2 observers on pigs that remained at 4 mo of age to assess evidence of abnormal outward curvature of the spinal column. Observers were blinded to the treatment groups as scores were assigned. A score of 1 represented no evidence of curvature, and a score of 3 represented clear evidence of abnormal spinal column curvature. An intermediate score 2 was assigned to pigs with marginal curvature. The final assessments of the prevalence of kyphosis were based only on animals which were assigned a score of 3 by both observers as reported previously (Rortvedt and Crenshaw 2012; Amundson et al., 2017).

Serology and Analysis

Blood samples were collected by venipuncture in the brachial region of pigs 21 d after the last vaccination during the nursery period (first injection for pigs vaccinated 1× and the second injection for pigs vaccinated 2×). A second sample was collected from the same pigs at 1 wk before market for the first group and 3 wk before market for the second group. At each collection period, pigs were restrained with a snare, and 10 mL of blood were collected into vacutainer tubes with no anticoagulant. Serum was collected after centrifugation (735 × g for 20 min). Aliquots were stored at 0 °C until shipment to the University of Minnesota Veterinary Diagnostic Laboratory for L. intracellularis serology analysis.

An immunoperoxidase monolayer assay (IPMA; Guedes et al., 2002a, 2002b) was used to determine L. intracellularis titers in serum samples. For the first group of serum samples, if positive titers were detected, a titration method was used to determine the level of titers present. Serum samples from the second blood collections were classified as either positive or negative L. intracellularis titers using the same IPMA assay. A 1:30 dilution value was used to identify a positive sample diagnosis. If the controls were not positive or negative at designated titers, the test was repeated with fresh reagents.

Statistical Analysis

Data were analyzed by regression analysis using mixed model procedures (SAS Inst. Inc., Cary, NC). The main effects of diet and vaccination protocol were analyzed as a 2-way ANOVA, which included the 9 treatments arranged as a 3 × 3 factorial design and initial pig weight block effects. During the nursery phase, the pen was considered the experimental unit for ADG, ADFI, and GF responses. Analysis of DXA scans of pigs, femurs, and the front feet was based on the individual pig as the experimental unit. As pigs were combined into a larger number of pigs per pen and mixed across sex and nursery treatments for the grower and finisher phases, the BW and ADG data for post-nursery responses were analyzed with mixed model procedures using the individual pig as the experimental unit. Feed consumption data were not recorded after the nursery trial. Weight block was included in all analysis as a random effect.

Serology and kyphosis scores, traits which involved categorical scores, were analyzed by RANK procedures to adjust the variables that were not normally distributed. Inferences for these traits were assessed using nonparametric one-way ANOVA procedures based on the Kruskal–Wallis test (Proc NPAR1Way Wilcoxon; SAS Inst. Inc., Cary, NC).

Results were reported as the average for each of the 9 treatment groups with a pooled SEM. The pooled SEM, calculated as the root mean square error (random error) of the ANOVA model divided by the number of observations within a treatment group, was reported as an estimate of the random variation within traits. Significant inferences were based on P < 0.05 and trends as P < 0.10. Inferences among treatment groups were based on orthogonal contrasts used to identify the main effects due to nursery diets (D effect; 0 vs. 100 and 800; and 100 vs. 800), effects due to vaccinations (0 × V vs. 1 × V and 2 × V; and 1 × V vs. 2 × V), and effects due to the interaction between nursery diets and vaccination protocols.

Results and Discussion

Two pigs were removed from the study at weaning because of poor growth over the nursery phase. In the interval from weaning to market, 1 pig died, and 2 pigs were euthanized due to morbidity. No other unexplained deaths or termination occurred over the duration of the study.

Growth Responses

Differences between the main effects of nursery diets and vaccination protocols were not detected in ADG or GF ratio over the 21-d nursery period (Table 2). However, ADFI tended to be lower (P < 0.10) in pigs fed diets with 0 vs. 100 and 800 IU/kg over the 21-d nursery period. The interaction effects between D and vaccination protocols were not significant for the traits analyzed.

Table 2.

Growth, feed intake, and feed efficiency responses of pigs from weaning to the end of a 3-week nursey phase¹

			Supplemental vitamin D3, IU/kg diet
Trait	Interval, d	Vaccine 2	0	100	800	SEM
Gain, kg/d
	0 to 21
		0 × V	0.230	0.252	0.236
		1 × V	0.207	0.237	0.226
		2 × V	0.234	0.255	0.223	0.021
Feed Intake, kg/d
	0 to 21a
		0 × V	0.352	0.376	0.405
		1 × V	0.318	0.359	0.359
		2 × V	0.351	0.392	0.352	0.027
Gain:Feed ratio
	0 to 21
		0 × V	0.657	0.666	0.582
		1 × V	0.643	0.671	0.631
		2 × V	0.662	0.640	0.623	0.028

¹Values are averaged across 6 pens per treatment, with 4 pigs per pen.

²Vaccine protocols included no injection (0 × V), a single injection (1 × V) of 2 mL L. intracellularis vaccine 2 wk after weaning, or 2 injections (2 × V) of 2 mL L. intracellularis vaccine at weaning and 1 wk after weaning.

^aDifference due to the main effect of diet, 0 vs. 100 and 800, P < 0.10.

During the first week, after pigs were switched to the standard diets (end of the nursery phase) and for 50 d after the nursery phase, the ADG of pigs previously fed diets with 0 IU/kg was lower (P < 0.01) compared with pigs fed 100 or 800 IU/ kg diets (Table 3). During the growth interval of 21 to 29 d after weaning, there was a trend (P < 0.10) for a decreased growth of pigs fed 100 vs. 800 IU/kg of D during the nursery phase, this trend was not observed after day 29. The difference in ADG from weaning until 70 d after weaning between pigs fed D levels of 0 vs. 100 and 800 IU/kg demonstrates a negative carry-over effect for pigs fed no supplemental D during the nursery phase. The magnitude and duration of the carry-over response in ADG was surprising. Over the entire period from weaning to market period (0 to 126 d), gain was reduced (P < 0.05) in pigs previously fed 0 vs. 100 and 800 IU/kg diets, except for the 2-wk period in the nursery phase during the phase that the dietary treatments were actually fed. No interactions between D supplements and vaccine protocols were detected in the subsequent growth phases. The reduction in growth of pigs fed diets with no supplemental D during the 3 wk nursery trial reduced (P < 0.05) final market weight (data not shown) of pigs by ~7 kg. Average pig weights were 112.1 vs. 119.4 kg (±2.7 SEM) for pigs fed 0 vs. 100 and 800 IU/kg of D during the nursery phase. No evidence for a compensation in growth was detected for pigs fed no D during the nursery phase, as these pigs did not catch up or exceed ADG during any of the weight intervals after day 70.

Table 3.

Growth rates of pigs from weaning to market weight¹

		Supplemental vitamin D3, IU/kg of diet
Intervals, d	Vaccine2	0	100	800	SEM
0 to 21
	0 × V	0.235	0.266	0.241
	1 × V	0.208	0.243	0.240
	2 × V	0.227	0.245	0.239	0.025
21 to 29 a, b
	0 × V	0.403	0.563	0.616
	1 × V	0.399	0.561	0.612
	2 × V	0.404	0.549	0.595	0.036
29 to 50 a, c
	0 × V	0.727	0.770	0.770
	1 × V	0.655	0.747	0.741
	2 × V	0.685	0.749	0.770	0.027
50 to 70 a
	0 × V	0.977	1.015	1.050
	1 × V	0.923	1.037	1.015
	2 × V	0.942	1.031	1.085	0.030
70 to 91
	0 × V	1.095	1.066	1.079
	1 × V	1.038	1.135	1.099
	2 × V	1.081	1.106	1.158	0.035
91 to 126
	0 × V	1.127	1.160	1.166
	1 × V	1.164	1.155	1.171
	2 × V	1.179	1.205	1.241	0.035
0 to 126 a
	0 × V	0.836	0.869	0.878
	1 × V	0.812	0.875	0.872
	2 × V	0.835	0.883	0.916	0.021

¹Values are averages of growth rates (ADG, kg/d) for 19 to 20 pigs per treatment during the weight intervals expressed as days after weaning. Values are based on growth rates of the individual pigs that completed the entire trial.

²Vaccine protocols included no injection (0 × V), a single injection (1 × V) of 2 mL L. intracellularis vaccine 2 wk after weaning, or 2 injections (2 × V) of 2 mL L. intracellularis vaccine at weaning and 1 wk after weaning.

^aDifference due to the main effect of diet, 0 vs.100 and 800, P < 0.05.

^bDifference due to excess vitamin D₃ in diets, 100 vs. 800, P < 0.10.

^cDifference due to the main effect of vaccination, 0 × V vs. 1 × V, and 2 × V, P < 0.10.

The concentrations of supplemental D (100 vs. 800 IU/kg) in nursery diets did not affect growth of pigs during the phases from weaning to market. If pigs fed 800 IU/kg of diet of D had an increased in ADG from the end of the nursery phase until market, then inferences that pigs required more D under an immune challenge might be supported; however, no improvement in growth was attributed to the extra D. The growth responses of pigs in this experiment implied that excess D, over 100 IU/kg, was not beneficial regardless of the vaccination protocol applied.

Vaccination treatments did not affect growth during the nursery period. A trend (P < 0.10) for a decreased growth was detected in pigs between vaccination protocols 0 × V vs. 1 × V and 2 × V from days 29 to 50, but this trend did not continue until market weight. At market, differences in growth due to the vaccination protocol were not detected. As diarrhea is commonly associated with L. intracellularis and positive serology titers have been prevalent in the SRTC herd after an outbreak in 2016, a decrease in growth rate was expected in nonvaccinated pigs. Johansen et al. (2013) reported that a log₁₀ increase in L. intracellularis bacteria load increased the odds for pigs to display a 2× reduction in growth rate. The negative consequences of reduced growth were not observed during this experiment. The observation that the vaccination did not improve overall growth compared with nonvaccinated pigs might be attributed to an inefficient antigen in the vaccine used, implying that the immune system was not sufficiently challenged to allow detection of a response to D supplementation. Another possible explanation for the lack of a difference between vaccinated and nonvaccinated pigs might be a lack of specificity of the vaccine antigen to the specific strain of L. intracellularis in the SRTC herd.

The results from the current experiment are not consistent with reports from other experiments in which growth from weaning to market was improved in vaccinated vs. nonvaccinated pigs in herds with subclinical infections of L. intracellularis (Henke et al., 2006). The results from the previous experiments found vaccinated pigs had increased gain and feed intake responses compared with responses of nonvaccinated pigs. However, the previous research involved administration of an oral, modified live vaccine rather than an injectable killed vaccine as used in the current experiment.

Whole Body, Femur, and Foot DXA Responses

Whole-body DXA scans are commonly used clinically as a noninvasive procedure to assess risks for developing osteoporosis in humans. Whole-body DXA scans of pigs have been used to assess skeletal recovery from periods of dietary Ca and P deficiencies (Aiyangar et al., 2010) and to assess whole skeletal bone mineral status in response to D supplements (Amundson et al., 2016). In the present study, differences in both whole-body BMC and BMD were detected among dietary treatment groups (Table 4). At 35 d after weaning (14 d after a 21-d treatment phase), whole-body BMD was increased (P < 0.05) by 10% in pigs fed either 100 or 800 IU/kg above that of pigs fed 0 supplemental D during the 3-wk nursery phase. Likewise, whole-body BMC tended to increase (P < 0.10) in pigs fed D levels of 100 and 800 vs. 0 IU/kg during the nursery phase. These results are consistent with earlier experiments that showed 100 IU/kg met requirements for maximum skeletal mineral content responses in growing pigs (Amundson and Crenshaw, 2015). No evidence was detected that the vaccination protocol altered BMD or BMC responses across vitamin D treatments in the nursery phase.

Table 4.

Whole-body and femur results from dual energy X-ray absorptiometry scans of pigs at 9 wk of age^{1, 2}

			Supplemental vitamin D3, IU/kg diet
Trait	Scan	Vaccine3	0	100	800	SEM
BMC, g
	Whole bodya
		0 × V	265	330	356
		1 × V	276	320	318
		2 × V	319	324	295	34
	Right femura
		0 × V	8.6	10.0	11.4
		1 × V	9.2	11.5	10.8
		2 × V	10.5	11.3	9.0	1.2
BMD, g/cm2
	Whole bodya*
		0 × V	0.392	0.426	0.451
		1 × V	0.402	0.438	0.443
		2 × V	0.408	0.435	0.440	0.023
	Right femur a
		0 × V	0.303	0.315	0.388
		1 × V	0.349	0.409	0.351
		2 × V	0.340	0.385	0.341	0.031

¹DXA scans were analyzed to determine bone mineral content (BMC, g) and bone mineral density (BMD, g/cm²).

²Values are averages of whole-body DXA scans of 4 barrows per treatment ( 1 each from weight blocks 1, 2, 5, and 6) at 9 wk of age (day 37 after weaning).

³Vaccine protocols included no injection (0 × V), a single injection (1 × V) of 2 mL L. intracellularis vaccine 2 wk after weaning, or 2 injections (2 × V) of 2 mL L. intracellularis vaccine at weaning and 1 wk after weaning.

^aDifference due to main effects of diet, 0 vs. 100 and 800, P < 0.10 and *P < 0.05.

From earlier work in our lab (Hoffman et al., 2007; Crenshaw et al., 2013), the femur was found to accurately predict whole-body BMC and BMD for 5 to 40 kg pigs, and DXA scans of the femur and foot were accurate predictors of BMC in pigs at market weight (120 kg). In the current study, the right femur from each pig was collected after whole-body DXA scans at day 37. The DXA scans of the right femur were conducted to further evaluate BMC and BMD (Table 4). At 9 wk of age, the BMC and BMD of femurs tended to be reduced (P < 0.10) in pigs fed 0 vs. 100 and 800 IU/kg during the nursery period; however, differences were not detected between pigs fed 100 vs. 800 IU/kg. This response was consistent with previous research which indicated that 100 IU/kg of diet of D provided concentrations for maximum ADG and bone ash accumulation in growing pigs (Amundson and Crenshaw, 2015).

In the current experiment, the left front foot from each pig was collected for DXA scans (Table 5). No differences were detected in the left foot BMD or BMC among D treatments or vaccination protocols. Assuming that the foot represents the whole body (Crenshaw et al., 2013), the lack of a difference implies that there was no carry-over effect in bone ash accumulation for pigs fed no D during the nursery phase. Thus, BMD was recovered by market weight for pigs fed no D during the nursery phase. Likewise, excess D (800 IU/kg) had no benefit on bone ash accumulation during an immune challenge or during subsequent phases. However, a limitation for this conclusion needs consideration. The D treatments were not equally represented across market groups at slaughter. The first group shipped to market included 22, 28, and 38 pigs in treatments 0, 100, and 800 IU/kg, respectively. The distribution of the number of pigs in the second shipment group included 36, 27, and 19 pigs for the respective D treatment groups. The different distributions confounded across shipment times reflected sorting based on pig weights. A greater number of pigs from the 800 IU/kg group were included in the first shipment, and more pigs in the 0 IU/kg group were in the second shipment. This skewed distribution of pigs from each treatment groups across shipment dates may have confounded the bone mineral results. One may infer that because more pigs fed 0 IU/kg were found on the second shipment, those pigs had an opportunity to catch up in bone ash with pigs fed the 800 IU/kg treatment due to the extra 2 wk before shipment. This age difference may explain the lack of a difference among treatment groups in BMC and BMD at market. If all the animals were marketed at the same time, a difference may have been detected in BMC due to dietary D treatments (Table 5).

Table 5.

Dual energy X-ray absorptiometry scans of the left front foot of pigs at market weight^1,2

		Supplemental vitamin D3, IU/kg diet
Item	Vaccine3	0	100	800	SEM
BMC, g
	0 × V	45.45	44.50	44.54
	1 × V	45.21	44.78	44.05
	2 × V	43.34	43.16	44.62	0.974
BMD, g/cm2
	0 × V	0.61	0.59	0.61
	1 × V	0.62	0.60	0.60
	2 × V	0.60	0.58	0.60	0.009

¹DXA scans were analyzed to determine BMC (g) and BMD (g/cm²).

²Values are averages of DXA scans of the left front foot of pigs at market weight. The averages are based on collected feet that retained all metacarpal and carpal bones during the collection process, n = 146 pigs with 13 to 19 pigs per treatment group.

³Vaccine protocols included no injection (0 × V), a single injection (1 × V) of 2 mL L. intracellularis vaccine 2 wk after weaning, or 2 injections (2 × V) of 2 mL L. intracellularis vaccine at weaning and 1 wk after weaning.

Prevalence of Kyphosis

Rortvedt and Crenshaw (2012) reported kyphosis, an abnormal spine curvature, was induced when no supplemental D was added to sow diets and the diets of their progeny.

In the current experiment, the prevalence of kyphosis appeared to increase in pigs from dietary treatments which provided 0 IU/kg during a 3-wk nursery period (Table 6). Pigs previously fed diets with 0 IU/kg had approximately a 10% prevalence (6 of 57 pigs) of grossly evident kyphosis. Only 1 of 59 pigs fed 800 IU/kg had evidence of kyphosis (5%). The prevalence of kyphosis (4 of 18 pigs, 22%) was the greatest in pigs fed 100 IU/g and subjected to 2 × V vaccination protocol (diet × vaccinations protocol interaction, P < 0.03). The interaction effect infers that 100 IU/kg diets did not provide adequate D for pigs subjected to the increased immune challenge of 2 vaccine injections. This inference should be interpreted with caution as the pigs fed diets with 0 IU/kg did not display the same prevalence and because of the low number of observations, 1 or 2 pigs can dramatically alter the inferences. However, as reported previously (Amundson et al., 2017), the expression of kyphosis appears to be independent of responses that involves measures of bone ash. Lesions associated with kyphosis are osteochondrosis-like lesions (Halanski et al., 2018) and may be a consequence of vitamin D metabolites in regulation of expression of matrix metalloproteinases genes that affect cartilage (Amundson et al., 2018). Whether direct effects of vitamin D metabolite signaling in cartilage tissues results in lesions or the effects are mediated by interactions with immune signals requires further research efforts.

Table 6.

Prevalence of kyphosis based on visual scores of pigs at 4 mo¹

		Supplemental vitamin D3, IU/kg diet
Item	Vaccine2	0	100	800
Kyphosis, n3
	0 × V	2 (20)	0 (20)	0 (20)
	1 × V	2 (19)	0 (18)	1 (19)
	2 × V	2 (18)	4 (18)	0 (20)
Kyphosis, %4 a
	0 × V	10.0	0	0
	1 × V	10.5	0	5.3
	2 × V	11.1	22.2	0

¹Visual scores for abnormal outward curvature of the spinal column (kyphosis) were recorded by 2 observers for pigs at 4 mo of age. Scores were 1, no abnormal curvature; 2, marginal curvature; 3 obvious curvature. Scores were averaged across both observers. Only pigs with a score of 3 were designated as kyphosis.

²Vaccine protocols included no injection (0 × V), a single injection (1 × V) of 2 mL L. intracellularis vaccine 2 wk after weaning, or 2 injections (2 × V) of 2 mL L. intracellularis vaccine at weaning and 1 wk after weaning.

³Values reported are the number of pigs assigned a kyphosis score of 3 by both observers. The total number of pigs in the respective treatment group is shown in parentheses.

⁴Values are the percentages of pigs with kyphosis within each treatment group. Statistical inferences are based on RANK analysis of visual scores of either 1 or 3.

^aInteraction between dietary vitamin concentrations and vaccination protocols, P < 0.03.

L. intracellularis Serology

A year before the current experiment, the SRTC herd “broke” with ileitis, which was attributed to L. intracellularis. The organism has likely remained endemic in the herd, although clinical symptoms are rarely observed. After the initial outbreak, the breeding herd and all pigs have been routinely vaccinated for L. intracellularis at weaning, using a single dose of the vaccine as used in the current experiment. The results from the first serology samples, inferred that the L. intracellularis was still present in the SRTC herd, as 17% of the nonvaccinated pigs expressed positive titers for L. intracellularis (Table 7). The first samples were collected 21 d after the first or second injection. Only 39% of the 1 × V pigs had positive titers, but 100% of 2 × V pigs had positive titers for L. intracellularis. All 2 × V pigs had L. intracellularis titers that were 3- to 4-fold greater than the 0 × V and 1 × V pigs, inferring that 2 vaccinations provided better protection against the bacterium. More than half of the 1 × V pigs did not display positive titers, implying that 1 vaccination did not provide adequate protection against L. intracellularis. Based on the age of pigs at the first blood collection, the positive titers in the 0 × V group may have reflected positive titers attributed to lingering maternal antibodies for L. intracellularis, as maternal antibodies were reported to still be detectable in pigs at 2 to 3 wk after weaning (Riber et al., 2015). Collections of blood samples at later times might have allowed detection of a larger percentage of positive titers in pigs from the 1 × V group, but a lower percentage in the 0 × V groups if maternal titers were still present. However, although peak L. intracellularis titers were reported to occur at 4 to 5 wk after injection (Guedes et al., 2003), the effects of dietary vitamin D on the expression of antibody titers were the primary interest, thus titers were assessed during the initial development phase at 21-d after vaccinations.

Table 7.

Serology analysis at 3 and 16.5 wk after vaccine injections^{1, 2}

Wk, post-inj.	Vaccine3 a, b	Positive, n4	Total, n5	Positive titers, %6
3
	0 × V	6	36	17
	1 × V	7	18	39
	2 × V	18	18	100
16.5
	0 × V	24	36	67
	1 × V	11	18	61
	2 × V	15	18	83

¹Serology data for the same 72 randomly selected pigs at 2 periods (3 and 16.5 wk) after vaccinations.

²Treatment groups were pooled across vaccination protocols at each age (n = 36 for 0 × V; n = 18 for 1 × V, and n = 18 for 2 × V).

³Vaccine protocols included no injection (0 × V), a single injection (1 × V) of 2 mL L. intracellularis vaccine 2 wk after weaning, or 2 injections (2 × V) of 2 mL L. intracellularis vaccine at weaning and 1 wk after weaning.

⁴The number of samples that were positive for L. intracellularis titers.

⁵Total number of samples for each vaccination protocol.

⁶Positive % is the number of positive samples per total samples within each respective treatment group expressed as a percent.

^aThe number of nonvaccinated (0 × V) pigs with positive titers differed between 3 and 16.5 wk post injection, P < 0.01.

^b The number of 2 × V-vaccinated pigs with positive titers differed between 3 and 16.5 wk post injection, P < 0.10.

By market weight, the increased prevalence (P < 0.01) of positive titers in nonvaccinated pigs confirmed the endemic presence of L. intracellularis in the herd. The samples were collected from the same pigs at each time. The 1 × V group had an increase prevalence of pigs expressing positive titers at the second blood collection period. In contrast, pigs in the 2 × V group trended to decrease (P < 0.10) the prevalence of positive titers at market weight. At the second blood collection, approximately two-thirds of pigs showed positive L. intracellularis serology titers regardless of vaccination or D treatment. Unfortunately, distinction between serology titers due to the L. intracellularis vaccine or L. intracellularis pathogens was not possible with the IPMA assay.

Apparently, all pigs in the 0 × V group did not seroconvert during the growth phase. The majority, 83%, of pigs in the 2 × V group displayed positive titers at market, which implied that the 2 × V vaccination protocol during the nursery phase protected the majority of pigs until market age. However, none of the pigs in the current study displayed clinical symptoms of ileitis during the experiment. Based on serology titers at market, the majority of pigs in all vaccination protocols and D treatment groups were exposed by either vaccinations or endemic pathogens, and experienced similar immune challenges. Typically, pigs exhibit clinical symptoms of ileitis at approximately 4 to 5 mo of age, which may explain why more pigs displayed positive titers at the second blood collection time.

Discussion

The roles of vitamin D metabolites in Ca and P homeostasis have been well established for 3 to 4 decades. A failure in Ca and P homeostasis as a result of a D deficiency leads to rickets in growing animals and children, and osteomalacia in adults as described in reviews by Deluca (2004), Dusso et al. (2005), and Plum and DeLuca (2010). More recently, the discovery of the vitamin D receptor (VDR) and regulatory effects of vitamin D in cells in addition to those involved in Ca and P homeostasis infers a more diverse role for vitamin D than just Ca homeostasis. Evidence which links vitamin D deficiencies to human diseases, such as multiple sclerosis and tuberculosis, implies potential immunoregulatory functions for this vitamin.

Only limited results that contribute to defining vitamin D requirements for growing pigs have been reported since the 1960s. Over a recent 5-year period (2009 to 2014), a dramatic increase in the diagnosis of swine lameness and sudden death of growing pigs in the Midwest was attributed to vitamin D deficiency (Arnold et al., 2015). The large increase in research on vitamin D over the past 10 yr can be attributed to an industry recall of swine feed associated with a vitamin D deficiency.

Research highlighting the carry-over effects of D supplementation during the nursery phase on growth and bone development during subsequence growth phases is not commonly reported. Even less information is available to define the relationship between D supplementation and interactions with immune challenges in pigs. The overall objective of this study was to characterize potential interactions between nursery dietary D supplements and the timing of vaccine injections on pig growth and bone mineral traits. Unfortunately, observations of growth and bone ash responses to nursery treatments in this experiment did not support inferences that the amount of dietary D needed by the growing pig was affected by the immune status. The dietary treatments fed during the nursery phase did not provide evidence that beneficial effects of excess dietary D boosted immune responses in vaccinated pigs nor was evidence provided to support inferences that excess D stimulated growth responses in pigs under a vaccine-induced immune challenge. In the current experiment, after pigs were switched to the standard SRTC grower and finisher diets, differences in growth due to the nursery phase dietary D treatments were still evident for about 7 wk after the nursery diets were fed. Such a long residual carry-over effect of a 3-wk D deficiency was surprising. A more rapid recovery from a short phase of a D deficiency was expected, especially in pigs fed diets with ~3× the required D.

Metabolic changes at the time of an immune challenge can alter nutrient requirements. Pigs exposed to increased disease-challenged environments had a reduced protein gain that was not completely compensated by increasing dietary lysine concentrations (Williams et al., 1997a; 1997b). Changes in protein accretion was attributed to an inefficiency of energy use associated with increased heat production from excess protein catabolism, not an inefficiency of lysine use. Huntley et al. (2017) found a 23% increase in heat production in immune challenged animals. The increased heat production resulted in a 26% decrease in ADG. The results from the current study showed nutrient requirements for growth, but not bone ash, were altered due to vitamin D treatment, but not vaccination protocol. This infers that vitamin D may have a role in the immune system, entirely separate from its function in renal tissues and Ca homeostasis. Recent results (Yang et al., 2019) have inferred alleviation of diarrhea severity, improvements in intestinal structure, and suppression of inflammatory responses in weaned pigs supplemented with mega-doses (>6,000 IU/kg diet) of 25-hydroxy vitamin D₃ supplements prior to a porcine epidemic diarrhea virus challenge. However, growth responses were not restored to that of control, nonchallenged pigs.

In conclusion, the current controlled experiment provided evidence that pigs fed no supplemental D during a 3-wk nursery period exhibited carry-over effects on subsequent growth to market weight. The magnitude of differences in pig growth was due to D, not the vaccine treatments. However, differences in growth were not evident until the grower-finisher phases. Carry-over effects on skeletal ash accumulation was not detected, inferring a priority of D for calcium homeostasis rather than protein accumulation. Limited evidence for D mediation of an immune challenge in the current experiment may be a consequence of limited vaccine effects on growth in the nursery and seroconversion of most pigs to L. intracellularis by market weight. The effects of supplemental D and altered immune status interactions on pig weight gain and bone growth require further examination during conditions that impose a greater immune challenge.

Supported by the National Institute of Food and Agriculture, United States Department of Agriculture, Hatch Project (1010347) and unrestricted gift funds to support swine nutrition research.

Conflict of interest statement

None declared.