| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Animal Sciences, University of Illinois, Urbana 61801
Corresponding author:
J. K. Drackley; e-mail:
drackley{at}uiuc.edu.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
Key Words: peroxisomes • fatty acid metabolism • fat • liver
Abbreviation key: LCFA = long-chain fatty acids, NA = nicotinic acid, PPAR
= peroxisome proliferator-activated receptor-, TG = triglyceride.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
Although the largest potential impact of dietary fat on hepatic metabolism may be expected to occur during the periparturient period (Grum et al., 1996b; Drackley, 1999), the impact of dietary fat on the liver during mid- to late lactation and on the subsequent lactation also is of interest. In a short-term experiment, Grum et al. (1996a) found that supplemental fat did not affect hepatic TG concentration, despite increases in capacity for total metabolism of palmitate by liver slices. Vazquez-Añon et al. (1997) reported that supplementation of fat during mid- to late lactation did not affect hepatic TG content in mid- to late lactation but did tend to increase hepatic TG in the next lactation.
Supplementation of NA has been used in an attempt to modulate the rate of body lipid mobilization because of the reported antilipolytic effects of NA in adipose tissue (Fronk and Schultz, 1979; Jaster et al., 1983). Supplemental NA also has increased milk protein content (Horner et al., 1986). Because NA may impact provision of NEFA to the liver by affecting lipolysis and because NA is a precursor to NAD, which is a cofactor in both mitochondrial and peroxisomal ß-oxidation of NEFA in liver, NA might affect hepatic capacity for ß-oxidation of NEFA. In previous studies (Skaar et al., 1989; Minor et al., 1998), NA tended (nonsignificantly) to increase liver TG concentrations during the early postpartal period.
The objectives of this experiment were 1) to determine how capacity for total and peroxisomal ß-oxidation of palmitate changes over an entire lactation and 2) to determine the effects of supplemental fat and NA on hepatic total and peroxisomal ß-oxidation of palmitate and on hepatic concentrations of total lipid, TG, and glycogen. The experiment was part of a larger study designed to determine production responses to long-term supplementation of fat and NA in dairy cows (Drackley et al., 1998).
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
Sampling and Analysis of Liver and Blood
Samples of liver (approximately 3 to 5 g) were obtained via puncture biopsy under local anesthesia (Drackley et al., 1991) at the end of wk 3, 6, 12, 24, and 42 of lactation. A portion of the liver was frozen immediately in liquid N2 until further analyses, and the remaining portion was placed immediately in ice-cold phosphate-buffered (0.015 M) saline (0.9% NaCl) at pH 7.4 and transported to the laboratory for in vitro metabolic assays.
Capacities for total and peroxisomal first-cycle ß-oxidation of [1-14C]palmitate (American Radiolabeled Chemicals, Inc., St. Louis, MO) were measured in liver homogenates using the procedures of Veerkamp and van Moerkerk (1986) modified as described previously (Grum et al., 1994, 1996a). Briefly, 0.5 g of liver was homogenized in a 10-ml homogenization buffer (pH 7.4). ß-oxidation was measured in the absence and presence of antimycin A and rotenone (Sigma Chemical Co., St. Louis, MO) as inhibitors of mitochondrial ß-oxidation. Radiolabel incorporated into CO2 and acid-soluble products was measured by liquid scintillation spectroscopy. Capacity for total first-cycle ß-oxidation of palmitate was that measured in the absence of mitochondrial inhibitors, whereas peroxisomal ß-oxidation was that measured in the presence of the inhibitors. Triplicate flasks of each tissue treatment (blank, total ß-oxidation, peroxisomal ß-oxidation) were incubated for 30 min at 37°C.
Samples of frozen liver were analyzed for concentrations of total lipid by gravimetric methods after chloroform-methanol extraction (Drackley et al., 1991), TG by a colorimetric procedure (Foster and Dunn, 1973), and glycogen by a colorimetric procedure (Lo et al., 1970). Plasma obtained from the coccygeal vein before the a.m. feeding was analyzed for concentrations of glucose using glucose oxidase (kit number 315, Sigma Chemical Co.), NEFA using a coupled enzymatic-colorimetric procedure (Drackley et al., 1991), and BHBA using an enzymatic procedure (Cant et al., 1993).
Statistical Analysis
A total of 44 cows completed the production experiment (Drackley et al., 1998); 40 of those cows had been selected a priori for liver measurements. Liver data were available for 38 cows (10 control, 10 control + NA, 9 supplemental fat, and 9 supplemental fat + NA); liver could not be obtained from two cows. Data for liver measurements, blood metabolites, and selected production data (DMI, milk yield, SCM yield, and energy balance) for the week preceding each liver biopsy were subjected to repeated-measures ANOVA for a continuous design, using the MIXED procedure of SAS (1996). The effects of fat supplementation, addition of NA, and the interaction of fat and NA were tested using cow as the subject, with covariance structures fit using the autoregressive procedure (SAS, 1996). Cow was considered a random factor. The effects of time postpartum and the interactions of dietary treatments with time were tested using the residual error. The respective measurements made during wk 3 (before dietary treatments were initiated) were included in the model as continuous variables. Least-squares means are reported throughout. Pearson correlations among variables were computed using the entire dataset, including the wk-3 (covariate) data, and also for the wk-3 data only. Statistical significance was set at P < 0.05; comparisons with P < 0.15 are discussed as trends.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
. The interaction of the main effect of dietary fat and week was significant (P < 0.02) for DMI (Figure 1); means were greater for cows fed fat-supplemented diets at wk 6 but lower for cows fed fat-supplemented diets at wk 24 and 42. Milk yield was not affected by diet, although the difference between cows fed control and NA (32.3 vs. 33.9 kg/d, respectively) approached significance (P < 0.11). Yield of SCM was increased (P < 0.05) by NA (31.5 vs. 29.8 kg/d for NA and control, respectively). A weak statistical tendency (P < 0.12) for an interaction of fat x week was detected for SCM (36.2, 34.3, 31.4, and 18.9 kg/d for control and 37.1, 35.4, 30.8, and 21.2 kg/d for fat-supplemented diets at wk 6, 12, 24, and 42, respectively). The interaction of fat x week was significant for calculated energy balance (Figure 1), demonstrating that energy balance was greater for cows fed fat-supplemented diets at wk 6, 12, and 24 but was greater for cows fed control diets at wk 42. Similar to results for the complete data set (Drackley et al., 1998), a tendency (P < 0.09) was detected for the three-way interaction of fat x NA x week for energy balance; cows fed the fat-supplemented diet without NA had greater energy balance during early lactation, whereas cows fed the control diet with NA remained in lower energy balance for a greater length of time than did cows fed the other diets.
|
|
), similar to data from the full data set reported earlier (Drackley et al., 1998). The concentration of NEFA was lower for cows fed the fat-supplemented diet plus NA than for cows fed the fat-supplemented diet without NA in early lactation, but cows fed the fat-supplemented diet plus NA had the greater NEFA concentration later in lactation. In contrast, NEFA for cows fed the control diet plus NA were greater than for unsupplemented cows at wk 6 but then were lower than unsupplemented cows at the other weeks. The three-way interaction of fat x NA x week also was significant for BHBA in plasma (data not shown) and followed the same pattern as NEFA, although mean concentrations of BHBA were less than 8 mg/dl for all groups at all time points.
|
). Accumulation of 14C in acid-soluble products was the primary fate of oxidation of [1-14C]palmitate; negligible amounts of 14CO2 were produced in the presence of antimycin A and rotenone. The rate of total ß-oxidation differed (P < 0.02) by week postpartum, with the greatest rates measured at wk 42 (Figure 3A). Although the fat x week interaction did not reach statistical significance (P < 0.19), means for cows fed the fat-supplemented diets were slightly lower than those for control cows at wk 12 but higher than controls at wk 42. The rate of peroxisomal ß-oxidation differed (P< 0.0001) by week postpartum, with means decreasing from wk 3 to 12 and then increasing at wk 24 and 42 (Figure 3B). As a result of these changes, the ratio of peroxisomal to total oxidation (Figure 3C) also was affected by week postpartum (P < 0.02); ratios were highest at wk 3 and then decreased by wk 12 before increasing again at wk 42. The weak tendency (P < 0.12) for interaction between fat and week postpartum indicates that the ratio tended to be lower for control cows at wk 12, but higher for controls at wk 24 and 42.
|
. For the concentrations of total lipid (Figure 4A) and TG (Figure 4B) in liver, the three-way interaction of fat x NA x week was significant (P < 0.01). The three-way interactions evidently arose mainly from the differences in response to NA at wk 6. Concentrations of total lipid and TG decreased for all treatments from wk 3 to 12 as expected. The concentration of glycogen (Figure 4C) tended to be lower (P < 0.09) for cows supplemented with NA, although differences were small (5.2 vs. 4.7% of wet weight for control and NA, respectively). Hepatic glycogen was lowest for all treatments during early lactation and increased throughout the remainder of lactation (P < 0.0001).
|
). Total and peroxisomal ß-oxidation rates were highly correlated (r = 0.748). Total ß-oxidation was correlated negatively with hepatic concentrations of total lipid (r = –0.362) and TG (r = –0.324) and positively (albeit weakly) with hepatic glycogen content (r = 0.148). In contrast, peroxisomal ß-oxidation was not correlated with liver composition. Modest positive correlations existed between total ß-oxidation capacity and DMI (r = 0.209) and energy balance (r = 0.307), but total ß-oxidation was correlated negatively with yields of milk (r = –0.179) and SCM (r = –0.204). The rate of peroxisomal ß-oxidation was correlated negatively with milk (r = –0.260) and SCM (r = –0.230) but was not correlated with DMI or energy balance. The ratio of peroxisomal to total ß-oxidation was correlated negatively with DMI (r = –0.324) and energy balance (r = –0.300) but was uncorrelated with yields of milk or SCM. Weak correlations also existed between total oxidation, but not peroxisomal oxidation, and plasma glucose (r = 0.234) and NEFA (r = –0.222).
|
Because of the dynamic changes in hepatic LCFA metabolism during the periparturient period, and the larger range of hepatic total lipid and TG contents in the pretreatment liver samples obtained at wk-3 postpartum, we also computed correlation coefficients among variables for wk-3 data only (Table 3). Correlations between total ß-oxidation in liver homogenates and peroxisomal ß-oxidation (r = 0.842), the ratio of peroxisomal to total ß-oxidation (r = –0.367), liver total lipid (r = –0.517), and liver TG (r = –0.468) were all larger than corresponding correlations for the full dataset. In contrast to the full dataset, peroxisomal ß-oxidation in liver homogenates was not correlated significantly with the ratio of peroxisomal to total ß-oxidation but was correlated negatively with concentrations of total lipid (r = – 0.450) and TG (r = –0.403) in liver. Relationships among other variables at wk 3 generally were similar to those for the full dataset.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
In the earlier experiment (Grum et al., 1996b), cows fed the fat-supplemented diet consumed less dietary NEL than cows fed the other diets. Starvation causes an increase of peroxisomal ß-oxidation per unit of liver mass in rodents (Mannaerts et al., 1979; Veerkamp and van Moerkerk, 1986). However, depriving nonlactating cows of all feed for 7 d did not increase hepatic peroxisomal ß-oxidation (Grum et al., 1994). Consequently, we speculate that some combination of increased LCFA from dietary or endogenous sources and the hormonal milieu characteristic of the periparturient period is required to elicit increases in capacity for peroxisomal and total ß-oxidation in bovine liver as observed earlier (Grum et al., 1996b).
The weak tendency (P < 0.12) for an interaction of fat x week postpartum for the ratio of peroxisomal to total ß-oxidation (Figure 3C
) resulted from the trend for cows fed fat-supplemented diets to have a slightly higher ratio at wk 12 but a lower ratio at wk 24 and 42. The changes in this ratio, however, appeared to be a result of changes in capacity for total ß-oxidation (Figure 3A) rather than changes in peroxisomal ß-oxidation capacity (Figure 3B). This result is opposite to what we observed previously when a fat-supplemented diet was fed to cows throughout the dry period (Grum et al., 1996b); in that experiment, total ß-oxidation capacity and the peroxisomal-to-total ratio tended to increase as a result of significantly greater peroxisomal ß-oxidation in response to the fat-supplemented diet. Results also differ from research with rodents; for example, in rats, hepatic peroxisomal ß-oxidation was increased 150% by feeding high-fat diets (Neat et al., 1980, 1981). Neat et al. (1980) fed a control diet (5% soybean oil) or diets containing 15% fat supplied as either marine oil or soybean oil. By d 17 of feeding, hepatic peroxisomal ß-oxidation was greater in rats fed diets containing either 15% marine oil or 15% soybean oil than for controls. The amount of supplemental fat fed to cows in our experiment may not have been high enough to elicit an increase of peroxisomal ß-oxidation, as reported for rats. Alternately, the profile of LCFA reaching the intestine as a result of ruminal biohydrogenation may not have provided the appropriate LCFA to cause induction of peroxisomal ß-oxidation.
In rodents, increases of hepatic peroxisomal ß-oxidation result from increased transcription of the enzymes secondary to activation and binding of a specific lipid-activated nuclear receptor called peroxisome proliferator-activated receptor- (PPAR). When activated by binding of a lipid ligand, PPAR binds to response elements in the promoter region of target genes to activate their transcription (see Desvergne and Wahli, 1999, for review). A variety of LCFA can activate PPAR, but polyunsaturated LCFA are much more potent activators than are saturated LCFA (Kliewer et al., 1997). Because the dietary LCFA provided in our experiment were a combination of unprotected fat sources that would be extensively hydrogenated in the rumen, the additional LCFA absorbed from the intestine from the fat-supplemented diet would be mostly saturated LCFA. Consequently, this profile of absorbed LCFA may not have been effective in increasing transcription of the genes encoding enzymes for peroxisomal and mitochondrial ß-oxidation.
Although correlations cannot demonstrate cause and effect, the capacity for total ß-oxidation was correlated negatively with liver total lipid and TG contents, plasma NEFA, and yields of milk and SCM (Table 2). Correlations were positive between total ß-oxidation capacity and liver glycogen, DMI, energy balance, and plasma glucose. These correlations, although statistically significant, were relatively weak compared with those between liver lipid or TG and DMI, energy balance, and NEFA. Negative correlations between either total or peroxisomal ß-oxidation and hepatic concentrations of total lipid and TG were much stronger when only wk-3 data were considered (Table 3), similar to our previous observations during the periparturient period (Grum et al., 1996b). Whether the negative associations between hepatic total or peroxisomal ß-oxidation capacity and total lipid or TG contents indicate that increased ß-oxidation capacity is protective against hepatic lipid accumulation or that hepatic lipid accumulation results in lower total ß-oxidation capacity cannot be determined from our data. Peroxisomal ß-oxidation capacity was not correlated significantly with any production or blood variables except for negative correlations with yields of milk and SCM in the full dataset (Table 2). We suggest that our data are consistent with an increased relative importance of peroxisomal ß-oxidation during times of negative energy balance, as shown previously during the periparturient period (Grum et al., 1996b).
The trend for total and peroxisomal ß-oxidation capacities to increase by wk 42 of lactation (Figure 3) is interesting, but we can offer no explanation for its cause or significance. Aiello et al. (1984) noted that ketogenic capacity of liver slices incubated with [1-14C]palmitate decreased from d 30 to d 60 to 90 of lactation, but then increased at d 180. Oxidation of [1-14C]palmitate to CO2 was unchanged between d 30 and 60 but then increased at d 90 and 180 (Aiello et al., 1984).
Main effects of fat and NA were not significant for concentrations of total lipid and TG in liver tissue, but the three-way interaction of fat x NA x week postpartum was significant for both variables (Figure 4A and 4B). The interactions arose mainly from differences among diets at wk 6. The pattern among diets is generally similar to that for the concentration of NEFA in plasma (Figure 2). Increased NEFA concentration in plasma in response to negative energy balance is the major factor that drives TG accumulation in liver (Vazquez-Añon et al., 1994; Drackley, 1999). Although differences in energy balance among treatments were not significant in the smaller data set used here, a pattern among treatments inverse to those for liver TG and plasma NEFA was observed for the full dataset (Drackley et al., 1998), in that energy balance was lowest during early lactation for cows fed the control diet supplemented with NA. Despite these transient effects in early lactation, supplemental fat did not affect the concentrations of total lipid or TG in liver during mid- to late lactation, which agrees with results of Vazquez-Añon et al. (1997). The effect of NA on the concentration of NEFA in plasma has been variable, including decreases (Dufva et al., 1983) or no change (Skaar et al., 1989; Minor et al., 1998;). Supplementation of NA beginning prepartum (Skaar et al., 1989; Minor et al., 1998) or postpartum (Minor et al., 1998) did not significantly affect liver TG concentrations during the peripartal period, although in both reports means were slightly higher for NA-supplemented cows.
The concentration of glycogen in liver tended (P < 0.09) to be lower for cows supplemented with NA. Because NA significantly increased production of SCM but did not affect DMI, the slightly lower overall mean for hepatic glycogen may reflect the greater demand for glucose for milk synthesis. Changes in hepatic composition with stage of lactation reflect the well-characterized increases of lipid and decreases of glycogen in early lactation, followed by decreases in lipid content and increases in glycogen content in later lactation (Skaar et al., 1989; Vazquez-Añon et al., 1994, 1997; Grum et al., 1996b). The strong correlations among contents of total lipid, TG, and glycogen in liver with DMI, energy balance, and plasma NEFA highlight the importance of these factors in determining the degree of lipid infiltration and glycogen depletion observed during the periparturient period and early lactation (Vazquez-Añon et al., 1994).
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
|
ACKNOWLEDGEMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
|
FOOTNOTES |
|---|
2 Current address: Department of Animal Sciences, The Ohio State University, 312 Plumb Hall, 2027 Coffey Road, Columbus 43210.
Received for publication February 14, 2002. Accepted for publication May 16, 2002.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
and 
. Proc. Natl. Acad. Sci. USA 94:4318–4323.This article has been cited by other articles:
|
|
 |
|
  B. Kuhla, D. Albrecht, S. Kuhla, and C. C. Metges Proteome analysis of fatty liver in feed-deprived dairy cows reveals interaction of fuel sensing, calcium, fatty acid, and glycogen metabolism Physiol Genomics, April 10, 2009; 37(2): 88 - 98. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. B. Andersen, C. Ridder, and T. Larsen Priming the Cow for Mobilization in the Periparturient Period: Effects of Supplementing the Dry Cow with Saturated Fat or Linseed J Dairy Sci, March 1, 2008; 91(3): 1029 - 1043. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. N. Douglas, T. R. Overton, H. G. Bateman II, H. M. Dann, and J. K. Drackley Prepartal plane of nutrition, regardless of dietary energy source, affects periparturient metabolism and dry matter intake in Holstein cows. J Dairy Sci, June 1, 2006; 89(6): 2141 - 2157. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Bobe, J. W. Young, and D. C. Beitz Invited Review: Pathology, Etiology, Prevention, and Treatment of Fatty Liver in Dairy Cows J Dairy Sci, October 1, 2004; 87(10): 3105 - 3124. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
) or diets supplemented with fat (
). Panel a: DMI. Largest SE of the least-squares means = 0.6 kg/d. Significant effects from the model: week (P < 0.0001), fat x week (P < 0.02), fat x NA x week (P < 0.13). Panel b: calculated energy balance. Largest SE of the least squares means = 0.9 Mcal/d. Significant effects from the model: week (P < 0.0001), fat x week (P < 0.0001), fat x NA x week (P < 0.09).
), the control diet plus 12 g/d of nicotinic acid [(NA, 
)], a diet supplemented with fat (•), or the diet supplemented with fat plus 12 g/d of NA (
). Largest SE of the least squares means = 19.1 µeq/L. Significant effects from the model: fat (P < 0.006), fat x NA (P < 0.09), week (P < 0.0001), fat x NA x week (P < 0.0001).
