Changes in Hepatic Microsomal Triglyceride Transfer Protein and Triglyceride in Periparturient Dairy Cattle

Changes in Hepatic Microsomal Triglyceride Transfer Protein and Triglyceride in Periparturient Dairy Cattle D. R. Bremmer,1 S. J. Bertics, S. A. Besong,2 and R. R. Grummer Department of Dairy Science, University of Wisconsin, 1675 Observatory Drive, Madison 53706

ABSTRACT We determined the relationship between microsomal triglyceride transfer protein (MTP) (activity, mass, and mRNA) and liver triglyceride concentration in 16 dairy cows (13 multiparous and three primiparous) from 27 d before expected calving (d –27) to 35 d postpartum (d 35), the time period when fatty liver is most likely to develop. In addition, dry matter intake, plasma nonesterified fatty acids (NEFA), and plasma glucose were monitored. There were no significant parity × time interactions. Dry matter intake, plasma NEFA, plasma glucose, and liver triglyceride were significantly affected by day of sampling. Dry matter intake was 10.7, 8.0, and 19.5 kg/d on d –27, 2, and 35, respectively. Plasma NEFA concentration was higher on d 2 (1113 µEq/L) compared with d –27 (201 µEq/L) and 35 (358 µEq/L). Plasma glucose concentration was 63.3, 54.3, and 57.8 mg/dl on d –27, 2, and 35, respectively. Hepatic triglyceride (TG) concentration increased from 1.8 to 11.8% liver TG (DM basis) on d –27 and 2, respectively. There was no difference between hepatic triglyceride concentration on d 2 and 35. There was a significant effect of day of sampling on hepatic MTP activity and mRNA. Hepatic MTP activity decreased from 2.08 to 1.79 nmole triolein transferred/ h per mg of microsomal protein on d –27 and 2, respectively, and increased from 1.79 to 2.17 nmole triolein transferred/h per mg of microsomal protein on d 2 and 35, respectively. Hepatic MTP mRNA increased from d –27 to 2 and remained elevated from d 2 to 35. There was no effect of day of sampling on MTP mass. There were no significant correlations between hepatic MTP activity, mass, or mRNA with either liver TG or plasma NEFA on any of the sampling days. The cause of a decrease in hepatic MTP activity and increase in

Received January 21, 2000. Accepted April 27, 2000. Corresponding author: R. R. Grummer; e-mail: grummer@ calshp.wisc.edu. 1 Present address: Land O’Lakes Dairy Feed Division, N10768 Hwy. 73 North, Greenwood, WI 54436. 2 Present address: 1000 ASU Dr., Alcorn State University, Alcorn State, MS 39096. 2000 J Dairy Sci 83:2252–2260

mRNA on d 2 is unknown. However, the lack of correlation between MTP activity, mass, or mRNA with either liver TG or plasma NEFA on d 2 postpartum suggests that MTP probably does not play a role in the etiology of fatty liver that occurs in dairy cows at calving. (Key words: microsomal triglyceride transfer protein, parturition, liver triglyceride, fatty liver) Abbreviation key: MTP = microsomal triglyceride transfer protein, TG = triglyceride, VLDL = very low density lipoprotein. INTRODUCTION Approximately one-half of multiparous dairy cows experience moderate to severe fatty liver at calving (15). Energy intake is reduced as the cow approaches calving, a time when nutrient demands by the fetus and maternal tissues are high, resulting in an increase in fat mobilization and high plasma NEFA concentration in blood and a high rate of fatty acid uptake by the liver. Fatty liver occurs when the rate of hepatic triglyceride (TG) synthesis exceeds the rate of TG disappearance through either hydrolysis or secretion via very low-density lipoproteins (VLDL). Triglyceride accumulates in the liver of cows near parturition and remains elevated for at least 4 wk (2, 34, 36). Fatty liver has been associated with increased incidence of mastitis (18), displaced abomasum (18), retained placenta and metritis (6, 18), poor reproductive performance (9, 31), and immune suppression (30). Cows with fatty liver are more prone to develop ketosis (6) and suffer production losses (14). Rates of hepatic TG synthesis are similar between ruminants and nonruminants, but for unknown reasons, ruminants have a very slow rate of hepatic VLDL secretion relative to nonruminants (22, 28). Lower concentrations of TG-rich lipoproteins in blood have been observed in cows with naturally occurring fatty liver (29), in cows with fatty liver induced by ethionine administration (35), and in cows with fatty liver induced by starvation (5). Serum concentrations of apolipoprotein B100, the major apolipoprotein of VLDL, are lower in cows experiencing metabolic disorders (20, 35). Because of the decrease in DMI that occurs at calving (2),

2252

CHANGES IN MTP AT PARTURITION

it has been speculated that the availability of essential nutrients could be limiting VLDL assembly. However, attempts to supplement nutrients that are essential for lipoprotein assembly have been unsuccessful at reducing the amount of TG that accumulates in the liver of dairy cattle (1, 10). This indicates that something other than the availability of essential nutrients may be a limiting factor for hepatic TG-rich lipoprotein assembly and secretion in ruminants. Microsomal triglyceride transfer protein (MTP) transfers neutral lipids from outside the endoplasmic reticulum to the lumen and plays a critical role in coordinating the assembly of VLDL in the liver and chylomicrons in the intestine of nonruminants (38). Microsomal triglyceride transfer protein is a heterodimer consisting of a large subunit and a smaller subunit (protein disulfide isomerase) with molecular weights of 97,000 and 58,000, respectively (38). Microsomal triglyceride transfer protein is responsible for the early stages of apolipoprotein B containing lipoprotein assembly and not the addition of the bulk core lipids that occurs during the later stages (38). Coexpression of MTP and apolipoprotein B in cell lines that previously did not express MTP or apolipoprotein B, or secrete apolipoprotein B containing lipoproteins, resulted in secretion of apolipoprotein B containing lipoproteins (12, 28). The inactivation of MTP, before translation and translocation of apolipoprotein B 100 is complete (with specific inhibitors of MTP) prevents the secretion of apolipoprotein B 100 containing lipoproteins (32). Hepatic MTP activity may be an important factor in determining the rate of VLDL secretion in dairy cows and the severity of fatty liver at calving. To our knowledge, no studies investigating MTP in periparturient dairy cattle had been conducted before this experiment. In this experiment, we determined whether hepatic MTP activity, mass, and mRNA change in dairy cows during the periparturient period, the period when fatty liver is most likely to develop. We also determined whether there is a relationship between hepatic MTP activity, mass, and mRNA and severity of fatty liver. We hypothesized that hepatic MTP activity, mass, and mRNA would increase at calving because of an increase in TG available for VLDL synthesis. In addition, we hypothesized that lower hepatic MTP activity, mass, and mRNA would be associated with more extensive hepatic TG accumulation at calving. MATERIALS AND METHODS Management of Cows and Heifers Thirteen multiparous and three primiparous Holstein cows due to calve from June to September 1997

2253

were monitored from 28 d prepartum (28 d before expected calving) (d –28) to 35 d postpartum (d 35). On d –28, all cows were transferred to stanchions and were fed once daily at 0800 h. From the start of the experiment to d 1 postpartum, cows were fed 2.0 kg of alfalfa hay DM and 1.2 kg of dry cow grain mix DM (98.32% ground shelled corn, 0.98% trace mineral salt with Se, and 0.7% vitamin premix that contained 1,500,000 IU of vitamin A/kg of premix, 500,000 IU of vitamin D/kg of premix, and 5,000 IU of vitamin E/kg of premix), and were fed alfalfa silage ad libitum. From d 2 to 4 postpartum, cows were fed a transition diet consisting of 2.0 kg of alfalfa hay DM and, ad libitum, a TMR consisting of 75% alfalfa silage, 5% whole fuzzy cottonseed, and 20% lactating cow grain mix (61.56% ground shelled corn, 7.50% soybean meal, 18.75% roasted soybeans, 1.46% CaCO3, 1.46% Ca2PO4, 0.42% MgO, 1.25% NaHCO3, 6.25% tallow, 1.04% trace mineral salt with Se, and 0.31% vitamin premix that contained 1,500,000 IU of vitamin A/kg of premix, 500,000 IU of vitamin D/kg of premix, and 5,000 IU of vitamin E/kg of premix) on a DM basis. From d 5 to 35 postpartum, cows were fed ad libitum a TMR consisting of 45% alfalfa silage, 7% whole fuzzy cottonseed, and 48% lactating cow grain mix on a DM basis. Dry matter content of forages was determined weekly at 100°C to adjust as-fed ratios of forage and concentrate. Feed offered and refused was recorded daily from d –28 prepartum to d 35 postpartum. After parturition cows were milked twice daily at 0400 and 1600 h. Sampling Blood samples were collected at 1130 h from the coccygeal vein on d –28, –27, 1, 2, 34, and 35 relative to day of calving. Blood was collected in vacutainers containing potassium oxalate and sodium fluoride and put on ice. Plasma was obtained after centrifugation at 850 × g for 15 min at 4°C. Aliquots were frozen at – 20°C until analyzed for NEFA [NEFA-C kit; Wako Fine Chemical Industries USA, Inc., Dallas TX; as modified by Johnson and Peters (21)] and glucose [Sigma procedure no. 510, Sigma Chemical Co., St. Louis, MO; based on the procedure described by Raabo and Terkildsen (30)]. Before analysis for glucose and NEFA, plasma samples from d –28 and –27, d 1 and 2, and d 34 and 35 were pooled so that there were three time points designated as d –27, 2, and 35. Liver biopsies were performed on d –27, 2, and 35 relative to day of calving as described by VazquezAnon et al. (38). Immediately after collection, a portion of the liver sample was rinsed in ice-cold 0.25 M sucrose, frozen in liquid nitrogen, and stored at –80°C until analyzed for MTP activity and mass. A second Journal of Dairy Science Vol. 83, No. 10, 2000

2254

BREMMER ET AL.

portion of the liver sample was rinsed in RNase-free water (diethyl-pyrocarbonate-treated water), frozen in liquid nitrogen, and stored at –80°C until analyzed for MTP mRNA. A third portion of the liver sample was rinsed in saline, frozen in liquid nitrogen, and stored at –20°C until analyzed for TG (36).

MTP Activity and Large Subunit Mass Microsomal triglyceride transfer protein activity was determined as described by Bremmer et al. (3). Microsomal triglyceride transfer protein large subunit protein mass (designated as MTP mass throughout the manuscript) was determined by Western blot analysis as modified by Lin et al. (25). After thawing on ice, liver was minced and a 0.1-g aliquot was homogenized in 1 ml of dissociation buffer (2.0 M NaCl, 0.05 M Na2HPO4, 2.0 mM Na2EDTAⴢ2H20, pH 7.4) with an IKA-Labortechnik tissue homogenizer for 20 s at 20,000 rpm. After homogenization, an additional 4 ml of dissociation buffer was used to rinse the homogenizer knives and homogenization tubes so that a final homogenate volume of 5 ml was obtained. The liver homogenates were assayed for protein (BCA Protein Assay, Pierce Chemical Co., Rockford, IL) before being frozen at –80°C. After thawing on ice, liver homogenates (30 µg of protein) and purified bovine MTP standard (gift from John R. Wetterau, Department of Metabolic Diseases, Bristol-Myers Squibb, Princeton, NJ) were subjected to 7.5% SDS-polyacrylamide gel electrophoresis followed by transfer to Immun-Blot PVDF Membrane (Bio-Rad Laboratories, Hercules, CA) in a trans-blot cell. The Immun-Blot PVDF Membrane was blocked overnight with a nonfat milk solution. The following day, the Immun-Blot PVDF Membrane was incubated for 1 h at 37°C with an anti-bovine MTP large subunit polyclonal antibody (gift from John R. Wetterau, Department of Metabolic Diseases, BristolMyers Squibb, Princeton, NJ) followed by incubation with horseradish peroxidase conjugated donkey antigoat IgG secondary antibody (Santa Cruz Biotechnology, Inc., Santa Cruz, CA). Bands that corresponded to the MTP large subunit were visualized by chemiluminescence (Santa Cruz Biotechnology, Inc.) using autoradiography followed by quantification using Collage Image Analysis Software (Fotodyne, Inc., Hartland WI). The relative MTP large subunit protein mass (MTP mass) was normalized against the purified bovine MTP standard on each membrane and expressed as micrograms of MTP large subunit per milligram of hepatic protein. Journal of Dairy Science Vol. 83, No. 10, 2000

Extraction of Total RNA from Liver Tissue and Transfer to Membranes Total RNA was isolated by homogenizing biopsies (50 to 100 mg of liver tissue) in 1 ml of phenol and guanidinium thiocyanate solution (RNA ISOLATOR, Genosys Biotechnologies, Inc., The Woodlands, TX) following a modification of the method of Chomczynski and Sacchi (7). The final RNA pellet was dissolved in RNase-free water (diethyl-pyrocarbonate-treated water) and quantified by measuring the absorbance at 260 and 280 nm and stored at –80°C. The purity and integrity of the RNA (10 µg of total RNA/lane) was verified on a 1.2% agarose, 2.2% formaldehyde gel electrophoresis. After photographing the ethidium bromide stained RNA, the gel was soaked in 6 × SSC (1 × SSC is 0.15 M NaCl plus 0.015 M sodium citrate) for 45 min before capillary transfer to a Hybond-N Membrane (Schleicher and Schuell, Keene, NH) overnight in 20 × SSC. After transfer, the membrane was soaked in 3 × SSC for 10 min, allowed to air dry for 10 min and then crosslinked using UV light (Gene Linker UV Chamber, BioRad, Hercules, CA) to fix RNA on the membrane. Synthesis of RNA Probe and Northern Blot Analysis of Bovine MTP mRNA Microsomal triglyceride transfer protein mRNA was quantitated by northern blot analysis. Polymerase chain reaction was used to amplify a 265-bp MTP cDNA fragment corresponding to nucleotides 1799 to 2064 of the bovine cDNA sequence, using two synthetic oligonucleotide primers, 5′-CCGTTTCTCCAGGAGTGGATC-3′ and 5′-GGCTGACATACCAGCATAGGAG3′ from the bovine gene sequence. The PCR amplification was carried out for 30 cycles of denaturation at 94°C for 60 s, annealing at 60°C for 60 s, and extension at 72°C for 60 s. The total PCR reaction mixture of 100 µl contained 2.5 units of Taq DNA polymerase, 50 mM KCl, 20 mM Tris-HCl (pH 8.4), 1.5 mM MgCl2, 1 µM each of forward and reverse primer, 1 µg of cDNA, and 200 µM each of dATP, dCTP, dGTP, and dTTP. To check for specificity, PCR products were analyzed by agarose gel electrophoresis. The PCR product was ligated with a T7 adaptor using a ligation kit (Lig’nScribe In Vitro Transcription Kit, Ambion, Austin, TX). The ligated PCR products were analyzed by agarose gel electrophoresis and used to synthesize a 32P-RNA probe (radiolabel MTP probe) using a Maxiscript T7 transcription kit (Ambion) according to the manufacturer’s instructions. Membranes were first probed for MTP mRNA by hybridizing overnight at 50°C in 50% formamide, 6 × SSC, 5 × Denhardt’s, 1% SDS, 0.1% EDTA, and 0.1 mg/ml of denatured salmon sperm

2255

CHANGES IN MTP AT PARTURITION

DNA. Washes were performed at 65°C in 1 × SCC, 0.5% SDS. Membranes were quantitated by phosphoimager analysis (BioRad Laboratories). Membranes were then stripped and hybridized again with glyceridealdehyde3-phosphate dehydrogenase cDNA probe labeled by random priming (Ready-to-Go DNA labeling kit, Pharmacia, Piscataway, NJ). Conditions for hybridization were similar to above, although the temperature was adjusted to 42°C. Microsomal triglyceride transfer protein mRNA values were normalized to the corresponding glyceridealdehyde-3-phosphate dehydrogenase mRNA levels. Data for d –27, 2, and 35 were expressed relative to d –27, therefore, MTP mRNA densitometry levels for all cows on d –27 are equal to 1, resulting in data that are not normally distributed. Statistical Analysis Before statistical analysis of MTP mRNA, all MTP mRNA densitometry values (d –27, 2, and 35) were treated as distribution free and ranked from smallest to largest using the nonparametric rank procedure of SAS (33). Data for all variables measured were analyzed as repeated measures using the mixed procedure of SAS (33) according to the following model: Yij = µ + Pi + Dj + PDij + eij where = = = =

dependent variable overall mean of the population mean effect of parity (i = 1 or 2) mean effect of day of sampling (j = d –27, 2, or 35) eij = unexplained residual element assumed to be independent and normally distributed.

Yij µ Pi Dj

Data were analyzed across all sampling days (d –27, 2, and 35) relative to day of calving with d 0 representing the day of calving. When day of sampling was determined to be significant, the Bonferroni multiple comparisons test was used to determine differences among lsmeans. Least squares means are reported in the tables. Correlations within each sampling day and over all sampling days combined were determined by the CORR procedure of SAS (33). Significance was declared at P < 0.05 and trends at P < 0.10. Three cows were removed from the experiment after d 2 postpartum. One cow was accidentally given a treatment being used in an experiment conducted in the barn at the same time as this experiment. One cow freshened with only three quarters of the mammary gland producing milk. One cow had multiple displaced abomasum sur-

geries after d 2 postpartum and was unable to continue the experiment. Data for these three cows from d –27 and 2 were included in the data set; however, data were not collected for these three cows on d 35 postpartum. Therefore, 13 multiparous and three primiparous cows were included in the data set for d –27 and 2 and 10 multiparous and three primiparous cows were included in the data set for d 35. There were no significant parity × day of sampling interactions; therefore, figures and tables represent pooled data from cows and heifers and the final model used was: Yij = µ + Pi + Dj + eij where Yij µ Pi Dj

= = = =

dependent variable overall mean of the population mean effect of parity (i = 1 or 2) mean effect of day of sampling (j = d –27, 2, or 35) eij = unexplained residual element assumed to be independent and normally distributed.

Because data for multiparous and primiparous cows were pooled, the term cow is used to represent both multiparous and primiparous cows. RESULTS AND DISCUSSION Dry Matter Intake, Plasma Metabolites, and Liver Triglyceride Dry matter intake, plasma NEFA and glucose, and liver TG (Table 1) were significantly affected by day of sampling (P < 0.0001). Dry matter intake decreased (P < 0.008) from 10.7 to 8.0 kg/d on d –27 and d 2, respectively (Table 1). Dry matter intake increased (P < 0.0001) from 8.0 to 19.5 kg/d on d 2 and d 35, respectively (Table 1). A decrease in DMI at calving has been reported in other experiments (2, 4, 36). The duration and the extent of prepartum DMI depression, may be an important factor in the development of fatty liver in the periparturient dairy cow (14). Dry matter intake is typically inversely related to adipose tissue mobilization, as indicated by an increase in plasma NEFA concentration during the final weeks before calving (2). Plasma NEFA concentration was higher on d 2 (1113 µEq/L) than on d –27 (201 µEq/L; P < 0.0001) and d 35 (358 µEq/L; P < 0.0005) (Table 1). Peak plasma NEFA concentration occurs 1 d postpartum, coinciding with the most severe depression in DMI (36). Plasma glucose concentration decreased (P < 0.0001) from 63.3 to 54.3 mg/dl on d –27 and 2, Journal of Dairy Science Vol. 83, No. 10, 2000

2256

BREMMER ET AL. Table 1. Changes in DMI, plasma NEFA, plasma glucose, and liver triglyceride (TG) concentration with day of sampling. Contrasts1 Day of sampling d −27 Variable measured 2

DMI (kg/d) Plasma NEFA (µEq/L)2 Plasma glucose (mg/dl)2 Liver TG (% DM basis)2

d2

d −27 vs. d2

d 35

X

SE

X

SE

X

SE

10.7 201 63.3 1.8

0.9 131 2.0 1.0

8.0 1113 54.3 11.8

0.9 131 2.0 1.0

19.5 358 57.8 10.2

0.9 140 2.1 3.6

d −27 vs. d 35

d 2 vs. d 35

P< 0.008 0.0001 0.0001 0.0001

0.0001 NS3 0.02 0.05

0.0001 0.0005 NS3 NS3

1 Contrasts were made among sampling days using the Bonferroni multiple comparisons test after determining that effect of sampling day was significant. 2 Significant effect of sampling day, P < 0.0001. 3 Nonsignificant.

respectively (Table 1). Plasma glucose concentration was higher (P < 0.02) on d –27 than on d 35 (63.3 vs. 57.8 mg/dl, respectively) (Table 1). There was no difference in plasma glucose concentration on d 2 and 35 (Table 1). On the day before calving, plasma glucose concentration increases sharply to peak at approximately 85 mg/dl (36). The peak in plasma glucose concentration represents a change in hormonal status to promote glycogenolysis and gluconeogenesis (19). During the first week postpartum, plasma glucose concentration decreases 25% (36), reflecting an increase in milk production and a drain of plasma glucose during the first week postpartum. After the first week postpartum until approximately 4 wk postpartum, plasma glucose has been shown to remain unchanged (2) or increase slightly because of a recovery of feed intake and improvement in energy balance (36). Hepatic TG concentration increased (P < 0.0001) from 1.8 to 11.8% liver TG (DM basis) on d –27 and 2, respectively (Table 1). There was no difference in hepatic TG concentration on d 2 and 35 (Table 1). Hepatic TG increases in cows near parturition and remains elevated for at least 4 wk (2, 34, 36). Similar to Vazquez-Anon et al. (38), postpartum hepatic TG concentration in this study was lower than those reported in previous experiments (2, 34). Vazquez-Anon et al. (38) attributed a low postpartum hepatic TG concentration to a short duration of feed intake depression before calving. In this experiment, the low concentration of hepatic TG is probably attributed to the inclusion of three primiparous cows in the data set. The mean hepatic TG concentration over all days (d – 27, 2, and 35) for multiparous cows was higher (P < 0.009) compared with primiparous cows (9.9 vs. 5.9% TG, DM basis). Other experiments have also reported low hepatic TG concentrations for primiparous compared to multiparous cows (4, 16). Journal of Dairy Science Vol. 83, No. 10, 2000

Hepatic MTP Activity, Mass, and mRNA There was a significant (P < 0.04) effect of day of sampling on hepatic MTP activity (Table 2). Hepatic MTP activity decreased (P < 0.06) from 2.08 to 1.79 nmole triolein transferred/h per mg of microsomal protein on d –27 and 2, respectively (Table 2). Hepatic MTP activity increased (P < 0.02) from 1.79 to 2.17 nmole triolein transferred/h per mg of microsomal protein on d 2 and 35, respectively (Table 2). The changes in hepatic MTP activity corresponded to a 14% decrease from d –27 to 2 and a 21% increase from d 2 to 35 (Table 2). It is tempting to hypothesize that the increase in hepatic TG concentration that occurred on d 2 was a direct result of a decrease in hepatic MTP activity after calving. However, there was no correlation between hepatic MTP activity and liver TG or plasma NEFA on any of the individual sampling days (Table 3). There was a trend for a significant (P < 0.07) negative correlation between hepatic MTP activity and plasma NEFA concentrations over all sampling days combined (Table 3), however, the correlation coefficient was low (r = –0.27). It should also be pointed out that although there was an increase in hepatic MTP activity from d 2 to 35 (Table 2), there was no change in liver TG from d 2 to 35 (Table 1). Therefore, the increase in hepatic MTP activity did not correspond to the depletion of hepatic TG. There was no effect (P < 0.17) of day of sampling on hepatic MTP mass (Table 2). There was an effect (P < 0.04) of day of sampling on hepatic MTP mRNA rank (Table 2). Mean hepatic MTP mRNA rank increased (P < 0.04) from 17.0 to 26.2 on d –27 and 2, respectively (Table 2). Mean hepatic MTP mRNA rank on d 35 was not different than d –27 or 2 (Table 2). This data conflicts with the decrease in hepatic MTP activity observed on d 2 (Table 2). It has been reported that decreases in MTP mRNA do not acutely regulate MTP

2257

CHANGES IN MTP AT PARTURITION Table 2. Changes in hepatic microsomal triglyceride transfer protein (MTP) activity, hepatic MTP mass, and hepatic MTP mRNA with day of sampling. Contrasts1 Day of sampling d −27 Variable measured MTP MTP MTP MTP

2,3

activity mass5 mRNA7,8 mRNA rank2,9

d2

d − 27 vs. d2

d 35

X

SE

X

SE

X

SE

2.08 6.83 1.00 17.0

0.15 1.27 0.00 3.54

1.79 7.86 1.39 26.2

0.15 1.27 0.15 3.54

2.17 7.78 1.11 24.3

0.15 1.28 0.12 3.81

d − 27 vs. d 35

d 2 vs. d 35

P< 0.06 ND6

NS4 ND6

0.02 ND6

0.04

NS4

NS4

1 Contrasts were made among sampling days using the Bonferroni multiple comparisons test after determining that effect of sampling day was significant. 2 Significant effect of sampling day, P < 0.04. 3 nmole triolein transferred/h per mg of microsomal protein. 4 Nonsignificant. 5 µg MTP large subunit/mg of hepatic protein. 6 Contrasts among days of sampling were not made because the effect of sampling day was not significant. 7 MTP mRNA densitometry values calculated from the ratio of densitometry readings for MTP mRNA glyceraldehyde-3-phosphate dehydrogenase mRNA. Means for MTP mRNA densitometry values were determined by the means procedure of SAS (33). 8 Statistical analysis was not performed on MTP mRNA densitometry values because of the lack of normal distribution after calculating the ratio of densitometry readings for MTP mRNA:glyceraldehyde-3-phosphate dehydrogenase mRNA. 9 Least squares means of ranks for MTP mRNA densitometry values.

protein levels; half-life for the MTP large subunit in HepG2 cells was 4.4 d (25). However, an observed decrease in hepatic MTP activity would be expected to coincide with a decrease in hepatic MTP mass. The lack of an effect of day of sampling on MTP mass further questions the biological significance of the small (14%) decrease in hepatic MTP activity observed on d 2 (Table 2). Because the half-life of MTP is 4.4 d, it is possible that the increase in hepatic MTP mRNA observed on d 2 may not have had sufficient time to increase levels of MTP activity and mass. However,

hepatic MTP mRNA rank was numerically higher on d 35 compared to d –27, which would have been sufficient time for an increase to be reflected in MTP activity and mass. In addition, the lack of change in liver TG from d 2 to 35 (Table 1) questions the biological significance of the elevated level of hepatic MTP mRNA postpartum. There was no significant correlation between hepatic MTP mass or mRNA and liver TG or plasma NEFA on any of the sampling days or over all sampling days combined (Table 3). There was a trend for a significant

Table 3. Pearson’s correlation coefficients for linear associations between plasma NEFA or liver triglyceride (TG) and hepatic microsomal triglyceride transfer protein (MTP) activity, hepatic MTP mass, or hepatic MTP mRNA on d −27, 2, 35, or over all sampling days combined. NEFA1 vs. MTP Activity3 Day of sampling

r

d − 27 d2 d 35 Over all days

−0.27 −0.14 0.13 −0.27

NEFA vs. MTP Mass4

P<

r

0.32 0.61 0.67 0.07

−0.10 −0.45 −0.07 −0.13

P< 0.72 0.09 0.83 0.39

Liver TG2 vs. MTP Activity

NEFA vs. MTP mRNA5 P<

r 6

ND −0.03 −0.16 0.16

r 6

ND 0.92 0.63 0.31

−0.24 0.09 0.09 −0.04

Liver TG vs. MTP Mass

P<

t

0.37 0.75 0.77 0.77

−0.13 0.15 0.28 0.20

P< 0.64 0.61 0.35 0.19

Liver TG vs. MTP mRNA P<

r 6

ND −0.04 −0.57 −0.06

ND6 0.91 0.06 0.72

1

µEq/L. % DM basis. 3 nmole triolein transferred/h per mg of microsomal protein. 4 µg of MTP large subunit/mg of hepatic protein. 5 Ranks for MTP mRNA densitometry values. Ranks were created by the rank procedure of SAS (1996) to normalize MTP mRNA densitometry values. 6 All MTP mRNA densitometry values were equal to 1 on d −27, therefore, all MTP mRNA ranks were equal on d −27 and correlations were not analyzed with MTP mRNA on d −27. 2

Journal of Dairy Science Vol. 83, No. 10, 2000

2258

BREMMER ET AL.

(P < 0.09) negative correlation between hepatic MTP mass and plasma NEFA concentrations on d 2 (Table 3). There was a trend (P < 0.06) for a negative correlation between hepatic MTP mRNA and liver TG concentrations on d 35 (Table 3). For our analysis, 24 correlation coefficients were determined. Therefore, by chance, one would expect 2.4 correlations to be declared significant or a trend with P < 0.10 as a criterion. Therefore, there is probably little biological significance in the trends for significant correlations found in this experiment. We hypothesized that hepatic MTP activity, mass, and mRNA would increase at calving because of an increase in substrate available for VLDL synthesis and that lower hepatic MTP activity, mass, and mRNA would be associated with more extensive TG accumulation at calving. The increase in hepatic MTP mRNA observed on d 2 compared with d –27 was not reflected in an increase in hepatic MTP activity and mass (Table 2) or a decrease in liver TG (Table 1), suggesting that during the transition period, there may be minimal regulation of MTP activity and mass. It is difficult to explain why hepatic MTP activity might decrease on d 2 (Table 2). In a study with hamsters (24), no effect of a 2-d fast on hepatic MTP mRNA was reported. Therefore, the low DMI observed on d 2 may be expected to cause the decrease in hepatic MTP activity (Table 2). Evidence indicates there is a surge of insulin in blood (approximately a threefold increase) on the day before calving (23). When HepG2 cells were incubated with 10 nM insulin (normal levels found in humans) for 6 h, MTP mRNA levels decreased 30% compared with control levels (25). After a 24-h incubation with insulin (1000 nM), there was no effect on MTP activity in HepG2 cells (25). The lack of an effect on MTP activity is probably related to the relatively long half-life (4.4 d) for MTP, as determined in HepG2 cells (25). Therefore, in our experiment, a surge of insulin that may have occurred on the day of calving would probably not have had sufficient time to exert negative effects on hepatic MTP activity. Hepatic MTP mRNA levels were increased on d 2 compared with d – 27 (Table 2), suggesting that the surge in insulin does not have short term negative effects on hepatic MTP after calving. Plasma glucagon concentration typically increases from 155 pg/ml during the dry period to 187 pg/ml during early lactation (8). Lin et al. (25) reported that there is no effect of glucagon on MTP mRNA or MTP activity in HepG2 cells. The effects of other lipolytic hormones and growth hormone on MTP have not been studied. Therefore, the decrease in hepatic MTP activity we report in this experiment on d 2 (Table 2) cannot be explained at this time. Regardless of the explanation for the decrease in hepatic MTP activity on d 2, Journal of Dairy Science Vol. 83, No. 10, 2000

the lack of correlation with liver TG or plasma NEFA on d 2 (Table 3) and the lack of a day of sampling effect on MTP mass (Table 2) makes one question whether the decrease has any biological significance in the etiology of fatty liver in the dairy cow at calving. The expression of hepatic apolipoprotein B 100 mRNA and concentrations of hepatic apolipoprotein B 100 have been determined in cows during the dry period and during a period of underfeeding in early lactation (13). Dot blot analysis revealed significantly lower (approximately 33% lower) levels of apolipoprotein B 100 mRNA at wk 1 of lactation compared with nonlactating pregnant cows (13). During wk 1 of lactation, hepatic apolipoprotein B 100 concentrations were 70% lower compared to concentrations in dry pregnant cows (13). Research indicates that a reduction in serum apolipoprotein B concentrations occurs in cows around the time of calving (20, 21) and in cows experiencing fatty liver (20, 26). Inactivation of MTP in cell culture experiments before translation and translocation of apolipoprotein B prevents the secretion of apolipoprotein B containing lipoproteins (11) and results in the degradation of apolipoprotein B (17). Inactivation of MTP after translation and translocation of apolipoprotein B into the lumen of the endoplasmic reticulum but before the addition of the bulk core lipids to the lipoprotein particle has no effect on secretion of TGrich apolipoprotein B containing lipoproteins (11), suggesting that MTP prevents the degradation of apolipoprotein B during the early stages of synthesis of apolipoprotein B containing lipoproteins. Rates of synthesis and degradation of hepatic apolipoprotein B 100 were not determined by Gruffat et al. (13); however, it appears that either upregulation of hepatic apolipoprotein B 100 occurs during different metabolic states of the dairy cow or an increase in degradation of hepatic apolipoprotein B 100 occurs during the first week after calving (13). A possible component affecting this situation is MTP. A significant effect of day of sampling on MTP mass was not observed in this experiment (Table 2). However, the decrease in MTP activity we report in this experiment on d 2 (Table 2) may have resulted in insufficient levels of hepatic MTP activity for the early stages of lipoprotein assembly to prevent degradation of hepatic apolipoprotein B 100 during the first week after calving (13). Degradation of hepatic apolipoprotein B 100 induced by insufficient hepatic MTP activity after calving could lead to reductions in serum apolipoprotein B concentrations that have been observed around the time of calving (20, 21). CONCLUSIONS To our knowledge this is the first experiment to study MTP in periparturient dairy cattle. In this ex-

CHANGES IN MTP AT PARTURITION

periment, we determined whether hepatic MTP activity, mass, and mRNA change in dairy cows during the time period when fatty liver is most likely to develop. We also determined whether there is a relationship between hepatic MTP and severity of fatty liver. We hypothesized that hepatic MTP activity, mass, and mRNA would increase at calving because of an increase in TG available for VLDL synthesis. Results did not support the hypothesis. Microsomal triglyceride protein probably does not play a role in the etiology of fatty liver that occurs in dairy cows at calving. ACKNOWLEDGMENTS Bob Elderbrook helped with feeding of cattle. Appreciation is also given to Peter Crump for statistical advice. We also thank John R. Wetterau (Department of Metabolic Diseases, Bristol-Myers Squibb, Princeton, NJ) for supplying purified bovine MTP and anti-bovine MTP large subunit polyclonal antibody for MTP mass determinations. Appreciation is expressed to Paul Bertics (Department of Biomolecular Chemistry, UW-Madison) for use of his autoradiography equipment. Appreciation is also expressed to Mary Grummer (Department of Pediatrics, University of Wisconsin-Meriter Perinatal Center) for assistance with mRNA analysis. REFERENCES 1 Bertics, S. J., and R. R. Grummer. 1999. Effects of fat and methionine hydroxy analog on prevention or alleviation of fatty liver induced by feed restriction. J. Dairy Sci. 82:2731–2736. 2 Bertics, S. J., R. R. Grummer, C. Cadorniga-Valino, and E. E. Stoddard. 1992. Effect of prepartum dry matter intake on liver triglyceride concentration and early lactation. J. Dairy Sci. 75:1914–1922. 3 Bremmer, D. R., S. J. Bertics, and R. R. Grummer. 1999. Differences in activity of hepatic microsomal triglyceride transfer protein among species. Comp. Bioch. Physiol. 124:123–131. 4 Bremmer, D. R., J. O. Christensen, R. R. Grummer, F. E. Rasmussen, and M. C. Wiltbank. 1999. Effects of induced parturition and estradiol on feed intake, liver triglyceride concentration, and plasma metabolites of transition dairy cows. J. Dairy Sci. 82:1440–1448. 5 Brumby, P. E., M. Anderson, B. Tuckley, J. E. Storry, and K. G. Hibbitt. 1975. Lipid metabolism in the cow during starvationinduced ketosis. Biochem. J. 146:609–615. 6 Bruss, M. L. 1993. Metabolic fatty liver of ruminants. Adv. Vet. Sci. Comm. 37:417–449. 7 Chomczynski, P., and N. Sacchi. 1987. Single-step method RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal. Biochem. 162:156–159. 8 de Boer, G., A. Trenkle, and J. W. Young. 1985. Glucagon, insulin, growth hormone, and some blood metabolites during energy restriction ketonemia of lactating cows. J. Dairy Sci. 68:326–337. 9 Gerloff, B. J., T. H. Herdt, and R. S. Emery. 1986. Relationship of hepatic lipidosis to health and performance in dairy cattle. J. Am. Vet. Med. Assoc. 188:845–850. 10 Gerloff, B. J., T. H. Herdt, W. W. Wells, J. S. Liesman, and R. S. Emery. 1986. Inositol and lipidosis. I. Effect of inositol supplementation and time from parturition on liver and serum lipids in dairy cattle. J. Anim. Sci. 62:1682–1692.

2259

11 Gordon, D. A., H. Jamil, R. E. Gregg, S. Olofsson, and J. Boren. 1996. Inhibition of the microsomal triglyceride transfer protein blocks the first step of apolipoprotein B lipoprotein assembly but not the addition of bulk core lipids in the second step. J. Biol. Chem. 271:33047–33053. 12 Gordon, D. A., H. Jamil, D. Sharp, D. Mullaney, Zemin Yao, R. E. Gregg, and J. R. Wetterau. 1994. Secretion of apolipoprotein B-containing lipoproteins from HeLa cells is dependent on the expression of the microsomal triglyceride transfer protein and is regulated by lipid availability. Proc. Natl. Acad. Sci. 91:7628–7632. 13 Gruffat, D., D. Durand, Y. Chilliard, P. Williams, and D. Bauchart. 1997. Hepatic gene expression of apolipoprotein B100 during early lactation in underfed, high producing dairy cows. J. Dairy Sci. 80:657–666. 14 Grummer, R. R. 1993. Etiology of lipid-related metabolic disorders in periparturient dairy cows. J. Dairy Sci. 76:3882–3896. 15 Grummer, R. R. 1995. Impact of changes in organic nutrient metabolism on feeding the transition dairy cow. J. Anim. Sci. 73:2820–2833. 16 Grummer, R. R., P. C. Hoffman, M. L. Luck, and S. J. Bertics. 1995. Effect of prepartum and postpartum dietary energy on growth and lactation of primiparous cows. J. Dairy Sci. 78:172–180. 17 Haghpassand, M., D. Wilder, and J. B. Moberly. 1996. Inhibition of apolipoprotein B and triglyceride secretion in human hepatoma cells (HepG2). J. Lipid Res. 37:1468–1480. 18 Herdt, T. H. 1988. Fatty liver in dairy cows. Vet. Clin. North Am. Food Anim. Pract. 4:269–287. 19 Herdt, T. H. 1988. Fuel homeostasis in the ruminant. Metabolic diseases of ruminant livestock. Vet. Clin. North Am. Food Anim. Pract. 4:213–231. 20 Itoh, H., K. Tamura, Y. Motoi, and F. Kawawa. 1997. Serum apolipoprotein B-100 concentrations in healthy and diseased cattle. J. Vet. Med. Sci. 59:587–591. 21 Johnson, M. M., and J. P. Peters. 1993. Technical note: An improved method to quantify nonesterified fatty acids in bovine plasma. J. Anim. Sci. 71:753–756. 22 Katoh, N., S. Minoura, E. Uchida, and K. Takahashi. 1993. Effect of estradiol administration and subsequent nonfeeding on liver estrogen receptor, serum apolipoprotein B-100, and serum triglyceride concentrations in steers. Am. J. Vet. Res. 54:1476– 1482. 23 Kleppe, B. B., R. J. Aiello, R. R. Grummer, and L. E. Armentano. 1988. Triglyceride accumulation and very low density lipoprotein secretion by rat and goat hepatocytes in vitro. J. Dairy Sci. 71:1813–1822. 24 Kunz, P. L., J. W. Blum, I. C. Hart, H. Bickel, and J. Landis. 1985. Effects of different energy intakes before and after calving on food intake, performance and blood hormones and metabolites in dairy cows. Anim. Prod. 40:219–231. 25 Lin, M. C. M., C. Arbeeny, K. Bergquist, B. Kienzle, D. A. Gordon, and J. R. Wetterau. 1994. Cloning and regulation of hamster microsomal triglyceride transfer protein. J. Biol. Chem. 269:29138–29145. 26 Lin, M.C.M., D. Gordon, and J. R. Wetterau. 1995. Microsomal triglyceride transfer protein (MTP) regulation in HepG2 cells: insulin negatively regulates MTP gene expression. J. Lipid Res. 36:1073–1081. 27 Mazur, A., M. Ayrault-Jarrier, Y. Chilliard, and Y. Rayssiguier. 1992. Lipoprotein metabolism in fatty liver dairy cows. Diabete Metab. 18:145–149. 28 Patel, S. B., and S. M. Grundy. 1996. Interactions between microsomal triglyceride transfer protein and apolipoprotein B within the endoplasmic reticulum in a heterologous expression system. J. Biol. Chem. 271:18686–18694. 29 Pullen, D. L., J. S. Liesman, and R. S. Emery. 1990. A species comparison of liver slice synthesis and secretion of triacylglycerol from nonesterified fatty acids in media. J. Anim. Sci. 68:1395–1399. 30 Raabo, E., and T. C. Terkildsen. 1960. On the enzymatic determination of blood glucose. Scand. J. Clin. Lab. Invest. 12:402–407. Journal of Dairy Science Vol. 83, No. 10, 2000

2260

BREMMER ET AL.

31 Rayssiguier, Y., A. Mazur, E. Gueux, I. M. Reid, and C. J. Roberts. 1988. Plasma lipoproteins and fatty liver in dairy cows. Res. Vet. Sci. 45:389–393. 32 Reid, I. M., A. M. Dew, and L. A. Williams. 1984. Haematology of subclinical fatty liver in dairy cows. Res. Vet. Sci. 37:63–65. 33 Reid, I. M., C. J. Roberts, and R. Manston. 1979. Fatty liver and infertility in high-yielding dairy cows. Vet. Rec. 104:75–76. 34 Rustaeus, S., P. Stillemark, K. Lindberg, D. Gordon, and S. Olofsson. 1998. The microsomal triglyceride transfer protein catalyzes the post-translational assembly of apolipoprotein B100 very low density lipoprotein in McA-RH7777 cells. J. Biol. Chem. 273:5196–5203. 35 SAS User’s Guide: Statistics, Version 7 Edition, 1999. SAS Inst., Inc., Cary NC.

Journal of Dairy Science Vol. 83, No. 10, 2000

36 Skaar, T. C., R. R. Grummer, M. R. Dentine, and R. H. Stauffacher. 1989. Seasonal effect of prepartum and postpartum fat and niacin feeding on lactation performance and lipid metabolism. J. Dairy Sci. 72:2028–2038. 37 Uchida, E., N. Katoh, and K. Takahashshi. 1992. Induction of fatty liver in cows by ethionine administration and concomitant decreases of serum apolipoproteins B-100 and A-I concentrations. Am. J. Vet. Res. 53:2035–2042. 38 Vazquez-Anon, M., S. J. Bertics, M. Luck, R. R. Grummer, and J. Pinheiro. 1994. Peripartum liver triglyceride and plasma metabolites in dairy cows. J. Dairy Sci. 77:1521–1528. 39 Wetterau, J. R., M.C.M. Lin, and H. Jamil. 1997. Microsomal triglyceride transfer protein. Biochim. Biophys. Acta 1345:136–150.