Propionibacterium acidipropionici P1691 and glucogenic precursors improve rumen fermentation of low-quality forage in beef cattle2 P. H. Sanchez,* L. N. Tracey,* J. Browne-Silva,* and S. L. Lodge-Ivey*3 *Department of Animal and Range Sciences, New Mexico State University, Las Cruces 88003
ABSTRACT: Cattle grazing dormant western rangelands may have a high ruminal acetate to propionate ratio (A:P) and may have low tissue clearance of acetate. Increasing propionate production could shift this ratio and improve animal performance. In Exp. 1, the effect of Propionibacterium acidipropionici P169 (PA) on forage digestibility and VFA production was evaluated in vitro using 2 substrates: 100% dormant warm-season grass extrusa and 50% sorghum–Sudan hay with 50% ground corn (DM basis). The objective of Exp. 2 was to evaluate the effect of PA or calcium propionate supplementation on digestibility, ruminal fermentation, acetate clearance, and BW change. Twelve 2-yr-old, pregnant Brangus heifers (BW = 416 ± 85 kg) were assigned to 1 of 3 treatments. All cattle were fed a basal ration of Old World Bluestem hay (Bothriochloa ischaemum; 5.8% CP and 76.5% NDF, DM basis) at 1.5% BW from d –10 to d 49. Treatments included a protein supplement (CON; 36% CP and 35% RUP, DM basis; 454 g/animal fed twice daily), CON plus 6 × 1010 cfu PA/animal (BACT), and CON plus 80 g calcium propionate (PROP). After initiation of treatments (d 0), rumen fluid was collected via oral lavage
every 3 d and analyzed for VFA, pH, and ammonia. Glucogenic potential of treatments was evaluated with an acetate tolerance test on d 49. In Exp. 1, PA addition increased (IVDMD; P < 0.001) and total VFA (P < 0.001) of 100% dormant warm-season grass extrusa but not 50% sorghum–Sudan hay with 50% ground corn (P ≥ 0.28). The addition of P169 decreased (P < 0.001) acetate, increased propionate (P < 0.001), and decreased A:P ratio (P < 0.001) for both substrates. In Exp. 2, total tract OM and NDF digestibility and ruminal pH, total VFA, and acetate did not differ (P ≥ 0.13) among treatments. Propionate concentration was least (P = 0.001) for CON, intermediate for P169, and greatest for PROP. Conversely, A:P ratio was greatest (P < 0.004) for CON, intermediate for P169, and least for PROP. Acetate clearance did not differ (P = 0.69) among treatments. Propionibacterium acidipropionici P169 increased IVDMD and total VFA of low-quality forage. Supplementation with PA and calcium propionate salts increased propionate and decreased A:P in the rumen. Supplementation of PA represents a potential way to increase ruminal propionate concentration when dormant forages are fed.
Key words: beef cattle, calcium propionate, forage, glucogenic precursors, Propionibacterium acidipropionici P169 © 2014 American Society of Animal Science. All rights reserved.
J. Anim. Sci. 2014.92:1738–1746 doi:10.2527/jas2013-7148 INTRODUCTION
1The authors would like to thank Danisco for their generous donation of Propionibacterium acidipropionici P169 and AG Research, LLC, for assistance with manuscript preparation. 2Research supported by the New Mexico Agriculture Experiment Station. 3Corresponding author: [email protected] Received September 12, 2013. Accepted January 29, 2014.
Volatile fatty acids represent the main supply of ME for ruminants (Van Soest, 1982). Acetate is the predominant VFA in most mammalian systems (Bergman, 1990) and is usually produced in greater amounts in ruminants consuming forages when compared to concentrate diets (Owens and Goetsch, 1988). Providing glucogenic precursors such as propionate to ruminants may influence metabolic dynamics by changing
1738 Downloaded from https://academic.oup.com/jas/article-abstract/92/4/1738/4703455 by guest on 14 January 2018
Glucogenic precursors improve fermentation
the balance of intermediates required in oxidative metabolism, influencing overall energy utilization, and improving overall energy balance during times of forage dormancy (Mulliniks et al., 2011). Supplementing sodium propionate or calcium propionate salts has increased propionate and decreased acetate concentration in the rumen of lambs consuming mature oat hay (van Houtert et al., 1993) and ammoniated barley straw (van Houtert and Leng, 1993). Supplementation with Propionibacterium acidipropionici P169 (PA) has also resulted in increased propionate and decreased acetate concentrations in the rumen of beef (Lehloenya et al., 2008) and dairy cattle (Weiss et al., 2008) consuming mixed rations. Few researchers have evaluated the effects of these PA on digestion and rumen fermentation in beef cattle consuming low-quality forages. We hypothesized that the addition of PA or calcium propionate to beef cattle diets high in structural carbohydrates would increase ruminal propionate production and thereby increase hepatic gluconeogenesis and allow for acetate to be used more efficiently by the body tissues. Two experiments were conducted to 1) evaluate the effect of PA on in vitro forage digestibility and VFA production and 2) compare the effect of supplementing PA versus calcium propionate on in vivo digestibility, ruminal fermentation, acetate clearance, and animal performance. MATERIALS AND METHODS Animal procedures were conducted at the New Mexico State University (NMSU) Large Animal Laboratory in Las Cruces, NM. The NMSU Institutional Animal Care and Use Committee reviewed and approved experiments that are described herein. Experiment 1 Growth of Propionibacterium acidipropionici P169. Propionibacterium acidipropionici P169 was obtained from the American Type Culture Collection (ATCC PTA-5272; Manassas, VA) and grown in a sodium lactate broth medium (Østlie et al., 1995) containing 1% tryptone (Difco Laboratory, Detroit, MI), 1% yeast extract (Difco Laboratory), 1.5% of 50% sodium lactate (Fisher Scientific, Fair Lawn, NJ), 0.25% K2HPO4 (Fisher Scientific), 0.02% MgSO4 (Sigma-Aldrich Inc., St. Louis, MO), and 0.02% MnSO4·H2O (Sigma-Aldrich, Inc.). All medium components were combined in a round bottom flask, heated until boiling in a microwave, and bubbled with CO2 until cool to touch. The flask was closed with a butyl rubber stopper and transferred to an anaerobic glove box (95% CO2 and 5% H2 atmosphere; Coy Laboratory Products, Grass Lake, MI). The medium was aliquoted (9.70 mL) to 25-mL (18 by 150 mm) Balch Downloaded from https://academic.oup.com/jas/article-abstract/92/4/1738/4703455 by guest on 14 January 2018
1739
tubes, capped with butyl rubber stoppers, removed from the glove box, sealed with aluminum seals, and sterilized by autoclaving (Hungate, 1960). Propionibacterium acidipropionici P169 was added aseptically through the stopper using a needle and syringe and cultures were incubated at 37°C in a shaking incubator (New Brunswick Scientific, Enfield, CT). Growth of PA was verified by measuring culture turbidity (600 nm) every 3 h for 24 h. Growth curves and doubling times were developed from these data using standard curves created by plating serial dilutions of the bacterium and growth medium and were used to determine when a cell density of 6 × 1010 cfu was obtained (approximately 8 h; data not shown). In Vitro Substrates. Two different feeds were used as fermentation substrates, simulating common ruminant production diets (DM basis): 100% dormant warmseason grass extrusa (4.6% CP and 56.1% NDF, DM basis) and 50% sorghum–Sudan hay (8% CP and 57% NDF, DM basis) with 50% ground corn (8% CP and 31.6% NDF, DM basis). Extrusa samples were collected at the Corona Range Livestock Research Center located 13 km east of Corona, NM (average elevation = 1,900 m), during the fall of 2009. Forages at this study site were primarily blue grama (Bouteloua gracilis), threeawns (Aristida spp.), and common wolf tail (Lycurus phleoides). Two ruminally cannulated cows were held off feed overnight but allowed ad libitum access to water. Rumen contents of the cows were evacuated at 0800 h following the overnight feed withdrawal. Cows were then allowed to graze for 1 h. The cows were then gathered and extrusa collected from the rumen and stored at –20°C until processed. Extrusa samples were dried in a forced-air oven (50°C) and ground using a Wiley Mill (Thomas Scientific, Swedesboro, NJ) equipped with a 2-mm screen. Triplicate samples of each substrate (0.5 g) were weighed and placed in 45-mL polypropylene in vitro tubes (Bel-Art, Wayne, NJ). Substrates were also analyzed for DM, ash (AOAC International, 2000), NDF using an ANKOM 200 fiber analyzer (ANKOM Technology Corp., Fairport, NY) with sodium sulfite and amylase, and CP by combustion using a LECO nitrogen analyzer (LECO Corp. St. Joseph, MO). In Vitro Incubations. Two cows were used as ruminal fluid donors and allowed ad libitum access to sorghum–Sudan hay. Ruminal fluid was obtained approximately 3 h after feeding and blended in equal parts. A metal suction strainer (Precision Machine Co. Inc., Hasting, NE) was used to collect ruminal fluid from the ventral and anterior sections of the rumen into a collection thermos previously heated to 37°C. Ruminal fluid was immediately transported to the laboratory, combined with equal parts McDougall’s buffer (Tilley and Terry, 1963), and gassed with CO2 to make the IVDMD inoculum. Each liter of McDougall’s buffer was composed of
1740
Sanchez et al.
9.8 g NaHCO3, 2.77 g Na2HPO4, 0.57 g KCl, 0.47 g NaCl, 0.12 g MgSO4·7H2O, and 0.16 g CaCl2·2H2O. In vitro tubes with substrate were filled with 20 mL of the inoculum, gassed with CO2, and capped with a plastic cap. Cultures of pure PA (6 × 1010 cfu in 500 μL incubated for 8 h at 37°C) were added to each tube as a sterile addition using a needle and syringe. In vitro tubes were incubated at 37°C in an anaerobic glove box (90% C02 and 10% H2 atmosphere) for 48 h. Tubes were manually agitated every 1 h for 12 h at the beginning of experiment and agitated every 12 h for the remainder of the experiment. Tubes were centrifuged at 5,500 × g for 15 min at 4°C on removal from the anaerobic glove box. Samples of supernatant were collected and stored at –20°C until analysis for VFA using gas chromatography (Goetsch and Galyean, 1983). In vitro DM digestibility was calculated from the amount of substrate remaining after digestion with ruminal fluid inoculum for 48 h. Statistical Analysis. Each in vitro incubation was repeated 3 times with treatments in triplicate for a total of 9 observations per treatment in a randomized block design. In vitro DM digestibility and VFA data were analyzed within each substrate using the MIXED procedures of SAS (SAS Inst., Inc., Cary, NC) with treatment (PA or no PA) as a fixed effect. Effects were considered significantly different when P ≤ 0.05 and a tendency when P > 0.05 and P ≤ 0.10. When a significant F-test was observed, least squares means were separated using the PDIFF option. There were no block × treatment interactions; therefore, only main effects will be presented. Experiment 2 Animals and Experimental Design. Twelve 2-yrold pregnant Brangus heifers (416 ± 85 kg BW) from the Chihuahuan Desert Rangeland Research Center (Las Cruces, NM) were used in a completely randomized design with 3 treatments (n = 4 heifers/treatment). Heifers were approximately 5 mo pregnant at the beginning of the experiment. Heifers were weighed and unshrunk BW were collected on a weekly basis starting on d 0. Diets and Treatments. Heifers were individually penned outdoors in dirt floor pens measuring approximately 6 × 6 m equipped with automatic water fountains and hand fed twice daily at 0600 and 1800 h. Heifers received a basal ration of chopped Old World Bluestem hay (Bothriochloa ischaemum; 5.8% CP, 76.5% NDF, and 12% ash, DM basis) at 1.5% BW (DM basis) for 10 d before the initiation of treatments (d –10 to 0) and for the duration of the experiment (d 0 to 49). Basal diet nutrient analysis was conducted by a commercial laboratory (SDK Labs, Hutchinson, KS). Heifers were supplemented with 1 of 3 treatments: 1) a 36% protein supplement (DM basis) containing 57% cottonseed meal, 21% Downloaded from https://academic.oup.com/jas/article-abstract/92/4/1738/4703455 by guest on 14 January 2018
wheat middlings, 10% soybean meal, 9% molasses, and 1.2% urea and fortified with trace vitamins and minerals (DM basis; CON), 2) CON protein supplement plus 6 × 1010 cfu of PA (Danisco Animal Nutrition, Waukesha, WI; BACT), or 3) CON protein supplement plus 80 g calcium propionate salt (NutroCal; Kemin Industries, Inc., Des Moines, IA; PROP). Per manufacturer’s recommendations, the P169 supplement was created by dissolving 60 g of PA freeze-dried cell powder in 1,000 mL of distilled water and 250 mL of this solution was sprinkled over the basal diet at each feeding to achieve a dosage level of 6 × 1010 cfu of PA. Calcium propionate salt was mixed with the protein supplement during manufacture. Cubed CON and PROP treatment supplements were commercially manufactured (Hi Pro Feed, Dexter, NM) and 454 g was fed to each heifer at the time of basal diet feeding using the same feeder. Sampling and Measurements. Subsamples of offered supplements and basal diet were collected, weighed, and recorded daily. Orts were not collected for this experiment due to complete consumption of the basal diet and treatments by all animals. Titanium dioxide (TiO2; Sigma-Aldrich, Inc., St. Louis, MO) was used as an external marker to estimate total tract OM digestibility of the diet (Titgemeyer et al., 2001; Myers et al., 2004). Titanium dioxide was weighed out (5 g) into number 11 Lock Ring Capsules (Torpac Inc., Fairfield, NJ) and capsules were administered via metal balling gun (Torpac Inc.) twice daily at 0600 and 1800 h for 10 d starting on d 34. Starting at 0600 h on d 39, fecal grab samples were collected every 12 h for a 5-d sampling period. Sample collection time was advanced by 2 h each day to represent every 2 h in a 24-h period. Approximately 200 g of wet material was collected at each sampling time. Fecal samples were composited on an equal weight basis for each heifer, dried in a forcedair oven (50°C), ground in a Wiley Mill (Thomas Scientific, Swedesboro, NJ) equipped with a 1-mm screen, and analyzed for OM (AOAC International, 2000), titanium (Myers et al. [2004] adapted to a microtiter plate [BioTek Instruments, Winooski, VT; absorbance measured at 410 nm]), and NDF using an ANKOM 200 fiber analyzer (ANKOM Technology Corp., Fairport, NY) with sodium sulfite and amylase. Apparent total tract digestibility of OM was calculated by dividing the amount of Ti dosed by the Ti concentration in the fecal sample (Titgemeyer et al., 2001). Digestibility of NDF (DM basis) was calculated using the following equation: % NDF digestibility = 1 00 – [100 × (% NDF in feed/% NDF in feces × % NDF in feces/ % NDF in feed)].
Glucogenic precursors improve fermentation
Rumen fluid was collected every third day starting at d 0 (initiation of treatments) via oral lavage with a metal suction strainer 4 h after the morning feeding into a stoppered side-arm flask (Lodge-Ivey et al., 2009). Ruminal pH was measured using a portable pH probe (Accumet AP72; Fisher Scientific, Waltham, MA) within 5 min of collection. Ruminal fluid samples were divided into 2- to 10-mL aliquots for subsequent analysis of ammonia-N and VFA, placed on ice, frozen, and stored at –20°C in 15-mL polypropylene conical tubes (Fisher Scientific, Pittsburgh, PA). Aliquots for analysis of ruminal ammonia-N were acidified with 2.5 mL of 5% HCl before freezing. Samples were kept on ice and transported to the laboratory for storage at –20°C until analysis. Ruminal fluid samples were thawed, centrifuged, and analyzed for VFA using HPLC according to the method of Guerrant et al. (1982). Molar concentrations of VFA were obtained by multiplying peak heights by response factors after the detector response value for each acid had been determined. Ruminal ammonia-N was determined using the phenol-hypochlorite procedure (Broderick and Kang, 1980) using a spectrophotometer at a wavelength of 630 nm. During the acetate tolerance test (ATT), animals had access to the basal diet and were fed their respective treatments at 0600 h. Heifers were fitted with an indwelling jugular catheter 30 min before ATT on d 49 to evaluate acetate clearance from blood using methods and procedures described by Mulliniks et al. (2011). A 20% acetic acid solution (pH 7.40) was filter-sterilized using a Steritop Filter Unit (Millipore, Billerica, MA) and infused into each animal via an indwelling jugular catheter at a dosage of 1.25 mL/kg of BW over a 20-min period. Blood samples were collected at –1, 0, 1, 3, 5, 7, 10, 15, 30, 60, and 90 min following the administration of the acetic acid. Infusion of acetate occurred between –1 and 0 min. Blood samples (10 mL) were collected into serum separator tubes (Corvac, Mansfield, MA) and allowed to clot at room temperature for 45 min. Blood samples were centrifuged at 2,000 × g for 25 min at 4°C. Serum was harvested (2 mL) and then filtered through a centrifugal filter (Millipore Amicon Ultra, Ultracel-10K; Billeria, MA) at 7,000 × g for 80 min at 4°C. Filtered serum was frozen and later analyzed for serum acetate concentrations using HPLC (Guerrant et al., 1982). Molar concentrations of acetate were obtained by multiplying peak heights by response factors after the detector response value for acetate been determined (Guerrant et al., 1982). Acetate half-life was calculated as the time required for a 50% decrease in peak serum acetate concentration (Kaneko, 1997). Serum acetate areas under the curve were calculated using the trapezoidal summation method. Statistical Analysis. Ruminal fluid data were analyzed as a completely randomized design using the Downloaded from https://academic.oup.com/jas/article-abstract/92/4/1738/4703455 by guest on 14 January 2018
1741
MIXED procedure of SAS with repeated measures using Akaike’s information criterion to select the most appropriate covariance structure (compound symmetry). The model included treatment, day, and treatment × day interactions. Effects were considered significant when P ≤ 0.05 and a tendency when P > 0.05 and P ≤ 0.10. When a significant F-test was observed, least squares means were separated using the PDIFF option. Individual heifer was considered the experimental unit. Acetate half-lives were estimated for each heifer by regressing the logarithmically transformed acetate concentrations over time (Kaneko, 1997). Area under the curve was determined for acetate concentrations using trapezoidal summation method. The MIXED procedure of SAS was used to test all main effects of the ATT. The model included treatment, heifer, and their interaction. Results and Discussion Experiment 1 Addition of PA increased ruminal IVDMD (P < 0.001) and total VFA (P < 0.001) of 100% dormant warm-season grass extrusa (Table 1) but did not alter (P ≥ 0.28) IVDMD and total VFA of 50% sorghum–Sudan hay with 50% corn (Table 2). Addition of PA decreased (P < 0.001) acetate and increased (P < 0.001) propionate concentration resulting in lower (P < 0.001) acetate to propionate ratio (A:P) for both substrates (Tables 1 and 2). Little research has been published that reports the effect of supplemental Propionibacterium on in vitro rumen fermentation and digestibility of forage. In agreement with our results, Lehloenya et al. (2008) and Weiss et al. (2008) reported decreased acetate, increased propionate, and decreased A:P in beef and dairy cattle fed total mixed rations (TMR) and supplemented with PA. Akay and Dado (2001) also reported that Propionibacterium P5 increased total VFA and propionate and decreased A:P when added to alfalfa silage, fescue hay, wheat straw, and corn grain but decreased IVDMD and increased acetate in vitro. Ghorbani et al. (2002) reported similar 24-h in situ dry matter disappearance of alfalfa hay, barley silage, concentrate, and wheat straw incubated in the rumen of cattle consuming feedlot diets supplemented with Propionibacterium P15 when compared to cattle consuming a control diet. Differences in Propionibacterium strain, substrates, method, and incubation time likely account for variation in results between studies. Our results suggest that PA has the capability of enhancing VFA production and improving digestion of low-quality forages by grazing ruminants.
1742
Sanchez et al.
Table 1. Effects of Propionibacterium acidipropionici P169 (PA) on in vitro fermentation of 100% dormant warm-season grass extrusa1 PA2 Item IVDMD, % Total VFA, mM Acetate, mol/100 mol Propionate, mol/100 mol Acetate:propionate
–
+
SEM
P-value
20.5 92.8 70.3 20.4 3.6
32.0 119.1 68.5 22.0 3.1
0.85 2.91 0.28 0.27 0.09
ABSTRACT: Cattle grazing dormant western rangelands may have a high ruminal acetate to propionate ratio (A:P) and may have low tissue clearance of acetate. Increasing propionate production could shift this ratio and improve animal performance. In Exp. 1, the effect of Propionibacterium acidipropionici P169 (PA) on forage digestibility and VFA production was evaluated in vitro using 2 substrates: 100% dormant warm-season grass extrusa and 50% sorghum–Sudan hay with 50% ground corn (DM basis). The objective of Exp. 2 was to evaluate the effect of PA or calcium propionate supplementation on digestibility, ruminal fermentation, acetate clearance, and BW change. Twelve 2-yr-old, pregnant Brangus heifers (BW = 416 ± 85 kg) were assigned to 1 of 3 treatments. All cattle were fed a basal ration of Old World Bluestem hay (Bothriochloa ischaemum; 5.8% CP and 76.5% NDF, DM basis) at 1.5% BW from d –10 to d 49. Treatments included a protein supplement (CON; 36% CP and 35% RUP, DM basis; 454 g/animal fed twice daily), CON plus 6 × 1010 cfu PA/animal (BACT), and CON plus 80 g calcium propionate (PROP). After initiation of treatments (d 0), rumen fluid was collected via oral lavage
every 3 d and analyzed for VFA, pH, and ammonia. Glucogenic potential of treatments was evaluated with an acetate tolerance test on d 49. In Exp. 1, PA addition increased (IVDMD; P < 0.001) and total VFA (P < 0.001) of 100% dormant warm-season grass extrusa but not 50% sorghum–Sudan hay with 50% ground corn (P ≥ 0.28). The addition of P169 decreased (P < 0.001) acetate, increased propionate (P < 0.001), and decreased A:P ratio (P < 0.001) for both substrates. In Exp. 2, total tract OM and NDF digestibility and ruminal pH, total VFA, and acetate did not differ (P ≥ 0.13) among treatments. Propionate concentration was least (P = 0.001) for CON, intermediate for P169, and greatest for PROP. Conversely, A:P ratio was greatest (P < 0.004) for CON, intermediate for P169, and least for PROP. Acetate clearance did not differ (P = 0.69) among treatments. Propionibacterium acidipropionici P169 increased IVDMD and total VFA of low-quality forage. Supplementation with PA and calcium propionate salts increased propionate and decreased A:P in the rumen. Supplementation of PA represents a potential way to increase ruminal propionate concentration when dormant forages are fed.
Key words: beef cattle, calcium propionate, forage, glucogenic precursors, Propionibacterium acidipropionici P169 © 2014 American Society of Animal Science. All rights reserved.
J. Anim. Sci. 2014.92:1738–1746 doi:10.2527/jas2013-7148 INTRODUCTION
1The authors would like to thank Danisco for their generous donation of Propionibacterium acidipropionici P169 and AG Research, LLC, for assistance with manuscript preparation. 2Research supported by the New Mexico Agriculture Experiment Station. 3Corresponding author: [email protected] Received September 12, 2013. Accepted January 29, 2014.
Volatile fatty acids represent the main supply of ME for ruminants (Van Soest, 1982). Acetate is the predominant VFA in most mammalian systems (Bergman, 1990) and is usually produced in greater amounts in ruminants consuming forages when compared to concentrate diets (Owens and Goetsch, 1988). Providing glucogenic precursors such as propionate to ruminants may influence metabolic dynamics by changing
1738 Downloaded from https://academic.oup.com/jas/article-abstract/92/4/1738/4703455 by guest on 14 January 2018
Glucogenic precursors improve fermentation
the balance of intermediates required in oxidative metabolism, influencing overall energy utilization, and improving overall energy balance during times of forage dormancy (Mulliniks et al., 2011). Supplementing sodium propionate or calcium propionate salts has increased propionate and decreased acetate concentration in the rumen of lambs consuming mature oat hay (van Houtert et al., 1993) and ammoniated barley straw (van Houtert and Leng, 1993). Supplementation with Propionibacterium acidipropionici P169 (PA) has also resulted in increased propionate and decreased acetate concentrations in the rumen of beef (Lehloenya et al., 2008) and dairy cattle (Weiss et al., 2008) consuming mixed rations. Few researchers have evaluated the effects of these PA on digestion and rumen fermentation in beef cattle consuming low-quality forages. We hypothesized that the addition of PA or calcium propionate to beef cattle diets high in structural carbohydrates would increase ruminal propionate production and thereby increase hepatic gluconeogenesis and allow for acetate to be used more efficiently by the body tissues. Two experiments were conducted to 1) evaluate the effect of PA on in vitro forage digestibility and VFA production and 2) compare the effect of supplementing PA versus calcium propionate on in vivo digestibility, ruminal fermentation, acetate clearance, and animal performance. MATERIALS AND METHODS Animal procedures were conducted at the New Mexico State University (NMSU) Large Animal Laboratory in Las Cruces, NM. The NMSU Institutional Animal Care and Use Committee reviewed and approved experiments that are described herein. Experiment 1 Growth of Propionibacterium acidipropionici P169. Propionibacterium acidipropionici P169 was obtained from the American Type Culture Collection (ATCC PTA-5272; Manassas, VA) and grown in a sodium lactate broth medium (Østlie et al., 1995) containing 1% tryptone (Difco Laboratory, Detroit, MI), 1% yeast extract (Difco Laboratory), 1.5% of 50% sodium lactate (Fisher Scientific, Fair Lawn, NJ), 0.25% K2HPO4 (Fisher Scientific), 0.02% MgSO4 (Sigma-Aldrich Inc., St. Louis, MO), and 0.02% MnSO4·H2O (Sigma-Aldrich, Inc.). All medium components were combined in a round bottom flask, heated until boiling in a microwave, and bubbled with CO2 until cool to touch. The flask was closed with a butyl rubber stopper and transferred to an anaerobic glove box (95% CO2 and 5% H2 atmosphere; Coy Laboratory Products, Grass Lake, MI). The medium was aliquoted (9.70 mL) to 25-mL (18 by 150 mm) Balch Downloaded from https://academic.oup.com/jas/article-abstract/92/4/1738/4703455 by guest on 14 January 2018
1739
tubes, capped with butyl rubber stoppers, removed from the glove box, sealed with aluminum seals, and sterilized by autoclaving (Hungate, 1960). Propionibacterium acidipropionici P169 was added aseptically through the stopper using a needle and syringe and cultures were incubated at 37°C in a shaking incubator (New Brunswick Scientific, Enfield, CT). Growth of PA was verified by measuring culture turbidity (600 nm) every 3 h for 24 h. Growth curves and doubling times were developed from these data using standard curves created by plating serial dilutions of the bacterium and growth medium and were used to determine when a cell density of 6 × 1010 cfu was obtained (approximately 8 h; data not shown). In Vitro Substrates. Two different feeds were used as fermentation substrates, simulating common ruminant production diets (DM basis): 100% dormant warmseason grass extrusa (4.6% CP and 56.1% NDF, DM basis) and 50% sorghum–Sudan hay (8% CP and 57% NDF, DM basis) with 50% ground corn (8% CP and 31.6% NDF, DM basis). Extrusa samples were collected at the Corona Range Livestock Research Center located 13 km east of Corona, NM (average elevation = 1,900 m), during the fall of 2009. Forages at this study site were primarily blue grama (Bouteloua gracilis), threeawns (Aristida spp.), and common wolf tail (Lycurus phleoides). Two ruminally cannulated cows were held off feed overnight but allowed ad libitum access to water. Rumen contents of the cows were evacuated at 0800 h following the overnight feed withdrawal. Cows were then allowed to graze for 1 h. The cows were then gathered and extrusa collected from the rumen and stored at –20°C until processed. Extrusa samples were dried in a forced-air oven (50°C) and ground using a Wiley Mill (Thomas Scientific, Swedesboro, NJ) equipped with a 2-mm screen. Triplicate samples of each substrate (0.5 g) were weighed and placed in 45-mL polypropylene in vitro tubes (Bel-Art, Wayne, NJ). Substrates were also analyzed for DM, ash (AOAC International, 2000), NDF using an ANKOM 200 fiber analyzer (ANKOM Technology Corp., Fairport, NY) with sodium sulfite and amylase, and CP by combustion using a LECO nitrogen analyzer (LECO Corp. St. Joseph, MO). In Vitro Incubations. Two cows were used as ruminal fluid donors and allowed ad libitum access to sorghum–Sudan hay. Ruminal fluid was obtained approximately 3 h after feeding and blended in equal parts. A metal suction strainer (Precision Machine Co. Inc., Hasting, NE) was used to collect ruminal fluid from the ventral and anterior sections of the rumen into a collection thermos previously heated to 37°C. Ruminal fluid was immediately transported to the laboratory, combined with equal parts McDougall’s buffer (Tilley and Terry, 1963), and gassed with CO2 to make the IVDMD inoculum. Each liter of McDougall’s buffer was composed of
1740
Sanchez et al.
9.8 g NaHCO3, 2.77 g Na2HPO4, 0.57 g KCl, 0.47 g NaCl, 0.12 g MgSO4·7H2O, and 0.16 g CaCl2·2H2O. In vitro tubes with substrate were filled with 20 mL of the inoculum, gassed with CO2, and capped with a plastic cap. Cultures of pure PA (6 × 1010 cfu in 500 μL incubated for 8 h at 37°C) were added to each tube as a sterile addition using a needle and syringe. In vitro tubes were incubated at 37°C in an anaerobic glove box (90% C02 and 10% H2 atmosphere) for 48 h. Tubes were manually agitated every 1 h for 12 h at the beginning of experiment and agitated every 12 h for the remainder of the experiment. Tubes were centrifuged at 5,500 × g for 15 min at 4°C on removal from the anaerobic glove box. Samples of supernatant were collected and stored at –20°C until analysis for VFA using gas chromatography (Goetsch and Galyean, 1983). In vitro DM digestibility was calculated from the amount of substrate remaining after digestion with ruminal fluid inoculum for 48 h. Statistical Analysis. Each in vitro incubation was repeated 3 times with treatments in triplicate for a total of 9 observations per treatment in a randomized block design. In vitro DM digestibility and VFA data were analyzed within each substrate using the MIXED procedures of SAS (SAS Inst., Inc., Cary, NC) with treatment (PA or no PA) as a fixed effect. Effects were considered significantly different when P ≤ 0.05 and a tendency when P > 0.05 and P ≤ 0.10. When a significant F-test was observed, least squares means were separated using the PDIFF option. There were no block × treatment interactions; therefore, only main effects will be presented. Experiment 2 Animals and Experimental Design. Twelve 2-yrold pregnant Brangus heifers (416 ± 85 kg BW) from the Chihuahuan Desert Rangeland Research Center (Las Cruces, NM) were used in a completely randomized design with 3 treatments (n = 4 heifers/treatment). Heifers were approximately 5 mo pregnant at the beginning of the experiment. Heifers were weighed and unshrunk BW were collected on a weekly basis starting on d 0. Diets and Treatments. Heifers were individually penned outdoors in dirt floor pens measuring approximately 6 × 6 m equipped with automatic water fountains and hand fed twice daily at 0600 and 1800 h. Heifers received a basal ration of chopped Old World Bluestem hay (Bothriochloa ischaemum; 5.8% CP, 76.5% NDF, and 12% ash, DM basis) at 1.5% BW (DM basis) for 10 d before the initiation of treatments (d –10 to 0) and for the duration of the experiment (d 0 to 49). Basal diet nutrient analysis was conducted by a commercial laboratory (SDK Labs, Hutchinson, KS). Heifers were supplemented with 1 of 3 treatments: 1) a 36% protein supplement (DM basis) containing 57% cottonseed meal, 21% Downloaded from https://academic.oup.com/jas/article-abstract/92/4/1738/4703455 by guest on 14 January 2018
wheat middlings, 10% soybean meal, 9% molasses, and 1.2% urea and fortified with trace vitamins and minerals (DM basis; CON), 2) CON protein supplement plus 6 × 1010 cfu of PA (Danisco Animal Nutrition, Waukesha, WI; BACT), or 3) CON protein supplement plus 80 g calcium propionate salt (NutroCal; Kemin Industries, Inc., Des Moines, IA; PROP). Per manufacturer’s recommendations, the P169 supplement was created by dissolving 60 g of PA freeze-dried cell powder in 1,000 mL of distilled water and 250 mL of this solution was sprinkled over the basal diet at each feeding to achieve a dosage level of 6 × 1010 cfu of PA. Calcium propionate salt was mixed with the protein supplement during manufacture. Cubed CON and PROP treatment supplements were commercially manufactured (Hi Pro Feed, Dexter, NM) and 454 g was fed to each heifer at the time of basal diet feeding using the same feeder. Sampling and Measurements. Subsamples of offered supplements and basal diet were collected, weighed, and recorded daily. Orts were not collected for this experiment due to complete consumption of the basal diet and treatments by all animals. Titanium dioxide (TiO2; Sigma-Aldrich, Inc., St. Louis, MO) was used as an external marker to estimate total tract OM digestibility of the diet (Titgemeyer et al., 2001; Myers et al., 2004). Titanium dioxide was weighed out (5 g) into number 11 Lock Ring Capsules (Torpac Inc., Fairfield, NJ) and capsules were administered via metal balling gun (Torpac Inc.) twice daily at 0600 and 1800 h for 10 d starting on d 34. Starting at 0600 h on d 39, fecal grab samples were collected every 12 h for a 5-d sampling period. Sample collection time was advanced by 2 h each day to represent every 2 h in a 24-h period. Approximately 200 g of wet material was collected at each sampling time. Fecal samples were composited on an equal weight basis for each heifer, dried in a forcedair oven (50°C), ground in a Wiley Mill (Thomas Scientific, Swedesboro, NJ) equipped with a 1-mm screen, and analyzed for OM (AOAC International, 2000), titanium (Myers et al. [2004] adapted to a microtiter plate [BioTek Instruments, Winooski, VT; absorbance measured at 410 nm]), and NDF using an ANKOM 200 fiber analyzer (ANKOM Technology Corp., Fairport, NY) with sodium sulfite and amylase. Apparent total tract digestibility of OM was calculated by dividing the amount of Ti dosed by the Ti concentration in the fecal sample (Titgemeyer et al., 2001). Digestibility of NDF (DM basis) was calculated using the following equation: % NDF digestibility = 1 00 – [100 × (% NDF in feed/% NDF in feces × % NDF in feces/ % NDF in feed)].
Glucogenic precursors improve fermentation
Rumen fluid was collected every third day starting at d 0 (initiation of treatments) via oral lavage with a metal suction strainer 4 h after the morning feeding into a stoppered side-arm flask (Lodge-Ivey et al., 2009). Ruminal pH was measured using a portable pH probe (Accumet AP72; Fisher Scientific, Waltham, MA) within 5 min of collection. Ruminal fluid samples were divided into 2- to 10-mL aliquots for subsequent analysis of ammonia-N and VFA, placed on ice, frozen, and stored at –20°C in 15-mL polypropylene conical tubes (Fisher Scientific, Pittsburgh, PA). Aliquots for analysis of ruminal ammonia-N were acidified with 2.5 mL of 5% HCl before freezing. Samples were kept on ice and transported to the laboratory for storage at –20°C until analysis. Ruminal fluid samples were thawed, centrifuged, and analyzed for VFA using HPLC according to the method of Guerrant et al. (1982). Molar concentrations of VFA were obtained by multiplying peak heights by response factors after the detector response value for each acid had been determined. Ruminal ammonia-N was determined using the phenol-hypochlorite procedure (Broderick and Kang, 1980) using a spectrophotometer at a wavelength of 630 nm. During the acetate tolerance test (ATT), animals had access to the basal diet and were fed their respective treatments at 0600 h. Heifers were fitted with an indwelling jugular catheter 30 min before ATT on d 49 to evaluate acetate clearance from blood using methods and procedures described by Mulliniks et al. (2011). A 20% acetic acid solution (pH 7.40) was filter-sterilized using a Steritop Filter Unit (Millipore, Billerica, MA) and infused into each animal via an indwelling jugular catheter at a dosage of 1.25 mL/kg of BW over a 20-min period. Blood samples were collected at –1, 0, 1, 3, 5, 7, 10, 15, 30, 60, and 90 min following the administration of the acetic acid. Infusion of acetate occurred between –1 and 0 min. Blood samples (10 mL) were collected into serum separator tubes (Corvac, Mansfield, MA) and allowed to clot at room temperature for 45 min. Blood samples were centrifuged at 2,000 × g for 25 min at 4°C. Serum was harvested (2 mL) and then filtered through a centrifugal filter (Millipore Amicon Ultra, Ultracel-10K; Billeria, MA) at 7,000 × g for 80 min at 4°C. Filtered serum was frozen and later analyzed for serum acetate concentrations using HPLC (Guerrant et al., 1982). Molar concentrations of acetate were obtained by multiplying peak heights by response factors after the detector response value for acetate been determined (Guerrant et al., 1982). Acetate half-life was calculated as the time required for a 50% decrease in peak serum acetate concentration (Kaneko, 1997). Serum acetate areas under the curve were calculated using the trapezoidal summation method. Statistical Analysis. Ruminal fluid data were analyzed as a completely randomized design using the Downloaded from https://academic.oup.com/jas/article-abstract/92/4/1738/4703455 by guest on 14 January 2018
1741
MIXED procedure of SAS with repeated measures using Akaike’s information criterion to select the most appropriate covariance structure (compound symmetry). The model included treatment, day, and treatment × day interactions. Effects were considered significant when P ≤ 0.05 and a tendency when P > 0.05 and P ≤ 0.10. When a significant F-test was observed, least squares means were separated using the PDIFF option. Individual heifer was considered the experimental unit. Acetate half-lives were estimated for each heifer by regressing the logarithmically transformed acetate concentrations over time (Kaneko, 1997). Area under the curve was determined for acetate concentrations using trapezoidal summation method. The MIXED procedure of SAS was used to test all main effects of the ATT. The model included treatment, heifer, and their interaction. Results and Discussion Experiment 1 Addition of PA increased ruminal IVDMD (P < 0.001) and total VFA (P < 0.001) of 100% dormant warm-season grass extrusa (Table 1) but did not alter (P ≥ 0.28) IVDMD and total VFA of 50% sorghum–Sudan hay with 50% corn (Table 2). Addition of PA decreased (P < 0.001) acetate and increased (P < 0.001) propionate concentration resulting in lower (P < 0.001) acetate to propionate ratio (A:P) for both substrates (Tables 1 and 2). Little research has been published that reports the effect of supplemental Propionibacterium on in vitro rumen fermentation and digestibility of forage. In agreement with our results, Lehloenya et al. (2008) and Weiss et al. (2008) reported decreased acetate, increased propionate, and decreased A:P in beef and dairy cattle fed total mixed rations (TMR) and supplemented with PA. Akay and Dado (2001) also reported that Propionibacterium P5 increased total VFA and propionate and decreased A:P when added to alfalfa silage, fescue hay, wheat straw, and corn grain but decreased IVDMD and increased acetate in vitro. Ghorbani et al. (2002) reported similar 24-h in situ dry matter disappearance of alfalfa hay, barley silage, concentrate, and wheat straw incubated in the rumen of cattle consuming feedlot diets supplemented with Propionibacterium P15 when compared to cattle consuming a control diet. Differences in Propionibacterium strain, substrates, method, and incubation time likely account for variation in results between studies. Our results suggest that PA has the capability of enhancing VFA production and improving digestion of low-quality forages by grazing ruminants.
1742
Sanchez et al.
Table 1. Effects of Propionibacterium acidipropionici P169 (PA) on in vitro fermentation of 100% dormant warm-season grass extrusa1 PA2 Item IVDMD, % Total VFA, mM Acetate, mol/100 mol Propionate, mol/100 mol Acetate:propionate
–
+
SEM
P-value
20.5 92.8 70.3 20.4 3.6
32.0 119.1 68.5 22.0 3.1
0.85 2.91 0.28 0.27 0.09
