Iowa State University Capstones, Theses and Retrospective Theses and Dissertations Dissertations

1988 Progressive changes in metabolism of cows during induction of ketosis and treatment of ketosis with intestinal administration of glucose Jeffrey James Veenhuizen Iowa State University

Follow this and additional works at: https://lib.dr.iastate.edu/rtd Part of the Agriculture Commons, and the Animal Sciences Commons

Recommended Citation Veenhuizen, Jeffrey James, "Progressive changes in metabolism of cows during induction of ketosis and treatment of ketosis with intestinal administration of glucose " (1988). Retrospective Theses and Dissertations. 9741. https://lib.dr.iastate.edu/rtd/9741

This Dissertation is brought to you for free and open access by the Iowa State University Capstones, Theses and Dissertations at Iowa State University Digital Repository. It has been accepted for inclusion in Retrospective Theses and Dissertations by an authorized administrator of Iowa State University Digital Repository. For more information, please contact [email protected]. INFORMATION TO USERS The most advanced technology has been used to photo­ graph and reproduce this manuscript from the microfilm master. UMI films the original text directly from the copy submitted. Thus, some dissertation copies are in typewriter face, while others may be from a computer printer. In the unlikely event that the author did not send UMI a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyrighted material had to be removed, a note will indicate the deletion. Oversize materials (e.g., maps, drawings, charts) are re­ produced by sectioning the original, beginning at the upper left-hand comer and continuing from left to right in equal sections with small overlaps. Each oversize page is available as one exposure on a standard 35 mm slide or as a 17" x 23" black and white photographic print for an additional charge. Photographs included in the original manuscript have been reproduced xerographically in this copy. 35 mm slides or 6" X 9" black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order.

•lUMIAccessing the World's Information since 1938 300 North Zeeb Road, Ann Arbor, Ml 48106-1346 USA

Order Number 8826462

Progressive changes in metabolism of cows during induction of ketosis and treatment of ketosis with intestinal administration of glucose

Veenhuizen, Jeffrey James, Ph.D. Iowa State University, 1988

UMI 300N.ZeebRd. Ann Arbor, MI 48106

PLEASE NOTE:

In all cases this material has been filmed in the best possible way from the available copy. Problems encountered with this document have been identified here with a check mark V .

1. Glossy photographs or pages

2. Colored illustrations, paper or print

3. Photographs with dark background

4. Illustrations are poor copy

5. Pages with black marks, not original copy

6. Print shows through as there is text on both sides of page

7. Indistinct, broken or small print on several pages

8. Print exceeds margin requirements

9. Tightly bound copy with print lost in spine

10. Computer printout pages with indistinct print

11. Page(s) lacking when material received, and not available from school or author.

12. Page(s) seem to be missing in numbering only as text follows.

13. Two pages numbered . Text follows.

14. Curling and wrinkled pages

15. Dissertation contains pages with print at a slant, filmed as received

16. Other

Progressive changes in metabolism of cows during

induction of ketosis and treatment of ketosis

with intestinal administration of glucose

by

Jeffrey James Veenhuizen

A Dissertation Submitted to the

Graduate Faculty in Partial Fulfillment of the

Requirements for the Degree of

DOCTOR OF PHILOSOPHY

Department: Animal Science

Major: Nutritional Physiology

Approved:

Signature was redacted for privacy.

Signature was redacted for privacy.

Signature was redacted for privacy.

For the Graduate College

Iowa State University Ames, Iowa

1988 il

TABLE OF CONTENTS

Page

INTRODUCTION 1

REVIEW OF LITERATURE 4

Etiology of lactation ketosls 4 Experimental Induction of ketosls 11 Histological organization of liver 15 Gluconeogenesls In the lactatlng cow 19 Gluconeogenic pathways 20 Gluconeogenic precursors 22 Regulation of gluconeogenesls 26 Lipid metabolism in ruminant liver 29 Ketogenesis in ruminant liver 33 Kinetics of glucose and propionate metabolism 39 Summary of literature 44

EXPLANATION OF DISSERTATION FORMAT 46

REFERENCES 47

SECTION I. BLOOD AND LIVER METABOLITES AND MILK YIELD DURING THE DEVELOPMENT OF EXPERIMENTAL KETOSIS AND EARLY TREATMENT OF KETOSIS IN COWS 61

ABSTRACT 62

INTRODUCTION 64

MATERIALS AND METHODS 66

Design of experiment of management of cows 66 Sampling and analytical procedures 70 Statistical analysis 73

RESULTS AND DISCUSSION 74

Milk production, feed composition, and body weight loss 74 Metabolites in blood 77 Liver composition 80

GENERAL DISCUSSION 84

ACKNOWLEDGEMENTS 89 ill

Page

REFERENCES 90

SECTION II. EFFECTS OF INDUCED AND TREATED KETOSIS ON IN VITRO HEPATIC GLUCONEOGENESIS AND IN VIVO GLUCOSE AND PROPIONATE METABOLISM IN COWS 93

ABSTRACT 94

INTRODUCTION 96

MATERIALS AND METHODS 98

Management of cows and design of experiment 98 Biopsies of liver 99 , In vitro Incubations 101 Protocol for single Injections of tracer 103 Analytical procedures 104 Statistical analysis 106

RESULTS AND DISCUSSION 107

In vitro Incubations 107 In vivo kinetics of propionate and glucose metabolism 114

GENERAL DISCUSSION 119

ACKNOWLEDGEMENTS 123

REFERENCES 124

GENERAL SUMMARY 128

REFERENCES 135

ACKNOWLEDGEMENTS 136

APPENDIX 138 iv

LIST OF TABLES

Page

SECTION I.

TABLE 1. Feed Ingredients 67

TABLE 2. Summary of sampling blood and liver and feeding schedule for ketosis-induced (KIP) and glucose-treated (GT) cows 71

TABLE 3. Milk production, feed consumption, and body weight changes in cows 75

TABLE 4. Metabolite concentrations in blood plasma from control, ketosis-lnduced (KIP), and glucose-treated (GT) cows 78

TABLE 5. Liver composition of control, ketosis- lnduced (KIP), and glucose-treated (GT) cows 81

SECTION II.

TABLE 1. Codes for stage of experiment and sampling of control, ketosis-induced (KIP), and glucose-treated (GT) cows 100

TABLE 2. Utilization of propionate for synthesis of glucose in slices of liver from control, ketosis-induced (KIP), and glucose-treated (GT) cows 108

TABLE 3. Metabolism of propionate and acetate to CO2 in slices of liver from control, ketosis-induced (KIP), and glucose-treated (GT) cows 111

TABLE 4. Ratio of substrate metabolized with Ca++ in media to that without Ca++ in media for control, ketosis-lnduced (KIP), and glucose-treated (GT) cows 113

TABLE 5. Kinetics of propionate metabolism in control and ketotic cows 115

TABLE 6. Kinetics of glucose metabolism in control and ketotic cows 116 V

Page

APPENDIX

TABLE Al. Acetate metabolized to triglycerides In liver slices of a control cow Incubated with different concentrations of glucose 139

TABLE A2. Concentrations of acetate In plasma of control, ketosls-Induced (KIP), and glucose-treated (GT) cows 140

TABLE A3. Concentrations of acetate In plasma of control, ketosls-Induced (KIP), and glucose-treated (GT) cows (omitting one cow) 141 vl

LIST OF FIGURES

Page

FIGURE 1. The liver acinus concept of the functional unit of liver parenchyma 16 1

INTRODUCTION

Metabolic priority In high-producing cows Is given to the demands

of milk production early in lactation at a time when appetite and

subsequent feed Intake is limited. The liver, being the primary site

of gluconeogenesls in the ruminant, is unable to synthealze adequate

amounts of glucose for lactose synthesis by the mammary gland.

Consequently, plasma glucose concentration decreases. This

hypoglycemia indirectly stimulates release of fatty acids and glycerol

by adipose tissue. The fatty acids are oxidized in the liver to ketones and CO2 and can provide an energy source for muscle and other

peripheral tissues so that glucose is spared. These metabolic processes seem to be adequate for normal cows to maintain maximum milk production and body health. In some cows however, a breakdown of homeostatic controls results in the metabolic disease known as lactation ketosls, which Is characterized by severe hypoglycemia and hyperketonemia.

Lactation ketosls can occur as early as 10 days postpartum and as late as 8 weeks postpartum (145), and can be divided into two stages, clinical, and sub-clinical. Clinical ketosls is characterized visually by decreased appetite, loss in body condition, and decreased milk production. Most ketotlc cows are lethargic, but occasionally some cows show extreme hyperexcltabllity. Subclinical ketosls is a nebulous disease state occurring before obvious clinical signs are 2

detected. Although much debated, It is thought that some decrease in

milk production does occur and that long-term productivity of the cow

is influenced negatively from subclinical ketosis. Incidence of

ketosis is estimated to range between 2 and 15% of lactating dairy

cows in the United States and Western Europe (11). The major economic

loss results from decreased milk production, and this loss, coupled

with costs of prevention and treatment, was estimated to have reached

$150 million a year In 1981 (110). However, including the subtle

losses during subclinical ketosis would cause total losses to be much

higher.

Although improvements in management of high-producing cows have

decreased the incidence of ketosis, increasing demands on cows by

improved genetic selection for milk production, treatment with

exogenous hormones that stimulate milk production, and increased use

of lesser-quality roughage make it seem unlikely that ketosis can be

eliminated. There is a need, therefore, to characterize the changes

that occur in metabolism during the development of ketosis, thus

enabling cause and effect relationships to be established. Such data would enhance the ability to develop techniques to prevent lactation ketosis. Because both carbohydrate insufficiency and adipose tissue mobilization are common characteristics of both spontaneously ketotic and fasted cows, much research has been done with fasted cows to characterize the metabolic effects of ketosis. Recent success of inducing ketosis gradually in overfed early-lactating cows by using 3

moderate feed restriction and feeding a ketone precursor, as opposed

to fasting, seems to provide an experimental protocol to stimulate

development of on-farm spontaneous ketosls (123).

The present studies were conducted to characterize changes in liver metabolism in more detail than has been accomplished previously in normal and ketotlc cows. An attempt was also made to give cows a possible ketosis-preventative and study liver metabolism. Specific objectives were:

1) to refine and verify the feed-restriction, butanediol feeding

protocol for inducing clinical ketosls in lactating cows,

2) to investigate the effects of duodenal infusions of glucose on

preventing the ketotlc state,

3) to measure capacities of hepatic gluconeogenesis and

lipogenesis in liver slices of normal cows and cows at

several stages of ketosls development and treatment,

4) to measure plasma and hepatic metabolite concentrations for

normal cows and cows at several stages of ketosls development

and treatment, and

5) to measure kinetics of metabolism of rumen propionate and

plasma glucose in normal and ketotlc cows in early lactation. 4

REVIEW OF LITERATURE

This review of literature Is Intended to provide an overview of

lactation ketosls and the physiological Implications of this metabolic

disease. Areas to be discussed are: 1) etiology of lactation ketosls, 2) experimental Induction of ketosls, 3) histological organization of liver, 4) gluconeogenesls In lactatlng cows, 5) lipid metabolism In ruminant liver, 6) ketogenesls In ruminant liver, and 7) kinetics of glucose and propionate metabolism.

Etiology of lactation ketosls

At parturition, marked changes in general metabolism and partitioning of nutrients occur to support the requirements of the mammary gland for milk synthesis. This process is called homeorhesls and is described as orchestrated changes to meet the priorities of a physiological state (22). Changes that occur after parturition include increased feed intake; Increased lipolysls and decreased llpogenesls in adipose tissue; Increased gluconeogenesls, glycogenolysis, and ketogenesls in liver; and increased protein mobilization and noncarbohydrate oxidation in muscle. These changes enable more glucose, fat, and protein to be available for milk synthesis. Normally, homeorhetlc control of milk synthesis causes stress but: not damage to the health of cows. However, the lack of systemic coordination can cause lactation ketosls, which is a 5

metabolic disease. Approximately 4% of cows In the United States and

Western Europe are afflicted with lactation ketosls (11, 49, 110).

Lactation ketosls Is a disorder that occurs early In lactation,

usually between 2 and 8 weeks postpartum. Cows past 75 days

postpartum rarely are affected by ketosls, possibly because of

returning to positive energy balance (144). Ketosls usually is

divided into two stages, clinical and subclinical. It generally is accepted that clinical ketosls is an extension and extreme manifestation of subclinical ketosls (11). Visual signs of clinical ketosls includes a sudden decrease in appetite, decreased milk production, rapid loss of body weight, and lethargy or sometimes hyperexcitability (143). Subclinical ketosls is almost impossible to detect visually because appetite and activity rate remain normal, but some decrease in milk production and an increase in loss of body weight are seen.

Hetabollcally, clinical ketosls is characterized by hyperketonemla, hypoglycemia, hypoinsulinemia, decreased hepatic glycogen. Increased hepatic triglyceride, and Increased nonesterlfled fatty acids (NEFA) in plasma (9, 80). Some cows have a sweet-smelling breath resulting from acetone exhaled from the lungs, thereby providing an excretory route for excess ketones. Although little research has been done with subcllnlcally-ketotlc cows for obvious reasons, data Indicate that they exhibit similar metabolic profiles as ketotic cows, except ketonemia and hypoglycemia are not as severe (12, 6

152).

Ketosls can occur spontaneously, I.e., for no obvious reason, or

as a result of other physiological perturbations. Schultz (145)

classified these as primary and secondary ketosls, respectively.

Curtis et al. (49) recently summarized Incidences of ketosls In 1374

lactations from 31 dairy herds In New York and found a 1.5% incidence

of primary ketosls and a 2.0% incidence of secondary ketosls. Primary ketosls is characterized by the classical signs already discussed but no Increase In body temperature. Secondary ketosls also is associated with the classical symptoms but is accompanied by a rise in body temperature. The Increase in body temperature may be associated with either metritis, retained placenta, nephritis, hardware disease, displaced abomasum, etc. (143). Kronfeld (100) emphasizes the importance of feed type and feed Intake in classifying ketosls. He has defined types of ketosls as primary underfeeding, secondary underfeeding (disease-related), alimentary or ketogenlc ketosls, and spontaneous ketosls. The value of this elaborate identification scheme is questionable.

Even though spontaneous ketosls has been characterized extensively, mechanisms responsible for development of ketosls are not well understood. Cause-and-effect relationships among measured plasma and hepatic metabolites have not been well established. Krebs (96) stated that severe ketosls develops because of limits in homeostatlc mechanisms. Baird (11) eloquently stated that giving metabolic 7

priority to demands of milk production at a time when appetite is

limited may lead to metabolic disorders such as ketosis.

The question remains as to why some cows develop ketosis and some

do not. Schultz (143) suggested a list of predisposing factors that

could lead to ketosis in high-producing cows. These include: a)

increased glucose drain for lactose production, b) endocrine

disorders, c) excess condition at calving, d) hepatic dysfunction, e)

inadequate energy intake after calving, f) deficiencies or excesses in

protein intake, g) mineral or vitamin deficiencies, and h) high intake

of ketogenic materials. The most plausible theory seems to be that a

combination of one or more of these predisposing factors results in a

shortage of carbohydrate. This view is supported by data that showed

that milk production in fasted cows during early lactation decreased

slower than when cows were fasted later in lactation (15). In a

review (11), Baird reported that a 6-day fast will result in severe

hypoglycemia and hyperketonemia before 7 wk postpartum but not after

10 wk postpartum. Thus, control mechanisms in lactation seem to push cows toward carbohydrate insufficiency and ketonemia in order to maintain milk production.

One plausible sequence of events that could occur during a carbohydrate insufficiency now will be presented. Increased milk production causes an increased demand for lactose and, subsequently, glucose. If the liver is unable to meet this demand, hypoglycemia results. In addition, decreased glucose availability in liver results 8

In decreases in amounts of glycogen and citric acid cycle

Intermediates, especially oxaloacetate (OAÂ) and citrate (9). OAÂ Is

important because it is both a precursor for glucose and is required

for the disposal of acetyl CoA (96). A decrease in available OAÂ may

inhibit gluconeogenesis and direct more fatty acid moieties to be

formed into ketones. The hypoglycemia in blood results in decreased

insulin release and a decreased insulin to glucagon ratio, which subsequently would allow greater release of NEFA from adipose tissue

(80, 146). The increased NEFA concentration in plasma would result in an increased influx of NEFA into the liver. In the liver, NEFA can be reesterified and packaged into very-low-density-lipoprotelns (VLDL) or deposited as triglyceride, NEFA also can be oxidized completely to

CO2 in the tricarboxylic acid (TCA) cycle or oxidized incompletely to stimulate ketogenesis. The increased ketogenesis and concurrent hypoinsulinemia can result in the ketonemlc state. Hyperketonemia Is thought to reduce appetite in ruminants (11). The fate of fatty acids in liver will be discussed in more detail in "Lipid metabolism in ruminant liver."

An alternative theory was proposed suggesting the ratio of glucogenic to llpogenic nutrients as being the critical factor (100).

Increased intake of glucogenic nutrients would provide substrates for gluconeogenesis and this supplies precursors for lactose synthesis in the mammary gland. However, because glucose does not contribute carbon to fatty acids found in milk fat (20), a diet with excess 9

carbohydrate would be deficient In precursors for fat synthesis In the

mammary gland, stimulating further llpolysls from adipose and

Increasing the ketonemia. Kronfeld (100) theorizes that the addition of fat to gluconeogenic diets will decreases the incidence of ketosls.

This theory, however, does not seem to account for either the fate of excess lipid not used by the mammary gland or the homeostatlc controls induced by hypoglycemia that are Independent of lipid availability to the mammary gland. Other theories of ketosls development have been proposed involving carnitine and glucocorticoids, and these have been reviewed (53).

As mentioned already, insulin and glucagon may play an important role in the development of ketosls. Insulin decreases after parturition (79) and also influences mobilization of NEFA from adipose tissue. Insulin is llpogenic, and a decrease in Insulin will decrease llpogenesis. Increase llpolysls from adipose tissue, and impair ketone body utilization by muscle (41). Glucagon, conversely, is lipolytic and consequently can be ketogenic (40). Glucagon influences regulation of glucose metabolism by accelerating glycogenolysis and gluconeogenesls in liver (21). The net effect of these two hormones in the cow is thought to be dependent on their ratio in blood plasma.

Insulin is a primary modifier of lipid and ketone body metabolism, and glucagon is secondary (41). While Insulin decreases after calving, glucagon is thought to increase (54, 113), thereby decreasing the insulin to glucagon ratio and stimulating llpolysls. deBoer et al. 10

(55), also showed that cows had increased plasma glucagon

concentrations after calving. Glucagon, however, has been shown to

decrease in spontaneously ketotic cows (35) and in feed-restricted

cows that were ketonemic (55). Changes seen in insulin and glucagon

concentrations during ketosis seem to support the existence of

hypoglycemia and hyperketonemia.

Treatment of ketosis has been successful, and the rationale centers upon an endeavor to increase the rate of supplying glucose to the cow. The main therapeutic measures used today include: a) subcutaneous or intravenous administration of glucose, b) oral administration of propionate or propylene glycol, and c) intramuscular injection of glucocorticoids (145). It should be noted that administration of glucose or propionate gives similar results, but glucose and propionate act in different ways metabolically.

Administration of glucose provides an immediate source of glucose to alleviate hypoglycemia and increase carbohydrate sufficiency, which is thought to ameliorate ketosis. Glucose administration in blood, however, actually decreases hepatic gluconeogenesis in normal, lactating cows (18). Administration of propionate provides a gluconeogenic precursor to stimulate gluconeogenesis and Increase hepatic output of glucose. Recently, nicotinic acid has been proposed as a treatment of ketosis because of its anti-lipolytic properties and its ability to Increase insulin concentrations (158).

Prevention of ketosis has focused on Improved feeding of 11

high-producing cows. To this end, Schultz (145), Baird et al. (16),

and Hibbitt (77) have provided guidelines to reduce the incidence of ketosis. It is not known, however, whether such guidelines decrease

incidences of subclinical ketosis as well as decreasing the more severe clinical ketosis. Further research is necessary to determine whether improving carbohydrate sufficiency in early lactating cows will reduce incidences of subclinical ketosis.

Experimental induction of ketosis

Detailed and controlled studies of the development of ketosis are severely limited by the lack of predictable and consistently-available ketotic cows. Much time and expense has been incurred to study ketotic cows. Some experimental approaches include sampling entire herds of cattle (115), relying on veterinarian-reported cases of on-farm ketosis (140), and random sampling of herds in local areas to find ketotic cows (91). Such studies result in significant experimental error from variable feeding practices and animal handling. Further, cows usually are sampled after detection of clinical ketosis, and samples do not provide a complete picture of the events leading to development of ketosis. Experimental procedures are needed that will cause cows to develop ketosis in a manner similar to on-farm ketosis and whereby feed intake, sampling, handling, and treatment of cows can be controlled.

The predisposing factors for development of ketosis in cows (143) have been the basis for ketosis-induction protocols. Briefly, these 12

factors include inadequate energy Intake after calving, excess Intake

of protein, high intake of ketogenlc materials, endocrine disorders,

and excess condition at calving. An experimental ketosls can be

produced easily in sheep by starvation during pregnancy (87, 131).

However, one of the first studies using starvation of cows to Induce

ketosls resulted in ketonemla but no obvious signs of ketosls (137).

Cows were starved for 6 days, milk production was decreased, and

appetites were not affected when cows were returned to pre-starvation

feeding levels. Since then, Baird et al. (15) used the starvation

technique to compare normal lactating, induced ketotlc, and

spontaneously ketotlc cows. Hlbbitt and Baird (78) also showed that

changes in blood and liver were similar between starvation ketosls and

spontaneous ketosls. The 6-day fast has been used extensively to

study ketosls and fatty liver, which is a related disease (15, 45,

132, 134, 135).

Increase of thyroxine have been implicated as increasing the

stress level of early lactating cows. Hlbbitt (76) used this suggestion to induce cows into ketosls by feeding a high-protein diet and injecting thyroxine from 6 weeks prepartum to 2 weeks postpartum.

Kellogg et al. (93) improved this method by feeding a ration that met requirements for the first 9 wk postpartum, then changing to a high-protein, low-energy ration and injecting thyroxine for 10 days.

Although 9 wk postpartum seems late for ketosls (144), cows decreased milk production rapidly, exhibited hyperketonemla, and most developed 13

visual signs of ketosls. Mills et al. (123) took a different approach

by combining moderate feed restriction with feeding 1,3-butanedlol.

In steers, 1,3 butanedlol Is at least partly oxidized to beta-hydroxybutyrate In the liver (121), and acetoacetate and beta-hydroxybutyrate concentrations In plasma are increased when

1,3-butanedlol Is fed at 4% of the diet (39, 74). Thus,

1,3-butanedlol seems to be an effective ketogenlc agent. A more complete review of 1,3-butanedlol and Its metabolic effects has been published (112). The four of five cows subjected to feed restriction and 1,3-butanedlol by Mills et al. (123) developed clinical ketosls at an average of 24 days after beginning of Induction. They exhibited hyperketonemla, hypoglycemia, Increased liver triglycerides and plasma

NEFA, decreased liver glycogen, and decreased In vitro hepatic gluconeogenic capacity (123, 124).

Some evaluation of the usefulness of each of these induction methods is appropriate. All of the methods seem to produce clinically or near-clinically ketotlc cows and therefore are useful for characterizing the metabolic condition of ketosls. The current need is to understand the development of ketosls and herein lies some concern with Induction protocols. Starvation has been successful in mimicking the biochemical and metabolic conditions of spontaneous ketosls (45, 78), but some questions exist about whether fasted cows show a true clinical ketosls. Also, the time course of a 6-day fast does not coincide with the development of spontaneous ketosls. 14

Spontaneous ketosls at times develops quickly (2 wk postpartum) but

usually develops more slowly, preceded by subclinical ketosls (145).

Therefore, It seems that starvation Is not an Ideal method to study

the development of on-farm ketosls, but Is acceptable to characterize

metabolic changes at the time of ketosls detection.

Methods utilizing feed restriction and thyroxine injection avoid

the drastic changes associated with complete feed withdrawal.

However, the time course for development of ketosls again is very

short and may not be representative of the development of on-farm

ketosls. Combining feed restriction with feeding 1,3-butanediol seems to show more promise. Plasma ketone body concentrations are Increased without a severe decrease in feed intake, and clinical ketosls seems to develop gradually without an immediate decrease in milk production

(123).

In summary, each induction method utilizes one or more of the suggested predisposing factors for ketosls (143) to trigger the disease state of ketosls. It is very Important to note that any method of induction "fixes" a given condition, such as lack of gluconeogenic substrates during starvation, and makes Interpretation of cause and effect difficult. Even with accepted limitations, ketosls-induction protocols, however, are a powerful tool with which to study the metabolic implications of lactation ketosls in a controlled environment. 15

Histological organization of liver

Histological examination of liver tissue from cows Is useful to

study structural changes that ketosls causes In hepatocytes. The

occurrence of changes In hepatic metabolite concentrations and

gluconeogenic capacity during ketosls (124) and in hepatic lipid

accumulation during fatty liver (132) raises the question of whether

structural changes of liver contribute to development of ketosls or

are they a result of clinical ketosls. A brief review of the

histological organization of liver is needed to provide a basis for

discussion.

The ruminant liver functions as an "exchanger" because it has an exocrine function and also synthesizes and stores nutrients for future use (60). The exocrine function centers upon detoxifying and packaging harmful products and by-products originating from metabolic processes thoughout the body. Synthesis and storage functions include synthesis of glucose for energy and as a precursor for milk lactose, glycogenesis, ketogenesis, and reesteriflcation of fatty acids into triglycerides. The liver Is structured to accomplish these functions.

Liver is composed primarily of parenchymal cells, which make up

80% of its volume and account for about 60% of its total number of cells (50). Blouin (38) subdivided rat liver parenchyma into epithelial hepatocyes (78% of total parenchyma), sinusoidal cells

(6%), fat-storing cells (1%), and spaces between the parenchymal 16

Terminal HeoaHcV Termhal Henatic V VCantral V), C\\ (CenfraJVJ

< Rwtal Space Brancher of RjrjalY and HepaKc A.

Sinusofds

Figure 1. The liver acinus concept of the functional unit of liver parenchyma

cells (15%). Epithelial hepatocytes are bathed with oxygenated blood

from the portal vein via sinusoids. Sinusoids are larger than

capillaries and are lined with sinusoidal cells and fat-storing cells.

Sinusoids are shown in Figure 1. There seems to be a directed flow of

blood through the sinusoids from the portal vein to the central vein

of liver (141).

Hepatocytes are mononucleated, although some are binucleated in rat liver (141), and serve as the primary site for 17

endocrine-controlled functions of the liver. In hepatic cytosol,

mitochondria are numerous and are estimated to make up 20% of the

cytoplasmic volume (111). Other major structures in the cytoplasm

Include the endoplasmic reticulum, lysosomes containing acid

phosphatase and hydrolase enzymes, and peroxisomes containing

peroxidase and catalase (141). Sinusoidal cells and fat-storage cells

vary In function, but can be characterized as having phagocytic

capacity (162), fat-synthesizing capacity (44), and lymphocytic

properties (86).

Understanding the architectural structure of the functional

arrangement of liver hepatocytes Is vital when discussing changes In

composition and function during ketosls. The most widely accepted

concept of liver Is the "liver acinus", first described by Rappaport

et al. (130) and later revised by Rappaport (129). The liver acinus

concept describes liver as adjoining functional units of parenchyma, known as lobules, oriented around the terminal afferent vessels

(portal vein and hepatic artery. Figure 1). These vessels plus the common bile duct have been called the portal triad. Blood flows from the portal triad to the periphery of the acinus and the terminal hepatic venule, formerly called the central vein. Figure 1

Illustrates the liver acinus concept and is redrawn from Rappaport et al. (130). Blood passing through the acinus gradually becomes less oxygenated. Three different zones have been proposed for this process

(Figure 1): the periportal zone (zone 1), which receives highly 18

oxygenated blood; the intermediate zone (zone 2); and the perivenous

or pericentral zone (zone 3), which receives blood after it exchanges

with cells in zones 1 and 2.

Studies of the functional areas of liver have expanded the theory

of structural zonatlon to include metabolic zonatlon (84), which has

been summarized by Fawcett (60). Using Figure 1 as an example, zone 1

is the "zone of permanent function," zone 2 is the "zone of varying

activity," and zone 3 is the "zone of permanent repose." Classifying

metabolic zones of liver is based upon the availability of oxygei\ for

cellular function. After the ingestion of a large meal in

monogastrlcs, glycogen first appears in the periportal region and

progressively fills other regions, reaching the pericentral region

during extreme overeating. This suggests that gluconeogenesls Is most

active in the periportal region of the liver. Studies have shown that

mitochondrial size and numbers are nearly double in the periportal versus the pericentral region (111).

Fawcett (60) describes that fat accumulation in the liver, during both pathological and normal physiological conditions, begins around the central vein in small droplets that gradually enlarge and the cell becomes distended. During dietary deficiencies, fat accumulation can extend to the periportal region of the liver. Such zoning of fat accumulation has been described for ruminants by Slmesen and Moller

(149). Olson and Thurman (125) recently studied hepatic ketogenesis

In periportal and pericentral regions of the rat liver. When 19

bromo-octanoate, which inhibits the enzyme ketothiolase I and thus is

a potent inhibitor of beta-oxidation, was infused, they found that

ketogenesis from endogenous long-chain fatty acids was nearly equal

for pericentral and periportal regions of liver of fasted rats.

However, when oleate was infused, ketogenesls was stimulated nearly

60% more in pericentral than in periportal regions. It is possible

that the zonal differences in ketogenesls Influence zonal differences

in triglyceride accumulation.

Gluconeogenesis in the lactating cow

Reviews of ruminant metabolism clearly have Identified gluconeogenesis as a major function of liver hepatocytes (106, 108,

168). Studies using sheep (7, 168) and steers (126) have established that the liver plus the kidneys synthesize at least 90% of the glucose used by the body dally. Glucose can be oxidized completely to CO2 through glycolysis and the TCA cycle, yielding ATP molecules through the electron transport pathway (104). Glucose therefore is a readily available source of energy for metabolic processes in the body, especially those in the nervous system, brain, and fetus. Other

Important fates of glucose include serving for a precursor for: a) glycogen in hepatic and muscle cells, b) glycerol-3-phosphate for llpogenesls, and c) lactose synthesis in the mammary gland during lactation (28). Gluconeogenesis obviously is of vital Importance for a lactating cow. Understanuing changes in gluconeogenesis during ketosls then may be a key to understanding and preventing the 20

development of both subclinical and clinical ketosls In cows. To

facilitate a knowledgeable discussion of gluconeogenesls relative to

ketosls, some review of the biochemical pathways, precursors, and

regulation of gluconeogenesls Is appropriate.

Gluconeogenic pathways The biochemical pathway of glucose

synthesis in the liver contains both cytosollc and mitochondrial

reactions. All glucose precursors except glycerol must enter the TCA

cycle, which is confined to the mitochondria. If citrate (formed from

OAA and acetyl CoA) is considered the starting point of the TCA cycle,

then the end products are CO2, NADH, FADH + H, and OAA. OAA serves

two functions: 1) it can be transported out of the mitochondria in

the form of malate, reoxidized to OAA via malate dehydrogenase, and be

converted to phosphoenolpyruvate (PEP) in the cytosol (104), or 2) it

can function as an acceptor of acetyl CoA to facilitate complete

oxidation of fatty acids and carbohydrates (28). PEP formed from OAA

in cytosol is converted to glucose by reverse glycolysis (104). A

limited availability of OAA in mitochondria therefore will decrease

gluconeogenesls as well as decrease oxidation of fatty acids.

Within the pathways of glucose synthesis, four reactions are

thought to be rate-limiting, each with a key enzyme whose activity

could control the rate of gluconeogenesls (28). These enzymes are: 1)

glucose-6-phosphatase (G6Pase), 2) fructose-l,6-diphosphatase

(FDPase), 3) pyruvate carboxylase (PC), and 4) PEP carboxyklnase

(PEPCK). High activities of both G6Pase and FDPase are Important in 21

driving the pathway toward glucose synthesis because they provide for

release of free glucose into plasma. PC activity is essential for the

conversion of pyruvate, lactate and certain amino acids to OÀÀ.

Finally, PEPCK catalyzes the conversion of OAA to PEP in cytosol and

therefore controls gluconeogenesis from all precursors except glycerol

(59).

How these four enzymes are affected by hypoglycemia and ketonemia in early-lactation cows has been studied but is not well understood.

Baird et al. (14) showed that activities of PC, PEPCK, and FDPase were not different between ketotic and normal cows. Other work also has shown that activities of PEPCK (19, 20), G6P (78), and FDPase (71) in liver are not affected by ketosis. Ballard et al. (19) did show an increase in PC activity in liver of ketotic cows, however, compared to liver of normal cows. Increased PC activity would direct more lactate and pyruvate towards glucose synthesis in early lactation cows. The lack of changes in key liver enzymes during ketosis suggests that the supply of gluconeogenic precursors is more important in maintaining concentrations of TCA cycle intermediates than is enzyme activity.

Mills et al. (124) reported that gluconeogenic capacity of liver decreased significantly during an experimental ketosis. The experimental ketosis was induced in part by feeding 1,3-butanedlol, which seems to be metabolized by the liver. Â uniform 66% decrease in gluconeogenesis from propionate, lactate, and gluconeogenic amino acids suggests that the hepatic capacity for gluconeogenesis is 22

regulated at the OAA to PEP step or even later. Mills et al. (123)

also showed a decrease in concentrations of hepatic citrate during

ketosis when compared with prepartum and recovery stages. The

decrease was only slightly more than that seen at two weeks

postpartum, however. More work on characterizing changes in

gluconeogenic capacity and in enzyme activities during "spontaneous"

on-farm ketosis is needed before further conclusions can be drawn.

Gluconeogenic precursors Precursors for gluconeogenesis are supplied by four main substrates in ruminants: 1) propionate from the rumen, 2) glycerol, 3) amino acids, or 4) lactate, which subsequently is oxidized to pyruvate (59, 95). Propionate absorbed from the rumen is considered to be the major gluconeogenic precursor in fed ruminants and is the only major volatile fatty acid (VFA) produced in the rumen that contributes carbon to gluconeogenesis (142, 159). About 90% of absorbed propionate is taken up by the liver during the first pass through the portal vein (18, 30). Propionate is activated to propionyl CoA, carboxylated to methylmalonyl CoA, and can enter the

TCA cycle as succinate. Acetate, however, is taken up by the liver at low rates and only provides acetyl-CoA, which does not contribute to the net synthesis of OAA or glucose (28). Butyrate has been shown to stimulate glycogenolysis resulting in hyperglycemia, but it cannot contribute net carbon for carbohydrate synthesis (37). Studies of propionate metabolism have allowed estimates that, in well-fed, non-pregnant sheep, theoretically propionate could provide 100% of the 23

dally requirement of glucose. However, studies have consistently

shown that only 27 to 60% of glucose Is derived from propionate In

sheep (83, 107, 151). Wlltrout and Satter (161) found that 45% of

glucose was derived from propionate In lactatlng cows. A majority of

the remaining propionate Is oxidized to CO2 In the liver (156).

Although propionate Is not used entirely for glucose synthesis, additional propionate stimulates gluconeogenesls. An Indirect measurement of this stimulation was presented by Jurenkova et al.

(85). They treated cllnlcally-ketotlc cows with propionic acid and glucose concentration Increased. Studies with the lonophore monensln

In cattle suggest again that Increasing propionate availability will stimulate gluconeogenesls (51, 155). Stimulation of gluconeogenesls by propionate was quantified in a recent study by Veenhuizen (156).

Steers being fed at maintenance were supplemented with dietary sodium propionate at amounts Intended to double propionate availability to the liver. Kinetic analyses of propionate and glucose metabolism in a four-pool model revealed that adding 6.3 mol proplonate/d to the diet resulted in an increased production of 1.3 mol glucose/d. Also, the percentage of glucose derived from propionate Increased from 43 to

67%. Thus, stimulation of gluconeogenesls by propionate In steers seems to be dependent only on concentration, and is not indicative of hormonal or enzymatic regulation. Therefore, increases seen in gluconeogenesls from propionate in lactatlng cows (161) may be primarily a result of increased feed Intake. 24

Glycerol is found primarily as the backbone of triglycerides In

adipose tissue and concentrations of glycerol In plasma are low (32).

In early lactation when Insulin Is decreased, the resulting llpolysls

will greatly increase the availability of free glycerol (68). Free

glycerol In plasma can be taken up by the liver and converted to

glycerol-3-phosphate, thereby entering the gluconeogenic pathway at

the step before fructose 1,6-diphosphate Is made (104). Glycerol is

the only precursor of glucose that can enter glucose without being

involved in some reactions that are a part of the TCA cycle. Lindsay

(108) estimated that glycerol only contributed 5% to gluconeogenesis

in fed ruminants. However, in fasted and ketotic sheep, up to 30% of

glucose irreversible loss was accounted for by glycerol (32). Also,

Ranaweera et al. (128) Infused large amounts of glycerol into fasted

sheep and found that 91% of total glucose entry rate was derived from

glycerol. These data show that, in the absence of other precursors,

glycerol can provide substrates for glucose in large amounts and

gluconeogenesis from glycerol seems to be dependent upon the concentration of glycerol in plasma. The contribution of glycerol to glucose most likely will remain low during Increased llpolysls in lactating cows because of the need for glycerol moieties for milk fat synthesis.

Contributions of amino acids to glucose synthesis are important, but variable, depending upon physiological state and nutritional status (28). Trenkle (153) suggested that up to 50% of glucose could 25

be derived from amino acids In ruminants. Although In vivo kinetic

experiments often are difficult to Interpret because of cross-over of

label (136), Wolff and Bergman (163) used In vivo kinetic techniques

to estimate that 11 to 30% of glucose irreversible loss is derived

from amino acids in fed sheep. Rellly and Ford (136) reported a 28%

contribution in sheep, and Egan and Black (56) estimated that between

30 and 50% of glucose is derived from lactate, glycerol, and amino

acids combined in maintenance-fed steers. Therefore, it seems that

about 30% of the glucose used is synthesized from amino acids, but

specific changes occurring during early lactation and ketosis have not

been quantified.

In early lactation cows, plasma glucagon concentration increase, which could stimulate amino acid uptake and protein degradation in the liver and in peripheral tissues (21). This availability of amino acids, coupled with Increased propionate from the rumen (161), could explain why no changes occur in the relative contributions of amino acids to gluconeogenesis during lactation. Individual amino acids used for gluconeogenesis listed in descending order of contribution are: 1) alanine, 2) glutamate, 3) aspartate, 4) serine, and 5) glycine (164). Alanine, serine, and glycine enter the gluconeogenic pathway via pyruvate; glutamate enters through alpha- ketoglutarate of the TCA cycle; and aspartate enters directly as OAA.

Pyruvate is the end product of aerobic glycolysis and in itself is not a net precursor of glucose. However, pyruvate can be 26

carboxylated to form OÂA in a reaction catalyzed by pyruvate

carboxylase (165). Because pyruvate is formed from lactate and some

amino acids, it is considered gluconeogenic. Lactate in plasma can originate from anaerobic reactions In the rumen and muscle. No net synthesis of glucose can occur from muscle lactate because it is formed from glucsoe and therefore represents recycling, which is known as the Cori cycle. Lactate from degradation of carbohydrates in the rumen does contribute to net glucose synthesis because it is derived from feedstuffs. Lindsay (109) reported that up to 28% of the glucose synthesized by liver and kidney originated from lactate in ruminants.

However, studies with fed sheep (33) showed only a 4 to 10% contribution of lactate, and in starved sheep there was only about a

15% contribution of lactate to glucose (3); however, Cori cycle recycling was not corrected.

If all the maximum contribution to glucose of the gluconeogenic precursors Just listed are summed, the result is about 130%.

Therefore, it is obvious that these precursors do not all contribute at maximum rates. Propionate generally is the predominant precursor, and infusions of propionate can decrease the percentage contribution of amino acids to glucose synthesis (31, 82). Glycerol and lactate are less important for gluconeogenesis, but they can become important when propionate and protein supplies from the rumen are decreased by either anorexia or fasting.

Regulation of gluconeogenesis Glucose synthesis is regulated 27

by substrate availability, hormone secretion, and enzyme activities.

Controls exerted by enzymes in the gluconeogenic pathway have been

discussed already. Substrate availability is primary for regulating

gluconeogenesis; whereas rates of hormone secretion are dictated by

physiological state (22) and substrate supply (36). Therefore, hormones act as secondary regulators of gluconeogenesis.

Substrate supply is both a potent stimulator and an Inhibitor of

gluconeogenesis. Positive correlations between digestible energy intakes and glucose-entry rates have been reported by Leng (106),

Lindsay (108), Steel and Leng (151), and Herbein et al. (73). Studies where propionate was infused into sheep (82) have shown a consistent increase in glucose turnover rate. Yost et al. (166) also showed increases in glucose turnover with increases in feed intake.

Veenhuizen (156) fed 450 g/d of propionate to steers and found an increase of 233 g/d in glucose production. Glucose infusions, however, inhibit gluconeogenesis. Judson and Leng (82) found decreases in glucose turnover in proportion to the amount of glucose infused Intravenously into sheep. Baird (10) found that hepatic release of glucose was decreased during intravenous infusions of glucose in cows. Not much is known, however, about the effects of exogenous glucose absorbed from the intestine. An experiment to evaluate the effects of intraduodenal and intravenous infusions of glucose on glucose turnover rate in cows currently is being completed by M. H. Cooley, Animal Science Department, Iowa State University, 28

Ames, lA.

Fasting Is an inhibitor of gluconeogenesis because substrate

availability becomes limiting. Lindsay (108) reported a 33% decline

in glucose production after a 24 h fast, and Katz and Bergman (88) reported a 50% decline in glucose production after sheep were fasted for 3 days. Although most of the in vivo glucose turnover research has been done with fasting sheep, results should be applicable to cows because of similar rumen metabolism.

Insulin and glucagon are the two most important hormones Involved in regulation of gluconeogenesis in ruminants (42); although glucocorticoids, thyroxine, and growth hormone are Involved also.

Insulin, being antillpoytlc, primarily acts to Increase peripheral uptake of glucose from plasma. Increases in insulin concentration in plasma inhibit hepatic gluconeogenesis, stimulate adipose tissue lipogenesls, and Increase net synthesis of proteins from amino acids

(21) by decreasing protein breakdown. Therefore, decreases in insulin concentration seen In early lactation in cows provide a regulation that is favorable for maximal rates of gluconeogenesis. Glucagon acts in opposition to insulin by stimulating glycogenolysls and gluconeogenesis, by stimulating amino acid uptake by the liver, and by increasing the rate of protein catabollsm by the liver (21). Whereas insulin primarily affectsadipose tissue, glucagon primarily affects hepatic tissue. Glucagon, however, does not seem to affect the relative proportion of propionate used for glucose synthesis (43). 29

Concentrations of glucocorticoids are Increased during fasting

(122) and ketosls (28) In cows. The effect of glucocorticoids may be

to spare glucose utilization by peripheral tissues or to Increase the

supply of glucose precursors (13). Glucocorticoids can Increase the

supply of glucose precursors by Increasing protein catabollsm In

muscle and thereby providing amino acids for TCÀ cycle Intermediates

even during an energy deficit. Plasma thyroxine concentrations have

been shown to decrease in early lactation and during ketosls.

Injections of thyroxine can Increase gluconeogenesls in lactatlng cows

by directing more substrates to glucose (72). Injections of thyroxine

have been used to Induce ketosls by increasing the drive to produce

glucose when substrates are limiting, thereby enhancing the

carbohydrate deficiency in liver that occurs early in lactation (93).

Growth hormone affects gluconeogenesls indirectly by antagonizing effects of insulin on llpogenesis in adipose tissue (21). Growth hormone decreases peripheral utilization of glucose, thereby providing substrates for the mammary gland and exerting somewhat of an inhibitory effect on hepatic gluconeogenesls.

Lipid metabolism in ruminant liver

Fat mobilization, or llpolysis, from adipose tissue in early-lactation cows seems to be the direct result of an Increased need for milk fat precursors coupled with an Inadequate Intake of energy (157). Llpolysis seems to be regulated partly by both insulin and growth hormone. Insulin is antllipolytlc and growth hormone 30

stimulates llpolysls (21). Because Insulin and growth hormone act In

opposition to each other, the insulin to growth hormone ratio is

important for regulation of adipose tissue metabolism (21).

Therefore, the decrease in insulin and the increase in growth hormone

seen during early lactation in cows stimulate increased lipolysis and

result in increased NEFA and glycerol release into blood plasma.

During ketosis, excessive lipolysis can lead to greatly increased

rates of NEFA influx into plasma.

Liver is a major site of NEFA uptake as shown by Bergman et al.

(34) for sheep where 25% of plasma NEFA were taken up by liver. NEFA

uptake by liver is not regulated by metabolic events in the liver

(24). When plasma NEFA concentrations increase, uptake of NEFA by

liver increases proportionally. Sheep liver also has been shown to take up small amounts of triglyceride fatty acids from plasma (34).

The short-chain fatty acid propionate is taken up readily by the liver, however, acetate uptake is very low (48, 114).

Intrahepatic NEFA first must be activated by acylation to CoA derivatives at the expense of ATP. Long-chain acyl CoA synthetase activity has not been measured directly in ruminants. In non-ruminants, however, the enzyme is associated with the endoplasmic reticulum and the outer mitochondrial membrane, and its activity exceeds that of beta-oxidation in mitochondria and/or esterification with alpha-glycerolphosphate (64). After activation, fatty acids have three fates: 1) complete oxidation to CO2 through the TCA cycle, 2) 31

Incomplete oxidation to ketone bodies and acetate, and 3)

reesterlflcatlon Into triglycerides.

For fates one and two Just mentioned, fatty acyl CoA derivatives

are transported Into mitochondria In the presence of carnitine. The

reaction Is catalyzed by carnitine acyltransferases I and II.

Beta-oxidation of long-chain fatty acyl CoA molecules yields

acetyl-CoA molecules for further metabolism. Acetyl-CoA can react

with OAA in the mitochondria to form citrate in the TCA cycle,

resulting in the complete, oxidation of acetyl-CoA to CO2 (104). As

discussed already, OAA also serves as an Important intermediate in the

gluconeogenic pathway, and its low concentration in mitochondria may be a crucial factor in the development of ketosls (28). Acetyl-CoA also can be converted into ketone bodies and acetate, with these being released into plasma. Ketone body production is defined as ketogenesls and will be discussed in more detail later with respect to the normal and pathological states of metabolism. De novo fatty acid synthesis, another fate of acetyl-CoA, has been shown to occur only at low rates or not at all in ruminant liver (20). The rate of llpogenesis from acetate and glucose in ruminant liver has been shown to be much lower than that of adipose tissue (33, 66). Although reasons are not fully understood, it has been suggested that an inability to activate acetate to acetyl-CoA (8) and a lack of extramitochondrial acetyl-CoA (24) may contribute to the lack of de novo fatty acid synthesis in ruminant liver. 32

The third fate of fatty acids, reesterlflcatlon Into

triglycerides, is thought to occur at significant rates in ruminant

liver. Bovine liver homogenates were found to reesterlfy fatty acids

into mono-, di-, and triglycerides as well as phospholipids in

proportion to hepatic fatty acid concentration (26). The rata of

fatty acid reesterlflcatlon, however, was low in sheep liver slices

(24). One hypothesis for low rates of fatty acid reesterlflcatlon is

a limited availability of glycerol-3-phosphate (154). Studies of

fatty liver in cows, however, show that lipid accumulation is mostly

in the form of triglyceride and rate of incorporation of fatty acids

into triglyceride may increase when uptake of NEFA by liver is

increased (45). Therefore, it seems that fatty acid reesterlflcatlon occurs at a rate compatible with the need to metabolize NEFA in the liver.

An important function of liver is to package fatty acids into lipoproteins for release into plasma after they have been reesterlfied. The very low density lipoproteins (VLDL) are the primary carriers of triglycerides in nonruminants (57). In ruminants,

VLDL concentrations in plasma are very low (47), but turnover rate is quite high (62). Thus, VIJ>L are thought to be important in ruminant lipid metabolism also. Bell (24) presented in his review that VLDL secretion in the rat is directly related to NEFA uptake by the liver but that the liver's ability to secrete VLDL is less than its ability to take up NEFA. Therefore, when NEFA uptake increases greatly in 33

cows during early lactation and ketosls, uptake of fatty acids may

exceed the maximal rate of secretion of VLDL and cause triglyceride to

accumulate In liver. Fatty acids then must be metabolized through

another pathway. Bergman et al. (34) support this Idea by showing

that very little labelled NEFA taken up by liver In fasted sheep

reappeared In circulating VLDL triglyceride fatty acids. Reld (133) . found that 66% of Holsteln-Frleslan cows sampled In England had

moderate or severe fatty liver at one week after calving. Many

questions concerning VLDL synthesis and secretion still are unanswered. There Is some evidence that symptoms of ketosls and associated changes In serum lipoproteins can be alleviated by treatment with methionine, which seems to result In Improved apoprotein synthesis and export of hepatic lipids (116).

Ketogenesls In ruminant liver

Ketogenesls In the liver Is described as the production and release of ketone bodies. The tenn ketone bodies refers to acetone, acetoacetate (ACAC), and beta-hydroxybutyrate (BHBA). Acetyl-CoA Is the building block of ketone body synthesis and can form ACAC via the formation of a 6-carbon Intermediate, hydroxymethylglutaryl-CoA

(HMG-CoA). This pathway therefore Is known as the HMG-CoA cycle

(120). ACAC can be reduced to BHBA, yielding NAD from NADH + H.

Acetone Is formed from the nonenzymatlc decarboxylation of ACAC, and its concentration In plasma Is usually low (120). However, during hyperketonemla In lactatlng cows, acetone increases In concentration 34

in blood and Is excreted through the lungs, resulting In a

sweet-smelling acetone breath.

Alimentary ketogenesls normally is the primary source of ketone bodies in fed, non-lactating ruminants. Butyrate, and to a lesser extent acetate, can be converted to BHBA and ACAC during absorption across the rumen epithelium (105). A major difference between alimentary and hepatic ketogenesls is the compartmentalization of beta-hydroxybutyrate dehydrogenase, an enzyme that catalyzes the reduction of ACAC to BHBA. In rumen epithelium the enzyme is 95% mitochondrial (105); whereas, in the liver, it is 95% cytosollc (94).

The implications are unclear, however it has been suggested that this difference contributes to the decrease in the ratio of BHBA to ACAC when fatty acid uptake is Increased during ketosis in cows (70).

In fed ruminants, the major source for hepatic ketogenesls is absorbed butyrate that escapes metabolism by the rumen epithelium

(24). Baird et al. (17) reported that 12% of the nonlactating dairy cow's digestible energy intake is metabolized through ketone bodies.

Heitmann et al. (70) calculated that 7% of the energy requirement of sheep was met by alimentary ketogenesls. Assuming these two results are comparable, almost 60% of the ketone bodies utilized by nonlactating, fed ruminants came from alimentary ketogenesls.

In both early lactation and fasted ruminants, hepatic ketogenesls becomes primary and alimentary ketogenesls Is secondary. Long-chain fatty acids are the major source of carbon for ketogenesls and are 35

oxidized to acetyl-CoA In hepatic mitochondria (120). As discussed

already, cows in early lactation or fasted cows have increased rates

of llpolysls from adipose tissue as partly Influenced by decreases in

insulin concentration and the Insulin to glucagon ratio. The

increased Influx of NEFA into the liver then is thought to saturate

the reesterlflcation pathway and the remaining NEFA are oxidized to

acetyl-CoA (24). Ketogenesls then becomes the primary route of

metabolism for acetyl-CoA, especially during fasting and ketosis.

An alternative to hepatic ketogenesls in ruminants is acetate

production and release from the liver. Bell (24) reviewed and summarized acetate release from liver for fed and fasted sheep and for lactating cattle and found a substantial and consistent release in vivo. Snoswell et al. (150) found that the enzymatic capacity of liver in vitro could account for the observed in vivo release of acetate in cows. The availability of free carnitine has been suggested to affect the balance between ketogenesls and acetate production, with increasing free carnitine favoring acetate production

(24). Therefore, endogenous production of acetate In liver could provide an end product of acetyl-CoA metabolism during fatty acid oxidation. Kronfeld (97, 99) has suggested that acetate release is

Increased during spontaneous ketosis, but Is decreased during fasting ketosis in lactating cows.

The fate of acetyl-CoA in ruminants suggests two forms of regulation of ketogenesls: 1) substrate supply, and 2) hepatic 36

adjustments that reroute the metabolism of fatty acids (118, 119). It

has been theorized that the greatly increased influx of NEFA

stimulates more fatty acids to go to ketogenesis and fewer to enter

the TCA cycle (24). This theory implies that TCA cycle activity is

inhibited, especially the activity of citrate synthase. Data from

Bergman (27) support this theory by showing that ketone body concentrations in sheep plasma increased exponentially with increasing

NEFA concentrations. However, the increase of NEFA metabolized to ketone bodies also has been described as an "overflow" process that results from acetyl-CoA concentrations that exceed the capacity for citrate synthesis in mitochondria, especially if the availability of oxaloacetate is decreased (24). Krebs (96) described conditions where deficiencies of TCA cycle intermediates, especially oxaloacetate, could occur when gluconeogenesis was stimulated and substrates were limiting. Thus, if the supply of NEFA increases and that of TCA cycle intermediates decreases simultaneously, as is likely during early lactation and ketosis, more NEFA can be directed into ketone bodies rather than into the TCA cycle. McGarry and Foster (117), in discussing the roles of the TCA cycle and fatty acid synthesis, suggested that decreased TCA cycle activity may result from increased ketogenesis, but it is not a prerequisite.

Discussion of hormonal effects on hepatic ketogenesis centers on

Insulin, growth hormone, and glucagon. As discussed already, decreases in concentrations of insulin and increases in growth hormone 37

stimulate llpolysls and Increase release of NEFÀ into plasma. In

nonruffllnants, insulin also has been shown to function intrahepatlcally

to slow ketogenesls by diverting fatty acids towards reesterificatlon

and away from oxidation (120). Conversely, glucagon has been shown to

be ketogenlc (120), and some Investigators have suggested it functions

to direct fatty acids towards oxidation and away from reesterificatlon

(70). This regulation may occur in ruminants also (1).

Ketone bodies have been shown to have autoregulatory properties.

Infusions of beta-hydroxybutyrate into sheep at amounts to simulate

maximum uptake (29) resulted in decreased concentrations of plasma free fatty acids, decreased uptake of hepatic fatty acids, and decreased ketogenesls. In two recent studies (52, 69), this effect of beta-hydroxybutyrate was shown to occur simultaneously with Increased concentrations of insulin and Increased production of pancreatic insulin, thereby suggesting that the autoregulatory function of ketone bodies is insulin-dependent. However, Sensenig et al. (147) recently reported that infusion of beta-hydroxybutyrate into untreated diabetic ewes inhibited ketogenesls from oleic but not from octanolc acid with no changes in production of pancreatic insulin. Because intramitochondrial transport of octanolc acid is independent of carnitine acyl transferase I (CAT-I), results from Sensenig et al.

(147) suggest that an autoregulatory mechanism that is independent of insulin occurs at the CAT-I level (70). Such a mechanism could be

Important for the regulation of ketogenesls in high-producing 38

lactatlng cows because low insulin concentrations are correlated

positively to milk production (36, 67).

A final point of discussion relating to ketogenesis in ruminants is the extrahepatic utilization of ketone bodies. Ketone bodies are released from the liver into the blood and can be utilized by other tissues as an energy source by converting them back to acetyl-Co for subsequent complete combustion to CO2 in the TCA cycle. Tissues that utilize ketone bodies are: heart, kidney, skeletal muscle, and the mammary gland (70). The key step in forming acetyl-CoA from ACAC and

BOHB is the activation of acetoacetate to acetoacetyl-CoA in a reaction with succinyl-CoA.

Utilization of ketone bodies is thought to be controlled in both ruminants and nonruminants by their concentrations in blood and not by enzymatic rates (70). Heitmann et al. (69) showed that 15% of ketone bodies were extracted across the hindquarters of both fed nonlactating and progressively fasted sheep. Heitmann et al. (70), using data published by Kaufman and Bergman (89, 90), also estimated that 15% of ketone bodies were being extracted across the hindquarters of lactating sheep. Bergman and Kon (29) showed a consistent 53 to 60% oxidation of acetoacetate to CO2 in fed, fasted, or pregnant sheep.

Thus, as ketonemia becomes more severe, extrahepatic tissues can utilize increasing amounts of ketones for energy.

Studies with nonruminants, however, have shown a decreased utilization of ketone bodies during prolonged starvation (138) and In 39

diabetes (92). A decreased activity of 3-oxoacld transferase in

skeletal muscle has been shown to accompany the decreased utilization

of ketone bodies (61). Such results suggest that, during ketosls or a

long-term fast, hyperketonemia and severe ketoacidosis can be the

result of over-production and/or under-utllizatlon of ketone bodies.

Kinetics of glucose and propionate metabolism

In vivo kinetic techniques are extremely useful and have been

used extensively to study glucose and propionate metabolism of sheep

(32, 83, 107, 151, 164) and cattle (5, 73, 112, 155, 156, 161).

Comprehensive texts (46, 63, 81, 148) and reviews (4, 156) of the

mathematical basis of kinetic analysis have been published and need

not be repeated. Some basic knowledge and definitions, however, are

needed to facilitate discussion of results from kinetic experiments.

The advent of computer programs (139) and simplified equations (160)

have made kinetic analyses more available; however, without a solid

understanding of the principles underlying the kinetic approach, the

full potential of this technique will not be realized.

Kinetic analyses are based upon the principle that metabolism is

dynamic and that metabolites are being synthesized and degraded

continually (75). A given metabolite is assumed to be distributed

homogeneously into a specific physical space in the body and this is

defined as a pool (65). A pool can be labeled Isotoplcally and the disappearance or appearance of label can be measured. The rate of appearance or disappearance of label can be described by mathematical 40

equations from which rates of physiological processes can be

calculated. A prerequisite for simple kinetic analyses is that the

animal be in metabolic steady state. Steady-state conditions dictate

that the metabolic processes measured can be described and solved

mathematically as a first-order process. Work with steers (6) has shown that ruminants can be assumed to be in steady state when small, frequent, and equal meals are consumed.

Kinetic analyses can be conducted by using single-injections, continuous Infusions, or primed-continuous Infusions of tracer (169).

For this review, only single-injection kinetics will be discussed.

More complete discussions of primed-continuous infusions of tracer are available (156, 169). By using a single-injection of radioactive tracer, the decrease of specific radioactivity of labeled compound can be described by this equation:

SRAta - Hi X e(-gi(t)), where SBA^g is the specific radioactivity of compound a at time t, is the intercept of an individual component of the multi-exponential curve of SRA vs. time, t is time post-injection, and Is the negative of the slope of the component (160). This equation describes, therefore, the dilution of Isotopic tracer as a function of the number of different compounds that contribute to production and utilization of tracee and the rates at which they contribute. A discrepancy in terminology should be noted. Shipley and Clark (148) 41

describe "g" as the the slope of the component, whereas White et al.

(160) describe "g" as the rate constant of disappearance of tracer.

Although the term "slope" commonly Is used, the term "rate constant"

is more accurate because the units of "g" are t'^ and, therefore, "g"

is not the slope of the line.

The most frequently reported variables that can be calculated

from single-injection kinetics are irreversible loss, total entry rate, recycling, pool size, and pool space (168). Calculations for these variables have been summarized eloquently by,White et al. (160).

Irreversible loss (XL) is the single most important variable obtained from kinetic analyses, and it represents the amount of tracee that leaves the pool never to return. If in vivo kinetic measurements are made while an animal is in steady state, the amount leaving the pool also equals the amount of new compound entering the pool at any given moment. Therefore, for glucose metabolism, XL represents the amount of new glucose formed by gluconeogenesls plus that absorbed from the gastrointestinal tract per unit time. Total entry rate (TER), however, is the total amount of tracee entering the sampled compartment. TER includes newly synthesized compounds, but it also includes compounds that are recycled within the body. An example of recycling for glucose metabolism Is glucose that is synthesized, stored as glycogen, and then released again at a later time as free glucose. Recycling is calculated by subtracting IL from TER (160) and can occur by both physical and chemical means. 42

Fool size is regarded as the total amount of tracee present in

the sampled compartment. This variable is limited mathematically to

the amount of non-isotopic tracee with which the isotopic tracer

completely mixes. Complete mixing of tracer with tracee therefore is

an important assumption for correct measurement of in vivo kinetics.

Pool space is defined as the volume of body water within which the

tracee is distributed and expressed as percent of body weight. The

assumption is made that wherever tracee is located, it is in the same

concentration as in the sampled fluid, usually blood.

The kinetic variables already discussed all are calculated after

labeling and sampling only one pool. When a second pool, or secondary

pool, is sampled after the first pool, or primary pool, is labeled,

further information can be obtained. The rate of transfer of tracer

into the second pool can be explained by an equation similar to

equation 1. Then, a transfer quotient (TQ) can be calculated by

dividing the area under the curve for the secondary pool by the area

under the curve for the pool originally labeled. TQ thus is defined

as the fraction of a secondary pool derived from the primary pool.

For example, when propionate is labeled and sampled in the rumen and

glucose is sampled in the blood, the percentage of glucose derived

from propionate can be determined. The equation and mathematical

theory of transfer quotients is described by Gurpide et al. (65).

There are few in vivo kinetic studies of glucose metabolism in lactating cows. The cost of cows and isotope, as well as the 43

unpredictability of obtaining cows in the ketotic state, have limited

the number of kinetic studies completed on ketotic cows in early

lactation. Studies that have been completed usually have very few

cows on experiment.

There is some evidence that in vivo gluconeogenesis per unit of

feed intake increases at the onset of lactation. Bennick et al. (25)

fed four cows a constant Intake from 45 days prepartum to 40 days

postpartum and found that glucose IL increased from about 1.5 kg/d

prepartum to about 2.5 kg/d at 40 days postpartum. Peel and Bauman

(127) also recently published glucose kinetic data from normal and somatotropin-treated cows. They found that the glucose XL of their cows increased from 2.17 kg/d with a placebo to 2.44 kg/d with injection of somatotropin. Other studies have concentrated on the effect of fasting on glucose XL in cows and show conclusively that glucose XL decreases drastically in fasted cows (98, 103). Studies with ketotic cows are less conclusive. Reports by Kronfeld et al.

(102) and Kronfeld and Raggi (101) suggest that cows in the early stages of spontaneous ketosls have nearly identical glucose pool sizes and glucose XL of about 1.7 to 2.0 kg/d, although data are variable.

However, some cows in the later stages of ketosls have shown a decrease in glucose XL coinciding with hypophagia (98).

Even fewer kinetic studies have been done to study propionate metabolism in lactatlng cows. Bauman et al. (23) reported propionate

XL of 31 moles/d In lactatlng cows fed about .5 Meal of digestible 44

energy in a high-grain, low-fiber diet. Propionate IL averaged only

13.3 moles/d for cows fed a high-forage, low-concentrate diet in the

same study. Esdale et al. (58) earlier had shown similar results with

a high-concentrate diet. Wiltrout and Satter (161), although using

estimates, reported 12.0 vs 25.0 moles of propionate produced daily in nonlactating vs lactating cows. Dry matter intakes were 9.5 and 17.0 kg/d for the nonlactating and lactating phases, respectively. Such limited studies suggest that propionate production in the rumen of cows is affected by both feed intake and type of feed consumed. This suggestion is supported by more extensive work in steers (73, 156,

167) and sheep (23, 83).

Summary of literature

The onset of lactation in cows creates great stress in the body and requires finely-tuned coordination of carbohydrate, lipid, and ketone body metabolism. Gluconeogenesis in the liver is of prime importance because of the lack of glucose absorbed from the gut and the tremendous drive of the mammary gland to produce milk lactose. In early lactation, this demand for glucose can exceed available substrates and lipolysis is stimulated in adipose tissue, thereby increasing fatty acids in plasma.- These fatty acids can be taken up by liver and metabolized for energy, packaged into triglycerides, or partly oxidized to ketone bodies. Any disruption in the precarious balance of these metabolic processes can lead to the metabolic disease known as lactation ketosis. 45

The characteristics of clinical ketosls In lactatlng cows have

been documented well. Visual signs of ketosls Include hypophagla,

decreased milk production, rapid loss of body weight, and lethargy or

hyperexcltablllty. Metabolic Implications are hyperketonemla,

hypoglycemia, hypoInsulInemla, decreased hepatic glycogen, increased

hepatic triglycerides, and Increased NEFA in plasma. The characteristics of subclinical ketosls are not understood very well.

This is partly due to the unpredictability of the incidence of ketosls, which makes experimental protocols difficult to perform.

There currently is a need for experimental protocols that provide a consistent source of ketotlc cows for research in cows with both subclinical and clinical ketosls. Therapeutic agents have been effective in reversing ketosls symptoms, but less is known about their ability to prevent ketosls from developing. Further data are needed to elucidate the cause-and-effect relationships of changes in metabolism during the development of ketosls. 46

EXPLANATION OF DISSERTATION FORMAT

This dissertation Is presented in the alternative format, as

outlined in the Iowa State University Graduate College Thesis Manual.

Use of the alternate format allows preparation of independent

sections, each of which is in a form suitable for submission for

publication in a scientific Journal.

Two separate papers have been prepared from research performed to

partly fulfill requirements for the Ph.D. degree. Each paper is complete in itself and has an abstract, introduction, methods, results and discussion, general discussion, and references. Although these papers are presented separately, the closeness of the subject matter allowed a generalized summary to be prepared. 47

REFERENCES

1. Alello, R. J., T. M. Kenna, and J. H. Herbein. 1984. Hepatic gluconeogenic and ketogenic interrelationships in the lactating cow. J. Dairy Sci. 67:1707.

2. Annison, E. F., and D. B. Lindsay. 1963. The measurement of entry rates of propionate and of butyrate in sheep. Biochem. J. 85:474.

3. Annison, E. F., D. B. Lindsay, and R. R. White. 1963. Metabolic interrelationships of glucose and lactate in sheep. Biochem. J. 88:243.

4. Armentano, L. E. 1982. Effects of monensin on volatile fatty acid kinetics in steers. Ph.D. Dissertation. Iowa State University library, Ames, lA.

5. Armentano, L. E., and J. W. Young. 1983. Production and metabolism of volatile fatty acids, glucose and CO2 in steers and the effects of monesin on volatile fatty acid kinetics. J. Nutr. 113:1265.

6. Armentano, L. E., S. E. Mills, and J. W. Young. 1984. Effects of feeding frequency on glucose concentration, glucose turnover, and Insulin concentration in steers. J. Dairy Sci. 67:1445.

7. Armstrong, D. G. 1965. Carbohydrate metabolism in ruminants and energy supply. Pages 272-288 ^ R. W. Dougherty, R. S. Allen, W. Burroughs, N. L. Jacobson, and A. D. McGllllard, eds. Physiology of digestion in the ruminant. Butterworths, Washington, DC.

8. Ash, R., and G. D. Balrd. 1973. Activation of volatile fatty acids In bovine liver and rumen epithelium. Biochem. J. 136:311.

9. Balrd, G. D. 1977. Aspects of ruminant intermediary metabolism in relation to ketosis. Biochem. Soc. Tran. 5:819.

10. Balrd, G. D. 1981. Metabolic modes indicative of carbohydrate status in the dairy cow. Fed. Proc. 40:2530.

11, Balrd, G. D. 1982. Primary ketosis in high-producing dairy cows; clinical and subclinical disorders, treatment, prevention, and outlook. J. Dairy Sci. 65:1. 48

12. Balrd, G. D., and R. J. Heltzman. 1970. Gluconeogenesls in the cow. The effects of glucocorticoid on hepatic intermediary metabolism. Blochem. J. 116:865.

13. Balrd, G. D., and R. J. Heitzman. 1971. Mode of action of a glucocorticoid in bovine intermediary metabolism: possible role in controlling hepatic ketogenesis. Biochim. Biophys. Acta 252:184.

14. Balrd, G, D., K. G. Hlbbltt, G. D. Hunter, P. Lund, H. Stubbs, and H. A. Krebs. 1968. Biochemical aspects of bovine ketosls. Blochem. J. 107:683.

15. Balrd, G. D., R. J. Heitzman, and K. G. Hlbbltt. 1972. Effects of starvation on intermediary metabolism in the lactatlng cow: A comparison with metabolic changes occurring during bovine ketosls. Blochem. J. 128:1311.

16. Balrd, G. D., R. J. Heitzman, K. G. Hlbbltt, and G. D. Hunter. 1974. Bovine ketosls. A review with recommendations for control and treatment. Part 2. Brit. Vet. J. 130:318.

17. Balrd, G. D., H. W. Symonds, and R. Ash. 1975. Some observations on metabolite production and utilization in vivo by the gut and liver of adult dairy cows. J. Agrlc. Sci. 85:281.

18. Balrd, G. D., H. A. Lomax, H. W. Symonds, and S. R. Shaw. 1980. Net hepatic and splanchnic metabolism of lactate, pyruvate and propionate in dairy cows in vivo in relation to lactation and nutrient supply. Blochem. J. 186:47.

19. Ballard, F. J., R. W. Hanson, D. S. Kronfeld, and F. Raggl. 1968. Metabolic changes in liver associated with spontaneous ketosls and starvation in cows. J. Nutr. 95:160.

20. Ballard, F. J., R. W. Hanson, and D. S. Kronfeld. 1969. Gluconeogenesls and lipogenesls in tissue from ruminant and nonruminant animals. Fed. Froc. 28:218.

21. Bassett, J. M. 1975. Dietary and gastro-intestinal control of hormones regulating carbohydrate metabolism in ruminants. Page 383 ijn 1. W. McDonald and A. C. I. Warner, ed. Digestion and metabolism in the ruminant. Univ. New England Publ. Unit, Armidale, Aust.

22. Bauman, D. E., and W. B. Currle. 1980. Partitioning of nutrients during pregnancy and lactation. A review of mechanisms involving homeostasis and homeorhesis. J. Dairy Sci. 63:1514. 49

23. Bauman, D. E., C. L. Davis, and H. F. Bucholtz. 1971. Propionate production in the rumen of cows fed either a control or high-grain, low-fiber diet. J. Dairy Sci. 54:1282.

24. Bell, A. W. 1981. Lipid metabolism in liver and selected tissues in ruminants. Pages 363-399 in W. W. Christie, ed. Lipid metabolism in ruminant animals. Pergamon Press, New York, NY.

25. Bennink, M. R., R. W. Hellenberger, R. A. Frobish, and D. E. Bauman. 1972. Glucose oxidation and entry rate as affected by the initiation of lactation. J. Dairy Sci. 55:712.(abstr.)

26. Benson, J. D., and R. S. Emery, 1971. Fatty acid esteriflcatlon by homogenates of bovine liver and adipose tissue. J. Dairy Sci. 54:1034.

27. Bergman, E. N. 1971. Hyperketonemia-ketogenesls and ketone body metabolism. J. Dairy Sci. 54:936.

28. Bergman, E. N. 1973. Glucose in ruminants as related to hypoglycemia and ketosis. Cornell Vet. 63:341.

29. Bergman, E. N., and K. Kon. 1964. Acetoacetate turnover and oxidation rates in ovine pregnancy ketosis. Am. J. Physiol. 206:449.

30. Bergman, E. N., and J. E. Wolff, 1971. Metabolism of volatile fatty acids by liver and portal-drained viscera in sheep. Am. J. Physiol, 221:586,

31. Bergman, E. N., W. E. Roe, and K. Kon. 1966. Quantitative aspects of propionate metabolism and gluconeogenesls In the normal and hypoglycemic ketotic sheep. Am. J. Physiol. 215:874.

32. Bergman, E. N., D. J. Starr, and S. S. Reulein. 1968, Glycerol metabolism and gluconeogenesls in the normal and hypoglycemia in ketotic sheep. Am. J. Physiol. 215:874.

33. Bergman, E. N., H, L, Katz, and C, F, Kaufman. 1970. Quantitative aspects of hepatic and portal glucose metabolism and turnover in sheep. Am. J. Physiol. 219:785.

34. Bergman, E. N., R, J, Havel, B, M. Wolfe, and T. Bohmer. 1971, Quantitative studies of the metabolism of chylomicron triglyceride and cholesterol by liver and extrahepatlc tissues of sheep and dogs, J, Clin. Invest. 50:1831. 50

35. Berzlns, R. 1975. Ruminant metabolism and hormone secretion during lactation. Ph.D. Dissertation. Natl. Library Can., Ottawa, Can. Order No. 31676.

36. Bines, J. W., I. C. Hart, and S. V. Morant. 1983. Endocrine control of energy metabolism in the cow; diurnal variations In the concentrations of hormones and metabolites In the blood plasma of beef and dairy cows. Horm. Metab. Res. 15:330.

37. Black, A. L., H. Klelber, and A. M. Brow. 1961. Butyrate metabolism In the lactatlng cow. J. Biol. Chem. 236:2399.

38. Blouln, A. 1977. Morphometry of liver sinusoidal cells. Pages 61-71 ^ E. Wlsse and D. L. Knook, eds. Kuppfer cells and other liver sinusoidal cells. Elsevier, North-Holland, Amsterdam.

39. Bonner, J. M., G. S. Hess, E. 0. Otchere, and J. W. Young. 1975. Physiological effects of 1,3-butanedlol fed to cattle. J. Dairy Scl. 58:56.

40. Brockman, R. P. 1976. Effects of glucagon and insulin on lipolysis and ketogenesis in sheep. Can. J. Comp. Med. 40:166.

41. Brockman, R. P. 1979. Roles of glucagon and insulin in the development of ruminant ketosis: A review. Can. Vet. J. 20:121.

42. Brockman, R. P., and C. Greer. 1978. Roles of glucagon and insulin in the regulation of metabolism in ruminants: A review. Can. Vet. J. 19:55.

43. Brockman, R. P., and C. Greer. 1980. Effects of somatostatin and glucagon on the utilization of 2-^^C propionate in glucose production in vivo in sheep. Am. J. Physiol. 229:1344.

44. Bronfenmajer, S., F. Schaffner, and H. Popper. 1966. Fat storing cells (lipocytes) in human liver. Arch. Pathol. 82:447.

45. Brumby, P. E., M. Anderson, B. Tuckley, J. E. Storry, and K. G. Hlbbltt. 1975. Lipid metabolism in the cow during starvation-induced ketosis. Blochem. J. 146:609.

46. Carson, E. R., C. Cobelli, and L. Finkelsteln. 1983. The mathematical modeling of metabolic and endocrine systems: formulation, identification, and validation. J. Wiley and Sons. New York, NY. 51

47. Christie, V. W. 1981. The composition, structure and function of lipids in the tissues of ruminant animals. Pages 95-191 jji W. V. Christie, ed. Lipid Metabolism In ruminant animals. Fergamon Press, New York, NY.

48. Cook, R. M., and L. D. Miller. 1965. Utilization of volatile fatty acids In ruminants. I. Removal from portal blood by the liver. J. Dairy Scl. 48:1339.

49. Curtis, C. R., H. N. Erb, C. J. Sniffen, R. D. Smith, and D. S. Kronfeld. 1985. Path analysis of dry period nutrition, postpartum metabolic and reproductive disorders, and mastitis In Holsteln cows. J. Dairy Scl. 68:2347.

50. Daoust, R. 1958. The cell population of liver tissue and the cytologlcal reference bases. Am. Inst. Biol. Scl. Publ, 4:3.

51. Davis, C. L., and A. B. Erhart. 1976. Effects of monensln and urea in finishing steer rations. J. Anim. Scl. 43:1.

52. Dawes, D. J., S. C. Sensenlg, and R. N. Heltmann. 1985. Autoregulatlon of ketogenesls in fasted sheep. Fed. Proc. 44:548.(abstr.)

53. deBoer, G. 1984. Glucagon, insulin, and growth hormone in the regulation of metabolism in dairy cows during lactation and ketosls. Ph.D. dissertation. Iowa State University, Ames, lA.

54. deBoer, 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 Scl. 68:326.

55. deBoer, G., A. Trenkle, and J. W. Young. 1986. Secretion and clearance rates of glucagon in dairy cows. J. Dairy Scl. 69:721.

56. Egan, A. R., and A. L. Black. 1968. Glutamine acid metabolism in the lactating dairy cow. J. Nutr. 96:450.

57. Eisenberg, S., and R. I. Levy. 1975. Lipoprotein metabolism. Adv. Lipid Res. 13:1.

58. Esdale, W. K., G. A. Broderlck, and L. D. Satter. 1968. Measurement of ruminai volatile fatty acid production from alfalfa hay or corn silage rations using a continuous infusion dilution technique. J. Dairy Scl. 51:1823.

59. Exton, J. H. 1972. Gluconeogenesis. Metabolism 21:945. 52

60. Fawcett, D. W. 1986. The liver and gall bladder. Pages 679-715 in K. Bloom and D. W. Fawcett, eds. A textbook of histology. W. B. Saunders Company. Philadelphia, PA.

61. Fenselau, A., and K. Wallis. 1976. 3-oxoacld-coenzyme a*transferase in normal and diabetic rat muscle. Biochem. J. 158:509.

62. Glascock, R. P., and V. A. Welch. 1974. Contribution of the fatty acids of three low density serum lipoproteins to bovine milk fat. J. Dairy Sci. 57:1364.

63. Godfrey, 0. 1983. Compartmental models and their application. Academic Press, New York, NY.

64. Groot, P. H. E., H. R. Scholte, and W. C. Hulsmann. 1976. Fatty acid activation: Specificity, localization, and function. Adv. Lipid Res. 14:75.

65. Gurpide, E., J. Mann, and S. Lieberman. 1963. Analysis of open systems by administration of tracers at a constant rate or as a single dose as illustrated by problems involving steroid hormones. J. Clin. Endocrinol. Metabol. 23:1155.

66. Hanson, R. W., and F. J. Ballard. 1967. The relative significance of acetate and glucose as precursors for lipid synthesis in liver and adipose tissue from ruminants. Biochem. J. 105:529.

67. Hart, I. C., J. A. Bines, and S. V. Morant. 1978. Endocrine control of energy metabolism in the cow: Correlations of hormones and metabolites in high and low yielding cows for stages of lactation. J. Dairy Sci. 62:270.

68. Hart, I. C., J. A. Bines, S. V. Morant, and J. L. Ridley. 1978. Endocrine control of energy metabolism in the cow: Comparisons of the levels of hormones (prolactin, growth hormone, insulin and thyroxine) and metabolites in the plasma of high- and low-yielding cattle at various stages of lactation. J. Endocrinol. 77:333.

69. Heitmann, R. N., S. C. Sensenig, C. K. Reynolds, J. M. Fernandez, and D. J. Dawes. 1986. Changes in energy metabolite and regulatory hormone concentrations and net fluxes across splanchnic and peripheral tissues in fed and progressively fasted ewes. J. Nutr. 116:2516. 53

70. Heltmann, R. N., D. J. Dawes, and S. C. Sensenlg. 1987. Hepatic ketogenesls and peripheral ketone body utilization in the ruminant. J. Nutr. 117:1174.

71. Heitzman, R. J. 1969. Hepatic enzyme activities early in lactation: a comparison of normal and ketotic cows. Res. Vet. Sci. 10:582.

72. Heitzman, R. J., K. G. Hibbitt, and I. Mather. 1971. The effects of thyroxine on hepatic gluconeogenesis and ketogenesis in dairy cows. Eur. J. Biochem. 21:411.

73. Herbein, J. H., R. W. Van Maanen, A. D. McGilliard, and J. W. Young. 1978. Rumen propionate and blood glucose kinetics in growing cattle fed isoenergetic diets. J. Nutr. 108:994.

74. Hess, G. S., and J. W. Young. 1972. Preventing and alleviating milk fat depression by feeding 1,3-butanediol to cows. J. Dairy Sci. 55:1097.

75. Hetenyi, G., Jr., and K. H. Norwich. 1974. Validity of the rates of production and utilization of metabolites as determined by tracer methods in intact animals. Fed. Froc. 33:1841.

76. Hibbitt, K. G. 1966. The induction of ketosis in the lactating dairy cow. J. Dairy Res. 33:291.

77. Hibbitt, K. G. 1979. Bovine ketosis and its prevention. Vet. Rec. 105:13.

78. Hibbitt, K. G., and G. D. Baird. 1967. An induced ketosis and its role in the study of primary spontaneous bovine acetonaemia. Vet. Rec. 81:511.

79. Hove, K. 1974. Nocturnal plasma insulin levels in cows with varying levels of plasma ketone bodies: Relations to plasma sugar and acetoacetate. Acta Endocrinol. 76:513.

80. Hove, K., and K. Halse. 1978, Absence of feeding-induced variations in plasma insulin in hypoglycemic-ketonaemic cows. Acta Vet. Scand. 19:216,

81. Jacquez, J, A. 1972. Compartmental analysis in biology and medicine, American Elsevier Publishing Co,, New York, NY,

82. Judson, G. J., and R. A. Leng, 1973, Studies on the control of gluconeogenesis in sheep: Effect of propionate, casein, and butyrate infusions, Br, J, Nutr, 29:175. 54

83. Judson, G. L., E. Anderson, J. R. Lulck, and R. A. Leng. 1968. The contribution of propionate to glucose synthesis in sheep given diets of different grain content. Br. J. Nutr. 22:69.

84. Jungermann, K., and D. Sasse. 1978. Heterogeneity of liver parenchymal cells. Trends Biochem. Sci. 3:198.

85. Jurehkova, G., D. Popovici, M. Raitaru, and E. Neacsu. 1980. Using the propionic acid as a mean of metabolic correction in the dairy cow with clinical ketosis. Lucr. Stiint. Cercet. Crestera. Taurin. 6:117.

86. Kaneda, K. and K. Wake. 1983. Distribution and morphological characteristics of the pit cells in the liver of the rat. Cell Tissue Res. 233:485.

87. Katz, H. L. and E. N. Bergman. 1966. Acid-base and electrolyte equilibrium in ovine pregnancy ketosis. Am. J. Vet. Res. 27:1285.

88. Katz, M. C., and E. N. Bergman. 1969. Hepatic and portal metabolism of glucose, free fatty acids, and ketone bodies in the sheep. Am. J. Physiol. 216:953.

89. Kaufman, C. F., and E. N. Bergman. 1971. Renal glucose, free fatty acid, and ketone body metabolism in the unanesthetized sheep. Am. J. Physiol. 226:827.

90. Kaufman, C. R., and E. N. Bergman. 1974. Renal ketone body metabolism and gluconeogenesis in normal and hypoglycemic sheep. Am. J. Physiol. Rev. 60:143.

91. Kauppinen, K. 1984. Annual milk yield and reproductive performance of ketotic and non-ketotic dairy cows, Zbl. Vet. Med. A. 31:694.

92. Keller, U., A. D. Cherrington, and J. E. Liljenquist. 1978. Ketone body turnover and net hepatic ketone production in fasted and diabetic dogs. Am. J. Physiol. 235:E238.

93. Kellogg, D. W., C. J. Balok, and D. D. Miller. 1971. Method for experimental induction of ketosis utilizing the interaction of low energy intake and thyroxine. J. Dairy Sci. 54:1499.

94. Koundakjian, P. P., and A. M. Snoswell. 1970. Ketone body and fatty acid metabolism in sheep tissues: 3-hydroxybutyrate dehydrogenase, a cytoplasmic enzyme in sheep liver and kidney, Biochem. J. 119:49. 55

95. Krebs, H. A. 1964. Gluconeogenesls. Froc. Royal Soc. 159:545.

96. Krebs, H. A. 1966. Bovine ketosls. Vet. Rec. 7:187.

97. Kronfeld, D. S. 1968. Acetate kinetics in normal and ketotlc cows. J. Dairy Sci. 51:397.

98. Kronfeld, D. S. 1971. Hypoglycemia in ketotlc cows. J. Dairy Sci. 54:949.

99. Kronfeld, D. S. 1972. Ketosls in pregnant sheep and lactating cows. A review. Aust. Vet. J. 48:680.

100. Kronfeld, D. S. 1982. Major metabolic determinants of milk volume, manunary efficiency, and spontaneous ketosls in dairy cows. J. Dairy Sci. 65:2204.

101. Kronfeld, D. S., and F. Raggi. 1964. Glucose kinetics in normal, fasting, and insulin-treated cows. Am. J. Physiol. 206:109.

102. Kronfeld, D. S., E. G. Tombropoulos, and M. Kleiber. 1959. Glucose biokinetics in normal and ketotlc cows. J. Appl. Physiol. 206:109.

103. Kronfeld, D. S., C. F. Ramberg, Jr., and D. M. Shames. 1971. Multlcompartmental analysis of glucose kinetics in normal and hypoglycemic cows. Am. J. Physiol. 220:886.

104. Lehninger, A. L. 1975. Biochemistry, pp 417-475. World Publishers, New York, NY.

105. Leighton, B., A. R. Nicholas, and C. J. Pogson. 1983. The pathway of ketogenesls in rumen epithelium of the sheep. Biochem. J. 216:769.

106. Leng, R. A. 1970. Glucose synthesis in ruminants. Adv. Vet. Sci. Comp. Med. 14:209.

107. Leng, R. A., J. W. Steel, and J. R. Lulck. 1967. Contribution of propionate to glucose synthesis in sheep. Biochem. J. 103:785.

108. Lindsay, D. B. 1970. Carbohydrate metabolism in ruminants. Pages 438-451 in A. T. Phllipson, ed. Physology of digestion and metabolism in the ruminant. Oriel Press, Newcastle upon Tyne, England. 56

109. Lindsay, D. B. 1978. Gluconeogenesls In ruminants. Blochem. Soc. Trans. 6:1152.

110. Llttledlke, E. T., J. W. Young, and D. C. Beltz. 1981. Common metabolic diseases of cattle: Ketosls, milk fever, grass tetany, and downer cow complex. J. Dairy Sci. 64:1465.

111. Loud, A. V. 1968. A quantitative stereological description of the ultrastructure of normal rat liver parenchymal cells. J. Cell. Biol. 37:27.

112. Lyle, R. R. 1983. Glucose kinetics and hepatic gluconeogenesls in ketotic and fasted steers. Ph.D. Dissertation. Iowa State University Library, Ames, lA.

113. Manns, J. G. 1972. Separation of pancreatic and gut glucagon-like immunoreactivity (G.L.I.) with observations on plasma concentrations of these hormones during lacatfon. Can. J. Physiol. Pharmacol. 50:554.

114. Mayfield, E. D., A. Bensadoun, and B. C. Johnson. 1966. Acetate metabolism in ruminant tissues. J. Nutr. 89:189.

115. McCarthy, R. D., P. T. Chandler, L. C, Griel, Jr., and G. A. Porter. 1968. Fatty acid composition of blood serum lipoproteins from normal and ketotic cows. J. Dairy Sci. 51:392.

116. McCarthy, R. D., G. A. Porter, and L. C. Griel, Jr. 1968. Bovine ketosls and depressed fat test in milk: A problem of methionine metabolism and serum lipoprotein aberration. J. Dairy Sci. 51:459.

117. McGarry, J. D., and D. W. Foster. 1971. The regulation of ketogenesis from octanoic acid. The role of the tricarboxylic acid cycle and fatty acid synthesis. J. Biol. Chem. 246:1149.

118. McGarry, J. D., and D. W. Foster. 1972. Regulation of ketogenesis and clinical aspects of the ketotic state. Metabolism 21:471.

119. McGarry, J. D., and D. W. Foster. 1977. Hormonal control of ketogenesis. Arch. Intern. Med. 137:495,

120. McGarry, J. D., and D. W. Foster. 1980. Regulation of hepatic fatty acid oxidation and ketone body production. Annu. Reve. Blochem. 49:395. 57

121. Hills. S. E. 1982. Metabolic charatexistics of the ketotlc state in the bovine. Ph.D. Dissertation. ISU. University Microfilms, Ann Arbor, MI. Order No. 82-21,209.

122. Mills, S. E., and B. F. Jenny. 1979. Effects of high concentrate feeding and fasting on plasma glucocorticoids in dairy heifers. J. Anim. Sci. 48:961.

123. Mills, S. E., D. C. Beitz, and J. tf. Young. 1986. Characterization of metabolic changes during a protocol for inducing lactation ketosls in dairy cows. J. Dairy Sci. 69:352.

124. Mills, S. E., D. C. Beitz, and J. W. Young. 1986. Evidence for Impaired metabolism in liver during induced lactation ketosls of dairy cows. J. Dairy Sci. 69:362.

125. Olson, M. J., and R. G. Thurman. 1987. Quantitation of ketogenesis in periportal and pericentral regions of the liver lobule. Arch. Biochem. and Blophys. 253:26.

126. Otchere, E. 0., A. D. McGllliard, and J. W. Young. 1974. Quantitation of a-linked glucose polymers passing to the small intestine in cattle. J. Dairy Sci. 57:1189.

127. Peel, C. J., and D. E. Bauman. 1987. Somatotropin and lactation. J. Dairy Sci. 70:474.

128. Ranaweera, A., E. J. H. Ford, and J. Evans. 1981. Gluconeogenesis from glycerol by ketotlc sheep pregnant with twins. Res. Vet. Sci. 30:303.

129. Rappaport, A. M. 1976. The mlcrocirculatory acinar concept of normal and pathological hepatic structure. Beitr. Pathol. 157:215.

130. Rappaport, A. M., Z. J. Borowy, W. M. Longheed, and W. N. Lotto. 1954. Subdivision of hexagonal liver lobules into a structural and functional unit: Role in hepatic physiology and pathology. Anat. Rec. 119:11.

131. Reid, R. L. 1968. The physiopathology of undernourishment in pregnant sheep, with particular reference to pregnancy toxemia. Adv. Vet. Sci. 12:163.

132. Reid, I. M. 1973. An ultrastructural and morphometrlc study of the liver of the lactatlng cow in starvation ketosls. Exper. and Mol. Pathol. 18:316. 58

133. Raid, I. M. 1980. Incidence and severity of fatty liver in dairy cows. Vet. Rec. 107:281.

134. Reid, I. M., A. J. Stark, and R. N. Iseor. 1977. Fasting and refeeding in the lactating dairy cow. 1. The recovery of milk yield and blood chemistry following a six-day fast. J. Comp. Path. 87:241.

135. Reid, I. M., R. A. Collins, G. D. Baird, C. J. Roberts, and H. W. Symonds. 1979. - Lipid production rates and pathogenesis of fatty liver in fasted cows. J. Agric. Sci. 93:253.

136. Reilly, P. E. B., and E. J. H. Ford. 1971. The effects of different dietary contents of protein on amino acid and glucose production and on the contribution of amino acids to gluconeogenesis in sheep. Br. J. Nutr. 26:249.

137. Robertson, A., and C. Thin. 1953. A study of starvation ketosis in the ruminant. Br. J. Nutr. 7:181.

138. Robinson, A. H., and D. H. Williamson. 1980. Physiological roles of ketone bodies as substrates and signals in mammalian tissues. Physiol. Rev. 60:143.

139. Russell, R. U., J. J. Veenhuizen, L. E. Armentano, J. W. Young, and J. L. Cornette. 1985. A computer program to solve for rates of exchange of material among pools in an open system kinetic model. J. Dairy Sci. 68:2115.

140. Saarinen, P. and J. C. Shaw. 1950. Studies on ketosis in dairy cattle. XI. Lipids, minerals, and ascorbic acid in the blood of cows with spontaneous ketosis. J. Dairy Sci. 33:496.

141. Saase, D. 1986. Liver structure and innervation. Pages 3-25 1^ R. G. Thurman, F. C. Kauffman, and K. Jungermann, eds. Regulation of hepatic metabolism, intra- and intercellular compartmentation. Plenium Press, New York, NY.

142. Satter, L. D., and D. W. Wiltrout. 1970. Utilization of ruminai propionate for glucose synthesis in the lactating and non-lactating cow. Fed. Proc. 29:692.

143. Schultz, L. H. 1968, Ketosis in dairy cattle. J. Dairy Sci. 51:1133.

144. Schultz, L. H. 1971. Management and nutritional aspects of ketosis. J. Dairy Sci. 54:962. 59

145. Schultz, L. H. 1974. Ketosls. pages 217-352 ^ fi. L. Larson and V. R. Smith, eds. Lactation: A comprehensive treatlce. Vol II. Academic Press, Inc., New York.

146. Schwalm, J. W., and L. H. Schultz. 1976. Relationship of Insulin concentration to blood metabolites in the dairy cow. J. Dairy Sci. 59:255.

147. Sensenig, S. C., D. J. Dawes, and R. N. Heltmann. 1986. A possible Intra-hepatic insulin independent mechanism of ketogenesis. Fed. Froc. 45:240.(abstr.)

148. Shipley, R. A., and R. E. Clark. 1972. Tracer methods for in vivo kinetics. Academic Press, New York, NY.

149. Simesen, M. G., and T. Holler. 1959. Liver biopsy on cattle. II. The patho-morphological picture of ketosls. Nord. Vet.-Med. 11:787.

150. Snoswell, A. H., N. D. Costa, J. G. McLean, G. D. Baird, H. A. Lomax, and H. W. Symonds. 1978. Interrelationships between acetylation and the disposal of acetyl groups in the livers of dairy cows. J. Dairy Res. 45:331.

151. Steel, J. W., and R. A. Leng. 1973. Effects of plane of nutrition and pregnancy on gluconeogenesis in sheep. 1. The kinetics of glucose metabolism. Br. J. Nutr. 30:451.

152. Treacher, R. J., G. D. Baird, ad J. L. Young. 1976. Antiketogenic effect of glucose in the lactating cow deprived of food. Biochem. J. 158:127.

153. Trenkle, A. H. 1980. Amino acid metabolism and hormonal control of lactation. Pages 505-522 ^ Y. Ruckebusch and P. Thivend, eds. Digestive physiology and metabolism in ruminants. HTP Press, Lancaster, England.

154. Tzur, R., E. Tal, and B. Shapiro. 1964. -glycerophosphate as regulatory factor in fatty acid esteriflcation. Biochlm. Biophys. Acta. 84:18.

155. Van Maanen, R. W., J. H. Herbein, A. D. McGilliard, and J. W. Young. 1978. Effects of monensln on in vivo rumen propionate production and blood glucose kinetics in cattle. J. Nutr. 108:1002.

156. Veenhuizen, J. J. 1983. Effects of propionate and phlorizin on kinetics of glucose and propionate metabolism in cattle. M. S. Thesis. Iowa State University Library, Ames, lA. 60

157. V«mon, R. 6. 1981. Lipid metabolism in the adipose tissue of ruminant animals. Pages 280-326 ^ U. W. Christie, ed. Lipid metabolism in ruminant animals. Fergamon Press, New York, NY.

158. Waterman, R., J. W. Schwalm, and L. H. Schultz. 1972. Nicotinic acid treatment of bovine ketosis. 1. Effects on circulatory metabolites and interrelationships. J. Dairy Sci. 55:1447.

159. Weigand, E., J. W. Young, and A. D. McGilliard. 1972. Extent of propionate metabolism during absorption from the bovine ruminoretlculum. Biochem. J. 126:201.

160. White, R. G., J. W. Steel, R. A. Leng, and J. R. Luick. 1969. Evaluation of three isotope-dilution techniques for studying the kinetics of glucose metabolism in sheep. Biochem. J. 114:203.

161. Wiltrout, D. W., and L. D. Satter. 1972. Contribution of propionate to glucose synthesis in the lactating and nonlactating cow. J. Dairy Sci. 55:307.

162. Wisse, E. 1977. Ultrastructure and function of kuppfer cells and other sinusoidal cells in the liver. Pages 33-60 ^ E. Wisse and D. L. Knock, eds. Kuppfer cells and other liver sinusoidal cells. Elsevier, North-Holland, Amsterdam.

163. Wolff, J. E., and E. N. Bergman. 1972. Metabolism and interconversion of five plasma amino acids by tissue of the sheep. Am. J. Physiol. 223:447.

164. Wolff, J. E., and E. N. Bergman. 1972. Gluconeogenesis from plasma amino acids in fed sheep. Am. J. Physiol. 223:455.

165. Wood, H. G. 1968. Mechanism of formation of oxaloacetate and phosphoenol pyruvate from pyruvate. J. Vitamin 14:59.

166. Yost, W. M., J. W. Young, S. P. Schmidt, and A. D. McGilliard. 1977. Gluconeogenesis in ruminants: Propionic acid production from a high-grain diet fed to cattle. J. Nutr. 107:2036.

167. Young, J. W. 1977. Gluconeogenesis in cattle; significance and methodology. J. Dairy Sci. 60:1.

168. Young, J. W., J. J. Veenhuizen, and R. W. Russell. 1987. Use of in vivo kinetic techniques to evaluate metabolism of nutrients. Fed. Proc. 46:295. 61

SECTION I: BLOOD AND LIVER METABOLITES AND MILK YIELD DURING THE DEVELOPMENT OF EXPERIMENTAL KETOSIS AND EARLY TREATMENT OF KETOSIS IN COWS 62

ABSTRACT

A protocol for Inducing ketosls and a technique for Its

prevention were used to characterize the metabolic effects of

development and early treatment of ketosls in cows. During the 9 wk

immediately after parturition, 18 high-producing cows were assigned

among three treatments: ad-lib-fed control, ketosls Induction by

restricting fed to 80% of ad lib intake plus supplementation with

1,3-butanedlol, and glucose-treated cows who received ketosls

induction plus continuous intraduodenal Infusion of glucose at 500 g/d

between d 25 and 40 postpartum. Ketosls induction was begun at d IS

postpartum and caused ketonemla and gradual development of ketosls over 25 to 30 d. Ketotlc cows maintained constant milk production until a 12% decrease 1 wk before and a 31% decrease at ketosls. Cows treated with glucose showed no decrease. Free fatty acid concentration in plasma of ketotlc cows was Increased by 3.0-, 2.6-, and 1.9-fold at 3 wk before, 2 wk before, and at ketosls, but glucose-treated cows Increased slightly then returned to baseline.

Beta-hydroxybutyrate increased 3.5-, 5.8-, and 8.4-fold for the same periods for the cows receiving ketosls Induction, but only a maximum of 1.6-fold for glucose-treated cows. Plasma glucose was decreased by

20% for ketosls and Increased by 9% for glucose-treated cows. Liver glycogen content was nearly zero by 2-wk before ketosls in ketotlc cows, but glycogen in glucose-treated cows increased to prepartal 63

values. Liver triglycerides averaged 2.0% across treatments at d 4

postpartum. For ketotlc cows, values reached 8.5, 9.0, and 9.8 for 2

wk before, 1 wk before, and at ketosls. Values for glucose-treated

cows at the same time periods were 1.3, 1.0, and 1.1*. Results

Indicate that development of ketosls results In abnormal blood metabolite concentrations up to 2 wk before the full manifestations of clinical ketosls are detected, which Includes a major decrease In milk production. 64

INTRODUCTION

The onset of lactation In cows results in many metabolic

adaptations to support milk production (6), and any breakdown in the

coordination of metabolism can lead to metabolic disease, such as

ketosis. Incidence of lactation ketosis in cows is thought to be

related to the severity of negative energy balance.

The characteristics of clinical ketosis have been documented well.

These Include visual signs of hypophagla, decreased milk production, rapid loss of body weight, and lethargy or hyperexcltability (30), and

metabolic conditions of hyperketonemia, hypoglycemia, hypoinsulinemla, decreased hepatic glycogen. Increased hepatic triglycerides, and

Increased fatty acids in plasma (2, 3, 20). Subclinical ketosis is understood less, however, and cause-and-effect relationships of metabolic changes occurring during the development of ketosis are unknown. Preventative measures have been suggested (17), but the effects of preventive agents on the metabolism of the early lacatating cow have not been studied.

Attempts to experimentally Induce ketosis by starvation have resulted in metabolic profiles that mlmick those of spontaneously-ketotic cows (5, 12, 28), but the protocols are

Inadequate to evaluate the development of ketosis. Recent success

(25, 26) has been achieved, however, in Inducing ketosis in cows by using moderate feed restriction and butanedlol feeding. This protocol 65

resulted In cows developing ketosis gradually during a 21-d period.

It shows the promise of providing a consistent source of cows that can

be studied In detail as ketosis develops, but has been proven only in

cows that were overfed in the dry period.

Increasing demands on cows by Improved genetic selection for milk

production, treatment with exogenous hormones that stimulate milk

production, and Increased use of roughage in diets make it seem

unlikely that ketosis can be eliminated. Our research has been

conducted, therefore, to further the understanding of the development

of ketosis and to study the metabolic effects of a prevention agent

during the development of ketosis in cows. Our specific objectives were to: a) experimentally Induce ketosis in cows fed normal diets in the dry period, b) evaluate progressive changes in blood glucose, free fatty acids and beta hydroxybutyrate, and liver composition during development of ketosis, and c) study the effect of exogenous glucose

Infusion into the duodenum on blood and liver metabolites and

Incidence of ketosis. 66

MATERIALS AND METHODS

Design of experiment and management of cows

One Brown Swiss and seventeen Holsteln cows, averaging 595 kg

body weight at 4 d postpartum, were used in a completely randomized

design with three treatments. One Holsteln cow did not respond to the ketosls induction protocol and was replaced to maintain equal numbers for the three treatments. Cows were housed and fed hay ad libitum in the dry lot for pregnant cows at the Iowa State University dairy farm for 30 d before calving. After 4 d postpartum, cows were housed in a tie-stall barn with other herd cows.

Cows were fed ad libitum a 60% forage, 40% concentrate diet starting at day 4. Feed ingredients are listed in Table 1. Cows were fed this mixture twice daily between 900 and 1000 h and between 1600 and 1800 h. Orts ware collected each day to determine exact feed

Intake. Long-stem alfalfa hay was fed before and after each pm milking and totaled 6 kg/d.

Six cows were assigned randomly to each of three treatments: control, a ketosls induction protocol (KIP), or a glucose-treated (GT) group. Two control and two KIP cows also were used for a separate in vivo kinetic study described separately (33). These four cows were housed in a separate facility, but otherwise were handled identically to other cows.

The experimental period consisted of 14 d before parturition until 60 d postpartum or two wk after ketosls detection for KIP cows. 67

TABLE 1. Feed ingredients

Forage and Ingredient grain mixture®

% of DM

Alfalfa haylage 33.3 Com grain 32.2 Com silage 20.6 Soybean oil meal 6.6 Com gluten 6.3 Vitamin and mineral suppl.b 1.0

Total 100.0

^Silage and grain mixture was achieved by combining the ISU herd silage mix with com grain mix in a 79:21 ratio.

^Vitamin and mineral supplement is a mixture of recommended amounts of sodium bicarbonate, magnesium oxide, dicalcium phosphate, calcium carbonate, sodium chloride, trace mineral premix, and vitamins A and 0.

Control cows were fed ad libitum throughout the period. KIP treatment

consisted of cows gradually being induced into clinical ketosis by

using the 20% feed restriction, 1,3-butanediol (FRBD) protocol

published by Mills et al. (25), with the exception that cows were not overfed during the dry period. At 2 wk postpartum, intake of silage plus grain mix was restricted to 80% of ad libitum intake of the previous 5 d and supplemented with the ketone body precursor. 68

1,3-butanedlol^ (BD) at 4% of dry matter (DM) intake. At 18 d

postpartum, BD supplementation was Increased to 8% of DM Intake. At

first, BD was Included at 16% of the DM Intake, but two cows quickly

passed through subclinical ketosls and began to show clinical ketosls

symptoms within 10 d after beginning FRBD. Supplementation with BD

then was decreased to 8% to bring about the more gradual development of ketosls, analogous to on-farm ketosls, for these and all other KIP and GT cows. FRBD continued until ketosls was detected. Contrary to

Mills et al. (25), supplements of BD were held at a constant percentage of DM intake after d 18 postpartum. An average of 900 g

BD/d was added, thereby resulting in 7.2% of the ME of the FRBD diet coming from BD.

Diagnosis of ketosls was based upon the presence of high concentrations of urine ketone bodies^ (16), altered behavior indicative of ketosls (30), decreases in milk production, and further decreases in consumption of the FRBD diet. Ketosls was diagnosed at a mean of 44 d postpartum and within a range of 24 to 56 d postpartum.

When cows were diagnosed as being clinically ketotic, FRBD was terminated, and cows were returned to ad libitum intake. No further

^The 1,3-butanedlol (1,3-butylene glycol) was donated by the Celanese Chemical Co., New York, NY.

^Determined by using KETO-DIASTIX, purchased from the Iowa State University Veterinary Service, Ames, lA. 69

treatment was Intended, but two cows required 500 ml of 50% dextrose^

given intravenously when severe clinical ketosls was detected.

GT cows were treated identically to those on KIP with one

exception. At day 23 of lactation, a small polyvinyl catheter was

placed in the proximal duodenum about 20 cm caudal to the pylorls.

Surgery was performed with cows standing upright and given Xylocalne^

at the 11th intercostal space in the region of the costochondral

junction. An 18-cm incision was made, blunt dissection was used to penetrate the peritoneal cavity, and the duodenum was externalized temporarily. The indwelling catheter was placed into the duodenum, sutured to the duodenal surface by using an absorbent sponge for support, and externalized at the 12th intercostal space. Cows decreased feed intake for one feeding after surgery but returned to normal by the next morning. Milk production was not affected, and inflammation at the incision site was minimal. At d 25 of lactation, tygon tubing was connected to the catheter and a peristaltic pump^ was used to infuse a .35 g/ml solution of dextrose at .96+.03 ml/min for

15 days. The infusion of glucose was intended to coincide with the

^Dextrose (50%) was purchased from the Iowa Sate University Veterinary Service, Ames, lA.

^Lidocalne hydrochloride (2%) was used as a local anaesthetic; Med Tech, Inc., Elkwood, KS.

^Slgmamotor, model TM 10. 70

time when subclinical and clinical ketosis was detected for KIP. The

FRBD protocol was continued for 7 d after the glucose infusion was

terminated.

One Infusion was terminated after 7 days because of

hyperexcitabllity and nervousness of the cow. Two other cows showed

extreme hyperexcitabllity during the last three days of glucose

infusion.

Sampling and analytical procedures

Milk production was recorded daily, and cows were weighed on alternate days. An explanation of terminology and abbreviations, timing of sampling blood and liver, and feeding protocols is listed in

Table 2. A blood sample was taken approximately 2 wk prepartum from the jugular vein. After parturition, weekly blood samples from the jugular vein were collected into heparlnlzed syringes containing 150

Ul of a 4% NaF solution, which acts as a glucose preservative.

Samples were taken just before the morning feeding, and plasma was prepared and stored at -20°G. Liver biopsies also were taken at same times as blood samples. Three to four grams of liver were taken by puncture biopsy (21) at each sampling. Approximately one gram was frozen immediately at -20°C for later determination of glycogen and triglyceride content. The remaining liver was used within an hour for in vitro incubations as reported (33). Effects of liver biopsies on these cows has been discussed (33). 71

TABLE 2. Summary of sampling blood and liver and feeding schedule for ketosls-Induced (KIP) and glucose-treated (GT) cows*

Stage of Feeding Sampling Average day experiment protocol time of sampling^

Prepartum (PP) ad lib hay. estimated 2 wk -11 + 1.5 prepartum.

Early postpartum ad lib feeding between 4 and 5.5 + .4 (EPP) of herd forage 7 d postpartum. and grain mix plus 6 kg alfalfa hay.

Pre-induction same as for EPP. 1 d before feed 13 + .3 (PI) restriction and 1,3- butanediol feeding began.

Early induction Restriction of about 1 wk after 21 + .3 (EI) feed intake to beginning FRBD. 80% of ad libitum; supplementation with 1,3-butanedlol at 8% of DM Intake (FRBD).

Mid-induction same as for EI. about 2 wk 28 + .3 (MI) beginning FRBD.

Late Induction same as for EI. about 3 wk 35 + .4 (LI) after beginning FRBD.

Ketotlc (K) same as for EI. within 48 h before 44 + 5.8 or after clinical ketosls as diagnosed.

Recovery (R) ad libitum feed about 2 wk after 60 + 3.7 intake. Same as diagnosis of clinical for EPP. ketosls.

^Sampling was similar for control cows, however, ad libitum feeding was continued throughout the period for the controls.

^Parturition is the reference point. Means of sampling day + SE for EPP, PI, EI, MI, and LI are from all 18 cows. Means of sampling day + SE for K and R are for the six KIP cows and control and treated cows were sampled at similar times. 72

Plasma was assayed for concentrations of glucose^ and

non-esterIfled fatty acids (NEFA)^. Protein-free filtrates of plasma

(32) were prepared and assayed for beta-hydroxybutyrate (BHBA) (34).

Frozen liver samples were thawed to remove them from sample

vials, dropped into liquid nitrogen, and pulverized by grinding in a

mortar and pestle. One-half of the sample then was homogenized with

10% perchloric acid, with a final ratio of solution to sample of 5 to

1 by weight. The resulting suspension was used for glycogen

determination (22), and a Gilford 2600 spectrophotometer was used for

detection of glucose during glycogen determination.

The remaining half of each sample was placed in 10 ml of a 2 to 1 chloroform/methanol solution. Samples were allowed to shake for at least 2 h (usually overnight), 4 ml H2O was added, and samples were vortexed vigorously. Each sample was centrifuged, the methanol-water layer was removed, the liver tissue was removed, and the chloroform layer was evaporated under nitrogen. The sample then was brought into solution by using 95% ethanol and assayed for triglycerides^. All chemicals were purchased from Sigma Chemical Co., St. Louis, MO.

^Glucose (Trinder) 500 kit was purchased from Sigma Chemical Co., St. Louis, HO.

^NEFA C kit was purchased from WAKO Pure Chemical Industries, Ltd., Osaki, Japan.

^Triglyceride (GPO-Trinder) kit was purchased from Sigma Chemical Co., St. Louis, MO. 73

Statistical analysis

Data were analyzed by using GLM (29) to compare treatment

differences at each stage of the experiment and stage differences

within treatments. A split-plot design (143 degrees of freedom) was used with treatments as the main effect (2 degrees of freedom) and with cows within treatment as error A (15 degrees of freedom), and with stage (7 degrees of freedom) and treatment by stage Interaction

(14 degrees of freedom) as subplots. One KIP cow only reached the subclinical ketosls stage of the experiment, and data for the K and R stages are not available. Also, three control cows were not sampled at the late induction stage. Therefore, LSMEANS were calculated and used for statistical analysis (29). 74

RESULTS AND DISCUSSION

Milk production, feed composition, and body weight loss

Data for milk production, feed consumption, and body weight loss

are shown In Table 3 for each treatment at each stage of the

experiment. Milk production Increased In all cows between EPF and PI, but only KIP and GT were significant (P<.05). Production of milk for both control and GT cows remained relatively constant for the remainder of the experiment. KIP cows did not show a significant decrease In production (P>.2) when compared to controls until a dramatic decrease at ketosls. This result hints that the FRBD protocol given to normal cows establishes conditions that are similar to on-farm ketosls. Values for milk production In KIP cows at LI were significantly different from those at PI (P<.05), but not from those at EI or MI. These changes represent an average of only .3 kg/d loss

In milk, emphasizing the subtle but significant losses In milk production that can occur during subclinical ketosls.

The persistency of milk production In GT cows gives the first

Indication that the duodenal glucose Infusion acts In some way to reverse the effects of FRBD and decrease the Incidence of ketosls.

During stage K, differences In milk production between KIP and GT were

9.0 kg/d. Assuming 5% lactose in milk, 450 g of glucose would be required to synthesize the extra milk, assuming all other milk precursors were available. If 65% of the Infused glucose is absorbed TABLE 3. Milk production, feed consumption, and body weight changes in cows^

Stage of experiment^ (days postpartum)

EPP PI EI MI LI K R SE (5) (13) (21) (28) (35) (44) (60)

Milk production (kg/d) Control 28.4 31.0 31.7 32.5 31.8 32.la 32.2 1.8 KIP 30.0 35.2 33.0 31.4 28.9 22.7b 34.2 1.2 GT 26.4 31.9 31.3 32.5 31.5 31.7a 33.0 1.1

Feed consumption (kg DM/d) Control 13.5 13.2 14.3a 15.03 16.7a 15.2a 17.2 1.2 KIP 13.7 14.8 11.lb 11.2b 11.5b 9.2b 14.9 1.3 GT 14.3 15.1 11.9b 11.lb 10.ob 10.9b 15.3 1.0

Cumulative body weight loss (kg)^ Control 0 29.5 21.9 36.4 31.0* 25.8a 33.4 9. 5 KIP 0 15.6 34.1 42.3 72.6b 78.9b 33.5 10. 8 GT 0 27.2 14.5 37.4 75.6b 59.3b 31.4 10. 2

^Treatment means with different superscripts within a column are significantly different P<.05. Means within a column without superscripts are not significantly different at P>.05.

^Stage effects within a treatment will be noted in text. The code for stage effects is found in Table 2.

^Early postpartal weight used as reference for body weight loss. 76

(15), only 340 g/d of glucose would be available to the mammary gland.

Further, it is estimated in low-producing cows that only 70% of

glucose available is used for milk lactose by the mammary gland (1),

and infusions of glucose into the abomasum of healthy cows in negative

energy balance did not increase milk production (15). Therefore, it seems that the effect of reversing ketosls by Intraduodenal infusions of glucose cannot be entirely because of the contribution of glucose to milk lactose.

Feed consumption data (Table 3) for KIP and GT cows reflect the feed restriction from EI through K stages. GT cows consistently ate all of the feed offered during the four stages of induction, but KIP cows showed a trend to decrease intake further at K (P<.15), with three of the six KIP cows becoming anorexic. Hypophagia is a common symptom of clinical ketosis (30).

Data for body weight (Table 3) indicate that endogenous nutrient stores were being used for energy in all groups of cows and the rate of body weight loss was similar up through the mid-induction stage.

During the last two induction stages, however, KIP and GT cows rapidly lost weight. It should be noted that GT cows continued to lose weight, even though milk production was maintained and they showed no outward signs of ketosis. The two-week recovery period, marked by increased feed intake, was sufficient to restore body reserves to control values for both KIP and GT cows. The body weight losses seen 77

In KIP cows were not as severe as, but more variable than, those seen

by Mills et al. (25) with cows overfed In the dry period.

Metabolites in blood

Plasma glucose concentrations (Table 4) were variable for control

and KIP treatments, but did show some trends. In general, glucose

decreased after calving in control cows, but was significant only in

MI and LI stages (P<.05), possibly when cows were in the most negative

energy balance. KIP cows had lower glucose concentrations at PI, but

this corresponded with higher milk production (Table 3). Glucose concentrations decreased significantly (P<.05) from PI and remained low at MI and K. Glucose concentration of KIP cows did not parallel the dramatic drop in milk production seen during K. Similar results were seen by Mills et al. (25). The increase at LI is unexplainable, but two cows had unusually high glucose values. After recovery, plasma glucose concentrations of KIP cows returned to EPP values but were significantly lower than controls (P<.01).

Glucose concentrations in GT cows were unexplainably higher than in KIP cows for EPP and PI. Glucose infusions, begun at d 25 postpartum, resulted in increases in plasma glucose at LI (P<.05), and the elevated glucose concentrations persisted into stage K even after glucose infusions were terminated. Two GT cows developed severe nervousness and hyperexcitability during the last week of infusion.

These two cows were shown later to have glucose concentrations greater TABLE 4. Metabolite concentrations in blood plasma from control, ketosis-Induced (KIP), and glucose-treated (GT) cows^

Stage of experiment^ (days postpartum)

pp EPP PI EI MI LI K R SE (-10) (5) (13) (21) (28) (35) (44) (60)

Glucose (mg/dl) Control 63 60a.b 62^ 58* 56* 50* 62* 67* 3 KIP 62 53a 55b 47b 42b 59a,b 44b 55b 4 GT 63 68^ 61* 63* 62* 73b 74* 67* 3

Non-esterified fatty acids (mH) Control .54 .60 .62*. b .46* .54* .45* .36* .15 ,12 KIP .51 .75 .83b 1.42b 1.25b 1.19b 1.02b .50 ,13 GT .48 .46 .49a .68* .67*,b .45* .43* .25 .10

Beta-hydroj^butyrate (mg/dl) Control 3.2 4.1 3.8 4.6* 4.4* 4.5* 4.7* 5.0 2.4 KIP 4.5 5.4 5.6 16.8b 27.ob 36.9b 33.5b 7.5 2.8 GT 5.3 4.2 5.5 8.2* 9.4* 8.9* 8.6* 4.1 2.2

^Treatment means with different superscripts within a column are significantly different, P<.05.

^Stage effects within a treatment will be noted in text. The code for abbreviations for stages is found in Table 2. 79

than 90 mg/dl. These results are contrary to those in sheep (24)

where glucose infusions into the blood are known to increase plasma

insulin concentrations, which decrease plasma glucose (11). Data for

GT cows further support the possibility that glucose absorbed from the

intestine alters metabolic regulation as well as providing glucose carbon for tissues.

Plasma non-esterified fatty acid (NEFA) concentrations reflect the net effect of lipolysis from adipose tissue and subsequent uptake of free fatty acids by the liver and other tissues. NEFA concentrations as high as 1.2 mH in plasma have been measured for ketotic cows (7, 31), and plasma NEFA concentrations have been suggested to be Inversely related to carbohydrate sufficiency in cows

(23). NEFA concentrations for control cows (Table 4) were relatively constant during the first 40 d postpartum, with a slight decrease during K and then a significant decrease (P<.01) at R. NEFA for KIP cows, however, significantly increased at the beginning of induction

(EI) and remained elevated through stage K. Stages EI and HI were significantly different (P<.05) from PI. Values at PI seem high, but could be explained by the low plasma glucose concentration in this group of cows. NEFA concentration at K seemed to decrease, although data were quite variable and individual values ranged from .5 to 1.5 mM. Values for GT cows reflected the trend of normalcy already shown for milk production and plasma glucose concentrations. NEFA values tended to Increase at the beginning of FRBD, but decreased after the 80

Initiation of glucose Infusions and remained low.

BHBA concentrations In plasma (Table 4) paralleled NEFA values

for all treatments, with the exception that BHBA generally continued

to Increase from EI to K rather than to decrease. For KIP cows, BHBA

concentrations at PI, EI, HI, and LI were progressively significantly

different from each other (P<.05), and stage K was different from R

and EI. Mills et al. (25), with limited sampling, reported a pattern of ketogenesls in liver slices that is similar to the plasma changes seen for BHBA with KIP cows. Our results suggest that even though

NEFA concentration Increases early, BHBA concentration in plasma

Increases gradually, peaking at about 1 wk before signs of clinical ketosis are detected. BHBA has been shown to exert a regulatory effect in plasma by inhibiting llpolysis of adipose tissue (13). The occurrence of this regulation could explain the gradual decrease in

NEFA concentration in the latter stages of ketosis.

Liver composition

Effects of KIP and GT treatments on liver composition are shown in Table 5, and provide further evidence for the progressive development of the ketotic state for KIP cows and the prevention of ketosis for GT cows. Glycogen content of liver decreased from the EPF to PI stage for control and KIP, but for unknown reasons not for GT cows. For KIP cows decreased glycogen content of liver may reflect the severity of negative energy balance. Therefore, assuming GT cows are in negative energy balance, the cause of increased glycogen at PI TABLE 5. Liver composition of control, ketosis-induced (KIP), and glucose-treated (GT) cows^

Stage of experiment^ (days postpartum)

PP EPP PI EI MI LI K R SE (-10) (5) (13) (21) (28) (35) (44) (60)

Glycogen (% wet wt) Control 3.0 2.1 2.1* 1.8* 2.4* 1.8* 2.1*,G 2.0 .5 KIP 3.4 3.1 1.4a 2.ia.b .23b .06b .82* 1.8 .5 GT 3.5 3.4 4.2b 3.0b 3.2° 3.9c 3.7b 2.8 .5

Triglyceride (% wet wt) Control 1,0 2.2 2.6* 2.5*.^ 3.7* 1.5* 1.4* 1.0* .7 KIP 1.1 2.6 3.2* 3.8* 8.9b 9.0b 9.8b 6.6b .7 GT 1.5 .9 l.lb 1.6b 1.3* 1.0* 1.1* .6* .6

^Treatment means with different superscripts within a column are significantly different, P<.05.

^Stage effects within a treatment will be noted in text. The code for abbreviations for stages is found in Table 2. 82

Is unclear. Glycogen content decreased significantly at HI, nearly

reached zero by LI, and then increased slightly at K. The timing of

these changes corresponded exactly inversely to changes in liver

triglycerides.

Triglycerides increased significantly at HI and remained steady

through K, nearly reaching 10% of liver wet weight. Studies with

ketotic cows have found liver triglyceride of up to 10% of wet weight

(4, 18). Unlike other metabolites, liver triglyceride did not return

to control values after a two-week recovery, indicating a continuing

inability to transport fat out of the liver. Liver triglycerides of

GT cows were lower at EPF and PI than control or KIP and increased

slightly at EI. They remained low, however, during glucose Infusion

and the recovery periods.

Two major differences can be noted between the present results

and those of Hills et al. (25) who used a similar FRBD protocol.

First, prepartum glycogen content of the present cows is about half

that of cows used by Hills et al. (25). This most likely reflects differences between normal feeding versus overfeeding in the dry period. Second, both control and KIP cows increased liver triglycerides slightly in early postpartum in this study; whereas.

Hills et al. (25) reported a 6-fold Increase in liver triglycerides after calving. They concluded that both a rapid decrease in glycogen and an increase in triglyceride content in liver Is a prerequisite for a cow to be susceptible to ketosis using the FRBD protocol. In 83

contrast, our results show that cows with normal liver glycogen and triglyceride in early postpartum can be induced into ketosis. Our results do concur with earlier work (25), however, suggesting that a depletion of glycogen and an accumulation of triglyceride in liver is a prerequisite for clinical manifestations of ketosis. 84

GENERAL DISCUSSION

Studies with fasted cows (5, 18) and with spontaneously-ketotlc

cows (4, 31) have shown similar blood and liver metabolite profiles as

seen during stage K for KIP cows. In addition, our studies with cows

that were not overfed in the dry period and given a revised FRBD

protocol gave results nearly identical to those of Mills et al. (25).

The impact of the present study, however, is not in characterizing the ketotic state of lactating cows, but it is in exploring the progressive development of ketosis and the metabolic implications of a preventative agent. Before discussing this impact, limitations of the ketosis-induction protocol must be presented.

The success of the induction protocol seems to require both feed restriction and a source of ketone bodies. Moderate feed restriction will enhance the hypoglycemic state, by limiting gluconeogenic substrates, thereby stimulating llpolysls without a drastic decrease in rumen fermentation products. Moderate feed restriction alone however will not cause ketosis to occur In cows (14). The inclusion of a ketone body source therefore provides a hyperketonemic state to enhance hypophagla. Preliminary studies in our laboratory have shown that neither feed restriction nor 1,3-butanediol supplementation given separately is sufficient to Induce ketosis (J. K. Drackley, Iowa State

University, unpublished data). Therefore, in our KIP cows given FRBD, moderate hyperketonemla is an experimental constant, not a response. 85

However, because fat accretion has been shown to occur in the liver of

most high-producing cows in early lactation (27), hyperketonemia seems

within physiological normalcy for a cow in early lactation.

Hills et al. (25) suggested that changes in blood and liver

metabolites that lead to ketosis occur sometime before signs of

clinical ketosis are seen. By using more intensive sampling, we have

been able to explore the timing of these changes in detail. From

Tables 4 and 5, we can summarize that: a) plasma NEFA concentrations

peak three weeks before ketosis and remain high, b) increases in BHBA

in plasma begin three weeks before and peak one week before ketosis,

c) liver glycogen content decreases dramatically two weeks before and is nearly zero one week before ketosis, and d) liver triglyceride content increases dramatically two weeks before ketosis and remains high. These results indicate that a metabolic state of preketosis, or ketosis susceptibility, exists up to two weeks before signs of clinical ketosis are detected.

Glucose infusions into the duodenum of GT cows effectively prevented the changes in blood and liver metabolites that occurred in

KIP cows (Tables 4, 5). The nonsignificant (P>.2), but consistent increase in BHBA concentration (Table 4) for GT when compared to control cows could account for the metabolism of butanedlol to BHBA in the liver and release of BHBA into plasma. The loss of body weight for GT cows (Table 3) was similar to KIP cows, therefore suggesting that body reserves still were being mobilized for energy in GT cows 86

during glucose infusion.

An attempt to explain the mechanism of action of glucose infusion

into the intestine of 6T cows centers on two main effects: 1)

providing glucose carbon for lactose synthesis in the mammary gland,

and 2) stimulation of regulatory mechanisms to increase

gluconeogenesis and increase carbohydrate sufficiency. The first

seems obvious, but, as mentioned already, it is inadequate to

completely account for the increased lactose synthesis when milk

production in GT cows is maintained at levels for control cows.

Four major differences in metabolism between KIP and GT cows were measured. First, glucose concentrations in plasma were much greater in GT cows. Second, be ta-hydroxybutyrate concentrations in plasma were much less when compared to KIP cows, suggesting a decrease in hepatic ketogenesis or an increase in extrahepatic utilization of ketone bodies. Third, glycogen content of liver from GT cows remained at levels seen for control cows whereas glycogen in liver of KIP cows was nearly depleted (Table 5). Fourth, there was no increased accretion of triglyceride in liver of GT cows. It does not seem possible that any one of these effects alone could be the reason why infused glucose reversed the ketotic condition in GT cows. Instead, it is feasible that these effects are a response to a stimulated regulatory mechanism not measured in this experiment. One possibility is a direct hormone response.

Glucose infusions into the duodenum and into blood have been 87

shown to Increase plasma concentrations of insulin (24) and decrease

plasma concentrations of glucagon (8, 9). These conditions would result in increased extrahepatlc utilization of glucose, decreased plasma concentrations of glucose, and decreased hepatic gluconeogenesis (10). Thus, such a hormone response would not help to explain the metabolic characteristics of GT cows. However, Hove (19) reported that the secretory response of insulin to glucose infusion into the blood was decreased when cows were ketonemlc. Therefore, another possible effect of the glucose infusions in GT cows is a decrease in glucagon concentrations in plasma with no increase In plasma insulin, which would result In Increased glycogenosis in liver and decreased llpolysls in adipose tissue (10).

A deficiency of OAÂ in mitochondria of liver has been suggested to decrease gluconeogenesis and to Impair complete oxidation of NEFÂ in ketotic cows (23). This hypothesis is supported by data from the present study that showed that liver glycogen was depleted from KIP cows before signs of clinical ketosls were detected. The glucose infusions into the intestine of GT cows, therefore, could be providing glucose to the liver that will spare OAA and prevent an OAA deficiency in the mitochondria. The repletion of OAA could enable fatty acids to be oxidized more completely and decrease ketone body production, a result reflected by decreased plasma concentrations of beta-hydroxybutyrate in GT cows. Increased fatty acid oxidation could explain the decrease in fat accretion in liver of GT cows. Also, NEFA 88

concentrations in plasma remain low even though the GT cows are losing

more weight than controls. This response suggests that llpolysls and

muscle catabollsm remain Increased In GT cows, that plasma NEFA are

taken up by the liver, but cows do not become hyperketonemlc.

Although this hypothesis has not been proven, It does explain the data

collected In GT cows during glucose Infusions.

In summary, the FRBD protocol for Inducing ketosls consistently

provided cows that mimicked spontaneously-ketotic cows. Further, ketosls developed at an average 44 d postpartum, but changes in blood

metabolites and liver composition occurred up to 2 wk before signs of

clinical ketosls were detected. Finally, 500 g glucose/d Infused continuously into the duodenum during the FRBD protocol effectively prevented both the metabolic aberrations and visual signs of ketosls in cows. Further investigations of the metabolic changes in liver of these eighteen cows are reported (33). 89

ACKNOWLEDGEMENTS

The authors thank M. D. Kenealy, T. E. Altchlson, and M. Shlpka for use of facilities and help with management of the cows. This research was supported partly by funds from Eastman Chemicals Co., and partly by Grant No. US-978-85 from BARD - The United States-Israel

Binational Agricultural Research and Development Fund. 90

REFERENCES

1. Annlson, E. F., R. Blckerstaffe, and J. L. Llnzell. 1974. Glucose and fatty acid metabolism in cows producing milk of low fat content. J. Agric. Sci. 82:87.

2. Baird, G. D. 1977. Aspects of ruminant intermediary metabolism in relation to ketosis. Blochem. Soc. Tran. 5:819.

3. Baird, G. D. 1982. Primary ketosis in high-producing dairy cows: clinical and subclinical disorders, treatment, prevention, and outlook. J. Dairy Sci. 65:1.

4. Baird, G. D., K. G. Hlbbltt, G. D. Hunter, P. Lund, M. Stubbs, and H. A. Krebs. 1968. Biochemical aspects of bovine ketosis. Blochem. J. 107:683.

5. Baird, G. D., R. J. Heitzman, and K. G. Hlbbitt. 1972. Effects of starvation on intermediary metabolism in the lactating cow; A comparison with metabolic changes occurring during bovine ketosis. Blochem. J. 128:1311.

6. Bauman, D. E., and W. B. Currie. 1980. Partitioning of nutrients during pregnancy and lactation. A review of mechanisms involving homeostasis and homeorhesis. J. Dairy Sci. 63:1514.

7. Bergman, E. N. 1971. Hyperketonemla-ketogenesis and ketone body metabolism. J. Dairy Sci. 54:936.

8. Berzins, R., and J. G. Manns. 1979. How concentrate feeding affects glucoregulatory hormones in ruminants: Implications in bovine ketosis. J. Dairy Sci. 62:1739.

9. Brockman, R. P. 1977. Glucose-glucagon relationship in adult sheep. Can. J. Comp. Med. 41:95.

10. Brockman, R. P. 1979. Roles of glucagon and insulin in the development of ruminant ketosis: A review. Can. Vet. J. 20:121.

11. Brockman, R. P. 1983. Effects of insulin and glucose on the production and utilization of glucose in sheep. Comp. Blochem. Physiol. 74A:681.

12. Brumby, P. E., M. Anderson, B. Tuckley, J. E. Storry, and K. G. Hlbbltt. 1975. Lipid metabolism in the cow during starvation-Induced ketosis. Blochem. J. 146:609. 91

13. Dawes, D. J., S. C. Sensenlg, and R. N. Heltmann. 1985. Autoregulatlon of ketogenesls in fasted sheep. Fed. Froc. 44:548.(abstr.)

14. deBoer, 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.

15. Elliot, J. M. 1976. The glucose economy of the lactating dairy cow. Froc. Cornell. Nutr. Conf. Ithaca, NY.

16. Fox, F. H. 1971. Clinical diagnosis and treatment of ketosis. J. Dairy Sci. 54:974.

17. Hibbitt, K. G. 1979. Bovine ketosis and its prevention. Vet. Rec. 105:13.

18. Hibbitt, K. G., and G. D. fiaird. 1967. An induced ketosis and its role in the study of primary spontaneous bovine acetonaemia. Vet. Rec. 81:511.

19. Hove, K. 1978. Insulin secretion in lactating cows: Responses- to glucose infused intravenously in normal, ketonemic, and starved animals. J. Dairy Sci. 61:1407.

20. Hove, K., and K. Halse. 1978. Absence of feeding-induced variations in plasma insulin in hypoglycemic-ketonaemic cows. Acta Vet. Scand. 19:216.

21. Hughes, J. P. 1962. A simplified instrument for obtaining liver biopsies in cattle. Am. J. Vet. Res. 23:1111.

22. Keppler, D., and K. Decker. 1974. Glycogen determination with amyloglucosidase. Page 1127 ^ H. U. Bergmeyer, ed. Methods of enzymatic analysis. Vol. 3. 2nd ed. Academic Press, New York, NY.

23. Krebs, H. A. 1966. Bovine ketosis. Vet. Rec. 78:187.

24. Manns, J. G., and J. M. Boda. 1967. Insulin release by acetate, propionate, butyrate, and glucose in lambs and adult sheep. Am. J. Physiol. 212:747.

25. Mills, S. E., D. C. Beitz, and J. W. Young. 1986. Characterization of metabolic changes during a protocol for inducing lactation ketosis in dairy cows. J. Dairy Sci. 69:352. 92

26. Mills, S. E., D. C. Beltz, and J. W. Young. 1986. Evidence for impaired metabolism In liver during Induced lactation ketosls of dairy cows. J. Dairy Scl. 69:362.

27. Reld, I. M. 1980. Incidence and severity of fatty liver In dairy cows. Vet. Rec. 107:281.

28. Reld, I. M., A. J. Stark, and R. N. Isenor. 1977. Fasting and refeedlng In the lactatlng dairy cow. 1. The recovery of milk yield and blood chemistry following a six-day fast. J. Comp. Path. 87:241.

29. SAS Institute, Inc. 1982. SAS user's guide: Statistics. Gary, NO.

30. Schultz, L. H. 1968. Ketosls In dairy cattle. J. Dairy Scl. 51:1133.

31. Schwalm, J. W., and L. H. Schultz. 1976. Relationship of Insulin concentration to blood metabolites In the dairy cow. J. Dairy Scl. 59:255.

32. Somogyl, M. 1945. Determination of blood sugar. J. Biol. Chem. 160:69.

33. Veenhulzen, J. J., J. K. Drackley, M. J. Richard, and J. W. Young. 1987. Effects of Induced and treated ketosls on in vitro hepatic gluconeogenesls and In vivo glucose and propionate metabolism In cows. J. Dairy Scl. (submitted)

34. Williamson, D. H., and J. Mellanby. 1974. D-(-)-3-hydroxybutyrate. Page 1836 ^ H. U. Bergmeyer, ed. Methods of enzymatic analysis. Vol. 3. Academic Press, London. 93

SECTION II: EFFECTS OF INDUCED AND TREATED KETOSIS ON IN VITRO HEPATIC GLUCONEOGENESIS AND IN VIVO GLUCOSE AND PROPIONATE METABOLISM IN COWS 94

ABSTRACT

During the 9 wk immediately after parturition, 18 high-producing

cows were assigned equally among three treatments: control,

ketosls-Induced, and glucose-treated cows. Ketosls was induced at an

average of 44 d postpartum and 29 d after Initiation of a protocol for

inducing ketosls that included restricting feed to 80% of ad lib and

supplementation with 1,3-butanediol. Glucose-treated cows received

the ketosls Induction protocol plus 500 g of glucose/d Infused

continuously into the duodenum from d 25 to 40 postpartum. Liver

biopsies were taken at eight stages: prepartal, early postpartum,

prelnduction, early induction, mid-induction, late induction, ketotlc,

and recovery. In vitro hepatic lipogenic rate from acetate was zero for all treatments, and gluconeogenic rates were not significantly different until ketosls when ketotlc cows decreased four-fold. Media without added calcium resulted in six-fold lower gluconeogenic rates than media with calcium for all stages of ketosls development.

Gluconeogenic rates for glucose-treated cows were not significantly different than for control cows. Simultaneous injections of

[U-l^C]propionate and [6-%]glucose were made into two control and two ketosls-induced cows. Results from kinetics were consistent with decreases in hepatic gluconeogenic capacity seen in vitro. Overall results suggest that Impairment of hepatic gluconeogenic capacity occurs only at the clinically-ketotic stage and corresponds with 95

decreases of In vivo glucose turnover. Glucose infusions during development of ketosis seem to prevent any decrease in gluconeogenic capacity of bovine liver. 96

INTRODUCTION

Lactation ketosis In dairy cows Is associated with hyperketonemla

and hypoglycemia during early lactation (38). It is accepted

generally that these conditions result from the Increased demands by

the mammary gland for nutrients to support milk production and the

limited feed intake during the first 3 wk of lactation. Decreases in

Insulin concentrations in plasma result in Increased lipolysls from

adipose tissue, thereby supplying fatty acids for ketogenesis in

bovine liver (10). Glucose is used preferentially by the mammary gland for lactose synthesis. Much less is known about intermediary metabolism in liver during ketosis.

A decrease In hepatic gluconeogenesls during ketosis has been suggested to result from a decrease in available substrates (3). This suggestion is supported by work with cows (5, 6, 7) showing that activities for key gluconeogenic enzymes are not decreased during ketosis. However, in a recent study with cows induced into ketosis,

Mills et al. (30) showed that in vitro gluconeogenic capacity of liver decreased during ketosis from all substrates when liver was incubated with separate saturating amounts of alanine, aspartate, glutamate, lactate, and propionate.

The cause of impaired hepatic metabolism is not known, but it occurred (30) when concentrations of ketone bodies and triglycerides were increased in livers of ketotlc cows (29). These changes seemed 97

to contradict data (25), which showed Increases In gluconeogenic

capacity during ketonemla and glucosurla In steers. The exact role of

decreased capacity for hepatic gluconeogenesls In the development of

ketosls In cows remains unclear.

The first part of the present study is to examine the progressive changes In hepatic gluconeogenesls and llpogenesls measured In vitro during ketosls-Induction and ketosls-prevention protocols (38).

Calcium has been suggested to have a role In Increased gluconeogenic capacity In ruminants (25) and nonrumlnants (18). Therefore, estimates of gluconeogenic capacity were made also with and without calcium In the media.

Limited studies have been conducted to measure glucose turnover

In vivo as affected by ketosls. Kronfeld et al. (22) reported that glucose Irreversible loss Is not affected by early stages of ketosls, but there Is a trend for decreased glucose turnover when clinical ketosls Is detected (20). Also, consistent increases in glucose distribution space have been seen in ketotlc cows (25, 21). The second part of the present study, therefore, was to make limited measurements of plasma glucose and rumen propionate irreversible loss in vivo at different stages of ketosls induction and to make direct comparisons to hepatic gluconeogenic rates measured in vitro. 98

MATERIALS AND METHODS

Management of cows and design of experiment

Management of the seventeen Holsteln and one Brown Swiss cow, the experimental protocol used to induce ketosls, and the sampling schedule have been described (38). Briefly, six cows were assigned randomly to each of three treatments: control, ketosis-induction protocol (KIP), or glucose-treated (GT). Ketosis induction was achieved by restricting feed Intake by 20% and supplementing the diet with 1,3-butanedloll (FRBD) from day 15 postpartum until clinical ketosis was detected. The GT cows received FRBD, and an indwelling catheter was placed into the duodenum on d 23 postpartum. Glucose was infused continuously at 500 g/d from d 25 to 40 postpartum. Clinical ketosis was detected in KIP cows at an average of 44 d postpartum.

Milk production, metabolic profiles in blood, and composition of liver have been reported (38).

Two control and two KIP cows were selected for in vivo kinetic measurements and treated identically to other cows with the following

^The 1,3-butanediol (1,3-butylene glycol) was donated by the Celanese Chemical Co., New York, NY. 99

exceptions. These four cows were fed 24 times a day by automatic

feeder (14) and housed away from the main herd at the Iowa State

University Dairy facility. Previous results from steers Indicate that

cattle are in metabolic steady state when fed at least every 2 h (2).

Cows were housed Indoors in tie stalls at all times.

One cow scheduled for kinetic studies on the KIP treatment was

diagnosed with clinical ketosis at d 56 postpartum. The other cow on

KIP, however, unexpectedly and quickly developed severe ketosis overnight after only 5 days on FRBD, and she sustained lacerations of the back legs and paralysis of the left foreleg when she fell during the night. She had to be allowed to recover, but after 42 d recuperation, she was started on the FRBD protocol again as if she had just calved. Therefore, samples were taken with a similar schedule to other cows, but 50 d later in lactation. After 35 days of FRBD protocol the cow showed nervousness, a weak gait, loss of body weight, ketonemla, and decreased milk production, but she did not show anorexia typical of lactation ketosis. Data from this cow are presented for all stages except when ketosis was detected in other cows (stage K).

Biopsies of liver

Biopsies of liver were taken 10 d prepartum and at approximately

5, 13, 21, 28, 35, 44, and 60 d postpartum. These times correspond to seven stages of the experiment (38), and are summarized in Table 1. 100

TABLE 1. Codes for stage of experiment and sampling of control, ketosls-induction (KIP), and glucose-treated (GT) cows

Code Stage of experiment* Day of sampling(+SE)b

EPP early postpartum 5±.4

PI pre-induction 13+.3

EI early induction 21+.3

MI mid-Induction 28+.3

LI late induction 35+.4

K ketotlc 44+5.8

R recovery 60+3.7

^Descriptions are for stages of ketosls induction for KIP cows. The same stage description is used for control and GT cows.

^Parturition is the reference point.

Three to four grams of liver were taken by puncture biopsy (16) just

before the morning feeding through a 20-mm incision in the skin at the

11th intercostal space. The incision was closed with suture staples^ and top-dressed with antibiotics^.

^Précisé* suture staples purchased from 3M Company, Minneapolis, MN.

^Furacin powder, purchased from Iowa State University Veterinary Services, Ames, lA. 101

Skin Incisions were made about 20 cm below the spinal vertebra

and postmortem analysis of two cows after the experiment revealed the following observations. In general, biopsies were made Into the

dorsal region of the central lobe, very close to the caudate lobe and the portal vein. It was apparent from postmortem that biopsies should be directed cranlally and ventrally, staying as close to the rib cage as possible.

Taking a liver biopsy did not depress feed Intake or milk production for most cows. Some cows, however, did refuse the morning feed, but resumed normal Intake by the evening feeding. For one cow, the biopsy was made at the 12th Intercostal space and Intestine was punctured. The wound In the Intestine was closed by the Iowa State

Veterinary Service, and the cow returned to normal within 5 days and continued on the experiment.

In vitro Incubations

Preparations of solutions and substrates for Incubations were as described by Mills (27). Immediately after being obtained by biopsy, liver was placed in cold .15 M sodium chloride buffered to pH 7.4 and stored in ice. Liver then was transported to the laboratory and

Incubations began within 30 to 60 min after biopsy. Thin slices of liver, approximately 100 mg, were prepared with a Stadie-Rlggs microtome (36), and were incubated in 25-ml erleruneyer flasks containing 3 ml of bicarbonate buffer (24), pH 7.4, and added 102

substrates. For estimations of In vitro gluconeogenic capacity, 30

Vlffloles of propionic acid plus approximately 1 yCl Na-[2-propionate

were used as the substrate with and without calcium in the buffer.

Ten ymoles of sodium acetate with approximately 1 y CI of

Na-[U-^^C]acetate added was used in flasks to test llpogenlc capacity

of liver slices. Bovine serum albumin (BSÂ) was added to the buffer

so that the final concentration was .03%, and 5 nH glucose was to be

added to the flasks containing sodium acetate. The concentration of

BSA was supposed to have been 3%, but, because long chain fatty acids

were not used as substrates, .03% BSÂ was acceptable. After

completing all incubations, a mathematical error was found resulting

in only .83 nH glucose being added to incubation flasks containing

acetate as the substrate. Later incubations that used 5 mM glucose,

however, did not show increased llpogenlc rates (Table A1 in

APPENDIX). Therefore, the .83 nH glucose used was sufficient for the

limited llpogenlc rates of bovine liver in vitro.

Liver incubations were performed in triplicate. Flasks were

gassed with 95% 02/5% CO2, hanging wells containing folded filter

paper were suspended, and flasks were stoppered. Flasks were

incubated in a Dubnoff metabolic shaker for 2 h at 37^C with constant shaking. Reactions were stopped by injecting .5 ml of 1.5 N H2SO4

into flasks. For ^^C02 collection, filter paper was wetted with 100 yl of 25% NaOH, and flasks were allowed to shake for 1 h after H2SO4 addition. Filter papers then were removed, dried, and counted for 103

radioactivity. Background samples were obtained by incubating tissue

in triplicate in media with .5 ml of 1.5 N H2SO4 added at the

beginning of the incubation.

Protocol for single injections of tracer

Each cow for kinetic studies was prepared with a rumen cannula

(15) about 3 d prepartum. Because cows were not selected for kinetic studies until one wk prepartum, cannulas could not be placed earlier.

Cannulas were not ready for use until d 20 postpartum; therefore, kinetic measurements could not be made earlier from the rumen. Single injections of radioactive tracer were made approximately every 10 days from d 20 until d 60 postpartum. Actual mean days of injections for cows were day 20, 30, 41, 49, and 61 postpartum. A left Jugular catheter was inserted one day before each injection and removed after sampling was completed.

On each injection day 1.6 mCi of [6-^H]glucose was mixed with 10 ml of sterile .15 H NaCl solution, and .9 mCi of Na-[U-^^C]proplonate was mixed with 40 ml distilled water. Aliquots were taken for scintillation counting. Rumen and blood samples were taken before

Injection of tracer to establish background radioactivity. The dose of [6-%]glucose was injected into the jugular catheter and the catheter was flushed immediately with 30 ml of sterile .15 M NaCl.

Two mln after glucose Injection, propionate was dispersed into the rumen with an apparatus designed to ensure adequate mixing of tracer with rumen contents (1). After dispersing the solution 104

containing tracer, 60 ml of water was flushed through the apparatus.

Blood was sampled from the left Jugular catheter at 5, 10, 15,

20, 25, 30, 40, 50, 60, 75, 90, 105, 120, 135, 150, 165, 180, 200,

220, 240, 260, 280, 300, 330, 360, 390, 420, 450, and 480 mln after

Injection of tracer. Blood tubes were heparlnlzed and sodium flourlde was added as a glucose preservative. Rumen fluid was sampled via the cannula with a stainless-steel probe equipped with a strainer on the distal end. Sampling times were at 15, 25, 35, 45, 55, 70, 85, 100,

115, 130, 145, 160, 175, 195, 215, 235, 255, 275, 295, 325, 285, 415,

445, and 475 mln, after injection. Mercuric chloride was added

Immediately to each rumen sample to stop microbial activity.

Analytical procedures

Recovery from media from tissue incubated with [2-^^C]propionate was calculated from an internal standard of either 7,000, 11,000, or

20,000 DPM of [6-3h]glucose, depending on the stock solution of labelled glucose. Because of the potential error in extrapolating data when two Isotopes are counted simultaneously, over 10,000 DFM of

[6-^H]glucose should be used. Media was neutralized with Ba(0H)2, centrifuged, and frozen at -20*'C. Glucose was isolated from thawed samples by using ion-exchange chromatography (28), and [^^C]glucose values were corrected for losses as determined by recovery of

[^H]glucose. Recovery of triglycerides from media incubated with acetate was determined as described by Pothoven and Beltz (32).

Radioactivity was determined with a Beckman LS-8000 scintillation 105

counter.

Glucose from blood plasma of cows used for kinetic studies was

Isolated (28), glucose concentration was determined^, and samples were

counted by liquid scintillation spectrometry to determine [^^C] and

[%]glucose specific radioactivity (SRA). Rumen fluid samples were

.assayed for propionate content by using high-pressure liquid

chromatography (1), and the isolated propionate fractions then were

counted by liquid scintillation spectrometry for calculation of

propionate SRA.

The line of best fit for the decline of plasma [%]glucose SRA

after intravenous injection of [^H]glucose tracer was calculated by

using the nonlinear regression procedure (NLIN) of SAS (34). Although

a 3-pool model has been fit for plasma glucose (23), an exponential

equation with only two terms was necessary to fit the data.

Calculations of irreversible loss, pool size, total entry rate,

recycling, and space for glucose were made from these equations (39).

Appearance and disappearance of [^^C]glucose in plasma after

intraruminal injection of propionate also was fitted to a

two-term equation using NLIN (34) and used to calculate a transfer

quotient, which gives the percentage of plasma glucose derived from

rumen propionate (13).

^Glucose (Trinder) kit was purchased from Sigma Chemical Co., St. Louis, MO. 106

Statistical analysis

Data from Incubations of liver slices were analyzed by using orthogonal comparisons of GLH (35) for treatment differences at each stage of experiment and for stage differences within treatments. A split-plot design (143 degrees of freedom) was used with treatments as the main effect (2 degrees of freedom) and with cows within treatment as error A (15 degrees of freedom), and with stage (7 degrees of freedom) and treatment by stage Interaction (14 degrees of freedom) as subplots. One KIP cow only reached the subclinical ketosis stage of the experiment, and data for the K and R stages are not available.

Also, three control cows were not sampled at the late induction stage.

Because of missing values at some stages and only two replications per treatment for cows on the kinetic study, statistical analyses were not run at all stages of the experiment for kinetic data. Instead, values have been included for a nonstatistical comparlsonand number of cows included are noted in the tables. 107

RESULTS AND DISCUSSION

In vitro Incubations

Effects of treatment and stage of ketosis on the capacity of

liver slices to metabolize propionate to glucose are shown in Table 1.

Data are reported as ymol propionate per h Incubation per g liver

tissue. Other work in our laboratory using similar protocols reported

data as nmol propionate per hour per mg DNA (30). Mills et al. (30)

reported DNA concentrations that ranged from .278 to .428 mg DNA per

100 mg of cow liver that contained from 1.1 to 11.0% fat. Differences

in DNA content between stages of ketosis development, however, were

not significant in that study, and DNA content of liver was not

determined for the present study. Euler and Hahn (12) reported DNA

content of rat liver also to be about .4 mg per 100 mg liver.

Therefore, a conservative conversion from data expressed per mg tissue

to data expressed per mg DNA is to multiply by .003.

Estimated in vitro gluconeogenic capacity (GC) from propionate in liver of control cows remained relatively constant for the first 28 days of lactation when Ca++ was included in the media (Table 2).

There was a consistent increase in GC starting from MI and continuing to R, but only HI and R were significantly different (P<.05). Liver slices from KIP cows also metabolized propionate to glucose at rates that were not significantly different from control slices from stages

EPF to LI (P>.2). At ketosis (stage K), however, a significant TABLE 2. Utilization of propionate for synthesis of glucose In slices of liver from control, ketosIs-induced (KIP), and glucose-treated (GT) cows^

Stage of experiment^ (days postpartum)

Substrate metabolized EPP PI EI HI LI K R SE and treatment group (5) • (13) (21) (28) (35) (44) (60)

ymol propionate converted to glucose/(h x g liver)

Propionate (with Ca++ in media)^ Control 6.9 7.6 6.8 5.3 8.9* 8.7* 7.3 1.3 KIP 6.0 4.6 4.4 3.9 3.1^ 2.0^ 6.2 1.4 GT 4.4 4.5 3.7 2.9 5.1*^ 8.0» 4.4 1.4

Propionate (without Ca++ in media) Control 2.5* 2.1 2.1* .9 2.1* 1.3 1.7 .5 KIP 1.1^ 1.4 .8b .5 .5^ .2 .9 .2 GT .gb 1.4 1.6*b 1.1 1.5*b .8 1.3 .3

^Treatment means with different superscripts within a column are significantly different, P<.05.

^Abbreviations for stages are listed in Table 1.

^Calcium was added to Incubation media to provide a final solution that was .02 M Ca++. 109

decrease In gluconeogenic capacity occurred for KIP cows, being

statistically significant at K when compared to stages EPP and R

(P<.01). The magnitude of decrease In GC seen at K Is greater than

that seen by Mills et al. (30) In cows Induced Into ketosls with a similar FRBD protocol. Gluconeogenic capacity of liver for KIP cows returned to near normal values at R, even though the recovery period for KIP cows Included the termination of FRBD protocol and return to ad lib feeding, but not clinical treatment of the ketotlc condition.

Differences between KIP and GT cows In gluconeogenic capacity of liver slices were significant at stage K (Table 2). Values at EI for

GT cows tended to be lower (P<.08) than those at PI, and remained low until stage LI. The subsequent Increase at K (P<.06) corresponded to the completion of the two-wk glucose Infusions Into the Intestine.

These data suggest that the negative effects of ketosls on gluconeogenic capacity are reversed by glucose Infusions. These data also match changes In other metabolic profiles already reported for GT cows (38).

Comparisons between propionate metabolized to glucose In media with and without Ca++ In Table 1 reveal differences In magnitude, but many similarities In the effect of treatment and stage of ketosls.

Values of gluconeogenic capacity for slices Incubated with Ca++ versus those Incubated without added Ca++ were statistically significant for each treatment at each stage of the experiment (P<.01). Differences between treatments, however, were similar between liver slices 110

Incubated with and without Ca-H-, except for an unexplainable

difference in gluconeogenic capacity between control and both KIP and

GT cows at EPF. The increase in gluconeogenic capacity in GT cows at

K when Ca-H- was included in the media was not seen when Ga++ was

omitted. It is possible that Ca-H could have a role in increased

gluconeogenic capacity of liver of GT cows, but the mechanism of action is unclear. In general, however, data from Table 1 indicate

that the effect of calcium on in vitro incubations is to increase gluconeogenic capacity in all samples, and that calcium does not have a specific therapeutic effect on liver from ketotic cows. Effects of

Ca-H on gluconeogenic capacity of ruminant liver will be discussed later.

Liver slices also were incubated with acetate plus glucose to determine the rate of fatty acid synthesis in ruminant liver. In normal lactating cows, the rate of fatty acid synthesis is nearly zero

(7), but fatty acid synthesis in liver has not been studied in cows with fatty liver. The rate of fatty acid synthesis in liver slices of control, KIP, and GT cows for the present experiment was essentially zero. Therefore, fatty acid synthesis was not induced in cows that were induced into ketosis and does not seem to contribute fatty acids for fat accumulation in liver during fatty liver disease and ketosis.

Metabolism of propionate and acetate to GO2 by liver slices Is shown in Table 3. In general, propionate oxidation to CO2 occurred at one-tenth the rate of propionate metabolized to glucose in vitro. TABLE 3. Metabolism of propionate and acetate to CO2 in slices of liver from control, ketosis-induced (KIP), and glucose-treated (GT) cows^

Stage of experiment^ (days postpartum)

Substrate metabolized EPF PI EI MI LI K R SE and treatment groups (5) (13) (21) (28) (35) (44) (60)

y mol substrate converted to C02/(h x g liver)

Propionate (with Ca++ in media)^ Control .61 .41 .38 .43 .31 .32 .29 .11 KIP .52 .58 .49 .55 .40 .53 .25 .12 GT .21 .30 .42 .24 .29 .16 .29 .13

Propionate (without Ca4+ in media) Control .36* .19 .26 .13 .09 .03* .05 .08 KIP .17* .15 .14 .12 .11 .33b .05 .08 GT .04b .13 .13 .13 .13 .08*^ .07 .08

Acetate Control .54 .61 .67 .87 .63 .76 1.08 .16 KIP .68 .71 1.16 .81 .79 .85 .47 .15 GT .40 .54 .51 .53 .45 .54 .58 .18

^Treatment means with different superscripts within a column are significantly different, P<.05.

^Abbreviations for stages are described in Table 1.

^Calcium was added to incubation media to provide a final solution that was .02 M Ca++ 112

Acetate oxidation to CO2 was greater than that of propionate with and

without added Ca++, although nonsignificantly so. This result most

likely reflects the fact that acetate is not being metabolized to

triglycerides in the liver. The decreases in magnitude of propionate

oxidation between slices incubated with and without calcium are

similar as seen with propionate metabolized to glucose (Table 2).

The Ca-H- effect on gluconeogenesis has been studied in rats and

has been shown to Increase gluconeogenesis by enabling the

alpha-adrenergic activation of enzymes in the mitochondria (18, 19).

Calcium also is thought to inhibit magnesium-regulated enzymes (11),

however, such inhibition would act to decrease gluconeogenesis. The

Ca-H- effect seen in the present study (Table 2) is summarized further

in Table 4, showing propionate metabolism to glucose and CO2 expressed

as a ratio of results from liver slices incubated with and without

Ca-H-. Although considerable variation is present in these ratios,

about a five- to six-fold increase in gluconeogenic capacity was seen with liver slices incubated with Ca-H- versus those without Ca-H-.

The effect of Ca-H- on ruminant liver in vitro was reported first by Lyle et al. (25), who saw a three- to five-fold increase in propionate metabolized to glucose in liver slices of steers when incubated with Ca-H during fasting and treatment with phlorizin and

1,3-butanediol. Lyle et al. (25), however, did not see this effect during the control period for steers and subsequently suggested that

Ca-H- might improve gluconeogenic capacity in liver during ketosis. TABLE 4. Ratio of substrate metabolized with Ca-H- in media^ to that without Ca-H- in media for control, ketosis-induced (KIP), and glucose-treated (GT) cows^

Stage of experiment^ (days postpartum)

Substrate metabolized EPP PI EI HI LI K R SE and treatment groups (5) (13) (21) (28) (35) (44) (60)

Propionate to glucose Control 2.8 3.6 3.3 5.9 4.2 6.6 4.3 5.1 KIP 5.5 3.3 5.6 7.8 6.2 6.0 6.8 7.0 GT 5.5 3.2 2.3 2.6 3.3 7.9 3.6 6.0

Propionate to CO2 Control 1.7 2.2 1.5 3.3 3.5 10.3 5.8 1.5 KIP 3.1 3.9 4.6 4.7 3.6 1.6 15.0 4.4 GT 5.2 2.3 3.2 1.8 2.3 2.0 4.1 1.8

^Calcium was added to incubation media to provide a final solution that was .02 M HCO3.

^Treatment means were not different within a column, P<.05

(^Abbreviations for stages are listed in Table 1. 114

Data from Table 2 and Table 4 Indicate that Ca++ has no unique effect

on liver from ketotlc cows during in vitro incubations. Calcium added

to the media resulted in uniform increases in gluconeogenic capacity

from propionate and did not alleviate the decreased gluconeogenic

capacity during stage K (Table 2).

In vivo kinetics of propionate and glucose metabolism

Kinetic measurements that were made on four of the eighteen cows corresponded to stages EI, MI, LI, K, and R. Dry matter intake and the kinetics of propionate metabolism are shown in Table 5. Number of cows are in parentheses, and kinetic measurements were not done during stage LI for control cows. There were no significant differences in testable stages (stages with more than one cow). Preliminary evaluations can be made, however, and provide some comparison to the data from in vitro incubation conducted with all eighteen cows.

Propionate Irreversible loss (PIL) for control cows was similar to values reported by Wiltrout and Satter (40) and Bauman et al. (8), and PIL seemed to Increase with Increased feed intake for both control and KIP cows. Control cows were fed ad libitum and, by chance, feed

Intake of control and KIP cows were similar, even though ketotlc cows were feed restricted from day 14 to 49 postpartum. A slight decrease in propionate irreversible loss at stage K for KIP cows accompanied a slight decrease in feed intake. Propionate pool size also decreased when ketosis was detected. Overall, the data suggest that the ability 115

Table 5. Kinetics of propionate metabolism in control and ketotic cows

Stage of Experiment* (Days postpartum)

Parameter and EI HI LI K R treatment for cows (20) (30) (41) (49) (61)

(number of cows in parentheses^) DM intake (kg/d) Control 16.4(2) 17.1(2) - 18.6(2) 21.4(1) Ketotic 17.6(2) 17.6(2) 17.6(2) 16.4(1) 23.5(2)

Propionate irreversible loss (mol Prop/d) Control 17.6 24.5 - 20.6 29.1 Ketotic 16.9 21.1 21.3 16.6 27.9

Propionate pool size (mol Prop) Control 1.45 1.48 - 2.35 3.12 Ketotic 1.41 1.47 1.60 .65 5.41

^Abbreviations for stages are listed for Table 1.

^Number of cows is constant for all measured parameters.

of the rumen to produce propionate from feed components is not affected directly by ketosis.

Glucose irreversible loss (GIL) values in Table 6 were more variable than PIL, but were similar to other values from cows in the first 60 to 146 days postpartum (9, 31). Limited numbers of cows make it difficult to interpret data further, and any trends must be verified by more comprehensive studies.

Glucose irreversible loss seemed to decrease when ketosis was 116

Table 6. Kinetics of glucose metabolism in control and ketotic cows

Stage of experiment* (Days postpartum)

Parameter and EI HI LI K R treatment for cows (20) (30) (41) (49) (61)

(pumber of cows in parentheses^) Glucose irreversible loss (mol glucose/d) Control 10.5(2) 8.8(2) - 11.4(2) 9.5(1) Ketotic 12.0(2) 14.7(1) 11.9(2) 9.5(1) 12.7(2)

Glucose pool size (mol glucose) Control .25 .12 - .15 .21 Ketotic .27 .12 .12 .38 .15

Total entry rate (mol glucose/d) Control 15.7 32.1 - 20.7 17.8 Ketotic 16.5 25.6 18.9 11.4 31.8

Recycling (mol glucose/d) Control 5.2 23.3 - 9.3 8.3 Ketotic 4.5 10.9 6.9 1.9 19.1

Glucose space (% body wt) Control 14.8 6.7 - 7.7 7.2 Ketotic 17.2 11.4 11.3 23.7 7.9

Transfer quotient (% glucose from propionate) Control 27.4 55.1 - 37.6 62.5 Ketotic 27.8 33.1 37.7 21.4 53.2

^Abbreviations for stages are listed in Table 1.

^Number of cows is constant for all measured parameters. 117

detected, and to be accompanied by a decrease In total entry rate and

recycling. A similar trend was seen in ketotic cows by Kronfeld (20)

and Kronfeld et al. (22). Glucose space was increased greatly in the

one KIP cow at stage K. Increases in glucose space have been observed

during ketosis in cows (21, 22), and could contribute to the

hypoglycemia that accompanies ketosis.

The transfer quotients expressing the percentage of glucose

derived from propionate (Table 6) initially were equal for both

treatments but thereafter were less for the KIP cows, especially at K.

Wiltrout and Satter (40) reported transfer quotients of 32 and 45% for

nonlactating and lactating cows, respectively. The kinetic

measurements were made at about d 60 postpartum. Therefore, the

possibility exists that an increase in the percentage of glucose

derived from propionate occurs gradually during the first 60 days of

lactation. Further research with a greater number of cows is needed.

Although glucose kinetic data reported in Table 6 are from a

small subset of cows, the data support the extensive in vitro data

presented already. Decreases in GIL, glucose recycling, and the

percentage of glucose from propionate plus the increase in glucose

space seen only at stage K for KIP cows correspond to the decreases in

gluconeogenic capacity from propionate in liver slices at stages K

(Table 2). These data suggest that decreases of in vivo

gluconeogenesis during ketosis is not a result of a decreased 118

availability of propionate from the rumen, although more research is needed. 119

GENERAL DISCUSSION

Data relating to the role of gluconeogenic capacity of liver In

the development of ketosls are conflicting. Measured activities of

key gluconeogenic enzymes In liver were not decreased In ketotlc cows

(5, 6, 7), suggesting that the reactions of the gluconeogenic pathway

were not limiting. Mills et al. (30), however, reported a significant

decrease of in vitro gluconeogenic capacity of liver from five

substrates in cows experimentally-induced into ketosls. Lyle et al.

(25) reported an increase in gluconeogenic capacity from propionate

when steers either were fasted; or were feed-restricted, fed

1,3-butanediol, and given injections of phlorizin.

Our results agree with those of Mills et al. (30), whose values

for propionate metabolized to glucose in media without Ca++ were

nearly Identical to those from liver slices of KIP cows Incubated without Ga++ from the present study. Because treatment and stage effects were almost identical between incubations with and without

Ca++ in the present study, comparisons therefore will be made between

Incubations with Ca++ in the present study and incubations without

Ca++ done previously (30).

Mills et al. (30) concluded that decreased rates of gluconeogenesls in liver of cows has a role in the etiology of lactation ketosls. They further suggested that impaired hepatic function slightly precedes ketosls. The cause for decreased metabolic 120

activity was unclear, although increased accumulation of fat and increased ketone body concentrations in liver (29), as well as possible changes in intracellular pH (17) were Implicated. Our data also show a decrease in gluconeogenic capacity only at ketosis (Table

2), and similar questions concerning cause and effect still can be asked.

An advancement of the present study over that of Mills et al.

(30) is more frequent sampling of liver and blood. Therefore, we were able to show that gluconeogenic capacity of liver in cows developing ketosis does seem to change until less than one week before ketosis is detected. This time is later than changes in other metabolic profiles presented already (38), which show that increased liver triglycerides, increased plasma ketone body concentrations, and decreased concentrations of blood glucose occurred up to two weeks before ketosis was detected. Liver glycogen decreased dramatically from EI to LI, and further research is needed to determine whether the depletion of glycogen is a prerequisite for a decrease in gluconeogenic capacity in liver during ketosis.

That gluconeogenic capacity of liver in GT cows did not decrease at stage K suggests that glucose positively affected gluconeogenic capacity in liver. Glucose infusions in GT cows, however, resulted in reversal of all metabolic conditions preceding ketosis except for mild ketonemia. Therefore, a specific cause for the maintenance of normal hepatic metabolism and prevention of ketosis by infused glucose can 121

not be determined.

Glucose Infusions Into the Jugular vein of normal cows have been

shown to decrease hepatic release of glucose (4). The existence of a

direct effect of glucose absorbed from the Intestine of GT cows,

therefore, on gluconeogenic pathways In liver Is questionable.

McGarry and Foster (26), studying ketogenesls In rat liver, suggested

that TCÂ cycle activity could be Inhibited during Increased ketogenesls. Reld et al. (33) found that the number of mitochondria in liver of cows suffering from fatty liver was decreased. Thus, the glucose effect of eliminating ketosls from GT cows in the present study could be mediated through the decreased fat and ketone body accumulation in the liver.

Data from In vivo kinetic experiments support the conclusions reached with in vitro incubations, but are insufficient to provide

Independent conclusions. Limited numbers of cows used in kinetic experiments is a recurring problem in literature. In general, a decrease in glucose turnover as a result of ketosls has been shown

(20, 22) and is suggested in the present study, but animal variation is great enough that results have not been statistically significant.

The cost of in vivo kinetic experiments with cows and the lack of ketotlc cows in controlled experimental conditions also contribute to this void in reliable literature values.

The present study verifies the finding (30) that gluconeogenic capacity is decreased in experimentally-Induced, cllnlcally-ketotlc 122

cows, and It further suggests that the decrease occurs abruptly within

1 wk of clinical ketosis. Glucose infusions into the duodenum of cows during the development of ketosis resulted in preventing any measured impairment of liver metabolism. It remains unclear, however, whether a sudden decrease in hepatic gluconeogenic capacity causes a cow to change from a state of subclinical to clinical ketosis, or whether impaired liver metabolism is a result of existing conditions after clinical ketosis occurs. 123

ACKNOWLEDGEMENTS

The authors thank M. D. Kenealy, T. E. Altchison, and H. Shlpka for use of facilities and help with management of the cows. This research was supported partly by funds from Eastman Chemicals, Co., and partly by Grant No. US-978-85 from BARD - The United States -

Israel Binational Agricultural Research and Development Fund. 124

REFERENCES

1. Armentano, L. E., and J. W. Young. 1983. Production and metabolism of volatile fatty acids, glucose and CO2 in steers and the effects of monesin on volatile fatty acid kinetics. J. Nutr. 113:1265.

2. Armentano, L. E., S. E. Mills, and J. W. Young. 1984. Effects of feeding frequency on glucose concentration, glucose turnover, and insulin concentration in steers. J. Dairy Sci. 67:1445.

3. Baird, G. 0. 1977. Aspects of ruminant intermediary metabolism in relation to ketosis. Biochem. Soc. Tran. 5:819.

4. Baird, G. D. 1981. Metabolic modes indicative of carbohydrate status in the dairy cow. Fed. Proc. 40:2530.

5. Baird, G. D., K. G. Hibbitt, G. D. Hunter, P. Lund, H. Stubbs, and H. A. Krebs. 1968. Biochemical aspects of bovine ketosis. Biochem. J. 107:683.

6. Ballard, F. J., R. W. Hanson, D. S. Kronfeld, and F. Raggi. 1968. Metabolic changes in liver associated with spontaneous ketosis and starvation in cows. J, Nutr. 95:160.

7. Ballard, F. J., R. W. Hanson, and D. S. Kronfeld. 1969. Gluconeogenesis and lipogenesis in tissue from ruminant and nonruminant animals. Fed. Proc. 28:218.

8. Bauman, D. E., C. L. Davis, and H. F. Bucholtz. 1971. Propionate production in the rumen of cows fed either a control or high-grain, low-fiber diet. J. Dairy Sci. 54:1282.

9. Bennick, M. R., R. W. Mellenberger, R. A. Frobish, and D. E. Bauman. 1972. Glucose oxidation and entry rate as affected by the initiation of lactation. J. Dairy Sci. 55:712.(abstr.)

10. Brockman, R. P. 1979. Roles of glucagon and Insulin in the development of ruminant ketosis: A review. Can. Vet. J. 20:121.

11. Cohn, H. 1963. Magnetic resonance studies of metal activation of enzymic reactions of nucleotides and other phosphate substrates. Biochemistry 2:623.

12. Euler, H. V., and L. Hahn. 1948. Concentrations of RNA and DNA in animal tissue. Arch. Biochem. 17:285. 125

13. Gurplde, E., J. Mann, and S. Lleberman. 1963. Analysis of open systems by administration of tracers at a constant rate or as a single dose as Illustrated by problems Involving steroid hormones. J. Clin. Endocrinol. Metabol. 23:1155.

14. Heath, S. P. 1978. Effect of zeranol on glucose kinetics in growing steers. M. S. Thesis. Iowa State University, Ames, lA.

15. Hecker, J. F. 1969. A simple method for Inserting rumen cannulae In sheep. Aust. Vet. J. 45:293.

16. Hughes, J. P. 1962. A simplified Instrument for obtaining liver biopsies in cattle. Am. J. Vet. Res. 23:1111.

17. Kashlwagura, T., C. J. Deutsch, J. Taylor, H. Ereclnska, and D. F. Wilson. 1984. Dependence of gluconeogenesis, urea synthesis, and energy metabolism of hepatocytes on intracellular pH. J. Biol. Chem. 259:237.

18. Kneer, N. H., M. J. Wagner, and H. A. Lardy. 1979. Regulation by calcium of hormonal effects on gluconeogenesis. J. Biol. Chem. 254:12160.

19. Kraus-Friedmann, N. 1986. What is the roll of Ca++ in the hormonal stimulation of gluconeogenesis? Trends Biochem. Sci. 11:276.

20. Kronfeld, D. S. 1971. Hypoglycemia in ketotic cows. J. Dairy Sci. 54:949.

21. Kronfeld, D. S. 1977. Glucose transport and recycling determined by means of two tracers and multicompartmental analysis. Fed. Pro. 36:259.

22. Kronfeld, D. S., E. G. Tombropoulos, and M. Kleiber. 1959. Glucose blokinetlcs in normal and ketotic cows. J. Appl. Physiol. 206:109.

23. Kronfeld, D. S., C. F. Ramberg, Jr., and D. M. Shames. 1971. Multicompartmental analysis of glucose kinetics in normal and hypoglycemic cows. Am. J. Physiol. 220:886.

24. Lasser, H. 1961. Salt solutions. Pae 58 ^ C. Lon, ed. Biochemist handbook. D. Van Nostrand Co., Inc., Princeton, NJ.

25. Lyle, R. R., K. D. Birkmeyer, and J. W. Young. 1984. In vitro hepatic gluconeogenesis and ketogenesls as affected by prolonged ketonemia-glucosuria and fasting in steers. J. Dairy Sci. 67:2283. 126

26. McGarry, J. D., and D. V. Foster. 1971. The regulation of ketogenesls from octanolc acid. The role of the tricarboxylic acid cycle and fatty acid synthesis. J, Biol. Chem. 246:1149.

27. Mills. S. E. 1982. Metabolic characteristics of the ketotic state in the bovine. Ph.D. Dissertation. ISU. University Microfilms, Ann Arbor, MI. Order No. 82-21,209.

28. Mills, S. E., L. E. Armentano, R. W. Russell, and J. W. Young. 1981. Rapid and specific isolation of radioactive glucose from biological samples. J. Dairy Sci. 64:1719.

29. Mills, S. E., D. C. Beitz, and J. W. Young. 1986. Characterization of metabolic changes during a protocol for inducing lactation ketosis in dairy cows. J. Dairy Sci. 69:352.

30. Mills, S. E., D. C. Beitz, and J. W. Young. 1986. Evidence for impaired metabolism in liver during induced lactation ketosis of dairy cows. J. Dairy Sci. 69:362.

31. Peel, C. J., and D. E. Bauman. 1987. Somatotropin and lactation. J. Dairy Sci. 70:474.

32. Pothoven, M. A., and D. C. Beitz. 1973. Effect of adipose tissue site, animal weight, and long-term fasting on lipogenesis in the bovine. J. Nutr. 103:468.

33. Reid, I. M., R. A. Collins, A. Patterson, and R. J. Treacher. 1981. Organelle changes in the liver cells of dairy cows around the time of calving. J. Comp. Pathol. 91:245.

34. SAS Institute. 1982. SAS User's Guide: Statistics, pp 15-39. Cary, NC.

35. SAS Institute. 1982. SAS User's Guide: Statistics, pp 139-201, Cary, NC.

36. Stadie, W. C., and B. C. Riggs. 1944. Microtome for the preparation of tissue slices for metabolic studies of surviving tissues in vitro. J. Biol. Chem. 154:687.

37. Schultz, L. H. 1968. Ketosis in dairy cattle. J. Dairy Sci. 51:1133.

38. Veenhuizen, J. J., A. D. McGilliard, and J. W. Young. 1987. Blood and liver metabolites and milk yield during the development of experimental ketosis and early treatment of ketosis in cows. J. Dairy Sci. (submitted). 127

39. White, R. G., J. W. Steel, R. A. Leng, and J. R. Lulck. 1969. Evaluation of three Isotope-dilution techniques for studying the kinetics of glucose metabolism In sheep. Blochem. J. 114:203.

40. Wlltrout, D. W., and L. D. Satter. 1972. Contribution of propionate to glucose synthesis In the lactatlng and nonlactatlng cow. J. Dairy Scl. 55:307. 128

GENERAL SUMMARY

The work completed for this dissertation Is a direct outgrowth of

the questions posed by S. E. Mills when finishing his studies of an

experimental protocol for the Induction of ketosls In cows (1). Some

of the questions were: 1) Is the liver biopsy technique safe for

regular sampling of high-producing cows, 2) can ketosls be Induced

Into cows that are not over-fed during the dry period, 3) what Is the

relative significance of Increased ketogenesls and Increased fat

accumulation in the liver on the development of ketosls, 4) does Ca++

have a regulatory effect ongluconeogenesls in the bovine liver, 5)

how do in vitro gluconeogenic capacities compare with in vivo

whole-body turnover of glucose, and 6) are decreases in gluconeogenic

capacity in livers of ketotlc cows a cause or an effect of ketosls?

The liver biopsy technique was used skeptically at first in this

project, and some Iowa State University administrators were concerned

about the health of the cows that would be sampled on a weekly basis

for seven consecutive weeks. Liver biopsies consistently were

successful, however, and over 200 biopsies were made without long-term

illness of cows. Most cows also showed no short-term effects from the

biopsy. One advantage of the present study over that of Mills et al.

(2) was the need for only 2 to 3 g of liver per biopsy. This was done

routinely with one or two punctures per sampling. Mills et al. (2)

needed about 5 to 6 g per biopsy. Since the completion of the present 129

study, another 200 liver biopsies have been made in two subsequent

projects in our laboratory. Only one cow has died from internal

hemorrhage and shock.

Use of the feed restriction, 1,3-butanediol supplementation

(FRBD) protocol in this study resulted in cows developing ketosis

gradually over a 30-d period after having been fed normal diets during

the dry period. Eighty-three percent of the cows on the

ketosis-induction protocol (KIP) developed clinical ketosis, and all

KIP cows developed a form of subclinical ketosis. The metabolic

profiles of these cows were nearly identical to those of Mills et al.

(2) that were over-fed in the dry period. Therefore, over-feeding does not seem to be a requirement for the success of the FRBD protocol in early lactation cows. However, as noted earlier in this dissertation, liver composition of normal cows just after parturition is different than that seen by Mills et al. (2) in cows over-fed in the dry period.

Weekly sampling of liver and blood of cows on the present study made it possible to evaluate the progressive changes in metabolism during development of ketosis in more detail than before. Ketosis induction was begun at day 14 postpartum and ketosis was detected at an average of 44 days postparum. Thus, about 4 wk elapsed during the development of experimental ketosis, which Is similar to the time sequence of on-farm ketosis (4).

A progressive sequence of changes in metabolic profile of cows 130

can be noted during the development of ketosls. First, about 1 wk

after start of Induction protocol, concentrations of

beta-hydroxybutyrate and non-esterifled fatty acids In plasma are

increased greatly. These changes were accompanied by a decrease In

concentrations of blood glucose. Second, during the second week of

FRBD and about 2 wk before signs of ketosls, triglyceride content In

liver increased and glycogen content in liver decreased. Finally,"

milk production and the gluconeogenic capacity of liver decreased 4 wk

after beginning induction protocol and only when ketosls was detected.

These changes suggest that an over-production of ketone bodies

and great demand for glucose can lead to carbohydrate insufficiency

and the accumulation of fat in liver. Further, clinical ketosls does not occur Immediately after onset of the abnormal metabolic conditions

Just described. Signs of ketosls do coincide, however, with measured decreases in gluconeogenic capacity of liver In vitro and possible decreases in gluconeogenesis in vivo. While the data imply direct relationships between the occurrence of abnormal metabolic conditions and the development of ketosls, cause-and-effect relationships cannot be determined until more specific measurements are made.

In addition to studying the development of ketosls, the present study enabled evaluation of the effect of providing exogenous glucose via the gut during the development of ketosls. The beneficial of glucose effect on ketosls has been documented when glucose is given as a therapeutic treatment after clinical ketosls has been diagnosed (2, 131

3, 4), but not as a preventative. Intravenous Infusions of 50% dextrose provide an immediate source of glucose to relieve hypoglycemia and the severe energy deficit. Cows usually will begin eating again shortly after treatment, but milk production remains decreased. Mills et al. (2) showed that blood metabolites and hepatic gluconeogenic capacity return to normal one week after ketotic cows had been given glucose and injections of glucocorticoids, plus returning to full feed.

The results from cows given glucose during FRBD protocol (GT cows) were promising from two aspects. First, the surgery techniques used were new, but proved to be successful. Catheters were implanted into the proximal duodenum while cows were standing, and swelling was minimal. Milk production usually was decreased for only one day and feed intake was not affected after 12 h. Also, catheters were removable with force, and complete closure of the intestinal puncture was found in two cows that were culled and slaughtered later in lactation. Thus, cows could be used for surgery and later returned to the herd without permanent scarring. Second, infusions of 500 g/d of glucose from days 25 to 40 postpartum seemed to prevent ketosis from developing when compared to KIP cows. As discussed in Section I of this dissertation, this effect seems to be more than just a mass action of providing glucose carbon for metabolism.

While some of the questions posed by Mills (1) were answered by the present study, other questions have surfaced. The data presented 132

on the metabolic changes occurring during development of ketosis point

toward carbohydrate insufficiency as a primary cause of ketosis. The

specific role of greatly-increased ketogenesis and lipid accumulation

in the liver in causing carbohydrate insufficiency only can be

speculated. Specific new questions generated by the present work are:

a) does increased ketogenesis or decreased glucose availability act as

a dictator of the severity of the ketotic state, b) does fat

accumulation in the liver contribute to liver disfunction and the

onset of ketosis, c) is there a specific biochemical breakdown in the

gluconeogenic pathway that brings about clinical ketosis, d) is the

decrease in gluconeogenic capacity of liver a direct cause of the

signs of clinical ketosis, or Just a response to ketosis, and e) is gluconeogenesis truly inhibited by ketosis in vivo, as hinted by previous studies?

The present study verifies the data of Mills et al. (2) and provides à strong foundation for discussion of the metabolic changes occurring during ketosis. However, measurements are not specific enough to provide strong conclusions of cause-and-effect mechanisms.

Future work now must concentrate on these mechanisms; work which admittedly is much more difficult. Three approaches seem apparent.

First, hormone concentrations must be characterized for all studies.

The fact that hormones are powerful regulators of metabolism is accepted, and although some data are available characterizing key hormone patterns early in lactation, we still do not know whether the 133

onset of clinical ketosls is mediated through changes in hormone

release. Second, changes in specific gluconeogenic reactions in liver

must be explored. Such measurements could include enzyme activities,

selected concentrations of TCA cycle intermediates in the cytosol and

mitochondria, and intracelluar pH. Third, future studies must manipulate the metabolism of cows with specific perturbations during the development of ketosls. This approach is needed to identify specific cause and effect mechanisms. One such perturbation could be the injections of phlorizin into subclinically-ketotic cows, with subsequent measurements of specific changes in metabolism on the whole-body and intracellular levels.

The relative success of glucose infusions into the intestine of

GT cows also stimulates many questions for future research. First and foremost, because there was some problem with hyperglycemia and nervousness in GT cows, verification of the glucose effect on prevention of ketosls is needed. Second, questions concerning method of introduction, length of treatment, and mechanism of action are primary.

The protocol of continuous infusion of glucose into the duodenum obviously is not acceptable for on-farm use. Therefore, one question is whether rumen-protection technology could be used to Include glucose or glucose precursors in the feed so they would by-pass the rumen and be absorbed directly from the small Intestine? Also, the minimum amount of glucose required to elicit responses seen with the 134

GT cows has not been researched. Further, glucose Infusions in GT cows were terminated an average of 1 wfc before ketosis was detected in

KIP cows, yet no symptoms of ketosis were observed. A question for future research is what length of treatment is needed during development of ketosis to prevent ketosis from occurring.

The mechanism of action of the glucose effect on ketosis is unclear because all symptoms were alleviated simultaneously. Some specific questions of Interest are: a) does glucose absorbed from the intestine elicit a hormonal response that could alleviate ketosis, b) what specific effect does glucose have on hepatic ketogenic and gluconeogenic reactions, c) is there a rate-limiting step that is affected by glucose, causing other symptoms to be alleviated, and d) are compounds other than glucose capable of eliciting similar responses. Future work with agents that could prevent ketosis should be useful because of both the potential for benefits to dairy producers and more detailed elucidation of the etiology of ketosis. 135

REFERENCES

1. Mills. S. E. 1982. Metabolic charaterlstlcs of the ketotlc state In the bovine. Ph.D. Dissertation. ISU. University Microfilms, Ann Arbor, MI. Order No. 82-21,209.

2. Mills, S. E., D. C. Beltz, and J. W. Young. 1986. Characterization of metabolic changes during a protocol for inducing lactation ketosis in dairy cows. J. Dairy Scl. 69:352.

3. Mills, S. E., D. C. Beltz, and J. W. Young. 1986. Evidence for impaired metabolism in liver during induced lactation ketosis of dairy cows. J. Dairy Scl. 69:362.

4. Schultz, L. H. 1974. Ketosis. pages 217-352 ^ B. L. Larson and V. R. Smith, eds. Lactation: A comprehensive treatice. Vol II. Academic Press, Inc., New York. 136

ACKNOWLEDGEMENTS

The completion of this Ph.D. degree culminates six years that I

have spent in Ames, lA, and this Journey has taken me far beyond my

own expectations, both personally and scientifically. To acknowledge

the many who have contributed to this Journey would be impossible.

Therefore, only a few will be mentioned.

I thank Dr. Jerry Young for six years of guidance and trust, and

for a working relationship that transcended into friendship. The

patience and understanding shown during some very difficult times have

not gone unappreciated. I hope I will be able to remember some of the vast grammatical lessons you have taught me.

I acknowledge the contributions of my committee members, Drs.

Beitz, Nissen, Berger, Zimmerman, and Cornette. I especially thank

Dr. Jim Cornette for being a living example that physiology does have a place in the field of mathematics.

For help in preparation of this dissertation, I thank Pam Davis for her skills as an artist, and Susan Prey for sharing computer time with me. I also acknowledge Dr. J. P. Atanasoff for inventing the computer, without which all this still would be in a mess on my desk.

The interaction with my fellow graduate students has been vital to my growth as a scientist and as a person. The dedication to excellence shown by students in the Nutritional Physiology Group has reinforced that in myself. I have enjoyed many stimulating 137

discussions, several "release" times, and particularly thank Lynnette

and Donna for their invested friendship.

As always, I thank my family for their never-ending support,

especially my brother who always seems to be there when needed.

Finally, I thank God and the faith community of St. Thomas Aquinas of

Ames for their challenge to keep a proper perspective of life and to continually search for the truth. 138

APPENDIX 139

TABLE Al. Acetate metabolized to triglycerides In liver slices of a control cow Incubated with different concentrations of glucose nH glucose In Incubation acetate metabolized to media triglycerides in liver

ymoles acetate/(h x g liver)

0 +.035

1.0 +.026

2.5 +.010

5.0 +.033 TABLE â2. Concentrations of acetate in plasma of control, ketosis-induced (KIP), and glucose-treated (GT) cows^

Stage of experiment^ (days postpartum)

PP EPP PI EI MI LI K R SE Treatment] (-10) (5) (13) (21) (28) (35) (44) (60)

pmol acetate/1

Control 531 742 962 745 780 861 802» 969 104 KIP 835 935 1060 1053 3669 2584 3397% 1051 741 GT 641 663 661 685 604 642 126iab 626 144

^Treatment means with different superscripts within a column are significantly different, P<.05.

^Abbreviations for stage of experiment are explained in Section I of dissertation.

%eans for KIP cows show a trend to be hi^er than control and GT at HI and LI, but large variation within KIP cows resulted in differences being not significant. TABLE A3. Concentrations of acetate in plasma of control, ketosis-Induced (KIP), and glucose-treated (GT) cows^ (omitting one cow)

Stage of experiment^ (days postpartum)

PP EPP PI EI MI LI K R SE Treatment^ (-10) (5) (13) (21) (28) (35) (44) (60)

Vimol acetate/1

Control 531 742* 952 745 780* 861* 802 969 104 KIP 882 1025^ 1106 926 1173% 1509% 1836 1078 198 GT 641 663* 661 685 604* 642* 1261 626 144

^Treatment means with different superscripts within a column are significantly different, P<.05.

^Abbreviations for stage of experiment are explained in Section I of dissertation.

%eans for KIP cows show a trend (P<.12) to be higher than control and GT at K, but large variation within KIP cows resulted in differences being not significant.