University of Calgary PRISM: University of Calgary's Digital Repository
Graduate Studies Legacy Theses
2000 Biochemical and functional interactions of methyltetrahydrofolate and homocysteine in vascular disease
Hyndman, Matthew Eric
Hyndman, M. E. (2000). Biochemical and functional interactions of methyltetrahydrofolate and homocysteine in vascular disease (Unpublished doctoral thesis). University of Calgary, Calgary, AB. doi:10.11575/PRISM/17191 http://hdl.handle.net/1880/40710 doctoral thesis
University of Calgary graduate students retain copyright ownership and moral rights for their thesis. You may use this material in any way that is permitted by the Copyright Act or through licensing that has been assigned to the document. For uses that are not allowable under copyright legislation or licensing, you are required to seek permission. Downloaded from PRISM: https://prism.ucalgary.ca NOTE TO USERS
Page(s) not included in the original manuscript and are unavailable from the author or university. The manuscript was microfilmed as received.
237 through 246
This reproduction is the best copy available.
UNIVERSITY OF CALGARY
Biochemical and Functional Interactions of Methyltetrahydrofolate and
Homocysteine in Vascular Disease
BY
Matthew Eric Hyndman
A THESIS SUBMITTED TO THE FACULTY OF GRADUATE STUDIES
IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE
DEGREE OF DOCTOR OF PHILOSOPHY
DEPARTMENT OF MEDICAL SCIENCE:
CALGARY,ALBERTA
NOVEMBER, 2000
8 Matthew Eric Hyndman 2000 Natianal Cl'brary Biblbthdy nationale 161 ofCanada du Cana a Acquisitions and Acquisitions et Biblibgraphin: Services services bibliiraphiques 395 WdhgWn StmN 395. rw WoUington OcEwwON KIA- CMawaON K1AW CMadr Canada
The author has granted a non- L'auteur a accorde une licence non exclusive licence allowing the exclusive permettaut a la National Ll'brary of Canada to Bibliotheque nationale du Canada de reproduce, loan, distn'bute or sell reproduire, prGter, distribuer ou copies of this thesis in microform, vendre des copies de cette these sous paper or electronic formats. la forme de microfiche/film, de reproduction sur papier ou sur format electronique.
The author retains ownership of the L'auteur conserve la propriete du copyright in this thesis. Neither the droit d'autew qui protege cette these. thesis nor substantial extracts fiom it Ni la these ni des extraits substantiels may be printed or otherwise de celle-ci ne doivent etre imprimes reproduced without the author's ou autrement reproduits sans son permission. autorisation. UNIVERSITY OF CALGARY
FACULTY OF GRADUATE STUDIES
The undersigned certify that they have read, and recommend to the Faculty of Graduate
Studies for acceptance, a dissertation entitled "Biochemical and Functional Interactions of Methyltetrahydrt~folateand Hornocysteine in Vascular Disease" submitted by Matthew
Eric Hyndman in partial fulfillment of the requirements for the degree of Doctor of
Philosophy.
supervisor, Dr. owi id Panons Department of wical Science
arfRespiratory Sciences
Departme nt of Biochemistry and Molecular Biology
Intefi~ai/External~xamifier Dr. Roy Gravel, Department of Cell Biology and Anatomy
External ~xaxpineh Dr. Jacques Gt nest, Department of Medicine, McGill University December 15,2000 ABSrnCT
In the last decade, hyperhomocysteinemia has become a well-established risk factor for vascular disease. Moderately elevated levels of the amino acid are associated with fatal and nonfatal cardiovascular disease. Although it is dear that elevations in homocysteine arise due to insufficiencies in the cofactors or enzymes involved in methionine metabolism, the mechanisms leading to the accelerated atherosclerosis in patients with hyperhomocysteinemia remains obscure.
The effects of high dose vitamin therapy were investigated in patients with renal failure. This study was expanded and the treatment effects were examined on the vitamin Bl2 dependent variable methylmalonic acid and in patients stratified by the common MTHFR 677 C+T genotypes. High dose vitamin therapy while sigruficantly lowering both homocysteine and MMA did not normalize levels irrespective of the MTHFR genotype.
The interactions between the MTHFR 677 C+T, MS 2756 A+G, and the
MTRR 66 A+G genotypes and methionine metabolism in a number of patients
with cardiovascular disease. The discovery of a highly signhcant decreased
incidence of recurrent cardiovascular events in those heterozygous for the MS G
dele prompted a more detailed investigation into the role of folate and
homocysteine in endothelial dysfunction and more specifically the impact of
folate and endothelial derived nitric oxide. Interestingly, folate attenuated oxidative stress and facilitated the release of nitric oxide both directly in vitro and indirectly in vivo.
The relationship between folate and nitric oxide prompted studies associating genetic polymorphism in eNOS and folate metabolism and their subsequent effect on blood pressure. We found a highly sigruhcant association between a very common point mutation in eNOS and blood pressure.
Furthermore, the MS 2756 G allele may attenuate the effects of decreased nitric oxide production induced by the point mutation in eNOS. Lastly, using an animal model of hypertension it was determined that treatment with methylated folates improved endothelial dependent nitric oxide release in the small resistance vessels,
Taken together, these studies provide new evidence that folate is an important mediator of vascular function. In addition, the studies herin suggests that one of the mechanisms leading to the association between hyperhomocysteinemia and vascular disease may be attributed to folate status. I would like to thank my supervisor Dr. Howard Parsons, who has given me the opportunity to pursue a career in science. I have greatly appreciated your support and friendhess during my time in the lab. Thank you.
In addition I would also like to extend a sincere thanks to my committee members, Dr. Floyd Snyder, Dr. Wayne Warnica and Dr. Henry Duff. These gentlemen always provided direction, support and guidance throughout my degree.
A special thanks is needed for the members of Dr. F. Snyder's and Dr. P.
Bridge's lab. Although, I still have many more questions for the lot of you thank you for answering all of my previous questions.
Three other individuals have also contributed to my studies and I have been lucky to interact with them. Maria Tesanovic, Subodh Verma and Todd
Anderson are three of the kindest people that I have come to know. Their
inspiration and dedication to teaching is unparalleled. Thank you so much for
all of you patience, guidance and friendship.
I must also thank Humberto Jijon and Ruth Malus. I have thoroughly
enjoyed working with the two of you. I will miss seeing your smiling faces each
day.
I would like to thank my friends Rebecca Nelson, Brent Harris and Steve
Feidc for their support and encouragement. I have been lucky to have such great siblings, The Itsy bitsy (Laura), Tiny thing (Jen)and my brother Scott. Thank you for your support during the last twenty or so years.
Findy, I would like to thank my parents, without whom I would quite literally not be here, which would have made this thesis much harder to write.
Your endless and continuing support from day one (especially day one) is appreciated and is an excellent example that I will aspire to follow.
The Heart and Stroke Foundation of Alberta and The Alberta Kidney
Foundation funded this thesis. Addition support was provided by William A.
Bell. TABLE OF CONTENTS
.. Approval Page ll Abstract iii Acknowledgements v Table of Contents vi List of Tables List of Figures Abbreviations
CIUPTER ONE: INTRODUCTION 1.I Homocysteine metabolism 1.2 Folate metabolism 1.3 Genetic and nutritional contributions to hyperhomocysteinemia 1.4 Homocysteine and atherosclerosis 1.5 Endothelial function 1.6 Tetrahydrobiopterin depletion 1.7 Folic acid and nitric oxide
CHAPTER TWO: ORAL VITAMIN B12 AND HIGH-DOSEFOLIC ACID IN HEMODIALYSIS PATIENTS WITH HYPERHOMCKYST(E)lNEMIA: A RANDOMIZED CLINICAL TRIAL 2.1 Introduction 38 2.2 Methods 40 2.3 Results 45 2.4 Discussion 51 2.5 Conclusions 55
CHAPTER THREE: PHARMACOLOGICAL SUPPLEMENTATION WITH VITAMIN BIZDECREASES BUT DOES NOT RETURN HOMOCYSTEINEE AND MMA TO CONTROL LEVELS IN PATIENTS WITH END STAGE RENAL DISEASE: A POSSIBLE WNK WITH GLYCINE METABOLISM
3.1 Introduction 3.2 Methods 3.3 Results 3.4 Discussion 3.5 Conclusion
vii CHAPI'ER FOUR: EFFECTS OF COMMON ONE-CARBON METABOLISM POLYMORPHISMS ON HOMOCYSTEINE, RBC FOLATE, VITAMIN €312 AND METHYLMALONIC ACID LEWIS
4.1 Introduction 4.2 Methods 4.3 Results 4.4 Discussion
CHAM'ER FIVE: EFFECT OF HETEROZYGOSITY FOR THE METHIONINE SYNTHASE 2756A-G MUTATION ON THE RISK FOR RECURRENT CARDIOVASCULAR EVENTS 98
CHAPTER SIX: THE EFFECTS OF COMMON OW-CARBON METABOLISM POLYMORPHISMS ON ENDO-L FUNCTION
6.1 Introduction 6.2 Methods 6.3 Results 6.4 Discussion
CHAM'ER SEVEN: 5-METHk-L~ROFOLATE ATTENUATES SUPEROXIDE PRODUCTION AND AUGMENTS NITRIC OXIDE PRODUCTION IN MODELS OF HYPERHOMOCYSTEINEMIA AND INSULIN RESISTANCE
7.1 Introduction 7.2 Methods 7.3 Results 7.4 Discussion
CHAlTER EIGHT: FOLATE AND HYPERTENSION: GENETICS AND PHARMACOLOGICAL STUDIES
8.1 Introduction 8.2 Methods 8.3 Results 8.4 Discussion
CHAPTER NINE: SUMMARY 181
CHAPTERTEN: FUTURE STUDE 10.1 Targeting endothelial function and hyperhomocysteinemia in ESRD 188 10.2 Folate modulation of platelet activity 198
STUDY ACKNOWLEDGEMENTS
BIBLIOGRAPHY
APPENDIX A
APPENDIX B LIST OF TABLES
CHAPTER TWO: ORAL VITAMIN B12 AND HIGH-DOSE FOLIC ACID IN HEMODIALYSIS PATENTS WITH HYPERHOMOCYSTQIhIEMIA: A RANDOMIZED CLINICAL TRIAL
Table 2.1 Baseline characteristics 46 Table 2.2 Homocysteine, RBC folate and vitamin B12 levels 48 Table 2.3 Homocysteine, folic acid and vitamin B12 levels during phase III 50
CHAJ?TER THREE: PHARMACOLOGICAL SUPPLEMENTATION WITH VITAMIN B12 DECREASES BUT DOES NOT RETLTRN HOMOCYSTEINEE AND MMA TO CONTROL LEVELS IN PATIENTS WITH END STAGE RENAL DISEASE: A POSSIBLE LINK WITH GLYCINE METABOLISM
Table 3.1 Mean and standard deviations of the screening levels of plasma homocysteine 64 Table 3.2 MMA and homocysteine levels during phase I, phase II and phase 111 stratified by MTHFR 677C-Tgenotype 65
CHAPTER FOUR: EFFECTS OF COMMON ONE-CARBON METABOLISM POLYMORPHISMS ON HOMOCYSTEWE, RBC FOLATE, VITAMIN B12 AND METHYLMALOMC ACID LEVELS
Table 4.1 Median fasting, post-methionine load and load-fast hornocysteine concentrations for the CAD patients subdivided by their MTHFR 677C-T,MS 2756 A-G and the MTRR 66A-G genotypes 89 Table 4.2 Median serum BIZplasma MMA and RBC folate levels for the CAD patients, subdivided by their MTHEX 677C-T, MS 2756 A-G and the MTRR 66A-G genotypes 90
CHAMER FIVE: EFFECT OF HETEROZYGOSITY FOR THE METHIONINE SYNTHASE 2756A-G MUTATION ON THE RISK FOR RECURRENT CARDIOVASCULAR EVENTS
Table 5.1 Mantel-Haenszel log rank test for equality of vascular events between patients with the A/A and A/G genotypes of methionine synthase 100
CHAPTER SIX: THEi EFFECI'S OF COMMON ONE-CARBON METABOLISM POLYMORPHISMS ON ENDOTHEUAL FUNCTION Table 6.1 Summary of demographic data 109 Table 6.2 Genotype distribution for MTHFR 677 C-Tand the MS 2756 A-G Polymorphisms 110 Table 6.3 Plasma homocysteine, RBC folate and serum folate stratified by MTHFR 677 C-T genotype 111 Table 6.4 Plasma homocysteine, RBC folate and serum folate stratified by MS 2756 A-G genotype 111 Table6.5 Meanhomocysteine,RBCfolate,senun&~andserurnfolate stratified by MTHFR 677 C-T and MS 2756 A-G genotypes 112
CHAPTER EIGHT: FOLATE AND HYPERTENSION: GENETICS AND P-COLOGICAL STUDIES
Table 8.1 Distribution of eNOS T-786-Cand MS 2756 A-G polymorphism Table 8.2 Patient demographics Table 8.3 Distibution of eNOS T-786-Cin those with essential hypertension Table 8.4 Telemetry parameters in the experimental groups LIST OF FIGURES
CHAPTER ONE: INTRODUCTION
Figure 1.1 Metabolic pathway of methionine 5 Figure 1.2 Structure of 5-methyl tetrahydrofolate 9 Figure 1.3 Intracellular metabolism of folate 12 Figure 1.4 Representation of the enzymatic pathways for the synthesis of nitric oxide and tetrahydrobiopterin 29
CHAPTER TWO: ORAL VITAMIN B12 AND HIGH-DOSE FOLIC ACID IN HEMODIALYSIS PATIENTS WITH HYPERHOMOCYSTQINEMIA: A RANDOMIZED CLINlCAL TRIAL
Figure 2.1 Trial profile 42
CHAPTER THREE: PHARMACOLOGICAL SUPPLEMENTATION WITH VITAMIN B12 DECREASES BUT DOES NOT RETURN HOMOCYSTEINEE AND MMA TO CONTROL LEVELS IN PATIENTS WITH END STAGE RENAL DISEASE: A POSSIBLE LINK WlTH GLYmMETABOLISM
Figure 3.1 Relationship between glycine and MMA levels in ESRD patients Figure 3.2 Intracellular vitamin B12 metabolism
CHAPTER FOUR: EFFECTS OF COMMON ONE-CARBON METABOLISM POLYMORPHISMS ON HOMOCYSTEINE, RBC FOLATE, VITAMIN B12 AND METHYLMALONIC ACID LEVELS
Figure 4.1 The relationship between MMA and 812 in CAD patients 84 Figure 4.2 The relationship between MMA and homocysteine in CAD Patients 85 Figure 4.3 Log-log plot examining the relationship between MMA and BIZ 86 Figu~e4.4 Log-log plot examining the relationship between homocysteine and B12 87
CHAMER FIVE: EFFECT OF HE'EROZYGOSITY FOR THE METHIONLNE SY'NTHASE 2756A-C MUTATION ON THE RISK FOR RECURRENT CARDIOVASCULAR EVENTS
Figure 5.1 Proportion of patients without an MI, HF, CBP 101
xii CHAPTER SIX: THE EFFECTS OF COMMON ONE-CARBON METABOLISM POLYMORPHISMS ON ENDOTHELIAL FUNCTION
Figure 6.1 Distribution of plasma homocysteine levels stratified by MTHFR 677 C-T genotype 113 Figure 6.2 Distribution of RBC folate levels stratified by the MTHFR 677 C-Tgenotype 114 Figure 6.3 Distribution of semfolate stratified by the MTHFR 677 C-T genotype 115 Figure 6.4 Flow-mediated vasodilatation stratified by A/A or A/G genotypes for MS 2756 A-G polymorphisms 117
CHAPTER SEVEN: 5-METHYLTETRAHYDROFOLATE ATTENUATES SUPEROXIDE PRODUCTION AND AUGMENTS NITRIC OMDE PRODUCTION IN MODELS OF HYPERHOMOCYSTEINEMIA AND INSULIN RESISTANCE
Figure 7.1 EaHy96 endothelial cells stained for eNOS with and without the pep tide inhibitor 130 Figure 7.2 A) Normal untreated bovine aortic endothelid cells 131 B) Co-incubation of primary antibody with inhibitory pep tide C) Endothelid cells stimulated with cyclosporin (24 hrs) Figure 7.3 Superoxide production hnrecombinant eNOS 133 Figure 7.4 Mean 123-dihydrorhodamine induced fluorescence by BAEC Incubated with normal media, DM,DAHP + sepiapterin and 5-MTHF 134 Figure 7.5 A) Mean DHE fluorescence signal from BAEC cells incubated with normal media and DAHP 135 B) Mead DHE fluorescence signal from BAEC cells incubated with 5-MTHE: and DAHP Figure 7.6 Acetylcholine-induced nitrite production by EaHY96 endothelial cells Incubated with homocys teine, methionine, NAS, with and without 5-W 136 Figure 7.7 Docking results of 5M"IT-E in the active site of eNOS compared to The actual B& binding site determined from the crystal structure 138 Figure 7.8 Acute Bfi improved endothefial function in aorta from fructose- hypertensive rats 140 Figure 7.9 Acute IMTHF improves endothelial function from fructose- hypertensive rats 141 Figure 7.10 Effects of two-hour methionine loading on vascular responses to acetylcholine in normal and insulin-resistant rat arteries 142 Figure 7.11 Effects of ten-hour methionine loading on vascular responses to
*** Xlll acetylcholine in normal and insulin-resistant rat arteries in the presence and absence of 5-MTHF 143 Figure 7.12 Effects of a chronic methionine load in the presence and absence of long-term folinic acid treatment 144 Figure 7.13 Effects of a chronic methionine load followed by acute treatment with B& 145
CHAPI*ER EIGHT: FOLATE AND HYPERTENSION: GENETICS AND PHARMACOLOGICAL STUDIES
Figure 8.1 A) Diastolic blood pressure stratified by the eNOS T-786-C genotype 165 B) Systolic blood pressure stratified by the eNOS T-786-C genotype 167 Figure 8.2 Systolic and diastolic blood pressures in subjects with the MS 2756 A/A or G/Gand the eNOS786 C/C genotype 170 Figure 8.3 Representative hernodynamic tracing of for the SHR study l7l Figure 8.4 Chronic folinic acid treatment improves endothelial function in mesenteric arteries of SHR rats 172
xiv MTRR Methionhe synthase reductase MTHFR Methylene tetrahydrofolate reductase MS Me thionine synthase 5-MTHF 5-me thy ltetrahydro folate SAH Sadenosyl homocy steine DAHP Diaminohydroxypteridine ESRD End stage renal disease MMM Methylmalonyl CoA mutase MMA Methy lmalonic acid RBC Red blood cell BH4 Tetrahydrobiopterin eNOS hdothelial nitric oxide synthase NOS Inducible nitric oxide synthase nNOS Neuronal nitric oxide synthase DHR 123 Dihy drorhodarnine LDL Low density lipoprotein HDL High density lipoprotein CVD Cardiovascular disease NO Nitric oxide SMC Smooth muscle cell PLP Pyridoxal-5-p hospha te CPS Cy stathionine-P-synthase FBP Fola te binding protein RFC Reduced folate carrier FPGG Folypolyglutamate synthase m/c SHMT Mitchondrial/cytosolic serine hydroxymethyl transferase GCS Glycine cleavage system TC Transcobdamin MTX Methotrexate IF Intrinsic factor CAD Coronary artery disease Asm Artherosclerotic vascular disease FMN Flavine mononudeotide FAD Flavine adenine dinucleotide CHAPTER 1 -1.0 INTRODUCTION
The purpose of this chapter is to provide some familiarity with homocysteine and folate metabolism and how they are related to vascular disease. Secondly the clinical implications assodated with the genetic polymorphisms will be addressed.
Cardiovascular diseases (CVD) are the leading cause of death in the developed world accounting for approximately 40% of mortality in both men and womenl. Although the clinical pathologies associated with CVD usually present themselves later in life, atherosclerosis is a lifelong vascular inflammatory disease beginning in Mdhood2. Diabetes>=,smoking6, infectious microorganisms, elevated LDL7, hypertension~loand plasma homocysteinen are known risk factors associated with vascular disease. In the earliest stages of atherosclerosis, lesion formation is initiated by the activation of the local endothelial cells resulting in the recruitment and subsequent migration of leukocytes in the sub-endotheliuml2. This is accompanied by inaeased endothelial permeability, promoting the passage and accumulation of lipids in the sub-endothelial space. Activated macrophages, lymphocytes and endothelial cells aIl synthesize growth factors and cytokines, which contribute to migration and proliferation of smooth musde cells (SMC) in the vascular intirna.
Production of extracellular matrix proteins is inaeased and if unabated the proliferating cells and chronic inflammation leads to the production of an 3 atheromatous obtrusion often associated with a lipid-rich core surrounded by a fibrous cap 13;14
Disruption of plaques occurs in approximately 9%of normal healthy individuals without artery occlusion 5. Disruption is a natural side effect of aging and depends upon the balance of hernodynamic forces and plaque stability. Lipid and macrophage rich cores with Little matrix support are typically the most vulnerable. The majority of disruptions are clinically silent and unpredictable, however, increased sensitivity to thrombosis can trigger an
uncontrolled clotting cascade surrounding the atheroma resulting in occlusion of
the vessel and ischemia of the downsheam tissue 4. Alternatively, stenosis may
also be caused by platelet aggregation on the surface of an intact atheroma
occluding the artery.
The vascular endothelium is an important mediator affecting both the
long term and short-term outcome of an atherosclerotic plaque. The vascular
endothelium plays an important role in regulating inflammation and
maintaining homeostasis. By directing leukocyte migration and platelet
activation through a variety of adhesion molecules and paracrine secretions,
proper endotheiial function can moderate the progression and the outcome of an
atherosclerotic lesion3. In contrast, endothelial dysfunction can exacerbate lesion
formation. Because of its wide-ranging effectson the vasculature and circulating
cells, nitric oxide (NO) is one of the more important substances produced and
secreted by the endothelial cek3. Decreased production of NO has been 4 associated with numerous risk factors for vascular disease such as diabetes's, hypercholesterolemiaE, smokin@'6, hypertension10;17, and is a hallmark of endothelial dysfunctionls. The underlying pathology of insufficient NO production is probably a result of both reduced nitric oxide production and inaeased degradation. Reactive oxidative species (ROS) are capable of interacting with nitric oxide and inhibiting its antiatherogenic effects.
Homocysteine is an atherogenic amino acid known to produce ROSl8 and has been recently become the subject of intense investigation. Conversely, evidence exists suggesting that homocysteine may be a reflection of folate status and that a deficiency in folate contributes to the synthesis of ROS19m.
1.1 Homocvsteine Metabolism
Plasma homocysteine levels can be a useful tool aiding in the identification of genetic mutations and more recently for assessing risk for vascular disease. The vast majority of homocysteine exists as a disulfide either attached to protein or bound to other cirmlating thiol groups such as cysteine,
another homocysteine or glutathioine 21. In the cell the situation is reversed and
homocysteine exists primariiy in its reduced free suifhydral form. Little
information is available about the cellular transport of homocysteine. A
transporter spdicto the disulfide from of homocysteine likely mediates uptake
as the oxidized form predominates in the circulation21. On the contrary export
has been proposed to occur by a transporter specific to the reduced form of
homocysteine2l;n. Because of the certainty of protein bound homocysteine some 5 indirect cellular absorption of protein bound homocysteine must also account for intra cellular transport *I.
The total concentration of circulating plasma homocysteine is a reflection of the balance between cellular metabolism, catabolism and urinary exaetion.
Metabolically, a high level is therefore due to either increased production of homocysteine from methionine or alternatively decreased excretion and/ or catabolism to cysteine (Fig. 1.1) R
methionine A /
AdoHcy
5,lQ methy lene THF 4
cysteine AdoMe *&
Figure 1: One carbon metabolism. 1. AdoMet dependent transmethylation 2. Glycine N-methyltransferase 3. Ado- homocysteine hydrolase 4. Cystathioine p-s ynthase 5. Me thioine synthase 6. kine-glycine hydroxymethyltransferase 7. Methy lenetetrahy drofoalte reductase 8. AdoMet synthase (MAT). 6 Methionine is required for both protein synthesis and as the backbone S adenosylmethionine (SAM)23. The remethylation cycle is initiated when methionine is adenosylated using ATand one of three methionine adenosyl- bansferase (MAT)isoenzymes, ultimately producing SAW. SAM does not circulate and therefore aIl cellular required SAM must be generated locally. Mat
II predominates in most tissues except the liver and RBP. Because MAT fI functions at near its maximal capacity at normal cellular methionine concentrations cellular SAM levels remain fairly constant. However, hepatic tissue contains an additional MAT 111 enzyme, which responds to elevated methionine levels by increasing hepatic SAM levels. Numerous transmethylation reactions require SAM and the demethylation of SAM irrespective of the isoenzyme forms Sadenosyl-homocysteine(SAH). The efficient removal of SAH by SAH hydrolase is dependent on the metabolism of its products homocysteine and adenosine24. Elevated homocysteine prevents the metabolism of SAH and therefore impairs transmethylation reactions. Removal of cellular homocysteine is dependent on the balance of three distinct processes;
1. Remethylation 2. Transsulfuration 3.Cellula.r export.
Remethylation back to methionine is dependent on the cofadors 8125- methy ltetrahy drofolate (MTHF) and the enzyme me thionine synthase.
Methylenetetrahydrofolate reductase (MTHFR) synthesizes the active 5-MlXF from 5-10 methylenetetrahydrofolate25.5-methyltetrahydrofolatedonates the methyl group to the cobdamin bound to MS, which is then transferred to the free 7 sulfhydral of a homocysteine molecule forming methionine. An additional remethylation pathway capable of generating methionine from homocysteine exists in the liver, kidney, brain 23 and lens 26 using the metabolite of choline, betaine, as the methyl-doner and be taine-homocy steine me thy1 transferase
(BHMT). The product of BHMT, dimethylglycine, can further drive remethylation by the previously mentioned MS by contributing ultimately to three methylated folates.
Transsulfuration of homocysteine competes with the remethylation cycle
for homocysteine and serine. The first step condenses se~eand homocysteine
using the heme-containing pyridoxal-phosphate dependent cystathioine-B
synthase27. y-cystathionase cleaves cystathionine forming a-ketobutyrate and
cysteine, irreversibly removing a homocysteine molecule 27.
Regulation of flow through the transsulfuration and remethylation
pathways is dependent on cellular SAM levels. SAM inhibits MTHFRB and
BHM"F9 while stimulating CPP. Therefore when methionine is in excess leading
to a corresponding elevation in SAM, the irreversible transsulhuation pathway is
favored and remethylation is inhibited. Conversely, if SAM levels are low then
the stimulation and inhibition of the transsulfuration and remethylation
pathways is removed, respectively, favoring the remethylation of homocysteine
and conserving methionine. This regulation of the transsulfuration and
remethylation pathways might be one of the mechanisms that contribute to the 8 stability of cellular SAM levels and enable hepatocytes to regdate sulfur amino acids.
Urinary excretion of homocysteine is limited31. First, only a small fraction of circulating homocysteine is free as opposed to protein bound and therefore available for filtration through the glomerulus. In addition, no difference in homocysteine concentration was found between the human renal artery and vein31. Nevertheless, sigruhcant excretion of homocys teine can occur in subjects with severe elevations that must saturate renal tubule uptake. It is possible that low levels of excretion of me thionine derivatives serve to regulate homocysteine levels. In addition, the kidney may contribute to the maintenance of homocysteine levels by means other than excretion. It has been shown that the kidneys produce a large portion of serine32. It is not known how or if this serine production contributes substantially to the methylation of folate or synthesis of cystathiortine.
1.2 Folate metabolism:
Homocysteine levels are highly dependent on Mate status and more specifically on the availability of 5-methyltehahydrofolate.The structure of 5-
MTHF is illustrated below. '~~25~7~6 Mcd. WL: 459.16
Figure 2. Structure of 5-methyl tetrahydrofolate
Cellular uptake of folate derivatives is mediated by two membrane bound proteins. The first is a glycosy 1-phosphatidy 1-inositol folate-binding protein receptor (FBP)FBP has a high affinity but a low capacity for folate transport.
However the majority of folate transport is carried out by the reduced folate carrier (RFC),which unlike the FBP has a high capacity but, lower affinity". In addition each carrier preferentially targets the absorption of various folate derivative. The RFC has a much higher affinity for the reduced forms of folate such that both 5-methyltetrahydrofolate and folinic acid bind and are transported with equal affinities, whereas folic add is 100 fold less Likely to be imported by the RFC. Conversely the folate receptor has a greater affinity for folate compared to 5-MTHF3. Very little is known about cellular folate export 10 other than it is an ATP dependent process and proceeds independently of the
RFC and FBP. Once inside the cell folates are rapidly polyglutamated by folate poly glu tamate synthase (FPGS). This poly glutama tion aids in maintaining cellular folate levels preventing export. Interestingly, FPGS prefers tetrahydrofolate, 5-10 methylenetetrahydrofolate and formyl-folates derivatives as substrates compared to 5-methyttettahydrofolate35-36.It has been hypothesized that this preferential binding by FPGS may play a role in causing the "methyl trap". During a cobalamin deficiency 5-MTHF accumulates because of inactivated methionine synthase. This increase in 5-Mcan then lead to a shift in folate distribution favoring circulating folates. Over time a folate deficiency arises because of the increased excretion of the circulating folates 3437's.Indeed, this is supported by studies, which have shown that supplementing with cobalamin causes a substantial drop in circulating folate levels39. It could be proposed that the reactivation of methionine synthase by the influx of B12 promotes the conversion of 5-methyltetrahydrofolate to THF, therefore enhancing the retention of folate within the cells by providing FPGS with a better substrate. Moreover, this suggests that cellular remethylation of already glutarnated THF may be cruaal in maintaining polyglutamated forms of 5-
MTHF required by MS. However, given that the majority of circulating folate is
5-h4T'E-Ethis polyglutamation preference likely only becomes a problem during a cobalamin deficiency. It is possible to speculate nonetheless that the preference 11 of FGPS for THF and fonnyl-folates serves as a mechanism facilitating the release of predominantly 5-Winto the circulation.
Intracellular methylation of folate is dependent on numerous processes occurring within the cytosol and mitochondria and is illustrated in figure 1.3 below. SIOIMWIUTHF Cly
Sarcoslnc
Figure 1.3. Intracellular metabolism of folate.
(Enzymes are underlined and protein carriers/receptors are shaded. RFC=reduced folate carrier, FBP=folate binding protein, DHFR=dihydrofoIate reductase, cSM=cytosolic serine-hydroxymethyUtransferase, MTHFR=methylenetetrahy drofolate reductase, MS=methionine synthase, BMTzBetaine-homocysteine methyltransferase, DMGD=dimethylglycine dehydrogenase, SD=sarcosine dehydrogenase, mSHMT=rnitochondrial serine- hy droxymethylhansferase, THF=tetrahydrofolate, DHF=dihydrofolate, THF=tetrahydrofolate,5-W-5-methyltetrahydro folate, Met-aeethioonine,Sam=Sadeno~yImethioine,S~=Sadenosyhornocysteine, Hcy s=Homocysteine, Cy sta=cysta thioine, Cys=cys teine, GIy=glydne, Ser=serine) 13 This cellular segregation of both the folate derivatives and the remethylation enzymes between the cytosol and mitochondria give us some clue as to what role each play. Methylation of tetrahydrofolate in the cytosol occurs primarily by the
3iarbon born serine being transferred by se~e-glycine hydroxymethyltransferse to THF generating 5-10-methylenetetrahydrofolate4*.
Mitochondxial remethylation of folate is somewhat more complex and is carried out by four distinct enzymatic processes. Mitochondria1 se~e hydroxymethyltransferase (mSHMT), glycine cleavage, dimethylglycine dehydrogenase and sarcosine dehydrogenase all contribute to the methylation of folate within the mitochondria. Although folate flux between the cytosol and mitochondria is limited, onetarbon aoss talk between the cytosol occurs via serine, glydne and possibly fonnate4l". It does appear that this interchange of glycine and serine is at least partially dependent on nutrient status as well as being tissue dependent. A large portion of glycine catabolism is carried out by the mitochondrial glycine cleavage system (GCS)and therefore must generate a substantial portion of its product, 5-10 methylenetetrahydrofolate. GCS is expressed primarily in the Liver's and kidney, perhaps highlighting the importance of these organs in folate metabolism. It would be interesting to measure blood folate status in the renal and hepatic arteries and veins.
Additional synthesis of 5-10 methylenetetrahydrofoiate within the mitochondrial can be spawned by the coupled dimethylglycine and sarcosine dehydrogenase system ultimately producing an additional glycine molecule. The generation of 5- 14 10 methylenetetrahydrofolate from three sources within in the mitochondria1 would favor production of serine by the mitochondria1 she-glydne hydroxymethyltransferase. Serine efflux out of the mitochondria could serve as the one carbon donor to homocysteine, mediated by the cytosolic serine-glycine hy droxymethylhansferase.
The majority of studies investigating the interactions in one carbon metabolism between the mitochondria and cytosol have been done using knockout yeast models. Nevertheless, the enzymes involved such as se~e- glycine hydroxymethyltransferse appear to be very well conserved and therefore presumably apply to higher level eukaryotic cells. Indirectly this relationship between serine and folate has been illustrated by a study conducted by Wilcken et al*. These authors demonstrated that that supplementation of folate to uremic patients resulted in a marked reduction in serine levels and a corresponding increase in glycine levels suggesting that re-me thy la tion of fola te is critically dependent on serine levels in patients on hemodialysis. This relationship in uremia will be discussed in more detail in chapter 3. It is not know if undiscovered polymorphisms affecting the reme thy la tion of folate are relevant to
homocysteine metabolism or if they influence the risk for vascular disease. The contribution of impaired methylation of tetrahydrofolate in the remethylation
cyde to hyperhomocysteinemia is unknown. 1.3 Genetic and nutritional contributions to hvperhomowsteinemia
Enzymatic deficiencies due to impaired cofactor activation and deranged protein stru-e are capable of inducing hyperhomocysteinemia both independently and in combination.
Vitamin 812 is an essential cofactor for two mammalian enzymes. The first involves methionine synthase and the second is methylrnalonyl mutase
(MMM)47. MS requires a methyl group bound to cobalt within B12 and as mentioned earlier catalyzes the remethy la tion of homocysteine to me thionine.
MMM is a mitochondrial protein that converts methylrnalonyCCoA to succinyl-
CoN7MMM uses an adenosylated version of 812. The absorption and subsequent activation of cobalamin is a complex process involving numerous physiological and biochemical steps before it is capable of interacting with either
MS or MMM47
The absorption process is dependent on a cascade of binding factors and receptors; 1. Intrinsic factor O 2. R-binders 3. Transcobalamin (TC) 4.
Membrane receptors and 5. Intracellular binders. The IF factor is a glycoprotein secreted by the parietal cells in the stomach, which binds free B12 forming an IF- cobalamin complex. Absorption is then mediated by receptors specific to the IF-
cobalamin complex on ileum enterocytes. Cobalamin is released into the
drmtation and binds either transcobalamin 11 or I. Although the exact role of TC
U and TC I is not well described, it is thought that TC 11 functions as a rapid
delivery mechanism and TC I mediates long term supply. The TC 1Icobalami.n 16 complex enters the cell by pinocytocis where it is cleaved from TC II in the
Severe genetic defects in cobalamin metabolism are rare enough not to be a plausible contributor to vascular disease risk for the general population. It is interesting to note nonetheless, that many of the genetic lesions present phenotypically with vascular complicationS18. Nutritional deficiencies in 812 metabolism on the other hand can, espeaally in the elderly dramatically alter both homocysteine and MMA levels49;50. In addition, and as previously discussed a deficiency in 812 also can contribute to increasing urinary folate excretion further depleting methionine synthase of cofactors. Depleted 812 may further a1ter fola te status and contribute to hyperhomocysteinemia by impairing the methylation of folate. MMA and propionyt Co-A, both of which become elevated when WIis impaired, may be capable of directly inhibiting the mitochondria1 glycine cleavage systems' or glycine transport into the mitdondria5m and therefore the generation of 5-10 methy lenetetrahydrofola te.
Intestinal absorption of folate is mediated by an active and passive mechanism. From the enterocyte, folate derivatives then enter the hepatic-portal system and are methylated by the liver. The SMTHF is then either secreted into the circulation, bile or incorporated into one carbon metabolism. Genetic lesions affecting intestinal absorption are rare and likely do not contribute to risk for vasdar disease in the general population. However, a common po1ymorphism affecting the cellular incorporation via in the reduced folate carrier has recently 17 been identified and is a result of a guanine residue being replaced by an adenine.
Individuals who were homozygous for the RFC 80 A/A genotype had sigmficantly elevated serum foiate levels compared to the G/G subjects. The ailelic frequency was 52.65% and 47.35% for the G and A alleles respectively. The effect of this mutation on cardiovascular disease remains to be determined 54.
Four other common poiymorphisms affecting the inhaceHular metabolism of folate and homocysteine and consequently the risk for vascular disease have been described. The first identified genetic risk factor contributing to mild hyperhomocysteinemia is caused by a missence mutation in MTHFR at base pair
677. The thymidine to cytosine switch causes an alanine to be replaced by a valine residue. By disrupting the secondary structure of MTHFR the valine residue predisposes the disassociation of the cofactor FAD inactivating the enzyme 25. Interestingly 5-MTHF, the product of MTHFR stabilizes bound FAD.
This is reflected in clinically by studies that have shown that the T allele only expresses its phenotype, elevating homocysteine 55-57when fola te levels are moderately compromised 5839. Although not conclusively, the T/T genotype has been shown to increase the risk for venous thrombosis -, coronary artery disease -cerebrovascular disease and neural tube defects69-72. At the same time, other investigations suggest that the TIT mutation may be protective against colon cancerni74 and leukemia75 .It is thought that decreased activity of
MTHFR shifts the methylated folate pool favoring nucleotide synthesis over
homocysteine methylation. The increased availability of nucleotides may prevent 18 defects in oncogenes by reducing the incidence of uracil misincorporation in the rapidly dividing tissues.
Another polymorphism affecting MTHFR has recently been discovered and is a result of a missence mutation at nucleotide 1298. An adenine is replaced by a cytosine translating to the replacement of a glutamine by an alanine. The frequency of this mutation is very similar to the 677 C+T polymorphism with homozygous prevalence being around 10%76 7.The evolution of these (MTHFR
1289 and 677) two mutations is likely to have been a distinct event as the two polymorphisms are rarely expressed in the same enzyme. This may have important implications in that dual heterozygotes will have a phenotypic expression similar to a homozygous mutant patient =?There is no published in vitro data investigating the effects of the 1298 A+C mutation on enzyme activity.
Nevertheless, clinical reports have shown that this polymorphism influences homocysteine levels and risk for neural tube defects59.
Methionine synthase uses the product of MTHFR, 5-
methyltetrahydrofolate to transfer methylgroups to homocysteine. A missence
mutation affecting MS due to an adenine being replaced by guanine at nudeotide
2756 causes a aspartate to be replaced by a glycine residue 78;79 group. Again to
date, only clinical reports indirectly show that the glycine residue deaeases
activity. One large study demonstrated that homozygosity for GIG genotype
mildly elevated risk for vascular disease in smokers? We have recently
examined the effect of this mutation on the incidence of recurrent cardiovascular 19 events and endothelid function. This is discussed in more detail in chapter 4 and
Spontaneous oxidation of the cobalt atom of cobalamin (I)to cobalamin
@) inactivates MS. Methionhe synthase reductase (MSR), reactivates MS via reductive methylation using SAM as the methyl dono+? A very common polymorphism in MSR at nucleotide 66 A+G introduces a isoleucine residue in place of rnethionine. Brown81 et a1 has recently determined the G/G genotype was slightly more prevalent in patients with premature cardiovascular disease m.
There is a limited number of studies examining the relationship between the various polymorphisms few of which have focused on examining their effects on risk for cardiovascular disease. For example, Christensen69 have shown that the MS 2756 A+G mutation lowers the incidence of neural tube defects an effect that was exactly opposite to the WR677 C-tT mutation. Tsaia* et a1
have recently investigated the effects of the MS 2756A-G polymorphism and
the 844ins64 alteration in CPS in 1031 individuals with CAD or subjects enrolled
in primary prevention for vascular disease. Interestingly, unlike and opposite to
the MTHFR 677 C+T mutation both the MS 2756 A+G and the CPS 844ns64
lowered homocys teine levels. While, homozy gosi ty for the MS G/G genotype
sipficantly lowered fasting levels the A/G genotype conferred some protection
against elevations in homocysteine induced by the methionine load. Given that
the MS 2756 A4mutation Likely has impaired activity 82 it is possible to
speculate that a slower MS enzyme could act to shift the folate pool increasing 20 the levels of 5-MTHF and thereby provide a reserve of methylgroups. This increase in 5-MTHF may act in two ways; 1. To delay the depletion of 5-MMF and therefore the corresponding inaease in homocysteine induced by a methionine load or 2. Facilitate the export of 5-MTHF by hepatocytes, and therefore increase one carbon units decreasing homocysteine export from other tissues. The CPS insertion had little or no effect on the fasting homocysteine levels but sigruhcantly lowered the post methionine load response82. Given the lowering effects of the MS 2756 A+G and the CPS insertion coupled with the opposite effects of the MTHFR 677 C+T and M''T'H.FR 1298 A+G mu tations on homocysteine levels, it is not surprising that a great deal of discrepancy exists where some reports demonstrate that the MTHFR 677 confers no deleterious effects and other demonstrate clear association with risk for vascular disease. The incidence and interaction between the MS, MSR and MTHFR polymorphisms between various ethnic populations and vascular disease pathologies remains obscure. Physiologically, any inaeased risk caused by the MTHFR polymorphisms (677 C+T, 1298 A+C) may be balanced out by a protective effect by heterozygosity for the MS 2756 A+G genotype. Furthermore, it is not known if the polymorphism in the 80 G+A point mutation in the RFC contributes to vascular diseases or how it interacts with the other previously mentioned polymorphism.
Redudng the bioavailability of folate also appears to influence cardiovascular morbidity. Methotrexate 0has been extensively used as a 21 chemotherapeutic agent because of its growth inhibitory effectson rapidly dividing tissues. MIX is a potent inhibitor of dihydrofolate reductase, preventing the synthesis of formyl-Mate derivatives and therefore also the production of nucleotides. It is these properties that have contributed to its success, making it the most commonly prescribed drug for alleviating rheumatoid arthritis.
Remethylation of folate is also impacted by MTX treatment. Adequate, THF is required before any methylation of folate can occur. This is reflected by a rise in homocysteine as early as two days after the initiation of low dose h4TX treatment
(25mgIday). Landewe et dB3 conducted a large clinical trial examining the effect of MTX on cardiovascular disease mortality. The investigators determined that patients with CVD started on M'TX treatment had 4.34 (p=0.0029) times the mortality compared to untreated subjects af. It was hypothesized that this may have been caused by an elevation in homocysteine. Nevertheless MTX treatment depletes THF folate and in order to increase homocysteine must also deplete 5-
MTHF.
1.4 Homocvsteine and Atherosclerosis:
Hyperhomocysteinemia was first associated with vascular disease in 1969 by Killmer McCully 84 in patients with homozygous deficienaes in cystathionhe-
P-synthase (Cp). This observation remained relatively unnoticed until Wildten and Wilcken85 demonstrated that homocysteine metabolism is altered in patients with angiographically documented coronary artery disease. Since these earlier 22 reports more than 1500 clinical and basic science papers have been published investigating mild hyperhomocysteinemia. Many of which have found elevated levels of homocysteine to be associated with and or an independent risk factor for vascular disease.
The most convincing evidence Linking homocysteine with vascular disease to data comes from a meta-analysis performed by Boushey et alM. The investigators summarized 38 studies and determined that each 5 umol/l incremental rise in homocysteine, was associated with an increase OR of 1.6 (95%
1.4-1.7)and 1.8 (95 % 1.3-1.9) for cardiovascular disease in men and women respectively. Similar associations were found between cerebral and peripheral artery diseases. In agreement Cattaneo et a187 and Eikelboom et ale8 came to similar conclusions. Cattaneo determined that an elevation in homocysteine increased the risk for CVD by a factor of 1.4 to 4.587. Moreover, it has been shown that patients with confirmed coronary artery and hyperhomocysteinernia have increased mortality 89. Al-Obaidigo et al have extended Nygard's observation by reporting an association between hyperhomocysteinemia and increased ischemic injury after a myocardial infarction. Cardiac-specific troponin, was used to assess myocardial damage, and peak levels were sigruficantly higher in patients in the highest homocysteine quintile. This study suggests that subjects with elevated homocysteine are predisposed to greater ischemia-reperfusion injury.
Intravenous, IV folinic acid is highly effective in restoring cellular folate levels and acutely lowering homocysteine. However there are no studies to date that 23 have investigated the effect of acute administration of homocysteine lowering therapies such as folinic acid on post myocardial infarction recovery.
Clearly, homocysteine or a mechanism contributing to it's elevation, increases the progression and outcome of atherosclerosis. However a definitive rnechanism(s) remains to be found. Indeed, evidence exists suggesting that a folic acid deficiency, know to contribute to an elevation in homocysteine levels, could increase risk for vascular disease independently of homocysteine. An elegant study by Rimm et a191 examined folate, 86 812 and alcohol intake on risk for fatal and non fatal myocardial infarction in 80 082 women. A higher intake of folate and vitamin 86 provided protection against vascular disease. Interestingly, the effect of folate was most evident in women who consumed > 15g/d of alcohol.
The authors speculated that acetaldehyde altered folate metabolisms by
inactivating methionine synthase. Therefore subjects with increased folate intake
were protected from the inactivation of methionine synthase, preventing the
corresponding rise in homocysteine. It is possible to speculate the t acetaldehyde
inactivation of methionine synthase coupled with increased folate consumption
functions in a manner similar to the effect of the 2756 A+G polymorphism in
methionine synthase, conferring protection by increasing 5-
methyltetrahydrofolate (see chapter 5).
The inverse relationship between homocysteine and 5-
methyltetrahydrofolate introduces a paradoxical situation complicating the
discovery of pathological mechanisms. Physiologically, if 5- 24 methyltetrahydrofolate is depleted homocysteine becomes elevated. However, alternative pathways such as the transsulfuration pathway do not require folate and therefore lesions affecting the transsulfuration of homocysteine may reveal clues separating this relationship between hornocysteine and folate. Patients with
CPS deficiency present with dislocation of the lenses, thrornboembolic events, mental retardation and osteoporosis. If left untreated the majority of patients will succumb to vascular thrombosis and vascular disease due to extreme elevations
(200-500umol/l) in homocysteine*. Given that patents heterozygous for CPS do have moderately elevated homocysteine levels, it would be logical to assume they would be over represented in vascular disease populations. Whitehead93 et al. determined that none of 100 Irish patients with vascular disease and two of sixty controls were heterozygous for the CPS mutation. A recent study by
Eberhardt and Fogione et a1 200094 investigated the effect of heterozygosity for
CPS using a genetically altered mouse model. Heterozygous (CPS -/+), mice had impaired endothelial function. Acetylcholine, A23187 and methylcholine all induced vasodilatation and vasoconstriction in the control CPS +/+ and CPS -/+ animals respectively. Sodium nitroprusside responses were normal in both groups. The CPS -/+ mice also had depleted cGMP, increased superoxide production and 3-nitrotyrosine immunostahing9? In addition, vasoconstriction was reversed with exogenous supplementation of Cu/Zn-SOD. Conversely, another study using a similar animal model determined that endothelial dysfunction in CPS deficient mice was dependent on folate status. It remains to 25 be established if the homocysteine elevation caused by dysfunction CPS aeated either a tissue specific and or a cellular folate deficiency. The other variables measured in the first study were not analyzed, but given that endothelial function is dependent on cGMP and superoxide/nitrotyrosine status it is reasonable to predict they also are dependent on folate. This however remains to be determined.
Vascular smooth muscle cell proliferation is an important aspect in the development of the atherosclerotic lesion, contributing to both the size and status. Arterial manipulation from angioplasty, stenting, or bypass surgery
%her increases SMC proliferation and is the primary cause of postoperative restenosis 95%. Hyperhomocysteinemia has dso been determined to be a independent predictor of intimal thickening of the carotid arteries 97-99. Southern et al 99 Merdemonstrated using a rat model that a diet deficient in folate and methionine, but supplemented with homocysteine caused a four fold increased in re-stenosis. However, because the diet was both deficient in folate and at the same time supplemented with homocysteine it is difficult to determine if the
SMC proliferation was a result of hyperhomocysteinemia or insufficient folate status 9. Human smooth musde cell cultures are also influenced by homocysteine concentrations. Clinically relevant elevations in homocysteine
(25umol/l) sigruficantly elevated thymidine incorporation and cellular proliferation of the isolated SMC. Supplementation with 20nmol/l folic acid reversed the dose dependent inaease in growth induced by homocysteine IW. 26 Although the beneficial effect of folic acid supplementation remains ill-defined it is likely that its antiproliferative properties are due to lower cellular homocysteine levels.
Numerous other studies have attempted to determine the link between homocysteine and vascular disease. Hyperhomocysteinemia has been reported to potentiate endothelial senescence, '0-1 thrombosis, activate immune cells 1% promote leukocyte rolling 'Qlm, interferes with factor W", induces hydrogen peroxide mediated endothelial damage '05;106, inhibits thrombomodulin and protein C 107, decreased t-PA 'Q,suppression of heparin sulfate ,109 and modulate the effect of platelet derived growth factorllo. Of great concern is the level of homocysteine used to achieve these results and the lack of proper control amino acids. Most of the above-mentioned studies used concentrations exceeding 0.5
mM and some were as high as 10 mM.One study did see an effect at O.lmM,
which could be compared to total physiological levels in CPS patients.
Furthermore, the addition of reduced free homocysteine, synthesized from
homocysteine thiolactone is not in any way representative of circulating levels. In
fact, less than two percent of circulating plasma homocysteine is in the reduced
form, and most dturestudies have used incubations with 1m.M free
homocysteine, which would comparatively represent a 10 000-fold elevation. Cell
culture media contains 10-20% serum and free protein bound sdhydrals are not
as prevalent compared to in vivo conditions. This would allow free
homocysteine to predominate perhaps even mimicking much higher in vivo 27 conditions. These studies may provide insight into the cause of vascular disease and thrombosis in untreated homozygous CPS patients however they do not convincingly support the relationship between mildly elevated homocysteine and vascular disease.
If homocysteine is not causative then it could be proposed that a mechanisms contributing to its elevation such as a folate, 812 or 86 deficiency could contribute to the progression of atherosclerosis. Folate, vitamin B6 and vitamin B12 all strongly correlated with plasma homocysteine levels. Rimmn et a1 determined that individuals with increased intake of folate and B6 had a lower risk for myocardial infarction but found no association with B12R. Furthermore, vegans and vegetarians have been shown to have sigruhcantly elevated homocysteine levels compared to omnivoresl~~and yet vascular risk would be expected to be lower in vegetarians. Vegetarians might have a mild deficiency in
812 but should have more than adequate folate levels. A recent study a Bunout et
a1 "* compared homocysteir~eand folate levels between age and sex matched
subjects with and without coronary artery disease. Serum homocysteine levels
were not sipficantly elevated in the CAD population. Conversely, the serum
folate levels were significantly lower in the CAD patients 112. Clinical evidence
suggests that elevated homocysteine is associated with vascular disease, but the
mechanisms involved are poorly understood. The idea that homocysteine may
be a marker rather than causative due to insufficient folate status should not be 28 ruled out and that the pathologies associated with hyperhomocysteinemia may be partially caused by depleted folate.
1.5 Endothelid Function:
Endothelial dysfunction has emerged as an important target for assessment and modulation since it represents a surrogate marker for atherosclerotic vascular disease"? As discussed earlier, diminished production of NO, appears to be the most important mechanism underlymg the development of endothelial dysfunction in cardiovascular disease states. A variety of genetic and environmental CAD risk factors, including hyperhomocys teinemia, inhibit endothelid Function18.
Nitric oxide is synthesized by three isoforms of nitric oxide synthase;l.
Endothelial nitric oxide synthase (eNOS) 2. Neuronal nitric oxide synthase
(nNOS) and 3. Inducible nitric oxide synthase (iNOS). As one might suspect the three enzymes are highly homologous in natur643. Nevertheless, even though all produce nitric oxide, sigruficant differences in physiological and tissue distribution exist "4;115. The activity of both nNOS and eNOS is regulated by calcium. The thud NOS isoform is also activated by calcium but remains fully active at baseline cellular levels. Catalytic activity of all three isoforms is also dependent on the cofactor tetrahydrobiopterin3.
In the endothelial cell, NO is synthesized from L-arginine in a reaction catalyzed by the enzyme eNOS Figure (1.4). \ \ nowOF ELECTRONS + N=----* of
Dihydroneoptcrin mphosphatc i 6-Pyruvoy l tctrahydropttrin Endothclium v Q-H, Biaptcrin H, Bioptcrin
TCGMP +RELAXATlON Smooth Muxlc
Figure 1.4. Representation of the enzymatic pathways for the synthesis of nitric oxide and tetrahydrobiopterin.
This reaction requires molecular oxygen, tetrahydrobiopterin (BH4),
NADPH, FAD,FMN, calmodulin and heme as cofactors. NOSiII (or endothelial eNOS) is composed of two identical subunits, each containing a reductase and an oxygenase domai.28.The N-terminal oxygenase domain binds both heme and 30 BFh, whereas the reductase domain contains the binding site for NADPH,which donates electrons to FAD/FMN and finally to L-arginine. Cellular regulation of eNOS is in part dependent on its location. The inactive form of eNOS is bound to caveolin-1 within the caveoiae of the plasma membrane293.1. Activation by the calcium calmodulin complex has a dual function to release the enzyme from the inhibitory caveolin-1 protein as well as directly activating the enzyme.
Myristylation of eNOS during translation followed shortly by the addition of two
palmitate fatty acids is required for caveolin binding"%In the unstimulated state
the majority of eNOS is bound within the caveolae and after a stimulus that
increases intracellular calcium levels, eNOS disassociates and is active within the
cytosol.
NO formation is critically dependent on the presence of the cofactor Bb.
By interacting with heme, B& shifts the activity of eNOS to favor the production
of NO"'-120. The role of BH4 within the active site of eNOS is not completely
understood, however it has been proposed to 1. Direct the flow of electrons
toward bound arginine and 2. Enhance eNOS's affinity for arginine. The
importance of BH4 is underscored by observations indicating that diminished
levels of this cofactor direct eNOS to produce superoxide over NO; an effect that
may contribute to the development of endothelial dysfunction lp. This
relationship has been demonstrated in vitro with puriiied NOS, where BH4
defitient enzyme results in increased superoxide production 203. Furthermore,
isolated canine arteries depleted of BH4 have sigruhcantly increased production 31 of oxygen-derived free radicalsln. In the absence of BH4 molecular oxygen is able to bind the heme moiety of the oxygenase domain and facilitate superoxide production. One could argue that because BH4 is essential for all three enzyme that a deficiency in BH4 could contribute to superoxide generation from all three isozymes. However it is unlikely that superoxide production because of BH4 deficient nNOS and iNOS is relevant in vascular tissue for a number of reasons.
First of all pterin binding to eNOS is not required for dimer formation of eNOS, as it is for NOSIB. Furthermore cystdographic studies of the isozymes have revealed that the pterin-binding site in eNOS is far more often empty compared to the other NOS isoforms (personal communication Dr. Libby). And lastly the transcription of GTP cyclohydrolase and NOS are both synergistically up regulated allowing ample BH4 for binding, during NOS expressionl*?
It has been demonstrated that stimulation of coronary arteries with Ach causes a dose and calcium dependent vasodilation125. Conversely, in patients with impaired endothelial function and atherosclerosis Ach induces
vasoconstriction^^. This constriction in the atherosclerotic artery is thought to be a result of depleted NO bioavailability and inaeased production of ROS from the stimulated endothelial cells. Indeed, NOS itself is an excellent causative candidate and is likely that the malfunction of eN05 during Ach stimulation contributes to free radicals production and potentiates the vasoconstriction.
Interestingly, Maier et a1 recently demonstrated that co-infusion of BH4 with Ach stimulation not only prevents this vasoconstriction but also allows Ach 32 vasodilatation in atherosclerotic coronary arteries's. It is well documented that endothelial cells treated with LDL produce an increased amount of superoxide and hypercholesterolemia enhances oxidative stress and impairs NO release
15;lu;la.Interestingly, brachial artery hction of hypercholesterolernic patients was improved by infusion of BH4, implicating eNOS derived free radicals as one of the culprits contributing to endothelid dysfunction in hypercholesterolernia
'29. Additional clinical studies in humans examining the effects of smokinglM and hyperiipidemia, have revealed that BH4 improves endothelial hction. There was no beneficial effect of BH4 in the controls.
Along the same lines, the importance of sufficient BH4 during ischemia- reperfusion injury '31-134 has been illustrated by Schmidl" et al. who determined that unilateral lung transplantation in swine with excess exogenous BH4 increased cGMP and sigtuhcantly lowered extravascular lung water, all of which were attributed to improved endothelial nitric oxide production. Ishiil32 et al. also have also examined supplementation of tetrahydrobiopterin and its effect on superoxide and nitric oxide in gastric ischemia-reperfusion injury in rats. They determined that supplementation with BH4 or sepiapterin (a precursor of BH4) decreased injury and replenished stomach BH4 levels. Moreover, an inhibitor of sepiapterin reductase prevented the protective effect of sepiapterin confirming that the derived benefit was a result of tetmhydrobiopterin. Ischemia reperfusion injury plays a large role in patient outcome after myocardial infarction. The effect 33 of tetrahydrobiopterin supplementation on post-ischemic injury following coronary infarction remains unclear.
The role of nitric oxide extends well beyond acute regulation of endothelial function and inflammation. Systematic effects of NO regulate blood pressure and cardiac output '"137. eNOS knock out mice are hypertensive, having blood pressure levels 20-30mmHg above control animals 138. Furthermore the importance of eNOS in hypertension is well illustrated by Consentino 139 who demonstrated that superoxide production was elevated in SHR rats compared to the WKY control group. Interestingly the investigators found that supplementation with tetrahydrobiopterin attenuated superoxide production, endothelial dysfunction and impaired nitric oxide release in response to the calcium ionophore A21387.
1.6 Tetrahvdrobiopterin Depletion:
A deficiency in BH4 can be a result of insufficient synthesis and/or recycling of pterin into its active reduced form coupled with increased degradation. De novo synthesis of H4B requires three enzymes. The first reaction is catalyzed by GTP cyclohydrolase, converting GTP into dihydroneopterin hiphosphate. A Zn and Mg dependent metalloprotein called 6- pyruvoyltetrahyciropterin synthase converts dihydroneopterin triphosphate into bpyruvoyl tetrahydrobiopterin. Finally tetrahydrobiopterin is formed by
NADPH sepiahin reductasel". 34 If the de novo synthesis of BH4 is not responsible for the BH4 deficiency in the endothelium then insufficient BH4 may be due to impaired recycling of the cellular biopterins. The replenishment of BH4 from the oxidized qHZB is performed primarily by dihydropterin reductase WHRP). Although the recycling of tetrahydrobiopterin has not been examined with vascular disease in mind, regeneration has proven to be important in neurological disorders'*.
Moreover, an elegant study by Vreko et al. has revealed that the conversion of qH2B is not saturated and can be stimulated by cofactors~41.Dopamine production, which would be reflective of BH4 levels, was increased in cultured rat cells by the addition of exogenous NADH. They also determined that the increase in synthesis of dopamine was not a result of activation of GTP cyclohydrolase I but rather increased conversion of qH2B to H4B by DHPR. This suggests that the reduction of qH2B is important in maintaining adequate cellular levels of H4B. These results did not examine eNOS function nor were endothelial cells used, however they do indicate that basal BH4 is at least
somewhat dependent on the active recycling.
1.7 Folic acid and Nitric oxide:
The effect of folate on nitric oxide production and endothelial function
was first discovered by Verhaa.9 et al. In vivo supplementation with 5-
methyltehahydrofolate to patients with hypercholesterolemia caused a
sigtuficant improvement in endothelial dependent vasodilatation, in a manner 35 identical to that supplementation with BH4. Title'* et a1 have expended on this obse~ationand examined endothelial function using a parallel-designed trial supplementing with placebo, folic acid and folic acid with antioxidant. Folic acid alone was the most effective treatment sigruhcantly improving vasodilatation in the CAD patients. Surprisingly, the antioxidant treatment although effective in reducing ROS, as it lowered malonyldialdehyde levels did not alter endothelial function. This finding is controversial given the prevailing hypothesis that hyperhomocysteinemia induces oxidative stress. Using the same technique Chao et all43 tried an opposite approach by measuring endothelial function before and after a methionine load. Interestingly, the authors concluded that even though the methionine load increased homocysteine levels the associated impaired endothelial function was not a result of acute oxidative stress. Previous studies have also shown that the methionine load not only increases homocysteine levels but also depletes 5-W44.One could argue that impaired endothelial function caused by a met!hionine load is a result of lowered folate status instead of hyperhomocysteinemia.
Given the almost identical structures and the similar clinical effects of 5- methyltetrahydrofolate and tetrahydrobiopterin, it is reasonable to hypothesize that the relationship between hyperhomocysteinemia and vascular disease may be in part caused by a defiaency in 5-methyltetrahydrofolate.This depletion of 5-
may, like tetrahydrobiopterin, shift eNOS favoring the production of superoxide over nitric oxide. 36 The objective of the studies herein were to investigate some of the potential effects of homocysteine and folate metabolism on cardiovascular disease and endothelid function, using dinical, moiaxlar, animal and cell culture experiments. CHAPTER 2
ORAL VITAMIN B12 AND HIGH-DOSEFOLIC ACID IN HEMODIALYSIS
PATIENTS WITH HYPERHOMOCYST(E)INEM[A:
A RANDOMIZED CLINICAL TRIAL
Braden Manns, M. Eric Hyndman, Ellen Burgess, Floyd Snyder, Howard G. Parsons and Naime Scott-Douglas.
Departments of Pediatrics' Medical Genetics2 and Medicin@* University of Calgary, Calgary, Alberta, CANADA
(This chapter is reprinted with permission exactly as it will be published in Kidney International 2000) 2.1 Introduction:
Atherosclerotic cardiovascdar disease (ASCVD) is the cause of death for 25960% of patients with ESRD and is also a major cause of chronic morbidity 145;146. This high prevalence of ASCVD is not fully explained by the increased rates of known risk factors such as hypertension, hypercholesterolemia and diabetes melli tuslJ7;1*. Hyperhomocy st (e)inemia is an independent risk factor for
ASCVD in patients without renal failurel49;la. More recently, it has been established as an independent risk factor for ASCVD in both retrospective
3347;151-19 and prospective studies lS1" for patients on dialysis. This is important since hyperhomocyst(e)inexnia(>go" percentile control value) is the most
prevalent risk factor for ASCVD in ESRD, occurring in 83-91 % of patientsld7;'s.
The physiological basis for hyperhomocyst(e)inemia in renal failure is not
clear '47157, but is important to consider since it forms the basis for therapy of
hyperhomocyst(e)inemia. Hornocyst(e)ine is produced during the metabolism of
methionine, an essential amino acid, and can be metabolized by three separate
pathways'". The remethylation pathway is the most important determinant of
the fasting homocyst(e)ine level and correlates best with atheroderotic risk159;1*.
Several mechanisms have been postulated to explain the reduction in activity of
the remethylation pathway in renal failure including a subclinical deficiency of
folic acid and vitamin B12 two cofactors needed in the remethylation
pathway147:161;162. Another possible explanation for hyperhomocyst(e)ineIItia is 39 the important role of the kidney in the normal metabolism of homocyst(e)ine
147;142. TWOrecent studies have shown a marked increase in the half-life of homocyst(e)ine in ESRD patients due to a sigruficant reduction in the clearance of homocy s t (e)ine by the kidney147;162. This reduction in homocyst(e)ine metabolism by the diseased kidneys may be exacerbated by the very low levels of serine (an amino acid produced in the kidney) commonly present in patients with ESRD, since serine is needed as a methyl donor in the remethylation pathway32;163;164.
Uncontrolled studies have suggested that the treatment of hyperhomocyst(e)inemia in ESRD patients requires higher doses of folk acid and vitamin B12 to achieve a considerable reduction in homocyst(e)ine levels as compared to patients with normal renal functionl65;'66. No studies have compared monotherapy with oral folic add at the conventional dose used in
ESRD patients (i.e.lmg folic acid per day) with higher doses such as those used by previous studies (i.e.5-15mg per day)163;*65;167. The role of isolated vitamin
B12 supplementation to reduce homocysteine levels has not been studied in patients with ESRD who are not vitamin B12 deficient. No large studies have examined the role of serine deficiency in the hyperhomocyst(e)hemia seen in
ESRD patients and whether normalization of serine levels through supplemen tation could reduce homocy st (e)ine levels.
In order to assess these possibilities, a fourteen-week hial was conducted consisting of three separate intervention phases to determine the most effective method of reducing homocyst(e)ine levels in ESRD patients receiving hemodialysis. Two phases were open-label (vitamin B12 and serine) and one was randomized, double-blind and placebocontrolled (folic acid).
2.2 Methods
Patients: Hernodialysis patients who had a total homocyst(e)ine level
>16umol/L were enrolled from a single tertiary care center in Calgary, Canada.
Patients were excluded if they had vitamin 812 (<13Opmol/L) or folate
(<250nmol/L) deficiency, were currently receiving or had received treatment with vitamin 812 injections or an average of greater than lmg of folic acid per day in the past 6 months, or if they were currently using anti-folate or anti- epileptic medications (i.e. pheny toin).
One hundred and fifty-seven patients were screened for entry into the study; 26 patients did not have hyperhomocyst(e)inemia (>16umol/L), 45 were unable or unwilling to participate, and hence 86 patients were enrolled in the study. At baseline, 92.6% of the enrolled patients were taking supplemental folic acid and Bvitamins in the form of Diavitem (R&D Laboratories, Marina del Rey,
CA) which contains folic acid lmg, vitamin 812 6ug vitamin B6 3mg and small amounts of other vitamins. Of those who were taking Diavitem, 56% used it daily and 44% used it three-times per week post-dialysis.
Protocol: The Conjoint Health and Research Ethics Board at The University of
Calgary approved the study protocol which is depicted schematically in Figure 41 2.1 Patients were screened for the presence of hyperhomocyst(e)inemia with a midweek predialysis total homocyst(e)ine level (Week 0). During phase 1, eligible patients who agreed to enrolment were switched to daily oral Diavitem for four weeks. Baseline blood work was drawn for homocyst(e)ine, red blood cell (RBC) folate and vitamin B12 four weeks later (Week 4). In phase 2, Diavitem was continued and oral vitamin 812 (Nature Mad&, Pharmavite Corporation,
Mississauga, Canada) at a dose of 1mg/day for four weeks was added in all patients. Bloodwork was repeated (Week 8). In phase 3, patients continued to receive Diavitem and vitamin B12 daily, and in addition, they were randomized
(using a computer-generated sequence) to receive either placebo, folic acid
(Apotex Inc., T~:22to, Canada) 5mg/day, or folic acid 20mg/day for four weeks.
Folic acid/placebo capsules were dispensed in serially numbered containers and were identical in size, shape and colour. Thus, patients, patient caregivers, data collectors and investigators were blinded to treatment allocation for folic acid therapy throughout the trial. Bloodwork was repeated (Week 12) (figure 2.1).
In phase 4, the final 24 patients enrolled in the trial (7 from "placebo", 8 from "5mg folic acid, and 9 from the "20mg folic add group") continued
Diavitem, vitamin 812, and placebo/folic acid (all in the same doses), and in
addition each received USP grade L-serine at a dose of O.OSg/kg twice per day
(total O.lg/kg/day) for a further two weeks. Following this, final bloodwork
was drawn at week 14. After each phase of the trial, compliance with therapy
was assessed by pill counts (figure 2.1). I Diavite once daily for 4 weeks (n=8 1) I
.. - ... - .- .- - .- .- .. - . -.. .-. I Diavite, 1 rng vitamin B 12 daily for 4 weeks (n=S I) I
Randomization
I 1 I I Diavite, vitamin B 12 I Diavite, vitamin B12 Diavite, vitamin B 12 and placebo daily for and folic acid 5 mg and folk acid 20 mg 4 weeks (n=25) for 4 weeks (n=28) for 4 weeks (n=28) -
placebo daily and Serine 49 b.i.d. for Serine 4g b.i.d. for Serine 4g b.i.d. for 2 weeks (n=8) I I 2 weeks (n=9)
Figure 2.1: Trial profde 43 Research Ouestions: The primary research question was whether short-term therapy with high dose folic acid (5 or 20mg) was more effective than placebo in lowering homocyst(e)ine in vitamin B12 supplemented hemodialysis patients with hyperhomocyst(e)inemiaon maintenance multi-vitamin therapy containing lmg of folic acid per day. There were three secondary research questions: 1) does short-term therapy with lmg of oral vitamin B12 daily lower homocyst(e)inein hemodialysis patients with hyperhomocyst(e)inemiaon maintenance mu1 ti-vi tamin therapy? 2) does supplementation with serine at
0.05g/kg twice per day lower homocyst(e)ine in hemodialysis patients with hyperhornocyst(e)inemiaafter supplementation with folk acid and vitamin B12? and 3) what factors, including baseline clinical characteristics, pre or post RBC folate and vitamin 812 levels, and allocation to placebo/folic acid predict a 50% lowering in saeening homocyst(e)inelevel?
Laboratory evaluation:
In all cases, bloodwork was drawn midweek and predialysis. Plasma homocyst(e)ine levels were measured using a fluorescence polarization immunoassay kit (IMX, Abbott Laboratories, Mississauga, Ontario). The standard deviation of the assay is 0.33umol/lt and the normal range for homocyst(e)ine in our laboratory is 4.9-13.7urnol/l in patients with normal renal
function. RBC folate and vitamin 812 levels were measured using a radiodilution
immunoassay kit (Ciba-Coming, Medicfield, Massachusetts). In our laboratory,
RBC folate and vitamin 812 deficiency are defined by levels Statistical Analvsis: All analyses were performed according to the intention to treat principle and included all enrolled patients who had at least two blood samples collected. Baseline variables were described using the mean (for variables with a normal distribution), median (used only to describe the screening homocyst (e)ine levels for individual groups since they were not normally distributed), or proportions and 95% confidence intervals where appropriate. Differences in baseline variables between the placebo, 5mg and 20mg folic acid groups were compared using one-way analysis of variance for normally distributed continuous variables and the Kruskal-Wallis test for the screening homocy s t (e)ine levels. One-way analysis of variance was used to compare the mean difference in homocyst(e)ine levels before and after treatment with placebo, 5mg or 20mg of folic acid. A paired two-sided t-test was used to compare the mean difference in homocyst(e)ine levels before and after four weeks of therapy with vitamin 812 and also before and after two weeks of therapy with serine. Multiple logistic regression was used to determine the fadon predictive of a 50% lowering in screening homocyst(e)ine level. The variables which were considered included age, number of years on dialysis, diabetes, screening 45 homocyst(e)ine level, baseline and post-treatment RBC folate and vitamin B12 levels, and treatment allocation. Backwards and forwards manual stepwise elimination was performed using existing dinical information and by sequentially removing predictor variables where the p value was >0.10. The sample size calculation was based on our primary hypothesis. With 25 patients in each arm, our study had 80% power at a two-tailed as0.05 to demonstrate a 25% difference in total homocyst(e)ine levels between the placebo, folic acid 5mg and folic acid 20mg groups. In addition, with 30 patients in the serine phase of the trial and using a paired t-test, our study had 80%power at a two-tailed az0.05 to detect a 25% difference in total homocyst(e)inelevels before and after serine supplementation. 23 Results Eighty-one (94.2%)and 78 (90.7%)patients completed phase two and phase three of the protocol, respectively. Two patients were lost to transplant, one was lost to death (pneumonia), two lost interest in the study, and four developed concurrent illness felt to be unrelated to the study medications, but which prompted their withdrawal from the study (two developed constipation, one developed nausea / vomiting due to diabetic gastroparesis, and one noted an increased tendency to dialyzer dotting). The results below are from the 81 patients who had at least two blood samples collected, thus enabling data analysis. Baseline dinical characteristics were similar aaoss the three treatment 46 arms for all variables tested (Table 2.1). No patient was deficient in RBC folate or vitamin B12 at baseline as defined by normal laboratory values. Table 21: Baseline characteristics Overall (n=81) P value comparing (95% CI) placebo, 5 rng and 20 mg folic acid groups Age (years) 63.5(60.1.66.8) Sex (% male) 47.6(36.4,58.9) Diabetes (%) 26.8(17.6,37.8) Years on dialysis 3.6(2.8,4.5) Baseline folate intake 0.70(0.62,0.77) (mg/ day Screening homocy st (e)ine 27.7(25.5,29.8) 0.39 (umol/ 1) Screening RBC folate 1676 (1436,1916) 0.43 (nmol/ 1) Screening vitamin B12 437(326.548) 0.54 @mol/ 1) a Kruskal-Wallis test The etiology of renal failure was diabetes mellitus 26.8%. glomerulonephritis 28.1 %, hypertension 13.4 %, interstitial nephritis (including reflux nephropathy) 8.6%, autosomai-dominant polycentric kidney disease 6.1%. other 7.3% and unknown 9.8% and was similar among the three treatment arms @ = 0.58). Average compliance by pill count was 91.8% and this was supported by a marked elevation in vitamin 812 and RBC folate levels during the trial to 230% and 158%of baseline levels, respectively. In phase 1, screening homocyst(e)ine levels (mean 27.7umol/ 1) decreased by 19.2% (95%CI12.6,25.9%; p<0.001) after four weeks of treatment with daily Diavitem. Patients who were not receiving DiaviteTMor folic add supplementation prior to the study experienced the greatest decrease in 47 homocyst(e)ine after the 4 weeks of treatment with DiaviteTM(-26.0%) compared to those who were receiving DiaviteTMdaily (-126%) prior to the study @=0.027). The decrease in homocyst(e)ine seen in those patients already receiving Diavitem daily may reflect enhanced drug compliance during the monitored phase of the study or regression to the mean. In phase 2 during which 1mg of vitamin 812 was given daily for two weeks, homocyst(e)ine levels were reduced from 22.4umol/ 1 to 18.6umol/l (mean reduction 16.7%, 95 %CI 11.8,21.6 %, p<0.001). During this period, vitamin 812 levels increased, as expected, from 470pmol/l to 893pmol/l@<0.001). Over the first 8 weeks of this study, the combination of daily DiaviteTMand vitamin 812 reduced homocyst(e)ine levels by 32.7% (95%C125.6,39.8%)compared to screening homocys t(e)ine levels. In phase 3, there was no difference in mean reduction of homocyst(e)ine levels after therapy with 5mg or 20mg of folic acid compared with placebo @=0.35)(Table 2.3). This suggests no advantage, in terms of homocyst(e)ine lowering, to supplementation of hernodialysis patients with >Img of folic acid per day in patients already receiving lmg of vitamin B12. When evaluating the placebo and folic add groups together, there was a trend towards a mean decrease in homocyst(e)ine of 5.9% @=0.08) during phase 3 which may have been due to ongoing treatment with vitamin 812 since vitamin B12 levels Table 2.2 Homocyst(e)ine, RBC folate and vitamin 812 levels at week 0,4,8,12, and 14 Week 0: Week 4: Per cent Week 8: Per cent Week 12: Week 14: Screening (after 4 weeks change (Week (after 4 weeks change (Week (after 4 weeks (after 2 weeks levels of Diavitem) 4 compared of vitamin 8 cornpard of placebo or of serine) (95%CI) (9596CI) to Wcek 0) B12) to Week 4) folic acidc) (95%CI) (95 %CI) (95% C1) (n=24) RBC folate 1676 2083 24 % 2286 9.7% 2600 2512 (nmal/ 1) (1436,191 6) (2006,2160) (21 88,2384) (2497,2702) (2337,2688) Vitamin Dl2 437 470 7.6% 893 90% 1009 1065 (pmol/ I) (326,548) (419,520) (812,973) (928,f091) (908,1221) p < 0.001 for comparison of week 4 and week 0 bp < 0.001 for comparison of week 8 and week 4 cplacebo, 5 mg and 20 mg folic acid groups were combined since no difference was demonstrated between the three groups in homocyst(e)ine reduction as shown in Table 3 dp = 0.15 for comparison of week 14 and week 12 continued to inaease during this phase from 890 to 1005 pmol/L @<0.001). After week 12 of the protocol, homocyst(e)ine levels had been reduced overall by 36.1 % (95%CI29.1,43.5%). During phase 4, the additionid treatment with se~etherapy for two weeks did not sigruhcantly reduce homocyst(e)ine levels (-0.62umol/ l,95 %CI - 1.5umol/ l,O.2umol/ 1; p=0.15). Serine levels increased from 0.077 mM (95X Q 0.059mM,0.095mM) to 0.15m.M (95XU 0.093mM,0.21mM) after the two weeks of serine therapy (p Using multiple logistic regression, the only variables that appeared predictive of a 50% lowering in homocyst(e)ine level after therapy with Diavitem and vitamin B12 were an elevated screening homocyst(e)ine level (odds ratio 4.1 for every lOurnol/l increment in screening homocyst(e)ine level, p=0.005) and a longer length of time on dialysis (odds ratio 1.14 for every one year increment in time on dialysis, p=0.12). Adverse events were rare and are noted above for those patients who dropped out of the trial. Of the patients who completed the trial, two patients experienced erythematous rashes, one felt to be due to vitamin 812, and one systemic rash that occurred in a patient while receiving folic add, 5mg. Table 2.3. Homocyst(e)ine, folic acid and vitamin 812 levels during phase 3 of the trial according to treatment allocation Week 8: Post vitamin B12 Week 12: Post placebo / folic acid Treatment Hcys. RBC folate Vitamin 812 Hcys. RBC folate Vitamin B12 Percent allocation (u&ol/ 1) (nmol/l) (pmol/l) (umol/ 1) (nmol/ l) (pmol/ 1) reduction in (95XCI) (95%CI) (95%CI) (95XC1) (95% CI) (95 %CI) Hcys. Placebo 19.4 2421 804 18.4 2477 875 -5.1 %a' (18.3,20.4) (2328,2511) (729,879) (1 7.3J9.5) (2374,2575) (795,955) Folic acid 18.1 2195 966 17.3 2565 1136 -4.4%a 5mg (1 7.2,lg.O) (2074,2316) (883,1049) (16.6,18.0) (2459,2669) (1069,12Gj Folic acid 20 18.5 2258 893 17.6 2743 995 -4.8%~ "95 (1 7.4J9.6) (2191,2323) (814,972) (16.6,18.6) (2641,2845) (908,1082) ahomocyst(e)ine values in week 12 compared with week 8 bp = 0.35 for comparison of percent reduction in homocyst(e)ine achieved with placebo, 5mg and 20 mg of folic acid Hcys.=Total hornocysteine 2.4 Discussion= Previous papers have emphasized the role of folic acid supplementation in the treatment of hyperhomocyst(e)inemia in patients with ESRD. This is the first prospective clinical trial to demonstrate that the addition of oral vitamin B12 to hemodidysis patients who are not vitamin 812 deficient by laboratory standards and who are receiving lmg of folic add, lowers total homocyst(e)ine levels siphcantly. It is unlikely that the 16.7% reduction noted in homocyst(e)ine levels during therapy with vitamin B12 was due to concomitant folic acid therapy since patients had been on a constant dose of folic acid for 4 weeks and RBC folate levels increased only minimally during the four weeks of vitamin 812 therapy (Table 2.2). Moreover, previous work has suggested that hornocyst(e)ine levels reach a new baseline after four weeks of treatment with a new dose of folic acid 161. Our study demonstrates no additional benefit to folic acid therapy above lmg per day in hemodialysis patients who are also receiving vitamin 812. RBC folate levels, which reflect tissue folate stores, increased by only 21.5% after 4 weeks of treatment with 20 mg of folic add. This suggests that, in hemodialysis patients, doses of folic acid above 1mg may not be integrated properly into cells. This may be due to saturation of cell receptors needed for entry of folic acid, or due to saturation of enzymatic pathways needed in the conversion of folic acid to 5-methyltetrahydrofolate '";169. If this is indeed the case, then measurement of RBC methylfolate levels (which more accurately reflect functional folate capacity) may correlate more closely with homocyst(e)ine levels. The results of our study need to be reviewed in light of previous clinical trials that have studied the homocyst(e)ine-lowering ability of folk acid and vitamin B12 in patients with ESRD 157;1"165;167 I6eln. Three uncontrolled trials have examined the use of folic acid (2.5-Smglday), in peritoneal and herno- dialysis patients on no previous folic acid supplementation demonstrating a 30- 55 % reduction in homocyst(e)ine levels with short-term therapy163;167. A recent uncontrolled study involving I1 hemodialysis patients on no folic acid or vitamin 812 supplementation who were administered a multivitamin containing folic acid lmg,vitamin 812 6 ug, and vitamin B6 lOmg (very similar to the multivitamin used in this trial) revealed a 24% reduction in homocyst(e)ine levelslm. In addition, two studies 157;'R were recently completed, including a small randomized trial by van Guldener et a1 157, showing that treatment of patients with either lmg or 5 mg of folic acid resulted in a similar reduction in homocyst(e)ine levels. These results, when considered with the current study, suggest that folic acid therapy above lmg per day has no additional benefits for homocyst(e)ine lowering in this patient population. Two trials have examined the role of vitamin 812 supplementation in patients with ESRD. One small uncontrolled study that enrolled 14 hemodialysis patients with low serum vitamin 812 levels demonstrated a 35% reduction in homocyst(e)ine after monotherapy with four weekly intravenous lmg doses of vitamin 81239. The only other randomized placebo-controlIed study done with 27 hyperhomocyst(e)inemichernodialysis patients compared a combination of vitamin 812 lmg/day and folic acid 15mg/day with vitamin B12 12ug/ day and folic acid lmg/day'fi. After 8 weeks, homocyst(e)ine levels were 26% lower in patients who received the high dsse vitamin combination. This is similar to the reduction noted in our study with vitamin 812 and DiaviteN therapy alone and may represent the effect of vitamin 812, rather than the postulated effect of high dose folic acid. We were unable to demonstrate any rela tionship between individual response to oral vitamin 812 and baseline serum vitamin B12 levels suggesting that serum vitamin B12 levels may not be a reliable marker of vitamin B12 statusln. Elevated methylmalonic acid (MMA) levels may idenhfy more accurately those patients with suboptimal total body vitamin B12 stores who would have a better response (in terms of homocyst(e)ine-lowering)to vitamin 812 supplements. However, as MMA is nonspecifically elevated in ESRD, this issue needs further clarification '74. It is interesting to speculate why vitamin 012 supplementation in patients who were already receiving the recommended dietary allowance of vitamin 812 and who had no laboratory evidence of vitamin 812 deficiency was effective in lowering homocyst(e)ine. Impaired gastrointestinal absorption seems unlikely since serum vitamin 812 levels were elevated It may be that the standard "normal" range reported by most laboratories is too low. Lindenbaurn et d176 have shown that, despite "normal" serum vitamin 812 levels, a si@cant proportion of individuals have physiological vitamin 012 deficiency as defined by an elevation in homocyst(e)ine or MMA levels 176. Alternatively, ESRD patients may have a defect in their ability to convert vitamin 812 (what the laboratory measures) into its active form (hydroxylcobalarnin) which is needed by methionine synthase in homocyst(e)ine metabolism 174. Finally, binding of vitamin 812 to its carrier protein, transcobalamin II, which is necessary for entry of vitamin 812 into tissues, may be impaired in ESRD. Defects in transcobalarnin I1 are associated with normal serum vitamin B2 levels but result in physiological vitamin B12 deficiency as evidenced by elevated homocyst(e)ine and MMA levels 1". At present, however, there is no direct evidence in support of this theory in patients with ESRD. Further research is clearly needed. Although we were able to demonstrate a 36.1% reduction in homocyst(e)ine levels using DiaviteTMand vitamin 812, only 13.6%of patients were able to achieve a normal homocyst(e)ine level((13.7 mol/ 1). Since our study was completed, Touaml69 et al, in an uncontrolled study, used 50mg of intravenous folinic acid (the active form of folic acid) once weekly, to treat hernodialysis patients with hyperhomocyst(e)inemia.They were able to normalize homocyst(e)ine levels in 78% of their patients over 11.2 months of therapy. Unfortunately, a recent randomized study by Bostomln et a1 comparing oral folinic acid with fohc acid did not demonstrate any difference in terms of homocyst (e)ine-lowering in hernodialysis patients with hyperhomocys t (e)inemia. Supplementation of patients with serine at 0.05g/ kg/ twice per day resulted in a minimal non-sigmficant reduction in homocyst(e)ine levels. This may have been due to the inadequate sample size in the serine phase since enrollment was not as large as planned. These results are similar to a hial done in 4 ESRD patients that showed no reduction in homocyst(e)ine after serine supplementation32. It seems unlikely that we missed a clinically important reduction in homocyst(e)ine levels. 2.5 Conclusions This study demonstrates that the current best oral regimen for lowering homocyst(e)ine in patients with ESRD on hernodialysis is lmg of folk acid and lmg of oral vitamin B12 daily. Supplementation with larger doses of folic acid is not therapeutically beneficial. However, other effective therapies need to be developed since only 13.6% of patients obtained normal homocyst(e)ine levels with this treatment, Whether treatment will result in a reduction in cardiovascular risk for ESRD patients requires elucidation by further clinical trials. CHAPTER 3 PHARMACOLOGICAL SUPPLEMENTATION WITH VITAMIN &? DECREASES BUT DOES NOT RETURN HOMOCYSTEINE AND METHYLMALONIC ACID (MMA) TO CONTROL LEVELS IN PATIENTS WITH END-STAGE RENAL DISEASE; A POSSIBLE LINK WITH GLYCINE METABOLISM. M. Eric Hyndman, Braden Manns, Floyd Snyder, Earnest Fung, Nairne Scott- Douglas, Howard G. Parsons. Departments of Pediatrics' Medical Genetics2 and Medicin@# University of Calgary, Calgary, Alberta, CANADA 3.1 Introduction Moderate elevations of total plasma homocysteine are an independent risk factor for cardiovascular disease in patients with and without ESRD In. The prevalence of hyperhomocysteinemia (85-100%)and death from atherosclerotic vascular disease (2540%) in ESRD is sigruhcantly greater compared to patients with normal renal function 146. Several attempts have been made to lower plasma homocysteine levels in patients with ESRD however the results of such interventions have been less than favorable. Supplementation with folk acid or it's active form, methyltetrahy drofolate (with or without cobalamin) are only partially attenuate hyperhomocyteinemia in ESRD 39;159:169;179;1s0. Homocysteine is generated by the demethylation of the essential amino acid, methionine through the intermediates Sadenosyl-methionine and S adenosylhomocy steine. Once formed homocysteine can either be remethy lated to methionine via the remethylation pathway or irreversibly converted to cysteine though the transsulfuration pathway. Hyperhomocysteinemia in ESRD is primarily a result of a defective remethylation and not transsulfuration Im.Both vitamin B12 and 5-methyltetrahydrofolate are essential cofactors required for remethylation of homocysteine to methionine. Remethylation in most tissues is catalyzed by methionine synthase (MS). The methyl group required for the reaction is donated by 5-methyltetrahydrofolate, which is generated by methylenetetrahydrofolate reductase v).MS uses the cofactor cobalamin (vitamin 812) to transfer a methyl group horn 5-methyltetrahydrofolate to homocysteine forming methionine and tetrahydrofolate. Severe and mild genetic mutations can independently or in combination with nutritional factors contribute to hyperhomocysteinemia. A common substitution of cytosine for thymidine at nucleotide 677 in the MTHFR gene has been reported". This substitution results in increased heat labiality and decreased MTHEX enzyme activity. Furthermore, in patients with a low plasma folate level, this mutation leads to an increase in plasma homocysteine levels. Although the MTHFR 677 C+T mutation has been implicated as an aggravating factor182 its prevalence does not explain the high incidence of hyperhomocysteinemia in ESRD. In addition to being a cofactor for methionine synthase, vitamin B 12 is required for the catabolism of rnethylmalonic acid, an internediary in the breakdown of the amino acids methionine, isoleucine and valine, odd chain fatty acids and cholesterol. The mitochondria1 protein, methylmalonyl-Co A-mutase (MMM) uses adenosylcobalamin in the isomerization conversion of methylmalonylCo A to succinyi-Co A and unlike methionine synthase its activity is independent of folate. Elevations in MMA are therefore a useful marker of vitamin &2 tissue deficiency ls3-185. The vast majority of MMA is metabolized and does not escape the mitochondria '86;lQ. In addition to vitamin BIZdeficiency, mutations in the &2 dependent enzyme MMM result in elevated concentration of MMA in the blood and urine 188. In the latter case, plasma glycine levels are abnormally high. The elevation in glycine is thought to be a result of decreased catabolism by the glycine cleavage system, due to insufficient transport of glycine into the mitochondria or through a direct inhibition of the glycine cleavage system. MMA inhibits both the transport 52~and the catabolism of glycin@;45;189. Previous studies have measured MMA in patients with end stage renal failure and have consistently documented elevated levels of this compound 1s;lw;191. Furthermore, ESRD patients also have sigtuhcantiy elevated glycine levels. Clearly the elevation in MMA is a metabolic deficiency rather than a genetic disorder however treatment with vitamin B 12 has been relatively ineffective. One study concluded that supplemental BIZdid not reduce MMA levels in ESRD subjects '90. In contrast a recent study by Diekes et al. s9 in a similar population determined that intravenous B12 supplementation sigruhcantly decreased but did not normalize MMA. A more detailed re- investigation by Moelby et al revealed that patients on dialysis require large supplementation with cyanocobalamin to maintain a constant stable level of both serum 812 and MMA levels. The pathophysiological mechanism(s) that result in both homocysteine and MMA elevation in the majority of rend failure patients is not well defined. Targeting the remethylation of homocysteine with folate and BIZare extremely effective in patients with normal renal function. However, in patients with ESRD, high dose combined folate and B12 therapy only partially attenuates the elevated MMA and homocysteine levels; these parameters still remain well above the control values. The aims of this study were two fold. First, to examine the influence of the 677C+T mutation (in MTHFR enzyme) on baseline homocysteine levels. Furlhermore, to explore the role of this mutation on homocysteine and MMA in hernodialysis patients receiving maintenance multi-vitamin therapy, maintenance vitamins plus high dose vitamin 812, and the latter plus 5 or 20 mg of folate. Our second aim was to examine the relationship between plasma MMA and plasma glycine levels. 3.2 Methods Subjects: One hundred and fifty patients with ESRD from a single tertiary care center in Calgary, Canada were screened for entry into the study. Exclusion criteria induded: plasma vitamin 812 < 133 pmol/L, red cell folate concentration < 450 nmol/L or a plasma total homocysteine < 16 umol/L. Subjects were also excluded if they were currently receiving or had received treatment with vitamin &2 injections within the last 3 months, more than lmg of folic acid per day in the past 6 months, or if they were currently using anti-folate or anti-epileptic medications (i.e. phenytoin). Of the patients screened, eighty one were encoued in the study since twenty-six patients did not have hyperhomocysteinemia (>16umol/L) and fifty were unable or un-g to participate. At baseline, 92.6% of the enrolled patients were taking supplemental DiaviteTM.Diavitem contains folk add 1mg, vitamin &2 6ug, vitamin 86 3mg and small amounts of other vitamins (Diavite R & D Laboratories, Marina del Ray, CA). Of those who were taking DiaviteN, 56% used it daily and 44% used it three-times per week pos t-dialysis. Studv Protocol: The study was divided into three phases of one month each. The first phase was to ensure standardization of conventional vitamin therapy. During Phase 1all enrolled subjects were supplemented daily with DiaviteTMfor four weeks. Phase two of the trial consisted of all subjects ingesting Diavite plus an additional oral supplementation of vitamin 812 (Nature Made@,Phannavite Corporation, Mississauga, Canada) at a dose of lmg/day for four weeks. In phase 3, subjects continued to receive Diavitem and vitamin B12 daily, and in addition, they were randomized (using a computer-generated sequence) to receive either placebo, folic acid 5mg or folic acid 20mg/day (Apotex hc., Toronto, Canada) for a period of four weeks. Folic acid/placebo capsules were dispensed in serially numbered containers and were identical in size, shape and color. The Conjoint Health and Research Ethics Board at The University of Calgary approved the study protocol. Plasma vitamin &2 and MMA, serum total homocysteine and RBC folate were measured prior to dialysis at the termination of the three study phases in all subjects. The plasma glycine concentration was determined at the end of the third phase only. Biochemical measurements: Plasma homocysteine levels were measured using a fluorescence polarization imrnunoassay (IMX,Abbott Laboratories, Mississauga, Ontario) as previously described154. The standard deviation of the assay is 0.33umoi/l, and the normal range for homocysteine in our laboratory is 4.9- 13.7umol/l in patients with nonnal renal function. RBC folate and serum vitamin B12 levels were measured using a commercial erythrocyte folate assay kit (Bio- Rad Corp., Mississauga, On)and a radiodilution immunoassay (Ciba-Coming, Medicfield, Massachusetts). In our laboratory, RBC folate and vitamin B12 deficiency are defined by levels <450nmol/l and <133pmol/l, respectively. Plasma was precipitated using 5-sulfosalicylic acid and the treated plasma samples were assayed for glycine using an automated amino add analyzer. MMA was recovered from plasma using solvent extraction using methyl-2H- malonic acid MMA as the internal standard and quantified using a previously described gashomatography-mass spectrometry stable isotope dilution methodl92. DNA was isolated from peripheral lymphocytes for the MTHFR 677C+T genotyping, and amplified using the primers and procedures of Frosst et a1.55. Statistical Analysis: Data are presented as medEM. One way analysis of variance with Scheffe post-hoc analysis was used to compare mean difference in homocysteine, 0 12, and MMA at the 4,8 and 12 weeks of study. Analysis of homocysteine at baseline was performed on the naturai logarithmic values because the data was not normally distributed. The above variables at 8 and 12 weeks were similarly compared for the placebo group and the 5 and 20mg folate treatment groups respectively. The results were determined to be signihcant if the p value was less that 0.05. Correlation coefficients between MMA and plasma glycine were determined using Pearson's coefficient and simple Linear regression analysis. 3.3 Results Eighty-one (94.2%)and 78 (90.7%) of subjects completed phase two and phase three of the study respectively. The mean and 95% confidence intervals (CI)of age and length of hemodialysis of the subjects were 63.5 (60.1,66.8)and 3.6 (2.8,4.5), respectively. The mean and 95% CI for RBC folate and serum B12 on entry to the study were 1676 (1436,1916) and 437 (326,5481, respectively. The etiology of renal failure was diabetes mefitus (26.8%), glomerulonephritis (28.1%), hypertension (13.4 %), interstitial nephritis (including reflux nephropathy) (8.6 %), autosomal-dominant polycy stic kidney disease (6.1 % ), miscellaneous (7.3%)and unknown (9.8%).The distribution of the MTHTR 677C-T was 51 % (n-4),37% (n=30), and 11 % (n=9) of the patients for the C/ C, C/T and T/T genotypes respectively. There was no statistical differences in homocysteine levels during the screening arm of the study. The mean homocysteine levels for the entire group on entry to the study was 27.7umol/L (95 %U 25.5,29.8). When segregated according to MTHFR genotype the homocysteine levels were higher in the TIT genotype but it did not reach a significant level (Table 3.1). Table 3.1 Mean and standard deviations of the screening levels of plasma homocysteine stratified by the MTHFR 677 C+T genotype. MTHFR 677 C-+T genotype Homocysteine umol/l at Screening 28.25 (9.3) 25.4 (6.6) 32.5 (17.1) Plasma homocysteine and methylmalonic acid levels after 4,8 and 12 weeks of treatment are shown in table 3.2 With daily DiaviteTMtreatment, the mean homocysteine of all subjects fell by 19.2%.Interestingly, during Phase I, while the homocysteine fell sigruhcantly in all MTHFR genotypes, homocysteine levels remained significantly higher in the MTHFR 677 T/T subjects compared to the C/C individuals @<0.03) despite treatment with Diavite (which contains lmg of folate). As is indicated in chapter 2 vitamin 812 and folate Levels were well above the deficient levels. Table 3.2 Methylmalonic acid and homocysteine levels during Phase I (DiaviteN), Phase I1 (B12) and Phase 111 (fohte/placebo) stratified by the MTHFR 677C+T genotype. Methylmalonic Acid (pmol/ L) Homocysteine (umol/ L) Phase I 785 (319)* 747(261)* 889(222)* 21.2(4.88)* 22.4(4.?) * 27.6(10.7)Ys (Baseline) Phase II 623 (246) a(226) 671 (113) 18.4(4.52) 18.9(4.5) 18.5(5.3) (Post B12) Phase I11 613 (263) n=15 561 (116)n=6 665(240) n=3 17.1(4.2)n=15 18.2(2.0)n=6 25.3(8.0) n=3 (Mac.) Phase I11 (5mg or 20 mg 618 (240) n=26 594(198) n=22 649(69) n=5 17.3(3.7)n=26 17.3(3.7)n=26 17.2(3.8)n=5 fola te) -- -- * p<0.05 ANOVA and Scheffe post hoc analysis Vs. Phase two (812 supplementation) $ pc0.03 ANOVA and Scheffe post hoe analysis T/T Vs C/C at baseline (log transformed data) During phase II (DiaviteTM + 1mg b2) the vitamin &2 level, as expected, increased from a mean of 470 pmol to 893 pmol/L @-0.001). With the increase in BIS homqsteine decreased in the entire group by 16.7% (95 CI%; 11.8,21.6 p<0.01). When the mean difference between Phase I and II were examined it revealed that the T/T subjects responded significantly better to B12therzpy (p<0.01 vs. C/C or C/T). This in turn, could be attributed to the highest homocysteine level present in those homozygous for the T/T MTHFR genotype prior to &2 therapy. In phase III, after therapy with 5 or 20 mg/day, there was no sigruficant reduction in the homocysteine level in any of the MTHFR genotypes. The ongoing treatment with vitamin B12 in phase caused an increase in vitamin 812 from 890 to 1005 pmol/L (p<0.001) The MMA levels, stratified for MTFHR genotype are depicted in table 3.2 After DiaviteTM treatment, MMA concentrations in the ESRD patients were 7 to 10 fold higher than the 90% percentile of our laboratories normal range (84-115 nmol/L). There was not a statistically signrhcant difference in the MMA concentrations among the MTHFR genotypes during any of the phases. Only additional supplementation with lmg of B12 (phase 11) lowered MMA. However the MblA level was not normalized in any subject during the treatment phases. Before treatment with vitamin there was a highly sigruficant negative relationship between BIZ and MMA (r = -0.04; p= 0.02). However B12 therapy abolished the correlation since the rise in serum Bl2 was not accompanied by a change in MMA levels. Because of the importance of the glycine c!eavage system on the methylation of folate we investigated the relationship between MMA and glycine. As illustrated in figure 3.1, a significant positive correlation between plasma MMA and glycine (r=O. 42; ~~0.03)was observed. - 1 I 1 1, .c 1.5 a 0 I - d 0 E3 s 0 z 0 .5 - C - 0 -, I 1 I 0 .2 .4 .6 Glycine mmol1L Figure 3.1 Relationship between glycine and MMA levels in ESRD patients (Phase III) 3.4 Discussion We have previously demonstrated that high dose B12 levels and folate do not -- normalize homocysteine levels in ESRD patien&. Data from the present study reveals that this treatment does not normalize MMA levels in ESRD either. In agreement with previous studies '93 we found that the MTHFR T-allele aggravates hyperhomocysteinemia but it's effects on homocysteine levels appear to be relatively minor compared to the metabolic complication associated with ESRD. In our study a folate intake greater than 1mg/day was not beneficial in lowering homocysteine levels in any of the C+T 677 MTHFR genotypes. Nevertheless, high dose vitamin B12was most effective in lowering homocysteine in the subjects with a homozygous mutation in MTHFR. We also present evidence that elevation of MMA is common in ESRD subjects. Unlike the study of Moelby et a1190 and in concurrence with others we did observe a sigruficant reduction in MMA levels after BIZtherapy. This study supports the notion that ESRD patients required high pharmacological doses of BIZto maintain stable levels of MMA and homocysteine. There was no evidence of preferential lowering of MMA among the 677 C+T MTHFR genotypes. High levels of MMA have previously been demonstrated to interfere with the glycine cleavage system. Our study is the first to demonstrate that MMA levels are positively correlated with serum glydne levels which may suggest a possible mechanism linking MMA to homocysteine metabolism in ESRD. Interestingly there was a strong correlation between senun Bt2 and plasma MMA levels prior to high dose B12 therapy but not after therapy. Nevertheless, despite the high serum B12 levels, further supplementation with pharmacological doses of 812 signrficantly lowered MMA concentrations. This response to B 12can be taken as evidence for a metabolic B12deficiency. The later finding strongly suggests that the plasma B12 level is a poor indicator of tissue availability of &2 in uremic subjects. Several possibilities for this finding exist and two events in the regulation of B12 metabolism need to be discussed. The fact that both homocysteine and MMA remain elevated suggest a primary defect in the cellular metabolism of B12 Uremia may interfere with B12 binding to serum transcobdamin I or IX. This could account for the high serum B12 levels and the evidence of a cellular deficiency in &2. Genetic lesions in transcobalamin are associated with normal serum 812 values and elevated plasma homocysteine/MMA concentrations. However, transcobdamin is reported to be normal in chronic renal failure 1%. The second possibility is that uremia alters an early reduction step in B12metabolism. Impairment at this level would result in both methylmalonic aciduria and homocysteinemia. Indeed this is the case in the in born errors of metabolism involving the cblC, cblD, and cblF genotypes (figure 3.2). Mtrbylmabnyl CoA Muruc Mrthylmdonyl CoA 4 Succinyl CoA Adcnosy lcobalamin /' cblB Cob(I)rhin I?cb~ Cob(I1I')almin MrrOCHONDRION 7lmin +&~ob(~ami~AdoMct M~I I-THF cblF cblE Mcthylcobolomin Cob(I)olamir Homocystcinc Metbionlae Synttuse Cob(1Il)ahin cblG Figure 3.2 Intracellular Vitamin Bn metabolism Prior to acceptance of an adenosyl or methyl group vitamin Bl2is reduced to hydroxycobalamin. A defect in reduction of BIZwill affect the binding of adenosyl and methyl to 812 and hence the binding to methylmalonyl CoA mutase and methionine synthase, respectively. The formex would result in an increased level of MMA and the latter an inaease in homocysteinemia. Treatment of the ESRD subject with hydroxycobalamin would help define the role of reduced intracellular metabolism of Bl2 in ESRD. In addition, an evaluation of the specific forms of cellular and circulating forms of cobalarnin might reveal a dysfunction in the metabolism of B12. Inherited deficiencies of methylrnalonyl-CoA mutase not only result in elevated MMA but sigruficantly affect glycine 1evels~;~~;~~g.Prior to the identification of MMA the disorder was classified under a group of disorders known as ketotic hyperglycinemia. Likewise ESRD patients also have elevated glycine and MMA levels. In contrast to the elevated levels of glycine and MMA, serine levels are depleted in patients with ESRW2. Both serine and glycine are indirectly linked to homocys teine metabolism through folate. Cy tosolic serine hydroxymethyl traiiferase (cSHMT) adds a methelene group to tetrahydrofolate (from serine) producing 5-10-methylenetetrahydrofolateand glycine. MTHFR uses 5-10 methylenetetrahydrofolate as a substrate to generate 5- methyltetrahydrofolate, which is then subsequently transferred to homocy steine to form methionine. The importance of serine in the remethylation of folate is highlighted by a study conducted by Wilcken et al.56, who demonstrated that supplementation of folate in uremic patients caused a reduction in serine and a coxresponding inaease in glycine levels. Serine-glydne hydroxymethyltransferase is a reversible enzyme. 5-10 methylenetetrahydrofolate is able to donate a methyl group to glycine, generating serine. In addition to &HMT there is also a mitochondria1 serine hydroxymethyltransferase (mSHMT) capable of generating or consuming 5-10 methylenetetrahydrofolate. However, in the mitochondria, the glycine cleavage system (GCS)catabolizes glycine into carbon dioxide, ammonia and 5-10- methylenetetrahydrofolate using tetrahydrofolate as the substrate. Therefore, additional methyl groups can be generated by GCS to form 5-10 methylenetetrahydrofolate and therefore drive serine synthesis from two glycine molecules 195. A recent study by Verleysdonk 40 has demonstrated that glycine catabolism directly produces serine. Furthermore, the much higher reaction rate of GCS would favor serine production by rnSHMT. MMA and its precursors have been shown to both inhibit the transport of glycine into the mitochondria 52 as well as directly impair the glycine cleavage system a. Therefore an inhibition of the GCS coupled with the loss of serine production by the kidney may impair the remethylation of folate and contribute to hyperhomocysteinemia. We found a positive correlation between glydne and MMA levels in our population (Figure 3.1) indicating that the elevation of MMA seen in ESRD is capable of interfering with glycine catabolism and therefore may alter intracellular and more specifically mitochondria1 folate. One limitation of this study is that we only measured glydne levels after all treatment phases were complete. We could therefore be underestimating the effect of MMA on glycine. It would be worthwhile to assess the relationship between MMA and glycine in acute renal failure or before vitamin therapy to shed light on this issue. Two recent studies support an impaired conversion of folic acid metabolism to 5-methyltetrahydrofolatein subjects with ESRD. Both intravenous folinic acid '69 and 5-methyltetrahydrofolate restored plasma homocysteine to normal levels in the majority of hernodialysis subjects. Folinic acid, the formyl form of tetrahydrofolate, will bypass any abnormality in glycine metabolism and is rapidly polyglutamated and incorporated into one carbon metabolism. Not all studies support this conclusion. Bostom et al.177 were not able to duplicate these findings. Both conclude that the mild homocysteinemia in hernodialysis subjects was refractory to treatment with N5 methylfolate or folinic acid respectively. Hence, the hypothesis of a defective folate metabolism needs to be pursued in ESRD studies in order to address whether an aheration in cobalamin metabolism should also be investigated. 3.5 Conclusion In conclusion, our study supports previous evidence that subjects receiving hernodialysis not only have hyperhomocysteinemia but also MMA aciduria. The MTHFR genotype, although moderately affecting homocysteine, is secondary to the metabolic complications. MMA levels correlated with serum glycine concentrations and we therefore proposed this as a mechanism, which may interfere with the methylation of folate by inhibiting the mitochondria glycine cleavage systern. Further investigations are needed to clarify this rela tionship. Clearly MMA levels within the uremic population should not be ignored and we would recommend that high dose BIZ supplementation be incorporated in addition to folic acid therapy to treat hyperhomocysteinemia. Alternatively, treatment with hydroxy or adenosyl-cobalamin rather than cyano-cobalomin might bypass and clanfy the altered BIZ metabolism. CHAPTER 4 EFFECTS OF COMMON ONE CARBON METABOLISM POLYMORPHISMS ON HOMOCYSTEINE,RBC FOLATE,VITAMIN Bu AND METHYLMALONIC ACID LEVELS M. Eric Hyndman, Ed Les, Floyd Snyder, Peter Bridge, Earnest Fung, Howard G. Parsons. Departments of Pediatrics' Medical Genetics2 and Medicin& University of Calgary, Calgary, Alberta, CANADA 4.1 Introduction Mild to moderately elevated plasma homocysteine levels are common in the general population, and have been implicated by many studies as an independent risk factor for atherosclerosis~;66;68.In the evaluation of patients with cardiovascular disease, assessment of plasma homocysteine levels is now becoming increasingly important. While identification of hyperhomocysteinemia seems appropriate, prospective clinical trials designed to lower homocysteine levels are required to demonstrate the importance of targeting homocysteine in terms of cardiovascular morbidity. Furthermore, understanding the etiology and mechanisms of hyperhomocysteinernia is crucial since it may serve as the basis for the development of treatment strategies to counter this important cardiovascular risk factor. Homocysteine, a sulfur-containing amino acid, is produced by demethylation of the essential amino acid methionine. Once formed homocysteine can be either remethylated to methionine or transsuhated to cysteine 23. In most tissues reme thylation of homocysteine to methionine occurs with 5-methyltehahydrofolate (5 Q-F3-THF) acting as the methyl donor. The process requires three independent enzymes, methionine synthase (MS), me thionine synthase reductase, (MTRR) and me thylenetetrahy drofolate reductase (MTHFR) 1%. MTHFR catalyzes the conversion of 5-10 MTHF into the active 5-h4THF form. The methyl group of 5-CFb-THF is then transferred to BIZ bound to methionine synthase. Hence, both ICH-THF and cobalamin are required for remethylation of homocysteine. However, spontaneous oxidation of MS bound cobdamin (I)inactivates the enzyme and AKRR is required for reactivation of MS 1". Severe mutations affecting these pathways contribute to impaired folate metabolism and hyperhomocysteinemia 1%. The incidence of the severe genetic lesions is low and therefore cannot account for the prevalence of moderate hyperhomocysteinemia in populations with cardiovascular disease. However three common mild mutation affecting MTHFR, MS and MTRR have been reported. A common substitution of cytosine for thymidine at nucleotide 677 in the MTHFR gene results in increased heat labiality, decreased enzyme activity and elevated plasma homocysteine levels"57 in those with a plasma level of foolate less than the median of the control values~sg.In addition, to this mutation, a point mutation at nucleotide 2756 in MS involving substitution of an adenine for a guanine residue has been reported. Recently, a mutation involving the replacement of aspartate by a glycine residue has been identified I%;'*. In another such mutation, the introduction of a guanine residue in place of an adenine residue at nucleotide 66 causes the replacement of isoleucine by a methionine residue in MTRR. Although, the effects of the T alleIe on MTHFR enzyme activity has been described25 the effects of the MS 2756 A+G or the MTRR 66 A+C mutations on enzyme activity are not dear. Cobalamin (&$ in addition to being a cofactor for rnethionine synthase is also a cofactor for L-methymalonyl-CoA mutaselal". When tissue levels of cobaIamin are low L-methylmalonyl-CoA activity is reduced resulting in elevated methyhalonic acid (MMA). Senun or plasma MMA levels have been proposed as a functional marker of cobalamin status since they reflect intracellular cobalamin status200. Furthermore, the level of MMA is not dependent upon fola te metabolism (unlike homocy steine) 200. Both fasting and methionine load induced homocysteine levels have been used to idenhfy patients with hyperhomocysteinemia. However, the role of the methionine load test to detect individuals at risk is still debated47 and most reports of hyperhomocysteinemia measure fasting homocysteine only86. As the methionine load test is time consuming and expensive further clarification of its role in identifymg those with mild hyperhomocysteinemia is needed. In the present study, we determined the incidence of hyperhomocysteinemia determined by a two-hour methionine load test versus the measurement of fasting homocysteine levels. Furthermore, we examined the effects of the common polyrnorphisms affecting the remethylation of homocysteine on plasma homocysteine, folate, B~Land MMA levels, Third we defined a low baseline level of MMA in our population, and developed a cut off level for B12 deficiency. Lastly, we examined the correlation between 812, homocysteine and MMA above and below the cut off level for BIZ deficiency. 4.2 Methods Subjects: One hundred and twenty-four patients, ranging from 37 to 74 years of age, were enrolled from a multi-center study examining the impact of lipid lowering therapy on recurrent cardiovascular events? Each patient had a documented myocardial infarction 0,defined as the presence of 1) chest pain consistent with MI, and 2) electrocardiogram changes consistent with MI? None of the patients had hypercholesterolemia (total cholesterol >62 (mmol/ L), renal, hepatic, or thyroid dysfunction. Twenty-seven volunteers in whom coronary angiography had been undertaken for atypical chest pain, and in whom coronary arteries and left ventricular function were normal, were recruited as controls for the fasting and methionine load homocysteine levels. These subjects did not have any renal, hepatic or metabolic disease. Plasma vitamin Biz, plasma MMA, RBC folate, and fasting and post- methionine load plasma homocysteine levels were measured in all study groups. The methionine load was administered according to the abbreviated oral methionine-loading test developed by Bostom et al*. Briefly, phlebotomy was performed after an overnight (12hour) fast, and at 2 hours after receiving an oral load (0.1 g per kg of body weight) of L-methionine in 200 mL of fruit juice. After the fasting phlebotomy and oral methionine load, subjects received a standardized snack low in methionine ( Plasma homocysteine concentrations were determined using SBD-F (7- benz~20xa-l-3-diazole-4-sulfonicacid) as previously described m.This method measures the total (free plus protein bound) homocysteine concentration. Serum B12 and erythrocyte folate were assayed by radiodilution assay using commercial radioimmunoassay kits (Ciba-Corning, Medfield, Mass) and erythrocyte folate assay kits (Bio-Rad Corp., Mississauga, ON). In our hospital laboratory, RBC folate and cobdamin deficiency are defined by levels <450nmol/L and 433 pmol/L, respectively. MMA was recovered from plasma using solvent extraction and rnethy1-W malonic acid as the internal standard and quantitated using a previously described gas - chromatography - mass spectrometry stable isotope dilution method*. DNA was extracted from peripheral lymphocytes. MTHFR 677 C+T genotyping was performed as previously described by Frosst et al.55. Methionine synthase 2756 A+G geneotype was performed as described by Leclec et al.". The MTRR analysis was determined using the forward primer 5'CCA AAG GCC ATC GCA GAA GAC AT-'3 and the reverse primer SGTG AAG ATC TGC AGA AAA TCC ATG TA-'3. The amplification process introduces a cut site for the restriction enzyme Nde I in the 66 bp product which is abolished by the G allele(appendix A). Statistical analysis: Continuous variables are described as median and twenty- fifth and seventy-fifth percentiles. An elevated homocysteine was defined as greater than the 80th percentile of our control population. The methionine load response was calculated as the difference in the hon~ocysteineconcentration at baseline and 2 hours after the methionine load. Our control fasting and methionine load 80th percentile homocysteine levels are similar to that reported by others'? An abnormal fasting, post methionine load, and methionine load response was defined as 13.86umo/ L, 31.88umol/ L and 21.04umol/ L respectively48. Correlation coefficients between MMA and Serum &2 levels were determined using Pearson's coefficients. A €312 deficiency was defined as a level <258 pmol/L. This value was chosen as it was similar to that proposed by previous authors 51. These authors suggested that cobalamin therapy in subjects with a serum cobalamin concentration up to but not beyond 258 pmol/L is associated with a reduction in serum MMA. Having established our cut off point for cobalamin deficiency at < 258 pmol/L we next determined the MMA concentrations in those subjects with a cobalamin level greater than 258 pmol/ L. We chose the 90th percentile of this value as our low baseline MMA value. This value (0.84 pmol/L) is in close agreement with a low baseline level reported by Naurath et alg subsequent to 8 injections of 1mg of LM. cobalamin over a 21-day period. Difference in means for continuous variables was assessed using either the Kruskal-Wallis test, one way analysis of variance followed by Scheffe's post- Hoc test, or the two sample Wilcoxon rank sum test, where appropriate. Results were considered to be sigruhcant if the two tailed p-values were less than 0.05. All analysis were conducted using the statistical and data analysis program STATA. 4.3 Results: Subiects Of the 124 CAD disease patients 109 agreed to complete the full study including the methionine load test. The majority who declined did so because of the time commitment for the methionine load test or the distance needed to travel to participate in the study. None of the CAD patients or the control subjects had any adverse effects from the methionine load test. As there were no differences between the treatment (lipid lowering therapy) and placebo groups in fasting, methionine load induced hyperhomocysteinemia, RBC folate, serum BIZor plasma MMA, the results were combined and reported as a single patient group. The mean age of the patients in both groups was comparable (CAD group were 57.%10 vs. controls 54. W 2.3, p>0.05). Biochemical rxwarneters: For the entire CAD group the median (25h and 75th percentiles) for fasting homocysteine, methionine load and the methionine load response were 11.55 (9.17,15.1), 27.14 (23.2,31.67)and 14.8 (11.8,17.7) umol/l. Overall, using the 80h percentile as an abnormal fasting and methionine load response, hyperhomocysteinemia was observed in 36% and 13%of the CAD patients respectively. While 13%(14 of 108) had an abnormal methionine load response only 5% (5 of 108) were not identified by an elevated fasting homocysteine. Median, 25th and 75th percentiles for vitamin 812, MMA and RBC folate were 239 (185-321) pmol/L, 0.075 (0.06-0.11) umol/L and 911 (714-1104) nmol/L, respectively. Interestingly, only one person was classified by our hospital laboratory as being deficient in folate (<410nmol/l) where as 7 patients were deficient in vitamin &2 (< 130 pmol/L). For the entire population RBC folate was significantly related to vitamin Bn levels (r=0.35 p<0.001) but not with homocysteine levels (I=-0.1 7;p<0.1). In a search for a functional BIZ deficiency we next investigated the relationship between &2 and the BI~-dependentmetabolite MMA. There was no clinical evidence of B12 deficiency from either a neurological or hematological standpoint. As stated above a vitamin BIZ deficiency was defined as a serum B12<258 pmol/L. Mean MMA and homocysteine levels were significantly different between the patients above and below a vitamin 812 of 258 pmol/L @<0.01).The median, 25th and 75th percentiles for MMA when BIZwas less than and greater than 258 pmol/L were 0.098 umol/L (0.68,0.15) and 0.072 umol/L (0.520.08) respectively. The 90th percentile of MMA in the patients with a B12 greater than 258 pmol/L was 0.084 pmol/l. No sigruhcant correlation between serum MMA and &2 concentrations (above and below 258 pmol/L) could be observed (figure 4.1). Homocysteine levels were 13.3 umol/L (10.1,16.6)and 10.6 umol/L (8.6, 14.2) in the patients with 812less than and greater than 258 pmol/L respectively. However, unlike MMA, a negative correlation (r= -0.29, p=0.03) did exist between homocysteine and BIZwhen the B12 level was <258 pmol/L (figure 4.2). Like MMA no correlation was found between homocysteine and B12 below 258 pmol/L. Serum Bl2 pmoVL Figure 4.1 The relationship between MMA and B12 in the CAD patients. There was not a sigdicant correlation between the MMA and Serum 812 above or below a 812 level of 258 pmol/l. Since the relationship between MMA, B12 and homocysteine were not linear we also investigated the relationship in the logarithmically transformed variables. There was a highly sigxuficant relationship between MMA and 812 (I= - 0.51; p 1 I 2 2.5 Log B12 Figure 4.3. Log-Log plot examining the relationship between MMA and vitamin B12. Log BIZ Figure 4.4 Log-Log plot examining the relationship between homocysteine and serum B12levels. Genotv~e-Nutrientinteractions: MTHFR thermolabile genotyping of the CAD patients indicated 42% to be C/C,43% C/T, and 16%TIT. Genotyping of the MS 2756 A+G polymorphism revealed that 7l%, 27 % and 0.87% had the A/ A, A/G and the G/Ggenotypes respectively whereas 19%,48% and 33% of the CAD population were A/A, A/G and GIG for the MTRR 66 A+G polymorphism respectively. Patients with the MTHFR TIT genotype had the highest fasting homocysteine levels and were signhcantly greater (p<0.001)than the C/T and C/Cgenotypes. The homocysteine level in response to the two-hour methionine load when segregated by MTHFR thermolabile genotype revealed similar results as the baseline homocysteine level (Tablel). When we focused on the difference between the baseline homocysteine and the value 2 hours after the methionine load the difference between MTHFR thermolabile genotypes was abolished. There was no significant difference between fasting, methionine load or the methionine load response in any of the genotypes for the MS 2756AjG (Wilcoxon rank sum p=0.42). Similarly the Kruskal-Wallis test revealed no differences between the MTRR 66A+G genotypes p=O.W (Table 4.1). Table 4.1 Median fasting, post-methionine load, and load minus fast homocysteine concentrations for the CAD patients subdivided by their MTHFR 677 C+T, h4S 2756 A+G and the MTRR 66A+G genotypes Genotypes Fasting homocysteine Methionine Load Methionine (mol/ 1) (~ol/1) load-Fas ting (umol/ 1) MTHFR 677 C+T C/C n=47 C/T n=42 T/T n=19 MS 2756 A+G A/ A n=76 A/G n=32 G/Gn=l MTEtR 66 A+G A/A n=19 A/G n=50 G/G n=34 *P The concentrations of vitamin El-12, RBC folate and MMA were not siphcantiy different between the various MTHFR 677 C-T, or the MTRR A+G genotypes. Similarly there was no difference between the MS 2567 A+G genotypes and MMA or vitamin BIZ however there was a sigruhcantly higher RBC folate level in the subjects with the A/G genotype compared those with the A/ A genotype (Table 4.2). Table 4.2 Median serum Bit plasma MMA, and RBC folate levels for the CAD patients subdivided by their M'TKFR 677 C+T, MS 2756 A+G, MTRR 66 A-BG genotypes. Genotype skum 812 Plasma MMA RBC Folate MTRR 66 A+G *p<0.05 Two-sample Wilcoxon rank sum test A/G Vs A/ A. 4.4 Discussion The primary purpose of the present study was to contrast the ability of 2 methods of identifying patients with mild elevations in homocysteine levels: fasting plasma homocys teine vs. me thionine-induced hyperhomocysteinemia. We determined that the homocysteine response to a 2-hour methionine load only marginally increased the prevalence of mild hyperhomocysteinemia in our population. The fasting homocysteine concentration identified 36% of our patients as having hyperhomocysteinemia and the methionine load response increased this to 41 %; differences that were not statistically sigruhcant. Furthermore, the 2-hour methionine response was not useful in ascertaining any differences in homocysteine metabolism in subjects stratified by the MTHFR 677 C+T, MS 2756 A+G or the MTRR A+G genotypes. Our results would suggest the 2-hour methionine load test needs further evaluation before recommending it as a screening tool for the diagnosis of hyperhomocysteinernia. We define BIZ deficiency as being below 133 pmol/L is undisputedly too low. As we demonstrate in figure 1and 2 both homocysteine and MMA are substantially higher in patients with senun BIZlevels below 258 pmol/L. Although, there is no evidence to suggest that supplementation with B12 will reduce the risk for cardiovascular disease, recent studies have demonstrated that moderate BIZ deficiencies may intluence smooth muscle cell proliferation and intirnal thickening of the carotid artery 97-99. Evidently there was a much stronger relationship between B12 and MMA compared to the association between BIZand homocysteine (figure 4.3 and 4.4). While the vitamin status of cobalamin influences the remethylation of homocysteine (and thereby homocysteine levels), MMA only accumulates when BIZlevels are compromised and are independent of folate status. We saw no improvement in the relationship between homocysteine and &2 when the MTHFR 677 TIT subjects were excluded from the analysis. This was somewhat surprising in light of the evidence that suggests that the C/Cand C/T patients have higher methylfolate levels 2m. One would expect that an elevation in cellular 5-MTHF would strengthen emphasis the relationship between homocysteine and BIZ. Indeed other polymorphisms that were not measured in this study, such as the MTHFR 1298 A+C, could contribute to the lower 5-MTHF in those patients who did not have the MTHFR 677 T allele. Notwithstanding the above discussion regarding MTHFR genotypes, folate status directly influences homocysteine levels. The high prevalence of hyperhomocysteinernia in our aging population with cardiovascular disease is consistent with that reported by multiple other studiesea? Our findings of a signhcantly elevated fasting homocysteine concentration in subjects with the T/T MTHFR genotype compared with the C/T and C/Cgenotypes is also consistent with most other studies 5537. While homozygosity was present in 16%of our CAD patients, 68 percent of these patients had an elevated fasting homocysteine, and they accounted for 36 percent of all patients with hyperhomocysteinemia. We next examined the ability of the methionine load test to discriminate among the various MTWR themolabile genotypes. Similar to the results obtained with the fasting homocysteine level, the post Zhour methionine plasma homocysteine level was sigrufrcantly elevated in the T/T subjects compared to the C/T and C/C MTHFR genotypes. This is not surprising since the fasting hornocysteine levels were elevated in the T/T group prior to the methionine ingestion. However, the relative increase (fasting minus the 2-hour post methionine load) was not different among the three MTHFR T subsets. This suggests that any differences in remethylation (due to h4THFR enzyme activity) cannot be identified at Zhours post ingestion of methionine. One likely explanation is that the flow of homocysteine is preferentially shunted through the transsulfuration pathway two hours-post methionine loading and consequently the 5-MTHF levels do not influence circulating homocysteine levels. Evidence suggests that any inhibition of remethylation after a methionine load is time dependent. In those with the T/T genotype compared to the C/T and C/C,Legnani et al." found no difference in homocysteine levels at 4 hours whereas after 6 hour Frost et al." have reported elevated homocysteine levels in patients with the T/T mutation. The effects of a Phour methioninc load on the MS 2756 A+G polymorphism was recently examined by Tsai et al.82. Interestingly the authors determined that homozygosity for the MS 2756 GIG genotype was associated with lower fasting levels of homocysteine. In addition, the MS A/G genotype also seemed to decrease the rise in homocysteine levels after the methionine load (p=0.053). Furthermore, there was a highly signihcant decease in the methionine load response in subjects with the bE 2756 A/G genotype and a 68 bp insert in exon 8 of CPS. Although, it is not dear why the MS mutation seems to be protective against post methionine load hyperhomocysteinemia it is reasonable to speculate that a shift in cellular 5-MTHF folate stores caused by an impaired MS enzyme activity helps to prevent a rise in homocysteine. Indeed, methionine loading, in addition to elevating circulating homocysteine, decreases 5-MTHF levels by as much as 50%lU, and possibly even more so intracellularly. We found a significant elevation in RBC folate levels in the subjects with the MS 2756 A/G genotype compared to patients with the A/ A genotype. Likewise in a smaller population Christensen et al.69 found a similar effect where patients with the MS 2756 A/ A genotype had mean RBC levels of 498(200) and the A/G had mean RBC folate levels of 610(295) (p=0.09). Therefore an increase in cellular SAM levels induced by the methionine load might be attenuated by higher concentrations of 5-MTHF in the A/G subjects with the MS mutation. Like the MTHFR 677C+T mutation, the effects of the MS 2756 G dele may similarly become a greater contributor, lowering post load hyperhomocysteinemia more significantly at 6 and eight hours. In light of our recent findingsj3, the MS G alleles could at least partially explain why subjects with lower post-load responses are at a decreased risk for vascular disease. We did not find any difference in fasting or methionine load response between the MTRR 66 A or G alleles. Furthermore.,we found no evidence to suggest that either allele alters serum Biz MMA or folate levels. Because of its direct involvement in the remethylation pathway it may be premature to conclude that one of the MTRR alleles does not affect the post methionine load homocysteine levels. It may contribute to an alteration in circulating homocysteine in the later stages of a methionine Load. Nonetheless, it does appear as if the GIG genotype is linked to another gene which influences or it directly influences the risk for vascular disease itself. Brown et a1 concluded that there was a sigruficantly higher proportion of the G/G genotype in patients with premature cardiovascular disease 8'. Our observation that the M'RR 66 A+G alleles do not influence fasting homocysteine are in agreement with their results. We also demonstrate that neither the A or the G 66 MTRR alleles appear to affect homocysteine after a two-hour methionine load nor do they affect MMA, B12 or folate levels. Interestingly, the distribution of the MTRR 66 A+G genotypes in the population described by Brown was essentially identical to what we found in this study and given that the mean age of our population was the same as their exdusion age one could speculate that the incidence of the MTRR 66 G/G is similarly elevated in subjects with later onset of vdardisease. In light of the foregoing discussion, we argue in favor of a single fasting homocysteine measurement compared to the two-hour methionine load as a screen for hyperhomocysteinemia. Future studies should address the effect of the MS 2756 A+G and the MTRR A+G polyrnorphisms on homocysteine levels after six/eight hours post-methionine loading. It does not appear as if the MS 2756 or the MTJXR 66 genotypes siphcantly alter senun B12 or MMA levels. However, the relationship between BIL MMA and homocysteine suggests that the definition of a 812 deficiency should be raised to 258 pmol/L from the current value of 133pmol/ L. CHAPTER 5 EFFECT OF HETEROZYGOSITY FOR THE METHIONINE SYNTHASE 2756 A+G MUTATION ON THE RISK FOR RECURRENT CARDIOVASCULAR EVENTS. Hyndman ME, Bridge PJ, Warnica JW,Fick G,Parsons HG Departments of Pedia tricsl Medical Genetics2 and Medicine31 University of Calgary, Calgary, Alberta, CANADA (Reprinted with permission from the American Journal of Cardiology) Genetic variations in folate metabolism are believed to contribute to the risk of cardiovascular disease although the mechanisms by which this may occur are not weil understood. Mildly elevated plasma homocysteine is increasingly being recognized as a risk factor for coronary artery disease. Two key enzymes in folate metabolism, methylenetetrahydrofolate reductase (MTHFR)and methionine synthase (MTR) are instrumental in the remethylation of homocysteine to methionin*. mRconverts folate to its active form, 5- methy ltetrahydrofola te (5-MTHF) and the latter product provides the methyl group for the remethylation of homocysteine. A common point mutation in the MTHFR gene (677 C+T) has been associated with depleted 5-MTHF, mild hyperhomocysteinernia and in some, but not all, studies an increased risk of cardiovascular disease2mm. MTR, a vitamin &2 dependent enzyme, catalyses the conversion of homocysteine to methionine using the product of MTHFR, 5- methyltetrahydrofolate78;2LJ3.A mutation in methionine synthase (MTR) has recently been identified and is the result of a 2756 A+G substitution, changing an aspartic acid to a glycine residue (D919G)78;? This polymorphism may alter folate metabolism and hence is a suitable candidate gene for modifying risk for cardiovascular disease. This obsemationd prospective study was designed to assess the susceptibility of patients with the MTR 2756 A +G mutation and a prior myocardial infarction for a recurrent cardiovascular event. Association was also sought between the MTR genotype and plasma homocysteine and red blood cell (RBC)foiate levels. The study included 109 patients with normal lipid levels who had a myocardial infarction at least 3 months prior to entry to the study. The incidence of recurrent cardiovascular events was documented over a mean of 5 years. A recurrent event was defined by the examining physicians as one of the following: angioplasty, angina, heart failure, myocardial infarction or bypass surgery. All patients participating in the study gave informed written consent and the University of Calgary ethics committee approved the study. AU statistical analysis were performed using Stata. Equality in survival data, defined as a vascular event was assessed using the tog rank (Mantel- Haenszel) test and Cox regression proportional hazards analysis. Differences in events were calculated between the wild type and those heterozygous for the M'TR A-G mutation. The Wilcoxon rank sum test was used to test for difference in red blood cells folate and homocysteine between the MTR wild and heterozygous genotypes. A p value of < 0.05 was considered significant The demographics for our population were as follows: 17%were diabetic, 92% male, and 8%female. Mean (and standard deviation) for age, total cholesterol, high density lipoprotein and low density lipoprotein were 57.9(10), 5.4 (0.44), 1.01 (0.26) and 3.6 (0.37) respectively. The overall (MTR 2756 G) allele tiequency was l5%, translating into 76,32 and 1patients with the A/ A, A/G and G/G genotype respectively. No demographic differences in age, gender, or lipid profiles were evident between the MIX wild and heterozygous genotypes. As there was only one patient homozygous for the MTR 2756 A+G genotype no statistical analysis is included for this grouping. Total recurrent events showed sigruficant difference between the 2 groups (table 1) @< 0.01). Patients with the A/G genotype had fewer vascular events than expected. Conversely the homozygous A/ A genotype patients had more vascular events than was expected (table 5.1). When the comparison was made using only three recurrent events (myocardial infarction, heart failure, and bypass surgery), a more strict definition of a recurrent event, the statistical sigruhcance between the two genotypes remained present (p=0.03). Table 5.1. Mantel-Haenszel log rank test for equality of vascular events between patients with the hornozygous normal (A/ A), and heterozygous (A/G) genotypes of methionine synthase. Vascular events (all1 MI, Heart Failure, bvvass MTR Expected* Observed ExpectedH Observed *p<0.01 *p=0.03 Indicating a significant difference in events between the A/ A and A/G genotypes for methionine synthase. Further analysis using Cox regression showed the MTR 2756 A+G genotype to be a sigruhcant predictor for the recurrence of cardiovascular events with the heterozygous patients 3.4 times less likely to have a recurrent MI, heart failure, or bypass surgery (95% a:1.09 to 10.9). The percentage of recurrent events in the wild type and those heterozygous of the MTR 2756 A +G genotype at various times after the start of the study is shown by Kaplan-Meier graph (figure 5.1). Kaplan-Meier cardiovascular event-free survival estimates I I I - 1.00 - A/A p Figure 5.1. A comparison of the probability of NOT having any of the following: heart failure, myocardial infarction, or coronary bypass surgery between the A/A (n=76)and A/G (n=32) MTR genotypes. (A/A= Wild Type, A/G= Heterozygotes for Methionine Synthase 2756 A+G mutation, MI= myocardial infarction. HF= heart failure, CBP= coronary bypass surgery, CAD= coronary artery disease) The Wilcoxon rank sum test revealed RBC folate levels were sigdicantfy elevated in patients with the A/G genotype compared to the A/ A genotype (~~0.05).Median (25h, 75th percentiles) were 970 nmol/L (739,1247) and 844.4 nmol/ L (704,1040) for the A/G, A/ A MTR genotypes respectively. The mutation did not sigmficantly affect homocysteine, or BIZlevels. This study demonstrates that subjects with a common mutation in the MTR gene, an A to G transition at base pair 2756, is associated with a reduced risk for a recurrent cardiovascular event. The reduction in risk was still evident when myocardial infarction, heart failure or bypass surgery were used as end points. The relationship was independent of age, gender, or plasma lipid profiles. Thus our study has revealed a genetic predisposition to a reduction in recurrent cardiovascular events in subjects heterozygous for the MTR 2756 A+G base pair transition. We also identified an increased RBC folate in those at reduced risk. The molecular mechanism that underlies the relationship between the MTR mutation and recurrent cardiovascular risk cannot be deduced from this study. However, it may be related to the RBC folate increase and type of folate in the red blood cell. One could argue that a reduction in MTR activity results in an increase in RBC methyltetrahydrofolate, a feature referred to as methyl trapping. In support of the mutation affecting MTR activity a recent report showed that homozygosity for the mutation altered homocysteine metabolism^. However, as homocysteine was not elevated in our heterozygous population we would have to argue the effect of reduced activity was mild or other confounding variables were sufficient to cause RBC methyl trapping but not an elevation in plasma homocysteine. The association between the methyltetrahydrofolate concentration and cardiovascular risk has not been clearly established. Nevertheless, evidence does exist which suggests an interaction between IMTHF and nitric oxide bioavailability~~.Vascular endothelial nitric oxide production is anti- atherogenic and it has been proposed that endothelial dysfunction mediated by a decreased production is a prelude to vascular disease. Verhaarm et al. have recently shown that infusion of Emethyltehahydrofolate improves endotheiial function in patients without disordered homocysteine metabolism but with familial hypercholesterolemia. The response was immediate and the idusion did not affect homocysteine levels, indicating the importance of folate and more specifically 5-MTHF improved endothelial function. In conclusion we report that heterozygosity for the MTR 2756 A+G mutation increases RBC folate and reduces secondary vascular events. Both outcomes may be a result of a mild methyl trapping of 5-hiTHF, an effect opposite to the MTHFR 677 C+T mutation. Further evaluation is needed to determine the effect of heterozygosity for the MTR A+G 2756 mutation on primary events, and the possible interaction with the MTHFR 677 C+T mutation. CHAPTER 6 EFFECTS OF COMMON ONE CARBON METABOLISM POLYMORPHISMS ON ENDOTHELIAL FUNCTION 6.1 Introduction The vascular endothelium has emerged as an important determinant of vaxular health and disease. Through the production of endothelium-derived relaxing and contracting factors the endothelium serves to regulate vascular tone, decrease oxidant production and inhibit platelet mi neutrophil adhesiont? The endothelium-derived relaxing factor NO appears to orchestrate the majority of these anti-atherogenic mechanisms and counter the pro-atherogenic potential of endothelium-derived vasoconstrpictive factors such as endothelin-1m-m. In states of endothelial dysfunction, this balance is altered in favor of increased thrornbogencitiy and atherosclerosis. Endothelial dysfunction plays a key role in the initiation and stabilization of the atherosclerotic plaque". In healthy coronary arteries infusion of acetylcholine causes a dose and calcium dependent dilation mediated by the release of nitric oxide from the endothelial cells. However infusion of acetylcholine into the coronary arteries of patients with atherosclerosis induces a paradoxical vasoconstriction~0;~~.This invasive technique used to assess endothelial function has now been replaced with flow- mediated endothelial dependent vasodilatation of the brachial artery. Importantly the vascular responses of the brachial artery correlate well with the extent and severity of coronary atherosclerosis "3. Impaired endothelid release of nitric oxide is thought to be one of the earliest signs and initiating event in the development of an atherosclerotic lesion. Hyperhomocysteinemia, like most risk factors for atherosclerosis is associated with impaired flow mediated endothelid function. Celermajer et a1 169 initially determined that patients with cIassical homocystineuria, had impaired brachial and femoral arterial endothelial function 6. Two other studies have investigated the effect of methionine-induced hyperhomocysteinemia in healthy adults '8;'". In both cases the methionine load induced hyperhomocysteinemia and impaired endothelial function. Interestingly, the methionine load at four hours was not associated with an increase in the oxidative marker P-selectin where as the impaired endothelial function 8 hours after a methionine load could be prevented by large dose of the antioxidant vitamin C. This discrepancy could be a result of insufficient time to upregulate P-selectin at four hours after a methionine load whereas eight hours may have been the time required to induce oxidant production and endothelial dysfunction. Alternatively the improved function caused by ascorbic acid may be mediated through increased cellular concentrations of tetrahydrobiopte~n2. It is well documented that methionine loads also deplete circulating 5-methyltetrahydrofolate and therefore an argument could be made that insufficient folate status could account for the impaired endothelid function. This argument is strengthened by studies that have shown that supplementation with folate improves not only endothelial function in patients with hyperhomocysteinemia 213 but also in those with hypercholesterolemia m;21.? coronary atherosclerosis 14 and following consumption of a high fat load. Over and above the indirect effects of folic add on endothelid function (mediated by lowering homocysteine), recent studies suggest that folic acid may exert direct endothelial protective effects mediated through augmenting NO production and release. Impaired flow mediated dilation (FMD) is believed to represent a surrogate marker for coronary atherosclerosis. Given that hyperhomocysteinemia has been shown to impair FMD and conversely, folic acid supplementation improves endothelid function we decided to investigate the effects of two common polyrnorphisms in homocysteine and folate metabolism on FMD and homocysteine metabolism in health adults. 6.2 Methods The study population reported herein is a sub-study from a larger clinical trial designed to assess the long-term effects of impaired brachial artery flow- mediated dilation on atherosclerotic vascular disease. The cohort consists of active and retired middle-aged men from the Calgary Fire department. Patients were excluded if they had a previous history of documented coronary artery disease peripheral vascular disease or cerebrovascular disease. Baseline endothelial measurement was performed as previously described by Anderson et a1.F Serum lipid profile and glucose were measured by standard methodology available in the FoothiUs Hospital. Homocysteine was measured using a fluorescence polarization imrnunoassay. Total homocysteine is reduced with dithiotheietol and homocysteine is then converted to Sadenosyl homocysteine (SAH) by bovine SAH hydrolase and excess adenosine. SAH and a labeled fluorescent tracer compete for the SAH antibody allowing quantification using polarized fluorescent light. The assay has a sensitivity of ~0.05~mol/l with a CV less than 5%. Both folate and 812 measurements were measured with microparticle enzyme imrnuno assay technology. Specific antibodies for either 812 or folate are bound to latex particles and then placed in a reaction chamber. A secondary alkaline phosphatase labeled conjugate antibody is added creating an anitbody-analyte-conjugatesandwich. 4-methylumbelliferyl phosphate (UMP) is hydrolyzed by the alkaline phosphatase into the fluorescent product methylumbellferone allowing quantification. The methodology is highly sensitive (2.0nmol/l) with a CV of 5.7% at the lower normal range. 6.3 Results Mean ages, blood pressure, insulin, fasting glucose, creatininel mg of protein, cholesterol, LDL cholesterol, HDL cholesterol, triglycerides, RBC folate, vitamin B12 and homocysteine levels are summarized in table 6.1. Table 6.1: Summary of demographic data Variable Mean ED) Age years n=591 Sex (Male:female) Weight kg Blood pressure mmHg Sys/ Dia n=249 Insulin n=570 Fasting glucose n=585 Creatine/ mg protein n=578 Cholesterol mrnol/ 1 n=586 LDL chol. mmol/l n=586 HDL chol. rnmol/ l n=586 Triglycerides mmol/ 1 n=586 RBC folate nrnol/ 1 n=581 Vitamin 812 pmol/ 1 n=583 Homocysteine umol/ 1 n=575 The vast majority of subjects had nodhomocysteine, serum folate, RBC folate and homocysteine levels. Homocysteine levels ranged from 4.1 to 31.7 umol/ l with the median value being 8.6 umol/ I. None of the 698 or 163 subjects in whom RBC folate and serum folates have been analyzed would have been classified as being deficient. However, 38% (n=27l) subjects, had vitamin B12 levels below 258 pmol/l (Level at which MMA is no longer elevated). In addition, 3%(n=24) of the individuals were below our hospitals laboratory lower limit of 143 pmol/L. Of the 591 patients who had their MTHFFt genotype analyzed, 231 (39%), 285(48 %), and 75(13 %) were C/C,C/T and T/T for the 677 C+T genotypes respectively. Of the 711 patients who had their MS genotype analyzed 483 (67.93%),205 (28.83%)and 23 (3.23%)were A/A, A/G and GIG for the MS 2756 A+G genotypes respectively (table 2). Table 6.2 Genotype distribution for the MTHFR 677 C+T and the MS 2756 A+G polymorphisms. Polymorphism Wild type Heterozygous Homozygous MTHFR C+T 39.09% 48.22% 12.69% MS 2756 A+G 67.93% 28.83% 3.23% Mean and standard deviations for homocysteine, RBC folate, serum folate, segregated by the M.THFR 677 C+T the MS 2756 A+G and both genotypes are summarized in table 6.3,6.4 and 6.5 respectively. There was no sigruficant difference in homocysteine, RBC folate, or serum folate levels between the MS 2756A-G genotypes (table 6.4). Conversely, homocysteine (figure I), serum folate (figure 6.2) and RBC folate levels (figure 6.3) were influenced by the thennolabile MTHFR 677C-T polymorphisms. Subjects with the T/T genotype had sigruficantly elevated homocysteine and RBC folate levels, but lower circulating serum folate concentrations. Patients with the lowest mean homocysteine were had C/Cfor the MTHFR C+T genotype and G/G for the MS 2756 genotype. Whereas subjects with the highest homocysteine levels had the exact opposite genotype, being TIT and A/ A for the MTHFR 677 and MS 2756 genotypes respectively (Table 6.5). Table 6.3 Plasma homocysteine, RBC folate, serum folate levels stratified by the MTHFR 677 C+T genotype. Senun folate levels were analyzed from m equal number of randomly selected samples from each genotype. MTHFR 677 C+T Genotype c/c C/T T/T n=223 n=278 n=74 Homocysteine umo/ 1 8.6(2.7) 8.7 (1.8) 9.5 (2.4)* RBC Folate 1127 (236) 1142 (249) 1320 (350)- Serum folate 35.5 (9.4) 34.2 (9.2) 28.6 (9.6)* n=54 n=54 n=55 *p=0.01T/T Vs C/C p=0.03 T/T Vs C/T, "p<0.01: ANOVA with scheffe's post hoe test. Table 6.4 Plasma homocysteine, RBC folate, serum folate levels stratified by MS 2756 A+G genotype. MS 2756 A+G Genotype A/ A A/G G/G n=473 n=200 n=21 Homocysteine umo/1 8.7(2) 8.8(2.5) 8.6(1.5) RBC Folate 1179(293) 1195(315) 1137(377) Serum folate 32.9(10) 32.5(9) 31.5(5) n=106 1149 n=8 Table 6.5 Mean homocysteine ,RBC folate, serum 812 and serum folate levels by MTHF 677 C+T and MS 2756 A+G genotypes. (From top to bottom in each cell: hornocysteine umoI/l, RBC folate nmol/L, vitamin 812 pmol/L, and number of subjects) Figure 6.1. Box plot illustrating the distribution of plasma homocysteine (umo]/L) levels stratified by the MTHFR 677 C+T genotype. (4Cn=231, n= 285 T/L=75) Figure 6.2. Box plot illustrating the distribution of RBC folate levels (mom) stratified by the MTHFR 677 C+T genotype. (gCn=231, n= 285 Tfl=75) In the 318 subjects analyzed to date there was no sigruhcant relationship between homocysteine, RBC folate, serum folate, vitamin 812 and endothelial function. The MTHFR 677 C+T genotype also did not significantly influence the FMDdilation in this population. However, FMD was significantly higher in the heterozygous MS 2756 (A/G) subjects compared to the homozygous MS (A/A) subjects (p=0.03; t-tes t unequal variances) (figure 4). Median diameter changes induced bv FMD were 8.6%(5.9%to 11.9) and 9.7%(7.6 % to 12.9%)for the A/A and A/G genotypes respectively. (There were insufficient individuals with the G/G genotype and they were therefore excluded from FMD analysis). The endothelial independent responses to nitroglycerin were not sigruficantly different between the A/A and A/G genotypes or in any of the other variables measured. I R H. Log % Diameter Change'1 1 Figure 6.4 Flow mediated vasodilatation (% change above baseline) stratified by either the A/ A or A/G genotypes for methionine synthase 2756A+G pol ymorphisms. 6.4 Discussion In this study we investigated the effect of two common polymorphisms involved in methionine metabolism on flow mediated endothelid function and their interactions with plasma homocysteine and circulating folate levels. The MS 2756 A+G genotype was associated with sigrhcantly enhanced flow-mediated endothelial dependent vasodilatation in a healthy adult male population compared to the homozygous wild type A/A subjects. This occurred without changes in homocysteine, RBC folate or serum folate levels. In contrast, the homozygous MTHFR T/T genotype, did not influence FMD but was associated with sigruhcantly elevated homocysteine, elevated RBC fola te and lower serum folate levels compared to the C/C or C/T patients. It is not clear how the methionine synthase 2756 A+G polymorphism alters enzyme activity and intracellular folate and/or homocysteine levels. Some reports have demonstrated that the homozygous G/Ggenotype increases circulating homocysteine levels where as others have found it sigruficant decreases plasma homocysteine. Furthermore we have previously demonstrated in a population with CAD, that A/G genotype siguhcantly increases RBC folate levels. However, in this study the mutation did not alter homocysteine, RBC folate or serum folate levels. Nevertheless, in this study the cohort with the lowest homocysteine levels were those with the MS 2756 GIG and the M. 677 C/C. This non-sigruficant observation must be confirmed in a larger population. One could speculate that the impaired MS enzyme predominantly affects intracellular folate levels while not signihcantly affecting circulating folates or homocysteine concentrations. Regardless of the impact on the biochemical variables, the A/G genotype improves endothelial function while having no effect on the endothelial independent response to nitroglycerin. This effect may extend to the homozygous G/Gpatients who had the highest vasodilatation of all three genotypes (non-sigxuficant). It would be logical to speculate that an impaired MS enzyme could shift the intracellular folate pool favoring 5- rnethyltetrahydrofolate. As previously mentioned evidence does exist which to suggest that 5-methytetrahydrofoiate improves endothelial function and facilitates the release of nitric oxide from endothelial cells 19. This will be discussed in more detail in the following chapter 7. Unlike the MS 2756 polymorphism the MTHFR 677-T mutation had no effect on endothelial function but influenced homocysteine, RBC folate and serum folate levels. Numerous other studies have reported that subjects with the MTHFR 677 T/T genotype have elevated homocysteine and lower serum folate levels. In our population of healthy adults, none of whom would have been classified as deficient in either RBC folate or serum folate, the T/T genotype still sigruhcantly increased homocysteine levels compared to either the C/C or C/T genotypes (n=163). Bagley and Selhub have demonstrated that subjects with the homozygous T/T genotype have increased formyl folate levels in their erythrocytes because of the decreased MTHFR activity. It is likely that this shift accounts for the differences between the serum and RBC folate levels. Traditionally folates have been measured using either microbiological assays or radiometric competitive binding assays. Recently an automated -1 method has been introduced in Canada (methodology described above), which, like the radioimmunoassay, depends on specific antibodiesa6. A previous investigation determined that the patients with the TIT genotype had signhcantly lower RBC folates if the microbiological assav was used and conversely had higher RBC folates with the radioimmunoassay. Two other studies have also reported higher erythrocyte folate levels in subjects with the T/T genotype "7. However, both serum and plasma folate levels are constantly lower in the T/T subjects when measured by either the microbiological or imrnunoassays. The immuno-based protocol used in our clinical laboratory seems to parallel the older generation radioimmunoassay giving lower senun and higher RBC folate measurements. The discrepancy between the serum/plasma folate and the RBC folate levels in the T/T patients may be a result of the preferential binding to formyl folate derivatives which are only present in subjects with the T/T genotype 109-. Moreover an agreement between the methodologies in the serum and plasma would be expected given that the vast majority of circulating folate is 5- methyltetrahydrofolate irrespective of the MTHFR genotype. This difference could have clinically relevant consequences and may be one of many possible explanations as to why RBC folate levels in our study did not correlate with endothelid function. Subjects with the TIT genotype had RBC folate levels of 192 nmol/L and 177 nrnol/L above the C/Cand C/T patients respectively (p<0.0001) yet had higher circulating homocysteine levels. Future studies are needed to address this issue especially because 545%of the population carries the homozygous T/T genotype. CHAPTER 7 5-METHYLTETRAHYDROFOLATE ATTENUATES SUPEROXIDE PRODUCTION AND AUGMENTS NITRIC OXIDE PRODUCTION IN MODELS OF HYPERHOMOCYSTEINEMIA AND INSULIN RESISTANCE 7.1 Introduction Impaired release of endothelial-derived nitric oxide has been proposed as a mediator and initiator of atherosclerotic disease. Nitric oxide (NO) can be produced by three homodimeric isozymes termed nitric oxide synthases. These include the endothelial (eNOS), neuronal (nNOS) and inducible (iNOS) forms of the enzyme. In the vascdature NO produced by eNOS is actively involved in homeostasis, regulating vasodilatation, leukocyte activity, platelet aggregation, and cellular proliferation. Under normal conditions the active site of eNOS efficiently catalyzes electron transfer from NADPH-FAD-FMN to 1-arginine generating citrulline and NO. However, under pathological conditions associated with cardiovascuIar risk factors, nitric oxide production is impaired '6;t~;~"~O.Decreased Bban important allosteric effector of eNOS may simultaneously contribute the lower production of NO and the generation of potent oxidants such as superoxide and hydrogen peroxide. Bfi deficient eNOS leads to the uncoupling the synthesis of NO from L-arginine, and facilitates the production of superoxide m-a. The mechanism leading to the decline in cellular B& levels that is associated with many of the risk factors for vascular disease is not entirely clear. Recent reports indicate that supplementation with BKimproves endothelial- dependent vasodilatation in experimental diabetes, reperfusion injury, hypercholesterolemia and in agarette smokers 1E;lal~.This may suggest that CAD risk factors lead to the spontaneous degradation of Bfi. BH4 is an unstable compound readily oxidizes in aqueous aerobic buffer within 20 min. Therefore, an increase in free radicals as is associated with the above-mentioned risk factors, may serve to deplete Ba levels and further exacerbate NO deficiency. Inhacellular B& degradation proceeds through three steps: conversion of BH4 to the quinonoid 6,7[8]dihydrobiopterin then to 7,s dihydrobiopterin and finally to biopterin. Interestingly, endothelid cells exclusively export the oxidized forms of biopterin, possibly preventing the accumulation of the oxidized pterins ~~~~. The structure of 5-methyltetrahydrofolate (5-MTHF) is extremely similar to that of Bfi with the exception of an extended tail attached to 5-MTHF.It has been proposed that the large size of SMTHF would exclude it from interacting with the active site of NOS (in a fashion similar to Bh). However, clinical data contradicts this argument because direct infusion of 5-MTHF into the brachial artery of patients with hyperlipidemia mimics the effects of BI+ on endothelial function20. Likewise, Maierls et d.demonstrated that direct infusion of acetylcholine and Bfi into the coronary arteries restored impaired endothelid function in patients with coronary artery disease. Secondly, Stroes19 et al. in a recent study demonstrated that supplementation with 5-MTHF to cultured endothelial cells stimulated with acetylcholine sigruhcantly improves nitric oxide release. The authors further concluded that 5-MTHF was not acting as an antioxidant but likely interacted with eNOS directly. This present study was designed to investigate the interactions between eNOS and 5-MTHF by addressing the following questions (1).In vascular endothelial cells, can the increased production of superoxide and the corresponding inhibition of nitric oxide that is associated with the inhibition of B& synthesis be ameliorated by supplementation with 5-MTHF ? (2). Using computer modeling technology, does 5-MTHF bind to the active site of eNOS in a similar fashion to Ba? Finally using two animal models, one known to be deficient in BH4 (fructose-fed rats) and the other in 5-MTHF (methionine load), determine if treatment with 5-MTHF and Bfi have similar effects on endothelium-dependent vasodilatation to acetylcholine. 7.2 Methods Recombinant eNOS superoxide production: Superoxide produced by bovine recombinant endothelial nitric oxide synthase (Isolated from baculovirus overexpression in Sf9 cells: Cayman Chemicals) was measured with bis-N- methylacridinium (Lucigenin) chemoluminescence. Recombinant eNOS (7.5 ug), was incubated with lOuM FAD,ImM CaCl, IrnM NADPH, 300 U/ml calmodulin, and either lOOuM or 500uM of 5-MTHF or tetrahydrobiopterin. Total photons produced by the reaction between superoxide and lucigenin were measured in a lumometer for 300 seconds. AU chemicals were made up immediately before use. eNOS was purchased from Cayman chemical all other reagents were from Sigma. Immunohistochemistrv: Both immortalized and bovine endothelial cells stained positively for eNOS protein. Cells were grown on German glass cover slips and then fixed with 10%formafin for 15 rnin. Slides were rinsed with lOOuM Tris buffer (pH 7.6,0.1%Triton-X 100) and incubated with primary polyclonal IgG rabbit anti NOS 3 (n-20;Santa Guze biotehology 1:W) ) in tris buffer containing 0.005% BSA for 24hrs at 4OC. Slides were then rinsed and incubated with anti rabbit IgG cyanored conjugated in tris buffer with 0.1% Triton-X 100 for 2hrs at 4°C.Fluorescence was measured by an epi-fluorescence equipped microscope. Measurement of ROS in cultured endothelial cells: Bovine endothelid aortic cells were isolated with lmg/rnl collagenase in PBS and cultured in MI99 contain 10% fetal calf serum and 200m.M glutatmine. Cells were then incubated in 5% FCS 24 hour before and with treatments. All experiments were used in cells between passages 2 to 6. After reaching confluence cells were loaded with 2uM of the fluorescent probe 123-dihydrorhodamine (DHR123)in six well plates for 30min in FCS free MI99 and then stimulated with lOuM acetylcholine for an additional 30 min. Cell were rinsed, trypsinised, re-suspended in Hanks buffer (pH 7.2) 5% FCS and then centrifuged and resuspended in Hanks without FCS. DHR123 is irreversibly converted to the green fluorescent membrane impermeable compound rhodamine-123. FACS was carried out with a flow cytometer equipped with an ion argon laser emitting at 488nm. Emission spectra was collected above 540x1111on 10 000 c& per sample in three separate experiments. Measurement of Suveroxide production- in cultured endothelial cells: Superoxide production was assessed using flow cytometry by comparing the mean DHE induced fluorescence of endothelial cells. Cells were culture as previously mentioned and then incubated in serum free MI99 and 25uM dihydroethidine @HE) and stimulated for an addition 30 min in lOuM ach. DHE reacts specifically with superoxide and is converted to ethidiurn. Cells were rinsed, trypsinised, re-suspended in PBS buffer (pH 7.2) 5%FCS and then centrifuged and re-suspended in PBS without FCS. Nitrite measurements: Nitrite levels were measured by modifying the HPLC methodology originally described by Muscara et d.m. After two hour of stimulation with ACh the incubation buffer (Hanks pH 7.2) was run through a C18 solid phase extraction column. lOOd of solution was then injected into a runring buffer consisting of 60 mM WC1and 35 rnM B0buffer (lml/min) and nitrite was separated with a SAX Waters 15an/5uM column. The column exit was connected to a zero volume tee introducing an equal mixture of 0.1 % naphthyl-ehtylene-diamineand I % sulfanilamide in 5% NP04.Before entering the detector the column and Griess effluent was allowed to reacted in a 50cm delay tube. Absorbance was monitored at 540 nm by a UV-Vis photodiode array detector. Isometric tension studies: Tension studies were performed on aorta isolated horn fructose-fed and control rats as previously described m. The strips were sectioned and then pre-constricted with the EDn of phenylephrine. After a stabilized constriction was obtained the isolated aortic segments were exposed to acetylcholine or sodium nitroprusside to generate dose response curves in the presence or absence of 5-Wand Bfi. In a separate group of rats, the acute effects of 5-MTHF on methionine-load induced vaxular responses was evaluated at two time points (2 and 10 hours post methionine load, O.lSmg/kg). In addition, the effects of 3 week treatment with folinic acid on long term methionine loading (for 2 weeks) were also examined. 5-MTHF used in the animal studies was a kind gft from Eprova pharmaceuticals. Computer Modeiing: Docking of tetrahydrobiopte~ and 5- methyltetrahydrofolate was performed using Autodock 3.0. Partial charges for each atom were assigned to the ligand and enzyme using SYBYL. Cubic affinity grid maps centered on the pterin site of bovine eNOS (dimensions 21 A x 21 A x 21 A, grid points spacing 0.35A) were calculated for each pterin atom type and electrostatics using autogrid. Rotatable bonds for the pterins were specified using the autotors tool. The pterins were individually docked to the eNOS pterin site using the Lamarkian genetic algorithm to search the precomputed affinity maps for low energy binding orienatations. Each docking trial was initiated with a randomly generated population of 50 binding orientations and completed after 1.5 million energy evaluations had been performed. Due to the stochastic nature of the docking process, 10 docking hials were performed. The resulting orienations were clustered with a 0.5 A tolerance and ranked according to binding energy predicted by docking. Independent results of both BH4 and 5- W were compared to the crystal structure of eNOS containing bound BH4. 7.3 Results Immunocytochemistry: Both endothelial cells isolated from the local abattoir and the immortalized human EaHy96 cell line expressed eNOS protein. Furthermore, incubation with cyclosporin, which has previously been shown to induce endothelial nitric oxide protein expression, clearly induced expression in the endotheiial cells. Secondly, incubation with an inhibitory peptide specific for the antibody-binding site on eNOS drastically reduced fluorescence, confirming specificim of the primary and minimal non-specific secondary antibody binding (figure 7.1 & 7.2). Figure 7.1. EaHy96 endothelid cells endothelial cell imrnunofluorescence for eNOS with and without peptide inhibitor. Figure 7.2 Bovine aortic endothelid cells A. Normal untreated B. Co-incubation of primary antibody with inhibitory peptide. C. Cells stimulated with 24hrs of Cyclosporin A. Attenuation of free radical species generated- from eNOS by 5-MTHF and BH4: Both 5-rnand BH4 attenuated lucigenin superoxide production by recombinant eNOS protein. Incubation with all reagents without eNOS demonstrated that the superoxide was being synthesized by eNOS and was not mediated by an independent reaction from any of the co-factors. Moreover, eNOS incubated with superoxide dismutase completely abolished the signal from lucigenin. Native enzyme not supplemented with either Bfi or 5-MTHF produced the largest amount of superoxide, and results from Bfi or treated 5- h4THF were compared to this value. There was a similar sigdicant dose dependent reduction by lOOuM and 500uM 5-MTHF and Bfi (figure 7.3). 5- MTHF and sepiapterin, (a precursor known to increase intracellular of BH-t production) significantly inhibited the production of reactive oxidative species within acetylcholine stimulated endothelial cells. 123-Dihydrorhodarnine is an uncharged and nonfluorescent dye capable of diffusing across cell membranes. It reacts with hydrogen peroxide "and peroxinitrite rsDto form the fluorescent and cell impermeable rhodamine 123. ANOVA with Scheffe post hoc analysis revealed a sigruhcantly lower mean cellular fluorescence in the 5-MTHF (500uM)+DAHP (5mM) (p<0.01) and sepiapterin (100uM) + DHAP (5mM) (p<0.02) compared to cells with only the GTPcycIohydrolase inhibitor DAHP (5mM) (figure 7.4). Similarly, DAHP caused an increased in the mean cellular florescence in cells incubated with the superoxide specific DHE florescent dye (figure 7.5A) Co-incubation with 100 uM 5-MTHF and DAHP reduced the fluorescent signal (figure 7.58) (p DAHP alone. I lucigenin SOD eNOS eNOS eNOS eNOS eNOS I noenzyme eNOS 100uM 500uM 1OOuM 500uM / BH4 BH4 SMTHF SMTHF ] 1 Figure 7.3. Superoxide production measured by lucigenin enhanced chemoiuminescence from recombinant eNOS total counts (300 sec) expressed as percent reduction from control over 300 seconds (n=4 per treatment). Normal DAHP DAHP +I00 DAHP + 500 umol sepiapterin umol 5-MTHF Figure 7.4. Mean 123-dihydrorhaodmineinduced fluorescence of BAEC incubated with normal media, DM,DAHP and sepiapterin (p<0.02 Vs. DM)and 5-MTHF (p<0.01 Vs DAHP ) ANOVA with Scheffe post hoc analysis 10 000 cells n=5. Nothing DAHP alone Figure 7.5. Mean DHE fluorescence signal from of BAEC cells incubated with A. Normal media and cells incubated with the GTP cydohydrolase inhibitor DAHP (5mM) for 24 hrs @<0.01: t-test on three experiments ). 8.Endothelid cells incubated with a lOuM, 1OOu.M and 500uM 5-MTHF and DAHP (5mM) for 24 hrs . Effects of 5-MTHF and Nitric mide production: Supplementation with 500uM methionine caused a non-signhcant reduction in nitrite release compared to cells incubated with 5-MTHF (100uM)+ methionine (500uM)@=0.09, t-test). However, incubation with 5-MTHF and n-acetylserotonine WAS: an inhibitor of sepiapterin reductase and therefore Bfi synthesis) sigruficantly improved the nitrite release from cultured endothelid cells compared to cells onlv incubated with NAS (figure 7.6)(Appendix B). Met load MetIoad+S- WS I MTHF I Figure 7.6. Acetylcholine induced nitrite production, by EaHY 96 endothelid cells incubated with 1mM homocysteine, 500uM methionine, lOOuM NAS with and with lOOuM 5-MTHF. Computer modeling: Both BH4 and 5-Wwere docked in the eN05 active site and their positions were compared to that of the position of the ori@ BI& determined from the crystal structure. The convergence of the Bfi dockings supports the docking application. Secondly, the average atomic distance between the top ranked Bfi dockings solutions was 1.26 and 1.01 respectively accounting for 6 of the 10 runs. Similarly, the docking solution for 5-MTHF was within 1.4 A (excluding the tail) of the actual pterin binding site. Moreover the 5-MTHF was pi stacked with trp 457, which correlates well with the report by Fishrnan et al. who found a similar interaction between Bfi and their analogous trp 44714. Furthermore, N3 and NA2 of the docked 5-MTHF were aligned identically to the corresponding B& atoms and were 3.1A0 and 3.0 A* from the propionate group of heme. This also is similar to what is reported by Fishman et al. who determined that BFb interacted with the same propionate group with reporting distances of 3.3 and 3.1 A. Finally, the relative binding energies between Bfi and 5-MTHF and were -16.92 and -13.96 respectively. The interaction between 5-MTHF and the eNOS active site is illustrated in figure 6.7 The actual pterin binding site is overlaid with 5-MITE for reference. Figure 7.7. Docking results of 5-methyltetrahydrofolate in the active site of eNOS compared to the actual tetrahydrobiopterin binding site determined form the crystal structure (ref). Heme iron is green with the propionyl group in red. Isometric studies: The addition of acetylcholine to the organ bath produced a dose dependent relaxation. A three hour incubation with both Bfi (100uM)and 5-MTHF (50uM; 150uM was no different than 50uM, data not shown) augmented the relaxation caused by ach in the rat aorta in the fructose fed insulin resistant rats but had no effect on the controls (figure 7.8 and figure 7.9). There was no sigruficant difference in vasodilatation between the insulin resistant and control aortae treated with and without 5-MTHF if they were harvested 2 hours after an acute methionine load (IP injection 0.15 g/kg) compared to rats given no load (figure 7.10). Conversely, control rats aorta harvested after a 10 hour methioine load had reduced responses to Ach compared to animals given no load. Incubation with 5-MTHF abolished the difference between the groups (figure 7.11). There was no difference between diabetic aortas treated with and without methionine or 5-MTHF after a 10-hour methionine load (Figure 7.11). Interestingly the impaired vasodilation induced by Ach could be attenuated by a co-addministration with 2rng of folinic acid (figure 7.12). Similarly, acute treatment with 100uM BH4 folling a chroninc methioine load also improved vasodilatory responces to Ach (figure 7.13). + Control ( n= 6) c"t Control I- BH, ( n= 7) A Fructose ( n= 5) Fructose + BH, ( n= 7) Acetylcholine (-IogM) SNP (-logM) Figure 7.8. Acute Effects of BH4 (100 pM, 3 hours) on Endothelium-Dependent and -Independent Vascular Relaxation in Control v Insulin-Resistant Aortae. *Fructose treated animals have a significantly (pC0.05) lower relaxation induced by Ach compared all other groups when stimulated with 10-5 and 10" M of acetylcholine. Acute treatment with BH4 improved the Ach induced vasodilatation in the insulin resistant animals while having no significant effect on the controls. There was no significant difference in dilation between any of the groups induced by sodium nitroprusside (SNP). + Control ( n= 5) Control + MTHF ( n= 5) A Fructose ( n= 7) Fructose + MTI-IF ( n= 6) 48-7-6545 98-76 Acetylcholine (-logM) SNP (-logM) Figure 7.9. Acute Effects of 5-MTHF (100 pM, 3 hours) on endothelium-dependent and independent vascular relaxation in control v isulin-resistant aortae. 'Fructose treated animals have a significantly (pC0.05) lower relaxation induced by Ach compared all other groups. Acute treatment with 5-MTHF improved the Ach induced vasodilatation in the insulin resistant animals while having no significant effect on the controls. There was no significant difference in dilation between any of the groups induced by sodium nitroprusside (SNP) 4 Control ( n= 5) Control+Methionine( n= 5) A Fructose (n= 7) Fructose+ Methionine ( n= 6) 8 8 -7 6 -545 Acetylcholine (-logM) Figure 7.10. Effects of a 2 hour methionine loading on control and insulin resistant rats. There was no signhcant difference between the rats treated with methionine in the insulin resistant rats. There was a sigruhcant difference between as was previously demonstrated between the insulin resistant and the control rats. +C (n=6) 4- Control + Methionine ( n= 6 ) #+ Control + Folinic ( n= 4 ) +Control + Methionine+Folinic ( n= 6 ) L -9 -8 -7 -6 -5 -4.5 Acetylcholine (-logM) Figure 7.12. Effect of chronic methionine load (0.15mg/kg/day IP) in the presence and absence of treatment with foiinic acid (2mg/kg/day). I +Control (n=6) -9 -8 -7 4 -5 -4.5 Acetylcholine (-logM) Figure 7.13. Effect of chronic methionine load followed by acute treatment with BH4 (100uM).The methionine load inhibited Ach induced relaxation. Acute treatment with BH4 sigruficantly improved the Ach induced dilation in the control rats after a methionine Ioad. 7.4 DISCUSSION This is the first study to provide evidence indicating that 5-MTHF is capable of binding the pterin site in eNOS. Furthermore, this direct binding to eNOS mimics the orientation and interactions of the natural cofactor Bl%. Secondly our additional in vitro data indicates that 5-MTHF prevents the increased oxidative radicals associated with a tetrahydrobiopterin deficiency in both recombinant and in cultured endothelial cells. Lastly, using animal models we demonstrate that both acute and chronic 5-MTHF supplementation restores nitric oxide dependent endothelial function in hyperhomocysteinernia and insulin resistant rats (BH4deficient). Likewise exogenous tetrahydrobiopterin supplementation also reversed the endothelial dysfunction induced by a methionine load. It is well documented that BH4 alters eNOS activity, however the exact mechanism and the role BH4 plays in directing eNOS to produce NO instead of superoxide remains ill-defineda'aarn. BH4 has been proposed to enhance the binding of arginine, activate heme into a high spin state, stabilize heme and donate electrons for the oxygenation reactionl16;=3. Various analogues of BH4 have also been reported to catalyze NOS activity. Interestingly, only fully reduced tetrahydro-analogues are active and support NO generation where as the dihydrobiopterins are inhibitory. One such compound reported to maintain NOS activity is 5-me thy1 tetrahydrobiopterin (5-methyl-BH4). Rie thmullerla et aI recently demonstrated that 5-methyl-BH4 exhibited a markedly reduced activity towards free oxygen while at the same time maintained catalytic activity, favoring NO production by nNOS. The authors speculated that the methyl group helped to stabilize the pterin ring'". These results support the notion that the analogous methyl group on 5-MTHF will not interfere with interactions in the active site. Secondly, the tetrahydro-reduced state of 5-MTHF also parallels the findings that only the fully reduced analogues of BH4 are catalvtically active. The docking of 5-MTHF to eNOS demonstrates that that (a) 5-MTHF is able to fit into the active site and (b) Interact in an almost identical manner to as the natural cofactor BH4. Although as is illustrated in figure 7.6 the tail of 5-MTHF must be oriented almost at a right angle to the pterin ring and cannot physically mimic the much shorter tail of BH4. W conformation would not be unique given that folate tail is also sharply bent in dihydrofolate reductase. In light of the confusion surrounding the interactions between BH4 and eNOS, it is not surprising that the relationship between folate and eNOS remains unclear. However, Stroes19 et al have recently demonstrated that 5-MTHF is unable to alter NOS activity in enzyme that is completely devoid of pterin. In reality, although a deficiency in BH4 can and does exist in vivo, some BH4 will always be present. Nevertheless, it does contradict evidence suggesting that 5-MTHF can replace BH4 as a cofactor for NOS.The authors speculated that SWcould facilitate the 1electron oxidation of BH4 generating a pterin radical, which may be neessary to for the formation of nitric oxide'? Such an interaction has been demonstrated to exist between ascorbic acid and tetrahydrobiopterin 113z4. Although 5-MTHF is able to scavenge superoxide its effects are 20 fold lower than vitamin C. Furthermore, the concentration of vitamin C and the SOD would be much relevant in vivo for scavenging superoxide compared to 5-M"IIWg. This description may need to be reassessed given that 5-MTHF is able not only fit in the active site but also mimics the orientation and interaction of BH4 and eNOS. In addition, pterin radical formation is thought to occur within the active site. Stabilization by 5-MTHF would have to occur in the active site making it a rather crowded place. Therefore another possible roles of 5-MTHF should be considered. The first is that 5-MTHF cannot bind unless BH4 is already bound to the other pterin-binding site. Such a relationship has been demonstrated by Goren et al who determined that the binding of the first BH4 has an anti-cooperative effect on the binding of the second BH4 '35. This would satisfy Stroes19 observation that 5-MTHF has no activity udess some pterin is present. Asses the activity of recombinant pterin free NOS co-incubated with 5- MTHF and a BH4 inhibitor analogue might resolve this interaction. Both previous clinical studies20 and now evidence from this study highlight the importance of 5-MTKF on reactive oxygen species produced by NOS. We and othersa have shown that 5-MTHF can attenuate superoxide production in recombinant eNOS. This important observation indicates that 5- MTHF is able to directly interact with eNOS rather than having an independent action such as lowering homocysteine, which could then possibly lower cellular oxidative stress. In addition, we have demonstrated that co-incubation with 5- MTHF or sepiapterin and DAHP significantly lowers reactive oxidative species compared to cells incubated with DAHP alone. Similarly, the superoxide specific conversion of DHE to the fluorescent ethidium can be attenuated by co incubation with 5-MTHF (figure 7.5). Interestingly, 5-MTHF also sigruhcantly increased nitric oxide production in cells cultured with the inhibitor of sepiapte~reductase, NAS (n-acetylseretonin) and therefore BH4 synthesis. The half-life of nitric oxide is dependent on primarily its reaction with superoxide or oxyheamoglobin~.The reaction between NO and superoxide is extremely favorable being three times more favorable than the reaction rate for superoxide and superoxide disrnutase236. Therefore, a nitric oxide synthase enzyme partially deficient in BH4 or 5-MTHF producing both superoxide and NO ensures the generation of the potent oxidant peroxynitrite. Furthermore excess superoxide production can lead to other oxidant such as hydrogen peroxide and the production of hydroxyl radicals. It is the dual nature of eNOS that prompted the appropriately named review after the famous Clint Eastwood movie, "The Good (NO) the Bad (peroxynitrite) and the Ugly (Hydroxyl radical)"? We have demonstrated that 5-MTHF is capable of inhibiting superoxide production by recombinant NOS and in endothelial cells deficient in BH4. It can therefore be postulated that 5-MTHF bind to NOS helps to convert eNOS from a peroxynitrite synthase back to a nihic oxide synthase. Indeed, the direct endothelial protective effects of acute and long-term treatment with 5-MlMF has been recentiy demonstrated in patients with hypercholesterolemia and normal homocysteine levels=? Moreover, supplementation of folate improved endothelial function during a methionine load test without affecting homocysteine levels in a group of healthy controls la. Here we demonstrate that 5-MTHF improves endothelial function acutely after a 10 hours methionine load. Interestingly these benefits were not present after 2 hours of methionine loading. Methionine loads in humans do not seem to affect the remethylation cycle until 6 to 8 hour post load. MTHFR 677 TIT subjects do not have elevated homocysteine levels compared- to the wild type or heterozygous individuals however they do have high post loads 6 and 8 hours after a load. Homocysteine levels two hours after a load are often double the normal range and above levels previously associated with increased risk for vascular disease5vj. Therefore, cellular 5-MTHF levels may not become depleted until the later stages of a methionine load as is reflected by an elevation in plasma homocysteine in the T/T subjects. Indeed it has been documented that T/T genotype coders lower cellular 5-MTHF levelsm2. We have also demonstrated that co-supplementation with a methylated form of folinic acid during a chronic methionine load prevented the ensuing endothelial dysfunction (figure 7.13). Although, foIinic acid is able to lower homocysteine levels, acute BH4 supplementation cannot alter cellular homocysteine concentration, but like 5-MTHF aIso prevented the post chronic load dysfunction. Recent studies have indicated that tehahydrobiopterin may be deficient in states of insulin resistance and diabetes while eNOS expression is inaeased by insulin1";~78;183;1". This combination of increased eNOS with the depletion of tetrahydrobiopterin would favor eNOS superoxide production rather than nitric oxide generation. Existing evidence suggests that the decreased nitric oxide production and increased superoxide generation can be linked to insulin resistance. Shinozakill?et a1 have shown that endothelial BH4 Ievels in fructose fed rats were sigruhcantIy decreased and that exogenous BH4 supplementation restored NO production paralleled by a decrease in superoxide. Using a different animal model, Pie per185 et a1 determined that 6-methyl- tetrahydrobiopterin could correct endothelial dysfunction in streptozotocin-induced diabetic rats. We determined that both acute administration of 5-MTHF and BH4 could restore endothelial function in diabetic fructose fed rats. Cellular changes in homocysteine cannot account for the improved endothelid function given that plasma homocysteine levels are lower in diabetic ratsl88. Similarly, 10 mgjkglday of BH4 for 8 weeks in fructose feed rats has just recently been demonstrated to improve endothelial function, insulin sensitivity and blood pressure all of which were attributed to the restoration of eNOS functionll8. Collectively, these studies indicate that exogenous folate supplementation is able to improve nitric oxide dependent vasodilatation independently of its well-known ability to lower homocysteine. Interestingly, in a model of hyperhomocysteinemia (3 weeks high methionine diet) which is known to induce endothelial dysfunction by increasing homocysteine levels, acute treatment with BH4 corrected the impaired nitric oxide dependent vasodilatation. Clearly, BHi cannot have any effect on homocysteine metabolism. Co-supplementation with folinic acid a methylated form of folate similar to 5- MTHF also prevented the hyperhomocysteinemia induced endothdial dysfunction. These data lend credence to the notion that rnethylated folates can directly improve endothelial function and that this may be due to a molecular mechanism linking 5-MTHF to the active site of eNOS in the endothelium. CHAPTER 8 FOLATE AND HYPERTENSION: GENETICS AND PHARMACOLOGICAL STUDIES 8.1 Introduction An elevation in blood pressure is a common clinical finding in patients with strokes, congestive heart failure and chronic renal disease198m;23&2*. Moreover, essential hypertension is believed to be a major contributor to the development of arteriosclerosis~~.It is interesting to note that although there is a large hereditary transmission of essential hypertension, to date there are no genetic lesions that have been identified that conclusively conhibu te to essential hypertension. Nevertheless, multiple mechanisms likely contribute to the pathogenesis of hypertension. For example, decreased vessel compliance in the conduit vessels and increased cardiac output are common contributors to an elevation in systolic blood pressure in the elderly? Other causes such as increased sympathetic tone or impaired renal function have also been implicated. In addition impaired nitric oxide bioavailability in the small resistance vessels can contribute to essential hypertension via increasing peripheral vascular resistanc$4tgQ. Even modest reductions in blood pressure result in sigtuCicant cardiovascular protection underscoring the importance of hypertension to overall mortality and morbidity201. Endothelial Function in Hvpertension: One of the hallmarks of hypertension is an increase in peripheral vascular tone leading to increases in systemic vascular resistance. Endothelid dysfunction, has been implicated in the pathogenesis of increased vascular tone in hypertension2~.Endothelial cells respond to mechanical stress and hormonal signals and subsequently contribute to basal vascular tone by regulating arterial tensionz*. It has been demonstrated that hypertensive patients have reduced vasodilatation to acetylcholine with preserved endothehum-independent relaxation to sodium nitroprusside Therefore decreased production of the vasodilator nitric oxide either contributes to or is a secondary event caused by hypertension. Furthermore, endothelid dysfunction antedates the development of hypertension; offspring's of parents with hypertension have diminished endothebum-dependent vasodilatation in the pre-hypertensive state2u. Genetics and hvpertension: The importance of nitric oxide in the regulation of resistance arteries coupled with the genetic basis of hypertension, suggests that mutations affecting eNOS might contribute to increased vascular resistance and in turn an elevation in blood pressure. Three polymorphisms in eNOS have described which have been shown to have important meaning in cardiovascdar diseasezz5;=29. The first involves a T-786+C point mutation in the AP-l binding site of eNOS. The dele frequency in a Japanese population with vascular disease was reported to be 9%". A luciferase reporter gene construct revealed that the T-786+C point mutation reduced promoter activity by 52%.Hypoxia induced activation of the construct was also reduced compared the wild type. Clinically subjects with the T-786jC mutation were at 3.05 times the risk for coronary spasm=. The relevance of the mutation in the Caucasian population has not been determined. However the authors concluded that the dele was likely to be over represented in the Japanese population because of their much high incidence of coronary spasm. Another missence mutation caused by a G+T substitution at nudeotide 894 of eNOS (cDNA) results in the replacement of a Glu298Asp has also been recently reported to impair eNOS functionmm. Shimasaki229 et al. determined that this mutation was siguhcantly over represented in patients with cardiovascular disease and like the T-786-C mutation increases the risk for coronary spasm *. Lastly, a tandem repeat in eNOS located at intron 4 was associated with a four or five 27 base pair repeat (4a). Recent studies have demonstrated that patients with the 4a repeat were at increased risk for coronary artery disease. However, a recent study by Yoshimuraa et a1 has revealed that the 4a repeat and the T-786+C mutation were in linkage disequilibrium. Foiate and hypertension: In light of our recent studies demonstrating the effect of 5-MTHF on nitric oxide bioavailability an argument could be made supporting the notion that impaired folate status and therefore hyperhomocysteinemia, could contribute to decreased nitric oxide production and hypertension. Few studies have investigated the relationship between homocysteine or folate and hypertension. There was a weak positive association found between homocysteine levels and blood pressure in the Hordaland heart study2a but this relationship was primarily evident in younger men and women. Sutton and Tyrell245 also found a positive relationship between fasting hyperhomocysteinemia and isolated systolic hypertension. Fiorina et al** determined that plasma homocysteine and folate levels were related to arterial blood pressure in type 2 diabetics. Interestingly, higher folate levels were sigruficantly associated with lower systolic and mean arterial pressure, although there was not a correlation between vitamin B12 and blood pressure. Another report also determined that subjects with hyperhomocysteinemia (>lBumol/L) were at three times the risk of having hypertension. The interactions between the common MTHFR 677 C+T and the MS 2756 A+G mutation were not investigated. It is also not clear if folate supplementation can alter blood pressure. However, it has just very recently been demonstrated that oral supplementation of Bh,which has similar effects as folate does on endothelial function, not only prevented endothelial dysfunction and oxidative stress but also lowered blood pressure in fructose induced insulin resistant ratstla. The objectives of this study were to investigate the associations and interactions between two common mutations invoived in 5-MTHF metabolism (MTHFR 677 C+T, MS 2756 A+G) and the novel eNOS T-786+C. In addition, the effects of acute and chronic treatment with folhlic acid on blood pressure and resistance artery endothelid function were examined in spontaneously hypertensive rats (SHR). 8.2 Methods Subjects: The study population reported herein is a sub study from a larger longitudinal dinical trial designed to assess the long-term effects of brachial artery flow-mediated dilation on cardiovascular end points (The FATE study). The cohort consists of active and retired middle-aged men from the Calgary Fire department. Patients were excluded if they had a previous history of documented coronary artery disease peripheral vascular disease or cerebrovascular disease. Blood pressure was measured in the supine position using a sphygmomanometer. Patients were considered to be hypertensive if they were being treated for hypertension or if their systolic and diastolic blood pressures were greater than 14OmmHg and 90rnmHg respectively. Subjects treated for hypertension were included in the odds ratio calculation but biochemical variables were exduded from analysis. Biochemical analvsis: Homocysteine was measured using a fluorescence polarization hunoassay. Total homocysteine is reduced with dithiotheietol and homocysteine is then converted to Sadenosyl homocysteine (SAH) by bovine SAH hydrolase and excess adenosine. SAH and a labeled fluorescent tracer compete for the SAH antibody allowing quantification using polarized fluorescent light. The assay has a sensitivity of <05umol/l. The CV below the normal range is less than 5%. Both folate and 812 were measured with microparticle enzyme irnmunoassay technology. Specific antibodies for either BIZ or folate are bound to latex particles and then placed in a reaction chamber. A secondary alkaline phosphatase labeled conjugate antibody is added creating an anitbody-analyte-conjugatesandwich. 4-methylumbelliferyl phosphate (UMP) is hydrolyzed by the alkaline phosphatase into the fluorescent product methy lumbellferone allowing quantification. The me thodology is highly sensitive (2.0nmol/l) with a CV of 5.7% at the lower normal range. Serum lipid profiles and glucose were measured by standard methodology available in the Foothills Hospital. Genetic analvsis: Genomic DNA was isolated as previously described. The eNOS T-7864mutation was analyzed by PCR reaction by using the following primers 5'ATG CTC CCA GGG CAT CA-3' and SGTC CTT GAG TCT GAC AlT AGG G-3' DNA. The reaction was carried out in 50 ul tubes containing: 5ul of 10xTaq buffer 2.5 ul of a 10 uM solution of each primer, 5.0~1of 2.0mM dNTP mix 1.0 ul of 50 mM MgClz 28.5111 of dH20 and 5 units of Taq polymerase and 50 ng of sample DNA. The reaction was run under the following conditions: 94OC 7 min and then cycled 30x at 940C 30 sec, 57 OC 30 sec, 72OC 30 sec, followed by 720C for 7m.h. 7.5 ul of the PCR product was digested with 10 units of Ngo MIV, 0.5~1of lox read 9 (gibco) and 1.5 ul of dHtO and then run on a 4% Nusieve agarose gel. Animal studies: SHR rats were purchased at 9 weeks of age from Charles River and housed in pairs. All surgeries were performed on anaesthetized rats according to a research protocol consistent with the standards of the Canadian Council on Animal Care and approved by the local Animal Care Committee of The University of Calgary. Mean, systolic, diastolic, heart rate, and activity were all recorded using telemetry implants from Data Sciences. Monitoring probes were implanted into the descending aorta and fixed with tissue glue. The probe is a fluid filled antithrombogenic catheter, which directs the impulse to a transmitter body sewn into the abdominal wall. Rats were then randomly separated into two groups one receiving saline and the other folinic acid (lOmg/ d/ kg, i. p) at the same time each morning for the duration of the study. Rats were separated and allowed to equilibrate in the monitoring cage for 15min. Data was then recorded and averaged over 40 min. Measurements were recorded at baseline (day I), 3 hours post treatment on day 1, day 7 and on day 14 in each rat. Blood for homocysteine analysis was collected by cardiac puncture. Pre~arationof Arteries: The small intestine were rapidly removed and placed in cold physiological salt solution (PSS) of the following composition (in rnM): NaCl 118; KC1 1.7; CaC12 2.5; KH~OJ1.2; MgSOd 1.2; NaHCa 25; dextrose 11.1. The pH of the PSS after saturation with 95% a-+ 5% Ca-gas mixture was 7.4. Segments of the second order branches of mesenteric artery were gently dissected from the mesenteric vascular bed and immediately placed in cold PSS. Adherent connective tissue was removed and a 2 mrn segment of artery was then prepared for recording isometric force development in a wire myograph as previously described247. Briefly, two tungsten wires (25 diameter each) were inserted through the lumen of the vessel and the tissue placed in a 10 rnl myograph chamber containing gassed 370C PSS. One wire was then attached to a force transducer and the other connected to a micrometer. After a 30-minute equilibration period, during which period the temperature was increased to 37C,arteries were stretched in a stepwise fashion to a resting tension of 2 mN, which in preliminary studies was found to be the optimal preload for force development in these blood vessels. Tissues were routinely allowed to equilibrate for one hour before the start of the experiments. Isometric tension was recorded at a rate of 5 measurements per second (sample interval 0.2 second) digital data was collected and analyzed commercial software (Axotape 2.0, Axon Instruments). Arterial rings were pretontracted with phenylephrine (PE, 2 VM) and subsequently relaxed by applying acetylcholine cumulatively. 8.3 Results Genetic analvsis: The distribution of the genotypes is tabulated below (Table 8.1). Fifteen percent, 3%and 13%of ow population was homozygous mutant for the eNOS T-786-+C, the MS 2756 A+G and the MTHFR 677 C+T genotypes respectively. The distribution of the MS and MTHFR genotypes is similar to what has been previously reported 69. The eNOS C allele frequency of 38% is much higher than the 306 reported in the control population of a Japanese cohort m. Table 8.1. Distribution of eNOS T-786+C and MS 2756 A+G polymorphisms. - - wildtype Heterozygous Mutant MTHFR 677 C+T 39.09%(231) 48.22% (285) 12.69%(75) Blood Pressure: Systolic and diastolic blood pressures ranged form 102mrnHg to 19OmmHg and, 57mmmHg to IOSmrnHG respectively. Mean BP levels are reported in table 1. Regression analysis revealed no association between vitamin Bla RBC folate, serum folate (n=150) and blood pressure. However, there was a strong relationship between homocysteine and age and diastolic and systolic blood pressures. In addition, there was a positive association between age and homocysteine levels. However, when age was adjusted, the association between homocysteine and blood pressure was not sigruficant. Complete data including all measured variables and blood pressures were available in 455 patients. 36 patients had a systolic and diastolic blood pressure above llOmmHg and 90mmHg respectively and were considered hypertensive. The demographics of the patient groups are depicted in table 2. Table 2 Patient Demographics (mean and (SD)). Normotensive Hypertensive Age years 44(9.3) 55(11) Vitamin 812 pmol/L 329(157) 285(117) Weight kg 90(12) 94(14) Cholesterol mmol/ L 5.13(0.936) 5.8(0.94) HDL mmol/L 1Z(0.27) l.lB(0.24) LDL mmol/ L 3.19(0.8) 3.6(0.9) Trig1ycerides rnmol/ L 1.55(1.08) 2.4(1.4) RBC Folate moI/L 1158(262) 1132(314) Serum Folate nmol/ L(n=152) 33(9.5) 32(9.3) Homocysteine urn011 L 8.7(2.4) 9.5(2) Systolic BP 122(10.8) 155(14) Diastolic BP V(8.4) gS(7.6) Relationship between ~enotvpesand Blood Pressure: The genotypes were then subdivided into two groups, those with and without hypertension. There was no signihcant difference between the allele frequencies for the MS 2756 A+G or the MTHEX 677 C+T genotypes between subjects with and without hypertension. In addition, both the MTHFR 677 and MS 2756 did not sigruhcantly affect mean blood pressure levels. Conversely, there were sigruficantly higher number of subjects in the hypertensive category with the C dele compared to the subjects with svstolic and diastolic pressure below 140 mmHg and 90 mmHg respectively (table 8.4). Subjects with the C/Cgenotype were 1.6 (95%CI: 1.08 to 2.31) times more likely to have hypertension compared to those with TIT and C/T genotypes. One-way ANOVA followed by Scheffe's post-hoc analysis revealed that diastolic pressures were significantly higher in the C/C vs. the T/C (p=0.005) or vs. the T/T subjeds (p<0.003).Similar differences were evident when systolic pressure was analyzed (figure 8.1 A and B). Table 8.3. Distribution of eNOS T-786+C in those with essential hypertension, defined as a systolic blood pressure above l4OmmHg and a diastolic above 9OmmHg. TIT T/C c/c Hypertensive 33%(n=12) 36%(n=13) 31 % (n=ll) Non-Hy pertensive 38 % (n=158) 47%(n=197) 15%(n=62) Figure 8.1. Distribution graphs stratified by the eNOST-786+C genotypes. A) Diastolic blood pressure B) Systolic blood pressures. One-way ANOVA with Scheffe post hoc analysis. Genotypes and Blood pressure: The effect of the MS, M'TH.FR, and eNOS genotypes on BP was determined by two way and three way ANOVA. There was not sigmficant effect between the eNOS C-786+T and MTHF 677 C+T and the MS 2756 A+G or any pair-wise comparison. However there was a significant difference in systolic blood pressures between the G/G-C/Cand A/ A-C/C MS eNOS genotypes (p=0.04).Subjects with the G/G-C/C genotype had mean systolic pressure of 132 MnHg compared to 118mmHg in the G/G-C/C (MS eNOS) genotypes. This must be interpreted with some caution given the low number of individuals with the G/G-C/C genotypes (figure 8.2). Animal studies: Telemetry data was recorded at baseline, 3 hours post and 7 and 14 days after folinic acid treatment in 14 SHR rats. Mean blood pressure, systolic, diastolic heart rate and activity were recorded and averaged over forty minutes as is illustrated in figure 8.3. All variables were assessed using repeated measures ANOVA. Acute treatment with folinic acid did not significantly lower blood pressure levels in the SHR animals. Mean pressure and standard deviations were 152 (17) in the treated and 139 (12.3) in the sa1i.egroups respectively. Activity, levels were also not sigruficantly different between the two groups (table 8.4). Interestingly, there was also not a sigruficant difference in homocysteine levels between the rats treated with folinic acid or saline. There was no difference in weight between the two groups at the beginning or at the end of the study. Table 8.4. Mean systolic, diastolic, blood pressure, heart rate, and activity over 40 min on day 1, three hours after first injection day 1, day 7 and day 14 in the treated (10 mg/kg/day folinic acid) and saline SHR rat groups. Comparisons were made using repeated measures ANOVA. Telemetry Data Group Day1 Day 1+3 Day 7 Day 14 hours Folinic 173(7.2) 178(15) blood pressure me 170(5.8) Diastolic Folinic 120(8) Blood pressure 118(3) Folinic 147(7) Blood Pressure Saline 144(4) Folinic 363(24) Heart Rate Saline 372(12) Folinic 6(3.9) Activity Saline 8(3) Telemetry Data i 450 , j 200 1 HeartRate I/ 1 150 i-.L,..--. j- Systolic 'Ii I yd:z\4%,-. 1 I ---.JI. -- 1 / 100 I I Mean Pressure i 1 ! I I I - - - -Diastolic I i I/ ~~lIlllbLl1 -Activity 1 ) i 00)oObU3V)~CVOQ)~1 I I TT~~~~b~~OO~I I F F F i I Measurment points I I Figure 8.3. Data recording from one rat recording session, heart rate, systolic pressure, mean pressure, diastolic pressure and activity (from top to bottom). -log (ACh), M 9 8.5 8 7.5 7 Figure 8.4 Chronic folinic acid treatment improves endothelial bction in mesenteric arteries oCSHR rats. p 8.4) 8.4 Discussion Here we report that the T-786jC mutation in the AP-I promoter binding site is strongly associated with hypertension in a healthy adult male population. Mean systolic and diastolic blood pressures were sigruficantly elevated in the homozygous mutants compared to the wild type and heterozygous patients. In addition, the odds ratio of having hypertension, defined as a systolic and diastolic blood pressure above l4OmmHg and 9OrnmHg was 1.6 for the C/C subjects compared to the T/ C and T/T individuals. Interestingly, patients with the homozygous C/C mutation combined with the G/G genotype for MS 2756 A+G had sigruficantly lower BP compared to those with the eNOS C/Cand wild type MS 2756 A/ A genotype. There was however no correlation between homocysteine, RBC folate, or the MTHFR 677 CjTpolymorphism and systolic or diastolic blood pressure. Nitric oxide is produced from the amino acid arginine in response to a number of physiological stimuliI7 Hypertensive patients have impaired NO mediated dilation of the small resistance vessels and it is though that this directly leads to an elevation in blood pressurea"? However the mechanisms leading to the impaired release of NO from the endothelial cells is complex and may involve a number of intracellular signaling pathways. The majority of signaling cascade leading to the activation of eNOS begins with cell surface receptors coupled to G- proteins, activating either phospholipase C or adenylyl cydase, and then ultimately leading to the subsequent production of NO. Given that two distinct pathways primarily activate eNOS one could speculate that if only one signaling cascade is inhibited that the stimulation of the other would lead to a normal release of NO. However, hypertensive patients have impaired NO vasodilatation when stimulated with either Ach or substance P suggesting an impairment down stream from these signal cascadesa~~l7.Evidence from eNOS knockout mice, who have increased vascular resistance and higher systematic blood pressure, confirms the direct role of eNOS derived NO in hypertension2". Moreover, treatment with NOS inhibitors in normal mice leads to an elevation in blood pressure while sigruficantly lowering blood pressure in the knockout micS48. It is important to mention that other homeostatic systems were unable to compensate for the knockout of eNOS and loss of endothelid derived NO, perhaps highlighting the importance of nitric oxide in the homeostasis regulation of blood pressure? Similar relationships between NO production and hypertension have been found in humans. The effects of L-NMMA, are diminished in hypertensive patients compared to healthy controls further corroborating evidence that NO availability and not the structural vessel changes are directly linked to hypertensions? Conversely, an argument could be made supporting the no tion that 1-arginine levels are diminished and result in the decreased production of NO. However, supplementation with L-arginine does not improve endothelid dependent responses to A&'*. In addition, this observation weakens the possibility that endogenous nitric oxide synthase inhibitors such as ADMA directly contribute to hypertension. Therefore, it is possible that disruption of eNOS expression caused by the eNOS T-786-+C as is demonstrated by this study perhaps in combination with other polymorphism such as the eNOS G1398Asp account for at least some of the decreased production of NO by the small resistance arteries. We may indeed be under representing the importance of this association because of our healthy population in whom only 36 of 455 patients had systolic and diastolic pressures above 140mmHG and 90 mmHg respectively. Although there is no direct evidence on how homozygosity for the MS 2756 mutation decreases blood pressure in C/CeNOS genotypes it is likely that an elevation in 5-MTHF caused by an impaired MS enzyme enhances nitric oxide production and at the same time decreases superoxide production. We have recently demonstrated that 5-MTHF is able to bind eNOS and reduce the production of superoxide within endothelial cells and stimulate NO production. Clinically we have also demonstrated that the MS G dele sigruficantly decrease the number of recurrent events in patients with cardiovascular diseasG3 and in addition improves flow mediated vasodilatation. The later observation emphasizes the effects of the MS 2756 gene on nihic oxide release form the endothelial cells. This interaction between a mutation that impairs protein expression and the other that likely increases 5-Wsuggests that co-factors such as 5-MTHF and BH4 are deficient in hypertension and contribute to impaired bioavailability of nitric oxide. Indeed, a recent report demonstrated that exogenous B& supplementation to fructose fed hypertensive insulin resistant rats si&cantly lowered blood pressurella. Given the similar effects between 5- MTHF and B& it is reasonable to hypothesize that if BH, is able to attenuate blood pressure then 5-MTHF might also be able to lower blood pressure in this model. This hypothesis is indirectly supported by the our previously mentioned evidence indicating that 5-MTHF like BH4 is capable of restoring endothelial dysfunction in the insulin resistant rats. The MTHFR 677 C+T mutation had no effect on blood pressure in our population. This is not entirely surprising given that clinical59 and structural studiesa have demonstrated that the MTHFR phenotype is only expressed when folate levels are marginally depleted. Clearly our population is not deficient in folate. AIthough as previously discussed the MTHFR T/T have sigruficantly lower serum folate levels and higher homocysteine levels indicating at least residually impaired activity even in our health population. These moderate effects on homocysteine and folate may nonetheless be contributing to the strengthening association between the MTHFR T allele and premature vascular disease. It is not dear why the MS 2756 G/G genotype attenuates and the WR677 C+T mutation has no effect on blood pressure. Perhaps unlike the MTHFR T/T genotype the MS G allele only expresses its phenotype in conditions where 5- MTHF is not limiting and therefore when folate levels are replete increases 5- MTHF. Homocysteine may directly contribute to an elevation in blood pressure by promoting smooth muscle cell proliferation249 and therefore inhibiting vessel compliance or may be a secondary maker for age. We determined that although homocysteine correlated with systolic blood pressure it was likewise correlated with age. It is well documented that aging is associated with an elevation in blood pressure. Nevertheless other studies have demonstrated that homocysteine contributes to intima thickness independently of age? Lower folate levels are undoubtedly associated with a rise in homocysteine levels and cell culture studies have demonstrated that folate supplementation can indeed reduce proliferation independently of endothelial cellslrn. It is therefore unlikely that this effect is induced by an improvement in nitric oxide bioavailability but rather through improved remethylation. Folate may well influence vessel wall remodeling by its well described homocysteine lowesing effects and by improving NO bioavailability'g-3. The effects of the MS 2756 A+G mutation and its interaction with the eNOS-786 C/C genotypes should and will be confirmed in a larger population. Furthermore, the high prevalence and interaction of the eNOS T-786+C point mutation certainly needs to be address in other population at risk for vascular disease. Animal studies: Although numerous factors such as smooth muscle cell hypertrophy increased vascular tone in resistance vessels, elevated sympathetic activity, disrupted catecholamine metabolism, altered central vascular regulation and increased oxidative stress have been associated with hypertension in the SHR, the underlying mechan.km(s) remain unclear. We found that chronic treatment with folinic acid while not improving blood pressure sigruficantly improved acetylcholine-mediated vasodilatation in the resistance mesentery arteries. Folinic acid is a stable methylated form of folate that is rapidly, absorbed and converted to 5-W.We have previously demonstrated that 5-Wis able to bind the pterin site of eNOS and like BH4 prevent the uncoupling of eNOS favoring the production of nitric oxide over superoxide. It is therefore likely that supplementation of folinic acid increases 5-MTHF levels and improves nitric oxide release from the mesenteric endothelium. Supplementation with folinic acid did not sigruficantly lower homocysteine levels and therefore it is not reasonable to speculate that the improved endothelial hction in the lolinic acid treated group is a result of decreased homocysteine. Nevertheless, two weeks of treatment did not lower systemic blood pressure in the SHR. This somewhat paradoxical situation where folinic acid improves resistance artery function while at the same time having no effect on systemic blood pressure may be a result of the secondary long-term effects of prolonged hypertension due to decreased NO bioavailability. Al tema tivel y, impaired mesentery artery response to acetylcholine may be a result of free radical production by sources other than eNOS that perhaps decrease intracellular Bfi levels. Recently, Cosentinol39 et al. demonstrated that dysfunctional eNOS is one source of superoxide in pre-hypertensive SHR rats. Supplementation with BH, significantly reduced superoxide production in the SHR while not having an effect on the control ratsl39. Similar, effects were found in the spontaneously hypertensive stroke prone rats '3.Surprisingly, neither study investigated chronic treatment on blood pressure. Evidence does exist which suggests that high intra-arterial pressure elicits the release of superoxide '0. Rerefore it remains to be determined if the increased production of superoxide by BKdeficient eNOS is causative or a response to other initiating events leading to hypertension. Xanthine oxidase has also been proposed as a contributor to oxidative stress in the SHR. Susuki et dzt demonstrated that NO is well synthesized by SHR rats but is eliminated by NO-scavenging systems within the microcirculation. Interestingly, the inhibitors of xanthine oxidase, oxypurinol and (-) BOF42Z both reduced the BP and improved endothelial function in SHR. Another report using a molybdenum deficient/tungsten replete diet also lowered mean blood pressure in SHR which can only be attributed to inhibition of xanthine oxidas@? Therefore, the B& deficiency may result because of increased superoxide production by xanthine oxidase further amplifying free radical production in the endothelium. The beneficial effects of folinic acid treatment evident in this study could be therefore due to an endothelial cellular deficiency in Bfi being overcome by increased 5-MTHF binding to the pterin site in eNOS. Studies investigating the long-term effects of folate and B& in pre-hypertensive rats are warranted to fully understand this issue. Similarly Bfi had been shown to reduce blood pressure in hctose fed hypertensive rats but the effect of 5- MTHF or folinic acid has not been investigated. Whatever, the exact mechanism may be the observation that chronic foIinic acid treabent improves endothelid function in hypertensive rats has important implications for the pathogenesis and treatment of systemic vascular resistance in hypertension. CHAPTER 9 SUMMARY Patients on hemodialysis suffer horn an excess risk of atherosclerotic vascular disease and thrombosisl*. Traditional risk factors such as smoking, hypertension, dyslipidemia, and diabetes mellitus cannot fully explain this enhanced riskla. Indeed, it is now well documented that an elevation in homocysteine levels is an independent risk factor for vascular disease. Patients undergoing hemodialysis invariably have elevated homocysteine levelsl~7.In addition, evidence exists that indicates that hyperhomocysteinemia in ESRD is primarily a result of impaired remethylation and not transsulfuration of homocysteine. This prompted us to investigate treatment of hemodialysis patients with pharmacological doses of vitamin Bt2 and folate. We saw a substantial lowering with increasing doses of 812 and little or no change with further treatment with folate above 1mg/day [Chapter 21. The sipficant lowering effects of BIZ despite highly elevated serum Bt2 suggested a fundamental impairment in the metabolism of BIZ. In Chapter 3, we addressed this hypothesis by examining the Blzdependent variable MMA. While high dose vitamin B12 therapy lowered MMA in ESRD, the levels were not normalized. Other reports have speculated that inborn errors of metabolism in MMA can lead to an impairment of the glycine cleavage enzyme. Furthermore, our previous studies similarIy found elevated glycine levels in the hemodialysis patients studied. InterestingIy, glycine levels in our study population correlated very well with MMA IeveIs. We have therefore proposed this as a possible mechanism of impairing remethylation of folate in ESRD. Ln addition, a decreased GCS a secondary confounding effect of depletion of serine could further amplify impaired remethylation of folate and the subsequent conversion of homocysteine to methionine. This proposal cannot however explain the extremely high circulating levels of MMA. It is possible that uremia impairs the reductive activation of 812 thereby causing an elevation in MMA as well as contributing to hyperhomocysteinemia. The common MTHFR 677 C+T mutation, while aggravating homocy steine levels did not substantially affect homocysteine levels in the ESRD population, supporting our proposed metabolic complication. Chapter 4 investigated the effects of three common polymorphisms involved in folate and homocysteine metabolism. Firstly, MTHFR C+T, as has been previously reported sigruficantly elevated fasting homoc ys teine levels, but did not signihcantly influence circulating homocysteine after a two-hour methionine load. In this population, neither the T nor the C alleles altered RBC folate, vitamin BE, or MMA levels. We also similarly investigated the effects of two other common polymorphisms in MS and MTR. We are the first to investigate the effects of the two hour methionine load on the MS 2756 A+G and MTRR A-G point mutation in subjects with cardiovascular disease. Likewise, both the MS 2756 A+C and the MTRR A+G mutation did not &ect the vitamin 812 or MMA status. However, heterozygousity for the MS 2756 G allele siguhcantly elevated RBC folate content. Chapter 5 examined the effect of heterozygosity for the MS 2756 G dele on recurrent cardiovascular events. We found a highly signihcant reduction in the number of recurrent events in the CAD patients with the MS 2756 A/G genotype compared to the A/ A subjects. Similarly, patients with the A/G genotype had fewer myocardial infarctions, bypass surgeries and heart failures compared to the A/ A subjects. In chapter 6 the effects of the MS 2756 A+G and the MTHFR 677 C+T on homocysteine, folate levels and endothelid function were examined. Interestingly, there was a large discrepancy between the results given by the serum and RBC folate levels in subjects with the homozygous T/T MTHFR 677 genotype. Serum folate levels were substantidy lower in the homozygous T/T patient and were conversely had significantly higher RBC folate levels. Given that a larger number of patients with the T/T genotype will have hyperhomocysteinernia and therefore an increased likelihood of RBC folate levels being measured this falsely high value in 1045%of the population suggests that our current methodology needs to be resolved or separate reference ranges be established according to the M'THFR 677 C+T genotypes. The protective effect of heterozygosity for the A/G genotype extended to a healthy adult male population. Subjects with the MS 2756 A/G genotype had a significantly greater vasodilatation compared to those the with A/ A genotype. Because the FMD is directly a result of nitric oxide produced by the vascular endothelial cek, it is reasonable to speculate that the MS 2756 A+G mutation either increases nitric oxide production or decreases nitric oxide degradation. The intriguing effect of the MS G dele on both recurrent cardiovascular events and endothelial function lead to the investigation between folate and endothelial derived nitric oxide production (Chapter 8). Using cell culture studies and animal models of hyperhomocysteinemia and BIZdeficiencies, we demonstrated that 5-MTHF functions similarly to tetrahydrobiopte~by attenuating superoxide production and facilitating nitric oxide release from endothelial cells. We further determined that 5-h4TKF is able to fit into the active site in eNOS. In addition, computer modeling suggests that the most energeticdy favorable orientation is almost identical to the binding and interactions of tetrahydrobiopterin. Our data is the first to suggest that 5-MTHF can bind directly to the pterin-binding site of eNOS. In addition, the decrease in superoxide and peroxynitrite produced within the endothefial cells supplemented with 5-MTHF complements the modeling results. Numerous reports have demonstrated that the uncoupling of eNOS is a fundamental contributor to endothelial dysfunction in hyperlipidemia, smoking, diabetes and hypertension. The discovery of the similar effects between 5-MTHF and BH4 has important clinicai implications. Firstly, the effects of Bfi on the restoration of endothelial function are well documented. However, supplementation with 5- MTHF or folic acid may be superior to exogenous Bhsupplementation because both of these compounds are more stable than B& and additionally they are non-toxic. Evidence, supporting the notion that 5-MTHF improves endothelial function, in addition to our studies that have demonstrated that the G allele of the MS 2765 genotype is protective, indicates that in addition to its homocysteine-lowering effects, folate status may reduce the risk for vdar disease through improved nitric oxide production. Indeed this is very well supported by a recent study that determined that CAD patients had lower serum folate levels compared to the controls "2. In the final chapter we investigated the role of homocysteine and folate on hypertension in a large population of healthy male adults. Because of the direct evidence that demonstrates a clear interaction between 5-MTHF and eNOS, we also examined the association between the surprisingly common point mutation in the 5' flanking region of eNOS. This is the first study to describe the incidence of the eNOS T-786+C mutation in a Caucasian popuiation. It is Likely that ethnic differences in the incidence of the C allele account for the differences between our population and what has been previously describedlrn. Furthermore, we found a sigruficantly lower systolic blood pressure in the eNOS C/Cpatients who were also homozygous at the MS 2756 GIG site, compared to patients with the eN-788 C/C and MS 2756 A/ A genotype. This observation is supported by our additional animal studies where nitric oxide-dependent vasodilatation of the small resistance mesenteric artery was improved in SHR rats treated for two weeks with folinic acid. CHAPTER 10 FUTURE STUDIES 1. TARGETING ENDOTHELIAL FUNCTION AND HYPERHOMOCYSTEINEMIA IN END STAGE RENAL DISEASE The present proposal is specifically targeted at examining the interaction and pharmacological modulation of endothelid function and hyperhomocysteinemia in patients with end stage renal disease (ESRD). Atherosclerotic cardiovascular disease is the leading cause of mortality in patients with ESRDm. Current estimates indicate that patients with chronic rend failure have up to a 19-fold increased risk of cardiovascular events compared with control subjects. Importantly, this mammoth burden of atherogenicity cannot be fully ascribed to the clustering of coronary artery disease (CAD) risk factors (diabetes, hypertension, dyslipidemia) frequently present in patients with ESRDW These observations have led to a surge of research efforts geared towards understanding and improving atheroscierotic vascular disease in patients with ESRD. One such factor that has attracted much current attention is the amino-acid homocysteine. Hyperhomocysteinemia has emerged as an independent risk factor for atherosclerosis. Plasma homocysteine levels are elevated in approximately 90% of patients with ESRD'R. Furthermore, homocysteine levels correlate closely with mortality in patients with ESRD. Despite the growing evidence implicating elevated homocysteine levels to vascular disease, the results of conventional treatment strategies (with folic acid) have been disappointing in ESRD patients~~l~~3~. Endothelid dysfunction represents the initiating event in atherosclerotic vascular disease. Endothelium-derived nitric oxide (NO) plays a vital role in vascular health and serves to inhibit the key processes involved in the development and stabilization of the atherosclerotic plaque. Derangements in the functional activity of the endothelium (demonstrated by a diminished NO- mediated dilation) result in a pro-thrombotic environment that facilitates vasospasm, platelet aggregation, leukocyte adhesion and the genesis of atherosclerosis~~~~6.Indeed, abnormalities in endothelium-dependent vasomotion are well documented in patients with ESRD and precede the development of atherosclerosis. Although the mechanisms linking ESRD to endotheIial dyshction are poorly defined, it is plausible that persistent and treatment resistant hyperhomocysteinemia is a marker for or represent one of the central tenants. Therefore, the assessment and treatment of hyperhomocysteinemia and endothelial dysfunction have important clinical and prognostic implications for patients with ESRD. The present proposal aims to examine the acute and chronic effects of folinic acid on endothelial hnction and hyperhomocysteinemia in ESRD patients. The central hypothesis is that folinic acid, the active/methylated form of folic acid, will improve endothelial function both directly (by augmenting NO production/ release) and indirectly by normalizing plasma homocysteine levels. This hypothesis is based on the following observations: First, defects in the re- methylation of homocysteine are involved in the pathogenesis of hyperhornocysteinemia in ESRD. The re-methyIation pathway serves to lower plasma homocysteine levels by the re-conversion of homocysteine into methionine by the transfer of a methyl group to homocysteine. Since folinic acid has an endogenous methyl group attached to its chemical structure it will augment the cellular re-methylation of homocysteine to a greater degree than folic acid (which requires to be methylated prior to activation). As a result, long term IV folinic acid treatment will normalize homocysteine levels and improve endothelial function. Second, as discussed in chapter 7, data from our group indicate that 5-methyltetrahydrofolateexerts a direct endothelial protective effects; Although, it has not been established if folinic acid itself can bind eNOS, in vivo supplementation with folinic acid has improved endothelial function in an animal model (Chapter 7). Therefore, folinic acid treatment may improve endo thelial function through a direct effect on NO availability. Through augmenting endothelial function, these studies may serve to uncover novel treatment strategies for improving one of the most consequential indicators of ESRD, i.e. accelerated atherosclerosis. Hyperhomocvsteinemia- - in ESRD: Disappointing Results of Conventional Treatment Modalities: Rational for using folinic acid: Why is it important to counter hyperhomocysteinemia in ESRD? The answer to this question lies in the accumulating evidence implicating homocysteine as an independent risk factor for atherosclerotic vascular disease. Indeed the effects of hyperhomocysteinemia on endothelid dysfunction are well documented in both experimental and clinical settings. In patients with hy perhomocysteinemia and preserved renal function, conventional treatment with folk acid lowers homocysteine levels and improves endothelium-dependent vasomotion~~31342). In contrast, folic acid supplementation does not notmalize homocysteine levels in ESRD nor does it improve endothelial hction~.The lack of high dose vitamin regimens to normalize homocysteine levels in ESRD has been confirmed by a variety of investigators '65. An underlying difference in the metabolic and cellular handling of homocysteine in ESRD (vs. control) may be responsible for this apparent treatment-resistance. In light of the limited success with folic acid and vitamin 812 regimens in lowering homocysteine levels, studies have attempted to address the potential benefit of treating ESRD patients with the active and methylated forms of folic acid (folinic acid and 5-methyltetrahydrofolate). Indeed, in one study, long term folinic acid treatment (50mg/ week, with vitamin 86) signihcantly lowered homocysteine in hernodialysis dependent patients '49. In another study, Pemass et al. observed similar beneficiai effects while employing 5- methyltetrahydrofolate, the active cellular form of form of folic acid. Importantly, the effects of this methy lated folate on homocys teine levels were markedly better when compared to folic acid. Conversely, Bostomln et al have demonstrated that oral supplementation with 5-MTKF conveyed no additional benefit compared to folic add treatment in terms of lowering homocysteine levels and the authors perhaps prematurely concluded that supplementation with methylfolates was ineffective in ESRD. As illustrated in chapter 3 it may indeed be possible that during a 812 deficiency 5-cellular MTHF cannot be maintained and/or is unavailable to the vascular tissues. Clearly, ESRD patients are B12 have a cellular deficiency in 812 despite elevated circulating senun 812. Other, reasons such as increased stability, decreased cost and weekly LV. access also favor the I.V. supplementation with folinic acid over 5-MTHF. In addition, increasing the dose of methyl.3ted folate may continue to lower homocysteine levels unlike foolic acid, which does not exert any additional benefits on homocysteine levels above the dose of 1 mg/day. The aforementioned studies, although limited in number, suggest that the active forms of folinic acid may counter hyperhomocysteinemia to a better degree than conventional treatment with folic acid. This effect may lie in the ability of methylated lolates to be rapidly incorporated into the cellular remethylation of homocysteine in ESRD patients SPECIFC AIMS: 1. To examine the direct vascular effects of lolinic acid on endotheliurn- dependent and endothelium-independent vascular function in patients with ESRD. Brachial artery endothelid hction will be assessed using plethysmography in the presence and absence of acute inha-arterial administration of folinic acid. Responses to ACh and sodium nitropmsside (SNP)will be used as markers of endothelium-dependent and -independent vasorelaxation. 2. To determine the chronic effects of folinic acid on homocysteine levels and endothelial function. A 12 week treatment protocol will be employed and endothelial hction and biochemical parameters (plasma homocysteine, RBC folate, serum 8-12hs C-reactive protein, serine, glycine) will be compared in the treated v untreated groups. METHODOLOGY Hypothesis: 1. The methylated folate derivative, folinic acid, will directly or indirectly augment NO production in the endothelium. Therefore, acute intra-arterial administration of folinic acid (500ug/min) will improve endothelium- dependent vasorelaxation to ACh. Endothelium-independent vascular function to sodium nitroprusside will remain unaltered in the presence of acute folinic acid. 2. The presence of a methyl group on folinic acid will facilitate the process of cellular homocysteine re-methylation and decrease homocysteine levels. Therefore, long term (12 week treatment) with folinic acid will normalize i~omocysteinelevels in patients with ESRD. Through decreasing the burden of hyperhomocysteinernia, chronic folinic acid treatment will restore endothelial function. Patient selection: Patients with greater than 1year of hemodialysis-dependent chronic renal failure will be enrolled in the study following written informed consent (see consent form attached). The following exclusion criteria wil! be employed: Failure to obtain informed consent, plasma homocysteine levels less than 15 mmoI/L, patients on homocysteine lowering therapies (other than Diavite), patients with left ventridar dysfunction (EF<30%)and unstable coronary syndromes, acute or chronic infection, changes in endothehum-modifying medications such as ACEI, HMG CoA reductase inhibitors and hormone replacement therapy and patients requiring vascular surgery for hemodialyis access. Since the primary outcome is improvement in endothelial function (vs. contribution of individual risk factors towards endothelial dysfunction), we will include patients with diabetes, hypertension and dyslipidemia in our study. This will make the results of our investigations more applicable to the patient population undergoing hemodialysis. Biochemical markers: Baseline blood work will include the following: total cholesterol, HDLchoIesterol, higlycerides and calculated LDL-cholesterol; homocysteine, red blood cell and serum folate, serum BI~,hs C-reactive protein, serine and glycine. The full compliment of blood work will be repeated at the 12 week study. Plethvsmomaphv studies protocols: Acute folinic acid study: The first 20 patients will undergo an infusion protocol that will assess the acute effects of intra-arterial folinic acid on vascular function. The following infusion scheme will be emp1oyed:l. saline control 2 ACh, at 3,10 and 30 pg/min for 6 minutes each, 3. SNP, at 1,3 and 10 pg/min for 6 minutes each, 4. recontrol saline infusion, 5. folinic acid at 500 pg/min for 10 minutes each achieving forearm concentrations of, v) the folinic acid will then be co-infused at 500 lg/min and the above infusion scheme repeated. In between infusions, a break of at least 20 minutes allows flow to return to baseline. All solutions are infused at 1.0 rnL/min into the brachial artery of the study arm via an epidural catheter. Chronic folinic acid study : Twenty patients will be included in the chronic study and the first run of acetylcholine and sodium nitroprusside will be used for the vascular end-points. An additional 20 patients will undergo the following infusion scheme: 1. saline control (20 minutes), 2. Ach, at 3,10 and 30 pg/min for 6 ininutes each, 3. SNP,at 1,3 and 10 pg/min for 6 minutes each. AU 40 patients will return after 12 weeks of randomized treatment and have the above infusion protocol repeated. The duration of the shortened protocol is 90 minutes. A total of 83 plethysmography studies will be required to complete the acute and chronic portions of the study. CONCLUSION: The increased rates of atherosclerosis in patients with ESRD contribute significantly towards mortality in this patient population. The initiating event in atherosclerosis is endothelial dysfunction and hence assessing and improving endothelial function has important prognostic implications. A growing body of evidence suggests that the increased rates of atherosclerosis and endothelial dysfunction in ESRD may be secondary to persistent and treatment resistant hyperhomocysteinemia. Despite the accumulating evidence associating elevated homocysteine levels with vascular disease, conventional treatment strategies with folic acid do not normalize homocysteine levels nor do they improve endothelium dependent vasomotion. In the present grant proposes to examine the effects of a methylated folate derivative, folinic acid, on hyperhomocysteinemia and endothelial function in ESRD patients. Several factors serve to strengthen the proposal and purport feasibility. First, the presence of a methyl group in folinic acid may serve to facilitate the cellular re- methylation of homocysteine and therefore cellular 5-MTHF levels. Since the re- methylation pathway is defective in ESRD patients, supplying a methylated and active form of folic acid may yield beneficial effects compared to folic acid (which requires endogenous methylation prior to activation). Second, folinic acid may improve endothelial function through a direct interaction with or by elevating 5-MTHF and thereby potentiate nitric oxide release. FUTURE STUDIES-II FOLATE MODULATION ON PLATELET ACTTVITY Thrombus formation within the vasculature is the initiating event in acute myocardial infarction259. Platelet derived thromboxane is elevated in patients with myocardial infarction implicating them as the important mediator of thrombus formation. Platelet adhesion and activation is in part regulated by endogenously produced and endothelid derived nitric oxide. The impact of platelet-derived nitric oxide on the relevance of NO in the activation of platelets was recently investigated in eNOS deficient mice. Isolated platelets from the eNOS deficient mice caused sigruhcantly reduced bleeding times compared to control mice when transferred to thrombocytopenic/eNOS deficient animals260. The relationship between tetrahydrobiopterin and platelet function has not been investigated. However, the presence of dihydropteridine reductase (DR) within platelets strongly supports the notion that B& is required for proper platelet function261. The enzyme, DR, catalyzes the reduction of oxidized dihyclrobiopterins to tetrahydrobiopterin. Therefore, it is possible that disease states such as diabetes deplete platelet Bfi levels, resulting in superoxide generation, which could facilitate platelet leukocyte interaction and further intensify ischemia-reperfusion injury. Endothelial derived NO is also capable of inhibiting the aggregation of platelets in vivo. It has been demonstrated that patients with an increased response to Ach have decreased platelets that are more resistant to collagen induced activation. We have recently demonstrated that 5-MTHF is able to restore endothelial function in an insulin resistant rat model in a manner identical to supplementation with BH4. No studies to date have investigated the effectof 5-MTHF on platelet aggregation. Nevertheless, studies have demonstrated that a folate deficiency not only induces an elevation in homocysteine but also enhances oralcontraceptive induced platelet aggregation and oxidative stress262. This finding coupled with evidence that hyperhomocysteinemia (and likely folate deficiency) induces a greater ischemic- reperfusion injury suggests that 5-MTHF influences platelet function. Therefore it would be of value to test the interaction between 5-MTHF and platelet aggregation with the intension of expanding toward acutely testing the effects of folinic or folic acid supplementation in patients immediately post MI. It is possible that the eNOS -786 T/T while being detrimental during the development and progression of atherosclerosis as well as during the initiating events leading to a myocardial infarction, nevertheless it may be protective during ischemia and during reperfusion. The previously mentioned protection by exogenous supplementation with Bfi indicates that during ischemia eNOS is capable of contributing to the oxidative stress that occurs with reperfusion. Decreased expression of eNOS should correlate with a inhibition of superoxide production. Therefore, acute treatment with 5-MTHF or BH4 could be more relevant in the majority of patients with the wild type and heterozygous eNOS - 786 genotypes. Study Acknowledgments: I would Like to thank Dr. Snyder for measuring the MMAs, Dr. Venna for his collaboration on the endothelial function data, Dr. Triggle's lab for help with mesenteric arteries, Robin Rosenfeld at Scripps for her collaboration and help with Autodock, Dr. Anderson for providing the FATE patients and dl biochemical data, and Dr. Warnica for the recurrent cardiovascular event data. Reference List 1 Cahill M, Karabatzaki M, Meleady R, Refsum H, Ueland P, Shields D et al. Raised plasma homocysteine as a risk factor for retinal vascular occlusive disease. Br. J.Ophthalmol.2000;84:154-57. 2 Steinberg D. A critical look at the evidence for the oxidation of LDL in atherogenesis. Atherosclerosis 1997;131 Suppl:S5-S7. 3 Khan MA, Collins AJ, Keane WF. Diabetes in the elderly population. Adv.Ren Replace.Ther. 2000;7:32-51. 4 Komowski R, Lansky AJ. Current perspectives on interventional treatment strategies in diabetic patients with coronary artery disease. Catheter,Cardiovasc.hte~.2000;50:245-54. 5 Mazzone T. Current concepts and controversies in the pathogenesis, prevention, and treatment of the macrovascular complications of diabetes. J.Lab Clin.Med. 2000;135:437-43. 6 CeIermajer DS, Sorensen KE, Georgakopoulos D, Bull C,Thomas 0, Robinson J et al. Cigarette smoking is associated with dose-related and potentially reversible impairment of endothehum-dependent dilation in healthy young adults. Circulation 1993;88:2149-55. 7 Libby P, Aikawa M, Kinlay S, Selwyn A, Ganz P. Lipid lowering improves endothelial functions. IntJ.Cardiol.2000;74 Suppll:S3-S10. 8 Safar ME, Blacher J, Mourad JJ, London GM. Stiffness of carotid artery wall material and blood pressure in humans: application to antihypertensive therapy and stroke prevention. Stroke 2000;31:782-90. 9 Huysmans K, Lins RL, Daelemans R, Zachee P, De Broe ME. Hypertension and accelerated atherosclerosis in endstage renal disease. J.Nephrol. 1998;ll:I 85-95. 10 John S, khmieder RE. Impaired endothelial function in arterial hypertension and hypercholesterolemia: potential mechanisms and differences. J.Hypertens. 2000;18:363-74. 11 Anderson JL, Muhlestein JE3, Home BD, Carlquist JF,Bair 12, Madsen TE et nl. Plasma homocysteine predicts mortality independently of traditional risk factors and C-reactive protein in patients with angiographically defined coronary artery disease. Circulation 2000;102:1227-32. 12 Chia MC. The role of adhesion molecules in atherosclerosis. Crit Rev.Clin.Lab Sci. 1998;35:573-602. 13 Arroyo LH, Lee RT. The unstable atheromatous plaque. Can.J.Cardio1. 1998;14 Suppl B:11B-3B. 14 Berkenboom G. Unstable atherosclerotic plaque. Pathophysiology and therapeutic guidelines. Acta Cardiol. 199853:235-41. 15 Harrison DG, Ohara Y. Physiologic consequences of increased vascular oxidant stresses in hypercholesterolenua and atherosclerosis: implications for impaired vasomotion [see comments]. Am.J.Cardiol.1995;75:758-81B. 16 Heitzer T, Ma-Herttuala S, Luoma J, Kurz S, Munzel T, Just H et al. Cigarette smoking potentiates endothelial dysfunction of forearm resistance vessels in patients with hypercholesterolemia. Role of oxidized LDL. Circulation 1996;93:1346-53. 17 Ignarro LJ. Physiology and pathophysiology of nitric oxide. Kidney Int.Suppl19%;55:S2-S5. 18 van Guldener C, Stehouwer CD.Hyperhomocysteinemia, vascular pathology, and endothelial dysfunction [In Process Citation]. Semin.Thromb.Hemost. 2000;26:281-89. 19 Stroes ES, van Faassen EE, Yo M, Martasek P, Boer P, Govers R et al. Folic acid reverts dysfunction of endothelial nitric oxide synthase. Circ.Res. 2000;86:1129-34. 20 Verhaar MC, Wever RM, Kastelein JJ,van Dam T, Koomans HA, Rabelink TJ.5-methyltetrahydrofolate, the active form of foiic acid, restores endothelial function in familid hypercholesterolemia. Circulation 21 Blom HJ. Consequences of homocysteine export and oxidation in the vascular system [In Process Citation]. Semin.Thromb.Hemost. 2000;26:227- 32. 22 Refsum H, Guttormsen AB, Fiskerstrand T, Ueland PM. Hyperhomocysteinemia in terms of steady-state kinetics. Eur.J.Pediatr. 1998;157 Suppl2:S45-S49. 23 Finkelstein JD. Pathways and regulation of homocysteine metabolism in mammals [in Process Citation]. Semin.Thromb.Hemost. 2000;26:219-25. 24 Ghosh S, Wolan D, Adak S, Gane BR, Kwon NS, Tainer JA et al. Mutational analysis of the tetrahydrobiopterin-binding site in induable nitric-oxide synthase. J .Biol.Chem. 1999;274:24100-12. 25 Guenther BD, Sheppard CA, Tran P, Rozen R, Matthews RG, Ludwig ML. The structure and properties of methylenetetrahydrofolate reductase from Escherichia coli suggest how folate ameliorates human hyperhomocysteinemia [see comments]. NatStruct.Biol. 1999;6:359-65. 26 Rao PV, Garrow TA, John F, Garland D, Millian NS, Zigler JS, Jr. Betaine- homocysteine methyltransferase is a developmentally regulated enzyme crystallin in rhesus monkey lens. J.Biol.Chem. 1998;273:30669-74. 27 Kery V, Bukovska G, Kraus JP. Transsulfuration depends on heme in addition to pyridoxal5'-phosphate. Cystathionhe beta-synthase is a heme protein. J.Biol.Chem. 1994;269:25283-88. 28 Crane BR, Arvai AS, Ghosh DK, Wu C, Getzoff ED, Stuehr DJ et al. Structure of nitric oxide synthase oxygenase dimer with pterin and substrate. Science 1998;279:2121-26. 29 Feron 0. Endothelial nitric oxide synthase expression and its functionality. Curr.0pin.Clin.Nutr.MetabCare 1999;2:291-96. 30 Andrew PJ, Mayer B. Enzymatic function of nitric oxide synthases. Cardiovasc.Res. 1999;43:521-31. 31 Michel JB, Feron 0, Sacks D, Michel T. Reciprocal regulation of endothelial nitxic-oxide synthase by Cd+-calmodulin and caveolin. J.Biol.Chem. 1997;272:15583-86. 32 Bostom AG, Shemin D, Lapane n,Miller JW,Sutherland P, Nadeau M et al. Hyperhomocysteinemia and traditional cardiovascular disease risk factors in end-stage renal disease patients on dialysis: a case-control study. Atherosclerosis 1995;114:93-103. 33 Shafai T, Sweetman L, Weyier W, Goodman SI, Fennessey PV, Nyhan WL. Propionic acidemia with severe hyperammonemia and defective glycine metabolism. J-Pediatr.1978;92:84-86. 34 Lavoie A, Tripp E, Hoffbrand AV. The effect of vitamin 812 deficiency on methylfolate metabolism and pteroylpolyglutamate synthesis in human cells. Clin.Sci.Mol.Med. 1974;47:617-30. 35 McGing P, Reed B, Weir DG,Scott JM. The effect of vitamin 812 inhibition in vivo: impaired folate polyglutamate biosynthesis indicating that 5- methyltehahydropteroylglutamate is not its usual substrate. Biochern.Biophys.Res.Commun. 1978;82:540-46. 36 Cichowicz DJ, Hynes JB, Shane B. Substrate specificity of mammalian folylpoly-gamma-glutamate synthetase for 5,B-dideazafolates and 5,s- dideaza analogues of aminopte~.BiochiaBiophys.Acta 1988;957:363-69. 37 McGing PG,Scott JM. Evidence that the decreased liver folate status following vitamin El2 inactivation in the mouse is due to inaeased loss rather than impaired uptake. Biodrim.Biophys.Acta 1981;673:594-97. 38 Lumb M, Sharer N, Deacon R, Jennings P, Purkiss P, Perry J et al. Effects of nitrous oxide-induced inactivation of cobalamin on methionhe and S- adenosylmethionine metabolism in the rat. Bi&.Biophys.Acta 1983;756:354-59. 39 Dierkes J,Domrose U, Ambrosch A, Schneede J, Guttormsen AB, Neumann KH et nl. Supplementation with vitamin 812 decreases homocysteine and methylmalonic acid but also serum folate in patients with end-stage renal disease. Metabolism 1999;48:631-35. 40 Verleysdonk S, Martin H, Willker W, Leibfritz D, Hamprecht 8. Rapid uptake and degradation of glycine by astroglial cells in culture: synthesis and release of serine and lactate. Glia 1999;27:239-48. 41 Kastanos EK, Woldman W, Appling DR. Role of mitochondria1 and cytoplasmic serine hydroxymethyltansferase isozyrnes in de novo purine synthesis in Saccharomyces cerevisiae. Biochemistry 1997;36:14956-64. 42 Appling DR. Compartmentation of folate-mediated one-carbon metabolism in eukaryotes. FASEB J. 1991;5:264551. 43 Barlowe CK, Appling DR. In vitro evidence for the involvement of mi tochondrid folate metabolism in the supply of cytoplasmic one-carbon units. Biofactors 1988;1:171-76. 44 Cybulski RL, Fisher RR. Mitochondria1 neutral amino acid transport: evidence for a carrier mediated mechanism. Biochemistry 1977;16:5116-20. 45 Kolvraa S. Inhibition of the glyci.cleavage system by branched-chain amino acid metabolites. Pediatr.Res. 1979;13:889-93. 46 Wilcken DE, Dudman NP, Tyrrell PA, Robertson MR. Folic acid lowers elevated plasma homocysteine in chronic renal insufficiency: possible implications for prevention of vascular disease. Metabolism 1988;37:697-701. 47 Bostom AG, Jacques PF, Nadeau IMR, Williams RR, Ellison RC, Selhub J. Post-methionine load hyperhomocysteinemia in persons with normal fasting total plasma homocysteine: initial results from the NHLBI Family Heart Study. Atherosclerosis 1995;116:147-51. 48 Bostom AG, Roubenoff R, Dellaripa P, Nadeau MR,Sutherland P, Wilson PW et al. Validation of abbreviated oral methionine-loading test [letter]. Clin.Chem. 1995;41:94849. 49 Robinson K, Arheart K, Refsum H, Brattstrom L, Boers G,Ueland P et al. Low circulating folate and vitamin 06 concentrations: risk factors for stroke, peripheral vascular disease, and coronary artery disease. European COMAC Group [see comments] [published erratum appears in Circulation 1999 Feb 23;99(7):983]. Cirdation 1998;97:437-43. 50 Naurath HJ,Joosten Ej Riezler Rj Stabler SP, Allen RH, Lindenbaurn J. Effects of vitamin 812, folate, and vitamin 86 supplements in elderly people with nonnal serum vitamin concentrations [see comments]. Lancet 1995;346:85-89. 51 Pennypacker LC, Allen RH, Kelly JP, Matthews LM, Grigsby JJ Kaye K et a/. High prevalence of cobdamin deficiency in elderly outpatients [see comments]. J.Am.Geriatr.Soc. 1992;40:1197-204. 52 Legnani C, Palareti G, Grauso F, Sassi S, Grossi G,Piazzi S et al. Hyperhomocyst(e)inernia and a common methylenetetrahydrofolate reductase mutation (AIa223VaI MTHFR) in patients with inherited thrombophilic coagulation defects. Arterioscler.Thromb.Vasc.Biol. 1997;17:2924-29. 53 Hyndman ME, Bridge PJ, Warnica JW,Fick G,Parsons HG. Effect of heterongosity for the rnethionine synthase 2756 A->G mutation on the risk for recurrent cardiovascular events [In Process Citation]. Am.J.Cardio1. 2000;86:1144-6,A9. 54 Chango A, Emery-Fillon N, de Courcy GP, Lambert D, Pfister M, Rosenblatt DS ct nZ. A polymorphism (80G->A) in the reduced folate camer gene and its associations with folate status and hornocysteinemia [In Process Citation]. Mol.Gene t.Me tab 2000;70:310-15. 55 Frosst P, Blom HJ, Milos R, Goyette P, Sheppard CA, Matthews RG et al. A candidate genetic risk factor for vascular disease: a common mutation in methylenetetrahydrofolate reductase [letter]. Nat.Genet. 1995;10:111-13. 56 Jacques PF, Bostom AG, Williams RR, Ellison RC, Eckfeldt JH, Rosenberg IH st nl. Relation between folate status, a common mutation in methylenetetrahydrofolatereductase, and plasma homocysteine concentrations [see comments]. Circulation 1996;93:7-9. 57 Harmon DL, Woodside JV, Yarnell JW, McMaster D, Young IS, McCmm EE rt nl. The common 'thermolabile' variant of rnethylene tetrahydrofolate reductase is a major determinant of mild hyperhomocysteinaemia [see comments]. QJM. 1996;89:571-77. 58 Harmon DL, Shieids DC, Woodside JV,McMaster D, Yarnell JW, Young IS rt nl. Methionine synthase D919G polymorphism is a sigruhcant but modest determinant of circulating homocysteine concentrations. Genet.Epidemio1. 1999;17:298-309. 59 Rozen R. Genetic modulation of homocysteinernia [In Process Citation]. Semin.Thromb.Hemost. 2000;26:255-61. 60 Boers GH.Hyperhomocysteinemia as a risk factor for arterial and venous disease. A review of evidence and relevance. Thrornb.Haemost. 1997;78:520- 22. 61 Donnelly JG,Rock GA. Genetic determinants of heritable venous thrombosis: geno;yping methods for factor V(Leiden)A1691G, methylenetetrahydrofolate reductase C677Tfprothrombin G2021OA mutation, and algorithms for venous thrombosis investigations. Clin.Biochem. 1999;32:223-28. 62 Eichger Sf Stumpflen A, Hirxhi M, Bialonczyk C, Herkner K, Stain M et nl. Hyperhomocysteinemia is a risk factor of recurrent venous thromboembolism. Thromb.Haemost. 1998;80:56669. 63 Eikelboom JW,Baker RI. Venous thrombosis and hyperhomocysteinaemia. Med.J.Aust. 1998;169:313-15. 64 Falcon CR, Cattaneo M, Panzeri D, Martinelli I, Mannucci PM. High prevalence of hyperhomocyst(e)inemiain patients with juvenile venous thrombosis. Arterioscler.Thromb. 1994;14:108&83. 65 Langman LJ,Ray JG,Evrovski J,Yeo E, Cole DE. Hyperhomocyst(e)inemia and the inaeased risk of venous thromboembolism: more evidence from a case-control study. Arch.Intem.Med. 2000;160:961-64. 66 Morita H,Taguchi J, Kurihara H, Kitaoka M, Kaneda H, Kurihara Y et al. [Gene Polymorphism of 5,lO-rnethylenetetrahydrofolate reductase as a coronary risk factor]. J.Cardiol.1997;29:309-15. 67 Kluijtmans LA, Kastelein JJ,hdemans Jl Boers GHl Heil SG, Bruschke AV et al. Thennolabile rnethylenetetrahydrofolate reductase in coronary artery disease. Circulation 1997;96:2573-77. 68 Gallagher PM, Meleady R, Shields DC, Tan KS, McMaster D, Rozen R et al. Homocysteine and risk of premature coronary heart disease. Evidence for a common gene mutation. Circulation 1996;94:2154-58. 69 Christensen B, Arbour L, Tran P, Leclerc D, Sabbaghian N, PIatt R et a!. Genetic poly morphisms in methylenetetrahydrofolate reductase and methionine synthase, folate levels in red blood cells, and risk of neural tube defects. Am.J.Med.Genet. 1999;84:151-57. 70 Ou CY, Stevenson RE, Brown VK, Schwartz CE, Allen WP, Khoury MJ et al. 510 Methy ienetetrahydrofolate reductase genetic polymorphism as a risk factor for neural tube defects. Am.J.Med.Genet. 1996;63:610-14. 71 Whitehead AS, Gallagher P, Mills JL, Kirke PN, Burke H, Molloy AM et al. A genetic defect in 5,10 rnethylenetetrahydrofolate reductase in neural tube defects. QjM. 1995;88:763-66. 72 Wilson A, Platt R, WuQ, Leclerc D, Christensen B, Ymg H et al. A common variant in methionine synthase reductase combined with low cobalamin (vitamin B12) increases risk for spina bifida. Mol.Genet.Metab 1999;67:317- 23. 73 Ma J, Stampfer MJ, Giovannucci E, Artigas C, Hunter DJ, Fuchs C et al. Methylenetetrahydrofolatereductase polymorphism, dietary interactions, and risk c f colorectal cancer. Cancer Res. 1997;57:1098-102. 74 Slattery ML, Potter JD, Samowitz W, Schaffer D, Leppert M. Methylenetetrahydrofolate reductase, diet, and risk of colon cancer. Cancer Epidemio1.Biomarkers Prev. 1999;8:513-18. 75 Skibola CF, Smith MT, Kane E, Roman E, Rollinson Sf Cartwright RA et al. Polymorphisms in the rnethylenetetrahydrofolate reductase gene are associated with susceptibility to acute leukemia in adults [see comments]. Proc.Natl.Acad.Sci.U.S.A1999;96:12810-15. 76 Weisberg I, Tran P, Christensen 8, Sibani S, Rozen R. A second genetic polymorphism in methylenetetrahydrofolate reductase (MTHFR) associated with decreased enzyme activity [see comments]. k1ol.Genet.Metab 1998;64:169-72. 77 van der Put NM, Gabreels F, Stevens EM, Smeitink JA, Trijbels FJ, Eskes TK et nl. A second common mutation in the rnethylenetetrahydrofolate reductase gene: an additional risk factor for neural-tube defects? [see comments]. hm.J.Hm.Genet. 1998;62:1044-51. 78 Lederc D, Campeau E, Goyette P, Adjalla CE, Christensen B, Ross M et al. Human methionine synthase: cDNA cloning and identification of mutations in patients of the cblG complementation group of folate/cobalamin disorders. Hum.Mol.Gene t. 1996;5:1867-74. 79 Li YN, Gulati Sf Baker PJ, Brody LC, Banerjee R, Kruger WD. Cloning, mapping and RNA analysis of the human methionine synthase gene. Hum.Mol.Genet. 1996;5:1851-58. 80 Wang XL, Cai HICranney G, Wilcken DE. The frequency of a common mutation of the methionine synthase gene in the Australian population and its relation to smoking and coronary artery disease. J.Cardiovasc.Risk 1998;5:289-95. 81 Brown CAI McKinney KQ, Kaufman JS, Gravel RA, Rozen R. A common polymorphism in methionine synthase reductase increases risk of premature coronary artery disease [In Process Citation]. J.Cardiovasc.Risk 2000;7:197-200. 82 Tsai MY, Bignell M, Yang F, Welge BG, Graham KJ, Hanson NQ. Polygenic influence on plasma homocysteine: association of two prevalent mutations, the 843ins68 of cy sta thionine be ta-sy nthase and A (2756)G of me thionine svnthase, with lowered plasma homocysteine levels. Atherosderosis 2000;149:131-37. 83 Landewe RB, van den Borne BE, Breedveld FC, Dijkmans BA. Methotrexate effects in patients with rheumatoid arthritis with cardiovascular comorbidity [letter]. Lancet 2000;355:1616-17. 84 McCully KS. Vascular pathology of homocysteinemia: implications for the pathogenesis of arteriosclerosis. Am.J.Pathol.1969;56:111-28. 85 Wilcken DE, Wilcken B. The pathogenesis of coronary artery disease. A possible role for methionine metabolism. J.Clin.hvest 1976;57:1079-82. 86 Boushey CJ, Beresford SA, Omem GS, Motulsky AG. A quantitative assessment of plasma homocysteine as a risk factor for vascular disease. Probable benefits of increasing folic acid intakes [see comments]. JAMA 1995;274:1049-57. 87 Cattaneo M. Hyperhomocysteinemia, atherosclerosis and thrombosis. Thromb-Haemost.1999;81:165-76. 88 Eikelboom JW, Lorn E, Genest J, Jr., Hankey G, Yusuf S. Homocyst(e)ine and cardiovascular disease: a critical review of the epidemiologic evidence [see comments]. Ann.htem.Med. 1999;131:363-75. 89 Nygard 0, Nordrehaug JE,Refsum H, Ueland PM, Farstad M, Vollset SE. Plasma homocysteine levels and mortality in patients with coronary artery disease [see comments]. N.Engl. J.Med. 1997;337:230-36. 90 Al Obaidi MK,Stubbs PJ, Collinson P, Conroy R, Graham I, Noble MI. Elevated homocysteine levels are associated with increased ischemic myocardial injury in acute coronary syndromes [In Process Citation]. J.Am.Col~.Cardiol.2000;36:1217-22. 91 Rimm EB, Willett WC, Hu FBI Sampson L, Colditz GA, Manson JE et al. Folate and vitamin 86 from diet and supplements in relation to risk of coronary heart disease among women [see comments]. JAMA 1998;279:359- 64. 92 Coppola A, Davi GI De St V, Mancini FP, Cerbone AM, Di Minno G. Homocysteine, coagulation, platelet function, and thrombosis [In Process Citation]. Semin.Thromb.Hemost. 2000;26:243-54. 93 Whitehead AS, Ward PI Tan S. The molecular genetics of homocystinuria hyperhomocysteinemia and premature vascular disease in Ireland. Methioine metabolism: Molecular mechanisms and clinical implications 1994;80-83. 94 Eberhardt RT, Forgione MA, Cap A, Leopold JA, Rudd MA, Trolliet M et RI. Endothelial dysfunction in a murine model of mild hyperhomocyst(e)inernia.J.Clin.Invest 2000;106:483-91. 95 kineC, Wilson YG, Cume IC, McGrath C, Scott J, Day A et al. Hyperhomocysteinaernia is a risk factor for vein graft stenosis. Eur.J.Vasc.Endovasc.Surg. 1996;12:304-09. 96 Jeremy JY, Rowe D, Emsley AM, Newby AC. Nitric oxide and the proliferation of vascular smooth muscle cells. Cardiovasc.Res. 1999;43:580- 94. 97 McQuillan BM, Beilby JP, Nidorf M, Thompson PL, Hung J. Hyperhomocysteinemia but not the C677I' mutation of methylenetetrahydrofolate reductase is an independent risk determinant of carotid wall thickening. The Perth Carotid Ultrasound Disease Assessment Study (CUDAS) [see comments]. Circulation 1999;99:2383-88. 98 Lupattelli G, Rufini S, Locati EH, Lombardini R, Ciuffetti G, Siepi D rt nl. Hyperhomocyst(e)inemia is associated with car0tid atherosclerosis. Angiology 1999;50:823-30. 99 Southern FN, Cruz N, Fink LM, Cooney CA, Barone GW, Eidt JF et nl. Hyperhomocysteinernia increases intimal hyperplasia in a rat carotid endarterectomy model. J.Vasc.Surg. 1998;28:909-18. 100 Cmody BJ, Arora S, Avena R, Cosby K, Sidawy AN. Folic acid inhibits homocysteine-induced proliferation of human arterial smooth musde cells. J.Vasc.Surg. 1999;30:1121-28. 101 Xu DINeville R, Finkel T. Homocysteine accelerates endothelial cell senescence. FEBS Lett. 2000;470:20-24. 102 Dudman NP, Temple SE, Guo XW, Fu W, Perry MA. Homocysteine enhances neutrophilsndothelial interactions in both cultured human cells and rats In vivo. Circ.Res. 1999;84:409-16. 103 Pruefer DIScalia R, Lefer AM. Homocysteine provokes leukocyte- endothelium interaction by downregulation of nitric oxide. Gen.Phmaco1. 1999;33:487-98. 104 Rodgers GM, Kane WH. Activation of endogenous factor V by a homocysteine-induced vascular endothelial cell activator. J.Clin.Invest 1986;77:1909-16. 105 Starkebaum G, Harlan JM. Endothelial cell injury due to copper-catalyzed hydrogen peroxide generation from hornocysteine. J.Clin.Invest 1986;77:1370-76. 106 Austin RC, Sood SKI Donvard AM, Singh GI Shaughnessy SG, Pamidi S et ~1.Homocy steine-dependent alterations in mitochondria1 gene expression, hcticn and structure. Homocysteine and H202 act synergistically to enhance mitochondria1 damage. J.Biol.Chem. 1998;273:3080&17. 107 Lentz SR, Sadler JE.Inhibition of thrombomodulin surface expression and protein C activation by the thrombogenic agent homorysteine. J.Clin.Invest 1991;88:1906-14. 108 Hajjar KA. Homocysteine-induced modulation of tissue plasminogen activator binding to its endothelial cell membrane receptor. J.Clin.lnvest 1993;91:2873-79. 109 Nishinaga M, Ozawa T, Shimada K. Homocysteine, a thrombogenic agent, suppresses anticoagulant heparan sulfate expression in cultured porcine aortic endothelial cells. J.Clin.hvest 1993;92:1381-86. 110 Nishio E, Watanabe Y. Homocysteine as a modulator of platelet-derived growth factor action in vascular smooth muscle cells: a possible role for hydrogen peroxide. Br.J.Phmacol.1997;122:269-74. 111 Mann NJ, Li D, Sinclair AJ, Dudman NP, Guo XW, Elsworth GR et a!. The effect of diet on plasma homocysteine concentrations in healthy male subjects. Eur.J.Clin.Nutr. 1999;53:895-99. 112 Bunout D, Petermann M, Hirsch S, de la MP, Suazo M, Barrera G et al. Low serum folate but nodhomocysteine levels in patients with atherosclerotic vascular disease and matched healthy controls. Nutrition 113 Heller R, Munscher-Paulig F, Grabner R, Till U. L-Ascorbic acid potentiates nitric oxide synthesis in endothelial cells. J.Biol.Chern. 1999;274:8254-60. 114 Fischmann TO, Hruza A, Niu XD, Fossetta JD, Lunn CA, Dolphin E et nl. Structural characterization of nitric oxide synthase isoforms reveals striking active-site conservation. Nat.Struct.Biol.1999;6:233-42. 115 Raman CS, Li H, Martasek P, Kral V, Masters BS, Poulos TL. Crystal structure of constitutive endothelial nitric oxide synthase: a paradigm for pterin function involving a novel metal center. Cell 1998;95:939-50. 116 Gorren AC, Bec N, Schramrnel A, Werner ER, Lange R, Mayer B. Low- temperature optical absorption spectra suggest a redox role for tetrahydrobiopterin in both steps of nitric oxide synthase catalysis [In Process Citation]. Biochemistry 2000;39:11763-70. 117 Shinozaki K, Kashiwagi A, Nishio Y, Okamura T, Yoshida Y, Masada M et nl. Abnormal biopterin metabolism is a major cause of impaired endothelium- dependent relaxation through nitric oxide/ 02- imbalance in insulin- resistant rat aorta. Diabetes 1999;48:2437-45. 118 Shinozaki K, Nishio Y, Okamura T, Yoshida Y, Maegawa H, Kojima H et al. Oral administration of tetrahydrobiopterin prevents endothelial dysfunction and vascular oxidative stress in the aortas of insulin- resistant rats [In Process Citation]. Circ.Res. 2000;87566-73. 119 Vasquez-Vivar J, Kalyanaraman B, Martasek P, Hogg N, Masters BS, Karoui H rt nl. Superoxide generation by endothelial nitric oxide synthase: the influence of cofactors. Proc.Nat1.Acad.Sci.U.S.A1998;95:9220-25. 120 Xia Y, Tsai AL, Berka V, Zweier JL. Superoxide generation from endothelial nitric-oxide synthase. A Ca2+/calmodulin-dependent and tetrahydrobiopterin regulatory process. J.Biol.Chem. 1998;273:2580408. 121 Schmidt K, Werner ER, Mayer 8, Wachter H, Kukovetz WR. Tetrahydrobiopterin-dependent formation of endothelium-derived relaxing factor (nitric oxide) in aortic endothelial cells. Bi0chem.J. 1992;281 ( Pt 2) 297-300. 122 Cosentino F, Katusk ZS. Tetrahydrobiopte~and dysfunction of endothelial nitric oxide synthase in coronary arteries [see comments]. Circulation 1995;91:1394. 123 Riethmuller C, Gorren AC, Pitters E, Hemmens B, Habisch HJ, Heales SJ rt al. Activation of neuronal nitric-oxide synthase by the imethyl analog of tetrahydrobiopterin. Functional evidence against reductive oxygen activation by the pterin co factor. J.Biol.Chem. 1999;274:16047-51. 124 Togari A, Arai M, Mogi M, Kondo A, Nagatsu T. Coexpression of GTP cyclohydrolase I and inducible nitric oxide synthase mRNAs in mouse osteoblastic cells activated by proinflammatory cytokines. FEBS Lett. 1998;428:212-16. 125 Maier W, Cosentino F, Lutolf RB, Fleisch M, Seiler C, Hess OM et al. Tetrahydrobiopterih improves endothelial function in patients with coronary artery disease. J.Cardiovasc.Pharmacol.2000;35:173-78. 126 Kojda G, Harrison D. Interactions between NO and reactive oxygen species: pathophysiological importance in atherosclerosis, hypertension, diabetes and heart failure. Cardiovasc.Res. 1999;43:562-7l. 127 Ohara Y, Peterson TE, Harrison DG. Hypercholesterolemia increases endothelial superoxide anion production. J.Clin.lnvest 1993;91:2546-31. 128 Ohara Y, Peterson TE, Sayegh HS, Subramanian RR, Wilcox JN, Hamson DG. Dietary correction of hypercholesterolemia in the rabbit normalizes endothelial superoxide anion production. Circulation 1995;92:898-903. 129 Stroes E, Kastelein 1, Cosentino F, Erkelens W, Wever R, Koomans H et ~1. Tetrahydrobiopterin restores endothelial function in hypercholesterolemia. J.Clin.Inves t 1997;99:4116. 130 Heitzer T, Brockhoff C, Mayer B, Warnholtz A, Mohau H, Heme S et al. Tetrahydrobiopterin improves endothelium-dependent vasodilation in chronic smokers : evidence for a dysfunctional nitric oxide synthase. Circ.Res. 2000;86:E36-E41. 131 Hillinger S, Schoedon G, Walter R, Zollinger A, Stammberger U, Speich R et RZ. Reduction of lung allograft ischernia/reperfusion injury by tetrahydrobiopterin coenzyme of NOS. Transplant.Proc. 1999;31:183-84. 132 Ishii MI Shimizu S, Nawata S, Kiuchi Y, Yamamoto T. Involvement of reactive oxygen species and nitric oxide in gastric ischemia-reperfusion injury in rats: protective effect of tetrahydrobiopterin. Dig.Dis.Sci. 2000;45:93-98. 133 Shimizu S, Ishii M, bIomose K, Yamamoto T. Role of tetrahydrobiopterin in the function of nitric oxide synthase, and its cytoprotective effect (Review). hi.J.h.Iol.Med. 1998;2:53310. 134 Tiefenbacher CP, Chilian \VM,Mitchell M, DeFily DV. Restoration of endothelium-dependent vasodilation after reperfusion injury by tetrahydrobiopterin. Circulation 1996;94:1423-29. 135 Schmid RA, Winger S, Walter R, Zollinger A, Stammberger U, Speich R et al. The nitric oxide synthase cofactor tetrahydrobiopterin reduces allograft ischemia-reperfusion injury after lung transplantation. J.Thorac.Cardiovasc.Surgg1999;118:726-32. 136 Keaney JF,Jr., Hare JM, Balligand JL, Loscalzo J, Smith TW,Colucci WS. Inhibition of nitric oxide synthase augments myocardial contrade responses to be ta-adrener gic stimulation. Am. J.Physiol1996;271:!+2646- H2652. 137 Kelly RA, Balligand JL, Smith TW. Nitric oxide and cardiac function. Circ-Res. 1996;79:363-80. 138 Shesely EG, Maeda N, Kim HS, Desai KM,Krege JH,Laubach VE et al. Elevated blood pressures in mice lacking endothelial nitric oxide synthase. Proc.Natl.Acad.Sci.U.S.A 1996;93:13176-81. 139 Cosentino F, Patton S, d1UscioLV, Werner ER, Werner-Felmayer G, Moreau P rt al. Tetrahydrobiopte~alters superoxide and nitric oxide release in prehypertensive rats. J.Clin.Invest 1998;101:1530-37. 140 Usui M, Matsuoka H, Miyazaki H, Ueda S, Okuda S, Imaizumi T. Endothelial dysfunction by amte hyperhomocyst(e)inaemia:restoration bv folic acid. Clin.Sci.(Colch.) 1999;96:235-39. 141 Heitzer T, Krohn K, Albers S, Meinertz T. Tetrahydrobiopterin improves endothelium-dependent vasodilation by inaeasing nitric oxide activity in patients with Tvpe I1 diabetes mellitus.[h Rocess Citation]. Diabetologia 2000;43:1435-38. 142 Title LM, Cumrnings PM, Giddens K, Genest JJ,Jr., Nassar BA. Effect of folic acid and antioxidant vitamins on endothelid dysfunction in patients with coronary artery disease. J.Am.Coll.Cardiol.2000;36:758-65. 143 Chao CL, Kuo TL, Lee YT. Effects of methionine-induced hyperhomocysteinemia on endothelium- dependent vasodilation and oxidative status in healthy adults. Circulation 2000;101:485-90. 144 Loehrer FM,Haefeli WE, Angst CP, Browne G, Frick G, Fowler B. Effect of methionine loading on 5-rnethyltetrahydrofolate, S adenosylmethionine and Sadenosy lhomocysteine in plasma of healthy humans. Clin.Sci.(Colch.) 1996;91:79-86. 145 Ibels LS, Stewart JH, Mahony JF,Neale FC, Sheil AG. Occlusive arterial disease in uraemic and haemodialysis patients and renal transplant recipients. A studv of the incidence of arterial disease and of the prevalence of risk factors implicated in the pathogenesis of arteriosclerosis. Q.J.Med. 1977;46:197-214. 146 Green D, Stone NJ, Krumlovsky FA. Putative atherogenic factors in patients with chronic renal failure. Prog.Cardiovasc.Dis. 1983;26:133-44. 147 Bostom AG, Lathrop L. Hyperhomocysteinernia in end-stage renal disease: prevalence, etiology, and potential relationship to arteriosclerotic outcomes. Kidney Int. 1997;52:10-20. 148 Wu G. Cardiovascular deaths among CAPD patients. Perit.Dial.Bull. 4, s23- 25.1983. Ref Type: Generic 149 Clarke R, Daly L, Robinson K, Naughten E, Cahalane Sf Fowler B et al. Hyperhomocysteinemia: an independent risk factor for vascular disease [see comments]. N.Engl.J.Med. 1991;324:1149-55. 150 Stampfer MJ, Malinow MR, Willett WC, Newcomer LM, Upson B, UUmann D et ~1.A prospective study of plasma homocyst(e)ine and risk of myocardial infarction in US physicians. JAMA 1992;268:877-81. 151 Friedman ]A, Dwyer JT. Hyperhomocysteinemia as a risk factor for cardiovascular disease in patients undergoing hemodialysis. Nutr-Rev. 1995;53:197-201. 152 Chauveau P, Chadefaux 8, Coude M, Aupetit J, Hamedouche T, Kamoun P rt nl. Hyperhomocysteinemia, a risk factor for atherosclerosis in chronic uremic patients. Kidney Int.Suppl1993;41:S72-ST. 153 Robinson K, Gupta A, Dennis V, Arheart K, Chaudhary D, Green R et a[. Hyperhomocysteinemia confers an independent inaeased risk of atherosclerosis in end-stage renal disease and is closely linked to plasma folate and pyridoxine concentrations. Circulation 1996;94274348. 154 Manns BJ, Burgess ED, Hyndman ME, Parsons HG, Schaefer JP, Scott- Douglas NW. Hyperhomocyst(e)inemia and the prevalence of atherosclerotic vascular disease in patients with end-stage renal disease. Am. J. Kidney Dis. 1999;34:669-77. 155 Bostorn AG, Shemin D, Verhoef P, Nadeau MR, Jacques PF, Selhub J et id. Elevated fasting total plasma homocysteine levels and cardiovascular disease outcomes in maintenance dialysis patients. A prospective study. Arterioscler.Thromb.Vasc.Biol. 1997;17:2554-58. 156 Moustapha A, Naso A, Nahlawi M, Gupta A, Arheart KL, Jacobsen DW rt al. Prospective study of hyperhomocysteinemia as an adverse cardiovascular risk factor in end-stage renal disease [published erratum appears in Circulation 1998 Feb 24;97(7):7ll].Circulation 1998;97:138-41. 157 van Guldener C, Janssen MJ, De Meer K, Donker AJ, Stehouwer CD. Effect of folic acid and betaine on fasting and postmethionine-loading plasma homocysteine and methionine levels in chronic haemodialysis patients. J.lntem.Med. 1999;245:175-83. 158 Mayer EL, Jacobsen DW, Robinson K. Homocysteine and coronary atheroscIerosis. J.Am.Coll.Cardiol.1996;27:517-27. 159 Selhub J, Miller JW.The pathogenesis of homocysteinemia: interruption of the coordinate regulation by Sadenosylmethionine of the remethylation and transsulfuration of homocysteine. Am.J.Clin.Nuh. 1992;55:131-38. 160 Kluijtmans LA, Boers GH, Stevens EM, Renier WO, Kraus JP,Trijbels FJ et nl. Defective cystathionine beta-synthase regulation by S adenosylmethionine in a partially pyridoxine responsive homocystinuria patient. J.Clin.Invest 1996;98:285-89. 161 Wilcken DE, Gupta VJ, Betts AK. Homocysteine in the plasma of renal transplant recipients: effects of cofactors for methionine metabolism. Clin.Sci. 1981;61:74339. 162 Guttomen AB, Ueland PM, Svarstad E, Refsum H. Kinetic basis of hyperhomocysteinemia in patient. with chronic renal failure. Kidney ht. 1997;52:495-502. 163 Janssen MJ, van Guldener C, de JongGM, van den BM, Stehouwer CD, Donker A]. Folic acid treatment of hyperhomocysteinemia in dialysis patients. Miner.Electrolyte Metab 1996;22:110-14. 164 Dudman NP,Tyrrell PA, Wilcken DE. Homocysteinemia: depressed plasma serine levels. Metabolism 1987;36:198-201. 165 Bostom AG, Shemin D, Lapane KL, Hume AL, Yoburn D, Nadeau MR rt al. High dose-B-vitamin treatment of hyperhomocysteinemia in dialysis patients. Kidney Int. 1996;49:147-52. 166 Smolin LA, Laidlaw SA, Kopple JD. Altered plasma free and protein-bound sulfur amino acid levels in patients undergoing maintenance hemodialysis. Am. J.Clin.Nutr. 1987;45:73743. 167 Arnadottir M, Brattstrom L, Simonsen 0,Thysell H, Hultberg B, Andersson A rt af. The effect of high-dose pyridoxine and folic acid supplementation on serum lipid and plasma homocysteine concentrations in dialysis patients. Clin.Nephrol.1993;40:236-40. 168 Antony AC, Kincade RS, Verma RS, Krishnan SR. Identification of high affinity folate binding proteins in human erythrocyte membranes. J.Clin.Invest 1987;80:7ll-23. 169 Touam M, Zingraff J, Jungers P, Chadefaux-Vekernans 8, Drueke T, Massy ZA. Effective correction of hyperhomocysteinemia in hemodialysis patients by intravenous folinic acid and pyridoxine therapy. Kidney ht. 1999;56:2292-96. 170 House AA, Do~elyJG. Effect of multivitamins on plasma homocysteine and folate levels in patients on hemodialysis. ASAIO J. 1999;45:94-97. 171 Dierkes J, Domrose U, Arnbrosch A, Bosselmann HP, k~rna~KH, Luley C. Response of hyperhomocysteinemia to folic acid supplementation in patients with end-stage rend disease. Clin.Nephrol.1999;51:108-15. 172 Spence JD, Cordy P, Kortas C, Freeman D. Effect of usual doses of folate supplementation on elevated plasma homocyst(e)inein hemodialysis patients: no difference between 1and 5 mg daily. Am.J.Nephro1. 1999;19:405-10. 173 Bernard MA, Nakonezny PA, Kashner TM.The effect of vitamin 812 deficiency on older veterans and its relationship to health [see comments]. J. Am.Geriatr.Soc. 1998;46:1199-206. 174 Thomas PK, Hoffbrand AV. Hereditary transcobalamin I1 deficiency: a 22 year follow up [letter]. J.Neurol.Neurosurg.Psy~ahy1997;62:197. 175 Marcuard SP, Albemaz L, Khazanie PG.Omeprazole therapy causes malabsorption of cyanocobalamin (vitamin 812) [see comments]. Ann.Lntem.Med. 1994;120:211-15. 176 Lindenbaum J, Rosenberg M, Wilson PW, Stabler SP, Allen RH. Prevalence of cobdamin deficiency in the Framingham elderly population [see comments]. Am. J.CIin.Nutr.1994;60:2-11. 177 Bostorn AG, Shemin D, Bagley P, Massy ZA, Zanabli A, Christopher K et al. Controlled comparison of L-5-methyltetrahydrofolate versus folic acid for the treatment of hyperhomocysteinemia in hernodialysis patients [published erratum appears in Circulation 2000 Aug 1;102(5):598]. CircuIation 2000;101:2829-32. 178 Shinozaki K, Nishio Y, Okamura T, Yoshida Y, Maegawa H, Kojima H et al. Oral administration of tetrahydrobiopterin prevents endothelial dysfunction and vascular oxidative stress in the aortas of insulin- resistant rats [In Process Citation]. Circ-Res. 2000;87:566-73. 179 Bostom AG, Shernin D, Lapane KL, Nadeau MR. Sutherland P, Chan J rt al. Folate status is the major determinant of fasting total plasma homocysteine levels in maintenance dialysis patients. Atherosclerosis 1996;123:193-202. 180 Bostom AG, Gohh RY, Beaulieu AJ, Nadeau MR, Hume AL, Jacques PF rt al. Treatment of hyperhomocysteinemia in renal transplant recipients. A randomized, placebo-controlled trial. Ann.htem.Med. 1997;127:1089-92. 181 van Guldener C, Kulik W, Berger R, Dijkstra DA, Jakobs C, Reijngoud DJ et nl. Homocysteine and methionine metabolism in ESRD: A stable isotope study. Kidney Lnt 1999;56:1064-71. 182 Fodinger M, Mannt-alter C, Wolfl G, Pabinger I, Muller E, Schmid R et al. Mutation (677 C to T) in the methylenetetrahydrofolate reductase gene aggravates hyperhomocysteinemia in hernodialysis patients. Kidney Int. 1997;52:517-23. 183 Shinozaki K, Kashiwagi A, Nishio Y, Okamura T, Yoshida Y, Masada M et al. Abnormal biopterin metabolism is a major cause of impaired endothelium- dependent relaxation through nitric oxide/02- imbalance in insulin- resistant rat aorta. Diabetes 1999;48:2437-45. 181 Vema S, Arikawa E, Yao L, Laher I, McNeill JH. Insulin-induced vasodilation is dependent on tetrahydrobiopterin synthesis. Metabolism 1998;47:1037-39. 183 Pieper GM. Acute amelioration of diabetic endothelid dysfunction with a derivative of the nitric oxide synthase cofactor, tetrahydrobiop terin. J.Cardiovasc.Pharmaco1.1997;29:8-15. 186 Rasmussen K. Studies on methylmalonic acid in humans. I. Concentrations in serum and urinary excretion in normal subjects after feeding and during fasting, and after loading with protein, fat, sugar, isoleucine, and valine. Clin.Chem. 1989;35:2271-76. 187 Rasmussen K, Vyberg B, Pedersen KO, Brochner-Mortensen J. MethVlmalonic acid in renal insufficiency: evidence of accumuiation and implications for diagnosis of cobdamin deficiency. Ch-Chem. 1990;36:1523-24. 188 Jacobs RL, House JD, Brosnan ME, Brosnan JT.Effects of streptozotocin- induced diabetes and of insulin treatment on homocysteine metabolism in the rat. Diabetes 1998;47:1967-70. 189 Wolf B, Paulsen EP. Valproate in the treatment of seizures associated with propionic acidernia [letter]. Pediatrics 1981;67:162-63. 190 Moelby L, Rasmussen K, Rasmussen HH. Serum methylmalonic acid in uraemia. Scand.J.Clin.LabInvest 1992;52:351-54. 191 Moelby L, Rasmussen K, Ring T, Nielsen G. Relationship between methylmalonic acid and cobalarnin in uremia. Kidney Int. 2000;57:265-73. 192 Fowlow SB, Holmes TM, Morgan K, Snyder FF. Screening for methylmalonic aciduria in Alberta: a voluntary program with particular sigruficance for the Hutterite Brethren. Am.J.Med.Genet. 1985;22:513-19. 193 Lee HA,Choi JS, Ha KS, Yang DH, Chang SK, Hong SY. Influence of 5,100 methylenetetrahydrofolate reductase gene polymorphism on plasma homocysteine concentration in patients with end-stage renal disease. Am. J.Kidney Dis. 1999;34:259-63. 194 Areekul S, Churdchu K, Cheerarnakara C, Wilairatana P, Charoenlarp P. Serum tranxobalamin II levels in patients with acute and chronic renal failure. J.Med.Assoc.Thai.1995;78:191-96. 195 Narkewicz MR, Sauls SD, Tjoa SS, Teng C, Femessey PV. Evidence for intracellular pattitioning of serine and glycine metabolism in Chinese hamster ovary cells. Bi0chem.J.1996;313 ( Pt 3):991-96. 196 Leclerc D, Wilson A, Dumas R, Gafuik C, Song D, Watkins D et al. Cloning and mapping of a cDNA for methionine synthase reductase, a flavoprotein defective in patients with homocystinuria. Roc.Nat1.Acad.Sci.U.S.A 1998;95:3059-64. 197 Wilson A, Leclerc D, Saberi F, Campeau E, Hwang HY, Shane B et ni. Functionally null mutations in patients with the cblG-variant fonn of me thionine s ynthase deficiency. Am. J .Hum.Genet. 1998;63:409-14. 198 Park JB,Sc1~fri.n EL. Effects of antihypertensive therapy on hypertensive vascular disease. Curr.Hypertens.Rep. 2000;2:280-88. 199 Carmel R, Green R, Jacobsen DW, Rasmussen K, Florea M, Azen C. Serum cobalamin, homocysteine, and methvlmalonic acid concentrations in a multiethnic elderly population: ethnic and sex differences in cobalamin and metabolite abnormalities [see comments]. Am. J.Clin.Nutr.1999;70:904-10. 200 Klee GG. Cobalamin and folate evaluation: measurement of methylmalonic acid and homocysteine vs vitamin B(12) and folate. Clin.Chem. 2000;46:1277-83. 201 Kannel WB.Elevated systolic blood pressure as a cardiovascular risk factor. Am. J.Cardiol.2000;85:251-55. 202 Bagley PJ, Selhub J. A common mutation in the methylenetetrahydrofolate reductase gene is associated with an accumulation of formylated tetrahydrofolates in red blood cells. Proc.Nat1.Acad.Sci.U.S.A 1998;95:13217- 20. 203 Gardemann A, Weidemann H, Philipp M, Katz N, Tillmanns H, Hehrlein FW ct a!. The TT genotype of the methylenetetrahydrofolate reductase C677T gene polymorphism is associated with the extent of coronary atherosclerosis in patients at high risk for coronary artery disease [see comments]. Eur.Heart J. 1999;20:584-92. 204 Schmitz C, Lindpaintner K, Verhoef P, Gaziano JM, Buring J. Genetic polymorphism of methylenetetrahydrofolate reductase and myocardial infarction. A case-control study. Circulation 1996;94:1812-14. 205 Matthews RG, Kaufman S. Characterization of the dihydropterin reductase activity of pig liver methylenetetrahydrofolate reductase. J.Biol.Chem. 1980;255:601417. 206 Kauhan S. Some metabolic relationships between biopterin and folate: implications for the "methyl trap hypothesis". NeurochemXes. 1991;16:1031-36. NOTE TO USERS Page(s) not included in the original manuscript and are unavailable from the author or university. The manuscript was microfilmed as received. 237 through 246 This reproduction is the best copy available. Appendix A: MTRR 66 MTHFR 677 Nitrite standard curve: 4 HPLC nitrite Standard Curve 400.00 ------ 350.00 - - - - y=0.1252x-2.0046,------300.00 ------ --A -----A- - - - - ------ ------A- 0 500 1000 1500 2000 2500 3000 3500 nM nitrite