Effects of Crystalline Amino Acid Concentrations With or Without Formaldehyde Treatment of Diets on Nursery Pig Growth Performance and Fecal Bacterial Concentration

A total of 1,235 nursery pigs (PIC 359 × 1050; initially 26.9 lb BW) were used in a 28-d study evaluating the effects of crystalline amino acid concentrations with or without formaldehyde treatment of diets on nursery pig growth performance, feed bacteria concentration, lysine content, and fecal microbial diversity. Sal CURB (Kemin Industries Inc., Des Moines, IA) is a commercial formaldehyde product that is commonly utilized in the poultry industry for Salmonella control in feed but has also been shown to reduce PEDV infectivity in swine diets. Pigs were weaned at approximately 21 d, fed a common starter diet for 10 d, and allotted to pens based on BW in a completely randomized design. Experimental diets were fed in 2 phases (phase 1, d 0 to 12; and phase 2, 12 to 28 post-weaning) in meal form. Experimental treatments were arranged as a 2 × 2 + 1 factorial with main effects of formaldehyde (none vs. 0.30% in all phases) and crystalline AA concentration (low vs. high) plus a positive control. The positive control represented this current production system’s formulated Lys requirement needed to maximize performance, whereas treatment diets were formulated at 80% of the positive control’s lysine concentration. Feed bacterial concentration was determined by performing aerobic plate, Enterobacteriaceae, and total coliform counts on composited feed samples collected from each batch of feed manufactured at the feed mill and directly from feeders at the farm. Total, available, and free Lys analyses were conducted on composited feed samples collected from each phase of the study to determine Lys content. A composite fecal sample was collected from 3 randomly selected pigs per pen on d 28 for each treatment, DNA isolated, and each sample assessed for bacterial community analysis. Overall, a significant crystalline AA × formaldehyde interaction (P < 0.05) was observed for ADFI and F/G. The interaction for ADFI was because added formaldehyde in high crystalline AA diets decreased feed intake; however, in low crystalline AA diets, ADFI was unchanged. For F/G, pigs had improved F/G in low crystalline AA diets without formaldehyde, but no difference was observed in high crystalline AA diets. Despite the interaction for ADFI and F/G, formaldehyde-treated diets reduced (P < 0.05) ADG, ADFI, and resulted in poorer F/G. Crystalline AA concentration did not impact performance. Added formaldehyde reduced or eliminated bacterial concentration of complete feed in phase 1 of the study. Formaldehyde reduced total and available Lys in both low and high crystalline AA diets, with a greater reduction occurring in low crystalline AA diets, but had no effect on free Lys. Added formaldehyde reduced (P = 0.001) Lactobacillaceae bacterial species, but increased (P = 0.001) Clostridiaceae bacterial species in fecal microbial samples. As expected, formaldehyde treatment reduced bacterial microflora of complete feeds. Overall, the level of crystalline AA did not impact performance while the nursery diet formaldehyde addition negatively influenced growth performance, AA utilization, and fecal microbial diversity.


Summary
A total of 1,235 nursery pigs (PIC 359 × 1050; initially 26.9 lb BW) were used in a 28-d study evaluating the effects of crystalline amino acid concentrations with or without formaldehyde treatment of diets on nursery pig growth performance, feed bacteria concentration, lysine content, and fecal microbial diversity.Sal CURB (Kemin Industries Inc., Des Moines, IA) is a commercial formaldehyde product that is commonly utilized in the poultry industry for Salmonella control in feed but has also been shown to reduce PEDV infectivity in swine diets.
Pigs were weaned at approximately 21 d, fed a common starter diet for 10 d, and allotted to pens based on BW in a completely randomized design.Experimental diets were fed in 2 phases (phase 1, d 0 to 12; and phase 2, 12 to 28 post-weaning) in meal form.Experimental treatments were arranged as a 2 × 2 + 1 factorial with main effects of formaldehyde (none vs. 0.30% in all phases) and crystalline AA concentration (low vs. high) plus a positive control.The positive control represented this current production system's formulated Lys requirement needed to maximize performance, whereas treatment diets were formulated at 80% of the positive control's lysine concentration.Feed bacterial concentration was determined by performing aerobic plate, Enterobacteriaceae, and total coliform counts on composited feed samples collected from each batch of feed manufactured at the feed mill and directly from feeders at the farm.Total, available, and free Lys analyses were conducted on composited feed samples collected from each phase of the study to determine Lys content.A composite fecal sample was col-

Introduction
Formaldehyde can be included in animal feed or ingredients to maintain complete feed and ingredients Salmonella negative for up to 21 d.Since the emergence of porcine epidemic diarrhea virus (PEDV) in the United States, formaldehyde products have received attention as a potential method to reduce the risk of PEDV transmission due to the ability of complete feed serving as a vector for the transmission of the disease. 4o reduce this risk, research using formaldehyde to reduce PEDV infectivity in contaminated feed and ingredients has been successful. 5,6However, formaldehyde is known to produce reactions with numerous groups of amino acid residues of proteins that can lead to the formation of methylol groups, Schiff-bases, and methylene bridges amongst these residues. 7Thus, inclusion in diets may reduce the availability of dietary AA for pigs, which may influence growth performance and nutrient utilization.Limited research exists regarding the effects formaldehyde treatment of diets has on pig performance, and no data exist that measure the influence it has on fecal microbial concentrations.Therefore, the objective of this study was to evaluate the effects of dietary crystal-4 Dee, S., Clement, T., Schelkopf, A., Nerem, J., Knudsen, D., Christopher-Hennings, J. and E. Nelson 2014.An evaluation of contaminated complete feed as a vehicle for porcine epidemic diarrhea virus infection of naïve pigs following consumption via natural feeding behavior: proof of concept

Materials and Methods
The Kansas State University Institutional Animal Care and Use Committee approved the protocol for this experiment.The trial was conducted in a tunnel-ventilated commercial grow-finish barn in central Iowa (Gourley Bros. Inc., Webster City, IA).Each pen (18.5 × 8 ft) had slatted flooring, one 4-hole self-feeder, and a pan waterer to provide ad libitum access to feed and water.
A The treatments were arranged in a 2 × 2 + 1 factorial with main effects of formaldehyde (none vs. 0.30% Sal CURB; Kemin Industries Inc., Des Moines, IA) and crystalline AA inclusion (low vs. high) plus a positive control diet.Sal CURB is a premix of 37% aqueous formaldehyde and propionic acid.All treatment diets were formulated to be 80% of the standardized ileal digestible (SID) Lys of that contained in the positive control, which was also 90 to 95% of the SID Lys requirement according to NRC.8A positive control was used in the experiment to represent diets that met the assumed SID Lys requirement for maximum growth performance in this system.The treatment ingredients were substituted for an equivalent amount of corn in the respective diets to form the experimental treatments (Table 1).
All diets were corn-soybean meal-based and were formulated in 2 phases.Diets were fed in meal form and were prepared at a commercial feed mill (Altoona, IA).Formaldehyde inclusion and application methods were conducted according to manufacturers' recommendations, with inclusion occurring in the mixer.Diets in each phase were collected from the mill and 6 randomly selected feeders, pooled within collection location, and submitted for analysis of DM, CP, Ca, P, propionic acid, and Lys content; specifically, total Lys, free Lys, and available Lys (Tables 2 and 4).Propionic acid was analyzed according to manufacturer's procedures and was analyzed to confirm correct inclusion rates of Sal CURB to treatment diets.
Feed bacterial concentration was tested for samples collected during manufacturing and from the farm using 3M Petrifilm plates (3M Microbiology, St. Paul, MN) with each of these plates selecting for: total coliforms (TC), aerobic plate counts (APC), or Enterobacteriaceae (EB).A plate reader was used to enumerate each plate for specific ranges, colony morphology, gas production, and acidification.Colony counts were expressed as colony forming units per gram of feed sample (cfu/g) and bacterial counts were expressed as an average of 2 separate runs processed in duplicate with a different feed sample.

Swine Day 2017
Fecal samples were collected into individual Whirl-Pak bags via rectal massage from 6 randomly pigs per pen on d 28.Samples were stored at 4°C and then transported to Kansas State University where d 28 samples were pooled within pen, for a final total of 12 samples per treatment.Samples were stored at -80°C until they were transported to the University of Nebraska-Lincoln for bacterial isolation and community analysis.
Fecal DNA from the pooled samples were isolated from 100 mg of each sample and PCR analysis was performed to amplify the 16S rRNA gene specific to bacterial communities.The amplified gene was then sequenced and subjected to bacterial community analysis.
Growth data were analyzed as a randomized complete block design using the PROC GLIMMIX procedure of SAS version 9.4 (SAS Institute, Inc., Cary, NC) with pen as the experimental unit.Pre-planned contrasts were utilized to compare the interactive and main effects of formaldehyde and crystalline AA inclusion, and the positive control vs. the other treatments.Results were considered significant at P ≤ 0.05 and marginally significant at P > 0.05 and P ≤ 0.10.
Bacterial community data were analyzed as a completely randomized block design using the PROC GLIMMIX procedure of SAS and the responses were presented as leastsquares means (± SEM).Additionally, OTU abundances at family level in the bacterial communities were analyzed using the GLIMMIX procedure of SAS.Results were considered significant at P ≤ 0.05 and marginally significant at P > 0.05 and P ≤ 0.10.

Results and Discussion
Results of the diet analysis closely matched those of formulated levels (Table 2).Propionic acid analysis of diets, used as an indicator of formaldehyde concentration as well, confirmed that targeted concentrations were at the correct levels in respective dietary treatments.
Analysis of both total Lys and available Lys in the positive control and treatment diets revealed lower levels than the formulated values (Table 3).In phase 1, the low crystalline AA formulated diets total and available Lys were reduced 8.7 and 10.4% when diets were treated with formaldehyde, respectively.In high crystalline AA diets, formaldehyde inclusion marginally reduced total and available Lys by 3.2% each.In phase 2, formaldehyde addition in low crystalline AA formulated diets reduced total and available Lys by 12.6 and 13.1%, respectively.Added formaldehyde in high crystalline AA formulated diets had little to no effect on total Lys with a reduction of 0.93 and 0.91%, respectively.Formaldehyde addition had no observed effect on free Lys, which can be considered an indicator of the amount of crystalline AA added to the diets.
For diet bacterial concentrations, as anticipated, formaldehyde treatment reduced the bacterial concentration of samples collected at both the feed mill and the farm compared to diets not treated with formaldehyde (Table 4).However, in phase 2 feed mill and the farm samples, bacterial loads for either non-treated or treated diets with Sal CURB were similar regardless of crystalline amino acid inclusion, except for the high crystalline treatments collected at the feed mill where formaldehyde treatment did reduce bacterial load.
From d 0 to 12, there was no evidence (P > 0.10) for a crystalline AA concentration × formaldehyde interactions or crystalline AA concentration main effects (Table 5).Pigs fed diets containing formaldehyde had poorer (P = 0.001) ADG and F/G compared to pigs fed diets that did not contain formaldehyde.Pigs fed the control diet had better (P = 0.05) ADG and F/G compared to pigs fed the other diets containing reduced Lys.
From d 12 to 28, a crystalline AA × formaldehyde interaction (P < 0.05) was observed for ADFI and F/G.This interaction for feed intake was a result of pigs fed high crystalline AA diets treated with formaldehyde having lower ADFI compared to the non-treated feed, while in the low crystalline AA diets formaldehyde treatment had no effect on ADFI.The F/G interaction was observed because pigs fed low crystalline AA diets without the inclusion of formaldehyde resulted in better F/G than pigs fed diets with formaldehyde; however, the inverse was observed in high crystalline AA diets.A tendency (P = 0.073) for a crystalline AA × formaldehyde interaction was observed for ADG, with pigs fed the low crystalline AA diets having a more dramatic decrease in ADG when formaldehyde was included compared to the high crystalline AA diets.Pigs fed formaldehyde-treated feed had reduced (P < 0.05) ADG and d 28 BW, they also tended (P = 0.052) to have reduced ADFI compared to pigs fed non-formaldehydetreated diets.Pigs fed the control diet had better (P < 0.05) ADG and F/G compared to pigs fed the other diets containing reduced Lys.There was no evidence of difference between diets containing low and high levels of crystalline AA for any growth criteria measured.
Overall (d 0 to 28), a significant crystalline AA × formaldehyde interaction (P < 0.05) was observed for ADFI and F/G.The interaction for ADFI occurred because pigs fed diets with high crystalline AA inclusions and formaldehyde treatment had poorer ADFI compared to pigs fed diets without formaldehyde, but in the low crystalline AA diets, ADFI was the same.The interaction for F/G was observed because pigs fed low crystalline AA diets without formaldehyde had better F/G than with the formaldehyde treatment, but pigs fed high levels of crystalline AA had similar F/G regardless of formaldehyde inclusion.Despite the interaction, the application of formaldehyde to diets resulted in reduced (P < 0.05) ADG, ADFI, and ending BW and poorer F/G compared to diets without the application of formaldehyde.Pigs fed the control diet had improved (P < 0.05) ADG, ending BW, and F/G compared to those fed other diets containing reduced Lys.There was no evidence of difference between diets containing low and high levels of crystalline AA for any response criteria measured throughout the trial.
For bacterial community abundance, no evidence of a difference (P > 0.10) in bacterial abundances amongst the dietary treatments for Methanobacteriaceae, Prevotellaceae, Lachnospiraceae, or Spirochaetaceae (Table 6).A significant crystalline AA × formaldehyde interaction (P = 0.003) was observed for Streptococcaceae abundances in the bacterial community of the gut, because pigs fed low crystalline AA diets had a more dramatic reduction in abundance when treated with formaldehyde compared to the high crystalline AA diets.The treatment of diets with formaldehyde decreased (P < 0.05) bacterial abundance for Paraprevotellaceae and Lactobacillaceae species, while formaldehyde treatment increased (P < 0.05) Clostridiaceae and Erysipelotrichaceae species within the bacterial community of the gut.Pigs fed formaldehyde-treated diets tended (P = 0.074) to have lower percentages of S24-7 bacteria species than pigs fed non-formaldehyde treated diets.Pigs fed low crystalline AA diets had increased (P < 0.05) abundance of Paraprevotellaceae, Lactobacilliaceae, Ruminococcaceae, and Veillonellaceae bacterial species compared to high crystalline AA diets.Pigs fed high crystalline AA diets had increased (P = 0.007) Clostridiaceae and tended (P = 0.080) to have increased Erysipelotrichaceae bacterial species compared to pigs fed low crystalline AA diets.Treatment diets fed to lower lysine levels than the control had increased (P = 0.009) Clostridiaceae bacterial species, while Paraprevotellaceae species tended (P = 0.091) to be lower in these diets compared to the positive control.
These data provide evidence that in late-nursery pigs the inclusion of formaldehyde in complete feeds has a negative impact on ADG, ADFI, F/G, and ending BW, when diets are fed below the Lys requirement of the pigs.Furthermore, it can be observed that inclusion of formaldehyde in complete nursery diets reduced the amount of total and available Lys within the diet, which suggest formaldehyde is affecting AA availability of the diet.Formaldehyde treatment of complete feeds also negatively impacts the gut microflora of late nursery pigs.This can be observed in the decrease of lactic acid bacterial species, specifically Lactobacillaceae that has the potential to improve gastrointestinal function.However, formaldehyde treatment increased Clostridiaceae bacterial species that can lead to enteric disruptions and promote proliferation of enteric disease.
These results suggest that the level of crystalline AA in the diets did not impact performance.In summary, formaldehyde treatment of feed reduced bacterial concentration within complete diets and affected fecal bacterial abundance.Also, these results suggest that formaldehyde is effective at reducing pathogen load within the feed, but formulation adjustments should be considered to reduce the negative impact on performance due to decreased AA availability.  1 Phase 1 diets fed from ~26.9 to 38.7 lb BW and phase 2 diets from ~38.7 to 60.5 lb BW.
2 Complete diet samples were obtained from each dietary treatment during manufacturing and from the farm feeder.Samples of diets were pooled and then submitted for analysis of DM, CP, Ca, and P (Ward Laboratories, Inc., Kearney, NE).

Table 3 .
Effect of formaldehyde-treated diets and crystalline amino acid level on dietary lysine content, % 1,2 Phase 1 diets fed from ~26.9 to 38.7 lb BW and phase 2 diets from ~38.7 to 60.5 lb BW. 2 Complete diet samples were obtained from each dietary treatment during manufacturing and from the farm feeder.Samples of diets were pooled and then submitted for analysis of total lysine, available lysine, and free lysine (Experiment Station Chemical Laboratories, University of Missouri-Columbia, Columbia, MO).Values represent average of duplicate analyses on pooled samples. 3Control diets were formulated to exceed the SID Lys requirement (NRC, 2012). 4Sal-CURB, Kemin Industries Inc., Des Moines, IA.

Table 4 .
Effect of formaldehyde-treated diets and crystalline amino acid level on complete feed bacterial concen-Indicates feed samples were collected directly from each individual batch of feed for each dietary treatment in each phase during manufacturing.Five equally spaced sub-samples were collected by passing sterile Whirl-Pak through stream of lot during manufacturing and pooled to create one composite sample for each dietary treatment in each phase to represent feed mill sample. 8Indicates feed samples were collected from 6 randomly chosen feeders from each dietary treatment in each phase.The 6 sub-samples were then pooled into one composite sample for each dietary treatment in each phase to represent farm sample.
2Complete feed samples from each dietary treatment and phase were collected during manufacturing and from the farm for enumeration of feed bacterial concentration (Department of Diagnostic Medicine/Pathobiology, College of Veterinary Medicine, Kansas State University, Manhattan, KS).3Feed bacterial concentrations are expressed as colony forming units per gram of feed sample (cfu/g).4Controldietswereformulated to exceed the SID Lys requirement (NRC, 2012).5Treatmentdiets were formulated to 80% of the control diet and contained low or high levels of crystalline AA. 6 Sal-CURB, Kemin Industries Inc., Des Moines, IA. 7

Table 5 .
Effect of formaldehyde-treated diets and crystalline amino acid level on nursery pig performance 1 ,b,c,d Means within same row with different superscripts differ (P < 0.05). 1 A total of 1,235 pigs (PIC 359 × PIC 1050, initially 26.9 ± 0.02 lb) were used in a 2-phase nursery study with 19 to 22 pigs per pen and 12 replications per treatment.Pigs were weaned at approximately 21 d, fed a common starter diet for 10 d post-weaning, and then fed experimental diets. 2 Control diets were formulated to exceed the SID Lys requirement (NRC, 2012).3Treatmentdiets were formulated to 80% of the control diet and contained low or high levels of crystalline AA. 4 Sal-CURB, Kemin Industries Inc., Des Moines, IA. a

Table 6 .
Effect of formaldehyde-treated diets and crystalline amino acid level on fecal bacterial abundances at phylum level1,2