UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Insulinotropic effects of certain blood metabolites in domestic sheep Ross, James Pelter 1973

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata

Download

Media
831-UBC_1973_A1 R68_6.pdf [ 6.38MB ]
Metadata
JSON: 831-1.0100971.json
JSON-LD: 831-1.0100971-ld.json
RDF/XML (Pretty): 831-1.0100971-rdf.xml
RDF/JSON: 831-1.0100971-rdf.json
Turtle: 831-1.0100971-turtle.txt
N-Triples: 831-1.0100971-rdf-ntriples.txt
Original Record: 831-1.0100971-source.json
Full Text
831-1.0100971-fulltext.txt
Citation
831-1.0100971.ris

Full Text

INSULINOTROPIC EFFECTS OF CERTAIN BLOOD METABOLITES IN DOMESTIC SHEEP by James Pelter Ross B.S.A., M.S.A., University of Bri t i s h Columbia, 1967 A THESIS SUBMITTED IN PARTIAL FULFI__NT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in the Department of. Animal Science We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA May 1973 In p r e s e n t i n g t h i s t h e s i s i n p a r t i a l f u l f i l m e n t o f the requirements f o r an advanced degree at the U n i v e r s i t y o f B r i t i s h Columbia, I agree t h a t the L i b r a r y s h a l l make i t f r e e l y a v a i l a b l e f o r r e f e r e n c e and study. I f u r t h e r agree t h a t p e r m i s s i o n f o r e x t e n s i v e copying of t h i s t h e s i s f o r s c h o l a r l y purposes may be granted by the Head of my Department or by h i s r e p r e s e n t a t i v e s . I t i s understood t h a t copying o r p u b l i c a t i o n o f t h i s t h e s i s f o r f i n a n c i a l g a i n s h a l l not be allowed without my w r i t t e n p e r m i s s i o n . JAMES PELTER ROSS The Department o f Animal Science The U n i v e r s i t y o f B r i t i s h Columbia, Vancouver 8, Canada. Date II Abstract Blood propionate, butyrate, glucose, acetoacetate, and 8-hydroxybutyrate were studied to determine i f these metabolites have a role in the release of insulin from the pancreatic i s l e t s i n domestic sheep. Since methods previously available for the measurement of plasma propionate and butyrate levels had a number of disadvantages, an improved method for the determination of blood v o l a t i l e fatty acids was devised. A study was conducted to determine i f blood v o l a t i l e fatty acids and glucose are related to plasma insulin in sheep consuming different feeds. Postprandial changes i n the concentrations of plasma v o l a t i l e fatty acids, glucose and insulin were measured at 2 hour intervals i n mature wethers fed barley, hay, or an equal mixture of barley and hay. The plasma constituents which shewed an increase after feeding were ins u l i n , glucose, and propionate for sheep fed barley; glucose, propionate, acetate, and isobutyrate for sheep fed hay; and glucose, acetate, propionate, butyrate and insulin for sheep fed barley and hay. Simple and multiple regression analyses were carried out relating plasma glucose and v o l a t i l e fatty acids to plasma insulin within each dietary regime. These results indicated that butyrate and isobutyrate were the most important blood v o l a t i l e fatty acids with respect to their relationship to plasma insulin levels i n this study. A study of diurnal variation in plasma insulin, v o l a t i l e fatty acids, and glucose levels i n lactating ewes revealed a similar pattern of diurnal variation for glucose, acetate and propionate. Whereas these metabolites had a single maximum at 5 pm. and a minimum at 8 am, the plasma butyrate le v e l appeared to reach a maximum at about 11 am as well as at 5 pm. The plasma insulin values appeared somewhat unreliable as indicated by the ..large Vii variation in the data. The pattern of diurnal variation in these blood metabolites i s discussed i n relation to the regulation of insulin secretion. The relationship between plasma glucose and insulin levels i n ad libitum fed and i n prefasted-refed suckling lambs was determined. The degree of correlation varied considerably depending upon the experimental condition. The relevance of a s t a t i s t i c a l l y significant relationship between plasma insulin and a blood metabolite is discussed in relation to the regulatory role of plasma metabolites i n the secretion of insulin. By experimentally altering the individual plasma propionate, butyrate and glucose levels within the estimated physiological range and measuring the corresponding plasma insulin levels, i t was anticipated that the relative importance of each of these metabolites i n the regulation- of insulin secretion could be assessed. A phlorizin infusion to fed wethers caused a significant decrease i n plasma glucose and insulin levels. Infusing glucose so as to elevate plasma glucose levels gradually to about 100 mg/100 ml f a i l e d to e l i c i t an insulin secretory response. Infusion of propionate and butyrate at rates sufficient to increase plasma levels considerably above the upper limit of the physiological range f a i l e d to e l i c i t an insulin secretory response. Only when plasma butyrate levels were elevated above 175 ymole/liter was an insulin secretory response obtained. These results suggest that the increase in plasma insul i n level that normally occurs in sheep following the ingestion of concentrate-type feeds i s not mediated by butyrate, propionate or glucose. However, the normal level of plasma insulin in fed sheep, at least in part, appears to be maintained by the plasma glucose lev e l . 6-hydroxybutyrate and acetoacetate were studied to determine i f physiological increases in the blood level of these metabolites would affect the level of plasma insulin. Only acetoacetate had an insulinotropic action when injected to fasted non-pregnant ewes. The injected acetoacetate appeared to be rapidly reduced within the blood to 3-hydroxybutyrate. The results are discussed in relation to an etiologic role of acetoacetate i n rurninant ketosis. TABLE OF CONTEOTS page Review of Literature . 1 Experiment I 24 Experimental procedure 25 Results and discussion 29 Experiment II 4 4 Part A. Experimental procedure 44 Results 47 D i s c u s s i o n 53 P a r t B. 1. L a c t a t i n g ewes Experimental procedure 5 7 Results and discussion 57 2. Lambs Experimental procedure 60 Results and discussion 60 Experiment III 64 Experimental procedure 64 Results 67 Discussion 79 General discussion 93 Experiment IV 100 Experimental procedure 100 Results 101 Discussion 101 Bibliography 114 LIST OF TABLES Table page 1. Drops of 85% orthophosphoric acid 30 2. Composition of 12 v o l a t i l e fatty acid standard mixtures analyzed by the method described 31 3. Recovery of v o l a t i l e fatty acids from blood plasma when added i n quantities within the physiological range 39 4. The composition and proximate analysis of the feeds used i n the experiments reported 4-5 5. The R squared expressed as a percentage and significance associated with simple and multiple linear regressions of plasma glucose, acetate, propionate, iscbutyrate and butyrate levels regressed on plasma insul i n levels for the three feeds 52 6. The R squared expressed as a per cent and significance associated with simple and multiple linear regressions of plasma glucose on plasma insulin levels i n suckling lambs = 62 7; Data on phlorizin infusion to non-pregnant ewes 65 8. Data on glucose infusions to non-pregnant fasted 66 ewes 9 . Data on VFA infusions to non-pregnant fasted ewes 67 10. Plasma VFA, glucose and insulin levels i n fed and 48hr fasted non-pregnant ewes 79 11. Comparison of ruminal, portal and peripheral propionate and butyrate entry rates with infusion rates employed in insulin secretion studies 88 LIST OF FIGURES Figure page 1. Arrangement of apparatus for microcondenser described in experimental procedure. 27 2. Graph showing relation between determined and standard acetic acid concentrations with regression line and confidence intervals explained i n text. 34 3. Graph showing relation between determined and standard propionic acid concentrations with regression line and confidence intervals explained in text. 35 4. Graph showing relation between determined and standard butyric acid concentrations with regression l i n e and confidence intervals explained i n text. 36 5. A. Chromatogram of sample from 10 ml. of sheep blood plasma to which 10 yEq of isovaleric acid had been added. 38 B. Chromatogram of sample of 10 ml of sheep blood" to which 1 ml of standard B had been added. 38 6. Graph shewing changes i n blood v o l a t i l e fatty acids of a single fed sheep over a fasted control sheep following feeding of chopped a l f a l f a hay. 41 7. Jugular plasma acetate, propionate, isobutyrate, butyrate, glucose and insulin concentrations measured postprandial in wethers consuming feed H. 48 8. Jugular plasma acetate, propionate, isobutyrate, butyrate, glucose and insulin concentrations measured postprandial i n wethers consuming feed B. 49 9. Jugular plasma acetate, propionate, isobutyrate, butyrate, glucose and insulin concentrations measured postprandial i n wethers consuming feed BH. 50 10. Diurnal variation in jugular plasma insulin, glucose and VFA i n lactating ewes 58 11. Jugular plasma glucose and insulin levels in suckling lambs. 61 12. Changes i n plasma glucose, insulin and VFA following a phlorizin injection or infusion to wethers i n a preliminary study. 68 13. Changes in jugular plasma glucose, insulin, FFA and VFA levels during the infusion of phlorizin to fed wethers. 14. Jugular plasma glucose, insulin and FFA levels during the infusion of glucose to non-lactating ewes at the lower infusion rate. 71 15. Jugular plasma glucose and insulin levels during the infusion of glucose to non-lactating ewes at the higher infusion rate. 72 16. Jugular plasma insul i n , glucose and propionate levels associated with the infusion of propionate to non-lactating ewes. 74 17. Jugular plasma insulin, glucose and butyrate levels associated with the lowest rate of butyrate infusion to non-lactating ewes. 75 18. Jugular plasma insulin, glucose and butyrate levels associated with the intermediate rate of butyrate infusion to ncn-lactating ewes. 77 19. Jugular plasma insulin, glucose and butyrate levels associated with the highest rate of butyrate infusion to non-lactating ewes. 78 20. Comparison of plasma ijisulin and butyrate levels for the different rates of butyrate infusion. 90 21. Changes i n jugular plasma insulin, glucose, FFA and 3-hydroxybutyrate following the intravenous injection of 3-hydroxybutyrate. 102 22. Jugular plasma acetone, acetoacetate and 3-hydroxybutyrate levels following the injection of 3-hydroxybutyrate. 103 23. Changes i n jugular plasma insulin, glucose, FFA and acetoacetate levels following the intravenous injection of acetoacetate. 105 24. Jugular plasma acetone, acetoacetate and 3-hydroxybutyrate levels following the acetoacetate injection. 106 25. Comparison of changes i n to t a l and individual ketone body levels i n plasma following g-hydroxybutyrate and acetoacetate injections. 109 1 REVIEW OF LITERATAJRE Introduction Value of ruminant digestion to man Among domesticated mammals the ruminants are unique in that they possess a stomach of four parts. The f i r s t three parts arise morphologically from the esophagus and i n the adult animal provide an environment suitable for the establishment of symbiotic microbial populations such as bacteria of the genera Ruminococcus and Butyrivibrio, and protozoa of the genera Dasytricha and Isotricha (Howard, 1967). The immediate advantage'to the host animal i s that dietary components such as cellulose and hemicellulose which cannot be degraded by mammalian digestive enzymes are degraded suf f i c i e n t l y by microbial enzymes so as to provide a useful source of nutrients. Microbial a c t i v i t y within the rumen and reticulum also modify dietary fats and protein as well as synthesizing essential dietary nutrients such as the B-complex vitamins (Annison and Lewis, 1959). These actions by microbes within the ruminoreticulum have made ruminants particularly valuable to man as a means of readily converting fibrous materials, largely indigestible i n man, into a form suitable for human food. Forestomach fermentation and glucose metabolism A l l mammals have an obligatory metabolic requirement for D-glucose and this i s generally met by the absorption of glucose from the small intestine (Allen, 1970). However, ruminant animals appear to have evolved a potential metabolic l i a b i l i t y with respect to the provision of metabolic glucose. Most of the soluble carbohydrate i n the diet that enters the 2 rwiinoreticulum i s degraded by the microbial enzymes to short chain fatty acids mainly of 2, 3 and 4 carbons i n length (Annison et a l . , 1959). Accordingly very l i t t l e soluble carbohydrate normally reaches the small intestine (Allen, 1970). The glucose required by the ruminant animal must therefore be provided endogenously by gluconeogenesis (Lindsay, 1970). In fe r a l ruminants the potential l i a b i l i t y of fermentative forestomach digestion i s probably of l i t t l e consequence since ruminants have been demonstrated to have a high gluconeogenic capacity (Lindsay, 1970). However, domestic ruminants may be incapable of meeting glucose requirements by endogenous means under conditions of very high productive performance, such as during early lactation in the cow and during late pregnancy i n the multiparous ewe (Kronfeld, 1970). The metabolic disorders associated with these apparent cxurbohydrate deficiencies are usually referred to as acetonemia and pregnancy toxemia respectively. Ruminants versus monogastric animals Many studies have been directed at improving the understanding of digestion and metabolism within the domestic rnjminaht,and many features of these processes appear to be common to a l l ruminants and different from similar processes encountered within man and other domestic mammals. In consequence, man and domestic mammals other than ruminants are frequently grouped together under the term monogastric or non-ruminant animals. Many comparisons between ruminant and non-ruminant digestion and metabolism have been made as a means of clari f y i n g the overall functional significance of various experimental findings. Numerous examples w i l l be found i n the following discussion. 3 An unqualified reference to ruminants in the discussion that follows refers only to the mature ruminant. This distinction i s necessary since the digestion and metabolism of the immature ruminant resembles more closely that of monogastric animals than i t does the mature ruminant. Since the av a i l a b i l i t y of glucose to the ruminant from the digestive tract i s normally very limited, glucose metabolism i n ruminants has been extensively studied to determine: hew glucose requirements are met, how important glucose is i n the metabolism of ruminants as compared to non-ruminants, and how changes occur i n carbohydrate metabolism during metabolic disorders i n ruminants. The pancreatic endocrine hormones insulin and glucagon are known to be intimately involved i n the regulation of carbohydrate metabolism i n monogastric animals, and there have been numerous studies i n recent years for the purpose of clari f y i n g the importance of these endocrines i n ruminant carbohydrate metabolism. Carbohydrate Metabolism and Ruminant Development The immature ruminant undergoes a transition from monogastric-like digestion to fermentative ruminant digestion i n the f i r s t few months after b i r t h . Accompanying changes i n digestion there are gradual alterations i n parameters of carbohydrate metabolism which have been the subject of numerous studies. As the young ruminant's diet changes from milk to solid feed there i s a gradual f a l l i n the level of plasma glucose from about 90-100 mg/100 ml to about 40-60 mg/100 ml (McCandless and Dye, 1950). Exactly how this change i n plasma glucose level i s mediated has not been firmly resolved. In some work i t has been suggested that this change i s at least partly the result of the loss of an exogenous glucose source 4 < (Ballard, Hanson and Kronfield, 1969). However, in studies with calves i t was shown that the decline in blood glucose occurred independent of rumen development, dietary changes or increases i n blood or rumen vola t i l e fatty acids (VTA) (Leat, 1970). The tolerance to intravenous glucose has also been demonstrated to decrease during the transition phase (McCandless et a l . , 1950), as well as the sensitivity to intravenous insu l i n (Jarrett and Potter, 1953). The decreased glucose tolerance has been shown to be due i n part to a decreased insulin secretory response to a standard intravenous glucose dose (Manns and Boda, 1967). The reduced tolerance to glucose is accompanied by a reduction in the size of the glucose pool and the rate of glucose u t i l i z a t i o n (Jarrett, Jones and Potter, 1964). Conflicting results have been reported with regard to changes in plasma insulin levels i n the developing ruminant (Leat, 1970; Manns et a l . , 1967). Considerable differences have been found i n hepatic metabolism between the immature and mature ruminant. Hepatic enzyme differences suggest a s h i f t from glucose to fat as the major metabolic fuel as the ruminant develops, and there i s a corresponding s h i f t i n hepatic metabolism from active hepatic glycolysis to gluconeogenesis (Leat, 1970). Although the endocrine pancreas has been demonstrated to be capable of modifying hepatic metabolism i n the direction observed during development, direct evidence has not as yet been obtained i n this regard. The end-products of fermentation i n the rumen may be important i n mediating the metabolic changes i n the l i v e r that occur during development. The coenzyme A derivatives of acetic, propionic and butyric acids have been demonstrated to stimulate gluconeogenesis by activating pyruvate carboxylase 5 (EC6.4.1.1) and an increased acetyl-CoA:CoA r a t i o would tend to in h i b i t some of the glycolytic enzymes (Leat, 1970). Other changes noted i n enzyme a c t i v i t i e s related to fat metabolism suggest that although the developing ruminant has a decreasing a b i l i t y to u t i l i z e glucose for fat synthesis there i s an increased overall a b i l i t y for fat synthesis, particularly from acetate (Ballard et a l . , 1969). Glucose Metabolism and the Endocrine Pancreas The endocrine pancreas secretes the hormones insulin and glucagon which appear to play a major role i n the regulation of metabolic processes involving glucose both for ruminant and monogastric animals (Dickson, 1970). The anatomical relation of the pancreas to the l i v e r suggests that the primary target organ for these hormones i s the l i v e r . Accordingly insulin has been shown to f a c i l i t a t e the oxidation of glucose by the isolated l i v e r , to reduce hepatic glucose output and hepatic gluconeogenesis, to enhance hepatic glucose uptake i n part by inducing glucokinase (EC 2.7.1.2.), and to enhance hepatic glycogen synthesis by inducing hepatic glycogen synthetase (EC 2.4.1.11) (Reiser, 1967). The changes caused by insulin i n hepatic glucose oxidation and gluconeogenesis are probably the result of a direct effect of insulin on hepatic enzymes since insulin has been demonstrated to induce phosphofructokinase (EC 2.7.1.11) and pyruvate kinase (EC 2.7.1.40) and to suppress glucose - 6 - phosphatase (EC 3.1.3.9), fructose -1, 6-diphosphatase (EC 3.1.3.11), phosphoenolpyruvate carboxykinase (EC4.1.1.32) and pyruvate carboxylase (EC 6.4.1.1) (Reiser, 1967). The effects of insulin on the l i v e r may not be as important i n r u T n i n a n t s as i n non-ruminants (Kronfeld, 1970). Kronfeld has pointed out that since 6 ruminants lack the high Kin glucokinase (EC 2.7.1.2), insulin may be less effective in altering hepatic glucose metabolism in rumxianti. In accordance with this suggestion the l i v e r of ruminants i s always producing glucose whereas the livers of non-ruminants may produce or take up glucose depending on the a v a i l a b i l i t y of exogenous glucose (Katz and Bergman, 1969). Although hepatic gluconeogenesis increases i n the monograstric animal as the a v a i l a b i l i t y of glucose from the gut is reduced, the exact means by which this metabolic adjustment i s brought about has not been resolved. It has been proposed that i n fasting monogastric animals this change i s the result of either increased a v a i l a b i l i t y of free fatty acids (FFA) to the l i v e r or a decreased level of insulin i n blood reaching the l i v e r (Williamson, Kreisberg and Felts, 1966; Vance, Buchanan and Williams, 1968). SdxLmmel and Knobil (1970) have demonstrated that i n rats under certain conditions of fasting, there i s an increased gluconeogenesis i n the absence of an increase i n plasma FFA or' a decrease i n plasma insulin, suggesting that some other factor must be responsible for enhanced gluconeogenesis during fasting. However, in studies on the effects of starvation on l i v e r enzymes in rats i t was found that some of the largest decreases i n enzyme act i v i t y following fasting occur with the NADP-linked dehydrogenases of the cytosol (Szepesi and Berdanick, 1971). Realimentation with high carbohydrate diets suggested that an increase in insulin secretibn. acts to induce glucose-6-phosphate dehydrogenase (EC 1.1.1.49) and malic enzyme (EC 1.1.1.38) and that the hepatic activity of these enzymes may be dependent upon insulin. Glucagon acts on the l i v e r to stimulate the enzyme system responsible for glycogen breakdown (Dickson, 1970). Glucagon is possibly the primary 7 hormone with respect to the prevention of hypoglycemia (linger and Eisentraut, 1970). Glucagon has also been demonstrated to stimulate gluconeogenesis from amino acids, pyruvate and lactate in the isolated perfused rat l i v e r (Garcia, Williamson and C a h i l l , 1966). A ketogenic action of glucagon has been attributed to activation of hepatic lipase (Bewsher and Ashmore, 1966) and enhanced u t i l i z a t i o n of FFA by the liver, has been demonstrated to be another action of glucagon (Gorman, Salter and Penhos, 1967). Apart from insulin actions on the l i v e r certain tissues are dependent upon insulin for the uptake of glucose. Muscle tissue i s insulin-dependent (Narahara and Cori, 1968). The primary event of insulin action on this tissue i s an increased rate of membrane transport of glucose (Reiser, 1967). The insul i n dependency of certain large masses of body tissue such as muscle tissue i s important since under conditions of reduced insulin a v a i l a b i l i t y or insul i n antagonism relatively much less glucose i s u t i l i z e d i n the metabolism of these tissues, thus sparing glucose for non-insulin dependent tissues such as the neural tissues of the brain which are to a large extent obligatory-glucose oxidizing (Cahill and Owen, 1968). Adipose tissue i s also insulin dependent. Insulin acts on adipose tissue to enhance glucose uptake, to stimulate the oxidation of glucose via the pentose phosphate pathway, and to inhibit l i p o l y s i s and FFA release by enhancing the formation of «-glycerophosphate necessary for esterifying FFA (McLean, Brown and Greeribaum, 1968). The inhibition of the l i p o l y t i c process may also involve a specific inhibition of an intracellular lipase by insulin (Kipnis, 1972). 8 Most of the effects which insulin and glucagon have in monogastric animals appear to be qualitatively the same for ruminants. However, since the relative amounts and types of digestive end-products absorbed from the digestive tract are quite different for monogastric animals than for ruminants considerable differences i n pancreatic endocrine secretion as well as i n hepatic and extrahepatic metabolism would be expected. Regulation of Pancreatic Endocrine Secretion Theoretical considerations Basically the mammalian body tissues u t i l i z e fat and carbohydrate in varying proportions as the body fu e l . However, as indicated neural tissues have an obligatory glucose requirement, and almost exclusively oxidize glucose. As a result the a v a i l a b i l i t y of exogenous glucose appears to be the most important single determinant of the type of hepatic and extrahepatic metabolism for both ruminants and monogastric animals. The adjustments i n metabolism appropriate to the a v a i l a b i l i t y of exogenous glucose appear to be largely mediated by the secretions of the endocrine pancreas. In monogastric animals i n which meals tend to be consumed and digested over rel a t i v e l y short time intervals there are periods of rapid nutrient influx from the gut as well as periods v i r t u a l l y devoid of nutrient influx. Insulin and glucagon are secreted i n proportions that allow for the maintenance of homeostatic conditions as well as for adjustments i n metabolism that optimize absorbed nutrient u t i l i z a t i o n . During the period of nutrient absorption i n non-ruminants there i s an increase i n the secretion of insulin 9 and possibly a decrease i n the secretion of glucagon resulting in an overall s h i f t i n metabolism from one that i s primarily catabolic and stored nutrient u t i l i z i n g to a metabolism that i s largely anabolic and absorbed nutrient u t i l i z i n g . How the endocrine pancreas i s affected by feeding i n rnjminants i s not clear. However, ruminants tend to digest meals over long periods of time and digestive end-products may be absorbed at a low but steady rate i n comparison to monogastric animals. In this regard the metabolism i n the l i v e r of ruminants would appear to proceed i n a hormonal milieu that has a low level of insulin since the hepatic oxidation and storage of glucose (i.e. a c t i v i t i e s enhanced by insulin) are negligible i n the ruminant (Katz et a l . , 1969). The scheme pertaining to glucose regulation of insulin secretion i n non-ruminants i s visualized as follows, A meal i n non-ruminants usually results i n an increase i n plasma glucose which stimulates pancreatic insulin output. The insulin secreted acts on the l i v e r to cause an overall net hepatic uptake of glucose and enhanced glyoogenesis while depressing hepatic glycogenolysis and gluconeogenesis , and at the same time stimulating glucose uptake, glycogenesis, and lipogenesis i n peripheral tissues. During the post-absorptive phase insulin secretion i s decreased and net hepatic glucose balance becomes negative in response to declining plasma glucose and insulin levels, while the peripheral glucose uptake decreases. However, this scheme may be much too simple i n that a large number of other factors besides glucose appear to influence insulin secretion. Also glucagon should be considered in any regulatory scheme for insulin secretion since" these hormones appear to affect similar metabolic processes 10 and are both secreted by c e l l s of the i s l e t s of Langerhan. A diverse group of compounds have been demonstrated to be capable of influencing, insulin secretion. Apart from glucose, amino acids (Fajans et a l . , 1967), ketone bodies (Madison, Mebane and Lochner, 1963), short (Horino et a l . , 1968), medium (Greenberger, Tzagournis and Graves, 1968) and long chain fatty acids (Crespin, Greenough and Steinberg, 1969), a diversity of neurohormones (Coore and Randle, 1964), steroids (Bassett, Wallace and Sydney, 1967) and peptides (linger et a l . , 1967) may also normally stimulate or in h i b i t insulin release and modify the insulin secretory response to glucose. These factors w i l l be discussed individually below. Although many compounds can be demonstrated to have insulinotropic properties under experimental conditions, the physiological significance of such findings i s often d i f f i c u l t to assess. The c r i t e r i a for physiological significance of insulinotropic compounds should include: that the compound normally occurs within the animal body, that the compound has an insulinotropic action within i t s normal plasma concentration range, and that the rate of increase i n concentration of the compound in plasma reaching the pancreatic i s l e t s be physiological when demonstrating the insulinotropic action of a compound. Seme workers have speculated that the primary regulatory metabolite(s) for endocrine pancreas secretion has evolved i n relation to the major energy supplying nutrients. (Kipnis, 1972; Horino et a l . , 1968). In monogastric animals this nutrient i s normally glucose. However VFA represent the major energy source for the ruminant and according to this theory these acids would provoke insulin secretion, and thus aid i n the conversion of VFA into two storage forms,namely glycogen and fat. However, on the basis of current 11 evidence this action of VFA cannot be considered established. Monogastric animals Glucose - Apparently glucose i s one of the most important insulinotropic compounds. In monogastric animals the plasma insulin level increases rapidly when the blood glucose concentration is elevated above the normal post-absorptive range (i.e. 60-100 mg/100 ml) by oral or intravenous glucose (Randle, Ashcroft and G i l l , 1968). Blood glucose levels have also been demonstrated to modify the release of glucagon. In dogs acute hypoglycemia induced by phlorizin led to an increase in plasma glucagon level (Randle et a l . , 1968). Glucose refeeding and hyperglycemia caused a f a l l i n plasma glucagon lev e l . In studies of post-prandial plasma glucose and insulin levels there i s a significant positive correlation between plasma glucose and insulin levels (Steffens, 1970), which further suggests that the influence of glucose on insulin release i s physiologically important. The unique role of glucose with respect to mammalian p - c e l l metabolism i s suggested by the fact that glucose is the only circulating blood metabolite which under physiological conditions i s known to stimulate insulin synthesis as well as secretion (Kipnis,' 1972). Lipids - The oral ingestion of fat or the acute elevation of plasma FFA do not e l i c i t any significant insulin secretory response in man. However the medium chain length fatty acids have been reported to e l i c i t a modest insulin secretory response particularly when given with an oral glucose load (Kipnis, 1972). Greenough, Crespin and Steinberg (1967) have demonstrated that experimentally raising plasma FFA levels by a rapid infusion of long 12 chain fatty acids into unanesthetized dogs increased plasma immune-reactive insulin (IRI) and decreased plasma blood glucose. Sanbar et a l . (1965) demonstrated that infusing octanoic acid into dogs induced a decline in blood glucose lev e l , and later demonstrated that this was associated with increased insulin secretion (Sanbar et a l . , 1967). The short.chain fatty acids propionic and butyric acid have been demonstrated to be without effect on the release pf insulin i n monogastric animals (Horino et a l . , 1968). Ketone bodies - The infusion of S-hydroxybutyric acid or acetoacetic acid into the a r t e r i a l blood supply of the pancreas of dogs caused a 200% r i s e in the concentration of insulin i n the pancreatic venous blood (Madison et a l . , 1964). These results indicated that ketone bodies have a direct stimulatory effect on the g-cells. Changes in hepatic glucose metabolism . following intravenous ketone body administration also suggests that ketone bodies stimulate insulin secretion (Mebane and Madison, 1964). Although ketone body infusions to dogs provoke prompt insulin secretory responses, the infusion of acetoacetate or . 6-hydroxybutyrate into normal human adults, in amounts sufficient to raise post-absorptive plasma ketone body levels by 10 to 50 fold for periods of 20 to 30 minutes^failed to e l i c i t a significant insulin secretory response as measured by plasma TJRI (Kipnis, 1972). However, other workers have reported that under suitable experimental conditions a stimulatory effect of ketone bodies on insulin secretion i n man could be detected for both acetoacetate and S-hydroxybutyrate (Balasse, Conns and Lambilliotte, 1970). It was found i n this study that abrupt elevations of the plasma ketone levels were more effective in stimulating insulin secretion than were gradual increases. Studies in v i t r o suggested that ketone bodies do not stimulate insulin secretion, but tend to inhibit this process 13 (Grcdsky and Forsham, 1966). Gastrointestinal hormones - A large body of conflicting evidence has been produced on the influence of gastrointestinal hormones on the secretion of insulin. The greater insulin secretory response provoked by o r a l l y administered compared to intravenously administered glucose may be caused by an intestinal hormone enhancing glucose-stimulated insulin release (Turner and Jarrett, 1970). This was suggested since certain of the enteric hormones have been reported to have insulinotropic properties. Chisholm et a l . (1971) have reported that plasma secretin levels were • significantly increased following oral but not intravenous glucose administration to human adults. Since the plasma secretin levels attained following oral glucose have been shown to cause an insulin secretory response, i t was concluded that secretin plays a major role i n the normal insulin secretory response by exerting direct releasing and glucose potentiating effects. The augmentation of the insulin secretory response by a gut factor(s) may require an elevated plasma glucose or amino acid level (Malaisse and Malaisse-Lagae, 1970). Similarly a glucagon-like factor isolated from the gut as well as a duodenal-jejunal extract appear to be quite potent stimuli to insulin release, but i n both cases this effect appears to be dependent upon an elevated blood glucose concentration (Turner et a l . , 1970). In this study the intestinal hormones secretin and pancreozyinin did not appear to exert any effect on glucose-induced insulin release. Consequently these authors suggested that either enteroglucagon or another uncharacterized hormone are both responsible for the enhanced insulin secretory response to oral as opposed to intravenous glucose. 14 However Blum and Linscheer (1971) obtained evidence that secretin is the hormone which augments the insulin secretory response to glucose absorbed from the small intestine. Similarly Dupre et al.(1967) have shown that secretin causes insulin release in man. This effect was found to be exaggerated during hyperglycemia and was associated with the accelerated disposal of glucose. A stimulation of insulin release resulted from the infusion of hydrochloric acid in physiologic amounts into the upper gastrointestinal tract. Gastrin stimulation of insulin secretion may be effected by this route. Pancreozymin was also found to stimulate insulin release and unlike secretin provoked a maintained enhancement of insulin secretion during intravenous infusion of arginine as well as causing an enhanced r i s e i n plasma glucagon-like immunoreactivi-Ly that occurred after the rapid intravenous injection of arginine. These results generally support the suggestion that the effects of digestive secretagogues on insulin secretion are physiologically important. Yet i n studies with dogs (Kahil, Mcllhaney and Jordan, 1970), the stimulation of endogenous secretin release by duodenal acid i n s t i l l a t i o n s resulted in a very transient r i s e in portal venous blood insulin levels. This occurred regardless of continued secretin production. Similarly Boynes, Jarret and Keen (1966) have shown that the induction of endogenous secretin release by the duodenal i n s t i l l a t i o n of hydrochloric acid i n human adults did not affect the blood glucose or plasma insulin levels. Also when glucose plus acid and glucose alone were i n s t i l l e d in the duodenum, almost identical changes in the blood glucose and insulin levels were observed. In a further study,the intraduodenal infusion of glucose i n man has been shown to have no effect on external pancreatic secretion (Sum and Preshaw, 15 1967). This suggests that, i t i s unlikely that either secretin or pancreozymin is involved in the insulin response to the introduction of glucose into the small intestine. Thus the role of enteric factors i n the regulation of insulin secretion remains uncertain although i t appears from evidence available that hyperglycemia may be a prerequisite for this insulinotropic action. Also, insulin secretion in response to oral glucose appears to be largely the result of the insulinotropic action of glucose. Pancreozymin, but not secretin or gastrin, were demonstrated to stimulate glucagon secretion when administered intravenously (linger et a l . , 1966). It was suggested that pancreozymin may act physiologically to enhance the glucagon stimulatory response during the absorption of amino acids. Cations - Although several cations including calcium, sodium,potassium and magnesium ions have been demonstrated to be involved in,or to modify the ins u l i n secretory response (Milner and Hales, 1970), i t appears unlikely that fluctuations i n plasma ion levels are important i n the normal regulation of insulin secretion. Amino acids - A prompt increase in plasma in s u l i n i s observed following the oral ingestion of a protein meal or the infusion of certain individual amino acids or a mixture of essential amino acids (Kipnis, 1972). Fajans et a l . (1967) found that amino acids have varying insulinotropic potency in man with decreasing potency in the order argixiine, lysine, leucine and phenylalanine, valine and methionine. The question of whether or not this effect of amino acids i s exerted directly on the g-cell has not been resolved. Since pancreozymin, a potent insulin secretagogue, i s released 16 following protein or amino acid ingestion, i t has been suggested that an enterc-0-cell axis exists for this class of nutrients (Kipnis, 1972). However, oral and intravenous routes of administration of a standard amino acid dose did not d i f f e r in the insulin secretory response e l i c i t e d . Since the insulinotropic potency of the amino acids i s largely dependent upon the concurrent plasma glucose l e v e l , these authors f e l t that, although protein and amino acids can affect insulin release, the plasma glucose level plays the dominant physiologic role i n the regulation of insulin secretion. In contrast the studies of Fajans et a l . (1967) led these workers to conclude that the stimulation of insulin secretion by amino acids obtained in their experiments indicated that this i s a physiologic rather than a pharmacologic phenomenon. The enteric hormone pancreozymin has been demonstrated to cause the release of glucagon i n the dog (Dupre, 1970), and may have a physiologic role i n the stimulation of glucagon secretion. This i s i n accord with the finding that various amino acids cause the level of glucagon-like immunoreactivity to increase in plasma when infused intravenously, butjthis effect i s enhanced when a given dose i s injected into the intestines (Dupre, 1970). The effect of alanine infusions on plasma glucagon levels revealed that a 50 to 100% increase in plasma alanine was associated with increased plasma glucagon levels (Muller, Faloona and Unger, 1971). These workers concluded that alanine does stimulate the secretion of glucagon while having very l i t t l e stimulatory effect on insulin secretion. The infusion of arg inine on the other hand tended to increase only insulin' release. It was 17 suggested that certain amino acids act on the pancreas so as to cause a release of hormones favouring protein synthesis, while others cause a release of hormones favouring gluconeogenesis. Another aspect of amino acid stimulation of endocrine release by the pancreas, which was revealed i n this study, was that certain amino acids can stimulate both insulin and glucagon release. However, the r a t i o of the two hormones released by the pancreas when stimulated by these amino acids may vary depending on the particular nutritional conditions. For example, alanine was found to cause the secretion of a low r a t i o of insulin to glucagon i n the presence of hypoglycemia but a high r a t i o during hyperglycemia. In other work, arginine has been demonstrated to induce a concurrent release of glucagon and insulin while modifying the plasma glucose level only s l i g h t l y (Cherrington and Vranic, 1971). More recent evidence suggests that hyperaminoacidemia causes a r i s e i n plasma glucagon and insulin l e v e l , and evidence has been presented which suggests that the role of glucagon during hyperaminoacidemia i s to prevent hypoglycemia due to aminogenic hyperinsulinemia (Unger et a l . , 1970). These authors speculate that glucose which enters insulin sensitive tissues i n the company of amino acids following aridno acid stimulated ijisulin release ,is replaced by hepatic glycogenolysis and gluconeogenesis stimulated by the simultaneous hyperglucagonemia. They also suggested that this would tend to be a useful metabolic adjustment particularly i n carnivorous animals or whenever protein is ingested without carbohydrate. Neural factors - Since the autotransplanted pancreas has been demonstrated to. perform near normally with respect to glucose homeostasis, 18 and since vagotomy in experimental animals or man did not significantly modify glucose tolerance, vagal innervation has been suggested to be unimportant with respect to pancreatic insulin release (Nelson et a l . , 1967). In a study of the effects of vagal stimulation, the only significant change resulting from vagal stimulation was a slight f a l l in the level of pancreatic vein IRI (Nelson et a l . , 1967). These authors suggested that since hypoglycemia stimulates sympathetic neural a c t i v i t y the stimulation of insulin secretion under these conditions would tend to aggravate the hypoglycemia. This suggests there is no rational basis for sympathetic neural stimulation of insulin secretion. However, other workers have provided evidence indicating that the nervous system may participate actively i n the stimulation of insulin secretion during the absorption of glucose from the gut (Dupre, 1970). Also Frohman et a l . (1967) found that the stimulation of the vagus nerve in the dog stimulated iBsulin secretion; The duration of insulin secretion was b r i e f and levels returned toward the baseline irrespective of continued stimulation. The results available would tend to suggest that i f the central nervous system influences the secretion of insulin through the vagus nerve, this action is probably of minor importance in the normal regulation of insulin secretion. A possible effect of the central nervous system on the secretion of insulin i s suggested by the presence of adrenergic receptors within the 3 -ce l l s (Loubatieres, 1971). A study of the influence of 6-adrenergic receptors on insulin secretion in the dog has been made by the administration of substances acting either to stimulate or inhibit these receptors (Loubatieres, 1971). The results of these studies indicated to these 1 9 authors that the g-adrenergic receptors of the 3-cells of the i s l e t s of Langerhan are of importance i n the process of insulin secretion. It has been postulated that the central nervous system might enhance the secretion of glucagon presumably through the autonomic nervous system (Vance, Buchanan and Williams, 1971). The presence of adrenergic and cholinergic innervation of the mammalian pancreas further suggests that the autonomic nervous system may play a role i n the regulation of glucagon release. Other hormones - As indicated the study of adrenergic receptors i n the pancreas suggested that the catecholamines may modify the secretion of insu l i n and of glucagon. In fact both of the catecholamines have been shown to inhibit glucose stimulated ijisulin secretion (Randle et a l . , 1968). Glucagon has been demonstrated to stimulate insulin secretion provided the plasma glucose level i s elevated (Kipnis, 1972). The direct effects of thyroxine, growth hormone and corticotrophin on insulin secretion are assumed to be minor cr insignificant at physiological hormone concentrations (Frohman, 1969). Similarly the hormones oxytocin, estrogens, vaspressin and angiotensin II have been demonstrated to stimulate insulin secretion directly or indirectly, but these effects were generally demonstrated with pharmacologic doses (Frohman, 1969). Ruminants Glucose - Runrinants have a relatively low steady concentraton of plasma glucose irrespective of feeding (McCandless et a l . , 1950), and this argues against glucose playing a role i n the regulation of pancreatic endocrine' secretion sirrdlar to that i n the non-ruminant. Also on the basis of the 20 relationship of glucose and insulin to time after feeding in sheep, Manns and Boda (1967) concluded that glucose probably does not regulate insulin secretion in sheep. These same workers have also shown that the insulin secretory response to a standard glucose load becomes progressively smaller as the young rwinant matures. Bassett (1972) has shown that injected glucose causes a f a l l in plasma glucagon. Volatile fatty acids - Manns and Boda (1967) have demonstrated that the VFA propionate and butyrate have a marked insulinotropic potential i n sheep. These workers indicated that these acids act directly on the pancreas to stimulate ins u l i n secretion, but f e l t that since glucagon has been shown to stimulate insulin secretion, the stimulation of insulin secretion by propionate and butyrate could also be due to an i n i t i a l release of glucagon. Horino et al.(1968) speculated that VFA may act indirectly to stimulate insulin secretion by stimulating the release of insulinotropic enteric hormones. Hertelendy et a l . (1968) i n v i t r o and Horino et a l . (1968) i n vivo confirmed the insulinotropic actions of propionate and butyrate. Horino et a l . (1968) also demonstrated that these acids were effective i i s u l i n o t r c p i c agents i n the cow. The insulinotropic a c t i v i t y of propionate and butyrate has been demonstrated in the new born lamb (Hertelendy, Machlin and Kipnis, 1969). These authors suggested that the responsiveness of the pancreas to these acids in sheep represents a constitutional characteristic present at b i r t h > which coincides with the concept that animals have evolved an insulin secretory mechanism that i s regulated by the major energetic substrate. Stern, Baile and Mayer (1970) have questioned the physiological 21 significance of VFA'stimulated insulin release. The effect o f loading the rumen with VFA in excess of normal concentrations on the plasma insulin levels suggested to these workers that VFA were not normally involved in the regulation of insulin secretion. Bassett, Weston and Hogan (1970). studied ruminal VFA production rates and the correlation to plasma insulin levels. The results indicated the VFA production in the rumen was not related to plasma insulin levels. Bartos, Skarda and Base (1970) attempted to determine the physiological significance of propionate and butyrate i n the stimulation of insulin secretion. Threshold plasma levels for stimulation of insulin secretion were found to be 74.9 ymole/liter and 45.0 ymole/liter for propionate and butyrate respectively. From their results these workers concluded that the secretion of insulin in sheep is regulated mainly by propionic and butyric acids. In earlier studies i n goats, P h i l l i p s (1966) demonstrated a butyrate-induced hyperglycemia following intravenous butyrate injection which could be abolished by pancreatectomy. These authors concluded that butyrate-induced pancreatic glucagon release acted to stimulate hepatic glycogenolysis. I t has also been reported that the fatty acid induced hyperglycemia i s chain length dependent with a maximum response induced by 6- and 8-carbon acids (Phillips et a l . , 1969). In studies with cattle short cr long-chain fatty acids f a i l e d to induce a h^'perglycemic response, and a similar result was found i n the dog ( P h i l l i p s , 1966). The physiological significance of VFA i n both insulin and glucagon secretion i s uncertain. With respect to the glucagon secretion, and subsequent glycogenolytic response to intravenous butyrate in the experiments of P h i l l i p s (196-6), i t i s worth considering the results of Horino et a l , 22 (1968). In this study butyrate was infused at a much lower level than i n the experiment of P h i l l i p s (1966), and there was a marked increase i n plasma insulin levels. However, regardless of the hyperinsulinemia there was neither hyperglycemia nor hypoglycemia. Horino et 31^(1968) suggested that the increased glucose u t i l i z a t i o n that would be expected following insulin secretion may have been compensated for by an increased glucose entry as a result of accelerated hepatic glycogenolysis presumably due to glucagon release. Thus glucagon as well as insulin secretion may occur following propionate or butyrate administration. Recently Bassett (1972) has measured directly an increase in plasma glucagon following the infusion of propionate, butyrate or valerate to sheep. The author concluded that this was a pharmacological rather than a physiological action of these compounds. Ketone bodies - Manns, Boda and Willes (1967) infused B-hydroxybutyrate into sheep and found there was no effect on the plasma :Lnsulin le v e l . Horino et al.(1968) injected 6-hydroxybutyrate and acetoacetate into sheep and found there was no significant effect on plasma insulin levels. Gastrointestinal hormones - Although secretin and pancreozymin have been indicated to be possible physiological stimulators of insulin secretion i n monogastric animals, studies by Baile, Glick and Mayer (1969) have indicated that these hormones do not serve a similar function i n ruminants. It was found that the pancreatic exocrine output of goats did not change following feeding. Secretin was found to be inactive while cholecystokinin-pancreozymin was active in stimulating pancreatic insulin secretion when injected intravenously. Although cholec^stoldLnin-pancreozymin was less active i n ruminants than non-ruminants the p o s s i b i l i t y remains that this hormone 23 plays a part in the regulation of insulin secretion. Amino acids - Hertelendy et a l . (1968) demonstrated that the amino acid arginine stimulates the release of insulin from the sheep pancreas in v i t r o . Bassett et a l . (1970) found a high correlation between the amount of protein digested in the intestines and the levels of plasma insulin i n sheep. However, the t o t a l amino acid level i n plasma and the individual plasma amino acid levels, except for tyrosine and valine,were not related to the plasma insulin level. These workers suggested that amino acids may act indirectly to stimulate insulin secretion by the stimulation of gastrointestinal hormone release. Cations - Littledike, Witzel and Whipp (19 68) have demonstrated that during postparturient hypocalcemia there was a marked reduction i n the insulin secretory response to intravenous glucose. The infusion of calcium' resulted i n a return to a normal insulin secretory response to glucose. Thus normal calcium levels i n plasma are required for ins u l i n secretion to occur i n ruminants. Neural factors - As in the non-ruminant the autotransplanted pancreas of the sheep has been demonstrated to retain the normal endocrine function (Bell et a l . , 1970), suggesting that neural effects on insulin secretion in the ruminant are probably no more important than i n the non-ruminant. Other hormones - Experiments by Bassettet a l . (1967) led these workers to conclude that glucagon acts to directly stimulate insulin secretion i n sheep as i t does in man. Hertelendy et a l . (1969) have demonstrated that epinephrine acts to block irisulin secretion in response to glucose as i t does in the non-ruminant. 1 24 Experiment . I One of the major objectives of this work was to determine the physiological significance of VFA i n the regulation of insulin secretion i n sheep. To carry out this study the accurate measurement of blood VFA levels was necessary. There were a'number of objections to methods previously available for blood VFA determination. Originally published concentrations of blood VFA were determined by l i q u i d partition chromatography (ELsden, 1946; McLymont, 1951; Moyle, Baldwin and Scarisbrick, 1948)which, though sensitive, proved tedious compared to the gas-liquid chromatographic (GLC) method of James and Martin (1952), or modifications of this method. Most methods currently in use for blood VFA determinations are of the • latte r type. Some disadvantages of these methods include: too elaborate (Bensadoun, • 1960), low sen s i t i v i t y (Ahnison, H i l l and Lewis, 1957; Annison, 1954; Baumgardt, 1964), short l i f e for colums in GLC analysis due to the relatively large sample volume (Annison, 1964; Erwin, 1961) and generally poor quantitative recoveries of the acids from blood when the amounts added were within the normal physiological range. Steam d i s t i l l a t i o n methods relying on t i t r a t i o n of the d i s t i l l a t e to measure the t o t a l acidity (Baumgardt, 1964; Bensadoun, 1960) may yi e l d erroneously high results due to other steam v o l a t i l e acids such as 6-hydroxybutyrate, HCl, or other such compounds i n the blood or i n trace amounts in chemicals used i n the analysis. A method was devised for the determination of blood VFA to suit the requirements of the experimental work of this thesis. 25 i Experimental Procedure A 20 ml sample of whole blood was collected for each analysis. The blood was centrifuged for 10 minutes at 3,000 xg and the plasma separated with a Pasteur pipette. Exactly 10 ml of plasma were added to 10 ml of d i s t i l l e d water and 5 ml of 85% orthophosphoric acid within a Markham s t i l l . This mixture was sprayed with an anti-foaming agent."'" A steam source was available, and steam entering the s t i l l at the time of sample addition was not passed through the sample but rather passed through the s t i l l drain. Steam d i s t i l l a t i o n of the sample was begun by f i r s t closing the sample entry port and then very cautiously closing the drain port (caution i s necessary since the samples w i l l foam i n i t i a l l y ) . If foaming continued for more than one minute, the sample was again sprayed b r i e f l y with the anti-foaming agent (the sample entry port must be opened as b r i e f l y as possible i n spray-ing the sample since sample acids may be los t ) . The volume d i s t i l l e d versus titratable acidity of the d i s t i l l a t e indicated that 150 ml of d i s t i l l a t e assured complete sample VFA d i s t i l l a t i o n . The 150 ml of d i s t i l l a t e collected i n a 250 ml Erlenmeyer flask were transferred with two 10 ml d i s t i l l e d water rinsings to a 250 ml beaker. Three"drops of 2N NaOH made the solution basic which was ve r i f i e d by add-ing- 2 drops of brom thymol blue (.1% i n dilute ethanol) indicator. This solution was then heated to dryness at 60 to 70 C. When dried, VFA salts were transferred to 5 ml beakers with 2 to 3 ml of d i s t i l l e d water i n .5 ml portions using a Pasteur pipette. Contents i n the beakers were then heated, to dryness at 50 to 60 C to avoid spattering. Beakers were capped with aluminum f o i l and stored for further analysis. Anti-foam. A s i l i c o n spray defoamer, Dow Corning. 26 To complete the analysis exactly 3 drops of 85% orthosphoric acid (89.8-1.8 y l i t e r s - SEM) were added with a Pasteur pipette through a small hole i n the f o i l l i d on the 5 ml beaker. Additional f o i l was immediately placed over the f i r s t to avoid v o l a t i l e acid loss. The phosphoric acid was then spread uniformly over the bottom of the beaker to dissolve a l l of the fatty acid salts. A small s l i t was made in the f o i l cover and a modified intravenous catheter needle placed through the s l i t about 1.0 cm from the bottom of the beaker (Fig. 1). A piece of masking tape was placed between the needle loop to close the hole i n the f o i l made by the needle. 2 The needle used was a 15 gauge, 15.24 cm intravenous catheter which had been separated from i t s plastic outer sheath and heated on a Bunsen burner to f a c i l i t a t e bending into a loop (the lumen of the needle must remain open). The plastic outer tubing was threaded onto the bent metal needle and the ends of each attached to rubber tubing (id 6 mm by 10 mm) with copper wire so that water could flow through the system (Fig. 1). Once attached to this apparatus (microcondenser), the beaker was gently heated by placing the t i p of a micro-Bunsen burner flame 10 cm below the beaker to avoid boiling the VFA solution. When a droplet had condensed and could be seen suspended from the needle, a small hole was punched i n the f o i l cover and a capillary tube (id 1.1 to 1.2 mm) inserted to collect the droplet. After collecting the droplet the sample could be analyzed by GLC or both ends of the capillary tube were sealed with Plastercine and labelled with masking tape. Samples to be immediately analyzed were retained i n the capillary tube 9 Jelco IV. Catheter Placement Unit. Jelco Laboratories, Raritan, New Jersey 08869. 27 Fig. 1. Arrangement of apparatus for micro-condenser described i n the-experimental procedure. A - rubber tubing; B - copper wire t i e ; 0 -transparent catheter; D - needle; E - aluminium f o i l ; F - masking tape; G - 5 ml beaker; H - location of vo l a t i l e fatty acids i n orthophosphoric acid; (-4—) - direction of water flow. 28 by sealing one end of the tube. Stored samples were opened by breaking the capillary tube off at one end. To analyze by GLC a Hamilton 1.0 y l i t e r syringe (7,000.series) was inserted into the capillary tube and .5 to 1.0 y l i t e r drawn up as required. This sample was then analyzed on a gas 3 4 <_hromatograph , with an auxiliary recorder , using the following settings: a i r 283 ml/min, nitrogen 50 ml/min, hydrogen 60 ml/min, temperature columns 160 C, i n l e t 195 C, detector (dual flame ionization) 195 C. The electrometer input attenuator was set at XI0 while the output attenuator was varied from 1 to 16X as required. The recorder was operated at 2.54 cm/min at a fixed span module of one m i l l i v o l t per 25.4- cm of chart. The column was 183 cm (6 f t ) long by .64 cm On in) and the packing material was the same as that used by Baumgardt (1964), with a so l i d support of fireb r i c k 60/80 mesh with neopentylglycolsuccinate (20%), and 2% H^PO^ the' l i q u i d phase. Relative responses determined for the acids detected were acetic 1.00, propionic 1.68, butyric 1.98, and isovaleric 2.21, agreeing closely with published results (McNair and Bonelli, 1969). The peak areas on the chromatograph were determined by triangulation (i.e.- height of peak in millimeters x width of peak at .5 height in mm). To quantitate the VFA from the chromatograph exactly 10 ymole of isovaleric acid were added to each plasma sample prior to steam d i s t i l l a t i o n . Isovaleric acid could be used as the internal standard since the amount of this acid i n blood i s neglible. To make the raw peak areas molar re l a t i v e , the raw areas were multiplied by correction factors for acetate 1.000, propionate .392, isobutyrate .270, butyrate .280, isovalerate .276 and valerate .239. ^icro-Tek Model GC 2000 MF • ' '-4 Westronics MT-21 Recorder 2 9 Although reagents were A.C.S. reagent grade, sodium isovalerate had a molar percentage composition for sodium isovalerate 97.53, acetate 1.38, isobutyrate .75, propionate 1.17 and isobutyrate .17. Consequently, another correction of the f i n a l acid concentration was necessary. The internal standard was routinely analyzed to verify the extent bf impurities. Pure salts of other acids in prepared VFA standard mixtures were free of contaminants with .the exception of isobutyrate which contained 2,3 and 3,8 mole per cent of acetate and propionate respectively. In practice i t was most expedient to operate the microcondenser and GLC simultaneously. Samples could i n this way be analyzed at an effective rate, through the microcondenser and GLC of about one sample per 10 minutes. Results and Discussion Since variations in procedure may affect the proportions of VFA condensing on the microcondenser, analyses of a series of standard VFA mixtures were made and the optimum operating conditions determined. One ml volumes of VFA standard solutions were analyzed (Table 1). Acidifying the VFA salts with 3 drops of phosphoric acid and analyzing the f i r s t droplet condensed gave the least variable and most accurate results. The 95% confidence interval for the mean of 4 determinations included the standard value for a l l 5 acids measured. Increasing or decreasing the number of drops of phosphoric acid added, or analyzing second or third droplets condensed, resulted i n lower recoveries as seen i n Table 1. Var i a b i l i t y of the method was evaluated by analyzing VFA mixtures in Table 2 . A wide range of standards was analyzed and probably extends TABLE 1 Drops of 85% orthcphosphoric acid 1 2 3 4 Micrcdenser 29.1-1.7 59.9-1.2 89.8^1.8 119.8^2.4 droplets collected 1 1 2 1 2 1 2 3 Acid standards (yEq) * Acetic 20.00 Mean 24.46 23.68 18.82 19.42 18.82 17.96 20.92 21.62 SEM .62 1.83 1.01 .29 1.27 .38 .98 1.90 Recovered % 122.1 114.4 94.1 97.1 94.1 89.8 104.6 108.1 SEM 2.5 7.7 5.4 1.5 6.8 2.1 1.8 8.8 Propionic .900 Mean 1.285 ' 1.294 .784 .876 .831 .851 .880 .824 SEM .155 .214 .024 .025 .038 .061 .065 .042 Recovered % 142.8 143.8 87.1 97.3 92.3 94.5 97.8 91.5 SEM 12.1 16.5 3.1 2.9 4.6 7.2 7.4 5.1 Isobutyric .800 Mean .875 .875 .587 .766 .658 .741 .650 .595 SEM .049 .202 .029 .018 .028 .016 .025 .017 Recovered % 109.4 126.2 73.4 95.8 82.2 92.6 81.3 74.4 SEM 5.6 23.1 4.9 2.3 4.3 2.2 3.9 2.9 Butyric .800 Mean .999 1.160 1.160 .808 .753 .802 .766 .721 SEM .120 .208 .010 .024 .030 .032 .021 .033 Recovered % 124.9 145.8 101.1 100.0 94.1 100.2 95.7 90.1 SEM 12.0 17.9 1.2 3.0 4.0 4.0 2.7 4.6 Valeric .800 Mean .866 .956 .876 .834 .842 .754 .861 .838 SEM .140 .220 .016 .013 .020 .013 .027 .031 Recovered % 110.7 119.5 109.5 104.5 10.5.2 94.3 107.6 104.7 SEM 16.4 23.0 1.8 1.6 2.4 1.7 3.1 3.7 a Mean and SEM for the volume in ndcroliters corresponding to the number of drops of acid. Ten pipettes were used for each mean and SEM Number of samples = 4 Table 2. Composition of 12 vo l a t i l e fatty acid standard mixtures analyzed by the method described. Acetic Propionic Isobutyric Butyric Valeric mmole/liter (ymoles/liter) .05e sg • 5 h 5 6 5 e .10* 10 e 10 f 10 g 10 h .20g 20 a 2 ^ 20 C 20 d .50a f 3CT 30 g 30 h 30 e .80f 50° 50 d 50 a 50 b 1.00b 70^ 70 e f 70 70 g 1.2CP 100 d iooa ioob 10 oc 1 . 5 0 C 130^ 13 0 k 130 1 13 0 h 1 . 8 0 * 180* 18 0 1 180 1 180 1 2.00d 200 b 200° 200d 200 a 2 . 2 0 1 20 0 1 200 1 200^ 200^ 2 . 5 0 1 22 0 1 220^ 220 k 220 k Acid concentrations with the same l e t t e r superscript formed a standard mixture. 32 considerably aboye the normal physiological range for each acid in blood. Each standard mixture in Table 2 was analyzed 8 times. With increasing standard concentration there was increasing variance in the determinations. Because of this lack of homogeneity of variance, regression analysis on the raw data was not possible. Therefore a log transformation of both determined and standard values was made since this gave uniform variance over the range. Regression analysis was on the transformed data. The regression lines and confidence limits were also calculated with the transformed data and then were converted to the i n i t i a l scale of measurement by antilogs. Regressions and confidence limits for the acids are found i n Figs. 2, 3 and 4. Similar graphs were obtained for valeric and isobutyric acids. The confidence limit closest to the regression line has been designated CL and represents the estimated 35% confidence interval for a mean determined value with a large number of standard samples analyzed. The confidence intervals corresponding to n = 1, 2, 3 and 4 have a s l i g h t l y different meaning. In using the method a series of determinations on an unknown sample are made and the mean determination related to the corresponding standard. Therefore the 95% confidence interval now applies to the standard rather than to the determined value. The confidence l i m i t n = 1 i s the 95% confidence interval about the standard concentration for a single determination. Similarly the confidence limit n = 2 is the 95% confidence interval about the standard concentration which corresponds with the mean of 2 determinations on an unknown sample. 33 For ease of comparison a l l 5 confidence limits are plotted to be read off the deterTiiined concentration axis even though n = 1, 2, 3 and 4 confidence limits actually apply to the standard axis. As an example, i f 3 determinations on an unknown sample have a mean acetate concentration of 1.012 m mole/liter the corresponding standard can be determined by substituting y = 1.012 m mole/liter in the regression equation and solving for x = 1.000 m mole/liter. By reading the confidence limit above and below the regression line at x = 1.000 m mole/liter on the y axis, the confidence limits about the mean value are 1.140 to. 0.860 m mole/liter. In practice the determined value can be assumed to equal the standard value since the constant in the regression equation is small and the regression coefficient i s near 1.00. The number of replicate analyses necessary to give a mean f a l l i n g within a given 95% confidence interval, can therefore be estimated from Figs. 2 to 4 provided the approximate concentration of an unknown i s available. Generally propionic, isobutyric and butyric acid occur in ruminant blood at below 50 ymole/liter and duplicate analyses are therefore adequate for the determination of these acids. Acetic acid has been reported i n sheep jugular blood at as high as 3.5 ymole/liter (Senel, 1967). However usually i t does not exceed 1.5 m mole/liter and duplicate analyses may therefore be acceptable. Correlation coefficients for the acids appeared high considering the dispersion of data about the regression lines at the higher concentrations. This was attributed to the small deviations from the line at the lower standard concentrations. 34 Fig. 2. Graph shewing relation between determined and standard acetic acid concentrations with regression line and confidence intervals explained in text. r 0.50 1.00 1.50 , 2 . 00 —SO Micromole/Liter x 10 S T A N D A R D C O N C E N T R A T I O N Fig. 3. Graph showing relation between determined and standard propionic acid concentrations with regression line and confidence intervals explained in text. 36 MicroM/UtOr „ 10 2 S T A N D A R D C O N C E N T R A T I O N Fig. 4. Graph showing relation between determined and standard butyric acid concentrations with regression line and confidence intervals explained in text. 37 Samples of bovine and ovine venous jugular blood were analyzed by this method and the results summarized in Table 3. A single large sample of bovine blood was collected and used for a l l of the analyses. A similar sample of sheep blood was collected and used for a l l analyses. The cow sampled had received feed (7 a l f a l f a hay: 4 beet pulp: 1 mixed grain) approximately 3 hours prior to sampling. The sheep had been fed (4 a l f a l f a : 1 barley) about 2 hours prior to sampling. A typical chromatograph obtained i s seenin Fig. 5. Accurate recoveries of the acids added at physiological levels were achieved for both cow and sheep blood plasma (Table 3). The largest deviations from 100% recovery were for acetic acid from sheep blood (110.3%) and for butyric acid from sheep blood (110.0%). However i n this type of recovery the variation of the procedure i s effectively doubled since variation about the difference i s the sum of the variation associated with the 2 mean determinations (i.e. plasma and standard A, and plasma alone). Also error in the per cent recovery values due to other causes besides procedural variation may be as much as doubled in the difference since errors from the 2 determinations are additive. The 95% confidence interval about the means for the recoveries includes the standard. However these intervals: were wide. The method therefore appears v a l i d . Volatile fatty acids in cow and sheep blood plasma are within range of those previously published (McLymont, 1951; Annison, 1954; Annison et a l . , 1957). A trace of isobutyric acid was also found i n the cow blood which was analyzed but neither isovaleric nor valeric acid could be detected. For sheep blood isobutyric acid was detected at 2.0- ymole/liter as well as a trace of isovaleric acid, but no 38 Fig. 5. (A) Chromatogram of sample from 10 ml of sheep blood plasma to which 10 yEq of isovaleric acid had been added. Fig. 5. (B) Chromatogram of sample of 10 ml of sheep blood to which 1 ml of Standard B (see Table 3 for composition) had been added. Output attentuation indicated beneath each acid with input attentuation set at XlO. Table 3. Recovery of vo l a t i l e f a t t y acids from blood plasma when added i n quantities within the physiological range CValues are means, and.standard errors). Plasma Acid Plasma analyzed Number of analyses CIO ml) t Standard B Plasma CIO ml) Difference Standard B (1 ml) Recovered Cmicro equivalents) % Acetic Cow 3 20.96±.18 10.27+.65 10.69+.83 10.00 106.917.8 Sheep 3 18.71+.28 7.681.83 11.03+1.11 10.00 .110.3+10.1 Propionic Cow 3 .94+.03 .11+0.1 ,83± .04 .90 92.214.8 Sheep 3 1.09±.03 .22+.02 .871.05 .90 96.615.7 Butyric Cow 3 .64±.03 .04101 .601 .04 .60 100.016.6 Sheep 3 ,851.02 .19+.01 .661 .03 .60 110.014.5 Isovaleric Cow 3 10.00a io.oo a 10.00a Sheep 3 10.00b 10.00a 10.00a a Isovaleric acid was the internal standard CO CO 40 valeric acid could be detected. To demonstrate the sensitivity of the method, jugular blood v o l a t i l e fatty acids were determined i n 2 sheep (Number 10 and 20) which had fasted 8 hr (32 hr since last feeding). Sheep 10 was then fed 400g of chopped a l f a l f a hay, and blood samples were taken again from both sheep at 2.5, 4.5 and 7.0 hr. At 4.5 hr another 400g of hay was fed to sheep 10. A l l feed offered was consumed within 30 minutes. The results are found i n Fig. 6. Sharp increases i n plasma acetate, propionate and isobutyrate but not in butyrate occurred at 2.5 hr in the fed sheep over the fasted one.. At 4.5 hr the circulating levels appeared to remain about constant or declined s l i g h t l y , but a l l 4 acids appeared to increase again following the second feeding. In Fig. 6, the horizontal line above and below the plotted means for the graph of acetic acid for the fed sheep indicate the 95% confidence interval for the mean of a duplicate determination. The confidence interval was determined as described for Fig. 2. A standard error appropriate to each mean was determined from the confidence l i m i t by the relationship t.05 x SEM = CL. -The standard error was included within the thickness of the graphed li n e for acetic acid for sheep 10 except for hr. 7.0 where the standard error i s inside the confidence limits. Changes in peripheral circulating VFA might be assumed to occur following feeding i n ruminants and have been measured for acetate (Reid, 1950; Bensadoun, 1960; Senel and Owen, 1967). Post-prandial changes in jugular blood propionate and butyrate have been determined in one other study (Thye, Warner and Mill e r , 1970) i n which changes i n acetate, propionate and butyrate were measured with 41 Fig. 6. Graph showing changes in blood v o l a t i l e fatty acids of a single fed sheep over a fasted control sheep following feeding of 400g of chopped a l f a l f a hay at 0 and 4.5 hr. 42 time after feeding i n lactating ewes. Peak plasma levels for acetate, propionate and butyrate were 3,5 and 5 times higher than values reported i n this work. Although these differences may have originated from differences in experimental conditions, differences i n the s p e c i f i c i t y of the methods cannot be excluded. Other factors known to affect blood VFA levels include metabolic (. Aafjes, 1964) and nutritional status (Annison, 1954), type of ration consumed (Reid, 1950), individual animal differences (Craine and Hansen,1952), species differences (McClym'ont, 1951), the type of blood analyzed (i.e. a r t e r i a l blood would be expected to have a relatively uniform VFA concentration whereas venous blood varies markedly depending upon the organs drained) and whether whole blood or blood plasma i s used for the analysis (Annison, 1954). This method does not include the determination of formic acid which is a normal constituent of ruminant blood (Annison, 1954; Bensadoun, 1960). It was considered that the method described here, for VFA analysis, would not be useful for formic acid. Some oxidative decomposition of formic acid would be expected either during heating for micrccondensation or when injected into the gas chromatograph at 195 C, because this acid normally decomposes at about 160 C (Fieser and Fieser, 1961). Therefore a procedure for the chemical determination of formic acid i n blood is preferable to. GLC analysis (Grant, 1948). However the method described permits accurate analyses for blood acetic, propionic, isobutyric and butyric acids of ruminants. Although valeric acid was not detected i n the blood analyzed for this work, estimation of valeric acid at as low as 5 ymoles/liter was possible with standard solutions. This method appears to have definite advantages over methods 1+3 previously reported, and the a b i l i t y to accurately recover VFA added to plasma in physiological amounts has been demonstrated. 44 Experiment II Part A. It has been suggested that blood VFA are of major importance in the regulation of insulin secretion in ruminants, and the plasma glucose level is relatively unimportant. Using the method of experiment I, i t was felt that this concept could be tested by measuring the extent of variation in plasma VFA, glucose and insulin, and the degree of correlation between these blood metabolites. Experimental Procedure Twelve mature Dorset Horn wethers weighing 59.1- 6.5 kg ± (SEM) were randomly divided into 3 equal groups and fed barley (B), or hay (H), or a mixture of equal parts of barley and hay (BH). The sheep were fed these diets for at least 3 weeks prior to an experiment. The composition and proximate analysis of the 3 feeds used are given in Table 4. The methods of AOAC (1960) were used for the determination of the feed proximate fractions. The feeding of equal amounts of feed BH was accomplished by offering only hay for 15 minutes. At the end cf t h i s period, the amount of hay consumed was determined and an equivalent amount of barley was.fed during the next 15 minutes. Hay was then fed for 1 hr followed by an equal amount of barley during the next 30 minutes. For each of the 3 remaining 2-hr intervals during which analyses were conducted, the sheep were fed hay for 1 hr and an equivalent amount of barley during the second hour. When 45 Table 4 The composition and proximate analysis of the feeds used in the experiments reported Feed Proximate Fractions % Rolled Earlev (B) Long-Hay (H ) Barley-Hay (BH) Dry Matter 89.0 89.4 89.2 Crude Fiber 5.0 30.3 17.6 Ether Extract 1.8 2.5 2.1 N-free Extract 68.2 45.3 56.8 Crude Protein (N x 6.25) 11.6 7.0 9.3 Ash 2.4 4.3 3.4 fed i n this manner, the sheep consumed a l l the barley offered within 10 minutes. Water was provided ad libitum throughout the experiments. Prior to the start of an experiment, the sheep were fasted 24 hr. to ensure a maximum response i n the parameters measured. A zero time blood sample of 50 ml was collected from the external jugular vein using vacutainer needles (Ross and K i t t s , 1969). The sheep were then fed the experimental diets as described above, and further blood samples collected at 2, 4, 6 and 8 hr after the commencement of feeding to sheep on feed H and B and an additional 10 ml sample at 30 min. for animals consuming the BH feed. Blood samples were immediately cooled i n water and then centrifuged at 3000 x g for 10 min. at 5°C and the plasma stored at about 0 C for not longer than 3 days. The plasma used for insulin determinations was stored 46 at -20 C for not more than 4 weeks prior to analysis. The method described was used for the determination of plasma VFA (Ross and Kitts, 1971), and 5 the glucose oxidase method for the determination of plasma glucose. The concentration of plasma insulin was determined using a radioimmunoassay g procedure . The precipitated insulin antibody complex was collected on millipore f i l t e r s which were placed on 4 cm square pieces of aluminum f o i l . After drying, the f o i l was folded to y i e l d f l a t 1 cm squares, which were then placed on a lead block and covered with the sodium iodide crystal of the 7 s c i n t i l l a t i o n apparatus . The effect of feed on the blood metabolites was tested using an analysis of variance. Duncan's new multiple range test was used to test the difference between means. The relationship of VFA and glucose to plasma insulin concentration within feeds was determined by simple and multiple regression analyses (Snedecor and Cochran, 1967). Changes in plasma concentration over time were tested by students t using zero time values as the control. To test for significant differences i n the R squared values, the reduction i n the residual sum of squares was employed (Harvey, 1960). 5 Worthington Biochemical Corporation, Freehold, N.J. Glucostat; prepared reagents for the enzymatic determination of glucose. g Amersham/Searle, Arlington Heights, 111. Insulin Immunoassay Kit; radioactive iodine labelled insulin and insulin anti-serum. 7 Crystal S c i n t i l l a t i o n Detector, Model DS5; and Decade Scaler, Model 151. Nuclear Chicago Corporation. 47 Results The amount of feed consumed and the concentrations of plasma glucose, VFA and insulin at various intervals after feeding are indicated i n Figs. 7, 8 and 9. During the f i r s t 2-hr interval about 800 g of feed were consumed by sheep fed BH (Fig. 9) as compared to 500 g by those fed B (Fig. 8) and 400 g by those fed H (Fig. 7). The mean tot a l feed consumed was 1550, 957 and 875 g for the BH, B and H feeds respectively. Under the conditions of ad libitum feeding, i t was not possible to distinguish between the effect of level of intake or the rate of consumption from that of the feed per se on the parameters studied. The concentration of insulin i n the plasma did not show any change at 2 hr for the animals studied. In sheep fed ration BH there was- a significant (P < .05) increase i n plasma insulin level at 4 hr followed by a further increase by 6 hr. Sheep fed B shewed considerable variation i n insulin levels as indicated by the re l a t i v e l y large standard errors (Fig. 8). Although the insulin concentration appeared to increase at 2 and 4 hr, these changes were not significant. At 8 hr, however, the concentration of insulin was significantly higher (P < .01) than that observed at zero time. Sheep fed only H did not shew significant changes in the insulin concentration throughout the period. In sheep fed BH, the concentration of glucose at 2 hr was less than at zero time (P < .10). However the other plasma constituents studied did not change significantly at 2 or 4 hr. At 4 hr, the concentration of glucose was not different from zero time, but was greater than at 2 hr (P < .05) and corresponded with the increase i n plasma insulin (P < .05) at this time, (Fig. 9). At 6 hr the concentrations of plasma glucose (P < .01), acetate (P < .05), propionate (P < .05) and butyrate (P < .05) were higher than at 48 S i 50 40 30 20 3g S i 90 8 0 70 6 0 UJ < Q Ul Ul U. E 1600 1200 8 0 0 4 0 0 F E E D H gm/2 hr interval 0 cumulative intake Q 0-0 2-0 4- 0  0 TIME hr > I i >1t i-r-7 6 0 > i > > 7- > 7-7-1 8 0 Figure 7. Jugular plasma acetate, propionate, isobutyrate, butyrate, glucose and insulin concentrations measured postprandial i n wethers consuming feed H. (* P <.10; ** P <.05; *** P <.01) F E E D B z e i s to z 50 40 30 20 LU < J— ? E LU LU U-1600 1200 aoo 4 0 0 gm/2 hr i n t e r v a l 0 c u m u l a t i v e i n t a k e • o-o V (212) (301) 2-0 4-0 T I M E hr / / / & 0 8 0 Figure 8. Jugular plasma acetate, propionate, isobutyrate, butyrate, glucose and insulin concentrations measured postprandial i n wethers consuniing feed B. (*P <.10; ** P <.05; *** P <. z e — V. _l to CO c z =» - =k 50 40 30 20 90 c o O 8 0 o o 70 = 0> -1 c O fc 6 0 12 LU w H « 8 < ~ CC >- ® J— O 4 O E CD ^ 1600 LU < 1200 Z O" 8 0 0 Q LU 4 0 0 LU U. FEED H B BUTYRATE • . ISOBUTYRATE g m / 2 h r in terva l c u m u l a t i v e Intake. WMkmrnn, 0 0 2-0 4 2-0 4-0 6-0 T I M E hr 8 0 Figure 9. Jugular plasma acetate, propionate, isobutyrate, butyrate, glucose and insulin concentrations measured postprandial in wethers consuming feed BH. (* P <.10; ** P <.05; ft** p <.01). 51 zero time and these values remained elevated at 8 hr. The concentration of plasma glucose was increased i n sheep fed B at t and 6 hr (P < .10). The plasma propionate concentration increased at 8 hr (P < '..05) and this corresponded with the increased insulin level (P < .10) at this time (Fig. 8). Changes i n plasma butyrate and isobutyrate were quite erratic, resulting i n large standard errors for isobutyrate at 4 hr. Although the mean plasma butyrate concentration was increased almost 30-fold at 8 hr, i t was not significantly different from zero time. For sheep fed H the level of plasma acetate (F *.01), propionate (P < .05) and isobutyrate (P < .05) were elevated at 6 hr. At 8 hr the level of glucose (P < .05), acetate (P <.01) and propionate (P < .05) remained high. There was no corresponding increase i n the concentration of plasma insulin at this time (Fig. 7). The relationship between the plasma metabolites was tested further by simple and multiple linear regression analysis. A plot of the data for each of the VFA and glucose against the level of plasma insulin indicated that non-linear relationships were not important. The results of the regression analyses are shown in part i n Table 5. A l l possible combinations of metabolites were regressed on insulin. Included i n this table are the R squared values which significantly increased the R squared values over those of the uncombined variables. The simple linear regression analysis indicated a relationship between the concentration of plasma insulin and isobutyrate for feed H, and between butyrate and plasma insulin for feed B. For the BH feed, glucose, acetate, propionate and butyrate were correlated with the plasma insulin l e v e l with 'butyr.oite.^ying the highest R squared value at 59.91%. 52 Table 5. The R squared expressed as a percentage and significance associated with simple and multiple linear regressions of plasma glucose (G), acetate (A), propionate (P), isobutyrate (I), and butyrate ( B ) levels regressed on plasma insulin levels for the three feeds. Metabolites df H BH B G 1+18 7.21 n.s. 50.45** .07 n.s. A 1+18 14.26 n.s. 22.61* .06 n.s. P 1+18 9.94 n.s. 26.14* 15.62 n.s. I 1+18 •22.02* .28 n.s. 6.12 n.s. B 1+18 4.41 n.s. 59.91** 59.12** GB 2+17 11.29 n.s. 71.17** .59.12** BI 2+17 26.40 n.s. 77.35** 59.28** PA 2+17 16.93 n.s. 27.00** 48.79** BA 2+17 18.07 n.s. 59.92** 67.39** BP 2+17 10.57 n.s. 60.48** 65.25** GBI 3+16 28.95** 83.64** 59.40** BIA . 3+16 38.51** 77.96** 71.00** R squared significantly greater than that of the uncombined variables for: Feed B: PA>P **; BIA>B** H: BIA>I* BH: GB>B**; BI>B**; GBI>GB** '* P .05 ** P .ol n.s. not significant 53 The multiple regression analysis which resulted in significantly increased R squared values are also indicated i n Table 5. Combining acetate with propionate on the barley feed increased the R squared value over that of propionate (P = 15.62% to 48.79%). Combining isobutyric and acetate with butyrate increased the R squared within feed B (B = 59.12% to BIA = 71.0%). There was a similar effect for feed H when butyrate and acetate were combined with isobutyrate (I = 22.02% to BIA = 38.51%). In the BH feed, a number of multiple regression analyses shewed increasing R squared values on combining metabolites. Combining glucose and butyrate resulted i n an increased R squared value (B = 59.9% to GB = 71.17%). Although the simple regression indicated that isobutyrate was not related to plasma insulin, combining isobutyrate with butyrate greatly increased the R squared value over butyrate alone (B = 59.91% to BI = 77.96%). The effect of combining isobutyrate with glucose and butyrate also resulted i n an increased R squared value (GB = 71.17 to GBI = 8 3 . 6 4 ) . Discussion The level of feed intake may have affected the extent to which the plasma metabolites changed i n this study, particularly f o r feed BH as compared to feed B or H. Feeds B and H were consumed i n approximately equal amounts and differences between them for the parameters studied may have been r e a l feed differences. In spite of f a i r l y large changes i n the mean plasma glucose levels for the H and B feeds for the periods studied, there was no relationship found between plasma insul i n level and that of glucose. However, plasma insulin level was significantly related to plasma isobutyrate for feed H and to 54 butyrate for feed B. When these 2 feeds were fed combined, a r i s e i n plasma glucose of similar magnitude to that of feed B was observed. However, for the BH feed the plasma glucose was related to the plasma insulin l e v e l . The almost 2-fold greater amount of feed BH consumed than of feed B or H may have influenced the relationships detected. For the BH feed there was a significant decline i n plasma glucose at 2 hr after feeding. A similar drop in plasma glucose level has been detected i n lactating ewes receiving a pelleted ration (45% concentrate: 55% roughage) (Thye et a l . , 1970). These workers suggested that the drop in plasma glucose may have been mediated by a r i s e i n plasma insulin i n response to measured increases i n levels of circulating propionate and butyrate. However, the plasma insulin l e v e l measured i n this study remained constant at the time of the drop in plasma glucose. Although relationships established in this study do not necessitate a dependency of plasma insulin level upon the plasma metabolites to which insulin i s related, Horino et a l . (1968) and Hertelendy et a l . (1968) have found that for sheep butyrate is the most potent insulinotropic VFA of butyrate, propionate and acetate. The relative potency of isobutyrate has not been studied. In a study by Bassett (1972), plasma insulin levels were measured in sheep following the feeding of 800 g of an equal mix of hay and grain. Plasma insulin levels increased from 15 to 50 yU/ml by 2 hr after feeding and gradually returned to prefeeding levels by about 10 hr after feeding. Plasma glucose levels significantly increased at 8 hr after feeding and then gradually returned to the prefeeding value by 24 hr after feeding. The t o t a l plasma VFA level was found to reach a maximum at about the same time as 55 the plasma insulin l e v e l . In the study of this thesis, butyrate and i t s structural isomer were the most important VFA with respect to their relationship to the plasma insulin level i n this study. Propionate, which was usually present i n plasma at higher levels than butyrate, was found to be relatively unimportant. Glucose was only related to the plasma insulin level for feed BH. Generally an increase in peripheral plasma insulin level is taken to be synonymous with increased insulin output by the pancreatic i s l e t , although other factors such as decreased insulin uptake by tissues could be responsible. Unfortunately, the extent to which insulin removal from blood i s affected by the plasma insulin level or other factors i s not well known, although the hepatic uptake of insulin from portal blood has been shown to increase with increasing insulin concentration (Field, Webster and Drapanas, 1968). However, the interpretation of changes in plasma insuli n levels i n terms of rates of insulin secretion seems warranted i n li g h t of the h a l f - l i f e of circulating insulin i n the order of 30 min (Randle et a l . , 1968). Also the fractional rate of insul i n degradation does not appear to be very sensitive to insulin dose levels within the physiological range or even within reasonable pharmacological dose limits (Yalow and Berson, 1965). Manns et a l . (1967) showed that with increased insulin secretion i n sheep there was a r i s e i n the circulating plasma insulin l e v e l , but found that the jugular plasma insulin level was a much less sensitive indicator of insulin secretion than was the portal venous plasma insulin l e v e l . If i t i s assumed that insulin secretion i s reflected by an elevated plasma insulin -level, and i f the secretion of insulin in ruminants is 56 dependent upon the circulating level of certain VFA, then the c r i t e r i a for a VFA that regulates insulin secretion should include that the plasma level of this VFA be related to the plasma insulin level. In this study only butyrate for feed B and BH and isobutyrate for feed H f u l f i l l e d this requirement. From this study i t would seem premature to exclude butyrate as a regulator of insulin secretion in ruminants. 57 Part B. In Part A, acute changes in plasma metabolites were induced by pre-fasting and refeeding ad libitum. The following experiments were designed to obtain data on normal plasma VFA, glucose and insulin levels in lactating ewes fed twice daily, and also on the relationship between plasma glucose and insulin levels in 2-week old lambs. 1. Lactating ewes Experimental Procedure Eight Dorset Horn ewes i n early lactation were divided into 2 groups of 4 animals. Feed was provided (1.36 kg rolled barley and 1.36 kg a l f a l f a hay) i n 2 equal feedings daily at 0800 and 1500 hr. Blood sampling was staggered so that alternate groups were sampled at each 2 hr interval through a 24 hr period. Blood was collected using vacutainers (Ross and K i t t s , 1969). The methods of blood handling and storage were the same as for Part A, as were the methods of blood glucose, VFA and insulin determination. Results and Discussion The results of this study are seen i n Fig. 10. Under the conditions of this experiment, the plasma glucose level reached a minimum between 0800 and 1000 hr followed by an increase after feeding that reached a peak between 1600 and 1800 hr, followed by a steady decline to the daily nadir. The mean concentration change for plasma glucose level was 10 mg/100 ml between 50 and 60 mg/100 ml. A similar diurnal change was observed i n the blood levels of acetate and propionate. The pattern of plasma butyrate change differed i n that there appeared to be a maximum shortly after each feeding. The f i r s t maximum for plasma butyrate level occurred at about 2 hr after the 0800 hr feeding, and was followed by a minimum at 1300 hr. A further increase following the 1500 hr feeding, and a maximum at about 1700 occurred, 58 Fig. 10. Diurnal variation i n jugular plasma insulin, glucose and VFA in lactating ewes. 59 followed by a slow decline to a irdnimum at 0800 hr. There appeared to be group differences for the plasma level of acetate, propionate and butyrate, since one group had consistently higher plasma concentrations. However, the patterns of VFA change were the same for both groups. The plasma isobutyrate level was low and relatively constant throughout the period of study. These results indicate that plasma VFA generally increase i n plasma following feeding, and thus could provide a normal stimulus for insulin secretion after a meal. The ingestion of food r i c h in carbohydrate has generally been demonstrated to be associated with increases i n plasma glucose and insulin levels in monogastric animals. The study of diurnal variation in mice (Steffens, 1970) suggests that post-prandial increases in plasma glucose and insulin are much greater i n the monogastric than the ruminant animal. Oral glucose loads given to humans provoke rel a t i v e l y large increases in plasma glucose and insulin (Metz and Feidenberg, 1970; Castro et a l . , 1970). Unfortunately, the insulin data reported i n this study was of dubious value, due to the inappropriate freezing of certain of the insulin assay reagents. Triplicate determinations showed considerable v a r i a b i l i t y . As a result, i t was not possible to determine i f plasma insulin levels show a diurnal variation, or i f plasma insulin level i s normally correlated with plasma glucose or VFA. This experiment did indicate that studies of this type would not y i e l d information on the regulatory role of VFA or glucose in the secretion of insulin, since these metabolites varied i n a similar way following feeding. A relationship established between plasma insulin level and one of the k metabolites measured in this study would probably exist for the other 3 metabolites. This point was discussed i n Experimental I I , Part A, and 60 possibly explains seme of the results obtained. 2. Lambs Experimental Procedure A t o t a l of 16 lambs, ranging i n age from 13 to 17 days, were used i n the following experiment. Blood samples were collected from the jugular vein by vacutainer (Ross and Ki t t s , 1969), and plasma glucose and insulin were measured using the methods previously described. The dams were fed 1.36 kg rolled barley and 1.36 kg a l f a l f a hay i n 2 equal portions twice daily. Four single and 4 twin lambs were allowed free access to milk from their dams, while an additional 4 single and 4 twin lambs were prevented from suckling by covering the udder of the dam for a 6 hr period. Blood samples were collected from a l l of the lambs at 30 min intervals for 2 hr. Zero time samples were collected from pre-fasted lambs prior to udder cover removal. A regression analysis was carried out, relating plasma glucose and insulin for each of the lamb groups. Results and Discussion The results of this experiment are seen i n Fig. 11. For both the single and twin lamb groups, refeeding following a fast resulted i n rapid increases in.both plasma insulin and glucose, whereas glucose and insulin remained rel a t i v e l y constant in ad libitum suckled lambs. The plasma glucose and ixisulin levels were lower for twin than for single lambs, and this may r e f l e c t a r e s t r i c t i o n i n the a v a i l a b i l i t y of milk to twin lambs. The results also indicate the immediate a v a i l a b i l i t y of exogenous glucose to the metabolic pool i n the suckling lamb. Accordingly, glucose i s a much more important physiological fue l i n the lamb than the adult (Jarrett et a l . , 1964). This i s further indicated i n the study of these workers by the finding that the glucose entry rate of the lamb was about 6 1 62 5 mg/min/kg compared to 1.3 mg/min/kg in the adult sheep. Also, the glucose pool determined for the lamb was 500 mg/kg body weight compared to 200 mg/kg in the adult sheep. A summary of simple and multiple regressions of plasma glucose on insulin levels i s found i n Table 6. Manns and Boda (1967) have demonstrated that injected glucose causes sharp increases i n insulin secretion i n the lamb, and plasma glucose i s assumed to play a major role i n the regulation of insulin secretion i n the lamb. The results of this experiment i n general support this hypothesis. Table 6. The R squared expressed as a per cent, and significance associated with simple and multiple linear regressions of plasma glucose on plasma insulin levels in suckling lambs. Treatment No. animals df R squared % Pre-fasted (single) 4 19 48.50** Fed (single) 4 19 O.KP.s. Pre-fasted (twin) 4 19 42.16** Fed (-twin) 4 19 44.96** Pre-fasted (single + twin) 8 39 51.23** Fed (single + twin) 8 39 8.62 A l l lambs 16 79 49.50* ** P <.01; n.s. not significant When a large increase in plasma glucose level was induced by the rapid ingestion of milk, there was a simultaneous r i s e i n the plasma insulin l e v e l . I f the plasma glucose level regulates insulin secretion rate, and i f the plasma insulin level reflects the rate of insulin secretion, then plasma glucose and insulin levels would be expected to be significantly correlated. As seen i n Table 6, the regression of plasma glucose on plasma insulin level 1 63 yielded significant R squared values in 3 of the 4 groups studied. The reason for a lack of correlation between plasma glucose level and that of insulin for ad libitum suckled lambs i n the groups of single lambs is not immediately evident. I t may be due to the fact that when plasma insulin and glucose levels are f a i r l y constant, variation i n plasma levels is due to procedural variation i n their determination, rather than to a cause and effect relationship. When a relatively wide range of plasma values are measured, -the procedural variation in their measurement i s relat i v e l y much less important, and under these conditions a cause and effect type of relationship, such as between plasma glucose and insulin level, w i l l result in a significant correlation being measured between these blood metabolites. The results of this experiment indicate that the plasma glucose and insulin level i n the lamb may significantly correlate , as would be expected ' for a hormone and the blood metabolite which regulates that hormone's release. Under certain conditions, the plasma glucose and insulin levels were found not to be correlated, but this does not obviate a role for glucose in the- regulation of insulin secretion. I f a plasma metabolite regulates insulin secretion, there should be some conditions under which a significant correlation between the "regulatory" metabolite and insulin level in plasma could be measured. However, there w i l l be conditions under which they would not be correlated, and i t i s possible for a "non-regulatory" blood metabolite to be correlated to the plasma insulin l e v e l . Therefore, the measurement of the degree of correlation between the plasma level of insulin and a blood metabolite i s of limited value i n assessing the regulatory role of a blood metabolite i n insulin secretion. 64 Experiment . I l l Data from the previous experiments was used as a basis for the estimation of normal variation in plasma VFA and glucose concentrations. By experimentally altering the individual plasma propionate, butyrate and glucose levels within the estimated physiological range, while simultaneously measuring the plasma insulin levels, i t was anticipated that the relative importance of each of these metabolites i n the regulation of insulin secretion could be assessed. Experimental Procedure In the experiments that follow, collected blood samples were handled and stored, and plasma glucose, VFA and insulin were measured i n blood plasma as previously described (Experimental II, Part A). Plasma FFA was measured using the method of Patterson (1963). Infusions and blood samplings were made through a single long polyethylene tubing, and were f a c i l i t a t e d by a 3-way stopcock. The volume of the tubing and cannula was determined prior to an experiment. To obtain a blood s a m p l e t h e i n f u s i o n was s t o p p e d , t h e tubing and cannula cleared of infusate a n d contaminated b l o o d , t h e b l o o d sample drawn, the tubing and cannula r e f i l l e d with infusate, a n d t h e i n f u s i o n continued. With practice this operation could be performed i n 30-45 sec. Thus a l l of t h e i n f u s i o n s w e r e b r i e f l y interrupted d u r i n g b l o o d s a m p l i n g . A l l sheep to be infused were cannulated at 4 hr before the start of g an experiment with an intravenous placement catheter (placed in the external jugular vein). Double male adapters were attached to the cannula and a 2 m 9 length of polyethylene tubing extended from the cannula to a syringe containing the infusate. The syringe was mounted on a constant speed syringe 10 pump . 8 Jelco IV Catheter Placement Unit. Jelco Laboratories, Raritan, N.J. 08869. 9PE90.id. .86 mm; od. 1.27 mm. B.C. Stevens Co. Ltd. "^Sage syringe pump. Model 249-2. Sage Instruments Inc. 65 A l l sheep were held i n metabolism cages 4 hr prior to, and throughout the course of an experiment. Water was available at a l l times. Eight non-pregnant ewes were sampled for blood to determine the normal blood values of the metabolites studied. These animals received 1.36 kg of grass hay daily in 2 equal portions. Blood samples were collected at about 3 hr after the AM feeding. The blood values determined in the 8 fed ewes were compared with values obtained in ewes which had been fasted 48 hr to determine the effect of a short fast. Comparison of mean values was made using student's t test throughout Experimental III and IV. a) phlorizin Six mature non-pregnant ewes, receiving 1.36 kg of hay daily were infused at varying rates with phlorizin dihydrate. In 2 preliniinary experiments phlorizin was infused to determine an appropriate infusion rate. The infusion data i s found in Table 7. Plasma glucose and insulin were determined at zero Table 7. Data on phlorizin infusion to non-pregnant ewes Animal no. Weight Phlorizin Infusion Duration kg Rate mg/hr/kg3/4 min 101 63.6 9.1 injected 113 65.5 18.2 10 1 54.5 55.4 10 549 59.1 55.4 10 211 54.5 55.4 10 545 68.2 55.4 10 time and at 5, 10, 15, 20, 30, 45, 60, 90 and 120 min after the start of a phlorizin infusion. Plasma VFA and FFA were determined at zero time and at 15, 30, 60 and 120 min. 66 b) glucose Eight mature non-pregnant ewes were infused with glucose. A l l animals infused were prefasted 48 hr. and cannulated as described. Two infusion rates were used. The lowest infusion rate was the estimated difference in glucose entry rate between fed and fasted sheep (Bergman, 1963). The infusion data i s found i n Table 8. Blood samples were collected at the time of the last feeding, at zero time, and at 5, 10, 15, 20, 30, 45 and 60 min. Blood samples were analyzed for glucose, insulin and FFA. Table 8. Data on glucose infusions to non-pregnant fasted ewes. Animal no. Weight Infusion Rate Duration kg 3/4 ymole/min/kg min' 218 24.2 6.5 15 38 17.7 6.5 15 35 17.4 6.5 15 34 15.6 6.5 15 22 17.1 19.5 15 24 17.7 19.5 15 25 22.4 19.5 15 42 19.9 19.5 15 c) propionate and butyrate Sodium propionate was infused at a single rate to 4 non-pregnant mature ewes, while sodium butyrate was infused at 3 different rates using 4 ewes for each infusion rate. The intermediate butyrate infusion rate and the single propionate infusion rate were equivalent to 0.2 mmole/min for a 50 kg sheep. A l l sheep were fasted 48 hr prior to an experiment. Blood samples were collected at the time of the last feeding and at zero time, 67 5, 10, 15, 20, 30, 45 and 60 min. Plasma glucose, insulin and VFA were determined , but VFA determinations were not made at 5, 10 or 20 min. The infusion data i s found i n Table 9. Table 9. ' Data on VFA infusions to non-pregnant; fasted ewes. Animal no. Weight 3/4 kg Acid infused Infusion rate 3/4 ymole/min/kg Duration min. 30 19.4 propionate 10.8 30 32 17.1 propionate 10.8 30 37 19.9 propionate 10.8 30 39 23.2 propionate 10.8 30 14 24.4 butyrate 1.08 30 133 22.4 butyrate 1.08 30 136 17.7 butyrate 1.08 30 35 17.4 butyrate 1.08 30 28 15.9 butyrate 10.8 30 26 15.3 butyrate 10.8 30 29 21.3 butyrate 10.8 30 27 21.3 butyrate 10.8 30 7 17.4 butyrate 32.4 30 48 19.6 butyrate 32.4 30 114 17.1 butyrate 32.4 30 120 16.2 butyrate 32.4 30 Results a) phlorizin The results obtained with the phlorizin infusions are summarized i n Fig. 12 and Fig. 13. In the preliminary infusions the plasma glucose and insulin levels appeared to decline during the f i r s t 10 min of the phlorizin 68 Z _ 2 0 0 L • ^ 150 CO C lOOL UJ E S i 9 0 8 0 7 0 6 0 LU » 2 - 0 0 < ® 1 - 5 0 ° 0 0 ° I •SO < z o 4 0 0 3 0 O 0) o 5". IT > 03 8 0 6 0 4 0 2 0 ANIMAL RATF F E E D 0 o |0I 9lmg./hr/kg 3 "fearley • • 113 (8-2 " Barley-hay (|:|) \ ° o o o • N ISOBUTYRATE 3 0 6 0 9 0 TIME min 120 Fig. 12. Changes in plasma glucose, insulin and VFA following a phlorizin injection or infusion to wethers i n a preliminary study. 69 z _ ~j e co E Z 3 - =k 2 0 0 1 0 0 I 1 5 5 - 4 m g / h r / k g 3/4 Ul < < L L . r -Li_UJ O < LU H < & 2 0 0 § 1 5 0 | 1 0 0 E - 5 0 4 0 - 0 3 0 0 2 0 0 1 0 - 0 . O 0. s • Q . — \ U J E 2 0 0 . < 1 5 0 _ tr > 1 0 0 . 3 CO 5 0 9 © A C E T A T E . . F F A \ f-3 0 6 0 T I M E min 9 0 1 2 0 Fig. 13. Changes in jugular plasma glucose, insu l i n , FFA and VFA levels during the infusion of phlorizin to fed wethers (*P<.10; **P<.05). 70 infusion. The i n i t i a l plasma glucose and insulin levels were considerably elevated above the normal concentration range, possibly due to the animals being i n an excited state (Fig. 12). Although the plasma glucose and insulin levels appeared to decline, the depressed levels were s t i l l above the estimated normal values. The plasma VFA levels were essentially unchanged throughout the phlorizin infusion. Since there appeared to be a strong resistance to phlorizin-induced hypoglycemia i n fed sheep, the rate of phlorizin infusion was increased considerably and four additional sheep were infused. Again there was a steady decrease i n plasma glucose and insulin levels during the f i r s t 10 min of the infusion (Fig. 13). The plasma glucose depression was significant at 10 min , but the apparent insulin depression was not significant. The plasma insulin and glucose levels appeared to make a slight recovery toward the pre-infusion level between 10 and 30 min , followed by a further decline at 45 min. which was significantly below the zero time value for the plasma insulin l e v e l , but not for the plasma glucose leve l . During the last 60 min of the experimental period the plasma glucose and insulin levels appeared to return to the pre-infusion levels. The plasma acetate, propionate and butyrate levels did not significantly change during the course of this study, but there may have been a tendency for these acids to increase gradually throughout the study. b) glucose The results of glucose infusions at two different rates are indicated 3/4 i n Fig. 14 and 15, The infusion of glucose at 6.5 ymole/min/kg caused a rapid r i s e i n plasma glucose levels to values significantly above the pre-infusion levels. The blood glucose levels remained elevated throughout the infusion. Levels then declined to the pre-infusion level by 25 min after Fig. 14. Jugular plasma glucose, insulin and FFA levels during the infusion of glucose to non-lactating ewes at the lower infusion rate (**P<.01). 72 GLUCOSE INFUSION 19-5 jjmole/mm/kg3/4 Insulin min Fig. 15. Jugular plasma glucose and insulin levels during the infusion of glucose to non-lactating ewes at the higher infusion rate 0':P<.05). 73 the infusion end. Plasma insulin levels remained unchanged throughout the period of study even though the mean plasma glucose level was increased by 15 mg/lOOml to values slightly above the normal blood glucose range. Plasma FFA levels were measured i n animals infused at the lower glucose infusion rate. There was considerable variation in the plasma FFA estimates. The plasma FFA appeared to increase slightly during the 48 hr -fast, but this was not significant. There was a decline i n plasma FFA toward the end of the infusion which was significant at 30 min. 3/4 When glucose was infused at 19.5 ymole/min/kg , there was a similar, but more pronounced, increase i n plasma glucose than for the lower glucose infusion rate. Mean plasma glucose levels increased from about 66 to 98 mg/ 100 ml, but there was no significant change i n plasma insulin l e v e l . c) propionate and butyrate The results of the propionate infusion are presented i n Fig. 16. 3/4 Infusion of propionate at 10.8 ymole/min/kg resulted i n a rapid significant r i s e i n the mean plasma propionate le v e l from 15 ymole/liter to 100 ymole/liter at 5 min of the infusion. This was followed by a slower rate of increase to a mean iraximum level of about 225 ymole/liter at 30 min. There was no significant change i n the plasma glucose or insulin level accompanying the changes in plasma propionate. 3/4 The infusion of butyrate at 1.08 ymole/min/kg resulted i n a significant increase i n the plasma butyrate level from a mean of 5.1 to a maximum mean level of 49.7 ymole/liter at 30 min (Fig. 17). There was a gradual f a l l i n plasma butyrate back to the pre-infusion level by 60 min. There was no significant change i n the plasma glucose or insulin levels from the pre-infusion values. 3/4 The infusion of butyrate at 10.8 ymole/min/kg resulted i n a large 74 PROPIONATE IN FUSION IO-8 umole/min/kg" Fig. 16. Jugular plasma insulin, glucose and propionate levels associated with the infusion of propionate to non-lactating ewes.("P<.05; **P<.01). 75 B U T Y R A T E I N F U S I O N l08jjmole/min/kg' Fig. 17. Jugular plasma insulin, glucose and butyrate levels associated with the lowest rate of butyrate infusion to non-lactating ewes (**P<.01) 76 increase in plasma butyrate, from a mean pre-infusion level of 5.7 to a maximum of 167.7 ymole/liter at 30 min (Fig. 18). Plasma butyrate level decreased rapidly following the infusion to near the pre-infusion level by 60 min. There was a small but non-significant r i s e i n plasma insulin, from 8.8 to 13.8 yUnits/ml at 5 min after the start of the infusion, but no other change occurred in plasma insulin. There was no significant change in plasma glucose. 3/4 The infusion of butyrate at 33.2 ymole/min/kg resulted i n a significant r i s e in plasma butyrate from a pre-infusion mean lev e l of 5.5 to 262 ymole/liter by 5 min of the infusion (Fig. 19). A maximum mean plasma butyrate level of 334 ymole/liter was measured at 30 min. The increase in plasma butyrate at 5 min was associated with a significant increase in plasma insulin from 8.1 to 26.9 yUnits/ml, which was the maximum mean plasma insulin measured during the infusion period. The plasma insulin level was significantly elevated above the pre-infusion level by 2.5 min after the start of the infusion. The plasma insulin l e v e l remained significantly elevated above the pre-infusion l e v e l u n t i l the end of the butyrate infusion, at which time there was a rapid decline to near the pre-infusion l e v e l . There appeared to be a gradual decline i n the plasma glucose l e v e l , but this was not significant. The infusion of propionate or butyrate at the rates reported i n this study did not affect the plasma levels of the non-infused VFA. The 48 hr fast significantly affected the levels of certain of the plasma metabolites studied as seen in Table 10. Plasma insulin declined during a 48 hr fast as did the level of plasma acetate, propionate and butyrate. The lev e l of plasma glucose did not significantly change. 77 BUTYRATE INFUSION IO-8^ mole/min/kg' Fig. 18. Jugular plasma insulin, glucose and butyrate levels associated with the intermediate rate of butyrate infusion to non-lactating ewes (**P<.01) 78 BUTYRATE INFUSION Fig. 19. Jugular plasma insulin, glucose and butyrate levels associated with the highest rate of butyrate infusion to non-lactating ewes (*P<.05). 79 Table 10. Plasma VFA, glucose and insulin levels i n fed and 48 hr fasted non-pregnant ewes. Plasma metabolite F e d a Fasted b (mean!SEM) (meant SEM) glucose (mg/100 ml) 56.6+1.0 58.0±2.4 n , s' insulin (yUnit/ml) 17.Oil.1 10.4+1.0'* acetate (nroole/liter) .8551.120 ft* .350+.020 propionate (ymole/liter) 17.0±2.3 7.1+.9* butyrate (ymole/liter) 11.0+1.3 ft 5.4+1.0 * (P<.05); ** (P<.01); n.s. (non-significant) mean of 8 sheep b mean of 12 sheep Discussion Blood glucose levels have been demonstrated to be of major importance in regulating the rate of insulin release i n non-ruminant animals (Randle et a l , 1968; Steffens, 1970). Physiological variation i n plasma glucose levels has been shown to cause changes i n the rate :of pancreatic insulin synthesis and release (Kipnis, 1972). However, Manns and Boda (1967) f e l t that since plasma glucose levels are f a i r l y constant i n ruminants, some other circulating metabolite may perform a regulatory function with respect to insulin secretion. Phlorizin was infused in an attempt to determine i f decreases in plasma glucose levels of fed non-pregnent ewes would result i n changes i n the insulin secretion rate as reflected by the plasma insul i n levels. The effectiveness of phlorizin i n inducing hypoglycemia i n this study was 80 rather limited. In other published work i t has been found that phlorizin administered intravenously or intramuscularly causes glycosuria in fed or fasted ruminants. However, hypoglycemia was only successfully induced i f phlorizin adndnistration was coupled with prolonged starvation or semi-starvation and pregnancy (Goetch and Pritchard, 1958; Staubus et a l , , 1960). These results indicate that ruminant animals have a large gluconeogenic reserve which allows for the maintenance of blood glucose homeostasis during periods of greatly increased glucose demand. Horino et a l . (1968) have reported 'that during prolonged fasting in sheep there i s a decline i n the plasma glucose level of about 10 to 15 mg/100 ml which corresponds to a f a l l in the plasma insulin l e v e l . This result indicates that the digestion and absorption of feed i n some way enhances the rate of insulin secretion and suggests that the plasma glucose level may be responsible. Similar results have been found i n monogastric animals (Kipnis, 1972). In the studies reported here the infusion of phlorizin to fed sheep resulted i n a transient decline i n plasma glucose lev e l , and a subsequent drop i n plasma insul i n l e v e l . The time lag in the nadir of the plasma insulin level i n relation to the minimum glucose level may have related to the h a l f - l i f e of plasma insulin. These results suggest that i n the fed sheep the plasma glucose l e v e l , at least i n part, maintains a normal insulin secretion rate. However, other factors would appear to be involved, since a short fast resulted in a significant decline in plasma insulin level i n the absence of any change i n plasma glucose le v e l (Table 10). 81 Grey at a l . (1969) has demonstrated that hypoglycemia of fasting i n the rat is associated with glucose intolerance. It was also shown that small interperitoneal glucose injections during a fast w i l l result in normal glucose tolerance during fasting. It has been shown that the drop i n plasma glucose levels associated with the onset of fermentation, and the loss of an exogenous glucose source in developing ruminants, is associated with decreasing glucose tolerance (McCandless et a l . , 1950). Boda (1964) found that fasting reduced glucose tolerance i n sheep. The decrease i n glucose tolerance during fasting or with development in the ruminant appear to have a common basis. This i s in part due to a diminished insulin secretory response to a standard glucose dose (Manns and Boda, 1967; Boda , 1964). From the published results available i t appears that the plasma insulin level declines as the normal blood glucose le v e l decreases, and prolonged low blood glucose levels i n some way impair the insulin secretory response to glucose. Thus the lower glucose tolerance of adult versus immature ruminants may relate to the differences i n normal plasma glucose levels and the corresponding effect on the 3-islet c e l l s . Another factor of importance i n the reduced glucose tolerance of fasting animals and mature ruminants i s the reduced glucose phosphorylating capacity of the l i v e r (Ashmore and Weber, 1968; Ballard and Olver, 1965). This change may be mediated by the a v a i l a b i l i t y of insulin to the l i v e r . In the glucose infusion experiment described i n this report, the infusion of glucose at rates sufficient to cause a 30 mg/100 ml r i s e i n the plasma glucose level f a i l e d to affect the plasma insulin level. The sheep used i n this study were fasted 48 hr and possibly there was some impairment of the insulin secretary response to glucose. The r i s e in plasma insulin level i n response to injected glucose i n fed sheep,has been found 82 to be quite low (Manns et a l . , 1967). A 4-fold r i s e in plasma glucose level following the injection of glucose at 1.11 mmole/kg was associated with an approximate 2.5-fold increase in plasma insulin level. Studies have indicated that injected glucose presents a square-wave stimulus to the 6-cell, while a glucose infusion presents a slow-rising stimulus (Curry, 1971). This worker demonstrated that the normal physiological stimulus to insulin secretion i s a slew-rising stimulus, which is much less effective i n causing an increase i n plasma insulin levels than the square-wave stimulus. It was also demonstrated with rat pancreatic g-cells, that a square-wave stimulus causes 2 phases of insulin secretion. There was a transient but greatly increased rate of insulin secretion followed by a gradual r i s i n g sustained secretion rate. The slow-rising physiological stimulus does not appear to be effective i n stimulating the rapid release phase of insulin secretion. In a series of glucose infusions,Horino et a l . (1968) found that the infusion of glucose at 14 ymole/min/kg to sheep caused an increase i n plasma glucose level to a maximum level of about 120 mg/100 ml. This increase i n plasma glucose level was associated with a significant increase in plasma insulin from about 30 to 70 yUnits/ml. Higher rates of glucose infusion caused higher glucose maximums and correspondingly higher plasma insulin levels. The lower rate of glucose infusion selected for study i n this report (Fig. 14) was the difference between estimated glucose entry rates for fasted and fed non-pregnant sheep (Bergman, 1963). The higher glucose infusion rate was three-fold higher than the lower rate. The infusion of glucose at the lower rate caused an apparent physiological increase in the plasma glucose l e v e l , but there was no change in the plasma insul i n level 83 (Fig. 14). The infusion of glucose at the higher rate caused a significant . increase in plasma glucose to a mean value of about 100 mg/100 ml, but there was no significant change in the plasma insulin level. This l a t t e r rate of glucose infusion i s comparable to the lowest rate of glucose infusion u t i l i z e d by Horino et a l . (1968). However, the maximum in glucose level obtained by these workers was about 120 mg/100 ml, and was associated with a significant increase i n plasma insulin l e v e l . Together these results suggest that a slow-rising glucose stimulus for insulin secretion becomes effective between 100 and 120 mg/100 ml of blood glucose. I t i s doubtful that under physiological conditions the plasma glucose would increase to this level following feeding. Therefore the increase i n plasma insulin following feeding in ruminants,that has been observed, does not appear to be caused by an increase i n plasma glucose. An increase i n plasma insulin following the intravenous infusion of glucose to sheep would appear to be a non-physiological effect. Since i n ruminants there is normally l i t t l e glucose absorBed from the gut, and since plasma glucose levels appear to be relatively constant, i t appears l i k e l y that i n fed ruminants there is a relatively constant endogenous glucose production. This i s also suggested by the finding that plasma glucose concentration is directly related to the plasma glucose entry rate (Bartley and Black, 1966; Annison and White, 1961). If exogenous glucose is provided intravenously to the ruminant i t i s possible to induce a monogastric-like metabolism. The intravenous infusion of glucose has been shown to result i n an increased glucose oxidation (Annison et a l . , 1961). Similar results have been obtained following the intraduodenal infusion of glucose to cows (Bartley et a l . , 19 66). In the lat t e r report up to 65% of the glucose infused was oxidized. There i s 84 also evidence that the u t i l i z a t i o n of glucose for lipogenesis can be increased by increasing the supply of glucose to the glucose pool (Lindsay, 1970). It has also been demonstrated that enzyme levels in adipose tissue of lactating cows fed maize were significantly increased so as to enhance lipogenesis from glucose (Lindsay, 1970). Certain types of rations may result in considerable amounts of glucose being absorbed from the small intestine, and i n an increase in the supply of exogenous glucose. Although the digestion of cellulose, hemicelluloses and soluble sugars i s essentially completed i n the rumen, the situation with dietary starch may be different. On hay type rations only a very small amount of starch escapes fermentation, at most a few g/day. For certain high energy feed, particularly barley or corn, considerable amounts of dietary starch may reach the small intestine. Wright, Grainger and Marco (1966) found high concentrations of starch i n the abomasum, and Tucker, Mitchell and L i t t l e (1968), studying digestion i n sheep, found that starch reaching the intestine may exceed 30% of that consumed, when cracked or ground maize was the principle dietary constituent. These workers had reported similar results ea r l i e r in steers (Karr, L i t t l e and Mitchell, 1966). However, these results appear to grossly over-estimate the amount of starch normally reaching the small intestine on high starch rations. Apart from these early high estimates of the extent of dietary starch reaching the small intestine, numerous other workers using enzymic methods for starch determination, obtained estimates of the percent dietary starch reaching the duodenum of from 3.0 to 12.0% (MacRae and Armstrong, 1969; Nicholson and Sutton, 1969; Orskov and Fraser, 1968; Sutton and Nicholson, 1968; Topps, Kay and Goodall, 1968; Topps, Whitelaw and Reid, 1968; MacRae and Armstrong, 1966). The average estimate of starch escaping fermentation 85 i n these studies was about 6.0%. Most of these studies were carried out on • animals consuming high levels of barley or maize. These studies indicate that as much as 20-30 g/day of starch reaches the small intestine in sheep fed certain high-starch rations. In cattle, the figure appears to be about 50 g/day. Therefore, most of the plasma glucose entry arises by gluconeogenesis even on high starch rations, since t o t a l glucose turnover in sheep has been estimated at about 100 g/day (Bergman, 1963). However, more starch reaches the intestine on high-starch as opposed to high-roughage feeds. Manns et a l . (1967) f i r s t demonstrated the insulinotropic action of propionate and butyrate in sheep and this was confirmed by Horino et a l . (1968). These workers concluded that propionate and butyrate are probably involved i n the normal regulation of insulin secretion in ruminants. A number of studies have attempted to evaluate the physiological significance of the insulinotropic action of propionate and butyrate. B e l l et a l . (1970) developed a method of transplanting a part of the pancreas into an exteriorized carotid artery-jugular vein loop i n the neck of sheep. Studies with this transplant revealed a consistent insulin secretory response following the infusion of butyrate at 2 yM/min into the blood supply of the transplant. The authors estimated an a r t e r i a l butyrate concentration of lOOyM, assuming the infusate mixed thoroughly within the blood approaching the transplant. Since normal butyrate levels i n peripheral blood of sheep have been reported to be in the range 10-50 yM (Manns et a l . , 1967), these workers concluded that butyrate is an important stimulus to insulin secretion in sheep under natural conditions. However, lower rates of butyrate infusion were ineffective, and the concentrations of butyrate i n blood reaching the transplant may have been different from the estimated levels. Also the normal 86 flow of blood to the pancreas i s about 1-2% of the cardiac output (Smith and Hamlin, 1970). This infers that at the lowest effective butyrate infusion rate, butyrate would be entering the general circulation at between 6.7 and 13.4 g/day which, as w i l l be discussed, i s probably above physiological butyrate entry rates. Bartos et a l . (1970) attempted to measure the threshold concentrations of propionate, butyrate and glucose i n blood, which can stimulate an insulin secretory response in goats. The level of propionate or butyrate in an infusate was gradually increased from 0 to 50 yM/min/kg within 60 min , and the plasma insulin and VFA level measured during an infusion. The authors concluded from the results that plasma insulin levels increased when the plasma propionate level was 74.9 ymole/liter or the butyrate level 45.0 ymole/liter. However, the results presented in this paper suggest that this is a particularly tenuous conclusion. Although there was no s t a t i s t i c a l analysis of the data, i t would appear that at 10 min of the propionate infusion, when plasma propionate was 74.9 ymole/liter, there was a small non-significant r i s e in plasma insulin. Only at 20 min of the propionate infusion, when plasma propionate averaged 180 ymole/liter, was there a large r i s e in plasma insulin. When the plasma butyrate level reached 45 ymole/liter, there was a small but non-significant increase i n plasma insulin. Only when the plasma butyrate level was above 200 ymole/liter was there a large increase in plasma insu l i n . Although these workers concluded that the secretion of insulin i s controlled mainly by propionic and butyric acids, the threshold levels for propionate and butyrate stimulated insulin release reported i n this paper would appear to be very crude estimates and subject to considerable error. Stem et a l . (1970) studied the effect of experimentally increasing the ruminal propionate and butyrate concentrations on plasma insulin levels i n 87 fasted goats. It was found that the injection of 5 times the hourly propionate production or 12 times the hourly butyrate production did not affeet•plasma insulin levels. Larger doses caused increases i n the jugular vein plasma insulin levels. I t was concluded that propionate and butyrate absorbed from the rumen do not normally influence the secretion of insulin in goats. A recent study by Bassett (1972) revealed that VFA mixtures infused into the rumen of sheep at rates equivalent to normal rumen production had no effect on plasma insulin or glucagon lev e l . It was.concluded that propionate and butyrate do not normally stimulate insulin or glucagon release in sheep. Propionate and butyrate infusions used in the studies reported here were designed to elevate the plasma propionate and butyrate levels of fasted sheep to near the normal fed levels. However, the entry rate of propionate and butyrate into the peripheral circulation of ruminants i s rather imprecisely known. As a result the infusion rates selected were somewhat arbitrary. The lowest rate of butyrate infusion was several fol d lower than that used in any other study. Many of the e a r l i e r infusions used were probably too high to be considered physiological. Estimates of ruminal production rates for propionate and butyrate are presented i n Table 11. Also in this table are some estimates of propionate and butyrate portal vein entry rates as well as the single estimate available of the peripheral plasma propionate and butyrate entry rates. Generally, these values were calculated by extrapolating estimates made over a 2 to 4 hr period to 24 hr for purposes of comparison. As a result these estimates probably over-estimate the actual 24 hr production and entry rates. Manns and Boda (1967) and Horino et a l . (1968) infused propionate and 88 butyrate at rates equivalent to the estimated portal entry rates of Annison et a l . (1962). A consideration of the data in Table 11 suggests that infusion rates which caused an insulin secretory response were probably non-phy siological. With respect to the estimate of Annison et a l . (1962)„of portal vein entry of propionate and butyrate, the determination was made using carbon-14 Table 11. Comparison of ruminal, portal and peripheral propionate and butyrate entry rates with infusion rates employed i n insulin secretion studies. Author Feed Ruminal Production Rates Faichney (1968) Bergman and Wol Weller et a l . (1967) propionate butyrate Wheat (500g)Lucerne (lOOg) 3770 g / 2 4 h r 3 4 ~ 4 Lucerne Pellets (lOOg) Wo f (1971) A l f a l f a Pellets (800g) Lucerne (600g) Hay (400g) 44.4 66.6 66.6 31.8 51.6 68.8 Portal Vein Entry Rate propionate butyrate g/24hr Annison et a l . (1962) 900g 239.7 49.5 Bergman and Wolf (1971) A l f a l f a Pellets (800g) 32.6 4.3 Peripheral Entry Rate propionate butyrate Bergman and Wolf (1971) A l f a l f a Pellets (800g) 6.4 g m h r^ Infusion Rates Route propionate effect 1 1 butyrate effect' (g/24hr) (g/24hr) Manns and Boda C1967) jugular 21.3 + 24.8 ± jugular 42.6 + 49.6 + portal 21.3 — 24.8 Horino et a l . (1968) jugular 106.6 + 123.8 + Study of this thesis jugular 20.6 — 73.9 + jugular 24.1 jugular 2.4 E f f e c t of infusion on plasma insulin levels (+) increase, (-) unchanged, (±) uncertain 89 labelled acids infused with sufficient carrier so as to raise the plasma propionate or butyrate levels sufficiently to allow specific activity, determinations. As seen in Table 11, the estimated rate of propionate entry was considerably greater than the estimated normal ruminal production rate, indicating that this estimate was probably too high. Also the estimated entry for butyrate was probably over-estimated, since this was about the same as the ruminal production rate. Absorbed butyrate is largely metabolized i n the ruminal epithelium (Pennington, 1952), and i t i s very unlikely that the portal vein entry rate for butyrate would approach the rate of butyrate production i n the rumen. The l i v e r also acts to prevent the entry of propionate and butyrate into the peripheral circulation as seen by the data of Bergman and Wolf (1971). These determinations were made by measuring arterio-venous differences i n conjunction with blood flow rates. Changes in plasma butyrate level following butyrate infusion and associated plasma insulin levels are readily compared i n Fig. 20 for the results of this thesis. The estimated upper limit for the normal plasma butyrate level i s indicated by the broken horizontal line at 50 ymole/liter. An increase in plasma butyrate following butyrate infusion to a level near the estimated upper limit of the normal range was associated with no significant change i n the plasma insulin l e v e l . Similarly, with the second infusion rate, the mean plasma butyrate level reached a maximum i n excess of 3 times the estimated normal plasma maximum, yet there was no significant increase i n the plasma insulin level. At the highest rate of butyrate infusion, the plasma butyrate level promptly rose at 5 min to about 5 times the estimated normal maximum. The increase i n plasma butyrate was associated with a rapid 3-fold significant increase in 90 BUTYRATE INFUSIONS 3 . A - 108 umole/min/kg • - 10-8 " • - 33-2 " Fig. 20. Comparison of plasma insulin and butyrate levels for the different rates of butyrate infusion. The normal plasma butyrate level i s indicated by the checkered area,(*P<.05; **P<.01). 91 the plasma insulin l e v e l . I t appears, therefore, as though the threshold for butyrate stimulated insulin release i s between 175 and 250 ymole/liter, which i s considerably above the upper limit for the normal plasma butyrate range. The single propionate infusion resulted in a maximum mean plasma propionate l e v e l of about 225 ymole/liter without associated changes in the plasma insulin level. Thus the threshold for propionate stimulated insulin release i s also greatly in excess of the upper limit of the normal plasma propionate range. Plasma glucose levels appeared to decline s l i g h t l y during the infusion of butyrate at 33.2 ymole/min (Fig. 19). However, this decrease in plasma glucose did not reach significance at the 5% l e v e l . In other butyrate and propionate infusions of similar magnitude to the highest butyrate infusion used in this study, i t was found that the plasma glucose level remained constant even though there were large increases in plasma insulin (Horino et a l . , 1968; Manns and Boda, 1967). In earl i e r studies, i t was demonstrated that the injection of large amounts of butyrate into sheep caused marked hyperglycemia (Ash et a l . , 1964). This was demonstrated to be mediated by the pancreas (Phillips et a l . , 1969; P h i l l i p s , 1966), and was suggested to be the result of an increased glucagon secretion. It has also been demonstrated that large amounts of butyrate introduced into the rumen caused hypoglycemia (Clark and Malan, 1956; Schultz and Smith, 1951; Schultz,Smith and Lardy, 1949). Precisely how this effect is mediated has not been established, but i t has been suggested that ketone bodies arising from butyrate metabolism within the ruminal epithelium may stimulate insulin release (Menahan et a l . , 1966a). Bassett(1972) demonstrated that intravenous infusion of propionate and butyrate simultaneously stimulate insulin and glucagon release in the sheep. 9 2 Therefore i t appears that propionate or butyrate infused or injected i n large amounts stimulates both glucagon and insulin release, but the glycogenolytic action of glucagon prevails over the hypoglycemic action of insul i n , and hyperglycemia ensues. With lower rates of intravenous propionate or butyrate administration, there is a balance between the actions of the two hormones released and the plasma glucose level remains unchanged. At s t i l l lower rates of acid administration within the physiological range, neither ins u l i n nor glucagon release is enhanced, and plasma glucose levels remain constant. 93 General Discussion The results of the experiments reported here, in agreement with earlier work, indicate that post-prandial glucose and insulin levels i n plasma of lambs change in a similar manner to non-ruminant animals. In adult sheep, these changes are much less pronounced. It was also" found that the type of feed consumed affects the insulin secretory response following feeding. Significant changes i n the plasma insulin l e v e l only occurred following feeding, in sheep consuming barley. In sheep consuming only hay, l i t t l e or no increase i n the plasma insulin level occurred following feeding. This l a t t e r finding may relate to the extent of glucose absorption from the gut. Judson et a l . (1968) found that although ruminal VFA production rates were about the same, increasing the proportion of concentrate fed to sheep resulted in decreased gluconeogenesis from propionate. The authors suggested that sheep consuming concentrate feeds may have a greater absorption of glucose from the gut, which indirectly through increased insulin secretion tends to inhibit gluconeogenesis. High concentrate rations have been demonstrated to be associated with increased plasma glucose levels (Bensadoun et a l . , 1963; Jorgensen and Schultz, 1963). The increase i n plasma insul i n level that was found to occur in sheep following the feeding of ooncentrate-type feeds has also been reported by Bassett (1972). A p a r a l l e l r i s e in plasma glucagon was also detected by this worker. To assess the physiological significance of post-prandial increases i n plasma insulin and glucagon, following feeding in ruminants, i t i s of value to consider the role of these hormones with respect to hepatic and extra-hepatiic metabolism. Ashmore and Weber (1968), i n a discussion of the hepatic effects of 94 of glucagon and insulin, indicate that glucagon stimulates gluconeogenesis t t by activating adenyl cyclase and increasing the level of 3 ,5 -c y c l i c AMP. Cyclic AMP appears to activate a tissue lipase which increases intracellular levels of FFA, acyl-, and acetyl-CoA esters. Acetyl-CoA has been demonstrated to be an essential cofactor for pyruvate carboxylase (E.C. 6.4.1.1.), and an increased level of acetyl-CoA would be expected to promote conversion of pyruvate to oxaloacetate. Also the increased levels of FFA appear to in h i b i t three regulatory enzymes of glycolysis, three of the pentose phosphate pathway, and three of the Kreb's cycle. Moreover, acyl-CoA has been demonstrated to inhibit citrate synthase (E.C. 4.1.3.7.), which would also act to decrease Kreb's cycle a c t i v i t y . Production of NADH through fatty acid oxidation provides fuel to respiratory chain phosphorylation. As a consequence, the ATP/ADP rat i o i n the c e l l increases, which results i n ATP inhibition of citrate synthase (E.C. 4.1.3.7.) and phosphofructokinase (E.C. 2.7.1.11), which acts to reinforce FFA :Lnhibition of these enzymes as well as to provide cofactor for pyruvate carboxylase (E.C. 6.4.1.1.). The increase in NADH/NAD+ r a t i o which probably accompanies fatty acid oxidation may act to further inhibit pyruvate kinase (E.C. 2.7.1.40) and phosphophructokinase (E.C. 2.7.11). Therefore, glucagon acts on the l i v e r not only to stimulate phosphorylase (E. C. 2.4.1.1.) acti v i t y and subsequently glycogenolysis, but also to decrease the catabolism of glucose through the pentose phosphate pathway, glycolysis, and the Kreb's cycle as well as decreasing lipogenesis from glucose and stimulating pyruvate carboxylase (E.C. 6.4.1.1.). A l l of the effects of glucagon on the l i v e r involve acute or rapid adaptations since the changes are caused largely by al l o s t e r i c effects. Whereas glucagon acts to cause only acute adaptive changes i n h e p a t i c metabolism, insulin has been shown t o cause both acute and c h r o n i c adaptive 95 changes. Insulin has been shown to lower the intracellular cyclic-AMP level... This should act to diminish hepatic lipase a c t i v i t y , reduce l i p o l y s i s , and increase the u t i l i z a t i o n of glucose through glycolysis. As a result, a-glycerophosphate formation and FFA esterification- would be increased, the FFA level decreases, and the enzymes.of the pentose phosphate pathway, the Kreb's cycle, and glycolysis, which are ijrhibited by FFA, would be reactivated. Also the inhibitory effects of acyl-CoA on lipogenesis would be largely removed, and the stimulatory effect of acetyl-CoA on pyruvate carboxylase (E.C. 6.M-.1.1) would be removed, since acyl-CoA and acetyl-CoA .".i levels would decline as the FFA lev e l decreased. The increased phosphorylation of glucose would lower the ATP/ADP r a t i o while the NADH/NAD+ r a t i o would decrease with decreased g-oxidation of fatty acids. These changes in cofactor concentrations would further act to enhance glycolysis and the Kreb's cycle as well as to oppose gluconeogenesis. Insulin has a more sustained effect on hepatic metabolism than glucagon, in that i t acts to induce the synthesis of certain key glycolytic enzymes as well as to inhibit the synthesis of key gluconeogenic enzymes. Insulin also acts to enhance the synthesis of certain key enzymes of lipogenesis, glycogenesis and of the pentose phosphate pathway. Apart from the actions on hepatic metabolism, insulin acts to f a c i l i t a t e the uptake and metabolism of glucose by muscle and adipose'tissue. The action of insulin on adipose tissue results i n a decreased l i p o l y s i s and increased FFA esterification which i s reflected i n decreased plasma FFA levels. In the fasted non-ruminant animal the metabolism of the l i v e r suggests a relatively low rate of insulin secretion and a high rate of glucagon 96 secretion, since hepatic gluconeogenesis and li p o l y s i s are the predominant metabolic a c t i v i t i e s . However, following a meal, gluconeogenesis is inhibited, the plasma FFA level f a l l s , while glycogenesis, glycolysis and glucose oxidation greatly increase. These alterations in hepatic metabolism following a meal in the non-ruminant animal suggest that the effects of insulin on hepatic metabolism predominate at this time, and this i s in accord with the observed increase i n insulin secretion. Similarly, post-prandial hepatic metabolism i n ruminants provides an indication of the possible changes in insulin and glucagon secretion rate. In contrast to monogastric animals, there i s an increase i n gluconeogenesis. following feeding, and an increased hepatic glucose output. These changes in hepatic metabolism occur even though there i s an increased secretion of insulin in some cases, which would tend to oppose these changes. A number of hepatic enzyme differences have been detected between ruminants and non-ruminant animals which may explain these findings. Ballard and Oliver (1965) have found that adult sheep l i v e r does not contain glucokinase ( E.C. 2.7.1.2.), and the a c t i v i t y of the hexokinase (E.C. 2.7.1.1.) in l i v e r from adult sheep i s much lower than the ac t i v i t y of glucokinase plus hexokinase i n l i v e r from adult rats. I t was also found that the glycogen content and the a c t i v i t i e s of UDP-glucose-glycogen glucosyl transferase (E.C. 2.4.1.1.) and UDP-glucose pyrophosphorylase (E.C. 2.7.7.9.) in l i v e r from adult sheep were less than i n l i v e r from young lambs. Since these enzymes are a l l induced by insulin, i t suggests that the overall rate of insulin secretion i s probably less in ruminants than non-ruminant animals. Accordingly, the ruminant l i v e r has a greatly reduced capacity to remove glucose from the blood or to store glucose as glycogen. Two other enzymes which are induced by insulin but which are found to 97 have very low a c t i v i t i e s in the ruminant l i v e r are ATP-citrate lyase ( E.C. 4.1.3.6.) and NADP-malate dehydrogenase (E.C. 1.1.1.38) (Hanson and Ballard, 1967). These enzymes are of importance in fat synthesis from glucose, and ruminants have a markedly reduced a b i l i t y to u t i l i z e glucose for lipogenesis. These results also suggest a lower secretion rate of insulin i n ruminants than non-ruminant animals. This is further supported by the finding that the prolonged intravenous infusion of glucose to sheep results i n a greatly reduced hepatic glucose output (Annison and White, 1961). Also, ruminants consuming rations high in starch have been found to have increased levels of ATP-citrate lyase (E.C. 4.1.3.6.) and NADP-malate dehydrogenase (E.C. 1.1.1.38) (Lindsay, 1970). It has been proposed that propionate and butyrate are involved i n the normal regulation of insulin secretion i n ruminants. Although the insulinotropic action of propionate and butyrate i s well established, the physiological significance has not been established. In the studies of this thesis, the insulinotropic action of butyrate was demonstrated, but the experimental elevation of plasma propionate or butyrate, within the physiological range, f a i l e d to stimulate insulin secretion as indicated by peripheral plasma insulin levels. Plasma insulin was demonstrated to increase in sheep fed concentrate-type feeds, but the cause of this r i s e i n plasma insulin level was not established. Bassett (1972) measured a r i s e in plasma insulin and glucagon i n sheep fed a concentrate-type feed. How such increases i n plasma insulin and glucagon level affect metabolism is uncertain. However, the post-prandial r i s e in plasma insulin following feeding in the ruminant does not appear to greatly affect metabolism in the l i v e r , since gluconeogenesis i s greatly increased at this time (Katz and Bergman, 1969). 98 The increased gluconeogenesis after feeding i n ruminants may be f a c i l i t a t e d by the increased rate of glucagon secretion; however, the increased levels of acetyl-, propionyl-, and butyryl-CoA would be expected to activate pyruvate carboxylase (E.C. 6.4.1.1.) and gluconeogenesis (Keech and Utter, 1963). I f propionate and butyrate normally stimulated insulin secretion, this would act to antagonize their direct action on gluconeogenesis. In the studies of this thesis, quite large increases i n plasma glucose levels following intravenous.glucose infusion, were associated with unchanged plasma insulin levels. However, the normal between-meal plasma insulin level may be maintained by the level of plasma glucose, since phlorizin-induced hypoglycemia was associated with a decrease i n the plasma insulin l e v e l . The results of this work, and others, tends to rule out the p o s s i b i l i t y of VFA or glucose normally stimulating insulin secretion after feeding of concentrate rations in ruminants. Certain amino acids have been demonstrated to stimulate insulin and/or glucagon secretion (Kipnis, 1972; Cherrington and Vranic, 1971; Muller, 1971; Unger et a l . , 1970). However, the p o s s i b i l i t y of amino acids having a role i n insulin •and/or glucagon secretion in ruminants remains to be evaluated. Gastrointestinal hormones have also been demonstrated to stimulate insulin and glucagon secretion in monogastric animals (Dupre e t - a l . , 1967; Unger et a l . , 1966). Baile et a l . (1969), on the basis of their studies, concluded that gastrointestinal hormones are unimportant i n the stimulation of insulin secretion i n goats, since pancreatic exocrine secretion did not appear to change after feeding. However, Bassett (1972), on the basis of evidence obtained with sheep, concluded that gastrointestinal hormones may play the major role i n the stimulation of insulin and glucagon secretion-99 following feeding i n sheep. In summary, ruminant animals receiving hay as a feed, have a very restricted exogenous glucose source, and the a v a i l a b i l i t y of glucose to the metabolic pool i s largely restricted by the rate at which gluconeogenesis proceeds. Glucose is normally required almost exclusively in certain tissues, such as erythrocytes and neural tissue as a metabolic f u e l , and in a l l tissues to a lesser extent as a precursor of many biological compounds such as amino sugars, uronic acids, and pentoses, to provide the necessary building blocks for the biosynthesis of essential c e l l components. A number of metabolic adaptations are made i n ruminants to ensure a metabolic glucose supply. These adaptations are mediated i n part by the substrates available to tissues, particularly the VFA, and also by a decrease in the secretion of insulin and possibly glucagon. When high-starch rations are fed to ruminants there are increases i n the rate of insulin and glucagon secretion, and the overall effect appears to be a slight inhibition of the normal increase i n gluconeogenesis which follows feeding, and an increased u t i l i z a t i o n of glucose in oxidative metabolism. The mediator(s) of the increased insulin and glucagon secretion on high-starch feeds does not appear to be glucose or VFA. Certain gastrointestinal hormones or amino acids may be responsible but evidence i s lacking. Evidence obtained in the studies of this thesis suggests that ruminants receiving only hay have plasma insulin levels that may re f l e c t a basal insulin secretion rate plus an increment dependent upon the normal level of plasma glucose. 100 Experiment IV The plasma levels of g-hydroxybutyrate and acetoacetate may increase quite markedly in ruminants under certain conditions associated with hypoglycemia (Berman, 1971). Bergman (1963) earl i e r demonstrated that acetoacetate induced hypoglycemia i n sheep. The effect of ketone bodies on insulin secretion i n ruminants has not been thoroughly studied. The following study was designed to test the insulinotropic action of g-hydroxybutyrate and acetoacetate. Experimental•Procedure Four non-pregnant ewes were fasted 48 hr prior to an experiment. Each ewe was f i t t e d with a cannula and a length of polyethviene tubing as described i n Experimental III, and held in metabolism cages. T o minimize excitement, animals were prepared about 4 hr before the experiment. DI^ g-hydroxybutyrate i n aqueous solution (11.5g/100ml) of pH 7.35 was injected at a rate of 75 mg/ 3/4 kg . Twenty ml blood samples were taken prior to the injection and at 5, 10, 15, 20 and 30 min post-injection. Thirty min after the last blood sample was collected following the 3/4 g-hydroxybutyrate injection, acetoacetate was injected at 75 mg/kg i n a solution (16.8g/100 ml) of pH 7.35. Blood samples were collected as for the g-hydroxybutyrate injection. Blood glucose and insulin were analyzed by the methods used i n Experimental 11, FFA were determined using the method of Patterson (1963), acetone and g-hydroxybutyrate by the method of Peden (1964), and acetoacetate by the method of Walker (1954). Standard acetoacetate solutions were prepared from acetoacetic anhydride by adding 1,2 N HC1 to the anhydride i n the proportion 1 anhydride: 4.5 HC1 101 and allowing the reaction to proceed at 0C for about 48 hr. The exact concentration of acetoacetate solutions was determined by decarboxylating the acetoacetate within a Warburg apparatus and measuring the carbon dioxide evolved (Edson, 1935). Results 3/4 The injection of 75 mg/kg of B-hydroxybutyrate to pre-fasted sheep caused a significant increase i n plasma B-hydroxybutyrate lev e l . There was no significant change in the plasma insulin, glucose, or FFA levels following this injection (Fig. 21), nor was there a significant change i n the plasma insulin, glucose, FFA or B-hydroxybutyrate associated with the pre-fast. The injection of B-hydroxybutyrate did not significantly affect the plasma acetone or acetoacetate levels (Fig. 22), and these levels were also unaffected by the pre-fast. 3/4 The injection of acetoacetate at 75 mg/kg caused a significant, although transient, increase in plasma acetoacetate level from .72 to 3.69 mg/100 ml at 5 min after the injection (Fig. 23). There was also a significant increase in the plasma insulin level from 16.0 to 23.8 yUnits/ml. However, no significant change occurred i n the plasma glucose or FFA for the same period, or i n the plasma insul i n , glucose, FFA or acetoacetate associated with the 48 hr pre-fast. The injection of acetoacetate caused significant increases i n the plasma acetone and B-hydroxybutyrate (Fig. 24). The plasma acetone level significantly increased from 1.89 to 3.87 and that of B-hydroxybutyrate from 10.0 to 17.0 mg/100 ml. Discussion The ketone bodies, acetoacetate and B-hydroxybutyrate, have been reported to have metabolic effects, in both ruminant and non-ruminant animals. Certain 102 3-hydroxy butyrate Injection mm Fig. 21. Changes i n jugular plasma insulin, glucose, FFA and 3-hydrcxybutyrate following the intravenous injection of 3-hydroxybutyrate to non-lactating ewes (**P<.Q1). 103 3-Hydroxybutyrote Injection 4 0 L 2 0 - H E 4 0 o o o . 2 0 2 0 10 r 75 m g / k g 3 /4 - L / L -48117^ 0 Acetone Acetoacetote 3-hydroxybutyrate -+ H J L J L 10 min 20 30 Fig. 22. Jugular plasma acetone, acetoacetate and 3-hydroxybutyrate levels following the 3-hydroxybutyrate injection (**P< .01). ion of the effects of ketone bodies on metabolism have been of interest to the runiinant physiologist because of the relevance of these effects to the ruminant metabolic disorders of acetonemia and pregnancy toxaemia (Kronfeld, 1970). These conditions are characterized by hyperketonemia and hypoglycemia, and i n this regard a number of workers have demonstrated that acetoacetate and g-hydroxybutyrate have insulinotropic properties i n monogastric animals (Madison et a l . , 1964; Mebane and Madison, 1964; Balasse et a l . , 1970). Related studies have also demonstrated a hypoglycemic and hypolipemic effect of ketone bodies infused to monogastric animals (Felig and Wahren, 1971; Pi-Sunyer et a l . , 1970; Mebane and Madison, 1964; Mebane and Madison, 1962). However, i t is not certain that insulin mediates these changes i n plasma glucose and FFA levels. Studies i n ruminants have indicated that ketone bodies have similar effects to those found i n non-ruminant animals, which suggests that ketone bodies could have etiologic significance i n the development of ruminant ketosis. I t has been demonstrated that increases i n plasma ketone bodies i n fasted ruminants depresses the plasma FFA le v e l (Mehahan, Schultz and Hoekstra, 1966a). However, acetone has been demonstrated to be ineffective i n this regard (Luthman and Jonson, 1968). The level of plasma FFA has been found to be lower i n cows with ketosis than i n fasting cows (Baird, Heitzman and Hibbitt, 1972), and suggests that ketone bodies may act to li m i t the rate of their own production by stimulating insulin secretion. Bergman, Kon and Katz (1963) demonstrated that the infusion of large amounts of acetoacetate caused a gradual development of hypoglycemia i n sheep. The development of hypoglycemia was accompanied by the onset of c l i n i c a l signs of ketosis such as drowsiness, stupor and occassional hyperesthesia. Fig. 23. Changes in jugular plasma insulin, glucose, FFA and acetoacetate levels following the intravenous injection of acetoacetate to non-lactating ewes (*P<.05; * * P < . 0 1 ) . 106 Acetoacetate Injection 4 0 2 0 E 4 0 o o |-20 20 10 _3/4 _L -48 hr 0 Acetone Acetoacetate 3-hydroxybutyrate J I I i ' t 10 . 20 mm 30 Fig. 24. Jugular plasma acetone, acetoacetate and 3-hydroxybutyrate levels following the acetoacetate injection (*P<.05; **P<.01) . 107 More recent studies i n sheep demonstrated that 3-hydroxybutyrate did not affect the plasma insulin level (Manns and Boda, 1967), while Horino et. a l . .(1968) reported that the injection of 3-hydroxybutyrate or acetoacetate to sheep did not result i n an insulin secretory response. Following the injection of 6-hydroxybutyrate to fasted sheep i n the work reported here, there was no change in plasma insulin level and no significant change in plasma glucose or FFA. In previous studies with B-hydroxybutyrate infusions, a decrease i n plasma FFA levels was recorded (Leng and West, 1969). The injection of acetoacetate caused a significant r i s e in plasma insulin level measured at 5 min after the injection (Fig. 23). This result may have differed from that of Horino et a l . (1968), i n that these workers measured the plasma insulin levels at about 10 min following an acetoacetate injection. The work described here shows no change in the level of plasma glucose or FFA. These results may d i f f e r from the findings of others, because of differences i n the levels of ketone bodies administered. As seen i n Fig. 25, the injection of 3-hydroxybutyrate into 48 hr fasted sheep was followed by an increased t o t a l ketone body level at 5 min of about 12 mg/100 ml. This increase i n ketone body lev e l was almost exclusively the result of an increased plasma 3-hydroxybutyrate l e v e l . Injecting acetoacetate under the same conditions and at the same rate resulted i n an increased t o t a l plasma ketone body level of about 12 mg/100 ml (Fig. 25). However, following the acetoacetate injection there were increases i n the levels of plasma acetoacetate, 3-hydroxybutyrate and acetone. Bergman et a l . (1963) previously reported that when acetoacetate was infused at high rates into sheep the acetoacetate was largely converted to 3-hydroxybutyrate, although part was converted to acetone. 108 A semilogarithmic plot of acetoacetate concentration against time was curvilinear. When the acetoacetate concentration was extrapolated to zero time, the acetoacetate concentration was found to be about 6.7 mg/100 ml. This result suggests that the injected acetoacetate was very rapidly reduced to B-hydroxybutyrate. On the basis of the h a l f - l i f e for plasma insulin of from 15 to'30 min (Trenkle,1970; Frohman, 1969), i t appears as though the plasma insulin levels did not r i s e to levels much above the values reported at 5 min after the injection. The conversion of acetoacetate to B-hydroxybutyrate has been demonstrated with sheep heart-muscle (Kulka et a l . , 1961), and Leng and Annison (1964) demonstrated the u t i l i z a t i o n of B-hydroxybutyrate by most tissues of the sheep. More recently, the enzyme B-hydroxybutyrate dehydrogenase (E.C. 1.1.1.30) has been demonstrated to be active in numerous tissues of the sheep. However, one of the most striking characteristics of ketogenic tissues observed was the relatively low acti v i t y of B-hydroxybutyrate dehydrogenase (E.C. 1.1.1.30) compared to that found in monogastric animals (Watson and Lindsay, 1972). Although the presence of an <*-hydroxybutyrate dehydrogenase has been detected i n the blood of the calf (Vagner et a l . , 1973), the extent of the metabolism of ketone bodies in blood does not appear to have been thoroughly studied. Clark and Malan (1956) injected acetoacetic ethyl ester into sheep, and monitored changes in the level of blood B-hydroxybutyrate. Very l i t t l e conversion was found. However, this substrate may not be acted upon by B-hydroxybutyrate dehydrogenase (E.C. 1.1.1.30). From the rapid rate at which acetoacetate reduction appeared to occur following the injection of acetoacetate (Fig. 25), i t appears that B-hydroxybutyrate dehydrogenase (E.C. 1.1.1.30) may be present i n the blood of sheep. However, since 15 to 20% of the blood pool flows to the l i v e r per 109 • Total Ketone Bodies Fig. 25. Comparison of changes i n t o t a l and individual ketone body levels i n plasma following 3-hydroxybutyrate and acetoacetate injections. 110 minute (Smith and Hamlin, 1970), the results would also be consistent with the l i v e r reducing the injected acetoacetate. This explanation appears less l i k e l y due to the low level of hepatic 3-hydroxybutyrate dehydrogenase (E.C. 1.1.1.30) i n ruminants (Watson and Lindsay, 1972; Baird et a l . , 1968). Leng and Annison (1964) considered acetoacetate to be the main ketone body formed by the ruminal epithelium, and hepatic tissue to be of major importance i n the conversion of acetoacetate to 3-hydroxybutyrate in the fed animal. However, Katz and Bergman (1969) have found that 3-hydroxybutyrate was v i r t u a l l y the only ketone body produced by the splanchnic bed in the fed sheep. The results reported here suggest that acetoacetate has an insulinotropic action in the sheep. This action i s d i f f i c u l t to demonstrate since acetoacetate administered to fed or short-term fasted sheep i s very rapidly reduced to 3-hydroxybutyrate which does not have insulinotropic properties i n the sheep. In recent years, a number of studies have been directed at determining the relation between alterations in ketone body metabolism and the development of ketosis in the domestic ruminant. A number of these studies indicate that plasma acetoacetate levels could have significance i n the development of ketosis through an effect on insulin secretion. In normal fed ruminants 3-hydroxybutyrate i s by far the predominant ketone body produced by ruminant epithelium and the l i v e r , and this i s reflected by 3-hydroxybutyrate:acetoacetate ratios of about 10.0 in blood and about 50.0 i n l i v e r (Baird et a l . , 1972). However, when hepatic ketogenesis from FFA i s greatly increased, as i n starvation or ketosis (Ballard et a l . , 1968), there is a s h i f t toward greater acetoacetate production (Watson and Lindsay, 1972; Baird et a l . , 1972; Baird et a l . , 1968). This increased acetoacetate production is usually associated with a decrease i n the blood and hepatic I l l tissue 3-hydroxybutyrate:acetoacetate r a t i o i n ruminants, but not under similar conditions i n non-ruminant animals i n which i t appears to sl i g h t l y increase (Bergman,1971). With ketosis in ruminants there are marked increases i n the plasma acetoacetate and B-hydroxybutyrate, but the B-hydroxybutyrate:acetoacetate ra t i o f a l l s . The exact cause of the change in this ratio has not been determined. It has been speculated that when acetoacetate production increases there is probably a change i n the redox state of certain tissues, and since pyridine nucleotides are known to be coupled with B-hydroxybutyrate dehydrogenase (E.C. 1.1.1.30), i t seems that a change i n the available r a t i o of NAD+/NADH could.be responsible for the s h i f t i n the B-hydroxybutyrate: acetoacetate ratio (Bergman, 19 69; Williamson et a l . , 1967). However, other factors may be responsible for the s h i f t observed in the ketone body ra t i o during starvation or ketosis. "In monogastries B-hydroxybutyrate dehydrogenase (E.C. 1.1.1.30) i s normally in equilibrium with i t s substrates and free nucleotide cofactors (Williamson et a l . , 1967). However, the very low level of this enzyme in ruminant tissues probably does not allow an equilibrium to be established between i t s substrates and the NAD couple i n li v e r (Baird et a l . , 1968). This situation may change during ketosis so as to modify the ketone body r a t i o produced by the l i v e r . During acetonemia Kronfeld et a l . (1968) have found that the a b i l i t y of the mammary gland to synthesize fatty acids i s reduced and have suggested that this may be the result of a decreased a v a i l a b i l i t y of reduced pyridine nucleotide cofactor. A further finding was that during acetonemia the fatty acid substrates acetate and B-hydroxybutyrate are removed from circulation while acetoacetate i s released into circulation. The net ketone body uptake by the mammary gland was found to be slightly positive, but the average 112 mammary acetoacetate release was measured at 250 g/day. I t has been suggested that the c l i n i c a l severity of acetonemia i s related more to the blood concentration of acetoacetate than to that of B-hydroxybutyrate (Thin and Robertson, 1953). Thus metabolic events leading to a decrease in the circulating B-hydroxybutyrate:acetoacetate r a t i o and an increase i n plasma acetoacetate levels, appear to contribute to the development of c l i n i c a l symptoms. Although hyperketonemia appears to be present i n pregnant ewes which otherwise appear normal, the development of pregnancy toxaemia can occur following a marked decrease i n the plasma glucose level (Kronfeld, 1970). The l a t t e r stages of the disease may be due to severe ketoacidosis. It might be speculated that increasing levels of plasma acetoacetate, during the early stages of ruminant ketosis,stimulate insulin release; plasma glucose levels decrease, and c l i n i c a l symptoms develop. However, Kronfeld (1970) has pointed out that such a mechanism i s counter-indicated, since a study of glucose kinetics i n sheep with pregnancy toxaemia revealed that the glucose pool, transfer rate and glucose space are a l l somewhat decreased. Yet i t remains possible that the insulinotropic action of acetoacetate has etiologic significance i n the development of pregnancy toxaemia. For example, a slight temporary increase i n insulin secretion i n response to an increased level of plasma acetoacetate may cause some impairment of gluconeogenesis, and thus i n i t i a t e a series of events leading to development of the disorder. There is very l i t t l e information available on changes in plasma insulin level in relation to ruminant ketosis. Leng (1965) found that plasma insulin-like a c t i v i t y tended to be elevated above normal in sheep with pregnancy toxaemia. Kinetic studies suggest that during acetonemia there i s an insulin excess (Kronfeld, 1970). Thus the concept of hypoglycemia resulting from ketone 113 body stimulated insulin release may have relevance to the development of acetonemia. Previous to the studies reported i n this thesis,there was considerable indirect evidence for an insulinotropic action of ketone bodies i n runiinants. However, attempts to measure directly this effect gave negative results, possibly because of the very rapid reduction of injected acetoacetate to B-hydroxybutyrate. The insulinotropic action of acetoacetate has been demonstrated in the experiments reported here. On the basis of this result, and studies on pregnancy toxaemia and acetonemia, the insulino-tropic effect of acetoacetate may have etiologic significance i n the development of these disorders. 114 BIBLIOGRAPHY Aafjes, J.H. 1964. Volatile fatty acids (VFA) in blood of cows with ketosis. Life Sciences 3: 1327-1332. ALlen, R.S. 1970. Carbohydrate metabolism, p. 547. In: M.J. Swenson (ed.) Dukes' physiology of domestic animals. Cornell Univ. Press, Ithaca. Annison, E.F., R.E. Brown, R.A. Leng, D.B. Lindsay and C.E. West. 1967. Rates of entry and oxidation of acetate, glucose, D(-)-0-hydroxybutyrate, palmitate, oleate, and stearate, and rates of production and oxidation of propionate and butyrate i n fed and starved sheep. Biochem. J. 104: 135-142. Annison,E.F., D.B. Lindsay, and R.R. White. 1962. The measurement of entry rates of propionate and of butyrate in sheep. Biochem. J . 85: 474-479. Annison, E.F. and R.R. White. 1961. Glucose metabolism i n sheep. Biochem. J. 80: 162-168. Annison, E.F. and D. Lewis. 1959. Metabolism i n the rumen. Methuen and Co. Ltd., London. 20p. Annison, E.F., K.J. H i l l , and D. Lewis. 1957. Studies on the portal blood of sheep. 2. Absorption of v o l a t i l e fatty acids from the rumen of the sheep. Biochem. J . 66: 592-597. Annison, E.F. 1954. Studies on the v o l a t i l e fatty acids of sheep blood with special reference to formic acid. Biochem. J . 58: 670-675. Ash, R.W., R.J. Pennington, and R.S. Reid. 1964. The effect of short chain fatty acids on blood glucose concentration in sheep. Biochem. J. 90: 353-357. Ashmore, J . , and G. Weber. 1968. Hormonal control of carbohydrate metabolism in l i v e r , p. 336-373. In: F. Dickens, P.J. Randle, and W.J. Whelan (ed). Carbohydrate metabolism and i t s disorders. Vol. 1. Academic Press, New York. Association of o f f i c i a l agricultural chemists. 1960. Washington, D.C. Baile, C.A., Z. Glick and J . Mayer. 1969. Effects of secretin and cholecystokinin-pancreozyniin on pancreatic juice and insulin secretion of goats. J . Dairy S c i . 52: 513-517. Baird, G.D., R.J. Heitzman, and K.G. Hibbitt. 1972. Effects of starvation an intermediary metabolism in the lactating cow. Biochem. J . 128: 1311-1318. Baird, G.D., K.G. Hibbitt, G.D. Hunter, P. Lund, M. Stubbs, and H.A.Krebs. 1968. Biochemical aspects of bovine ketosis. Biochem. J . 107:683-689. 115 Balasse, E.O., H.A. Conns, and J.P. Lambilliotte., 1970. Evidence for a stimulatory effect of ketone bodies on insulin secretion in sheep. Horm. Metab. Res. 2: 371-372. Ballard, F.J., R.W. Hanson, and D.S. Kronfeld. 1969. Gluconeogenesis and lipogenesis i n tissue from ruminant and non-ruminant animals. Federation Proc. 28: 218-231. Ballard, F.J., R.W. Hanson, D.S. Kronfeld, and F. Raggi. 1968. Metabolic changes i n l i v e r associated with spontaneous ketosis and starvation i n cows. J . Nutrition 95: 160-172. Ballard, F.J., and L.T. Oliver. 1965. Carbohydrate metabolism i n l i v e r from f e t a l and neonatal sheep. Biochem. J . 95: 191-200. Bartley, J.R. and A.L. Black. 1966. Effect of exogenous glucose on glucose metabolism i n dairy cows. J . Nutrition 89: 317-322. Bartos, S., J . Skarda, and J . Base. 1970. Effect of glucose propionate and butyrate on the secretion of immune-reactive insulin in goats. Endocrinol. Exp. 4: 151-157. Bassett, J.M. 1972. Plasma glucagon concentrations i n sheep: their regulation and relation to concentrations of insulin and growth hormone. Australian J . B i o l . S c i . 25: 1277-1287.' Bassett, J.M., R.H. Weston, and J.P. Hogan. 1971. Dietary regulation of plasma insul i n and growth hormone concentrations i n sheep. Australian J . B i o l . S c i . 24: 32;-330. Bassett, J.M. and A.L.C. Wallace. 1967. . Influence of Cortisol on plasma insulin in the sheep. Diabetes 16: 566-571. Baumgardt, B.R. 1964. Practical observations on the quantitative analysis of free v o l a t i l e fatty acids (VFA) i n aqueous solutions by gas-liquid chromatography. Dep. Dairy Sci . Bull. 1. Univ. of Wisconsin. B e l l , J.P., L.A. Salamonsen, G.W. Holland, E.A. Espiner, D.W. Beaven and D.S. Hart. 1970. Auto-transplantation of the pancreas i n sheep: Insulin secretion from the transplant. J . Endocrinol. 48: 511-525. Bergman, E.N. and J.E. Wolf. 1971. Metabolism of v o l a t i l e fatty acids by l i v e r and portal-drained viscera in sheep. Amer. J . Physiol. 221: 586-592. Bergman, E.N. 1971. Hyperketonemia-ketogenesis and ketone body metabolism. J . Dairy Sci . 54: 936-948. Bergman, E.N., K. Kon, and M.L. Katz. 1963. Quantitative measurements of acetoacetate metabolism and oxidation in sheep. Amer. J. Physiol. 205: 658-663. 116 Bergman, E.N. 1963. Quantitative aspects of glucose i n pregnant and non-pregnant sheep. Amer. J. Physiol. 204: 147-156. Bensadoun, A., O.L. Paladines and J.T. Reid. 1962. Effect of level of intake and physical form of the diet on plasma glucose concentration and v o l a t i l e fatty acid absorption in ruminants. J. Dairy Sci. 45: 1203-1208. Bensadoun, A. 1960. Direct estimation of the absorption of v o l a t i l e fatty acids from the gastrointestinal tract of ruminants. Ph.D. Thesis. Cornell Univ. Library, Ithaca, New York. Bewsher, P.D. and J. Ashmore. 1966. Ketogenic and l i p o l y t i c effects of glucagon on l i v e r . Biochem. Biophys. Res. Commun. 24: 431-436. Blum, A.L. and W.A. Linscheer. 1971. Determinants of insulin response to glucose absorption from the human small bowel. Metab. C l i n . Exp. 20: 666-677. Boda, J.M. 1964. Effect of fast and hexose injections on serum insulin concentrations of sheep. Amer. J . Physiol. 206: 419-424. Boynes, D.R.,.R.J. Jarrett, and H. Keen. 1966. Intestinal hormones and plasma insulin. Lancet 1: 409-411. C a h i l l , G.F. J r . , and O.E. Owen. 1968. Seme observations on carbohydrate . metabolism in man, p. 497-522. In: F. Dickens, P.J. Randle, and W.J. Whelan (ed.). Carbohydrate metabolism and i t s disorders. Vol. 1. Academic Press, New York. Castro, A., J.P. Scott, D.P. Grettie, D. MacFarlane, R.E. Bailey. 1970. Plasma insulin and glucose responses of healthy subjects to varying glucose loads during three-hour oral glucose tolerance tests. Diabetes .19:842-851. Cherrington, A. and M. Vranic. 1971. Role of glucagon and insulin i n control of glucose turnover. Metab.. C l i n . Exp. 20: 625-628. Cole, J.F., R.A. Curtis, B.J.Mc Sherry, J. McD. Robertson, and D.S. Kronfeld. 1969. Bovine ketosis: frequency of c l i n i c a l signs, complications and alterations in blood ketones, glucose and free fatty acids. Can. Vet. J. 10: 179-187. Chisholm, D.J., E.W. Kraegen, J.D. Young, and L. Lazarus. 1971. Comparison of secretion response to oral intraduodenal or intravenous glucose administration. Harm. Metab. Res. 3: 180-183. Clark, R. and J. R. Malan. 1956. Alterations i n the blood sugar and ketone levels by dosing acetate, propionate and butyrate into the rumen of the sheep. Onderstepoort J . Vet. Res. 27: 101-109. Coore, H.G. and P.J. Randle. . 1964. Regulation of insulin secretion studied with pieces of rabbit pancreas incubated i n v i t r o . Biochem. J . 93: 66-78. 117 Craine, E.M. and R.G. Hansen. 1952. The short chain fatty acids of the peripheral blood of goats. J. Dairy Sci. 35: 631-638. Crespin, S.R., W.B. Greencugh, and D. Steinberg. 1969. Stimulation of insulin secretion by infusion of free fatty acids. J. C l i n . Invest. 48: 1934-1943. Curry, D.I. 1971. Insulin secretory dynamics in response to slow r i s e and square wave stimuli. Amer. J . Physiol. 22: 324-328. Dickson, W.M. 1970. Endocrine glands, p. 1216-1230. In: M.J. Swenson (ed.) Dukes'/'physiology of domestic animals. Cornell Univ. Press, Ithaca. Dupre, J. 1970. Regulation of the secretions of the pancreas. Ann. Rev. Med. 21: 299-316. Dupre, J . , J.D. Curtis, R. Waddell, and J.C. Beck. 1967. Effects of. digestive secretagogues on the endocrine pancreas in man. Advan* Exp. Med. Bi o l . 2: 297303. Edson, N.L. 1935. Ketogenesis - antiketogenesis. I. The influence of ammonium chloride on ketone body formation in l i v e r . Biochem. J . 29: 2082-2094. Elsden, S.R. 1946. The application of the s i l i c a gel partition chromatogram to the estimation of v o l a t i l e fatty acids. Biochem. J. 40: 252-255. Erwin, E.S., G.J. Marco, and E.M. Emery. 1961. Volatile fatty acid analyses of blood and rumen f l u i d by gas chrcmatographv. J . Dairy Sci. 44: 1768-1769. Faichney, G.J. 1968. The production and absorption of v o l a t i l e fatty acids (VFA) from the rumen of the sheep. Australian J . Agr. Res. 19: 791-802. Fajans, S.S., J.C. Floyd J r . , R.F. Knopf, and J.W. Conn. 1967. Effect of amino acids and proteins on insulin secretion i n man. Recent Progress Hormone Res. 23: 617-662. Fel i g , P. and J . Wahren. 1968. Insulin concentrations i n the foetal plasma and foetal fluids of sheep. J. Endocrinol. 40: 389-390. Field, J.B., M. Webster, and T. Drapanas. 1968. Evaluation of factors regulating hepatic control of insulin hcmoestasis. Amer. Soc. Cl i n . Invest. (Abstr.) 99. Fieser, L.F., and M. Fieser. 1961. Advanced organic chentistry. Reinhold Pub. Corp., Chapman and Hall Ltd., New York. F i r s t ed., p363. 118 Frohman, L.A. 1969. The endocrine function of the pancreas. Ann. Rev. Physiol. 31: 353-382. Garcia, A., J.R. Williamson, and G.F. Cah i l l Jr. 1966. Studies on the perfused rat l i v e r . II. Effect of glucagon on gluconeogenesis. Diabetes 15: 188-193. Gagliardino, J.J., R.E. Hernandez. 1971. Circadian variation of the serum glucose and immunoreactive insulin levels. Endocrinology 88: 1529-1531. Goetch, G.P., and W.R. Pritchard. 1958. Effects of oral administration of short chain fatty acids on certain blood and urine constituents of fasted phlorizin treated ewes. Amer. J. Vet. Res. 19: 637-643. Gorman, CK. , J.M. Salter, and J.C. Perihos. 1967. Effect of glucagon on lipi d s and glucose i n normal and eviscerated rats, and on isolated perfused rat li v e r s . Metabolism 16: 1140-1157. Grant, W.M. 1948. Calorimetric micrcdetermination of formic acid based on reduction to formaldehyde. Anal. Chem. 20: 267-272. Greenberger, N.J., M. Tzagournis, and T.M. Graves. 1968. Stimulation of insulin secretion in man by medium chain triglycerides. Metabolism 17: 796-801. Greenough, W.B., S.R. Crespin, D. Steinberg. 1967. Hypoglycemia and hyperinsulinemia in response to raised free fatty acids (FFA) levels. Lancet 7530: 1334-1336. Grey, N.J., S. Goldring and D.M. Kipnis. 1969. Evidence for a glucose-inducible glucoreceptor for insulin secretion i n the rat. p.155-171. In: E. Cerasi and R. Luft (ed.) Pathogenesis of diabetes mellitus. John Wiley and Sons, New York. Grodsky, G.M. and P.H. Forsham. 1966. Insulin and the pancreas. Ann. Rev. Physiol. 31: 347-380. Hanson, R.W. and -F.J. Ballard. 1967. The relative significance of acetate and glucose as precursors for l i p i d synthesis in l i v e r and adipose tissue from ruminants. Biochem. J . 105: 529-536. Harvey, W.R. 1960. Least square analysis data with unequal subclass numbers. Publ. ARS, 20-8. U.S.D.A. B e l t s v i l l e , Md. Hertelendy, F., L. Machlin and D.M. Kipnis. 1969. Further studies on the regulation of insulin and growth hormone secretion in the sheep. Endocrinology 84: 192-199. Hertelendy, F., L.J. Machlin, Y. Takahashi and D.M. Kipnis. 1968. Insulin release from sheep pancreas i n v i t r o . J . Endocrinol. 41: 605-606. Horino, M., L.J. Machlin, F. Hertelendy, and D.M. Kipnis. 1968. Effect of short chain fatty acids on plasma insulin i n ruminants and non-r\uninant species. Endocrinology 83: 118-128. 119 Howard, B.H. 1967. Intestinal microorganisms of ruminants and other vertebrates, p. 317-385. In: S.M. Henry (ed.) Symbiosis. Academic Press, New York. James, A.T. and A.J.P. Martin. 1952. Gas-liquid partition chromatography: the separation and micro-estimation of v o l a t i l e fatty acids from formic acid to dodecanoic acid. Biochem. J. 50: 679-685. Jarrett, I.G., G.B. Jones, and B.J. Potter. 1964. Changes in glucose u t i l i z a t i o n during development of the lamb. Biochem. J . 90: 189-194. Jarrett, I.G. and B.J. Potter. 1953. Insulin tolerance and hypoglycemic convulsions in sheep. Australian J. Exptl. B i o l . Med. S c i . 31: 311-318. Jorgensen, N;A. and L.H. Schultz. 1963. Ration effects on rumen acids, ketogenesis and milk composition. J. Dairy Sci. 46: 437-443. Judson, G.J., E. Anderson, R.J. Luick and R.A. Leng. 1968. The contribution of propionate to glucose synthesis in sheep given diets of different grain content. B r i t . J . Nutrition 22: 69-75. Kahil, M.E., G.L. Mcllhaney, and P.H. Jordan. 1970. Effect of enteric hormones on insulin secretion. Metabolism 19: 50-57. Karr, M.R., C.O. L i t t l e , and G.E. Mitchell. 1966. Starch disappearance from different segments of the digestive tract of steers. J . Anim. Sc i . 25: 652-654. Katz, M.L. and E.N. Bergman. 1969. Hepatic and portal metabolism of glucose, free fatty acids and ketone bodies i n the sheep. Amer. J . Physiol. 216: 953-960. Keech, D.B. and M.F. Utter. 1963. Pyruvate carboxylase. I L Properites. J . B i o l . Chem. 238: 2609-2614. Kipnis, D.M. 1972. Nutrient regulation of insulin secretion in human subjects. Diabetes 21: (suppl.2) 606-616. Kronfeld, D.S. 1970. Ketone body metabolism, i t s control and i t s implications i n pregnancy toxaemia, acetonemia and feeding standards, p. 566-583. In: A.T. Phillipson (ed.) Physiology of digestion and metabolism in the ruminant. Oriel Press, Newcastle upon Tyne. Kronfeld, D.S., F. Raggi and C.F. Ramberg, J r . 1968. Mammary blood flow and ketone body metabolism in normal fasted and ketotic cows. Amer. J. Physiol. 215: 218-227. Kulka, R.G., H.A. Krebs, and L.V. Eggleston. 1961. The reduction of acetoacetate to B-hydroxybutyrate i n animal tissues. Biochem. J. 78: 95-106. 120 Leat, W.M.F. 1970. Carbohydrate and l i p i d metabolism in the ruminant during post-natal development, p.211-222. In: A.T. Phillipson (ed.) Physiology of digestion and metabolism i n the ruminant. O r i e l Press, Newcastle upon Tyne. Leng, R.A. and C.E. West. 1969. Contribution of acetate, butyrate, palmitate, stearate and oleate to ketone body synthesis in sheep. Res. Vet. Sci.10:57-63. Leng, R.A. 1965. Ketone body metabolism i n normal and underfed pregnant sheep and i n pregnancy toxaemia. Res. Vet. S c i . 6: 433-441. Leng, R.A. and E.F. Annison. 1964. The metabolism of D(-)-B-hydroxybutyrate in sheep. Biochem. J . 90: 464-469. Lindsay, D.B. ' 1970. Carbohydrate metabolism in the ruminant, p. 438-451. In: A.T. Phillipson (ed.) Physiology of digestion and metabolism i n the ruminant. Oriel Press, Newcastle upon Tyne. Littledike, E.T., D.A. Witzel and S.C. Whipp. 1968. Insulin: Evidence for inhibition of release i n spontaneous hypocalcemia. Proc. Soc. Exp. B i o l . Med. Loubatieres, A., M.M. Mariani, G. Sord and L. Sari. 1971. The action of B-adrenergic blocking and stimulating agents on insulin secretion. Characterization of the type of B-receptor. Diabetalogia 7: 127-132. Luthman, J. and G. Jonson. 1968. The glycogenic properties of butyric acid. Acta Vet. Scand. 9:168-173. McCandless, E.L. and J.A. Dye. 1950. Physiological changes i n intermediary metabolism of various species of ruminants incident to functional development of the rumen. Amer. J . Physiol. 31: 434-445. MacRae, J.C. and D.G. Armstrong. 1966. Investigations of the passage of a-linked glucose polymers into the duodenum of sheep. Proc. Nutrition Soc. 25: x x x i i i . MacRae, J.C. and D.G. Armstrong. 1969. Studies on intestinal digestion i n the sheep. 2. Digestion of some carbohydrate constituents in hay cereal and hay-cereal rations. B r i t . J . Nutrition 23:. 377-387. McLean, P., J. Brown and A.L. Greenbaum. 1968. Hormonal control of carbohydrate metabolism of adipose tissue, p.398-425. In: F. Dickens, P.J. Randle and W.J. Whelan (ed.) Carbohydrate metabolism and i t s disorders. Vol. 1. Academic Press, New York. McClymontjG.L. 1951. Identification of the v o l a t i l e fatty acid in the peripheral blood and rumen of cattle and the blood of other species. Australian J. Agr. Res. 2: 92-96. 121 McNair, H.M. and E.J. Bonelli. 1969. Basic gas chromatography, Varian aerograph. Consolidated printers, Berkeley, California. 5th ed. p.143. Madison, L.L., D. Mebane, B. linger and A. Lochner. 1964. The hypoglycemic action of ketones.: I I . Evidence for a stimulatory feedback of ketones on the pancreatic beta c e l l . J . Cl i n . Invest. 43: 408-415. Madison, L.L., D. Mebane and A. Lochner. 1963. Evidence for a stimulatory feedback of ketone bodies on pancreatic beta c e l l s . J. C l i n . Invest. 42: 955 (Abstr.) Malaisse, W. and F. Malaisse-Lagae. 1970. Biochemical pharmacological and physiological aspects of the adenyhyclase-phosphodiesterase system in the pancreatic 3-cell, p. 435-444. In: S. Faulkner, B. Hellman, and I.B. Taljedal (ed.) The structure and metabolism of the pancreatic i s l e t s . Pergamon Press, Toronto. Manns, J.G. and J.M. Boda. 1967. Insulin release by acetate, propionate, butyrate and glucose in lambs and adult sheep. Amer. J . Physiol. 212r 747-755. Manns, J.G., J.M. Boda, R.F. Willes. 1967. Probable role of propionate and butyrate i n control of insulin secretion i n sheep. Amer. J . Physiol. 212: 756-764. Mebane, D. and L.L. Madison. 1964. C l i n i c a l and experimental hypoglycemic action of ketones. I. Effects of ketones on hepatic glucose output and peripheral glucose u t i l i z a t i o n . J . Lab. C l i n . Med. 63: 177-192. Mebane, D. and L.L. Madison. 1962. The hypoglycemic effect of ketone bodies. J . Clin. Invest. 41: 1383-1384. Menahan, L.A., L.H. Schultz and W.G. Hoekstra. 1967. Metabolism of butyrate-3-carbon-14 in the ruminant under various metabolic states. J . Dairy S c i . 50: 1417-1429. Menahan, L.A., L.H. Schultz and W.G. Hoekstra. 1966a. Factors affecting ketogenesis from butyric acid i n the ruminant. J . Dairy S c i . 49: 835-845. Menahan, L.A., L.H. Schultz and W.G. Hoekstra. 1966b. Relationship of ketone body metabolism and carbohydrate u t i l i z a t i o n to fat mobilization in the ruminant. J . Dairy S c i . 49: 957-961. Metz, R. and R. Friedenberg. 1970. Effects of repetitive glucose loads on plasma concentrations of glucose , insulin and free fatty acids: Paradoxical insulin responses i n subjects with mild glucose intolerance. J . C l i n . Endorinol. Met. 30: 602-608. 122 Milner, R.D.G. 1970. The stimulation of insulin release by essential amino acids from rabbit pancreas i n vi t r o . J . Endocrinol. 47: 347-356, Moyle, V., E. Baldwin and R. Scarisbrick. 1948. Separation and estimation of saturated C2-Cg fatty acids by buffered partition columns. Biochem. J . 43: 408-412. Muller, W.A., G.R. Faloona and R.H. Unger. 1971. The effect of alanine on glucagon secretion. J . C l i n . Invest. 50: 2215-2218. Muller, W.A. 1971. The effect of experimental insulin deficiency on glucagon secretion. J . Cl i n . Invest. 50: 1992-1999. Nahara, H.T. and C F . Cori. 1968. Hormonal control of carbohydrate metabolism i n muscle, p. 375-396. In: Carbohydrate metabolism and i t s disorders. Vol. 1. Academic Press, New York. Nelson, N.C, W.G. Blackard, J.C. Cocchia, and J.A. Labat. 1967. Influence of the vagus nerves on pancreatic insulin secretion (dog). Diabetes 16: 852-857. Nicholson, J.W.G. and J.D. Sutton. 1969. The effect of diet composition and level of feeding on digestion in the stomach and inestines of sheep. B r i t . J . Nutrition 23: 583-601. Orskov, E.R. and C Fraser. 1968. Dietary factors influencing starch disappearance in various parts of the alimentary tract and caecal fermentation i n early-weaned lambs. Proc. Nutrition Soc. 27: 37A. Patterson, D.S.P. 1963. Some observations on the estimation of non-esterified fatty acid concentrations i n cow and sheep plasma. Res. Vet. S c i . 4: 230-237. Peden, V.H. 1964. Determination of individual serum "ketone bodies" with normal values i n infants and children. J . Lab. C l i n . Med. 63: 332-343. Pennington, R.J. 1952. The metabolism of short chain fatty acids i n the sheep. I. Fatty acid u t i l i z a t i o n and ketone body production by ruminal epithelium and other tissues. Biochem. J. 51: 251-258. P h i l l i p s , R.W., W.A. House, R.A. Mi l l e r , J.L. Mott, and D.L. Sooby. 1969. Fatty acid, epinephrine, and glucagon hyperglycemia in- normal and depancreatized sheep. Amer. J . Physiol. 217: 1265-1268. P h i l l i p s , R.W. 1966. Fatty acid induced glyoogenolysis in ruminants. Physiologist 9: 265 (Abstr.) Pi-fSunyer, F. Xavier, R.G. Campbell, and S.A. Hashim. 1970. Experimentally induced hyperketonemia and insulin secretion i n the dog. Metab. C l i n . Exp. 19: 263-270. Randle, P.J.?S.J.H. Ashcroft and J.R. G i l l . 1968. Carbohydrate metabolism and release of hormones, p. 427-447. In: F. Dickens, P.J. Randle and W.J. Whelan (ed.) Carbohydrate metabolism and i t s disorders. Vol. 1. Academic Press, New York. 123 Reid, R.L. 1950. Studies on the carbohydrate metabolism of sheep. II. The uptake by the tissues of glucose and acetic acid from peripheral circulation. Australian J. Agr. Res. 1:338-343. Reiser, P. 1967. Insulin metabolism and membranes, p. 59-98. The Williams and Wilkins Co., Baltimore. Ross, J.P. and W.D. Kitts. 1971. Improved method for the determination of v o l a t i l e fatty acids i n ruminant blood plasma. J . Dairy Sci. 54: 1824-1831. Ross, J.P. and W.D. Kitts. 1969. Concentration of certain blood metabolites in obese pregnant and non-pregnant ewes. Can. J . Anim. S c i . 49:91-95. Saribar, S.S., J.R. Evans, B. Lin and G. Hetenyi Jr. 1967. Farther studies on the effect of octanoate on glucose metabolism i n dogs. Can. J . Physiol. Pharm. 45: 29-38. Sanbar, S.S., G. Hetenyi J r . , N. Forbath, and J.R. Evans. 1965. Effects of infusion of octanoate on glucose concentration in plasma and the rates of glucose production and u t i l i z a t i o n in dogs. Metabolism 14: 1311-1323. Schimmel, R.J. and E. Khobil. 1970. Insulin, FFA and stimulation of hepatic gluconeogenesis during fasting. Amer. J . Physiol. 218: 1540-1547. Schultz, L.H., V.R. Smith, and H.A. Lardy. 1949. Blood sugar studies in .• relation to ketosis in ruminants. J. Dairy Sci. 32: 718. (Abstr.) Schultz, L.H. and V.R. Smith. 1951. Experimental alterations of the sugar and ketone levels of the blood of ruminants i n relation to ketosis. J. Dairy S c i . 34: 1191-1199. Senel, S.H. and F.G. Owen. 1967. Relation of dietary acetic and butyric acids to intake, d i g e s t i b i l i t y , lactation, performance arid ruminal and blood levels of certain metabolites. J. Dairy Sci. 50: 327-332. Smith,- C.R. and Hamlin. 1970. Regulation of the heart and blood vessels, p. 170-171. In: M.J. Swensen (ed.) Dukes' physiology of domestic animals. Cornell University Press, Ithaca. Snedecor, G.W. and W.G. Cochran. 1967. S t a t i s t i c a l methods. 6th ed. Iowa State University Press, Ames, Iowa. Staubus, J.R., R.E. Brown, C L . Davis and W.O. Nelson. 1960. Effect of phlorizin, insulin and butyrate on the concentration of glucose and ketones in the blood and urine of fasted steers. J . Dairy S c i . 43: 1796-1808. Steffens, A.B. 1970. Plasma insulin content i n relation to blood glucose level and meal pattern i n the normal and hypothalamic hyperphagic rat. Physiol. Behav.5: 147-151. 124 Stem, J.S., CA. Baile and J . Mayer. 1970. Are propionate and butyrate physiological regulators of plasma insulin i n ruminants? Amer. J. Physiol. 219:84-91. Sum, P.J. and R.M. Preshaw. 1967. Intraduodenal glucose infusion and pancreatic secretion in man. Lancet 2: 340-341. Sutton, J.D., and J.W.G. Nicholson. 1968. The digestion of energy and starch along the gastro-intestinal tract of sheep. Proc. Nutrition Soc. 27: 49A.. Szepesi, B. arid CD. Berdanier. 1971. Time course of the starve-refeed response i n rats: the possible role of insulin. J . Nutrition 101: 1563-1574. Taylor, J.A. and H.D. Jackson. 1968. Formation of ketone bodies from palmitate-carbon-14, and glycerol-carbon-14 by tissues from ketotic sheep. Biochem. J. 106: 289-292. Thin, C and A. Robertson. 1953. Biochemical aspects of ruminant ketosis, J . Gomp. Pathol. Therap. 63: 184-189. Thye, F.W., R.G. Warner and P.D. Mi l l e r . 1970. Relationship of various blood metabolites to voluntary feed intake i n lactating ewes. J. Nutrition 100: 565-572. Topps, J.H. and R.N.B. Kay. 1969. A comparison of methods for measurement of starch i n food and gut contents. Proc. Nutrition Soc. 28: 23A. Topps, J'i H., R.N.B. Kay, E.D. Goodall. 1968. Digestion of concentrate and of hay diets i n the stomach and intestines of ruminants. I. Sheep. B r i t . J . Nutrition 22: 261-280, Topps, J.H., F.G. Whitelaw, and R.S. Reid. 1968. Digestion of concentrate and of hay diets in the stomach and intestines of ruminants. 2. Cattle. B r i t . J . Nutrition 22: 281-290. ' Trenkle, A. 1970. Effects of short chain fatty acids, feeding, fasting and type of diet on plasma insulin levels i n sheep. J . Nutrition 100: 1323-1330.. Tucker, R.E., G.E. Mitchell, and C.O. L i t t l e . 1968. Ruminal and postruminal starch digestion i n sheep. J . Anim. Sci. 27: 824-826. Turner, D.S. and R.J. Jarrett. 1970. Intestinal hormones and insulin release and biosynthesis, p. 463-468. In: S. Faulkner, B. Hellman and I.B. Taljedal (ed.) The structure and metabolism of the pancreatic i s l e t s . Pergamon Press, Toronto. Unger, R.H. and AiM. Eisentraut. 1970. Regulation of glucagon release in vivo, p. 141-148. In: S. Faulkner, B. Hellman, and I.B. Taljedal (ed.) The structure and metabolism of the pancreatic i s l e t s . Pergamon Press, Toronto. 125 Unger, R.H., H. Ketterer, J. Dupre, and A.M. Eisentraut. 1967. The effects of secretin, pancreozymin, and gastrin on insulin and glucagon secretion in anaesthetized dogs. J . C l i n . Invest. 46: 630-645. Vagner, J.P., S. Blatt, and M. Kay. 1973. Biochemical and hematological values i n male Holstein-Friesian calves. Amer. J . Vet. Res. 34: 273-277. Vance, J.E., K.D. Buchanan, and R.H. Williams. 1971. Glucagon and insulin release: Influence of drugs affecting the autonomic nervous system. Diabetes 20: 78-82. Vance, J.E., K.D. Buchanan, and R.H. Williams. 1968. Effect of starvation and refeeding on serum immunoreactive glucagon and insulin levels. J . Lab. C l i n . Med. 72: 290-297. Walker, P.G. 1954. A colorimetric method for the estimation of acetoacetate. Biochem. J . 58: 699-703. Watson, H.R., and D.B. Lindsay. 1972. 3-hydroxybutyrate dehydrogenase in tissues from normal and ketonaemic sheep. Biochem. J . 128: 53-57. Weller, R.A., F.V. Gray, A.F. Pilgrim and G.B. Jones. 1967. The'rates of production of v o l a t i l e fatty acids i n the rumen. IV. Individual and total v o l a t i l e fatty acids. Australian J . Agr. Res. 18: 107-118. ' Williams, R.H. and J.W. Ensick. 1966. Secretion, fates and actions of insulin and related products. Diabetes 15: 623-654. Williamson^ D.H., P. Lund, and H.A. Krebs. 1967. The redox state of free nicotinamide-adenine dinucleotide in the cytoplasm and mitochondria of rat l i v e r . Biochem. J . 103: 514-527. Wright, P.L., R.B.Grainger, and G.J. Marco. 1966. Post-ruminal degradation and absorption of carbohydrate by the mature ruminant. J. Nutrition 89: 241-246. Yallow, R.S. and S.A. Berson. 1965. Dynamics of insulin secretion i n hypoglycemia. Diabetes 14: 341-349. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            data-media="{[{embed.selectedMedia}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0100971/manifest

Comment

Related Items