AJVR, Vol 70, No. 5, May 2009 563

During early lactation, high-producing dairy cows are in a negative energy balance. Because of this, large

amounts of NEFAs from adipose tissue are mobilized to the liver. When the excessive influx exceeds the metabolic capacity of the liver, the overwhelming accumulation of triglycerides causes the development of hepatic lipidosis (ie, fat cow syndrome).1–8 Hepatic lipidosis is associated with several postparturient diseases, including ketosis, left displacement of the abomasum, retained placenta, mastitis, postparturient hypocalcemia, and downer cow syndrome. The effective control of such diseases is crucially important for modern dairy husbandry because affected cows, even after recovery from those metabolic diseases, have a high risk for reproductive and infectious diseases.3,7,9–15

Evaluation of changes in serum lipoprotein and apolipoprotein patterns in cows with ketosis

Tülay İleri-Büyükoğlu, DVM, PhD; Şima Şahinduran, DVM, PhD; Kenan Sezer, DVM, PhD; Tayfun Güldür, DVM, PhD

Objective—To detect changes in serum lipoprotein and apolipoprotein profiles via pre-cipitation and electrophoresis in ketotic cows and in those cows treated with different methods.Animals—21 cows with clinical and subclinical ketosis, 7 healthy cows in the early lactation period, and 7 healthy cows in the nonlactation period.Procedures—Ketotic cows were allocated into 3 groups; the first group was treated with dextrose and dexamethasone, the second group with dextrose and prednisolone, and the third group with dextrose and insulin. The β and α lipoproteins were precipitated with dextran sulfate-magnesium chloride in ketotic cows after treatment and healthy cows in the nonlactation and lactation periods. The serum samples, precipitates, and supernatants were examined via agarose gel electrophoresis for detection of alterations in serum lipo-proteins. Subsequently, alterations in serum apolipoproteins were detected via SDS-PAGE of precipitates.Results—Compared with serum β and α lipoprotein concentrations in healthy cows during nonlactation, those in cows during lactation were higher; however, those in cows with keto-sis were lower. The SDS-PAGE analysis of serum β lipoproteins revealed that apolipoprotein E (approx 36 and 40 kDa) decreased in ketotic cows, in comparison with healthy cows in the nonlactation and lactation periods, but increased after treatment. Decreases in apolipo-protein B (approx 222 kDa), apolipoprotein A-I (19 and 24 kDa), apolipoprotein A-IV (55 kDa), apolipoprotein C-III (8.8 and 10.2 kDa), and albumin (66 kDa) concentrations were detected in ketotic cows, in comparison with the healthy cows in the lactation period.Conclusions and Clinical Relevance—Serum lipoprotein and apolipoproteins may rou-tinely be determined via precipitation and electrophoresis in the diagnosis and treatment of ketosis. (Am J Vet Res 2009;70:563–570)

Ketosis is a major metabolic disorder that develops during the postparturient period. Cows with ketosis are likely to have high liver lipid concentrations. Evidence indicates that hepatic lipidosis precedes ketosis.10–14,16–18 Susceptibility to such disorders is greatest during copious milk secretion, during which glucose loss occurs and glu-cose and insulin concentrations in blood are decreased.13

Lipids are mainly stored in the adipose tissue and liver in the form of triglycerides. Lipids are transported from one tissue to another within the blood stream in the form of lipoproteins, after first combining with a protein carrier called an apolipoprotein.19 In cows with hepatic lipidosis and ketosis, triglyceride accumulation results in impaired metabolism of lipoproteins, most of which are produced by the liver. Studies on impaired lipoprotein metabolism are therefore important to elu-cidate the pathogenesis of hepatic lipidosis and hepatic

Received April 3, 2008.Accepted August 21, 2008.From the Departments of Biochemistry (İleri-Büyükoğlu) and Veteri-

nary Internal Medicine (Şahinduran, Sezer), Faculty of Veterinary Medicine, Mehmet Akif Ersoy University, Burdur 15100, Turkey; and the Department of Biochemistry, Faculty of Medicine, İnönü University, Malatya 44280, Turkey (Güldür).

Supported by the Akdeniz University Scientific Research Projects Unit.

The authors thank Dr. Metin Koray Albay for technical assistance.Address correspondence to Dr. İleri-Büyükoğlu.

AbbreviAtionsDS-MgCl

2 Dextran sulfate-magnesium chloride

HDL High-density lipoproteinLCAT Lecithin cholesterol acyltransferaseLDL Low-density lipoproteinNEFAs Nonesterified fatty acidsVLDL Very–low-density

08-04-0114r.indd 563 4/24/2009 1:03:35 PM

564 AJVR, Vol 70, No. 5, May 2009

lipidosis–related diseases, such as ketosis. The major biochemical changes detected in serum of cows with ketosis are an increased concentration of NEFAs2,13,17,20 and decreased lipoprotein16,18 and apolipoprotein10–12,14 concentrations.

Plasma lipoproteins are classically characterized by use of techniques such as precipitation, electro-phoresis, density ultracentrifugation, affinity chroma-tography, and immunologic techniques. Among those techniques, precipitation methods are widely used in clinical laboratories. They are fast, simple, and relative-ly inexpensive to perform. Lipoproteins can generally be separated from each other within 10 minutes to 2 hours, depending on the precipitation conditions and lipoprotein class of interest; therefore, these methods can be readily applied for screening, clinical analysis, and other purposes for which analyses are required rapidly or in a large number of samples.21 Many meth-ods used for determination of the lipoprotein pattern of cattle are based on ultracentrifugal3,5,6,22–26 separa-tion. Nonetheless, ultracentrifugal procedures are time consuming and equipment needed is not available in many laboratories. Moreover, it should always be kept in mind that lipoprotein separation procedures may re-quire modification before they are applied to the plasma of a given species.27 In previous studies,28,29 different precipitation methods were adapted for serum of sev-eral animal species.

The aim of the study reported here was to deter-mine changes in serum lipoprotein and apolipoprotein profiles via DS-MgCl

2 precipitation and electrophoresis

(agarose gel and SDS-PAGE) in ketotic cows, compared with cows treated with different methods.

Materials and MethodsAnimals and serum specimens—Holstein cows with

suspected ketosis that were brought to the Meh-met Akif Ersoy University Veterinary Clinical Center by their own-ers were used as principals in this experiment. Cows (n = 21) were of different ages and were in early lactation (2 to 37 days after parturition). Milk yield was approximately 8,000 kg/y/cow and was similar in healthy and diseased cows. Urine and milk samples were collected from each cow and the Rothera test30 and test stripsa were applied to diagnose ketosis. Routine clinical examination was car-ried out, and rectal temperature and heart and respiratory rates were recorded for each animal before and after treat-ment. Clinical ketosis (diagnosed via positive results of the Rothera test in the milk and urine, high concentration of β-hydroxybutyrate in the serum, and clinical signs of disease, such as decrease of milk yield or loss of appetite) was diagnosed in 21 dairy cows that were allocated into 3 groups equally for treatment by use of 3 procedures.

Each cow in the first group received 750 mL of 30% serum dextrose IV and a dose of 4 mg of dexa-methasone/100 kg of body weight IM for 2 consecutive days. Each cow in the second group received 750 mL of 30% serum dextrose IV plus 0.2 mg of prednisolone/kg IM daily for 2 days. Furthermore, 750 mL of 30% dex-trose (IV, for 2 days) and 200 units of insulin (IV, twice, with 48 hours between treatments) were administered to each cow in the third group. Two control groups (7 healthy cows in the early lactation period and 7 healthy

cows in the nonlactation period [> 300 days after par-turition]) were used for comparison with cows with ke-tosis and between control groups. These healthy cows were obtained from a dairy farm in Burdur Province. Animals in control groups did not receive any treat-ment during the study.

Blood samples from each cow were collected in the morning (before feeding) and were allowed to clot at 15° to 25°C for 1 hour, and subsequently, the serum was separated by centrifugation at 500 X g at 20°C for 15 minutes. Lipoprotein precipitation procedures were carried out on the same day. The serum samples were stored at –20°C until total cholesterol and β-hydroxy-butyrate analyses were performed.

Precipitation of serum β lipoproteins with DS-MgCl

2—A 20 g/L solution of dextran sulfateb and a 1.0

mol/L solution of MgCl2-6H

2O, adjusted to a pH of 7.0,

were prepared. Both were stored at 4°C. The working reagent was prepared freshly each week by mixing equal volumes of the 2 solutions. Then, 75 µL of the working reagent was added to 1.0 mL of serum, and the solution was mixed thoroughly and incubated at 15° to 25°C for 15 minutes and subsequently at 4°C for 1 hour before centrifugation at 2,500 X g for 15 minutes.31,32 The su-pernatant was separated, the precipitate was dissolved in 200 µL of saline (0.9% NaCl) solution, and magne-sium ions were removed by the addition of 50 µL of a solution of 12.7% potassium oxalate and centrifugation at 500 X g for 5 minutes.33 One part each of precipitate and supernatant was separated, and the other part was stored at –20°C for SDS-PAGE analysis.

Precipitation of serum α lipoproteins with DS-MgCl

2—Precipitation was made with modifications

of methods developed by Burstein et al34 and Bachorik and Albers.21 A 50 g/L solution of dextran sulfate and a 2.0 mol/L solution of MgCl

2-6H

2O were prepared

and adjusted to a pH of 7.0. Both were stored at 4°C. The working reagent was prepared freshly each week by mixing equal volumes of the 2 solutions. Then, 200 µL of the working reagent was added to 1.0 mL of the supernatant, and the solution was mixed thor-oughly and incubated at 15° to 25°C for 15 minutes and subsequently at 4°C for 1 hour before centrifugation at 2,500 X g for 15 minutes. The supernatant was removed by decantation, and the precipitate was dissolved in 400 µL of 0.5M potassium oxalate35 and centrifuged at 500 X g for 5 minutes. One part each of precipitate and supernatant was separated, and the other part was stored at –20°C for SDS-PAGE analysis.

Separation of serum lipoproteins via agarose gel electrophoresis—Serum and the lipoproteins precipi-tated were separated and identified by use of an aga-rose gel electrophoretic system.c Agarose was used at a concentration of 0.8%. Fifteen microliters of serum was applied directly to agarose gel. After mixing 200 µL of precipitate and supernatant with 50 µL of sample buffer (16 mL of barbital buffer at a pH of 8.6, 5 mL of glyc-erol, and 100 µL of 1.0% bromophenol blue were mixed and made up to 100 mL with distilled water), 18 µL of precipitate and supernatant was placed in the slot. Elec-trophoresis was performed in barbital buffer at a pH of 8.6 for 45 minutes at 83 V. After electrophoresis, the

08-04-0114r.indd 564 4/24/2009 1:03:36 PM

AJVR, Vol 70, No. 5, May 2009 565

gels were fixed in a solution of ethanol-acetic acid-wa-ter (ratio of 60:10:30 vol/vol/vol) for 10 minutes, oven dried (60°C for 30 minutes), and stained overnight with a Fat Red 7B solution. Gels were destained in a solution of methanol-water (ratio of 1:4 vol/vol) until the lipo-protein bands were clear.36,37 An image processing pro-gramd was used for densitometric scanning of stained electrophoretograms with automatic integration of the peaks of optical density.

SDS-PAGE—Apolipoprotein composition was de-termined via discontinuous SDS-PAGE,e as described by Laemmli.38 Four percent stacking and 7.5% and 12% re-solving gels (1.0 mm thick) were prepared. Apolipopro-teins at high molecular weights were evaluated in the 7.5% gel, and apolipoproteins at low molecular weights were evaluated in the 12% gel. The precipitates were diluted 1:1 (vol/vol) with the sample buffer containing 5.0 mL of Tris-HCl buffer (pH, 6.8), 4.0 mL of glycerol, 8.0 mL of 10% SDS, 2.0 mL of 2-mercaptoethanol, 1.0 mL of 0.05% bromophenol blue, and 20 mL of distilled water. After adding the sample buffer, the precipitates were incubated at 100°C in a boiling water bath for 5 minutes. A 15-µL aliquot of precipitates (indicated as P

1 and P2) and a

5- to 10-µL aliquot of standard protein markersf,g were ap-plied on gels. Electrophoresis was performed at a constant current of 20 mA/gel for 90 minutes. After electrophoresis, the gels were stained overnight with 0.025% Coomassie brilliant blue G-250 in methanol-acetic acid-water (ratio of 40:7:53 [vol/vol/vol]), destained with methanol-acetic acid-water (ratio of 40:7:53 [vol/vol/vol]), and stored in 5% acetic acid. An image processing programd was used for estimation of molecular weight in gels.

Other methods—Serum concentration of β-hy-droxybutyrate was determined by use of a commercial-ly available test kit.h Serum concentration of cholesterol was measured by use of an automatic analyzer.i

Statistical analysis—Data for cholesterol and β-hydroxybutyrate concentrations were analyzed by use of the Kruskal-Wallis and Mann-Whitney U tests. Val-ues are expressed as mean ± SD.j For all comparisons, values of P < 0.05 were considered significant.

Results

Agarose gel electrophoresis of serum β and α li-poproteins isolated via DS-MgCl

2 precipitation—Se-

rum lipoprotein patterns in healthy cows during the lactation and nonlactation periods were compared with those in ketotic cows. In general, lipoproteins migrated into β (ie, LDL) and α (ie, HDL) regions. Pre-β li-poproteins (ie, VLDL) were not detectable in serum samples because of their small amounts. In agarose gel electrophoresis, lipoproteins with α

1 mobility

were observed in some β regions during lactation but were absent during nonlactation. Compared with mean percentage areas for serum β and α lipopro-teins in healthy cows during nonlactation, those in cows during lactation were higher, whereas those in cows with ketosis were lower (Table 1). Decreases were clearer for serum β lipoproteins than for α li-poproteins in cows with ketosis. In terms of ketotic cows, the serum lipoproteins migrated farther and in a much more widespread manner than those of non-ketotic cows. In response to treatment, dextrose and insulin were more effective, regarding the recovery of serum β lipoproteins of ketotic cows, than were other treatment methods.

SDS-PAGE—Two protein bands, whose molecular weights were estimated to be 36 and 40 kDa, were found via analysis of bovine serum β lipoproteins. These pro-teins, identified as bovine apolipoprotein E, significant-ly increased during the lactation period, compared with the nonlactation period, whereas in the ketotic cows, the proteins decreased and even disappeared. After 3 days of treatment with prednisolone and dexamethasone, apolipoprotein E was increased in ketotic cows. Small amounts of the 19- and 24-kDa proteins, presumably apolipoprotein A-I, a major apolipoprotein of α lipo-proteins, were detected in bovine serum β lipoproteins. Furthermore, in healthy cows during lactation, these dimeric proteins were in higher concentration than in cows during the nonlactation period and were in lowest concentration in ketotic cows. The same protein bands were also detected in bovine α lipoprotein in high con-centration, and they were reduced in the ketotic cows (Table 2). However, decreases were not as pronounced for α lipoproteins as for β lipoproteins. The β lipopro-tein fraction contained a decreased albumin band (66 kDa) in the ketotic cows and returned to basal concen-tration after treatment for 3 days. In contrast, in the α lipoprotein fraction of ketotic cows, increased concen-trations of albumin were detected, compared with the lactation period, and the albumin band was slightly de-

After After After treatment treatment treatment Ketotic with with withAnalyte Lipoprotein Nonlactation Lactation cows insulin dexamethasone prednisolone

Serum β 2.29 2.68 1.23 1.48 0.47 0.89 α1+α 20.40(α) 22.76 19.92 19.10 20.20 21.03

Precipitate(P1) β 2.05 1.73 0.72 0.98 1.80 0.58 α1 0.54(α) 3.2 2.11 — — 3.47

Precipitate(P2) α 11.70 16.82 14.32 16.10 11.81 15.19

—=Notapplicable.

Table 1—Mean percentage areas of agarose gel electrophoretograms of serum lipoproteins isolated with a DS-MgCl

2 precipitation method in healthy cows (during lactation and nonlactation), ketotic cows,

and ketotic cows after treatment with insulin, dexamethasone, and prednisolone.

08-04-0114r.indd 565 4/24/2009 1:03:36 PM

566 AJVR, Vol 70, No. 5, May 2009

creased after treatment with insulin for 3 days. A 10.2-kDa protein and a 8.8-kDa protein (near the 14.2-kDa molecular weight marker) were detected in β lipopro-teins. These protein bands, which were considered to be similar to bovine apolipoprotein C-III, increased in cows during lactation, compared with the magnitude in cows during nonlactation. These protein bands de-creased in ketotic cows and also after treatment. A 55-kDa protein band, probably apolipoprotein A-IV, was found in both β lipoprotein and α lipoprotein fractions. The band was decreased in the β lipoprotein fraction of ketotic cows but increased in the α lipoprotein fraction, compared with protein bands in cows during the lacta-tion and nonlactation periods. After treatment, the pro-tein band was restored. In the β lipoprotein fraction, a 222-kDa protein band was detected, which was absent in the α lipoprotein fraction. The band corresponding to apolipoprotein B was not detected in ketotic cows.

The cholesterol concentration was lower in the ke-totic cows than in healthy cows during lactation and increased after treatment (Table 3). Concentrations of β-hydroxybutyrate in the ketotic cows were higher than those in healthy cows and returned to the val-ues of healthy cows after treatment with insulin and dexamethasone.

Discussion

The electrophoretic distribution of bovine serum indicated that α lipoproteins were the major lipopro-tein and that β lipoproteins were also present in lower amounts. Pre-β lipoproteins are associated with VLDL but were not detectable in serum samples because of the small amount.22 In cows, ultracentrifugally isolated LDL overlapped with a light form of HDL. These large

α lipoproteins associated with hyperalphalipoprotein-emia were detected in some LDL fractions during lac-tation and lactation-pregnancy but were absent during nonlactation.3,22,25,39 As detected by use of agarose gel electrophoresis, α

1 (HDL

1) was associated with β mo-

bility during nonlactation and with α and β mobility throughout lactation. The variety in HDL

1 mobility is

related to lipoprotein lipase activity as blood triglycer-ides are hydrolyzed for milk fat synthesis.22 In bovine studies, dextran sulfate has been used to precipitate li-poproteins, but as Burstein et al34 noted, this polysac-charide also can be used to precipitate human LDL and HDL. These authors note that the lipid-to-protein ratio is a determinant for the relative ease of formation of an insoluble lipoprotein-polyanion complex. As a result, α-LDL and HDL

1 in high concentrations in lactating

cows, after interacting with dextran, coprecipitate with β lipoproteins. The increased concentration of α-LDL and HDL

1 in lactating cows then would be products of a

primary particle secreted by the liver, rather than a rem-nant of VLDL catabolism.40 In the present study, bovine lipoproteins migrated into the β and α region and pre-β lipoproteins were not detectable via agarose gel electro-phoresis of precipitates with dextran sulfate. During the lactation period, β and α bands were stronger than those in cows during the nonlactation period. Lipoproteins with α

1 mobility were evident in the β region during lactation

but were absent during nonlactation.In cows with ketosis, LDL (β lipoproteins) and

HDL (α lipoproteins) concentrations are reported to be lower.16,18 Severe hepatic lipidosis has been associated with a large decrease in LDL concentration and a small-er change in HDL concentration.6,8,16 Results of our study indicated that bovine serum β and α lipoproteins isolated via DS-MgCl

2 precipitation were decreased in

Precipitate(P1;β+α1) Precipitate(P2;α)

After After After After treatment treatment treatment treatment Ketotic with with with Ketotic withLipoprotein Nonlactation Lactation cows insulin dexamethasone prednisolone Nonlactation Lactation cows insulin

ApoB 2.64 4.24 — — — — — — — —Albumin 30.83 30.38 27.73 30.17 33.10 33.95 25.92 26.75 33.08 31.90ApoA-IV 7.83 8.97 5.84 8.04 6.73 4.00 1.98 3.12 4.41 2.80ApoE 2.16 3.13 — — 2.55 5.16 — — — —ApoA-I 13.34 26.88 12.13 15.55 10.54 9.51 39.62 45.95 42.70 47.52ApoC-III 3.40 5.69 4.17 — 3.32 4.08 — — — —

Apo=Apolipoprotein.SeeTable1forremainderofkey.

Table 2—Mean percentage areas of SDS-PAGE of lipoprotein fractions in the same cows as in Table 1.

Group No. Cholesterol β-Hydroxybutyrate

Lactatingcows 7 4.130.83a 0.880.04a

Nonlactatingcows 7 3.040.32b 0.610.27a

Ketoticcows 21 3.111.00b 1.760.60b

Cowsaftertreatmentwithinsulin 7 4.891.26a,c 0.990.14a

Cowsaftertreatmentwithprednisolone 7 4.620.44a,c 1.160.56a,b

Cowsaftertreatmentwithdexamethasone 7 3.961.35a,b 0.990.18a

a–cWithinacolumn,valueswithdifferentsuperscriptlettersaresignificantly(P0.05)different.

Table 3—Mean SD serum cholesterol and α-hydroxybutyrate concentrations (mmol/L) of healthy cows, ketotic cows, and cows after treatment.

08-04-0114r.indd 566 4/24/2009 1:03:36 PM

AJVR, Vol 70, No. 5, May 2009 567

ketotic cows. Via agarose gel electrophoresis, decreases in serum β lipoproteins of cows with ketosis were more distinct than in serum α lipoproteins. The dramatic de-crease in serum β lipoproteins could result from several mechanisms, including decreased VLDL secretion, de-creased conversion of VLDL to LDL, or increased LDL uptake.6 In the ketotic cows, the serum lipoproteins mi-grated farther than those in nonketotic cows. This sug-gested that NEFAs associated with serum albumin in α and β lipoprotein fractions of the ketotic cows increased as the severity of the disease increased. In humans, apo-lipoproteins have been recognized as obvious markers for identifying discrete lipoprotein particles. However, limited research has been conducted on ruminant apo-lipoproteins.41 The analysis of lipoproteins in ruminant animals is complicated because of an overlap of α li-poproteins (containing apolipoprotein A-I as the major apolipoprotein) and β lipoproteins (containing apoli-poprotein B as the major apolipoprotein). Contrary to humans, in bovine serum, most of the circulating lipids are associated with apolipoprotein A-I–containing lipo-proteins. In cattle, concentrations of β lipoproteins are about one-fifth to one-fourth those in healthy humans. This agrees with the low concentrations of plasma tri-glycerides in ruminants resulting from their nutritional and metabolic particularities, but the importance of these lipoproteins for milk fat formation has been rec-ognized in dairy cows.41

In cows with hepatic lipidosis, notable decreases were determined in lipid and lipoprotein consentra-tions.8 Furthermore, in cows with fatty liver and keto-sis, decreases of serum apolipoprotein B-100,12 apoli-poprotein A-I, apolipoprotein C-III, cholesterol esters, albumin concentrations, and LCAT activity are pro-nounced.3,4,7,10,11,13,15,36 It is possible to assume that β-hy-droxybutyrate, hepatic triglycerides, and NEFAs reduce apolipoprotein synthesis or secretion.10,15 Moreover, in a recent study,42 liver from ketotic cows was character-ized by substantial downregulation of genes involved in protein synthesis, protein trafficking, protein ubiq-uitination, and molecular transport. The fact that the apolipoprotein B gene (APOB) was upregulated along with downregulation of genes associated with protein ubiquitination-degradation and triglyceride accumula-tion points at other mechanisms limiting VLDL synthe-sis and export (eg, lower de novo synthesis of cholester-ol and oleic acid, rather than deficient apolipoprotein B synthesis). The functional consequences of impaired intracellular protein degradation in liver might include alterations in cellular turnover, effects on gene regula-tion, modulation of cell signaling, induction of apop-tosis and necrosis, release of reactive oxygen species, and loss of gene-protein function. Such changes could be a primary factor involved in the impairment of hepatic function that contributes to the onset of clinical ketosis in dairy cows. Also, in the present study, the cholesterol concentration was significantly lower in the ketotic cows than in the lactating cows. The increased cholesterol in the lactating cows appeared to be associated with dynamic changes in and among the various lipoprotein classes, with the LCAT enzyme playing a pivotal role.40

Marcos et al4 and Grummer13 reported that serum concentrations of albumin are decreased in cows with

hepatic lipidosis, which may indicate a reduction in the ability of the liver to synthesize protein when tri-glyceride content is high. However, in other studies,2,16 the albumin concentrations of ketotic cows were sig-nificantly higher than in the early lactation cows. In contrast, protein concentrations of ketotic cows were significantly lower, compared with the protein concen-trations of the control cows.

The α lipoprotein fraction contains an increased concentration of NEFAs correlated with the severity of the ketosis. Free fatty acids associated with serum albumin are included in the extracts of the α lipopro-teins. Increased concentrations of NEFAs in the serum of ketotic cows have been reported.2 In severe cases of ketosis, the β lipoproteins also contained increased concentrations of NEFAs.

In our view, the β lipoprotein fraction contained a decreased concentration albumin band (66 kDa) in ke-totic cows and returned to basal value after treatment. However, in α lipoproteins of ketotic cows, increased concentrations of albumin were detected, in comparison with lactating cows, and the albumin band was slightly decreased after treatment with insulin for 3 days.

Apolipoprotein B-100 and apolipoprotein A-I are essential for transportation of triglycerides from the liver to the mammary gland and of cholesterol to steroidogenic tissues (ovary and adrenal cortex).10,12 Compared with values in cows during early lacta-tion, apolipoprotein B-100 and apolipoprotein A-I concentration further decreased in cows with keto-sis. However, the decreases in apolipoprotein B-100 concentration were more distinct than those of apo-lipoprotein A-I. The preferential decrease in apoli-poprotein B-100 concentration indicated that apoli-poprotein B-100 is synthesized by the liver, whereas apolipoprotein A-I is synthesized by the liver and the intestine. The intestinal synthesis appears not to be considerably affected in diseased cows.5,10,11 In a previous study43 in which bovine lipoproteins were isolated in plasma via gel filtration and apolipopro-tein composition was determined via SDS-PAGE, the 22-kDa protein band was detected in bovine triglyc-eride-rich lipoproteins. This low–molecular-weight protein may be analogous to nonruminant apolipo-protein A-I. In the same study, the 220- and 290-kDa proteins were predominantly detected; however, > 25 proteins were detected between 110 and 370 kDa in triglyceride-rich lipoproteins. Also, bovine HDL con-tained proteins analogous to apolipoprotein A-I and apolipoprotein C. Differences in apolipoprotein pro-files between nonlactating and lactating cows were not apparent. In another study, the apolipoprotein A-I band mainly distributed in HDL was detected in LDL and VLDL. The band appeared to be a mixture of apolipoprotein A-I (24 kDa) and γ-globulin light chain (23 kDa).44 Because bovine HDL

1 (α

1) particles

were detected in the LDL range, HDL1 contained the

major protein component of HDL (ie, apolipoprotein A-I).45 However, the major apolipoprotein of bovine α-LDL and HDL

1 is apolipoprotein A-I.40

In the β lipoprotein fraction, a 222-kDa protein band was detected, which was similar to apolipoprotein B, in cows in lactation and nonlactation; however, this

08-04-0114r.indd 567 4/24/2009 1:03:36 PM

568 AJVR, Vol 70, No. 5, May 2009

was not detected in ketotic cows. The 19- and 24-kDa proteins (presumably apolipoprotein A-I) were reduced in β and α lipoprotein fractions of cows with ketosis. Moreover, in the α lipoprotein fraction of ketotic cows, apolipoprotein A-I had indistinct decreases.

One study46 found that bovine apolipoprotein A-IV concentrations determined by use of ELISA during the lactation period in cows are significantly higher than during the nonlactation period. After 4 days of food withholding, the concentration of plas-ma apolipoprotein A-IV had decreased significantly at days 3 and 4 and was returned to the basal value by 3 days of feeding. In the same study, a minor band (42 kDa) was found in VLDL, LDL, and HDL fractions via SDS-PAGE. Another band (43 kDa) was found in the HDL fraction, and a 42-kDa protein band was identi-fied as bovine apolipoprotein A-IV.46 Puppione et al25 suggested that the 46-kDa apolipoprotein in the HDL fraction might also reflect the conversion of triglyc-eride-rich lipoproteins into α lipoproteins if it cor-responds to a bovine form of apolipoprotein A-IV. In the present study, a protein of approximately 55 kDa (probably apolipoprotein A-IV) in the α lipoprotein fraction was higher in concentration in cows during lactation than during nonlactation and was increased in ketotic cows. However, the same protein band was decreased in the β lipoprotein fraction during the lactation period, compared with the nonlactation period, and also was decreased in cows with ketosis. After treatment, the protein band was restored.

Apolipoprotein C-III is recognized as one of the most important serum triglyceride regulatory factors. Bovine apolipoprotein C-III is a low molecular mass protein mainly distributed in the HDL fraction, and it consists of at least 2 isoforms, 8.2 and 7.3 kDa; the lat-ter is the major species.1 Also, apolipoprotein C-III has been detected in chylomicrons and LDL and VLDL frac-tions.24,44 Apolipoprotein C-III inhibits the activation of lipoprotein lipase by apolipoprotein C-II and further reduces hepatic uptake of the remnant lipoprotein. Lecithin cholesterol acyltransferase is also activated by apolipoprotein C-III. Bovine serum apolipoprotein C-III concentration has the lowest value in the nonlac-tation period, gradually increases around the early lac-tation period, and reaches the maximum concentration in the midlactation period. The serum apolipoprotein C-IIII concentration is decreased in cows with hepat-ic lipidosis, ketosis, left displacement of abomasum, hypocalcemia, and retained placenta, compared with healthy cows in early lactation.1,9,11,14,44 In addition to LCAT and apolipoprotein B-100, determination of the apolipoprotein C-III concentration during the peripar-tum period appears to be helpful in diagnosing peripar-turient diseases.1,9,11,14 It was also found that the amount of 10.2- and 8.8-kDa proteins, which are similar to bo-vine apolipoprotein C-III, increased during lactation, compared with that in cows during nonlactation, in the β lipoprotein fraction. However, a faint protein band was detected in ketotic cows and was not changed after treatment. Apolipoprotein C-III could not be detected in the α lipoprotein fraction because the β lipoprotein fraction used in this study was contaminated to some extent by HDL (probably HDL

1).

Apolipoprotein E is believed to have an impor-tant role in cholesterol metabolism, mediating hepatic uptake of HDL, which would be the final phase of the hypothesized reverse transport of cholesterol.47 The ex-istence of apolipoprotein E was reported in VLDL and LDL of lactating cows.43 However, apolipoprotein E in lipoprotein fractions was not detected in calves and adult cows.47,48 Takahashi et al48 found an increase of a 38-kDa protein in the bovine VLDL fraction caused by withholding of food, and the lipoprotein was iden-tified as bovine apolipoprotein E by determination of its N-terminal amino acid sequence. The concentration of bovine plasma apolipoprotein E was low, compared with that in other species. Apolipoprotein E was mark-edly increased in VLDL, LDL, and nonlipoprotein frac-tions by withholding of food. The concentration of apo-lipoprotein E in the HDL fraction was low, compared with those of VLDL or LDL, and was not changed by withholding of food. Increased concentrations of plas-ma apolipoprotein E in a cow from which food has been withheld imply the possible relevance of apolipoprotein E in the development of lipid-related disorders such as hepatic lipidosis in the peripartum period. Detection of an unusually low apolipoprotein E concentration in the peripartum period may be useful for early diagnosis of such disorders.

In another study,47 it was found that the apolipopro-tein E concentrations evaluated by use of ELISA during lactation were significantly higher than during nonlac-tation. Because most of the plasma apolipoprotein E is thought to be of liver origin, it was suggested that liver production had been upregulated during lactation. Via agarose gel electrophoresis, bovine apolipoprotein E was found in the β lipoprotein fraction, whereas it could not be found in the α lipoprotein fraction. The slow β particles (α

1) that had a high concentration of

apolipoprotein E were probably deficient in neutral lip-ids.47 The HDL

1 appears to be metabolically related to

HDL, and the predominant apolipoprotein of HDL1 is

apolipoprotein E.45

By use of SDS-PAGE, it was determined that in the β lipoprotein fraction, 2 protein bands, with molecular weights estimated to be 36 and 40 kDa (probably apo-lipoprotein E), had a higher concentration during the lactation period than during the nonlactation period, whereas in the ketotic cows, they decreased and even disappeared. Apolipoprotein E could not be detected in the α lipoprotein fraction because apolipoprotein E was the predominant apolipoprotein of HDL1 (α1). After ke-totic cows were treated for 3 days with prednisolone and dexamethasone, apolipoprotein E concentration increased. Therefore, detection of apolipoprotein E in ketotic cows may indicate a better prognosis, and more studies on apolipoprotein E should be carried out.

In the present study, ketotic cows were treated with glucocorticoid (prednisolone and dexamethasone) plus dextrose and insulin plus dextrose. Insulin inhibits the release of NEFAs from adipose tissue, which is followed by a decrease in circulating plasma NEFAs. Insulin en-hances lipogenesis and the synthesis of acylglycerol and increases oxidation of glucose to CO

2 via the pentose

phosphate pathway. All these effects are dependent on glucose and can be explained, to a large extent, on the

08-04-0114r.indd 568 4/24/2009 1:03:37 PM

AJVR, Vol 70, No. 5, May 2009 569

basis of the ability of insulin to enhance the uptake of glucose into adipose cells. A principal action of insulin in adipose tissue is to inhibit the activity of hormone-sensitive lipase, reducing the release not only of NEFAs but of glycerol as well.49 Reduction in glucose output by the liver could decrease blood glucose concentra-tions and insulin secretion, which, in turn, could lead to greater lipid mobilization from adipose tissue and in-creased rates of hepatic fatty acid uptake and ketogen-esis. During ketosis, plasma NEFAs and liver triglycer-ide concentrations typically increase and liver glycogen content decreases.13

The glucocorticoids (11-oxysteroids) are secreted by the adrenal cortex and increase gluconeogenesis. This is a result of enhanced hepatic uptake of amino acids and increased activity of aminotransferases and key enzymes of gluconeogenesis. In addition, glucocor-ticoids inhibit the use of glucose in extrahepatic tissues. In all these actions, glucocorticoids act in a manner an-tagonistic to insulin.49

Prednisone is the most widely prescribed of the glucocorticoids. Its short half-life, low cost, and neg-ligible mineralocorticoid effect make it useful for most immunosuppressive and anti-inflammatory indications. Prednisolone is the active hepatic metabolite of predni-sone and is useful for treatment of liver failure. Dexa-methasone is the long-acting glucocorticoid of choice, being about 7 times as potent on a weight basis as pred-nisone and having a half-life of 24 hours. Prednisolone is preferred because of the low prevalence of adverse ef-fects. A substantial decrease in milk production should be expected when administrating a corticosteroid to lactating cattle.

In an earlier study,7 cows with hepatic lipidosis were treated with insulin to suppress further adipose fat mobi-lization and to help increase lipoprotein output. Ketotic cows were treated with glucocorticoid plus insulin by McCarthy et al.2 Glucocorticoids promote the breakdown or mobilization of tissue protein, whereas insulin sup-presses the formation of gluconeogenic enzymes in the liver, and the dual hormonal action increases the avail-ability of amino acids by promoting their mobilization but slowing their conversion to glucose. In the present study, it was observed that lipoprotein profiles and cholesterol and β-hydroxybutyrate concentrations of cows treated with insulin plus dextrose returned to reference range val-ues more quickly than those of cows treated with dexa-methasone and prednisolone plus dextrose.

From a practical aspect, our results suggested that precipitation of bovine serum lipoproteins with DS-MgCl

2 can be useful for diagnosis of ketosis and esti-

mating a prognosis. In the same way, determination of bovine lipoprotein and apolipoprotein profiles might be useful for the study of other metabolic disorders. Further investigations are required for identification of apolipoproteins in ketotic cows.

a. Spinreact urin-10, S.A.Ctra.Santa Coloma, Sant Esteve De Bas, Spain.

b. Dextralip 50, Sigma-Aldrich Chemie GmbH, Munich, Germany.c. Mini-sub cell GT, agarose gel electrophoresis system, 7 X 10 cm,

Bio-Rad Laboratories, Hercules, Calif.d. Rasband WS. ImageJ, US National Institutes of Health, Bethesda,

Md. Available at: rsb.info.nih.gov/ij/. Accessed Jul 10, 2008.

e. Mini-Protean 3 cell vertical slab gel system, Bio-Rad Laborato-ries, Hercules, Calif.

f. Board range standard protein mixture, Bio-Rad Laboratories, Hercules, Calif.

g. Wide molecular weight range, Sigma-Aldrich Chemie GmbH, Munich, Germany.

h. Randox Laboratories Ltd, Crumlin, County Antrim, Northern Ireland.

i. VET TEST 8008, IDEXX Laboratories Inc, Westbrook, Me.j. SPSS for Windows, release 10.0, SPSS Inc, Chicago, Ill.

References1. Nitanai A, Katoh N, Oikawa S, et al. Decreases in serum apoli-

poprotein C–III concentration in cows with ethionine-induced fatty liver. J Vet Med Sci 2004;66:1113–1118.

2. McCarthy RD, Porter GA, Griel LC. Bovine ketosis and depressed fat test in milk: a problem of methionine metabolism and serum lipoprotein aberration. J Dairy Sci 1968;51:459–462.

3. Mazur A, Marcos E, Rayssiguier Y. Plasma lipoproteins in dairy cows with naturally occurring severe fatty liver: evidence of al-teration in the distribution of apo A-I-containing lipoproteins. Lipids 1989;24:805–811.

4. Marcos E, Mazur A, Cardot P, et al. Serum apolipoproteins B and A-I and naturally occurring fatty liver in dairy cows. Lipids 1990;25:575–577.

5. Uchida E, Katoh N, Takahashi K. Induction of fatty liver in cows by ethionine administration and concomitant decreases of se-rum apolipoproteins B-100 and A-I concentrations. Am J Vet Res 1992;53:2035–2042.

6. Rayssiguier Y, Mazur A, Gueux E, et al. Plasma lipoproteins and fatty liver in dairy cows. Res Vet Sci 1988;45:389–393.

7. Gerloff BJ, Herdt TH. Hepatic lipidosis from dietary restriction in nonlactating cows. J Am Vet Med Assoc 1984;185:223–224.

8. Sevinç M, Başoğlu A, Güzelbektaş H, et al. Lipid and lipopro-tein levels in dairy cows with fatty liver. Turk J Vet Anim Sci 2003;27:295–299.

9. Katoh N, Nakagawa-Ueta H. Concentrations of apolipoprotein C–III in healthy cows during the peripartum period and cows with milk fever. J Vet Med Sci 2001;63:597–601.

10. Oikawa S, Katoh N, Kawawa F, et al. Decreased serum apolipo-protein B-100 and A-I concentrations in cows with ketosis and left displacement of the abomasum. Am J Vet Res 1997;58:121–125.

11. Katoh N. Relevance of apolipoproteins in the development of fatty liver and fatty liver-related peripartum diseases in dairy cows. J Vet Med Sci 2002;64:293–307.

12. Itoh H, Tamura K, Motoi Y, et al. Serum apolipoprotein B-100 concentrations in healthy and diseased cattle. J Vet Med Sci 1997;59:587–591.

13. Grummer RR. Etiology of lipid-related metabolic disorders in periparturient dairy cows. J Dairy Sci 1993;76:3882–3896.

14. Yamamoto M, Nakagawa-Ueta H, Katoh N, et al. Decreased concentration of serum apolipoprotein C–III in cows with fatty liver, ketosis, left displacement of the abomasum, milk fever and retained placenta. J Vet Med Sci 2001;63:227–231.

15. Katoh N, Oikawa S, Oohashi T, et al. Decreases of apolipo-protein B-100 and A-I concentrations and induction of hap-toglobin and serum amyloid A in nonfed calves. J Vet Med Sci 2002;64:51–55.

16. Sevinç M, Başoğlu A, Öztok İ, et al. The clinical-chemical pa-rameters, serum lipoproteins and fatty infiltration of the liver in ketotic cows. Turk J Vet Anim Sci 1998;22:443–447.

17. Brumby PE, Anderson M, Tuckley B, et al. Lipid metabo-lism in the cow during starvation-induced ketosis. Biochem J 1975;146:609–615.

18. Kennerman E. Serum lipoprotein levels in cows with subclinical ketosis. Veteriner Cerrahi Dergisi 2000;6:52–55.

19. Mahley RW, Innerarity TL, Rall SC, et al. Plasma lipoproteins: apolipoprotein structure and function. J Lipid Res 1984;25:1277–1294.

20. Mills SE, Beitz DC, Young JW. Characterization of metabolic changes during a protocol for inducing lactation ketosis in dairy cows. J Dairy Sci 1986;69:352–361.

08-04-0114r.indd 569 4/24/2009 1:03:37 PM

570 AJVR, Vol 70, No. 5, May 2009

21. Bachorik PS, Albers JJ. Precipitation methods for quantification of lipoproteins. Methods Enzymol 1986;129:78–100.

22. Raphael BC, Dimick PS, Puppione DL. Electrophoretic charac-terization of bovine serum lipoproteins throughout gestation and lactation. J Dairy Sci 1973;56:1411–1414.

23. Raphael BC, Dimick PS, Puppione DL. Lipid characterization of bovine serum lipoproteins throughout gestation and lactation. J Dairy Sci 1973;56:1025–1032.

24. Katoh N. In addition to the high-density lipoprotein fraction, apolipoprotein C–III is detected in chylomicrons and the very low-density lipoprotein fraction from serum of normolipidemic cows. J Vet Med Sci 2001;63:95–97.

25. Puppione DL, Kunitake ST, Toomey ML, et al. Physicochemical characterization of ten fractions of bovine alpha lipoproteins. J Lipid Res 1982;23:371–379.

26. van Dijk S, Wensing T. Comparison of the lipoprotein pattern of the horse, the pony and the lactating and non-lactating cow ob-tained by a combination of an ultracentrifugation and a precipi-tation technique. Comp Biochem Physiol B 1989;94:735–738.

27. Cornwell DG, Kruger FA. Molecular complexes in the isola-tion and characterization of plasma lipoproteins. J Lipid Res 1961;2:110–134.

28. İleri T, Güldür T. Comparative isolation of serum β + pre-β li-poproteins of human and various domestic animal species by dextran sulfate/MnCl

2 and sodium phosphotungstate/MgCl

2

precipitation methods. Turk J Vet Anim Sci 2002;26:729–740.29. İleri T, Güldür T. Comparative isolation of serum β + pre-β

lipoproteins of human and various domestic animal species by heparin/MnCl

2 precipitation method. FÜ Sağlık Bil Derg

2003;17:27–33.30. Geishauser T, Leslie K, Kelton D, et al. Evaluation of five cow-

side tests for use with milk to detect subclinical ketosis in dairy cows. J Dairy Sci 1998;81:438–443.

31. Sjoblom L, Eklund A. Determination of HDL2 cholesterol by pre-

cipitation with dextran sulfate and magnesium chloride: estab-lishing optimal conditions for rat plasma. Lipids 1989;24:532–534.

32. Warnick GR, Nguyen T, Albers AA. Comparison of improved precipitation methods for quantification of high-density lipo-protein cholesterol. Clin Chem 1985;31:217–222.

33. Sakagami T, Zilversmit DB. Separation of dog serum lipopro-teins by ultracentrifugation, dextran sulfate precipitation, and paper electrophoresis. J Lipid Res 1961;2:271–277.

34. Burstein M, Scholnick HR, Morfin R, et al. Rapid method for

isolation of lipoproteins from human serum by precipitation with polyanions. J Lipid Res 1970;11:583–595.

35. Burstein M, Scholnick HR. Lipoprotein-polyanion-metal inter-actions. Adv Lipid Res 1973;11:67–108.

36. Naito HK. Clinical chemistry theory, analysis and correlation. In: Kaplan LA, Pesce AJ, eds. Lipoprotein electrophoresis. 2nd ed. St Louis: CV Mosby Co, 1989;1195–1207.

37. Karcher RE, Nuttall KL. Lipoprotein agarose gel electrophore-sis. In: Burtis CA, Ashwood ER, eds. Tietz textbook of clinical chemistry. Philadelphia: WB Saunders Co, 1999;154–155.

38. Laemmli UK. Cleavage of structural proteins during the assem-bly of the head of bacteriophage T4. Nature 1970;227:680–685.

39. Hollanders B, Mougin A, N’Diave P, et al. Comparison of the lipoprotein profiles obtained from rat, bovine, horse, dog, rab-bit and pig serum by a new two-step ultracentrifugal gradient procedure. Comp Biochem Physiol B 1986;84:83–89.

40. Puppione DL. Implications of unique features of blood lipid transport in the lactating cow. J Dairy Sci 1978;61:651–659.

41. Marcos E, Mazur A, Cardot P, et al. Quantitative determination of apolipoprotein B in bovine serum by radial immunodiffusion. Comp Biochem Physiol B 1989;94:171–173.

42. Loor JJ, Everts RE, Bionaz M, et al. Nutrition-induced ketosis alters metabolic and signaling gene networks in liver of peripar-turient dairy cows. Physiol Genomics 2007;32:105–116.

43. Grummer RR, Meacham CA, Hurley WL, et al. Apolipoprotein composition of bovine lipoproteins isolated by gel filtration chromatography. Comp Biochem Physiol B 1987;88:1163–1174.

44. Yamamoto M, Katoh N, Adachi Y, et al. Identification and purifi-cation of apolipoprotein C-III from the serum of cows. Am J Vet Res 1998;59:667–672.

45. Chapman MJ. Comparative analysis of mammalian plasma lipo-proteins. Methods Enzymol 1986;128:70–143.

46. Takahashi Y, Konishi H, Sato K, et al. Enzyme-linked immu-nosorbent assay for bovine apolipoprotein A-IV. J Vet Med Sci 2004;66:1199–1204.

47. Takahashi Y, Itoh F, Oohashi T, et al. Distribution of apolipopro-tein E among lipoprotein fractions in the lactating cow. Comp Biochem Physiol B 2003;136:905–912.

48. Takahashi Y, Sato K, Itoh F, et al. Bovine apolipoprotein E in plasma: increase of ApoE concentration induced by fasting and distribution in lipoprotein fractions. J Vet Med Sci 2003;65:199–205.

49. Mayes PA, Botham KM. Lipid transport & storage. In: Murray RK, Granner DK, Mayes PA, et al, eds. Harper’s illustrated bio-chemistry. 28th ed. New York: McGraw-Hill Co, 2003;205–218.

08-04-0114r.indd 570 4/24/2009 1:03:37 PM