| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
ANIMAL NUTRITION |
The Ohio State University, and The Ohio Agricultural Research and Development Center, Columbus 43210-1095
 |
Abstract |
|---|
 Top Abstract INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Key Words: mineral • pig • reproduction • sow
|
INTRODUCTION |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
). Inadequate mineral intake may affect hormonal secretion, enzyme activity, muscle function, bone mineral content, and other body mineral functions.
Mineral requirements for swine have largely been determined using inorganic mineral salts. Elevated dietary Ca and P along with greater trace mineral concentrations are commonly recommended for high-producing sow lines. Possible chelation interactions may occur between macro- and microminerals in the lumen of the digestive tract that may be more pronounced when greater mineral concentrations are fed. This could possibly affect the absorption and biological function of minerals (Morris and Ellis, 1980; O’Dell, 1997; Ammerman et al., 1998).
Several trace minerals can be sequestered with peptides, AA, or carbohydrates, and such mineral products (i.e., organic) are available for feeding livestock. Adding organic trace minerals to the diet may minimize chelation interactions with minerals and other dietary factors. The effect of organic trace minerals on sow reproduction has not been widely investigated.
This experiment evaluated sow reproductive performance over 6 parities when corn-soybean diets containing either inorganic or organic sources of trace minerals at concentrations at or above NRC (1998) requirements are fed. Adding trace minerals at high dietary Ca and P levels was investigated, as well as determination of mineral intakes during each of the reproductive cycles of the sow.
|
MATERIALS AND METHODS |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Experimental Designs and Treatments
Dietary trace minerals (Cu, Fe, Mn, Se, and Zn) of organic or inorganic origins were fed to developing gilts, whereupon the females were continued into the reproductive phase. Experiment 1 evaluated the effects of these 2 trace mineral sources and levels in developing gilts to evaluate performance responses and to possibly modify body tissue trace mineral reserves. Experiment 1 was a 2 x 2 factorial in a randomized complete block design using 3 groups of gilts in 6 total replicates. The first factor evaluated trace mineral source (organic or inorganic), whereas the second factor evaluated 2 trace mineral levels. The organic mineral source (Bioplex, Alltech Inc., Nicholasville, KY) was purchased from a commercial premix company. The organic trace minerals were metal proteinates, sequestered with enzymatically hydrolyzed soybean protein, whereas organic Se was a yeast protein largely as selenomethionine (Sel-Plex, Alltech Inc.). Most inorganic trace minerals were salts in the sulfate form, except Se, which was as Na selenite and Mn oxide. Chromium was added as Cr picolinate at 0.20 mg/kg to all treatment diets throughout the study. The first dietary trace mineral level was added at the NRC (1998) total requirement level, whereas the second level contained each trace mineral added at a level considered typical to that used in feed industry (IND) diets, as determined from an industry survey (D. C. Mahan, unpublished data).
After the gilt development phase, Exp. 2 was conducted as a 2 x 3 factorial arrangement in a split-plot design with the sow serving as the main plot. Each animal remained on the same trace mineral treatments as during their development phase, except that 2 additional treatments were added at the time of the initial breeding. To accomplish the 2 additional treatments, 2 pens of developing gilts had been fed the IND trace mineral levels of both inorganic and organic trace mineral sources. The treatments diets that were fed during the development period were provided until breeding, whereupon additional Ca and P were added to the 2 additional IND treatment diets (IND + Ca:P) that were fed for the duration of the reproductive study.
Sows were fed their treatment diets (gestation and lactation) for 6 total parities in 3 replicates in the 2 x 3 factorial arrangements of treatments. The first factor evaluated 2 dietary trace mineral sources (organic or inorganic), whereas the second factor contained 3 dietary mineral levels that were provided during both gestation and lactation.
Grower Phase
A total of 216 gilts (Yorkshire x Landrace) were obtained (Temple Genetics, Gentryville, IN) at weaning (21 d of age) in 3 groups. Upon arrival, each pig group was placed in isolation and fed conventional starter diets for a 5-wk period. At approximately 30 kg of BW, gilts were transferred to partially slotted (i.e., 40%) concrete floors that contained 6 pigs, one 4-hole stainless steel feeder and 1 stainless steel nipple waterer, and iron rods as pen separators. Gilts were allotted to experimental treatments based on BW and ancestry. A total of 2 replicates per group (n = 6 total replicates) were involved in this study. Because 2 of the diets during the reproductive phase were not begun until breeding, 2 additional pens within each replicate for each trace mineral source were fed the IND dietary trace mineral levels. Consequently, each replicate contained 6 pens, and each pen was fed their treatment diet from 30 kg of BW to breeding. Gilts were exposed to adult boars once daily from approximately 115 kg of BW to breeding, but records of estrus onset were not recorded. The gilts in the 2 additional IND pens were transferred to the IND + Ca:P treatment diets at the time of breeding, remaining on this treatment regimen throughout the study.
Breeding and Postbreeding Period
At 8 to 9 mo of age, gilts from each treatment group were artificially inseminated at the onset of estrus and 12 h later. Sows were also artificially inseminated on their first estrus postweaning and 12 h later. Sows not exhibiting estrus within 10 d of weaning were placed in the next breeding group, and if they failed to cycle in this subsequent group (i.e., 45 d postweaning), they were removed from the study. Semen used for breeding was from a PIC genetic boar line (280) and obtained from a commercial stud farm twice weekly, stored at 18°C, rotated twice daily, and used within 4 d of collection. Gilts and sows not pregnant, as determined by ultrasound examination after 30 d postcoitum, were removed from the experiment.
Bred females were housed in individual gestation stalls with concrete slotted floors with each stall containing a stainless steel feeder and nipple waterer. Body weights were collected at breeding and 110 d postcoitum. Backfat measurements using an A-mode ultrasound single transducer (Lean Meater, Renco Corp., Minneapolis, MN) was used at the last rib approximately 40 mm off the midline at each weighing.
Selected sows from each treatment group were removed from the study after weaning their litters at parities 1, 2, 4, and 6 and killed for body composition (data not reported). Consequently, the lower number of sows completing the 6-parity study reflected this loss as well as sows removed for other reasons.
Upon weaning, sows were moved to individual gestation stalls and offered their gestation treatment diet at 1.5 times the quantity of gestation feed intake until bred or for a maximum of 10 d. At breeding, the gestation treatment diet was provided once daily at 85% of their projected gestation feed allowance for 15 d, whereupon it was increased for the remainder of gestation. Feed intake from 15 d postcoitum to farrowing was 2.1 kg/d (6,540 kcal/d) for parity 1 gilts, but their daily allowance increased by 135 g (i.e., 440 kcal/d) for each successive parity to parity 4. Because of declining sow body condition, the gestation feed allocation was increased by 270 g/d (i.e., 880 kcal/d) for parities 5 and 6.
Farrowing and Lactation Period
At 110 d, gestation sows were moved to individual farrowing crates and fed their lactation treatment diets at the same quantity as during gestation. Upon parturition, sows were fed their diet at 2.0 kg on d 1, which was then cumulatively increased by 2.0 kg/d such that by d 3 to 4, sows were provided ad libitum access to diets.
Within 12 h postpartum, sows were weighed, back-fat thickness (last rib) measured, and litters processed including the injection of 200 mg of Fe (Fe dextran). Litter size was standardized among sows across treatment groups by 3 d postpartum within each farrowing group. Although transferring pigs across treatment confounded sow treatment responses, it was assumed that milk production and its composition would be more consistent if litter size was standardized. Utilization of sow tissue mineral reserves would also be expected to be more uniform during lactation between sows if litter size was approximately equal. Sow BW, feed intake, and litter weights were measured at d 7 postpartum and at 17 d (weaning). Backfat thickness (last rib) was determined at weaning.
Colostrum was collected from all functional mammary glands within 12 h of parturition and milk at weaning. Samples were collected (30 to 50 mL) after an i.m. injection (40 USP units) of oxytocin, frozen, and stored at –;4°C for subsequent fat analysis.
Premix and Diet Composition
Addition of both treatment trace mineral levels ignored the indigenous contribution from the feed grains or other exogenous sources from diet preparation. The dietary premix treatment levels and their analyses are presented in Table 1. Treatment diets were formulated to total dietary concentrations of Ca (0.85%) and P (0.65%) during the development phase for gilts. Treatment diets were initially provided at 30 kg of BW.
|
). These latter treatment diets were also fed from 110 kg of BW to the initial breeding with the amount fed adjusted on an individual gilt basis to attain average breeding BW of approximately 135 kg. The minimum amount fed was 1.8 kg/d.
|
nutrient requirements. The percentage composition of the basal diets for gestation and lactation are presented in Table 3. Samples of diets within parity were collected, composited, and subsequently analyzed for macro- and microminerals.
|
Daily intakes of trace minerals from the premixes as well as the total amount of macromineral consumed from treatment diets were calculated for individual animals during gestation and lactation. The dietary level of each mineral was multiplied by the ADFI of the sow for each reproductive phase when expressed on the total amount of each mineral consumed per day and on an amount per kilogram of sow BW basis. Only the trace minerals contributed from the premixes were used for these calculations, whereas the macrominerals used total dietary levels. Breeding and 12-h farrowing weights were averaged and used as the mean gestation BW, whereas the mean lactation BW was the average weight at farrowing and weaning.
Analytical Methods
Premixes and diets were analyzed for Cu, Fe, Mn, Zn, Ca, P, Mg, K, S, and Na using inductively coupled plasma equipment (PS 3000, Leeman Labs Inc. Hudson, NH). Chloride was analyzed using the Cl ion electrode (LaCroix et al., 1970) by a chloridometer (Model 4–2008, Buchlert-Cotleve, Saddle Brook, NJ). Selenium samples of feed, tissue, and milk were wet-ashed in nitric and perchloric acid and analyzed using the fluorometric method (AOAC, 2000). Milk from all weaned sows was analyzed for fat content using the Babcock method (AOAC, 2000).
Performance data for the period from 30 to 110 kg of BW of Exp. 1 were analyzed as a 2 x 2 factorial arrangement of treatments in a randomized complete block design in 6 replicates using the MIXED model procedure (SAS Inst. Inc., Cary, NC). The 2 additional IND treatment pens were averaged with their respective treatment group and were included in the statistical analysis of performance data. The statistical model included the effects of trace mineral source and level, blocks (replicates), and treatment x block as the error term. Block was considered a random effect. Pen was the experimental unit used for analysis of performance data.
For Exp. 2, reproductive performance and sow mineral intake data were analyzed as a 2 x 3 factorial in a split-plot design with sow as the main plot, in 3 replicates using the MIXED model procedure of SAS. Sows were repeated measures between parities and they or their litter served as the experimental unit. The data were analyzed according to the following model: Yijklm = µ + Si + Lj + SLij+ Pk + SPik + LPjk + sl + gm + g(SL)ijm + eijklm, where Yijklm = the dependent continuous variable; µ = the overall mean; Si = the fixed effect of the ith trace mineral source (i = 1,2); Lj = the fixed effect of the jth trace mineral level (j = 1,2,3); SLij = the fixed effect of the ith trace mineral source x the jth trace mineral level interaction; Pk = the fixed effect of the kth parity (k = 1,...,7); SPik = the fixed effect of the ith trace mineral source x kth parity interaction; LPjk = the fixed effect of the jth trace mineral level x kth parity interaction; sl = the random effect of the lth sow; gm = the random effect the mth block (m = 1,...,7); g(SL) ijm = the random effect of the mth block x the ith trace mineral source x the jth trace mineral level interaction; and eijklm = the residual error.
The block x treatment interaction was included, because sows from specified replicates were killed after they had weaned a predetermined number of litters (i.e., 1, 2, 4, and 6); therefore, the number of observations was unequal among replicates and parities. Repeated measures were included in the analysis. The subject of the repeated measure was individual sow nested within treatment, and the first-order autoregressive covariance structure was used. This structure consistently gave the lowest Bayesian information criteria for the covariance structures tested. Individual sow and litter measurements were considered the experimental unit. Parity effects were partitioned into linear and other curvilinear components using orthogonal polynomial contrasts. Least squares means for treatments are presented in tabular form.
|
RESULTS |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
. Analytical values for both premix sources were similar to calculated levels. Mineral contents in the complete diet were greater than that from the premix contribution. The indigenous minerals in the feed grains, macromineral sources (limestone and dicalcium phosphate), and other exogenous contaminants from feed mixing equipment all contributed to the greater trace mineral levels in the complete diet mixtures. The Ca and P analyses of complete diets were generally within 0.10% of calculated values. In most cases, when the trace minerals from the premix were subtracted from the total analytical values, the indigenous contributions of trace minerals were close to meeting the animals NRC (1998) requirement, although their bioavailability is largely unknown. The Fe content in the final diet mixtures averaged about 150 mg/kg of complete diet above the premixes and around 250 mg/kg greater when additional dicalcium and limestone was used (IND + Ca:P treatment). In the NRC treatment gestation diet, the Cu content was about 5 mg, Mn 25 mg, Zn 38 mg, and Se 0.04 mg/kg of diet greater than that provided by the trace mineral premixes. Reproductive Performance
During gilt development (30 to 110 kg of BW), there was no effect of trace mineral source or level on ADG, ADFI, or G:F ratio responses (Table 4). Reproductive performance response in Exp. 2 to dietary trace mineral sources and levels presented in Table 5 demonstrated a greater (P < 0.05) number of pigs born (total and live) when sows were provided organic trace minerals rather than the inorganic trace mineral source. There was a trend (P < 0.10) for a fewer number of pigs born (total) at the greater trace mineral level. Although there was a small increase in the number of live pigs born when organic trace minerals were fed and a decline when inorganic minerals were fed, and fewer numbers of total pigs born when the diets contained elevated dietary Ca and P, the interaction was not significant. Stillbirths (P < 0.10) tended to be greater when organic trace minerals were fed, whereas the number of mummies (P < 0.05) was greater at the IND trace mineral level. The trace mineral source x level interaction for the number of pigs born (total, live, stillbirths) was not significant.
|
|
Litter size was standardized within 3 d postpartum; thus, the number of pigs per litter at d 7 and at weaning (17 d) was similar among the 6 treatment groups. Individual pig ADG responses during the nursing period tended (P < 0.10) to be greater when sows were fed organic trace minerals, implying that these sows may have had greater milk productions.
Gilt and Sow Weight, Feed, and Backfat Measurements
There was no effect of trace mineral source on sow breeding or gestation BW or gestation gains (Table 6). Sows fed organic trace minerals had a somewhat lighter (P < 0.05) weaning BW, but there was no difference in lactation weight change. Lactation feed intake (total or ADFI), estrus rebreeding interval, and milk fat content were similar between trace mineral sources, trace mineral levels, or when the elevated dietary Ca:P levels were fed.
|
Parity Responses
The parity reproductive and performance results presented in Table 7 demonstrated that sow BW at breeding, 110 d postcoitum, and weaning BW increased linearly (P < 0.01) with advancing parity. Gestation gains and farrowing BW were lowest (P < 0.01) during parity 3 and 4 but increased thereafter to parity 6 resulting in a quadratic response (P < 0.01). This latter response was largely attributed to the additional feed allowance provided to sows during parity 5 and 6. Return-to-es-trus intervals were similar among parities. Sows lost more BW during lactation 1 and 2 than from parity 3 to 5, whereupon weight loss increased in parity 6 resulting in an overall quadratic (P < 0.05) weight change for the 6-parity period. As a result, backfat thickness declined from parity 1 to 4 but increased to parity 6, resulting in an overall quadratic parity response (P < 0.01). A quadratic decline in milk fat content occurred at weaning to parity 4, whereupon it increased to parity 6 (P < 0.10), responses also consistent with changes in backfat thickness.
|
). Live pigs born increased quadratically (P < 0.01) to parity 4; however, there was a decline thereafter. There was no apparent difference in the number of stillbirths or mummies by parity. Litter birth weights increased in a quadratic manner (P < 0.01) to parity 5, followed by a decline by parity 6, the latter response also reflected by their lower litter size. Pig birth BW, however, increased by parity (P < 0.05).
|
Sow Daily Mineral Intake
Calculated daily mineral intakes of sows fed the various treatment diets during gestation and lactation by parity, along with their calculated daily requirement intakes predicted by the NRC (1998) model, are presented in Table 9. The NRC (1998) model includes the contribution from the indigenous grain mineral sources, whereas our dietary treatment calculations only used the trace minerals derived solely from the added premix. Because gestation feed intakes were similar for all treatments within parity, there was a lack of individual variation that prevented statistical evaluation, whereas because lactation mineral intakes differed, they were statistically analyzed (Table 9).
|
model. Thus, when the indigenous minerals are further added to these daily intake values, the total trace mineral intakes are substantially greater than the NRC (1998) model predictions. When the IND level was fed, the trace mineral intakes resulted in even greater intakes. As parity advanced, daily gestation mineral intakes increased, reflecting the greater feed intakes of older sows. Sows fed NRC treatment diets for each parity generally had intakes similar to that predicted by the NRC (1998) model, particularly during parities 1 to 4, but they were greatly increased during parity 5 and 6, responses attributable to the greater gestation feed intakes provided to the adult sows. Trace mineral intakes were correspondingly greater when the IND trace minerals were fed as parity advanced.
During lactation, trace mineral intakes in the NRC treatment group were somewhat lower than the animal requirements predicted by the NRC (1998) model, but as previously indicated, the values (Table 9) did not include the indigenous or other exogenous minerals. Lactating sows fed the IND or the IND + Ca:P treatment diets consumed considerably more trace minerals than predictions from the NRC (1998) model. Calcium and P intakes for the NRC and IND treatment groups were similar to those that were predicted from the NRC model, whereas when the IND + Ca:P treatment diets were provided, Ca and P intakes were substantially greater than NRC model predictions (Table 9).
During lactation, daily mineral intakes for each trace mineral level were approximately 2-fold greater than gestation, reflecting the lactating sows greater feed intakes (Table 9). Lactating sow trace mineral intakes generally increased quadratically (P < 0.01) to parity 4 and declined thereafter. The decline in trace mineral intakes to parity 6 was greater for the NRC and IND + Ca:P trace mineral levels resulting in a trace mineral level x parity interaction response (P < 0.01).
The calculated amount of each macro- and micromineral when expressed on a milligrams per kilogram of BW basis is presented in Table 10. These results demonstrated that when the treatment diets were fed, trace mineral intakes during gestation were generally similar within each dietary trace mineral level by parity, even though feed intakes increased. Mineral intakes per kilogram of BW were obviously even greater (P < 0.01) for the IND and IND + Ca:P treatment levels (Table 10) than the NRC treatment group. Gestation macro- and micromineral intakes per kilogram of sow BW decreased slightly in a quadratic (P < 0.01) manner. Consequently, although total mineral intakes increased during each subsequent gestation, the amount consumed per unit of BW declined as sow BW or parity increased. During lactation, trace mineral intakes per kilogram of BW increased quadratically (P < 0.01) to parity 3 but declined thereafter. Because of decreasing lactation feed intakes and heavier sow BW in sows during latter parities, mineral intakes relative to sow BW declined as parity increased. Sow body mineral depletion would therefore be expected during lactation as the number of parities increase.
|
|
DISCUSSION |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
when grower-finisher gilts were fed supplemental inorganic trace minerals at or above NRC requirements.
Feeding the organic or inorganic trace minerals sources at the various levels, however, did affect sow reproductive performance, particularly litter size. Our experiment indicated that the organic trace mineral source resulted in more (P < 0.05) total (12.2 vs. 11.3) and live (11.3 vs. 10.6) pigs born compared with sows fed inorganic trace minerals. Although the data imply that when the dietary level of inorganic trace minerals increased, litter size declined, litter size increased when sows were fed organic trace minerals and the results were not significant. It should be noted, however, that the organic minerals fed in this experiment included a complete replacement of all essential trace minerals, and the effects of individual trace minerals could not be ascertained. The recent reproductive history in our herd has shown an average 11.4 total pigs and 11.0 pigs live born (Mahan and Peters, 2004). Although a direct comparison cannot be made, the diet composition, management practices, genetic composition of gilts and sows, and facility conditions used in the previous report were comparable to the 2 treatment groups that contained the IND inorganic trace mineral plus the elevated Ca and P levels (i.e., IND + Ca:P). Thus this experiment implies an improvement in sow reproductive performance from the organic trace mineral source.
A limited amount of research has been conducted to investigate the effectiveness of organic trace minerals for reproducing sows. Partial replacement of inorganic trace mineral sources of Cu, Mn, and Zn with organic trace mineral sources has resulted in more live fetuses and fewer dead embryos at 30 d postcoitum (Mirando et al., 1993). In that study, the number of corpora lutea was similar between treatments, but the number of live embryos was greater when organic trace minerals were fed. Greater dietary levels of Cu, Mn, and Zn have been reported in the conceptus products than in the surrounding endometrial tissues and ovaries between 12 and 30 d postcoitum (Hostetler et al., 2000). This indicates an increased uptake or improved utilization, or both, of trace minerals by the embryo and fetus during early pregnancy.
When the NRC treatment diet was fed, the calculated daily intake of trace minerals during gestation generally met the requirements of the sow as predicted by the NRC (1998) model, but the intake was somewhat below the requirement during lactation. However, when expressed on a milligrams per kilogram of BW basis, the mineral intake was below the requirement and was lowered as parity advanced, particularly during lactation. By adding indigenous supply to the trace mineral, intakes would be even greater than the NRC (1998) model in all treatment groups.
Our results further indicated that milk fat content at weaning declined to parity 4, whereupon it increased to parity 6 (P < 0.10). These responses are consistent with the greater backfat thickness during the latter parities and the greater feed and energy intakes of the sow during gestation. The corresponding decline in both milk fat and backfat thickness to parity 4 implies that body fat was mobilized and served as a major contributor to milk fat content and that gestation feed intake may be greater than that recommended by NRC (1998). The negative relationship between body fat content and backfat thicknesses is well documented (Mullan and Williams, 1989; Dourmad, 1991; Weldon et al., 1994). Lactation sow feed intakes were shown to decline during the latter parities, and the potentially lower milk production was the probable reason for the lower litter BW and gains by weaning, responses consistent with those of King and Dunkin (1985, 1986).
In conclusion, gilt growth performance between 30 and 110 kg of BW was not affected by trace mineral source and level. Feeding reproducing sow organic trace minerals resulted in a greater number of total and live pigs born. Increasing dietary Ca and P above NRC levels was not beneficial and may have been detrimental for sow reproductive performance. We would further conclude that the indigenous minerals contributed from feed grains should be ignored in formulating sow diets, at least when using current NRC (1998) requirement levels.
|
Footnotes |
|---|
2 Appreciation is expressed to K. Mays and L. Warnock for animal care and data collection, H. Zerby and G. Dunlap for animal slaughter and processing, and K. Jewell (Star Labs, Wooster, OH) for mineral and B. Bishop for statistical analyses. All personnel were from Ohio Agricultural Research and Development Center and The Ohio State University.
3 Current address: P&G Pet Care, 6571 State Route 503 N., Lewisburg, OH 45338.
4 Corresponding author: mahan.3{at}osu.edu
Received for publication July 17, 2007. Accepted for publication April 7, 2008.
|
LITERATURE CITED |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
This article has been cited by other articles:
|
|
 |
|
  J. C. Peters and D. C. Mahan Effects of neonatal iron status, iron injections at birth, and weaning in young pigs from sows fed either organic or inorganic trace minerals J Anim Sci, September 1, 2008; 86(9): 2261 - 2269. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
