Effect of storage of wet brewer’s grains with incremental levels of salt on apparent total tract nutrient digestibility and purine derivative excretion in dairy heifers

Abstract

Objectives of this study were to evaluate apparent total tract nutrient digestibility and purine derivative (PD) excretion in dairy heifers limit-fed diets containing wet brewer’s grains (WBG) treated with salt. A 12-wk replicated 4 × 4 Latin square was conducted using 8 Holstein heifers of 224.5 ± 19.4 d of age, and body weight (BW) of 219.2 ± 28.1 kg (mean ± SD). Fresh WBG were treated with 0%, 0.8%, 1.6%, and 2.4% salt and stored for 4 d before being fed. Salt was added either to the WBG or separately to equalize the amount of salt in the diet. The diet contained 9% grass silage, 47% corn silage, 19% corn meal, 17.6% WBG and salt, 2% soybean meal, and 3% mineral mix. Diets were formulated to be limit-fed at 2.15% of BW, provide 14% crude protein (CP) and 2.27 Mcal metabolizable energy (ME)/kg of dry matter (DM). Heifers were adapted to diets for 14 d followed by a 7-d collection period. Dry matter intake (DMI) was recorded daily during the collection week while BW was recorded once a week. Urine and fecal samples were collected during the last 4 d of the collection period. Acid insoluble ash was used as an internal marker to determine apparent nutrient digestibility. Weight loss of WBG during storage was determined from days 1 to 11 and initial and final yeast and mold counts were determined. Final yeast counts were similar among treatments while final mold counts tended to be lesser (P = 0.07) for the 0.8% and 1.6% salt treatments. Urinary volume was similar among treatments while allantoin (P = 0.14), and uric acid (P < 0.01) and total PD excretion tended to increase (P = 0.13) quadratically. DMI was varied by treatment (linear, quadratic, and cubic effects P < 0.01). Heifers fed the 0.8% treatment had the least DMI. Nonfiber carbohydrate (NFC) digestibility linearly decreased (P < 0.04) as salt increased. Digestibilities of DM, and organic matter (OM), tended to decrease (P < 0.10) with increasing levels of salt added to WBG. Fat digestibility was quadratic with the greatest value for the 1.6% treatment. Treating WBG with salt reduced its deterioration based on lesser mold counts for the 0.8% and 1.6% treatments. These treatments had resulted in greater fat digestibility and tended to have increased PD excretion suggesting improved microbial protein synthesis.

Introduction

The high moisture content of wet brewer’s grains (WBG) is associated with high costs of transportation and a short storage shelf-life. Different preservation strategies have been examined to increase the length of storage of WBG, such as drying (Pereira et al., 1998), ensiling with dried feeds (Kung, 2005), and the addition of bacterial inoculants (Marston et al., 2009). However, most of the conservation techniques of WBG have not been evaluated for their effect on nutrient digestibility and utilization when fed to animals. In a recent study, Hatungimana and Erickson (2019) reported that storing WBG with salt improved in vitro and in situ DM digestibility as well as intestinal protein digestibility. In another experiment, it was reported that including 20% WBG in the diet of yearling dairy heifers resulted in similar growth performance and nutrient digestibility compared with heifers fed diets containing soybean and corn meal concentrates (Hatungimana et al., 2020).

Based on these results, further research is needed to improve our knowledge of the effect of treating WBG with salt on in vivo nutrient digestibility and urinary purine derivative (PD) secretion. Urinary PD are indicative of microbial protein synthesis (Chen and Gomes, 1992). Besides the known nutritional value and physiological role of common salt (sodium chloride) for cattle (Berger, 2006), salt has been proven to be associated with increased liquid dilution rate, the efficiency of microbial protein synthesis, fiber digestibility, milk fat synthesis, and organic matter (OM) utilization in cattle fed high concentrate diets (Harrison et al., 1975; Rogers et al., 1982; Schneider et al., 1988). We hypothesize that treating and storing WBG with salt would enhance dietary nutrient digestibility and improve microbial crude protein (CP) synthesis.

The current study aims at evaluating the apparent total tract nutrient digestibility and urinary PDs excretion as an indicator of microbial CP synthesis in dairy heifers limit-fed diets containing WBG treated with different levels of salt.

Material and Methods

Experimental design

This experiment was reviewed and approved by the University of New Hampshire Animal Care and Use Committee (protocol # 180506). Eight Holstein heifers with a mean age of 224.5 ± 19.4 d and body weight (BW) of 219.2 ± 28.1 kg (mean ± SD) were used in a replicated 4 × 4 Latin square and randomly assigned to 4 treatments and periods. Each period lasted for 21 d, of which there were 14 d of adaptation and 7 d of sample collection. Fresh WBG were collected weekly from a local dairy farm and transported (18.3 km) to the Fairchild Dairy Research Center at the University of New Hampshire. One sample of fresh WBG was collected from the weekly batch and sent to ANALAB (Fulton, IL) for yeast and mold count analysis (AOAC International, 1999, method 997.02, as described by Hatungimana and Erickson, 2019). Another fresh sample of WBG was dried in an air-forced oven for 48 hr and sent to the Rock River Laboratory (Watertown, WI) for nutrient analysis. The study was conducted from July 25 to October 30, 2019. The ambient temperature ranged from 1 to 32 °C.

After arrival at the research center, the WBG was treated with 4 amounts of salt (0%, 0.8%, 1.6%, and 2.4%). The salt amounts and the feeding rate of WBG were based on previous studies which reported that treatment of WBG with 2.6% and 3.8% salt decreased mold growth and improved in vitro and in situ DM digestibility (Hatungimana and Erickson, 2019). Moreover, it was reported that heifers fed 20% WBG had similar growth performance with heifers supplemented with soybean meal (SBM) and corn-based supplements (Hatungimana et al., 2020). Treatments were formulated considering the maximum tolerable salt in diets of dairy cattle which is 40 g NaCl/kg dry matter (DM) or 1.40 g NaCl/kg BW (NRC, 2001). Our diets provided ~29 g NaCl/kg DM.

WBG and the salt were mixed using a motorized feeding vehicle (Data Ranger, American Calan Inc. Northwood, NH) for 5 min and stored for 4 d in plastic tubs (cylinder shape: 42 cm of height and 42 cm of diameter) inside a commodity shed. Approximately 72.6 kg of as-fed WBG was placed in 2 tubs (36.3 kg/tub) per treatment. Initially, WBG was intended to be stored for 1 wk which is a typical feedout length of WBG in the Northeast region of the United States. However, the storage time was reduced to 4 d due to the rapid spoilage of WBG observed during high environmental temperatures. The total number of days from receiving to final feedout was 11 d (4 d for initial storage, 7 d of feeding). Plastic tubs containing treatments were weighed on a platform scale (Cardinal, Northeast Scale Co. Inc., Hooksett, NH) before storage and at the end of the collection period to measure the weight loss of WBG (fresh basis) due to mold and yeast growth or water evaporation. Samples were also taken at the end of the period (day 11) for analysis of molds and yeasts. Because this was a preservation experiment under aerobic conditions to mimic feeding WBG on the farm, the pH and fermentation end products were not evaluated, and plastic tubs were not sealed.

After 4 d of storage, WBG and supplemental salt was included in the totally mixed ration (TMR) at the rate of 20% of the dietary DM at the time of feeding. Amounts of WBG removed from each feeding were also recorded to keep track of how much WBG was fed relative to the amount stored. The weight loss of WBG was calculated at the end of each feeding period considering the initial and the final weight of WBG stored, the amount fed, and the amount left after the feeding period was over. Individual nutrient analysis of TMR components is presented in Table 1 while the diet composition is presented in Table 2. During feeding, additional salt was added to diets at the time of mixing to achieve a similar salt concentration across all diets (2.4%, Table 2).

Table 1.

Average (mean ± SD) nutrient composition of major ingredients used in the experiment

Item, % DM	Grass silage	Corn silage	WBG	Corn meal	Soybean meal
DM	33.5 ± 2.90	34.3 ± 0.7	22.6 ± 0.90	86.90	87.60
CP	12.4 ± 2.40	7.0 ± 0.3	32.1 ± 2.16	8.25	53.30
aNDF1	59.1 ± 5.40	42.4 ± 2.60	43.5 ± 8.40	9.80	9.0
ADF	39.7 ± 3.70	25.2 ± 1.80	22.2 ± 3.60	4.25	6.0
Lignin	6.67 ± 0.87	3.3 ± 0.30	11.2 ± 4.40	1.33	0.7
Starch	0.77 ± 0.48	32.8 ± 2.60	3.4 ± 2.70	74.01	8.05
Fat2	3.61 ± 0.75	2.98 ± 0.48	8.83 ± 1.07	3.95	1.90
Ash	7.74 ± 1.2	3.49 ± 0.20	4.79 ± 0.24	1.52	6.36
Ca	0.70 ± 0.12	0.18 ± 0.03	0.17 ± 0.01	0.02	0.50
P	0.28 ± 0.03	0.25 ± 0.02	0.59 ± 0.04	0.27	0.65
K	2.35 ± 0.42	1.08 ± 0.05	0.07 ± 0.04	0.39	2.48
Mg	0.22 ± 0.04	0.16 ± 0.005	0.18 ± 0.01	0.10	0.34
S	0.21 ± 0.03	0.10 ± 0.01	0.35 ± 0.02	0.11	0.41

¹α-Amylase NDF.

²Ether extract.

Table 2.

Ingredient composition (% DM) of experimental diets containing 0%, 0.8%, 1.6%, and 2.4% salt in WBG limit-fed to dairy heifers

	Treatment1 (% salt in WBG)
Ingredient, % DM	0	0.8	1.6	2.4
Grass silage	9	9	9	9
Corn silage	47	47	47	47
Corn meal	19	19	19	19
SBM	2	2	2	2
WBG + salt	17.6 + 0	17.6 + 0.8	17.6 + 1.6	17.6 + 2.4
Salt2	2.4	1.6	0.8	0.0
Mineral and vitamin mix3	3	3	3	3

¹Salt treatments: 0% salt, WBG stored 4 d without addition of salt. 0.8% salt, WBG were mixed and stored 4 d with 0.8% salt. 1.8% salt, WBG were mixed and stored 4 d with 1.6% salt. 2.4%, WBG were mixed and stored 4 d with 2.4% salt.

²Salt treatment were added in reverse order to diets during feeding to equalize amounts of salt consumed by heifers.

³Contained 19.05% Ca; 6.01% P; 3.51% Mg; 20.00% salt; 7.80% Na; 0. 26% Fe; 0.26% Zn; 0.26% Mn; 12.30% Cl; 602 mg/kg Cu; 15.0 mg/kg Co; 25.09 mg/kg Se; and 15.00 mg/kg I; 267,800 IU/kg vitamin A; 111,071 IU/kg vitamin D; and 2,207 IU/kg vitamin E.

Animal care and feeding

Heifers were group-housed in a naturally ventilated free-stall barn bedded with mattresses in a single pen (8 × 4.8 m). Heifers were fed once daily at 0900 hours using the Calan gate feeding system (American Calan Inc., Northwood, NH) in individual wooden boxes to measure feed intake. The feed was mixed and distributed using a motorized feeding vehicle (Data Ranger, American Calan Inc., Northwood, NH). Rations were limit-fed at 2.15% of BW (DM basis) to provide 14% CP and 2.27 Mcal metabolizable energy (ME)/kg of DM using NRC (2001). Rations were adjusted every week based on BW measurements and DM analysis of feed ingredients. Heifers were adapted to diets for 14 d followed by 7 d of collection. Heifers had ad libitum access to water through automatic refilling water troughs. Heifers were watched every day for any health problem according to the routine management of the research center. BWs of heifers were measured weekly before feeding using a platform scale (Cardinal, Northeast Scale Co. Inc., Hooksett, NH).

Feed and fecal sampling and analysis

Samples of TMR for each treatment were collected from days 14 to 17 of the collection period and stored at –20 °C for future nutrient analysis. Individual feed ingredients were sampled for nutrient analysis every 2 wk or any time new silos of grass and corn silages were opened. Fecal samples were collected from days 17 to 21, twice a day at a 12-hr interval offset by 3 hr on subsequent days to provide representative samples over a 24-hr period. Rectal fecal grabs samples (~200 g/sample) were collected via gloved hand by manually stimulating defecation or collecting directly from the rectum. Fresh fecal samples over the 4-d of the collection period were combined to obtain a single composite and were frozen at –20 °C.

Feed and fecal samples were then thawed at room temperature and emptied into aluminum trays to be dried in a forced-air convection oven (Binder, Bohemia, NY) at 55 °C for 48 hr (for feed samples) and at least 96 hr for fecal samples until completely dried. The dried TMR and fecal samples were ground through a 1-mm screen using a Wiley mill (Model 3, Arthur H. Thomas, Philadelphia, PA). Ground samples were sent to Rock River Laboratory (Watertown, WI) for nutrient analysis.

Feed and fecal samples were analyzed for neutral detergent fiber (NDF; method 6 in an Ankom Fiber Analyzer A2000 with α-amylase and sodium sulfite, Ankom Technology; Van Soest et al., 1991), and acid detergent fiber (ADF; method 5 in an Ankom Fiber Analyzer A2000, Ankom Technology, Fairpoint, NY; method 973.18, AOAC International, 1998). The determination of those fiber concentrations (NDF and ADF) was not performed sequentially. Nitrogen was analyzed via Dumas combustion (AOAC International 2002; method 968.06) on a Rapid N cube (Elementar Analysensystem, GmbH, Hanau Germany). Nitrogen was then multiplied by 6.25 to calculate CP concentration assuming that proteins contain 16% N.

Starch concentration was analyzed using a modified method of glucose analysis (Bach Knudsen, 1997) completed on a YSI 2700 select Biochemistry Analyzer (YSI Biochemistry analyzer, YSI Inc., Yellow Springs, OH). Fat concentration was analyzed with the ether extraction technique (method 2003.05; AOAC International, 2006). Neutral detergent insoluble CP (NDICP) was determined by analyzing CP concentration on the insoluble residue of an NDF extraction. Acid detergent insoluble CP was determined similarly as the CP concentration associated with the insoluble residue of an ADF extraction.

Ash concentration was determined by incinerating 1 g of sample for 8 hr at 450 °C in a muffle furnace (AOAC International, 2002; method 942.05). Nonfiber carbohydrate (NFC) concentration was calculated as: NFC, % = 100 – (CP% + (NDF% – NDICP%) + fat% + ash%). Acid insoluble ash concentration was determined according to Van Keulen and Young (1977). Mineral composition analysis included Ca, P, Mg, K, Na (AOAC International, 1998; method 985.01), and S (AOAC International, 1998; method 923.01). Energy concentration of dietary treatments was determined using NRC (2001) during diet formulation.

Yeast and mold counts in WBG were analyzed (AOAC 1999, method 997.02), using culture plates of dry medium supplemented with antibiotics and dye to enhance visualization of growth and cold H₂O-soluble gelling agent. Undiluted or diluted suspensions were added to plates at a rate of 1 mL/plate. The suspension was spread over a 30-cm² growth area. The gelling agent was allowed to solidify, plates were incubated, and yeasts and molds were counted and reported as colony forming units (CFU).

Urine sampling and analysis and digestibility evaluation

Urine samples were collected at the same time as fecal samples, via manual stimulation of the pudendal nerve. Urinary samples were immediately transported to the laboratory where a 1-mL subsample of urine was pooled over 4 d into centrifuge tubes containing 32 mL of 0.072 N H₂SO4 for subsequent analysis of creatinine, allantoin, and uric acid.

Subsamples of urine were thawed at room temperature and analyzed for: concentrations of creatinine (assay kit # 5007001, Cayman Chemical Co., Ann Arbor, MI) using a microplate reader (Epoch Bio Tek Instruments, Inc., Winooski, VT) set at a wavelength of 492 nm, and concentrations of allantoin (Chen and Gomes, 1992); uric acid (Assay kit # 700320, Cayman Chemical Company, Ann Arbor, MI) using a fluorometer (SpectraMax M2e, Sunnyvale, CA), set at an excitation wavelength of 530 nm and an emission wavelength of 585. Urinary volume was estimated from the urinary concentration of creatinine and creatinine excretion (CE) using a regression equation for growing heifers. Creatinine excretion rate was calculated using the following equation (CE, mg)/kg BW = 0.28 ± 0.01– 0.000097 ± 0.000015 × BW (Chizzotti et al.,2008) and was converted to millimole per day. Urinary volume was calculated using the equation: (CE, mmol/d)/(urinary creatinine, mmol/L). Urinary excretion of total PD was calculated by adding allantoin and uric acid excretion.

Acid detergent insoluble ash was used as an internal digesta marker to estimate 24 hr fecal excretion and apparent total tract nutrient digestibility was determined using the following equation:

$${\displaystyle \begin{array}{l}{\displaystyle \mathrm{a}\mathrm{p}\mathrm{p}\mathrm{a}\mathrm{r}\mathrm{e}\mathrm{n}\mathrm{t}\mathrm{n}\mathrm{u}\mathrm{t}\mathrm{r}\mathrm{i}\mathrm{e}\mathrm{n}\mathrm{t}\mathrm{d}\mathrm{i}\mathrm{g}\mathrm{e}\mathrm{s}\mathrm{t}\mathrm{i}\mathrm{b}\mathrm{i}\mathrm{l}\mathrm{i}\mathrm{t}\mathrm{y}{\displaystyle \left(\mathrm{\%}\right)=100-100}}\\ \times \frac{\mathrm{\%}\mathrm{A}\mathrm{I}\mathrm{A}\mathrm{i}\mathrm{n}\mathrm{f}\mathrm{e}\mathrm{e}\mathrm{d}\times \mathrm{\%}\mathrm{n}\mathrm{u}\mathrm{t}\mathrm{r}\mathrm{i}\mathrm{e}\mathrm{n}\mathrm{t}\mathrm{i}\mathrm{n}\mathrm{f}\mathrm{e}\mathrm{c}\mathrm{e}\mathrm{s}}{\mathrm{\%}\mathrm{A}\mathrm{I}\mathrm{A}\mathrm{i}\mathrm{n}\mathrm{f}\mathrm{e}\mathrm{c}\mathrm{e}\mathrm{s}\times \mathrm{\%}\mathrm{n}\mathrm{u}\mathrm{t}\mathrm{r}\mathrm{i}\mathrm{e}\mathrm{n}\mathrm{t}\mathrm{i}\mathrm{n}\mathrm{f}\mathrm{e}\mathrm{e}\mathrm{d}}\end{array}}$$

Statistical analysis

Data were analyzed as a replicated 4 × 4 Latin square. Initial BW served as a covariate for BW, average daily gain (ADG), and gain:feed ratio using the Mixed procedure of SAS 9.4 (SAS Institute Inc., Cary, NC). Actual mold and yeast counts (CFU) were analyzed statistically and then logarithmically transformed. The model included square, period, and treatment according to the following model:

$${\displaystyle \begin{array}{l}{\displaystyle Y_{ijkl}=\mathrm{\mu }+S_i+P_j+X_{ijk}+T_l+S{T}_{il}+E_{ijkl}}\end{array}}$$

where Y_ijkl = the dependent variable, µ = the overall mean, S_i = random effect of square (i = 1, 2), P_j = the random effect of period (j = 1, ... , 4), T_k = the fixed effect of the kth treatment (k = 0; 0.8%, 1.6%, and 2.4% salt in WBG), X_ijk = the covariate measurement; ST_il = fixed effect of interaction between the ith square and lth treatment, and the E_ijkl = the residual error. Degrees of freedom were calculated using the Kenward–Roger approximation option of the Mixed procedure. Single degree of freedom contrasts for linear, quadratic, and cubic effects were estimated. If the probability of the covariate parameter estimate was >0.25, the covariate was removed from the model. Square × treatment interaction was analyzed and removed in the model when not significant (P > 0.05). For all variables, the least-square means for each treatment were reported. For all measured parameters, data outliers that were more than 2.5 SD units from the mean were removed before statistical analysis. Significant treatment effects were declared at P ≤ 0.05 and tendencies at 0.05 < P ≤ 0.15.

Results

Feed nutrients and diet composition

The average nutrient composition of major feed ingredients used in this study is presented in Table 1. During the study, the DM and CP concentration of WBG averaged 22.6 ± 0.90% and 32.1± 2.16%, respectively. Table 2 illustrates the dietary ingredient composition (% DM). Dietary ingredients were similar in all treatments except the level of salt mixed with fresh WBG during storage. Table 3 contains the nutrient composition of diets based on laboratory analysis. The nutrient contents of the diets were similar across treatments.

Table 3.

Average (mean ± SD) nutrient composition of experimental diets containing WBG treated with different amounts of salt

	Treatments1, % salt in WBG
Item %, DM	0%	0.8%	1.6%	2.4%
DM	35.2 ± 1.2	34.3 ± 1.46	35.5 ± 1.19	35.2 ± 1.33
CP	13.7 ± 0.61	13.6 ± 0.82	13.2 ± 0.06	13.4 ± 0.90
aNDF2	34.6 ± 0.74	35.1 ± 1.48	32.8 ± 3.12	33.3 ± 1.06
ADF	18.4 ± 0.80	19.0 ± 1.53	18.6 ± 0.90	17.4 ± 1.06
Fat3	3.0 ± 0.54	3.1 ± 0.48	3.3 ± 0.85	3.6 ± 0.83
Starch	29.4 ± 1.54	28.9 ± 2.06	31.2 ± 1.14	30.1 ± 1.40
NFC4	42.6 ± 1.68	42.0 ± 2.24	44.5 ± 3.20	43.2 ± 1.40
Ash	7.5 ± 0.35	7.6 ± 0.47	7.51 ± 0.20	7.8 ± 0.47
ME5, Mcal/kg	2.57 ± 0.20	2.57 ± 0.30	2.57 ± 0.20	2.57 ± 0.30

¹Salt treatments: 0% salt, WBG stored 4 d without addition of salt. 0.8% salt, WBG were mixed and stored 4 d with 0.8% salt. 1.8% salt, WBG were mixed and stored 4 d with 1.6% salt. 2.4%, WBG mixed and stored 4 d with 2.4% salt.

²α-Amylase NDF.

³Ether extract.

⁴NFC, % = 100 – (CP% + (NDF% – neutral detergent insoluble CP%) + fat% + ash%).

⁵ME was calculated using NRC (2001).

The yeast and mold counts (reported as logarithm₁₀ CFU/g), as well as losses of the weight of stored WBG, are presented in Table 4. Initial pretreatment yeast CFU were 4.90, 4.89, 4.22, and 4.22 log CFU/g for periods 1 through 4, respectively. Initial pretreatment molds were 2.68, 1.00, 1.78, and 1.78 log CFU/g for periods 1 through 4, respectively. Mold counts tended to be different by period (P = 0.09) and decreased from periods 1 to 4. Final day yeast counts were not different (P = 0.76) among treatments. There was no effect of period on yeast counts. However, final day mold counts tended to be quadratically (P = 0.07) lower for the 0.8% and 1.6% treatments. WBG weight losses (% initial weight of WBG tubs) exhibited a linear effect (P = 0.03). The greatest difference was between the 0.8% salt treatment and the 1.6% and 2.4% salt treatments (P = 0.03). As salt treatments increased, the weight loss of WBG decreased. There was a trend for weight loss by period (P = 0.051). The greatest weight loss occurred in period 2 (10.13%) with period 1 (3.88 %) having the least weight loss. Period 3 (7.72%) and 4 (8.78%) were intermediate.

Table 4.

Yeast and mold growth in WBG treated with different amount of salt

	Treatments1, % salt in WBG					Contrast P-values
Item	0	0.8	1.6	2.4	SEM2	L3	Q4	C5
Yeast, log CFU6	8.05	8.00	8.37	8.03	8.88 × 107	0.76	0.53	0.45
Mold, log CFU7	7.33	6.70	6.74	6.83	6.83 × 106	0.62	0.07	0.68
WBG weight loss8, %	8.62	11.02	5.72	5.12	1.37	0.03	0.30	0.52

¹Salt treatments: 0% salt, WBG was mixed and stored 4 d without the addition of salt. 0.8% salt, WBG was mixed and stored 4 d with 0.8% salt. 1.8% salt, WBG was mixed and stored 4 d with 1.6% salt. 2.4%, WBG was mixed and stored 4 d with 2.4% salt. Yeast and mold counts are logarithmically transformed and were statistically analyzed before they were logarithmically transformed.

²Standard error of the mean (not logarithmically transformed).

³Linear effect.

⁴Quadratic effect.

⁵Cubic effect.

⁶Yeast count, data were not transformed before statistical analysis.

⁷Mold count data were not transformed before statistical analysis.

⁸Tubs containing treatments were weighed at the beginning of the storage and the end of every collection period. The amount of WBG removed to be fed was daily recorded. The weight losses of WBG were calculated as follows: [(Weight of WBG stored - (weight of WBG fed + weight of WBG left after the collection period)) / weight of WBG stored] × 100.

Table 5 illustrates the least-square means for urinary volume and PD excretion in heifers. Detectable linear, quadratic, or cubic contrasts were not obtained (P ≥ 0.20) across treatments for volumes of urine. Urine volume was estimated from creatinine output which responded similarly to treatments (P = 0.83; P = 0.90, and P = 0.95 for linear, quadratic, and cubic responses). Urinary uric acid concentration had a quadratic effect (P < 0.01) and was greater for heifers fed the 0.8% and 1.6% salt treatments compared with heifers fed the control and 2.4% salt treatment. Allantoin and total PD excretion tended to be greater for 0.8% and 1.6% salt treatments resulting in a tendency for a quadratic response in each case (P ≤ 0.14).

Table 5.

PD excretion of heifers limit-fed with diets containing WBG treated with different amounts of salt

	Treatment1, % salt in WBG					Contrast P-values
	0	0.8	1.6	2.4	SEM	L2	Q3	C4
Urinary volume, l/d	13.1	14.8	13.4	13.1	1.12	0.69	0.20	0.36
Allantoin, mmol/d	117.5	136.4	134.0	119.5	11.7	0.95	0.14	0.85
Creatinine, mmol/d	66.6	67.2	67.3	67.4	1.40	0.83	0.90	0.95
Uric acid, mmol/d	4.56	5.99	5.52	4.64	0.68	0.88	<0.01	0.43
Total PD, mmol/d	121.6	141.9	139.5	123.8	11.1	0.94	0.13	0.84

¹Salt treatments: 0% salt, WBG were mixed and stored 4 d addition of salt treatment. 0.8% salt, WBG was mixed and stored 4 d with 0.8% salt. 1.8% salt, WBG was mixed and stored 4 d with 1.6% salt. 2.4%, WBG was mixed and stored 4 d with 2.4% salt.

²Linear effect.

³Quadratic effect.

^.4Cubic effect.

The DMI and total tract nutrient digestibilities in heifers limit-fed with diets containing WBG treated with different amounts of salt are presented in Table 6. The DMI was lesser in heifers fed the diet containing WBG treated with 0.8% salt but this difference was numerically too small to be interpreted (linear P < 0.01; quadratic P < 0.01; cubic P < 0.01). The apparent digestibilities of NFC linearly decreased (P < 0.04) and DM, and OM tended to linearly decrease (P ≤ 0.10) as salt added to the WBG increased. Fat digestibility reacted quadratically (P = 0.03) with the greatest fat digestibility at the 1.6% treatment. NFC digestibilities improved from periods 1 through 4 (P = 0.01). There was a trend (P = 0.10) for DM and OM digestibilities to differ by period. As the ambient temperature got cooler, DM and OM digestibilities improved (periods 1 through 4). While not the intent of the study, growth performance in terms of BW (P < 0.04), ADG (P < 0.04), and gain/DMI (P = 0.06) (data not reported) improved as salt addition increased.

Table 6.

DMI and apparent total tract nutrient digestibility of heifers limit-fed diets containing WBG treated with different amounts of salt

	Treatment1, % salt in WBG					Contrast P-values
	0	0.8	1.6	2.4	SEM	L2	Q3	C4
DMI, kg/d Digestibility, % DM	5.10	4.98	5.13	5.10	0.01	<0.01	<0.01	<0.01
DM	64.4	61.3	60.2	53.6	4.50	0.10	0.68	0.27
OM	65.6	62.1	61.1	52.9	4.50	0.06	0.58	0.18
CP	56.1	51.5	50.0	51.2	4.84	0.41	0.49	0.68
aNDF5	44.5	51.6	34.4	39.9	6.62	0.24	0.89	0.31
ADF	38.2	45.4	31.0	37.4	6.36	0.49	0.94	0.24
Fat6	65.3	71.7	74.8	60.7	5.00	0.57	0.03	0.38
Starch	98.9	95.8	98.9	98.6	1.60	0.77	0.40	0.30
NFC7	83.9	83.8	81.6	79.0	2.07	0.04	0.45	0.42

¹Salt treatments: 0% salt, WBG stored 4 d without the addition of salt. 0.8% salt, WBG was mixed and stored 4 d with 0.8% salt. 1.8% salt, WBG was mixed and stored 4 d with 1.6% salt. 2.4%, WBG was mixed and stored 4 d with 2.4% salt.

²Linear effect.

³Quadratic effect.

⁴Cubic effect.

⁵α-Amylase NDF.

⁶Ether extract.

⁷NFC = 100 – (CP% + (NDF% – NDICP%) + fat% + ash%).

Discussion

The goal of this experiment was to evaluate the storage life, PD excretion, and apparent total tract nutrient digestibility in Holstein heifers limit-fed diets containing WBG treated with different amounts of salt. Much research has been conducted to evaluate the effects of different preservation techniques on improving the shelf-life of WBG; however, few of those techniques were evaluated for their effect on apparent nutrient digestibility and urinary PD excretion. As this experiment was conducted under aerobic conditions to mimic feeding WBG on the farm, the pH and fermentation end-products were not evaluated.

The nutrient concentration of different batches of WBG did not vary much because WBG was collected from the same source for the entire course of the experiment. The nutrient composition of WBG varies from brewery to brewery and depending on grain source (barley, wheat, corn, etc.), or processing technology (Robertson et al., 2010; Muthusamy, 2014). Therefore, a constant nutrient analysis is suggested for a better inclusion in diets.

The evaluation of yeast and mold count in WBG treated with salt showed no difference in yeast count and a trend in mold reduction. The lack of difference in yeast count among treatments was probably because yeasts are tolerant to high salt concentrations. This is because yeasts have on the surface of their membrane lipid bilayers a high density of negative charges that allow them to counter the high cation concentration of Na⁺ on the outside, and K⁺ on the inside (Masui et al. (1979; Russell, 1989). Mold counts tended to be less in the intermediate treatments (0.8% and 1.6% salt). Mold count reduction in WBG treated with salt agrees with the previous research by Hatungimana and Erickson (2019), who reported a decrease in mold growth when WBG was treated with 2.6% and 3.8% salt and stored for 28 d. According to Csonka (1989), salt inhibits the growth of mold by plasmolysis causing shrinkage of the cytoplasm. Salt also decreases water activity and disrupts microbial and enzymatic processes (Lawrence et al., 2004; Albarracin et al., 2011). The lack of a more pronounced difference in mold counts could be due to a shorter storage time of 11 d compared with 28 d in the study of Hatungimana and Erickson (2019). Moreover, it was observed that adding salt reduced WBG loss over the 11 d periods, and this could be due to lesser mold development, along with reduced evaporation losses. Nadi and Shokri (2012) reported that water evaporation rate decreases as NaCl concentration increases at a certain level (1.5 M) but any further increase in NaCl concentration may result in a higher evaporation rate. There was a trend for WBG weight loss to differ by period with the greatest weight loss being during period 2 (August). For reasons that are not apparent, the lowest weight loss occurred during period 1. Mold formation was greatest during the initial periods of the study (1 and 2) when the temperature was hot compared with the cooler periods (3 and 4) where mold counts were lower.

The urinary and creatinine output among heifers fed diets containing WBG treated with different levels of salt were similar. This response in urine and creatinine output occurred because urine output was a calculated value, based on the concentration of creatinine (Chizotti et al., 2008). Additionally, the fact that heifers consumed the same amount of salt in their diets explains the lack of difference in the urinary output. Normally, salt intake leads to increased water intake and urination rate to prevent the surplus of sodium in the body (NRC, 2007b). According to NRC (2001), the maximum level of NaCl in feed for dairy cattle is ~40 g NaCl/kg DM. Our diets provided around 29 g of salt/kg DM which was much less than the maximum level to be detrimental to heifers’ health.

For allantoin and uric acid concentration in urine, a quadratic trend response was observed for the intermediate salt treatment levels. These results may be related to less mold growth in the WBG in those treatments. Heifers consumed the same amount of salt; therefore, the difference in PD would be attributed to the reduced mold growth observed in these treatments or a greater fraction A (rumen soluble protein fraction) production in the intermediate salt treated WBG as observed by Hatungimana and Erickson (2019). More importantly, soluble rumen N is required for rumen microbial synthesis (Harun and Sali, 2019). Molds produce mycotoxins which can adversely affect rumen fermentation (Hussein and Brasel, 2001). Possibly, the intermediate concentrations of salt reduced the mycotoxin produced in the WBG as observed with the decreased mold CFU and the trend for increased PD for the intermediate treatments. The increased PD excretion suggests greater microbial protein synthesis in these treatments which is beneficial for heifer growth performances.

The decreased digestibility of the DM, OM, and NFC was probably due to the increased passage rate due to the increased ruminal dilution rate (Amaral et al., 1985). Also, the digestibilities of these nutrients were greater in the later periods compared with the early periods. This could be attributed to greater mold counts in the hotter periods (1 and 2) compared with the cooler periods (3 and 4). Molds would likely negatively affect apparent nutrient digestibility. In a study by Hatungimana and Erickson (2019), the in situ degradation rate of WBG treated with 2.6% salt was greater than that of WBG without treatment or WBG treated with a commercial preservative. The soluble fraction A was greater in WBG treated with salt in that experiment as well. In the current study, treating WBG with salt may have increased nutrient solubilization in WBG and decreased the residence time of digesta in the rumen resulting in tendencies for reduced NFC digestibility, and trends for reduced DM, and OM nutrient digestibility.

While not the purpose of the study, BW, ADG, and gain/DMI was enhanced as salt inclusion in the WBG was increased. Dry matter intake (DMI) was slightly reduced for heifers fed 0.8% salt treatment, however since the heifers were limit-fed, the reduction in intake was a function of BW. This could be due to more microbial protein synthesis due to less spoilage of the salt-treated WBG, but a larger study needs to be conducted to confirm this. According to Uddin et al. (2015), an increased rate of passage reduces the maintenance expenses of rumen microbes because they contribute less time inside the rumen. Even if the nutrient digestibility was less as salt inclusion in diets increased, the enhanced growth performance observed in heifers fed a greater amount of salt may be explained by the efficiency of microbial protein synthesis.

Conclusion

Treatment of WBG with salt may offer some advantages but also a few disadvantages. In the current study, treating and storing WBG with incremental levels of salt before being included in diets resulted in a decreased mold growth and a decreased loss in WBG, suggesting that salt treatment would be efficient for WBG preservation as it was observed in previous studies. In this experiment, feeding WBG treated with intermediate salt levels resulted in increased total PD, suggesting improved microbial protein synthesis which is beneficial for heifer’s growth. However, it was observed that feeding diets containing WBG treated with high levels of salt resulted in a decreased nutrient digestibility which may limit the growth performance of animals if maintained on the diets for a long period. While not the purpose of this study, the growth performance of heifers seemed to be improved as salt inclusion was increased, which suggest that these effects need to be evaluated in larger feeding studies.

Glossary

Abbreviations

ADF

acid detergent fiber

ADG

average daily gain

BW

body weight

CE

creatinine excretion

CFU

colony forming unit

CP

crude protein

DM

dry matter

DMI

dry matter intake

ME

metabolizable intake

NDF

neutral detergent fiber

NFC

nonfiber carbohydrate

OM

organic matter

PD

purine derivative

SBM

soybean meal

TMR

totally mixed ration

WBG

wet brewer’s grains

Conflict of interest statement

The authors declare no real or perceived conflicts of interest.