Rumen-protected choline reduces hepatic lipidosis by increasing hepatic triacylglycerol-rich lipoprotein secretion in dairy cows

Objectives were to determine the effects of supple-menting rumen-protected choline (RPC) on hepatic composition and secretion of triacylglycerol-rich lipoprotein when cows were subjected to feed restriction to develop fatty liver. It was hypothesized that RPC reduces hepatic triacylglycerol by enhancing secretion of hepatic lipoprotein. Pregnant, nonlactating parous Holstein cows (n = 33) at mean (± standard deviation) 234 ± 2.2 d of gestation were blocked by body condition (3.79 ± 0.49) and assigned to receive 0 g/d (CON), 25.8 g/d choline ion from a RPC product containing 28.8% choline chloride (CC; treatment L25.8), or 25.8 g/d of choline ion from a RPC product containing 60.0% CC (H25.8). Cows were fed for ad libitum intake for the first 5 d and restricted to 41% of the net energy for lactation required for maintenance and pregnancy from d 6 to 13. Intake of metabolizable methionine was maintained at 18 g/d during feed restriction by supplying rumen-protected methionine. Hepatic tissue was sampled on d 6 and 13 and analyzed for triacylglycerol and glycogen, and mRNA expression of hepatic tissue was investigated. On d 14, cows were not fed and received a 10% solu-tion of tyloxapol intravenously at 120 mg/kg of body weight to block hydrolysis of triacylglycerols in very low density lipoprotein (VLDL). Blood was sampled sequentially for 720 min and analyzed for concentration of triacylglycerol and total cholesterol. Lymph was sampled 6 h after tyloxapol infusion, and analyzed for concentrations of fatty acids, β-hydroxybutyrate, glucose, triacylglycerol, and total cholesterol. A sample of serum collected at 720 min after tyloxapol was assayed for the


INTRODUCTION
Hepatic lipidosis is prevalent in dairy cows, affecting 50 to 60% of the transition cows in the first 3 wk postpartum (Bobe et al., 2004).Cows with hepatic triacylglycerol accumulation beyond 4 to 7% on a tissue wet weight basis have associated impaired productive performance, health, and survival (Arshad and Santos, 2022).It is assumed that hepatic lipidosis impairs performance and health because excessive hepatic triacylglycerol induces inflammation in the hepatic tissue (Li et al., 2015) and impairs hepatocyte functions such as gluconeogenesis and ureagenesis (Strang et al., 1998;Mashek et al., 2002).Some of the underlying mechanisms of increased susceptibility to hepatic lipidosis in transition cows include excessive lipomobilization coupled with inability to extensively assemble and secrete triacylglycerol-rich lipoproteins by the liver (Emery et al., 1992;Bobe et al., 2004).Ruminants are known to have reduced secretion of very low density lipoproteins (VLDL), which impairs hepatic export of triacylglycerols (Emery et al., 1992), and it has been suggested that choline is a limiting factor for VLDL triacylglycerol export from the hepatic tissue (Grummer, 2013).Indeed, supplementing choline as rumen-protected choline (RPC) reduced hepatic triacylglycerol content in cows induced to develop hepatic lipidosis (Cooke et al., 2007;Zenobi et al., 2018a), and the response was dose dependent up to at least 25.8 g/d of choline ion (Zenobi et al., 2018a).Despite the benefits of choline in reducing hepatic triacylglycerol content, the mechanisms for such effects remain unclear in dairy cows.It is likely that the mechanisms resemble those observed in other species (Lombardi et al., 1968;Rinella et al., 2008) or in bovine hepatocytes in vitro (Chandler and White, 2017).
Choline is a trimethyl amine that is substrate for de novo synthesis of phosphatidylcholine and the sphingolipid sphingomyelin, which are components of cell membranes and play a role in lipid transport.Nevertheless, choline in feed ingredients or supplemented as choline salts to diets is extensively degraded by rumen microbes (Dawson et al., 1981;Sharma and Erdman, 1988), thus limiting the dietary supply for absorption in the intestine.An alternative to increase the supply of dietary choline is feeding it in a rumen-protected form, RPC (de Veth et al., 2016).Supplementation of RPC to transition cows improved productive performance and some aspects of health in early lactation (Arshad et al., 2020), presumably in part by reducing hepatic lipidosis (Cooke et al., 2007;Zenobi et al., 2018a).In pre-ruminants and monogastric species, supplementing choline to the diet reduces hepatic lipidosis by enhancing hepatic VLDL secretion rate (Lombardi et al., 1968;Rinella et al., 2008;Chandler and White, 2017).Nevertheless, in lactating dairy cows, only in vitro data exist that supplementing RPC might alter the fate of fatty acids taken up by the liver with reduced re-esterification and presumably increased export as lipoproteins (Piepenbrink and Overton, 2003).
We hypothesize that feeding choline as RPC reduces hepatic lipidosis by enhancing the export of triacylglycerol from the hepatic tissue.Therefore, the objectives were to study the effects of supplementing choline ion from 2 sources of RPC on hepatic composition and secretion of triacylglycerol-rich lipoprotein.Additional objectives were to evaluate hepatic mRNA expression of transcripts involved in multiple pathways linked to choline and lipid metabolism and the composition of serum and lymph in dairy cows that might help explain changes in hepatic composition.

MATERIALS AND METHODS
The experiment was conducted at University of Florida Dairy Unit (Hague, FL) from May to October of 2019, and all procedures with cows were approved by the Institutional Animal Care and Use committee of the University of Florida, protocol number 201910612.

Sample Size Calculation
A 2-tailed sample size calculation was performed using the POWER procedure of SAS (version 9.4; SAS/ STAT, SAS Institute Inc., Cary, NC).The assumption was that supplemental RPC would increase the mean serum triacylglycerol over time by 8 mg/dL, when the standard deviation of serum triacylglycerol is 6 mg/dL (Zenobi et al., 2018a).The sample size calculated (α = 0.05; β = 0.20) resulted in a minimum of 10 cows per treatment.Because of the uncertainty of concentrations of triacylglycerol after infusion of tyloxapol, 11 cows per treatment was the sample size selected.Given the experimental design with use of pre-planned orthogonal contrasts, the power to detect a difference of 8 mg/dL would be greater than the calculated 80% for the effect of RPC, or a smaller difference between treatments would be detected as significant.

Experimental Design, Dietary Treatments, Cows, and Housing
The experiment followed a randomized complete block design.Pregnant, nonlactating parous Holstein cows (n = 33) with a mean (±SD) 234 ± 2.2 d of gestation, BCS 3.79 ± 0.49, and BW 728 ± 72 kg were enrolled in the experiment.Every 14 d, a cohort of 6 cows were ranked by BCS, and each 3 cows were assigned a block.Within each block, cows were assigned randomly to receive 1 of 3 treatments that supplied either 0 or 25.8 g/d of choline ion as choline chloride (CC; ~74.6% choline ion).Treatments were 0 g/d choline ion (CON; n = 11), 25.8 g/d choline ion from a RPC product containing 28.8% CC (L25.8, n = 11; ReaShure, Balchem Corp., New Hampton, NY), or 25.8 g/d of choline ion from a RPC product containing 60.0% CC (H25.8,n = 11;prototype,Balchem Corp.).Treatments were supplemented as a top-dress, and CON cows received 200 g/d of a top-dress comprising coating material containing hydrogenated triacylglycerol, used to protect CC in the RPC products, and ground corn in a 43:57 ratio.The rationale for supplementing coating material in Cows were housed in a cross-ventilated tiestall barn with individual feed bins and water troughs equipped with flow meters.Stalls had air mattresses bedded with sand, and bedding material was cleaned twice daily and replaced thrice a week.The experiment lasted 14 d: 5 d of ad libitum feeding followed by 8 d of feed restriction, and 1 d of no feed offered.The schematic representation of the experimental design is shown in Figure 1.

Ad Libitum Feeding Period
During the ad libitum period, cows were fed a diet designed using the NRC (2001), and it was anticipated that they would consume 10 kg DM/d to achieve a positive NE L balance.Because of the updates made in nutrient requirements and supply with the development of the NASEM Dairy 8 (NASEM, 2021), the nutrient supply was recalculated using the observed DMI and the nutrient composition of the diet based on the chemical analyses of the ingredients.The recalculated values (NASEM, 2021) resulted in a diet with NE L content of 1.62 Mcal/kg that resulted in negative energy balance because of the increased energy required for maintenance and the less-than-anticipated DMI, which averaged 8.3 kg/d.The compositions of the diets and supplies of nutrients are shown in Supplemental Table S1 (https: / / figshare .com/s/ bed22cad5ca6e2460771).
Cows were fed once daily at 0900 h, and the amounts of feed offered to individual cows were adjusted daily to ensure at least 5% refusals, which were weighed before the morning feeding.Twice weekly, dietary ingredients were sampled and dried at 105°C for 24 h for adjustment of the amounts of DM offered.

Feed Restriction Period
During the feed restriction period, the diet was formulated and offered in amounts to supply approximately 50% of the NE L required for maintenance and pregnancy (NRC, 2001).The mean BW and gestation length of each block of 3 cows were used to calculate the metabolic BW with the associated maintenance requirement, and gestation length was used to generate the energy needs for pregnancy, using a calf with 42 kg BW at birth.Based on NRC (2001), the feed restriction imposed resulted in exactly 50% of the NE L required; however, the recalculated values using the NASEM Dairy 8 (NASEM, 2021) resulted in a supply of NE L of 41% of the energy required for maintenance and pregnancy.
Feed restriction was used to induce lipomobilization and increase accumulation of triacylglycerol into the hepatic tissue, as previously reported (Cooke et al., 2007;Zenobi et al., 2018a).The diet during feed restriction was formulated including rumen-protected methionine (Smartamine M, Adisseo USA, Alpharetta, GA) to result in a similar intake of metabolizable methionine anticipated for the ad libitum feeding period (Supplemental Table S1).The mineral and vitamin premix comprised a greater proportion of the diet during the feed restriction period compared with the ad libitum period, to supply similar amounts of micronutrients and ionophore (Supplemental Table S1).

Dietary Ingredient Sampling and Chemical Analyses
Feed ingredients were sampled twice weekly and dried at 105°C for 24 h in a forced-air oven, and dry weights were recorded to calculate the DM content of feeds and to adjust the amounts of feed offered to cows.Intake of DM by cows was based on the moisture content of feeds dried at 105°C.A second set of samples were dried at 55°C for 72 h, ground to pass a 4-mm screen of a Wiley mill (Thomas Scientific, Swedesboro, NJ), and stored for later chemical analyses.Dried samples were composited into 3 specimens per ingredient and analyzed using wet chemistry procedures (Dairyland Laboratories Inc., Arcadia, WI).The mean (±SD) compositions of the diets are reported in Supplemental Table S1.

Body Condition Score and Body Weight
Cows were weighed in the morning before feeding using a walk-through calibrated electronic scale on 2 consecutive days at the beginning (d 1 and 2) and end of the experiment (d 13 and 14).Values were averaged into single means for the beginning and end of the experiment, and the change in BW calculated.Concurrent with BW, body condition was scored by a single individual using a 1-to-5 scale with increments of 0.25 units (Ferguson et al., 1994), as depicted in the Elanco BCS chart (Elanco Animal Health, 2009), and the mean value was calculated for the beginning and end of the experiment.The change in BCS during the experiment was calculated.

Hepatic Tissue Collection
Hepatic tissue was collected on d 6 and 13 of the experiment.Briefly, the cow was restrained, and the areas covering the 10th to the 12th right intercostal spaces were scanned by ultrasonography (Aloka SSD-500V equipped with a 3.5-MHz convex transducer, Aloka Co. Ltd., Tokyo, Japan) to determine the location for sampling hepatic tissue.The area was shaved and thoroughly disinfected with 3 alternating rounds of povidone-iodine scrubbing followed by a 70% isopropyl alcohol rinse.Upon disinfecting the surgical site, 15 mL of a solution containing 2% lidocaine hydrochloride was administered in the subcutaneous space and intercostal muscles.The surgical site was disinfected again as described previously.After a 2-cm incision of the skin, a stainless-steel percutaneous hepatic biopsy tool (Aries Surgical, Davis, CA) was introduced, aiming for the hepatic tissue previously visualized by ultrasonography, and approximately 1 g of tissue was collected.The tissue was placed in filter paper, rinsed with sterile saline, sliced into 3 sections, transferred into 3 separate cryovials, snap-frozen in liquid N, and stored at −80°C until subsequent analyses.

Analysis of Hepatic Tissue
The content of triacylglycerol was analyzed on hepatic tissue sampled on d 6 and 13.Blocks of 3 cows containing all 3 treatments were analyzed in the same assay.Approximately 300 mg of hepatic tissue were processed in triplicate for determination of DM and for extraction of triacylglycerols using chloroform: methanol (Folch et al., 1957).Triacylglycerol was quantified using a colorimetric method (Foster and Dunn, 1973).Details of the procedures for quantification of DM and triacylglycerol in the hepatic tissue are depicted in Arshad and Santos (2022).Intra-and interassay coefficients of variation (CV) were, respectively, 4.2 and 14.0% for hepatic triacylglycerol on wet weight basis.
Hepatic glycogen also was analyzed on d 6 and 13.Approximately 100 mg of frozen hepatic tissue was split into 2 specimens and ground to a powder in liquid N 2 using a mortar and pestle, and then each 50 mg of tissue was diluted with 500 µL of 2 M HCl based on the procedure of Passonneau and Lauderdale (1974).Samples were homogenized with a Precellys 24 Homogenizer (Bertin Instruments, Hialeah, FL) at 5,000 rpm for three 10-s intervals using zirconia beads (2.0 mm, catalog no.11079124zx, BioSpec Products, Bartlesville, OK).Homogenates were heated at 95°C for 2 h and frequently vortexed to convert glycogen into glucose units.Samples were then centrifuged at 10,000 × g for 10 min at 4°C and the supernatant transferred to a new tube.A total of 100 µL of supernatant was mixed with 300 µL of a 2 M solution of HCl and 400 µL of a 2 M solution of NaOH.A glucose standard was treated with equal volumes of the same 2 M solutions of HCl and NaOH.A reaction buffer containing 100 mM triethanolamine, 6.8 mM EDTA (pH 7.6), 9.8 mM MgCl 2 , 0.5 mM NADP, and 1.2 mM ATP was added to a series of standards and the unknown samples.Initial absorbance was determined at 340 nm using a microplate reader.Then a cocktail of 1.98 IU of hexokinase and 0.99 IU of glucose-6-phosphate dehydrogenase was added to each well, and a second absorbance was determined at 340 nm.The differences between the second and first absorbance readings were determined for the standards and the unknown samples.Using the standard curve, the concentration of glucose per milliliter was calculated for the unknowns and then expressed per unit of hepatic tissue used in the sample (Ramos et al., 2021).Intraand interassay CV were 5.4 and 7.2%, respectively.

Hepatic RNA Extraction and Transcript Expression
Hepatic tissue was collected on d 6 and 13 of the experiment, and mRNA extraction was performed using Trizol (TRIzol LS Reagent, Invitrogen, Waltham, MA).In summary, approximately 25 mg of hepatic tissue was added to a microtube containing zirconium oxide beads (CKMix, Bertin Corp., Thermo Fisher Scientific, Waltham, MA) containing 800 µL of Trizol.Chloroform was added to establish a 20% solution.The hepatic tissue was homogenized vigorously by hand for 15 s and incubated at room temperature for 3 min.Samples were homogenized 3 times for 20 s at 6,200 rpm using a homogenizer (Precellys 24, Bertin Corp., Thermo Fisher Scientific).Samples were then centrifuged at 12,000 × g for 15 min at 4°C, and the colorless, aqueous supernatant containing the RNA was transferred to a new microtube.Purification of RNA was performed using the Quick-RNA 96 kit (Zymo Research, Irvine, CA) according to the manufacturer's instruction.Purity and concentration were evaluated using a NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific).Samples on d 6 and 13 had mean (±SD) 260:280 nm ratios of 2.02 ± 0.02, and 2.02 ± 0.02, respectively (Supplemental Table S2; https: / / figshare .com/s/ bed22cad5ca6e2460771).Samples from d 6 and 13 had mean (±SD) 260:230 nm ratios of 2.12 ± 0.06, and 2.12 ± 0.06, respectively (Supplemental Table S2).A subset of samples were analyzed for RNA integrity numbers (Supplemental Table S2).The mRNA for a selected set of genes was quantified by the Fluidigm Biomark HD quantitative PCR microfluidic system (Fluidigm Co., San Francisco, CA).The PCR primers were designed by Fluidigm Delta Gene assays and synthesized by Fluidigm (Fluidigm Co., San Francisco, CA).Details of genes and primers are provided in Supplemental Table S3 (https: / / figshare .com/s/ bed22cad5ca6e2460771).A pooled sample containing mRNA from bovine hepatic tissue from 10 different samples was used for primer validation.Primers were validated using Fluidigm primer quality control criteria (R 2 ≥ 0.97; efficiency of 80 to 130%; slope = −3.92 to −2.76) applied to cDNA serially diluted by 12× and evaluated in 8 replicates.Reference genes used were ACTB, GAPDH, PGK1, RPL19, RPS9, and YWHAZ (Goselink et al., 2013;Kessler et al., 2014;Chandler and White, 2017;Arshad, 2023).
Primers targeting 5 genes (FABP3, IL6, SLC27A1, SREBF1, and VLDLR) failed to pass quality control in the qualification run and were excluded from the analyses.Primer efficiency in the genes passing quality control and used in this experiment ranged between 89.6 and 126.6% (Supplemental Table S3).Statistical analyses were performed on delta cycle threshold (dCt) values as described by Steibel et al. (2009).Fold changes relative to reference treatment (CON) were calculated using the method described by Yuan et al. (2006), whereby fold changes were calculated from least squares means (LSM) difference according to the formula 2 −ddCt , where dCt = Ct Target gene − geo-metric mean of Ct Reference genes , and ddCt = dCt Treatment A − dCt Treatment B .Heatmaps were generated using the Heatmapper online tool (Babicki et al., 2016).

Tyloxapol Infusion and Blood Sampling and Processing
On d 13, an intravenous 14-gauge and 9-cm long catheter (MilacathExtended Use, Mila International Inc., Florence, KY) was fitted into the jugular vein of each cow.Patency was maintained by flushing with 15 mL of sterile heparinized saline containing 10 IU of heparin per milliliter.Starting at 1400 h of d 13, any feed left in the bunk was removed, and cows remained off-feed until 2100 h of d 14, with only access to water.On d 14 of the experiment, only treatments were fed at 0600 h, followed by tyloxapol infusion at 0900 h.Cows received an intravenous solution of 10% tyloxapol (product T8761, Sigma-Aldrich, St. Louis, MO), a 4-(1,1,3,3-tetramethylbutyl) phenol polymer with formaldehyde and oxirane, at a dose of 120 mg/kg BW.The weight of cows was calculated as BW minus 50 kg, with the latter representing the expected pregnant uterus weight at 230 d of gestation (Bell et al., 1995).Cows remained off feed throughout the 12-h blood sampling period after tyloxapol dosing, to ensure that the measured serum triacylglycerol concentrations reflected hepatic secretion of VLDL triacylglycerols and not influx of dietary triacylglycerol as chylomicrons.Tyloxapol is a nonionic detergent that blocks the action of lipoprotein lipase in vivo, thus preventing lipolysis of triacylglycerols in lipoprotein particles (Schotz et al., 1957).It has been shown that acute increases in the blood concentrations of fatty acids are re-esterified to triacylglycerol in the hepatic tissue and exported as a component of VLDL particles (Lombardi et al., 1968), and the proportion of triacylglycerol in the bovine VLDL is 64% of total lipid moieties (Stead and Welch, 1975).Thus, the rationale for administering tyloxapol was to induce hyperlipidemia and quantify triacylglycerol accumulation over time as a measure of hepatic VLDL secretion rate, as shown in guinea pigs (Fernandez et al., 1997) and mice (Rinella et al., 2008).
Each 1 mL of tyloxapol product contained 1.25 g of 4-(1,1,3,3-tetramethylbutyl) phenol polymer with formaldehyde and oxirane, and a 10% tyloxapol solution was prepared by dissolving and gently mixing the calculated amount of tyloxapol product of each cow in 0.9% sodium chloride solution for 2 h for proper homogenization.All procedures were performed aseptically.The dose of tyloxapol of 120 mg/kg of BW was chosen to achieve hyperlipidemia, as described previously in Holstein cows (Wrenn et al., 1971) and guinea pigs (Fernandez et al., 1997).
Blood was sampled immediately before the administration of tyloxapol, and at 10,20,40,60,120,180,240,480, and 720 min after infusion.At each sampling, 20 mL of blood was aspirated with a syringe and discarded, and then another 20 mL of blood was aspirated as the specimen used for subsequent assays.Blood was immediately transferred to one 10-mL tube containing no anticlotting agent for serum separation, and to one 10-mL tube containing lithium heparin for plasma separation (Vacutainer, Becton Dickson, Franklin Lakes, NJ).Tubes were placed in ice and centrifuged within 1 h of collection at 2,000 × g for 20 min at room temperature for serum and plasma separation.Serum and plasma samples were aliquoted into multiple vials and frozen at −20°C until analyses.

Lymph Sampling and Processing
Lymph was collected 6 h after infusion of tyloxapol from 30 cows: 10 CON, 11 L25.8, and 9 H25.8 cows.We were unable to obtain a sample from 1 CON and 2 H25.8 cows.The posterior udder of the cow was shaved and the skin thoroughly cleansed with an iodophor scrub and 70% alcohol.Superficial lymph vessels were identified by their contour, lack of movement when the skin was moved, and orientation crossing under the superficial blood vessels, as depicted in Khol et al. (2012).An 18-gauge (1.27-mm) needle connected with a 3-mL syringe was placed at an angle of approximately 45° through the skin and inserted a few millimeters into the lymph vessel, and lymph was gently aspirated.The fluid was transferred to a sterile tube containing 0.75 mL of K 2 EDTA to prevent clotting.Tubes were centrifuged for 5 min at 1,000 × g at room temperature, and lymph samples were transferred into multiple aliquots of 1.0 mL and stored frozen at −20°C until subsequent analyses.

Blood and Lymph Measurements
Serum samples collected on d 1 (covariate) and d 14 of the experimental period after tyloxapol infusion were analyzed for concentrations of triacylglycerol and total cholesterol.All assays followed the initial randomization with blocks, such that samples from a given block were analyzed in the same assay.Concentrations of triacylglycerol (Stanbio Triglycerides LiquiColor Procedure No. 2100, Stanbio Laboratory, Boerne, TX) and total cholesterol (Stanbio Cholesterol LiquiColor Procedure No. 1010, Stanbio Laboratory) were analyzed using colorimetric enzymatic assays.Intra-and interassay CV were, respectively, 3.7 and 6.4% for triacylglycerol and 1.4 and 4.8% for total cholesterol.were approximated using the following equation: VLDL cholesterol = triacylglycerol/5, as reported previously (Friedewald et al., 1972;Kessler et al., 2014).
Targeted metabolomic analyses were performed in the serum samples collected at 720 min after tyloxapol infusion, using an MxP Quant500 Kit (Biocrates Life Sciences AG, Innsbruck, Austria) to identify and quantify 107 small molecules and 523 lipids.The assay was performed in serum samples from 10 CON and 20 RPC-fed cows.The serum sample was prepared as directed in the manufacturer's instructions (Biocrates Life Sciences, AG, Innsbruck, Austria) for metabolomic analysis.Serum samples and calibration standards were thawed on ice, mixed for 10 s, and centrifuged at 10,000 × g at 4°C for 10 min.Calibration standards and quality controls were dissolved in 100 µL of H 2 O and mixed at 1,200 rpm for 15 min.The sample volume of 10 µL and 10 µL of calibration standards, quality controls, and PBS were added to the 96-well plate.The plate was dried under nitrogen for 30 min.All samples and standards were included in a premix of phenyl isothiocyanate at room temperature for 60 min for derivatization purposes and subsequently dried under nitrogen for 60 min.Samples were extracted in 5 mM ammonium acetate in methanol for 30 min using an orbital shaker, and the extracts were collected by centrifuging the preparation plate at 500 × g for 2 min at room temperature.Sample extracts were diluted with H 2 O (1:1) for the LC phase of the analysis.For the flow injection analysis, 50 µL of sample extract was mixed with 450 µL of the kit solvent on a separate plate.The LC and flow injection analysis plates were sealed, mixed for 10 min at 600 rpm at room temperature, and placed into the thermostatically controlled autosampler for analysis.
Serum extracts were analyzed using an Exion LC unit coupled with a Waters Xevo-TQ-S (Waters Corporation, Milford, MA).Sample extracts were separated using the MxP Quant500 C18 column with an attached guard and precolumn mixer (Biocrates Life Sciences AG, Innsbruck, Austria).The mobile phase consisted of 2 components: (A) H 2 O and 0.2% formic acid, and (B) MeCN mixed with 4 parts of 0.2% formic acid.The mobile phase was delivered at a flow rate of 0.8 mL/ min with a gradient of B from 0 to 100% over 4.5 min.Eluent percentage of B was increased to 1.0 mL/min flow rate and maintained at 100% for 30 s, followed by a rapid return to the initial conditions for 70 s to equilibrate the column.Both positive-and negativemode gradients were 5.8 min long.The negative-mode acquisition gradient differed from positive mode with a difference in percentage B composition between 2.0 and 4.5 min.The injection volume was 5 µL for positive data acquisition and 15 µL for the negative run.Wash solvent composition consisted of equal volumes of water, methanol, acetonitrile, and isopropanol.
The Q500 Kit offers direct-flow injections for lipid analysis.An isocratic method was performed using the kit-provided solvent (290 mL methanol: 1 ampule of flow-injection analysis additives).The isocratic mobile phase (B: 100% methanol) was delivered at slow flow rate of 0.03 mL/min.The injection volume was 20 µL for both positive-and negative-mode acquisitions.All data were extracted using the MetIDQ software following Biocrates instructions (Biocrates, Innsbruck, Austria).Three different concentration ranges of quality control samples (low, mid, and high) were provided by the manufacturer (Biocrates, Innsbruck, Austria).

Statistical Analyses
The experiment followed a randomized complete block design with cow as the experimental unit.The areas under the curve (AUC) for serum concentrations of triacylglycerol, VLDL cholesterol, and total cholesterol for the first 720 min after tyloxapol infusion were calculated for each cow by the trapezoid method us-ing the EXPAND procedure of SAS (SAS STAT, SAS Institute Inc.).
Normality of residuals and homogeneity of variance were examined for each continuous dependent variable analyzed after fitting the statistical model.Concentrations of free fatty acids and glucose in lymph violated the assumptions of normality and were subjected to power transformation according to the BoxCox procedure (Box and Cox, 1964) using a macro (Piepho, 2009) for mixed models in SAS (SAS STAT, SAS Institute Inc.).Whenever, the significance of parameters and interpretation of the data did not change between the results of the analyses of the data on the original scale and the results after power transformation, we opted to keep the results from the analyses in the original scale of the data to avoid back-transformation of the standard error of the mean.
Data were analyzed by linear mixed-effects models using the MIXED procedure of SAS (SAS STAT, SAS Institute Inc.), and analyses were performed separately for the ad libitum and feed restriction periods.The statistical models included the fixed effects of treatment (CON vs. L25.8 vs. H25.8),BCS and BW at the time of enrollment, whether the cow had twins or not (yes or no), covariate measurement taken before treatment implementation whenever available, and random effect of block.For responses with repeated measures within cow, the models also included the fixed effects of minute of measurement and the interaction between treatment and minute, and the random effect of cow nested within treatment.For blood metabolites, the measurement at d 1 was used as covariate in the statistical analyses.Minute was the term in the REPEATED statement.Because of unequal spacing between measurements, the covariance structure used to model the correlated errors of repeated measurements from the same experimental unit was a spatial power structure.For analysis of serum cholesterol, the final Hessian matrix was not positive definite, resulting in issues with model convergence.For that, 3 other covariance structures were evaluated: unstructured, compound symmetry, and compound symmetry heterogeneous; selection was based on model fit, assessed using the smallest Akaike's information criterion.In all mixed-effects models, the Kenward-Roger method was used to approximate the denominator degrees of freedom to compute the Ftests.When an interaction between treatment and time resulted in P < 0.10, means at different minutes were partitioned using the SLICE command of SAS (SAS STAT, SAS Institute Inc.).
In all analyses, single degrees of freedom orthogonal contrasts were evaluated to assess the effects of RPC (CON vs. [1/2 L25.8 + 1/2 H25.8]), and source of RPC (L25.8 vs. H25.8).Data are reported as LSM and stan-dard error of the mean.Evidence of statistical significance against the null hypothesis was considered at P ≤ 0.05, and tendency was considered at 0.05 < P ≤ 0.10.
The metabolome data on serum samples 12 h after infusion of tyloxapol were compared between CON and RPC-fed cows.All absolute metabolites concentrations quantified were normalized to the respective cows' total protein content in the serum and expressed as nM/g of protein.Univariate t-tests with false discovery rate adjusted to P ≤ 0.10 were performed to compare concentrations of metabolites between 2 treatments using R (version 4.1.3,R Development Core Team, 2022, https: / / www .r-project .org).Statistical significance was considered at P ≤ 0.05, and the tendency was assessed at 0.05 < P ≤ 0.10.All metabolomics data were processed and analyzed using the MetaboAnalyst 5.0 software (Xia et al., 2009).Recommended statistical procedures for metabolomics analyses were followed according to previously published protocols (Xia et al., 2009(Xia et al., , 2015)).Metabolites that were frequently (>30%) below the detection limit or with more than 30% missing values were excluded from the data sets.Otherwise, missing values were imputed by the k-nearest neighbor model with MetaboAnalyst software.Data normalization of metabolite concentration was performed before the statistical analysis, and metabolic pathway analysis was used to create a Gaussian distribution (Xia et al., 2009).Log-transformation and autoscaling were used for metabolite values.
Principal component analysis, and partial least squares-discriminant analysis (PLS-DA) were performed via MetaboAnalyst.In the PLS-DA model, variable importance in projection (VIP) plot was used to rank the metabolites based on their importance in discriminating RPC-fed cows from the CON treatment.The VIP score is a measure of the importance of each variable in the PLS-DA model used and summarizes the contribution of each variable to the model.Metabolites with large VIP values, typically >1, are those most important in discriminating CON from RPC-fed cows.

Ad Libitum Feeding Period
Hepatic Composition.Feeding RPC or 1 of the 2 sources of RPC did not affect the content of hepatic triacylglycerol and glycogen on as-is basis, or the ratio of triacylglycerol to glycogen in the hepatic tissue on d 6 of the experiment (Table 1).
Hepatic Tissue mRNA Expression.A heatmap with the transcripts differentially (P ≤ 0.10) expressed during the ad libitum period was created using the fold change relative to CON to illustrate the patterns according to treatment (Figure 2A).The transcripts affected (P ≤ 0.10) by treatment and the respective fold changes are reported in Table 2, and LSM for the dCt are presented in Supplemental Table S4 (https: / / figshare .com/s/ bed22cad5ca6e2460771).
Feeding RPC increased (P ≤ 0.01) the expression of transcripts involved in the synthesis of phosphatidylcholine (PCYT1A), cellular uptake of fatty acids (CD36) and carnitine (SLC22A5), and ER stress response (ATF4), and tended (P ≤ 0.09) to upregulate transcripts involved in hepatocellular transport of fatty acids (FABP5), synthesis and assembly of VLDL particles (MTTP), and cellular autophagy (ATG5).However, feeding RPC decreased (P ≤ 0.04) the expression of transcripts associated with cytokines (IL1B) and lipophagy (RAB10), and tended (P = 0.09) to downregulate a transcript involved in mitochondrial oxidation of fatty acids (PCCA).Feeding RPC had mixed effects on expression of nuclear factor kappa B transcripts, tended (P = 0.07) to reduce expression of NFKB1, and tended (P = 0.06) to increase the expression of NFKB2.
Feeding L25.8 increased (P ≤ 0.03) the expression of SLC22A5, transcripts pertaining to acute phase re-sponse (HP, SAA3) and cellular antioxidant protection (GPX3), and tended (P ≤ 0.07) to upregulate PCYT1A, ATF4, and a transcript associated with protein folding and quality control in the ER lumen (HSPA5) compared with H25.8, whereas expression of a transcript associated with the activation of medium-chain fatty acids (ACSM1) tended (P = 0.10) to be upregulated in cows fed H25.8 compared with those fed L25.8.

Feed Restriction Period
Hepatic Composition.Feeding RPC reduced (P < 0.001) the concentration of triacylglycerol (CON = 9.0 vs. RPC = 4.3 ± 0.6% wet weight basis) and increased (P < 0.001) that of glycogen (CON = 1.9 vs. RPC = 3.8 ± 0.2% wet weight basis) in the hepatic tissue (Table 1), which resulted in a ratio of triacylglycerol to glycogen that was 5.3-fold smaller (P < 0.001) for cows fed RPC compared with CON (CON = 5.62 vs. RPC = 1.06 ± 0.6).The same effects of RPC were observed on hepatic composition on dry tissue basis (data not shown).Source of RPC (P < 0.01) affected the glycogen content of the hepatic tissue, and cows fed H25.8 had greater glycogen content than those fed L25.8.
Hepatic Tissue mRNA Expression.A heatmap with the transcripts differentially (P ≤ 0.10) expressed during the feed restriction period was created using the fold change relative to CON to illustrate the patterns according to treatment (Figure 2B).The transcripts affected (P ≤ 0.10) by treatment during the feed restriction period and the respective fold changes are Cows were fed for ad libitum intake on d 1 to 5 and hepatic tissue collected on the morning of d 6, before feed restriction.
Feeding RPC increased (P = 0.009) the expression of a transcript involved in inflammation (TNFA) and tended (P ≤ 0.10) to upregulate MTTP and transcripts pertaining to cellular autophagy (ATG3) and ER stress response (XBP1); however, feeding RPC decreased (P = 0.03) the expression of a transcript associated with biogenesis and stabilization of lipid droplets (PLIN2) and tended (P ≤ 0.09) to downregulate transcripts involved in mitochondrial oxidation of fatty acids (HADHA, MLYCD) and lipophagy (RAB18).

Responses After Tyloxapol Infusion
Cows received their last meal at 0900 h of d 13, 24 h before dosing tyloxapol.On that morning, intakes of DM and fatty acids were, respectively, 4.8 kg and 85 g for CON, 4.3 kg and 75 g for L25.8, and 4.9 kg and 86 g for H25.8.This limited intake of DM and fatty acids 24 h before tyloxapol was designed to restrict chylomicron appearance in plasma on d 14.

Concentrations of Metabolites in Serum.
The concentrations of triacylglycerol in serum were affected by the interaction (P < 0.001) between treatment and time (Figure 3A).The interaction was caused by the greater increase in concentrations of triacylglycerol over time in cows fed RPC compared with CON, which resulted in a larger (P = 0.03) AUC in cows fed RPC compared with CON (CON = 21,741 vs. RPC = 30,511 ± 3,706 mg/dL × min; Figure 3B).The same responses to feeding RPC were observed for VLDL cholesterol (Figures 3C and 3D), which resulted in larger (P = 0.03) AUC in cows fed RPC compared with CON (CON = 4,348 vs. RPC = 6,103 ± 741 mg/dL × min); however, treatment did not affect the concentrations of total cholesterol in serum or the AUC for total cholesterol after infusion of tyloxapol (Figures 3E and 3F).Source of RPC had no effect on the concentrations of triacylglycerol, VLDL cholesterol, or total cholesterol in serum of cows after tyloxapol infusion.
Lymph Composition.Treatment did not affect concentrations of fatty acids, BHB, glucose, and total cholesterol in lymph; however, feeding RPC tended (P = 0.07) to reduce the concentrations of triacylglycerol in lymph (CON = 16.7 vs. RPC = 12.9 ± 1.9 mg/ dL) compared with CON (Table 4).Serum Metabolome.Treatment affected (P ≤ 0.10) the concentrations of 89 metabolites in serum after adjusting for false discovery, and feeding RPC increased the concentrations of those metabolites (Figure 4A).Specifically, 3 acylcarnitines, 1 AA-related metabolite, 11 bile acids, 1 ceramide, 6 diacylglycerols, 2 dihydroceramides, 1 glycerophospholipid, and 64 triacylglycerols with varying chain lengths differed between treatments and were in greater concentrations in RPC than in CON cows.Table 5 depicts the mean (±SD) concentrations (nM/g of total protein) of the 20 serum metabolites according to their largest fold change and affected by treatment, their associated P-values after adjusting for false discovery rate, and the direction of change in concentration.The 20 metabolites, including 6 bile acids, 4 diacylglycerols, and 10 triacylglycerols with varying chain lengths, were the compounds with the greatest effects in discriminating RPC from CON serum metabolome and are shown according to their VIP scores (Figure 4B).The principal component analysis shows limited segregation of the serum composition between CON and RPC-fed cows (Figure 4C).When PLS-DA were performed, including the 89 metabolites that differed between treatments, a slight discrimination between treatments was observed (Figure 4D).The heatmap (Figure 5) shows the 20 most abundant metabolites in the serum metabolome clustered by the treatment, and the abundance of metabolites increases from blue to red.A schematic summary of the findings of the present experiment and proposed mechanisms by which RPC alters hepatic metabolism and reduces the risk of hepatic lipidosis are depicted in Figure 6, which combines findings from this experiment and those of others in which dairy cows were supplemented with RPC (Goselink et al., 2013;Zenobi et al., 2018a;Arshad, 2023).

DISCUSSION
Feed restriction induced hepatic lipidosis with more than a 4-fold increase in hepatic triacylglycerol and a reduction of more than 50% in hepatic glycogen in CON cows; however, supplementing 25.8 g/d of choline ion as RPC minimized the increase in hepatic triacylglycerol during feed restriction and maintained hepatic glycogen concentrations.The changes in hepatic composition caused by feeding RPC were compatible with a lipotropic effect of choline on the hepatic tissue and likely mediated by the changes in the ability of hepatocytes to assemble and export triacylglycerol-rich lipoproteins.Indeed, cows fed a diet supplemented with RPC had increased triacylglycerol accumulation in serum, based on the progressive hypertriglyceridemia after infusion of tyloxapol, a response compatible with increased secretion of nascent VLDL by the liver (Fernandez et al., 1997).The increased AUC for triacylglycerol and VLDL cholesterol in serum suggest increased packaging of triacylglycerol in the core of VLDL particles and increased hepatic export of those compounds in cows fed RPC.These responses were accompanied by increased expression of MTTP and ATG3 involved in the synthesis and assembly of VLDL particles and autophagy, whereas decreased expression of PLIN2, which is associated with the stabilization of lipid droplets during feed restriction.These results corroborate previously published data in rats (Lombardi et al., 1968) and mice (Rinella et al., 2008) in which choline displays a lipotropic effect enhancing hepatic export of triacylglycerol, thus reducing the risk of hepatic lipidosis.
Feeding choline as RPC has been effective to prevent and alleviate hepatic triacylglycerol infiltration in nonlactating pregnant cows under feed restriction (Cooke et al., 2007;Zenobi et al., 2018a).Cooke et al. (2007) proposed reduced lipolysis as one of the potential mechanisms by which feeding RPC might attenuate the degree of hepatic lipidosis; nevertheless, this hypothesis has not been aligned with the findings by Zenobi et al. (2018a) and Arshad et al. (2023), who demonstrated that increasing intake of choline ion as RPC reduced the concentrations of hepatic triacylglycerol independent of changes in plasma fatty acids or BHB.In fact, when Goselink et al. (2013) fed 12.9 g/d of choline ion as RPC during the transition period, they reported that RPC reduced hepatic lipidosis without affecting the expression of genes involved in adipose tissue lipolysis, thus reinforcing the idea that the reduction in hepatic triacylglycerol from supplementing choline is mediated either by reduced hepatic uptake of fatty acids or by increased export of lipoprotein-rich triacylglycerols, and not by reduced adipose tissue lipolysis.In the present experiment, feeding RPC had no effect on the expression of genes involved in the uptake or hepatocellular transport of fatty acids during feed   restriction; nevertheless, feeding RPC increased the expression of MTTP, a gene that regulates the increased secretion of apolipoprotein B-100 (APOB100)-containing lipoproteins in hepatic cells (Jamil et al., 1998).Microsomal transfer triglyceride protein transfers neutral lipids between membrane vesicles and functions as a chaperone for the synthesis of hepatic VLDL, and these steps are considered rate-limiting for the secretion of VLDL triacylglycerols and APOB100 in mice (Tietge et al., 1999).Administration of pharmacological inhibitors of the MTTP protein resulted in decreased secretion of triacylglycerol and APOB100 from hepatic cells (Jamil et al., 1996), whereas an overexpression of hepatic MTTP in vivo resulted in increased secretion of VLDL triacylglycerols and APOB100 by the liver (Tietge et al., 1999).Also, cows fed RPC during the transition period showed a reduction in postpartum hepatic triacylglycerol content (Elek et al., 2013;Goselink et al., 2013), coinciding with an increase in the expression of MTTP and APOB100 in hepatic tissue (Goselink et al., 2013).In the dry cow model, Arshad (2023) showed that hepatic expression of MTTP and APOB100 increased with the supplementation of RPC in cows induced to develop hepatic lipidosis.Collectively, data from the present experiment and those of others in which dairy cows were fed diets supplemented with RPC (Goselink et al., 2013;Arshad, 2023) support the concept that supplying choline during periods of negative nutrient balance alters hepatic expression of genes that favor assembly and secretion of VLDL to enhance hepatic disposal of triacylglycerol.Nevertheless, it is important to mention that the lipotropic effects of RPC in reducing hepatic triacylglycerol postpartum in lactating cows are not consistent (Arshad et al., 2020), perhaps because the increase in milk yield is not necessarily followed by a concurrent increase in DMI in the first weeks postpartum (Zenobi et al., 2018b;Bollatti et al., 2020).In Figure 6, we propose a model to explain the changes in hepatic composition using data originated in the present experiment and those of others that support the reduced hepatic triacylglycerol content caused by RPC in dairy cows (Piepenbrink and Overton, 2003;Goselink et al., 2013;Zenobi et al., 2018a;Arshad, 2023).Jia et al. (2019) demonstrated that dairy cows affected by fatty liver had increased abundance of the PLIN5 gene and the PLIN5 protein in hepatic lipid droplets compared with cows without fatty liver.Those authors also showed that an overexpression of PLIN5 in bovine primary hepatocytes promoted hepatic steatosis by increasing lipid biogenesis and compromising the synthesis and assembly of VLDL particles.In nonruminants, nonalcoholic fatty liver disease (NAFLD) is associated with increased accumulation of lipid droplets in the hepatic tissue concurrent with reduced cellular autophagy (Mashek et al., 2015;Schulze et al., 2017).Lipid droplets carry a cargo that includes perilipins (PLIN), and a growing body of literature shows that increased activity of perilipins allows sequestering of lipids by protecting the lipid droplets from cytosolic lipase actions (Sztalryd and Brasaemle, 2017).In the schematic model in Figure 6, we show that cows fed RPC had reduced expression of PLIN2 and increased expression of ATG3 during feed restriction.Thus, it is possible that reduced expression of PLIN2 led to the mobilization and export of lipid droplets as a component of VLDL particles (Gluchowski et al., 2017), or that increased expression of ATG3, a component of ubiquitination-like systems favored the fusion of lipid droplets into autophagosomes that favor their degradation (Schulze et al., 2017).Andrejeva et al. (2020) showed that synthesis of phosphatidylcholine is required for the formation of autophagosomes, and feeding RPC increases plasma concentrations of phosphatidylcholines and sphingolipids in plasma (Zenobi et al. 2018a).It is plausible to suggest that supplementing RPC increases the synthesis of phosphatidylcholines that might facilitate the disposal of lipid droplets in autophagosomes during autophagy in hepatic tissue.Feeding RPC increased the concentrations of glycogen in the hepatic tissue, combined with the reduced content of triacylglycerol, which resulted in a marked reduction in the ratio of hepatic triacylglycerol to glycogen during the feed restriction period.These results corroborate those of Piepenbrink and Overton (2003), who showed that supplementation of choline as RPC during the transition period increased the hepatic glycogen content in early-postpartum lactating Holstein cows.Furthermore, when Zenobi et al. (2018a) subjected cows to feed restriction and supplemented increased amounts of choline ion as RPC, they showed that concentration of hepatic glycogen increased in a quadratic fashion in response to dose of RPC.Altogether, results from the present experiment and those of Piepenbrink and Overton (2003) and Zenobi et al. (2018a) strongly suggest that supplementing RPC to cows during periods of negative nutrient balance either increase glycogenesis or reduce glycogenolysis.In the present experiment, RPC did not affect the expression of glycogen synthase kinase 3 β (GSK3B), which regulates glycogenesis by phosphorylation (inactivation) of GSK3A.Chandler and White (2019) isolated primary hepatocytes from neonatal Holstein calves and treated them with increasing concentrations of choline chloride, from 0 to 4,528 µM, which resulted in increased con- centrations of cellular glycogen.Increasing the amount of choline ion increased the expression of transcripts such as betaine-homocysteine S-methyltranferase, 5-methyltetrahydrofolate-homocysteine methyltransferase, and methylenetetrahydrofolate reductase, which code for enzymes that play a role in the regeneration of methionine in 1-carbon metabolism (Chandler and White, 2017;Arshad, 2023).Perhaps methyl groups originating from choline facilitated regeneration of methionine or spared methionine, and the latter can enhance gluconeogenesis by exerting control on PPARG coactivator 1 α (Tavares et al., 2016), thus favoring the flux of carbon toward glucose and glycogen synthesis in hepatocytes.In rats, feeding choline increases the glycogen content in hepatocytes (Munck and Koritz, 1964), but the opposite effect was observed with methionine.Moreover, Baquet et al. (1990) showed that increasing weight of hepatocytes was positively associated with the synthesis of glycogen, and choline induces cell swelling by hypo-osmolarity of hepatocytes, coinciding with an increased synthesis of glycogen.It is noteworthy to mention that choline can be phosphorylated to synthesize phosphatidylcholine, and the latter can affect the fluidity of plasma membranes (Hensley et al., 2000).Also, hydrolysis of phosphatidylcholine to glycerophos-phocholine, or oxidation of choline to betaine, which act as osmoregulatory compounds, might increase the swelling of hepatocytes to stimulate glycogen synthesis (Baquet et al., 1990).Finally, feeding RPC reduced hepatic triacylglycerol, and increased lipidosis can compromise hepatic gluconeogenesis (Mashek et al., 2002), which is needed to supply glucose for glycogen synthesis.It is possible that the reduced lipidosis observed in the present experiment in cows fed RPC might have increased gluconeogenesis to supply glucose-carbon for glycogen synthesis.
Bile acids are relatively hydrophilic cholesterol derivatives that are synthesized in the liver, and cholesterol is mainly exported from hepatic tissue in the forms of biliary cholesterol, bile acids, or cholesteryl esters as components of lipoprotein particles.Cows fed RPC had increased hepatic secretion of VLDL cholesterol and concentration of primary and secondary bile acids in serum.Both choline and phosphatidylcholine are known to stimulate hepatic synthesis and secretion of bile acids (LeBlanc et al., 1998).Rats fed a choline-enriched diet had hepatic tissue with increased concentrations of cholesteryl esters, coinciding with increased hepatic secretion of bile acids and cholesterol compared with rats fed a choline-free diet (LeBlanc et al., 1998).In humans, altered hepatic cholesterol metabolism and accumulation into hepatocytes are involved in the development of steatosis during NAFLD (Arguello et al., 2015).Humans with hepatic steatosis have increased hepatic cholesterol synthesis and de-esterification but reduced export, concurrent with reduced bile acid synthesis (Arguello et al., 2015).It is possible that such alterations are also present in dairy cows that develop hepatic lipidosis and that elevated concentrations of cholesterol in the hepatic tissue exacerbate the degree of fatty liver in dairy cows.Accumulation of hydrophobic bile acids in hepatocytes activates Kupffer cells, which produce reactive oxygen species (Ljubuncic et al., 1996).The increased cellular oxidative stress might exacerbate injury to hepatocytes and, therefore, further limit the ability to export lipids.Cows fed RPC had increased serum concentrations of primary and secondary bile acids, and choline is known to stimulate bile acid synthesis and cholesterol export by the liver (LeBlanc et al., 1998).In the present experiment, we observed an increase in serum concentrations of glycine or taurine conjugates of cholic acids, deoxycholic acids, and chenodeoxycholic acids, which suggest either an increased synthesis or recycling of bile acids in the liver-gut axis.Increased concentrations of primary or secondary bile acids may act on their nuclear receptors such as farnesoid X receptors (FXR), which may exhibit a protective role against the progression of NAFLD (Gottlieb and Canbay, 2019).It has been shown that the agonists of FXR receptors increased the expression of transcripts and protein abundance of fibroblast growth factors, which are known to increase hepatic oxidation of fatty acids, decreased hepatic lipogenesis, and reduced inflammation and steatosis in hepatic tissue (Kalaany and Mangelsdorf, 2006;Gottlieb and Canbay, 2019).We did not measure the abundance of bile acids in hepatic tissue or transcripts or enzymes involved in the synthesis and regulation of bile acids, so such mechanisms require further investigation.Indeed, it would be interesting to determine whether supplementation of choline to dairy cows can alter the activity of FXR to regulate bile acids in the attenuation of severity of hepatic lipidosis in dairy cows.The composition of lymph was mostly unaltered by feeding RPC, except for the tendency of reduced concentration of triacylglycerol.Cows were under feed restriction for 8 d, and the last meal occurred approximately 30 h before lymph collection, which might be because of reduced synthesis of chylomicrons, which carry triacylglycerols in lymph.Perhaps the limited intake of fatty acids by those cows, only 75 to 80 g/d, consumed 30 h before lymph sampling did not require additional phospholipids for intestinal uptake of those lipids.It was anticipated that feeding RPC would facilitate absorption of dietary fatty acids and increase lymph content of triacylglycerols; however, lymph was not collected from the mesenteric lymphatic duct or vena cava to determine the representative concentrations of metabolites in intestinal tissue.Therefore, the concentrations of metabolites in supra-mammary lymph might not necessarily reflect the concentrations of lymph draining the intestinal tract.

CONCLUSIONS
Feed restriction induced hepatic lipidosis in dairy cows, but supplementing RPC minimized the increase in hepatic triacylglycerol and maintained hepatic glycogen concentrations.The changes in hepatic composition caused by RPC were compatible with a lipotropic effect of choline and were linked with an increased secretion of hepatic VLDL.The reduction in hepatic triacylglycerol caused by feeding RPC was observed concurrent with increased hepatic expression of genes involved in synthesis and assembly of lipoproteins and autophagy, and reduced expression of a gene involved in the stabilization of lipid droplets.We propose a model suggesting The PBA synthesized from cholesterol (CH) in hepatic tissue are transported via bile duct to the small intestine, and after enzymatic activity (EA) of microbes in the small intestine, they are converted to secondary bile acids (SBA; L).Feeding RPC provides substrate for synthesis of phosphatidylcholine, and both choline and phosphatidylcholine might alter the activity of farnesoid X receptors to regulate the hepatic synthesis, recycling, or flow of bile acids in the liver-gut axis, which contributes to hepatic cholesterol metabolism and secretion and, thus, might help limit the progression of fatty liver in dairy cows.
that cows fed RPC had increased mobilization of lipid droplets to be exported as a component of VLDL particles or to fuse with autophagosomes for subsequent degradation.Supplementing RPC increased the export of cholesterol as VLDL cholesterol and bile acids that might reduce the severity of fatty liver.Collectively, RPC increased hepatic triacylglycerol secretion to promote lipotropic effects that reduce hepatic lipidosis in dairy cows.
Arshad et al.: CHOLINE AND HEPATIC TRIACYLGLYCEROL SECRETION CON was to supply similar amounts of lipids from the top-dress in each treatment.Cows fed L25.8 received 200 g/d of a top-dress containing ReaShure, hydrogenated triacylglycerol, and ground corn at a 60:0:40 ratio, respectively.Cows fed H25.8 received 200 g/d of a top-dress containing the prototype, hydrogenated triacylglycerol, and ground corn at a 29:31:40 ratio, respectively.Investigators were not blind to treatments.
Figure 1.Diagram of the experiment.Cows were blocked by BCS and, within block, randomly assigned to receive CON (0 g/d choline ion; n = 11) or to be supplemented with rumen-protected choline (RPC) to supply 25.8 g/d of choline ion from either L25.8 (28.8% choline chloride in the product; n = 11) or H25.8 (60% choline chloride; n = 11).Panel A: cows were fed for ad libitum intake on d 1 to 5 and restricted to 50% of the NE L required for maintenance and pregnancy from d 6 to 13.No feed was offered on d 14.Panel B: d 14, cows were off feed and received intravenously a 10% solution of tyloxapol at 120 mg/kg of BW.
Arshad et al.: CHOLINE AND HEPATIC TRIACYLGLYCEROL SECRETION Figure2.Heat maps including differently (P ≤ 0.10) expressed transcripts by rumen-protected choline (RPC).Genes were separated using the average linkage clustering method and following a Euclidean distance measurement approach.The mRNA expression increases from red to green, and it is presented on a Z-score-based scaling system.Panel A: hepatic tissue sampled on d 6. Panel B: hepatic tissue sampled on d 13 during feed restriction.Cows were assigned to CON (0 g/d choline ion) or to be supplemented with RPC to supply 25.8 g/d of choline ion in either low (L25.8)or high concentration (H25.8) of choline chloride in the rumen-protected product.

Figure 3 .
Figure 3. Concentrations of triacylglycerol (A), area under the curve (AUC) for triacylglycerol (B), concentrations of very low density lipoprotein (VLDL) cholesterol (C), AUC for VLDL cholesterol (D), concentrations of total cholesterol (E), and AUC for total cholesterol (F) in serum after an intravenous administration of tyloxapol in dairy cows.Treatments are CON (0 g/d choline ion), L25.8 (25.8 g/d of choline ion from a rumen-protected choline [RPC] product with 28.8% choline chloride), or H25.8 (25.8 g/d choline ion from a RPC product with 60.0% choline chloride).Panel A: interaction of treatment and time (P < 0.001).Panel B: effects of RPC (P = 0.03) and source of RPC (P = 0.43).Panel C: interaction of treatment and time (P < 0.001).Panel D: effects of RPC (P = 0.03) and source of RPC (P = 0.43).Panel E: interaction of treatment and time (P = 0.69).Panel F: effects of RPC (P = 0.91) and source of RPC (P = 0.28).Error bars depict SEM.

Figure 6 .
Figure6.Proposed model illustrating the effects of rumen-protected choline (RPC) on hepatic metabolism in dairy cows under negative nutrient balance.The model includes findings from the present experiment and those of others in which cows were fed RPC.Inadequate intake of nutrients induces lipolysis that increases concentrations of fatty acids (FA) in blood (A), which increases the uptake of those FA by hepatocytes (B) and predisposes cows to hepatic lipidosis.Fatty acids taken up by the hepatic tissue can be re-esterified to triacylglycerol (TAG; C), thus increasing synthesis of lipid droplets (LD; D); nevertheless, cows fed RPC have reduced expression of PLIN2, which might increase hydrolysis of LD (D).Feeding RPC increases the expression of MTTP, which transfers neutral lipids to very low density lipoprotein (VLDL) particles (E) and APOB100, the main lipoprotein in VLDL, both required for hepatic export of TAG (F).Feeding RPC increases hepatic secretion of triacylglycerol-rich lipoprotein (G), which reduces hepatic lipidosis.Choline increases the expression of ATG3 (H), possibly increasing vesicle elongation (VE) and formation of autophagosomes (AP) to carry out LD degradation during autophagy (I).Hepatic tissue in cows fed RPC has increased glycogen content because of either increased glycogenesis or reduced glycogenolysis (J).Cows fed RPC have increased concentrations of primary bile acids (PBA; K) in serum.The PBA synthesized from cholesterol (CH) in hepatic tissue are transported via bile duct to the small intestine, and after enzymatic activity (EA) of microbes in the small intestine, they are converted to secondary bile acids (SBA; L).Feeding RPC provides substrate for synthesis of phosphatidylcholine, and both choline and phosphatidylcholine might alter the activity of farnesoid X receptors to regulate the hepatic synthesis, recycling, or flow of bile acids in the liver-gut axis, which contributes to hepatic cholesterol metabolism and secretion and, thus, might help limit the progression of fatty liver in dairy cows.
Arshad et al.: CHOLINE AND HEPATIC TRIACYLGLYCEROL SECRETION

Table 1 .
Arshad et al.:CHOLINE AND HEPATIC TRIACYLGLYCEROL SECRETION Effects of rumen-protected choline on hepatic composition during the ad libitum and feed restriction periods 3

Table 2 .
Arshad et al.: CHOLINE AND HEPATIC TRIACYLGLYCEROL SECRETION Effects of rumen-protected choline on hepatic relative mRNA expression of transcripts affected by treatment during the ad libitum period (fold change [95% CI]) 1Choline chloride (CC) was supplemented as rumen-protected choline (RPC) with either low (L, 28.8% CC) or high concentration (H, 60.0% CC) of CC in the RPC product.The amounts supplemented were 0 (CON) or 25.8 (L25.8 and H25.8) g/d of choline ion.Results are depicted as fold change relative to CON treatment with respective 95% CI in parentheses.

Table 3 .
Arshad et al.: CHOLINE AND HEPATIC TRIACYLGLYCEROL SECRETION Effects of rumen-protected choline on hepatic relative mRNA expression of transcripts affected by treatment during the feed restriction period (fold change [95% CI]) 1Choline chloride (CC) was supplemented as rumen-protected choline (RPC) with either low (L, 28.8% CC) or high concentration (H, 60.0% CC) of CC in the RPC product.The amounts supplemented were 0 (CON) or 25.8 (L25.8 and H25.8) g/d of choline ion.Results are depicted as fold change relative to CON treatment with respective 95% CI in parentheses.

Table 4 .
Arshad et al.:CHOLINE AND HEPATIC TRIACYLGLYCEROL SECRETION Effects of rumen-protected choline on lymph composition in dairy cows 1

Table 5 .
Arshad et al.: CHOLINE AND HEPATIC TRIACYLGLYCEROL SECRETION Effects of rumen-protected choline on 20 serum metabolites (mean ± SD) based on the largest fold change (FC) in dairy cows 1 1Intravenous administration of a 10% tyloxapol solution (Triton WR-1339, 120 mg/kg of BW) to block hydrolysis of triacylglycerols in very low density lipoprotein particles.Serum metabolome was assayed 12 h after infusion of tyloxapol.