UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Effects of medium chain fatty acids and ketones on leucine metabolism in astrocytes : towards an understanding… Townsend, Marria May 2001

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

Item Metadata

Download

Media
831-ubc_2001-0587.pdf [ 3.59MB ]
Metadata
JSON: 831-1.0090247.json
JSON-LD: 831-1.0090247-ld.json
RDF/XML (Pretty): 831-1.0090247-rdf.xml
RDF/JSON: 831-1.0090247-rdf.json
Turtle: 831-1.0090247-turtle.txt
N-Triples: 831-1.0090247-rdf-ntriples.txt
Original Record: 831-1.0090247-source.json
Full Text
831-1.0090247-fulltext.txt
Citation
831-1.0090247.ris

Full Text

E F F E C T S O F M E D I U M C H A I N F A T T Y A C I D S A N D K E T O N E S O N L E U C I N E M E T A B O L I S M IN A S T R O C Y T E S : T O W A R D S A N U N D E R S T A N D I N G O F T H E A N T I - E P I L E P T I C E F F I C A C Y O F T H E K E T O G E N I C DIET Marr ia M a y Townsend B . S c . (Nutr. Sci.) The University of British Co lumbia , 1997 A T H E S I S S U B M I T T E D IN P A R T I A L F U L F I L L M E N T O F T H E R E Q U I R E M E N T S F O R T H E D E G R E E O F M A S T E R O F S C I E N C E In T H E F A C U L T Y O F G R A D U A T E S T U D I E S (Human Nutrition) Department of Food, Nutrition and Health W e accept this thesis as conforming to the required standard T H E U N I V E R S I T Y O F BRIT ISH C O L U M B I A 2001 © M A R R I A M A Y T O W N S E N D , 2001 UBC Special Collections - Thesis Authorisation Form Page 1 of 1 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 req u i r e m e n t s f o r an advanced degree a t t h e 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 c o p y i n g o f 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 g r a n t e d by the head o f my department or by h i s or her r e p r e s e n t a t i v e s . I t i s u n d e r s t o o d t h a t c o p y i n g 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 n ot be a l l o w e d w i t h o u t my w r i t t e n p e r m i s s i o n . The U n i v e r s i t y o f B r i t i s h Columbia Vancouver, Canada http://www.library.ubc.ca/spcoll/thesauth.html 9/26/01 A B S T R A C T A high fat, low g lucose diet, termed "ketogenic" because it results in e levat ions in circulat ing ketones, has been used for over 75 years as a treatment for pediatr ic epi lepsy. The mechan ism by which the ketogenic diet supp resses epi lept ic se izures is not understood. Fundamental ly , the diet must involve an effect on brain metabol ism but there is a lack of information about the metabol ic impact of a change in fuel source at the level of the brain cel l . Th is study examined the effect of medium chain fatty ac ids ( M C F A ) and ketones on the oxidation of leucine in astrocytes. The first ser ies of exper iments measured the production of 1 4 C 0 2 , f r o m [U 1 4 C]- leuc ine, in the p resence of no addit ional substrate (control) and increasing concentrat ions of octanoate (an M C F A ) and p-hydroxybutyrate. T h e second ser ies of exper iments measured 1 4 C 0 2 product ion from oxidative decarboxylat ion of [1- 1 4 C]- leucine and 1 4CC> 2 production from the chemica l decarboxylat ion of [1-1 4 C] - leuc ine der ived a-keto isocaproate (a-KIC) in the p resence and absence of P-hydroxybutyrate and octanoate. Inclusion of p-hydroxybutyrate caused a 6 0 - 7 0 % reduction in 1 4 C 0 2 product ion from [U- 1 4 C]- leuc ine; with octanoate the inhibition was even more dramatic with 8 0 % reduction compared to control. Exper iments using [1 - 1 4 C]-leucine did not find a statistically signif icant change in 1 4 C 0 2 production when p-hydroxybutyrate was included, but did f ind an increased level of label led a-KIC in the media, reflecting leucine that had been transaminated but had not p roceeded through to the second step of metabol ism. The amount of residual a -KIC was increased by up to 54%. Octanoate did inhibit oxidative decarboxylat ion of [1-ii A B S T R A C T 1 4 C]- leuc ine , with 5.0 m M octanoate reducing the product ion of 1 4 C C » 2 by 94%. In contrast to the accumulat ion of a -K IC seen in exper iments using p-hydroxybutyrate, incubation with octanoate resulted in a dec reased product ion of a -K IC. Th is f inding suggests that octanoate and p-hydroxybutyrate may inhibit leucine metabol ism by different mechan isms. T h e s e f indings support the hypothesis that M C F A s and ketones alter leucine metabol ism in astrocytes. They may have implications for the understanding of integrated fuel metabol ism within the brain and for the mechan ism of act ion of the ketogenic diet. T A B L E O F C O N T E N T S T A B L E O F C O N T E N T S A B S T R A C T T A B L E O F C O N T E N T S iv L IST O F T A B L E S vi L IST O F F I G U R E S viii L IST O F A B B R E V I A T I O N S ix A C K N O W L E D G E M E N T S x 1 I N T R O D U C T I O N 1 2 L I T E R A T U R E R E V I E W 3 2.1 Pediat r ic Ep i lepsy 3 2.2 Diet Therapy for the Treatment of Pediatr ic Ep i lepsy 4 2.3 Advan tages and D isadvantages of the Ketogenic Diet 8 2.4 Potential Mechan i sms of Act ion of the Ketogenic Diet 10 2.5 T h e Excitatory Neurotransmitter Sys tem in Pediat r ic Ep i lepsy . . .13 2.6 Brain Metabol ism and the Ketogenic Diet 15 2.7 T h e Relat ionship of Leuc ine and Glutamate Metabol ism in Brain. 18 3 S T U D Y O V E R V I E W 26 4 P U R P O S E 26 4.1 Object ives 26 4.2 Hypothesis 27 5 E T H I C S 27 6 M E T H O D S 28 6.1 Mater ia ls 28 6.2 An ima ls 29 6.3 Astrocyte Preparat ion and Culture 29 6.3.1 T i ssue Dissociat ion 30 6.3.2 R e l e a s e and Separat ion of Ol igodendrocytes 31 6.3.3 Purif ication of Ast rocytes 32 6.4 Col lect ion of 1 4 C 0 2 from [ 1 4 C]-Labe l led Substrates 33 6.4.1 Measurement of [U- 1 4 C] Leuc ine Oxidat ion 34 6.4.2 Measurement of Oxidat ive Decarboxylat ion of [1 - 1 4 C ] -Leuc ine 37 6.4.3 Measurement of the Product ion of [1 - 1 4 C ] - L e u c i n e Der ived a-Keto isocaproate 38 iv T A B L E O F C O N T E N T S 6.5 Ce l l Protein Determination 39 7 D A T A A N A L Y S E S 40 7.1 Data Handl ing and Calcu la t ions 40 7.2 Statist ical A n a l y s e s 42 8 R E S U L T S 43 8.1 Metabo l ism of [U - 1 4 C] -Leuc ine by Ast rocytes 43 8.1.1 Effect of p-Hydroxybutyrate on [U- 1 4 C] -Leuc ine Oxidat ion. 44 8.1.2 Effect of Octanoate on [U- 1 4 C] -Leuc ine Oxidat ion 45 8.2 Metabol ism of [1 - 1 4 C ] - L e u c i n e by Ast rocytes 46 8.2.1 Effect of p-Hydroxybutyrate on the Oxidat ive Decarboxylat ion of [1 - 1 4 C]-Leuc ine 47 8.2.2 Effect of p-Hydroxybutyrate on the Rate of Product ion of a -Keto isocaproate from [1- 1 4 C]-Leuc ine 49 8.2.3 Effect of p-Hydroxybutyrate on the Rate of Net Transaminat ion of [1 - 1 4 C]-Leuc ine 51 8.2.4 Effect of Octanoate on the Oxidat ive Decarboxylat ion of [1 -1 4 C] -Leuc ine 52 8.2.5 Effect of Octanoate on the Rate of Product ion of a -Keto isocaproate from [1 - 1 4 C]-Leuc ine 53 8.2.6 Effect of Octanoate on the Rate of Net Transaminat ion of [1 -1 4 C] -Leuc ine 54 9 D I S C U S S I O N 56 9.1 Inhibition of Astrocyt ic Leuc ine Metabol ism by Octanoate and p-Hydroxybutyrate 56 9.1.1 Effect of p-Hydroxybutyrate on [1 - 1 4 C ] - L e u c i n e Metabol ism in Astrocytes 58 9.1.2 Effect of Octanoate on the Metabol ism of [1 - 1 4 C ] - L e u c i n e in Astrocytes 60 9.2 Hypothes is of the Mechan ism of Inhibition of Leuc ine Oxidat ion by Med ium C h a i n Fatty A c i d s and Ketones 62 9.3 Alternat ive Hypotheses of the Mechan i sm of Inhibition of Leuc ine Oxidat ion by Med ium C h a i n Fatty A c i d s and Ketones 65 9.4 Towards an Understanding of the Ant iepi lept ic Eff icacy of the Ketogen ic Diet 67 9.5 Limitations of the Study 70 9.6 Future R e s e a r c h Direct ions 72 10 B I B L I O G R A P H Y 74 11 A P P E N D I X 87 v LIST O F T A B L E S LIST O F T A B L E S 2.1 Fatty ac id composi t ion of bovine milk 7 2.2 Fatty ac id distribution of medium cha in tr iacylglycerol oil 7 8.1 Effect of addit ional substrates on the production of 1 4 C 0 2 from [U- 1 4 C]- leuc ine in astrocytes 43 8.2 Effect of p-hydroxybutyrate on the product ion of 1 4 C 0 2 from [ U - 1 4 C ] -leucine in astrocytes 45 8.3 Effect of octanoate on the production of 1 4 C 0 2 from [U- 1 4 C]- leuc ine in astrocytes 46 8.4 Effect of p-hydroxybutyrate on the rate of oxidative decarboxylat ion of [1 -1 4 C] - leuc ine in astrocytes cultured in D M E M media 48 8.5 Effect of p-hydroxybutyrate on the rate of oxidative decarboxylat ion of [1 -1 4 C] - leuc ine in astrocytes cultured in P B S 48 8.6 Effect of p-hydroxybutyrate on the rate of production of 1 4 C 0 2 der ived from chemica l decarboxylat ion of a-keto isocaproate in astrocytes cultured in D M E M media 50 8.7 Effect of p-hydroxybutyrate on the rate of production of 1 4 C 0 2 der ived from chemica l decarboxylat ion of a-keto isocaproate in astrocytes cultured in P B S 50 8.8 Effect of p-hydroxybutyrate on the rate of net t ransaminat ion of [1 - 1 4 C ] -leucine in astrocytes cultured in D M E M 51 8.9 Effect of p-hydroxybutyrate on the rate of net t ransaminat ion of [1 - 1 4 C ] -leucine in astrocytes cultured in P B S 52 vi LIST O F T A B L E S 8.10 Effect of octanoate on the rate of oxidative decarboxylat ion of [1 - 1 4 C ] -leucine in astrocytes cultured in P B S 53 8.11 Effect of octanoate on the rate of production of 1 4 C 0 2 der ived from chemica l decarboxylat ion of a-keto isocaproate in astrocytes cultured in P B S 54 8.12 Effect of octanoate on the rate of net t ransaminat ion of [1 - 1 4 C] - leuc ine in astrocytes cultured in P B S 55 vii LIST O F F I G U R E S LIST O F F I G U R E S 2.1 Hepat ic ketone production due to increased fatty ac id oxidation 11 2.2 Schemat ic representat ion of the glutamate-glutamine cyc le 20 2.3 Schemat ic representat ion of leucine metabol ism 22 2.4 Schemat ic representat ion of the research hypothesis 25 7.1 Schemat ic representat ion of the first two steps of leucine metabol ism .. 41 9.1 Schemat i c representat ion of [1- 1 4 C]- leucine oxidation 6 2 viii LIST O F A B B R E V I A T I O N S LIST O F A B B R E V I A T I O N S A E D antiepi leptic drug B C A A branched chain amino ac id B C K A branched chain keto ac id B H B P-hydroxybutyrate D M E M Du lbecco 's modif ied eag le medium D M E M / F 1 2 Du lbecco 's modif ied eag le medium + F12 (Ham) 1:1 D P M disintegrat ions per minute G A B A y-amino butyric ac id G L U glutamate G L N glutamine a - K G a-ketoglutarate a -K IC a -keto isocaproate L C F A long chain fatty ac id L E U leucine M C F A medium chain fatty ac id M C T medium chain tr iacylglycerol P B S phosphate buffered sa l ine T C A tr icarboxyl ic ac id cyc le T P P thiamin pyrophosphate ix A C K N O W L E D G E M E N T S A N D D E D I C A T I O N A C K N O W L E D G E M E N T S A N D D E D I C A T I O N I would like to express my gratitude to the fol lowing persons whose gu idance and support have contributed to the complet ion of this project. Thank you to Dr. She i l a Innis for al lowing me the opportunity to work with her and for gu idance and adv ice a long the way. To Roger Dyer, for his incredible pat ience whi le I was learning the necessary laboratory techniques. To Dr. J im Thompson for a lways being interested and avai lable when I needed to d i scuss aspec ts of my work. To Dr. Zhaoming X u and Dr. Kev in Farrel l for serv ing on my thesis committee. To my partner Todd , and my daughters A y l a and Sad ie , who have provided much love, support and encouragement . I would a lso like to thank The University of Brit ish Co lumb ia and the Gert rude Langr idge Fel lowship Fund for their f inancial support. Th is thesis is dedicated to the memory of Ma rc Lafreniere. His love and support throughout our twelve years of fr iendship, and his unfail ing belief in me have been crit ical to my success . Thank you Marc. x I N T R O D U C T I O N 1 Introduction Epi lepsy is a condit ion character ized by recurrent brain se izures that occurs primarily in chi ldren. How and why se izures are produced is not well understood, but ev idence suggests that high levels of glutamate, the major excitatory neurotransmitter, may be involved. The ketogenic diet is a high fat, low g lucose diet that has been used as a treatment for pediatr ic ep i lepsy for over 75 years (Swink et a l . 1997). The mechan ism by which the ketogenic diet supp resses epi lept ic se izures is not known. Ketogenic diets result in e levated levels of circulating ketones. W h e n medium chain tr iacylglycerols (MCT) are used as the dietary fat, e levated circulat ing concentrat ions of both ketones and medium chain fatty ac ids ( M C F A s ) develop. Ketones, and perhaps M C F A s , pass from the cerebral capi l lar ies into astrocytes where they become the major fuel source when g lucose is limited (Auestad et a l . 1991). It is poss ib le that the change in primary fuel substrate in the astrocyte produces subsequent changes in brain metabol ism, which ultimately lead to a reduction in se izures . C h a n g e s in the metabol ism of the excitatory neurotransmitter glutamate, occurr ing in response to a high fat, low g lucose diet, could be responsib le for the antiepi lept ic eff icacy o f the ketogenic diet. Brain glutamate uptake is negl igible (Grill et a l . 1992) and studies indicate that the branched chain amino ac id leucine is an important source of brain glutamate nitrogen. Yudkoff et a l . (1994a) demonstrated that up to 3 0 % of astrocytic glutamate nitrogen is der ived from leucine; suggest ing that changes to leucine metabol ism will have important consequences for the levels of this excitatory neurotransmitter. There is very little information on integrated fuel metabol ism in astrocytes in the literature. Ketone bod ies have been shown to inf luence some aspec ts of amino ac id 1 I N T R O D U C T I O N metabol ism in astrocytes (Yudkoff et a l . 1997), but the impact of different fuel substrates, such as fatty ac ids and ketones, on the metabol ism of leucine in astrocytes has not been reported. An imal studies have demonstrated that metabol ic consequences of a ketogenic diet include e levated levels of acetyl C o A and a high A T P : A D P ratio (De V ivo et a l . , 1978). High levels of A T P are known to inhibit some enzymat ic react ions including the rate-limiting step of leucine oxidation, which is cata lyzed by branched chain ketoacid dehydrogenase (Lehninger et a l . , 1993). Increased acetyl C o A from fatty ac id and ketone oxidation would a lso be expected to inhibit this step in leucine metabol ism (Lehninger et a l . , 1993). Inhibition of b ranched chain ketoacid dehydrogenase shou ld cause an accumulat ion of a-keto isocaproate (a-KIC), the cognate keto-acid of leucine. Increased levels of a -KIC are known to cause a reversal of the reaction cata lyzed by branched chain amino ac id t ransaminase, with the transfer of an amino group from glutamate to a-KIC result ing in the formation of leucine and a-ketoglutarate (a-KG) (Yudkoff et a l . 1994a). Ultimately, a reversal of the t ransaminase reaction results in dec reased astrocyt ic glutamate and glutamine (Yudkoff et a l . 1994a and Yudkoff et a l . 1996b). Th is study examines the hypothesis that fatty ac id and ketone metabol ism will inhibit complete oxidation of leucine in astrocytes by interfering with the rate-limiting step of leucine catabol ism, cata lyzed by branched chain ketoacid dehydrogenase. 2 L I T E R A T U R E R E V I E W 2 LITERATURE REVIEW 2.1 Pediatric Epilepsy Epi lepsy is a condit ion character ized by recurrent brain se izures , which usual ly deve lops in chi ldhood. Se izu res are the result of abnormal and excess ive d ischarg ing of the neurons and can be accompan ied by alterations in sensat ion, behavior or consc iousness (Freeman, 1995). Se izu res can be c lassi f ied into five major types: absence , myoclonic, genera l ized tonic-clonic, partial onset, and others. Ch i ldhood se izures can be precipitated by a neurological insult, such as infection or t rauma, but in most c a s e s they are idiopathic. A lmost 1 % of all chi ldren will deve lop epi lepsy by the age of fifteen (Annegars, 1993). Ch i ld ren are bel ieved to be more suscept ib le to se izures than adults because developing neurons tend to be more exci table (Johnston, 1996). The mechan isms by which se izures are produced in the epi lept ic brain are not fully understood. The excitability of neurons may be related to the level of glutamate, the major excitatory neurotransmitter. The epi lept ic brain appears to have increased levels of this amino ac id , which cou ld be due to increased production, dec reased catabol ism, or both. Ant iepi lept ic drugs often act at the level of glutamate by compet ing with it for its receptor or by mimicking the effects of the inhibitory counterpart y-aminobutyric ac id ( G A B A ) (Chapman, 2000). The word ep i lepsy is der ived from the Latin word epilepsius which literally means "a taking hold" (Shafer and Sa lmanson 1997). Th is implies that the person affected by ep i lepsy is overwhelmed by a "mysterious, supernatural power" (Shafer and Sa lmanson , 1997). Th is literal meaning is embedded in the st igma faced throughout history by those affected with this d i sease . The impact of ep i lepsy on the chi ld and 3 L I T E R A T U R E R E V I E W family can be severe and is related to the physica l effects of se izures , treatment-related effects, and soc ia l implications. The standard mortality rate for people with ep i lepsy is two-four t imes higher than normal, with c a u s e s of mortality including the direct effects of se izures and status epi lept icus, acc idents occurr ing during a se izure, and su ic ide (Guberman and Bruni, 1999). S o m e addit ional r isks assoc ia ted with ep i lepsy are lower I.Q., learning disabi l i t ies, and mental retardation. Behaviora l problems, such as anxiety and aggress ion are a lso common in epi lept ic chi ldren. Cogni t ive and behavioral problems are partially due to underlying pathology in the central nervous system, but are exacerbated by the effects of recurrent se izures (P rasad et a l . , 1996). Ant iepi lept ic drugs may a lso contribute to cognit ive and behavioral problems, such as an inability to concentrate (Dodson, 1993). Convent iona l treatments for ep i lepsy include a variety of antiepi lept ic drugs (AEDs) , and surgery. Despi te the cont inued development of new drugs, it has been est imated that from 20 -30% of chi ldren have se izures that are not fully respons ive to any of the avai lab le A E D s , or exper ience intolerable s ide effects from drug therapy (Whe less , 1995). Alternative treatments for ep i lepsy include the ketogenic diet, immunoglobul ins and steroids. Of these options, the ketogenic diet is the only one with proven eff icacy for the treatment of pediatr ic ep i lepsy (Prasad et al . 1996). 2.2 Diet Therapy for the Treatment of Pediatric Epilepsy Dietary interventions such as fast ing were recommended for the control of se izu res as far back in time as Hippocrates and G a l e n (Prasad et al . , 1996). Var ious anecdota l reports throughout history have supported the notion that fast ing inhibits epi lept ic se izures. The impracticality of using fast ing as a long-term treatment for ep i lepsy led 4 L I T E R A T U R E R E V I E W physic ians to try to understand how food restriction could inhibit se izures . In the early part of this century Geye l in (1921) suggested that beneficial effects of fast ing might be related to the result ing ac idos is . In the s a m e year, Wi lde r (1921) not iced that fast ing led to an increased level of circulat ing ketones (acetoacetate and p-hydroxybutyrate) and specu la ted that they might somehow be related to the se izure control. Both of these phys ic ians were c lever enough to real ize that a diet high in fat and low in carbohydrate would provoke a physio logical response similar to that induced by complete fast ing. The c lass i c ketogenic diet was thus des igned to produce ketosis and ac idos is , thereby mimicking the effects of fast ing, while still providing adequate protein and calor ies for growth. It did so by providing a ratio of 3:1 or 4:1 fat to carbohydrate plus protein in the diet. In the period fol lowing the introduction of the ketogenic diet, and prior to the development of pharmacologica l therapies, many phys ic ians used this diet with a high degree of s u c c e s s (Peterman, 1925; Helmholz, 1927). The c lass i c ketogenic diet is composed mainly of dairy fats, whi le carbohydrate is severe ly restricted and only the minimum amount of protein needed to support growth is included (K insman et al . , 1992). Dairy fats contain some short and M C F A s , and large amounts of long cha in saturated fatty ac ids (Table 2.1). Fatty ac ids are hydrocarbon cha ins varying in length from 4-36 carbons with a carboxyl group ( C O O H ) at one end and a methyl group ( C H 3 ) at the other. Long chain fatty ac ids ( L C F A s ) have greater than twelve carbons and are insoluble in water. A s a result of their insolubility, L C F A s must be transported through the circulat ion in l ipoproteins, complexes in which tr iacylglycerols form an internal core that is so lubi l ized by phosphol ip ids and speci f ic apoproteins. M C F A s have between six and twelve carbons. M C F A s have unique propert ies related to their short chain length and resultant water solubility. Thei r 5 L I T E R A T U R E R E V I E W digest ion is s impler than for L C F A s because they do not require the compl icated system to transport the water insoluble L C F A s . Unl ike L C F A s , they eas i ly p a s s through cel l membranes and enter the mitochondria independent of the carnit ine acyl t ransferase system (Bach and Babayan , 1982). It is a lso poss ib le that, unl ike L C F A s , M C F A s c ross the blood brain barrier and enter astrocytes. W h e n it was d iscovered that a diet high in M C T s resulted in a greater elevat ion of p lasma ketones than dairy fats, a M C T vers ion o f the diet was deve loped (Huttenlocher, 1976). The M C T diet provides fat as M C T oil (Table 2.2), composed of tr iacylglycerols containing mainly the M C F A s octanoate (8:0) and decanoate (10:0). B e c a u s e these M C T s are more ketogenic than dairy fats, a higher amount of carbohydrate can be incorporated into the diet, thereby improving its palatabil i ty and acceptabil i ty. The s u c c e s s rates of the c lass ic and M C T diets in control l ing se izures , however, have been reported to be very similar (Schwartz et a l . , 1989). U s e of the ketogenic diet dec reased fol lowing the development of ant i -seizure medicat ions such as phenytoin in the late 1930's. R e s e a r c h related to the diet, wh ich had focused on d iscover ing its mechan ism of act ion a lso dec l ined. S ince that time, many advances have been made in the pharmacologica l and surgical treatment of epi lepsy, but there are still a signif icant number of chi ldren who cannot be helped by these therapies. Between 2 5 - 3 0 % of people with epi lepsy in the United States have se izu res that are unrespons ive to pharmacologica l therapies and only a smal l proportion of patients with ep i lepsy are suitable for surgery (So, 1993 ; Huttenlocher and Hapke 1990). In addit ion, antiepi leptic drugs (AEDs) may cause severe s ide effects such as drowsiness, headaches , cognit ive impairment, ataxia and depress ion 6 L I T E R A T U R E R E V I E W (Guberman and Bruni, 1999) that make these drugs intolerable to the chi ld or their family. Table 2.1: Fatty acid composition of bovine milk 1 Fatty Acid Weight % 4:0 3.32 6:0 2.34 Z C<6 5.66 * 8 : 0 1.19 * 10:0 2.81 12:0 3.39 14:0 11.41 14:1 2.63 16:0 29.53 16:1 3.38 18:0 9.84 18:1 27.39 18:2 2.78 E C > 1 2 90.35 1 Adap ted from J e n s e n and Newburg (1995) E = sum of fatty ac ids of carbon cha in length (C) * Compare to Table 2.2 Table 2.2: Fatty acid distribution of medium chain triacylglycerol o i l 1 Fatty Ac i d % by weight E < C 6 < 6 8:0 60-80 10:0 18-32 I >C12 <4 1 A s manufactured by M e a d Johnson (Bel levi l le, Ontario) I , sum of fatty ac ids of carbon chain length (C) 7 L I T E R A T U R E R E V I E W Recent ly, there has been a resurgence of interest in the ketogenic diet, partly because of increased media attention to this form of treatment, and partly due to the increasing popularity of alternative medical therapies (Whe less , 1995). Despi te the interest, the increasing cl inical use of the ketogenic diet, and its long history of ef fect iveness, surprisingly little is known about how it works, or when it should be implemented. Th is is understandable cons ider ing the fact that research into the mechan ism of act ion of the ketogenic diet was virtually non-existent between 1966 and 1990 (Stafstrom, 1999). It is c lear that ketogenic diets provide an important treatment alternative in c a s e s where ep i lepsy does not respond to A E D s , or when the s ide effects of A E D s are intolerable. U s e of the ketogenic diet in less severe situations has yet to be evaluated. Further research is warranted to clarify the role of the ketogenic diet in the management of pediatr ic epi lepsy. 2.3 Advantages and Disadvantages of the Ketogenic Diet The ketogenic diet is thought to be benef ic ial for between 1/3 and 2/3 of chi ldren with intractable epi lepsy. In 1972, Liv ingston reported that 5 2 % of 1001 patients who were treated with the ketogenic diet had their se izures completely control led, and 27 % had a signif icant improvement in se izure f requency (Livingston, 1972). In December of 1998, researchers at the Johns Hopkins Med ica l Institute reported results of a prospect ive study of 150 chi ldren with intractable ep i lepsy (Freeman et a l . , 1998). The chi ldren were treated with a ketogenic diet and se izure f requency was measured. A signif icant number of chi ldren were able to successfu l ly fol low the ketogenic diet for a full year and 2 7 % of these had at least a 9 0 % reduction in se izure frequency. T h e s e 8 L I T E R A T U R E R E V I E W chi ldren had previously tried an average of six A E D s without success . Despi te the prevai l ing opinion that the ketogenic diet should only be used as a last resort, it actual ly has a higher rate of s u c c e s s than many of the avai lable drug therapies. A n A E D is cons idered effective if it reduces se izures by 5 0 % in half of the recipients; studies of the recently deve loped A E D s indicate that none are able to ach ieve this level of se izure control (Chadwick, 1997; Faught, 1997). In the prospect ive study at Johns Hopkins, 75 of the chi ldren (50%) maintained a 5 0 % reduction in their se izures after one year, and 41 of these (27%) ach ieved better than 9 0 % reduction in se izures (Freeman et a l . , 1998). Recent ly , a systemat ic review of all publ ished results regarding the eff icacy of the ketogenic diet for the treatment of pediatr ic ep i lepsy was conducted (Leferre and Aronson , 2000). Th is ana lys is showed that the ketogenic diet results in complete cessat ion of se izu res in 16% of cases , more than 9 0 % reduction in se izure f requency in 3 2 % of c a s e s and more than 5 0 % reduct ion in se izure f requency in 5 6 % of cases . T h e authors conc luded that despi te the absence of control led trials, the ev idence to date strongly supports the eff icacy of the ketogenic diet for intractable pediatr ic epi lepsy. If the ketogenic diet is so effective in inhibiting se izures , why is it not more widely u s e d ? Many textbooks on epi lepsy do not even mention the ketogenic diet, and when it is ment ioned its va lue is general ly minimized. O n e reason for the hesitation to recognize it as a val id form of treatment could be the lack of knowledge about its mechan ism of act ion. It must be noted, however, that very little is known about how some A E D s prevent se izures as well . Another reason may be that the diet has a reputation for being unpalatable and difficult to prepare. It is an extremely restrictive diet and can be difficult for parents or caregivers to manage. Wi th proper support and training this difficulty can be minimized (Freeman et a l . , 1994) and some parents appreciate the opportunity to be 9 L I T E R A T U R E R E V I E W more involved in the management of their chi ld 's i l lness (Whe less , 1995). F reeman et a l . (1998) found that ef fect iveness was the most important factor determining whether a chi ld would remain on the diet. In their study, the probabil ity of chi ldren remaining on the diet after one year was 8 0 % for those whose se izures were reduced by more than 50%. Chi ldren who d iscont inued the diet did so not because it was unpalatable or difficult to prepare, but rather because it was not working. A s with any therapy, the ketogenic diet does have potential s ide effects that may limit its tolerability. Short-term compl icat ions, normally surfacing within a month of diet initiation may include dehydrat ion, hypoglycemia, diarrhea, vomiting and refusal to eat (Freeman and Vin ing, 1994). Long-term compl icat ions may include urolithiasis, e levated cholesterol , irritability and metabol ic d is turbances such a s ac idos is (Vining et a l . , 1996; Schwar tz et a l , 1989; Herzberg et a l . , 1990). Vi tamin supplements are required to prevent def ic iencies. There have been reports of subjective improvements in the cognit ion and behavior of chi ldren on ketogenic diets, some of which may be due to a reduction in their A E D s (Nigra et a l . , 1995). R e s e a r c h is needed to properly clarify the effects of the diet on cognit ion and behavior (P rasad et a l . , 1996). 2.4 Potential Mechanisms of Action of the Ketogenic Diet There is a ser ious lack of information about how the ketogenic diet works. Ear ly hypotheses suggested effects related to ketosis, ac idos is , hydration, elevat ion of serum lipids and electrolyte imbalance. Schwartzkro in (1999) recently summar ized updated hypotheses including effects on the nature of brain metabol ism, dec reased excitability due to alterations in cel l properties, effects on neurotransmitters and synapt ic t ransmiss ion, impact on neuromodulat ing "circulating factors" such as insulin, and 10 L I T E R A T U R E R E V I E W changes to the extracel lular mil ieu. Whatever the exact mechan ism, it is c lear that the ketogenic diet must ultimately impact on brain neurotransmitter metabol ism. It is known that when there is a shortage of g lucose, fatty ac ids are rapidly ox id ized in the liver and high levels of acetyl C o A are generated (Owen et a l . , 1967). Rap id production of acetyl C o A results in ace ty l -CoA concentrat ions that exceed the capaci ty of the T C A cyc le (Figure 2.1). Figure 2.1 Hepatic ketone production due to increased fatty acid oxidation G l u c o s e HIGH FAT / LOW GLUCOSE DIET Pyruvate Fatty acid oxidation in Liver TT Acetyl CoA Oxaloacetate Citrate a-ketoglutarate Isocitrate Fumarate Succ ina te S u c c i n y l - C o A 11 L I T E R A T U R E R E V I E W A s a result, ace ty l -CoA condenses to form the ketone bodies, acetoacetate and p-hydroxybutyrate, which are re leased from the liver. During high fat diets, p lasma ketone levels r ise dramatical ly. During starvation, the dec rease in p lasma insulin and increase in g lucagon results in the mobil izat ion of ad ipose t issue fatty ac ids. The fatty ac ids are taken up and rapidly ox id ized in the liver with the generat ion of ketones. It is well known that the brain is ab le to util ize ketones for energy and does so preferentially when g lucose is in short supply (Auestad et a l . , 1991; Bixel and Hamprecht, 1995). In addit ion, the capaci ty to extract ketones from the cerebral capi l lar ies and use them as a fuel source is particularly high in the young brain (Nehl ig, 1999). The antiepi leptic effect of the ketogenic diet is bel ieved to be related to the switch from g lucose to ketones and fatty ac ids as the primary energy fuel (Nordli and De V ivo, 1997 and P rasad et a l . , 1996). O n e of the most popular theories predicts that a direct or indirect effect of the ketones, acetoacetate and p-hydroxybutyrate, is responsib le for the ant i-convulsant action of the ketogenic diet (Huttenlocher, 1976). Others have suggested that increased p lasma levels of the M C F A s octanoate and decanoate may be directly involved (Si l ls et a l . , 1986a). Resul ts of severa l studies have indicated that se izure control is not necessar i ly correlated with p lasma ketone or M C F A concentrat ions (Schwartz et a l . , 1989; Si l ls et a l . , 1986a). It must be acknowledged that the very short half-life of these subs tances makes it difficult to accurately measure them, and p lasma concentrat ions are not necessar i ly indicative of turnover. Further, the concentrat ions of ketones and M C F A s in the peripheral circulat ion may not reflect their levels in the central nervous 12 L I T E R A T U R E R E V I E W system. A recent study by Bough et al . (1999) demonstrated se izure control in calor ie restricted, but non-ketot ic rats, implying that ketosis is not necessary for se izure control. Ear ly animal studies showed that ingest ion of a ketogenic diet resulted in e levated blood ketone levels and increased res is tance to induced se izures (Uhlemann and Neims, 1972; Apple ton and De Vivo, 1974). Us ing an animal model o f t he ketogenic diet, De V ivo et a l . (1978) showed that rats fed high fat diets had an increased threshold for e lectroconvuls ive shock. Thei r results indicated that something about the change to using fat as the major fuel source conferred upon the rats an increased res is tance to se izures . They hypothes ized that the increased A T P : A D P ratio, assoc ia ted with chronic ketosis improves neuronal stability, thereby preventing se izures . Subsequent animal studies have conf irmed that the ketogenic diet ra ises the se izure threshold (Nakazawa et a l . , 1983) and have demonstrated the effect to be particularly robust in young an imals (Uhelmann and Neims, 1972; Bough et a l , 1999). 2.5 The Excitatory Neurotransmitter System in Pediatric Epilepsy: Most se izure problems begin in chi ldhood and a greater variety of se izure types are seen in chi ldren, than in adults (Johnston, 1996). Th is ra ises the quest ion of what it is about the immature brain that makes it particularly suscept ib le to se izures . Exper imental ev idence suggests that the development of ch i ldhood epi lepsy may be related to unique aspec ts of the excitatory neurotransmitter system during brain development. Glutamate is the major excitatory neurotransmitter in the mammal ian brain (Er ic inska and Si lver, 1990). There are two types of receptors for glutamate, the metabotropic receptors, which are l inked to second messenger systems, and ionotropic receptors which are assoc ia ted with ion channe ls (Johnston, 1996). E n h a n c e d activity 13 L I T E R A T U R E R E V I E W of both types of glutamate receptors has been demonstrated in the develop ing brain as compared to the adult brain (Blue and Johnston, 1995; Nicoletti et a l . , 1986). For instance, the N-methyl-D aspartate (NMDA) receptor is more respons ive to glutamate during the post-natal period than it is later in life. It has a lso been shown that glutamate binding si tes on the receptors are more numerous and they are less easi ly inhibited in young animals (McDona ld and Johnston, 1990; Tremblay et a l . , 1988). Th is enhanced activity of glutamate receptors in the young brain may explain why ep i lepsy is most common in chi ldhood. It may a lso provide c lues as to why some types of therapies, including the ketogenic diet, are more effective in chi ldren than in adults. A s d i scussed previously, animal studies have demonstrated a higher degree of se izure protection and higher blood ketone levels in young animals ingesting a ketogenic diet than in older an imals (Uhlemann and Neims, 1972 and Bough et al . 1999). Ev idence to support the theory that epi lept ic se izures are related to the excitatory neurotransmitter system has been found in both animal and human studies. Exper iments us ing young rats have demonstrated that se izures and excitotoxic brain injury can be induced by injecting ana logues of glutamate (McDona ld et a l . , 1992). Similarly, overst imulation of glutamate receptors has been shown to produce se izures and even to lead to long-term changes that resemble those s e e n in chronic epi lepsy. Signif icantly e levated concentrat ions of glutamate have been found in brain t issue removed from humans during surgical treatment of ep i lepsy (Sherwin et a l . , 1988). Measurement of brain t issue removed from chi ldren with ep i lepsy revealed spontaneous bursts of electr ical activity (Wuar in et a l . , 1990; Avol i and Ol iv ier 1987). T h e s e spontaneous bursts of activity cou ld be inhibited by using a competit ive inhibitor of the N M D A glutamate receptor (Wuar in et a l . , 1990). Pharmaco log ica l therapies have 14 , L I T E R A T U R E R E V I E W been deve loped based on the results of these and other experiments. The cumulat ive ev idence suggests that the excitatory neurotransmitter system, and in particular glutamate, is involved in the product ion of se izures . B e c a u s e it is known that the excitatory neurotransmitter system is more act ive in chi ldren and exper ience has shown that chi ldren respond best to a ketogenic diet, we can hypothesize that the ketogenic diet is act ing at the level of glutamate metabol ism. 2.6 Brain Metabolism and the Ketogenic Diet The fol lowing summary of the metabol ic consequences of a diet high in fat and low in carbohydrate and protein provides the necessary background to support a hypothesis that the ketogenic diet inf luences brain glutamate metabol ism. During consumpt ion of a high fat diet, there is a high level of oxidation of the dietary fatty ac ids in the liver. Rap id oxidation of fatty ac ids results in the generat ion of ace ty l -CoA in amounts that exceed the capaci ty of the T C A cyc le for metabol ism (Figure 2.1). W h e n this occurs, the excess acetyl C o A condenses to form the ketone bodies, acetoacetate and p-hydroxybutyrate, which are re leased from liver result ing in a r ise in p lasma ketones. In the c a s e of the M C T diet, serum octanoate and decanoate levels a lso rise (Si l ls et a l . , 1986b), reflecting at least in part the transport of these M C F A s from the gastrointestinal tract to the liver v ia the portal vein. In the liver, M C F A s have multiple fates. They can be ox id ized to ace ty l -CoA which is converted to ketones, used for de novo synthesis of longer cha in fatty ac ids, or ox id ized in the T C A cyc le to CO2. S o m e unesteri f ied M C F A s bypass the liver and consequent ly a signif icant amount of octanoic and decano ic ac id is present in the peripheral b lood, and potentially avai lable for uptake by the brain (Fernando-Warnaku lasur iya et a l . , 1981). 15 L I T E R A T U R E R E V I E W Energy substrates, including ketones and perhaps M C F A s , enter the brain by pass ing from the cerebral capi l lar ies to the astrocytes. Astrocytes are s tar -shaped cel ls posi t ioned between the cerebral capi l lar ies and the neurons. They can be thought of as process ing plants, which take up a variety of nutrients from the blood and convert them to substrates that can be used by neurons (Bixel and Hamprecht, 1995). S o m e metabol ic s teps such as the convers ion of glutamate to glutamine, are located exclusively in astrocytes (Norenberg, 1979). Th is div is ion of metabol ic p rocesses is referred to as compartmentat ion and it means that diet- induced changes in the uptake and metabol ism of nutrient substrates by astrocytes can have subsequent effects on neuronal metabol ism (Daikhin and Yudkoff, 2000). In the c a s e of the ketogenic diet, the nutrients being taken up by the astrocytes are primarily ketone bodies and potentially M C F A s . Ast rocytes in culture can oxid ize the M C F A s with the generat ion of ketones that are exported to the neurons (Auestad et a l . , 1991). The ability of astrocytes in culture to oxid ize octanoate is unique among the cel ls of the brain (Edmond et a l . , 1987). With in astrocytes, ketone bodies can be reconverted to acetyl C o A and enter the T C A cyc le for the production of energy. Alternatively, ketones can be exported to meet neuronal energy needs. The most important metabol ic result of the switch to fats from g lucose as the primary fuel source is that rapid oxidation of fat p roduces large amounts of acetyl C o A and ketones. The impact of the increased availabil i ty of acetyl C o A and ketones on the metabol ism of other fuels, particularly amino ac ids, has not been fully estab l ished. An imal studies have provided some insight into the metabol ic consequences of ingesting a ketogenic diet. De V ivo et a l . (1978) fed rats a high fat, low g lucose diet and measured levels of cerebral metaboli tes. Chronica l ly ketotic rats had increased cerebral 16 L I T E R A T U R E R E V I E W levels of g lucose-6-phosphate, lactate, pyruvate, p-hydroxybutyrate, ct-ketoglutarate, a lanine and citrate, and dec reased levels of fructose 1,6-diphosphate and aspartate. Cerebra l energy reserves were signif icantly higher in the ketotic rats, as reflected by a high A T P A D P ratio. It was suggested that the high levels of A T P inhibited key enzymes, including pyruvate dehydrogenase, and a-ketoglutarate dehydrogenase with subsequent accumulat ion of pyruvate and a-ketoglutarate. Higher levels of lactate probably reflected a diversion of pyruvate through lactate dehydrogenase, exp la ined by the inability of pyruvate to proceed to acetyl C o A and enter the T C A cyc le. Increased levels of pyruvate can be expected to shift the fol lowing react ion to the right: pyruvate + glutamate a - K G + alanine, consistent with the e levated levels of a lanine and a - K G . Thei r results indicated that g lycolys is was inhibited and normal activity of the T C A cyc le was maintained. The most signif icant f inding was increased cerebral energy reserves in the ketotic rats, as reflected by a high A T P : A D P ratio. It was suggested that the high levels of A T P inhibited key enzymes, leading to signif icant alterations in the levels of T C A cyc le intermediates (De V ivo et a l . , 1978). The metabol ic changes observed in the study by De V ivo et a l . (1978) reflect b iochemical changes in brain that result from a diet with fat as the major fuel source. The increased amounts of acetyl C o A from oxidation of ketone bod ies in the brain inhibited glycolysis, and changes in the concentrat ions of T C A cyc le intermediates were the result of the increased A T P : A D P ratio. The results of this study support the hypothesis that util ization of ketones as an energy substrate alters brain intermediary metabol ism. However, because these studies involved ana lys is of whole brain, no information on the metabol ic changes at the level of the astrocyte is provided. 17 L I T E R A T U R E R E V I E W Resea rch into the metabol ic consequences of a ketogenic diet, which incorporate current knowledge of the interactions between astrocytes and neurons is needed. Yudkoff et a l . (1997) have demonstrated that ketone bodies can inf luence amino ac id metabol ism in cultured astrocytes. Astrocyt ic transaminat ion of glutamate to aspartate was inhibited, glutamine levels dec reased and citrate concentrat ions increased in response to 5 m M acetoacetate or p-hydroxybutyrate (Yudkoff et a l . , 1997). The latter f indings provide support for the hypothesis that the oxidation of ketones ultimately affects neurotransmitter levels v ia alterations in cel l metabol ism. However, Yudkoff et a l . (1997) used a s ingle concentrat ion of ketone bodies (5.0 mM) likely to be severa l fold higher than the physio logical concentrat ion ach ieved in the brain, even during ketogenic diet therapy. No information on the poss ib le effects of fatty ac ids, including M C F A s and L C F A s , on the oxidation of amino ac ids in astrocytes has been publ ished. The relat ionship between glutamate and the branched cha in amino ac id leucine is extremely important in brain metabol ism and does not appear to have been cons idered in regards to the metabol ic effect of the ketogenic diet. The fol lowing sect ion descr ibes the interaction between the metabol ism of leucine and glutamate. It provides the background needed to rat ional ize the need for research to determine the impact of different fuel substrates, such a s ketones and M C F A s on leucine metabol ism in the brain. 2.7 The Relationship of Leucine and Glutamate Metabolism in the Brain A s d i scussed previously, glutamate is the major excitatory neurotransmitter in the brain. Glutamate in the brain has one of the fol lowing fates, incorporation into 18 L I T E R A T U R E R E V I E W protein, oxidation or use as a neurotransmitter. The brain der ives very little, if any glutamate or glutamine directly from the bloodstream, in fact a net efflux of glutamine from astrocytes to the cerebral capi l lar ies has been demonstrated (Grill et a l . , 1992 Smith et a l . , 1987). Th is is a bel ieved to be a protective mechan ism des igned to keep intracellular glutamate concentrat ions low, thereby preventing cel lu lar damage and facil itating synapt ic t ransmission (Huang et a l . , 1997). Glutamine, syn thes ized in the astrocytes from glutamate by the act ion of glutamine synthetase, is re leased to the neurons where it serves as the precursor to the neurotransmitter glutamate (Figure 2.2). Fol lowing neurotransmission, glutamate is taken up by the astrocytes again, complet ing the "glutamine-glutamate cyc le" (Mar t inez-Hernandez et a l . , 1977). B e c a u s e glutamate transport into the brain is negl igible, a means of replenishing glutamate lost to oxidation and protein synthesis is required. 19 L I T E R A T U R E R E V I E W Figure 2.2 Schematic Representation of the Glutamate-Glutamine Cycle 20 L I T E R A T U R E R E V I E W Yudkoff et a l . (1990) have demonstrated that a signif icant proportion of the glutamate in neurons is the result of t ransaminat ion of leucine and the other branched cha in amino ac ids, iso leucine and val ine within astrocytes. In the brain, glutamate carbon is der ived from g lucose, and nitrogen is primarily der ived from the branched chain amino ac ids (Daikhin and Yudkoff, 2000). Leuc ine uptake from the cerebral capi l lar ies into astrocytes exceeds that rate of uptake of all other amino ac ids (Smith et al . 1987). The metabol ism of leucine results in the formation of glutamate, and ketones which are then ox id ized in the T C A cyc le (Figure 2.3). Yudkoff et a l . (1994a) est imated that 2 5 - 3 0 % of astrocyt ic glutamate/glutamine nitrogen is der ived from leucine a lone. In these experiments, other amino ac ids that could have provided nitrogen for glutamate synthesis were avai lable to the astrocytes, demonstrat ing that the ce l ls used leucine preferentially. Th is suggests that alterations in the metabol ism of leucine will have important consequences for the levels of glutamate/glutamine in brain. The metabol ism of leucine occurs in three steps (Figure 2.3). The first step is transaminat ion with a-ketoglutarate (a-KG) to yield glutamate and a -keto isocaproic ac id (a-KIC). Th is t ransaminat ion is reversible but under normal condit ions the production of glutamate is favored. The second step involves the oxidative decarboxylat ion of a-KIC to isovaleryl C o A . Th is step, cata lyzed by the mult ienzyme complex "branched chain ketoacid dehydrogenase" ( B C K A dehydrogenase) , is the rate-limiting step in leucine oxidation and is irreversible. The final phase of leucine metabol ism involves a ser ies of react ions that ultimately result in the production of the ketone acetoacetate. The acetoacetate is then converted to acetyl C o A which can enter the T C A cyc le , with complete oxidation of leucine result ing in the product ion of CO2. 21 L I T E R A T U R E R E V I E W Figure 2.3 Schematic representation of leucine metabolism -Ketoglutarate Glutamate (x-Ketoisocaproate LOUCJnG Aminotransferase BCKA dehydrogenase Isovaleryl-CoA 1 i Acetyl-CoA ^ Acetoacetyl-CoA 22 L I T E R A T U R E R E V I E W The direction of the transaminat ion reaction between leucine and a - K G is control led by the levels of the substrates and the energy needs of the cel l (Lehninger et a l . , 1993). In the reverse direction (a -KIC + Glutamate H> a - K G + Leucine) glutamate is consumed, raising the possibi l i ty that a reversal of this react ion may be an important way of modulat ing the level of glutamate in the brain (Yudkoff, 1997). It is poss ib le that the changes in brain metabol ism, occurr ing in response to the ketogenic diet, affect one or more of the steps in leucine metabol ism, with the end result being a reduction in brain glutamate. T h e enzyme complex B C K A dehydrogenase, which ca ta lyzes the rate-limiting step of leucine oxidation, is strikingly similar to the pyruvate dehydrogenase complex, which cata lyzes the decarboxylat ion of pyruvate (Yudkoff, 1997). Pyruvate dehydrogenase is known to be inhibited by ace ty l -CoA and a high A T P (Lehninger et a l . , 1993). Further, the metabol ic changes demonstrated in the brain of ketotic rats (De V ivo et a l . , 1978) suggest inhibition of this enzyme. It is poss ib le that the c lose ly-homologous B C K A dehydrogenase complex may a lso be inhibited by the high A T P : A D P ratio and high ace ty l -CoA levels generated from ketone and M C F A oxidation. If so, consumpt ion of a ketogenic diet with subsequent elevat ion of acetyl C o A and A T P : A D P ratio can be expected to inhibit the rate-limiting B C K A dehydrogenase step in leucine metabol ism. Th is inhibition would cause a -K IC and glutamate to accumulate, driving the react ion in the reverse direction, with subsequent consumpt ion of glutamate. Yudkoff et a l . (1994b) have demonstrated that providing astrocytes with a high concentrat ion of a -KIC results in a reversal of the transaminat ion reaction, and increased oxidation of the a - K G formed. The net result of this is a signif icant reduction in astrocyt ic glutamate and glutamine (Yudkoff et a l . , 1994b; Yudkoff et a l . , 1996b). Z ie lke and co l leagues (1997) have a lso shown an increased rate of glutamate oxidation in response to e levated 23 L I T E R A T U R E R E V I E W extracel lular a -K IC concentrat ions in vivo us ing microdialysis. Ultimately, a dec rease in export of glutamine to the neurons will lead to a reduction in neuronal synthes is of the excitatory neurotransmitter glutamate. Th is ser ies of events could provide the key explanat ion for the mechan ism of act ion of the ketogenic diet (Figure 2.4). Prev ious studies have demonstrated the important link between glutamate and leucine metabol ism in astrocytes, but the impact of dietary fat-derived fuel substrates (ketones, octanoate) on this relat ionship has yet to be determined. In particular, the impact of using fatty ac ids or ketones as the primary fuel substrate in astrocytes on the metabol ism of leucine has not been examined. 24 L I T E R A T U R E R E V I E W Figure 2.4: Schematic representation of the research hypothesis 25 S T U D Y O V E R V I E W . P U R P O S E A N D E T H I C S 3 STUDY OVERVIEW The establ ishment of primary astrocyte cultures and subsequent metabol ic exper iments were conducted at the B C R e s e a r c h Institute for Chi ldren 's and W o m e n ' s Health. The metabol ic studies were done in two phases . The first exper iments involved measur ing 1 4 C 0 2 produced in astrocytes from uniformly 1 4 C label led leucine ( [U- 1 4 C]-leucine) and compar ing the effects of two different fuel substrates, |3-hydroxybutyrate and octanoate, on the rate of oxidation. A second ser ies of exper iments uti l ized [1 - 1 4 C]-leucine to speci f ical ly address the effect of p-hydroxybutyrate and octanoate on the first two steps of leucine metabol ism during which the number one carbon is re leased as 1 4 C 0 2 . ' 4 PURPOSE The purpose of this study was to determine whether or not octanoate and p-hydroxybutyrate inhibit astrocyt ic leucine metabol ism to C 0 2 . 4.1 Objectives • To establ ish primary cultures of cerebral cort ical astrocytes for use in metabol ic studies. • To establ ish a method for the col lect ion of label led products of leucine metabol ism in astrocyte cel l cultures. • To compare the oxidation of [U- 1 4 C]- leuc ine to 1 4 C 0 2 in astrocytes cultured with the M C F A octanoate (8:0), the ketone p-hydroxybutyrate, or g lucose (control) as the primary energy substrate. 2 6 S T U D Y O V E R V I E W , P U R P O S E A N D E T H I C S • To determine the effect of p-hydroxybutyrate on the production of a -KIC from [1 -1 4 C] - leuc ine in astrocytes. • To determine the effect of octanoate on the product ion of a -KIC from [1- 1 4 C]- leuc ine in astrocytes. • To determine the effect of p-hydroxybutyrate on oxidative decarboxylat ion of [1 - 1 4 C]-leucine in astrocytes. • To determine the effect of octanoate on oxidative decarboxylat ion of [1- 1 4 C]- leuc ine in astrocytes. 4.2 Hypothesis A change in primary fuel substrate from g lucose to M C F A s or ketones will inhibit astrocyt ic leucine metabol ism. Speci f ical ly, use of octanoate or p-hydroxybutyrate as the primary fuel source will inhibit the rate-limiting step of leucine metabol ism cata lyzed by the B C K A dehydrogenase complex. Inhibition will be reflected in dec reased 1 4 C 0 2 production from [1- 1 4 C]- leucine concurrent with increased rate of production of leucine-der ived a-keto isocaproate. G lucose , in contrast, will not effect the metabol ism of leucine. 5 ETHICS The study protocol and procedures were approved by The University of Brit ish Co lumb ia Commit tee on An imal Care . 27 M E T H O D S 6 METHODS 6.1 Materials Tissue culture supplies: Fa lcon tubes (Blue Max™ 50 ml and Blue Max™ Jr., 15 ml); f lasks (250 ml) and plates (Mult iwell™6-well) were purchased from Becton Dick inson Labware (Frankl in Lakes , NJ) . Sero log ica l pipettes (2 ml, 10 ml and 25 ml) were purchased from V W R (West Chester , PA) , and Na lgene filters used in media preparat ion were from Na lge Nunc International (Rochester , NY) . Tissue Culture Media: Med ia was purchased from G ibco BRL /L i f e Techno log ies (Grand Island, NY) . Med ia used in the preparat ion and maintenance of astrocyte cultures was Du lbecco 's Modi f ied Eag le Medium: Nutrient Mixture F-12 (Ham) 1:1 #12400, ( D M E M / F 1 2 ) . It was purchased as a powder, which w a s prepared in purified water with 1.2 g of sod ium bicarbonate added per litre, and ster i l ized by filtration us ing Na lgene filters. The media used in exper iments was Du lbecco 's Modi f ied Eag le Med ium #10317, which has a lower concentrat ion of g lucose than D M E M / F 1 2 (5 m M vs 25 mM) and contains no glutamine. The ready-to-use liquid preparat ion was purchased. Chemicals and Reagents: Ster i le reagents such as fetal calf serum, antibiotics, trypsin, and versene were purchased from G ibco BRL /L i f e Techno log ies (Grand Island, NY) . Tr ichloroacet ic ac id and toluene were from F isher Scient i f ic (Nepean, Ontario) and O S C liquid scinti l lation fluid was from Amersham/Sear le (Arlington Heights, Illinois). A l l other chemica ls and media including [U - 1 4 C] -28 M E T H O D S leucine were purchased from S igma (St. Louis, MO) , with the except ion of [1-1 4 C]- leuc ine , which was from ICN Pharmaceut ica ls (Aurora, Ohio). 6.2 Animals Male and female Sprague Dawley rats were purchased from U B C Animal C a r e and maintained in the animal care unit at the B C R e s e a r c h Institute for Ch i ld ren 's and W o m e n ' s Heal th. T h e an imals were housed under s tandard condit ions in a temperature and humidity control led animal room, with ad libitum a c c e s s to Laboratory Rodent Diet #5001 (PMI Feeds , Inc., R ichmond, IN) and water. The animals were bred and newborn rats taken for preparat ion of astrocytes within 48 hours of birth. The timing of this was chosen because at this s tage of brain development the rat brain is particularly enr iched in astrocyt ic ce l ls (Co le and d e V e l l i s , 1989). 6.3 Astrocyte Preparation and Culture Primary cultures of cerebral cortical astrocytes were prepared from the forebrains of newborn Sprague Dawley rat pups ( less than 48 hours old) based on the method of McCar thy and DeVel l i s (1980). Th is procedure can be div ided into three phases : t issue dissociat ion, re lease and separat ion of o l igodendrocytes, and purif ication. Al l instruments and solut ions used in these procedures were pre-ster i l ized. Efforts were made to work asept ical ly and quickly in order to maximize cel l harvest and viability, and prevent contaminat ion. The media used in the preparat ion and maintenance of astrocyte cultures was 29 M E T H O D S Du lbecco 's Modi f ied Eag le Medium:Nutr ient Mixture F-12(Ham) 1:1, ( D M E M / F 1 2 ) . 6 . 3 . 1 T i s s u e d i s s o c i a t i o n : Fol lowing cervical d is locat ion, the brains were removed from the pups. A cut was made from the base of the skul l to the mid-eye a rea and the skin f laps fo lded back, reveal ing the underlying skul l . A n incision was then made through the midline f issure of the skul l by lifting slightly upwards whi le cutting with the sc issors . Bra ins were removed from the skul l cavity with a spatu la and p laced in a 60 mm petri d ish containing steri le D M E M / F 1 2 media with 1 0 % fetal calf serum ( F C S ) and 1% penici l l in/streptomycin, and kept warm with a heating pad. W h e n all the brains had been removed, the d ish was moved to a laminar f low hood for micro-dissect ion in a steri le environment. Us ing forceps, the meninges were careful ly removed from each brain and d iscarded. The cerebral hemispheres were separated and the cort ices pee led off and transferred to a steri le petri d ish containing f resh media. A n y remaining meninges were removed from the cort ices. T h e cort ices were then poured into a steri le nytex bag and the med ia and ce l ls col lected in a 100 mm petri d ish containing 20 ml of media. Wh i l e holding the bag c losed with forceps, and keeping the bag immersed in the media, light strokes of a g lass rod were used to gently push the t issue through the mesh bag. A syr inge fi l led with media was used to careful ly r inse free cel ls adher ing to the bag. The cel l /media suspens ion was then poured through a #60 s ieve into a steri le cup and a syr inge fi l led with media used to wash over the cel ls as they 30 M E T H O D S filtered by gravity. The filtrate from the #60 s ieve was then poured through a #100 s ieve into a second steri le cup and washed with 10 ml of F C S , which helped to loosen any adher ing cel ls. The cel ls with the media and F C S were then transferred to 15 ml steri le plast ic tubes and centr i fuged in an IEC Cent ra -4B centri fuge (Needham Heights, MA) at 800 rpm for 5 minutes. The supernatant w a s poured off and the cel ls resuspended in D M E M / F 1 2 media containing 1 0 % F C S and 1% penici l l in/streptomycin. The ce l ls were counted using a hemacytometer and plated in 75 c m 2 t issue culture f lasks, at a concentrat ion of 1.5 x 1 0 7 ce l ls per f lask (in 9-10 ml of media). The approximate yield was one f lask per brain. Ce l l s were incubated at 37°C for 72 hours without moving the media to a l low time for the astrocytes to adhere to the bottom of the flask. Fol lowing this, the med ia was changed every 48-72 hours. 6.3.2 Release and Separation of Oligodendrocytes: S e v e n to nine days after the initial plating, speci f ic procedures were used to remove o l igodendrocytes and select ively retain astrocytes. First, the media was changed and then the f lasks were secured in the horizontal posit ion to a Lab-L ine Junior Orbit Shake r p laced in the incubator. The cel ls were shaken at 200 rpm for 6 hours, the media containing loose cel ls (ol igodendrocytes, astrocytes and macrophages) was poured off, 9-10 ml of f resh media added, and shak ing cont inued for 18 hours. At the end of this 18-hour period, the med ia was changed aga in and the cel ls were shaken cont inuously for a further 24 hours at 200 rpm. 31 M E T H O D S 6.3.3 Purification of Astrocytes After 48 hours of shak ing, the astrocytes remained in a confluent monolayer on the bottom of the f lask, and the majority of o l igodendrocytes and macrophages were re leased. E n h a n c e d purity of the astrocyte cultures was then ach ieved by further shak ing, fol lowed by a change to the nutrient media. T h e media was rep laced, the cel ls were shaken at 100 rpm for a further 48 hours, and then the media was rep laced with media containing 5%, rather than 10%, fetal calf serum. The cel ls were then maintained at 37°C and the media changed every 48-72 hours. O n e week prior to the metabol ic studies, the ce l ls were replated in 6-well t issue culture plates with D M E M / F 1 2 media containing 5 % F C S , but without antibiotics. P a s s a g i n g was performed by versene-t rypsin treatment as descr ibed by C o l e and de Ve l l i s (1989). Ce l l s were briefly w a s h e d with versene solut ion (2 ml per f lask), fol lowed by wash ing with a 0 .25% trypsin solut ion (1.5 ml per f lask). The trypsin was poured off and the cel ls incubated at 37°C for 5-10 minutes until the confluent layer ran freely upon inversion of the flask, indicating that the cel ls had d issoc ia ted. The cel ls were transferred to a 15 ml conica l tube with 10 ml of D M E M / F 1 2 media and centr i fuged at 800 rpm for 5 minutes. The media w a s then poured off and the cel ls resuspended in antibiotic-free media, and plated in 6-well t issue culture plates with approximately 2 x 1 0 5 cel ls per well (one 75 c m 2 f lask was passaged and replated into one 6-well plate). Viabi l i ty of cel ls used in metabol ic studies was conf irmed with Trypan B lue staining and light microscopy. 32 M E T H O D S 6.4 Collection of 1 4 C 0 2 from [14C]-Labelled Substrates Methods for a s s a y of astrocyt ic leucine metabol ism were deve loped based on publ ished procedures of Aues tad et a l . (1991) and Yudkoff et a l . (1994a). The two-step method used by Aues tad et a l . (1991) to col lect 14C02 from the metabol ism of label led fatty ac ids in astrocytes was initially tested. The method was successfu l ly used to col lect 14C02 from [U- 1 4 C]-octanoate (8:0) and leucine. The inter-assay variabil ity was high, so reagents were modif ied until variabil ity was minimized. The procedure of Aues tad et a l . (1991) used methylbenzethonium hydroxide (hyamine hydroxide) for the col lect ion of CO2, however, only methylbenzethonium chlor ide but not the hydroxide is currently avai lable. Severa l methods of preparing methylbenzethonium hydroxide were attempted and eventual ly, a 0.5 M solut ion in 1 M sodium hydroxide was found to be ideal for the purpose of col lect ing CO2. A d isadvantage of the method of Aues tad et a l . (1991) is that it involves passag ing cel ls with the use of versene and trypsin immediately prior to the metabol ic experiments. The step is necessary to transfer cel ls to g lass vials for the metabol ic studies, but potentially damages cel l integrity, and microscopic evaluat ion of cel ls did suggest changes to the cel l morphology fol lowing passag ing . To avoid the potentially cel l damaging effects of passag ing and unknown effects on metabol ism, the metabol ic studies were attempted in 6-well t issue culture plates as descr ibed by Yudkoff et al . (1994a). The cel ls were passaged and then replated in 6-well t issue culture plates where they were a l lowed 1 week to adapt to their new environment prior to the metabol ic studies. E a c h well se rved as a separate trial in 33 M E T H O D S the experiments. To minimize background counts, addit ional procedures used by Aues tad et a l . (1991) were a lso fol lowed, as descr ibed in detail in sect ion 6.4. Prel iminary studies using octanoate and leucine as the substrates, found that this combinat ion of methods provided the greatest recovery of 1 4 C 0 2 , with the least inter-assay variability, and the lowest background counts. A n addit ional advantage of us ing 6-well plates was that the cel ls were undisturbed, adhered to the bottom of the t issue culture plate, rather than re leased and resuspended in media. 6.4.1 Measurement of [U-14C]-Leucine Oxidation The first ser ies of exper iments measured astrocyt ic 1 4 C 0 2 production from [U-1 4 C] - leuc ine in the p resence of octanoate (0.0, 0.5,1.0 and 5.0 mM), p-hydroxybutyrate (0.0, 0.5, 1.0 and 5.0 mM), and g lucose (addit ional 5.0 mM). The media used in these exper iments was Du lbecco 's Modi f ied Eag le Med ium (DMEM) , which contains approximately 5 m M g lucose, and no glutamine. A n incubation time of 90 minutes and a leucine concentrat ion of 0.2 m M were se lec ted for the exper iments based on previous studies by Yudkoff and co l leagues (1994a). The appropr ia teness of these was conf irmed with time course and concentrat ion studies, the results of which are conta ined in the appendix (Figures A.1 and A.2). Preparation and Preincubation: In preparat ion for the experiments, the media was removed and 0.9 ml of D M E M media was added to each wel l . Then , 20 u.l of 34 M E T H O D S the substrate mix, containing leucine, 10.0 mM; a-ketoglutarate, 10.0 m M ; thiamin pyrophosphate (TPP) , 5.0 mM; coenzyme A, 5.0 m M ; carnit ine, 6.2 m M ; was added to each wel l . Therefore, the final concentrat ions of these metabol i tes in the 1 ml reaction were 0.2 m M leucine, 0.2 m M a -KG, 0.1 m M coenzyme A , 0.1 m M T P P and 0.12 m M carnit ine. Then , the potentially compet ing substrates octanoate, p-hydroxybutyrate were added to produce final concentrat ions of 0.5 mM, 1.0 m M or 5.0 m M in the 1 ml volume. To test whether addit ional g lucose would effect leucine oxidation, g lucose was added to increase the basel ine concentrat ion by 5 mM, again in a final vo lume of 1 ml. Fol lowing substrate and competitor addit ion, the t issue culture plate lids, with bal ls of g lass wool (weighing 0.05 g) g lued to the unders ide of the lids such that a ball of g lass wool was suspended over e a c h wel l , were rep laced. The cel ls were then preincubated for 15 minutes at 37°C. First Incubation: At the end of the preincubat ion, 8 u.l of [U- 1 4 C]- leuc ine (approximately 0.9 u.Ci of [U- 1 4 C]- leuc ine) was added to each wel l , providing a final vo lume of 1.0 ml in each wel l . Us ing a micropipetter, 0.20 ml of 1M N a O H was added to each ball of g lass wool , then the lids were p laced on the plates and the cel ls were incubated at 37°C for 90 minutes. Second Incubation: At the end of the 90-minute incubation, 0.25 ml of 0.5 M H2SO4 was added to each well to stop cel l metabol ism and re lease CO2. The t issue culture d ish lids were returned to their original posit ion and plates 35 M E T H O D S incubated for another 90 minutes at 4°C. The re lease of the 1 4 C 0 2 product from the acidi f ied media at 4°C rather than room temperature reduced the chemica l breakdown of acetoacetate, and resulted in lower background counts (Auestad et a l . , 1991). During this incubation, C 0 2 produced from the oxidative decarboxylat ion of leucine was re leased from the media and col lected in the sodium hydroxide-moistened g lass wool . Third Incubation: At the end of the second incubation, each ball of g lass wool was removed from the lids and transferred to a g lass vial containing 5 ml of H 2 0 . The v ia ls were then sea led with a rubber cap fitted with a suspended centre well containing fluted filter paper. By inserting a needle through the rubber cap us ing a 1-ml hypodermic syr inge, 0.3 ml of 0.5 M hyamine hydroxide was added to the filter paper and 0.5 ml of 5.0 M H 2 S 0 4 was added to the water. The v ia ls were then incubated at 37°C for 30 minutes. Quantification of UC02'. The centre wel l and its contents were then transferred to a 10-ml plast ic scinti l lation vial containing 8 ml of O C S liquid scinti l lation fluid and 2 ml of toluene. The 1 4 C 0 2 col lected in the centre wel ls was then quantif ied using a Beckman Liquid Scinti l lat ion Counter (Fullerton, C A ) . Contro ls (blank react ions), containing no cel ls, were carr ied out concurrent ly in all experiments. Labe led CO2 produced in the control react ions reflected chemica l breakdown of leucine, and was therefore subtracted from the va lue for C 0 2 re lease by the cel ls. E a c h sample was counted at least twice and where the dpm va lues var ied 36 M E T H O D S by more than 5%, the sample was counted a third time. The average dpm for each sample was used in the ana lys is of the data, exc luding erroneous results identified by repeated counting. 6.4.2 Measurement of Oxidative Decarboxylation of [1 -14C]-leucine The next ser ies of exper iments measured astrocyt ic 1 4 C 0 2 product ion from [1-1 4 C] - leuc ine in the p resence of p-hydroxybutyrate (0.0, 1.0 and 5.0 mM) and octanoate (0.0, 0.5, 1.0 and 5.0 mM), and quantif ied [1- 1 4 C]- leuc ine-der ived <x-ketoisocaproate by chemica l decarboxylat ion and measurement of 1 4 C 0 2 f r o m the same trials. In the c a s e of p-hydroxybutyrate two sets of exper iments were performed, the first in low g lucose D M E M / F 1 2 media and the second in phosphate buffered sa l ine (PBS) . Prel iminary studies using low g lucose med ia indicated that 1 4 C 0 2 col lect ion from [1- 1 4 C]- leucine was quite low and signif icant inhibition of oxidat ion by p-hydroxybutyrate was not evident. However,- g lucose concentrat ions in low g lucose media are not actual ly limiting, as 5.0 m M is normal physio logical level. The presence of physio logical levels of g lucose may alter the util ization of p-hydroxybutyrate, or the effects of its metabol ism on leucine. Therefore, addit ional studies with p-hydroxybutyrate and octanoate were done in phosphate buffered sal ine (PBS) . Preparation and Preincubation: A s previously, the media was removed from the wel ls and rep laced with 0.9 ml of P B S or standard media. The leucine substrate mix (20 u.l) and the substrate (P-hydroxybutyrate or octanoate) were added a s 37 M E T H O D S descr ibed in sect ion 6.4 to give a final vo lume of 1.0 ml. Addi t ional preparatory procedures were as descr ibed in sect ion 6.4. Incubations: T h e incubat ions were conducted exactly as descr ibed for exper iments with [U- 1 4 C]- leuc ine in sect ion 6.4 with the except ion that the labeled substrate was [1- 1 4 C]- leucine, 0.9 uC i in 0.9 ul. 6.4.3 Measurement of the Production of [1 -14C]-Leucine-Derived a-Ketoisocaproate Leuc ine-der ived a -keto isocaproate (a -KIC) was measured in the same trials by chemica l decarboxylat ion of a -K IC us ing H 2 0 2 a n d subsequent col lect ion of 1 4 C 0 2 . Fol lowing the second incubation and removal of g lass wool to v ia ls (see above), 0.25 ml of H 2 0 2 was added to each well and new lids with f resh g lass wool containing 0.30 ml of hyamine hydroxide were added. The plates were incubated at 37°C for 30 minutes. Quantification of 14C02. Labe led C 0 2 from the oxidation of [1- 1 4 C]- leuc ine and chemica l decarboxylat ion of [1 - 1 4 C ] - a -K IC was quantif ied by liquid scinti l lation counting. The filter paper and g lass wool were each transferred to 10-ml plast ic scinti l lation vials with 8 ml of O C S liquid scinti l lation fluid and 2 ml of toluene. The 1 4 C 0 2 co l lected in the hyamine hydrox ide-soaked filter paper and g lass wool was then quantif ied by liquid scinti l lation counting. Contro ls (blanks), containing no 38 M E T H O D S cel ls, were run concurrently in all exper iments and used to correct for non-speci f ic chemica l breakdown of the leucine substrate or leucine-der ived a -K IC . 6 . 5 Cell Protein Determination Cel l protein was determined for all experiments. Immediately fol lowing the removal of the g lass wool from the lids, the media was removed from the plates and the cel ls f rozen at - 2 0 ° C until ana lyzed . For ana lys is of protein the cel ls were thawed and recovered from the wel ls by scraping. Potential interfering subs tances were removed by addit ion of 1 ml sod ium deoxycholate to each sample, fo l lowed by the addit ion of 1 ml tr ichloroacetic ac id (12%), and the cel l protein recovered by centrifugation at 3000 rpm in a Sorval l T 6 0 0 0 B Centr i fuge (Newton, CT ) at 4°C for 30 minutes. The filtrate (sodium deoxycholate, tr ichloroacetic ac id and any interfering substances) was removed and the cel l protein a s s a y e d by the method of Lowry (Lowry et a l . , 1951) at an absorbance of 660 nm in a Beckman D U 6 4 0 Spectrophotometer (Fullerton, C A ) . 39 D A T A A N A L Y S E S 7 DATA ANALYSES 7.1 Data Handling and Calculations The rate of CO2 production was calcu lated as dpm/hr based on the incubation time of 90 minutes in all experiments. Calcu lat ion of speci f ic radioactivity in the incubation is required to convert dpm va lues to the amount of C 0 2 . The calculat ion of speci f ic radioactivity was based on the actual amount of leucine in the system (0.2 urnol/1ml total incubation) equivalent to 2 x 1 0 5 p m o l , and the amount of radioact ive leucine added in each experiment. The amount of radioactivity added in each experiment was 0.9 [id for [U 1 4 C]- leuc ine, and 0.8 u.Ci for [1- 1 4 C]- leuc ine. Radioact iv i ty was converted to dpm va lues, 1 u.Ci = 2.2x 1 0 6 dpm, thus 1.998 x 1 0 6 d p m and 1.760 x 1 0 6 d p m were added in the exper iments with [U 1 4 C]- leuc ine and [1- 1 4 C]- leuc ine, respect ively. The speci f ic radioactivity o f t he label led leucine in the react ions was then calcu lated by dividing the dpm value by the amount of leucine in pmoles. Spec i f i c Radioact iv i ty = radioactivity (dpm) / amount of leucine (pmol) The rate of oxidation of [U- 1 4 C]- leuc ine and oxidative decarboxylat ion of [1 - 1 4 C]-leucine was then calcu lated by dividing the rate of C 0 2 production (dpm/hr) by the speci f ic activity of radioact ive leucine (dpm/pmol). In the c a s e of uniformly labeled leucine ( [U- 1 4 C]- leucine), the oxidation rate was then div ided by six because e a c h leucine molecule can potentially form six molecu les of 1 4 C 0 2 . Ra tes of leucine oxidation and oxidative decarboxylat ion (pmol/hr) were then expressed per mg of protein, using the results of the ana lys is of cel l protein. 40 D A T A A N A L Y S E S Oxidat ion rate (pmol/hr) = 1 4 C 0 2 product ion (dpm/hr) / speci f ic activity (dpm/pmol) The amount of 1 4 C 0 2 re leased by the chemica l decarboxylat ion of a -KIC (Step 1, Figure 7.1) represents the amount of [1- 1 4 C]- leucine that had been t ransaminated, but had not p roceeded through subsequent s teps of leucine metabol ism v ia branched chain ketoacid dehydrogenase. The amount of a -KIC (pmol) and rate of a -KIC production (pmol/hr) were ca lcu la ted a s exp la ined above. T h e rate of net t ransaminat ion of leucine to a-KIC is equal to the sum of the rate of production of 1 4 C 0 2 from branched cha in ketoacid dehydrogenase (Step 2, Figure 7.1) plus the rate of production of 1 4 C 0 2 from chemica l decarboxylat ion of a-keto isocaproate (Stepl, Figure 7.1). Figure 7.1 Schematic representation of the first two steps of leucine metabolism a - K G Glutamate BCKA dehydrogenase [1 - 1 4 C]-Leu 14, C 0 2 41 D A T A A N A L Y S E S 7.2 Statistical Analyses Al l data were ana lyzed using the Statist ical P a c k a g e for the Soc ia l S c i e n c e ( S P S S Inc. vers ion 7.5 for Windows , Ch icago , Illinois). For exper iments using [U - 1 4 C] -leucine, means and standard deviat ions were calculated for the oxidation of leucine to 1 4 C 0 2 (pmol/mg protein/hr). For exper iments with [ 1 - 1 4 C]- leuc ine, means and standard deviat ions were ca lcu lated for oxidation of leucine to 1 4 C 0 2 ( n m o l / m g protein/hr), rate of production of a - K I C from leucine (nmol/mg protein/hr) and rate of net t ransaminat ion of leucine (nmol/mg protein/hr). One-way ana lys is of var iance was used to determine signif icant di f ferences in outcome var iables between exper iments with differing amounts of substrates. W h e n signif icant di f ferences were found, the Post H o c Least Signif icant Dif ference Test was used to determine which of the means were different. T h e level of s igni f icance was P =0.05 in all tests. 42 R E S U L T S 8 RESULTS 8.1 Metabolism of [U14C]-Leucine by Astrocytes The production of 1 4 C 0 2 f r o m [U- 1 4 C]- leuc ine in astrocytes was signif icantly inhibited when 0.5, 1.0 and 5.0 m M octanoate or 1.0 and 5.0 m M 0-hydroxybutyrate (BHB) were included a s addit ional substrates (Table 8.1). In contrast, the inclusion of an addit ional 5 m M of g lucose in the react ion mix had no effect on leucine oxidation (P=0.612J. Table 8.1: Effect of additional substrates on the production of 14C02 from [U14-C]-leucine in astrocytes Addit ional Substrate % of control n P-va lue 1 None 100 14 -5.0mM g lucose 106 14 0.612 0.5mM B H B 73 9 0.067 1.0mM B H B 40 9 <0.001 5 .0mM B H B 31 9 <0.001 0 .5mM octanoate 50 11 O . 0 0 1 1 .OmM octanoate 21 12 <0.001 5 .0mM octanoate 35 12 O . 0 0 1 1 rate of 1 4 C 0 2 production compared to control (no addit ional substrate) 43 R E S U L T S 8.1.1 Effect of p-Hydroxybutyrate on [U-14C]-Leucine Oxidation The mean rate of oxidation of [U 1 4 C]- leuc ine to 1 4 C 0 2 w a s 14.44 pmol/mg protein/hr (Table 8.2). The addit ion of 0.5 m M p-hydroxybutyrate dec reased the oxidation of leucine to C 0 2 by 27%, however, this was not of statistical s igni f icance (P=0.067). A probable reason for this is the interassay variat ion result ing in the high standard error. Addi t ion of 1.0 m M and 5.0 m M p-hydroxybutyrate dec reased the rate of leucine oxidation C 0 2 by 6 0 % (P<0.001) and 6 6 % (P<0.001), respect ively (Table 8.2). Increasing concentrat ions of p-hydroxybutyrate caused correspondingly greater inhibition of leucine oxidat ion, but di f ferences between the concentrat ions were only signif icant between 0.5 m M and 1.0 m M p-hydroxybutyrate (P= 0.041) and 0.5 m M and 5.0 m M p-hydroxybutyrate (P=0.011). 44 RESULTS Table 8.2: Effect of p-hydroxybutyrate on the production of 1 4 C02 from [U-14C]-leucine in astrocytes Concentration of p-hydroxybutyrate (mM) Trials DPM Cell protein (mg) Oxidation rate (pmol/mg/hr) 0.0 14 2.82 x 102 ±30.2 0.222 ± 0.012 14.44 + 1.60 0.5 11 2.22 x10 2 + 39.4 0.236 ± 0.016 10.57 + 1.76 1.0 12 1.23 x10 2 ± 32.4 * 0.227 ± 0.011 5.77 ± 1.30* 5.0 12 94.3 ± 13.7* 0.244 ± 0.017 4.84 + 0.790* 1 values expressed as mean + standard error of 11-14 separate trials * values statistically different from control (0.0 mM p-hydroxybutyrate) in same column (P<0.05) 8.1.2 Effect of Octanoate on [U-14C]-Leucine Oxidation The addition of 0.5 mM, 1.0 mM and 5.0 mM octanoate decreased the rate of leucine oxidation to C 0 2 b y 5 0 % (P<0.001), 79% ( P O . 0 0 1 ) and 6 5 % ( P O . 0 0 1 ) , respectively (Table 8.3). The results show maximal inhibition of leucine oxidation occurred with the addition of 1.0 mM octanoate, with no further effect of increasing octanoate concentration. The dose response noted with p-hydroxybutyrate was not evident with octanoate and significant differences between concentrations were only found when comparing 0.5 mM octanoate with 1.0 mM octanoate (P=0.036). 45 R E S U L T S Table 8.3: Effect of octanoate on the production of 1 4 C 0 2 from [U-1 4C]-leucine in astrocytes Concent ra t ion of oc tanoate (mM) Tr ials D P M Ce l l protein (mg) Ox idat ion rate (pmol/mg/hr) 0.0 14 2.82 x 1 0 2 0.222 14.44 ± 3 0 . 2 ± 0 . 0 1 2 ± 1 . 6 0 0.5 9 2.04 x 1 0 2 0.256 7.27 + 59.2 ± 0 .040 ± 1 . 8 5 * 1.0 9 55 .5 0 .223 3.05 ± 8.74 * ± 0 . 0 1 4 ± 0.652 * 5.0 9 96 .8 0 .223 5.03 ± 2 0 . 3 * ± 0 . 0 1 1 ± 0 .799 * 1 va lues e x p r e s s e d a s m e a n ± s tandard error of 9-14 separa te trials * va lues statistically different f rom control (0.0 m M octanoate) in s a m e co lumn (P<0.05) 8.2 Metabolism of [1-14C]-Leucine by Astrocytes Initially, exper iments with [1- 1 4 C]- leucine were conducted in D M E M media containing about 5 m M g lucose. Recovery of 14C02 was low in some experiments, interassay variabil ity was high and statistically signif icant di f ferences with addit ion of B H B were not detected, either in rates of leucine oxidation to CO2 or rates of a -K IC production from leucine. The dpm obtained in some exper iments were not sufficiently above those of the b lanks to a l low for data analys is . W h e n a s s a y e d in P B S , rates of oxidation of leucine to C02and production of a -KIC from leucine were 7-8 and 4-5 fold higher than in D M E M media containing 5 m M glucose. Thus the effects of B H B on the oxidative 46 R E S U L T S decarboxylat ion of [1- 1 4 C]- leuc ine were a lso quantitated in P B S . For compar ison, . results are shown for exper iments in D M E M media and P B S . Further experiments, conducted only in P B S , tested the effect of octanoate on the oxidative decarboxylat ion of [1- 1 4 C]- leucine. 8.2.1 Effect of (3-Hydroxybutyrate on the Oxidative Decarboxylation of [1-14C]-Leucine The mean rate of oxidation of [1- 1 4 C]- leuc ine to 1 4 C 0 2 was 0.132 nmol/mg/hr in D M E M media and 0.583 nmol/mg/hr in P B S . The addit ion of 1.0 m M and 5.0 m M p-hydroxybutyrate did not produce a statistically signif icant effect on the rate of oxidative decarboxylat ion of leucine (Table 8.4 and 8.5). 47 R E S U L T S Table 8.4: Effect of R- hydroxybutyrate on the rate of oxidative decarboxylation of [1-14C]-leucine in astrocytes cultured in DMEM media1 Concentration of p-hydroxybutyrate (mM) Trials DPM Cell protein (mg) Rate of 1 4 C 0 2 production (nmol/mg/hr) 0.0 4 2.60 x 102 ± 6 1 . 9 0.171 ± 0.024 0.132 ± 0.044 1.0 3 2.51 x10 2 ± 8 9 . 6 0.183 ± 0 . 0 1 9 0.112 ± 0.045 5.0 6 2.89 x10 2 ± 4 7 . 6 0.179 ± 0 . 0 1 4 0.130 ± 0 . 0 1 1 1 values expressed as mean ± standard error of 3-6 separate trials 2 no statistically significant effects of addition of p-hydroxybutyrate Table 8.5: Effect of p-decarboxylation of [1 • hydroxybutyrate on the rate of oxidative 14C]-leucine in astrocytes cultured in PBS 1 Concentration of P-hydroxybutyrate (mM) Trials DPM Cell protein (mg) Rate of 1 4 C 0 2 production (nmol/mg/hr) 0 6 1.08 x 103 ± 1.93 x 102 0.138 ± 0.005 0.583 ± 0 . 1 0 3 1.0 6 0.887 x10 3 ± 2.44 x10 2 0.110 ± 0.008 * 0.629 ± 0 . 1 8 6 5.0 6 1.33 x 103 ± 3 . 7 9 x 1 0 2 0.106 ± 0.005 * 0.936 ± 0.267 1 values expressed as mean ± standard error of 6 separate trials 2 no statistically significant effects of addition of p-hydroxybutyrate 48 R E S U L T S 8.2.2 Effect of (3-Hydroxybutyrate on the Rate of Production of a-Ketoisocaproate from [1-14C]-Leucine The amount of 1 4 C C » 2 p roduced by the chemica l decarboxylat ion of [1 - 1 4 C ] - a -K IC remaining in exper iments after the 90-minute incubation, reflects the amount of leucine that had been transaminated to a -K IC, but had not p roceeded through the subsequent step of oxidative decarboxylat ion. The rate of production of 1 4CC>2 from chemica l decarboxylat ion of leucine-der ived a -K IC in exper iments in D M E M media w a s 16.2 nmol/mg/hr. The addit ion of 1.0 m M and 5.0 m M p-hydroxybutyrate did not cause a statistically signif icant change in the rate of production of a -K IC (Table 8.6). In P B S , the rate of production of leucine-der ived a -K IC was determined to be 60.3 nmol/mg/hr. The addit ion of 1.0 m M and 5.0 m M p-hydroxybutyrate increased the rate of a -K IC production by 5 4 % (P=0.008), and 3 6 % (P=0.062), respect ively (Table 8.7). Thus , the increase in production of a -K IC by 1.0 m M p-hydroxybutyrate was statistically significant, whereas the apparent increase by 5.0 m M p-hydroxybutyrate was not. However, there was no statistically signif icant dif ference between the rate of product ion of a -K IC in the presence of 1.0 m M p-hydroxybutyrate compared with 5.0 m M p-hydroxybutyrate (P=0.329). 49 RESULTS Table 8.6 Effect of p - hydroxybutyrate on the rate of production of 1 4 C 0 2 derived from chemical decarboxylation of a-ketoisocaproate in astrocytes cultured in DMEM media1 Concent ra t ion of p-hydroxybutyrate (mM) Tr ials D P M Ce l l protein (mg) Ra te of 1 4 C 0 2 product ion (nmol/mg/hr) 0.0 6 3.19 x 1 0 4 ± 2 . 1 7 x 1 0 3 0.155 ± 0 . 0 1 8 16.2 ± 1.61 1.0 4 4 .48 x 1 0 4 ± 1.48 x 1 0 3 * 0.191 ± 0 . 0 1 6 17.8 ± 1.88 5.0 6 4.21 x 1 0 4 ± 3.59 x 1 0 3 * 0 .179 ± 0 . 0 1 4 17.7 ± 0.622 1 va lues e x p r e s s e d a s m e a n ± s tandard error of 4-6 separa te trials * va lues statistically different from 0.0 m M p-hydroxybutyrate in s a m e co lumn (P<0 Table 8.7 Effect of p - hydroxybutyrate on the rate of production of 1 4 C O derived from chemical decarboxylation of a-ketoisocaproate in astrocytes cultured in PBS 1 Concent ra t ion of p-hydroxybutyrate (mM) Tr ials D P M Ce l l protein (mg) Ra te of 1 4 C 0 2 product ion (nmol/mg/hr 0 6 1.08 x 1 0 s ± 1.33 x 1 0 4 0.138 ± 0 .005 6 0 . 3 ± 8 . 6 1 1.0 6 1.35 x 1 0 s ± 8.62 x 1 0 3 0.110 ± 0 . 0 0 8 * 93.1 ± 5 . 7 9 * 5.0 6 1 . 1 6 x 1 0 5 ± 1 . 1 5 x 1 0 4 0.106 ± 0 .005 * 82.2 ± 8 . 2 9 1 va lues e x p r e s s e d a s m e a n ± s tandard error of 6 separa te trials * va lues statistically different f rom 0.0 m M p-hydroxybutyrate in s a m e co lumn (P<0.05) 50 RESULTS 8.2.3 Effect of p-hydroxybutyrate on the Rate of Net Transamination of [1-1 4C]-Leucine The net rate of [1- 1 4 C]- leuc ine transaminat ion by branched cha in amino ac id t ransaminase is equal to the rate of 1 4CC>2 production from oxidative decarboxylat ion, plus the rate of production of 1 4 C 0 2 from the chemica l decarboxylat ion of [1- 1 4 C]- leuc ine-der ived a-keto isocaproate. No signif icant di f ferences were found in the rate of leucine transaminat ion for exper iments conducted with increasing concentrat ions of R-hydroxybutyrate in D M E M media (Table 8.8 and 8.9). The rate of net leucine transaminat ion was 3-4 fold higher in P B S compared to D M E M media. The rate of net leucine transaminat ion in P B S was 61.0 nmol/mg cel l protein/hr and it increased signif icantly when 1.0 m M and 5.0 m M p-hydroxybutyrate were included (P=0.008 and P= 0.05, respect ively). Table 8.8 Effect of p-hydroxybutyrate on the rate of net transamination of [1-14C]-leucine in astrocytes cultured in DMEM 1 Concentration of p-hydroxybutyrate (mM) Trials DPM Cell protein Net transamination rate (nmol/mg/hr) 0.0 4 3 .47x 1 0 4 ± 1.87 x 1 0 3 0.171 ± 0.024 16.3 ±2 .41 1.0 3 4 . 4 0 x 1 0 4 ± 1.36 x 10 3 0.183 ±0.019 18.3 ±1 .53 5.0 6 4 . 2 4 x 1 0 4 ± 3.61 x 1 0 3 0.179 ±0.014 17.7 ±0.676 1 values expressed as mean + standard error of 3-6 separate trials * values statistically different from control (0.0 mM p-hydroxybutyrate) in same column (P<0.05) 51 R E S U L T S Table 8.9 Effect of p-hydroxybutyrate on the rate of net transamination of [1-14C]-leucine in astrocytes cultured in PBS 1 Concentrat ion of P-hydroxybutyrate (mM) Tr ia ls D P M Ce l l protein (mg) Net t ransaminat ion rate (nmol/mg/hr) 0.0 6 1 .10x 1 0 5 ± 1.32 x 1 0 4 0.138 ± 0.005 61.0 + 8.69 1.0 6 1.36 x 1 0 5 + 8.49 x 1 0 3 0.110 ± 0.008 * 92.1 + 5 . 1 8 * 5.0 6 1 . 1 7 x 1 0 5 ± 1 .13x 1 0 4 0.106 ± 0 . 0 0 4 * 83.0 + 8.51 1 values expressed as mean ± standard error of 6 separate trials * values statistically different from control (0.0 mM p-hydroxybutyrate) in same column (P<0.05) 8.2.4 Effect of Octanoate on the Oxidative Decarboxylation of [1- C]-Leucine In these trials, the mean rate of oxidative decarboxylat ion of [1- 1 4 C]- leuc ine was 2.35 nmol/mg/hr. Addi t ion of octanoate to the incubation media resulted in signif icant reduct ions in 1 4 C 0 2 product ion, with 0.5 m M octanoate caus ing a 5 4 % reduction to 1.09 nmol/mg/hr (P<0.001), 1.0 m M octanoate caus ing a 7 7 % reduction (P<0.001) to 0.527 nmol/mg/hr, and 5.0 m M octanoate caus ing a 9 4 % reduction to 0.120 nmol/mg/hr (P<0.001) (Table 8.10). 52 R E S U L T S Table 8.10 Effect of octanoate on the rate of oxidative decarboxylation of [1-14C]-leucine in astrocytes cultured in PBS 1 Concentration of Octanoate (mM) Trials DPM Cell protein (mg) Rate of 1 4 C 0 2 production (nmol/mg/hr) 0.0 3 7.62 x 103 ± 1.20 x 103 0.240 + 0.017 2.35 ± 0 . 2 5 0 0.5 5 3.59 x 103 ± 1.20x103* 0.214 ± 0 . 0 4 1 1.09 ± 0 . 2 1 9 * 1.0 5 8.09 x 102 ± 1.98 x 102* 0.144 ± 0 . 0 0 8 0.527 ± 0.029 * 5.0 5 2.16 x 102 ± 84.2 * 0.121 ± 0 . 0 4 5 * 0.120 ± 0 . 0 1 1 * 1 values expressed as mean ± standard error of 3-5 trials (# of trials in brackets) * values statistically different from control (0.0 mM octanoate) in same column (P<0.05) 8.2.5 Effect of Octanoate on the Production of a-Ketoisocaproate from [1-1 4C]-Leucine A s stated previously, the amount of CO2 produced by the chemica l decarboxylat ion of [1- 1 4 C]-a-KIC remaining in exper iments after the 90-minute incubation, reflects the amount of leucine that has been t ransaminated to a-KIC, but has not p roceeded through the subsequent s teps of leucine metabol ism. The rate of product ion of leucine-der ived a-KIC occurr ing in trials with no addit ional substrate, conducted in P B S was 30.3 nmol/mg/hr (Table 8.11). W h e n 0.5 m M of octanoate was added the rate of production dec reased to 28.4 nmol/mg/hr (6% decrease , P=0.81). Wi th 1.0 m M octanoate, product ion of l euc ine -der i ved a-KIC 53 RESULTS rose to 46.3 nmol/mg/hr (53% increase, P=0.10) whi le with 5.0 m M octanoate it w a s 5.31 nmol/mg/hr (83% decrease , P= 0.008). Table 8.11 Effect of octanoate on the rate of production of "CO2 derived from chemical decarboxylation of a-ketoisocaproate in atrocytes cultured in PBS 1 Concentration of octanoate (mM) Trials DPM Cell protein (mg) Rate of 1 4 C 0 2 production (nmol/mg/hr) 0.0 3 9.72 x 104 ± 6.74 x 1 0 3 0.240 ± 0.017 30.3 ±2.96 0.5 5 7.05 x 104 ± 9.34 x 1 0 3 * 0.214 ± 0.004 28.4 ± 7.26 1.0 3 8.34 x 10 4 ± 1.14 x 1 0 4 0.133 ± 0.007 46.3 ±4.84 5.0 5 3.06 x10 3 ± 8.90 x 10 2 * 0.128 ± 0.045 5.31 ±2.94* 1 values expressed as mean ± standard error of 3-5 trials (# of trials in brackets) * values statistically different from control (0.0 mM octanoate) in same column 8.2.6 Effect of Octanoate on the Net Transamination of [1-14C]-Leucine A s d i scussed previously, the rate of net t ransaminat ion of [1- 1 4 C]- leuc ine is equal to the rate of production of 1 4 C 0 2 col lected from oxidative decarboxylat ion plus the rate of 1 4 C 0 2 production from the chemica l decarboxylat ion of [1 - 1 4 C]-leucine-der ived a-keto isocaproate. The transaminat ion rate calculated in this manner w a s 33.0 nmol/mg/hr in trials with no addit ional substrate. Addi t ion of 0.5 m M and 1.0 m M octanoate resulted in non-signif icant changes to the transaminat ion rate with 0.5 m M octanoate caus ing a 9 .7% dec rease to 29 .5 54 R E S U L T S nmol/mg/hr (P=0.69) and 1.0 m M octanoate caus ing a 43.5% increase to 42.3 nmol/mg/hr (P=0.13) (Table 8.12). Inclusion of 5 m M octanoate resulted in 83.4% dec rease in the transaminat ion rate to 5.36 nmol/mg/hr (P=0.004). Table 8.12 Effect of octanoate on the rate of net transamination of [1-14C]-leucine in astrocytes cultured in PBS 1 Concentration of octanoate (mM) Trials DPM Cell protein (mg) Net transamination rate (nmol/mg/hr) 0.0 3 1.05 x 10s ± 6.58 x 103 0.241 ± 0 . 0 1 8 33.0 ± 2 . 5 7 0.5 5 7.41 x 1 0 4 ± 9.95 x 103* 0.214 ± 0 . 0 4 1 29.5 ± 7 . 0 4 1.0 3 8 . 4 4 x 1 0 4 ± 1.15x 104 0.133 ± 0 . 0 0 7 42.3 ± 4 . 8 5 5.0 5 3.28 x 1 0 3 ± 8.53 x 102* 0.128 ± 0 . 0 4 5 5.36 ± 3.02 * 1 values expressed as mean ± standard error of 3-5 trials (# of trials in brackets) * values statistically different from 0.0 mM octanoate in same column (P<0.05) 55 DISCUSSION 9 DISCUSSION The ketogenic diet has been used to treat pediatr ic ep i lepsy s ince the 1920s and the knowledge that fast ing inhibits se izures dates back much further. Despi te the large number of anti-epi leptic drugs avai lable for ep i lepsy treatment, many chi ldren do not respond to drug therapy or exper ience intolerable s ide effects. The ketogenic diet provides an alternative and highly success fu l form of treatment in such cases . Repor ts in the literature indicate a marked reduction in se izure f requency for up to two thirds of chi ldren using a ketogenic diet. The lack of understanding of the mechan ism of act ion of the ketogenic diet has impeded the accep tance of it as a val id form of therapy. In v iew of the fact that ep i lepsy has been l inked to an overact ive excitatory neurotransmitter system, it is important to explore ways in which a high fat, low g lucose diet can potentially impact on neurotransmitter metabol ism. O n e such way might be through an impact on leucine metabol ism, which is an important source o f the a-amino group for the product ion of brain glutamate. Th is study was undertaken to determine the effect of medium cha in fatty ac id and ketone metabol ism on the oxidation of leucine in astrocytes. 9.1 Inhibition of Astrocytic Leucine Metabolism by Octanoate and f3-Hydroxybutyrate The results of this study demonstrate that astrocyt ic leucine metabol ism is signif icantly al tered when M C F A s or ketones replace g lucose as the primary fuel source. Octanoate and p-hydroxybutyrate signif icantly inhibited the production of 56 DISCUSSION 1 4 C 0 2 from [U- 1 4 C]- leuc ine (Table 8.1). Product ion of 1 4 C 0 2 from the oxidation of [U- 1 4 C]- leuc ine w a s reduced by as much as 6 6 % with 5.0 m M R-hydroxybutyrate ( P O . 0 0 1 ) and 7 9 % with 1.0 m M octanoate ( P O . 0 0 1 ) (Tables 8.2 and 8.3). The 1 4 C 0 2 produced from [U- 1 4 C]- leuc ine inc ludes 1 4 C 0 2 from the decarboxylat ion of a-KIC and 1 4 C 0 2 from oxidation of the remaining carbon skeleton in the T C A cyc le. Thus the dec rease could be related to dec reased transaminat ion of leucine, dec reased flux through B C K A dehydrogenase, inhibition of the entry of leucine carbon into the T C A cycle, or some combinat ion of these possibi l i t ies. Wh i le the results of exper iments using [U- 1 4 C]- leuc ine indicate that R-hydroxybutyrate and octanoate inhibit leucine oxidation, they provide no information about the particular step in the metabol ism that is affected. Addi t ional exper iments using [1- 1 4 C]- leuc ine were conducted to elicit more speci f ic information about the inhibition of leucine oxidation by M C F A s and ketones. Leuc ine labeled only on the first carbon was chosen for this purpose because 1 4 C 0 2 col lected from metabol ism of [1- 1 4 C]- leuc ine is speci f ical ly der ived from the decarboxylat ion of a -KIC by B C K A dehydrogenase, whereas 1 4 C 0 2 from [U- 1 4 C]- leuc ine is potentially from decarboxylat ion, and from oxidation of leucine-der ived ace ty l -CoA in the T C A cycle. Another advantage of using [1-1 4 C] - leuc ine was that [1- 1 4 C]-a-KIC could be chemical ly decarboxylated to produce 1 4 C 0 2 , which could then be col lected and used to est imate the level of leucine der ived-a-KIC in the media at the end of the 90 minute incubation. The amount of a -KIC reflected leucine that had been transaminated to a-KIC, but had not p roceeded through the branched chain ketoacid dehydrogenase step, thus 57 DISCUSSION providing information about the net transaminat ion of [1- 1 4 C]- leucine. For the trials with [1- 1 4 C]- leucine, the results obtained us ing octanoate were quite different from those with p-hydroxybutyrate, thus they will be d i scussed separately. 9.11 Effect of p-Hydroxybutyrate on [1-14C]-Leucine Metabolism in Astrocytes Unlike exper iments using [U- 1 4 C]- leuc ine, the results of exper iments with [1 - 1 4 C]-leucine indicate that p-hydroxybutyrate does not cause a dec rease in 1 4 C 0 2 product ion from leucine in astrocytes (Table 8.5). However, the high interassay variability, reflected by the high standard deviat ion, suggests interpretation of these data must be caut ious. It is poss ib le that there was a dif ference in 1 4 C 0 2 production, but it was not detected. In light of the marked effect of p-hydroxybutyrate in inhibition of the product ion of 1 4 C 0 2 from [U- 1 4 C]- leuc ine, however, it s e e m s likely that [1- 1 4 C]- leuc ine oxidation shou ld be similarly inhibited. T h e results of the studies with [1- 1 4 C]- leucine further demonstrate that p-hydroxybutyrate increased leucine transaminat ion by brain astrocytes (Table 8.9). The net transaminat ion rate is based on 1 4 C 0 2 product ion from oxidative decarboxylat ion plus the rate of a -K IC product ion as der ived from chemica l decarboxylat ion of label led a -K IC. In these studies, as d i scussed above, a -K IC production was increased, but 1 4 C 0 2 w a s not different. A higher amount of 1 4 C in a -K IC at the end of the 90-minute incubation could potentially be expla ined by a 58 DISCUSSION higher transaminat ion of leucine. Alternatively, it could be the result of an inhibition of the second step of leucine metabol ism, which would result in accumulat ion of a-KIC, as reflected in the increased 1 4 C in a-KIC at the end of the 90-minute incubation (Table 8.7). The rate of product ion of 1 4 C 0 2 from a-KIC increased by 5 4 % (P=0.012) and 3 6 % (P=0.06), respect ively in exper iments with 1.0 m M and 5.0 m M p-hydroxybutyrate. In light of the f indings from exper iments with [U- 1 4 C]- leuc ine, it s e e m s more likely that the higher amount of a-KIC, and subsequent ly higher rate of a-KIC product ion, ref lected inhibition of the B C K A dehydrogenase enzyme, rather than an increased in B C A A t ransaminase activity (Figure 9.1). Rates of net t ransaminat ion were increased in the in vitro exper iments with 1.0 or 5.0 m M p-hydroxybutyrate compared to no p-hydroxybutyrate, but the dif ference was statistically signif icant only at the 1 m M level (Table 8.9). If the increased a-KIC was due to increased transaminat ion of leucine, without inhibition of B C K A dehydrogenase, then we would expect to s e e a cor responding increase in 1 4 C 0 2 . O n the other hand, if it were due to inhibition of B C K A dehydrogenase, the production of 1 4 C 0 2 should be dec reased . The results of the exper iments with [1- 1 4 C]- leuc ine show that the increased levels of a -KIC were not accompan ied by a change in 1 4 C 0 2 product ion. Aga in , it must be noted that interpretation of these results must be caut ious because of the high variabil ity in the data on 1 4 C 0 2 production. The net transaminat ion rates calculated were more than 100 t imes the rate of production of 1 4 C 0 2 from [1- 1 4 C]- leucine; this is much higher than the 17-fold dif ference reported by Yudkoff et al . (1994a). The reason 59 DISCUSSION for this is not known but cou ld relate to methodological problems in the col lect ion of 1 4 C 0 2 from [1- 1 4 C]- leuc ine in the current study. Poss ib ly , further exper iments involving higher concentrat ions of leucine and longer incubation t imes might identify changes in 1 4 C 0 2 production from [1- 1 4 C]- leucine in response to f3-hydroxybutyrate. However, the results of exper iments done here with [1 - 1 4 C]-leucine and us ing [U- 1 4 C]- leuc ine as the substrate suggest that if there was a change in 1 4 C 0 2 production this would be a reduction rather than an increase in 1 4 C 0 2 . The exper iments with [U- 1 4 C]- leuc ine demonstrate a very signif icant reduction in 1 4 C 0 2 production when ketones are provided as the primary fuel for astrocyte metabol ism. Th is suggests dec reased oxidation o f the carbon skeleton of leucine in the p resence of fatty ac ids and ketones, consistent with entry of ace ty l -CoA from these substrates into the T C A cycle. It is not c lear why an inhibition of 1 4 C 0 2 production from leucine in the p resence of p-hydroxybutyrate was demonstrated with [U- 1 4 C]- leuc ine, but not with [1- 1 4 C]- leucine. A poss ib le explanat ion is that it was simply a limitation of the method used. 9.12 Effect of Octanoate on the Metabolism of [1-14C]-Leucine in Astrocytes The results of studies with [U- 1 4 C]- leuc ine suggest that the impact of octanoate on the metabol ism of leucine may be more profound than that of p-hydroxybutyrate (Table 8.1). Consis tent with this possibil i ty, results of exper iments with [1- 1 4 C]- leuc ine indicate that octanoate does c a u s e a signif icant dec rease in 1 4 C 0 2 product ion in astrocytes, whereas such an impact was not detected with p-hydroxybutyrate. Al l concentrat ions of octanoate resulted in 60 D I S C U S S I O N signif icant dec reases in 1 4 C 0 2 production with the most dramatic decrease , 94%, observed in trials with 5.0 m M octanoate (Table 8.10). The effect of octanoate on the net transaminat ion of leucine, ca lcu lated by adding 1 4 C 0 2 production from oxidative decarboxylat ion and the rate of a-KIC product ion, was a lso different from what was observed with p-hydroxybutyrate. Wi th p-hydroxybutyrate, an increase in residual a -KIC was observed and thus an increase in net transaminat ion was calculated. Wi th octanoate, a -KIC was slightly dec reased with 0.5 mM, increased with 1.0 m M and dec reased with 5.0 mM. The only effect that was statistically signif icant was the 8 2 . 5 % dec rease seen with the inclusion of 5.0 m M octanoate (Table 8.11). W h e n the a-KIC data are used to calculate a net transaminat ion rate (Table 8.12), the trials with 5.0 m M octanoate are signif icantly reduced compared to those with no addit ional substrate (5.36 nmol/mg/hr vs. 33.0 nmol/mg/hr, P= 0.004). Thus , unlike the situation with p-hydroxybutyrate, there does not appear to be an accumulat ion of a-KIC and there is a dramatic dec rease in the 1 4 C 0 2 product ion. Th is may suggest an inhibition at the level of the B C A A t ransaminase result ing in both dec reased [1-1 4 C]- leuc ine-der ived a-KIC and dec reased 1 4 C 0 2 . It is therefore poss ib le to speculate that octanoate may be act ing v ia a different mechan ism than p-hydroxybutyrate to inhibit leucine metabol ism. 61 DISCUSSION BCAA transaminase BCKA dehydrogenase [ l - ^ C R e u + a - K G T C A cyc le - • G lu + [1- 1 4 C]-a-KIC °1 0 , • Isovaleryl-C o A 14, co 2 A c e t y l - C o A + Key: 1 = chemica l decarboxylat ion 2 = enzymat ic decarboxylat ion 4 C 0 2 | Ace toace ty l -C o A Figure 9.1 Schematic representation of [1-14C]-leucine oxidation 9.2 Hypothesis of the Mechanism of Inhibition of Leucine Oxidation by MCFAs and Ketones The f inding that fatty ac ids and ketones inhibit the production of C 0 2 from leucine in astrocytes is consistent with the research hypothesis. The increased levels of a -KIC seen in exper iments with p-hydroxybutyrate supports the hypothesis that the inhibition occurs at the second step of leucine metabol ism. 62 DISCUSSION O n the other hand, the dec rease in a-KIC seen observed in exper iments with octanoate is not consistent with this hypothesis and suggests that another explanat ion may be needed for those part icular data. The fol lowing d iscuss ion will descr ibe a hypothetical chain of events that could lead to inhibition of the decarboxylat ion of a -KIC by branched cha in ketoacid dehydrogenase and result in the observed increase in leucine-der ived a-KIC and dec rease in CO2 product ion from leucine seen in exper iments with p-hydroxybutyrate. The second step of leucine metabol ism, decarboxylat ion of a -KIC cata lyzed by branched chain ketoacid dehydrogenase, is sensi t ive to the energy state of the cel l . Th i s type of dehydrogenase enzyme is known to be inhibited when A T P , acetyl C o A , or fatty ac ids are high (Lehninger, 1993) Acety l C o A levels were not determined in this study. However, because acetyl C o A is the product of fatty ac id and ketone oxidation, we can a s s u m e that under the condit ions of this experiment, with p-hydroxybutyrate or octanoate provided as fuel sources , the acetyl C o A level would be high. It is a lso poss ib le that the A T P level was e levated by the oxidation of fatty ac ids and ketones. De V ivo et a l . (1978) found an e levated A T P : A D P ratio in the brains of rats fed a high fat diet. Thus , high levels of acetyl C o A , and perhaps high A T P , could have an inhibitory effect on the branched chain ketoacid dehydrogenase enzyme, result ing in the observed accumulat ion of leucine-der ived a-KIC. A block at this step would a lso explain the reduced CO2 production in the p resence of octanoate or p-hydroxybutyrate. 63 DISCUSSION The transaminat ion of leucine by B C A A t ransaminase is a reversible reaction, the direction of which is control led by the concentrat ions of the reactants. The forward reaction in which glutamate and a -K IC are formed from leucine and a - K G is known to be favoured under normal c i rcumstances. Th is react ion, however, is readily reversible when the concentrat ions of glutamate or a -K IC are increased (Yudkoff et al . , 1994a). Even smal l inc reases in the level of a -K IC have been shown to dramatical ly affect the branched cha in amino ac id transaminat ion reaction, result ing in the consumpt ion of glutamate. Us ing astrocytes in culture, Yudkoff et a l . (1994a) demonstrated that the transaminat ion reaction rapidly responded to changes in a -K IC concentrat ion, with 0.05 m M a -KIC caus ing a reduction in astrocyt ic glutamate within only 5 minutes of a -K IC addit ion to the media. At a level of 1.0 m M a -K IC, the intra-astrocytic glutamate concentrat ion was dec reased by 5 0 % in the same period of time. A later study by the same group found that increasing extracel lular a -K IC concentrat ion resulted in increased transaminat ion of a -K IC with glutamate (reverse transamination) and increased oxidation of a-ketoglutarate ( a -KG) v ia the T C A cyc le (Yudkoff et a l . , 1996b). The latter authors specu la ted that removal of a - K G would further pull the transaminat ion to the left, thereby consuming even more glutamate. F lux through glutamine synthetase was a lso dec reased due to the lower levels of glutamate. The end result was lower intracellular glutamine. Thus , it is reasonab le to speculate that the increase in a -K IC observed in the present study may have caused a reversal of the transaminat ion reaction, with subsequent consumpt ion of glutamate and glutamine. In vivo, glutamine 64 DISCUSSION produced in the astrocytes is re leased to neurons where it is precursor to neurotransmitter glutamate. Thus , if these metabol ic events were to occur in vivo, increases in astrocyt ic a -K IC could ultimately result in a reduction in the major excitatory neurotransmitter glutamate. Th is hypothetical ser ies of events may expla in the effect of R-hydroxybutyrate on leucine metabol ism but the effect of octanoate appears to need another explanat ion. The dramat ic effect of octanoate on astrocyt ic 1 4 C 0 2 product ion was observed in both the exper iments with [U- 1 4 C]- leuc ine and those with [1 - 1 4 C]-leucine. The exper iments with [1- 1 4 C]- leucine did not indicate that a -K IC had accumulated in the incubation media, in fact the only signif icant result sugges ts that a -K IC was reduced. T h e s e f indings are not consistent with the initial research hypothesis, and they may suggest that octanoate is exerting its inf luence at a separate step in the pathway of leucine metabol ism, perhaps by inhibiting the initial t ransaminat ion of leucine. 9.3 Alternative Hypotheses of the Mechanism of Inhibition of Leucine Oxidation by MCFAs and Ketones A s descr ibed previously, it is reasonab le to bel ieve that the inhibition of b ranched chain ketoacid dehydrogenase is related to high levels of acetyl C o A from fatty ac id and ketone metabol ism. Other explanat ions, however, are possib le, including a direct effect of ketones or fatty ac ids on the enzyme, or inhibition secondary to changes in the concentrat ion of other T C A cyc le intermediates such as citrate, or changes to the internal C o A pool. Yudkof f et a l . 65 DISCUSSION (1997) found increased levels of citrate in astrocytes cultured with ketones. Increased citrate concentrat ions were assoc ia ted with inhibition of glutamine synthetase. It is poss ib le that citrate was similarly increased in this study, and this may have contributed to changes in leucine metabol ism. Consis tent with this possibil i ty, De V ivo et a l . (1978) found increased concentrat ions of citrate in the brains of rats fed a very high fat diet. Citrate is known to inhibit the ct-ketoglutarate dehydrogenase complex, which is homologous to the B C K A dehydrogenase complex (Lehninger, 1993). Thus , if citrate levels were increased in astrocytes incubated with p-hydroxybutyrate, then this may have had an inhibitory effect on B C K A dehydrogenase, which would result in increased a -K IC and dec reased 14C02 production from [1- 1 4 C]- leucine. However, no speci f ic information of the effects of increased citrate concentrat ions on B C K A dehydrogenase are avai lable. Another possibi l i ty is that the oxidation of fatty ac ids and ketones may have affected the intra-mitochondrial pool of C o A , al though inclusion of C o A in the react ion mix makes this unlikely. It is a lso poss ib le that the dec reased production of CO2 and increased a -KIC observed when astrocytes were provided with octanoate or p-hydroxybutyrate for fuel were not due to inhibition of B C K A dehydrogenase, but rather to some other effect on leucine oxidation. For example, the dec rease in 14C02 product ion from [U- 1 4 C]- leuc ine in exper iments with octanoate or p-hydroxybutyrate could be due to an inability of the carbon skeleton of leucine to enter the T C A cycle. Acety l C o A is the end product of leucine metabol ism and it normally p roceeds into the T C A cyc le with the product ion of CO2. The high levels 66 DISCUSSION of acetyl CoA coming from the oxidation of fatty acids and ketones could conceivably inhibit [U-14C]-leucine-derived acetyl CoA from entering the T C A cycle and thus result in decreased production of 1 4 C02. This hypothesis would not explain the finding of increased a-KIC unless it was acting in addition to the effect on BCKA dehydrogenase. 9.4 Towards an Understanding of the Antiepileptic Efficacy of the Ketogenic Diet Children are more susceptible to epileptic seizures than adults (Johnston, 1996) and also have a better response to the ketogenic diet (Prasad et al., 1996). The developing nervous system, described by Johnston as "hyperexcitable", is more prone to seizures induced by fever or injury, and those of an idiopathic nature. Receptors of the excitatory neurotransmitter system are more easily opened, and more difficult to inhibit in the immature brain (Nicoletti et al., 1986). This enhanced activity probably contributes to the tendency of children rather than adults to develop epileptic seizures. The ability of ketogenic diets to suppress seizures is generally believed to be related in some way to the switch from glucose to fatty acids and ketones as the primary fuel source in the brain. The greater efficacy of the ketogenic diet in children, as compared to adults, may be related to unique aspects of brain and/or fatty acid metabolism during development. Children achieve a higher level of blood ketones than adults in response to fasting (Haymond et al., 1983) and younger children attain ketosis in response to a ketogenic diet more easily than older children (Schwartz et al., 67 DISCUSSION 1989). The uptake of ketones into the brain is a lso greater in young animals and chi ldren (Nehl ig, 1999; Pe rsson , et a l . , 1972). In addit ion, express ion o f the enzymes involved in ketone metabol ism is higher in the young, and thus util ization of ketones is higher (Dahlquist, et a l . , 1972; Hawkins et a l . , 1971). The cumulat ive ev idence suggests that the enhanced ability to extract and use ketones may be related to the greater eff icacy of the ketogenic diet in chi ldren. Interestingly, the rate of brain leucine oxidation is a lso higher in chi ldren compared to adults (Shambaugh and Koehler , 1983); e levat ions in leucine metabol ism may result in e levat ions in glutamate and thus may be related to the greater neuronal excitability seen in chi ldren. It is poss ib le that the use of ketones and M C F A s as the primary fuel source alters brain metabol ism via inhibition of leucine oxidat ion, which subsequent ly dec reases the excitability of neurons through a reduction in brain glutamate. The results of the two sets of exper iments descr ibed here suggest leucine metabol ism is signif icantly altered when astrocytes are provided with ketones or M C F A s as their major fuel source. A n important relat ionship exists between the metabol ism of leucine and that of the major excitatory neurotransmitter glutamate in the brain. Glutamate uptake at the b lood brain barrier is negl igible and leucine has been identif ied as a major source of nitrogen for the replenishment of glutamate lost to oxidation or anabo l ic p rocesses . Thus changes to the metabol ism of leucine will have important consequences for the synthesis of glutamate, and for neurotransmission. Data obtained from the oxidation of [U-1 4 C] - leuc ine in the p resence of different fuel sources suggest that the oxidation of 68 DISCUSSION leucine is inhibited when fatty ac ids or ketones are provided as primary fuel sources for astrocytes. The reduction in leucine metabol ism was observed with concentrat ions of p-hydroxybutyrate and octanoate as low as 0.5 m M a va lue that may be physiological ly significant. Huttenlocher determined the concentrat ion of p-hydroxybutyrate in the p lasma and cerebral sp inal fluid of chi ldren on a ketogenic diet and found them to be 2.5 m M and 0.4 m M respect ively (Huttenlocher, 1976). Circulat ing levels of octanoate in chi ldren receiv ing an M C T diet have been reported to increase from a normal of 0.04 m M to 0.6 m M (Schwartz et a l . , 1989). If octanoate does indeed c ross the blood brain barrier we would expect to s e e a corresponding increase in cerebral octanoate in chi ldren on M C T diets. It must be acknowledged that concentrat ions of p-hydroxybutyrate and octanoate within the astrocyte are not known and may differ from va lues reported in whole brain. The data a lso suggest that octanoate inhibits the oxidation of [1- 1 4 C]- leuc ine al though the mechan ism of how this is accompl ished is unc lear and is not consistent with the hypothetical mechan ism proposed in this thesis. Further, the results of the exper iments with [1- 1 4 C]- leucine suggest that when p-hydroxybutyrate is used as the primary fuel source by astrocytes, the second step of leucine metabol ism is inhibited and a -K IC accumulates. A n accumulat ion of a -K IC should c a u s e a reversal of the react ion cata lyzed by B C A A t ransaminase, ultimately result ing in dec reased astrocyt ic glutamate and glutamine. Ast rocytes export glutamine to the neurons where it is converted back to glutamate and used in neurotransmission. If the level of glutamine reaching 69 DISCUSSION the neurons is decreased secondary to inhibition and/or reversal of BCAA transaminase, then we can expect a decrease in the excitability of those neurons. Although not desirable in the normal brain, this would be efficacious in an epileptic brain where glutamate levels are high and receptors are extraordinarily sensitive. Antiepileptic drugs generally work by interfering with the excitatory neurotransmitter system. For example, they may compete for the glutamate receptor, or increase the conversion of glutamate to GABA, an inhibitory neurotransmitter (Chapman, 2000). It is possible that the ketogenic diet also exerts its effects at the level of glutamate metabolism. Glutamate levels were not measured in these experiments, but in view of the relationship between leucine and glutamate in brain, it is reasonable to hypothesize that the observed changes in astrocytic leucine metabolism would be accompanied by changes in astrocytic glutamate and glutamine. The finding that leucine oxidation is inhibited by fatty acids and ketones in astrocytes contributes to the current understanding of integrated fuel metabolism in brain, and may provide a clue to the mechanism of action of the ketogenic diet. Further research, however, is needed to confirm an effect on astrocytic and neuronal glutamate and glutamine. 9.5 Limitations of the Study Several limitations must be acknowledged when interpreting the results of this study. The experiments were done in an in vitro, single cell system, the astrocyte culture, which is very different from the situation occurring in the whole brain. It is possible that events occurring in cultured astrocytes are different than 70 DISCUSSION what would occur in the whole brain. Wh i l e this must be recognized a s a limitation, it is a lso c lear that useful information can be obtained in such a s imple system. The highly control led experimental system al lows manipulat ion of substrates and observat ion of the direct effects on leucine metabol ism within astrocytes, without interference by the neuroendocr ine system. Al though the results must be interpreted with caut ion, studies can now be des igned based on the results. The hypothesis can be tested in more complex sys tems such as co -cultures of astrocytes and neurons, whole brain extracts or s l ices, whole an imals and cl inical studies in humans. The results of the exper iments conducted in P B S are limited by the fact that the only fuels avai lab le to the astrocytes were leucine and R-hydroxybutyrate. The cel ls did not have a cho ice of amino ac ids as they normally would, or as they did in exper iments with D M E M , and l ikewise there was no g lucose avai lable to them. Wh i le this situation is not realistic, it did reveal an accumulat ion of a -K IC , which has been interpreted to reflect inhibition of B C K A dehydrogenase, when astrocytes were using R-hydroxybutyrate a s the primary fuel source. Th is may in fact be a more accurate reflection of what occurs as a result of a high fat, low g lucose diet, when g lucose is limiting. W h e n D M E M media was used, g lucose was abundant (5 mM) and because astrocytes will use g lucose in preference to other fuel substrates, this may have limited the ability to study the effect of R-hydroxybutyrate on leucine metabol ism. A further limitation of this study is that al though the hypothesis l inks the eff icacy of the ketogenic diet to its effect on glutamate and glutamine, these 71 DISCUSSION neurotransmitters were not measured. Unfortunately, the methods used in this study did not allow for the determination of glutamate and glutamine in the astrocytes cultures. The results, however, clearly show an effect on the metabolism of leucine, and reasonable hypotheses can be made based on the data obtained and established knowledge ofthe relationship between brain leucine and glutamate metabolism. At the same time, studies to measure glutamate and glutamine concentrations are needed before definitive conclusion can be made that the metabolism of MCFAs and ketones ultimately lowers astrocytic glutamate/glutamine. 9.6 Future Research Directions Cell Culture: Future research using cell cultures should aim to clarify the effect of MCFAs and ketones on the metabolism of leucine. The effect of octanoate on the oxidation of [1-1 4C]-leucine appears to differ from the effect of p-hydroxybutyrate. The results indicate that octanoate may actually be a more potent inhibitor of leucine metabolism and further studies are needed to determine the mechanism of its effect. As well, the effect of acetoacetate, and a combination of acetoacetate and p-hydroxybutyrate, on leucine oxidation should be tested to determine whether the effect is similar to that obtained with only p-hydroxybutyrate. Studies are necessary to determine whether or not astrocytic glutamate concentration is actually decreased by MCFA and ketone metabolism. The impact of fatty acid and ketone metabolism on the levels of leucine-derived glutamate and glutamine could be tested using the stable isotope [15N]-leucine. 72 DISCUSSION Cel l cultures could a lso be used to determine the levels of acetyl C o A , citrate and the A T P A D P ratio in astrocytes using fatty ac ids and ketones as the primary fuel source. Animal Studies: Future research in animals should determine whether the events occurr ing at the level of the astrocyte will a lso occur in vivo. Microdia lys is cou ld be used to show that increased a -KIC results in increased oxidation of glutamate/glutamine in vivo (Zielke et a l . , 1997), and to test whether infusion of M C F A s and ketones will inhibit leucine metabol ism and dec rease glutamate concentrat ion in brain in vivo. An imal studies could a lso be used to explore whether or not chronic feeding of a high fat diet leads to the changes in brain leucine metabol ism observed in short term metabol ic incubations. An ima ls could be fed high fat diets such as those used by De V ivo et a l . (1978) and metabol ic studies done to trace the metabol ism of label led leucine and the production of a-KIC and glutamate. Human Studies: Exper iments should be des igned to measure the levels of leucine and a -KIC in the circulat ion of chi ldren on the ketogenic diet. Stable isotope methods are safe for use in pediatr ic research (Koletzo et a l . , 1998) and could a lso be used to trace leucine metabol ism to a -KIC and glutamate. 73 BIBLIOGRAPHY 10 BIBLIOGRAPHY Annegars , J . F . (1993). Epidemio logy of Ch i ldhood Onset Se izures , C h . 6 in Dodson , W . E . and Pel lock, J . M . (eds.): Pediatr ic Epi lepsy: D iagnos is and Therapy. N e w York, Demos Publ icat ions: pp 57-61. Appleton, D.B. and De V ivo, D.C. (1974). A n An imal Mode l for the Ketogen ic Diet. Epilepsia 15: 211-217. Aues tad , N., Korsak, R.A., Morrow, J . W . and Edmond, J . (1991). Fatty Ac i d Oxidat ion and Ketogenes is by Astrocytes in Pr imary Cul ture. J of Neurochem 56:1376-1386. Avo l i , M. and Olivier, A . (1987). Burst ing in Human Epi leptogenic Neocor tex is D e p e r e s s e d By an N M D A Antagonist. Neurosci Lett 76. 249-254. B a c h , A . C . and Babayan , V . K . (1982). Medium-cha in tr iglycerides: A n Update. Am J Clin Nutr 36(5): 950-962. Ba l laban-Gi l , K., Ca l l ahan , C , O 'Del l , O , Pappo , M., Moshe , S. and Shinnar, S . (1998). Compl icat ions of the Ketogenic Diet. Epilepsia 39 (7): 744-748. Bixel , M .G . , and Hamprecht, B. (1995). Genera t ion of Ketone Bod ies from Leuc ine by Cul tured Astrogl ia l Ce l l s . J of Neurochem 65: 2450-2461 . 74 BIBLIOGRAPHY Blue, M.E. , and Johnston, M.V. (1995). The Ontogeny of Glutamate Receptors In Rat Barrel F ie ld Cortex. Brain Res Dev Brain Res 84: 11 -25. Bough, K .J . , Doug las , A . and Eag les , D A . (1999). A Ketogenic Diet Increases the Res is tance to Penty lenetet razole- lnduced Se izu res in the Rat. Epilepsia 40 : 138-143. Chadwick, D.W. (1997). A n Overv iew of the Eff icacy and Tolerabi l i ty of N e w Ant iconvulsant Drugs. Epilepsia 3 8 : S 5 9 - S 6 2 . C h a p m a n , A G . (2000). Glutamate and Epi lepsy. J of Nutr 130: 1043S-1045S . Co le , R. and de Vel l is , J . (1989) Preparat ion of Ast rocytes and Ol igodendrocyte Cul tures From Primary Rat Gl ia l Cul tures, in Shahar , A . , de Vel l is , J . , Vernadak is , A . and Haber, B., (eds.): A Dissect ion and T i ssue Cul ture Manua l of the Nervous Sys tem. N e w York, A l an R. L iss : pp 121-133. Daikhin, Y . and Yudkoff, M. (2000). Compartmentat ion of Brain Glutamate Metabol ism in Neurons and G l ia . J of Nutr 130: 1026S-1031S . 75 BIBLIOGRAPHY Dahlquist, G . , Pe rsson , U. and Pe rsson , B. (1972). The Activity of D-Beta-Hydroxybutyrate Dehydrogenase in Feta l , Infant and Adult Rat Brain and the Influence of Starvat ion. Biol Neonate 20: 40-50. De V ivo , D.C. , Leck ie , M.P. , Ferrendel l i , J . S . and McDouga l , D.B. (1978). Chron ic Ketos is and Cerebra l Metabol ism. Ann Of Neurol 3:331 -337. Devl in, T . M . (ed.) (1992). Textbook of Biochemistry with Cl in ica l Corre lates. John Wi ley and S o n s Inc, N e w York: pp 249-252. Dodson , W . E . (1993). Ep i lepsy and I.Q., C h . 35 in Dodson , W . E . and Pel lock, J . M . (eds.): Pediat r ic Epi lepsy: D iagnos is and Therapy. N e w York, Demos Publ icat ions: pp 373-385. Edmond , J . , Robb ins , R.A., Bergstrom, J .D . , Co le , R A . and de Vel l is , J . (1987). Capac i ty for Substrate Uti l ization in Oxidat ive Metabol ism by Neurons, Astrocytes, and Ol igodedrocytes From Develop ing Brain in Pr imary Cul tures. J of Neurosci Res 18:551 -556. Er ic inska, M. and Si lver, I A . (1990). Metabol ism and Ro le of Glutamate in Mammal ian Brain. Prog Neurobiol 35:245-296. 76 BIBLIOGRAPHY Faught, E. (1997). Eff icacy of Topiramate as Adjunct ive Therapy in Refractory Part ial Se izu res : United States Trial Exper ience. Epilepsia 3 8 : S 2 4 - S 2 7 . Fernando-Warnaku lasur iya , G . J . , Staggers, J . E . , Frsot, S.c . and We l l s , M.A. (1981). Stud ies on Fat Digest ion, Absorpt ion and Transport in the Suck l ing rat. I. Fatty Ac i d Composi t ion and Concentrat ion of Major Lipid Components . J of Lipid Res 22(4): 668-674. F reeman, J . M . , Kel ly, M T . and Freeman, J . B . (1994). The Ep i lepsy Diet Treatment. N e w York: Demos. F reeman, J . M . (1995). A Cl in ic ian 's Look at the Developmenta l Neurobio logy of Epi lepsy, C h . 1 in Schwarzkro in et al (eds.): Brain Development and Epi lepsy. N e w York, Oxford University P ress Inc: pp 9-33. F reeman, J . M . , V in ing , E . P . G . , P i l las, D.J . , Pyzik , P . L , C a s e y , J C . and Kel ly, M T . (1998). T h e Eff icacy of the Ketogen ic Diet-1998: A Prospect ive Evaluat ion of Intervention in 150 Chi ldren. Pediatrics 102(6): 1358-1363. Geye l in , H.R. (1921). Fast ing A s a Method For Treat ing Epi lepsy. Medical Record 99 :1037-1039. 77 BIBLIOGRAPHY Gri l l , V. , Bjorkhem, M., Gutniak, M. and Lindqvist, M. (1992). Brain Uptake and R e l e a s e of Amino Ac ids in Nondiabet ic and Insul in-Dependent Diabet ic Subjects: Important Ro le of Glutamine R e l e a s e for Nitrogen Ba lance . Metabolism 41 : 28 -32. Guberman A . and Bruni, J . (1999). Essent ia ls of Cl in ica l Epi lepsy, 2 n d edit ion, Boston: Butterworth-Heinemann. Hawkins, R A . , Wi l l i amson, D.H. and Krebs, H A . (1971). Ke tone-Body Uti l ization by Adult and Suck l ing Rat Brain In V ivo . Biochem J 122: 13-18. Haymond, M.W. , Howard, C , Ben -Ga l im , E. and De V ivo, D.C. (1983). Effects of Ketos is on G l u c o s e Flux in Chi ldren and Adul ts. Am J Physiol 245 : E373-378 . Helmholz, H.F. (1927). The Treatment of Ep i lepsy in Ch i ldhood. F ive Y e a r s Exper ience Wi th the Ketogenic Diet. JAMA 88: 2028-2032. Herzberg , G .Z . , F ivush, B A . , K insman, S .L . (1990). Urol i thiasis Assoc ia ted Wi th the Ketogenic Diet. J Pediatrics 117:743-745. Huang, R., S o c h o c k a , E. and Hertz, L. (1997). Ce l l Cul ture Stud ies o f t he Ro le of E levated Extracel lu lar Glutamate and K + in Neuronal Ce l l Death During and After Anox ia / Ischemia. Neurosci Biobehav Rev 21 : 129-134. 78 BIBLIOGRAPHY Huttenlocher, P .R. (1976). Ketonemia and Se izures : Metabo l ic and Ant iconvulsant Effects of Two Ketogenic Diets in Ch i ldhood Epi lepsy. Pediat Res 10:536-540. Huttenlocher, P .R. and Hapke, R . J . (1990) A Fo l low-Up Study of Intractable Se izu res in Ch i ldhood. A n n . Neurol . 28(5): 699-705. J e n s e n , R . G . and Newburg, D.S. (1995). Bov ine Milk Lipids, in Handbook of Milk Composition ( Jensen , R . G . , ed.). A c a d e m i c P ress , S a n Diego. Johnston, M.V. (1996). Developmenta l Aspec t s of Ep i leptogenes is . Epilepsia 37(Suppl .1) : S 2 - S 9 . K insman, S .L . , V in ing, E . P . G . , Quaskey , S.A. , Mell i ts, D. and F reeman, J . M . (1992). Eff icacy of the Ketogenic Diet for Intractable Se izu re Disorders: R e v i e w of 58 C a s e s . Epilepsia 33(6): 1132-1136. Koletzo, B., Demmelmair , H., Hartl, W. , K indermann, A. , Koletzo, S . , Sauerwa ld , T. and Szi tany i , P. (1998). The U s e of Stable Isotope Techn iques for Nutrit ional and Metabo l ic R e s e a r c h in Paediat r ics. Early Hum Dev 53: S77 -97 . 79 BIBLIOGRAPHY Leferre, F. and A ronson , N. (2000). Ketogen ic Diet for the Treatment of Refractory Ep i lepsy in Chi ldren: A Systemat ic Rev iew of Eff icacy. Pediatrics. 105(4): E46 . Lehninger, Ne lson , and C o x (1993). Pr inc ip les of Biochemistry, 2 n d edit ion (Lehninger, Ne lson and Cox , eds.) , pp. 527-536. Wor th Publ ishers , N e w York. Liv ingston, S . (1972). Comprehens ive Management of Ep i lepsy in Infancy, Ch i ldhood and Ado lescence . Springfield,IL., Char les C. Thomas : pp 378-405. Lowry, O .H . , Rosebrough , N.J . , Farr, A . L . and Randa l l , R . J . (1951). Protein Measurement with the Fol in Pheno l Reagent . J Biol Chem 193: 265-275. Mar t inez-Hernandez, A . , Bel l , K .P . and Norenberg, M.D. (1977). Glutamine Synthetase: Gl ia l Local izat ion in Brain. Science 195:1356-1358. McCar thy , K.D. and de Vel l is , J . (1990). Preparat ion of Separa te Astrogl ia l and Ol igodendrogl ia l Ce l l Cul tures From Rat Cerebra l T issue . J. Cell Biol. 85: 890-902. McDona ld , J . W . and Johnston, M.V. (1990). Phys io log ica l and Pathophys io log ica l Ro les of Excitatory Amino A c i d s During Centra l Nervous Sys tem Development. Brain Res Brain Res R e v 15: 41-70. 80 BIBLIOGRAPHY McDona ld , J .W. , Trescher , W . H . , C h e n , C-K . , and Johnston, M.V. (1992). Induced Neurotoxicity Transient ly P e a k s During Ear ly Postnatal Development. Brain Res 583: 54-70. Nakazawa , M., Kodama , S. and Matsuo, T. (1983). Effects of Ketogen ic Diet on Elect roconvuls ive Thresho ld and Brain Contents of Adenos ine Nucleot ides. Brain Dev 5: 375-380. Nehl ig , A . (1999). Age-Dependen t Pa thways of Bra in Energy Metabo l ism in the Suck l ing Rat, A Natural Mode l of the Ketogenic Diet. Epilepsy Research 37 :211 -221 . Nicoletti, F., ladarola, M.J . , Wrob lewsk i , J.T. , Cos ta , E. (1986). Excitatory Amino Ac id Recogni t ion Si tes Coup led Wi th Inosital Phospho l ip id Mertabol ism: Developmenta l C h a n g e s and Interaction With u-adrenoreceptors. Proc Natl Acad Sci USA 83 :1931-1935. Nigra, M.A., Vent imigl ia, J . , Se l cen , D. and Beierwaltes, P. (1995). Se izu re Frequency, Behav iora l , and Per formance Effects of the Ketogen ic Diet. Ann Of Neurol 38(3): 549-550. 81 BIBLIOGRAPHY Nordl i , Jr., D R . and De V ivo, D.C. (1997). The Ketogenic Diet Revis i ted: Back to the Future. Epilepsia 38(7): 743-749. Owen , O . E . , Morgan, A . P . and Kemp, H .G. (1967). Brain Metabol ism During Fast ing. J of Clin Invest 46: 1589-1595. Pe rsson , B., Settergren, G . and Dahlquist, D. (1972). Cerebra l Ar ter io -Venous Difference of Acetoaceta te and D-Beta-Hydroxybutyrate in Chi ldren. Acta Paediatr Scand 61 : 273-278. Peterman, M . G . (1925). The Ketogen ic Diet in Epi lepsy. JAMA 84 :1979-1983. P rasad , A . N . , Stafstrom, C . F . , and Holmes, G . L (1996). Alternative Ep i lepsy Therap ies : The Ketogen ic Diet, Immunoglobul ins, and Steroids. Epilepsia 37(Suppl1) : S 8 1 - S 9 5 . Schwartz, R .M. , Boyes , S. , and Ayns ley -Green , A. (1989). Metabol ic Effects of Three Ketogen ic Diets in the Treatment of Seve re Epi lepsy. Dev Med & Child A/euro/31:152-160. Scwartzkro in, P A . (1999). Mechan i sms Under ly ing the Ant i -Epi lept ic Eff icacy of the Ketogen ic Diet. Epilepsy Research 37:171-180. 82 BIBLIOGRAPHY Shafer, P .O . and Sa lmanson , E. (1997). "Psychosoc ia l Aspec t s of Epi lepsy" , Chapter 5 in Schacter , S . C . and Schomer , D.L. (eds.): The Comprehens ive Evaluat ion and Treatment of Epi lepsy. S a n Diego: A c a d e m i c P ress . Shambaugh , G . E . and Koehler , R A . (1983). Fetal fuels. VI. Metabol ism of a -Keto isocapro ic Ac i d in the Fetal Rat Brain. Metabolism 32: 421-427. Sherwin, A . , Robitai l le, Y . and Quesney , F. (1988). Excitatory Amino A c i d s are E levated in Human Epi lept ic Cerebra l Cortex. Neurology 38: 920-923. Si l ls , M A . , Forsythe, W.I. and Haidukewych, D. (1986a). Ro le of Oc tano ic and Decano ic A c i d s in the Control of Se izures . Arch of Dis Child 6 1 : 1 1 7 3 - 1 1 7 7 . Si l ls , M A . , Forsythe, W.I., Ha idukewych, D., Macdona ld , A . and Rob inson , M. (1986b). The Med ium Cha in Tr ig lycer ide Diet and Intractable Epi lepsy. Arch of Dis Child 61 : 1168-1172. Smith, Q. R., Momma, S. , Aoyag i , M. and Rapoport , S . I. (1987). Kinet ics of Neutral Amino Ac id Transport A c r o s s the B lood-Bra in Barrier. J Neurochem 49:1651-1658. S o , E L . (1993). Update on Epi lepsy. Med. Clin. North. Amer. 77(1): 203-214. 83 BIBLIOGRAPHY Stafstrom, C . E . (1999). An imal Mode ls of the Ketogenic Diet: Wha t Have W e Learned, Wha t C a n W e Learn? Epilepsy Res 37:241-259. Swink, T.D. , V in ing, E . P . G . , F reeman, J . M . (1997). The Ketogen ic Diet: 1997. Adv. Pediatr. 44: 297-329. Tremblay, E., Ro is in , M.P. , R e p r e s a , A . , Charr iaut -Mar langue and Ben-Ar i , Y . (1988). Transient Increased Density of N M D A Binding S i tes in the Deve lop ing Rat H ippocampus. Brain Res 461 : 393-396. Trimble, M.R. (1993). Behaviora l and Cogni t ive Issues in Ch i ldhood Epi lepsy, C h . 36 in Dodson , W . E . and Pel lock, J . M . (eds.): Pediatr ic Epi lepsy: D iagnos is and Therapy. N e w York, Demos Publ icat ions: pp 387-407. Uh lemann, E .R. and Neims, A . H . (1972). Ant iconvulsant Propert ies of the Ketogen ic Diet in M ice . J Pharmacol Exp Ther 180: 231 -238. V in ing, E . P . G . , Kwiterovitch, P. Hs ieh , S. , C a s e y , J . and F reeman, J . (1996). The Effect of the Ketogenic Diet on P l a s m a Cholestero l . Epilepsia 37(suppl 5): 107. W h e l e s s , J . W . (1995). The Ketogenic Diet: Fa(c)t or Fict ion. J of Child Neurol 10(6): 419-423. 84 BIBLIOGRAPHY Wilder , R . M . (1921). The Effects of Ketonemia on the Cou rse of Epi lepsy. Mayo Clin Proc 2 :307-308. Wuar in , J . - P . , K im, Y.I. and C e p e d a , C . (1990). Synapt ic Transmiss ion in Human Neocor tex R e m o v e d for Treatment of Intractable Ep i lepsy in Chi ldren. Ann of Neurol 28: 503-510. Yudkoff, M., N iss im, I., K im, S . U . , P leasure , D., Hummeler, K. and Sega l , S . (1983). [ 1 5 N]-Leuc ine as a Source of [ 1 5 N]-Glutamate in Organotyp ic Cerebe l la r Explants. Biochem Biophys Res Commun 115:174-179. Yudkoff, M., N iss im, I., and Hertz, L. (1990). Precursors of Glu tamic Ac i d Nitrogen in Pr imary Neuronal Cul tures: Stud ies Wi th 1 5 N . Neurochem Res 15:1191-1196. Yudkoff, M., Daikhin, Y . , Lin, Z . - P . , Stern, J . , P leasure , D. and N iss im, I. (1994a). Inter-relationships of Leuc ine and Glutamate Metabol ism in Cul tured Astrocytes. J of Neurochem 62 :1192-1202. Yudkoff, M., Dakhin, Y . , P leasure , D., Stern, J . and Niss im, I. (1994b). Inhibition of Astrocyte Glutamine Product ion by a -Keto isocapro ic Ac id . J of Neurochem 63:1508-1515. 85 BIBLIOGRAPHY Yudkoff, M., Daikhin, Y . , Grunste in, L., Stern, J . , P leasure , D. and Niss im, I. (1996a). Astrocyte Leuc ine Metabol ism: Sign i f icance of B ranched-Cha in Amino Ac id Transaminat ion. J of Neurochem 66:378-385. Yudkoff, M., Daikhin, Y . , N iss im, I., P leasure , D. Stern, J . and Niss im, I. (1996b). Inhibition of Astrocyt ic Glutamine Pduct ion by a -Keto isocapro ic Ac id . J of Neurochem 63 : 1508-1515. Yudkoff, M. (1997). Brain Metabol ism of B ranched-Cha in Amino Ac ids . GLIA 21 : 92-98. Yudkoff, M., Daikhin, Y . , N iss im, I., Grunste in, R. and Niss im, I. (1997). Effects of Ketone Bod ies on Astrocyte Amino Ac id Metabol ism. J of Neurochem 69: 682 -692. Z ie lke, H.R., Huang , Y . , Baab , P . J . , Col l ins, R .M. , Z ie lke, C . and Ti ldon, J .T. (1997). Effect of c t -Keto isocaproate and Leuc ine on the In V i vo Oxidat ion of Glutamate and Glutamine in the Rat Brain. Neurochem Res 22(9): 1159-1164. 86 APPENDIX 11 APPENDIX 12000 0 50 100 150 200 Time (minutes) Figure A.1: Time course ofthe production of 1 4 C 0 2 production from [U - 1 4 C ] -leucine, as reflected by DPM production in astrocyte cell cultures 8 7 APPENDIX Q_ Q 0.05 0.1 Leucine concentration (mM) Figure A.2: Production of 1 4 C 0 2 as reflected by DPM production in astrocytes cultured with increasing concentrations of [U-1 4C]-leucine. 8 8 

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}]}"
                            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:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0090247/manifest

Comment

Related Items