Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Environmental modification of the expression of neural tube defects in SELH/Bc mice Hall, Jennifer Lynn 1996

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

Item Metadata

Download

Media
831-ubc_1996-0469.pdf [ 9.03MB ]
Metadata
JSON: 831-1.0087678.json
JSON-LD: 831-1.0087678-ld.json
RDF/XML (Pretty): 831-1.0087678-rdf.xml
RDF/JSON: 831-1.0087678-rdf.json
Turtle: 831-1.0087678-turtle.txt
N-Triples: 831-1.0087678-rdf-ntriples.txt
Original Record: 831-1.0087678-source.json
Full Text
831-1.0087678-fulltext.txt
Citation
831-1.0087678.ris

Full Text

ENVIRONMENTAL MODIFICATION OF THE EXPRESSION OF NEURAL TUBE DEFECTS IN SELH/Bc MICE by JENNIFER LYNN HALL B.Sc, The University of British Columbia, 1992 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES (Department of Medical Genetics) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA June 1996 © Jennifer Lynn Hall, 1996 In presenting this thesis in partial fulfillment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of Medical Genetics The University of British Columbia Vancouver, Canada DE-6(2/88) ABSTRACT The objective of this work was to explore the environmental modification of the expression of neural tube defects (NTDs) in SELH/Bc mice. One approach was to examine the potential of SELH/Bc mice as an animal model for the reduction in recurrence and occurrence rates of NTDs in women given periconceptional folic acid supplementation. Neither folic acid nor methionine supplementation produced a detectable reduction of the exencephaly frequency in SELH/Bc mice. However, the frequency of exencephaly was consistently higher in SELH/Bc dams fed a chemically-defined Harlan Teklad diet than in SELH/Bc dams fed the standard ration of Purina Lab Chow. This observation was directly tested. The exencephaly frequency was 7-fold higher on Harlan Teklad diet than on the Purina Lab Chow diet (21% versus 3%). This finding represents the first demonstration of the nutritional modification of the expression of NTDs in SELH/Bc mice, and it affords a unique opportunity to study the mechanism of prevention of exencephaly in the SELH/Bc mouse model, something that is not possible to examine directly in human studies of NTDs. Another approach tested for the presence of a genotype-teratogen interaction of SELH/Bc mice with valproic acid as an animal model of the effect of genetic liability to NTDs on the liability to anticonvulsant induced birth defects. The SELH/Bc response to valproic acid treatment was compared to the response of two inbred strains of mice that rarely have spontaneous exencephaly. When the data were transformed according to the developmental threshold model with an underlying normally distributed scale of liability, the response to valproic acid treatment was found to be additive with the genetic liability. The genetic liability to exencephaly in S E L H B c mice greatly increased the absolute risk after valproic acid treatment. At the peak response time, the frequency of exencephaly in SELH/Bc mice was 69% compared to 35% and 40% respectively in SWV/Bc and ICR/Be mice. This observation may have clinical significance for women with a positive family history of NTDs taking valproic acid during the first trimester of pregnancy. ii TABLE OF CONTENTS Abstract ii Table of Contents iii List of Tables v List of Figures x Acknowledgements xi INTRODUCTION I. Neural Tube Defects A. Incidence 2 B. Etiology 3 II. Models for the Inheritance of Neural Tube Defects A. Genetic Models 13 B. Environmental Models 17 III. The Development of Neural Tube Defects A. The Normal Process of Neurulation 29 B. Lessons from the Study of Neural Tube Defects 33 EXPERIMENTAL STUDIES I. General Methods 36 II. Study of the Effect of Methionine Supplementation on the Expression of NTDs in SELH/Bc mice A. Introduction 37 B. Materials and Methods 38 C. Results 39 D. Discussion 41 iii III. Studies of the Effect of Folic Acid Supplementation on the Expression of NTDs in SELH mice A. Introduction 41 B. Experiment #1: Supplemental folic acid added to drinking water 42 C. Experiment #2: Supplemental folic acid added to dry food 46 D. Discussion 51 IV. Study of the Effect of the Harlan Teklad chemically-defined diet (compared to Purina Lab Chow) on the Expression of NTDs in SELH/Bc mice A. Introduction 52 B. Materials and Methods 52 C. Results 54 D. Discussion 55 V. Study of the Effect of Valproic Acid on the Expression of NTDs in Three Strains of mice with Different Genetic Liabilities to Exencephaly A. Introduction 61 B. Materials and Methods 61 C. Results 65 D. Discussion 73 VI. Study of the Effect of Folic Acid Supplementation on the Expression of NTDs in SELH/Bc mice Treated with Valproic Acid A. Introduction 79 B. Materials and Methods 79 C. Results 80 D. Discussion 81 DISCUSSION 84 References 89 Appendix A: Data Analysis 99 Appendix B: Composition of the Purina Lab Chow and Harlan Teklad Diets 146 iv LIST OF TABLES T A B L E 1 Classification of NTD in genetic models (mouse unless stated otherwise) 14 T A B L E 2 Effects of methionine supplementation on the in utero mortality rate and frequency of exencephaly in SELH/Bc mice scored on day fourteen of gestation. 39 T A B L E 3 Effects of folic acid supplementation (0.20 mg/L in drinking water ad libitum) on the in utero mortality rate and frequency of exencephaly in SELHA/Bc mice scored on day fourteen of gestation. 45 T A B L E 4 Effects of folic acid supplementation (0 ppm, 2 ppm, or 10 ppm in Harlan Teklad Diet ad libitum) on the in utero mortality rate and frequency of exencephaly in SELH/Bc mice scored on day fourteen of gestation. 49 T A B L E 5 Effects of folic acid supplementation (0 ppm, 2 ppm, or 10 ppm in Harlan Teklad Diet ad libitum) on the level of folate in the maternal red blood cells. 50 T A B L E 6 Effects of the Harlan Teklad chemically-defined diet (compared to Purina Lab Chow) on the in utero mortality rate and frequency of exencephaly in SELH/Bc mice scored on day fourteen of gestation. 54 T A B L E 7 Dietary composition of Purina Lab Chow (PLC) and the Harlan Teklad chemically-defined diet (HT TD93013). 58 T A B L E 8 A comparison of the dietary sources of protein, fat, carbohydrate, and fiber in the Purina Lab Chow and the Harlan Teklad chemically-defined diet. 59 T A B L E 9 Vitamin and mineral composition of Purina Lab Chow (PLC) and the Harlan Teklad chemically-defined diet (HT TD93013). 60 T A B L E 10 Effects of valproic acid treatment on the in utero mortality rate and frequency of exencephaly in SELH/Bc, SWV/Bc, and ICR/Be mice scored on day fourteen of gestation. 66 T A B L E 11 Mean litter arcsine percent moles ± S.E. among implants and percent exencephaly ± S.E. among scoreable embryos in SELH/Bc, SWV/Bc, and ICR/Be mice after treatment with 600 mg/kg of valproic acid or an equivalent volume of distilled water. 69 T A B L E 12 Effects of folic acid supplementation on the in utero mortality rate and frequency of exencephaly in SELH/Bc mice treated with 600 mg/kg of valproic acid on D8hl2 of gestation. 80 T A B L E A. 1 Testing for difference between the in utero mortality data for SELH/Bc mice treated with either vehicle or methionine on D8h08 of gestation in the study of the effect of methionine supplementation on the expression of NTDs in SELH/Bc mice. 100 v T A B L E A.2 T A B L E A 3 T A B L E A.4 T A B L E A.5 T A B L E A.6 T A B L E A.7 T A B L E A.8 T A B L E A.9 T A B L E A. 10 Testing for difference between the exencephaly frequency data for SELH/Bc mice treated with either vehicle or methionine on D8h08 of gestation in the study of the effect of methionine supplementation on the expression of NTDs in SELH/Bc mice. 101 Estimation of the daily water intake of a SELHA/Bc female. 102 Testing for difference between the in utero mortality data for SELHA/Bc mice given either untreated drinking water or drinking water with a 0.20 mg/L folic acid supplement in the study of the effect of folic acid supplementation on the expression of NTDs in SELH mice. 103 Testing for difference between the exencephaly frequency data for SELHA/Bc mice given either untreated drinking water or drinking water with a 0.20 mg/L folic acid supplement in the study of the effect of folic acid supplementation on the expression of NTDs in SELH mice. 104 Testing for difference between the in utero mortality data for SELH/Bc mice fed the base Harlan Teklad diet with a 0 ppm, 2 ppm, or 10 ppm folic acid supplement in the study of the effect of folic acid supplementation on the expression of NTDs in SELH mice. 105 Testing for difference between the exencephaly frequency data for SELH/Bc mice fed the base Harlan Teklad diet with a 0 ppm, 2 ppm, or 10 ppm folic acid supplement in the study of the effect of folic acid supplementation on the expression of NTDs in SELH mice. 108 Testing for difference between the exencephaly frequency data for SELH/Bc mice fed either Purina Lab Chow or Harlan Teklad Diet TD93013 in the study of the effect of the Harlan Teklad chemically-defined diet (compared to Purina Lab Chow) on the expression of NTDs in SELH/Bc mice. I l l Testing for difference between the in utero mortality data for SELH/Bc mice fed either Purina Lab Chow or Harlan Teklad Diet TD93013 in the study of the effect of the Harlan Teklad chemically-defined diet (compared to Purina Lab Chow) on the expression of NTDs in SELH/Bc mice. 113 Testing for difference between the exencephaly frequency data for SELH/Bc mice treated with either vehicle or valproic acid on D8h08 of gestation in the original study of the effect of valproic acid on the expression of NTDs in three strains of mice with different genetic liabilities to exencephaly. 114 T A B L E A. 11 Testing for difference between the exencephaly frequency data for SELH/Bc mice treated with either vehicle or valproic acid on D8hl2 of gestation in the original study of the effect of valproic acid on the expression of NTDs in three strains of mice with different genetic liabilities to exencephaly. 115 vi T A B L E A. 12 T A B L E A. 13 T A B L E A. 14 T A B L E A. 15 T A B L E A. 16 T A B L E A. 17 T A B L E A. 18 T A B L E A. 19 T A B L E A.20 Testing for difference between the exencephaly frequency data for SELH/Bc mice treated with either vehicle or valproic acid on D8hl6 of gestation in the original study of the effect of valproic acid on the expression of NTDs in three strains of mice with different genetic liabilities to exencephaly. 116 Testing for difference between the exencephaly frequency data for SWV/Bc mice treated with either vehicle or valproic acid on D8h08 of gestation in the original study of the effect of valproic acid on the expression of NTDs in three strains of mice with different genetic liabilities to exencephaly. 117 Testing for difference between the exencephaly frequency data for SWV/Bc mice treated with either vehicle or valproic acid on D8hl2 of gestation in the original study of the effect of valproic acid on the expression of NTDs in three strains of mice with different genetic liabilities to exencephaly. 118 Testing for difference between the exencephaly frequency data for SWV/Bc mice treated with either vehicle or valproic acid on D8hl6 of gestation in the original study of the effect of valproic acid on the expression of NTDs in three strains of mice with different genetic liabilities to exencephaly. 119 Testing for difference between the exencephaly frequency data for ICR/Be mice treated with either vehicle or valproic acid on D8h08 of gestation in the original study of the effect of valproic acid on the expression of NTDs in three strains of mice with different genetic liabilities to exencephaly. 120 Testing for difference between the exencephaly frequency data for ICR/Be mice treated with either vehicle or valproic acid on D8hl2 of gestation in the original study of the effect of valproic acid on the expression of NTDs in three strains of mice with different genetic liabilities to exencephaly. 121 Testing for difference between the exencephaly frequency data for ICRVBc mice treated with either vehicle or valproic acid on D8hl6 of gestation in the original study of the effect of valproic acid on the expression of NTDs in three strains of mice with different genetic liabilities to exencephaly. 122 Testing for difference between the exencephaly frequency data for SELH/Bc mice treated with valproic acid on D8h08, D8hl2, or D8hl6 of gestation in the original study of the effect of valproic acid on the expression of NTDs in three strains of mice with different genetic liabilities to exencephaly. 123 Testing for difference between the exencephaly frequency data for SWV/Bc mice treated with valproic acid on D8hl6 of gestation in the original or follow-up study of the effect of valproic acid on the expression of NTDs in three strains of mice with different genetic liabilities to exencephaly. 125 vii T A B L E A.22 T A B L E A.23 T A B L E A.24 T A B L E A.25 T A B L E A.21 Testing for difference between the exencephaly frequency data for SWV/Bc mice treated with valproic acid on D8h08, D8hl2, D8hl6, or D8h20 of gestation in the original and follow-up studies of the effect of valproic acid on the expression of NTDs in three strains of mice with different genetic liabilities to exencephaly. 126 Testing for difference between the exencephaly frequency data for ICR/Be mice treated with valproic acid on D8hl6 of gestation in the original or follow-up study of the effect of valproic acid on the expression of NTDs in three strains of mice with different genetic liabilities to exencephaly. 128 Testing for difference between the exencephaly frequency data for ICR/Be mice treated with valproic acid on D8h08, D8hl2, D8hl6, or D8h20 of gestation in the original and follow-up studies of the effect of valproic acid on the expression of NTDs in three strains of mice with different genetic liabilities to exencephaly. 129 Testing for difference between the in utero mortality data for SELH/Bc mice treated with either vehicle or valproic acid on day eight of gestation in the original study of the effect of valproic acid on the expression of NTDs in three strains of mice with different genetic liabilities to exencephaly. 131 Testing for difference between the in utero mortality data for SWV/Bc mice treated with either vehicle or valproic acid on day eight of gestation in the original or follow-up study of the effect of valproic acid on the expression of NTDs in three strains of mice with different genetic liabilities to exencephaly. 134 Testing for difference between the in utero mortality data for ICR/Be mice treated with either vehicle or valproic acid on day eight of gestation in the original or follow-up study of the effect of valproic acid on the expression of NTDs in three strains of mice with different genetic liabilities to exencephaly. 137 Testing for difference between the pooled in utero mortality data for SELH/Bc, SWV/Bc, or ICR/Be mice treated with vehicle on day eight of gestation in the original study of the effect of valproic acid on the expression of NTDs in three strains of mice with different genetic liabilities to exencephaly. 140 Testing for difference between the exencephaly frequency data for SELH/Bc, SWV/Bc, or ICR/Be mice treated with valproic acid at the peak response time on day eight of gestation, in the original and follow-up studies of the effect of valproic acid on the expression of NTDs in three strains of mice with different genetic liabilities to exencephaly. 143 Testing for difference between the in utero mortality data for SELH/Bc mice fed the base Harlan Teklad diet with either a 2 ppm or 10 ppm folic acid supplement in the study of the effect of folic acid supplementation on the expression of NTDs in SELH/Bc mice treated with valproic acid. 145 T A B L E B. 1 Composition of the Harlan Teklad Diet TD92053 (0 ppm Folic Acid). 147 T A B L E A.26 T A B L E A.27 T A B L E A.28 T A B L E A.29 viii T A B L E B.2 Composition of the Harlan Teklad Diet TD92054 (2 ppm Folic Acid). 149 T A B L E B.3 Composition of the Harlan Teklad Diet TD93013 (10 ppm Folic Acid). 151 T A B L E B.4 Composition of Purina Lab Chow based on the latest ingredient analysis information. 153 ix LIST OF FIGURES FIGURE 1 Intracellular pathways of folate-mediated one-carbon metabolism and their relation to methionine/homocysteine metabolism and vitamin B12 function. 7 FIGURE 2 Schematic representation of the multifactorial threshold model of inheritance. 11 FIGURE 3 General structure of tetrahydrofolate polyglutamate cofactors (a) and folic acid (b). 19 FIGURE 4 Organization of intracellular pathways of folate metabolism in eucaryotes. 22 FIGURE 5 A spontaneous apical blister in a vehicle treated SELH/Bc mouse. 40 FIGURE 6 Schematic representation of the preventative effect of the Purina Lab Chow diet on the frequency of exencephaly in SELH/Bc mice. 55 FIGURE 7 Comparison among strains of the relative change in the frequency of exencephaly after treatment with 600 mg/kg of valproic acid over a twelve hour period on day eight of gestation. 67 FIGURE 8 Comparison among strains of the relative change in the in utero mortality rate after treatment with 600 mg/kg of valproic acid over a twelve hour period on day eight of gestation. 68 FIGURE 9 Valproic acid-induced exencephaly and midfacial clefts in ICR/Be (a,g), SWV/Bc (b,h), and SELH/Bc (c) mice. 71 FIGURE 10 A valproic acid-induced malformation observed only in SELH/Bc (a) and ICR/Be (b) mice characterized by a distinct gap between the lobes of the prosencephalon. 72 FIGURE 11 Comparison among strains of the relative change in the frequency of exencephaly after treatment with 600 mg/kg of valproic acid at the peak response time on day eight of gestation. 77 FIGURE 12 Schematic representation of the expected change in the frequency of exencephaly if the effect of diet and valproic acid treatment were additive. 82 x A C K N O W L E D G E M E N T S Special thanks to Dr. Virginia Diewert who was instrumental in planning the folic acid supplementation studies. Dr. Diewert openly shared her unpublished results and freely gave of her time and expertise. I would also like to thank Eunah Makoni and Norah Moorhead for taking care of the animals. I am indebted to Dr. Diana Juriloff and Dr. Muriel Harris, who approached me about graduate studies and entertained my dream of studying medicine. I wish to acknowledge their commitment to the education of their students and their enthusiasm for science - it is inspiring. And to my friends and family, who never stopped believing in me. xi INTRODUCTION I. Neural Tube Defects The neural tube is the embryonic progenitor of the brain and spinal cord (Elwood etai, 1992a). Neural tube defects (NTDs) encompass a spectrum of disorders all due to the failure of the neural tube to close properly toward the end of the fourth week of development (Copp et al., 1990). The severity of the disorder depends on the level of the lesion and the degree of neural involvement. Failure of the rostral end of the neural tube to close results in a condition termed anencephaly - literally meaning "absence of the brain" (Elwood et al., 1992a). Anencephaly is lethal before or soon after birth (Elwood et al., 1992a). Failure of the caudal end of the neural tube to close results in a condition termed spina bifida (Elwood etal., 1992a). The prognosis of infants with spina bifida depends on the degree of disruption of the vertebrae and spinal nerve pathways, but the infant usually suffers from multiple neurological deficits affecting the musculoskeletal, gastrointestinal, and urinary systems (Elwood etal., 1980). Failure of the neural tube to close along most of its rostral-caudal axis results in a lethal condition termed cranio-rachischisis (Elwood etal., 1992a). Less severe, skin-covered lesions (e.g., encephalocele) have also been described (Elwood etal, 1992a). Anencephaly and spina bifida account for approximately 90% of NTDs (Holmes etal., 1976; Khoury etal., 1982; Hall etal, 1988; Van Allen etal, 1993). Most of these are considered to be "isolated" defects (Khoury etal, 1982). This means that other defects if present are considered minor or secondary to the defect of neural tube closure (Khoury etal, 1982). Adrenal hypoplasia in anencephalic infants (Lemire, 1975) and hydrocephaly (with the Arnold-Chiari anomaly) in infants with spina bifida (Elwood etal, 1992a; Lemire, 1975) are examples of secondary defects. Unrelated congenital anomalies do occur among probands with anencephaly (14.7%) and spina bifida (10.1%) but are less frequent than among probands with cranio-rachischisis (62%), encephalocele (30%), or multiple NTDs (25%) (Hall etal, 1988). A small proportion of NTDs occur as part of a recognized syndrome (Holmes etal, 1976; Khoury etal, 1 1982): the classical triad of occipital encephalocele, cystic kidney dysplasia, and Polydactyly associated with Meckel's syndrome is an example (Fraser and Lytwyn, 1981). A. Incidence Despite advances in antenatal diagnosis and the choice of selective termination, neural tube defects (NTDs) remain a major public health problem (Elwood etal, 1992b). During the 1980s, the incidence of NTDs was around 1-2 per 1000 births (excluding terminations) in most of the British Isles, continental Europe, the United States, and Australia - making NTDs one of the most prevalent congenital malformations in humans (Elwood et al., 1992b). Remarkable variations in the incidence of NTDs have been observed. For example, during the twelve month period between October 1986 and September 1987, the Chinese Birth Defects Monitoring Program reported a 15 fold variation in the overall rate of NTDs (anencephalus, spina bifida, and encephalocele combined) among the 29 provinces studied: the lowest rate (0.7 per 1000) was reported in Hubei; the highest rate (10.6 per 1000) was reported in Shanxi (Little and Elwood, 1992a). Within the British Isles, where many surveys based on multiple sources of ascertainment are available, the incidence of NTDs in Northern Ireland has been consistently higher than in England and Wales: for example, during the period from 1974 to 1979, rates of 6.3 per 1000 (anencephalus and spina bifida combined) were recorded in Northern Ireland compared to 2.5 per 1000 in England and Wales (Little and Elwood, 1992a). In addition to geographic variations, large variations over time have been observed; for example, increased rates of NTDs were reported following the Great Depression in the late 1920s in the United States and the Dutch famine after the Second World War (Scott etal, 1995). Elwood etal. (1992c) summarized the epidemiology of NTDs in the following way: "Their distribution in populations shows virtually every variation possible. By time, there are short-term variations; by season, there are variations over a few years, and some remarkable very long-term fluctuations; by place, the defects have shown large variations between and within countries; and by person, their frequency varies with the mother's racial and ethnic background, social and economic circumstances, age and previous reproductive history, and is influenced by some drugs she may have taken and possibly by her occupational and environmental exposures." 2 B. Etiology It has long been suspected that there are environmental causes of neural tube defects (NTDs). Epidemiological studies have found strong evidence of non-genetic factors on the etiology of NTDs (see Introduction Section I.A). In the British Isles, most studies up to the mid-1970s showed that socioeconomic class was correlated with the prevalence of NTDs (Little and Elwood, 1992b). Women in low socioeconomic groups had a two to four fold higher risk of having a pregnancy affected by anencephaly or spina bifida (Little and Elwood, 1992b). Seasonal fluctuations in the incidence of NTD pregnancies were also observed (Little and Elwood, 1992c), with most studies reporting peak prevalences during the winter months (corresponding to infants conceived in spring and early summer). As the seasonal fluctuations were not limited to one socioeconomic class and dietary deficiencies were known to be teratogenic in experimental animal studies - nutritional factors, particularly the availability of fresh fruit and vegetables, were suspected of being at least partially responsible for these changes in frequency (Elwood etal., 1992d). The hypothesis that vitamin insufficiency could be an etiological factor in the development of NTDs was furthered by Smithells and colleagues. Smithells etal. (1976) assayed prospectively red blood cell folate, serum folate, white blood cell vitamin C, riboflavin, and serum vitamin A of women in the first trimester of pregnancy and correlated their findings with the outcome of the pregnancy. Mothers known to be taking vitamin supplements were excluded (Smithells et al., 1976). Six of the 1290 mothers recruited into the study gave birth to infants with central nervous system defects, one of which was not a NTD (microcephaly) (Smithells etal., 1976). Compared to the other participants, these six mothers were found to have significantly lower mean values for first trimester red blood cell folate (p < 0.001, n = 965) and white blood cell vitamin C (p < 0.05, n = 1104) (Smithells etal, 1976). Several lines of evidence were consistent with the notion that folic acid insufficiency might be an important cause of NTDs. Firstly, folic acid deficiency causes multiple severe congenital malformations in experimental animals although it is extremely hard to induce and the result is 3 usually embryonic death (Nelson 1960). Secondly, congenital malformations have been reported in the abortuses of women given therapeutic abortions with the folic acid antagonist 4-aminopteroyglutamic acid (for review, see Seller [1994] citing Thiersch, 1952). Thirdly, a study by Hibbard and Smithells (1965) suggested that folic acid metabolism as measured by the formiminoglutamic acid (FIGLU) excretion test was more often defective in mothers of infants with congenital malformations (although it is now recognized that about one third of all patients admitted to hospital also have modestly increased excretion of FIGLU due to disorders other than defective folate metabolism) (Chanarin, 1986). These observations led Smithells etal. (1980 and 1983) and Laurence etal. (1981) to conduct small clinical trials of folic acid supplementation in women who had a previous pregnancy complicated by a NTD and were thereby at increased risk of having a recurrence. The first trial administered 0.36 mg of folic acid and a mixture of seven vitamins (A, 4000 U; D, 400 U; B i , 1.5 mg; B2, 1-5 mg; B6, 1.0 mg; C, 40 mg; and nicotinamide, 15 mg) and minerals (ferrous sulphate, equivalent to 75.6 mg iron and calcium phosphate, 480 mg) to two cohorts of women (Smithells et al., 1980 and 1983); whereas, the second trial administered 4 mg of folic acid alone (Laurence et al., 1981). Both studies showed fewer recurrences in the supplemented group (Smithells etal., 1980: 0.6% [1/178] versus 5% [13/260], p < 0.01; Smithells etal, 1983: 0.9% [2/234] versus 5% [11/215], p < 0.007; Laurence etal, 1981: 0% [0/44] versus 9% [6/67], p = 0.04); however, statistical significance was achieved in the Laurence etal. (1981) trial only after 16 non-compliant women allocated to receive folic acid - including 2 recurrences - were added to the placebo group. Despite criticism of the imposed design of Smithells etal.'s (1980 and 1983) non-randomized trial and the small numbers and method of analysis in Laurence etaVs (1981) double-blind randomized placebo trial (Wald and Polani, 1984) - the possibility of prevention of recurrence of NTDs raised by these studies was taken sufficiently seriously for the Medical Research Council (M.R.C.) in Britain to initiate a multinational, multicenter double-blind randomized placebo trial to determine whether supplementation with folic acid or the other vitamins used by Smithells etal. (1980 and 1983) around the time of conception could reduce the recurrence 4 risk of NTDs (M.R.C. Vitamin Study Research Group, 1991). To ensure that a negative result could not be attributed to using too low a dose of folic acid, the 4 mg dose of folic acid used by Laurence etal. (1981) was administered in the M.R.C. Vitamin Study rather than the 0.36 mg dose of folic acid used by Smithells etal. (1980 and 1983) (M.R.C. Vitamin Study Research Group, 1991). One thousand eight hundred seventeen women were recruited for the M.R.C. study and randomly allocated to receive mineral supplements (ferrous sulphate, dried 120 mg; dicalcium phosphate, 240 mg) containing folic acid (group A), other vitamins (group D), both (group B), or neither (group C) (M.R.C. Vitamin Study Research Group, 1991). One thousand one hundred and ninety-five women had completed pregnancies in which the fetus or infant was known to have or not to have a NTD (M.R.C. Vitamin Study Research Group, 1991). There were 27 recurrences of NTDs among this group of women: 6 in the folic acid supplemented groups (groups A and B: 6/593 or 1.0%) and 21 in the other two groups (groups C and D: 21/602 or 3.5%) - corresponding to a 72 percent (relative risk 0.28) lower incidence of NTDs among women who had received supplements containing folic acid (M.R.C. Vitamin Study Research Group, 1991). The relative risk among women who received supplements containing nicotinamide and vitamins A, D, B\, B2, B6, and C (groups B and D: 12/597 or 2.0%) compared to women who received supplements without these vitamins (groups A and C: 15/598 or 2.5%) was not statistically significant (M.R.C. Vitamin Study Research Group, 1991). The M.R.C. Vitamin Study provides the first conclusive evidence that folic acid supplementation can reduce the recurrence risk of NTDs, but over 95% of infants born with NTDs are first occurrences (Czeizel and Dudas, 1992). Czeizel and Dudas initiated a double-blind randomized placebo trial in 1984 to determine whether vitamin supplementation around the time of conception could prevent the first occurrence of NTDs (Czeizel and Dudas, 1992). A total of 7540 women planning a pregnancy (in most cases their first) were randomly allocated to receive trace-element supplements containing either 0.8 mg of folic acid and a mixture of eleven other vitamins (A, 6000 U until the end of 1989 and 4000 U thereafter; D, 500 U; B1, 1.6 mg; B 2 , 1.8 mg; B 6 , 5 2.6 mg; B12,4pig; C, 100 mg; E, 15 mg; nicotinamide, 19 mg; calcium pantothenate, 10 mg; biotin, 0.2 mg) and four minerals (calcium, 125 mg; phosphorus, 125 mg; magnesium, 100 mg; iron, 60 mg) or 7.5 mg of vitamin C (Czeizel and Dudas, 1992). Four thousand one hundred and fifty-six women had completed pregnancies in which the fetus or infant was known to have or not to have a NTD (Czeizel and Dudas, 1992). Of the 2104 women who received the multivitamin supplement, there was a significantly lower prevalence of neural tube defects (0 versus 6, P = 0.029) compared to the 2052 women who received the control supplement (Czeizel and Dudas, 1992). Substantial corroborating evidence exists for folic acid reducing the occurrence risk of NTDs. Four case-control studies (Bower and Stanley, 1989; Martfnez-Frias and Rodriguez-Pinilla, 1992; Werler etal., 1993; Shaw etal., 1995) and one cohort study (Milunsky etal., 1989) have demonstrated that pregnant women taking vitamins containing folic acid are at lower risk of having children with NTDs than women not taking supplements. One case-control study found no association (Mills etal., 1989). Shaw etal. (1995) studied a population similar to Mills etal. (1989) but used a more extensive case ascertainment system and interviewed a greater percentage of case mothers among those eligible for ascertainment. Shaw etal. (1995) found that reduced risks were less marked or not observed among certain subsets of the population. In theory, the availability of dietary folate could account for the seasonal variations in the frequency of NTDs and may be partially responsible for the temporal and socioeconomic variations, but variations in the frequency of NTDs by ethnic group, maternal parity, sex of the infant, and family history are not easily explained by a dietary mechanism (Elwood et al, 1992d). In the search for genetic factors underlying susceptibility to NTDs, most attention has been given to genes affecting homocysteine metabolism. Figure 1 illustrates the intracellular pathways of folate-mediated one-carbon metabolism and their relation to methionine/homocysteine metabolism and vitamin B12 function. The possibility that homocysteine was important in the etiology of NTDs was first raised 6 7 by the work of Steegers-Theunissen etal. (1991). They documented elevated levels of plasma homocysteine following methionine loading in two cohorts of non-pregnant women who had a previous pregnancy complicated by a NTD (Steegers-Theunissen et al., 1991 and 1994) (Figure 1). In the initial cohort, 5 of the 16 women who had given birth to an infant with a NTD had methionine intolerance, that is, total plasma homocysteine levels six hours after methionine loading exceeding the mean plus two standard deviations in the controls (Steegers-Theunissen etal., 1991). Methionine intolerance was not observed in the 15 control women who had given birth to normal offspring (Steegers-Theunissen etal., 1991). The level of homocysteine observed in the methionine intolerant women was comparable to that observed in heterozygotes for homocystinuria (Steegers-Theunissen etal., 1991). Steegers-Theunissen and colleagues repeated their initial study with a second cohort of 41 non-pregnant women who previously had a pregnancy complicated by a NTD and 50 control women who previously had normal offspring (Steegers-Theunissen etal., 1994). The mean for total plasma homocysteine in both the fasting state and six hours after methionine loading was significantly increased in the group of women who previously had infants with NTDs. Methionine intolerance was reported in 9 of the 41 women and in 2 of the 50 control women (Steegers-Theunissen etal, 1994). None of the methionine intolerant women were heterozygotes for classic homocystinuria upon skin biopsy: the levels of cystathionine synthase (Figure 1) in skin fibroblasts were within the normal range (Steegers-Theunissen etal., 1994). Reduced metabolism of homocysteine could result from either: 1.) an inherited defect of a.) cystathionine synthase (or metabolism of its cofactor pyridoxal phosphate [vitamin FJ6]), b.) methionine synthase (or metabolism of its cofactors 5-methyl tetrahydrofolate [folate] and methylcobalamin [vitamin B12]), or c.) 5,10-methylene tetrahydrofolate reductase or 2.) an aquired deficiency of a.) vitamin B6, b.) folate, or c.) vitamin B12 (Figure 1). The explanation offered by Steegers-Theunissen and colleagues for their data suggested that an inherited derangement of folate or vitamin B12 metabolism may be responsible for the observed hyperhomocysteinemia by reducing the activity of methionine synthase (Steegers-Theunissen etal., 1994) (Figure 1). Earlier 8 reports of an association between maternal vitamin B12 status and NTDs (Kirke etal., 1993 and Adams etal., 1993: cited by Mills etal., 1996) provided corroborating evidence that the reaction catalysed by methionine synthase - one of only two enzymes known to use vitamin B12 as a cofactor (Stokstad etal., 1988) - was important in the etiology of NTDs. The relevance of the homocysteine hypothesis during pregnancy was demonstrated by Mills etal. (1995). In a retrospective study analysing blood samples collected during pregnancy, women who gave birth to infants with NTDs (case mothers, n = 76) had significantly higher plasma concentrations of homocysteine compared to the other pregnant women (control mothers, n = 315) (8.62 ^mol/L versus 7.96/<mol/L, p = 0.03) (Mills etal, 1995). However, plasma concentrations of vitamin B i 2 (263 ng/L versus 297 ng/L, p = 0.008) and folate (3.52 /<g/L versus 4.54 pg/L, p = 0.004) were significantly lower in the case mothers confounding this result (Mills etal, 1995). To remove the effect of varying plasma levels of vitamin B12, the case mothers were stratified (into deciles) based on plasma concentrations of vitamin B12 and compared to vitamin Bi2-matched controls, including additional control mothers with vitamin B12 levels below the mean plasma concentration in the case mothers (< 243 ng/L) (Mills etal, 1995). The activity of the vitamin Bi2-dependent enzymes methylmalonyl CoA isomerase and methionine synthase was evaluated by comparing the plasma levels of their substrates, methylmalonic acid and homocysteine respectively, in case and control mothers representing each decile of vitamin B12 (Mills etal, 1995). Plasma levels of methylmalonic acid were within the normal range and not different between the groups indicating that methylmalonyl CoA isomerase function was normal in both groups (Mills etal, 1995). In contrast, a significant relation was found between case status and plasma levels of homocysteine at plasma concentrations of vitamin B12 below 243 ng/L (p = 0.007) (Mills etal, 1995). Because methionine synthase is not the only enzyme that uses homocysteine as a substrate, the accumulation of homocysteine at low levels of vitamin B12 could not apriori be attributed to the decreased activity of methionine synthase in case mothers, despite the compelling logic for such an association. No significant relation was found between case status and plasma levels of either of methionine synthase's cof actors (folate and 9 vitamin B12) below vitamin B12 levels of 243 ng/L supporting the conclusion that an inherited defect of homocysteine metabolism was responsible for the hyperhomocysteinemia observed in the case mothers (Mills etal., 1995), although an aquired defect of vitamin B 6 could not be ruled out. Preliminary evidence suggests that a mutation (677 C—>T) in 5,10-methylene tetrahydrofolate reductase (Figure 1) may be associated with some cases of spina bifida (van der Put etal, 1995). Sixteen percent of mothers (n = 70), 10% of fathers (n = 60), and 13% of patients with spina bifida (n = 55) were homozygous for the mutation compared to 5% of controls (n = 207) (van der Put etal, 1995). Expanded studies are needed to confirm the frequency of the 677 C—>T mutation in the general population (de Franchis etal, 1995; Wilcken and Wang, 1996). The mutation reduces the activity of 5,10-methylene tetrahydrofolate reductase in both the homozygous and heterozygous states and is associated with elevated plasma homocysteine concentrations (van der Put etal, 1995). It is not known whether elevated maternal plasma concentrations of homocysteine are teratogenic or simply serve as a biochemical marker for the causative defect. Elevated levels of homocysteine in amniotic fluid of pregnancies complicated by a NTD has been observed (Steegers-Theunissen etal, 1995). Excess homocysteine has been shown to have teratogenic effects in whole embryo culture studies of the rat, but only at very high doses, and the vast majority of the induced malformations are not NTDs (VanAerts et al, 1994) making it more likely that elevated maternal levels of homocysteine are indirectly related to elevated risk for NTDs. One of the theoretical models that best fits the epidemiological data for the common NTDs is referred to as the multifactorial threshold model of inheritance (Little and Nevin, 1992). The term multifactorial is used to refer to the multiple factors, both genetic and environmental, each with only a relatively small effect that contribute to the predisposition or liability to disease (Thompson etal, 1991). The model is based on the assumption that there is an underlying normally distributed liability to disease with a threshold value beyond which the liability is expressed as an abnormal phenotype (Thompson etal, 1991). Figure 2 is a schematic representation of the multifactorial threshold model of inheritance. 10 Threshold Total liability (genes + environment) Figure 2. Schematic representation of the multifactorial threshold model of inheritance. The proportion of affected individuals in the general population (solid line) and the first degree relatives (broken line) is used to determine where the mean of the sampled population falls on the underlying scale of liability. Shading reflects the increased proportion of affected individuals in first degree relatives (speckled black) compared to the general population (solid black). Multifactorial inheritance has the following characteristics: 1.) The prevalence of multifactorial traits in the first degree relatives of the proband is approximately equal to the square root of the prevalence, p, in the general population; whereas, the prevalence in the second and third degree relatives is approximately equal to EVIDENCE: Although substantial epidemiological data has been collected on the recurrence risk of NTDs in first, second, and third degree relatives - the effects of both the declining prevalence of NTDs at birth and the bias of ascertainment on this data needs to be clarified (Little and Nevin, 1992). Overall, risk estimates based on the prevalence of NTDs in the general population have not been accurate predictors of the number of affected sibs, aunts/uncles, or first cousins (Little and Nevin, 1992). 2.) The higher the prevalence in the general population, the lower the relative increase in recurrence risk for multifactorial traits in first degree relatives (Thompson etal., 1991). EVIDENCE: In a series of epidemiological studies done in the British Isles between 1937 and 1972, the ratio of NTDs in sibs to the prevalence of anencephalus in the general population 11 does appear to decrease as the prevalence at birth of anencephalus in the general population increases (Little, 1992). 3. ) The recurrence risk for multifactorial traits is higher with more severe malformations (Thompson^ al, 1991). EVIDENCE: The concept of severity is not well defined for NTDs. If the presence of additional major anomalies were regarded as a measure of increased severity, then the existing data refutes this prediction (Little and Nevin, 1992). In pooled epidemiological data, Little (1992) calculated that the sib risk for probands with NTDs and other major anomalies was not greater than the sib risk for probands with isolated NTDs (2.44% [6/246] versus 2.83% [50/1764]). 4. ) The recurrence risk for multifactorial traits is higher after two or more affected births in contrast to Mendelian inheritance where the recurrence risk is independent of the number of affected births (Thompson et al, 1991). EVIDENCE: Strong evidence exists that the recurrence risk for NTDs after two affected sibs is higher than the recurrence risk for NTDs after a single affected birth (Little and Nevin, 1992). In pooled epidemiological data (excluding biased data from hospital studies), Little (1992) calculated the recurrence rate for NTDs after two affected sibs among 372 families as 64/578 or 11.1% (95% confidence intervals 8.5% to 13.6%). 5. ) The recurrence risk for multifactorial traits is higher for relatives of patients of the less susceptible sex, if the trait is more frequent in one sex than in the other (Thompson etal, 1991). EVIDENCE: Anencephaly is more common in females; however, the data do not support the prediction that the risk to sibs of male probands is greater than the risk to sibs of female probands (Little and Nevin, 1992). In pooled epidemiological data comparing sib risk by sex of the proband, the recurrence risk in sibs of male probands was not greater than the recurrence risk in female probands (4.4% [131/2949] versus 4.5% [205/4565]) (Little, 1992). 12 6.) The recurrence risk for multifactorial traits is higher when the parents are consanguineous (Thompson et al., 1991). EVIDENCE: The available epidemiological data indicate a small increase in parental consanguinity among probands with NTDs^ but the data are inconsistent and of varying quality (Little and Nevin, 1992). In summary, the epidemiological data available for NTDs support (2.) and (4.), refute (5.), and are insufficient to support or refute (1.), (3.), and (6.) (for review, see Little and Nevin [1992]). Modifications to the multifactorial threshold model have been proposed, for example, the inclusion of major gene effects on susceptibility to environmental factors, but it is difficult to distinguish among these variants using goodness of fit analyses because many make similar predictions at the extreme end of the distribution. For discrete traits like NTDs only the extreme end of the distribution is visible, that is, the proportion of affected individuals; thus, more than one variant of the multifactorial threshold model is consistent with the epidemiological data. II. Models for the Inheritance of Neural Tube Defects A. Genetic Models Eighty to 88% of human neural tube defects (NTDs) occur as isolated malformations and are considered to be multifactorial in origin (Khoury et al., 1982). Other identifiable causes of NTDs in humans include single-gene defects with pleiotropic expression (eg., Meckel's syndrome), chromosomal abnormalities (eg., trisomy 13, trisomy 18, and triploidy), and teratogens (eg., valproic acid) (Hunter, 1993). It is not clear what proportion of NTDs are due to these other causes. In a prospective study of 27 stillborn and liveborn infants with NTDs, Holmes etal. (1976) found 1 case (3.7%) with a cause (unknown) not attibuted to multifactorial inheritance. By contrast, in a retrospective study of 79 infants with NTDs studied post mortem, Holmes etal. (1976) found 12 cases (15.2%) with a cause not attributed to multifactorial inheritance: Meckel's syndrome (6.3%), trisomy 13 (1.3%), and unknown causes (7.6%). 13 Another retrospective study by Khoury etal. (1982) of 563 and 9403 stillborn and liveborn infants with NTDs ascertained from two birth defect surveillance programs found less than 1% of the cases could be attributed to a known cause. This situation is reversed in genetic animal models of NTDs (Table 1): the vast majority of TABLE 1. Classification of NTD in Genetic Models (Mouse Unless Stated Otherwise)3 Neural Gene Neural tube defectb crest Other Genetic Model symbol Chromosome Cranial Spinal defects malformations Ref.c i. Chromosomal Abnormalities a. Snell's translocation T(2;4)lSn EX SB Yesd 1 b. Trisomy 12 and 14 EX 2 ii. Mutant Gene with Mendelian Inheritance a. Loop-tail Lp 1 Cranio-rachischisist Yese 3 b- Cramo-myeloschisis^ rafJ cms Cranio-rachischisist Yesf 4 c. Cranioschisis crn EXt Yesi 5 d. Crooked Cd 6 EXt Yesh 6 e Exencephaly* xn EX 7 f • Extra-toes* Xt, Xt b P h 13 EXt Yes1 8 g. Rib fusions Rf EXt Yesi 9 h. Curly tail ct 4 EX SBt Yesk 10 i- Splotch* Sp,Spd 1 EX SBt Yes Yes1 11 j. Axial defects Axd EX SB Yesm 12 k. Curtailed fc / tw5 17 SB Yesn 13 1. Vacuolated lens vl 1 SB Yes Yes° 14 iii. Mutant Genotype a. New Zealand white mice EX 15 b. MT/Hokldr EX 16 c. SELH/Bc EX 17 a Expanded variation of Table 4 by Copp etal. (1990). b E X , exencephaly; SB, spina bifida. c References: 1, O'Shea, 1986 citing O'Shea and Kaufman, 1983; 2, Pntz and Morriss-Kay, 1981; 3, Strong and Hollander, 1949; 4, Layton and Smith, 1977; 5, Kalter, 1981; 6, Morgan, 1954; 7, Wallace etal., 1978; 8, Johnson, 1967; 9, Theiler and Stevens, 1960; 10, Gruneberg, 1954; 11, Auerbach, 1954 ; 12, Essien, 1992; 13, Park etal., 1989; 14, Wilson and Wyatt, 1986 and 1988; 15, Vogelweidera/. , 1993; 16, Matsuda, 1990; 17, Juriloff etal, 1989; MacDonald et al, 1989. d Craniofacial abnormalities, distended neural tube. e Tail, vagina, behavioral. f A x i a l skeleton, tail. 8 Eye. n Tooth, tail, tail fur, eye, vertebra, behavioral. 1 Eye, snout, axial and appendicular skeleton, edema. J Rib, vertebra. k Tai l . 1 Tail, neural crest (pigment, dorsal root ganglia). m Tai l . n Tail, vertebra, notochord, ventral duplication of neural tube, hindgut, hindlimb paralysis, bladder distension, fecal impaction. ° Lens, tail, neural crest. t Neural tube defects are a recessive manifestation of the mutant gene, t The mutation is extinct. * Homologous human gene known. Extra-toes is caused by a mutation in the mouse GU3 gene; mutations in the homologous human gene GLI3 cause Greig cephalopolysyndactyly syndrome (Hui and Joyner, 1993). Splotch is caused by a mutation in the mouse Pax-3 gene; mutations in the homologous human gene HuP2 cause Waardenburg syndrome (Tassabehji etal, 1992). 14 genetic animal models of NTDs involve either chromosomal abnormalities (e.g., Snell's translocation, trisomy 12, and trisomy 14) or, more commonly, single-gene defects in which the NTDs occur as a recessive manifestation of the mutant gene. Only three genetically complex NTD mutants - MT/Hokldr, New Zealand White, and SELH/Bc - have been described, and of these, only the SELH/Bc mouse model has been extensively studied. Human homologues for two of the single gene mutants - Splotch and Extra-toes - are known. Splotch is caused by a mutation in the mouse Pax-3 gene; mutations in the homologous human geneHuP2 cause Waardenburg syndrome (Tassabehji etal., 1992). Extra-toes is caused by a mutation in the mouse GU3 gene; mutations in the homologous human gene GLI3 cause Greig cephalopolysyndactyly syndrome (Hui and Joyner, 1993). Neural tube defects are not a recognized feature of either of these syndromes (Winter and Baraitser, 1991); however, if NTDs are a recessive manifestation of the mutant gene (as in Splotch and Extra-toes), NTDs would be rare occurrences except in the offspring of consanguinous parents with the respective syndromes. A report of the prenatal diagnosis of lumbosacral spina bifida in a fetus whose mother was diagnosed afterwards as having Waardenburg syndrome is interesting because the parents were third cousins (Moline and Sandlin, 1993; Lindhout, 1994). To test for an association between NTDs and Waardenburg syndrome, Hoi etal. (1995) screened patients with spina bifida (37), encephalocele (1), and cranio-rachischisis (1) identified from families with two or more affected individuals for mutations in the HuP2 gene. One patient with lumbrosacral spina bifida and mild signs of Waardenburg syndrome was heterozygous for a frameshift mutation in the HuP2 gene (Hoi etal., 1995). This mutation segregated with the signs of Waardenburg syndrome in the family; however, DNA from the other family member with lumbrosacral spina bifida (deceased) was not available to screen (Hoi etal., 1995). Most human NTDs are considered to be multifactorial in origin, making it unlikely that single gene mutations identified with very high penetrance or chromosomal abnormalities will account for more than 10 to 20% of the total human burden of NTDs. However, there is evidence of genetic background effects for some of the single gene mutants (e.g., Curly-tail: Neumann et 15 al, 1994; Splotch: Estibeiro etal, 1993; Axial defects: Essien etal., 1990) which would suggest the existence of one or more polymorphic modifier loci; these loci could alter the penetrance of a mutation in a given population. Several putative modifier loci of the mutation in curly tail mice have been mapped to chromosome 4 (Neumann etal., 1994). The genetics of the complex mutant systems - MT/Hokldr, New Zealand White, and SELH/Bc - may be more representative of the genetics of the isolated human NTDs. However, the MT/Hokldr and New Zealand White inbred strains have not been exploited as a resource in the study of NTDs. This may be due to the lower incidence of NTDs in the MT/Hokldr and New Zealand White strains: most estimates of the frequency of NTDs in SELH/Bc mice are in the range of 10 to 20% (e.g., Macdonald etal, 1989; Juriloff etal, 1989; Tom etal, 1991); by contrast, 6% of MT/Hokldr mice (Matsuda, 1990) and 7% of New Zealand White mice have NTDs (Vogelweidera/., 1993). SELH/Bc mice are genetically liable to the cranial neural tube closure defect, exencephaly (Juriloff etal, 1989). Two to three genes are thought to cause this liability (Juriloff etal, 1989). Like most cases of the analogous human neural tube defect, anencephaly (Khoury etal, 1982) -the exencephaly observed in SELH/Bc mice is non-syndromic and genetically complex (Macdonald etal, 1989; Juriloff etal, 1989). The origin of the SELH/Bc mouse stock traces back to a cross made in 1977 between a partially inbred stock of mixed genetic background (BALB/cGa, 129/-, CBA/-) and "random bred BLU:Ha(ICR)" mice obtained from the Arbor Scientific Company Limited (Port Credit, Ontario) (Juriloff etal, 1989). Two subsequent backcrosses were made to the "random bred BLU:Ha(ICR)" mice at which time 87.5% of the genes in the stock would be expected to be of ICR origin with the remaining 12.5% of mixed BALB/cGa, 129/-, and CBA/- origin (Juriloff et al, 1989). The stock was inbred, and exencephaly was first observed in 1981 in the F5 newborns of breeding pairs that all traced back to one F3 breeding pair (Macdonald, 1988). The SELH/Bc mouse stock was created and maintained by selecting siblings of exencephalic newborns for 16 breeding in subsequent generations (Juriloff etal., 1989). SELH/Bc mice are descended from one exencephaly-producing F6 breeding pair (Juriloff etal., 1989). In the late 1980s, twenty-eight polymorphic enzyme loci (ldh-1, Pep-3, Mod-1, Mod-2, Mor-1, Mpi-1, Gpi-1, Lap-1, Adh-3, Hbb, Ahd-1, Akp-1, Gpd-1, Pgm-1, Pgm-2, Es-10, Np-1, Alp, Trf, Gpt-1, Pgk-2, Car-2, Pre-1, Ldr-1, Mup-1, Ggc, Xld-1, and Aco-1) were typed in one to twenty-two SELH/Bc mice (mean = five) (Juriloff etal., 1989). Only the Gpd-1 locus on chromosome 4 was found to segregate (Juriloff etal., 1989). In 1991, the coefficient of inbreeding (F) of the SELH/Bc colony exceeded 0.9 (Tom et al, 1991). One family of the SELH/Bc mouse stock, SELHA/Bc, was formally inbred by another twenty-one consecutive generations of brother-sister inbreeding and has an estimated F value of 0.99 (Gunn etal., 1993). Although the frequency of exencephaly among scoreable embryos varies, all SELH/Bc mice tested produce some exencephalic progeny (Macdonald etal., 1989), and most estimates of the frequency of exencephaly are in the range of 10-20% (e.g., Macdonald etal., 1989; Juriloff et al, 1989; Tom etal., 1991). The genetic liability to exencephaly is thought to be carried by every SELH/Bc embryo, and environmental and/or stochastic effects determine whether a given SELH/Bc embryo will express this liability and be born with exencephaly (Macdonald et al., 1989). B. Environmental Models The number of environmental agents known to cause neural tube defects (NTDs) when administered to pregnant mammals are numerous and diverse (Copp et ah, 1990). Copp et al. (1990) lists eighty-three such agents - ranging from certain vitamins, minerals, and drugs to frank laboratory and environmental teratogens. Many cause a spectrum of defects depending on the dose and timing of exposure during pregnancy (Copp etal., 1990). Environmental agents which reduce the frequency of NTDs or cause isolated NTDs have the greatest potential to advance our understanding of the pathogenesis of NTDs by allowing the elucidation of key gene-regulated steps 17 in neural tube closure. Three such agents are discussed in more detail below. All are thought to be relevant to the human situation. i . Folic Acid Folic acid is a water soluable vitamin present in leafy green vegetables, most fruits, and liver (Davis and Nicol, 1988). It is an essential nutritional requirement for humans and other mammals, even though there are folate producing bacteria in the gut of mammals (Davis and Nicol, 1988). The biochemical role of folate in cellular metabolism is to carry one-carbon units (Wagner, 1995) . Folate cofactors are involved in the synthesis of purines and pyrimidines which are required for the production of DNA and RNA and in the remethylation of homocysteine to methionine which is required for protein synthesis and the production of S-adenosylmethionine (Figure 1). S-adenosylmethionine is the universal methyl group donor for the methylation reactions that occur in the cell (Wagner, 1995). Primary prevention of NTDs has been demonstrated in humans by periconceptional supplementation with folic acid and multivitamins (occurrence: Czeizel and Dudas, 1992) or folic acid alone (recurrence: M.R.C. Vitamin Study Research Group, 1991) (for review, see Introduction Section LB). However, it is not known how folic acid prevents NTDs in humans. It may be significant that folic acid is not the naturally occurring form of the vitamin (Figure 3). The oxidized form of folate is used in pharmaceutical preparations since it is stable, but it is rarely present in vivo (Scott et al., 1995). Moreover, the bioavailability of folic acid is much (50%) greater than natural food folates (Bailey, 1992). Most food folates are polyglutamate derivatives that must be hydrolyzed to their monoglutamate forms prior to absorption across the intestinal mucosa (Shane, 1995). A recent study examining the effect of increasing dietary folate on red blood cell folate assessed the effectiveness of consumption of 400 ug/day of extra folate as natural food folates compared with folic acid supplements and fortified food over a three month period (Cuskelly et al., 1996) . Sixty-two non-pregnant women with no personal history of NTDs or history of NTDs in their first degree relatives were randomly assigned to one of five groups: folic acid supplement 18 a. b. COOH —i n pteridine p -aminobenzoic poly-Y-glutamate acid chain COOH Figure 3. General structure of tetrahydrofolate polyglutamate cofactors (a) and folic acid (b). R4 and R2 are either hydrogens or one-carbon units: -CH3 (methyl; most reduced); -CH2- (methylene; intermediate); 0=CH-, -CH=NH, -CH= (formyl, formimino, methenyl; most oxidized). Slight adaptation and expansion of Figure 1 by Appling (1991). (group I), folic acid-fortified foods (group II), natural food folates (group III), dietary advice (group IV), or control group (no intervention, group V) (Cuskelly etal., 1996). None of the study participants were taking supplements containing folic acid nor on medication or had chronic illnesses known to interfere with folate metabolism (Cuskelly etal., 1996). All of the study participants were asked to exclude folic acid-fortified foods from their diets for three months prior to the start of the intervention (Cuskelly etal., 1996). After this exclusion period, there was no significant difference between the groups in either red blood cell or dietary folate (Cuskelly et al., 1996). 19 Women in groups I, II, and III were contacted weekly and provided with a supply of their choice of foods (II and III) or with supplements (I) (Cuskelly etal., 1996). Women in group IV were given dietary advice to eat folate-rich foods (Cuskelly et al., 1996). Of the 62 women who were recruited, 49 proceeded to intervention (Cuskelly etal., 1996). Forty-five women completed the intervention and of these 3 did not meet dietary acceptance criteria and 1 was excluded due to technical difficulties in the laboratory (Cuskelly etal., 1996). All four interventions significantly increased dietary folate intakes based on a diet history and a food frequency questionnaire; however, folate status as assessed by red blood cell folate concentrations improved only in the groups consuming the extra folate as folic acid (group I: p < 0.01, n = 9 and group II: p < 0.01, n = 6) (Cuskelly etal., 1996). No significant increase in red blood cell folate concentrations was observed in the groups consuming the extra folate as natural food folates (group III: n = 10 and group IV: n = 7) (Cuskelly etal., 1996). This difference was attributed to the greater bioavailability of folic acid over natural food folates (Cuskelly etal., 1996). In humans, the main transport form of folate is 5-methyl-tetrahydrofolate (Shane, 1995) (Figure 1). Although some metabolism of folic acid, primarily to 5-methyl-tetrahydrofolate, occurs in the intestinal mucosa - it is not required for transport: at pharmacological doses, most of the transported vitamin appears unchanged in the portal circulation (Shane, 1995). Given that pharmocological doses of folic acid were used in the randomized trials which demonstrated the preventative effect of periconceptional folic acid supplementation (see Introduction Section I.B), one would expect high levels of folic acid in the blood of supplemented women. Without knowledge of the mechanism of action of folic acid few conclusions can be drawn about the nature of the genetic liability in susceptible individuals; nevertheless, it is interesting to speculate on how folic acid may mediate its preventative effect. The process of neural fold development and elevation is expected to create a very high demand for the nucleic acid endpoints of the folate metabolic pathway to form the new cells in the nucleus dense neural epithelium. This could be viewed as a tissue specific stress on the folate metabolic pathway. It is possible that the oxidized form of folate rescues the embryo during this period of elevated metabolic activity by 20 providing an alternate source of the vitamin. Unlike the breakdown product of natural food folates (5-methyl-tetrahydrofolate), folic acid enters the coenzymatically active tetrahydrofolate pool through the activity of dihydrofolate reductase (Figure 1). A mutation in the embryo's methionine synthase gene which reduced the conversion rate of 5-methyl-tetrahydrofolate to tetrahydrofolate could theoretically be complemented by maternal folic acid supplementation given sufficient dietary levels of methionine and/or possibly betaine (a degradation product of choline) (Figure 1). Betaine can act as an alternate donor of one-carbon units to regenerate the carbon sulphur skeleton of methionine (Figure 1); otherwise, methionine supplementation would be required. Under physiological conditions, the conversion of 5,10-methylene tetrahydrofolate to 5-methyl tetrahydrofolate is irreversible (Figure 1), and the activity of methionine synthase is essential to permit the retention and utilization of dietary folates by the cell (Shane, 1995). The metabolic consequences of a rate limiting mutation in the methionine synthase gene is effectively mimicked by vitamin B12 deficiency: Stokstad et al. (1988) have demonstrated in male Sprague-Dawley rats that methylmalonyl CoA isomerase, the only mammalian enzyme other than methionine synthase known to utilize vitamin B12 as a cofactor, is relatively resistant to the depletion of vitamin B i 2 stores. Thus, vitamin B i 2 deficiency selectively inhibits flux through the methionine synthase pathway mimicking the metabolic consequences of a rate limiting mutation in the methionine synthase gene. If folic acid side steps the methionine synthase pathway, folic acid supplementation would be predicted to restore folate levels in vitamin B i 2 deficient rats. Indeed, folic acid supplementation of vitamin B i 2 deficient rats was shown to return liver folate levels to normal; however, the rate of histidine oxidation - a reaction dependent on tetrahydrofolate -apparently was not restored by folic acid supplementation (Stokstad et al, 1988). Figure 4 highlights the pathway used by Stokstad etal. (1988) to quantify the in vivo rate of histidine oxidation. The vitamin B12 deficient rats were given an intraperitoneal injection of 0.5 piC\ of [ring-2-14C] -histidine along with 1 yimoX of cold histidine per gram of body weight and the level of radioactivity released as C 0 2 was measured (Stokstad etal., 1988). It is now known that one of the enzymes in the assay pathway (5,10-methenyl tetrahydrofolate cyclohydrolase) is part of 21 22 a afunctional enzyme consisting of 5,10-methylene tetrahydrofolate dehydrogenase, 5,10-methenyl tetrahydrofolate cyclohydrolase, and 10-formyl tetrahydrofolate synthetase (Figure 4). This enzyme complex, known as CI-tetrahydrofolate synthase, exists as "a homodimer containing two functionally independent domains, with dehydrogenase/cyclohydrolase activities sharing an overlapping active site on the NH2-[amino]-terminal domain and synthetase activity associated with the COOH-[carboxy]-terminal domain" (Appling, 1991). Substrate channelling between the dehydrogenase and cyclohydrolase activities of C l -tetrahydrofolate synthase would account for the apparently low level of histidine oxidation detected by the assay protocol. The radiolabeled 5,10-methenyl tetrahydrofolate derived from the oxidation of [ring-2-14C]-histidine cannot compete as a substrate for the cyclohydrolase reaction with the 5,10-methenyl tetrahydrofolate produced in the dehydrogenase reaction (Appling, 1991), resulting in the elimination of unlabelled CO2 by 10-formyl tetrahydrofolate reductase and leading to an erroneously low measured flux through the histidine oxidation pathway (Figure 4). It is exciting to speculate that individuals with reduced methionine synthase activity would be unable to recycle/regenerate the methionine carbon skeleton as quickly as normal individuals and would accumulate homocysteine (despite normal cystathionine synthase activity) when challenged with an oral methionine load. It would be interesting to repeat the experiments of Steegers-Theunissen etal. (1991 and 1994) testing both mothers and fathers to determine if any of the affected pregnancies were associated with methionine intolerant fathers. It should not be surprising that some methionine intolerant individuals were found among the control group given that NTDs result from such an intimate interaction of genetic and environmental susceptibilities that even known teratogens do not cause NTDs in every exposed embryo (Fraser, 1976). In summary, it appears that vitamin B12 deficiency would lead to increased levels of homocysteine and a relative deficiency of tetrahydrofolate and probably methionine, the product of the methionine synthase reaction. Therefore, maternal vitamin B i 2 status should be considered in hypotheses explaining the folic acid effect on NTDs. Although the experiments of Stokstad et al. (1988) show that folic acid supplementation appears not to rescue folate metabolism, their 23 conclusion appears to be flawed for the following reason. It is now known that the C l -tetrahydrofolate synthase exhibits substrate channelling between its dehydrogenase and cyclohydrolase activities (Appling, 1991). This means that the radiolabelled 5,10-methenyl tetrahydrofolate, derived from the oxidation of the exogenous [ring-2-14C] -histidine, cannot compete as a substrate with the unlabelled 5,10-methenyl tetrahydrofolate produced in the dehydrogenase reaction - leading to an erroneously low measured flux through the tetrahydrofolate-dependent histidine oxidation pathway (Figure 4). Elevated plasma homocysteine levels after methionine loading in non-pregnant women who had a previous pregnancy complicated by a NTD (Steegers-Theunissen etal., 1991 and 1994) is also consistent with the hypothesis that folic acid administration side steps a defect in methionine synthase. As shown in Figure 1 the enzyme methionine synthase simultaneously converts homocysteine to methionine and recycles a dead-end folate cofactor to tetrahydrofolate. Therefore, a defect in methionine synthase activity would be expected to result in an accumulation of homocysteine upon methionine loading. If a genetic model could be identified that responded to maternal supplementation with folic acid, it would provide a means to study the mechanism of prevention. To date, folic acid has been administered to only two genetic mutants - curly tail mice (Seller, 1994) and axial defects mice (Essien and Wannberg, 1993 a,b) - neither of which have responded to maternal supplementation with folic acid. ii. Methionine Methionine is an essential amino acid. Its availability is necessary for the normal growth and development of mammals (Finkelstein, 1990). It is involved in several fundamental biological processes including protein synthesis, the methylation of proteins, nucleic acids, and phospholipids, and the biogenesis of homocysteine and cysteine (Finkelstein, 1990) (Figure 1). Methionine's role in neurulation was first demonstrated in whole embryo culture studies of the rat. Rat embryos cultured on heterologous sera from various sources - for example, from women with histories of fetal wastage (Ferrari etal., 1986), from dogs (Flynn etal., 1987), and from cows (Coelho etal., 1989; Coelho and Klein, 1990) - demonstrated a specific requirement 24 for only methionine. The addition of methionine to the heterologous sera was shown to significantly reduce the frequency of NTDs in the cultured embryos relative to controls. Coelho and Klein (1990) showed that the tips of the neural folds failed to turn inwards in rat embryos cultured in unsupplemented cow serum suggesting that methionine may be required for normal microfilament contraction. Protein analysis of the neural tubes of these embryos revealed a reduction in microfilament-associated methylated amino acids (Coelho and Klein, 1990). Methionine supplementation has been shown to reduce the frequency of NTDs in one neural tube mutant, axial defects (Essien, 1992; Essien and Wannberg, 1993 a,b) but not in another neural tube mutant, curly tail (van Straaten etal., 1995). Axial defects mice carry an autosomal mutation (Axd) which has dominant effects on the neural tube and vertebral column in most strains but exhibits reduced penetrance and variable expressivity (Essien, 1992). BALB/cByJ mice are the most sensitive to the expression of the Axd mutation with 22% of embryos exhibiting curly tails and 1% of embryos displaying sacral spina bifida in crosses with Axdl+ mice (Essien, 1992). The Axd mutation was backcrossed onto a BALB/cByJ background and the current penetrance is approximately 60% in day fourteen embryos, with 31% of embryos exhibiting curly tails and 25-30% of embryos displaying open NTDs in heterozygous (Axdl+) matings (Essien, 1992; Essien and Wannberg, 1993a). Since no obligate heterozygote Axd/+ embryo has been observed with a persistent open NTD, open NTDs are thought to occur only in homozygous AxdIAxd embryos (Essien, 1992). The administration of 70 mg/kg of methionine to Axdl+ dams by intraperitoneal injection on day eight and nine of gestation resulted in a 41% reduction (from 29% to 17%) in the incidence of open NTDs in matings with Axd/+ sires (Essien, 1992). The response was dose dependent in that Axdl+ dams treated with 28 mg/kg of methionine had a lower reduction (28% versus 41%) in the incidence of open NTDs (Essien, 1992). The protective effect of methionine was not observed in Axd/+ dams treated on day seven and day eight of gestation or on day nine and day ten of gestation (Essien, 1992). The primary prevention effected by methionine was specific and was not replaced by the folate derivative 5-formyl tetrahydrofolate (folinic acid, a metabolically related 25 compound converted to 5,10-methenyl tetrahydrofolate in the cell [Wagner, 1995]) or vitamin B12 (Essien and Wannberg 1993 a,b) (Figure 1). Higher doses of methionine (> 70 mg/kg) did not provide a greater level of protection (Essien and Wannberg, 1993 a,b). Methionine is thought to partially ameliorate whatever process is affected by the Axd gene during neurulation; however, the cellular and biochemical basis for this effect is unknown (Essien, 1992; Essien and Wannberg, 1993 a,b). In light of the recent data suggesting that a mutation in the 5,10-methylene tetrahydrofolate reductase gene is associated with spina bifida in humans (van der Put etal., 1995), it is exciting to speculate that axial defects mice have a mutation in the homologous mouse gene. A mutation in the mouse homolog of 5,10-methylene tetrahydrofolate reductase would be expected to reduce the intracellular levels of 5-methyl tetrahydrofolate (Figure 1). During periods of metabolic stress on the methionine synthase pathway, the reduction in the intracellular levels of 5-methyl tetrahydrofolate may be sufficient to affect the activity of methionine synthase and cause a transient deficiency of methionine (Figure 1). Extra methionine during periods of rapid cell division would relieve some of the demand for the 5-methyl tetrahydrofolate produced by 5,10-methylene tetrahydrofolate reductase, explaining why methionine supplementation reduced the frequency of NTDs in Axdl+ matings. Levels of tetrahydrofolate would not be affected as mouse feed is supplemented with folic acid which is converted directly to tetrahydrofolate (Figure 1), explaining why folinic acid supplementation had no effect on the frequency of NTDs in Axd/+ matings. A putative mutation in 5,10-methylene tetrahydrofolate reductase could also account for the failure of higher doses of methionine to further reduce the frequency of NTDs in Axd/+ matings. Five-methyl tetrahydrofolate is not only an essential cofactor of methionine synthase, it also limits the utilization of S-adenosylmethionine by glycine methyltransferase (Finkelstein, 1990; Appling, 1991). The glycine methyltransferase pathway is postulated to be important in the catabolism of excess methionine (and S-adenosylmethionine) (Finkelstein, 1990). It is a "non-productive", high-capacity transmethylation reaction that uses a "nonessential substrate" (glycine) to produce a "nontoxic product" (sarcosine) which can be oxidized by sarcosine dehydrogenase (Figure 4) to 26 regenerate glycine and recycle the methyl group (Finkelstein, 1990; Appling, 1991). Five-methyl tetrahydrofolate is bound to glycine methyltransferase in vivo and is a powerful inhibitor (Appling, 1991). High levels of S-adenosylmethionine inhibit 5,10-methylene tetrahydrofolate reductase, reducing the concentration of 5-methyl tetrahydrofolate and thereby relieving the inhibition of glycine methyltransferase (Finkelstein, 1990). The cellular response to a decrease in the concentration of 5-methyl tetrahydrofolate is to increase the synthesis of 5-methyl tetrahydrofolate by 5,10-methylene tetrahydrofolate reductase. If this is not possible, as is postulated to be the case during periods of rapid cell division in axial defects mice, glycine methyltransferase would compete with essential methyl transferases for S-adenosylmethionine (Finkelstein, 1990) - limiting the protective effect of methionine supplementation. The extent to which a similar interaction could occur in human embryos with a homozygous mutation in 5,10-methylene tetrahydrofolate reductase is not known. A lot would depend on the nature of the human mutation. If high concentrations of substrate could increase the rate of the reaction catalysed by the mutant enzyme, then folic acid supplementation - due to its significantly increased bioavailability over natural food folates (Bailey, 1992; Cuskelly etal., 1996) - would be expected to increase the intracellular concentration of 5-methyl tetrahydrofolate and thereby have a protective effect. It should be noted that the intracellular synthesis of 5-methyl tetrahydrofolate cannot be replaced by dietary sources of 5-methyl tetrahydrofolate. The activity of 5,10-methylene tetrahydrofolate reductase is necessary to produce the polyglutamate form of 5-methyl tetrahydrofolate which is required for the inhibition of glycine methyltransferase (Appling, 1991). i i i . Valproic Ac id Valproic acid is an anticonvulsant drug used to treat epilepsy. Women taking valproic acid during the first trimester of pregnancy are known to have an elevated risk of a number of congenital malformations, including spina bifida (for review, see Lammer et al., [1987]). The estimated risk of a valproic acid exposed fetus having spina bifida is 1-2%, based on case-control studies (Lammer et al., 1987). 27 Although valproic acid is teratogenic in animal studies, rodent embryos exposed to valproic acid develop exencephaly not spina bifida (Nau etal., 1991). In addition to this species difference, there are also marked strain differences in the susceptibility to valproic acid induced exencephaly. Finnell etal. (1988) studied the teratogenic effects of valproic acid in three inbred strains of mice with no unusual incidence of spontaneous malformations. The same dose of valproic acid administered at the critical period for NTDs resulted in strain differences in frequency of exencephaly observed in the exposed litters (Finnell etal., 1988). Interestingly, the same hierarchy of susceptibility was observed for heat-induced exencephaly, that is, the strain that was least sensitive to the teratogenic effects of valproic acid was most resistant to heat-induced exencephaly (Finnell et al., 1986). Wegner and Nau (1992) have suggested that alterations in embryonic folate metabolism may be one of the mechanisms of valproic acid teratogenesis based on structure activity assays. A closely related structural analog of valproic acid which exhibited antiepileptic activity but not teratogenicity did not influence embryonic folate metabolism (Wegner and Nau, 1992). Teratogenic doses of valproic acid significantly reduced the levels of formylated tetrahydrofolates (5-formyl tetrahydrofolate and 10-formyl tetrahydrofolate) and increased the level of tetrahydrofolate in exposed rodent embryos (Wegner and Nau, 1992). These alterations were consistent with valproic acid inhibiting the formation of 5-formyl tetrahydrofolate from tetrahydrofolate, a reaction catalysed by glutamate formyltransferase (Wegner and Nau, 1992). The specific inhibition of glutamate formyltransferase by valproic acid was confirmed by in vitro studies (Wegner and Nau, 1992). Five-formyl tetrahydrofolate (folinic acid) has been shown to be a potent inhibitor of many folate utilizing reactions and may play a regulatory role in one-carbon metabolism (Wagner, 1995). The coadministration of 5-formyl tetrahydrofolate and valproic acid reduced the teratogenicity of valproic acid in mouse embryos in vivo (Trotz etal., 1987) but not rat embryos in vitro (Hansen and Grafton, 1990). 28 III. The Development of Neural Tube Defects A. The Normal Process of Neurulation The purpose of this section is not to give an exhaustive description of neurulation but rather to highlight those aspects that are especially relevant to our current understanding of the pathogenesis of neural tube defects (NTDs). Because human embryos cannot be examined directly and most pathological specimens are obtained after neural tube closure and suffer some degree of degeneration, much of our understanding of the normal process of neurulation comes from work on early development in a few laboratory species of amphibia, birds, and mammals (Copp etal., 1990). It is generally accepted that although early embryology appears to be very similar in laboratory species, the role of a given force in neurulation may be different in different species and at different levels of the rostrocaudal body axis (eg., brain versus spinal cord) (Copp etal., 1990). Moreover, significant differences in embryology do exist between the model systems further complicating the interpretation of the emerging picture of the process of neurulation (Schoenwolf and Smith, 1990). This review will focus on work done in mammalian, principally rodent, embryos. Neurulation, the process of neural tube formation, is the initial event in the development of the central nervous system (Schoenwolf and Smith, 1990). It is divided into two distinct phases in mammalian embryos (Morriss-Kay etal, 1994). The first phase, primary neurulation, begins with the formation of the neural plate and ends with the closure of the neural tube (Schoenwolf and Smith, 1990). For simplicity, this review will consider the process to occur in four recognizable stages - a.) formation of the neural plate, b.) shaping of the neural plate, c.) bending of the neural plate, and d.) closure of the neural groove (Schoenwolf and Smith, 1990); however, it should be noted that these discrete stages do not exist during primary neurulation. The second phase, secondary neurulation, begins after the closure of the caudal neuropore and involves the formation and subsequent canalization of a solid aggregate of neuroepithelial cells associated with the tail bud 29 (Schoenwolf, 1994). Secondary neurulation will not be discussed further because later stage human embryos and fetuses only have a diminutive caudal spine (Copp et al., 1990). a. Formation of the neural plate Gastrulation - the formation of the three primary germ layers - ectoderm, mesoderm, and endoderm - occurs before and during the stage of neural plate formation (Morriss-Kay etal., 1994). Although the process of neural induction is not understood in mammals (Morriss-Kay et al, 1994), the "morphogenetic movements of gastrulation [are postulated to] play an essential role in the formation of the neural plate and in neuraxis patterning by reshuffling cell layers and thereby bringing different tissues together, allowing new cell-cell inductive interactions to occur" (Schoenwolf, 1994). At the late presomite stage, the neural plate can be seen to differentiate within the epiblast, amultipotent stem cell population (Morriss-Kay etal., 1994; Copp etal., 1990). It forms a shield-shaped area of pseudostratified columnar epithelium with a midline groove overlying its attachment to the notochord (Morriss-Kay et al., 1994; Copp etal., 1990). At this stage the neural plate consists only of tissue that will develop into the prosencephalon, mesencephalon, and the most rostral part of the rhombencephalon (Morriss-Kay etal., 1994). As the embryo elongates, gastrulation continues at the caudal end of the embryo (Sulik and Sadler, 1993) and new neuroepithelium continues to differentiate from epiblast just rostral to the primitive streak until the closure of the caudal neuropore at around the thirty somite stage (Morriss-Kay et al, 1994). The thinner epithelium lateral and rostral to the neural plate forms the surface epithelium which will become epidermis later in development (Morriss-Kay etal, 1994; Copp et al, 1990). b. Shaping of the neural plate As the first few pairs of somites form, there is a change in the epithelial structure from pseudostratified to cuboidal epithelium resulting in the transverse expansion of the neural plate (Morriss-Kay etal., 1994). The cell number in the transverse plane of the mesencephalic and rostral rhombencephalic regions remains constant throughout neurulation (Morriss-Kay etal., 1994) . By contrast, the prosencephalic region which initially exists as a small amount of tissue at 30 the rostral edge of the neural plate undergoes tremendous growth in both transverse and longitudinal planes (Morriss-Kay etal., 1994). This size increase cannot be accounted for by cell division alone because the neuroepithelial cell cycle time is the same in all regions (Morriss-Kay et al., 1994). In rat embryos, the expansion of the prosencephalic region has been demonstrated in cell labelling studies to be due to the rostral flow of cells from the mesencephalic and rostral rhombencephalic regions (Morriss-Kay etal., 1994). The preotic sulcus, the future position of the boundary between rhombomeres 2 and 3, marks the caudal boundary of the forward flow (Morriss-Kay et al., 1994). c. Bending of the neural plate Concomitantly with the transverse expansion of the neural plate and the formation of the foregut together with the caudal movement of the heart, the flat neural plate is converted into a pair of convex neural folds as the amount of mesoderm beneath them increases (Morriss-Kay etal., 1994). The mesoderm in the prospective brain region is less condensed than more caudal regions owing to the relatively larger amounts of extracellular matrix (Morriss-Kay etal., 1994). Morriss-Kay etal. (1994) speculate that this is an important mechanism in the elevation of the broader neural folds associated with the development of the mammalian brain. The boundary between the extracellular matrix rich mesoderm rostrally and the more condensed mesoderm caudally is roughly at the level of the preotic sulcus (Morriss-Kay etal, 1994). The continued elevation of the neural folds into a V-shaped formation involves the conversion of the cuboidal neuroepithelium into a pseudostratified form which increases in thickness until closure is complete (Morriss-Kay etal., 1994). Development and maintenance of the concave curvature that precedes closure is associated with the formation and contraction of junction-related microfilament bundles termed adherens (Morriss-Kay etal., 1994). Disruption of the microfilament bundle integrity by cytochalasin D results in the concave curvature being lost (Morriss-Kay etal., 1994). Regeneration of the concave curvature is only possible if neural tube closure has been initiated at the junction of the prosencephalon and mesencephalon (Morriss-Kay etal., 1994). 31 d. Closure of the neural groove In the mouse, neural tube closure begins on day eight or nine of gestation (fertilization being day zero of gestation) when the embryo has developed about four somite pairs (Golden and Chernoff, 1993). At this stage, the mouse embryo is "cup shaped" with its dorsal aspect pointing inward (Jacobson and Tarn, 1982). Closure is initiated at the point of greatest rostro-caudal flexure, near the myelencephalon/cervical junction (Jacobson and Tarn, 1982). Human embryos, unlike mouse embryos, do not undergo axial rotation to assume the fetal position; nevertheless, the initial site of fusion of the neural folds is remarkably similar in both species (O'Rahilly and Muller, 1994). As Closure 1 progresses bidirectionally from the initial point of neural fold apposition (Golden and Chernoff, 1993), the mouse embryo elongates and rotates to assume the typical fetal position with its dorsal aspect pointing outward (Poelmann etal., 1987). During axial rotation, the neural folds at the level of the prosencephalic/mesencephalic junction approach each other and form a second fusion point (Golden and Chernoff, 1993). Closure 2 proceeds bidirectionally to meet up rostrally with Closure 3 and caudally with Closure 4 (Golden and Chernoff, 1993). Closure 3 is initiated from the most rostral point of the rostral neuropore and progresses caudally (Golden and Chernoff, 1993). A closure pattern similar to Closure 3 in the mouse has also been described in human embryos (Golden and Chernoff, 1993). Closure 4 is initiated over the caudal rhombencephalon and proceeds rostrally (Golden and Chernoff, 1993). Less is known about the mechanism of closure of the rhombencephalon (Closure 4) although it is thought to involve a membranous closure, that is, the surface ectoderm is believed to fuse before the neuroepithelium (Geelen and Langman, 1977). Cranial neural tube closure is complete by the time the mouse embryo has developed about twenty somite pairs on day nine of gestation (Golden and Chernoff, 1993). Primary neurulation is complete by the time the mouse embryo has developed about twenty-nine somite pairs on day ten of gestation (Copp etal., 1990). The role of other morphogenetic activities - such as formation of the germ layers, heart, gut, body folds, neural crest, and regional subdivisions of the mesoderm - in primary neurulation has only begun to be appreciated (Schoenwolf and Smith, 1990). For example, the development 32 of ventral cephalic flexure in the mouse appears to pause during the time of cephalic neurulation thereby preventing the converging neural folds from being splayed apart (Jacobson and Tarn, 1982), and the emigration of neural crest cells from mesencephalic and rostral rhombencephalic neural folds appears to be essential for the flexibility of the lateral edges allowing them to curve medially and fuse in the dorsal midline (Morriss-Kay etal., 1994). The cell proliferation defect in the hindgut endoderm and notochord of curly tail mutant mice is a dramatic example of the effect that other morphogenetic activities can have on neurulation: the abnormally reduced rate of cell proliferation in the hindgut endoderm and notochord heightens ventral curvature of the neural axis and opposes the force directing the convergence of the caudal neural folds, resulting in delayed closure of the caudal neuropore and lumbrosacral spina bifida in roughly 20% of homozygous diet embryos (van Straaten, 1993). Much still remains to be learned about the extrinsic forces in the mechanism of neurulation. The contemporary view of neurulation is a "multifactorial process" driven by the integration of a number of "diverse, fundamental cell behaviours" both within the neuroepithelium and the surrounding tissues (Schoenwolf, 1994). Such fundamental cell behaviours include "changes in cell shape, size, position, and number, as well as changes in cell-cell and cell-extracellular matrix interactions" (Schoenwolf, 1994). B. Lessons from the Study of Neural Tube Defects Basic research into the etiology of neural tube defects (NTDs) coupled with the results of epidemiological and pathological studies of human NTDs has resulted in the emergence of the hypothesis that "closure sites in humans can be identified, that the specific genes which regulate closure can be determined, and (that) specific environmental components can be identified which influence each closure site" (Hall, 1994). The multi-site initiation of closure of the neural tube was first described by Golden and Chernoff in an abstract published in 1983. Prior to this date, several investigators had documented more than one site of fusion in the cranial region of rodent embryos, but no one had summarized 33 these findings. Since Golden and Chernoff's initial report (1983), a striking variant of the multi-site initiation model of neural tube closure has been described. SELH/Bc mice are genetically liable to the cranial neural tube closure defect, exencephaly (Juriloff etal., 1989). The genetic liability to exencephaly is thought to be carried by every SELH/Bc embryo, and environmental and/or stochastic effects determine whether a given SELH/Bc embryo will express this liability and be born with exencephaly (Macdonald etal., 1989). Developmental studies of SELH/Bc embryos have demonstrated that SELH/Bc embryos lack the normal zone of contact and fusion between the neural folds at the level of the prosencephalon/mesencephalon junction (Closure 2) (Macdonald et al., 1989). Delayed closure of this region is completed by the caudal extension of Closure 3 (Macdonald et al., 1989). All but 10-20% of SELH/Bc embryos manage to close the cranial neural tube by this abnormal mechanism (e.g., Macdonald etal., 1989; Juriloff etal., 1989; Tom etal., 1991). An attempt to apply the multi-site initiation model of neural tube closure to interpret the pattern of NTDs observed in human pathological specimens was largely successful, except in the region of the lumbo-sacral spine (Van Allen etal., 1993). The four closure sites described in mouse embryos by Golden and Chernoff appeared to account for the pattern of human NTDs without having to invoke events like the secondary re-opening of the neural tube (Van Allen etal., 1993). Proponents of this approach suggest that a fifth closure site may exist in human to explain the observed pattern of NTDs in the region between the second lumbar vertebra and the second sacral vertebra (Van Allen et al., 1993). There are several reasons for thinking that specific genes can be determined which regulate closure of the neural tube. The existence of neural tube mutants (see Introduction Section II.A) and the recognition of characteristic differences in the timing and initiation of neural tube closure in normal inbred strains (Sakai, 1989; Macdonald etal., 1989; Juriloff etal., 1991; Golden and Chernoff, 1993) suggests that closure is under genetic control. Moreover, the demonstration that rhombomeres are not only morphological segments but also boundaries of cell movement, gap junctional permeability, and gene expression illustrates the genetic patterning of the neural axis 34 during neurulation (Morriss-Kay etal., 1994). But the most convincing evidence yet comes from studies of genetic correlation that conclusively demonstrate that the lack of Closure 2 in SELH/Bc mice is directly related to their risk of exencephaly (Gunn etal., 1995). Whether or not specific environmental components can be identified which influence each closure site, large-scale randomized interventional trials with folic acid in humans have demonstrated that environmental components exist (M.R.C. Vitamin Research Group, 1991; Czeizel and Dudas, 1992). It remains to be seen what the effect of widespread folic acid supplementation will be on the distribution of NTDs in the human population. If folic acid supplementation reduces the incidence of certain types of NTDs and not others, such a finding would be consistent with folic acid affecting some closure sites more than others. However, the closure site per se is not likely to be the determining factor; certain levels of the neural axis may rely upon a folic acid dependent or enhancable process more than others to facilitate closure of the neural tube. Basic research has provided important insights into possible mechanisms involved in the generation of NTDs; the hope is that the lessons learned from model systems will enable researchers to identify human candidate loci and devise preventative strategies. 35 EXPERIMENTAL STUDIES I. General Methods a. Animal maintenance All the mice were maintained in the Medical Genetics Mouse Unit at the University of British Columbia (Vancouver, British Columbia). The mice were housed together in groups of one to five in standard polycarbonate cages with dried crumbled corncob bedding and fed Purina Lab Chow and acidified water (pH 3.1, HC1) ad libitum unless otherwise stated. The temperature of the mouse unit was controlled (22 ± 2°C) (Juriloff etal., 1991), and the animal rooms were on a twelve hour light (6:00 am to 6:00 pm standard time) and dark cycle. For the purposes of timed matings, 8:00 am standard time on the day of the plug was taken to be day zero, hour eight, of gestation. b. Examination of embryos Females with timed pregnancies were killed by carbon dioxide gas on day fourteen of gestation. The uterus was immediately removed and placed in a petri dish on a black wax background. Isotonic saline (0.85% NaCl) was added to the petri dish to aid in the examination of the embryos. The uterus was secured and cut open under a binocular dissection microscope to reveal the chorionic sacs of the embryos. The isotonic saline (0.85% NaCl) was changed, and embryos were examined for major external malformations of the abdomen, digits, tail, ears, eyes, snout, and lower jaw with special emphasis placed on the morphology of the neural tube. Any embryonic abnormalities or moles (dead early postimplantation embryos) were recorded along with the position of the embryo or mole in the uterus. Any maternal abnormalities were noted when observed. Embryos were fixed and stored in Bouin's fixative. c. Data analysis The analysis of litter data is problematic because embryos within a litter share a common maternal environment and are therefore more like each other than like embryos from other litters; 36 consequently, each embryo is not an independent unit (D.M. Juriloff, personal communication). This phenomenon is attributed to differences in the uterine environment that might arise from differences in maternal physiology or biochemistry or, when the mother is treated, to small errors in dosage delivery or small differences between litters at the time of treatment in mean developmental age (Tom etal., 1991). Frequency data based on individual embryos were compared using the Chi-square test of independence (Zar, 1984). If the Chi-square test rejected the null hypothesis at the 0.05 significance level, a second test based on litters as a unit was also used to diminish the possible contributions of litter effects. In the second approach, the proportions of moles out of implantations or exencephalic embryos out of scoreable embryos were normalized by transformation to their Freeman-Tukey arcsine values (Mosteller and Youtz, 1961). The mean litter arcsine values for each group were compared (Mosteller and Youtz, 1961). If two groups were being compared, an unpaired t-test was used (Zar, 1984). To compare more than two groups, a one-way analysis of variance was used (Zar, 1984). If the one-way analysis of variance rejected the null hypothesis at the 0.05 significance level, a multiple comparison test - the Tukey test - was used (Zar, 1984). The Tukey test determined the relationship between the mean litter arcsine values for each group (Zar, 1984). Litters from females with fewer than five implantations, aborted litters, and litters from females with solid tumors were removed from the data as being atypical. The data were analysed statistically on an Apple Macintosh computer using the Analysis Tools and Functions available on Version 4.0 of Microsoft Excel from the Microsoft Corporation (Redmond, Washington). II. Study of the Effect of Methionine Supplementation on the Expression of NTDs in SELH/Bc mice A. Introduction The purpose of this experiment was to determine if methionine supplementation produced a reduction in the frequency of exencephaly in SELH/Bc mice, as had been seen for open neural tube 37 defects (NTDs) in another mouse neural tube mutant (axial defects). For review, see Introduction Section II.B.b. B. Materials and Methods a. Mice SELH/Bc mice were obtained from our breeding colony (for strain history, see Introduction Section II.A). Females were nulliparous and approximately 3 to 5 months of age when bred. Ten sires were used in the matings. b. Animal maintenance and breeding The mice were maintained under standard conditions (see Experimental Studies Section I). Timed pregnancies were obtained by placing one to three virgin females with singly-caged males overnight. Females were checked for copulation plugs before 10:00 a.m. using a blunt metal probe. Mated females were assigned alternately to either the methionine treated group or the vehicle control group. Unmated females and females with copulation plugs were removed from the male cages and housed separately in groups of four or less. Unmated females were returned to their respective male cages after 4:30 p.m. each day. c. Methionine L-Methionine was obtained from BDH Laboratory Supplies (Poole, England) in a 100 g bottle and stored at 4°C. Each week, approximately 10 mis of a 14 mg/ml solution of methionine was prepared by dissolving a known amount of L-methionine powder in freshly distilled water. One to 1.5 ml aliquots were immediately transferred to 2 ml serum vials which were then stoppered and stored at 4°C in a container wrapped in tin foil. A few minutes before use, a new vial was removed from the refrigerator and warmed to room temperature. Females were weighed on day eight, hour eight of gestation and given an intraperitoneal injection of 50 ul/10 g of body weight to administer a dose of 70 mg/kg of L-methionine or an equivalent volume of distilled water. This was administered at the same dosage and route of delivery as reported in the literature (Essien, 38 1992; Essien and Wannberg, 1993 a,b). After treatment, the females were housed according to treatment group and given a clean cage and cage lid and fresh food (Purina Lab Chow) and water. d. Data analysis The effect of methionine supplementation on the in utero mortality rate and frequency of exencephaly in SELH/Bc mice was tested by a Chi-square test of independence at the 0.05 significance level (Zar, 1984). C . Results Maternal administration of 70 mg/kg of methionine on day eight, hour eight of gestation had no detectable effect on the in utero mortality rate or the frequency of exencephaly in SELH/Bc mice (Table 2) (Appendix Tables A. 1 and A.2). The in utero mortality rate was 5% (5/107) in the vehicle control group and 4% (5/111) in the methionine treated group (Table 2). The frequency of exencephaly was 10% (10/102) in the vehicle control group and 8% (9/106) in the methionine treated group (Table 2). TABLE 2. Effects of methionine supplementation on the in utero mortality rate and frequency of exencephaly in SELH/Bc mice scored on day fourteen of gestation.1 Moles Exencephaly No. of No. of Meanlitter No. of Mean litter Treatment litters2 implants Number3 Percent arcsine embryos4 Number3 Percent arcsine Vehicle 92 107 5 4.7 14.7 1002 10 9.8 19.0 Methionine n i 111 5 4.5 14.6 106 9 8.5 18.8 1 A dose of 70 mg/kg body weight was administered by intraperitoneal injection on day eight, hour eight of gestation. Vehicle controls were injected with an equivalent volume of distilled water. 2 Litters with less than five implantations (1, control group), aborted litters (1, control group), and litters from females with solid tumors (1, methionine group: ovarian) were removed from the data and shown as superscripts. Differences between the vehicle and methionine treated groups were tested by a Chi-square test at the 0.05 significance level. 4 Embryos with odd head development are shown as superscripts. Two embryos with apical blisters were observed in the control group. Two embryos with apical blisters were observed in the vehicle control group (Figure 5) (Table 2). Apical blisters occur spontaneously at a low frequency in the SELH/Bc mouse stock. As these defects do not occur spontaneously at any appreciable frequency in other strains of mice (eg., SWV/Bc and ICRVBc mice), it is likely that they are a rare manifestation of the SELH/Bc 39 Figure 5. A spontaneous apical blister in a vehicle treated SELH/Bc mouse. Apical blisters occur spontaneously at a low frequency in the SELH/Bc mouse stock. As these defects do not occur spontaneously at any appreciable frequency in other strains of mice (eg., SWV/Bc and ICR/Be mice), it is likely that they are a rare manifestation of the SELH/Bc genotype. 40 genotype. The apical blisters are not always in the same place and are frequently offset from the midline. The defects range in size from approximately 1 mm to 3 mm in diameter. Although the neuroepithelium appears to be involved in some of the larger apical blisters upon macroscopic examination, neuroepithelial involvement has only been conclusively shown in a few embryos that have been examined microscopically (unpublished data). As it is not clear whether all apical blisters involve the underlying neuroepithelium, embryos with these defects were not considered to have NTDs for the purposes of this study. D. Discussion The purpose of this experiment was to determine if methionine supplementation produced a reduction in the frequency of exencephaly in SELH/Bc mice as had been seen for open NTDs in another mouse neural tube mutant (axial defects) when administered at the same dosage and route of delivery as reported in the literature (Essien, 1992; Essien and Wannberg, 1993 a,b). In contrast to the reduction in open NTDs observed in axial defects mice following methionine supplementation, there was no detectable effect of methionine supplementation on the in utero mortality rate or frequency of exencephaly in SELH/Bc mice (Table 2). This finding is probably a reflection of the different genetic etiologies of the two mutant systems. Methionine is thought to partially ameliorate whatever process is affected by the Axd gene during neurulation (Essien, 1992; Essien and Wannberg, 1993 a,b: for review, see Introduction Section II.B.b). A similar gene-environment interaction either does not occur in SELH/Bc mice, or it is not sufficient to facilitate a reduction in the incidence of NTDs in SELH/Bc mice at this dose and time of treatment. III. Studies of the Effect of Folic Acid Supplementation on the Expression of NTDs in SELH mice A. Introduction Primary prevention of neural tube defects (NTDs) has been demonstrated in humans by periconceptional supplementation with folic acid and multivitamins (occurrence: Czeizel and 41 Dudas, 1992) or folic acid alone (recurrence: M.R.C. Vitamin Study Research Group, 1991) (for review, see Introduction Section I.B). However, it is not known how folic acid prevents NTDs in humans. If a genetic model could be identified that responded to maternal supplementation with folic acid, it would provide a means to study the mechanism of prevention. The purpose of these experiments was to determine if the frequency of exencephaly in SELH embryos is reduced by periconceptional folic acid supplementation. B. Experiment #1: Supplemental folic acid added to the drinking water As a first approach, SELH mice were supplemented with the mouse equivalent of the human daily recommended dose of 4 mg per day of folic acid for women of childbearing age who have had a fetus or child with a NTD that is not attributable to the action of a teratogen (U.S. Centers for Disease Control, 1991). i. Materials and Methods a. Mice SELHA/Bc mice were obtained from our breeding colony (for strain history, see Introduction Section II.A). Mice from the seventeenth to twentieth inbred generation were used in this study. The females were nulliparous and approximately 2 to 7 months of age when bred with a mean age and standard deviation of 5 ± 2 months. Ten sires were used in the timed matings. b. Estimation of the daily water intake of SELHA/Bc females The conversion of the human oral dose of 4 mg per day of folic acid to an equivalent dose administered to mice in the drinking water ad libitum required an estimate of the daily water intake of a SELHA/Bc female per gram of body weight. This quantity was estimated from the mean water intake per gram of body weight of twelve randomly selected, singly-caged, virgin SELHA/Bc females. The mice were maintained under standard conditions (see Experimental Studies Section I) with the exception of the water bottle. A plastic 50 ml test tube covered with a layer of tinfoil and 42 brown packing tape and corked with a size six stopper and standard sipper tube replaced the regular clear glass water bottle. The test tubes were filled with 50 ml of water, and the water intake was monitored by measuring the weight of the water bottle with a top-loading digital balance every twenty-four hours [± 30 minutes] for four consecutive days. The weight of each mouse was monitored daily and recorded. The measurements made on the first day were removed from the data to minimize the potential effects of the novelty of the new water containers on the amount of water that the mice drank. As the density of water is approximately 1 g/ml at room temperature (pWater = 0.99823 g/ml at 20°C and for practical purposes changes very little with temperature, for example, pWater = 0.95838 g/ml at 100°C [Handbook of Chemistry and Physics, 1963]), the difference in weight between consecutive measurements of the water bottle was equivalent to the difference in the volume of water, reflecting the water intake of the female over a twenty-four hour period. The mean daily water intake per gram of body weight for each female was averaged and found to be 0.28 ± 0.01 ml per gram of body weight (Appendix Table A.3). c. Conversion of the human oral dose of folic acid to an equivalent dose administered to mice in the drinking water ad libitum Given that the mean daily water intake of SELHA/Bc females was estimated to be 0.28 ± 0.01 ml per gram of body weight (or 0.28 ± 0.01 L per kilogram of body weight), the recommended human oral dose of 4 mg per day of folic acid was converted to an equivalent dose administered to mice in the drinking water ad libitum as follows. Assuming a 70 kg woman, the daily recommended dose of folic acid was estimated to be approximately 0.06 mg per kilogram of body weight. The concentration of folic acid in the drinking water required to achieve this dose was calculated to be 0.2 mg/L by dividing the recommended dose of folic acid (0.06 mg/kg) by the daily water intake of a SELHA/Bc female (0.28 L/kg). d. Animal maintenance and breeding The mice were maintained under standard conditions (see Experimental Studies Section I) with the exception of the water bottle. The regular clear glass water bottles were covered with a 43 layer of tinfoil and brown packing tape to prevent the photodegradation of the folic acid in the water. Control water bottles were also covered. Mice in the folic acid supplementation group were given unacidified tap water with a 0.2 mg/L folic acid supplement. Because the addition of folic acid to the tap water was expected to promote bacterial growth, the folic acid supplemented water bottles were changed three times a week. As the mice in the control group were given the regular acidified tap water (pH 3.1), the control water bottles were changed only once per week. Females were categorized according to date of birth in sibships and sequentially assigned blind to either the folic acid supplementation group or the control group. Females in the folic acid supplementation group were supplemented ad libitum with 0.2 mg/L of folic acid in their drinking water for at least four days prior to placement with males, and supplementation continued until the pregnancy was terminated. Timed pregnancies were obtained as described in Experimental Studies Section II.B.b. The preconceptional supplementation period was 6 to 20 days with a mean and standard deviation of 11 ± 4 days. e. Folic acid Folic acid was obtained from the Sigma Chemical Company (St. Louis, Missouri) in a 25 g bottle and stored wrapped in tinfoil at room temperature. On the days that the folic acid supplemented water bottles were changed, 10 L of the folic acid stock solution was made up fresh from 0.002 g aliquots of folic acid stored at room temperature in weighing boats covered with tinfoil. f. Data analysis The effect of folic acid supplementation on the in utero mortality rate and frequency of exencephaly in SELHA/Bc mice was tested by a Chi-square test of independence at the 0.05 significance level (Zar, 1984). ii. Results Maternal supplementation of 0.20 mg/L of folic acid in the drinking water ad libitum had no detectable effect on either the in utero mortality rate or the frequency of exencephaly in SELHA/Bc 44 mice (Table 3) (Appendix Tables A.4 - A.5). The in utero mortality rate was 7% (7/103) in the control group and 8% (10/133) in the folic acid supplemented group (Table 3). The frequency of exencephaly was 15% (14/96) in the control group and 16% (20/123) in the folic acid supplemented group (Table 3). TABLE 3. Effects of folic acid supplementation (0.20 mg/L in drinking water ad libitum) on the in utero mortality rate and frequency of exencephaly in SELHA/Bc mice scored on day fourteen of gestation.1-2 Moles Exencephaly No. of No. of Meanlitter No. of Mean litter Group3 litters4 implants Number5 Percent arcsine embryos6 Number5 Percent arcsine Untreated 10 103 7 6.8 15.7 96 14 14.6 23.0 FAtreated 12 133 10 7.5 16.8 1212 20 16.3 23.9 1 Females were supplemented ad libitum with 0.20 mg/L of folic acid in the drinking water for at least four days prior to placement with males until the fourteenth day of gestation. The mean supplementation period was twenty-five days. 2 One litter was scored on D l 1. 3 UT, untreated; FA, folic acid. 4 Litters with less than five implantations (1, FA treated group) and litters from females with solid tumors (2, UT group: 1 ovarian, 1 putative pancreatic) were removed from the data. ^ Differences between the UT and FA treated groups were tested by a Chi-square test at the 0.05 significance level. 6 Embryos with odd head development are shown as superscripts. One embryo with an apical blister and one D l 1 embryo with a distinct midline cleft in the prospective cerebellum were observed in the FA treated group. One embryo with an apical blister was observed in the folic acid supplemented group (Table 3). Apical blisters occur spontaneously at a low frequency in the SELH/Bc stock from which the SELHA/Bc strain was further inbred (for strain history, see Introduction Section II.A). For a discussion of the significance of apical blisters in SELH/Bc mice, see Experimental Studies Section II.C. One embryo with a distinct midline cleft in the prospective cerebellum was observed on gestational day eleven in the folic acid supplementation group (Table 3). The litter was collected early, upon consultation with Teresa Gunn, due to the marked size of the mother given the presumed gestational date of her litter (it was assumed that an earlier copulation plug had been missed). As exencephaly can be scored on gestational day eleven, the litter data were included in this study. The embryo was given to Dr. Muriel Harris and Dr. Diana Juriloff who had described abnormal cerebella, lacking the vermis and characterized by a midline fissure, in the adult brains of ataxic SELH/Bc mice (Juriloff etal., 1993) and who were examining histological sections of embryonic cerebella at the time. Harris et al. (1994) subsequently described the midline cleft in the 45 prospective cerebellum as the embryological precursor of the abnormal cerebellar defects observed in the adult brains of ataxic SELH/Bc mice. C. Experiment #2: Supplemental folic acid added to dry food As a second approach, SELH mice were fed a Harlan Teklad chemically-defined diet with a 0,2, or 10 ppm folic acid supplement added. The Harlan Teklad base diet and level of folic acid supplementation were recommended by Dr. Virginia Diewert who had used them in her studies of folic acid supplementation (Diewert et ah, 1994 and 1995) and had demonstrated changes in serum folate levels that correlated with the level of folic acid in the diet (V.M. Diewert, personal communication). The base Harlan Teklad diet with a 2 ppm folic acid supplement was considered a normal diet. i. Methods a. Mice SELH/Bc mice were obtained from our breeding colony (for strain history, see Introduction Section II.A). Females were nulliparous and approximately 3 to 9 months of age when bred with a mean age and standard deviation of 5 ± 2 months. A total of 42 sires were used in the matings. b. Animal maintenance and breeding The mice were maintained under standard conditions (see Experimental Studies Section I) with the exception of the food. Instead of the standard ration of Purina Lab Chow, the mice were fed a Harlan Teklad chemically-defined diet with 0,2, or 10 ppm of folic acid added: TD92053 (0 ppm folic acid), TD92054 (2 ppm folic acid), and TD93013 (10 ppm folic acid). To prevent the photodegradation of the folic acid in the food, small quantities of fresh food were added to the cages three to four times a week; it was continuously available ad libitum. Females were placed on the new diet for at least two weeks prior to placement with males. The mean preconceptional supplementation period and standard deviation for females in the highest (10 ppm) folic acid 46 dietary group was 49 ± 18 days as compared to 27 ± 7 days in the 2 ppm folic acid control dietary group and 23 ± 6 days in the 0 ppm folic acid deficient dietary group. The longer times to breeding in the highest (10 ppm) folic acid dietary group was thought to be a consequence of the procedure of rotating males (see below) and attributable to the relative inexperience of the males added to this group. Females were grouped blind according to date of birth in sibships and alternately assigned to one of the three dietary groups. Timed pregnancies were obtained by placing one to two virgin females with singly-caged males after 4:30 p.m each day. Females were checked for copulation plugs before 10:00 a.m. using a blunt metal probe. Female mice with copulation plugs were housed together in groups of three or less according to their dietary groups and weighed on a digital scale to the nearest gram at 9:30 a.m. [± 45 minutes] on day zero and day fourteen of gestation. The unmated female mice remained in the cage with the male mouse and were checked for copulation plugs after 4:30 p.m. using the procedure described above. There were no females with afternoon plugs. Replacement females were set up with males during the afternoon plug check. To minimize the effect of diet on the SELH/Bc sires, males were rotated among the dietary groups as follows. Each week the males from the folic acid deficient dietary group were discarded, and new males were added to the folic acid supplemented dietary group. The remaining males were rotated from the folic acid supplemented dietary group to the folic acid control dietary group and from the folic acid control dietary group to the folic acid deficient dietary group. c. Harlan Teklad Diets TD92053, TD92054, and TD93013 • The 0 ppm (TD92053), 2 ppm (TD92054), and 10 ppm (TD93013) folic acid diets were obtained from Harlan Teklad (Madison, Wisconsin) in 25 kg quantities and stored in closed cardboard drums with closed plastic liners at room temperature. Refrigeration facilities were not available. The study was completed in 14 weeks. See Appendix Tables B. 1 to B.3 for the exact composition of the diets. 47 d. Maternal red blood cell folate determination To test that the 0 ppm, 2 ppm, and 10 ppm folic acid supplemented Harlan Teklad diets were having an effect on maternal folate status, blood was collected from the last 10 dams in each dietary group. On day 14 of gestation, females were bled and immediately killed by cervical dislocation. The uterus was removed and stored in physiological saline (0.85% NaCl) at 4 °C until the sample preparation was completed and the blood was delivered to the Department of Laboratory Medicine at the Vancouver Hospital and Health Sciences Center (U.B.C. Pavilions) for red blood cell folate determination. Blood was collected (by Dr. M.J. Harris) from the infraorbital sinus using a nonheparinized microhematocrit capillary tube to perforate the membrane and shunt the blood into an Eppendorf tube. A second capillary tube was utilized if the first tube became plugged. Capillary tubes were not reused. The blood was not sent for assay if the female had less than 5 implantations or if the sample contained a large blood clot. After all the samples were collected, the Eppendorf tubes were wrapped with parafilm and delivered at room temperature to the Laboratory Medicine receptionist for red blood cell folate determination. To prevent the blood from clotting during collection, a one third dilution of a 500 mM EDTA stock solution at pH 8.0 was used as an anticoagulant as follows (for preparation details, see Sambrook et al., 1989). The capillary tubes were submersed vertically in the 0.17 uM solution of EDTA, removed, and dragged across a paper towel to coat the inside of the tube and leave a 1 cm plug of EDTA in the tip of the capillary tube that would come in contact with the mouse. For the first two sample runs, a 20 ul drop of the 0.17 uM EDTA solution was pipetted into the tip of the Eppendorf tube to mimick the standard hospital procedure. However because one of the samples in the first sample run contained a large blood clot, the volume added to the Eppendorf tube was doubled to 40 ul of 0.17 uM EDTA solution for the remaining sample runs. To minimize blood clot formation, samples were shaken vigorously immediately after collection to dislodge the drop of EDTA in the tip of the Eppendorf tube and to facilitate mixing. 48 e. Data analysis The effect of folic acid supplementation on the in utero mortality rate and frequency of exencephaly in SELH/Bc mice was tested by a Chi-square test of independence at the 0.05 significance level (Zar, 1984). V ii . Results Variation in the level of folic acid in the Harlan Teklad diets from 0 ppm (TD92053) to 2 ppm (TD92054) to 10 ppm (TD93013) had no detectable effect on either the in utero mortality rate or the frequency of exencephaly in SELH/Bc mice (Table 4). The observed in utero mortality rate TABLE 4. Effects of folic acid supplementation (0 ppm, 2 ppm, or 10 ppm in Harlan Teklad Diet ad libitum) on the in utero mortality rate and frequency of exencephaly in SELH/Bc mice scored on day fourteen of gestation.1 Moles Exencephaly No. of No. of Mean litter No. of Mean litter Group2 litters3 implants Number4 Percent arcsine embryos5 Number4 Percent arcsine 0 ppm FA 27 295 16 5.4 14.6 279 66 23.6 30.8 2 ppm FA 281 310 19 6.1 15.6 2901 ,60 20.6 26.7 10 ppm FA 301 315 32 10.2 19.2 2821 54 19.1 26.5 1 Females were supplemented ad libitum with 0, 2, or 10 ppm of folic acid added to the Harlan Teklad base diet for at least fourteen days prior to placement with males until the fourteenth day of gestation. The mean preconceptional supplementation period and standard deviation was 23 ± 6 days for the 0 ppm FA group, 21 ±1 days for the 2 ppm FA group, and 49 ± 18 days for the 10 ppm FA group. 2 FA, folic acid. 3 Litters with less than five implantations (1, 2 ppm FA group; 1, 10 ppm FA group) were removed from the data and shown as superscripts. 4 Differences between the 0 ppm, 2 ppm, and 10 ppm FA groups were tested by a Chi-square test at the 0.05 significance level. 5 Embryos with odd head development are shown as superscripts. One embryo with an apical blister was observed in the 2 ppm and 10 ppm FA groups. was slightly higher in the 10 ppm folic acid supplemented dietary group (10%, 32/315) relative to the 0 ppm folic acid deficient dietary group (5%, 16/295) and the 2 ppm folic acid control dietary group (6%, 19/310), but this difference was not statistically significant (P=0.05 for x 2 test, P>0.05 for one-way analysis of variance; Appendix Table A.6) (Table 4). The frequency of exencephaly was 24% (66/279) in dams fed the 0 ppm folic acid deficient diet and slightly lower at 21% (60/291) in dams fed the 2 ppm folic acid control diet and 19% (54/283) in dams fed the 10 49 ppm folic acid supplemented diet, but these differences were not statistically significant (Table 4) (Appendix Table A.7). Although the frequency of exencephaly did not significantly differ between the 0, 2, and 10 ppm folic acid dietary groups, the observed frequency of exencephaly was considerably higher in the SELH/Bc dams fed the chemically-defined Harlan Teklad diet than in untreated control SELH/Bc dams fed the standard Purina Lab Chow ration in a separate study, run concurrently in the same mouse room with mice from the same litters (D.M. Juriloff, unpublished data); a difference I pursued in Experimental Studies Section IV. The level of folate in the maternal red blood cells correlated with the dietary intake of folic acid (Table 5) which suggests that the lack of response to the amount of folic acid added to the Harlan Teklad base diet was not due to the malabsorption of folic acid. Dams fed the 0 ppm folic acid deficient diet had red cell folate levels ranging from 709 nmol/L to 1098 nmol/L; whereas, dams fed the 10 ppm folic acid supplemented diet had red cell folate levels ranging from 1469 nmol/L to 5324 nmol/L (Table 5). Dams fed the 2 ppm folic acid control diet had intermediate red cell folate levels ranging from 1326 nmol/L to 2837 nmol/L (Table 5). Interassay variability was high and probably accounts for the apparent overlap in red cell folate levels between the 2 ppm and 10 ppm folic acid dietary groups. No overlap was observed among the dietary groups within an assay run (Table 5). TABLE 5. Effects of folic acid supplementation (0 ppm, 2 ppm, or 10 ppm in Harlan Teklad Diet • ad libitum) on the level of folate in the maternal red blood cells. Amount of No. of folic acid females in the diet assayed Red blood cell folate (nmol/L)1-2 0 ppm 10 709d 757a 900d, 932a, 938c, 965c, 1018c, 1033c, 1085d, 1098c 2 ppm 9 I326b, 2048d, 2213c, 2305d, 2514d, 2643d 2716d, 2837c, (4850)t 10 ppm 10 1469", 1 7 1 6 b 3005a 3 4 1 0 a 4190^ 4220c, 4232c 4612c, 4907c, 5324c 1 Red blood cell folate status was accessed by the Department of Laboratory Medicine at the Vancouver Hospital and Health Sciences Center (U.B.C. Pavilions) using the Quantaphase II® Bi 2/Folate Radioassay kit from Bio-Rad Laboratories, Ltd. (Mississauga, Ontario). The mean preconceptional supplementation period and standard deviation was 19 ± 3 days for the 0 ppm FA group, 23 ± 6 days for the 2 ppm FA group, and 60 ± 11 days for the 10 ppm group. 2 Assay run number is indicated by a superscript t Sample thawed and re-frozen for assay the following week because the hospital lab ran out of the Microbead Reagent. As freezing and thawing a sample is known to increase the folate level (Brown etal., 1990), the measurement was removed from the data. 50 One embryo with an apical blister was observed in the 2 ppm and 10 ppm folic acid dietary groups (Table 4). Apical blisters occur spontaneously at a low frequency in SELH/Bc mice. For a discussion of the significance of apical blisters in SELH/Bc mice, see Experimental Studies Section II. C. D. Discussion Although interest in the effects of folic acid in utero was initially stimulated by the teratogenic action of folic acid deficiency in experimental animals (Nelson, 1960) and in women given therapeutic abortions with the folic acid antagonist aminopteroyglutamic acid (for review, see Seller [1994] citing Thiersch, 1952), the reduction in recurrence and occurrence frequencies of NTDs in humans with folic acid supplementation is not generally believed to be due to folic acid deficiency in unsupplemented women. A deficiency model is fundamentally different from a model that states that supplementation will prevent NTDs in susceptible individuals. The former implies a purely environmental etiology of NTDs; the latter implies a multifactorial etiology of NTDs in which the expression of the underlying genetic liablity can be modified by the environment. SELH mice may represent a subgroup of the human burden of NTDs that is not responsive to folic acid supplementation. Periconceptional folic acid supplementation had no detectable effect on the in utero mortality rate or the frequency of exencephaly in SELH mice using two different dosages and two different routes of administration. Changes in maternal red blood cell folate levels correlated with the dietary intake of folic acid indicating that the lack of response was not due to the malabsorption of folic acid by SELH/Bc mice. Although variation in the level of folic acid in the Harlan Teklad diets from 0 ppm (TD92053) to 2 ppm (TD92054) to 10 ppm (TD93013) had no detectable effect on the frequency of exencephaly in SELH/Bc mice, the observed frequency of exencephaly was considerably higher in the SELH/Bc dams fed the chemically-defined Harlan Teklad diets than in SELH/Bc dams fed 51 the standard Purina Lab Chow ration in a separate study, run concurrently in the same mouse room with mice from the same litters (D.M. Juriloff, unpublished data). The only difference between the studies was the diet that the mice were fed. As the base Harlan Teklad diet did not elevate the frequency of exencephaly in strains of mice that rarely have spontaneous exencephaly (e.g., A/J, A/WySn, CD1, and CL/Fr mice: V .M. Diewert, personal communication), it appeared that Purina Lab Chow contained something that was interacting with the mutant SELH genotype to prevent NTDs. This observation needed to be confirmed in a separate experiment. IV. Study of the Effect of the Harlan Teklad chemically-defined diet (compared to Purina Lab Chow) on the Expression of NTDs in SELH/Bc mice A. Introduction The purpose of this study was to determine in a controlled experiment if the frequency of exencephaly in SELH/Bc mice fed Harlan Teklad Diet TD93013 was significantly different than the frequency of exencephaly in SELH/Bc mice fed Purina Lab Chow; that is, if the difference in the frequency of exencephaly in SELH/Bc embryos from dams fed either the Harlan Teklad chemically-defined diet (TD92053, TD92054, or TD93013) or Purina Lab Chow was repeatable. B. Materials and Methods a. Mice SELH/Bc mice were obtained from our breeding colony (for strain history, see Introduction Section II. A). Females were nulliparous and approximately 4 to 5 months of age when bred. Ten sires were used in the matings. b. Animal maintenance and breeding The mice were maintained under standard conditions (see Experimental Studies Section I) with one exception: one half of the mice were fed Harlan Tekad Diet TD93013; the other half were fed Purina Lab Chow. The mice were fed small quantities of fresh food three to four times a 52 week. The females were fed the diets ad libitum for at least two weeks prior to placement with males. Females were categorized according to date of birth in sibships and sequentially assigned blind to either the Harlan Teklad TD93013 dietary group or the Purina Lab Chow dietary group. Timed pregnancies were obtained as described in Experimental Studies Section Ill.C.i.b., with the exception that sires were not rotated between the dietary groups. c. Har lan Teklad Diet The question of how much folic acid to add to the Harlan Teklad base diet to make it comparable to Purina Lab Chow had no obvious answer. There is no evidence (see Experimental Studies Section III) that the level of folic acid has any affect on the frequency of exencephaly in SELH/Bc mice. Nevertheless, the choices were 1.) to use one of the diets used in Experimental Studies Section III.C or 2.) to try to match the level of folic acid in Purina Lab Chow. The "latest ingredient analysis" of Purina Lab Chow gave 5.9 ppm of folic acid (Appendix Table B.4); however, because the nutrient composition of the natural ingredients varies, the exact level of folic acid in any given lot of Purina Lab Chow is unknown. Thus, it was impossible to duplicate the level of folic acid in Purina Lab Chow with any degree of certainty. Consequently, it seemed preferable to use the level of folic acid in one of the diets used in Experimental Studies Section III.C. An advantage of this approach was the ability to compare the results of this study with the earlier study. The choice between 2 ppm and 10 ppm of folic acid was somewhat arbitrary, but as the folic acid estimate for the standard ration (Purina Lab Chow) in our animal unit is 5.9 ppm, it seemed wise to use the 10 ppm folic acid formulation (Harlan Teklad Diet TD93013) to ensure a sufficient amount of folic acid in the diet. A new batch of Harlan Teklad Diet TD93013 was obtained from Harlan Teklad (Madison, Wisconsin) in a 20 kg quantity and stored in a closed cardboard drum with a plastic liner at room temperature. Refrigeration facilities were not available. The study was completed in 6 weeks. 53 d. Purina Lab Chow Purina Lab Chow is regularly obtained from Jamieson's Pet Food Distributors Ltd. (Delta, British Columbia) by our animal unit. A new 50 lb. (22.6 kg) bag of Purina Lab Chow dated March 4, 1994 was opened on May 25, 1994 and used in this study. The food was dispensed into an opaque plastic bag and stored in a closed cardboard drum at room temperature. C. Results The change in maternal diet significantly affected the frequency of exencephaly in SELH/Bc mice (P<0.0001 for y} and t-tests, Appendix Table A.8): the frequency of exencephaly dropped from 21% (26/126) in the Harlan Teklad TD93013 dietary group to 3% (4/152) in the Purina Lab Chow dietary group (Table 6). The change in maternal diet had no effect on the in utero mortality rate which was 6% in both dietary groups (Table 6) (Appendix Table A.9). T A B L E 6. Effects of the Harlan Teklad chemically-defined diet (compared to Purina Lab Chow) on the in utero mortality rate and frequency of exencephaly in SELH/Bc mice scored on day fourteen of gestation.1 Moles Exencephaly No. of No. of Mean litter No. of Mean litter Group 2 litters3 implants Number5 Percent arcsine embryos6 Number5 Percent arcsine'1 PLC 13 161 9 5.6 14.8 1493 4* 2.6 11.9* HTTD93013 134 134 8 6.0 15.4 1251 26 20.6 28.2 1 Females were fed the diets ad libitum for fourteen days prior to breeding until the fourteenth day of gestation. 2 PLC, Purina Lab Chow; HT TD93013, Harlan Teklad Diet TD93013. 3 Litters with less than five implantations (3, HT TD93013 group) and litters from females with solid tumors (1, HT TD93013 group: 1 ovarian) were removed from the data and shown as superscripts. 5 Differences between the PLC and HT TD93013 groups were tested by a Chi-square test at the 0.05 significance level. 6 Embryos with odd head development are shown as superscripts. One embryo with an apical blister, one embryo with exencephaly, spina bifida, and varying degrees of 2,3 syndactly on the hindfeet, and one embryo with a closed neural tube and no eyes, no lateral ears, no tongue, no lower jaw, no medial or lateral facial prominences, vestidual maxillary prominences (with a few verbrissae on the left and three rows on the right), a missing digit on the right forefoot, and a short tail were observed in the PLC group. One embryo with an apical blister was observed in the HT TD93013 group. 7 Differences between the PLC and HT TD93013 groups were tested by an unpaired t-test at the 0.05 significance level. * P<0.0001. A few unusual embryos were observed in the Purina Lab Chow dietary group. One embryo had a closed neural tube and no eyes, no lateral ears, no tongue, no lower jaw, no medial or lateral facial prominences, vestidual maxillary prominences (with a few verbrissae on the left and three rows on the right), a missing digit on the right forefoot, and a short tail. The other 54 embryo had exencephaly, spina bifida, and varying degrees of 2,3 syndactly on the hindfeet (Table 6). These embryos may be the products of new dominant lethal mutations. In addition, two embryos with apical blisters were observed, one in each dietary group (Table 6). Apical blisters occur spontaneously at a low frequency in SELH/Bc mice. For a discussion of the significance of apical blisters in SELH/Bc mice, see Experimental Studies Section II.C. D. Discussion The principle of primary prevention of neural tube defects (NTDs) has been demonstrated in the SELH/Bc mouse model (Figure 6). The frequency of exencephaly in SELH/Bc dams fed Threshold Scale of liability (standard deviation units) Figure 6. Schematic representation of the preventative effect of the Purina Lab Chow diet on the frequency of exencephaly in SELH/Bc mice. The data in Table 6 are presented in terms of the multifactorial threshold model. The proportion of exencephalics in the Harlan Teklad TD93013 dietary group (solid line) * and the Purina Lab Chow dietary group (broken line) was used to determine where the mean of the sampled population fell on the underlying scale of liability. Shading distinguishes the Purina Lab Chow-preventable exencephaly (speckled black) from the non-Purina Lab Chow preventable exencephaly (solid black). Purina Lab Chow was seven-fold lower than in SELH/Bc dams fed a "normal" Harlan Teklad chemically-defined diet in a controlled study. This result confirms the observation made in Experimental Studies Section Ill.C.ii. The dietary response was attributable to the relative preventative effect of Purina Lab Chow based on the finding that the Harlan Teklad diet did not 55 elevate the frequency of exencephaly in strains of mice that rarely have spontaneous exencephaly (e.g., A/J, A/WySn, CD1, CL/Fr mice: V . M . Diewert, personal communication). Figure 6 is a schematic representation of the preventative effect of the Purina Lab Chow diet on the frequency of exencephaly in SELH/Bc mice. The data in Table 6 are presented in terms of the multifactorial threshold model of inheritance. The term multifactorial is used to refer to the multiple factors, both genetic and environmental, each with only a relatively small effect that contribute to the predisposition or liability to disease (Thompson etal., 1991). The model is based on the assumption that there is an underlying normally distributed liability to disease with a threshold value beyond which the liability is expressed as an abnormal phenotype (Thompson et al., 1991). All SELH/Bc embryos carry the genetic liability to exencephaly (Macdonald etal., 1989); those embryos that develop exencephaly (whether 3% on the Purina Lab Chow diet or 21% on the Harlan Teklad diet) are not genetically different from those that develop normally. Environmental and/or stochastic effects determine whether a given SELH/Bc embryo will express the genetic liability and be born with exencephaly (Macdonald etal., 1989). By this model, the change in the environment (in this case, diet) causes a shift in the liability (Figure 6). Although the frequency of exencephaly is reduced on the Purina Lab Chow diet, the liability is still not zero (Figure 6). This reflects the role of stochastic effects in the genesis of NTD, such that even in a genetically homogeneous population of embryos, the response to an environmental stimulus is not always the same (Macdonald etal., 1989). Primary prevention has only been demonstrated in four genetic models - curly tail (retinoic acid and certain inhibitors of DNA synthesis: Seller, 1994), Splotch (retinoic acid: Kapron-Bras and Trasler, 1985; Moase and Trasler, 1987), Axial defects (methionine: Essien, 1992; Essien and Wannberg, 1993 a,b), and SELH/Bc (Purina Lab Chow: Experimental Studies Section IV.C). Unlike the other genetic models, primary prevention in the SELH/Bc mice does not require intervention at a specific stage of development. A difference in diet is sufficient to mediate the preventative effect in SELH/Bc mice, demonstrating that the preventative factor(s) does not have harmful effects at other stages of development, at least at the dosage level in the Purina Lab Chow 56 diet. In contrast, the preventative agents identified in the curly tail (retinoic acid and certain inhibitors of DNA synthesis: Seller, 1994) and Splotch (retinoic acid: Kapron-Bras and Trasler, 1985; Moase and Trasler, 1987) mouse models are known to have harmful effects at other stages of development. The preventative agent identified in the axial defects mouse model (methionine: Essien, 1992; Essien and Wannberg, 1993 a,b) has not been extensively studied at other stages of development; however, preliminary data suggested that +/+ dams exposed to 70 mg/kg of methionine on day seven and day eight of gestation had a higher percentage of resorbed embryos compared to controls. The nutritional factor(s) in Purina Lab Chow mediating the reduction in the frequency of exencephaly in SELH/Bc mice has not been identified. It is presumably a normal dietary constituent because Purina Lab Chow was formulated as a standard ration for life-cycle nutrition (Appendix Table B.4). Whatever the relevant factor(s) is in Purina Lab Chow, it is either absent from the Harlan Teklad diet or present at too high or too low a level to produce the preventative effect in SELH/Bc mice and is either not required or present at sufficient levels in the Harlan Teklad diet to permit normal development of other inbred strains of mice. Because Purina Lab Chow is only guaranteed to contain not less than 23.0% crude protein and 4.5% fat and not more than 6.0% crude fiber, 8.0% ash, and 2.5% added minerals - it is not possible to make exact comparisons with the Harlan Teklad diet. Nevertheless, a crude comparison of the two diets is given in Table 7 based on the "latest ingredient analysis" information available on the Purina Lab Chow diet (Appendix Tables B.3 and B.4). The major differences between the diets are the higher levels of fat and carbohydrate in the Harlan Teklad diet and the presence of crude fiber in the Purina Lab Chow diet (Table 7). A comparison of the dietary sources of fat and carbohydrate suggests that differences may also exist in the levels of polyunsaturated and saturated fats, as well as the levels of simple and complex carbohydrates (Table 8). Variation in the dietary levels of a number of vitamins and minerals was also noted (Table 9). Such crude dietary comparisons cannot identify the preventative agent(s) in Purina Lab Chow but could provide a means to screen candidate agents. 57 TABLE 7 Dietary composition of Purina Lab Chow (PLC) and the Harlan Teklad chemically-defined diet (HT TD93013).§ Component PLC HT TD93013 Protein, % 23.4 19.90 "Vitamin-free" casein 19.70 Arginine, % 1.38 Cystine, % 0.32 Glycine, % 1.2 Histidine, % 0.55 Isoleucine, % 1.18 Leucine, % 1.7 Lysine, % 1.42 Methionine, % 0.43 0.20t Phenylalanine, % 1.03 Tyrosine, % 0.68 Threonine, % 0.91 Tryptophan, % 0.29 Valine, % 1.21 Fat, % 4.5 10.93 Soybean Oil, % 5.93 Lard, % 5.00 Cholesterol, ppm 270 Carbohydrate, % 49.0 64.81 Sucrose, % 36.41 Corn starch, % 25.00 Cellulose, % 3.40 Nitrogen-Free Extract (by difference) includes 49.0* 16.0% Neutral Detergent Fiber (approximately equal to cellulose, hemicellulose, and lignin) and 8.2% Acid Detergent Fiber (approximately equal to cellulose and lignin), % Other, % 23.1 4.36 Vitamins, % 0.52 Minerals, % 3.84 Fiber (Crude), % 5.8 Ash, % 7.3 Moisture, % 10.0* § The composition of PLC is based on the "latest ingredient analysis" information available from Jamieson's Pet Food Distributors Ltd. (Delta, British Columbia) (Appendix Table B.4). The total digestible nutrient is 76.0% and the estimated physiological fuel value is 3.30 kcal/g (Appendix Table B.4). Since the nutrient composition of the natural ingredients varies, the chemical composition will differ accordingly. The composition of HT TD93013 is based on the diet specifications available from Harlan Teklad (Madison, Wisconsin) (Appendix Table B.3). t The 0.20% constitutes only the DL-methionine supplement; to estimate the actual percentage of methionine in the HT TD93013 diet, the methionine content of casein would have to be determined, •f Estimate based on the nitrogen-free extract (Appendix Table B.4). * For the purpose of the calculations, the moisture content was assumed to be 10% (Appendix Table B.4) 58 Table 8 A comparison of the dietary sources of protein, fat, carbohydrate, and fiber in the Purina Lab Chow and the Harlan Teklad chemically-defined diet.t Diet Dietary Source Purina Lab Chowt Harlan Teklad TD93013 Protein soybean meal casein fish meal DL-methionine supplement meat meal DL-methionine supplement Fat dried whey soybean oil animal fat preserved with BHA lard Carbohydrate ground yellow com sucrose ground oats com starch cane molasses wheat germ meal Fiber dried beet pulp cellulose alfalfa meal wheat middlings t Taken from the dietary specifications in Appendix Tables B.3 and B.4. $ Many of the natural ingredients are dietary sources of more than one nutrient 59 TABLE 9 Vitamin and mineral composition of Purina Lab Chow (PLC) and the Harlan Teklad chemically-defined diet (HT TD93013),t Component |Vitamins PLC HT TD93013 Carotene, ppm Thiamin, ppm Riboflavin, ppm Niacin, ppm Pantothenic Acid, ppm Choline, ppm x 100 Folic Acid, ppm Pyridoxine, ppm Biotin, ppm Vitamin B12, mcg/kg Vitamin A, IU/g Vitamin D, IU/g Vitamin E, IU/kg Ethoxyquin (antioxidant), ppm Menadione Sodium Bisulfite Complex, ppm Minerals 4.5 15 8 95 24 22.5 5.9 6 0.07 22 15 4.5 40 9 6 55 19 17.2 10 5 0.2 20 30 33* 40 10 28 Calcium, % Phosphorus, % Potassium, % Magnesium, % Sodium, % Chlorine, % Fluorine, ppm Iron, ppm Zinc, ppm Manganese, ppm Copper, ppm Cobalt, ppm Iodine, ppm Chromium, ppm Selenium, ppm 1 0.61 1.1 0.21 0.4 0.56 35 198 70 64.3 18 0.6 0.7 1.83 0.2 0.82 0.40 0.3 0.05 0.1 0.16 35 29 58 5 0.2 2.0 0.1 The imposition of PLC is based on the "latest ingredient analysis" information available from Jamieson s Pet Food Distributors Ltd. (Delta, British Columbia) (Appendix Table B 4) Since the nutrient composition of the natural ingredients varies, the chemical composition will differ accordingly The composition of HT TD93013 is based on the diet specifications available from Harlan Teklad (Madison, Wisconsin) (Appendix Table B.3). * Vitamin D3 60 The dietary response is the first demonstration of the nutritional modification of the expression of NTDs in SELH/Bc mice, and it affords a unique opportunity to study the mechanism of prevention of exencephaly in the SELH/Bc mouse model, something that is not possible to examine directly in human studies of NTDs. V. Study of the Effect of Valproic Acid on the Expression of NTDs in Three Strains of mice with Different Genetic Liabilities to Exencephaly A. Introduction Valproic acid is an anticonvulsant drug used to treat epilepsy. Women taking valproic acid during the first trimester of pregnancy are known to have an elevated risk of a number of congenital malformations, including spina bifida (for review, see Lammer etal. [1987]). The estimated risk of a valproic acid exposed fetus having spina bifida is 1-2% based on case-control studies (Lammer etal., 1987). Valproic acid has been previously shown to cause neural tube defects (NTDs) in normal inbred strains of mice (Finnell etal., 1988), but its effect on neural tube mutants, like SELH/Bc mice, had not been studied. The purpose of this study was to explore the potential genotype-teratogen interaction of SELH/Bc mice with valproic acid as an animal model of the effect of genetic liability to NTDs on the liability to anticonvulsant induced birth defects. B. Materials and Methods i. Original Study a. Mice SELH/Bc mice were obtained from our breeding colony (for strain history, see Introduction Section II.A). Females were nulliparous and approximately 2 to 6 months of age when bred with a mean age and standard deviation of 3 ± 1 months. Fifteen sires were used in the matings. 61 ICR/Be mice were obtained from our breeding colony. The ICR/Be mouse strain is an inbred strain maintained in our laboratory by brother-sister breeding pairs. It was created in our laboratory from "random bred BLU:Ha(ICR)" mice received in 1977 from the Arbor Scientific Company Limited (Port Credit, Ontario) (Juriloff etal., 1989) and is closely related to the SELH/Bc mouse strain (for strain history, see Introduction Section II.A). Unlike SELH/Bc mice, ICR/Be mice have no unusual incidence of spontaneous malformations (e.g., Juriloff etal., 1991; Tom etal., 1991). Mice from the forty-seventh to forty-eighth inbred generation of the ICR/Be mouse strain were used in this study. Females were nulliparous and approximately 2 to 5 months of age when bred with a mean age and standard deviation of 3 ± 1 months. Thirteen sires were used in the matings. SWV/Bc mice were obtained from our breeding colony. The SWV/Bc mouse strain is an inbred strain maintained in our laboratory by brother-sister breeding pairs. It was created by Dr. J.R. Miller at U.B.C. from mice obtained from the Canadian Defense Research Board (Suffield, Alberta) in 1949 (Staats, 1980; Festing, 1989). The stock was maintained in a closed colony for ten years before inbreeding was commenced in 1959. SWV/Bc mice have been widely used in studies of development and have no unusual incidence of spontaneous malformations (e.g., Juriloff etal., 1991; Tom etal., 1991). Mice from the ninety-eighth to hundredth inbred generation of the SWV/Bc mouse stock were used in this study. Females were nulliparous and approximately 2 to 5 months of age when bred with a mean age and standard deviation of 3 ± 1 months. Twelve sires were used in the matings. The reason that the ICR/Be and SWV/Bc mouse strains were used in this study is that although both are normal strains, the morphology of the cranial neural tube closure differs (Juriloff etal, 1991). b. Animal maintenance and breeding Mice were maintained under standard conditions (see Experimental Studies Section I). They were fed Purina Lab Chow ad libitum. Timed pregnancies were obtained by placing one to four virgin females with singly-caged males of the same strain at 5 p.m. standard time (6 p.m. 62 daylight savings time) [± 45 minutes]. Females were checked for copulation plugs before 9 a.m. standard time (10 a.m. daylight savings time) using a blunt metal probe. Female mice with copulation plugs were housed together according to their strain in groups of four or less and weighed on a digital scale to the nearest gram unit at 8:30 a.m. standard time (9:30 a.m. daylight savings time) [± 45 minutes] on day zero and day fourteen of gestation. The unmated female mice remained in the cage with the male mouse and were checked for copulation plugs at 5 p.m. standard time (6 p.m. daylight savings time) [± 45 minutes] using the procedure described above. The few females with afternoon plugs were not included in the study. Replacement females were set up with males during the afternoon plug check. Within each strain, mated females were assigned in turn to one of three treatment times on the eighth day of gestation: D8h08 (8 a.m. standard time or 9 a.m. daylight savings time), D8hl2 (12 p.m. standard time or 1 p.m. daylight savings time), or D8hl6 (4 p.m. standard time or 5 p.m. daylight savings time). Within a given treatment time, females were assigned blind to one of two treatment groups: vehicle treated or valproic acid treated. c. Valproic acid The dosage of valproic acid, route of administration, and treatment time paralleled the experiment by Finnell etal. (1988) that showed that valproic acid causes NTDs in normal inbred strains of mice. Sodium valproate was obtained from the Sigma Chemical Company (St. Louis, Missouri) and stored in desiccant at room temperature until it was used. The valproic acid solution and distilled water vehicle control were prepared less than an hour before treatment. Approximately 5 ml of a 0.06 g/ml solution of valproic acid was prepared by dissolving a known amount of valproic acid powder in freshly distilled water. On day eight of gestation, females were weighed at the time of treatment and were given an intraperitoneal injection of 100 ul/10 g of body weight to administer a dose of 600 mg/kg of valproic acid or an equivalent volume of distilled water. After treatment, the females were housed according to strain and treatment group and given a clean cage and cage lid and fresh food and water. Initially, females were observed for over an 63 hour to ensure that the dose of valproic acid administered was not lethal, that the mice did not suffer, and that any effects of treatment were temporary. d. Data analysis Within each strain, all the in utero mortality data were pooled, and a Chi-square test was used to determine if there was significant heterogeneity. To test if SELH/Bc, SWV/Bc, and ICR/Be mice significantly differed in their in utero mortality rates, the pooled vehicle control values for the three strains were compared using a Chi-square test and one-way analysis of variance. A Tukey test was used to determine which pairs of strains differed significantly in their in utero mortality rates. The effect of valproic acid treatment on the frequency of exencephaly in SELH/Bc mice was examined using two methods. First, a Chi-square test was used to compare the frequency of exencephalic embryos in each valproic acid treatment time with the appropriate vehicle control. Second, an unpaired t-test was used to compare the mean litter arcsine values for each treatment group with the appropriate vehicle control. This procedure was not applicable to the exencephaly frequency data for SWV/Bc and ICR/Be mice as no exencephalic embryos were observed in the SWV/Bc vehicle control groups and only two exencephalic embryos were observed in the ICR/Be vehicle control groups. To determine if susceptibility for valproic acid induction of exencephaly varied over time of treatment, the exencephaly frequency data for the different valproic acid treatment times within a strain were compared using a Chi-square test and a one-way analysis of variance. For SWV/Bc and ICR/Be mice, the D8hl6 replicates were first compared using a Chi-square test to determine if the data from set one (Original Study) and set two (Follow-up Study) could be pooled. A Tukey test was used to determine if there was a peak susceptibility time for valproic acid induction of exencephaly. To test if SELH/Bc, SWV/Bc, and ICR/Be mice responded differently to valproic acid treatment, a Chi-square test and a one-way analysis of variance were used to compare the exencephaly frequency data for each strain at its peak susceptibility time for valproic acid induction of exencephaly. A Tukey test was used to determine, firstly, whether the mutant SELH/Bc 64 response to valproic acid treatment was significantly different than the normal strains, and, secondly, whether the response of the two normal strains - ICR/Be and SWV/Bc - was significantly different from each other. The overall increase in the frequency of exencephaly within each strain after valproic acid treatment was determined using Abbott's formula (Finney, 1971). ii . Follow-up Study The purpose of the follow-up study was to ensure that the peak of the critical period for induction of exencephaly by valproic acid in SWV/Bc and ICR/Be mice was not later than D8hl6. The methods used in the follow-up study were the same as those employed in the original study (see Experimental Studies Section V.B.i.), except that mated females of the same strain were assigned alternately to one of two valproic acid treatment times on the eighth day of gestation: D8hl6 (4 p.m. standard time or 5 p.m. daylight savings time) or D8h20 (8 p.m. standard time or 9 p.m. daylight savings time). Mice from the hundredth inbred generation of the SWV/Bc mouse stock and from the forty-eighth to forty-ninth inbred generation of the ICR/Be mouse stock were used in this study. SWV/Bc females were nulliparous and approximately 4 to 6 months of age when bred. ICR/Be females were nulliparous and approximately 2 to 6 months of age when bred with a mean age and standard deviation of 4 + 1 months. Seven SWV/Bc sires and 9 ICR/Be sires were used in the matings. C. Results Maternal administration of 600 mg/kg of valproic acid on day eight of gestation induced exencephaly in all three strains of mice - SELH/Bc, SWV/Bc, and ICR/Be (Table 10). In SELH/Bc mice, the frequency of exencephaly rose from 7-11% in the vehicle control groups to 42-69% in the valproic acid treated groups (P<0.0001 for y} and t-tests, Appendix Tables A. 10 -A. 12). In SWV/Bc mice, the frequency of exencephaly rose from 0% in the vehicle control groups to 20-42% in the valproic acid treated groups (Appendix Tables A. 13 - A. 15). In ICR/Be mice, the frequency of exencephaly rose from 0-1% in the vehicle control groups to 12-40% in the 65 TABLE 10. Effects of valproic acid treatment on the in utero mortality rate and frequency of exencephaly in SELH/Bc, SWV/Bc, and ICR/Be mice scored on day fourteen of gestation. Moles Exencephaly Genotype, No. of No. of No. of No. of litters Treatment§ Treatment time litters implants Percent embryos affected Percent^ (n) SELH/Bc WT D8h08 (set 1) 12 140 5.7 132a 7 7.6 (10) D8hl2 (set 1) 15 179 5.6 169a 11 11.2 (19) D8hl6 (set 1) 13 168 1.8 165 7 6.7 (11) VA D8h08 (set 1) 12 145 7.6 134 12 42.5* (57) D8hl2 (set 1) 17 178 5.0 169 17 69.2* (117) D8hl6 (set 1) 12 138 2.2 135 12 59.2* (80) SWV/Bc WT D8h08 (set 1) 9 126 8.7 115 0 0 (0) D8hl2 (set 1) 8 107 12.1 94 0 0 (0)i D8hl6 (set 1) 9 118 5.1 112 0 0 (0) VA D8 h08 (set 1) 8 104 8.6 95b 6 20.0 (19) D8hl2 (set 1) 8 110 6.4 103c 8 30.1 (31) D8hl6 (set 1) 8 107 11.2 95d 8 34.7 (33) (set 2) 9 124 12.1 109e 9 42.2 (46) D8h20 (set 2) 8 103 10.7 92f 8 25.0 (23) ICR/Be WT D8h08 (set 1) 8 92 16.3 778 1 1.3 (1) D8hl2 (set 1) 8 89 15.7 75 1 1.3 (1) D8hl6 (set 1) 8 84 13.1 73 0 0 (0) VA D8h08 (set 1) 8 89 14.6 76" 5 11.8 (9) D8hl2 (set 1) 8 90 12.2 79> 6 25.3 (20) D8hl6 (set 1) 8 95 24.2 72 8 40.3 (29) (set 2) 9 103 19.4 83J 9 37.3 (31) D8h20 (set 2) 10 102 20.6 81 8 19.8 (16) § WT, water (vehicle); VA, valproic acid. t Differences within a strain tested by a Chi-square test at the 0.05 significance level. The effect of treatment on the frequency of exencephaly in SELH/Bc mice was examined using two methods. First, a Chi-square test was used to compare the frequency of exencephalic embryos in each group with the appropriate vehicle control at the 0.05 significance level. Second, an unpaired t-test was used to compare the mean litter ' Freeman-Tukey arcsine values for each treatment group with the appropriate vehicle control at the 0.05 significance level. This procedure was not applicable to the exencephaly frequency data for SWV/Bc and ICR/Be mice as no exencephalic embryos were observed in the SWV/Bc vehicle control groups and only two exencephalic embryos were observed in the ICR/Be vehicle control groups. a Two embryos with apical blisters. b Five embryos with exencephaly and a midfacial cleft. c Seven embryos with exencephaly and a midfacial cleft d Ten embryos with exencephaly and a midfacial cleft. e One embryo with exencephaly and a midfacial cleft. f One embryo with spina bifida, and three embryos with shrivelled tails. 8 One embryo with exencephaly and holoprosencephaly. n One embryo with an apical blister. 1 Two embryos with an apical blister, two embryos with exencephaly and a midfacial cleft, and one embryo with an isolated midfacial cleft. J One embyro with an isolated midfacial cleft. * PsO.0001. 66 valproic acid treated groups (Appendix Tables A. 16 - A. 18). The peak response to valproic acid treatment was obtained on D8hl2 of gestation for SELH/Bc mice (P< 0.025 for Tukey test, Appendix Table A. 19) and D8hl6 of gestation for SWV/Bc (P<0.05 for Tukey test, Appendix Tables A.20 - A.21) and ICR/Be mice (P<0.01 for Tukey test, Appendix Tables A.22 to A.23) though differences between the last two treatment times, four hours apart, were not large for any of the strains (Figure 7). S o D8h08 D8hl2 D8hl6 D8h20 Treatment time Figure 7. Comparison among strains of the relative change in the frequency of exencephaly after treatment with 600 mg/kg of valproic acid over a twelve hour period on day eight of gestation. The data in Table 11 are presented graphically. The replicate data points were averaged. The in utero mortality rate was not significantly different between the vehicle and valproic acid treated groups in any of the three strains - SELH/Bc, SWV/Bc, or ICR/Be (Table 10; Appendix Tables A.24 - A.26) (Figure 8). In pooled vehicle control groups, the overall in utero mortality rate was found to be 4% (21/487) for SELH/Bc, 8% (30/351) for SWV/Bc, and 15% (40/265) for ICR/Be mice. In pooled valproic acid treated groups, the overall in utero mortality rate was found to be 5% (23/461) for SELH/Bc, 10% (54/548) for SWV/Bc, and 18% (88/479) 67 § 5 t o- 4 1 r. 1 : — ' • 1 D8h08 D8hl2 D8hl6 D8h08 D8hl2 D8hl6 D8h20 D8h08 D8hl2 D8hl6 D8h20 Treatment time Figure 8. Comparison among strains of the relative change in the in utero mortality rate after treatment with 600 mg/kg of valproic acid over a twelve hour period on day eight of gestation. The data in Table 11 are presented graphically. Replicate data points are indicated by crosses; the average data point is plotted. Shaded lines distinguish the vehicle control groups (speckled lines) from the valproic acid treated groups (black lines). for ICR/Be mice. The strain difference in the vehicle control in utero mortality rate was significant between the highest and lowest strains, ICR/Be and SELH/Bc, ( P « 0 . 0 0 1 for Tukey test, Appendix Table A.27). The vehicle control in utero mortality rate of SWV/Bc mice, which was intermediate, was not significantly different from either of the other two strains (Appendix Table A.27). Four SELH/Bc embryos with isolated apical blisters were observed in the vehicle control groups (Table 10). Apical blisters occur spontaneously at a low frequency in the SELH/Bc mouse stock. As these defects do not occur spontaneously at any appreciable frequency in other strains of mice (eg., SWV/Bc and ICR/Be mice), it is likely that they are a rare manifestation of the SELH/Bc genotype. For further discussion of the significance of apical blisters in SELH/Bc mice, see Experimental Studies Section II.C. 68 TABLE 11. Mean litter arcsine percent moles ± S.E. among implants and percent exencephaly ± S.E. among scoreable embryos in SELH/Bc, SWV/Bc, and ICR/Be mice after treatment with 600 mg/kg of valproic acid or an equivalent volume of distilled water. Moles Exencephaly Genotype, No. of Mean litter Mean litter Treatment^  Treatment time litters arcsine ± S.E. arcsine^ ± S.E. SELH/Bc WT D8h08 (set 1) 12 14.8 (± 2.0) 11.1 (± 2.3) D8hl2 (set 1) 15 15.0 (± 1.7) 20.6 (± 2.2) D8hl6 (set 1) 13 10.5 (± 1.5) 15.5 (± 2.4) VA D8h08 (set 1) . 12 15.9 (± 2.3) 40.0 (± 4.3) D8hl2 (set 1) 17 15.3 (± 2.2) 55.5* (± 3.6) D8hl6 (set 1) 12 10.8 (± 1.6) 52.1 (± 4.3) SWV/Bc WT D8h08 (set 1) 9 18.2 (± 2.3) 7.9 (± 0.2) D8hl2 (set 1) 8 22.0 (± 2.8) 8.3 (± 0.4) D8hl6 (set 1) 9 14.5 (± 2.2) 7.9 (± 0.1) VA D8h08 (set 1) 8 19.0 (± L8) 24.9 (± 4.7) D8hl2 (set 1) 8 16.2 (± 18) 34.4 (± 3.7) D8hl6 (set 1) 8 21.3 (± 2.3) 36.9* (± 2.5) (set 2) 9 20.4 (± 3.0) 40.5* (± 4.8) D8h20 (set 2) 8 18.6 (±3.1) • 31.1 (± 3.9) ICR/Be WT D8h08 (set 1) 8 24.5 (± 3.8) 10.8 (± 1.7) D8hl2 (set 1) 8 21.9 (± 4.2) 10.6 (± 1.3) D8hl6 (set 1) 8 21.7 (± 3.0) 9.2 (± 0.2) VA D8h08 (set 1) 8 23.8 (± 2.6) 20.4 (± 3.5) D8hl2 (set 1) 8 20.4 (± 3.5) 28.8 (± 5.0) D8hl6 (set 1) 8 30.4 (± 2.8) 40.3* (± 5.7) (set 2) 9 26.6 (± 3.4) 39.3* (± 3.8) D8h20 (set 2) 10 28.7 (± 3.5) 28.4 (± 3.7) § WT, water (vehicle); VA ; valproic acid. t The effect of treatment time on the frequency of exencephaly was examined as follows. A Chi-square test was used to compare the frequency data for each treatment time (see Table 10) within a treatment group. If the Chi-square test rejected the null hypothesis at the 0.05 significance level, a one-way analysis of variance was used to compare the mean litter Freeman-Tukey arcsine values for each treatment time. If the one-way analysis of variance rejected the null hypothesis at the 0.05 significance level, a Tukey test was used to determine if there was a peak susceptibility time. This procedure was not applicable to the exencephaly frequency data for the SW V/Bc and ICR/Be vehicle control groups as no exencephalic embryos were observed in the SWV/Bc vehicle control groups and only two exencephalic embryos were observed in the ICR/Be vehicle control groups. For the SWV/Bc and ICR/Be valproic acid treated groups, the D8hl6 replicates were first compared using a Chi-square test to determine if the data from set one and set two could be pooled. * Peak susceptibility time for valproic acid induction of exencephaly. 69 Although the most common defect caused by maternal administration of 600 mg/kg of valproic acid in SELH/Bc, SWV/Bc, and ICR/Be mice was exencephaly (Figure 9[a,b,c]), some other malformations were observed in a small proportion of treated embryos (Table 10). Twenty-three of the 494 valproic acid treated SWV/Bc embryos (4.7%) had faces that were split completely along the midline, in addition to, exencephaly (Figure 9[h]). Midfacial clefts were also observed in four of the 391 valproic acid treated ICR/Be embryos (1.0%), but in two of these embryos the neural tube had closed over the cranial region (Figure 9[g]). In contrast, no midfacial clefts were observed in any of the 438 valproic acid treated SELH/Bc embryos. The lower jaws were normal in the embryos with midfacial clefts. Another malformation characterized by a distinct gap between the lobes of the prosencephalon was only observed in valproic acid treated embryos of the SELH/Bc and ICR/Be strains. Most of the embryos identified with this malformation were ascertained by the presence of a large fluid-filled blister between the eyes of affected embryos (Figure 10). Scoring this malformation was difficult, and estimates of the frequency varied considerably depending on whether unfixed or fixed specimens were scored; consequently, it will not be discussed further. Three embryos with apical blisters were observed in the ICR/Be valproic acid treated groups. One embryo with spina bifida and three embryos with shrivelled tails were observed in the SWV/Bc D8h20 valproic acid treated group. No cleft lip was observed. The response of the mutant SELH/Bc mice to valproic acid treatment was compared to the normal inbred strains SWV/Bc and ICR/Be which rarely have spontaneous exencephaly. At its peak response time SELH/Bc mice had 69% (117/169) exencephaly; whereas, both SWV/Bc and ICR/Be mice had 39% (79/204 and 60/155) exencephaly (Table 10). The frequency of exencephaly in SELH/Bc mice at the peak response time was significantly higher than SWV/Bc or ICR/Be mice (P<0.005 for the Tukey test, Appendix Table A.28) which were not different from each other. If the spontaneous frequency of exencephaly in SELH/Bc mice (8.5% [40/466], estimated from the average frequency of exencephaly in vehicle controls [Table 10]) is removed by * Abbott's formula (Finney, 1971) - P= " , where P is the induced response rate, P* is the observed response rate, and C is the spontaneous response rate - then the frequency of induced 70 d. ICR/Be e. SWV/Bc f. SELH/Bc Figure 9. Valproic acid-induced exencephaly and midfacial clefts in ICR/Be (a,g), SWV/Bc (b, h), and SELH/Bc (c) mice. The strain differences in the susceptibility to valproic acid-induced midfacial clefts appeared to be related to the normal developmental pattern of cranial neural tube closure (d,e,f). Legend: 'P' = prosencephalon, ' M ' = mesencephalon, 'R' = rhombencephalon, '2' = Closure 2, '3' = Closure 3. Arrows indicate direction of neural tube closure. 71 Figure 10. A valproic acid-induced malformation observed only in SELH/Bc (a) and ICR/Be (b) mice characterized by a distinct gap between the lobes of the prosencephalon. Most of the embryos identified with this malformation were ascertained by the presence of a large fluid-filled blister (arrow) between the eyes of affected embryos. 72 exencephaly would be 66% which is still clearly higher than the induced frequency of exencephaly in SWV/Bc (39%) and ICR/Be mice (38%, similarly corrected for 0.9% [2/225] spontaneous exencephaly). D. Discussion The purpose of this study was to compare the SELH/Bc response to valproic acid to that of normal inbred strains that rarely have spontaneous exencephaly. If one observes the frequency of exencephaly after valproic acid treatment (Figure 7), SELH/Bc mice clearly have a higher frequency of exencephaly than the normal strains at all the treatment times. In fact, the frequency of exencephaly in SELH/Bc mice is almost twice as high as the normal strains at the peak response times (Table 10). This observation may have clinical significance for women with a positive family history of NTDs taking valproic acid during the first trimester of pregnancy. These women, like SELH/Bc dams, would be expected to transmit an elevated genetic liability to NTDs, and if they respond to valproic acid treatment like SELH/Bc dams, the women would be at much greater risk of having children with NTDs than women with a negative family history of NTDs. The similarity of the two normal strains in susceptibility to valproic acid induction of exencephaly was surprising. Figure 7 shows that the SWV/Bc and ICR/Be response to valproic acid treatment did not differ from each other across a twelve hour period on day eight of gestation. This observation was unexpected given the independent genetic histories of the two strains (for strain histories, see Experimental Studies Section V.B.i.a) and the previously observed differences in the morphology of closure of their cranial neural tubes (Juriloff etal., 1991). Given the known origins of the two strains, it was unlikely that the SWV/Bc and ICRVBc strains were closely related. To assess the degree of relationship between the two strains, an estimate of the proportion of typed polymorphic enzyme loci (Aco-1, Adh-3, Ahd-1, Akp-1, Alp-1, Car-2, Es-10, Ggc, Gpd-1, Gpi-1, Gpt-1, Hbb, Idh-1, Lap-1, Ldr-1, Mod-1, Mod-2, Mor-1, Mpi-1, Mup-1, Np-1, Pep-3, Pgk-2, Pgm-1, Pgm-2, Pre-1, Trf, Xld-1) that differed between the strains was calculated (based on data in Juriloff et al. [1987] and unpublished data for Car-2) and 73 compared with similar estimates of unrelated strains. The SWV/Bc and ICR/Be strains were found to share 87% (20/23) of the typed alleles. By contrast, the DBA/2J and C57BL/6J strains which are the most widely divergent strains known (Taylor, 1972) were found to share 44% (11/25) of the typed alleles (based on data in Roderick and Guidi [1990] and Staats [1980]). Although the peak response to valproic acid treatment was quantitatively very different between SELH/Bc and the normal strains, the pattern of response to valproic acid treatment over time was similar among the three strains. The differences in the frequency of exencephaly observed between consecutive treatment times after four hours were not large for any of the strains, as depicted in Figure 7 by the relatively flat lines connecting the data points for each strain. In fact, the cranial neural tube of the mouse embryo appears to be susceptible to interference by valproic acid over the entire eight (SELH/Bc) to twelve hour (SWV/Bc and ICR/Be) period studied. There are at least two reasons why the critical period for valproic acid induction of exencephaly might be so broad for SELH/Bc, SWV/Bc, and ICR/Be embryos. The real critical period for valproic acid induction of exencephaly may be very sharp, and valproic acid may only have an effect if it is present at a sufficient concentration during a relatively short developmental period. Such a hypothesis can be reconciled with the observed data only if the clearance rate of valproic acid from the maternal circulation is low, so that as long as the valproic acid is administered before the critical period in development, it can elicit a similar response across consecutive treatment times spanning several hours. However, if the clearance rate of valproic acid from the maternal circulation was low, one would expect to observe other birth defects caused by valproic acid that have later critical periods than cranial neural tube closure. As maternal administration of valproic acid did not cause a syndrome of defects, the valproic acid must have been cleared rapidly from the maternal circulation and the process of neural tube closure in the cranial region must be open to interference by valproic acid over a relatively long developmental period. This interpretation is consistent with data on the pharmakinetics of valproic acid in mice which demonstrate that valproic acid is cleared very rapidly from the maternal circulation with a half-life between 0.7 and 1.3 hours (Nau etal., 1991). 74 Midfacial clefts were the only scoreable malformation other than exencephaly and a few apical blisters observed in the D8h08, D8hl2, and D8hl6 valproic acid treated groups (Figure 9). This defect, as suggested by Tom etal. (1991), was consistent with the phenotype expected if the extreme rostral end of the neural tube (Closure 3) had not closed. SELH/Bc, SWV/Bc, and ICR/Be mice displayed widely differing susceptibility to the induction of midfacial clefts following exposure to valproic acid. This suggests that the susceptibility to the induction of midfacial clefts, like exencephaly, has a strong underlying genetic component; however, the liability genes involved were not the same for the two traits as the opposite hierarchy of susceptibility was observed. Interestingly, midfacial clefts were not observed in isolation in SWV/Bc mice (Figure 9[h]): all twenty-three of the valproic acid treated embryos with midfacial clefts had exencephaly. In contrast, two of the four ICR/Be valproic acid treated embryos with midfacial clefts had no exencephaly (Figure 9[g]). No midfacial clefts were observed in valproic acid treated SELH/Bc mice. The strain differences in the susceptibility to valproic acid-induced midfacial clefts (SWV/Bc 4.7%, ICR/Be 1.0%, SELH/Bc 0%) appeared to be related to the normal developmental pattern of cranial neural tube closure (Figure 9[d,e,fJ). A midfacial cleft appears to involve a failure of fusion at the rostral initiation site, Closure 3 (Tom et al., 1991). In normal mouse strains, Closure 3 begins at the most rostral end of the neural tube and proceeds caudally until it meets with Closure 2 (Macdonald etal., 1989)). In SELH/Bc mice, Closure 2 is omitted and Closure 3 closes the whole of the cranial neural tube except the rhombencephalon (Macdonald et al., 1989). In SWV/Bc mice the initiation site, Closure 2, is located over the prosencephalon region and is rostrally offset compared with ICR/Be, LM/Bc, and AEJ/RkBc mice, where Closure 2 initiates at the prosencephalon-mesencephalon junction (Juriloff etal., 1991). If valproic acid tends to interfere with closure over the prosencephalon it would also interfere with Closure 2 in SWV mice and thus remove the mechanism for closure over the mesencephalon. In contrast, in IRC/Bc mice the Closure 2 site and subsequent closure over the mesencephalon would be left intact. This may account for why the association of midfacial clefts and exencephaly is not as 75 common in ICR/Be mice as it is in SWV/Bc mice. Closure 3 in SWV/Bc mice must be relatively susceptible to disruption by valproic acid. In addition, the observation that no cleft face was observed in SELH/Bc supports the hypothesis (Gunn etal., 1995) that Closure 3 is more robust in SELH/Bc mice than in normal strains. Conceptually for multifactorial threshold traits the frequency of affected individuals is considered part of a normally distributed population that straddles a threshold and, therefore, represents the proportion of the population that falls beyond a given threshold of liability (Fraser, 1976). Liability is essentially a statistical concept to account for the combination of genetic and environmental factors that predispose an individual to the trait, but the liability scale may represent a quantitative biological trait that is directly related to the risk of the multifactorial trait (Fraser, 1976). For example, in cleft palate the stage at which the palatine shelves become horizontal is continuously distributed and directly related to the risk of the defect (Fraser, 1976). Given the properties of a normal curve, the frequency of affected individuals can be used to determine where the mean of the sampled population falls on the underlying scale of liability (Fraser, 1976). The closer the distribution is to the threshold, the higher the genetic liability. The response to a teratogen can be visualized for a population by examining the shift in the mean of the treated population relative to the control population (Fraser, 1976). In order to determine if a response to a teratogen is greater in one genotype than in another, one compares the relative shift of the distributions (eg., Tom etal., 1991). This comparison for this study is shown in Figure 11 for the peak response times of the three strains. Although no spontaneous exencephaly was observed in the vehicle controls of the SWV/Bc strain, if the strain has a low non-zero risk, one would expect to observe 0 exencephalics in 321 embryos quite often. To place the SWV/Bc control distribution on scale, it was necessary to use additional data from other studies (cited by Gunn et al, 1992: 0/168, Tom etal, 1991; 2/401, Juriloff, D.M. , unpublished data; 1/116, Finnell etal, 1986; 0/79, Finnell etal, 1988) to get an overall frequency of approximately 0.3% (3/1085). Comparison of the distances that the means of the control distributions shifted upon valproic acid treatment (Figure 11) indicated that the shift of the SELH/Bc distribution was no more than the 76 SELH/Bc 1 8 6 1 / \ / / \ / / V / A / ' \ ' / / \ J J \ \ SS* \ ^^^^hk££»S: 1 1 1 I I •1 1 1 SWV/Bc ^ - K - — >WS / \ ' I \ i \ 1 / \ ' / \ ' / \ / / \ ' / \' / A / / \ / / \ / / \ / ' X y x. '1 1 "^T—• | m •\ \ \ \ \ \ S\ s„ss:::::::j\ \ \ \ \ \ \ \ 1 1 1 1 • —1 1 1 ICR/Be 2 0 9 ^ / \ / / \ i / \ ' / \ / / \ ' / \' / \ / 1 \ / / \ / / \ / ' \ / ' \ / / \ y / \ m t l 1 =1 1 • : \ \ ^ ^ \ \ s s s m mmim \ \ \ V \ \ \ V Threshold Scale of liability (standard deviation units) Figure 11. Comparison among strains of the relative change in the frequency of exencephaly after treatment with 600 mg/kg of valproic acid at the peak response time on day eight of gestation. The data in Table 10 are presented in terms of the multifactorial threshold model. The proportion of exencephalics, in pooled vehicle controls (solid line) and in the peak valproic acid treatment group (broken line), was used to determine where the mean of the sampled population fell on the underlying scale of liability. To place the SWV/Bc vehicle treated distribution on scale, it was necessary to use additional data from other studies (cited by Gunn et al., 1992) to get an overall frequency of exencephaly of approximately 0.3% (3/1085). Shading distinguishes the spontaneous exencephaly (solid black) from the valproic acid-induced exencephaly (speckled black). 77 shift of the normal strains, suggesting that the response of the mutant SELH/Bc mice to valproic acid treatment was not greater than the normal strains. This suggests that the developmental pathways impinging on valproic acid effects are not abnormal in SELH/Bc mice. Closure of the neural tube in the cranial region appears to involve a number of forces both intrinsic and extrinsic to the neural folds that promote their elevation and convergence (for review, see Introduction Section III.A). The stage at which the neural folds elevate and converge is directly related to the risk of developing exencephaly. The delayed closure of the cranial neural tube in SELH/Bc embryos relative tonormal inbred strains (eg., ICR/Be and SWV/Bc) is a good example of this phenomenon (Macdonald etal., 1989; Gunn etal., 1993). Timing of neural tube closure is important because of the involvement of extrinsic forces in neurulation. For example, the development of ventral cephalic flexure in the mouse appears to pause during the time of cephalic neurulation thereby preventing the converging neural folds from being splayed apart (Jacobson and Tarn, 1982), and the emigration of neural crest cells from mesencephalic and rostral rhombencephalic neural folds appears to be essential for the flexibility of the lateral edges allowing them to curve medially and fuse in the dorsal midline (Morriss-Kay etal., 1994). Possibly, valproic acid interferes with one or more of the intrinsic and/or extrinsic forces of neurulation to delay the closure of the cranial neural tube in all exposed embryos as retinoic acid has been shown to do (Tom etal., 1991). In embryos from normal inbred strains of mice, the delay in neural tube closure caused by valproic acid treatment is not sufficient to place the majority of embryos beyond the critical threshold for neural tube closure. Most of the valproic acid exposed embryos recover and are born without exencephaly. In SELH/Bc embryos cranial neural tube closure is already delayed and valproic acid treatment acts to further delay closure in exposed embryos, placing the majority of exposed embryos beyond the critical threshold for neural tube closure. Only a minority of valproic acid exposed SELH/Bc embryos recover and are born without exencephaly. This relationship between genotype and teratogen is referred to as an additive interaction and is schematically represented in Figure 11 by the constant shift in the mean of the control distribution relative to the valproic acid treated distribution for each strain. There is no 78 variation in valproic acid susceptibility present among the three strains. In this model, the elevated liability of SELH/Bc mice to the induction of exencephaly by valproic acid results from the shape of the normal curve and the proximity of the control distribution to the threshold (Figure 11). VI. Study of the Effect of Folic Acid Supplementation on the Expression of NTDs in SELH/Bc mice Treated with Valproic Acid A. Introduction The administration of anticonvulsant drugs, including valproic acid, depress folate levels in the blood of patients with epilepsy (Wegner and Nau, 1992 citing Hendel etal., 1984). Given that primary prevention of neural tube defects (NTDs) has been demonstrated in humans by periconceptional supplementation with folic acid and multivitamins (occurrence: Czeizel and Dudas, 1992) or folic acid alone (recurrence: M.R.C. Vitamin Study Research Group, 1991), it has been hypothesized that valproic acid may be exerting its effect on the embryo through maternal blood folate levels. Because the means were available to test this hypothesis in SELH/Bc mice while running Experimental Studies Section III.C, eight female SELH/Bc mice in the control (2 ppm folic acid) and the supplemented (10 ppm folic acid) dietary groups were treated with 600 mg/kg of valproic acid at the peak response time on day eight of gestation to determine if maternal supplementation with folic acid has a preventative effect on valproic acid-induced NTDs in SELH/Bc mice. B. Materials and Methods This study was run simultaneously with Experimental Studies Section III.C. The only deviation from the methods described in Experimental Studies Sections Ill.C.i.a. to Ill.C.i.c. was that eight SELH/Bc females from the control (2 ppm folic acid, Harlan Teklad Diet TD92054) and supplemented (10 ppm folic acid, Harlan Teklad Diet TD93013) dietary groups were treated with 600 mg/kg of valproic acid by intraperitoneal injection on D8hl2 of gestation, as described in Experimental Studies Section V.B.i.c. Females were nulliparous and approximately 3 to 8 months 79 of age when bred with a mean age and standard deviation of 6 ± 1 months. The effect of folic acid supplementation on the in utero mortality rate in valproic acid-treated SELH/Bc mice was tested by a Chi-square test of independence at the 0.05 significance level (Zar, 1984). C. Results Maternal administration of 600 mg/kg of valproic acid at the peak response time on day eight of gestation resulted in a high in utero mortality rate in both folic acid dietary groups. The dose was lethal before day twelve of gestation in 38% (35/93) and 50% (43/86) of the implants in the folic acid control and supplemented dietary groups, respectively (Table 12). The rate did not vary significantly between the groups (Appendix Table A.29). TABLE 12. Effects of folic acid supplementation on the in utero mortality rate and frequency of exencephaly in SELH/Bc mice treated with 600 mg/kg of valproic acid on D8hl2 of gestation.1 Moles Exencephaly No. of No. of Mean litter No. of Mean litter Group2 litters3 implants Number4 Percent arcsine embryos Number5 Percent arcsine 2 ppm FA 8 93 35 37.6 36.8 58 40 1 7 98.3 77.2 10 ppm FA 8 86 43 50.0 46.6 43 3 2 1 0 97.7 73.8 Females were fed the diets ad libitum for fourteen days prior to breeding until the fourteenth day of gestation. FA, folic acid. Litters were scored on day fourteen of gestation. Differences between the 2 ppm FA and 10 ppm FA groups were tested by a Chi-square test at the 0.05 significance level. Embryos with other malformations are shown as superscripts. Six embryos with no external genitalia and a deficiency of tissue between the hindlimbs, five embryos with midfacial clefts, one embryo with a midfacial cleft, no external genitalia, and a deficiency of tissue between the hindlimbs, one embryo with a midfacial cleft and a missing eye, one embryo with a deficiency of tissue on the right side of the face, a missing digit on the right fore- and hindfeet, and a shriveled tail, one embryo with no external genitalia, a deficiency of tissue between the hindlimbs, and a shriveled tail, one embryo with an apical blister over the posterior neuropore, and one embryo with a tail flexion defect were observed in the 2 ppm FA group. One embryo with a midfacial cleft and a missing eye, one embryo with a midfacial cleft, one embryo with a deficient snout, one embryo with an extra digit on the right hindfoot, one embryo with a midfacial cleft, no external genitalia, and a deficiency of tissue between the hindlimbs, one embryo with no external genitalia and a deficiency of tissue between the hindlimbs, one embryo with a small, abnormally shaped genitalia and a short tail, one embryo with no external genitalia, a deficiency of tissue between the hindlimbs, and a short tail, one embryo with no external genitalia, a deficiency of tissue between the hindlimbs, and a tail flexion defect, and one embryo with a short tail were observed in the 10 ppm FA group. All scoreable embryos on both diets developed exencephaly with the exception of one dead embryo in each dietary group. Other malformations were observed in an additional 29% (17/58) and 23% (10/43) of scoreable embryos in the folic acid control and supplemented dietary groups, respectively (Table 12). The malformations associated with exencephaly were varied, but all were severe (Table 12). Multiple malformations were observed in a number exencephalic embryos, 80 some involving different regions of the body. Only a few malformations were observed in more than one embryo within a dietary group. In the 2 ppm folic acid control group, eight embryos had no external genitalia and a deficiency of tissue between the hindlimbs, seven embryos had midfacial clefts, and three embryos had tail defects (Table 12). The same three malformations were observed in more than one embryo in the 10 ppm folic acid supplemented group. In the 10 ppm folic acid supplemented group, four embryos had no external genitalia and a deficiency of tissue between the hindlimbs, three embryos had midfacial clefts, and four embryos had tail defects (Table 12). D. Discussion The data from this experiment were not interpretable as the SELH/Bc response to valproic acid treatment was off-scale: essentially no normal embryos were observed in any of the litters examined. Valproic acid treatment was lethal in more than one third of all implants before day twelve of gestation, and any scoreable embryos had exencephaly with the exception of one dead embryo that closed its neural tube in each of the dietary groups. The response to treatment was surprising given that the valproic acid was administered at the same dosage, route of delivery, and gestational age as reported in Experimental Studies Section V.B.i.c. The only difference between this study and the one described in Experimental Studies Section V was the diet that the mice were fed: the SELH/Bc mice in this study were fed the base Harlan Teklad diet with a 2 ppm or 10 ppm folic acid supplement; whereas, the SELH/Bc mice in the other study were fed the standard ration of Purina Lab Chow. The role of the diet in modifying the frequency of exencephaly in SELH/Bc mice was suspected in Experimental Studies Section III and later demonstrated in Experimental Studies Section IV. The question was whether it was sufficient to explain the SELH/Bc response to valproic acid treatment in this study. Figure 12 is a schematic representation of the expected change in the frequency of exencephaly if the effect of diet and valproic acid treatment were additive. 'A' represents the effect of valproic acid treatment on the frequency of exencephaly in 81 SELH/Bc A / \ \ 1 ——rrr^ . in -^j | i ; ] 1 ! 1 | Threshold Scale of liability (standard deviation units) Figure 12. Schematic representation of the expected change in the frequency of exencephaly if the effect of diet and valproic acid treatment were additive. The data are presented in terms of the multifactorial threshold model. 'A' represents the effect of valproic acid treatment (solid lines) on the frequency of exencephaly in SELH/Bc mice fed Purina Lab Chow in Experimental Studies Section V (Figure 11). 'B' represents the effect of diet on the frequency of exencephaly in SELH/Bc mice fed Harlan Teklad Diet TD93013 in Experimental Studies Section IV (Figure 6). The mean of the hypothetical distribution (broken line) was determined by adding the expected effect of diet ('B') to the observed effect of valproic acid treatment ('A') in SELH/Bc mice. Shading distinguishes the expected proportion of normal embryos (solid black) from the expected proportion of exencephalic embryos (speckled black). SELH/Bc mice fed Purina Lab Chow in Experimental Studies Section V (Figure 11). 'B' represents the effect of diet on the frequency of exencephaly in SELH/Bc mice fed the base Harlan Teklad diet with a 10 ppm folic acid supplement in Experimental Studies Section IV (Figure 6). The mean of the hypothetical distribution was determined by adding the expected effect of diet ('B') to the observed effect of valproic acid treatment ('A') in SELH/Bc mice. Given the properties of a normal curve, the expected proportion of normal embryos (shaded solid black) was estimated to be 5.05% (Zar, 1984) or 2 out of 43 scoreable embryos. Although the likelihood of observing 1 (dead) normal embryo out of 43 scoreable embryos is good, if the expected proportion is 2 out of 43 - the high embryonic mortality rate is not predicted by a simple additive interaction. The valproic acid treatment and the Harlan Teklad chemically-defined diet individually had no detectable effect on the in utero mortality rate in SELH/Bc mice. In SELH/Bc dams fed Purina Lab .Chow and treated with 600 mg/kg of valproic acid of D8hl2 of gestation, the in utero mortality rate was 5% - which was not significantly 82 different from the vehicle controls (Experimental Studies Section V.C). In SELH/Bc dams fed the base Harlan Teklad diet with a 2 ppm or 10 ppm folic acid supplement and treated with 600 mg/kg of valproic acid on D8hl2 of gestation, the in utero mortality rates were 38% and 50% respectively (Table 12). The high embryonic mortality rate was not attributable to the Harlan Teklad diet as it was compared to Purina Lab Chow in Experimental Studies Section IV.C and found to have no detectable effect on the in utero mortality rate in SELH/Bc mice. Malformations other than exencephaly were observed in 29% (17/58) and 23% (10/43) of scoreable embryos from valproic acid treated SELH/Bc dams fed the base Harlan Teklad diet with a 2 ppm or 10 ppm folic acid supplement respectively (Table 12). The malformations associated with exencephaly were varied, but all were severe (Table 12). Multiple malformations were observed in a number of exencephalic embryos, some involving different regions of the body (Table 12). The marked severity and syndromic nature of many of the defects as well as the involvement of different regions of the body distinguish the malformations observed in this study from those observed in Experimental Studies Section V. Together with the high embryonic mortality rate, these observations suggest that the Harlan Teklad diet potentiates the teratogenic effects of valproic acid. It is not clear whether this phenomenon is specific to SELH/Bc mice. 83 D I S C U S S I O N Inbred strains of mice that carry neural tube defect-producing mutations each represent single etiological groups; in contrast, multiple etiological groups are thought to contribute to the human burden of neural tube defects (NTDs). Herein lies the power of the mouse model. The expression of specific liability genes can be studied independently and in the context of other liability genes and different environments to determine the relative contribution of the various genetic and environmental factors responsible for the development of NTDs. The hope is that an understanding of the basic mechanisms elucidated in mouse models will help researchers to identify the specific genes which regulate and the specific environmental factors which influence neural tube closure in humans. Once the relevant human liability genes are identified, predictive tests could be developed to determine which individuals in a population are predisposed to produce children with NTDs. An understanding of the environmental factors that influence neural tube closure could help those individuals identified as carriers to minimize their risk. The objective of this work was to use the SELH/Bc mouse model to explore the environmental modification of the expression of NTDs. SELH/Bc mice are genetically liable to the cranial neural tube closure defect, exencephaly (Juriloff etal., 1989). Like most cases of the analogous human NTD, anencephaly (Khoury etal., 1982) - the exencephaly observed in SELH/Bc mice is non-syndromic and genetically complex (Macdonald et al, 1989; Juriloff et al., 1989). The genetic liability to exencephaly is thought to be carried by every SELH/Bc embyro, and environmental and/or stochastic effects determine whether a given SELH/Bc embryo will express this liability and be born with exencephaly (Macdonald etal, 1989). The genetic complexity of the SELH/Bc mouse model makes it a potentially valuable model of the common human NTDs. The initial studies explored the role of diet on expression of NTDs in SELH/Bc mice. Supplementation of methionine to SELH/Bc mice was explored for the following reasons: a.) methionine had been shown to prevent NTDs in another NTD mutant (axial defects: Essien, 1992) 84 and b.) the addition of methionine to heterologous sera had been shown to significantly reduce the frequency of NTDs in cultured rat embryos relative to controls (Ferrari etal., 1986; Flynn etal., 1987; Coelho etal., 1989; Coelho and Klein, 1990). Experimental Studies Section II examined the effect of methionine supplementation on the expression of NTDs in SELH/Bc mice. In contrast to the reduction in open NTDs observed in axial defects mice following methionine supplementation, there was no detectable effect on the frequency of exencephaly in SELH/Bc mice. This finding is probably a reflection of the different genetic etiologies of the two mutant systems. A similar gene-environment interaction either does not occur in SELH/Bc mice, or it is not sufficient to facilitate a reduction in the incidence of NTDs in SELH/Bc mice. The effect of folic acid supplementation on the expression of NTDs in SELH/Bc mice as a potential animal model for the reduction in recurrence and occurrence rates of NTDs in women given periconceptional folic acid supplementation was examined (Experimental Studies Section III). Folic acid supplementation has no effect on the frequency of exencephaly in SELH/Bc mice. Maternal red blood cell folate levels reflected folic acid levels in the diet. Low folic acid levels did not significantly increase exencephaly frequencies. The lack of response of the SELH/Bc mouse model to maternal folic acid supplementation is consistent with the findings of other mouse mutants (curly tail: Seller, 1994; axial defects: Essien, 1992; Essien and Wannberg, 1993 a,b). It may not be surprising that no mouse model has been identified that responds to folic acid supplementation given the high level of folic acid in standard mouse diets. The standard ration in our animal unit (Purina Lab Chow) contains 5.9 ppm of folic acid (Appendix Table B.4); considering that folic acid has twice the bioavailability of natural food folates (Bailey, 1992), it is unlikely that such a high level of supplementation could be achieved in a natural environment. If one postulates that additional folic acid in the diet is required to overcome a metabolic block to meet the cellular demand during periods of elevated metabolic activity, then standard mouse diets like Purina Lab Chow would be expected to mask the genetic liability. To reveal the liability one would have to decrease the amount of folic acid in the diet to a level just sufficient for the growth and 85 development of normal strains. A study with female Swiss Webster mice indicated that 0.4 ppm of folic acid in the diet is sufficient for reproductive viability (Heid et al., 1992). Interestingly, the mean red blood cell folate of the Swiss Webster mice fed the 0.4 ppm folic acid diet (992 ± 29 nmol/L: Heid etal., 1992) was on the order of that measured in SELH/Bc mice fed the folate deficient Harlan Teklad Diet (TD92053) (Table 5 in Experimental Studies Section Ill.C.ii). The response of maternal red blood cell folate levels to dietary folic acid levels indicates that SELH/Bc mice do not have a defect in folic acid uptake. The lack of exencephaly response to the low folic acid diet and the lowered maternal red blood cell folate suggests that SELH/Bc mice do not have a rate limiting mutation in a folate metabolic pathway, as does the lack of response to high levels of folate. It is possible that the SELH/Bc liability genes represent a segment of the human population that does not respond to folic acid supplementation; only susceptible genotypes may benefit from the extra folic acid in the diet. Although SELH/Bc mice did not respond to folic acid supplementation, the observation that the frequency of exencephaly in SELH/Bc mice fed the Harlan Teklad chemically-defined diet (TD92053, TD92054, or TD93013) was consistently higher than that observed in SELH/Bc mice fed Purina Lab Chow (Experimental Studies Section III.D.ii) led to the first demonstration of the nutritional modification of the frequency of exencephaly in SELH/Bc mice (Experimental Studies Section IV.C), confirming the hypothesis that diet can mask genetic liability. The fact that Harlan Teklad Diet TD93013 is considered a normal diet (with added folate) is important. Unlike the other known genotype-environment interactions in neural tube mutants, the treatment does not have adverse effects at other times in development. If the liability genes in SELH/Bc mice have human homologs, it may be possible to prevent non-folate responsive NTDs by other dietary interventions. The Harlan Teklad diet used was of the same composition as that made by Harlan Teklad for Dr. M.C. Johnson (Dr. V . M . Diewert to Dr. M.J. Harris, personal communication). It was formulated to match as closely as possible Purina Mouse Chow in nutrients, but to be of known chemical composition and amenable to alteration of vitamin content (Dr. V . M . Diewert to Dr. M.J. 86 Harris, personal communication). It was therefore intended to be nutritionally complete. Neither the Harlan Teklad diet compared with the Purina Mouse Chow nor the supplementing levels of folate in the Harlan Teklad diet affected the frequency of spontaneous cleft lip in various strains (Dr. V . M . Diewert to Dr. D.M. Juriloff, personal communication). Based on this previous experience with the Harlan Teklad diet, the nutritional composition of the Harlan Teklad diet appeared to be complete. The seven-fold lower frequency of exencephaly in SELH/Bc mice fed Purina Lab Chow compared with the chemically-defined Harlan Teklad diet provides a foothold to study the embryonic mechanisms causing reduction of exencephaly risk. In general the mechanisms underlying environmental beneficial effects upon NTD risks are not well understood. The mechanism by which retinoic acid reduces the frequency of NTDs in Splotch homozygotes is unknown. Evidence for induced embryonic mortality of affected embryos is contradictory (Kapron-Bras and Trasler, 1985 and Moase and Trasler, 1987). The mechanism of prevention of NTDs in axial defects embryos is not known (Essien, 1992; Essien and Wannberg, 1993 a,b). The mechanism of prevention of spina bifida in curly tail homozygotes has been investigated extensively. The effect on neural tube closure appears to be indirect. Agents that delay growth of the neural tube correct an imbalance in abnormal curvature of the embryo (which holds the neural tube open) due to a defect in growth of the hindgut (Copp etal., 1988a and 1988b). By analogy, the effects of diet on the frequency of exencephaly in SELH/Bc may be mediated through an indirect effect. Another approach tested for the presence of a genotype-teratogen interaction of SELH/Bc mice with valproic acid as an animal model of the effect of genetic liability to NTDs on the liability to anticonvulsant-induced birth defects (Experimental Studies Section V). The teratogenic effect of valproic acid was found to be additive with the genetic liability, but the effect on absolute risk was non-linear and profound. Thus, the risk of valproic acid-induced exencephaly was greater in the strain with a genetic predisposition to exencephaly. If the SELH/Bc liability genes have homologs in the human population, it could mean that women taking valproic acid who have a positive family 87 history of NTDs could have a significantly higher risk of having a child with a NTD than women taking valproic acid who have a negative family history of NTDs. The surprising effect of diet on the frequency of valproic acid-induced exencephaly and embryonic mortality in SELH/Bc mice (Experimental Studies Section VI) suggests that the nutritional composition of Purina Lab Chow (compared with the Harlan Teklad diet) prevents some valproic acid-induced exencephaly in SELH/Bc mice. The nature of the preventative factor(s) and the mechanism of action are not known. It may be that this phenomenon is an indication that the Harlan Teklad diet has a borderline nutritional deficiency that is revealed by the additional developmental stress caused by valproic acid. It is not known if this response is specific to SELH/Bc mice, and in the future it will be important to test normal strains for a similar dietary effect on valproic acid-induced NTDs. If normal strains respond similarly it should be possible to identify the dietary factor (not folate) that reduces valproic acid teratogenesis and apply this knowledge to human valproic acid exposed pregnancies. 88 REFERENCES Adams, M.J., M.J. Khoury, R. Stevenson, J. Haddow, G. Knight, and R.H. Allen (1993) Midtrimester serum methylmalonic acid as a risk marker for neural tube defects. Society for epidemiologic research annual meeting 84 (abstract). Appling, D.R. (1991) Compartmentation of folate-mediated one-carbon metabolism in eukaryotes. FASEB J., 5(12):2645-2651. Auerbach, R. (1954) Analysis of the developmental effects of a lethal mutation in the house mouse. J. Exp. Zool., 127:305-329. Bailey, L.B. (1992) Evaluation of a new recommended dietary allowance for folate. J. Am. Diet. Assoc., 92(4):464. Bower, C , and F.J. Stanley (1989) Dietary folate as a risk factor for neural-tube defects: evidence from a case-control study in Western Australia. Med. J. Aust., 150:613-619. Brown, R.D., R. Jun, W. Hughes, R. Watman, B. Arnold, and H. Kronenberg (1990) Red cell folate assays: some answers to current problems with radioassay variability. Pathology, 22:85. Chanarin, I. (1986) Folate deficiency. In: Folates and Pterins, Volume 3. R.L. Blakley and V . M . Whitehead, eds. John Wiley and Sons (New York, New York) p. 88. Coelho, C.N.D., and N.W. Klein (1990) Methionine and neural tube closure in cultured rat embryos: morphological and biochemical analyses. Teratology, 42:437-451. Coelho, C.N.D., J.A. Weber, N.W. Klein, W.G. Daniels, and T.A. Hoagland (1989) Whole rat embryos require methionine for neural tube closure when cultured on cow serum. J. Nutr., 119:1716-1725. Copp, A.J., F.A. Brook, and H.J. Roberts (1988a) A cell-type-specific abnormality of cell proliferation in mutant (curly tail) mouse embryos developing spinal neural tube defects. Development, 104:285-295. Copp, A.J. , F.A. Brook, J.P. Estibeiro, A.S.W. Shum, and D.L. Cockroft (1990) The embryonic development of mammalian neural tube defects. Prog. Neurobiol., 35:363-403. Copp, A.J. , J.A. Crolla, and F.A. Brook (1988b) Prevention of spinal neural tube defects in the mouse embryo by growth retardation during neurulation. Development, 104:297-303. Cuskelly, G.J., H. McNulty, and J.M. Scott (1996) Effect of increasing dietary folate on red-cell folate: implications for prevention of neural tube defects. Lancet, 347(9002): 657-659. Czeizel, A .E . and I. Dudas (1992) Prevention of the first occurrence of neural-tube defects by periconceptional vitamin supplementation. N. Engl. J. Med., 327(26): 1832-1835. Davis, R.E., and D.J. Nicol (1988) Folic acid. Int. J. Biochem., 20(2): 133, 135-136. de Franchis, R., G. Sebastio, C. Mandato, G. Andria, and P. Mastroiacovo (1995) Spina bifida, 677T—>C mutation, and role of folate. Lancet, 346(8991/8992): 1703. 89 Diewert, V . M . Department of Clinical Dental Sciences. University of British Columbia, Vancouver. Diewert, V . M . , C.S. Tse, and B. Tait (1994) Folic acid affects cleft lip and resorption rates in mice. J. Dent. Res., 73:179, Abstract 623,1ADR Program. Diewert, V . M . , J. Sarnowski, and B. Tait (1995) Folic acid does not accelerate primary palate formation in mice. J. Dent. Res., 74:71, Abstract 480, A A D R Program. Elwood, J.M. and J.H. Elwood (1980) Spina bifida and results of treatment. In: Epidemiology of anencephalus and spina bifida. Oxford University Press (Toronto, Ontario) pp. 59-66. Elwood, J.M., J.H. Elwood, and J. Little (1992a) Classification, anatomy, and embryology. In: Epidemiology and control of neural tube defects. Oxford University Press (Oxford, England) pp. 10-24. Elwood, J.M., J. Little, and J.H. Elwood (1992b) Conclusions. In: Epidemiology and control of neural tube defects. Oxford University Press (Oxford, England) p. 757. Elwood, J.M., J. Little, and J.H. Elwood (1992c) Preface. In: Epidemiology and control of neural tube defects. Oxford University Press (Oxford, England) p. vii. Elwood, J.M., J.H. Elwood, and J. Little (1992d) Diet. In: Epidemiology and control of neural tube defects. Oxford University Press (Oxford, England) pp. 521, 523-524. Essien, F.B. (1992) Maternal methionine supplementation promotes the remediation of axial defects in Axd mouse neural tube mutants. Teratology, 45:205-212. Essien, F.B. and S.L. Wannberg (1993a) Methionine but not folinic acid or vitamin B 1 2 alters the frequency of neural tube defects in Axd mutant mice. J. Nutr., 123:27-34. Essien, F.B. and S.L. Wannberg (1993b) [Erratum.] Methionine but not folinic acid or vitamin B12 alters the frequency of neural tube defects in Axd mutant mice. J. Nutr., 123:973-974. Essien, F.B., M.B. Haviland, and A . E . Naidoff (1990) Expression of a new mutation {Axd) causing axial defects in mice correlates with maternal phenotype and age. Teratology, 42:183-194. Estibeiro, J.P., F.A. Brook, and A.J. Copp (1993) Interaction between splotch (Sp) and curly tail (ci) mouse mutants in the embryonic development of neural tube defects. Development, 119:113-121. Ferrari, D., N. Klein, R. Fogelsonger, C. Lammi-Keefe, L. Marella, R. Hillman, D. Mair, P. Olsen, and G. Gross (1986) Use of rat embryo cultures to identify amino acid absorption defects in association with human reproductive failures. Teratology, 33:84c. Festing, M.F.W. (1989) Inbred strains of mice. In: Genetic Variants and Strains of Laboratory Mouse, 2nd Ed. M.F. Lyon and A.G. Searle, eds. Oxford University Press (Oxford, England) p. 646. Finkelstein, J.D. (1990) Methionine metabolism in mammals. J. Nutr. Biochem., L228-237. 90 Finnell, R.H., G.D. Bennett, S.B. Karras, and V.K. Mohl (1988) Common hierarchies of susceptibility to the induction of neural tube defects in mouse embryos by valproic acid and its 4-propyl-4-pentenoic acid metabolite. Teratology, 38:313-320. Finnell, R.H., S.P. Moon, L.C. Abbott, J.A. Golden, and G.F. Chernoff (1986) Strain differences in heat-induced neural tube defects in mice. Teratology, 33:247-252. Finney, D.J. (1971) Probit Analysis, 3rd Ed. Cambridge University Press (New York, New York) p. 125. Flynn, T .G. , L. Friedman, T.N. Black, and N.W. Klein (1987) Methionine and iron as growth factors for rat embryos cultured on canine serum. J. Exp. Zool., 244:319-324. Fraser, F.C. (1976) The multifactorial/threshold concept - uses and misuses. Teratology, 14:267-280. Fraser, F.C. and A. Lytwyn (1981) Spectrum of anomalies in the Meckel Syndrome, or: "Maybe there is a malformation syndrome with at least one constant anomaly". Am. J. Med. Genet., 9:67-73. Geelen, J.A.G., and J. Langman (1977) Closure of the neural tube in cephalic region of the mouse embryo. Anat. Rec, 189:625-640. Golden, J.A., and G.F. Chernoff (1983) Anterior neural tube closure in the mouse: fuel for disagreement with the classical theory. Clin. Res., 31_: 127A. Golden, J.A., and J. Chernoff (1993) Intermittent pattern of neural tube closure in two strains of mice. Teratology, 47:73-80. Gunn, T .G. , D.M. Juriloff, W. Vogl, M.J. Harris, and J.E. Miller (1993) Histological study of the cranial neural folds of mice genetically liable to exencephaly. Teratology, 48:459-471. Gunn, T . M . , D.M. Juriloff, and M.J. Harris (1992) Further genetic studies of the cause of exencephaly in S E L H mice. Teratology, 45:679-686. Gunn, T . M . , D.M. Juriloff, and M.J. Harris (1995) Genetically determined absence of an initiation site of cranial neural tube closure is causally related to exencephaly in SELH/Bc mouse embryos. Teratology, 52:101-108. Hall, J.G. (1994) Introduction. In: Neural Tube Defects, Ciba Foundation Symposium 181. G. Bock and J. Marsh, eds. John Wiley & Sons (Toronto, Ontario) p. 3. Hall, J.G., J.M. Friedman, B.A. Kenna, J. Popkin, M . Jawanda, and W. Arnold (1988) Clinical, genetic, and epidemiological factors in neural tube defects. Am. J. Hum. Genet., 43:827-837. Handbook of Chemistry and Physics (1963) 44th Ed. C D . Hodgeman, ed. The Chemical Rubber Publishing Co. (Cleveland, Ohio) pp. 2191, 3273. Hansen, D.K., T.F. Grafton. (1990) Lack of attenuation of valproic acid-induced effects by folinic acid in rat embryos in vitro. Teratology, 43:575-582. Harris, M.J. Department of Medical Genetics. University of British Columbia, Vancouver. 91 Harris, M.J., D.M. Juriloff, T .M. Gunn, and J.E. Miller (1994) Development of the cerebellar defect in ataxic SELH/Bc mice. Teratology, 50:63-73. Heid, M.K. , N.D. Bills, S.H. Hinrichs, and A.J. Clifford (1992) Folate deficiency alone does not produce neural tube defects in mice. J. Nutr., 122:888-894. Hibbard, E.D., and R.W. Smithells (1965) Folic acid metabolism and human embryopathy. Lancet, LT254. Hoi, F.A. , B.C.J. Hamel, M . R A . Geurds, R.A. Mullaart, F.G. Barr, R.A. Macina, and E .C .M. Mariman. (1995) A frameshift mutation in the gene for PAX3 in a girl with spina bifida and mild signs of Waardenburg syndrome. J. Med. Genet., 32(l):52-56. Holmes, L.B. , S.G. Driscoll, and L. Atkins. (1976) Etiologic heterogeneity of neural-tube defects. N. Engl. J. Med., 294:365-369. Hui, C , and A.L . Joyner (1993) A mouse model of Greig cephalopolysyndactyly syndrome: the extra-toes mutation contains an intragenic deletion of the GU3 gene. Nature Genetics, 3:241-246. Hunter, A.G.W. (1993) Brain and spinal cord. In: Human Malformations and Related Anomalies, Volume II. R.E. Stevenson, J.G. Hall, R.M. Goodman, eds. Oxford University Press (New York, New York) pp. 116-118. Jacobson, A . G . , and P.P.L. Tarn (1982) Cephalic neurulation in the mouse embryo analysed by S E M and morphometry. Anat. Rec., 203:375-396. Johnson, D.R. (1967) Extra-toes: a new mutant gene causing multiple abnormalities in the mouse. J. Embryol. Exp. Morphol., 17(3):543, 545-561, 579. Juriloff, D.M. Department of Medical Genetics. University of British Columbia, Vancouver. Juriloff, D.M. , K.B. MacDonald, and M.J. Harris (1989) Genetic analysis of the cause of exencephaly in the SELH/Bc mouse stock. Teratology, 40:395-405. Juriloff, D.M. , M.J. Harris, C. Tom, and K.B. MacDonald (1991) Normal mouse strains differ in the site of initiation of closure of the cranial neural tube. Teratology, 44:225-233. Juriloff, D.M. , M.J. Harris, M.L. Harrod, T .M. Gunn, and J.E. Miller (1993) Ataxia and a cerebellar defect in the exencephaly-prone SELH/Bc mouse stock. Teratology, 47:333-340. Juriloff, D .M. , S. Benson, M.J. Harris, D. Koehn, V. Magel, and B. Macdonald (1987) Research News. Mouse News Letter, 77:113-114. Kalter, H. (1981) Eye defects associated with a gene for exencephaly in mice. Teratology, 23:44A. Kapron-Bras, C M . and D.G. Trasler (1985) Reduction in the frequency of neural tube defects in splotch mice by retinoic acid. Teratology, 32:87-92. Khoury, M.J., J.D. Erickson, and L . M . James (1982) Etiologic heterogeneity of neural tube defects: clues from epidemiology. Am. J. Epidemiology, 115(4):538-548. 92 Kirke, P.N., A . M . Molloy, L . E . Daly, H. Burke, D.G. Weir, and J.M. Scott (1993) Maternal plasma folate and vitamin B 1 2 are independent risk factors for neural tube defects. Q. J. Med., 86:703-708. Lammer, E.J., L . E . Sever, and G.P. Oakley, J.R. (1987) Teratogen update: valproic acid. Teratology, 35:465-473. Laurence, K . M . , N. James, M.H. Miller, G.B. Tennant, and H. Campbell (1981) Double-blind randomised controlled trial of folate treatment before conception to prevent recurrence of neural-tube defects. Br. Med. J., 282:1509-1511. Layton JR., W.M., and J.M. Smith (1977) Craniomyeloschisis: a spontaneous mutation of the rat. Teratology, 15:171-177. Lemire, R.J. (1988) Neural tube defects. JAMA, 259(4):560. Lemire, R.J., J.D. Loeser, R.W. Leech, and C A . Ellsworth (1975) Neurulation. In: Normal and Abnormal Development of the Human Nervous System. Harper and Row Publishers Inc. (Hagerstown, Maryland) pp. 64, 66-68, 159-162. Lindhout, D. (1994) Waardenburg syndrome and neural tube defects. Am. J. Med. Genet., 50(4) :387. Little, J. and J.M. Elwood (1992a) Geographical variation. In: Epidemiology and control of neural tube defects. Oxford University Press (Oxford, England) pp. 101,124. Little, J. and J.H. Elwood (1992b) Socio-economic status and occupation. In: Epidemiology and control of neural tube defects. Oxford University Press (Oxford, England) p. 458. Little, J. and J.M. Elwood (1992c) Seasonal variation. In: Epidemiology and control of neural tube defects. Oxford University Press (Oxford, England) pp. 210-211, 217, 246. Little, J. and N. Nevin (1992) Genetic models. In: Epidemiology and control of neural tube defects. Oxford University Press (Oxford, England) pp. 678-679, 688-691, 709-710. M.R.C. Vitamin Study Research Group (1991) Prevention of neural tube defects: results of the Medical Research Council Vitamin Study. Lancet, 338(8760): 131-137. Macdonald, K.B. (1988) The genetics and embryopathology of exencephaly in SELH/Bc mice. University of British Columbia, Vancouver, p.36. MacDonald, K.B. , D.M. Juriloff, and M.J. Harris (1989) Developmental study of neural tube closure in a mouse stock with a high incidence of exencephaly. Teratology, 39:195-213. Martinez-Fnas, M.L. and E. Rodriguez-Pinilla (1992) Folic acid supplementation and neural tube defects. Lancet, 340(8819): 620. Matsuda, M. (1990) Comparison of the incidence of 5-azacytidine-induced exencephaly between MT/Hokldr and Slc:ICR mice. Teratology, 41:147-154. 93 Mills, J.L., G.G. Rhoads, J.L. Simpson, G . C Cunningham, M.R. Conley, M.R. Lassman, M.E. Walden, O.R. Depp, H.J. Hoffman, and The National Institute of Child Health and Human Development Neural Tube Defects Study Group (1989) The absence of a relation between the periconceptional use of vitamins and neural-tube defects. N. Engl. J. Med., 321:430-435. Mills, J.L., J.M. McPartlin, P.N. Kirke, Y.J. Lee, M.R. Conley, D.G. Weir, and J.M. Scott (1995) Homocysteine metabolism in pregnancies complicated by neural tube defects. Lancet, 345(8943): 149-151. Mills, J.L., J.M. Scott, P.N. Kirke, J.M. McPartlin, M.R. Conley, D.G. Weir, A . M . Molloy, and Y.J. Lee (1996) Homocysteine and neural tube defects. J. Nutr., 126:756S-760S. Milunsky, A. , H. Jick, S.S. Jick, C.L. Bruell, D.S. MacLaughlin, K.J. Rothman, and W. Willett (1989) Multivitamin/folic acid supplementation in early pregnancy reduces the prevalence of neural tube defects. JAMA, 262(20):2847-2852. Moase, C.E. and D.J. Trasler (1987) Retinoic acid-induced selective mortality of Splotch-delayed mouse neural tube defect mutants. Teratology, 36:335-343. Moline, M.L. and C. Sandlin (1993) Waardenburg syndrome and meningomyelocele. Am. J. Med. Genet., 47(1): 126. Morgan JR., W.C. (1954) A new crooked tail mutation involving distinctive pleiotropism. J. Genet., 52(2):371-372. Morriss-Kay, G., H. Wood, and W. Chen. (1994) Normal neurulation in mammals. In: Neural Tube Defects, Ciba Foundation Symposium 181. G. Bock and J. Marsh, eds. John Wiley & Sons (Toronto, Ontario) pp. 51-63. Mosteller, F., and C. Youtz (1961) Tables for the Freeman-Tukey transformations for the binomial and Poisson distributions. Biometrika, 48:433-440. Nau, H , R.-S. Hauck, and K. Ehlers (1991) Valproic acid-induced neural tube defects in mouse and human: aspects of chirality, alternative drug development, pharmacokinetics and possible mechanisms. Pharmacology & Toxicology 69:310-312. Nelson, M . M . (1960) Teratogenic effects of pteroylglutamic acid deficiency in the rat. In: Congenital Malformations, Ciba Foundation Symposium #59. G.E.W. Wolstenholme, and C M . O'Connor, eds. J. and A. Churchill Ltd. (London, England) pp. 134-138. Neumann, P.E., W.N. Frankel, V.A. Letts, J.M. Coffin, A.J. Copp, and M . Bernfield (1994) Multifactorial inheritance of neural tube defects: localization of the major gene and recognition of modifiers in ct mutant mice. Nature Genetics, 6:357-362. O'Rahilly, R. and F. Muller (1994) Neurulation in the normal human embryo. In: Neural Tube Defects, Ciba Foundation Symposium 181. G. Bock and J. Marsh, eds. John Wiley & Sons (Toronto, Ontario) p. 70-77. O'Shea, K.S. (1986) Ultrastructural analysis of chromosome translocation-induced neural tube defects. Brain Res. Bull., 16:833-835. O'Shea, K.S. and M.H. Kaufman (1983) Chromosome translocation (T(2;4)lSn)-induced neural tube defects in the mouse embryo. J. Neurogenet., L29-38. 94 Park, C.T., J.H. Pruitt, and D. Bennett (1989) A mouse model for neural tube defects: the Curtailed (Tc) mutation produces spina bifida occulta in Tc/+ animals and spina bifida with meningomyelocele in Tc/t. Teratology, 39:303-312. Poelmann, R.E., M.M.T. Mentink, and J.L. van Leeuwen (1987) Axial rotation of murine embryos, a study of asymmetric mitotic activity in the neural tube of somite stages. Anat. Embrvol., 176:99-103. Putz, B., and G. Morriss-Kay (1981) Abnormal neural fold development in trisomy 12 and trisomy 14 mouse embryos. J. Embryol. Exp. Morphol., 66:141-158. Roderick, T.H. and J.N. Guidi (1990) Strain distribution of polymorphic variants. In: Genetic Variants and Strains of Laboratory Mouse, 2nd Ed. M.F. Lyon and A . G . Searle, eds. Oxford University Press (Oxford, England) pp. 678-705. Sakai, Y. (1989) Neurulation in the mouse: manner and timing of neural tube closure. Anat. Rec, 223:194-203. Sambrook, J., E.F. Fritsch, and T. Maniatis (1989) Molecular Cloning: A Laboratory Manual Book 3, 2nd Ed. Cold Spring Harbor Laboratory Press (Cold Spring Harbor, New York) p. B . l l . Schoenwolf, G.C. (1994) Formation and patterning of the avian neuraxis: one dozen hypotheses. In: Neural Tube Defects, Ciba Foundation Symposium 181. G. Bock and J. Marsh, eds. John Wiley & Sons (Toronto, Ontario) pp. 26, 34, 36. Schoenwolf, G . C , and J.L. Smith (1990) Mechanisms of neurulation: traditional viewpoint and recent advances. Development, 109(2):243-270. Scott, J.M., D.G. Weir, A. Molloy, J. McPartlin, L. Daly, and P. Kirke (1994) Folic acid metabolism and mechanisms of neural tube defects. In: Neural Tube Defects, Ciba Foundation Symposium 181. G. Bock and J. Marsh, eds. John Wiley & Sons (Toronto, Ontario) p. 182. Scott, J.M., D.G. Weir, and P.N. Kirke (1995) Folate and neural tube defects. In: Folate in Health and Disease. L.B. Bailey, ed. Marcel Dekker Inc. (New York, New York) pp. 329-331. Seller, M.J. (1994) Vitamins, folic acid and the cause and prevention of neural tube defects. In: Neural Tube Defects, Ciba Foundation Symposium 181. G. Bock and J. Marsh, eds. John Wiley & Sons (Toronto, Ontario) pp. 162-165,169. Shane, B. (1995) Folate chemistry and metabolism. In: Folate in Health and Disease. L.B. Bailey, ed. Marcel Dekker Inc. (New York, New York) pp. 3-4, 11-13. Shaw, G.M. , D. Schaffer, E .M. Velie, K. Morland, and J.A. Harris (1995) Periconceptional vitamin use, dietary folate, and the occurrence of neural tube defects. Epidemiology, 6(3):219-226. Smithells, R.W., M.J. Seller, R. Harris, D.W. Fielding, C J . Schorah, N . C Nevin, S. Sheppard, A.P. Read, S. Walker, and J. Wild (1983) Further experience of vitamin supplementation for prevention of neural tube defect recurrences. Lancet, i: 1027-1031. 95 Smithells, R.W., S. Sheppard, and C.J. Schorah (1976) Vitamin deficiencies and neural tube defects. Arch. Dis. Child., 51:944-950. Smithells, R.W., S. Sheppard, C.J. Schorah, M.J. Seller, N.C. Nevin, R. Harris, A.P. Read, and D.W. Fielding (1980) Possible prevention of neural-tube defects by periconceptional vitamin supplementation. Lancet, i:339-340. Staats, J. (1980) Standardized nomenclature for inbred strains of mice, Seventh listing. Cancer Res., 40:2103, 2115. Steegers-Theunissen, R.P., G.H. Boers, H.J. Blom, J.G. Nijhuis, C.M.G. Thomas, G.F. Borm, and T.K. Eskes (1995) Neural tube defects and elevated homocysteine levels in amniotic fluid. Am. J. Obstet. Gynecol., 172:1436-1441. Steegers-Theunissen, R.P.M., G.H.J. Boers, F.J.M. Trijbels, and T.K.A.B. Eskes (1991) Neural-tube defects and derangement of homocysteine metabolism. N. Engl. J. Med., 324:199-200. Steegers-Theunissen, R.P.M., G.H.J. Boers, F.J.M. Trijbels, J.D. Finkelstein, H.J. Blom, C.M.G. Thomas, G.F. Borm, M.G.A.J. Wouters, and T.K.A.B. Eskes (1994) Maternal hyperhomocysteinemia: a risk factor for neural-tube defects? Metabolism, 43(12): 1475-1480. Strong, L . C . , and W.F. Holander (1949) Hereditary loop-tail in the house mouse. J. Hered., 40:329-331. Sulik, K.K. and T.W. Sadler (1993) Postulated mechanisms underlying the development of neural tube defects. In: Maternal Nutrition and Pregnancy Outcome, Volume 678. C.L. Keen, A. Bendich, and C.C. Willhite, eds. The New York Academy of Sciences (New York, New York), pp. 10, 13. Tassabehji, M . , A.P. Read, V.E. Newton, R. Harris, R. Balling, P. Gruss, and T. Strachan (1992) Waardenburg's syndrome patients have mutations in the human homologue of the Pax-3 paired box gene. Nature, 355:635-636. Taylor, B.A. (1972) Genetic relationships between inbred strains of mice. J. Hered., 63:83-86. Theiler, K., and L.C. Stevens (1960) The development of rib fusions, a mutation in the house mouse. Am. J. Anat., 106:171. Thiersch, J.B. (1952) Therapeurtic abortion with folic acid antagonist, 4-aminopteroylglutamic acid (4-amino P.G.A.) administered by the oral route. Am. J. Obstet. Gynecol., 63:1298-1304. Thompson, M.W., R.R. Mclnnes, and H.F. Willard (1991) Genetics of disorders with multifactorial inheritance. In: Genetics in Medicine, 5th ed. W.B. Saunders Company (Toronto, Ontario) pp. 349-357. Tom, C , D.M. Juriloff, and M.J. Harris (1991) Studies of the effect of retinoic acid on anterior neural tube closure in mice genetically liable to exencephaly. Teratology, 43:27-40. Trotz, M . , C. Wegner, and H. Nau (1987) Valproic acid-induced neural tube defects reduction by folinic acid in the mouse. Life Sci., 41:103-110. U.S. Centers for Disease Control (1991) Use of folic acid for prevention of spina bifida and other neural tube defects - 1983-1991. M.M.W.R., 40(30):513-516. 96 Van Allen, M.I., D.K. Kalousek, G.F. Chernoff, D. Juriloff, M. Harris, B.C. McGillivray, S.-L. Yong, S. Langlois, P.M. MacLeod, D. Chitayat, J.M. Friedman, R.D. Wilson, D. McFadden, J. Pantzar, S. Ritchie, and J.G. Hall (1993) Evidence for multi-site closure of the neural tube in humans. Am. J. Med. Genet., 47:723-743. van der Put, N.M.J., R.P.M. Steegers-Theunissen, P. Frosst, F.J.M. Trijbels, T .K.A.B . Eskes, L.P. van den Heuvel, E .C .M. Mariman, M. den Heyer, R. Rozen, and H.J. Blom (1995) Mutated methylenetetrahydrofolate reductase as a risk factor for spina bifida. Lancet, 346(8982): 1070-1071. van Straaten, H.W.M., H. Blom, M.C.E. Peeters, A.M.J. Rousseau, K.J. Cole, and M.J. Seller (1995) Dietary methionine does not reduce penetrance in curly tail mice but causes a phenotype-specific decrease in embryonic growth. J. Nutr., 125:2733. van Straaten, H.W.M., J.W.M. Hekking, C. Consten, and A.J. Copp (1993) Intrinsic and extrinsic factors in the mechanism of neurulation: effect of curvature of the body axis on closure of the posterior neuropore. Development, 117:1163-1172. VanAerts, L .A .G.J .M. , H.J. Blom, R.A. Deabreu, F.J.M. Trijbels, T .K.A.B . Eskes, J.H.J.C. Peereboom-Stegeman, and J. Noordhoek (1994) Prevention of neural tube defects by and toxicity of L-homocysteine in cultured postimplantation rat embryos. Teratology, 50(5):348-360. Vogelweid, C M . , D.W. Vogt, C L . Besch-Williford, and S.E. Walker (1993) New Zealand white mice: an experimental model of exencephaly. Lab. Anim. Sci., 43(l):58-60. Wagner, C. (1995) Biochemical role of folate in cellular metabolism. In: Folate in Health and Disease. L.B. Bailey, ed. Marcel Dekker Inc. (New York, New York) pp. 23-42. Wald, N.J., and P.E. Polani (1984) Neural-tube defects and vitamins: the need for a randomized clinical trial. Br. J. Obstet. Gynaecol., 91:517-518. Wallace, M.E. , P.J. Knights, and J.R. Anderson (1978) Inheritance and morphology of exencephaly, a neonatal lethal recessive with partial penetrance, in the house mouse. Genet. Res., 32:135,144,148-149. Wegner, C. and H. Nau (1992) Alteration of embryonic folate metabolism by valproic acid during organogenesis: implications for mechanism of teratogenesis. Neurology, 42(4 Suppl. 5): 17-24. Werler, M.M. , S. Shapiro, and A.A. Mitchell (1993) Periconceptional folic acid exposure and risk of occurrent neural tube defects. JAMA, 269(10): 1257-1261. Wilcken, D.E.L. and X.L . Wang (1996) Relevance to spina bifida of mutated methylenetetrahydrofolate reductase. Lancet, 347(8997):340. Wilson, D.B., and D.P. Wyatt (1988) Closure of the posterior neuropore in the vl mutant mouse. Anat. Embryol., 178:559-563. Wilson, D.B., and D.P. Wyatt (1986) Pathogenesis of neural dysraphism in the mouse mutant vacuolated lens (vl). J. Neuropathol. Exp. Neurol., 45(l):43-55. Winter, R.M. and M. Baraitser (1991) Multiple Congenital Anomalies. Chapman and Hall Medical (London, England) pp. 256-257, 635-636. 97 Zar, J.H. (1984) Biostatistical Analysis, 2nd Ed. Prentice-Hall Inc. (Englewood Cliffs, New Jersy) pp. 61-64, 83, 162-163, 167-168,185-186, 483. 98 APPENDIX A: Data Analysis 99 h3 B O I? PJ > CN co a o-i u p CO in ON O d ON rH O ON CN in o o ON ON O in o o i n i n cu CU e 'S . o 1 is ja > S CU cu 43 co CU _> ex •c: o CD Q cd cu M 3 H s <u (U c 'So T3 CU 2 » o m ( N O CN CN ^r ' CN o d o d d S r S v o CM <N O VD h O M N VQ rH If) S •rf r-H o 00 o 3f i n <- .2 fa S W Q CU o a 3 CU g CO.3 S £ £ > £ S S 00 CU o c •n (33 > 3 a* o oo 3 CO CO (U CU H I CO O U o —^» o ^1 tS pa Cj <3 . I s i >n :m i n r > ri o c i n i r i o v D ^ ' c r ^ t in r-< N i o o i n r S r S o N o S c i S r-n r i CN CN r-S^^^^^oooo^2 — I O O O C N ^ H O O O ^ H O 3 o o o \ 3 o i n ^ o \ i ^ v q H ^ o o m ' d oS od od od CN rH t CN Q\ 00 N rH H • " t Q ^ C N ^ r H C N C N C N o o o O q VO o CO CN CO I>n § ll It 1 100 •c u <D 3 13 > i a. T3 O CM ra £; <u O -*-> <D t-i a-X5 U 1^  o O X CN CO CN V£> ON ON CN 00 CO o\ ON cs ON ON S O N •a CO a ?a J 3 O I/) o > '-3 CX •c g <S Q -a CQ H S3 ca £ CO o •*-» 00 _g •a XI VN CO ON CN OO -vT oo <n co ON 00 CN 00 O CN S ON r~~ CN CN ON *o oo oo u-i co ON o o oo r> vo 00 CO ON c o 2 > W Q CN T3 ~0 <D a CU ca e i c -a -a SJ SJ.S . | o a ca •c > <u 60 c s 3 •4—» CZ> <D £ ca o H U cu "a. o 2 « ca 2 »5 '•3 e ca ca .H a 73 w a c o 2 ^1 1 5 3 in O r> 00 (N o VO rS CN ON 00 ON CN i—i I—1 CN O rfr oo r-. ON ^ CO ON CO ON ON CN CN CN 3 ^ CN O oo oo 2 2 oqoqcor^co^ovoo'^-odcoodS^HCNONodd CN CN CN CO <-< CN S oo 1 CO O 1 CN CN i C l H O O H f O H O H o CO o CN ON 00 00 p CN o 101 T A B L E A 3 Estimation of the daily water intake of a SELHA/Bc female.§ Volume per Mean water Mouse Weight of water bottlet gram of intake per SELHA/Bc weight Start Finish Difference * mouse mouse identification no. (g) (8) (8) (8) (ml/g) (ml/g) A5720 19 95.0 90.2 4.8 0.253 19 90.2 85.5 4.7 0.247 19 85.5 80.2 5.3 0.279 0.26 A5714 28 91.0 84.6 6.4 0.229 28 84.6 77.7 6.9 0.246 28 77.7 70.7 7.0 0.250 0.24 A5709 26 90.2 81.0 9.2 0.354 25 81.0 73.4 7.6 0.304 25 73.4 65.3 8.1 0.324 0.33 A5693 24 94.2 89.3 4.9 0.204 23 89.3 83.2 6.1 0.265 23 83.2 75.4 7.8 0.339 0.27 A5682 29 90.5 82.2 8.3 0.286 29 82.2 74.8 7.4 0.255 28 74.8 66.9 7.9 0.282 0.27 A5660 27 91.1 85.0 6.1 0.226 26 85.0 77.0 8.0 0.308 26 77.0 68.2 8.8 0.338 0.29 A5671 29 92.4 86.4 6.0 0.207 29 86.4 79.4 7.0 0.241 30 79.4 74.1 5.3 0.177 0.21 A5664 26 93.2 84.1 9.1 0.350 26 84.1 75.5 8.6 0.331 27 75.5 67.5 8.0 0.296 0.33 A5649 29 91.9 85.4 6.5 0.224 28 85.4 72.0 13.4 0.479 28 72.0 61.2 10.8 0.386 0.36 A5636 28 91.0 83.3 7.7 0.275 28 83.3 76.6 6.7 0.239 28 76.6 69.3 7.3 0.261 0.26 A5630 26 91.6 85.4 6.2 0.238 26 85.4 78.7 6.7 0.258 27 78.7 71.4 7.3 0.270 0.26 A5607 27 92.9 86.8 6.1 0.226 27 86.8 80.7 6.1 0.226 27 80.7 73.9 6.8 0.252 0.23 MEAN 0.28 STANDARDERROR 0.01 § The data collected on the first night was removed to minimize the potential effects of the new feeders on the amount of water that the mice drank. f Measurements were made over a twenty-four hour period [±30 minutes]. $ Because the density of water (pwater) is approximately 1 g/ml at room temperature (pwater = 0.99823 g/ml at 20°C and pwater = 0.95838 g/ml at 100°C) (Handbook of Chemistry and Physics, 1963), the weight difference in grams of water is equivalent to the volume difference in milliliters of water. 102 00 s 3 °c T3 ca cd CD O UH • rn «2 3 <*-CD s<u > '5b cu o 6 o PQ o it: ca co 43 r co "3.2 >> S 5 P J a •a 3 -t—» co CD 43 a . a £ a 5 S a o T3 3 o . co .3 — co C O CO o ^ 43 fed c c B — o 3 O 3 cd § g-s a w ^ RH CO 3 3 £ 3 a oo a is 3 -3 fi 00 O 3 t, s I < W J PQ < co » CO CTJ Si l 43 u C N X co 3 13 > I CM < N Tj CO o ST PH I ; ri s o CO cn 0 0 i n o ice § <n i n :ON C N 0 0 - t f i n -a CO co i— 3 T3 co oo 3 CO 43 3 00 CO > 'n3 D. •c co oo CO Q PQ s CM in r - l r- r-H VO CN C N i n r - ON <n C N ^ o\ r- o vo cn ON m m S cn oS od od i n r H C N o P J Q - O T3 r^ oo o cn C N oo ON S vd oo C N 3 O CO oo oo ¥ CN cn 3 o o . § 3 ^ CO co . . „ to <TJ (Tj £-i£ a a a J -J | 2 cn tzi > & 2 2 co CO o 3 C3 "S > 13 3 CT CO 00 3 a 3 co co CO CO co CO O H U co —H 43 Cd U "a, o Z CO 3 . H §*' a CO CO 4=1 00 fi 'co 3 T3 CO 43 <r- fi o c 2 o5 '*3 3 of c3 .a S 3 « O 2 CO & 00| 3 3 C Q "S3 S §: O s CN ri ^ThMJOVOhmrHOO cJodododoNododod'^  C N r H r H r H C N VO o CO r H 0 0 r i CN r H ("OrHOOCnCNOrHTtcnO VO CN r H O O r H r H O O r H CN CN O 9 i n i n ON o i n » n T i - oo o c n i n c n o r - r-H CN 0 0 CN d os *—i CN r-H CN i n r H m i n w~) T—l r—( ^ ' o\ r> ^  oo C n O O r H O O r H O C N O Iin § II •-3 ha l l 103 t3 CD cd fi C 3 u u_ CD O •S ~ .« O CD U S <u > CD 'SO'S CD O O o CQ CZ! £ < 43 £ W GO S W • -CD -S s M CD Q S z 3 cd TJ 3 -r <s a 3 cd err1 o P -3 i ,o 43 oo u ° o g CN O o g cd .2 43 3 CD X CD CD 43 O 3 O cn CD >H i £ ! CD 3 CD CD & -4-4 CD 42 CD O 3 2-.0) 5 CM * 3 00 5a TJ .a TJ O i b >-O CD 00 S 3 > '-3 00 oa r" CD .5 ^• i IT) C < ^ PT J PQ < oa CD •4—> cd 3i 43 u TJ CD 4—4 CJ PJ •c X (D 3 *cd > • CM CN . cu p O CO p CN r-H d ;Q0 o ON 8 o o ON r-H •rf' oS CN CO T t o CN TJ CD +H cd CD -4-» ^ M TJ CD cn I CD 43 oa O oa CD > '-3 OH •c £ eg « ts Q TJ CQ J3 3 H I 3 cd S CD CD it CD 43 -4-» 00 3 • HH cn 3 !-4 S ICS cn r-i IQ CN ON Tl; ON VO CO CO r-H CN CN rH rf co ON oo r> q vq vq TT CO CO rH VO CN rH cn i n i r i o 0 \ 0 0 vd od i n CO T t Q 00 00 ON T f CO CO S r-H CO Tf 3 o 2 > W Q 3 S 3 C O 3 g « « 5 G u G E S i l j j j g 2 co oo > pt! 2 2 oa CD O 3 cd "S > cd 3 cr CD 00 3 3 oa cd cn CD CD I cn O & H U o "a. cd -4-4 o 3 cd TJ 3 CD O 2 • PH -3 4-J *H cd 3 n & 2 a* ••3 3 cd cd ^ -B S C 5a CD td & 00| 3 fi Ci Q "^3 •4HA CU S _ 1-3 CM -2 »5 ca 3^ C3 c^  c< I £ " H ca 00 VO . . . . _ . . TtobododcN ' in 'dodd CO CN CO -<t CN rH cN _ i > >n co o 5j- >n M m h O VD O Tf -5^2 00 rH CO rH rHOOOCNrHOrHCOrH 00 VO r H C O C N O O C O i n r - H r H r H r n O 00 CO i n 00 3 CO 00 rH 00 00 Tf rH CN 1> o » n o rj-oq co i n co d CN o\ if- 2 2 2i ^ 00 i> ON 00 CO O CO t H Q o VO oo CN CO CN o CN T t CO d CN VO ON Ii n j « 104 o T3 co -i—» o w . CM i<3 TD CO co co CO a 3 a-ce i u co -P ON ON i n i n o 1* 00 ON | in CN CN CN NO IT) O co rS CN r> 00 ON CN CN CN o 2 o 2 2 o 3 00 ITi ON CN CN CN CN CN ON 1 CO o\ 00 CN CN CN NO ON CN —< CO e a e , Q, CX QJ O. CX CM O CN O co c 'be 0 i_ 03 =H CO M 3 1 c 03 e co CO CO JS -*-» 00| _B 'to 3 03 T3 CO J3 O 2 S S 5-e s a CJ fe e • 1 I 5-J s i ^ ST ^ .5 Hi- &a m i n o r-00 S NO CO o NO >n i n o O N - r j - i n c N c x i i n o i n i n h ; h h o o h o n > c o o o \ o o o S S c x J ^ O N c N o S o S o r j i n c x J c N CN CM H N rHrH co M 03 a > x CO c co J3 2 2 ^ £ 2 2 £ o o m o N i n 2 2 2 ^ ~ O O C N O O O C N O O O <—1 CN o T3 CO 3 a a o CO >n ON in 3- ON in O CO o 00 00 00 00 00 0 00 o CN CN CN 3 * 5 . , in in TJ- in in r^ 'srr^oooooN.oo oSooSoooooSoooo CN ^ -H M C N ' - I C N ^ H C O C O ^H Q Q ^ 00 CO CO CO CN CN O O O 1—1 CN ^ H O ' - I O - H ' — I C O O O cNONONt^incoininoincNONONOin o q ^ O O h h O O t ^ O ^ O M r c i O O ^ O NO 00 ON 00 r- 00 --i ci CN r-ScNoooN^- ' toSoo CN CN CN CN £ 2 ON O CO •<* CO >-H O CO CN CN C N O O c o O c N - — 1 O O CN >—1 O 105 co H3 "3 32 co H ca *G ca X T3 ca CD C 'to 0 co 3 s co P H CD - C •*-> oo| 'to 3 1 T3 CO 4 3 o 2 s £ 5-1 £ §J ft, "^ 3 s £ §J ics 'S a •13 3 §•1 I 1 N O o O 5 O in in ol N O N O O N o o o in O N d ON oS 00 S CN CN >—i CN I—1 CN CN CN •st co co in r-~ in h oo *o O h o O N 00 CN S 00 S CN CO CN CN r^, CN CN CN CO CN o ON >—ICOCNOCNOCOCOCNOCN in CN in O N o O N 5 00 in O N in N O o CO in >—I CN 00 CO o o 00 CN 00 d S CO S 00 ci s CN CN I—1 CN CO I—1 9 CN 2 oo 2 2 ON2;z!^r-2-l!>n '• C N O O O ^ H ^ H ' — I ^ H O C N O O O CO 00 ca 2 in in o^  in o 1 3 a " co •4-» o T 3 co 3 e •a o e o o m 7 o H d S CN - H «> It in N O O N in o O N in op o od CN oS S od S O O O O ^ H O — I < — I O O O e o O N O < PJ < 106 co •s co H c 03 03 £ GO 3 co CO -3 * 2 CO O " co -a PQ -a E T i « I • ° • PJ ^ P J OT "2 OT O <M CO co 3 £ Q •3 .s g is e c c CX CD ~ 3 g co co 43 •4—» c o c o l-i oo . a -a * o C 03 ~ £ co g CO CX-J3 £ S3 oo c H CO O co E cx co o a CO CO CX s I*-. CL oo <+M E CX 00 cx •B ° to « 43 H .t3 •4—! e o NO < PJ J PQ < B ccj T3 T3 co co S3 co 43 O CO o (3 •c 03 O CO 03 S3 03 C^  0 s E 3 U S > 04 oT co 5lu o 0\ O N CN CO "fr ,t d vi in c x O -H fC) NO NO CN «n oS vo NO O 00 •^ r S NO ON co r> co «n O 00 o CN CN CO S E E , CX CX DJ CX (X CX] O CN O CO CO u 3 O &o P H 00 3 03 T3 co co C 03 I-. -*-> co 43 B co co > : | CO co CO Q CQ I SI CN O NO o ND CN «fr 00 oS i n od 00 CN r> r> or~ co CO 00 o >n in CN 00 CN r- NO CN CN 00 CN ON ON NO CN NO "fr i co in m CO CX 3 2 o cx 3 O c co CO & *-» CO PQ o S3 4= CA "fr CN <n in "fr 00 CN oo H h o *o h NO CN CN r-» ON t~ CN ON "fr CO CN O 00 00 ^ S "fr co NO cK >n CN CN 3 o > co Q •a a G B B oo oo CO co c 03 •c 03 > 3 co g •c u r-l CO 3 73 > CN T3 CO o CO CO -rn co 3 O " co i 43 u -a CO > CO I co £• O O 1M ON as CO 00 CN CO ON 00 in CN o oS co CN cN cN CN CN CN r-< CN oo "fr r -loo —! ON <n vo in 'CO -H ON CO CN CN CN CN NO O "fr NO NO in e s s , CX CX C U cx ix a] lO CN O CO In -4-> 0 e 'co 1 1^ O 3 H £ s CO rg UH CO 43 OX) £3 '55 3 ca T3 co 43 O 2 c s r-, r- S-lo^ cj a CJ £ a"1 e g lo -G 1^  S 41 o »n o O N oo O N in od CN CN <-H CO O N N O 3 O N 00 o 00 o N O o in 00 CN "fr N O N O •n CN N O cn O N o "fr d d r-H O N CN CO O N 00 in CN CN •n CN CO CN CN CN CN r-H CO 4—1 o r-H r-H 00 in O N "fr CN o CN 1—1 in CN £ 3 od d C N C N N O r-l <M 1 — 1 o C N C N o O N C O C O N O o O N 00 C N o 00 in o o N O O N C O N O in O N O N r—5 C N C N C O in C N O N O N C O "fr C O C O C N 00 O N r—1 I—5 C N C O in CO 00 ca cx •*-» X co a CO 3 O T 3 co 3 a *a a o CO n H n o H N o m o ^ N n o N M O <—ICO'^ COCN"frOCO'—' O I in oo 55 ^ I—1 in 00 "fr r-H CO r-l O N O N o O N O N O 8.39 o r-H 00 00 r- in in oo N O N O 00 00 N O 8.39 O N N O 00 CO O N "fr "fr O N r-H O N O N O N CN 8.39 in CN r-H CO CO CO CO CO CN CO r-H CN CN "fr 8.39 CN r ^ ^ I - S S ^ S O N O O ^ O N O N - - ^ CO CN"fr,sr"fr"frCNcN—iCNCNinOCN 108 3 X <u GO ea -D CD TJ OH Cu) O 'S o PQ CO CO CO o T3 !3 S a w w_ -5 co 3 ca TJ s c ^ co § « < ST OH CD ^ Or, <» _ rT T"? CO OH CD co ir". 3<° CO CO CD co co s OH OH O CD 43 1=1 7! CD ta, CO o * 1 8 <M 3.CD 3 _o <S 3 CD s CD OH 3 , , oa O H - O 6 OH O ca P 43 .a to .a CD TJ •O-H 3 CD O f-H w 1> <i P H J PQ < ca CD _g o CD 44 3 H 3 ca £ <D 2 3 oo| 3 oa 3 ca TJ co 43 CM O 3 o '-3 ca N o 2 CJ 0,"C3 CM - C £ I C'H 'Q ^ 5 ? *5 *-< CU ^ c N ^ p c n c n v ) v q c » v p t ^ ( X 5 < n c ^ Q ^ S , n ' o d - - i r i a\0 \ ^ S c ^ c N r H O \ T} CO Tf T]- CNCNrncNcOCOCNCNCN oo o\ 2 o> Nm ' O ^ O H H H t S c N ' f H H H O VO co 00 CN VO O CO O od »o CN Tf O0\'tH;VDCMOatri00 o\oor--o\coo\ooovovo uooNcKcKod^cNTfcNaN CNCN CN CN CO rH cN CO CN 2 oo CN ^ 0 0 2 2 ^ ^ 2 ^ ^ CN CN Tj- CN CN O CO CN Tf o CN CO rH CO 00 ca OH oa 2 <^  > d CD rH CD 43 «n oo CO rH >o CO CN CO coo»ncN<na \Oco rHV000c0OO\CN00 T? voooor-corHooco COrHVOCNCNcOCNV) TJ CO 3 3 Z3 O rH \f (SJ f-l r rj- r | (\| (T, (T) H; 3 o o 109 "ul a X cu tn Cd CD "S <<-i -o ° "5 cu 43 cu CJ PQ cu £ Z-2 P H 3w o 2 W C*H -5 co £3 cd o e CU 3 CJ cu e c § P cu z o 3 cn "H-H•a '3 s -fi £-I s l U OH (-, <U DH O HS O C 3 ^ O cu b '-3 u o eg & s" o * I s OH § CU _, —i a o-2 3 3 OH ™ ^ ° 2 to .2 ^ -a _ cd 3 cu O f_ w < W J PQ < 3 cd T3 T J CU cn 3 2 CU J3 CU CJ 3 cd "S > cn "cd 3 cd >> cd cu 3 o u cd OH cd O Z 5 cd -a cu 3 <§ cu J3 B cn CU % CJ cn CU Q PQ Si si CM S^  O rj N h ON Tf vo oo " d od " ON VO CO O O C N ON C N —< C N >/i rH C N Tf VO Tj- in CO rH C N TJ- ON C N rH VO H rH C N U O 3 O ON VO CO C N od d <n >n ON O ON ON C N co ro od d vo cd CU O UJ Q TH rH CU 3 C O cd cd fi s T J -a cd cd C fi § c-^ i3 s a cd 11 2 co co > C4 2 2 o d II 43 CH •3 o 110 •c ci 3 73 > i fN co H 00 o CO CN 73 Ci -+—» O 4i . P J 73 Ci > Ci , GO X) CO Ci •+-» a 3 a-CO t u o u o xen. o ND vi ci No WON- CO I—1 r-H r-H c o NO CO 3 l-H r-H NO CN 43 T3 U 03 43 2 3 H 03 C c <3 3 o3 3 03 73 73 Ci Ci 43 3 co Ci > o co Ci Q CQ >» Ci M 3 H i G o3 S Ci Ci Ci 43 -*-» 00 3 co 3 03 73 Ci 43 03 £3 O g o ts 03 03 N B 73 ci g c CO O «J Z 03 p in CN co in oo CN ON vi H d S ON CO '-H "fr 00 00 CN NO O —i 00 ON NO 00 CO ON "fr CO ^ vi "fr co S ^ CO <—i CN I-& Q •rt 73 Ci i3 3 j3 a , 1 1 00 00 > OH 2 2 Ci . . . . O 03 03 G 73 73 03 e 3. -c 00 G 4 1 •» Is a 4 1 ON ooovioooinNoincocNNoo NO OONDcONOcOOCNOOOCNCOin <-< S o \ d o \ H K h S h \ d o d c i CO 1 , ^ H - H \ f M ( N ( N M f r i M ( N loo ON r- £ NO ^ !2 ON 2 ON CN «0 »—IO'-HCOCN'-H-CNCOCOCN'— i O ND ON 00_ ^  ON V) ON _ . r in NO'^P"fr"frcOPcO ONSodSd'-Hodod <-i CN CN in ON in "fr o ro t r~~-00 00 S ON CN CN "t CN "fr —i O C) -H CJ ^  CO 00 r-H r - H O O O ^ > - H O O O O O O NO NO CO NO CN NO CN "fr Ol >n 111 CD - 3 CD TJ .CD o CD O O PQ E CD . 4 3 CD CD CD 1 3 E IS PQ H £ c 3 to -g H j-, S3 co £ I a >, O o «-» 3 o cr CD cd O -5 a o 35 cn e a, CD ° - T J « 3 <  B « " a CD <D CD CD o Q -° E-1 3 o o 43 u 43 cd s .2 -a L2 "9 6^ 2 £ H ^ s oo cd s "C • . 3 cd cn CD " f—I IH CT £ o o U < 3 PQ < i— TJ g DH E o o CD TJ TJ CD s cd CD TJ cd 3 •c CD O 3 cd "S > "cd 3 cr CD 00 s JD " O H E cd cn O U rH CN cn CN o '~H od iri CN ON « fee* 3 ° CN ^ K O oo o cn oo co <—i cn y E E 3 CO CD CD CD -c _ > cd U < > CD o 3 3 cd CD 82 I "8 CD o OH |PH CD 3 I d > TJ 4-> OH 4-> b b H -H V V I ft ft 11 u u ' S3 El t£) l'c» CO 112 ?< CO 3 13 > OH C N TJ Cu) -4—» co TJ Cu) > Cu) CO 4 3 O i l l Cu) •*-» Cu) = =3 CT co i -Si O o c o O N oo CN o d CN 00 ' CN vd in CN 00 CN CN i> ON S CN VO m CN O N 00 4=1 TJ o «» 47} 3 • Cu) H T J TJ Cu) 3 cd -*-» Cu) 4 3 3 CO Cu) & CD cz> Cu) Q C Q Cu) Cu) 44 3 H £ s Cu) K PH cu 43 •4—1 00 3 '53 3 Cd TJ Cu) 43 <-• cd <*- 3 2 o3 '•3 c cd cd 13 co 3 3 O 2 Cu) 3 ON co T J - rH T-J- in o in r- oo CN o in M vo O M > r in rH CN rH C O V O rH 00 O N oo wo vo in TF ON ^ vo CN vo r-T f —I 3 O P-l Q TJ TJ fu) In U. " cd cd B 3 • -d s g,s a I I | f-II co co > C H 2 2 co cu o 3 cd •§ > 13 3 cr CD 00 3 • i-H a 3 co CO cd co CD CD 5 cd CO O 6 H U CD CJ "DH DH cd o O N 00 • n i n i n i n o o o o N T t i n T t r i pOOcOOOOVOrHcOCNO'^-CN S o d d o d c o O N r H o d S S d o N N H H CNrHrH CNCN rH CO CN O O ^ V O - ^ - - O N l^ 1 0>H!» O C N r H O O r H r H O O O C N r H O O N O N O o i n o N O T i - i n o N i n o N O N T t O N ON-^roinvo-q-oocoocor-ir^ c o c o S o d S o N d o d o d o d o d o d o N CN CN rH rH CN rH rH Tt Tt'^-CNincNrHcOrHCNrHrt 00 CN C N O O r H r H r H r H O O O r H O 00 00 O o CN C O 00 I in i> r i O N O N Iiri II ll 113 o CO J 3 O > CO O -=5 3 •tn O <u -3 ~ CU ' ^  u< £ CX 73 g <o ca _c co o p. O —j CO g O ^ o CS ca <u co 69 •J -S" CO pq ca co 3 73 a> >> co o J 3 a ~ co 3 O co 2 co x-• o '-3 CO a CO 00 CO !113 "2 CO c 5 co ° o l«J CO .S CO J 3 .5 o 3 co 3 co _ O 3 CO \E3 -3 £ 3 ja <0 i _ C U O J3 3 ^-J s O .S 1 3 z £ o ^ o o 3 s .a H g i—I ccj <> U CQ < CO CO •4—» co Rj 3 CO-CO t IS u •c X co 3 ca > 3 CO i* o o d o o o in 73 CO 4—» o in ^ 00 ON o > I 5> 0 5 . o in in C N CO CO CO CO cN ^ x< S o r> in CO ca > a ca -a 73 CO CO 3 e •+—» CO 4 3 3 co CO > '•3 a, •c CO • CO CO Q CQ 3 H 3 ca a CO CO tSu CO J 3 - u * 00 G '33 3 ca 73 3 co O ca o E 2 'a ••g 3 ca ca . N u 7a w S C 'co o g Z ca O N C N O N C O O N "t C O 00 o O CO S C N « " H h O h O N 00 in N O — ; " t CO CJ O N CN <—i CN >-! N O CN N O in o in in O N C O C O O S C O O N S S N O - H CN o ca T3 3 0 •{3 •2 co Q 1 •a co o 3 ca ca c3 S a co g 3 00.3 -a « ca S S3 55 £ S S S co o 3 ca "S > 73 §1 co 00| 3 g 3 co co cn a, 3 o u. O "^3 cj 2 g 3 t/3 O N CN O N 00 CO o J CN oo in p CO S co H VO co « oo g s e I 3 CO C o > ca U <> 00 in * ' CN ^ O CN co CO 3 3 ca CO 8S •a co N co CO 'C ca > «co o O IP H H 1 73 *-o co | * 13 -> c Cu t 7 3 cj a to s s .s is E5 a 1^ 4 1 S 8 Is 55 N O "t C O N O 00 r - i C N N O ^: so oo - h h O h m O inoqori-HNOoir^ON d Q ro iri r i c i O N cN r - ' i n c N ^ - c o " t i n N O ' - i i n c o c N OO ^HCN t >CN^incN'—i H M CN 00 ^ ifl M >C h ^ -H h r . in in O N C N O N O N O O N i > i n o N i n | p p COCNcOOONOcor^r^COO CN CN C N CN CN C N - H T - H O Q ^ O C O C N O O C O ' - I C N CN C) T - H O > — I C N ^ H ^ H O O O O CN "t I o b 4—1 H 11 I % X 114 •c 0 CN X <u 3 13 > 1 D H CN X T3 CD -4—» o <^. T 3 <u CO -O so CO +-» co S3 §1 -fi u I . cu p CO D © *H H - i V o o d o 8 i n xen O q O q *u r-H rH No o rH o rH o o si o o ca 00 00 1 2 VO V D q no -a • »-H O cd > 5 ca T3 CO CO r C CO CO > • r-H •4—» O H •c o C O CO Q CQ CO H4 3 H CO e PH CO - 3 •4H 00 _c 'co 3 B CO -a CO r 3 '• a 1, rS o c % £ 2 » '•3 3 S « N is 13 jo £ £ C O ° 2 Z c3 vp wo i n 3 >n i-» wo CN CO CO CO t-H i n V D rn i n CN r -CN CN vo 00 rH 5 ^ i n CN rH 3 8 d CN 00 rH i n S CN CN i> CN CO S-H O & •p -p 3 o '-3 .2 '> co a3 T 3 ca s a 8 CO 10 CO co P 0 0 - 3 S co co > Cc, S S co CO 3 ca •c ca > 13 §1 <D 00) a 3 cz> ca CO f ca o U CZ> O H 3 O u, o 2 s s 3 CO i n i n 3] CN CN V D CN O rH CN i> CO X 00 s 2 § co -c > ca i n # * 00 ^ 0 0 V CN 3 CN CO CO fi EH .<o Q CO -a co N 'a co rfi o o. w O CO ^ 3 i i .-a > c C H do s 1 CJ C J *>* 2 rS as ^H Ol |00 rH CN i n r-H i> CO o 1> V D CN i n vd CN d V O CN vo vo 2 8 S 8 2 5 ^ CN v j i n oS S V D T J - T J - vo ^ r n O N ^ ^ v o t - ^ v o ^ ^ J ^ v o o O ^ ^ ^ CO CN CN rH CN CN rH C O O S 3 CN S CO 3 00 •rj \0 d oS i n rH CN rH i n rH rH o O N CO O A O rH 00 00 CO vo rH 00 O N rH CN CN CN rH rHrH rH^'^rHrHrHrHrH O N O N O N O N in 00 00 CO rH o <-N CO O CN ""•N rH rH i-H rH m O rHrHOCNCOOCNrHCNCNrHrHO O N rv op' o o rH rH K K rH rH PH PH -g u u * JC-115 • c u X cu 3 73 > 73 <U -u* CJ <^  . ra C J CO b CO u 2 ccj § 1 J3 U o o d iO o o >n in ON c IS o > § 5> p 3 CO in in q ON in °-T-H O 00 cj -a -3 O cj C cj 73 l> > a ctf. 73 73 cj to 3 03 •*-» C J 43 3 CO C J > '•4—» OH • c i cj CO C J Q PQ >^ C J M 3 H 3 03 6 C J C J i t C J & 00 3 'to 3 03 73 3 cj O 43 '-3 03 ^ 3 o B 2 " 03 03 73 u u to ° a £ ol 1^3 !00 o O CO o ON CN TT "fr i n <—i CO CN O 00 00 CN in co in CN ^ vo >n — i ^ M h o CN 00 S NO r~ CN CO 3 o '•0 ^ .2 2 > fa C J PJ Q 73 03 3 73 o3 3 CJ 3 T3 U ' CJ 03 73 03 on .3 S « S 3 2 00. OO > DH 2 2 s <3 cj s 5 ON in "fr in •S r - H r - H ON 00 CN r - H CN in "fr CO O r> CO CO No. r - H 1 8 CO C J C J 3 03 •c 03 > 73 §1 cj oo| 3 cj 03 GO O H O CO CX 3 O u o 73 cj a I & B B 3 00 CN 00 2 CN in o 5 co "ci-in CN r - H C J 8 00 g S 2 § 3 C J 'C O > 03 u <> b ON ^ V ^ cj o 3 e C J 5 73 3 C J ^ 'to C J |73 $ O rEV<— DH MH 73 C J 3 i t > c o o in in CN o o\ CO O O I S M CO CO O <—' ON in x r - H S co CN CN CN N o o o i n N O N o r ^ i > i - - N o m r - H in in ON in ON in in 00 o in o r - H 00 NO 00 NO 00 S 00 00 od r - H r - H CO r - H o CN co>n ' - i c o ( N O c o c o o i n c o " t c o p c N ^ — i " c r o o o o o r-l ^ H o o V V |(X cx 11 U CJ w oo I a * 116 3 TJ -a CD 03 i CD 43 3 VI CD > CD tn CD Q PQ 2 i> CN IT) oo t» <n CN co u O E PJ T J CN r- o oo 00 00 00 i> CO VO i > —i M r H 1/| h d d vo <n t-CN in CN i> O H I S O O s o '-3 2 '> CD Q •p Rj C3 s TJ TJ si C C «3B3HS CD g 00-3. 3 c S t / 3 CO > P4 2 S 03 CD CD s •c > CT CD 00 c 03 CTJ 1/1 CD OH 03 O E-U CD JO CCi a •4—» o Z CD •4—» CD l-H 3 c r 03 l u CD cd "a, •4-* O Z 3 H 3 CO g CD P H CD 4 3 -*-» 00 _g VI 3 8 d TJ CD 4 3 « H 3 o E 2 ^ cd . N u 13 <D 3 3 5H * **H C oa Z c3 1^ "5b 1, H fe o "S < s 1^ I t in H o o N i n ^ o i n p q CN cn o oo o >S vd —JrHoddvood CN cn co cN CN ^ H (N C N O C N r H l O C N CN-^f C O — l O r H O O O in ^ O CN in op in in ON ^ h o h h m D oo r-- i> oo oo oo cNin m ri m o H i/i d o o o o o o o o O N ON 00 ON in O N O N CN d 117 $ ca 7 3 7 3 CU C U 43 & C D > O cc CD Q PQ CM r**» ON vo r> ON NO in o "t o\ ^ co o C N r*» <n <—i '— 1 1— 1 C N C N in o\ r-~ in o —< ^ m CN CO O rH r-H CN CO loo d H H ^ S d o 7 3 IH nj (3 7 3 R J C 3 . o <u 3 CD P 0 0 - 3 . ca. vr3 1 I > «2 S S on (L> O £ 3 ccj •G cd > 3 cr <u 0 0 3 B 3 czi C/3 B J N-J </) <U U s 03 cn O U cu 3 ca o 11 9-o Z (U •4—» CU ca 3 cr O l I 43 u cu ca o " D , ca o Z 3 H 3 ca B cu cu cu vS 0 0 _ 3 CO 3 -B ca 7 3 3 cu o '-3 ca u. 3 O C % £ 2 »5 ca c3 . N £ 73 JJ s C '35 o o Z ca 73 <3 cj £ r*t Cj 1 8 v. ^ 41 * - i ca 88$ CN in r S 0 0 o> ON in oo ON ON " f r r -H C N C N C O "fr co in co "fr C N "fr o o ON r - H o ON ON o in in C O in C N C N C N CO»n C N "fr *0 C N C N CN "fr in in ON in CN CN co o co r--oo ON S d od oo S "fr in C N o in C N Q , <n ^ C N H co u ' r - H 1 r - H r - H r - H O O O O O O O O CO 118 >H cu •a '53 43 CU .-3 -3 CU « n *H U O 3 •tt o <u CU . n O 2 2 « o ca cu W cS a o ^ ca i*-, CU 43 B "B O ca c*_ '-3 T 3 O cu cu § • § "1 cu 3 03 3 3^ 13 P 2 O 43 ^ 4 3 1 U 3 ? fi C <•> x 5 S u o R 43 £3 ° « M 5S 3 cu —, cu « cu ca 3 ''c3 H-1 ± 3 O to 4J VO CU 8^5 3 Q fi 2 c ' S ca et <*H .2 o • 2 « ._ - i 2 1.11? u, ,cu 0 0 c - cu o °A jo % * ' • cu. PJ PQ < 43 <U > oa cu -+-» CU u, ca 3 cr oa i 43 u •S ca -4-* o Z 3 ca CU CU 43 a oa cu > o oa CU Q PQ >> 44 3 H • e ca a cu P H CU 43 •+-» 00 _fi '53 3 ca X ) fi cu o 43 '3 ca <HH fi 2 '3 « ca ca la cu 3 _ f i C '53 O cu Z ca C N C N C O VO O O O 0\ r H 00 IT) 0\ t VD C N S d H ui S m »n C N C N -<t 3" m OV r H rH ON 00 Ov N C O r- o o o o oo 3 o '-3 .2 "> cu Q -a T J cu t; c o ca ca c ^ c s i a°fi .. ^ 3 B § J I 3 2 co co > D-! 2 2 § 2 cu fi B oo.S -r-03 CU o > 13 3 cr cu 00 3 3 oa ca cu cu 3 ca o i i o z C5 I!© CJ cj 1 2 i n . ^ •IH* Cj ^ Si < s r l 8 r i »5 rC} s -5. *) •st m o >n h g n vo a\ C N i n rj o\ S vd d mm (NxfcN^m^-° ^ H 2 H H ^ 2 •<t m C N vo •<-]- vo m i n 00 Ov - T m IS od »n \r, o\ in o\ »n oo m m o S S od S od od S m m r H C O r H (N ^ d o o o o o o o o m i n Ov CN IT) Ov m m 9 CN 119 B CS 73 73 CD CD -a 3 CJ > o CO <u a CQ 13 <J £ 2 ^ co d ^ o fc w 73 Ix cS 3 73 Co CJ 00 NO C N O C N oS NO NO 00 ^ ON C N cn o m in in o "fr" "fr "fr* 00 C N C N C N s o CJ cj H * ! +3 iH iri 2 w w>c<:22 00 CJ O a 3 cj 00 a 3 CO CO CS » CO 1> CJ a CS CO O U J 3 ccj O "3, o co CJ a 3 cr - 3 o cj ii o 3, es o ja 3 H 3 CS a CJ CJ 1-1 P H CJ vS £50 G 'co 3 B cS 73 3 cj O 4=1 '-3 +-> cS t4-< 3 O g O t o '3 H fS cS . N u , Is « a .s CO O cS o 2 ^1 a 41 65 o, 00 00 C N CO in in in o NO C O o o in C O d 00 C N 00 T-H C N C N C N OcO H Q O N H ( S ON cNOinoNinr^ r~co C N in o co o odod ON C N ri od od od C N r H ^ ^ ON ON m ~ C N O p o o o o o o C O 00 C N NO NO ON NO 00 120 8 13 s3 T3 T3 CD S o •3-<a GO 3 1 r ^ P S3 <U CU H hicl O Ve ai o • •—j ' • ^ IT* 3 CU > cx •c o 1/1 <U Q CQ r - H £3 od in CN in VO ^ o> o CN ON VO m O CN vd od in ^ o m ^ i -CN VO H O 00 rH ON O in co r~; vo CN cn vo d H co co r-J od ON g w cd S3 3 CU 3 '-3 .2 > cu Q c3 T3 cu O CU 2 oo co > pt, 2 2 CO CU O c S3 •c cd > 13 3 cr cu o o s £ i/i oo S3 to cu cu S3 i/i o U cu 1 o "cx cx S3 •4—» o Z CO CU S3 3 cr CO I U cu si CU Ix o z cu 3 H i 3 S3 £ cu cu cu J3 •i—» 00 a 'co 3 3 S3 T3 CU J 3 +-• S3 <H 3 o B % £ 2 « 'ts c S3 s3 13 cu 3 C C o o o Z S3 C5 I a CJ s, 5 1. ^ CO CN *=< CU CN CN CN CN ON vd CO ON ON ON ON O O O CO CO 00 O O 00 VO 00 ON in in* CO C N C N "tf -ct CN CO rn ^  CN O —1 00 p co o c N v o i n r H c o in C N CN ON O CO t-» C N C N CO VO !ONON d od ON ON od ON ON oo 2 ON ON ^ £ j O O O O O O O r H in rH d CO C N O N o C N in in 121 B ca x) x) co co 43 co > CD co CO Q CQ I rn e'-en vo 8 >n vo o -3- >n CN VO o vDrHina\-f CN CN in cn cn cn r-ON d d 0 <-H 00 oS o 3 0 .2 '> CO Q 1 T3 E 6 g u s 6 2 X) ca c XI ca 3 ^ B « «j H H 2 tn>tt!22 "5 c ca ca CO CO CO s > 3 cr CD 60 3 B 3 CO ca CD CD 11 co O & H U 43 ca 0 a , ca -4-* o Z czi CO -4-> CD i-H ca 3 cr 43 U CO ca 11 fr •4—» o Z >> CD 4-3 3 ca E CD , 2 UH CD 43 •*-» 60 t3 ' K 3 B ca x) 3 CO o 43 '-3 ca <— 3 o £ o "S3 "3 3 ca ca E c 'S o Z ca ^1 "eijl u a CJ 1 2 ^ CU ^ cu %) .C 1^ ca o o co 00 00 CO •n CN CO CN CO co O 3; H rn p« vo in ^ CO CO VO CN 00 00 in o\ ON VO—I N tf r f (T, QO rt "tf ON C N C N ON t-; cn C N C N cn r~; ON ON ON 00 ON ON 00 ON C N C N 00ON 00 "I; ON ON 00 0 0 0 0 0 0 0 0 in VO ri C N cn CN ON C N 122 2 C H 13 > - i •d o >, .§•18 co O CD o PQ oo CD CD _S .3 o '•3 CD 3 CD , OJj| -+-» G , C D 3 §1 CD cn 3 _ o-1 ctj CD 3 '5b 43 13 ° 43 CD > Q,43 ^ CD CD u 3 O S '5 e o g 8»1 1° S fD (H Int -° 43 ^CD CX) t, g Q . S CD In CO <N H ;3 '—1 Z 7 £ ° 9oeT.2 S o M CD 5S b i H Q gJ ON § U ^ CD Pn rt J PQ < CO CD CD i-, <a §1 CO I 43 U •c C M ° CD 3 13 > <N •a CD +-> o PJ St ;ON in © "X. V I* 00 ON CN CN ajoo 2 a ON NO r H NO in o in o r-o o r-~ oo oo ON CN in in in r- £ o in ^ oo ,-- CN NO O r -H 43 43 43 00 00 00 D Q Q CD 3 'co 1 CD M 3 • 8 S CD r 2 P H CD 43 W>| c 'co 3 73 CD 43 O Z x j c 41 ~x cy l ^ s t ONinTf inoNNoooin incNp o d c j ^ r i f f i ^ o d o c i ^ h i n h - H i n r - ^ i n i n r - i n ^ t T f i n ^ t 2 2 2 2 2 0 0 N O 0 0 2 2 2 O N^voooinNONor^r-r^No 41 h o o \ 0 \ t N ^ i n H T t o o o o s ? i S 0 ; P ^ ^ ^ c 4 p ^ c N i n " 3 cNinr^NocNNONOrxNO"t"t"tN6-<cr-<t o o in' ^  ON 33 3; 2ON CO oo 2 2 cn o c N O N ^ t i n - H ^ i n ^ S ^ ^ ^ S t ^ N D 3 1 NO cn od - H d CN NO CN v o o o p r ^ r - o r ^ c n o inoopcNr-HvocNr^ON Pco'incNcNONCNSin M-cn ^ r i n N o^inmcN ci oo in o NO t^ - - H M- CN 123 s o o o s cx •a a ca fi <u o a o C Q d £ ca 7 3 >> O T 3 CU - 3 -3 co S> g CO CU •8 O - 3 cu 3 cu 00 . 3 . 3 ^ 00 cu C •3 S & | 3 cu X CU 7 3 .S s o CO CU 00 CU 4=1 -4—» 3 CU u l * H •5 o aj NO 4 3 r - H o 0 0 cu O E f H !t3 <N U, 4 3 O 0 0 Q 00 „ 3 0 0 \ 3 O <u x ^ Q o s 2.2 CO s '?3 CU CU < PJ J C Q < S CX 13 > cx C U C U 4 3 ca 7 3 7 3 CU ca 4 3 cx CU CD s CU cu O *3 co 3 CU 4 3 ( U O •c ca > CO 13 ca 5: 6 a u a 0 0 7 3 cu * a CO £ Q -£ -8 s .0 * CO CO CU CU CD 3 ca "5 > 04 c£r u. CU < 3 3I6 in m C O <—1 " t CN r - i cN CN CN 0 0 o in C N in in ON o 0 0 ON C N "fr "fr C N ON ND CN l> CN C N ND r - H r - H 4 3 4 3 0 0 0 0 Q Q 3 o •3 •c ca > CU O >H 3 O P H CO CO 8 C N o m "fr CO CO r-5 ON X —< 0 0 C N CN 0 0 CO ON ON ND 0 0 C N 'fr NO C O r-~ co 0 0 ci O • o 3 <U CU CU C Q 3 2 o 3 4 3 CO CU _ > tx •c Q 7 3 C Q 0 0 0 O C O 0 ON C O r - H ON r - H O ON C N in "fr "fr r - H C N C N C N d NO 0 0 r - H 0 0 O $ in in s in C O 0 C O in C O C N r~ in in C O "fr r - H "fr r - H C N r - H in in C N NO O ON ON C N C O NO ON r - H 0 0 0 NO r - H ON C O "fr C N C N 3 ON r - H C N NO 3 O -I 2 > W Q 7 3 7 3 IH Ul ca ca <u 3 7 3 7 3 ca 8b.5 - 3 cu & a S CO > OH S S 0 0 in ON O o a o H too ca 7 3 7 3 CU CO 3 <§ CU 43 co s M 3 H 2 o 0 a-PJ CU O CU > CQ 3 O co o u C N ? CO NO ? CO p -t— CN NO CO ON 0 0 0 0 CO d CN in in 00 ON ON CN CO CO " t ND 0 0 - H O 4 3 4 3 0 0 0 0 Q Q cn CO 3 3 CO CO I H 1H CU cu > > C N NO r - H r - H 4 3 4 3 00 00 Q Q NO - 3 73 r. r. * y y o o a 124 T) B o TJ CU t cu c« O co 5 cd CT co i u •a CU 3 > CN S cu . cu 1 / 1 . p CO CN d ov ,rH O V CN r-od vd in vo ON <"H r- CN VD CN CO "«t CN CO VD VD CO VD co rH cN 4H cu cu GO CO VD VD I—I rH 4=1 43 00 X Q Q cd -a TJ cu CO c cd iH •4—> CU cu I o CO CU Q PQ cu M 3 H I 3 cd s cu cu £ cu 43 •4-4 60 a 'co 3 cd TJ cu -3 ^ co b ^ 'tf CN r- as vo rH co *o 00 rH OS M CN VO CN CN CO VD G\ 10 rH oo VD CN co o m O O P in o\ CN CN 3 *-3 fl Q TJ TJ S3 S3 •a TJ i3 .«J s 3 cu g 60.3 3 3 cd 2 CO CO > 04 2 2 3 cd TJ TJ cu CO 3 e •4^  <u 43 CU o > a) >> r—H cd 3 cd cd cu 3 o u cu 3 cd o "SL ex cd -4-* o Z cd 2 »5 'td § 13 cu g E '53 o P Z S3 cu .5, 05 ^ si ^•5 ^ cu inoovoooinrHQOv cNO\ooincNOin*"Tco S i n v d o N T t S - ' t S r H cocN-q-voco^tco'ctcN •rJ-CNinrHOCOCOrHO incNooScor^-^-vorH ovoovovON-^ cNin ON ci in ci CT\ S vo d COCOCN^CN-NfcO^ P rH CN CO l*- rH ON CN h t C O C N V O ^ V D C O i n CO Ov o CO in o\ CN in Ov CO CO 110 § II i ct) I li 125 co 3 a > i D H CN x> CO •*-» CO <". X) co t: CO 1/1 4=1 O CO CO •4—> 3 cr I u . cu 3 2 o oo in o o d ON NO C N C O r- -H p- co C O C N r H _ j . » r-H co ON r— r H C N NO t~~; C O ON r H CN 00 C N C O NO C N O N IT) C N 4=1 00 Q rH ON CO co C N x) CO 8 a, C N NO o rH rH C N 43 -3 43 X X X Q Q D CO 3 'oo a >A ( U 4*i 3 i fi s C U U £ C U 43 •4-^  fi '53 3 CCS X) C U 43 O Z "—A O CN "S3 e It " H CU <u %) • r * C J 1^ I t 5* **» NO o CO o X X 3 IT) rH NO X NO X C N X rH ON 00 ON Tj" 00 X rH C N rH C O C O C N T J - C N H H ^ X co NDrHCOrHir iCOCOrH CO N O o O N in O N 5; O N C N in C N ON «s CO C O C N C N c N N o i n o o N o x i n r H C N in n ON x in r i o in C O ^ C O C N ^ N O C O ^ C O Q ON C O S r^  C N O r H C N C O r - r H Q ^ C N ' ^ - C N i n r H O C O C O r H O • ^ • C O C N N O ' c t N O c O i n i n c N x 2 < r l ' > ' < t N O r H CO 3 O N O N in O o O N X X O N O N o O N O N X rH CN in in CO CN in CN CO CN CN & 2 on co <n co C N rj-^ rH rH r—t rH rH rH co <n C N NO C N C N 3 o in •3- x o in r H o ON in o O C N C O o r»' NO r H r H X C N C O C O C N r H C N C N O C N X C N - r j - C O r H O r H O O O 126 2 ! \ a > 2 •S II I n B! o U CD 5 CJ i s e x — i « u x CD CD CD '£•£ o CD ^ co CQ u CD > 2 a £ * 1 3 o 3 ca 73 cj 3 73 CD C 3 3 fD ca £ s l a S 43 ° CX CD CD J3 CJ +-• <=! 3 CD -3 CD g CD - r -43 CD 00 -+-» 3 e .CD 73 43 5 £3 3 CD CD £ -*-» CD - O CD CD 3 CD u, ,CD O co 3 _ ' S CD co 00 u CD oo Q NO c ° 0 0 O R ^ r o C N 3 < ° •-J ca PQ < cx CD CD 43 -*-» 3 O CJ ca 5 ca 73 73 CD co 3 ca CD 43 CD CJ 3 ca •c ca > CD cn 3 co >> l a u r>l ca o u B g 3 oo CD CJ •c ca > opl 2 CD O NO in ON C N O ON ON od CN IS cn I S H tN oo oo cn C N cn O N »—I O N O od iri O N I— r -H C N 00 00 O N O N fx O oo —; cn cn r - l T t in r-~ O N 00 NO C N 00 S J 0 0 C N o 43 43 00 00 Q Q 73 CD "5 o w NO O r - H C N 43 43 00 00 Q Q CD 2 3 O 00 P H 00 NO 00 C N l O o d sc-at O 00 C N o l> co cn cn |in O N C N NO C N r - H O N r - H OO CX lo 3 CD CD +-» CD PQ 3 O H o 3 43 5 ca 73 73 CD CD 43 a CO CD > CD co CD Q PQ o C N oo Q O N "t "d" O N CO in X O O N I-H O N p 00 00 <-< cn cn ON od S cn - H o N - H T f NO C N Q oo oo O N "t t - i IT) r» CO O N 00 CN r - H S C O —< C N ON NO ^ co in 00 r - H O N C N NO |00 "t ON NO 1^ |co r~; w:| p «t co d C N S coi *-< co <-H <-H C N C N <n ON NO ON C N >n o m in o oo f» co C N r» o oo . •<* cn od od od NO C N r - H C O " T 3 O '0 n .2 2 > C u P H Q T3 T3 <D ca ca g 3 CD g 3 a n . 3 3 73 73 ca Sb .S CD 3 B 2 OO OO > P H S S 3 O N NO O NO 3 O H CO a H CD 4-3 H -I colm </-> 0 d 01 CO !<-if * . CO C N PJ OO CD CJ -3 -CI fDlX CD ' •H B 3 co u CD > m 3 O oo •n ^, r o O ON O C N C N CO C N C N >n >—i <—i ON O N in in o o CO C O "t "t CO CD CD CN ON in C N C N o 43 X Q CO 3 00 u CD > 17? CD o O C X O C N C N - - i 43 43 X X Q Q 00 00 CO to OU - H l H P CD CD 2 > > X T3 73 l — 1 CD CD oo Q O S O >-i IND NO NO C N a O X CN O 43 43 X X Q Q co co 3 3 CO 00 43 43 43 X X X Q Q Q C N O r - H CN 43 43 X X Q Q • 4=1 o n ii X (X oo oo * * 127 cd O rH IS ° co P *-' b •r- CX T ) 0 S - c i ^ cd +-» cd 3 - 1 3 , 5 P-l P CJ CJ CJ CU W o o — ) I n +-> U 13 .2 1 — 1 > .ts H—< •P o eg o -~ •3 cu -3 fa .a >,w ts cj 2 « S « BO c r ° c P P t i l cx ? -S O S > C O CU CU cu Or, -g <-> b c 43 O c*-_ o 3 5 w P. -2 c S "6 J3 cu O co 43 cu CU CU 43 U O - b 3 3 -S 2 e cu 5 •** "6 3 R u co E O 4> Z <*3 0 0 ^ O O C * - O 3 O '•3 vo S "^  CN Q CN < PJ J PQ < F—1 3 O cu ca 3 1 43 u •c o CN CU 3 13 CN TJ CU o PJ . cu O O CO d co 00 d >n 5 r- co 00 r-H S CN CN CO CO CN O N ^ CN CO r H C N 4-» -4-» cu cu CO CO VO VO r H r H [43 43 00 00 Q Q 8 cd TJ cu CO 3 cu 43 3 GO cu I CJ CO <U Q PQ cu 4x1 3 t—1 i 3 cd S cu <u £ cu 43 -4-» OO c 3 cd T J 3 cu O 43 '3 3 O C 2 ^ '•3 3 cd cd 13 2 S .2 GO p ^ O N vo iiO O N co C N co f - co O N C N 00 r n ON CO r n od S 00 vd CO r H cN CN C N WO VO o 5 in co r— co CO vo o g in vd t< r- o t< r H IT) CN VO C N 3 O P > fa « PJ Q 73 -o cu IH IH U -S 3 S <u T J TJ cd ao 3 3 - r -, 3 8 3 * § ~ « 2 co co > a! 2 2 3 cd TJ T J cu cn 3 <U 43 CU O 3 cd co 13 3 cd c^ I 3 o u cu 1 o "cx cx cd •4-» o Z o Z cd , cu G»l ^ EM i t ^ cu O N VO CN O N 00 CN O |VO CN r-H NO 00 r H CN r i O VO r H CO VO r H |co »o uo co co >n co CO 00 00 CN od CN CO loo CN O CN vo in CN CN in co co co O O CO CN CO CO O CO 00 rH r-H t--- ND U0 00 CO d CN S CN S WO CN ^ -"t CO CO vo 5 d CN 00 00 in O N ON X V D r H l N t J - r f ^ Q J - H CO CO CO 00 C O in VD CN CN CO CN ON CN 128 CU 3 13 > CN 0 0 X.I H I* 0 0 o\ r—I C O T 3 o T 3 cu c: cu 4 0 O CO 3 cr 4 3 u . cy 3 1 O ON ON i n 'fx i n . ' i n i n oo CO *-< - H C N i n "fr C N NO r-^  O i—I <-H I—J C N C N "fr C N ON i n m NO i n o \ NO f x o o vo CN NO *-H T3 d) "o o 3? 0 0 C N NO O P H H M 4 3 4 3 4 3 4 3 X 0 0 0 0 X Q Q Q Q CD a 'co 1 CD 3 H i 3 BJ B CD r 2 HH CD 3 '55 3 8 ccj T J CD 4 3 O Cj Ix) C 41 CU o |CN C § NO " * x in -fr "1 co o x o in in o in NO O C O C O x o N X c o r ^ x x i n i n x CN CO CN CO CO "fr CN CN 2 oo2 t- O O W M h C N O O C N C N C O C O ^ ^ C N o o m t N n m o ? r O 0 0 r - l r - H t ~ x N O i n ' f r X C O in C N O N N O C N O N X C N O C O X N D c N r - i N D X — I C N X C N " f r c o c o N o c N c o i n i n c o c o i n c o c N c o X X i n ON ON ^ X ^ 1^ X f x C N "fr O NO C N " f r " f r c O X ^ C N N O i n c N C N i n c O C O C O 41 * x cu CU ^ x> i*4 cj fell 41 I CU C N C N O N O N Q Q O C O C N C N C O X C D O X N O ON NO x ON i n i n r o C N CO CN "fr "fr cN co ON ON ^ ON ^ 2 0 0 O c o O C N N O i n ^ c O " f r x o > n i n i n o N o r ^ c N i n m o o i n c o O N " f r ' c N C > x S c N X CO CN i CN CN CN xSa . t x . cN^^o O m H O O t N r H O l ON i n C O x C N X CN rx NO r x NO i n i n 8 i n i - H d CO CN ON o CN CO X C N NO NO rx ON 129 0 2 1 2 .-3 ca * > CD o -5 cx CD cj s cj CJ o CJ CJ CD o o PQ on cj cn CD 3 ca U H CX cj CJ 3 CD 00 o • ca a £ 3 3 _ _ CD ° 1? Is c 2 1 ^  "T .2 43 43 O > CX CD CD O CJ 43 • --j CD 3 " X — cD 3 " 2 3 43 "-g .5 - $ 2 8 co -3 CD CD co c j oo m £ c*jj JU C 00 w M Q Q '•3 NO" •H Ox 00 ooQ 8 "+3 CN O! r-H « 43 E-i oo _ e' o o a 3 S | co Q 3 ^ 3 & PQ 1 ,3 < o 3 O CO CO 2 CX X CD CD 3 ca T3 T3 CD co 3 CD 43 CD c j •a ca > CO u ca a =3' 3 o s s 3 CO •c ca > 00^  ca^  u CD > < O NO ON NO ON r - i NO CN S ON O r~~ CN r-H 00 ojoo 43 00 Q t- H NO t> 00 CO 00 O N 00 CN CO CN IT) C N O N h iri C O NO C O co t oo C N NO C N oo>_;2 T3 CD "o o CN NO O H H N 43 43 43 00 00 00 Q Q Q CD 2 3 O CO P H 00 CN o d It O N 00 NO NO in od in O r-H C O O N C O "fr o C O oo in H CN oo C O — ' C N NO C X 3 cx 2 =» 6 2 3 CD CD •4—» CD PQ 3 CO CD > O co CD Q PQ <2> C N • * oo NO 13 i-x ca CN NO "fr r-H a C O o C O 00 C N o 00 C O r-H ON NO 00 in C N r-H C O r-H C O "fr oo o 00 r-H CO r-H ON NO 00 CN CO <q lO\ co co S S |co <—i f- "fr 5 in d S CN NO h 3 in \o r i o\ o C"» O CN O N NO CO O od in CN NO od in C N r-< O C O "fr C N co o co in oo 00 NO C N O C N O C O O N ND NO 00 "fr C N O N C N C O 3 O i - .2 2 > b " PJ Q H H ca ca CJ CJ 3 s 3 S "2 V, .2 ob.5 3 is s cj 3 E 2 CO CO > D H 2 2 CN "fr C O r-H CO r-H in oo i3 o H • ca -a T3 CJ 3 H CD 43 co CD •*-» H CD 3 "fr o d al if ^_ H— -t— ON C O ON ON o C N 00 oq in C O r-H r-H PJ C N in in C N 00 in "fr C N CO C O C O "fr "fr CD CJ — v 3 < CJ I six* CO 3 co Ui CD > PQ 3 O co •c ra. o U ON l> C O C O CO v. c « CD in in o o ^ 00 ^ Q 0 0 00 o 43 00 Q CO 3 CO u CD > CD O o cx O C N CN r-H 43 43 00 00 a a CO 3 CO I H CD > T ? CD CZ) S 00 l H O CD j H > 00 CD co O 00 CN O 43 43 00 00 Q Q o oo o cx cx ND NO NO CN CN 43 43 43 00 00 00 Q Q Q 00 00 Q Q 130 CN o 3 > CN T 3 co •4—» co l l T3 10 it CO oa -Q O 03 CO 3 cr oa 'S 1 C ? cu 111 O O CN CQ vo od wo o a o > WO ND CN CN r» CO CO ON wo b b od as H r-^  ND co NO co O ^ o CO ND <—i wo co oo t-» CN NO 00 S NO 00 NO CN ON WO ON WO CO ND NO CO ND CO 00 2 CO ON CO 00 CN N O 00 CN V O O r H r H O r H r H |rC 43 43 43 43 43 00 00 00 00 00 00 D D Q f l Q Q 3 O cd oa I CO 3 CTJ >> CO M 3 H • 3 m B co co CO 43 •4H 00 3 • rH 3 co 43 oa, <+- -r o o § 00| •43 "d cd « .a ta 4 4 CO 3 ~ g CO • CO < > *r*» t j <3 is £ WOTtCAWOWO^WOONTt-rj-WOOO h M m h O ^ h H T j - r | f x r f h S 00 h CN 00 o r-~ r - r-rH CN wo m W 0 r H ( r ) C N r H c r ) T j - r H U 0 C 0 r l - C O 0 HS -> ST fe -S .= •8 s i O 00 CO NO ^ 00 ON S 00 0 O r H O r H 1 — 1 0 O O wo O O N wo O wo wo ON wo 0 r H r H 0 N O r- 00 CO 0 CN r H 00 ON 00 r-' r H r H r H r H r H r H CN b S 06 CN CN ^ • ^ W O N D ^ ^ C N c O C O O c O r H r H r H ^ ' ^ T — \ r H r H r H r H T — i r H r H -H CN rHOCOOOrHrHrHOrHrHOrHOO O ON NO CO ON 00 CN r- 00 W0 5 wo 0 ON CN rH O NO ON ON 00 <*\ CN 00 b CN 00 ON rH ci CN CN rH rH ON OWOCNrHOCOCN^  rHOrHOOCNOrHOOrHCN CN CN NO CO 00 NO CO NO WO CN CN ON o CN |od 3 131 u cu 4=1 +-> '<u 4=1 H o "£ -™ 43 " -2 •*-> a cu o 2 " .2 3 § g x c > cu PQ O & M - M GO K o cu £ £ « s s £ •S — 7 3 "2 § >>3 So 3 13 § T H c i_ O -S cu 2 - c ° f a O O CU JU 43 t= 43 -ti .s.s S CU 3 .2 § 3 ° CU CU co £ 00 3 4D O p cu b J? IS -a .2 T3 r- GO , ° -A G C*H T3 oo'o Z 3 cd i4_ 3 o O <U O 5 L_| In O 3 > S cu < 9 PJ J PQ < cu cu 3 O cd GO (U 3 ' 55 CO S3 cu 4-1 3 H £ 2 CU (U £ CU 43 00 3 'GO 3 cd T3 cu & H-c o 3 O CO i oo| -a cd N O 3 ° c o Z ^ ^ cd 3 wo O N O N 00 ON wo 3 U0 W0 wo O N wo o CO r-H N O N O CO CO o o CO CO X* I— - 00 ON r—1 d wo* 00* 00 00 wo* I—I CN CO CN CN r | r i r i oooowowowo'<tfO'N-wowoooN ^ r ^ r - r - o o r - r - i t ^ r - r ^ O N c o S S S S o d c K r H O N ' S r - w o o d RT H CN O O O O — ' O O O O O C N O ON WO ON ON O co ON* CO 00* 00 CN O O r H r H r H C N O O r H O O O — l O C N O O r - r - O N w o w o o N O N ^ O N O N O c o O r - C O O O C O O N C O C N O N O O N O X x ' x ' x x d c o x S c o ' t f O N S — I r H CN CN CN r H CN O O r H C O ^ T J - r H l ^ T t C O C N W O O O O — i O C N O O c N C N CO 00 N O ,oS CN X CO CO ON W0 CN X !ON ND q r-i O N wo 132 cu 43 cu & O CTj re — S .2 u § 3 cu s i o PQ 6 * 00 CJ CJ CJ X PJ £ — >pa So • - H cn + J re cj U C UH O -3 cj CJ fj 43 13 43 -ti 3 +^  > .s.s S CJ 3 .2 3 £ cj *^  cj on , , cj oo •>-> ^ 00 3 're CJ 43 _ IS -o -S "O r- oo 5 Q ,o ~ L H oo'o Z 3 cc) i*_ 8 'SS H .2 3 O Q « X O CJ fr o 2 g cx CJ cu CN < « PQ < A3 -a CU 8 00 cj > 1 CJ 00 cj Q PQ o Rl O O cx| re > CJ ^ in CN in CN oo o oo ND T - j - fr ON d v i CN od S in >-i <n r - • CN o r- in - H oo ON CN <—i ON '—< cN NO |in CN oo d fr t> - H oo CN cn CN O NO -"fr O N "fr cn ON cn ON cn in CN oo in CN S cn d r> S NO CN CN 3 ON i n o "fr " t "fr CN ON CN CN "fr p in t> cn CN —i o CN CN r-» in in oo . . r> NO in cn fr oo CQ in NO 5 CN 00 CN o "fr in ON O 00 C N f- ON in NO 00 O N "fr - H NO iri NO S cn fr H C N 3 O '0 o > fc 5 PJ Q - 3 Uc i_ cj 3 d _, -3 -3 S cj 3 3 ^ 1 3 | J | 3 5 oo oo > & 2 5 •a cj oo CJ 43 CJ CJ O 00 '53 >•. "re 3 re >> re • CJ 3 o u CJ 3 re o O Z -o ca « u 3 2 (X -a > < > o o o i> "3 a u > in o d n cd 43 ^ CX g l 3 o g be .'S c ° 133 •c X cu 3 CN B a <u. il CJ > CJ 4? C U 4—• 3 O " CO i 43 u 1^ 3 2 o 3 b v o in VD S3 O CN" 00 C N O 2 •> g "fr ON S ON 2 ON ON ON "fr 0 0 C N ^H Q r H 00 CN d O ON o in d H ^"frf^mgmgcN " H O N ^ O N S O N S O N CN U-) rH C N VD 00 C N v o VD O r H r H O r H r H r H CN 43 43 43 43 43 43 43 00 00 00 00 00 00 00 Q Q Q Q Q Q Q s o ' -3 cd 73 3 Cd .3 C D 3 CO B. CO 3 H 3 cd B CJ CJ & 43 » 00 13 3 cd -a CU 43 CU CO CO , IH o 00| -a B -3 ^ Cd - r * B « o 43 CU > 3 O ' •3 cd N O , cu s i cu C3 oo ON oo ON o in |"fr CO r H r H NO f » 00 CO 00 ON S C N r H r H VD 00 ON 00 O C3 s i in "fr CN CO CN CO ^ * O O C N r H r H O r H i n o N i n i n o o o N O N i n p o N o p c N O N o o r -Sco'SS"frco"frS CN CN rn co CN CN 'VO'TtrHVOO'rj-COCO r H CN CN r H CO CN CN o in "fr ON ON CO 00 o ON in 00 CN r H uo oo r H ON CO o 00 00 CO in 00 S r H r H CN CN CN CO in "fr in uo uo CN r H VD I ^ H O r H r H C O C N C N O r H C N V O d CO 00 VD in o v d r-co CN CN 3 VD CN 134 o JD -3 CD > <u 43 « "2 >. S -a re ra H a, o <u c o cx s re » ^ <u 43 -a cu CD I I •§ 3 <J— -3 C+H • ~H o <D r3 ^ o.2 &0 b "a -3 O 3 CD <*H -3 00 , H cn re -*-> y a c -S => <D T3 T 3 . o ^ 112 g o t$ 2 g 8 CD G • — .5 o o CD CD cn 43 -3 (3 CD Li CD 43 e S S w e » O tu CD tn * ^ CD CD e O 00 - " C CM 10 „ O Q CD 4-> L_! H-1 43 ET, h3 obZ t« J- o re OO**3 t3 3 o co — CD ^3 3 o CD 3 .o '53 CO „ CX CD X re <D <D 43 CD 'S CN < 3 PQ < f3 2 "•4—» re CO CD C CO 3 re >> CD 4H 3 H i 3 re g CD CD 00 _g co 3 re -a CD 43 CD co §<' s, 00| -a Cu CD ° * 3 CD O ±3 re "re o Z o > H , 3 1 . re > o cj, fe i t 1 O C3 O -"4, fe .s s i • **** I oooNoocNONinin O N ^ D C A ' H O O O O O M |<n" ovco co NO fr t> od CN ^ CN CN CN CN r-n C N C N fr in t> CO >—'CO CN CN CN O CN CN fr ON t> r-H t> ON 00 IT) O N O N ON o r» in in fr tS CO ON r-H CN CN r—1 linoofrNOcoinfrin r - H ^ r — ( r — ( r - H r - H i — l r — I r-HOr-Hr-HOr-HCNr-H >n in O 00 5 in O N in O N CN 00 00 o O N d 00 fr CO r-H CN r-H CN CN in r—1 O N 00 00 C N C N C O r ^ C O C O N O f r r-lOr-Hr-Hr-HCNr-HCN CN ON CN CD CO CO o 00| T3 CD re 2, re > cu . s t3 O ca 41 ca K 1 53 rS 4 1 O N CO 00 O N CN >n oo fr _ ' r-H CN _ . . O fr CO O N CN O N t> O N CO O N in O N N O CN CN CN I x ^ S S O N ^ O O ^ r ^ C O O C N O C N O c O O N c n f r O t ^ O N O O O N f x inoocNNot>r-ir-ioox SrSHoNOOoo'cofrin r-H CN r-H r-H CN CN CO i n i n i n c N O f r i n c o i n H m O r H O - H t N t N I O 00 NO 8 oo fr CN 135 o CD 2 CU > CU X! T J CU -4—» cd <U CD cd cu -x| cd l> & cd T J -a cu Vi is o X C i CM oo Q ho -a cu '+•» cd cu to cn cu ^> CX •c o Vi CU Q CQ CN ho p CN -« oo oo O vo CN oo ,o\ •> ON 3 in oo co |V0 O VO "T CO NT 0 0 od co od uo d S S ^ CN CN W"> ON co CO "tf t-» CO ON ON U0 VO CN 0 0 d ci od d od r-" uo CN 0 0 CN CO ON O H UO CO CN UO |CN CO U0 CO CN 0 0 O H CN vd CN d vd S CN ^ ri CN ON ^ co co "t uo ON r H 0 0 r H CO CN O N |vd r n uo vd vd r~- co CN r H CN f - CN UO " C t U0 ON ON 0 0 r n uo 0 0 O 0 0 od r-i uo vo vd od " t CN r H CN O 0 0 CO i > O 0 0 0 0 U0 r H U0 VO f - r H "t CN vd CN uo S co * H CN CN r—• CN co uo oo oo ON uo uo r- CN CN CN r~ CO VO CN VD CN r- "t CO UO " t CO r H ON T t CO CN CO o •N" UO CN QO 0 0 <N S ON O S S " t CN CN c o 0 & 1 cd '> (U Q cd T J S I CO CO co 3 S i i C fi fl j cd cd va-J3 > 04 2 2 I T J T J CU Vi G cd l-c cu X cu CO •c cd c» _^ "cd cd i cu a O U cu 3 cd o "cx fr o Z PJ J CQ < 136 •c CU 3 re > CN T3 (U O T3 CU t cu CO 43 O cn CU i n re 3 o-cn i -3 u O* cu 3 O O o fr CO d C N NO ON in ON C N NO 00 ir, NO >n NO IT) NO C N fr NbcoONcofrodinfr c n r - H i r i ' - i o o f r c N i n o o c o f r c o f r c o i > i n i n ' i n f r i n i n ' N d S S t > i n c n N o o N C N c n ^ t ^ t > t>i> l ^ t>0000 •nfr'-HCOrHCOO'-H C N C N C N C N NO 00 C N ND NO O T—I r—I O «-H »—I CN 4=! 4= 43 -3 43 43 43 00 00 00 00 00 00 00 Q Q Q Q Q Q Q 3 O cn CU 3 CO a >> ja 3 E— i 3 ccj B CU P H CU 43 00 3 co 3 B c0 -a (U 43 (U co 00| T3 CU -+-» CO CU re j -§ ju "o 43 CU > O O "3 N O 2 ca ca fe IP s i C3 O CJ, fe.-5 41 ca .?? Cj C N C N C N O o in C O in fr in o C N NO C O fr o C N S ci d d iS r-H C O cn 1 — 1 C N C N C O 4 1 co in o^ ON C N NO O C O 00 C N C N S od fr ON C O C N C N C N C N o O fx NO in fr O N C N C O * — l C N rH rH O CO O N O N CN O N Ifr ^ fx in co o ON in co o H H N H r i C N co o o fr o m o o O N in fr N O o ci o 00 00 d O N S r-H in fr 00 CN CN CO fr CN H rH CN r-H CN o CO CO C N co in C N H o 3 in 137 o _CD 73 43 CJ > <D •S T3 ^ ! '53 O - V -fi ^ 43 • - a DH & O a> _L C cu -a OH c cd > •o o •2 8 „ e s s o 0 is cu • M O cz> '-3 cu O o .2 ~ >»2 O TJ 3 £3 M « cd O H C T J 3 cu . I U H > V > ,<D - S O 1 * - 1 Sis _ oo S o ^ £ fi .2 3 "§> s •S g" <u CD 03 " (3 c3 g.s fi 5 .2 43 £3 43 ob-S M czi O Q is co cu 8 CD C M G 43 T J T=i t ^ -00 T J c c o • 3 c •a s <u T J CO czi CD M OH cd cu O CD 43 fi O fcj -•3 o N O CN > < W CQ < Cd •s . 0 O 4= CU - M CD (3 'co 2 cd >> (U M 3 H I 3 cd 6 CD CU CD 43 00 T _ l 3 .3 CD CZI C O 3 ' £3 cd T J <D 43 0 3 , 3^ O l-H , 00| TJ C M CD o -a 3 o '-3 cd N Id O Z I. cd > ^ ca ^ 1 1^  1^  H r t f 4 . i f i 0 0 \ 0 0 O O H O V D O \ N V O V D V 0 N S 0 ' r 4 r H ' 0 \ m c < T t ( N r H ( f i m H CNCN.-j.r-1 CN UO CN CN rMrH r^HrMrHrHrM " t " t " t C N ' - H T t c n r H | O U 1 0 \ t N i 0 O \ 0 0 \ C N O O O C N O O N c O C O ^S^frcKodi/oixob CO CN CN CN CN N rl C I flv CN CN U H - 1 M r-1 UN H rH M C C  O rn C O C N C N O O C N C N O : \ooo\o\»ooinTt I C O O N C O O O O V D O O N jod uo r-n ON 0 0 ON H oS C N C N C N C N rn C N C N O CN O -T\ CN CN rH rH UN rH rH CO C N C N rH C N O rH C N C O CO CN H CD C O C O , £H 00| T J CD • M cd CD T J CD S, > 1 a l l & l 4~- ^  a s i O vo ON ON CO VO ON co ON 0 0 co vo co u o o d o N O N o d c N o d c o c o u o CN CN CN CN CO CN U0 CN CN Q O O 0 0 ON C N C N c O C N O c O C N ' c t ' - H C N vo-<tcocoovouooo coONr^vooouooO'*? odoSScNc^godScD CNCNcOCOCN-N^rM CN 5 p C O r H r H j j p C N C O ' ^ ' - H C N C O ^ C O r H l l O r M O r H uo vd 0 0 CN CN o C N O N r-ON c^  CN 8 o CN 138 o <D > In CU "o _^ 'S o i f .tj u a £ O 4> — C O •a o, c » o o . ^ s g fa S § J H ! ^ & oo 5 CO ^ s a c T3 3 CD > . > , c j 3 O "H § 1 8 -*-» '2 3 op g e 5 <~ •- O O 2 CD co •5 -3 3 3 cu cu H-» cu 43 <U O 3 cu .*& 3 M O cu • — CD •3 3 £ cu c 00-3 ^ co o Q fa Js E O r > ° 00 T3 • S 3 <U - o 3 O 'co CO CU _ X CO cu * J O <u 3, & g < PJ J CQ < 3 ccj T3 cu CO 3 H cu -3 3 CO CU > O CO CU Q PQ o 5 1-. 31 ca > CN ho 00 o^ ~x Sci CN ""1 00 o 00 NO "-H CN "-H •« 00 Q 00 o 00 Q NO t> NO O —1 ON O NO fr ON —1 o CO fr 0 0 co C D d in 0 0 co CN CN fr >n fr CN in o 0 0 0 0 NO NO fr CN —1 o fr in NO co d ui co S d CN —1 o co fr tn CO O ON CN O CN fr 0 0 O ON in NO d cN 0 0 CO CN ON cN CO NO CN —1 fr 0 0 co r- 0 0 in co >n ON fr - H in o O CN O co ON ON CO 0 0 —H CN ON CN CO >—1 ON 0 0 ON in fr t> NO co in 0 0 O ON CO CN S fr --H 0 0 ON CN in CN CN 0 0 fr ^ t> rH CN CO NO p NO —< fr CN NO " H co 0 0 fr co ON CN CN t> CN CO CN t> ON ON o in in ON rH t> O CN O CN ,rH fr rH ON ON 0 0 t> |CN rH CO CN CO 0 0 NO CO ON CO t> O fr i> NO o CN rx o fr CO d CO NO 0 0 vj CN rH rH CO fr 3 o '-3 >H .2 PJ Q -a -a <u „ l-H ca ca a 3 S "2 "2 .2 Sb.5 ca 3 a 3 cu g 3 C 3 3 ca ca a 2 CO CO > D H 2 2 -a -a CU CO 3 & CU cu CJ •c ca > o co '55 >~> "ca 3 ca ca & cj 3 o u cu O "SH O-ca -*-» o Z, if \$ o o 0 a 11 I S 3 I 139 CN° T J B o « . PJ B cu CCS 3 cr U CN C? cu cu ICQ R= PJ O co ON o vo vd CN CN oo vo vo —< O N X ' 00 r^ CN CN CN rH V l CN CN CO CN o CO o CQ > CO a '55 cu cu M 3 H £ s cu cu £ cu -3 -*-» Ofi| 3 'czi 3 CTJ -a CU 4 3 O 1 3 cu S i |Cj ^  <u —> CJ ^ C N o 1 3 cu CJ o , u 1^ OQ VO CN ~ H oo 00 Q s'l | i n t o \ o \ f f i o o o o M n i n o M O ^ » O N | oo tNHi^ooHO\rnooo\ootNO\ rH rH O CM M r H r H CN CN r H CO CN 1^  CJ ©J S i ' T V O O C N O X X r - C N V O c O d o N ' S r H y o ^ o d o d S c N c ' c o S r J C N r H C N C 0 T H r C N r H COCOrHCNCNCO 3 in o N O CN rHCNrHCNOCOCOO-^- r H r H CN r H ON CU C^ 00 cx •4-* X cu 3 cu 43 rHrHCNCOUOCNrHOWOCOO CN CO 3 o T3 CU 3 fi o c o o cowo^j-wououocNrHNovo-^-rHvoo^-rH O rH rH CO CI Cl O rH -H C) CN rH CO Cl 3 M h oo ^ 3 h - . 3 c i h H o 3 h h v o o N i > o O ON |NO co O N X O O N X C O X O X CN CN CN W0O ON o gco ON CO ON r-H CN rH X 00 CN CN rH rH ON O V0 CN '— O f I CI \t rHOrHOOCNOrHOOrHCNrHOCO 140 CU o s cu CQ U cu . 3 •t—» ' c o s o < PJ •J CQ < II ed C j 8 s cu 3 '55 cu u. cd >> c cd a cu cu cu 00 3 '55 3 cd T3 cu -3 3 _o '-3 cd .H "cd o| 2 %i .5? <3 1 O eg © ©J ^.3 1 O C3 ^ .s IS— *a Ik 1 O Q I f | i in o 00 t N CO ITI vD O N (N O O N O o co oo ci CN vo in r i S od O N O N O N CN CO CN CN CN rH cN 2 2 ON as H ON o CO CN CN CN o o rH O N 00 in O N co 00 r-H O N T—1 o vo in o vo in 00 CN S 00 CN 00 rH O N rH S O N rH 00 rH CO CO •^•rHin'Tl-CNcOCNCO^I" C U w> cd tx CZI 3 O cu i-cx cu -3 •4—> 6 p T J C U 3 3 CN O o o CN rH rH o in O O N in o in in O N o rH rH o vo i> 00 CO ri rH 00 S O N 00 rH rH rH rH rH rH rH CN CN rH CN tZ-, VO ' ^ ^ C ^ C O C O O C O r H T H - C N C O i n r H C O C N r H C O T t r H i n C O " t C O 'C O O r H r H r H O r H r H O r H O O O O O O O r H O r H r H O O O O 3 O o vo So CN O N r-d 5t in CO o CO 8 CN 141 o CJ PQ EH U r-H u. O o" PQ > CO PQ P J CO 43 •*-» 3 o o re CJ 2 ex " r e > — o 4—< o .CJ re CX <U . CJ c <u X .O -3 43 re CJ „ 5 cj 3 re "0 '-3 H O U o 3 •£ S en 3 —i CJ re 3 2 3 ^3 -a CJ "o o cx CJ 43 •4-H 3 cj cj £ CJ 40 CJ O 3 cj u< .<D u. .CJ 0 0 -3 O 43 •*-> CJ -rH -3 £ +-» *^ 3 u . 3 CJ .2 * S O « 3 CJ Koo're O tn 43 cj 00 G 'S 33 >i 3 re 00 43 3 <U '£3 cj -S 3 O < PQ < -a CJ -*-» re tn Q H O 3 _o 'tn cn 2 cx X CJ 3 re -a -a CJ cn 3 ra cj 43 CJ U 3 re •c re > o tn '35 >> 5 re 2 ra1 3 o u s a 3 co 8 re *o CJ cj 43 8 tn CJ > : | CJ tn CJ Q PQ cj CJ 3 re 'C re > 2 CJ > < S O ON 00 ON CN NO C D fr >n O NO fr r i cs cj PQ CO cj 2 3 O CO P H CO fr •5 !d I r - H in oq r - H NO CN r-00 o in co in CN CN ON 00 CN CO in tn CX 3 O i— O 3 CJ CJ CJ PQ tn CX 3 2 o 3 43 »3 ON in in ON in in o N O o O ON ON o o CN d d N O 00 in CS r - H O r - H CO fr , ° CD |in o CO 00 fr O fr fr oo r - H fr in oo o CN CN 00 r-H S NO S t> fr in CN co 00 fr ON fr co fr c5 in CN oo NO CN r - H fr CN 3 O f» 00CN 2 > PJ Q -a -a cj ^ v o 3 u u. re re S £ 3 g s -a -a re' gb.B .5 re 3 3 -c s s * " «3 3 ra re ;a J3 2 co co > D H 2 2 CJ ON oo in oo NO 3 o r-H 8 re T3 T3 cj cj 43 tn CJ 3 r-H NO CO 13 P J CO CJ cj -3 l k -<| 2'* CJ > PQ 3 O cn •c re S o u r - H ON ON CN NO CN CO fr-fc ON ON r -H ^ H in NO ON fr fr D J P Q 3 e CJ > > o £ G9 PQ ^ o o cr -a O SO m" o d cr a u u H u 4 3 u o 4H r< CO a ii •3 •a $ <u 3 -3 >• -3 u •3 •c u u a o -a u ca ca 3 > -a u •3 •c u I ON ca I 6X) 4H 4H X ca a -I -a in rj-o o r H r H r H r H cx cx % s o a |3 <3 c*5 t/5 N a ca • CX >q §« H (( U O rfl O o ft-a d ~S CX S cc • o >o 3 -d o £ a a-^ t» a> | o i tw Z c/5 u PQ ca § -d I 3^ •d c/2 I a u > 53 o CQ . | i Efi -2 a * •3 "s » d ca ca d o ca j 'd S » fi 'S M 5 O h " 2 1 1 3 u S - 3 rd 4 ^ S -a ° = H r i l l > (M H ca ca .S CQ •S I S - 5 § cs u ca " P Q § "S > r* a ca o M fa ' ^ -3 « .& g s Ji U , . O i T J «a o t-s H J ! " t) a •5 I JTJ S S ^ H S ca -a r*> ca <a TJ S 3 o ca U r d •a ° H "S '5 •C a rd ^  2 ca i3 * * * * 142 T3 B <~ a o cu is o '3 w 2 C D o - S CQ « "2 o o 3 *-CM W 3 . - 3 o CQ > CO -fi I to, C D cu 3 CU * -9 O CO T3 CD CQ co .'5b • -c o cd 3 O o • - H 4—» CU 3 cu, 00| 3 e CU 43 -*-» 3 £3 cO T J >> « § a-B cu cz> cu 8 ^ -3 6 0 <*> i v ^ fi CX-fi t i CU CM) M o •« cu 3 ° cu X - f i 2 3 -S O <« cu Q cu £ H £ cu < M cu ss o r> 3 _ e M S P M CU 1- CU ' M c=Ll a * H O « c ° £ 3 ° ^ g •J3 T3 T3 a "a "§ x 'o 'o •*"-< CO cO PJ > > CQ -5 5 3 czi CU cu CO §1 CO I -3 u •c o CN cu 3 "CO > i OH CN •CJ "CJ CU +-» cu PJ (, cu w cu . cu -"5 • CU 3^ O 8 O N O N |vo P o d o o o wo CN h o ON' 00 2 l > - H in |ON O N <-H od in X O N r-> in CN O N O N D cu fi '53 CJ S3 cu M 3 H I fi CO s CU P H CU 43 +-» eo| fi '53 3 B c0 T3 CU 43 O z ^3 CN 1 3 C <3 cu vo 00 ^3 cu O 8j CN 1 3 K cu NO ~H -e 00 Q ^ cu Si •§ .5 3 ca ^ * BN " H 6> " H (U o o C O C N C O C O o 5 O N N O C N O N X C N o C O X C O X i-H r-H N O in N D C N r-H N O X r-H C N X C N X C N C N d H d r-H C O r-H 00 in C N C O C O N O C N C O in C O C O C O C N C O X X in O N ^ ^ O N x 2 o oo r-N N ^ t P 1 r-H r-H r-H N O r H C N T f T t c O X r H C N N D i n c N C N i n c O c O C O CO o in 3 CN N O UO o o N O X WO r—| O N N O O N O N O N CN uo CN O N X WO CN o wo CO ON CN in r j O N N O wo O N r-H CO CO CN CN CO CO CN N O CO CO CN O H t N m h H a ( v i 3 M i n r H O m m H O ' ^ - C O C N N O - < C f N O c O W O W O C N x 2 c O r ~ ' v t N O r - H 00 rH NS O ^ O N D ^ i r i H r f Q O O O O O t Q x -N- r- co O N WO N O C N wo C N wo N O C N N O CN O N O N O t 3 N O o o CN WO WO O N S ^ VO r—] rH O N C O ^ N o r - ^ N o C N c N r H ^ ^ r j c N ^ S c N O N T t W O r H f N i ^ ^ O ^ ^ M - O ^ ^ 143 Tj CD •4—* a & •« CD ° .2 3 £ U 2 CQ CD -S y ^ fi . C D U B S O CD g o"^ O CQ 3 -t t l CO 03 > cx-2 £ * T 3 OD r\ S fi co — oo 1— ?3 o S fi £3 'C fr* -fi cr £ « CD .2 2 $ B >~> CO C|_ 73 <D O X 0 0 co CX'W 3 «T ° -a C 43 CD 01} co X! -rj CD CD CD u S?43 43 J2 *3 C 3 -5 « O co I I I CD -3 hr J5 ~ Z *v CD <*H § « 2 § a l CD jn co is a s •»-> S 3 cd o H T J -a O cd cd e .2 2 o o o CD i-. i_ w a. cx j * CQ 5 cd T J TJ CD 3 cd CD 43 CD CD 3 cd •c cd CO >> "cd cd • CD a O U £3 cd TJ T3 C D C D 43 £3 CO CD > O co CD Q CQ e e 3 C O CD CD 3 cd 'C cd > oof IT CD i n i n o \ f l CN H H N h CN rH r-H ON •—I r - o o 00 ON co co H CN i n r ~ ON NO i n r ~ NO NO r- r- r-» CD CQ ~ViX PJ C O o CQ > C O CD CD 3 O C O P H co ON CO o o o 00 CN 3 i n CN uo oq ON r-H NO rH ON CO CN 00 co C X 3 O >-b CO C X 3 O 3 CD CD C D CQ o 3 43 uo |uo co uo - H C O 1 O N NO ^ O 00 C N C O r - H O U0 ON co C O S S b S co — 1 CN NO ON T]- r -H uo f > ON NO I co ON r -H co uo 00 (N S 00 H ON CO ' — 1 CN M CN NO £ < 0 3 uo uo CN co co co Tf rH U0 NO rn uo CN r-CN 3 o '•3 n .2 2 > C CD a Q Tj TJ CD CD cd cd _ TJ TJ J3 fi S -c *2 £3 £3 S3 cd M — 2 co co > ft, 2 2 o uo co rH 00 CN £3 o £3 cd T J T J CD co 3 +-» CD 43 co CD 3 H CO <o CHCN PJ C O CD CD ' fi.v CD > is' CQ 3 O co 'C ? O U ON ON co co * * * H— CN r—i cN ' ON co ,uo d O N O N O N rH rH rH CO CO CO r-~ NO ^ ^ ^ o NO uo CD CQ > C O co 3 co I H C D > CD CQ 1 PJ C O £ CD « CQ % C O co co in 3 CD co > <H CD CD > CQ 0 I CQ r-1 ft? u ca u cn TJ O" NO " "O I o o o " cr x o S « ^-43 >o > o u cr S* c u u is • u CJ 3 > •a u •a •c 0 1 S o •a u u u u 43 U a i5 N a ca TJ ca a o TJ Cd a I U PQ U u PQ > r* C / 3 ca 3 ca 3 > ca .g i O 43 ca is o u ca 73* ^  ?cj O CJ CJ Vr, ^ H CU) CM) ci) t/2 on CW * * * * * * ca ca Q ^ -g 0 o -M ° ^ ed U ^ 1 *5 a ° > & cr -j '53 CJ 3 r « _3 43 T3 S cj .a e S H 'S ^ i i 3 b ca r3 to • 13 i •a "S * ! S a fjj "» g W 3 I 0 r C 3 H U t) ^ -2 "73 p a s-„ 9 ca • d o " ca o J3 *j CO H CO O a a S3 144 •c u 3 "re > T3 CD <^> . Q-J PJ fe I; co CD CD S3 3 C7 43 u 2 o IS co 00 f-CN CN IT) co r~-«n fr § CO 00 CO IT) fr in co co fr CM O "a CD CO 3 CO CD 43 CD 3 CO CD > '•3 •c g a Q -a PQ CD 441 3 E—1 c3 S CD CD CD vS 00 3 'co 3 re -a CD 4=1 +-• re <** 3 o 6 s £ 2 »5 re re "re CD E CO O CD Z re -3 ico <-H co NO vo ON in |NO co o r- NO ON NO vo in in in d co fr fr r-t CN fr CN NO CN >n 00 NO CO O fr ND NO in —i r~- '—' ro co '—1 in NO d S S in r -H NO in NO CN 3 O »H .2 2 > PJ Q -O T3 JD i _ i _ o re re c T3 - a c3 3 3 -in •S3 E 3 CD £ 00.3 3 3 2 CO CO > Pri 2 2 CD CD 3 re 'C re > "re 3 CD 00 3 3 co CO re -*-» CO CD CD e re CO o E—1 U CD re o 11 a. re -*-> o Z e CD s """H C J 1 a s * * * * I O -5 |1 r*-4 '©•-Si s i N D O ON CO r H 00 m H O O N V O r-H co N O in co CN d r-H fr fr fr CN CO N O fr N O CN fr CN fr ON r-NO NO CN co r- co NO O 00 o CN r -H in in in CN r -H o r- ON NO in i> S CN CO NO CN fr 00 fr CO CO ON CO CO CO CO O r -H CO CN o NO in co in CN O P co NO 00 3 NO o fr O N CN 8 CO 145 APPENDIX B: Composition of the Harlan Teklad and Purina Lab Chow Diets 146 T A B L E B. 1 Composition of the Harlan Teklad Diet TD92053 (0 ppm Folic Acid). Description Harlan Teklad Diet TD92053 is a folate deficient mouse diet. Guaranteed Casein, "Vitamin-Free" Test Analysis* DL-Methionine Sucrose Corn Starch Soybean Oil Lard Cellulose Mineral Mix, AIN-76 (170915) Calcium Carbonate CaC03 Ethoxyquin (antioxidant) Choline Dihydrogen Citrate Dry Vitamin E Acetate (500 U/g) Dry Vitamin A Palmitate (500 000 U/g) Dry Vitamin D 3 (500 000 U/g) Biotin Vitamin B12 (0.1 % in mannitol) Calcium Pantothenate Menadione Sodium Bisulfite Complex Niacin Pyridoxine HC1 Riboflavin Thiamin HC1 197.0 g/Kg 2.0 g/Kg 359.9668 g/Kg 250.0 g/Kg 59.3 g/Kg 50.0 g/Kg 34.0 g/Kg 35.0 g/Kg 7.5 g/Kg 0.01 g/Kg 4.88 g/Kg 0.07 g/Kg 0.06 g/Kg 0.066 g/Kg 0.0002 g/Kg 0.02 g/Kg 0.022 g/Kg 0.028 g/Kg 0.055 g/Kg 0.006 g/Kg 0.006 g/Kg 0.01 g/Kg Mineral Mix, AIN-76 (170915): Calcium Phosphate, dibasic (CaHPC>4) Sodium Chloride (NaCl) Potassium Citrate, monohydrate Potassium Sulfate (K2SO4) Magnesium Oxide (MgO) Manganous Carbonate Feme Citrate, USP (16.7% Fe) Zinc Carbonate Cupric Carbonate Potassium Iodate KIO3 Sodium Selenite Na2SeQ3- 5H2O Chromium Potassium Sulfate CrK(S04)2- I2H2O Sucrose, finely powdered 500.0 g/Kg 74.0 g/Kg 220.0 g/Kg 52.0 g/Kg 24.0 g/Kg 3.5 g/Kg 6.0 g/Kg 1.6 g/Kg 0.3 g/Kg 0.01 g/Kg 0.01 g/Kg 0.55 g/Kg 118.03 g/Kg Chemical Protein,% Composition* Casein "Vitamin-Free" Test, % DL-Methionine, % 19.90 19.70 0.20 Carbohydrate, % Sucrose [includes sucrose in Mineral Mix AIN-76], % Corn Starch, % 61.41 36.40978 25.00 147 T A B L E B. 1 (cont.) Composition of the Harlan Teklad Diet TD92053 (0 ppm Folic Acid). Fat, % 10.93 Soybean Oil, % 5.93 Lard, % 5.00 Fiber, % 3.40 Cellulose, % 3.40 Minerals, % 3.84 Mineral Mix, AIN-76 (170915) [less sucrose], % 3.09 Calcium Carbonate CaC03, % 0.75 Vitamins, % 0.52 Ethoxyquin (antioxidant), % 0.001 Choline Dihydrogen Citrate, % 0.488 Dry Vitamin E Acetate (500 U/g), % 0.007 Dry Vitamin A Palmitate (500 000 U/g), % 0.006 Dry Vitamin D 3 (500 000 U/g), % 0.0066 Biotin, % 0.00002 Vitamin B12 (0.1% in mannitol), % 0.002 Calcium Pantothenate, % 0.0022 Menadione Sodium Bisulfite Complex, % 0.0028 Niacin, % 0.0055 Pyridoxine HC1, % 0.0006 Riboflavin, % 0.0006 Thiamin HC1, % 0.001 Taken from diet specifications available from Harlan Teklad (Madison, Wisconsin). Percentages are based on the guaranteed analysis. 148 T A B L E B.2 Composition of the Harlan Teklad Diet TD92054 (2 ppm Folic Acid). Description Harlan Teklad Diet TD92054 is a normal mouse diet. Guaranteed Casein, "Vitamin-Free" Test Analysis* DL-Methionine * Sucrose Corn Starch Soybean Oil Lard Cellulose Mineral Mix, AIN-76 (170915) Calcium Carbonate CaCQ3 Ethoxyquin (antioxidant) Choline Dihydrogen Citrate Dry Vitamin E Acetate (500 U/g) Dry Vitamin A Palmitate (500 000 U/g) Dry Vitamin D 3 (500 000 U/g) Biotin Vitamin B12 (0.1% in mannitol) Calcium Pantothenate Menadione Sodium Bisulfite Complex Niacin Pyridoxine HC1 Riboflavin Thiamin HC1 *Folic acid 197.0 g/Kg 2.0 g/Kg 359.9648 g/Kg 250.0 g/Kg 59.3 g/Kg 50.0 g/Kg 34.0 g/Kg 35.0 g/Kg 7.5 g/Kg 0.01 g/Kg 4.88 g/Kg 0.07 g/Kg 0.06 g/Kg 0.066 g/Kg 0.0002 g/Kg 0.02 g/Kg 0.022 g/Kg 0.028 g/Kg 0.055 g/Kg 0.006 g/Kg 0.006 g/Kg 0.01 g/Kg 0.002 g/Kg Mineral Mix, AIN-76 (170915): Calcium Phosphate, dibasic (CaHP04) Sodium Chloride (NaCl) Potassium Citrate, monohydrate Potassium Sulfate (K2SO4) Magnesium Oxide (MgO) Manganous Carbonate Ferric Citrate, USP (16.7% Fe) Zinc Carbonate Cupric Carbonate Potassium Iodate KIO3 Sodium Selenite Na2Se03' 5H2O Chromium Potassium Sulfate CrK(S04)2- I2H2O Sucrose, finely powdered 500.0 g/Kg 74.0 g/Kg 220.0 g/Kg 52.0 g/Kg 24.0 g/Kg 3.5 g/Kg 6.0 g/Kg 1.6 g/Kg 0.3 g/Kg 0.01 g/Kg 0.01 g/Kg 0.55 g/Kg 118.03 g/Kg Chemical Protein,% Composition* Casein "Vitamin-Free" Test, % DL-Methionine, % Carbohydrate, % Sucrose [includes sucrose in Mineral Mix AIN-76], % Cora Starch, % 19.90 19.70 0.20 61.41 36.40958 25.00 149 T A B L E B.2 (cont.) Composition of the Harlan Teklad Diet TD92054 (2 ppm Folic Acid). Fat, % 10.93 Soybean Oil, % 5.93 Lard, % 5.00 Fiber, % 3.40 Cellulose, % 3.40 Minerals, % 3.84 Mineral Mix, AIN-76 (170915) [less sucrose], % 3.09 Calcium Carbonate CaC03, % 0.75 Vitamins, % 0.52 Ethoxyquin (antioxidant), % 0.001 Choline Dihydrogen Citrate, % 0.488 Dry Vitamin E Acetate (500 U/g), % 0.007 Dry Vitamin A Palmitate (500 000 U/g), % 0.006 Dry Vitamin D 3 (500 000 U/g), % 0.0066 Biotin, % 0.00002 Vitamin B12 (0.1% in mannitol), % 0.002 Calcium Pantothenate, % 0.0022 Menadione Sodium Bisulfite Complex, % 0.0028 Niacin, % 0.0055 Pyridoxine HC1, % 0.0006 Riboflavin, % 0.0006 Thiamin HC1, % 0.001 Folic acid, % 0.0002 * Taken from diet specifications available from Harlan Teklad (Madison, Wisconsin). $ Percentages are based on the guaranteed analysis. * Harlan Teklad Diet TD92054 is a modification of Harlan Teklad Diet TD92053. The differences between the two diets are indicated by an asterisk. 150 T A B L E B.3 Composition of the Harlan Teklad Diet TD93013 (10 ppm Folic Acid). Description Harlan Teklad Diet TD93013 is a folate supplemented mouse diet. Guaranteed Casein, "Vitamin-Free" Test Analysis* DL-Methionine * Sucrose Corn Starch Soybean Oil Lard Cellulose Mineral Mix, AIN-76 (170915) Calcium Carbonate CaC03 Ethoxyquin (antioxidant) Choline Dihydrogen Citrate Dry Vitamin E Acetate (500 U/g) Dry Vitamin A Palmitate (500 000 U/g) Dry Vitamin D 3 (500 000 U/g) Biotin Vitamin B12 (0.1% in mannitol) Calcium Pantothenate Menadione Sodium Bisulfite Complex Niacin Pyridoxine HC1 Riboflavin Thiamin HC1 *Folic acid 197.0 2.0 359.9568 250.0 59.3 50.0 34.0 35.0 7.5 0.01 4.88 0.07 0.06 0.066 0.0002 0.02 0.022 0.028 0.055 0.006 0.006 0.01 0.01 g/Kg g/Kg g/Kg g/Kg g/Kg g/Kg g/Kg g/Kg g/Kg g/Kg g/Kg g/Kg g/Kg g/Kg g/Kg g/Kg g/Kg g/Kg g/Kg g/Kg g/Kg g/Kg g/Kg Mineral Mix, AIN-76 (170915): Calcium Phosphate, dibasic (CaHP04) Sodium Chloride (NaCl) Potassium Citrate, monohydrate Potassium Sulfate (K2SO4) Magnesium Oxide (MgO) Manganous Carbonate Ferric Citrate, USP (16.7% Fe) Zinc Carbonate Cupric Carbonate Potassium Iodate KIO3 Sodium Selenite Na2Se03' 5H2O Chromium Potassium Sulfate CrK(S04)2- I2H2O Sucrose, finely powdered 500.0 g/Kg 74.0 g/Kg 220.0 g/Kg 52.0 g/Kg 24.0 g/Kg 3.5 g/Kg 6.0 g/Kg 1.6 g/Kg 0.3 g/Kg 0.01 g/Kg 0.01 g/Kg 0.55 g/Kg 118.03 g/Kg Chemical Protein,% Composition* Casein "Vitamin-Free" Test, % DL-Methionine, % Carbohydrate, % Sucrose [includes sucrose in Mineral Mix AIN-76], % Corn Starch, % 19.90 19.70 0.20 61.41 36.40878 25.00 151 T A B L E B.3 (cont.) Composition of the Harlan Teklad Diet TD93013 (10 ppm Folic Acid). Fat, % 10.93 Soybean Oil, % 5.93 Lard, % 5.00 Fiber, % 3.40 Cellulose, % 3.40 Minerals, % 3.84 Mineral Mix, AIN-76 (170915) [less sucrose], % 3.09 Calcium Carbonate CaCC»3, % 0.75 Vitamins, % 0.52 Ethoxyquin (antioxidant), % 0.001 Choline Dihydrogen Citrate, % 0.488 Dry Vitamin E Acetate (500 U/g), % 0.007 Dry Vitamin A Palmitate (500 000 U/g), % 0.006 Dry Vitamin D 3 (500 000 U/g), % 0.0066 Biotin, % 0.00002 Vitamin B12 (0.1% in mannitol), % 0.002 Calcium Pantothenate, % 0.0022 Menadione Sodium Bisulfite Complex, % 0.0028 Niacin, % 0.0055 Pyridoxine HC1, % 0.0006 Riboflavin, % 0.0006 Thiamin HC1, % 0.001 Folic acid, % 0.001 * Taken from diet specifications available from Harlan Teklad (Madison, Wisconsin), •t- Percentages are based on the guaranteed analysis. * Harlan Teklad Diet TD93013 is a modification of Harlan Teklad Diet TD92053. The differences between the two diets are indicated by an asterisk. 152 T A B L E B.4 Composition of Purina Lab Chow based on the "latest ingredient analysis" information.1 Description "Laboratory Rodent Diet™ is a constant-formula rodent diet recommended for rats, mice, and hamsters. The constant formula feature is designed to minimize nutritional variables in long-term studies. It is formulated for life-cycle nutrition; however, it is not designed for maximizing production in breeding colonies. This product has been the standard of bio-medical research for approximately fifty years." Guaranteed Crude protein not less than 23.0% Analysis Crude fat not less than 4.5% Crude fiber not more than 6.0% Ash not more than 8.0% Added minerals not more than 2.5% Ingredients Ground yellow corn, soybean meal, dried beet pulp, fish meal, ground oats, brewers' dried yeast, alfalfa meal, cane molasses, wheat germ meal, dried whey, meat meal, wheat middlings, animal fat perserved with BHA, salt, calcium carbonate, vitamin B12 supplement, DL-methionine, calcium pantothenate, choline chloride, folic acid, riboflavin supplement, thiamin, niacin supplement, pyridoxine hydrochloride, ferrous sulfate, vitamin A supplement, vitamin D3 supplement, vitamin E supplement, calcium iodate, ferrous carbonate, manganous oxide, cobalt carbonate, copper sulfate, zinc sulfate, zinc oxide. Chemical Protein, % 23.4 Composition2 3 Arginine, % 1.38 Cystine, % 0.32 Glycine, % 1.20 Histidine, % 0.55 Isoleucine, % 1.18 Leucine, % 1.70 Lysine, % 1.42 Methionine, % 0.43 Phenylalanine, % 1.03 Tyrosine, % 0.68 Threonine, % 0.91 Tryptophan, % 0.29 Valine, % 1.21 Fat, % 4.5 Cholesterol, ppm 270.0 Fiber (Crude), % 5.8 Neutral Detergent Fiber4, % 16.0 Acid Detergent Fiber5, % 8.2 Total Digestible Nutrient, % 76.0 Nitrogen-Free Extract (by difference), % 49.0 Gross Energy, KCal/gm 4.25 Physiological Fuel Value6, KCal/gm 3.30 153 T A B L E B.4 (cont.) Composition of Purina Lab Chow based on the "latest ingredient analysis" information.1 Chemical Composition2 3 (cont.) Ash, % 7.3 Minerals Calcium, % 1.00 Phosphorus, % 0.61 Potassium, % 1.10 Magnesium, % 0.21 Sodium, % 0.40 Chlorine, % 0.56 Fluorine, ppm 35.0 Iron, ppm 198.0 Zinc, ppm 70.0 Manganese, ppm 64.3 Copper, ppm 18.0 Cobalt, ppm 0.6 Iodine, ppm 0.7 Chromium, ppm 1.83 Selenium, ppm 0.20 Vitamins Carotene, ppm 4.5 Menadione (added), ppm _ Thiamin, ppm 15.0 Riboflavin, ppm 8.0 Niacin, ppm 95.0 Pantothenic Acid, ppm 24.0 Choline, ppm x 100 22.5 Folic Acid, ppm 5.9 Pyridoxine, ppm 6.0 Biotin, ppm 0.07 Vitamin B12, mcg/kg 22.0 Vitamin A, IU/gm 15.0 Vitamin D, IU/gm 4.5 Vitamin E, IU/kgt 40.0 Ascorbic Acid, mg/gm Taken from the dietary specifications available from Jamieson's Pet Food Distributors Ltd. (Delta, British Columbia). Since the nutrient composition of the natural ingredients varies, the chemical composition will differ accordingly. Moisture content is assumed to be 10.0% for the purpose of the calculations. Neutral Detergent Fiber (NDF) is approximately equal to cellulose, hemicellulose, and lignin. Acid Detergent Fiber (ADF) is approximately equal to cellulose and lignin. Physiological Fuel Value (KCal/gm) is the sum of the decimal fractions of protein, fat, and carbohydrate (use Nitrogen Free Extract) multiplied by 4, 9, and 4 KCal/gm respectively. Early editions of the Purina dietary specifications erroneously referred to the units for Vitamin E as "IU/gm". The correct units for Vitamin E are "IU/kg" (Dr. R. Rose at Harlan Teklad to Dr. M.J. Harris, personal communication). 154 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0087678/manifest

Comment

Related Items