UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Interaction of pyridoxal phosphate with acetyl-CoA carboxylase Elliott, Jason Edward 2003

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

Item Metadata

Download

Media
831-ubc_2003-0407.pdf [ 5.01MB ]
Metadata
JSON: 831-1.0090953.json
JSON-LD: 831-1.0090953-ld.json
RDF/XML (Pretty): 831-1.0090953-rdf.xml
RDF/JSON: 831-1.0090953-rdf.json
Turtle: 831-1.0090953-turtle.txt
N-Triples: 831-1.0090953-rdf-ntriples.txt
Original Record: 831-1.0090953-source.json
Full Text
831-1.0090953-fulltext.txt
Citation
831-1.0090953.ris

Full Text

INTERACTION OF PYRIDOXAL PHOSPHATE WITH ACETYL-CoA CARBOXYLASE JASON EDWARD ELLIOTT B.Sc, The University of British Columbia, 2001 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES DEPARTMENT OF BIOCHEMISTRY AND MOLECULAR BIOLOGY We accept this thesis as confoirning to the required standard THE UNIVERSITY OF BRITISH COLUMBIA August 2003 © Jason Edward Elliott, 2003 In presenting t h i s thesis i n p a r t i a l f u l f i l m e n t of the requirements for an advanced degree at the University of B r i t i s h Columbia, I agree that the Library s h a l l make i t f r e e l y a v a i l a b l e for reference and study. I further agree that permission for extensive copying of t h i s thesis for s c h o l a r l y purposes may be granted by the head of my department or by his or her representatives. It i s understood that copying or pu b l i c a t i o n of t h i s thesis for f i n a n c i a l gain s h a l l not be allowed without my written permission. Department of D ><3illfrniS-|ry k lioWolftf B.Q/Q The University of B r i t i s h Columbia Vancouver, Canada A B S T R A C T Acetyl-CoA carboxylase (ACC) catalyzes the conversion of acetyl-CoA to malonyl-CoA, the first committed step in de novo fatty acid synthesis. Two mammalian A C C isoforms exist, with many distinctive properties and differential tissue distribution. A C C isoforms exist as inactive dimers that can be allosterically activated by citrate; activation being associated with polymerization, in the case of ACC-1 . Inhibition of other citrate-utilizing enzymes by pyridoxal phosphate (PLP) suggested that PLP might also inhibit A C C . This thesis describes work done to assess the interaction of PLP with A C C . PLP was found to be a potent inhibitor of A C C , with an apparent IC50 of 225 u M against A C C purified from liver. The effects of PLP were seen in affinity-purified A C C , suggesting its effects are direct. PLP is effective against both A C C isoforms, but ACC-1 shows greater sensitivity to PLP than ACC-2. PLP inhibition of A C C is rapid, occurring within 1 minute of addition. Pre-treatment of A C C with citrate provided substantial protection against PLP inhibition at low inhibitor concentrations; the protection being more evident with ACC-1 than ACC-2. This suggests PLP and citrate bind to the same site on A C C . Treatment with PLP and sodium borohydride leads to irreversible A C C inhibition, confirming that PLP forms a Schiff base with A C C that can be reduced with sodium borohydride. In the absence of borohydride reduction PLP interacts reversibly with A C C . Selective labelling of A C C using [3H]-borohydride indicated that PLP reacts directly with the A C C subunits. Numerous analogs of PLP were tested for inhibitory action against A C C . From these studies it was concluded that the aldehyde and negative charges on PLP are especially important for its inhibitory action on A C C . The pyridine ring is also important, but to a lesser degree than the aldehyde or phosphate. The ability to use PLP to determine the sequence of the citrate binding site and to provide the basis for inhibitor design is discussed. ii TABLE OF CONTENTS A B S T R A C T ii T A B L E OF CONTENTS i i i LIST OF T A B L E S vi LIST OF FIGURES vii LIST OF ABBREVIATIONS ix A C K N O W L E D G E M E N T S x CHAPTER 1 INTRODUCTION 1 1.1 PREFACE 1 1.2 LIPID M E T A B O L I S M 2 1.2.1 Triacylglycerol Metabolism 2 1.3 A C E T Y L - C o A C A R B O X Y L A S E 7 1.3.1 Structure and Function of A C C 7 1.3.2 ACCIsoforms 8 1.3.3 A C C Regulation - Long Term 11 1.3.4 A C C Regulation - Short Term 12 1.3.4.1 Allosteric Regulation of A C C 12 1.3.4.2 Regulation of A C C by a Protein Regulator 14 1.3.4.3 A C C Phosphorylation 15 1.4 P Y R I D O X A L 5'-PHOSPHATE 17 1.4.1 Pyridoxal Phosphate Structure and Function 17 1.4.2 Experimental Use of PLP 20 1.4.2.1 Measuring PLP in the Body 22 1.4.2.2 PLP and Reactive Lysine Residues 23 1.4.2.3 PLP and Metabolite Transport Measurement 24 1.4.2.4 PLP and Enzymatic Activity Determination 25 1.5 THESIS INVESTIGATION 26 i i i CHAPTER 2 E X P E R I M E N T A L PROCEDURES 28 2.1 MATERIALS 28 2.2 METHODS 29 2.2.1 Tissue Preparation 29 2.2.1.1 Isolation 29 2.2.1.2 Adipose Tissue Incubation 29 2.2.1.3 Homogenization 29 2.2.2 Purification of Acetyl-CoA Carboxylase 30 2.2.2.1 Ammonium Sulphate Precipitation 30 2.2.2.2 Avidin Affinity Chromatography 31 2.2.3 A C C Activity Assays 32 2.2.4 Inhibition Assays 33 2.2.5 Sodium Borohydride Reductions 33 2.2.6 SDS-PAGE 34 2.2.7 PLP Spectrophotometry : 35 CHAPTER 3 PLP EFFECTS ON A C C 36 3.1 RATIONALE 36 3.2 RESULTS A N D DISCUSSION 37 3.2.1 A C C Activation by Citrate 37 3.2.2 A C C Detection by Electrophoresis 39 3.2.3 Inhibitory Effect of PLP on A C C 41 3.2.4 Time Dependence of PLP Inhibition 48 3.2.5 Citrate Protection Against PLP Inhibition of A C C 55 3.2.6 Effect of PLP on A C C Activated by Glutamate 65 3.2.7 Potential Substrate Protection Against PLP Inhibition of A C C 69 3.2.8 Effect of PLP Reactive Compounds on PLP-Inhibited A C C 71 3.2.9 Effect of PLP Reduction by Sodium Borohydride 74 3.2.10 Spectrophotometric Analysis of PLP Reduction 79 3.3 S U M M A R Y 82 iv CHAPTER 4 PLP A N A L O G S ACCOUNT FOR PLP SPECIFICITY 83 4.1 RATIONALE 83 4.2 RESULTS A N D DISCUSSION 84 4.2.1 Possible PLP Moieties That Could Account For A C C Inhibition 84 4.2.2 PLP's Reactive Aldehyde 85 4.2.3 PLP's Phosphate Group 89 4.2.4 PLP's Anionic Poles and Pyridine Ring 91 4.2.5 Non-Aromatic PLP Analogs 99 4.2.5 Use of A C A and an Analog 103 4.3 S U M M A R Y 109 CHAPTER 5 CONCLUSIONS & FUTURE GOALS 110 REFERENCES 114 v LIST OF TABLES Table 1.1 Acetyl-CoA Carboxylase Phosphorylation 16 Table 3.1 Effect of A C C Pre-Incubation With Citrate on PLP Inhibition 56 Table 3 2 Comparison of A C C Pre-Incubation With Citrate and Glutamate on PLP 68 Inhibition Table 3.3 Effect of Hydroxylamine and Aminooxyacetate on PLP-treated A C C Activity... 75 Table 4.1 Summary of PLP Analogs and Their Effects on A C C 106 vi LIST OF FIGURES Figure 1.1 Insulin-related control points for lipid biosynthesis in adipose tissue 6 Figure 1.2 ACC-Related Control of Fatty Acid Oxidation 11 Figure 1.3 Structures of F36 vitamers 19 Figure 1.4 PLP-dependent Transamination Mechanism 21 Figure 3.1 Activation of A C C by Citrate 38 Figure 3.2 A C C Detection by Electrophoresis and Coomassie Blue Staining 40 Figure 3.3 PLP Inhibition of Liver A C C 42 Figure 3.4 PLP Inhibition of Various A C C Preparations 43 Figure 3.5 PLP Inhibits Highly Purified A C C 47 Figure 3.6 Time Dependence of PLP Inhibition of A C C - Long-Term 49 Figure 3.7 Time Dependence of PLP Inhibition of Liver, Heart and Skeletal Muscle A C C . 51 Figure 3 8 Effect of Citrate Pre-Incubation on Time Dependence of PLP Inhibition of 53 Liver and Skeletal Muscle A C C Figure 3.9 Effect of Citrate Pre-Incubation on PLP Inhibition of A C C 57 Figure 3.10 Citrate Activation Curves at Constant PLP Concentrations 62 Figure 3.11 Double Reciprocal Plot and K i Determination 64 Figure 3.12 Effect of PLP on A C C Activated by Glutamate 65 Figure 3.13 Effect of Increasing A C C Substrates on PLP Inhibition of A C C 70 Figure 3.14 Structures of PLP-Reactive Compounds 72 Figure 3.15 Effect of Hydroxylamine and Aminooxyacetate on PLP-treated A C C Activity... 73 Figure 3.16 Effect of Sodium Borohydride Reduction of PLP-Labelled A C C 77 Figure 3.17 PLP/BH4 Incorporation in A C C 78 Figure 3 18 Spectrophotometric Analysis of PLP Reduction by B H 4 With Lysine & 81 Proteins Figure 4.1 Possible PLP Moieties Accounting For Specificity and Binding Efficiency 84 Figure 4.2 Effect of Aldehyde-Lacking PLP Analogs 86 Figure 4.3 Effect of Pyridinecarboxaldehydes on A C C 88 Figure 4.4 Effect of Phosphate-Lacking PLP Analogs 90 vii Figure 4.5 Comparison of PLP and Citrate Structures 92 Figure 4.6 Effect of Pyridinedicarboxylic Acids on A C C 93 Figure 4.7 Effect of Benzenetricarboxylic Acids on A C C 95 Figure 4.8 Effect of Phthalic Acid and Nicotinic Acid on A C C 97 Figure 4.9 Effect of Non-Ring PLP Analogs on A C C 100 Figure 4.10 Structures of Some of the PLP Analogs Tested 101 Figure 4.11 Structures of Some of the PLP Analogs Tested 102 Figure 4.12 Effect of A C A and Acetylsalicylic Acid on A C C 104 Figure 4.13 Structures of A C A and Acetylsalicylic Acid 105 LIST OF ABBREVIATIONS A### Absorbance at ### nanometers A C A 1 'acetoxychavicol acetate A C C Acetyl-CoA carboxylase A M P Adenosine monophosphate A M P K AMP-activated protein kinase A O A Aminooxyacetate B H 4 Borohydride (typically the sodium salt) a-ME a-mercaptoethanol or 2-mercaptoethanol BR Bradford reagent cAMP Cyclic-AMP CaMK Calcium/Calmodulin dependent protein kinase C K Casein kinase CPT-1 Carnitine Palmitoyl Transferase I D N A Deoxyribonucleic acid FAS Fatty acid synthetase FFA Free fatty acid GLUT4 Glucose transporter 4 H A Hydroxylamine H C O 3 " Bicarbonate HPLC High performance liquid chromatography HRP Horseradish peroxidase HSL Hormone sensitive triacylglycerol lipase IC50 Concentration of a compound required to show 50% inhibition of a target Lys Lysine PDA Piperazine diacrylamide PDH Pyruvate dehydrogenase PDHP Pyruvate dehydrogenase phosphatase PEP Phosphoenolpyruvate P K A cAMP-dependent protein kinase PLP Pyridoxal 5'-phosphate R E A Radioenzymatic assay R N A Ribonucleic acid SDS Sodium dodecyl sulphate S.E.M. Standard error of the mean Ser Serine T A G Triacylglycerol T C A Tricarboxylic acid cycle (also known as Krebs Cycle) V L D L Very low density lipoproteins ix ACKNOWLEDGEMENTS I would like to thank Professor Roger W. Brownsey for his generous offer of a student researchship during my degree. As well, his guidance, suggestions and availability to discuss ideas made this degree a challenging, interesting and fulfilling experience for me. He is a great scientist and a great supervisor and I learned a lot from him. Other members of the lab also helped me in my work. Dr. Jerzy Kulpa was always there to lend a hand, and his assistance with several experiments was immensely helpful. Kioumars Jelveh and Weissy Lee, fellow graduate students in the lab, gave helpful pointers and kept things interesting through conversation, jokes and lunches in the hallway. Mary Pines, Kenneth Chan and Jeff Chen, undergraduate members of our lab, also helped me with my work, and helped keep the lab lively. I would also like to thank the members of my Research Committee, Dr. Robert Molday and Dr. Michel Roberge, for their suggestions and criticisms. These helped me make the project more complete and focussed and for this I extend to them my thanks. Finally, I would like to thank my family and friends for their support during my degree. Their encouragement and insistence that I "get out of that lab!" helped me explore other interests of mine, which made the degree as a whole an enjoyable experience. x C H A P T E R O N E I N T R O D U C T I O N 1.1 PREFACE An unfortunate but very evident fact exists: people in developed countries, especially North America, have higher rates of obesity and Type II diabetes than elsewhere in the world. And the prevalence of these conditions is on the rise. While living in developed countries does not predispose one to becoming overweight, the generally sedentary lifestyle and high calorie, high fat foods that have been and are being consumed in these countries do contribute to this phenomenon of increasing body weight and diabetic prevalence in the population. While the primary deficit in Type I diabetes (seen most often in children) is autoimmune destruction of pancreatic (3-cells (which produce insulin), the type II form of the disease is most prevalent. Type II diabetes is characterized by impaired insulin action, largely due to defects in insulin sensitivity. Type II diabetes accounts for about 90% of the cases of this disease, and being overweight or obese is strongly correlated with development of type II diabetes. In the United States in 2001, 58% of the population was overweight, and those overweight had over 7 times the risk of developing diabetes than those with normal body mass [1]. The link between diabetes and obesity is clearly important. Equally important is understanding how the body becomes overweight. This is not an easy task, as there are numerous factors which can lead to weight gain. However, since most of the increase in body mass is associated with accumulation of fat, an understanding of fat metabolism is vital. 1 1.2 LIPID METABOLISM The metabolism of lipids represents a complex and highly regulated process, which has three main roles in the body. First is storage and release of energy in the form of triacylglycerols, which act as long-term energy stores and are most abundant of the lipids in the body. Second, glycerophospholipids and sphingolipids comprise the bulk of the lipid content of biological membranes, with the latter especially prevalent in nerve cells. Thirdly, the steroids, including cholesterol, represent components of biological membranes, are present in blood plasma in the form of various lipoproteins, including very low-density, low-density and high-density lipoproteins (VLDL, L D L and HDL), and can act as cellular messengers or neurotransmitters. The intricate balance of biosynthesis and breakdown of these lipids is important in maintaining health, and imbalances can lead to cellular damage, disease and death. Therefore, controlling the rate of intake, production, utilization and breakdown of lipids is fundamental to correct operation of the body. The mechanisms by which the different classes of lipids are controlled are both similar and distinct. The metabolism of glycerophospholipids and steroids is complex and involves many integrated control systems, and will not be discussed here. However, fatty acid and triacylglycerol metabolism is a central aspect of this thesis, and knowledge of the relevant pathways and controls is important in understanding the relevance of this work. 1.2.1 Triacylglycerol Metabolism Triacylglycerols (TAGs), composed of a glycerol backbone with three esterified fatty acid residues, make up the largest proportion of lipids in the body and are the primary long-2 term energy storage supply. In times of excess caloric intake, unused food "energy" is converted into TAGs and stored, primarily in the adipocytes of adipose tissue, but also to a small degree in hepatocytes of the liver. In times of low caloric intake, these TAGs are metabolized to provide energy. This T A G breakdown is promoted by the action of the hormone sensitive triacylglycerol lipase (HSL), so named due to its sensitivity to regulation by phosphorylation and dephosphorylation in response to hormonally controlled cyclic A M P (cAMP) levels. Catabolic hormones such as glucagon and norepinephrine increase adipose tissue cAMP concentrations, thereby allosterically activating cAMP-dependent protein kinase (PKA). P K A then phosphorylates certain enzymes, such as hormone-sensitive lipase and acetyl-CoA carboxylase (ACC). Phosphorylation of HSL activates the enzyme, and together with phosphorylation of perilipin, lead to stimulation of triacylglycerol hydrolysis. This conversion of T A G into free fatty acids (FFA), raises blood FFA levels, and stimulates (3-oxidation of the fatty acids in other tissues such as liver and muscle. Before p-oxidation, the fatty acids are first esterified with coenzyme A, making acyl-CoA esters, and the acyl group is then transferred to carnitine (via CPT-I), producing acyl-carnitines. These acyl carnitines are transported across the mitochondrial membrane by the translocase associated with CPT-I. Once inside the mitochondria, the acyl group is once again transferred back to CoA (via CPT-II) and the acyl-CoA undergoes (3-oxidation. In parallel with cAMP-mediated activation of lipid catabolism, phosphorylation of A C C (by P K A and other kinases) inactivates the enzyme, thus preventing fatty acid synthesis. T A G synthesis is accomplished via a highly controlled pathway. Assuming that de novo lipid synthesis begins with the supply of glucose, several control points exist in the 3 generation of cytoplasmic pyruvate, notably GLUT4, PFK-1 and pyruvate kinase. Another crucial control point is the production of acetyl-CoA from pyruvate, by pyruvate dehydrogenase (PDH). This acetyl-CoA can then enter the T C A cycle or can be converted to malonyl-CoA by A C C , arguably the first committed step in fatty acid biosynthesis. This malonyl-CoA product is then used as the 2-carbon substrate of fatty acid synthetase (FAS), which produces an acyl chain end-product, usually 16 carbons long (palmitate). Clearly, several control points exist in the overall pathway, and those relevant to this thesis will be discussed further. Insulin, in many respects, antagonizes the effects of glucagon and norepinephrine. Insulin decreases cAMP levels, thus inactivating P K A and preventing both activation of HSL and inactivation of A C C . Insulin also results in activation of GLUT4, PFK-1, PDH and A C C , and although the mechanism of these activations is still not completely understood, they appear to involve a variety of processes. The first step is promotion of glucose uptake via stimulation of the insulin receptor, which activates signalling cascades resulting in translocation of intracellular vesicles from subcellular locations to the plasma membrane. These vesicles contain the GLUT4 glucose transporter (insulin sensitive), and this increase in glucose transporters on the plasma membrane leads to an increase in glucose uptake by the cell, resulting in available fuel for fat synthesis in adipose tissue. In general, regulation of GLUT4 shows many similarities to that of synaptic vesicle trafficking (For reviews, see [2-4]). The regulation of PFK-1 in response to insulin, like the other metabolic effects, is dependent on PI3-kinase activation. The generation of PIP3 leads to activation of downstream kinases including PDK-1, PDK-2 and PKB. PFK-1 appears to be activated following phosphorylation by P K B or a related kinase. The next major control point is pyruvate 4 dehydrogenase (PDH). Insulin activates PDH phosphatase (PDHP), which maintains PDH in an active, dephosphorylated state, thus ensuring continued production of acetyl-CoA for fatty acid synthesis [5]. Figure 1.1 illustrates these control points. Of considerable interest, acetyl-CoA carboxylase (ACC) represents the first committed step of fatty acid synthesis and is highly regulated. As mentioned, insulin promotes A C C activation, but the enzyme is also controlled by other phosphorylation and dephosphorylation events, as well as allosterically and through association with at least one regulator protein. These varied control mechanisms together with other control features suggest A C C is a key point in fat metabolism, and understanding of how this enzyme is regulated can provide new insight into localized and body-wide fat metabolism. 5 Figure 1.1 Selected Insulin-related control points for lipid biosynthesis in adipose tissue. Insulin Synthesis Visual representation of selected insulin-mediated control points for lipid biosynthesis in adipose tissue. The first control point is through the insulin receptor, mediating glucose uptake via GLUT4 vesicular transport up-regulation (1). Activation of PI3-kinase and production of PIP3 result in activation of P K B (not shown). P K B (or a related kinase) then phosphorylate and activate PFK-1 (2). Insulin activation of pyruvate dehydrogenase phosphatase ensures that PDH remains active (3). Insulin decreases cellular cAMP levels, resulting in decreased activity of acetyl-CoA carboxylase (4) via inactivated P K A , and inactivation of hormone sensitive lipase (5). 6 1.3 ACETYL-CoA CARBOXYLASE (ACC) 1.3.1 Structure and Function of A C C Acetyl-CoA carboxylase (EC 6.4.1.2) is a biotin-containing carboxylase, catalyzing the first committed step in de novo fatty acid synthesis. The catalysis occurs in two steps. The first is transfer of a carboxyl group ( C O 2 ) from an activated bicarbonate substrate to a biotin prosthetic group in the enzyme, producing carboxy-biotin. This energy dependent step requires ATP, and is shown as (1). The second step involves carboxyl group transfer from carboxy-biotin to an acetyl-CoA substrate, producing the product, malonyl-CoA (2). Both of these partial reactions are blocked by pre- or co-incubation with avidin [6]. (1) ACC-biotin + ATP + HCO3" -> ACC-carboxy-biotin + ADP + Pi (2) ACC-carboxy-biotin + acetyl-CoA -> ACC-biotin + malonyl-CoA A C C is considered to contain 3 functional domains: the biotin carboxylase domain, the carboxyltransferase domain and a 'flexible' biotin arm which can access both distinct active sites. In most prokaryotes, A C C catalysis is carried out by a complex, with the functional catalytic domains present on two or more polypeptides. For example, in Escherichia coli, A C C is comprised of four polypeptides, each encoded by a distinct gene. Several groups have crystallized and determined the three-dimensional structures of these component polypeptides individually, but as of yet, no structure exists for the whole complex [7, 8]. Yeast A C C is encoded by a single gene and is highly related to other eukaryotic forms. It is expressed as a tetramer with a subunit size of -265 kDa [9]. Animal A C C exists in two isoforms, each encoded by a separate gene on separate chromosomes, and each 7 comprised of a single polypeptide containing the three functional domains. The isoforms/subunits range in size from 265-280 kDa and the native proteins are dimeric under non-denaturing conditions. Unlike other A C C forms, these animal A C C ' s are induced to assemble into polymers of 5-20 dimers under the influence of di- and tri-carboxylic acids. As well, unlike the prokaryotic forms of A C C , no single-polypeptide form has been structurally characterized, mostly due to the large size of the protein. However, very recently, a fragment of the yeast enzyme, comprising the carboxyltransferase domain was crystallized and the structure solved [10]. The significant similarity in primary sequence between yeast and mammalian enzymes suggests that this structure may be the first step towards a complete structure for the mammalian enzyme. 1.3.2 ACCIsoforms As mentioned, mammalian A C C exists as either of two isoforms. The smaller, (ACC-1 or A C C - a or ACC-265) at 265 kDa, was the first isoform characterized. The second, larger isoform at 280 kDa (ACC-2 or ACC-(3 or ACC-280) was discovered more recently, and has 142 additional N-terminal residues [11]. This isoform was first identified in heart tissue and this discovery was unexpected, since heart tissue is highly catabolic and non-lipogenic and A C C was not thought to be required in this tissue. The increased molecular weight of ACC-2 was also confusing, since no A C C isoforms were known at the time, but using immunological and other classical techniques, ACC-2 was shown to indeed be an A C C isoform found most prevalently in the heart [12], although it was shown shortly thereafter that it is also found in liver, skeletal muscle, mammary gland and brown adipose tissue [13]. 8 Although sharing over 60% primary sequence homology, the two isoforms show significant differential tissue distribution, as alluded to previously. In rats, ACC-1 is the sole isoform expressed in white adipose tissue, testes, bladder, uterus and placenta. In tissues such as the liver, lactating mammary and pancreas, about 75% of the expressed A C C is ACC-1 , with ACC-2 as the other 25%. However, in heart and skeletal muscle, ACC-2 is the predominantly expressed isoform. Differential distribution of protein isoforms often suggests differing roles for each isoform, and this was suspected for A C C . The metabolic nature of the tissue in which each is expressed, as well as the sequence differences between them indicates possible roles for each isoform. Since lipogenic tissues (liver, adipose) express mainly ACC-1 , it has been speculated that ACC-1 is largely responsible for de novo fatty acid synthesis, producing malonyl-CoA for use by fatty acid synthase (FAS) to make palmitate. However, as there is little de novo fatty acid synthesis in heart and skeletal muscle, due to absence of FAS, the role of ACC-2, which is more highly expressed in these tissues, is unlikely to be fatty acid production. Instead, ACC-2 is believed to be involved in regulation of P-oxidation in muscle tissue. This is achieved through the action of the A C C product, malonyl-CoA. Malonyl-CoA is a potent inhibitor of carnitine palmitoyl transferase-I, which catalyzes the formation of acyl-carnitines prior to translocation into the mitochondria for P-oxidation. Some recent evidence suggests that the additional N-terminal sequence on ACC-2 is a targeting motif for the outer mitochondrial membrane, and immunofluorescence evidence shows that an N -terminal fragment of ACC-2 does localize to the mitochondrial membrane [14]. This co-localization is believed to place ACC-2 in close proximity to CPT-I, thus allowing tight control over fatty acid transport into the mitochondria. When ACC-2 is active, increased 9 production of malonyl-CoA may inhibit CPT-I and inhibit fatty acid oxidation. The reverse is true when ACC-2 is inactivated. One significant difference between the two A C C isoforms is their requirement for survival. Disrupting one A C C allele in diploid yeast prevents vegetative growth, suggesting that A C C is essential for survival [15]. Surprisingly, long-chain fatty acids cannot, alone, restore growth, suggesting function beyond formation of CI6 (palmitate). In animals, given that two A C C isoforms exist (unlike in yeast where there is only one form of the enzyme), it was hypothesized that deletion of ACC-2, given its potential role in (3-oxidation control, might not result in a lethal phenotype, whereas ACC-1 deletions likely would. Indeed, recent evidence shows that a deletion of ACC-2 in mice is non-lethal [16]. In fact, the knockout mice were more lean and consumed more food than wild-type mice, illustrating the effect that ACC-2 has on regulating fatty acid oxidation. Fat oxidation control is shown in Figure 1.2. The important concept regarding the role of A C C in metabolism is that i f the enzyme is active, then malonyl-CoA can be used to synthesize lipids and inhibit lipid oxidation; whereas i f A C C is inactivated, decreased malonyl-CoA reduces lipid synthesis and relieves inhibition of carnitine palmitoyl transferase I, allowing |3-oxidation of lipids to proceed. In short, active A C C contributes to net lipid production, while inactive A C C should contribute to net fat oxidation. 10 Figure 1.2 ACC-Related Control of Fatty Acid Oxidation. Fatty Acids • complex lipids 1.3.3 A C C Regulation - Long Term ACC-2 expression in muscle tissue is not altered during conditions of altered diet [13, 17], as shown using immunoprecipitation with A C C isoform-specific antibodies [13], or in streptozotocin-induced diabetes [18]. It does show repression during hibernation [19], and concomitant repression with ACC-1 in liver tissue during starvation and diabetes [20]. However, ACC-1 shows considerable variation in expression under different physiological states. For example, ACC-1 is repressed in adipose and liver tissue in rats fed high-fat diets, during starvation, and in diabetes induced by insulin deficiency [21]; conversely, the repression of ACC-1 in these conditions could be reversed by low-fat feeding, feeding, and 11 insulin treatment, respectively [20]. It has been suggested that several factors, including mRNA transcription and stability and protein translation are involved in varying the ACC-1 levels under repression conditions [22, 23]. Three promoters control ACC-1 gene expression. The first responds to hormonal and dietary conditions [24], with the second expressed constitutively [24] or under control of SREBP transcription factors [25], and the third induced during lactation, under influence of the STAT5 transcription factor [26, 27]. The single A C C -2 promoter is activated by transcription factors such as MyoD, Myf4 and Myf6 [28]. 1.3.4 A C C Regulation - Short Term In addition to the longer term regulation of A C C through control of gene expression, shorter term controls are also responsible for regulating the enzyme. Three different mechanisms seem to account for the short term regulation of A C C : small allosteric ligands, interacting proteins and A C C phosphorylation and dephosphorylation by several protein kinases and phosphatases. As mentioned, hormones can exert effects on A C C activity, and this process occurs very rapidly. Glucagon and epinephrine result in decreased A C C activity, coinciding with lipid oxidation, and insulin induces A C C activation, concomitant with lipid biosynthesis. These hormone controls are primarily exerted through phosphorylation events on A C C . 1.3.4.1 Allosteric Regulation of A C C As previously indicated, animal A C C ' s have the ability to form polymers, under the influence of di- and tri-carboxylic acids. The polymeric form represents active A C C , while the dimeric form is inactive. This polymerization was first observed in 1962, using sucrose 12 density centrifugation, which indicated that A C C which had been pre-activated by citrate (the most effective carboxylate activator [12, 17, 29-31]) sedimented at a higher velocity than enzyme lacking citrate treatment [30]. Since the polymerization was complete within minutes (and difficult to study with the techniques available at the time), the depolymerization process was studied instead, due to its slower rate. This promoted the discovery that depolymerization and loss of A C C activity were positively correlated. As well, treatment of A C C dimers with avidin resulted in inactivation, while A C C polymers were resistant to avidin. This indicated either that the dimer-dimer interaction regions were in close proximity to the biotin binding region (and avidin binding blocks these regions from interacting) or that polymerization indirectly results in shielding of biotin on A C C from avidin [32]. Citrate is the best allosteric activator of A C C , but glutamate also activates A C C effectively [33]. Although the site of binding of these acids is unknown, it probably does not occur in the active sites, nor at the binding site for biotin, nor at the sites of phosphorylation, but this has not been established. Other molecules can have the reverse effect: CoA [34], fatty acyl-CoA esters [12, 31, 35], and chloride [36] all promote A C C depolymerization and deactivation. A C C is also inhibited by malonyl-CoA [12, 31] and a biotin analog, CABI (chloroacetylated biotin derivative) has also been shown to be a fairly potent inhibitor of the bacterial A C C carboxyltransferase subunit [37]. Several pentenedioic acid-based metabolites from the fungus Gongronella butleri have been shown to inhibit the enzyme as well [38]. Additionally, herbicides of the aryloxyphenoxypropionic acid class inhibit plant ACCs [39], possibly by acting on the carboxyltransferase domain in a eukaryotic form of the enzyme (usually present in plants lacking an insensitive prokaryotic form of ACC) [40]. Although significant information regarding their structural features and how these affect their potency 13 on plant ACCs is known, the effect of these herbicides on animal ACCs is not well studied. One study suggests that CoA-esters of aryloxyphenoxypropionates act as potent inhibitors of rat liver A C C [41], but these results have not been sufficiently validated to establish these herbicides as mammalian A C C inhibitors. Nonetheless, there is also insufficient evidence to rule out the inhibitory actions of these herbicides on mammalian A C C . Additionally, as with many other enzymes, A C C is inhibited by high salt and low protein concentrations as well as alkaline pH [42], although very low salt (less than 100 uM) leads to irreversible A C C aggregation and inhibition. Surprisingly, even though ACC-2 is activated by citrate [12, 17, 29] it does not form polymers like ACC-1 [43]. Although ACC-2 does not seem to form homodimers, some evidence exists for the formation of heterodimers with ACC-1 [13, 44]. This suggests the two isoforms may be co-regulated, through a form of direct linkage, at least in some tissues. 1.3.5.2 Regulation of A C C by a Protein Regulator Experiments with size exclusion and ion-exchange chromatography revealed two proteins which affected the activity of A C C , one in liver [45] and one in adipose tissue [46]. These are 75 and 130 kDa, respectively, but any similarity between the properties of these proteins has not been elucidated. In the absence of these proteins, the sensitivity of A C C to citrate, is significantly lower than when they are present. Since polymerization is partially preceded by citrate binding, it is possible that these regulator proteins somehow act to facilitate A C C polymerization or citrate binding, but the exact mechanism of their action remains unclear. 14 1.3.5.3 A C C Phosphorylation The phosphorylation events that regulate A C C activity are numerous, complicated and still not completely defined. As mentioned earlier, A C C is activated by insulin and inhibited by glucagon and epinephrine. How these hormones exert their effects is through distinct phosphorylation events on A C C by different protein kinases, thus altering the activity of the enzyme. Most evidence concerning identity of specific sites relates to ACC-1 , with sites on ACC-2 still not defined. Since A C C phosphorylation does not represent an important component of this thesis, only a brief discussion of the process will be included. Several inactivating phosphorylation sites and kinases for ACC-1 have been elucidated, as well as sites with an unknown or no clear effect on A C C activity. These are detailed in Table 1.1. Phosphorylation of the T-peptide' is insulin-dependent, but neither the phosphorylating kinase nor the phosphorylation site has been characterized. A M P K phosphorylates A C C at several sites, but site-directed mutagenesis has shown an inhibitory effect is only seen through phosphorylation of Ser-79 on ACC-1 [47]. P K A phosphorylates A C C only on serine residues [43], and on ACC-1 , these have been determined to be Ser-77 and Ser-1200. Mutagenesis has shown that PKA-mediated inactivation of A C C is a result of phosphorylation of Ser-1200 [47]. P K A phosphorylates ACC-2 much more efficiently than ACC-1 [43, 44]. CaMK shows similar substrate sequence specificity as A M P K , but CaMK phosphorylates ACC-1 at Ser-25, and this does not seem to affect A C C activity in vitro [43, 48]. Casein kinase 2 (CK2) phosphorylates ACC-1 at Ser-29, but this has no apparent effect on A C C activity directly. Finally, P K C has been shown to phosphorylate ACC-1 at Ser-77 and Ser-95, with some indication that PKC-mediated phosphorylation of Ser-77 leads to A C C activation. This contrasting result with P K A phosphorylation of the same Ser-77 could be 15 due to impure kinase preparation, and the role of PKC phosphorylation of this site remains undetermined. Table 1.1 Acetyl-CoA Carboxylase-1 Phosphorylation. 1 Phosphorylation Site(s) Phosphorylating Kinase Kinase Control 2 Effect of Phosphorylation Reference Ser-25 CaMKII Ca2 + /CaM, Insulin (all +) None [49, 50] Ser-29 CK2 ? None [48-50] Ser-77 P K C D A G , PS, C a2 + (all +) ? [48,49,51] P K A Catecholamines, cAMP (both +) Insulin (-) Possible inhibition [48, 49,51, 52] Ser-79 A M P K A M P (+) Insulin (-) Inhibition [49, 52] Ser-95 P K C D A G , PS, C a2 + (all +) ? [49,51] Ser-1200 A M P K A M P (+) Insulin (-) Possible Inhibition [49, 52] P K A Catecholamines, cAMP (both +) Insulin (-) Inhibition Ser-1215 A M P K A M P (+) Insulin (-) Possible Inhibition [52] T-peptide' T-peptide kinase' Insulin (+) Possible activation [51, 53] Adapted from [54]. Activating signals are given by a (+) and inhibitory signals by a (-). 16 1.4 PYRIDOXAL 5'-PHOSPHATE (PLP) Although several compounds, both endogenous and exogenous, can inhibit acetyl-CoA carboxylase, some of these, especially exogenous inhibitors are either generally not physiologically relevant to the mammalian enzyme, or limited in their use. Since A C C represents an important control point for lipid metabolism, it is a potential target for pharmacologic inhibitors, whether these be synthesized or naturally derived. Additionally, it is of great interest to fully understand endogenous compounds and mechanisms that regulate A C C activity, since these can offer insight into drug design for targeted A C C inhibition. Pyridoxal phosphate (PLP), the phosphorylated and most common form of vitamin FJ6 in vivo, has previously been shown to inhibit enzymes using citrate as a substrate or as an allosteric activator of enzyme activity. Given these previous results and the understanding that citrate acts allosterically on A C C , it seemed possible that PLP might act on A C C . 1.4.1 Pyridoxal Phosphate Structure and Function Pyridoxal 5'-phosphate (pyridoxal phosphate or PLP) is the biologically active form of vitamin B6 (pyridoxine). Vitamin B 6 functions as a coenzyme in over 100 enzymatic reactions, involved in amino acid, lipid, carbohydrate and neurotransmitter metabolism. The major forms of vitamin B6 in mammalian tissues and fluids are: pyridoxine, pyridoxal, pyridoxamine and their phosphorylated derivatives, pyridoxal 5'-phosphate (PLP) and pyridoxamine 5'-phosphate, as well as the end product of vitamin B6 metabolism, 4-pyridoxic acid. The structures of these vitamers (one of two or more related chemical substances that fulfill the same specific vitamin function) are shown in Figure 1.3. PLP contains a pyridine ring with a reactive aldehyde para to the ring nitrogen, at position C4'. A 17 methyl group is ortho to the ring nitrogen at position C? and a hydroxyl is meta, at 03-. Also meta is a carbon bound to a phosphate group, at C5'. These structural features are believed to be important for the role of PLP as an inhibitor of certain enzymes, as will be discussed later. PLP cannot be biosynthesized by mammals, and must be obtained from the diet. Major sources include certain vegetables, beans, fish, seeds and nuts. Vitamin B6 is typically found in the diet as pyridoxal, pyridoxamine and pyridoxine [55]. It is passively absorbed by the intestine and delivered to the liver, where it is converted to it's active form, PLP. For most cells, PLP cannot freely cross the cell membrane and must be hydrolyzed to pyridoxal by alkaline phosphatases. Pyridoxal can pass through the cell membrane or be transported into the cell, whereupon pyridoxal kinases phosphorylate pyridoxal back to PLP [56]. Vitamin B6 deficiency is rare in developed countries, but this condition could pose a risk for cardiovascular disease since vitamins B6 and B12 are cofactors of enzymes used in the breakdown of homocysteine. Some preliminary evidence exists showing that hyperhomo-cysteinemia results in negative outcomes for patients with high plasma levels of homocysteine with regards to developing cardiovascular disease and risk of death for those with coronary heart disease [57, 58]. It is widely recommended that the diet include sources of vitamins B6, although it is currently not clear i f there are long-term effects (on plasma vitamin B6 concentrations) of daily pyridoxine supplementation, as used for treating diseases such as cystathioninuria and homocystinuria, seizures, and carpal tunnel syndrome [59]. As mentioned, once in the cell, PLP acts as a cofactor in many enzymatic reactions. Most familiar is its role in transaminases, common in amino acid metabolism. The binding of PLP to an enzyme is through its reactive aldehyde residue to an e-amino group of an enzymatic lysine residue, forming an "internal" aldimine. 18 Figure 1.3 Structures ofBf, vitamers. Pyridoxal Phosphate Pyridoxal Pyridoxine 4-Pyridoxic Acid To illustrate the principle of PLP-mediated catalysis, a transamination reaction will be described, as illustrated in Figure 1.4. Incoming amine-containing substrates react with the labile double bond, forming an "external" aldimine linkage to PLP. A proton on the substrate is removed by a reactive group on the enzyme (often the e-amino group to which PLP was first linked with), leading to a tautomerization of PLP. A reactive molecule then interacts with the bound amine substrate, and this leads to hydrolysis of the substrate, leaving pyridoxamine phosphate. The second substrate, lacking an amino group then reacts with the reactive amine, and the reverse reactions mentioned above occur, regenerating PLP and producing a new product containing an amine and one lacking an amine, hence 'transamination'. In general PLP acts as an "electron sink", storing electrons from cleaved substrate bonds, and dispensing them to allow the formation of new linkages with incoming protons or secondary substrates [60]. 1.4.2 Experimental Use of PLP While PLP is naturally used as a cofactor in enzymes, it can also be used experimentally for numerous purposes. These range from using it as a measure of vitamin B6 absorption and usage [55, 61, 62], determining reactive lysine residues and PLP/substrate/ allosteric binding sites of various enzymes [63-71], measuring transport of metabolites across membranes [72-75], and for determining enzymatic activity [76, 77]. 20 Figure 1.4 PLP-dependent Transamination Mechanism. Enzyme ( C H 2 ) 4 H - C O O H N H , OPC?-Enzyme C O O H O P Q ' Enzyme ( C H , ) 4 N H , - C O O H ^ N : T , H H ^-y H O XT opor H,C N r E n z y m e - P L P S c h i f f b a s e a n d a m i n o a c i d G e m i n a l d i a m i n e i n t e r m e d i a t e A m i n o a c i d - P L P S c h i f f b a s e ( a l d i m i n e ) C a r b i n o l a m i n e P M P - E n z y m e A l p h a - k e t o a c i d Adapted from [78]. 21 1.4.2.1 Measuring PLP in the Body Although PLP is the most important of the vitamin B6 vitamers, and required for survival, only minute amounts are present in blood plasma, or free in the cytoplasm of cells. This presents a challenge for diagnostic determination of PLP concentration, as would be required for estimating the clinical status of an individual. Typically, all measurements made thus far have been with blood plasma, due both to ease of procurement, as well as greater PLP abundance in plasma than as a free molecule in the cell. Several methods have been developed for estimating PLP concentration in plasma. These include microbiological, enzymatic and chemical methods, as well as HPLC-based approaches [61]. The most common assays use either activated apo-tyrosine decarboxylase, which requires PLP to convert [l-1 4C]tyrosine to l4CC»2, which is then counted for radioactivity, or a 3 H radioenzymatic assay (REA) which also uses PLP-dependent tyrosine decarboxylase and HPLC [79]. Unfortunately, these methods typically are complex, non-homogeneous and require radioactive reagents and specialized equipment, which often makes them inappropriate for clinical diagnosis, since the techniques and equipment are expensive and not widely available. However, some new, non-radioactive approaches are emerging which are both faster, less complicated, comparatively accurate and less toxic than those currently employed. One such approach uses the apo-form of PLP-dependent, recombinant homocysteine a,y-lyase cloned from Trichomonas vaginalis [62]. This enzymatic method is simple, rapid, sensitive, specific and requires only small volumes of plasma. The range of techniques used for the assay of PLP have yielded very variable results as illustrated by a comparison of reports of plasma PLP levels [61]. Early work, prior to 1990, showed similar plasma PLP concentrations, but results since then have generally 22 differed from the earlier data. Whether this is due to improved sensitivity of the techniques, or more specific procedures is unknown. Clearly though, additional work must be done in order to develop a simple, rapid and accurate method for determining plasma PLP concentration, since this would provide a very useful diagnostic tool, both in the laboratory and in a clinical setting. 1.4.2.2 PLP and Reactive Lysine Residues Because PLP interacts with enzymes through a Schiff base linkage to e-amino groups of lysine residues, this feature can be exploited experimentally to determine which lysine residues account for enzymatic functionality, either through use of PLP as a natural cofactor, or through its reaction with enzymes that depend upon catalytic lysines even i f they function without a PLP cofactor. The key step in this procedure is the reduction of the aldimine linking PLP and the enzyme lysine using the common reducing agent, sodium borohydride (BH4). The now covalently linked PLP-lysine can then be determined by sequencing the labeled peptide after protein digestion, using either peptide sequencing with traditional N - or C-terminal solution approaches, or more recently using mass spectrometry. PLP binding is also measured using spectrophotometric and spectrofluorometric methods, since borohydride reduction shifts the absorbance and emission maximum of PLP, giving a diagnostic signal. Numerous reports of PLP used in these ways are found in the literature. For example, the important PLP-binding lysine residues and novel PLP-binding motifs in two PLP-dependent aminomutases from two different bacterial species were determined using PLP reduction with BH4, followed by HPLC and mass spectrometry [63, 67]. The interaction of PLP with the heme-containing protein cystathionine (3-synthase showed how PLP-dependent catalysis 23 proceeded by localizing the lysine residue to which PLP binds [65]. PLP has also been used to determine the reactive lysine residues in enzymes lacking a natural PLP cofactor, by adding it exogenously. An essential lysine residue in the substrate-binding site of retinal oxidase was determined using PLP [68], as was a lysine believed to localize near the N A D P binding site of glutathione reductase [64]. 1.4.2.3 PLP and Metabolite Transport Measurements PLP has been successfully used to determine important residues in metabolite transporters, as well as to measure the kinetics of such transport. PLP seems to be most effective as an inhibitor of transporters of di- and tri-carboxylic acids such as oxaloacetate [75], a-ketoglutarate [73] and citrate [72, 74]. Implications from such studies is that PLP is a very effective inhibitor, showing IC50 values in the range of 200-500 uM, and complete inhibition at concentrations above about 5 mM. Why PLP interacts so specifically with these particular transporters is not completely understood, but it has been suggested that that since citrate is highly negatively charged, its binding site would be expected to have positively charged amino acids (such as lysines), allowing charge stabilization and ionic interactions. Since PLP also contains a negatively charged phosphate group and an amine-reactive aldehyde, it should also interact favourably with the citrate binding sites on these enzymes [69, 70]. In most of these studies PLP inhibition was concentration dependent, and PLP inhibition could be successfully blocked using specific substrates. 24 1.4.2.4 PLP and Enzymatic Activity Determination One particularly interesting use of PLP was to show that a gain-of-function mutant of Escherichia, coli was due to an uncompensated lysine residue. This lysine showed substrate-protectable inhibition by PLP of the sugar phosphate transporter UhpT. This also showed that the ability of the mutant to utilize a new phosphoenolpyruvate (PEP) substrate was due to the interaction of the lysine residue with the carboxyl group of PEP [66]. Similarly, PLP has been used to determine the activity of both succinic semialdehyde dehydrogenase [76], and succinic semialdehyde reductase [77]. 25 1.5 THESIS INVESTIGATION Although it is firmly established that citrate acts as an allosteric activator of both ACC-1 and ACC-2, its exact binding location has remained a mystery. Determination of the citrate binding site would be highly beneficial in understanding how citrate can induce polymerization of ACC-1 , and result in activation of both isoforms. The current progress in structural studies of A C C combined with citrate binding site information would allow a much better understanding of A C C function and regulation. As mentioned, PLP can act as an inhibitor of di- and tri-carboxylate transporters, competitively interacting at the active site, although why PLP acted in this fashion was not explained [72-75]. As described in other reports, PLP was probably effective because it can provide a combination of a highly negatively charged phosphate group and a reactive aldehyde. The phosphate group would mimic the negative charges of the carboxylate substrates, and the aldehyde would enable an interaction with an active site lysine residue, providing additional stability. Studies of the allosteric citrate binding site of rabbit muscle phosphofructokinase (PFK) utilized PLP as a key component [69-71]. These suggested that the rationalization for using PLP was that since citrate contains three negatively charged moieties, the citrate binding site is likely to have multiple positive charges. If one of these was due to a lysine residue, it could potentially interact with PLP. This was shown to be the case, as evidenced by citrate protection of PLP inhibition. Using sodium borohydride, [ H]pyridoxal phosphate was reduced, and the covalently labelled protein digested with trypsin. The peptides were separated by HPLC and labelled peptides isolated. One peptide contained 70% of the radioactivity recovered, and was sequenced using Edman degradation. 26 Since A C C is also allosterically activated by citrate, it was hypothesized that perhaps PLP would show a similar response with A C C as it did with PFK. The major motivation for considering examination of PLP effects on A C C was that it might provide a means to localize the elusive citrate binding site of A C C . Furthermore, positive results with PLP would open up the possibility of structure-function studies to determine the important structural aspects of PLP that enabled A C C inhibition. This thesis addresses two of these concepts. In Chapter 3, characterization of the effects of PLP on A C C is given. In Chapter 4 the effects of PLP analogs are examined and how the different moieties of PLP account for its specificity to A C C and other enzymes. Finally, Chapter 5 concludes the thesis, and also provides some insight into future work, including a brief discussion of the use of PLP in determining the citrate binding site of A C C . 27 CHAPTER TWO EXPERIMENTAL PROCEDURES 2.1 MATERIALS Male Wistar rats were obtained from the University of British Columbia animal care facility at weights ranging from 120-250 g. A l l animal procedures were carried out within the guidelines of the Canadian Council of Animal Care and with the approval of the U B C Committee on Animal Care. Rats were maintained on a 12 hour light/12 hour dark cycle at a controlled temperature with free access to water and laboratory rat chow. Solvents and laboratory chemicals were mostly obtained from Fisher Scientific or B D H Chemicals Canada Ltd. HEPES, glycine, EDTA, EGTA, PMSF, ATP, PLP (& analogs), glutathione, Pepstatin A and BSA were from Sigma-Aldrich. ICN supplied fatty-acid free BSA, and Roche supplied coenzyme A. Tris and MOPS were from USB (Amersham). Peptides International supplied the protease inhibitor leupeptin. Streptavidin-HRP, [ l 4 C ] - K H C 0 3 , pre-stained SDS-PAGE high molecular weight Rainbow Markers and ACS scintillation fluid were from Amersham Pharmacia Biotech. PVDF membrane and Ultrafree-4 and Ultrafree-15 centrifugal filters (Biomax-30) were from Millipore. SDS-PAGE reagents and Affigel-10 were from BioRad. Hen egg-white avidin was from Canadian Inovatech, and anti-phosphorylated A C C (S79) antibody was supplied by Upstate Biotech. [ H]-NaBH 4 was from Perkin Elmer Life Sciences and SoftLink™ Soft Release Avidin Resin was from Promega. 28 2.2 METHODS 2.2.1 Tissue Preparation 2.2.1.1 Isolation Rats were killed by C O 2 asphyxiation usually between 9:00 and 10:00 A M , and liver was immediately removed onto ice, with skeletal muscle and heart immediately frozen in liquid nitrogen. Adipose tissue was not chilled or frozen immediately, but instead was incubated in pre-gassed (0 2:C0 2;95:5) Krebs-Henseleit buffer (25 m M NaHC0 3 , 1.2 m M K H 2 P O 4 , 120 m M NaCl, 5 m M KC1, 1.2 mM MgS0 4 , 1.3 m M CaCl 2 and 11 m M glucose) pre-warmed to 37°C. This buffer was made up freshly each day, gassed while at 4 °C then warmed to 37 °C. 2.2.1.2 Adipose Tissue Incubation Epididymal and perirenal fat pads were dissected into smaller pieces (50-100 mg) and rinsed once with warmed, pre-gassed Krebs-Henseleit buffer, then incubated with gentle shaking in fresh buffer for approximately 30 minutes to counteract the effects of any residual hormones. Following incubation, the pieces were blotted to remove excess buffer and then frozen in liquid nitrogen. 2.2.1.3 Homogenization Liver was homogenized in 3-7 volumes of homogenization buffer using a fitted Teflon-glass Potter-Elvehjem homogenizer kept on ice. The buffer contained 20 m M MOPS (pH 7.2), 250 m M sucrose, 2mM EDTA, 2 m M EGTA, 2.5 m M benzamidine HC1, 3 u M Pepstatin A , 5 u M leupeptin, 2.5 m M glutathione and 0.5 m M PMSF. Following freezing in 29 liquid nitrogen, skeletal muscle, heart and adipose tissue were powdered and to the powder was added 4 volumes (muscle), 9 volumes (heart) and 5 volumes (adipose) of homogenization buffer. These three tissues were homogenized using a Polytron homogenizer and as with liver, all homogenization steps were carried out on ice. 2.2.2 Purification of Acetyl-CoA Carboxylase Tissues were isolated and homogenized as described above and the homogenate was then centrifuged for 90 seconds at 1000 rpm to pellet cell debris. The supernatant was filtered through glass wool to help remove any residual floating fat and centrifuged for 20-30 minutes at 12000 rpm (11000 x g) to pellet larger organelles, microsomes and other cellular components. Following this, the supernatant was again filtered through glass wool and subjected to ultracentrifugation at 55000 rpm (215000 x g) for 60-90 minutes. A l l centrifugation steps except the first 90 second spin (room temperature) were performed at 4°C. The ultracentrifuged supernatant was then treated with ammonium sulphate to precipitate A C C . 2.2.2.1 Ammonium Sulphate Precipitation Powdered ammonium sulphate was slowly added to the high speed supernatant on ice to give a final saturation of 40%. The mixture was stirred on ice for a minimum of one hour, then the precipitate recovered by centrifugation at 14000 rpm (15000 x g) for 20-30 minutes. For samples used primarily in activity assays, the pellet was resuspended in homogenization buffer lacking protease inhibitors. For samples to be subsequently applied to an avidin affinity column, the pellet was resuspended in HB buffer (50 m M Tris, 500 m M KC1, 2 m M 30 EDTA, 2 mM EGTA, 0.02% (v/v) sodium azide). Samples of A C C recovered by ammonium sulphate precipitation were typically stored at either -20 °C (short-term) or -80 °C (long-term) until use or further purification. If further purification was required, samples were thawed and centrifuged at 13200 rpm (16000 x g) for 3 minutes to clarify the enzyme sample. Any enzyme dilutions required for further purification were made at this point. 2.2.2.2 Avidin Affinity Chromatography Due to the high affinity of biotin to tetrameric avidin (~ 10"15 M), reversal of this binding can only be achieved using denaturing conditions such as boiling in SDS buffer containing (3-mercaptoethanol. These conditions destroy A C C activity and thus tetrameric avidin cannot be used to recover active enzyme. However, using a monomerized form of avidin, and saturating the high affinity sites with free biotin, the affinity of the avidin-biotin interaction can be lowered and A C C desorbed under mild conditions - simply by adding free biotin. This allows the "lower affinity" biotin binding sites on the avidin to be used in affinity purification of biotin-containing proteins, such as A C C . SoftLink™ Soft Release Avidin Resin from Promega was used according to the manufacturer's instructions. Briefly this involved washing the column with HB buffer followed by binding of free biotin to the reactive sites on the resin using 5 m M biotin in HB buffer. The column was then "stripped" of biotin bound to the lower affinity sites using 10%> acetic acid, followed by washing in HB buffer. Following this, a diluted and briefly centrifuged sample of the resuspended ammonium sulphate pellet was added to the column and either incubated on the column, or mixed with the resin under shaking for a minimum of one hour, but usually for two to three hours, to ensure complete binding of A C C . Unbound 31 proteins were then washed from the column with HB buffer until the absorbance of the washing buffer, monitored at 280 nm was constant and very low (typically less than 0.01 A28o)- A C C was then eluted using HB buffer containing 5 m M biotin and the eluate was concentrated using the "aquacide" method. This involved pouring the sample of A C C into a dialysis sac and incubating the sac in a container of powdered polyethylene glycol to extract moisture from the dialysis sac. Alternatively, A C C samples were concentrated with Ultrafree (4 or 15) centrifugal filter units (Millipore). The flow-through "Void" fraction was also retained and concentrated using the same method as the A C C eluate. The avidin affinity column was regenerated with 10% acetic acid and washed with PBS containing 0.02% (w/v) sodium azide for storage. 2.2.3 A C C Activity Assays The [ l 4 C]-HC0 3 fixation method [30, 80] was used to determine A C C activity. The first step involved pre-incubating the enzyme for 30 minutes at 37°C, to allow A C C activation/polymerization. The pre-incubation buffer consisted of homogenization buffer containing 10 or 20 mM sodium citrate (allosterically activates ACC) , 2 mg/mL B S A and the A C C sample. 50 m M sodium glutamate was used instead of citrate in some instances. Following pre-incubation, a 50 (iL aliquot of the activated A C C was added to 450 u\L pre-warmed (37 °C) assay buffer (50 mM HEPES pH 7.2, 10 mM MgS04, 0.5 m M EDTA, 5 m M ATP, 7.5 m M glutathione, 2 mg/mL BSA, 150 uM acetyl-CoA and 7.5 m M [ 1 4C]-KHCO3) . Assay time varied by tissue and experiment, but all assays were terminated by the addition of 200 pX of 2M HC1. Following a brief centrifugation at 13200 rpm {16000 x g}for one minute, 600 uX of the assay mixture was transferred to a liquid scintillation vial and 32 evaporated under a stream of air, to remove unincorporated [ C]-KHCC>3. The remaining dried material contains [ l 4C]-malonyl-CoA, which is acid stable and non-volatile. Each scintillation vial was then shaken for at least one hour with 400 uX of added water, followed by addition of 4 mL of ACS liquid scintillation fluid (Amersham). Samples were counted using a Beckman LS6000IC Scintillation Counter. To determine the substrate specific activity 10 uL of the assay buffer was added to 4 mL of ACS fluid containing 200 uL of 2-phenylethylamine (a CGvtrapping compound). The specific activity of the assay buffer was determined in triplicate for each experiment. One unit of A C C activity is defined as the amount of A C C needed to convert one umole of acetyl-CoA to malonyl-CoA in one minute. 2.2.4 Inhibition Assays Inhibition of A C C by various compounds was assessed by two protocols. The first involved treating A C C samples with pre-incubation buffer lacking sodium citrate, but containing the desired inhibitor, typically for 30 minutes at either 25 or 37 °C. This was followed by addition of citrate and incubation for a further 30 minutes at 37 °C. Incubation with assay buffer then followed as described above. The second variation was the reverse of the first and involved pre-incubating first with citrate (30 minute, 37 °C), then adding the inhibitor and incubating for a further 30 minutes at 37 °C prior to assaying (as above). 2.2.5 Sodium Borohydride Reductions Reduction of potential aldimine bonds between pyridoxal 5'-phosphate and A C C using sodium borohydride was carried out using one of two methods. In the first method, using unlabelled borohydride, A C C was pre-incubated (30 minute, 25 °C) with PLP. Sodium 33 borohydride (1 to 10 mM) was then added and the incubation continued for a further 15 minutes at 25 °C. Unreacted PLP and borohydride were then removed by two wash steps in which A C C was diluted by adding 10 mL of homogenization buffer (containing or lacking 20 m M sodium citrate), and then concentrated at 4 °C using a 15 mL Biomax-50 centrifugal filter unit (Millipore). The centrifugations were performed according to the manufacturer's instructions, which involved centrifugation at 8000 x g in an SS-34 Sorval fixed angle rotor and a Sorval 5CB refrigerated centrifuge. Following washing and concentration, A C C activity assay were carried out as described previously, using 50 uL of concentrated enzyme sample. Appropriate controls lacking PLP, borohydride, and citrate were also performed. The second method of A C C reduction was carried out using [3H]-NaBH4 in order to test for A C C protein-labelling. The enzyme was reacted with tetrameric avidin beads for 45 minutes at 4 °C, followed by thorough washing. The beads were then treated with 0.5 m M PLP for 30 minutes at 25 oC, followed by addition of 1 m M BH4 for 15 minutes. The beads were again thoroughly washed and the bound proteins ready for elution by heating in Sample loading buffer, as described below. 2.2.6 SDS-PAGE Using the procedure of Laemmli [81], discontinuous sodium dodecyl sulphate polyacrylamide gel electrophoresis (SDS-PAGE) was used to separate proteins of interest. Gels run solely to observe A C C running patterns, such as in an ammonium sulphate preparation, were typically 3% stacking and 5% separating acrylamide, using piperazine diacrylamide as the crosslinker instead of bis-acrylamide, to increase gel strength. Run times were typically 1.5 hours at 175 volts and 50 milliamps. Following electrophoresis, gels were 34 fixed, stained with Coomassie blue, de-stained, stabilized with glycerol and placed between two sheets of cellulose film and dried at room temperature. To separate A C C from other proteins following treatment with PLP and [3H]-NaBH4, 4-15% gradient acrylamide mini-gels were employed (BioRad). Prior to electrophoresis, samples were heated for 10 minutes at 95 °C in SDS loading buffer (125 m M Tris (pH 6.8), 10% (w/v) SDS, 20% (w/v) sucrose, 0.02% (w/v) bromophenol blue and 5% (v/v) (3-mercaptoethanol). Molecular weight of the protein bands was estimated by co-loading pre-stained high molecular weight Rainbow markers on each gel (Amersham). Gel run times were typically 2-3 hours at 135 volts and about 20-30 milliamps. Gels were fixed, stained with Coomassie blue, de-stained, stabilized with glycerol and placed between two sheets of cellulose film and dried at room temperature. Quantitation of incorporated [3H] into proteins due to PLP reduction by [ H]-NaBH4 was determined similar to the method of Brownsey and Denton [53]. Briefly, stained bands were cut from the gel and digested in 30% H2O2 at 80 °C for 3 hours and then evaporated to dryness. The dried residue was then dissolved in 200 uX 2M HC1, 4 raL of ACS liquid scintillation fluid added, and counts determined. 2.2.7 PLP Spectrophotometry Because free PLP exhibits a significantly different absorbance maximum than the borohydride-reduced adduct, spectrophotometric analysis of PLP covering wavelengths from 200 to 600 run provides a useful analytical tool. Scans of free and reduced PLP, as well as free B H 4 and lysine were used as controls for PLP conformation. Measurements were made in acid, water and base, initially, and at neutral pH in subsequent observations. A l l measurements were made using a Perkin Elmer Lambda 35 spectrophotometer. 35 CHAPTER THREE PLP EFFECTS ON ACC 3.1 RATIONALE Tri-carboxylic acids such as citrate and isocitrate [12, 17, 29-31] and di-carboxylic acids such as glutamate [33] are well known to promote polymerization of ACC-1 and activation of both A C C isoforms. As discussed in Chapter 1, PLP has been shown to interact at the citrate binding site of several proteins, including phosphofructokinase. The major goal of the work described in this chapter was to test the hypothesis that PLP can interact at the allosteric citrate binding site of A C C . Because PLP interferes with effects of citrate on PFK [69-71], as well as inhibiting tri- and di-carboxylate transporters, it was also speculated that PLP would have inhibitory effects on A C C . Given the differences in citrate sensitivity of A C C isoforms, it was speculated that PLP may show different effects in A C C prepared from different tissues. In this chapter I describe effects of citrate and PLP on A C C isoforms and the ability of citrate and other analogs to protect A C C against potential effects of PLP. 36 3.2 RESULTS AND DISCUSSION 3.2.1 A C C Activation by Citrate Citrate treatment results in rapid activation of ACC-1 and ACC-2, and, at least in the case of ACC-1 , activation is accompanied by polymerization [30]. A C C activation is concentration-dependent, and typically exhibits maximal effect at citrate concentrations around 10-20 mM. Liver typically contains the largest amount of A C C per gram of wet weight tissue (on the order of 200-300 mU/g), followed next by adipose tissue (100-150 mU/g). Both heart and skeletal muscle contain much smaller amounts of A C C than do either liver or fat tissue (usually less than 10 mU/g). The same rank order of activities also apply when A C C activity is expressed as mU/mg protein recovered following ammonium sulphate precipitation. This is illustrated in Figure 3.1, where liver contains more units per milligram than a comparable preparation from muscle. This figure also shows that A C C from both tissues exhibits maximal activation at around 20 m M citrate (no significant further activation occurs with 30 or 40 m M citrate, results not shown). Finally, the sensitivity of A C C to citrate is much higher in extracts from liver (initial steepness of liver curves), largely due to the significant amount of ACC-1 present in this tissue. Over a larger number of experiments, K a values for citrate with liver/fat A C C are indeed lower than that of A C C from heart or skeletal muscle. 37 Figure 3.1 Activation of ACC by Citrate. 2 0 0 0 5 1 0 1 5 2 0 Citrate Concentration (mM) Citrate Concentration (mM) Preparations of A C C purified by ammonium sulphate precipitation from liver (o) and skeletal muscle (A), were incubated with citrate and assayed as described. A C C activity is expressed as mU/mg of protein (A) or as % maximum activity (B). Results for liver are expressed as the mean ± S.E.M. of three experiments, while skeletal muscle data represents a single experiment typical of 3 experiments. 38 3.2.2 A C C Detection by Electrophoresis Confirmation and quantitation of A C C in enzyme preparations is usually accomplished by activity assays. However, certain conditions, such as enzyme inactivation, or detection of enzyme in degraded preparations can be difficult using an assay of catalytic activity. Furthermore, the purification of A C C may be easily followed using other methods. Electrophoresis can overcome some of the A C C assay limitations, due mainly to staining sensitivity, and visualization of protein content of a given enzyme preparation. Typical results of electrophoresis of A C C is given in Figure 3.2. In part (A), a liver A C C ammonium sulphate precipitation sample is run alongside molecular size markers (using a 5% separating gel). In part (B), a sample of another liver A C C ammonium sulphate precipitate was purified over tetrameric avidin beads and electrophoresed (on a 4-15% gradient gel). Although both preparations show A C C 1 and ACC2 bands at approximately the same spot on the gel (relative to the markers), the other bands which stain are different. This is partly due to the differences in run time between the two, as well as differences in the type of gel used and the purification steps. It is important to note that these gels are presented to illustrate the ability to detect the two isoforms by electrophoretic separation, following different purification steps, and for different end purposes. The gel in part (A) was used for visualization purposes only, while the gel in part (B) was used to detect bound PLP, as described in 3.2.9. Western blot analysis can also be performed, but such an analysis is not useful in this thesis and are not described. Note that the apparent sizes of ACC1 and ACC2 as seen on these gels do not correspond with known values from cDNA sequencing. The reasons for this are due to technical limitations as well as other as yet unknown causes. 39 Figure 3.2 ACC Detection by Electrophoresis and Coomassie Blue Staining. A B Liver A C C prepared by ammonium sulphate precipitation was run on a 3%/5% stack/separating acrylamide gel using a PDA cross-linker, for 1.5 hours at 175 volts and 50 milliamps, alongside molecular weight standards (A). A separate liver ammonium sulphate preparation was purified using tetrameric avidin beads and eluted using Sample loading buffer prior to running alongside molecular weight standards on a 4-15% gradient acrylamide gel at 135 volts and 30 milliamps (B). Size standards are indicated as are the positions of A C C l and ACC2 bands. 40 3.2.3 Inhibitory Effect of PLP on A C C It has been shown that PLP can inhibit phosphofructokinase in a concentration-dependent manner, with an IC50 of about 200uM and complete inhibition at around 2 mM, when reduced with sodium borohydride [71]. Similar inhibitory potency was observed in studies of tricarboxylate transporters [72-75] suggesting that PLP might be examined for potential inhibitory effects on A C C over this concentration range. The results of such experiments on liver A C C are shown in Figure 3.3. In this experiment, A C C was partially purified from rat liver and then treated with PLP before citrate activation of the enzyme. Clearly, PLP has a strong, concentration-dependent inhibitory effect on liver A C C , with an IC50 value around 225 uM, and 90% inhibition of enzyme activity seen at PLP concentrations less than 1 mM. Also evident from Figure 3.3 is that 10 mM citrate was unable to reverse the effects of PLP at concentrations as low as 25 uM. While PLP shows strong inhibitory effects on liver A C C , would it be expected to show similar responses in A C C from other tissue types? Since liver contains both ACC-1 and ACC-2 in significant amounts (3:1 ratio, respectively) it might be concluded that both isoforms are being affect by PLP. However, as shown in Figure 3.1, ACC-2 (from skeletal muscle) is not as citrate sensitive as ACC-1 , and this citrate sensitivity could be reflected in ACC-2 sensitivity to PLP inhibition. Experiments were therefore carried out to test the effect of PLP treatment on various A C C preparations and results are shown in Figure 3.4. 41 Figure 3.3 PLP Inhibition of Liver A CC. [PLP] (mM) A C C partially purified by ammonium sulphate precipitation of liver was incubated with the PLP concentrations shown for 30 minutes, followed by a 30 minute incubation with 10 mM citrate to activate A C C . Samples were then assayed as described. Results are expressed relative to the maximum activity obtained in each individual experiment (no added PLP) and represent the mean ± S.E.M. for 6 separate A C C and PLP preparations. Some error bars fall within the symbols. 42 Figure 3.4 PLP Inhibition of Various ACC Preparations. Tissue preparations of skeletal muscle (A, €), heart (B, • ) , liver (C, o) and adipose (D, A) A C C purified by ammonium sulphate were treated with PLP for 30 minutes, followed by activation with citrate for 30 minutes and subsequent assay for enzyme activity (all at 37 °C). For comparative purposes, (A) to (D) are combined in (E), with the same symbols used to represent each tissue type as in (A) to (D). Results for (A) and (C) are expressed as the mean ± S.E.M. for 3 and 6 separate experiments, respectively. Results from (B) are expressed as the mean ± range of 2 separate experiments. (D) represents a single experiment. Results are represented as the percentage of the maximum obtained in each individual experiment (no added PLP). Some error bars fall within the symbols. Skeletal Muscle 0 1 2 3 4 5 PLP Concentration [mM] A C C from oxidative tissues (muscle and heart) generally shows less sensitivity to PLP than does enzyme from lipogenic tissues (liver and adipose). Comparison of the curves in Figure 3.4 shows that skeletal muscle A C C (Figure 3.4A) shows the least PLP sensitivity of all tissue types tested, with an IC50 around 630 pJvl, compared to around 225 u\M or less for liver and adipose tissue (Figure 3.4C and 3.4D). Heart tissue, which contains slightly more ACC-1 than does skeletal muscle, shows PLP sensitivity between liver and skeletal muscle, with an IC50 value near 500 p M (Figure 3.4B). Also significant is the observation that even at high concentrations of PLP (2-5 mM) skeletal muscle and heart A C C , unlike fat and liver A C C , show considerable residual activity. In fact, unlike other tissues, heart A C C is not completely inhibited even at PLP concentrations in excess of 7 mM. These experiments therefore provide the first evidence for PLP action on A C C , and illustrates at least two important points. First, A C C from all tissue types examined did respond to PLP and all isoforms are inhibited by this compound, even at relatively low concentrations. Second, the A C C isoform composition of the tissues is reflected in the overall sensitivity of the enzyme to PLP. More specifically, tissues containing a greater proportion of ACC-2 seem to show less sensitivity to PLP than those tissue expressing more ACC-1 . Whether this is due directly to the difference in citrate sensitivity between the two isoforms or whether other proteins mediate the interaction between PLP and A C C in different tissue preparations is not evident from this experiment. However, as shown in Figure 3.5, the response of liver A C C purified by ammonium sulphate and more highly purified by avidin affinity chromatography show little difference in response to PLP. This suggests that in liver at least, direct binding of PLP to A C C is likely, and that the response of A C C to PLP is not mediated by another interacting protein. 46 Figure 3.5 PLP Inhibits Highly Purified A CC. 0 1 2 3 4 5 PLP Concentration [mM] Preparations of liver A C C purified by ammonium sulphate (A) or avidin-affinity chromatography (o) were treated with PLP for 30 minutes, followed by activation with 10 m M citrate for 30 minutes and subsequent assay for enzyme activity (all at 37 °C). Results for ammonium sulphate purified A C C are expressed as the mean ± S.E.M. for 6 separate experiments, while those for avidin purified A C C are for a single experiment. Results are represented as the percentage of the maximum obtained in each individual experiment (no added PLP). Some error bars fall within the symbols. 47 3.2.4 Time Dependence of PLP Inhibition Regardless of whether PLP is acting at the citrate binding site, or another distinct site, certain aspects of its inhibition on A C C can be explored. One of these is the time dependence of inhibition; specifically, how fast does PLP inhibit ACC? This question was addressed by incubating A C C with PLP, followed by a citrate incubation and assay. The results of such experiments are given in Figures 3.6 , 3.7 and 3.8. Figure 3.6 describes the effects of A C C incubated 0 to 25 minutes with 500 pJVI PLP, prior to citrate addition and subsequent assay. A C C activity loss with PLP incubation appeared to occur in two distinct phases; an initial rapid loss of activity (< 2 minutes of PLP incubation), followed by a slower, continuing decline in A C C activity (2 to 25 minutes of PLP incubation). Interestingly, liver A C C preparations showed the most extensive initial, very rapid decrease in A C C activity (~ 70% loss), followed by a linear, gradual decrease in activity as PLP incubation time increased. In contrast, heart and skeletal muscle A C C seem to show a less extensive initial rapid loss of activity (25-35% loss); followed by a further gradual decrease in activity with increasing incubation time, in a roughly linear fashion. Conclusions from this experiment must be drawn cautiously because only one trial of each tissue type was analyzed. Also, since these results occur for A C C incubated with PLP, then followed by citrate incubation, the effective incubation time with PLP is actually the time given, plus the 30 minutes of citrate incubation. Additionally, whether the liver, heart and skeletal muscle A C C responses truly differ and whether this difference can be attributed to the isoform composition differences between the two tissues is not certain from this data. However, given the observation that muscle A C C in previous experiments appears to give significantly different results than A C C in liver, heart or adipose tissue, the isoform 48 differences cannot be ignored as a possible explanation of the differences in this experiment. Additional experiments were therefore carried out to focus on shorter times of PLP treatment. Figure 3.6 Time Dependence of PLP Inhibition of ACC - Long-Term. 0 5 10 15 20 25 PLP Incubation Time (min) Preparations of A C C partially purified by ammonium sulphate precipitation of liver (o), muscle (•) and heart (A), were incubated with 500 u M PLP for the indicated time (1 to 25 minutes), followed by a further incubation of 30 minutes with 20 m M citrate and subsequent assay. Results are all of single experiments and are represented as the percentage of the maximum obtained in each individual experiment (no added PLP). 49 In Figure 3.7, the short-term PLP incubation effects on A C C inhibition were examined for liver, skeletal muscle and heart A C C . In all of the short incubations, the effects of PLP are seen within 30 to 60 seconds, with very little subsequent inhibition of A C C seen from 1 to 4 minutes, indicating that PLP's effect on A C C is indeed very rapid. As well, the lack of variation in activity after the 30 second time point in most samples indicates that following the initial inhibition, PLP does not cause further inhibition, at least in the short term. The greatest difference between tissues is evident between liver (Figure 3.7A) and skeletal muscle (Figure 3.7C), with liver A C C activity remaining largely constant after the initial inhibition. Comparison of the results at 500 uM show that for liver, the results correlate very well with those shown in Figure 3.6 (also 500 uM) (both ~ 30% inhibition). Short-term skeletal muscle and heart (Figure 3.7B) inhibition levels do not correlate as well to long-term data as compared to liver, but are not considerably different. As mentioned previously, since these results represent only single experiments, some variability is to be expected. 50 Figure 3.7 Time Dependence of PLP Inhibition of Liver, Heart and Skeletal Muscle ACC. A C C partially purified by ammonium sulphate precipitation of liver (A), heart (B) and skeletal muscle (C) was incubated with PLP for 30 seconds to 4 minutes, followed by 30 minutes of 20 m M citrate incubation and subsequent assay. PLP concentrations tested were 100 pJVI (€), 500 \iM (A), 1 m M (•) and 5 m M (•). Results are all of single experiments and are represented as the percentage of the maximum obtained in each individual experiment (no added PLP). 51 0.0 0.5 1.0 1.5 2.0 2.5 3.0 Time (min) Figure 3.8 Effect of Citrate Pre-incubation on Time Dependence of PLP Inhibition of Liver and Skeletal Muscle A CC. A C C partially purified by ammonium sulphate precipitation of liver (A) and skeletal muscle (B) was incubated with 20 m M citrate for 30 minutes, followed by incubation with PLP for 1 to 25 minutes, and subsequent assay (open symbols). Results shown by filled symbols represent the same curves as are shown in Figure 3.6, and have A C C incubated first with 0.5 m M PLP, followed by 30 minute incubation with 20 m M citrate. PLP concentrations tested were 500 uM. Results are all of single experiments and are represented as the percentage of the maximum obtained in each individual experiment (no added PLP). 53 In Figures 3.8 the effect of citrate pre-incubation on PLP inhibition time was explored in liver (Figure 3.8A) and muscle (Figure 3.8B), respectively. These results are somewhat surprising, since the absolute degree of protection against PLP inhibition offered by citrate is very similar between the two tissues, somewhat contrary to the results shown in Figure 3.9 (see 3.2.5). However, citrate does offer better protection against PLP in liver compared to muscle, which supports the results in Figure 3.9. It is also interesting that the rate of PLP inhibition of A C C for both liver and muscle is slowed significantly with citrate pre-incubation, with maximal inhibition seen at 5 minutes, compared to 30 seconds without citrate pre-incubation. This suggests that citrate somehow prevents PLP inhibition of A C C . Whether this is simply due to citrate physically blocking the binding site, or due to a conformational change induced by citrate binding is unclear. An important distinction with these experiments and those shown in Figure 3.7 is that in Figure 3.8, A C C was treated for 30 minutes with citrate, and then incubated with PLP. This means that PLP incubation time with A C C is actually the time shown (i.e. there is no additional 30 minutes due to subsequent citrate incubation). Another interesting result from Figures 3.7 and 3.8 confirms the concept that ACC-2 is less sensitive to PLP than is ACC-1 . More precisely, at lower PLP concentrations, muscle A C C does not show the same level of inhibition seen in liver and heart. For example, in Figure 3.7A it can be seen that with 1 m M PLP, liver is inhibited to about 15% residual activity and heart to about 20% (Figure 3.7B), while muscle shows less inhibition, at around 35% residual A C C activity (Figure 3.7C). This result agrees with that seen in Figure 3.4. 54 3.2.5 Citrate Protection Against PLP Inhibition of A C C The experiments described so far provide evidence that PLP does inhibit A C C , but do not conclusively show that PLP interacts at the citrate binding site. Although there is evidence that PLP does interact at the allosteric citrate binding site of PFK [69-71], and at the active site of tricarboxylate transporters [72-74], there is no guarantee that PLP acts at the allosteric citrate binding site on A C C . Definitive proof of this is not easily obtained. In the longer-term it will be important to attempt studies such as those used with enzymes containing known citrate binding sites [69-74]. In those cases, PLP has been covalently linked to the enzyme, the enzyme digested and the PLP-labelled peptide sequenced so that the PLP and citrate binding sites could be compared. Since in A C C the citrate binding site is unknown, a comparison between the PLP and citrate binding sites cannot be performed. Instead, other experimental data must be obtained, such that convincing evidence exists to support the idea that PLP is in fact binding at the citrate binding site on A C C . One common approach used to address this problem is to use protection experiments, in which one molecule binding at the site of interest can protect the enzyme from the inhibitory or stimulatory effects of another molecule potentially binding at the same site. In this event, citrate might offer some protection to A C C against inhibition by PLP i f the two molecules indeed bind at the same site on A C C . Using this approach, A C C preparations were pre-treated with citrate, prior to incubation with PLP. If citrate binding to A C C could prevent PLP binding, A C C activity should remain higher than in assays where A C C was treated with PLP first. The results of such experiments are shown in Figure 3.9. Citrate offers significant protection to A C C against PLP inhibition, especially at low PLP concentrations for liver A C C (Figure 3.9A) and to a lesser degree for heart A C C (Figure 55 3.9B). Conversely, in muscle tissue (Figure 3.9C), citrate does not seem to offer any substantial protective effect at any of the concentrations tested. These data provide further support for the idea that PLP and citrate might compete at the same binding site and also further support the fact that A C C from different tissues (likely resultant from the different isoform expression) responds differently to both citrate and PLP. A comparison of IC50 values, as well as the concentrations of PLP required to inhibit A C C to a level of 10 and 90% under the conditions used in the experiment shown in Figure 3.9 is given in Table 3.1. Table 3.1 Effect of ACC Pre-Incubation With Citrate on PLP Inhibition. Data for PLP inhibition of A C C is represented as IC50 and the concentration of PLP required to show 10 and 90 % inhibition of A C C for the experiments performed as described in Figure 3.9. A l l results are given as PLP concentrations in millimolar (mM) amounts. 'PLP First' refers to treatment of A C C with PLP prior to citrate activation, while 'Citrate First' implies A C C pre-incubation with citrate before PLP incubation. Tissue 10% Inhibition IC 5 0 for PLP 90% Inhibition PLP First Citrate First PLP First Citrate First PLP First Citrate First Liver 0.04 0.03 0.22 0.48 0.90 2.0 Heart 0.07 0.22 0.50 1.08 4.90 7.20 Sk. Muscle 0.10 0.08 0.63 0.77 3.60 4.15 56 Figure 3.9 Effect of Citrate Pre-incubation on PLP Inhibition of ACC. A C C was prepared by ammonium sulphate precipitation for liver (A, squares), heart (B, triangles) and skeletal muscle (C, circles). In one set of experiments (open symbols) A C C was treatment first with citrate (10 mM for 30 minutes at 37 °C), followed by a further incubation with PLP (30 minutes at 37 °C at concentrations show) and subsequent enzyme assay. In another set of experiments (filled symbols) A C C was treated first with PLP, then with citrate under the same conditions as the other experiment set. Results for (A) and (C) for filled symbols are expressed as the mean ± S.E.M. for 6 and 3 separate experiments, respectively, while filled-symbol results in (B) are expressed as the mean ± the range for two separate experiments. Open circles in (C) are expressed as mean ± S.E.M. for 3 separate experiments, while those in (A) are the mean ± range for two experiments. Open symbols in (B) are for a single experiment. Some error bars fall within the symbols. Results are represented as the percentage of the maximum obtained in each individual experiment (no added PLP). 57 Heart 0 2 4 6 [PLP] (mM) 58 Several interesting results are evident from the data in Table 3.1. First, as expected, liver A C C shows the strongest affinity for PLP, with or without prior citrate treatment. In fact, the concentration of PLP required to inhibit A C C at the levels indicated ranges from 2.5 to 7 fold lower for liver A C C compared to heart and muscle A C C . Secondly, liver A C C also seems to show the best protection against PLP by citrate pre-incubation, especially at the 50 and 90 % inhibition levels. Third, citrate pre-treatment is least effective in skeletal muscle A C C (of the three tissues examined), with the best citrate protection on muscle being a 22 % increase in PLP concentration at IC5o. Importantly, these experiments cannot show definitively that PLP and citrate are binding at the same site. Although citrate does protect A C C from PLP inhibition somewhat at lower concentrations in liver and heart, it does not totally block PLP inhibition at higher concentrations in these tissues, or to any degree in muscle. There are several possible reasons for this. The first could be that PLP can bind much more strongly to the enzyme than can citrate, possibly through it's highly reactive aldehyde, forming a Schiff base. This would imply that even i f citrate does bind to A C C , PLP's favourable binding characteristics would cause it to "knock out" citrate and take its place in the binding site. A second possibility is that PLP is actually binding to a site distinct from the citrate binding site, and this somehow causes inactivation of the enzyme. In this scenario, PLP would likely cause a conformational change in A C C such that citrate could not correctly bind and induce enzyme activation, or which prevents citrate activation of the enzyme by some other means, such as affecting substrate binding, or the flexible biotin arm. Previous evidence with other enzymes and proteins would suggest that PLP is binding to the same site as citrate on A C C , but the second 59 concept cannot be ruled out at this point. Further work which will be essential for determining the identity of the PLP/citfate binding site is discussed later. Additionally, it would be tempting to assume that citrate induced polymerization of ACC-1 would account for the loss of rapid PLP effect and protection against PLP on citrate pre-treated liver A C C (and to a lesser degree heart ACC) , by somehow preventing PLP binding. However, given the observation that ACC-2 does not appear to form polymers [43], and citrate treatment also slows down PLP inhibition of A C C in muscle (Figure 3.8B), albeit less significantly than in liver, it seems that this hypothesis is unlikely and polymerization state of A C C does not significantly affect PLP inhibition. Further work needs to be done to clarify this issue. An additional experiment was performed, with the goal of establishing the minimum concentration of citrate required to regain maximal A C C activity for a given concentration of PLP. More specifically, following a PLP incubation, various concentrations of citrate were added to the ACC/PLP mixture for 30 minutes and subsequently assayed. The results of such experiments with liver A C C are shown in Figure 3.10. The information in Figure 3.10 is interesting mainly because it shows that as the PLP concentration increases, citrate is less able to overcome the inhibitory effects. For example, comparison of 50 and 500 u\M PLP treatment shows that even at maximal citrate (20 mM), only approximately 12% A C C activity is recovered (at 500 u\M PLP), while inhibition caused by 50 LLM PLP can be recovered approximately 60%> by maximal citrate treatment. This has at least two possible implications. First, although strong inhibition of A C C by PLP can be accomplished, and significant loss of enzyme activity can occur even at low PLP 60 concentrations, there is likely an equilibrium taking place between PLP bound to A C C and free in solution. This view is supported by previous work which showed that PLP binding to succinic semialdehyde dehydrogenase was reversible and reached equilibrium [76]. Since an equilibrium state is probably reached, it is logical that at low PLP concentrations the amount of A C C with PLP bound is going to be smaller than at higher PLP concentrations. Coinciding with this is the fact that as the citrate concentration increases, citrate is better able to compete for binding sites on A C C , regardless of PLP's greater binding efficiency; simply put, there is just too much citrate present for PLP to compete for all of the binding sites on A C C . However, as the PLP concentration increases, it can in a sense 'saturate' the binding sites, and prevent citrate from binding. Nonetheless, even at lower PLP concentrations (50 pM), citrate treatment cannot recover full enzyme activity, indicating that PLP does show preferential binding to A C C compared to citrate. Based on these results, it might be instructive to examine effects of even higher concentrations of citrate. Although A C C activation, per se, is not further increased above 20 m M citrate, it is possible that the ability to overcome PLP inhibition might show continued dose-dependency above 20 m M citrate. Such experiments are ongoing. 61 Figure 3.10 Citrate Activation Curves at Constant PLP Concentrations. [Citrate] (mM) A C C prepared from rat liver by ammonium sulphate precipitation was incubated with PLP at 0 m M (€) (control), 0.050 mM (•), 0.10 m M (o), 0.25 m M (•) and 0.50 m M (A) for 30 minutes each, followed by a 30 minute incubation with varying concentrations of citrate as shown, with a subsequent assay for A C C activity (all at 37 °C). Results are represented as the percentage of the maximum obtained with no added PLP and 20 m M citrate. Results are typical for 2 experiments. 62 Additional information can be taken from the data shown in Figure 3.10. Since the curves shown can be represented in terms of reaction rate against PLP concentration, a double reciprocal plot (Lineweaver-Burk) can allow a determination of the inhibition type seen using PLP. Further manipulation of this plot, by taking the slopes of the lines and plotting them against the concentration of PLP at which they occur, can give an estimate of the apparent Kj for PLP on A C C . Results of such manipulations are shown in Figure 3.11. As shown in Figure 3.11 A, curves of inverse reaction rate against the inverse citrate concentration, for 5 different concentrations of PLP suggest that the type of inhibition exhibited by PLP on A C C is most likely mixed. Both V m a x and K m values change with variations in the concentration of PLP, and the curves intersect to the left of the vertical axis, consistent with mixed inhibition. Figure 3.1 IB shows the result of plotting the slopes from the curves in Figure 3.11A against the PLP concentrations at which these slopes occur. The curve generated is linear and the intercept of this curve with the x-axis is an estimate of the -K; value for PLP on A C C . This value is estimated to be about 30 (iM. Care must be taken with these plots, since they are the result of only two experiments, and more importantly, PLP and citrate are not substrates for A C C . Kinetics of allosteric activators are considerably more complex than with substrates, and defining kinetic effects of an inhibitor of an allosteric activator is equally challenging. Figure 3.11 is presented only as introductory example of further kinetic analysis which will be required in order to define more precisely the interaction of PLP with A C C . 63 Figure 3.11 Double Reciprocal Plot and Ki Determination. Data from the curves in Figure 3.10 was re-plotted as the reciprocal of the rate (min-nmol") against the inverse citrate concentration (mM"')(Lineweaver-Burk) in (A) and as the slopes of the curves in (A) against the concentration of PLP at which these slopes occur (B). The x-intercept in (B) represents - K ; . The curves in (A) are generated from an average of two experiments. 64 3.2.6 Effect of PLP on A C C Activated by Glutamate In view of the evidence that glutamate also causes allosteric activation of A C C coupled with changes in enzyme polymerization [33], the effects of PLP on A C C activated by glutamate were examined. The results of these experiments are shown in Figure 3.12. Figures 3.12A and 3.12B show that PLP inhibits glutamate activation of liver and muscle A C C in a similar manner to the citrate activation. Also evident from Figures 3.12C and 3.12D is that pre-incubation with glutamate affords A C C approximately the same level of protection against PLP inhibition as does citrate pre-incubation of the enzyme. Comparison of the IC50 values of PLP against A C C alone and A C C pre-incubated with 10 mM citrate or 50 mM glutamate are given in Table 3.2 A general trend in this data is that 50 m M glutamate yields a higher IC50 for PLP compared to 10 m M citrate. As previously described, glutamate shares a structure similar to citrate, with two carboxyl groups, and since there is no additive activating effect with citrate and glutamate, there is a high likelihood that both citrate and glutamate bind at the same site [33]. Figure 3.12 Effect of PLP on ACC Activated by Glutamate. A C C purified by ammonium sulphate precipitation from liver (circles) and skeletal muscle (triangles) was incubated with PLP for 30 minutes followed by a 30 minute incubation with 50 m M glutamate or 10 m M citrate, with subsequent assay for A C C activity (filled symbols). Alternatively, the PLP and glutamate/citrate incubations were reversed (open symbols). (A) represents liver A C C with glutamate activation, (B) is skeletal muscle with glutamate activation, (C) is liver and skeletal muscle activated with citrate, and (D) is liver and muscle activated with glutamate. Results in (A and B) are for single experiments each, and those in (C) are as described for Figure 3.9A and 3.9C. (C) and D are presented for comparative purposes between citrate (C) and glutamate (D) activation of liver and skeletal muscle A C C . Results are represented as the percentage of the maximum obtained in each individual experiment (no added PLP). 65 The similarity of PLP inhibition of A C C when activating with citrate and glutamate suggests that both of these activators is binding to the same site. If they were binding to different sites, and PLP interacted at the binding site of one or the other, the results obtained would be expected to differ. The similarity of the inhibition curves provide some additional evidence that these activators likely bind at a common site on A C C , and that both provide some protection from inhibition by PLP. Table 3.2 Comparison of ACC Pre-Incubation With Citrate and Glutamate on PLP Inhibition. Data for PLP inhibition of A C C is represented as IC50 for A C C prepared by ammonium sulphate precipitation and pre-incubated with 10 m M citrate or 50 mM glutamate, for the experiments shown in Figures 3.9 and 3.12, respectively. Tissue IC 5 0 for PLP Citrate after PLP Glutamate after PLP Citrate before PLP Glutamate before PLP Liver 220 (iM 295 uM 480 uM 530 uM Sk. Muscle 630 (IM 700 uM 770 uM 1100 uM 68 3.2.7 Potential Substrate Protection Against PLP Inhibition of A C C Although the experiments described previously cannot definitively show that PLP and citrate are indeed binding at the same site on A C C , it is possible to show, in principle, that PLP is not causing inhibition of A C C by interfering with substrate binding or catalysis. By increasing the amount of substrate present in the assay, A C C is guaranteed to be saturated with substrate. If PLP binds reversibly at or near one of the substrate binding sites, the increased substrate concentration should show some protection against PLP, since substrate would impair the ability of PLP to bind to A C C , simply through physical hindrance. The effects of doubling and tripling the substrate concentration in the A C C assay, with regard to PLP inhibition of A C C , is shown in Figure 3.13. The data in Figure 3.13 clearly indicates that doubling and tripling the substrate concentration in the A C C assay buffer gave little protective effect to A C C against PLP inhibition. Although this information is based only on one experiment, it suggests that PLP is not binding at or near the active site of A C C . Additional studies need to be done to clarify this result, and a further modification of this experiment currently being examined is to pre-treat A C C with its substrates (individually) and see i f incubation prior to PLP treatment affords any protection against inhibition. 69 Figure 3.13 Effect of Increasing ACC Substrates on PLP Inhibition of ACC. 120 -r 100.0 • Control • 2 X • 3 X A C C purified by ammonium sulphate precipitation of liver was treated with 0.5 m M PLP 30 minutes at 25 °C in the dark, followed by a 30 minute incubation with 10 m M citrate and subsequent activity assay. The assay buffer contained either the normal concentration of substrates (control), twice the normal amount or three times the normal amount of substrates, as indicated. Results are represented as % inhibition, relative to the control (normal assay buffer), for single experiments. 70 3.2.8 Effect of PLP Reactive Compounds on PLP-Inhibited A C C While pyridoxal phosphate has been shown to bind to many different proteins, it has also been shown to establish an equilibrium between the bound and free states [76]. This implies that PLP is not permanently bound to the target protein, and under certain circumstances can be removed. If PLP is indeed reacting to form a Schiff base with free amino group(s) of A C C , then it should be possible to prevent or reverse this effect with compounds that provide amino groups to react with both free PLP and PLP attached to proteins through a Schiff base. Two such PLP-reactive compounds are hydroxylamine (HA) and aminooxyacetate (AOA), whose structures are shown in Figure 3.14. Because H A and A O A both contain terminal amines, they react with free PLP, forming Schiff base products. H A and A O A thereby compete with the e-amino group of protein lysyl residues to which PLP normally forms a Schiff base. As well, since PLP is present in a bound and unbound equilibrium, removal of free PLP should induce bound PLP to detach from any proteins to maintain the bound:free equilibrium. In fact, A O A is an effective inhibitor of transaminases and is used to achieve transaminase inhibition, both on purified enzymes in vitro and in intact cells [82]. With A C C , loss of PLP binding should relieve inhibition caused by PLP, and activity of the enzyme should be regained when treated with either H A or A O A . The results of experiments testing this idea are shown in Figure 3.15. 71 Figure 3.14 Structures of PLP-Reactive Compounds. o H O - N H 2 H 2 N O Hydroxylamine Aminooxyacetate From Figure 3.15 it is evident that although neither H A or A O A affects the maximum activity of A C C when incubated with the enzyme alone, enzyme previously treated and inhibited by PLP shows recovery of activity when reacted with both H A and A O A . This indicates that both of these compounds are capable of reacting with PLP and in so doing, release some of the inhibition on A C C induced by PLP. This is consistent with previous results using these compounds with other enzymes labelled with or endogenously containing a PLP co-factor [65, 76, 77]. As shown, equal concentrations of PLP and either H A or A O A results in approximately 43% recovery of A C C activity. A summary of the data in Figure 3.15 as well as additional experiments is shown in Table 3.3. Reducing the concentration of H A and A O A 10-fold resulted in no recovery of A C C activity. As well, higher concentrations of hydroxylamine did not improve recovery of A C C activity, while aminooxyacetate at concentrations above 2 m M showed very poor recovery of A C C activity. An additional experiment using 100 p M PLP showed a similar percentage recovery of A C C activity with 1 m M H A and A O A (compared to Figure 3.15), but less recovery when using 100 u M concentrations of these compounds (Table 3.3). This is interesting, because it suggests that equal concentrations of PLP and either of H A and A O A is not sufficient to recover similar amounts of A C C activity. It seems as though a minimum concentration of 1 m M H A and A O A is required in order to achieve acceptable recovery of A C C activity following PLP inhibition. 72 Figure 3.15 Effect of Hydroxylamine and Aminooxyacetate on PLP-treated ACC Activity. A C C , partially purified by ammonium sulphate precipitation of liver extracts, was treated for 30 minutes with 1 m M PLP, followed by 30 minute incubation with either 1 m M hydroxylamine (HA) or 1 mM aminooxyacetate (AOA), with a subsequent 10 m M citrate activation and A C C activity assay. (1) and (2) are A C C treated with only H A and A O A , respectively. (3) is A C C treated with only PLP. (4) and (5) are A C C treated with PLP followed by HA or A O A treatment, respectively. Results are expressed as the percentage of the maximum obtained in the absence of PLP, HA or A O A . Results are for single experiments. 73 The key understanding that arises from these experiments is that PLP does seem to exist in an equilibrium with bound and unbound forms. Addition of H A and A O A can remove free PLP and promote at least some de-inhibition of A C C , by removal of PLP. Results with H A and A O A also provide a possible explanation for the fact that citrate can induce recovery of some A C C activity when added following PLP incubation. Addition of enough citrate, especially at low PLP concentrations can force some of the bound PLP into solution, thus relieving some inhibition of A C C . 3.2.9 Effect of PLP Reduction by Sodium Borohydride As shown in Figure 3.15 and described elsewhere, PLP seems to exist in an equilibrium between bound and unbound forms [76]. Using this rationale, i f A C C treated with PLP were dialyzed or washed thoroughly, it is reasonable to expect that since free PLP would be washed away, some PLP bound to A C C would dissociate, thus relieving inhibition on the enzyme. This hypothesis was tested using an incubation with 0.5 m M PLP followed by incubation with some samples with sodium borohydride at concentrations of 0.5 mM, 1.0 mM and 2.5 mM. Duplicates of these samples were either desalted or assayed directly. Desalting used a buffer containing 20 mM citrate and was performed using 15 mL Biomax-50 centrifugal filter units (Millipore), with multiple buffer changes to ensure complete washing of A C C . Washed enzyme was then concentrated in the same unit and assayed directly for A C C activity. The results of this experiment are shown in Figure 3.16. 74 Table 3.3 Effect of Hydroxylamine and Aminooxyacetate on PLP-treated A CC Activity. A C C , partially purified by ammonium sulphate precipitation of liver extracts, was treated for 30 minutes with PLP at the concentrations shown, followed by 30 minute incubation with either hydroxylamine (HA) or aminooxyacetate (AOA) at the concentrations shown. The enzyme was then incubated with 10 m M citrate and then subsequently assayed. (% Activity Recovered) refers to the percentage of total A C C activity (of control) recovered following treatment with either H A or A O A . Results are for single experiments. Some results are also presented in Figure 3.15. [PLP] [HA] [AOA] % Activity (mM) (mM) (mM) Recovered 0.1 1.0 - 44 0.1 - 1.0 26 0.1 0.1 - 27 0.1 - 0.1 31 1 1.0 - 43 1 - 1.0 44 1 0.1 - 0 1 - 0.1 0 1 10 - 36 1 - 6 4 75 Results of this experiment lead to several conclusions. First, it supports the hypothesis described above, that PLP interacts reversibly with A C C and the removal of PLP by washing PLP-labelled A C C can result in recovery of enzyme activity (Figure 3.16 lanes 3 and 4). Second, it shows that reduction of PLP-treated A C C with sodium borohydride can render A C C irreversibly inhibited, thereby preventing loss of PLP inhibition during subsequent washing, even in buffer containing citrate (Figure 3.16 lanes 5-7). This experiment also shows that borohydride reduction of ACC-PLP can lead to the same or greater inhibition as with A C C - PLP which is unreduced and unwashed. Finally, it illustrates that increasing the BFL^PLP ratio beyond 5:1 does little to improve A C C inhibition. In fact, a 2:1 ratio works about equally well, suggesting that at those concentrations, BH4 can adequately react with most, i f not all, PLP present as Schiff bases with A C C . Looking ahead to future work, one particularly important aspect of the experiment shown in Figure 3.16 is the observation that PLP is indeed forming a Schiff base with A C C , and borohydride reduction allows a covalent linkage between PLP and A C C to be formed which is resistant to washing. This is an important consideration, since covalent coupling of PLP has been used by others to determine binding sites and essential lysines; specifically by proteolytic digestion and analysis of PLP-labelled peptides. Thus, showing that this technique works with A C C is key to further experiments used to characterize the PLP binding site, which is likely also the citrate binding site on A C C . An initial example of this work on PLP is shown in Figure 3.17. This figure clearly shows that under the conditions used, PLP can be successfully linked to A C C , and through use of radioactive borohydride, its incorporation can be followed. The segment with the highest radioactive counts represents the single band on an SDS-PAGE gel corresponding to 76 A C C . Additional work regarding the level of PLP incorporation and the appropriate concentrations to use is currently ongoing. Figure 3.16 Effect of Sodium Borohydride Reduction of PLP-Labelled ACC. A C C partially purified by ammonium sulphate precipitation of liver extracts was allowed to react with 0.5 m M PLP for 30 minutes followed by a 20 minute incubation with varying borohydride concentrations. Samples were then washed with buffer containing 20 mM citrate using Biomax-50 centrifugal filter units and assayed for A C C activity. Results are represented as % activity of A C C treated with only 20 m M citrate (1). (2) is A C C treated with 0.5 mM borohydride; (3), A C C treated with 0.5 mM PLP without washing, (4) A C C treated with 0.5 m M PLP with washing, (5), (6) and (7) are A C C treated with 0.5 mM PLP followed by 0.5 mM, 1.0 mM and 2.5 mM borohydride and then washing, respectively. (2), (4) and (5) are the average of two experiments, with all other results coming from single experiments. 77 Figure 3.17 PLP/BH4 Incorporation in A CC. A C C purified by ammonium sulphate precipitation of liver was incubated with tetrameric avidin beads for 45 minutes at 4 °C. The beads were then washed thoroughly with buffer and incubated with 5 mM PLP for 20 minutes at 25 °C in the dark, followed by reaction with 15 mM [3H]-BH4 for 10 minutes at 25 °C. The beads were then again washed thoroughly with buffer, followed by digestion and SDS-PAGE on a 4.5% gel. Following electrophoresis, the gel was stained with Coomassie Blue and dried. The gel lane was then sliced into segments and each segment digested and subjected to scintillation counting as described in Methods. 30000 25000 20000 E % 15000 x CO 10000 5000 4-ACC Band f f t t f t t f f t 8 9 10 11 12 13 14 15 16 17 18 19 20 Lane Segment # 78 3.2.10 Spectrophotometric Analysis of PLP Reduction It is well established that free PLP has an absorbance maximum around 395 nm. Significantly, when reduced by borohydride, the PLP wavelength maximum shifts from 395 nm to around 325 nm, characteristic of the formation of a phospho-pyridoxyl-lysyl group, with an extinction coefficient of 9710 M" 1 cm"1 at 325 nm [77]. As a complementary method to studying PLP interaction with A C C , spectrophotometric methods were therefore employed to exploit the absorbance characteristics of PLP. In order to first test the feasibility of using spectrophotometric detection to monitor the binding state of PLP, the absorbance of PLP was studied when free in solution as well as following reduction with borohydride in the presence of lysine. Following this, tests with PLP combined with proteins were performed. Spectra from both experiments are shown in Figure 3.18. In Figure 3.18A, two curves are shown. One is the spectrum of a solution containing 5 uM PLP and 5 u M lysine; the second is that of PLP/lysine combined with B H 4 , all at 10 uM. Both spectra were measured at neutral pH. The key point to observe in this figure is that PLP bound to lysine exhibits the expected absorbance maximum at about 390 nm, while the spectrum is markedly shifted and has a sharper absorbance maximum near 330 nm, following reduction by sodium borohydride. Because PLP shows the same spectrum in the absence or presence of lysine (but not borohydride) this also shows that borohydride treatment of a PLP-lysine mixture leads to the shift of the absorbance maximum due to reaction of the reversible Schiff base to produce a permanent covalent PLP-lysyl compound. In Figure 3.18B, the spectrum of partially purified liver A C C in the presence of 5 u M PLP is shown alongside that of A C C and 5 uM PLP following the addition of 5 uM B H 4 . As expected, protein + PLP 79 shows an absorbance maximum at 390 nm, and addition of borohydride to the protein/PLP mix results in reduction of the PLP-protein Schiff base and causes a shift in the absorbance maximum from 395 nm to 325 nm. The power of these experiments lies in the fact that a detectable change in PLP absorbance can be measured even at low 5 LtM concentrations (7.5 nmoles of PLP in the cuvette during measurement), both with free lysine and more importantly, with proteins. The ability to detect a PLP signal at low concentrations is an important consideration, since this method may be applied to tracking and purification of A C C peptides labelled with PLP. The sensitivity of the instrumentation available makes using a spectrophotometric approach to monitor PLP labelling an attractive and viable method. It is important to note that although A C C is present in the sample treated with PLP and BH4 in Figure 3.18B, the preparation is still far from homogeneous and it is most likely that PLP is also reacting with other proteins in this sample. Further work using avidin affinity purified A C C will be required to show definitively that PLP-labelling of A C C can be followed spectrophotometrically. 80 Figure 3.18 Spectrophotometric Analysis of PLP Reduction by BH4 With Lysine & Proteins. In (A), PLP (5 | lM) was incubated with lysine (5 uM) for 35 minutes at 25 °C, and then the absorbance spectrum measured (- - -), or 10 u M PLP and lysine were reacted for 35 minutes at ambient temperature and then reacted for 10 minutes with 10 uM BH4, and the absorbance spectrum recorded ( ). In (B) a similar approach was taken, except lysine was replaced with ammonium sulfate purified liver A C C , with PLP and B H 4 present at 5 uM. A l l spectrophotometric measurements were made at 37 °C. 81 3.3 SUMMARY Based on evidence that PLP can bind to citrate active and/or allosteric sites and thereby inhibit a variety of enzymes [64, 69-77], we initially argued that PLP might provide a useful probe for the citrate binding site of A C C . PLP strongly inhibited A C C , with particular potency against enzyme from tissues containing predominantly ACC-1 (adipose, liver, heart). Skeletal muscle A C C (largely ACC-2) seemed less sensitive to PLP inhibition. PLP inhibition of A C C was rapid, with near maximal A C C inhibition for a given PLP concentration seen within 30 seconds for liver tissue. Heart and muscle A C C showed slower response to PLP, but short term inhibition of A C C occurred rapidly, even in these tissues. Binding of PLP to the citrate-binding site of A C C was supported by at least partial protection against PLP inhibition using citrate/glutamate pre-activation. Citrate/glutamate afforded slight protection at high PLP concentrations, and greater protection at low PLP concentrations. A C C from liver/heart shows greater protection than A C C from muscle. Treatment of PLP-labelled A C C with compounds known to react with PLP, via Schiff base formation, resulted in recovery of A C C activity (approximately 50%). H A and A O A (at 1 mM) achieved the best recovery of A C C activity, against both 100 p M and 1 m M PLP. Treatment of PLP-labelled A C C with sodium borohydride resulted in covalent and irreversible linkage of PLP to A C C , so that PLP could no longer be removed and A C C no longer re-activated by washing. Washing of borohydride-treated PLP/ACC revealed inhibition similar to or greater than that seen simply by adding PLP with no reduction, indicating that BH4 treatment was reducing PLP bound to A C C in its inhibitory position. Spectrophotometric analysis of borohydride reduction of PLP produced characteristic changes in absorbance maximum, and might be used to follow PLP-labelled A C C peptides. 82 CHAPTER FOUR PLP ANALOGS ACCOUNT FOR PLP SPECIFICITY 4.1 RATIONALE Previous work has demonstrated the utility of PLP as an enzyme inhibitor [72-75], and as a probe for the binding site of citrate or other anionic molecules [69-71, 76, 77] or reactive lysine residues [63-68]. It has been speculated that the main reason that PLP will bind to certain sites, notably citrate binding sites, is due to the presence of positively charged residues at these sites, which ordinarily allow for stabilization of the negative charges on citrate. While an attractive idea, conclusive evidence for this is really only available in the case of PLP binding to the active site of transaminases (for which crystal structures are known). Some studies have examined several PLP analogs for functionality, and although these point towards the importance of several moieties on PLP, they do not comprehensively define the portions of PLP that account for its greater effectiveness than its analogs in these studies [71, 74,76,77]. The goal of the work reported in this chapter was to perform an analysis of the effects of a range of different PLP analogs, as well as other compounds containing similar moieties, in an attempt to understand which portions of PLP account for its effect on ACC. This was achieved by performing inhibition assays of ACC. 83 4.2 RESULTS AND DISCUSSION 4.2.1 Possible PLP Moieties That Could Account For A C C Inhibition As shown in Figure 4.1, there are several possible moieties on PLP that could account for both its specificity of binding, as well as the efficiency and strength of binding to A C C and other proteins. These include one or more of four moieties, as designated in Figure 4.1: (1) the reactive aldehyde at position 4', (2) the anionic phosphate group at position 5', (3) the ring hydroxyl group at position 3 (which should be ionized under most conditions tested), and (4), the pyridine ring itself. These four moieties and the role they play in PLP binding will be considered and discussed below. The methyl group on the pyridine ring of PLP is expected to be non-reactive and is not considered in the following analyses. Figure 4.1 Possible PLP Moieties Accounting For Specificity and Binding Efficiency. Structural characteristics on PLP which might account for its binding specificity and efficiency with A C C . (1) is the reactive aldehyde, (2) is the phosphate group, (3) is the hydroxyl group and (4) is the pyridine ring. 84 4.2.2 PLP's Reactive Aldehyde The aldehyde at position 4' on PLP is the reactive component of the molecule which forms the Schiff base with lysine residues on proteins. Since this moiety is essential in forming the Schiff base, enzymes which use PLP as a cofactor are not expected to function as efficiently using analogs lacking this aldehyde. However, when PLP is able to inhibit other enzymes, notably those with a citrate binding site, the absence of this aldehyde may or may not affect this inhibitory capacity. Two previous studies have shown that replacing the aldehyde with an amino group (pyridoxamine phosphate) resulted in loss of inhibition against succinic semialdehyde dehydrogenase [76] and succinic semialdehyde reductase [77]. In an effort to see i f the aldehyde group accounted for some of the inhibitory action of PLP on A C C , several compounds which are similar to PLP were tested for their inhibitory action on A C C . These include pyridoxamine, pyridoxamine phosphate, and pyridoxine. The structures of these three compounds are shown in Figure 1.3. Since PLP shows very strong inhibition of A C C at a concentration of 1 m M (approximately 90% inhibition - 10% residual activity) this concentration was chosen as the standard to assess the effects of analogs in this and other sections. Inhibition at concentrations higher than 1 m M were not considered, since these concentrations are not physiologically relevant in most cases, nor achievable pharmacologically, in general. A l l experiments were performed using ammonium sulphate purified liver A C C , since this preparation typically showed the best response to PLP inhibition and was most easily obtained. Results of treating A C C with 1 mM of pyridoxamine, pyridoxamine phosphate and pyridoxine are compared with 1 m M PLP inhibition in Figure 4.2 85 Figure 4.2 Effect of Aldehyde-Lacking PLP Analogs. • PLP a Pyridoxamine y Pyridoxamine Phosphate s Pyridoxine A C C partially purified by ammonium sulphate precipitation of liver was treated with PLP or the indicated analogs at concentrations of 1 mM for 30 minutes, followed by a 30 minute incubation with 10 m M citrate and subsequent activity assay. Results are represented as % inhibition, compared to a sample of A C C treated with 10 m M citrate alone (0% inhibition) for each separate experiment. Results for PLP, pyridoxamine and pyridoxamine phosphate are given as the mean ± S E M for 6, 2 and 2 experiments, respectively. Pyridoxine results are for a single experiment. 86 As can be seen in Figure 4.2, PLP analogs lacking the aldehyde show very poor inhibition of A C C . The strongest of the three tested, pyridoxamine, shows only 21% inhibition of A C C , compared to the 92% inhibition with PLP itself. These results indicate that PLP's aldehyde group is important for its inhibitory action on A C C . The result best showing this is pyridoxamine phosphate, since this compound has an amine replacing the aldehyde, but with the phosphate group and phenolic oxygen still intact. Pyridoxamine phosphate shows very poor inhibition, at only 16%, suggesting that the aldehyde on PLP is binding to a nearby lysine residue on the enzyme, or that the amino group in pyridoxamine was repelled by the positively charged residues in the binding site. The results of pyridoxamine and pyridoxine have to be analyzed carefully, because they both lack not only the aldehyde, but the phosphate group as well. From these results it is interesting that these two compounds, which differ at the group off of position 4', show reasonable difference in inhibitory action on A C C . Pyridoxamine shows 14% better inhibition that pyridoxine, which perhaps indicates that the amino group on pyridoxamine is better able to hydrogen bond or at least deal with the high positive charges hypothesized to surround the citrate/PLP binding site, than is the hydroxyl group of pyridoxine. Two additional compounds, both with a pyridine ring and an aldehyde, show that the aldehyde alone is not sufficient for inhibition. Both 2- and 4-pyridinecarboxaldehyde poorly inhibit A C C , as shown in Figure 4.3. The structures of these molecules is given in Figure 4.10. Surprisingly, of the two compounds, 4-pyridinecarboxaldehyde, which more closely resembles PLP, shows less inhibition of A C C than does 2-pyridinecarboxaldehyde. This also confirms that the aldehyde and pyridine ring alone are not sufficient to account for A C C inhibition, regardless of the position of the aldehyde group on the ring. 87 Figure 4.3 Effect of Pyridinecarboxaldehydes on ACC. 100.0 90.0 80.0 v £ 60.0 o si c o JO !E c 50.0 40.0 30.0 20.0 10.0 0.0 91.5 16.3 3.3 H PLP • 2-Pyridinecarboxaldehyde s 4-Pyridinecarboxaldehyde A C C partially purified by ammonium sulphate precipitation of liver was treated with PLP or the indicated analogs at concentrations of 1 m M for 30 minutes, followed by a 30 minute incubation with 10 m M citrate and subsequent activity assay. Results are represented as % of A C C inhibition, compared to a sample only treated with citrate (0% inhibition) for each separate experiment. Results for PLP are given as the mean ± S E M for 6 experiments, while those for the two analogs are for single experiments each. 88 4.2.3 PLP's Phosphate Group Although the aldehyde group on PLP may contribute a significant portion of the specificity and efficacy of PLP, it is not the only component that plays a role. Several studies have previously shown that pyridoxal can only weakly, i f at all substitute for PLP as an inhibitor [71, 76, 77]. In order to verify the effect of pyridoxal on A C C , similar inhibition assays to those testing the effect of the PLP aldehyde were performed, using pyridoxal and another analog, 4-pyridoxic acid. The structures of these compounds are shown in Figure 1.3, and the result of the assays is given in Figure 4.4. As can be seen in Figure 4.4, pyridoxal inhibits A C C at a reasonable level, but is still far less effective than an equal concentration of PLP. 4-pyridoxic acid shows poor inhibition characteristics similar to other analogs such as pyridoxamine and pyridoxamine phosphate. The results from these experiment are interesting, because they show that loss of a phosphate group on pyridoxal does not decrease the inhibitory properties against A C C by nearly as much as a similar analog such as pyridoxamine. This clearly indicates that it is indeed the aldehyde group on PLP which plays the most important role in its binding and inhibitory characteristics against A C C . Analogs lacking both the aldehyde and the phosphate group, such as pyridoxine and 4-pyridoxic acid, show very poor inhibition, and even pyridoxamine phosphate, which contains the phosphate group, shows only minor improvements over the other two. Thus although the phosphate group may play a minor role in determining PLP specificity and efficacy, it is not the primary moiety accounting for PLP's inhibitory action on A C C . 89 Figure 4.4 Effect of Phosphate-Lacking PLP Analogs. A C C partially purified by ammonium sulphate precipitation of liver was treated with PLP or the indicated analogs at concentrations of 1 mM for 30 minutes, followed by a 30 minute incubation with 10 m M citrate and subsequent activity assay. Results are represented as % of A C C inhibition, compared to a sample only treated with citrate (0% inhibition) for each separate experiment. Results for PLP and pyridoxal are given as the mean ± S E M for 6 and 2 experiments, respectively. 4-Pyridoxic acid results are for a single experiment. 90 4.2.4 PLP's Anionic Poles and Pyridine Ring Direct testing of the efficacy of the hydroxyl group on PLP is very difficult, since analogs lacking this group are not commercially available, and considerable expense and time would be required to synthesize them. However, since both ends of the PLP molecule can be negatively charged (the hydroxyl and the phosphate), it seemed relevant to examine i f PLP's spaced negative charges could account for some inhibition of A C C . This can be rationalized by comparing the structures of PLP and citrate, as shown in Figure 4.5. Since both molecules have spaced negative charges (5 carbons apart in citrate and 6 carbon equivalents in PLP), these could account for the ability of PLP to enter the citrate binding site and thus position itself to form a Schiff base with the enzyme. In order to test the importance of the spaced negative charges, several compounds were used, each with slightly different spacing or orientation of negatively charged groups. The most common group used to produce a negative charge were carboxyl groups, since these are also present on citrate and maintain the most structural similarity to the natural ligand, as well as generally being negatively charged at neutral pH. Compounds with separated negative charges best thought to resemble PLP were pyridine rings with two carboxylic acid moieties. The results of A C C treatment with these compounds is shown in Figure 4.6, and their structures in Figure 4.10. 91 Figure 4.5 Comparison of PLP and Citrate Structures. Figure 4.6 Effect of Pyridinedicarboxylic Acids on A CC. 120 100 if 80 O ° 60 o '•P S | 40 5 5 20 91.5 74.4 [ C O O H 74.1 , T "XT N 79.9 T 0 -• PLP H 2,5-Pyridinedicarboxylic Acid • 3,4-Pyridinedicarboxylic Acid • 3,5-Pyridinedicarboxylic Acid A C C partially purified by ammonium sulphate precipitation of liver was treated with PLP or the indicated analogs at concentrations of 1 mM for 30 minutes, followed by a 30 minute incubation with 10 m M citrate and subsequent activity assay. Results are represented as % of A C C inhibition, compared to a sample only treated with citrate (0% inhibition) for each separate experiment. Results for PLP, 2, 5-, 3, 4-, and 3, 5-pyridinedicarboxylic acids are given as the mean ± S E M for 6, 2, 4 and 4 experiments, respectively. 93 As shown in Figure 4.6, all three pyridinedicarboxylic acids tested appear to inhibit A C C very well at 1 mM concentrations; far better in fact than any other analogs tested. This result is interesting because it indicates that while the aldehyde portion of PLP may account for strong binding, hence citrate cannot adequately remove PLP at higher concentrations, it is also likely that the negative charges on PLP allow it to enter the citrate binding site, due to favourable ionic interactions with positively charged residues in the binding site. In addition, the difference in location on the ring of the carboxylic acid groups appears to be rather unimportant, although the 3,5-pyridinedicarboxylic acid form seems to show slightly better A C C inhibition than the other two. This is likely due to favourable and similar charge placement on the pyridine ring relative to citrate and PLP, whereas the others differ slightly. Having 5 carbon equivalents separating the negative charges appears to be best (matching with citrate), while 4 and 6 equivalents between negative charges show no real difference. Further, the second carboxyl group of this set of analogs appears to be better for binding to A C C than the corresponding hydroxyl group on PLP. The pyridine ring in both PLP and the three pyridinedicarboxylic acids could play a role in PLP binding to A C C . The ring itself is fundamental for PLP-dependent catalysis, but it's role in PLP acting as an inhibitor is unclear. In order to ascertain the function of the pyridine ring, three benzenetricarboxylic acids were tested for A C C inhibition. These compounds lack the ring nitrogen atom, but are still aromatic, and thus fairly similar in overall size to the pyridinedicarboxylic acids tested previously. The results of these experiments are shown in Figure 4.7, and the structures of these compounds in Figure 4.10. 94 Figure 4.7 Effect of Benzenetricarboxylic Acids on ACC. 100.0 90.0 80.0 "5 70.0 £ 60.0 *- 50.0 o c •g 40.0 !a | 30.0 3? 20.0 | 10.0 0.0 91.5 a: H O O C C O O H 32.4 PLP • 1,2,3-Benzenetricarboxylic Acid 11,2,4-Benzenetricarboxylic Acid • 1,3,5-Benzenetricarboxylic Acid A C C partially purified by ammonium sulphate precipitation of liver was treated with PLP or the indicated analogs at concentrations of 1 mM for 30 minutes, followed by a 30 minute incubation with 10 m M citrate and subsequent activity assay. Results are represented as % of A C C inhibition, compared to a sample only treated with citrate (0% inhibition) for each separate experiment. Results for PLP are given as the mean ± S E M for 6 experiments, while the 3 analog results are for single experiments. 95 From Figure 4.7 it is evident that all three benzenetricarboxylic acids inhibit A C C to a fairly substantial degree at 1 mM (30-50%). Interestingly, even the best of the three benzenetricarboxylates, the 1,3,5-form, is not as effective at inhibiting A C C as even the least effective of the pyridinedicarboxylic acids, 3,4-pyridinedicarboxylic acid (greater than 70% inhibition). This can be interpreted to mean that although the negative charge placement on the ring is important, it is possible that either the pyridine ring itself is required to account for the additional effective inhibition shown by the pyridinedicarboxylic acids, or the additional carboxylic acid group on the benzenes negatively affects the inhibition properties of these compounds, relative to the pyridines. It is interesting too, that for 1,3,5-benzenetricarboxylic acid, the negative charges are equally 5 carbon equivalents apart, while the 1,2,3-form does not have this same spacing, and the 1,2,4-form has only one of these 5 carbon equivalents. This suggests that the placement of the negative charges could play a role with the benzene compounds, although this appeared not to be the case with the pyridinedicarboxylic acids. In an attempt to clarify these ambiguities, two additional compounds were tested. These include a benzenedicarboxylic acid (phthalic acid) and a pyridinecarboxylic acid (nicotinic acid or niacin). The structures of these compounds are given in Figure 4.11 and the results of the experiments are shown in Figure 4.8. 96 Figure 4.8 Effect of Phthalic Acid and Nicotinic Acid on ACC. •5 s 100.0 90.0 80.0 4 70.0 60.0 4 50.0 c | 40.0 £ - 30.0 4 ss 20.0 10.0 0.0 91.5 COOH COOH 51.2 v : : !-'' 'OH 2.3 I PLP B Pthalic Acid s Nicotinic Acid A C C partially purified by ammonium sulphate precipitation of liver was treated with PLP or the indicated analogs at concentrations of 1 mM for 30 minutes, followed by a 30 minute incubation with 10 m M citrate and subsequent activity assay. Results are represented as % of A C C inhibition, compared to a sample only treated with citrate (0% inhibition) for each separate experiment. Results for PLP are given as the mean ± S E M for 6 experiments, while the 2 analog results are for single experiments. 97 The results of the experiments shown in Figure 4.8 illustrate three points, and help to address the uncertainties described above. First, since phthalic acid is 1,2-benzenedicarboxylic acid (with 4 carbon equivalents between negative charges), and it shows an A C C inhibition level comparable to 1,3,5-benzenetricarboxylic acid (~ 50%) this indicates that the third carboxylic acid groups at positions 3 and 4 of 1,2,3- and 1,2,4-benzenetricarboxylic acids, respectively, do not account for any increase in inhibitory activity. In fact, these third carboxylic acids appear to decrease the inhibition of those two benzenetricarboxylic acids relative to phthalic acid. Secondly, it shows that having 5 carbon equivalents between negatively charged groups is not absolutely essential, at least on benzene rings and when only two negatively charged groups are present. Introduction of the third negatively charged group appears to decrease inhibitory effect, especially i f that group is directly adjacent on the ring to one of the others. Perhaps some charge repulsion is occurring in these molecules, which distorts the ring structure or the ability to interact with the positively charged residues of the citrate binding site. Finally, this result shows that having two carboxylic acids on a benzene ring does not produce the same level of A C C inhibition as a pyridinedicarboxylic acid. This suggests that the pyridine ring may be contributing to the inhibitory effect of the pyridinedicarboxylic acids and PLP. The observation that nicotinic acid (3-pyridinecarboxylic acid) shows almost no inhibition of A C C also indicates that a pyridine ring with only one negatively charged substituent is ineffective, in sharp contrast to a pyridine ring with two carboxylic acid moieties. Thus it appears that a combination of a pyridine ring, with two negatively charged groups accounts for substantial efficacy of PLP inhibition of A C C . Significantly, the second 98 carboxyl group may confer a stronger influence than the ring hydroxyl of PLP, which is probably only partially ionized at neutral pH. 4.2.5 Non-Aromatic PLP Analogs A l l of the analogs tested so far have had either a benzene ring or a pyridine ring as the central unit. Although the results to date seem to indicate that a pyridine ring, as opposed to a benzene ring, is beneficial for conferring A C C inhibitory characteristics to these molecules, these experiments have not addressed whether a ring structure is needed for A C C inhibition. To address this issue, several molecules were tested for A C C inhibitory capability. These molecules either did not contain rings at all, or had rings which were separate from the negatively charged groups on the molecule. These molecules included: pimelic acid (a straight chain, saturated, dicarboxylic acid), 1,3-acetonedicarboxylic acid (a dicarboxylic acid with a carbonyl group between the two carboxylic acids), benzylmalonic acid, and butylmalonic acid. The structures are shown in Figure 4.11, and the results in Figure 4.9. From Figure 4.9 (compared to Figure 4.7), it is clear that pimelic acid inhibits A C C slightly better than 1,3,5-benzenetricarboxylic acid and phthalic acid. Benzylmalonic acid and butylmalonic acid did not inhibit A C C at all (data not shown). This indicates the two carboxyl groups alone provide a substantial inhibitory potential, provided the two negatively charged groups are spaced apart. For pimelic acid the spacing is by 5 carbons, and inhibition was evident, whereas essentially no spacing is present between the carboxyl groups of the malonic acid species. Furthermore, pimelic acid is a less effective inhibitor than either PLP or the pyridinedicarboxylates, demonstrating that the pyridine ring does still contribute to inhibition of A C C . 99 Figure 4.9 Effect of Non-Ring PLP Analogs on A CC. 100.0 90.0 80.0 & 70.0 > £ 60.0 o o c o ••p Ja !E c 50.0 40.0 30.0 20.0 10.0 0.0 91.5 55.2 HOOC C O O H 3.4 I PLP • Pimelic Acid s 1,3-Acetonedicarboxylic Acid A C C partially purified by ammonium sulphate precipitation of liver was treated with PLP or the indicated analogs at concentrations of 1 m M for 30 minutes, followed by a 30 minute incubation with 10 m M citrate and subsequent activity assay. Results are represented as % of A C C inhibition, compared to a sample only treated with citrate (0% inhibition) for each separate experiment. Results for PLP are given as the mean ± S E M for 6 experiments, while the 2 analog results are for single experiments. 100 Figure 4.10 Structures of Some of the PLP Analogs Tested. N 2 -Pyri d in eca rb o xal d e hyd e o N 4-Pyridinecarboxaldehyde H O O C ^ ^ H O O C N C O O H . C O O H N C O O H C O O H 2,5-Py ri d i n ed icarb o xy I i c Aci d 3,5 -Pyr id in ed i c arb o xy Ii c A ci d 3,4-Pyridinedicarboxylic Acid C O O H C O O H H O O C C O O H C O O H 1,2,3 -B en zenetricarboxylic Acid 1,2 y4 -B en zenetricarboxylic Acid 1,3,5 -B en zenetricarboxylic Acid 101 Figure 4.11 Structures of Some of the PLP Analogs Tested. C O O H ' C O O H H O O C C O O H Pthalic Acid 1,3 Acetonedicarboxylic Acid o N H O O C C O O H Nicotinic Acid Pimelic Acid . C O O H C O O H C O O H C O O H Benzylmalonic Acid Buylmalonic Acid 102 4.2.5 Use of A C A and an Analog Upon the advice of a colleague, an additional compound, which was thought to bear structural similarity to PLP, was tested for A C C inhibitory properties. This compound, 1'acetoxychavicol acetate (ACA) is isolated from the rhizomes of Languas galanga (Zingiberaceae), a common plant in Thailand. It has been shown in several studies to inhibit xanthine oxidase, and prevent tumours in several tissue types in rats [83]. Although the available evidence did not strictly suggest any activity with regard to inhibiting A C C , it was tested like the other analogs. In fact, A C A did cause substantial A C C inhibition, and so one of its analogs, acetylsalicylic acid (aspirin) was also tested for A C C inhibition. The results for both compounds are shown in Figure 4.12 and the structures in Figure 4.13. The data in Figure 4.12 is very interesting, because A C A appears to be almost as potent an inhibitor of liver A C C as is PLP. This is a potentially important result, because although there is some structural similarity between the two compounds, how A C A would favourably interact with A C C is not clear. Notably, there are no groups with negative charge present on A C A , although the two ester functionalities could potentially hydrogen bond with residues in the citrate binding site. Nonetheless, this molecule does inhibit A C C with high potency. This raises the question as to whether this compound is also acting at the citrate binding site, or i f it is inhibiting A C C through other means. Unfortunately, due to limited supply of this compound, inhibition of A C C following citrate activation of the enzyme could not be performed, thus precluding an examination of citrate protection of A C C against A C A . . This might be an interesting candidate compound to consider in future work. It might also be interesting to test the effects of some of the compounds already examined, in which the carboxyl groups are esterified, as in A C A . 103 Figure 4.12 Effect of ACA and Acetylsalicylic Acid on ACC. 2 c o IE c 140.0 130.0 120.0 110.0 100.0 90.0 80.0 70.0 60.0 50.0 40.0 30.0 20.0 4 10.0 0.0 T 91.5 86.7 COOH 56.1 PLP H ACA S Acetylsalicylic Acid A C C partially purified by ammonium sulphate precipitation of liver was treated with PLP or the indicated analogs at concentrations of 1 m M for 30 minutes, followed by a 30 minute incubation with 10 m M citrate and subsequent activity assay. Results are represented as % of A C C inhibition, compared to a sample only treated with citrate (0% inhibition) for each separate experiment. Results for PLP are given as the mean ± S E M for 6 experiments, while the 2 analog results are for single experiments. 104 Figure 4.13 Structures of ACA and Acetylsalicylie Acid. 1 "acetoxychavicol acetate (ACA) Acetyls alicylic Acid Interestingly, aspirin also shows reasonable inhibition of A C C at 1 mM, although this is about the same level as other PLP analogs tested, such as phthalic acid and 1,3,5-benzenetricarboxylic acid. The carboxylic acid and benzene ring structure may account for this action. In order to allow easier comparison of PLP analogs, their structural components and their effect on A C C , this information is presented in Table 4.1. 105 o o -£2 K e S § S-i <u » J3 > -4-> T3 ° iS ^ 'C ^ B -4—» 43 in" en o3 I"8 ' « £ O ~ ^ s s O ^ B a) <u ° a. « a? 03 i> <4H U 43 ^ r 6 3 ° 0) 43 B § g - oo "B B 3 O U T3 43 " U O 3 * i O -B B oo <D f-c 03 <U o 3 5 1) O t_ • H n 43 43 <+H B o o3 B CO CD »H PH 1) o *-£ > ii 00 •S 3 a S —. B 13 « 5b 00 03 O 2 ° B B B 03 oo C ! * o 43 <U oo O CD ~ 43 S 2 -§ CD <u> oo . S CD O -4—* . 03 "O o <D ^ fa a 0JQ a o <D 03 IB oo o3 2 .2 03 *-»-* o3 oo o o3 oo -*-> ii oo ^ G O £ 2 a ^ 5 "O B o •'H1-1 13 « ^ g O B 2 c O C U U < o o ^6 0> S M N g pa .S 0X) T3 S3 "B •* s CN CM 0> 08 J3 a. «> o -B a. <u T3 ^. -B T3 B B O Q. a o U PH 03 X O "U B 1 o -4—» 03 OH OO O 43 PH <U B I X o "T3 D B ' X O "O 'o < o 'x o "O PH I 0) x o •§ o ii B PH CN 43 13 x o •s o D B 106 o -a M 3 es H .S 03 on 0) C3 O CD to < O c o S U -2 <-> S3 m IT) o "3-CM <N CM in in S 2 03 .2 WD PH o JS 2 CM CM CM 4> « OH s» O CM a> T3 a o I o U '3 < X o 1 o • I—I -e CM" T3 '3 < o o •e o o <: X o •I o in 12 '3 < o X O 1 •a -*-» <u C <u N m rn CM 12 '3 < o c <u N <D CQ 4 CM" 12 '3 < o o 1 o •£ c <u N ° ? m -a 12 '3 < o "3 a • i-H OH 3 et H £ J3 t§ ^ ^ O 'C S3 & c CD 1 -4—» 03 ^ & 00 133 o * S e 11 o « 1.3 S3 CD ^3 rH CD S3 CD CD c 1) ° ft 03 -!=: o S3 CD 3 CO 03 CD ,CD S-, c+-i CD oo X 2 <+* ^ 9 J i CD i -S3 03 r. T3 S3 03 S3 =3 O oo o 03 CD S3 CD oo CD i-t ft ft - ~ S g O 3 J3 i o 03 CD 3^ £ " < + H <D o 'S >, S3 S3 c 03 •2 S 3 oo <D .2 03 S3 CD 00 <D S-i ft CD <D oo ^ S3 oo o3 a* =3 « •33 « % oo o3 O „ oo i o O J3 CD oo O CD S3 ' CD _ 2 -3 03 CD 2 ° 03 + J 03 00 60 03 O ft 03 g ft 03 S ^ OH <J CD <u £ -o oo -S3 CD O 03 T 3 O CD I S 03 £ 60 a .a -o -2 ^ 00 00 M S3 O a a O S3 03 CD ^ a o S3 O S u •2 u 2 << s= o o o © oo in cu S 2 a* "^ 2 W> a. O ! -o S3 a oo O •S3 PH a> T3 J3 T3 13 S3 3S O a = o U • I-H O < O 'S o a N S3 CD PQ T3 "3 < o 'S o 13 a T3 O < ^ o X o •e 03 O -3 CD S3 O CD O '3 < o - —H ^—• o o U CD » 03 CD O 03 "o CD "> 03 •s >. X! O +-» CD O < S3 •g 'ft 00 < '3 < o 00 CD O 4.3 S U M M A R Y From the experiments presented in this chapter, it becomes clear that PLP contains multiple structural components which regulate and affect its ability to inhibit A C C . Since it contains all of the favourable components of its analogs, PLP inhibits A C C better than any of the analogs tested. Individually, the phosphate, aldehyde and ring all contribute to the inhibitory characteristics of PLP. The most important moiety appears to be the aldehyde at position 4', since removal of this component substantially reduces the ability of these compounds to inhibit A C C . The phosphate group at position 5' was not as potent as the aldehyde group. Minimally, a benzene or pyridine ring with two carboxyl groups is rather effective as an A C C inhibitor. A pyridine ring with only one carboxyl group is not effective, but in general, compounds containing a pyridine ring showed stronger inhibition of A C C than compounds with a benzene ring. Compounds containing two carboxyl groups are probably better at binding to A C C than the combination of phosphate and the ring hydroxyl of PLP, but direct testing of this hypothesis requires an additional PLP analog to be synthesized. The unusual compound A C A gives very high A C C inhibition at 1 mM, although little structural similarity between A C A and PLP makes A C A ' s method of A C C inhibition a little unclear. Conceivably, the carbonyl group could be effective binding determinants even i f esterified. 109 CHAPTER FIVE CONCLUSIONS & FUTURE GOALS Previous work has shown that PLP can bind to allosteric citrate binding sites on phosphofructokinase [69-71], and that it can inhibit both PFK and citrate transport proteins [72-75]. Using this information, it was hypothesized that PLP might also influence A C C activity by binding to the allosteric citrate binding site. The information presented in this thesis suggests that PLP is a potent inhibitor of acetyl-CoA carboxylase, with an IC50 around 225 p M . On balance, based on the evidence obtained thus far, it is likely that PLP does indeed bind to the citrate binding site, but further work will be required to establish this more definitely. Experiments aimed at using citrate as a competitive inhibitor of PLP, notably by adding citrate first to pre-activate A C C , did show substantial protection against PLP, especially when PLP concentrations were below 2 m M and citrate was used at higher concentrations (20 mM). A more extensive dose-response matrix (varying both citrate and PLP concentrations) is possibly warranted. Tentatively, then, given that citrate treatment of liver A C C preparations affords some protection against PLP, it is concluded that PLP is indeed binding to the citrate binding site on A C C . Why PLP should have such strong affinity for this site is likely due to several of its structural features. Since citrate is highly negatively charged, it's binding site should contain positively charge amino acid residues, at least one of which could be a lysine, such as in phosphofructokinase [69]. Since the aldehyde on PLP can form a Schiff base with the e-amino group of a lysine, and the phosphate and hydroxyl groups can mimic the negatively charged carboxylic acids on citrate, PLP should be able to (a) bind in a stable complex and (b) 110 form a temporary covalent linkage with A C C through the Schiff base. It is also possible that the pyridine ring structure can interact with aromatic amino acid side chains adjacent or in the citrate binding site, adding additional stabilization not seen with citrate. An interesting result reported previously found that although citrate protected PFK against inhibition by PLP, it did not significantly protect the enzyme from PLP incorporation. This was speculated to be a result of either many partial interactions at various e-amino groups, or a reaction at a particular group made more reactive upon citrate binding to PFK [70]. Whether such a situation is occurring in A C C is currently unknown, but would be important to investigate. This could be achieved in a number of ways, but the simplest is to have sufficient enzyme to monitor the incorporation of PLP on A C C by reduction with borohydride. The amount of PLP incorporated could then be determined by spectrophotometric analysis of the change in absorbance at 325 nm, characteristic of the reduced Schiff base, or by using tritiated borohydride and measuring the counts incorporated. This second approach has been attempted, although initial results were inconclusive, mainly due to imprecision in the procedure, and have thus not been included in this thesis. More recent attempts using tetrameric avidin agarose beads have been more successful, but are still not refined enough for conclusive results, as will be discussed below. Initial tritium incorporation suggested a large excess of PLP binding to A C C , but as mentioned, these results are not reliable, and more recent results show far less tritium incorporation, indicating much less PLP is actually binding to A C C . Typical results reported previously show about a 1 or 2:1 PLP:enzyme ratio [70], although one report found a huge PLP excess of up to 40:1 of PLP:PFK[71]. I l l Use of structural analogs has pointed towards the aldehyde of PLP being crucial for A C C inhibition. The phosphate and the negative charges are also important. An interesting result not reported in this thesis is the observation that treating A C C with PLP followed by either 3,4- or 3,5-pyridinedicarboxylic acid does not produce the same level of A C C inhibition i f the enzyme is treated first with the analog and then with PLP. This suggests that the analog binding first is somehow altering the structure of A C C sufficiently to promote additional PLP inhibition. This also suggests that the hypothesis that citrate alters the conformation of PFK such that it can bind more PLP may also hold true in A C C , although in this case the analog may promote both more PLP binding and greater PLP inhibition of A C C . This will again require further tests with promising analogs and PLP, as well as measurements of PLP incorporation. As mentioned, PLP has been successfully used to determine the sequence of the citrate binding site in PFK, as well as active sites in other proteins. This is a key goal for studies with A C C as well. As shown in Figure 3.16, PLP can be covalently linked to A C C , using sodium borohydride reduction. Refinements in the procedure for determining PLP incorporation are still ongoing. Knowing this, we are using a similar approach to produce labelled peptides. Following the procedure detailed in Figure 3.16, instead of digesting the beads in Sample Loading Buffer, A C C is instead digested with trypsin and the peptides produced are analyzed by spectrophotometry for a reduced Schiff base signal at 325 nm. The spectrophotometric analysis has not been successful, although this was due to technical problems with a batch of beads. These procedures will be refined in the near future. Upon successful spectrophotometric determination of a reduced Schiff base signal in the peptide mix, these will be purified on a reverse-phase HPLC column and monitored for absorbance at 112 325 nm. The isolated peptide(s) can then hopefully be sequenced using mass spectrometry. Ideally, one peptide will be produced, and upon sequencing will yield the identity of several of the residues in the citrate binding site. Although three-dimensional structures of mammalian A C C are not yet known, primary sequence homology between mammalian and prokaryotic forms with known structures could enable identification of additional residues in the citrate binding site, and help end the mystery of where citrate binds on A C C , and allow new insight as to how this binding causes shifts in A C C activity and polymerization state in mammalian A C C . So even i f PLP is not a relevant endogenous A C C regulator (additional work will be needed to observe the effects of PLP on in vivo lipid synthesis), it could prove to be a useful tool in determining the location of citrate binding on A C C . At the very least it can be used as a way to study additional kinetic and polymerization properties of A C C . For example, questions regarding the polymerization state of PLP-labelled A C C still need to be answered, as this would be very useful in coming to understand i f PLP can reverse the polymerization effects of citrate, and thus lower A C C activity. Such work could be accomplished using both size exclusion chromatography and sucrose density centrifugation. These experiments are currently ongoing. 113 REFERENCES 1. Mokdad, A .H . , Ford, E.S., Bowman, B.A. , Dietz, W.H., Vinicor, F., Bales, V.S., Marks, J.S., Prevalence of obesity, diabetes, and obesity-related health risk factors, 2001. J A M A , 2003. 289(1): p. 76-79. 2. Hunter, S., Garvey, WT, Insulin action and insulin resistance: diseases involving defects in insulin receptors, signal transduction, and the glucose transport effector system. American Journal of Medicine, 1998. 105(4): p. 331-345. 3. Saltiel, A . , Kahn, CR, Insulin signalling and the regulation of glucose and lipid metabolism. Nature, 2001. 414(December): p. 799-806. 4. Saltiel, A. , Pessin, JE, Insulin signalling pathways in time and space. Trends in Cell Biology, 2002. 12(2): p. 65-71. 5. Denton, R .M. , Midgley, P.J., Rutter, G.A., Thomas, A.P., McCormack, J.G., Studies into the mechanism whereby insulin activates pyruvate dehydrogenase complex in adipose tissue. Ann. N . Y . Acad. Sci., 1989. 573: p. 285-296. 6. Moss, J., Lane, M.D. , Acetyl coenzyme A carboxylase. IV. Biotinylprosthetic group-independent malonyl coenzyme A decarboxylation and carboxyl transfer: generalization to other biotin enzymes. J. Biol. Chem., 1972. 247(16): p. 4952-4959. 7. Waldrop, G.L., Rayment, I., Holden, H.M. , Three-dimensional structure of the biotin carboxylase subunit of acetyl-CoA carboxylase. Biochemistry, 1994. 33(34): p. 10249-10256. 8. Roberts, E.L., Shu, N . , Howard, M.J. , Broadhurst, R.W., Chapman-Smith, A. , Wallace, J.C., Morris, T., Cronan, J.E. Jr., Perham, R.N., Solution structures of apo and holo biotinyl domains from acetyl coenzyme A carboxylase of Escherichia coli determined by triple-resonance nuclear magnetic resonance spectroscopy. Biochemistry, 1999. 38(16): p. 5045-5053. 9. Al-Feel, W., Chirala, S.S., Wakil, S.J., Cloning of the yeast FAS3 gene and primary structure of yeast acetyl-CoA carboxylase. Proc. Natl. Acad. Sci. U.S.A., 1992. 89(10): p. 4534-4538. 10. Zhang, H. , Yang, Z., Shen, Y . , Tong, L., Crystal structure of the carboxyltransferase domain of acetyl-coenzyme A carboxylase. Science, 2003. 299(5615): p. 2064-2067. 11. Abu-Elheiga, L., Almarza-Ortega, D.B., Baldini, A. , Wakil, S.J., Human acetyl-CoA carboxylase 2. Molecular cloning, characterization, chromosomal mapping, and evidence for two isoforms. J. Biol. Chem., 1997. 272(16): p. 10669-10677. 114 12. Thampy, K.G. , Formation of malonyl coenzyme A in rat heart. Identification and purification of an isozyme of A carboxylase from rat heart. J. Biol. Chem., 1989. 264(30): p. 17631-17634. 13. Bianchi, A. , Evans, J.L., Iverson, A.J. , Nordlund, A.C. , Watts, T.D., Witters, L.A. , Identification of an isozymic form of acetyl-CoA carboxylase. J. Biol. Chem., 1990. 265(3): p. 1502-1509. 14. Abu-Elheiga L., B.W.R., Zhong L., Chirala S.S., Woldegiorgis G., Wakil S.J., The subcellular localization of acetyl-CoA carboxylase 2. Proc. Natl. Acad. Sci. U.S.A., 2000. 97(4): p. 1444-1449. 15. Hasslacher, M . , Ivessa, A.S., Paltauf, F., Kohlwein, S.D., Acetyl-CoA carboxylase from yeast is an essential enzyme and is regulated by factors that control phospholipid metabolism. J. Biol. Chem., 1993. 268(15): p. 10946-10952. 16. Abu-Elheiga L., M . M . M . , Abo-Hashema K.A . , Wakil S.J., Continuous fatty acid oxidation and reduced fat storage in mice lacking acetyl-CoA carboxylase 2. Science, 2001. 291(5513): p. 2613-2616. 17. Winder, W.W., MacLean, P.S., Lucas, J.C., Fernley, J.E., Trumble, G.E., Effect of fasting and refeeding on acetyl-CoA carboxylase in rat hindlimb muscle. J. Appl. Physiol., 1995. 78(2): p. 578-582. 18. Lopaschuk, G.D., Gamble, J., The 1993 Merck Frosst Award. Acetyl-CoA carboxylase: an important regulator of fatty acid oxidation in the heart. Canadian Journal of Physiology and Pharmacology, 1994. 72: p. 1101-1109. 19. Belke, D.D., Wang, L.C., Lopaschuk, G.D., Acetyl-CoA carboxylase control of fatty acid oxidation in hearts from hibernating Richardson's ground squirrels. Biochim. Biophys. Acta, 1998. 1391(1): p. 25-36. 20. Iverson, A.J . , Bianchi, A. , Nordlund, A.C. , Witters, L .A. , Immunological analysis of acetyl-CoA carboxylase mass, tissue distribution and subunit composition. Biochem. J., 1990. 269(2): p. 365-371. 21. Allmann, D.W., Hubbard, D.D., Gibson, D .M. , Fatty acid synthesis during fat free refeeding of starved rats. J. Lipid Res., 1965. 6: p. 63-74. 22. Girard, J., Perdereau, D., Foufelle, F., Prip-Buus, C , Ferre, P., Regulation oflipogenic enzyme gene expression by nutrients and hormones. FASEB J., 1994. 8(1): p. 36-42. 23. Iritani, N . , Nutritional and hormonal regulation of lipogenic-enzyme gene expression in rat liver. Eur. J. Biochem., 1992. 205(2): p. 433-442. 24. Lopez-Casillas, F., Kim, K . H . , Heterogeneity at the 5' end of rat acetyl-coenzyme A carboxylase mRNA. Lipogenic conditions enhance synthesis of a unique mRNA in liver. J. Biol. Chem., 1989. 264(13): p. 7176-7184. 115 25. Magana, M . M . , et al., Sterol regulation of acetyl coenzyme A carboxylase promoter requires two interdependent binding sites for sterol regulatory element binding proteins. J Lipid Res, 1997. 38(8): p. 1630-8. 26. Barber, M.C. , Travers, M.T., Elucidation of a promoter activity that directs the expression of acetyl-CoA carboxylase alpha with an alternative N-terminus in a tissue-restricted fashion. Biochem. J., 1998. 333(1): p. 17-25. 27. Mao, J., Molenaar, A . J., Wheeler, T. T., Seyfert, H. M . , STAT5 binding contributes to lactational stimulation of promoter III expressing the bovine acetyl-CoA carboxylase alpha-encoding gene in the mammary gland. J Mol Endocrinol, 2002. 29(1): p. 73-88. 28. Lee, J.J., et al., Cloning of human acetyl-CoA carboxylase beta promoter and its regulation by muscle regulatory factors. J Biol Chem, 2001. 276(4): p. 2576-85. 29. Winder, W.W., Hardie, D.G., Inactivation of acetyl-CoA carboxylase and activation of AMP-activatedprotein kinase in muscle during exercise. Am. J. Physiol., 1996. 272(2 Pt. 1): p. E299-304. 30. Martin, D.B., Vagelos, P.R., The mechanism of tricarboxylate acid cycle regulation of fatty acid synthesis. J. Biol. Chem., 1962. 237: p. 1787-1972. 31. Trumble, G.E., Smith, M.A. , Winder, W.W., Purification and characterization of rat skeletal muscle acetyl-CoA carboxylase. Eur. J. Biochem., 1995. 231(1): p. 192-198. 32. Moss, J., Lane, M.D. , Acetyl coenzyme A carboxylase. 3. Further studies on the relation of catalytic activity to polymeric state. J. Biol. Chem., 1972. 247(16): p. 4944-4951. 33. Boone, A . N . , Chan, Andy, Kulpa, Jerzy E., Brownsey, Roger W., Bimodal Activation of Acetyl-CoA Carboxylase by Glutamate. J. Biol. Chem., 2000. 275(15): p. 10819-10825. 34. Moule, S.K., Edgell, N.J., Borthwick, A.C. , Denton, R .M. , Coenzyme A is a potent inhibitor of acetyl-CoA carboxylase from rat epididymal fat-pads. Biochem. J., 1992. 283(1): p. 35-38. 35. Goodridge, A .G . , Regulation of the activity of acetyl coenzyme A carboxylase by palmitoyl coenzyme A and citrate. J. Biol. Chem., 1972. 247(21): p. 6946-6952. 36. Allred, J.B., Roehrig, K .L . , Inhibition of rat liver acetyl CoA carboxylase by chloride. J. Lipid Res., 1980. 21(4): p. 488-491. 37. Levert, K . L . , Waldrop, G.L., Stephens, J.M., A biotin analog inhibits acetyl-CoA carboxylase activity and adipogenesis. J. Biol. Chem., 2002. 277(19): p. 16347-16350. 116 38. Endo, A. , Takeshima, H. , Kuwabara, K. , Acetyl CoA carboxylase inhibitors from the fungus Gongronella butleri. J. Antibiot. (Tokyo), 1985. 38(4): p. 599-604. 39. Konishi, T., Sasaki, Y . , Compartmentalization of two forms of acetyl-CoA carboxylase in plants and the origin of their tolerance toward herbicides. Proc. Natl. Acad. Sci. U.S.A., 1994. 91(9): p. 3598-3601. 40. Turner, J.A., Pernich, D.J., Origin of enantiomeric selectivity in the aryloxyphenoxypropionic acid class of herbicidal acetyl coenzyme A carboxylase (ACCase) inhibitors. J. Agric. Food Chem., 2002. 50(16): p. 4554-4566. 41. Kemal, C. and J.E. Casida, Coenzyme A esters of 2-aryloxyphenoxypropionate herbicides and 2-arylpropionate antiinflammatory drugs are potent and stereoselective inhibitors of rat liver acetyl-CoA carboxylase. Life Sci, 1992. 50(7): p. 533-40. 42. Gregolin, C , Ryder, E., Warner, R.C., Kleinschmidt, A .K . , Lane, M.D. , Liver acetyl-CoA carboxylase: the dissociation-reassociation process and its relation to catalytic activity. Proc. Natl. Acad. Sci. U.S.A., 1966. 56: p. 1751-1758. 43. Boone, A. , Regulation of acetyl-CoA carboxylase-2., in PhD. Thesis, University of British Columbia. 2000. 44. Winz R., H.D., Aebersold R., Brownsey R.W., Unique structural features and differential phosphorylation of the 280-kDa component (isozyme) of rat liver acetyl-CoA carboxylase. J. Biol. Chem., 1994. 269(20): p. 14438-14445. 45. Quayle, K . A . , Denton, R .M. , Brownsey, R.W., Evidence for a protein regulator from rat liver which activates acetyl-CoA carboxylase. Biochem. J., 1993. 292: p. 75-84. 46. Heesom, K.J . , Moule, S.K., Denton, R .M. , Purification and characterisation of an insulin-stimulated protein-serine kinase which phosphorylates acetyl-CoA carboxylase. FEBS Lett., 1998. 422: p. 43-46. 47. Ha, J., Daniel, Samira, Broyles, Steven S., Kim, Ki-Han, Critical Phosphorylation Sites for Acetyl-CoA Carboxylase Activity. J. Biol. Chem., 1994. 269(35): p. 22162-22168. 48. Hardie, D.G., Carling, David, Ferrari, Stefano, Guy, Paul S., Aitken, Alistair, Characterization of the phosphorylation of rat mammary ATP-citrate lyase and acetyl-CoA carboxylase by Ca2+ and calmodulin-dependent multiprotein kinase and Ca2+ andphospholipid-dependentprotein kinase. Eur. J. Biochem., 1986. 157: p. 553-561. 49. Lopez-Casillas, F., Bai, D.H., Luo, X .C . , Kong, I.S., Hermodson, M.A. , Kim, K . H . , Structure of the coding sequence and primary amino acid sequence of acetyl-coenzyme A carboxylase. Proc. Natl. Acad. Sci. U.S.A., 1988. 85(16): p. 5784-5788. 117 50. Haystead, T.A., Campbell, D.G., Hardie, D.G., Analysis of sites phosphorylated on acetyl-CoA carboxylase in response to insulin in isolated adipocytes. Comparison with sites phosphorylated by casein kinase-2 and the calmodulin-dependent multiprotein kinase. Eur. J. Biochem., 1988. 175(2): p. 347-354. 51. Haystead, T.A., Hardie, D.G., Insulin andphorbol ester stimulate phosphorylation of acetyl-CoA carboxylase at similar sites in isolated adipocytes. Lack of correspondence with sites phosphorylated on the purified enzyme by protein kinase C. Eur. J. Biochem., 1988.175(2): p. 339-345. 52. Munday, M.R., Campbell, D.G., Carling, D., Hardie, D.G., Identification by amino acid sequencing of three major regulatory phosphorylation sites on rat acetyl-CoA carboxylase. Eur. J. Biochem., 1988. 175(2): p. 331-338. 53. Brownsey, R.W., Denton, R .M. , Evidence that insulin activates fat-cell acetyl-CoA carboxylase by increased phosphorylation at a specific site. Biochem. J., 1982. 202: p. 77-86. 54. Collins, S.E., An Investigation of the Potential Role of PKC Isoforms in the Regulation of Acetyl-CoA Carboxylase-1. MSc. Thesis, University of British Columbia, 2001. 55. Bor, M.V. , Refsum, H. , Bisp, M.R., Bleie, O., Schneede, J., Nordrehaug, J.E., Ueland, P.M. , Nygard, O.K., Nexo, E., Plasma vitamin B6 vitamers before and after oral vitamin B6 treatment: a randomized placebo-controlled study. Clin. Chem., 2003. 49(1): p. 155-161. 56. Merrill, A.H.J. , Henderson, J.M., Vitamin B6 metabolism by human liver. Ann. N . Y . Acad. Sci., 1990. 585(1): p. 110-117. 57. Nygard, O., Nordrehaug, J.E., Refsum, H. , Ueland, P.M., Farstad, M . , Vollset, S.E., Plasma homocysteine levels and mortality in patients with coronary artery disease. N . Engl. J. Med., 1997. 337(4): p. 230-236. 58. D'Angelo, A. , Mazzola, G., Crippa, L., Fermo, I., D'Angelo, S.V., Hyperhomocysteinemia and venous thromboembolic disease. Haematologica, 1997. 82(2): p. 211-219. 59. Merrill, A.H.J. , Henderson, J.M., Diseases associated with defects in vitamin B6 metabolism or utilization. Annu. Rev. Nutr., 1987. 7: p. 137-156. 60. Christen, P., Mehta, P.K., From cofactor to enzymes. The molecular evolution of pyridoxal-S'-phosphate-dependent enzymes. Chem. Rec, 2001. 1(6): p. 436-447. 61. Kimura, M . , Kanehira, K. , Yokoi, K. , Highly sensitive and simple liquid chromatographic determination in plasma ofB6 vitamers, especially pyridoxal 5'-phosphate. J Chromatogr A, 1996. 722(1-2): p. 296-301. 118 62. Han, Q., Xu, M . , Tang, L., Tan, X . , Tan, X . , Tan, Y . , Hoffman, R .M. , Homogeneous, nonradioactive, enzymatic assay for plasma pyridoxal 5-phosphate. Clin. Chem., 2002. 48(9): p. 1560-1564. 63. Tang, K . H . , Harms, A. , Frey, P. A. , Identification of a novel pyridoxal 5'-phosphate binding site in adenosylcobalamin-dependent lysine 5,6-aminomutase from Porphyromonas gingivalis. Biochemistry, 2002. 41(27): p. 8767-8776. 64. Pandey, A. , Katiyar, S.S., Essential lysine residue in glutathione reductase: chemical modification by pyridoxal 5'-phosphate. Biochem. Mol . Biol. Int., 1995. 36(2): p. 347-354. 65. Kery, V. , Poneleit, L., Meyer, J.D., Manning, M.C. , Kraus, J.P., Binding of pyridoxal 5'-phosphate to the heme protein human cystathionine beta-synthase. Biochemistry, 1999. 38(9): p. 2716-2724. 66. Hall, J.A., Maloney, P.C., Pyridoxal 5-phosphate inhibition of substrate selectivity mutants ofUhpT, the sugar 6-phosphate carrier of Escherichia coli. J. Bacteriol., 2002. 184(13): p. 3756-3758. 67. Chen, D., Frey, P.A., Identification of lysine 346 as a functionally important residue for pyridoxal 5'-phosphate binding and catalysis in lysine 2, 3-aminomutase from Bacillus subtilis. Biochemistry, 2001. 40(2): p. 596-602. 68. Huang, D.Y. , Ichikawa, Y . , Identification of essential lysyl and cysteinyl residues, and the amino acid sequence at the substrate-binding site of retinal oxidase. Biochim. Biophys. Acta, 1995. 1243(3): p. 431-436. 69. Kemp, R.G., Fox, Ronald W., Latshaw, Steven P., Amino Acid Sequence at the Citrate Allosteric Site of Rabbit Muscle Phophofructokinase. Biochemistry, 1987. 26: p. 3443-3446. 70. Colombo, G., Kemp, R.G., Specific modification of an effector binding site of phosphofructokinase by pyridoxal phosphate. Biochemistry, 1976. 15(8): p. 1774-1780. 71. Uyeda, K. , Reaction of phosphofructokinase with maleic anhydride, succinic anhydride, and pyridoxal 5'-phosphate. Biochemistry, 1969. 8(6): p. 2366-2373. 72. Gremse, D., A. , Dean, Brenda, Kaplan, Ronald S., Effect of Pyridoxal 5'-Phosphate on the Function of the Purified Mitochondrial Tricarboxylate Transport Protein. Archives of Biochemistry and Biophysics, 1995. 316(1): p. 215-219. 73. Natuzzi D., D.L., Stipani V. , Cappello A.R., Miniero D.V., Capobianco L., Stipani I., Inactivation of the reconstituted oxoglutarate carrier from bovine heart mitochondria by pyridoxal 5'-phosphate. Journal of Bioenergetics and Biomembranes, 1999. 31(6): p.535-541. 119 74. Genchi, G., Spagnoletta, A. , De Santis, A. , Stefanizzi, L., Palmieri, F., Purification and characterization of the reconstitutively active citrate carrier from maize mitochondria. Plant Physiol., 1999! 120(3): p. 841-847. 75. Planning, I., Baumgarten, K., Schott, K. , Heldt, H.W., Oxaloacetate transport into plant mitochondria. Plant Physiol., 1999. 119(3): p. 1025-1032. 76. Blaner, W.S., Churchich, J., Succinic Semialdehyde Dehydrogenase. The Journal of Biological Chemistry, 1979. 254(6): p. 1794-1798. 77. Hong, J.W., Cho, S.W., Yoo, J.S., Yoo, B.K. , Lee, K.S., Choi, S.Y., Modulation of the catalytic activity of brain succinic semialdehyde reductase by reaction with pyridoxal 5'-phosphate. Eur. J. Biochem., 1997. 247(1): p. 274-279. 78. Voet, D., Voet, J.G., Pratt, C.W., Fundamentals of Biochemistry. 1999, Toronto: John Wiley & Sons. Inc. 79. Camp, V . M . , Chipponi, J., Faraj, B.A. , Radioenzymatic assay for direct measurement of plasma pyridoxal 5'-phosphate. Clin. Chem., 1983. 29(4): p. 642-644. 80. Halestrap, A.P., Denton, R .M. , Insulin and the regulation of adipose tissue acetyl-coenzyme A carboxylase. Biochem. J., 1973. 132(3): p. 509-517. 81. Laemmli, U.K. , Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature, 1970. 227: p. 680-685. 82. Cornell, N.W., Zuurendonk, P. F., Kerich, M . J., Straight, C. B., Selective inhibition of alanine aminotransferase and aspartate aminotransferase in rat hepatocytes. Biochem J, 1984. 220(3): p. 707-16. 83. Moffatt, J., Hashimoto, M . , Kojima, A. , Kennedy, D.O., Murakami, A. , Koshimizu, K., Ohigashi, H. , Matsui-Yuasa, I., Apoptosis induced by I'-acetoxychavicol acetate in Ehrlich ascites tumor cells is associated with modulation of polyamine metabolism and caspase-3 activation. Carcinogenesis, 2000. 21(12): p. 2151-2157. 120 

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.831.1-0090953/manifest

Comment

Related Items