"Land and Food Systems, Faculty of"@en . "DSpace"@en . "UBCV"@en . "Snow, Mary Elizabeth"@en . "2010-01-18T20:40:16Z"@en . "2006"@en . "Doctor of Philosophy - PhD"@en . "University of British Columbia"@en . "Maternal ethanol intake during pregnancy results in impairments in general growth and skeletal development. However, it is unknown if ethanol\u00E2\u0080\u0099s effects on skeletal development result from generalized growth retardation or effects specific to bone. Additionally, the level of ethanol exposure required to produce skeletal effects is unknown. The objective of this thesis was to determine if prenatal ethanol exposure has specific effects on bone development, in addition to its effects on general growth, in the fetal [i.e. f\u00C5\u0093tal ] rat. The studies in this thesis were designed to test the hypotheses that (1) prenatal ethanol exposure affects skeletal development at doses of ethanol lower than those required to affect general growth (assessed by fetal body weight and length), (2) skeletal sites differ in their sensitivity to the effects of prenatal ethanol exposure and (3) prenatal ethanol exposure disrupts the morphology of the growth plate of the fetal tibia. The first study examined the effect of different doses of ethanol (designed to approximate low, moderate and high levels of exposure) on fetal growth and skeletal development. This study showed that endochondral ossification was affected by a moderate level of ethanol exposure, whereas body weight and length were only affected by a high level of exposure. Furthermore, the effects of ethanol varied by bone, with bones that undergo more development in utero being more sensitive than bones that are less developed in utero. Taken together, these data indicate the ethanol\u00E2\u0080\u0099s effects on endochondral ossification are independent of ethanol\u00E2\u0080\u0099s effects on general growth and that ethanol may affect the later, rather than the earlier, stages of endochondral ossification. The second study examined the effect of a high level of ethanol exposure on the histological stages of fetal bone development in the tibia, one of the bones found to be most affected by ethanol exposure. Ethanol exposure resulted in decreased tibial length, which was due to a decrease in the length of the diaphysis rather than the epiphyses. In addition, ethanol exposure resulted in a decrease in the resting zone length (which was proportional to the decrease in total bone length) and an enlargement of the hypertrophic zone length. As there is increasing evidence to suggest that the intrauterine environment may influence long-term bone health, the effects of prenatal ethanol exposure on fetal skeletal development could potentially increase the offspring\u00E2\u0080\u0099s risk of osteoporosis later in life."@en . "https://circle.library.ubc.ca/rest/handle/2429/18569?expand=metadata"@en . "THE EFFECTS OF PRENATAL ALCOHOL EXPOSURE ON ENDOCHONDRAL BONE DEVELOPMENT IN THE FETAL RAT by MARY ELIZABETH SNOW B.Sc. (Hons.), McMaster University, 1999 M.Sc, University of Guelph, 2000 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Human Nutrition) THE UNIVERSITY OF BRITISH COLUMBIA August 2006 \u00C2\u00A9 Mary Elizabeth Snow, 2006 ABSTRACT Maternal ethanol intake during pregnancy results in impairments in general growth and skeletal development. However, it is unknown if ethanol's effects on skeletal development result from generalized growth retardation or effects specific to bone. Additionally, the level of ethanol exposure required to produce skeletal effects is unknown. The objective of this thesis was to determine if prenatal ethanol exposure has specific effects on bone development, in addition to its effects on general growth, in the fetal rat. The studies in this thesis were designed to test the hypotheses that (1) prenatal ethanol exposure affects skeletal development at doses of ethanol lower than those required to affect general growth (assessed by fetal body weight and length), (2) skeletal sites differ in their sensitivity to the effects of prenatal ethanol exposure and (3) prenatal ethanol exposure disrupts the morphology of the growth plate of the fetal tibia. The first study examined the effect of different doses of ethanol (designed to approximate low, moderate and high levels of exposure) on fetal growth and skeletal development. This study showed that endochondral ossification was affected by a moderate level of ethanol exposure, whereas body weight and length were only affected by a high level of exposure. Furthermore, the effects of ethanol varied by bone, with bones that undergo more development in utero being more sensitive than bones that are less developed in utero. Taken together, these data indicate the ethanol's effects on endochondral ossification are independent of ethanol's effects on general growth and that ethanol may affect the later, rather than the earlier, stages of endochondral ossification. The second study examined the effect of a high level of ethanol exposure on the histological stages of fetal bone development in the tibia, one of the bones found to be most affected by ethanol exposure. Ethanol exposure resulted in decreased tibial length, which was due to a decrease in the length of the diaphysis rather than the epiphyses. In addition, ethanol exposure resulted in a decrease in the resting zone length (which was proportional to the decrease in total bone length) and an enlargement of the hypertrophic zone length. As there is increasing evidence to suggest that the intrauterine environment may influence long-term bone health, the effects of prenatal ethanol exposure on fetal skeletal development could potentially increase the offspring's risk of osteoporosis later in life. ( iii TABLE OF CONTENTS ABSTRACT ii TABLE OF CONTENTS iv LIST OF TABLES vii LIST OF FIGURES viii LIST OF ABBREVIATIONS x ACKNOWLEDGEMENTS.: xiv CHAPTER 1: INTRODUCTION AND LITERATURE REVIEW 1 1.1 INTRODUCTION 1 1.2 LITERATURE REVIEW 4 1.2.1 FETAL ALCOHOL SPECTRUM DISORDER 5 1.2.1.1 Diagnosis and Prevalence 5 1.2.2 ANIMAL MODELS OF PRENATAL ETHANOL EXPOSURE... 9 1.2.3 GROWTH RETARDATION 15 1.2.3.1 Prenatal Ethanol Exposure and Growth Retardation 18 1.2.4 FETAL BONE DEVELOPMENT 19 1.2.4.1 The Process of Endochondral Bone Development 22 1.2.4.2 Ethanol's Effects on Skeletal Development 29 1.2.5 FETAL PROGRAMMING/FETAL ORIGINS OF ADULT DISEASE 32 1.2.5.1 Potential Mechanisms of Fetal Programming .33 1.2.5.2 Fetal Programming and Bone Health.. 34 1.2.6 POTENTIAL MECHANISMS OF ETHANOL'S EFFECTS ON FETAL BONE 37 1.3 THESIS OBJECTIVE AND HYPOTHESES 41 CHAPTER 2: METHODS 43 2.1 Experiment #1 43 2.1.1 Breeding and Feeding of Animals 43 2.1.2 Determination of Blood Ethanol Concentrations 46 2.1.3 Termination of Dams and Fetuses 47 2.1.4 Fetal Growth and Skeletal Development 47 2.1.5 Statistical Analyses 49 2.2 Experiment #2 51 2.2.1 Breeding and Feeding of Animals 51 2.2.2 Determination of Peak Blood Ethanol Concentrations 52 2.2.3 Histology 52 2.2.4 Statistical Analyses 57 CHAPTER 3: RESULTS 58 3.1 Experiment #1 58 3.1.1 Fetal Growth and Skeletal Development 58 3.1.2 Normal Development and Degree of Ethanol-lnduced Delay 67 3.1.3 Maternal Parameters 75 3.2 Experiment #2 81 3.2.1 Fetal Parameters 81 3.2.2 Maternal Parameters 92 CHAPTER 4: GENERAL DISCUSSION 97 4.1 SUMMARY AND DISCUSSION 97 v 4.2 STRENGTHS AND LIMITATIONS 113 4.3 FUTURE DIRECTIONS 115 4.3.1 Elucidating Mechanisms 115 4.3.2 Fetal Programming (Potential for Long Term Effects) 118 REFERENCES 119 APPENDICES 142 Appendix A , 142 Appendix B 144 Appendix C 147 Appendix D 151 vi LIST OF TABLES Table 1.1: Common terminology used with reference to growth retardation 16 Table 3.1: The average number of fetuses per litter for Experiment #1 59 Table 3.2: The estimated delay in development of body weight and skeletal ossification in ethanol-exposed and pair-fed fetuses (compared with C fetuses on d 17-d21 gestation). Level of significance represents analysis of differences in fetal weight and ossification between treated fetuses and d 20 and d 21 C fetuses 74 Table 3.3: Maternal food intake (g food per g body weight) by week for Experiment #1 76 Table 3.4: The average number of fetuses per litter for Experiment #2 82 Table 3.5: Intra- and inter-rater reliability for the histological morphometric measures assessed using the intraclass correlation coefficient (ICC) 93 Table 3.6: Maternal food intake (g food per g body weight) by week in Experiment #2 94 vii LIST OF FIGURES Figure 1.1: Schematic diagram of the process of endochondral ossification 21 Figure 1.2: The cartilaginous zones of the growth plates of long bones 24 Figure 1.3: Temporal patterns of factors involved in the process of endochondral ossification 27 Figure 2.1: Experimental feeding regimen for the rat dams for Experiment #1. Dams were fed a liquid diet containing ethanol at a level of 15%, 25% or 36% ethanol-derived calories (E15, E25 and E36, respectively) or a liquid control diet (Pair Fed, PF, and Control, C, groups) for 6 weeks.. 45 Figure 2.2: Experimental feeding regimen for the rat dams for Experiment #2 53 Figure 2.3: Morphometric parameters analyzed in fetal rat tibia stained with hematoxylin and eosin 56 Figure 3.1: Fetal body weight (A) and length (B) in Experiment #1 60 Figure 3.2: Ossification of the fetal ulna (A), radius (B), tibia (C) and sacrum (D)... 62 Figure 3.3: Ossification of the fetal femur (A), humerus (B), and scapula (C) 63 Figure 3.4: Ossification of the fetal sternum (A) and metatarsals (B) 64 Figure 3.5: The increase in fetal body weight in control (C) fetuses from d 17-21 gestation (\u00E2\u0080\u00A2) 68 Figure 3.6: The increase in ossification of the radius (A), sternum (B), ulna (C), tibia (D), sacrum (E), femur (F), humerus (G), scapula (H) and metatarsals (I) in control (C) fetuses from d 18-21 gestation (\u00E2\u0080\u00A2) 69 Figure 3.7: Maternal weight gain throughout gestation 78 Figure 3.8: Maternal ethanol intake and blood ethanol concentrations 79 Figure 3.9: Fetal body weight from Experiment #2 83 Figure 3.10: Typical tibiae from ethanol (E), pair-fed (PF) or ad lib control (C) fetuses 84 Figure 3.11: Total length of the fetal tibia... 85 Figure 3.12: Diaphysis length of the fetal tibia 86 Figure 3.13: Proximal and distal epiphysis length 88 viii Figure 3.14: Resting, proliferative and hypertrophic zone lengths in the proximal epiphysis 89 Figure 3.15: Area of the resting, proliferative and hypertrophic zones in the proximal epiphysis 91 Figure 3.16: Maternal weight gain during gestation in Experiment #2 96 ix LIST OF ABBREVIATIONS AGA appropriate for gestational age ANOVA analysis of variance ARBD alcohol-related birth defects ARND alcohol-related neurodevelopmental disorder BEC blood ethanol concentration BMC bone mineral content BMD bone mineral density BMP bone morphogenetic protein C control (ad libitum fed) cm centimetre CNS central nervous system CV coefficients of variation d day dL decilitre E ethanol E15 ethanol (15% ethanol derived calories) E25 ethanol (25% ethanol derived calories) E36 ethanol (36% ethanol derived calories) EDC ethanol derived calories EDTA ethylenediaminetetraacetic acid ELBW extremely low birth weight FAS Fetal Alcohol Syndrome FASD Fetal Alcohol Spectrum Disorder FGF fibroblast growth factor FGFR fibroblast growth factor receptor Fig. figure g gram GH growth hormone GR glucocorticoid receptors h hour Hox homeobox HPA hypothalamic-pituitary-adrenal HZ hypertrophic zone iCa ionized calcium ICC intraclass correlation coefficient IGF-1 insulin like growth factor 1 Ihh Indian hedgehog IL-1 interleukin-1 IUGR intrauterine growth retardation kg kilogram LBW low birth weight LGA large for gestational age mg milligram microCT microcomputed tomography mm millimetres um micrometres n number per group NCAM neural-cell adhesion molecule 9C degrees Celsius oz ounce PF pair fed PF15 pair fed group (paired to the E15 group) PF25 pair fed group (paired to the E25 group) PF36 pair fed group (paired to the E36 group) PI ponderal index Ptc patched PTH parathyroid hormone PTHrP parathyroid hormone-related peptide PZ proliferative zone Runx runt-related transcription factor RZ resting (or reserve) zone SD standard deviation SE standard error SGA small for gestational age Shh Sonic hedgehog Sox sex determining region Y (SRY)-box TGF-B transforming growth factor beta US United States xii VEGF vascular endothelial growth factor VLBW very low birth weight Wnt Wingless/integration site wt weight A C K N O W L E D G E M E N T S First and foremost, I would like to thank my supervisor, Dr. Kathy Keiver, for taking me into her lab and providing her support and encouragement throughout this process. I would like to thank the members of my committee - Dr. Jim Thompson, Dr. Wayne Vogl, Dr. Joanne Weinberg, and Dr. Zhaoming Xu - for their advice and guidance. Special thanks to Dr. Wayne Vogl for the use of his microscope and to Dr. Joanne Weinberg for the provision of facilities in which to conduct my research and for always making me feel like a part of her lab. Thanks also to Dr. T. Michael Underhill for his invaluable advice. I could not have completed my research without the technical assistance of a great number of people - many thanks to Linda Ellis & Wayne Yu for their assistance with the rats and other aspects of the research, George Spurr for teaching me so much about histology and for the use of his histology facilities, and Alexandra Foldes, Elietha Bocskei, Markus Purtzki, Wendy Woo, and Pauline Wong for their help with my research. Thanks to all the members of the Weinberg lab with whom I had the pleasure of working - Dr. Maria Glavas, Alison Halpert, Dr. Candace Hoffman, Dr. Joanna Sliwowska, Ni Lan, Joyce Leo, Fiona Yamashita, and Dr. Xingqi Zhang. Special thanks must go to my \"partner in crime\" - Shalu Duggal. I could not have asked for a better lab mate and friend with whom to share my time at UBC. Thanks for everything! Thank you to the Faculty of Land and Food Systems and the University Graduate Fellowship for funding during my graduate training. I am also grateful to the Research Society on Alcoholism and the Fetal Alcohol Spectrum Disorder Study xiv Group for providing financial support to attend their conferences to present my work and network with other scientists, and for being so supportive of scientists in training. I am especially grateful to my parents - Ann & Jack Snow. Thank you for encouraging me to pursue my dreams and always telling me that I can do anything I set my mind to. To my sister, Nancy, I am grateful for your support, your encouragement and your willingness to listen to me go on about my research. Thanks to all of my friends for your support throughout the years. Special thanks to Sarah MacDonald-McLean for her encouragement, her friendship and her editorial assistance, to Kalev Hunt for making sure I actually got away from my computer once in a while, and to Dr. Kaede Ota for always being there for me. Thanks to my team - the Aggiettes. Late night hockey games were just what I needed after a hard day in the lab or hours of thesis writing. And, finally, to David - thank you for things too innumerable to mention, not the least of which was bringing coffee and muffins to the lab on early \"sampling\" mornings. xv CHAPTER 1: INTRODUCTION AND LITERATURE REVIEW 1.1 INTRODUCTION There is increasing evidence that the prenatal environment can influence the risk of developing some chronic diseases (e.g., cardiovascular disease, diabetes and osteoporosis) later in life (Barker, 1998; Cooper et al, 2002; Godfrey et al, 2001; Hanson et al, 2004; Jones et al, 1999). This is thought to occur through fetal programming, which refers to the ability of changes in environmental factors (e.g., nutrition, stress, exposure to toxicants) at critical periods during development to permanently alter the structure, physiology or metabolism of the body, resulting in a lifelong effect on the organism (Barker, 1998). One such environmental factor is alcohol (ethanol). Exposure to ethanol in utero is known to have serious long-term implications for the offspring and can lead to a diverse array of health problems, described collectively as Fetal Alcohol Spectrum Disorder (FASD) (Chudley et al, 2005). Fetal Alcohol Syndrome (FAS), the most severe manifestation of FASD, is characterized by a distinctive craniofacial dysmorphology, central nervous system abnormalities and growth retardation. Exposure to ethanol in utero also has a number of effects on the developing skeleton, and these effects do not appear to normalize after birth. Research using animal models has demonstrated that prenatal ethanol exposure retards fetal skeletal ossification (Keiver et al, 1996, 1997; Keiver & Weinberg, 2004; Lee and Leichter, 1983; Weinberg et al, 1990), and that this effect persists into postnatal life (Lee & Leichter, 1980; Leichter & Lee, 1979). Moreover, maternal ethanol intake has 1 effects on the growth plate of the offspring's tibia two weeks after birth (Miralles-Flores and Delgado-Baeza, 1992). Prenatal ethanol exposure has also been shown to affect functional measures of bone health, as bone strength was altered by maternal ethanol intake in fetal sheep (Given et al., 2004). It is unknown if prenatal ethanol affects bone function in post-natal life. Other than effects on bone length, it is not known if the observed effects of prenatal ethanol exposure on bone development are permanent or affect long-term bone health. However, recent studies suggest that low birth weight and impaired bone growth and mineralization in utero may decrease peak bone mass and increase the risk of osteoporosis later in life (Cooper et al, 1995, 1997, 2002; Godfrey et al, 2001; Jones & Dwyer, 2000; Jones et al, 1999). Further, the prenatal environment can program bone cells, resulting in altered responses to stimuli (Migliaccio et al., 1996; 2000) as well as alterations in bone composition and morphology (Mehta et al, 2002; Swolin-Eide et al., 2004) in adulthood. Thus, it is possible that prenatal exposure to ethanol may influence the long-term bone health of the offspring. Due to the colinearity between body size and bone mass, it is difficult to dissociate general effects on body size from specific effects on skeletal development (Godfrey et al, 2001). It is currently unknown if the effects of prenatal ethanol exposure on bone are the result of generalized growth retardation, or if they represent a specific effect of ethanol on the process of bone development that could occur in the absence of growth retardation. This is an important question, because many ethanol-exposed babies do not exhibit growth retardation. Although growth retardation in early life in itself has been associated with decreased skeletal mass 2 and increased risk of fracture later in life (Harvey & Cooper, 2004), this is apparently due to a decrease in overall bone size (bone mineral content) (Harvey & Cooper, 2004). On the other hand if ethanol has specific effects on bone development, then bone cells may be programmed and their response to stimuli altered. This could potentially affect both the acquisition of bone mass as well as the rate of bone loss in later life. Although it has not been established if prenatal ethanol exposure has specific effects on bone development, there is evidence to suggest that this is likely. Ethanol exposure affects bone at all other life stages examined, including pregnancy (Keiver & Weinberg, 2003), during post-natal growth (Sampson et al, 1996; Turner et al., 1988), adulthood (Olszynski et al., 2004), and old age (Sampson, 1998; Sampson et al, 1998). Whether or not these effects result from direct effects of ethanol on bone cells, and/or indirect effects on hormonal and other systems involved in bone metabolism is unknown. Importantly, the effects of prenatal ethanol exposure on the post-natal tibial growth plate were still apparent after adjustment for effects on body weight or tibial length, suggesting that the effects were independent of those on general growth (Miralles-Flores and Delgado-Baeza, 1992). Osteoporosis is a debilitating disease that is estimated to cost the Canadian health care system $1.3 billion per year (Osteoporosis Canada, 2005). As the prevalence of FASD is significant (approximately 1% of live births in the United States (May & Gossage, 2001; Sampson et al, 1997), adverse effects of prenatal ethanol exposure on long-term bone health could represent a significant health care issue. 3 The objective of this thesis was to determine if prenatal ethanol exposure has specific effects on bone development, in addition to its effects on general growth, in the fetal rat. The studies presented in this thesis were designed to test the hypotheses that 1) prenatal ethanol exposure affects skeletal development at doses of ethanol lower than those required to affect general growth (assessed by fetal body weight and length), 2) skeletal sites differ in their sensitivity to the effects of prenatal ethanol exposure, and 3) prenatal ethanol exposure disrupts the morphology of the growth plate of the fetal tibia. 1.2 LITERATURE REVIEW The studies in this thesis were designed to determine if prenatal ethanol exposure has specific effects on fetal bone development, in addition to its effects on general growth, in the fetal rat. This literature review begins with a discussion of Fetal Alcohol Spectrum Disorder, including the definition and prevalence of this disorder and issues surrounding drinking during pregnancy. Next, a discussion of the use of the animal model used in this thesis, as well as other animal models commonly used to study the effects of prenatal alcohol exposure, is presented. This is followed by an examination of growth retardation in general, and ethanol-induced growth retardation in particular. Then a discussion of the process of fetal bone development and what is known about ethanol's effects on bone is provided. As there is evidence that the in utero environment may influence long-term bone health, an overview on the concept of fetal programming and the fetal origins of adult 4 disease is presented. Finally, the potential mechanisms by which ethanol may affect bone are discussed. 1.2.1 FETAL ALCOHOL SPECTRUM DISORDER Maternal ethanol intake during pregnancy can have devastating effects on the offspring. Fetal Alcohol Syndrome was first described in the medical literature in French in 1968 (Lemoine et al, 1968) and in English in 1973 (Jones & Smith, 1973; Jones et al, 1973). Children with FAS were described as having a characteristic facial dysmorphology, growth retardation and central nervous system (CNS) abnormalities (Jones & Smith, 1973). More recently, it has become apparent that prenatal ethanol exposure at levels lower than those which cause full-blown FAS also have adverse effects on the fetus, prompting the coining of the term Fetal Alcohol Spectrum Disorder (FASD), an umbrella term which includes a wide range of effects of prenatal ethanol exposure (Chudley et al, 2005). 1.2.1.1 Diagnosis and Prevalence More specific diagnoses under the umbrella of the FASD diagnosis include: FAS, partial FAS, alcohol-related birth defects (ARBD) and alcohol-related neurodevelopmental disorder (ARND) (IOM, 1996). Diagnosis of FAS can occur with or without confirmed maternal alcohol exposure and requires evidence of (a) the characteristic pattern of facial dysmorphology, including short palpebral fissures (opening for the eyes), flattened philtrum (grooved area between the nose and upper lip) and thin upper lip, (b) pre- or post-natal growth retardation, and (c) CNS abnormalities, such as small head size, brain structure abnormalities (e.g., 5 microcephaly [small brain size]), or neurological signs (e.g., impaired fine motor skills) (IOM, 1996). Diagnosis of the other disorders does not require the full pattern of defects necessary for a diagnosis of FAS. The exact level or pattern of ethanol intake that is required to cause FAS/FASD is not known. Data on human ethanol intake during pregnancies that have resulted in FAS/FASD are derived from self-reports, which are subject to recall bias and intentional non- or under-reporting due to the social stigma against drinking while pregnant (Floyd & Sidhu, 2004). As with any area of toxicology, the duration, level and pattern of ethanol intake (and the resulting blood ethanol concentration [BEC]) may influence the effects that occur. Although the exact level of ethanol exposure required to result in FAS is not known, it is clear that low to moderate ethanol intake does not result in FAS, but may have more subtle effects on the offspring. The Institute of Medicine's description of high risk drinking required for a diagnosis of FAS and alcohol-related effects is \"a pattern of excessive intake characterized by substantial, regular intake or heavy episodic drinking\" (IOM, 1996, p. 20); i.e., \"heavy\" consumption of ethanol which results in FAS can occur on a regular basis or may be the result of repeated binge drinking. It has been estimated that an average of at least 4-6 drinks (or 2-3 oz of absolute ethanol) per day are required to result in full-blown FAS (Sokol et al, 1986), while an average of more than 1 drink (or 0.5 oz of absolute ethanol) per day (or less if consumed in a binge pattern) is enough to potentially damage the offspring (Sokol et al, 2003). As recent evidence has demonstrated that even light drinking (0.5 drinks per day) can have adverse 6 effects on the offspring (Sood et al, 2001), it is clear that a threshold for ethanol's deleterious effects has not been adequately identified (Sokol et al, 2003). Not all fetuses exposed to heavy levels of maternal drinking will be born with FAS. It has been estimated that only 4.3% of women who drink heavily during pregnancy will give birth to a child with FAS, where \"heavy\" drinking was defined as one or more of the following: an intake of an average of >2 drinks per day, an intake of 5-6 drinks per occasion (i.e., binge drinking), a positive score on an alcoholism screening survey, or a clinical diagnosis of alcohol abuse (Abel, 1995). A number of risk factors for FASD have been identified, including maternal age, socioeconomic status, genetics, maternal alcohol metabolism, early onset of regular drinking, and pattern of drinking (e.g., frequent binge drinking [> 5 drinks per occasion on > 2 days per week] and frequent drinking [e.g., daily or every weekend]) (Maier & West, 2001; May & Gossage, 2001; Warren & Foudin, 2001). The peak BEC achieved appears to be a more significant risk factor for prenatal damage than amount of alcohol consumed (Pierce & West, 1986; Maier & West, 2001; Warren & Foudin, 2001) and these risk factors may play a role by their effect on BEC. Other manifestations of FASD, such as ARND and ARBD, occur at more moderate levels of maternal alcohol intake, where \"moderate\" is defined as an average of 1-2 drinks (or 0.5-1 oz absolute ethanol) per day (Jacobson & Jacobson, 1999). However, it should be noted that these effects of \"moderate\" drinking appear to be more significant when more drinks are consumed per occasion than in cases where the pregnant woman actually consumes 1-2 drinks each day (Jacobson & Jacobson, 1999). Moreover, there are more subtle effects associated with even lower 7 levels of maternal alcohol intake. For example, growth deficits have been observed in the offspring of mothers drinking an average of less than one drink (or < 0.5 oz absolute ethanol) per day during pregnancy, with a pattern of continuous ethanol intake, rather than binge-type drinking, appearing to be the most important (Day et al, 2002). There is no known safe level of alcohol intake during pregnancy and the current recommendation is that women completely abstain from drinking during pregnancy (Health Canada, 1996; US Surgeon General, 2005). Despite recommendations to avoid ethanol intake during pregnancy, the prevalence of women drinking at levels high enough to cause observable defects in the offspring is high. While there are currently no estimates of FASD prevalence in Canada, it is estimated that FASD occurs in approximately 1% of live births in the United States (May and Gossage, 2001; Sampson et al, 1997). Moreover, since FASD is believed to be underdiagnosed, these estimates are likely to be lower than the actual prevalence (Chudley et al, 2005). The estimated prevalence of FAS ranges from 0.05-0.3% of live births (May and Gossage, 2001; Chudley et al, 2005), which is comparable to that of Down Syndrome or spina bifida (Roberts & Nanson, 2000). Furthermore, the prevalence of FAS in specific Aboriginal communities within Canada has been reported to be between 2.5% (Chudley et al, 2005) and 19% of live births (Robinson et al, 1987). Research in Canada and the United States shows that many pregnant women do consume alcohol. Most of those who do consume alcohol during pregnancy, however, do so at low to moderate levels. Approximately 15-20% of pregnant women report consuming at least one drink during their pregnancy (Ebrahim et al, 8 1998; Flynn et al, 2003; Roberts & Nanson, 2000). Drinking throughout pregnancy has been reported by 7-9% of pregnant women (Roberts & Nanson, 2000) and at least one episode of binge drinking during pregnancy was reported by approximately 2% of women (Ebrahim et al, 1998; Flynn et al, 2003). Importantly, research shows that despite widespread knowledge of the risks of drinking while pregnant, ethanol use during pregnancy is not declining (Floyd & Sidhu, 2004). In addition, it is important to recognize that alcoholism is a serious addiction and that recommendations alone will not prevent addicted women from consuming alcohol. Furthermore, a substantial number of fetuses are at risk of ethanol exposure prior to pregnancy recognition. Half of all pregnancies are unplanned (Forrest, 1994) and more than half of all women of childbearing age report drinking, with approximately 15% of those consuming at binge levels (>5 drinks on one occasion), in the past month (Floyd & Sidhu, 2004). 1.2.2 A N I M A L M O D E L S O F P R E N A T A L E T H A N O L E X P O S U R E Most of the progress in understanding the underlying mechanisms of ethanol's teratogenicity has come from the study of animal models rather than from human subjects. Pregnant women who drink tend not to volunteer for studies, making it difficult to obtain an adequate number of participants for human studies. From those women who do take part in studies, it is difficult to determine the ethanol exposure experienced by the fetus, as ethanol intake data are usually determined by recall and obtaining BECs would require medical contact during a period of intoxication. To our knowledge, BECs of pregnant women have not been reported in the literature. 9 Importantly, ethical considerations prohibit the administration of ethanol to pregnant women; thus, animal models are essential in the study of ethanol's effects on the developing fetus. Animal models allow for control over the dose, timing and duration of ethanol exposure and are not confounded by socioeconomic factors, genetic differences, or polydrug use that often complicate the interpretation of human studies. Many species have been used in assessing various effects of prenatal ethanol exposure, including mice, rats, guinea pigs, zebrafish, chick, sheep, and nonhuman primates. There is no one ideal animal model for prenatal ethanol exposure research as no animal model exhibits all of the criteria required for a diagnosis of FAS (Cudd, 2005); thus, the choice of animal model must be made based on which species models the effects in question, among other considerations. Each species has its own strengths and limitations. Nonhuman primates, while most similar to humans in terms of fetal development, are expensive and have a long gestation, a small number of fetuses per pregnancy, and are associated with ethical concerns. Similarly, cost and lower number of fetuses per pregnancy make sheep a less popular choice than other animal models. The benefits of using rodents include their low cost, short gestation, large litters, high fertility throughout the year under laboratory conditions and the fact that they are well characterized as a model for prenatal ethanol exposure (Riley & Meyer, 1984). In comparison with mice, rat fetuses are larger and thus provide a greater blood and tissue sample size with which to work. However, it is important to note that rodents are born at a stage of development that is premature relative to humans; 10 a term rat fetus is equivalent to approximately an early second trimester human fetus in terms of bone development (Strong, 1925). Since different stages of gestation represent different critical periods for development (e.g., the first trimester in humans is the critical period for the formation of the major organs, whereas the second and third trimesters represent the critical period for growth and brain development) (Coles, 1994), it is important to be cognizant of the relative stage of development of the animal model in question in comparison with humans. Another important consideration in using animal models for the study of prenatal ethanol exposure is the method of ethanol administration. It is imperative that one chooses a method of ethanol administration that models the pattern of drinking which the study intends to model. Furthermore, all methods have strengths and weaknesses that must be considered when selecting a model. Methods that have been employed include inhalation, injection, gavage, gastrostomy, ethanol in the drinking water and ethanol-containing liquid diets. Inhalation using a vapor chamber can result in high BECs and thus can be a useful model for binge drinking, but has the drawback that this route of administration is dissimilar from the human method of alcohol intake. Injection, gavage and gastrostomy allow for tight control over the dose of ethanol and can result in high BECs, but these methods are fairly stressful and, since prenatal stress is known to adversely affect fetal development (Kapoor & Matthews, 2005), stress can potentially be a confounder in studies of prenatal ethanol exposure. Ethanol given via drinking water is problematic in that most rodents find the taste of ethanol aversive and so reduce their fluid intake, and concomitantly their food intake, in response to ethanol in the drinking water. This 11 can result in low BECs and possibly dehydration due to low fluid intake, and suboptimal nutrition due to low food intake (Riley & Meyer, 1984). Providing ethanol as part of a liquid diet, which is given as the sole source of food, results in significant BECs and minimal stress for the dam (Weinberg, 1984), and is the method used in the experiments in this thesis. When conducting studies on prenatal ethanol exposure, it is important to include appropriate control groups in the study design. Regardless of the route of administration, ethanol is known to have an anorexigenic effect and so rats fed a liquid diet containing ethanol tend to have a decreased food intake compared with ad lib control rats, especially when first placed on the experimental diet. Thus, it is important to include a pair fed (PF) group, in which rats are fed an isocaloric liquid control diet in an amount equal to the amount consumed by the ethanol (E) dams. Specifically, the amount of food, on a g per kg body weight basis, consumed by an E rat on d 1 of its gestation is the same amount of food that is consumed by its PF partner on d 1 of its gestation; for this reason, pregnancies are staggered such that PF rats are always at least one calendar day behind their respective E partners. This group controls for the decreased intake of nutrients in the ethanol-exposed group. Ethanol has a high-energy value (7 kcal/g) and the ethanol-derived calories in the liquid E diet are replaced by carbohydrates (typically maltose-dextrin) in the liquid control diet. The ethanol-derived calories are not replaced with protein as differences in dietary protein content have been shown to result in differences in BEC and in fetal effects, despite identical amounts of dietary ethanol (Weiner et al, 1981). Similarly, fat is not used to replace ethanol-derived calories in the liquid control diet i 12 as high fat intake is known to result in fatty liver and the effects of such pathology are a potential confounder. Unfortunately, pair feeding is not a perfect control for the E group. Pair feeding is, in itself, an experimental treatment. PF rats usually have their food intake restricted and consequently alter their food intake pattern, consuming their daily ration quickly, and are subsequently food deprived for the remainder of the day (Weinberg, 1984). Thus, PF rats are hungry and this stressor can potentially affect the developing fetus. Therefore, it is important to also include an ad libitum fed control (C) group, which serves as a control for the PF group. Thus, differences among the E, PF and C groups allow one to separate effects resulting from ethanol and those resulting from reduced nutrition. It should be noted that we start our rats on their respective diets 3 weeks prior to gestation. This allows the E rats to adapt to the diet prior to breeding, thus minimizing the food restriction experienced by the PF (and E) rats during gestation. Previous research in our lab has shown that this strategy can result in PF rats consuming the same amount of food as ad lib control (C) rats during gestation (Keiver & Weinberg, 2003). In addition, investigators should be aware that ethanol can affect the digestion, absorption and metabolism of a number of nutrients and it is not possible to control for these effects. Regardless of the method of ethanol administration used, it is important to determine the BEC of the animals in any study on the effects of prenatal ethanol exposure. Species vary in their rate of ethanol metabolism (for example, rats, mice and humans metabolize alcohol at 550, 300 and 100 mg/kg/hr, respectively) and so a given dose of ethanol will result in significantly different BECs and duration of ethanol 13 exposure in different species (Abel, 1980). Furthermore, the relationship between BEC and time after administration has been shown to differ between mice and rats, between different methods of ethanol administration (Livy et al, 2003), and between pregnant and non-pregnant animals (Abel, 1980). As peak BEC has been shown to have a stronger relationship to effects of ethanol exposure than dose of ethanol per se (Pierce & West, 1986), it is imperative to determine BECs in any study on the effects of prenatal ethanol exposure to allow for comparison of the results with the human situation. Administration of ethanol via a liquid diet, the method used in the experiments in this thesis, typically yields peak BECs of 100-200 mg/dL when given at 36% EDC (Keiver & Weinberg, 2003; Weinberg, 1985; Weinberg et al, 1990). This level is higher than the legal BEC limits for drunk driving (50-100 mg/dL) in Canada and the United States, and would generally be considered to correspond to heavy, but not binge level, drinking. As human alcoholics have been found to reach BECs of >500 mg/dL (Urso et al, 1981), this level of BEC is well within the range of BECs in humans consuming ethanol. As there is not just one single pattern of ethanol consumption among human alcoholics (e.g., some alcoholics consume in a sporadic binge pattern, others drink daily), it is not possible to have one single method of ethanol administration that will model all patterns of human drinking. This liquid diet method of ethanol administration results in a chronic, steady intake throughout gestation, and thus models a pattern of daily drinking. This model has been used for over 20 years to study a variety of FASD effects in rats. ( 14 1.2.3 G R O W T H R E T A R D A T I O N One of the characteristic and most consistent effects of prenatal ethanol exposure in humans and animal models of FASD is growth retardation, resulting in decreased body weight and length (Day et al, 2002; Detering et al, 1979; Hannigan et al, 1993; Jones et al, 1973; Keiver & Weinberg, 2004; Lee, 1987; Lochry et al, 1980; Streissguth et al, 1991). Growth retardation, regardless of the cause, is associated with both short-term morbidity and increased susceptibility to chronic disease in later life (Brodsky & Christou, 2004). When discussing growth retardation, it is important to be aware of the terminology and concepts associated with this area. Table 1.1 provides definitions of a number of terms commonly used when referring to intrauterine growth retardation (IUGR). While the term IUGR is often used synonymously with small for gestational age (SGA), it should be noted that there is, in fact, a conceptual difference between these two groups of infants. IUGR refers to infants with clinical evidence of restricted fetal growth, whereas SGA is a statistical term (Han, 1999; Sparks et al, 1998) that includes normal, small infants (i.e., by definition, 10% of infants with normal growth should fall below the10 th percentile). Thus, SGA infants include both growth retarded and normal children. Intrauterine growth retardation can result from a number of causes, including prenatal ethanol (and other drug) exposure, maternal malnutrition and disease, genetic disorders and congenital viral infection. Rubella, cytomegalovirus and herpes virus infections are all known to result in IUGR (Ergaz et al, 2005). Genetic disorders resulting in IUGR include trisomy 13, trisomy 18, trisomy 21, XO (Turner) 15 Term Definition intrauterine growth retardation (IUGR) fetal size less than the 10 , n percentile of the normal size for gestational age with clinical evidence of growth restriction (e.g., morbidity/mortality associated with growth restriction) severe IUGR fetal size less than the 3 r d percentile (i.e., less than 2 standard deviations below the mean) for gestational age small for gestational age (SGA) infant with a birth weight less than the 10 t h percentile of the normal size for gestational age large for gestational age (LGA) infant with a birth weight greater than the 90 t n percentile of the normal size for gestational age appropriate for gestational age (AGA) infant with a birth weight between the 10 t h and 90 t h percentiles of the normal size for gestational age low birth weight (LBW) infant with a birth weight less than 2500 g, regardless of gestational age at birth very low birth weight (VLBW) infant with a birth weight less than 1500 g, regardless of gestational age at birth extremely low birth weight (ELBW) infant with a birth weight less than 1000 g, regardless of gestational age at birth (Bernstein et al, 2002; Han, 1999; Sparks et al, 1998) Table 1.1: Common terminology used with reference to growth retardation. 16 syndrome, Dubowitz syndrome and Brachmann-De Lange syndrome (Das & Sysyn, 2004). Maternal protein and/or energy malnutrition has been shown to result in IUGR, as has prenatal stress and glucocorticoid administration (Ergaz et al, 2005). Growth restriction can occur in two major patterns: symmetric or asymmetric. Symmetric growth retardation refers to infants in which growth retardation is proportional, with low body weight, length and head circumference, and proportionally small organs (Han, 1999; Sparks et al, 1998). Asymmetric growth retardation refers to infants with low birth weight, but normal body length, with some organs (e.g., brain) of normal size whereas others (e.g., liver) are small (Han, 1999; Sparks et al, 1998). To differentiate between symmetric and asymmetric growth retardation, the ponderal index (PI) is used. The PI is calculated as: PI = birth weight (g) X 100 / body length (cm) 3 A low PI, where the infant has a low birth weight relative to body length, reflects asymmetric growth retardation, whereas an infant with symmetric growth retardation would have a normal PI. In general, asymmetric growth retardation is thought to result from a decrease in nutrient supply later in pregnancy (e.g., due to placental insufficiency) (Han, 1999; Sparks et al, 1998). In contrast, symmetric growth retardation is thought to occur as a result of genetic disorders, congenital infections, severe maternal malnutrition and exposure to teratogens (including ethanol) (Han, 1999; Sparks et al, 1998) and usually represents a limited growth potential (Sparks etal , 1998). More recently, 17 however, it has been proposed that rather than representing two distinct groups (i.e., asymmetric and symmetric), growth retardation occurs along a continuum of changes in body proportions, and that the assumptions regarding the timing of fetal nutritional limitation related to type of growth retardation may not be completely accurate (Harding, 2001). Growth retardation from prenatal ethanol exposure, while symmetric, has been shown to be characterized by a \"relative brain sparing\" in rats (i.e., while absolute brain size in ethanol-exposed fetuses is reduced, relative to total body weight, brain sizes were increased) (Weinberg, 1985). 1.2.3.1 Prenatal Ethanol Exposure and Growth Retardation Maternal ethanol intake at high levels results in decreased body weight and length in the fetus or neonate (Abel, 1996, 2000; Abel & Dintcheff, 1978; Breese et al, 1994; Hannigan et al, 1993; Keiver et al, 1996, 1997; Keiver & Weinberg, 2004; Mauceri et al, 1994; Pullen et al, 1988; Reyes et al, 1985; Sanchis & Guerri, 1986; Singh & Snyder, 1982; Vavrousek-Jakuba et al, 1991; Weinberg, 1985; Weinberg et al, 1990). Although significant effects on body weight tend to be associated with maternal ethanol intakes that result in BECs >100 mg/dL (Abel, 1996; Gallo & Weinberg, 1982, 1986; Savage et al, 2002), the BECs required to decrease body length have not been as well characterized. Studies in the rat and in the mouse suggest that significantly shorter body length also occurs at high (Detering et al, 1979; Keiver & Weinberg, 2004; Lee, 1987; Lee & Leichter, 1980; Leichter & Lee, 1979; Lochry et al, 1980), but not at lower (27% ethanol-derived calories [EDC]) (Samson, 1981) levels of ethanol exposure. However, Lochry et al (1980) reported a significant linear trend between body length at birth and prenatal ethanol exposure 18 (0, 12, 23 and 35% EDC), but individual group differences were not reported. In humans, prenatal ethanol exposure is also known to affect body weight and length in a dose-dependent manner, with effects being detected even at low levels of exposure (e.g., <0.2 drinks [or 0.1 oz of absolute ethanol] per day) (Day et al, 2004). In a retrospective analysis of the birth weights of rat litters, Hannigan et al (1993) demonstrated that prenatal ethanol exposure consistently results in growth retardation in all pups in an exposed litter, as opposed to affecting only some pups in a litter. Moreover, children and animal models exposed to ethanol in utero do not experience catch up growth postnatally and so continue to remain smaller than non-exposed children (Day et al, 2004; Geva et al, 1993; Klug et al, 2003; Leichter & Lee, 1979). It is clear that high levels of prenatal ethanol exposure result in an overall growth retardation. Determining whether the effects of prenatal ethanol exposure on fetal bone development are merely a result of this generalized growth retardation or represent a specific effect of ethanol on bone was a focus of this thesis. 1.2.4 FETAL BONE DEVELOPMENT The fetal skeleton develops by two distinct processes: intramembranous and endochondral ossification. Intramembranous ossification, which forms flat bones (e.g., bones of the face and skull), occurs when mesenchyme condenses, becomes vascularized and mesenchymal cells differentiate directly into bone-forming osteoblasts. The osteoblasts deposit bone matrix and calcification of the matrix occurs, with mesenchymal cells that remain between sections of developing bone 19 differentiating into bone marrow (Moore & Persaud, 1998). Endochondral ossification, which forms most bones (including all long bones), also begins with the condensation of mesenchyme. However, unlike intramembranous ossification, the mesenchymal cells differentiate into chondrocytes, which form a cartilaginous template. This template becomes calcified and vascularized, and chondroclasts/osteoclasts, which are derived from blood monocytes (Carlson, 1996), begin to resorb the cartilage. Osteoblasts, which are recruited from the perichondrium (Colnot et al, 2004; Colnot 2005), gradually replace the resorbed cartilage with bone (Moore & Persaud, 1998). Figure 1.1 provides a schematic representation of the process of endochondral ossification in a long bone, which will be described in more detail in section 1.2.4.1. The process of endochondral ossification begins in utero and continues into postnatal life. Different bones within the skeleton start their ossification at different times, resulting in some bones being more developed in utero than others. Furthermore, there are species differences in the degree of ossification that occurs in utero, with some species (e.g., humans) showing greater skeletal ossification than others (e.g., rats). There is evidence that prenatal ethanol exposure delays skeletal ossification in bones that form by both intramembranous (Lee & Leichter, 1983) and endochondral ossification (Keiver et al, 1996, 1997; Keiver & Weinberg, 2004, Weinberg et al, 1990). This thesis focused on the effects of prenatal ethanol exposure on endochondral ossification and so this review will focus on endochondral, rather than on intramembranous, ossification. 20 Mesenchymal Cartilage Condensation Model Vfm Si f i f e mesenchymal cells chondrocytes Hypertrophy Angiogenesis Ossification Secondary Ossification Bone calcified matrix hypertrophic chondrocytes apoptosis blood vessel bone collar bone proliferative chondrocytes hypertrophic chondrocytes marrow secondary cavity ossification centre -epiphysis -diaphysis growth plate F i g u r e 1.1: Schematic diagram of the process of endochondral ossification. First, mesenchymal cells form a condensation and these mesenchymal cells then differentiate into chondrocytes, resulting in the formation of a cartilage \"model\" of the bone. Chondrocytes at the centre of the diaphysis of this model undergo hypertrophy and this cartilage becomes calcified. Subsequently, blood vessels invade the region of hypertrophic chondrocytes, columns of proliferative chondrocytes develop adjacent to the hypertrophic chondrocytes and the bone collar forms around the diaphysis by intramembranous ossification. The region of calcified cartilage becomes resorbed by chondroclasts/osteoclasts and osteoblasts start to form bone tissue. Next, secondary ossification centres form in the epiphyseal regions and the marrow cavity develops within the diaphysis. Finally, a growth plate consisting of cartilage separates each epiphysis from the diaphysis. These growth plates allow for continued bone growth throughout childhood. (Adapted from Goldring et al, 2006; Kronenberg, 2003; Moore & Persaud, 1998). 1.2.4.1 The Process of Endochondral Bone Development The limbs begin as limb buds, which appear on the sides of the developing embryo and consist of mesenchyme covered with ectoderm. Mesenchymal cells within the limb bud condense in a proximal to distal progression, first to form the stylopod (humerus/femur), then the zeugopod (radius & ulna/tibia & fibula), and finally the autopod (carpals/tarsals and phalanges) (Tuan, 2004). These precartilage condensations can be identified as densely packed mesenchymal cells compared with the more loosely packed surrounding mesenchyme. The limb bud initially contains blood vessels, which regress at the site of mesenchymal condensation (Colnot, 2005). Condensations provide a scaffold that determines the shape, size, position and number of skeletal elements that will eventually form (Tuan, 2004). The initial size of a condensation is a reflection of the growth potential of the bones that will result from that condensation (Hall & Miyake, 2000). Following condensation, mesenchymal cells undergo differentiation into chondrocytes; an exception to this is the mesenchymal cells within the interdigital regions of the autopod, which undergo apoptosis, allowing the development of the digits (Tuan, 2004). Some mesenchymal cells at the boundary of the skeletal element remain undifferentiated and form the perichondrium (Karsenty & Wagner, 2002). The end result of this process of chondrogenesis is the formation of a cartilage template of the skeletal element. The cartilaginous template then matures and is replaced with bone tissue. First, chondrocytes at the centre of the template differentiate into hypertrophic chondrocytes (Karsenty & Wagner, 2002). Simultaneously, cells of the 22 perichondrium differentiate into osteoblasts, which form the bone collar around the centre of the cartilaginous template (St. Jacques et al, 1999). The bone collar serves to stabilize the developing shaft and constricts the width of the bone such that the subsequent chondrocyte proliferation proceeds along the long axis of the bone, thus facilitating its lengthening. The cartilaginous portions of the bone are organized into distinct sections, or zones (Figure 1.2): the resting (or reserve) zone (RZ), the proliferative zone (PZ) and the hypertrophic zone (HZ). The zones can be differentiated histologically based on cell morphology (described below). Morphology is commonly used as the first step in examining effects on the growth plate (Gaiser et al, 2002; Gress & Jacenko, 2000; Mehta et al, 2002; Sanchez & He, 2002; Smith et al, 2005; Smink et al, 2003a, 2003b; van Buul-Offers et al, 2005). The morphological examination can then be further extended to include immunohistochemical staining of stage-specific markers to investigate functional consequences of various manipulations (e.g., transgenic mice, drug treatments). In this thesis, examination of fetal bone was limited to morphological assessments because the purpose of the thesis was to determine if ethanol has specific effects on bone. If ethanol is shown to have specific effects on bone, then future studies can be conducted to investigate the underlying mechanisms and functional consequences. The RZ (which is also referred to as the reserve or germinal zone) consists of small, spherical chondrocytes, alone or in pairs, which are separated from one another by matrix (Figure 1.2). This zone is a remnant of the original embryonic cartilage that differentiated from the condensed mesenchyme (Abad et al, 2002). 23 F i g u r e 1.2: The cartilaginous zones of the growth plates of long bones. A. A photomicrograph of the proximal end of a gestation day 21 fetal rat tibia, illustrating the resting zone, the proliferative zone and the hypertrophic zone. B. A schematic diagram of these zones. (B. Adapted from Kronenberg, 2003). Resting chondrocytes are stem-like cells that can give rise to proliferative, and ultimately hypertrophic, chondrocytes (Abad et al, 2002). Evidence suggests that the cells of the RZ also functions to orient the developing PZ and prevent chondrocyte hypertrophy from occurring too near to the RZ (Abad et al, 2002). The PZ consists of flattened chondrocytes, arranged in columns parallel to the bone's long axis (Figure 1.2); these cells undergo cellular proliferation and produce matrix, contributing to the longitudinal growth of the bone (Hunziker, 1994). The proliferative chondrocytes at the bottom of each column then cease to proliferate, becoming larger and more spherical as they differentiate into hypertrophic chondrocytes. Chondrocytes of the HZ mineralize the matrix by releasing calcium and phosphate ions (as well as enzymes that degrade inhibitors of mineralization) from matrix vesicles (Lian et al, 1999). This calcified cartilage is subsequently invaded by blood vessels due to the production and release of angiogenic factors by hypertrophic chondrocytes (Kronenberg, 2003). Hypertrophic chondrocytes undergo apoptosis and chondroclasts/osteoclasts are recruited to the zone to degrade the calcified matrix (Kronenberg, 2003). The remaining cartilaginous matrix provides a scaffold on which osteoblasts begin to secrete bone matrix, resulting in the replacement of cartilage with bone tissue (Kronenberg, 2003). This process of chondrocyte hypertrophy, matrix mineralization, resorption, vascular invasion and bone formation is referred to as \"ossification\" (Goldring et al, 2006) and the bone resulting from this process at the centre of the diaphysis is referred to as the primary ossification centre. Subsequently, secondary ossification centres develop in the 25 epiphyses as the chondrocytes there cease to proliferate and undergo hypertrophy and apoptosis, and attract vascular invasion and osteoblasts (Kronenberg, 2003). The cartilage remaining between the primary and secondary ossification centres is referred to as the growth plate (or epiphyseal plate) and allows for longitudinal growth until it closes at the end of puberty (Karsenty & Wagner, 2002). A large number of factors (e.g., local growth and differentiation factors, transcription factors, systemic hormones) have been shown to regulate the many stages of this complicated process of endochondral bone formation. The role of local regulatory factors at different stages is shown in Figure 1.3. The size of a condensation is a reflection of the growth potential of the resultant bones and is determined by factors that regulate the boundary of the condensation (Hall and Miyake, 2000). The boundaries of the condensation are regulated by matrix components whose synthesis is controlled by growth factors (Goldring et al., 2006; Hall and Miyake, 2000; Koyama et al., 1995). The differentiation of mesenchymal cells into chondrocytes and the maturation of chondrocytes through the resting, proliferative and hypertrophic phases is controlled via the interaction of three major signaling pathways: the bone morphogenetic protein (BMP), fibroblast growth factor (FGF) and Indian hedgehog (Ihh)-parathyroid hormone related peptide (PTHrP) pathways. These pathways act independently and in concert to control and balance the rates of cell proliferation and differentiation, and thus control the pace of endochondral bone formation. 26 Mesenchymal Condensation Cartilage Model TGF-B Wnt-3A,7A FGF-2,4,8,10 Shh BMP-2,4,7 HoxA, HoxD Sox9 Gli3 Fibronectin-cadherin-Tenascin ^ C o l l a g e n l - ^ \u00C2\u00A3>\u00E2\u0080\u00A2 i) i 0 9% > 4 \u00E2\u0080\u00A2Hyaluronan-IGF-1 FGF-2/FGFR2 BMP-2,4,7,14 Sox9,5,6 Aggrecan \u00E2\u0080\u00A2 Collagen II B, IX, XI \u00E2\u0080\u00A2 Hypertrophy FGF-19/FGFR3 BMP-2,7 PTHrP Ihh/Ptc Runx2, Runx3 \u00E2\u0080\u00A2Collagen X-Angiogenesis Ossification VEGF FGF2/FGFR1 Wnt14/B-catenin Runx2 _ Collagen I Osteoclacin / Collagen I IA V \ Collagen XI / Figure 1.3: Temporal patterns of factors involved in the process of endochondral ossification. The temporal patterns of growth and differentiation factors (above the arrows), transcription factors (below the arrows) and extracellular matrix proteins that distinguish the different stages (below) are indicated on the figure. (Figure adapted from Goldring et al, 2006). In addition to these local factors, systemic factors, such as insulin (Fowden, 1995), glucocorticoids (Mushtaq et al, 2004), and parathyroid hormone (PTH) (Miao et al., 2002) may also play a role in the regulation of chondrocytes. PTH is involved in the regulation of both angiogenesis and ossification, and is known to be physiologically important for prenatal skeletal development. PTH is also a major regulator of fetal calcium homeostasis and its effects on mineralization result from its role in maintaining blood ionized calcium (iCa) levels (Kovacs et al, 2001a). Insulin-like growth factor 1 (IGF-1) was previously believed to act systemically in regulating fetal bone development, but has since been shown to act on fetal bone in a paracrine, rather than endocrine, manner (Robson et al, 2002). Whether or not it also acts systemically to regulate fetal bone development is presently unclear. In addition to local and systemic hormones, the maintenance of maternal/fetal calcium homeostasis is also essential for normal fetal bone development (Chalon & Garel, 1985; Lima et al, 1993; Loughead et al, 1990; Rebut-Bonneton et al, 1983a, 1983b; Sinclair 1942). Maternal/fetal hypocalcemia results in decreased fetal bone mineralization (Chalon & Garel, 1985; Lima et al, 1993; Sinclair, 1942). Major regulators of fetal calcium homeostasis include PTH and PTHrP (Kovacs et al, 1996, 2001b), and there is evidence that vitamin D is also important (Rummens et al., 2002). Vitamin D deficiency during pregnancy in the guinea pig results in an expansion of the HZ compared with fetuses whose mothers are fed a vitamin D replete diet (Rummens et al, 2002). This defect is corrected by feeding pregnant dams a calcium supplemented diet, indicating that vitamin D exerts its effects on fetal bone via its affects on calcium homeostasis (Rummens et al, 2002). 28 1.2.4.2 Ethanol's Effects on Skeletal Development While much research into FASD has focused on the CNS, behavioral and cognitive effects, there is evidence that prenatal ethanol exposure also affects the developing skeleton. Children with FAS have a delayed mean bone age (Habbick et al, 1998), which is evaluated by comparing the size and shape of skeletal elements viewed on a radiograph of the hand to reference radiographs, and is used to assess the skeletal maturity in children. In the rat, prenatal ethanol exposure has been shown to decrease fetal body weight, total body length and individual bone length, and to reduce or delay skeletal ossification, as assessed by Alizarin red and Alcian blue staining (Keiver et al, 1996, 1997; Keiver & Weinberg, 2004; Lee & Leichter, 1983; Weinberg et al, 1990). As in the human, the effects of prenatal ethanol exposure on bone in the rat persist into postnatal life; a maternal intake of 30% v/v in drinking water has been shown to result in decreased bone size and skeletal maturity scores in 2-4 week old rat pups (Lee & Leichter, 1980; Leichter & Lee, 1979; Ludeha et al., 1983). Like bone age in children, skeletal maturity scores in rats are determined by examining radiographs and comparing the size and shape of skeletal elements. However, unlike in humans, the entire skeleton is assessed (Hughes and Tanner, 1970). As no studies have examined the effects of prenatal ethanol exposure on bone beyond age 14 in humans or 4 weeks of age in rates, it is unclear if bone development is merely delayed (and thus will eventually catch up) or if it is permanently retarded. In early studies on ethanol's effects on skeletal development, ethanol was administered in the drinking water provided to the animals (Lee and Leichter 1980, 29 1983; Leichter and Lee, 1979; Ludeha et al., 1983). As this method of administration can result in reduced maternal water and food intake, it was hypothesized that ethanol's effects on bone development might be due to dehydration and malnutrition. However, it has since been shown that although inadequate nutrition can exacerbate ethanol's effects, providing adequate maternal nutrition and hydration does not eliminate its effects on the developing skeleton. Prenatal ethanol exposure retards skeletal development in studies in which ethanol is administered to rats in a liquid diet (Weinberg et al., 1990; Keiver et al., 1996, 1997; Keiver and Weinberg, 2003), which does not result in dehydration and minimizes and controls for reduced maternal nutrient intake. Moreover, Weinberg and co-workers demonstrated that increasing maternal protein intake above recommended levels could ameliorate some, but not all, of the effects of ethanol (36% EDC) on fetal skeletal ossification (Weinberg et al, 1990). The effects of prenatal ethanol exposure on postnatal skeletal development are also not likely to be due to decreased neonatal nutrition. Although maternal ethanol intake during gestation could affect neonatal nutrient intake through effects on milk quantity and quality during lactation, Lee and Leichter (1980) demonstrated no significant differences in bone size or skeletal maturity scores of pups raised in litters of three compared with litters of eight. Despite similar levels of maternal ethanol intake and methods of administration, there is variability among studies in the degree of severity of ethanol's effects on fetal skeletal development. Our lab has previously found that 5 wk (2 wk before and 21 d throughout gestation) of maternal ethanol intake at 36% 30 EDC (Keiver et al, 1997) resulted in much more severe effects on fetal skeletal development than 3 wk (21 d gestation only) at the same dose (Keiver et al, 1996). We hypothesized that the severity of the effects of maternal ethanol intake on fetal skeletal development depended on duration of intake, but results of a subsequent study designed to examine this hypothesis suggested that the variation is likely due to differences in maternal/fetal BECs achieved, rather than duration per se (Keiver & Weinberg, 2004). The animal studies discussed above have all utilized fairly high levels of ethanol administration (36% EDC in liquid diet or 30% v/v ethanol in drinking water) and model a continuous pattern of maternal ethanol intake. There is only one preliminary report on the effects of different doses of ethanol administered in a binge pattern on fetal bone (Given et al., 2004). In that study, prenatal ethanol exposure had effects on the bone strength of fetal sheep that varied in a dose-dependent manner, with lower doses actually increasing bone strength (Given et al., 2004). No study to date, however, has examined the question of whether the effects of ethanol on the fetal skeleton reflect generalized growth retardation (or recovery from transient growth retardation) or if ethanol specifically affects the process of bone formation. Only one study has examined the effects of prenatal ethanol exposure on the morphology of the growth plate during postnatal bone development. Using a rat model, Miralles-Flores and Delgado-Baeza (1992) found that a maternal ethanol intake of 36% EDC for 51 d (30 d prior to and throughout 21 d gestation) resulted in a decrease in total tibial length, but had no effect on the organization of the zones 31 within the growth plate at birth. However, they did observe a decrease in the length of the HZ, which resulted in a decrease in the length of the total growth plate and a decrease in total bone length, on postnatal d 15. These results suggest that ethanol does have specific effects on bone development, but it is unclear if the effects are limited to postnatal development or could also account for the skeletal effects seen in the fetus with ethanol exposure. The implications of the effects of prenatal ethanol exposure on the developing skeleton for long-term bone health (e.g., risk of osteoporosis later in life) has not been investigated and is an important area for investigation. 1.2.5 FETAL PROGRAMMING/FETAL ORIGINS OF ADULT DISEASE Evidence is accumulating that the intrauterine environment can exert profound effects on the long-term health of an organism. Originally proposed by David Barker, the \"Fetal Origins of Adult Disease\" hypothesis (alternatively referred to as the \"Barker Hypothesis\" or the concept of \"fetal programming\") suggests that conditions (e.g., nutrition) during fetal life can have profound effects on long-term health. The hypothesis centres around the concept of developmental plasticity, a phenomenon where differences in environmental factors during critical periods of embryonic and fetal development can result in a range of different physiological or morphological states for a given genotype (Barker, 1998, 2004; Cooper et al, 2002; Hanson et al, 2004; Harvey & Cooper, 2004). Although the response may be adaptive for the fetus during those particular conditions, they may predispose the individual to disease later in life when the conditions are no longer present 32 Prenatal environmental factors that result in low birth weight have been shown to be associated with increased risk of heart disease, hypertension and type II diabetes later in life (Barker, 1998; Hanson et al, 2004; Leon 1998). Moreover, there is evidence that the skeletal system may also be subject to programming and thus, the environment experienced in utero may influence the risk of developing osteoporosis in later life (Cooper et al, 2002; Harvey & Cooper, 2004). 1.2.5.1 Potential Mechanisms of Fetal Programming There are several possible mechanisms by which the in utero environment may program the offspring. First, the fetal environment can have permanent effects on gene expression through effects on DNA methylation. For example, reduction in uterine blood flow results in IUGR and alterations in the DNA methylation status of genes in the kidneys of the offspring that are thought to underlie the hypertension seen in these animals in adulthood (Pham et al, 2003). Second, the fetal environment influences cell number. Maternal protein restriction results in IUGR in which offspring possess significantly smaller hearts, due to a significant decrease in cardiomyocytes compared to controls (Corstius, 2005). This reduction in cardiomyocyte number is permanent and may result in impaired cardiac function later in life (Corstius, 2005). Third, the intrauterine environment also influences the offspring's hormone level. For example, prenatal exposure to glucocorticoids has been shown to permanently alter basal hormone levels as well as hormonal responses to stimuli (Fowden & Forhead, 2004). These permanent effects on the functioning of the endocrine system may have long-term effects on the health of the offspring. Evidence suggests that the structure and function of cell receptors may be 33 permanently altered as a result of alterations in hormone exposure in utero. For example, maternal protein restriction during gestation results in elevated maternal and fetal glucocorticoid levels, and increased numbers of glucocorticoid receptors (GR) in the kidney of offspring during adulthood (Bertram et al, 2001). This increase in GRs may mediate the hypertension seen in these animals in adulthood (Bertram etal, 2001). 1.2.5.2 Fetal Programming and Bone Health Osteoporosis is a skeletal disease characterized by low bone mass and bone microarchitecture deterioration that results in bone fragility and susceptibility to fractures (Genant et al, 1999). Risk of osteoporosis is dependent on both the peak bone mass achieved during young adulthood and the subsequent rate of bone loss later in life. Peak bone mass is achieved by 20-30 years of age in humans (Eastell, 1999). After menopause, women experience bone loss that occurs in two phases. First, a phase of rapid bone loss that lasts approximately 5 years occurs as a result of decreased estrogen levels. This rapid phase is followed by a slower phase of bone loss, which is attributed to age-related factors such as increased PTH levels and osteoblast senescence (Eastell, 1999). Men also experience the slow phase of bone loss starting at approximately 55 years of age (Eastell, 1999). Thus, prevention of osteoporosis requires a slowing of the bone loss in later life and/or maximizing the peak bone mass achieved in young adulthood. Factors that affect peak bone mass and/or the response of bone cells to stimuli, including programming of skeletal cells during fetal life, are important in decreasing the risk of osteoporosis later in life. 34 Epidemiological studies provide correlative evidence of a link between body weight and later skeletal health. Low birth weight and body weight at 1 year of age are associated with low bone mineral content (BMC) later in life (Cooper et al, 1995, 1997; Dennison et al, 2005; Gale et al, 2001) and a low rate of childhood growth has been shown to be associated with the risk of later hip fracture (Cooper et al, 2001). Moreover, animal studies provide more definitive evidence of a causal relationship between prenatal environment and long-term effects on bone health. Maternal protein restriction during pregnancy in the rat results in decreased BMC and increased epiphyseal growth plate height in the offspring in adulthood (Mehta et al, 2002). As well, administration of interleukin-1 (IL-1) to pregnant dams results in significantly shorter tibia in male and female offspring, and significantly decreased height, areal bone mineral density (BMD) and BMC in the vertebrae of male offspring, at 10-12 weeks of age (Swolin-Eide et al, 2004). Moreover, adult mice that were exposed to a synthetic estrogen during prenatal life exhibited increased bone mass, resulting from a decrease in osteoclast number and increased mineral apposition rate (Migliaccio et al, 1996, 2000). In the adult mice, circulating estrogen levels were not altered by prenatal estrogen exposure, indicating that in utero estrogen exposure permanently programmed the bone cells. The effects were only partially ameliorated by prepubertal ovariectomy (Migliaccio et al, 1996, 2000), suggesting that the response of bone cells to estrogen deficiency in the adult mice was modified by prenatal estrogen exposure. Thus, changes in the prenatal environment may alter the programming of bone cells during fetal life and result in long-term effects on bone. 35 The mechanisms by which the prenatal environment may program bone development are not yet known, but it has been suggested that alterations in the set point of the growth hormone (GH)-IGF-1 axis (Dennison et al, 2003; Fall et al, 1998) and/or the hypothalamic-pituitary-adrenal (HPA) axis may be involved (Harvey and Cooper, 2004) may be involved. In addition, it has been hypothesized that the stimulation of PTH/PTHrP activity in the fetus, in response to calcium deficiency, may have permanent effects on the growth trajectory of bone during childhood (Tobias & Cooper, 2004). As prenatal ethanol exposure has been shown to affect fetal bone development and there is evidence that the skeletal system is susceptible to fetal programming, the effect of prenatal ethanol exposure on fetal bone development could potentially increase the offspring's risk of osteoporosis later in life. For example, if ethanol's effects on fetal bone are merely a reflection of general growth retardation, it may result in a decrease in peak bone mass achieved and thus, an increased risk of osteoporosis. Alternatively, or in addition, if ethanol exerts specific effects on the fetal skeleton and those effects program the bone in a way that alters the rate of bone loss later in life (such as programming its response to decreased estrogen), this could also increase osteoporosis risk. Osteoporosis significantly decreases quality of life and is estimated to cost the Canadian health care system $1.3 billion per year, with projections that this cost will rise to $32.5 billion by 2018 (Osteoporosis Canada, 2005). The possibility that the in utero environment may increase the risk of osteoporosis during later life thus represents an important area of research. 36 1.2.6 POTENTIAL M E C H A N I S M S O F E T H A N O L ' S E F F E C T S ON F E T A L B O N E The mechanisms by which ethanol may exert its effects on fetal bone development are unknown. Early in the field of FAS research, there was an interest in finding a single underlying mechanism for ethanol's effects on the fetus. There was speculation that maternal/fetal undernutrition, rather than alcohol per se, was the causative factor, but research with animal models has shown that alcohol itself is a teratogen (Weinberg, 1996). Pair-feeding regimens, which control for reduced maternal food intake resulting from ethanol consumption, demonstrate that maternal nutrition does not account for the effects of ethanol on the developing fetus, although undernutrition can exacerbate ethanol's effects. As ethanol can interfere with placental function, the possibility is raised that ethanol may exert its effects via placental insufficiency. However, this cannot fully explain ethanol's effects because models that eliminate the placenta as a factor, such as in vitro and chick model studies, also demonstrate that alcohol affects development. Furthermore, ethanol affects a number of systems, including maternal and fetal endocrine systems, protein expression, and cell proliferation and function, and thus likely is acting via more than one mechanism. Ethanol freely crosses the placenta (Guerri & Sanchis, 1985) and therefore may exert direct effects on the cells involved in bone development. For example, ethanol has been shown to inhibit osteoblast proliferation and differentiation in vitro (Chavassiex et al, 1993; Farley et al, 1985; Klein, 1997) and thus prenatal ethanol exposure may affect fetal bone development through direct effects on osteoblasts. In addition, chronic ethanol exposure in the adult results in increased total bone and 37 bone marrow fat content in both humans and animals (Sampson et al, 1996, 1998; Wezeman & Gong, 2001, 2004; Gong & Wezeman, 2004), and ethanol has been shown to inhibit osteogenic differentiation (Gong & Wezeman, 2004) and increase adipogenesis (Wezeman & Gong, 2004) in mesenchymal stem cells. Interestingly, preliminary data show that the tibia of fetal guinea pigs prenatally exposed to ethanol also has increased fat content (Keiver and Brien, unpublished data), suggesting that ethanol may alter the balance between adipogenesis and osteogenesis in the fetus. Disruption in the expression of any of the many local factors that regulate the process of fetal bone development results in impairments in bone development; thus, ethanol may exert its effects on fetal bone development via effects on local factors. For example, as the initial size of a condensation is a reflection of the growth potential of the resultant bones (Hall & Miyake, 2000) and exposure to high levels of ethanol has been shown to significantly reduce bone size (Lee & Leichter, 1983; Miralles-Flores and Delgado-Baeza, 1992), it is possible that some of the factors regulating mesenchymal condensation size may be involved in mediating ethanol's effect on bone size. Similarly, if specific stages of chondrocyte development during endochondral bone development are found to be disrupted by ethanol, it raises the possibility that the factors regulating that stage are involved in mediating ethanol's effects. However, almost nothing is known regarding specific effects of ethanol on fetal bone morphology, and the effects on levels of local factors in fetal bone have not been investigated. In addition to possible direct effects, ethanol may also exert indirect effects on bone through effects on systemic hormones or nutritional factors. Ethanol 38 consumption affects the hormone systems known to be involved in fetal skeletal development. For example, ethanol appears to impair insulin signaling (Wan et al, 2005) and thus could exert its effects through the effect of insulin on chondrocyte proliferation (Fowden, 1995). Ethanol exposure during pregnancy also results in decreased maternal PTH levels, despite the presence of maternal hypocalcemia (Keiver & Weinberg, 2003), and ethanol-exposed fetuses appear to exhibit impairment in the ability of PTH to correct the resultant fetal hypocalcemia (Keiver et al, 2004). As PTH is essential to fetal bone development (Miao et al, 2002), ethanol may be exerting its effects via effects on PTH function. Furthermore, maternal ethanol intake increases maternal corticosterone levels and has been shown to program the HPA axis of the fetus, permanently altering the offspring's stress response (Zhang et al, 2005). As glucocorticoids can alter both pre- and postnatal bone growth, ethanol may exert effects on bone through its effects on the HPA axis. Maternal ethanol intake also results in maternal and fetal hypocalcemia (Baran et al, 1982; Keiver et al, 1996; Keiver & Weinberg, 2003, 2004) and, as hypocalcemia has been shown to decrease fetal bone mineralization (Chalon & Garel, 1985; Lima et al, 1993; Sinclair, 1942), this may underlie some of ethanol's effects on fetal bone. However, it is not yet known if the ethanol-induced fetal hypocalcemia is sufficient to result in decreased fetal bone mineralization or if the ethanol-induced disruptions of maternal/fetal PTH functioning are sufficient to result in impaired fetal bone development. Furthermore, the effect of prenatal ethanol exposure on PTHrP, which is also involved in the regulation of both fetal calcium homeostasis and bone development, has not been investigated. 39 Another possible mechanism by which ethanol may affect fetal bone is via oxidative stress (Goodlett et al, 2005). Oxidative stress is a condition in which there is a high level of reactive oxygen species (such as superoxide, hydroxyl ion, hydrogen peroxide), which result in damage to cells. Ethanol can induce oxidative stress by causing mitochondrial dysfunction, by the oxidation of ethanol by cytochrome P-450 2E1, or during the oxidation of acetaldehyde (a breakdown product of ethanol) (Goodlett et al, 2005). Moreover, oxidative stress has been shown to be associated with impairments in skeletal ossification. For example, lipopolysaccharide treatment, which induces oxidative stress, has been shown to result in IUGR and retarded ossification. Co-treatment with ascorbic acid, an antioxidant, has been shown to ameliorate these effects (Chen et al, 2006). Further evidence of the effects of oxidative stress on developing bone has been demonstrated in the case of maternal diabetes. Fetuses of diabetic rats exhibit delayed ossification (Verhaeghe et al, 1999) and treatment with ascorbic acid ameliorates the reduction in skeletal ossification in these fetuses (Braddock et al, 2002). Clearly, there are a number of possible mechanisms by which prenatal ethanol exposure may affect the developing skeleton and it is quite possible that ethanol is acting via multiple mechanisms. Since very little research has been conducted on the effects of prenatal ethanol exposure on developing bone, it is important to closely examine ethanol-exposed bone to determine what specific effects ethanol may have, in order to direct future work in this field. 40 1.3 THESIS OBJECTIVE AND HYPOTHESES The objective of this thesis was to determine if prenatal ethanol exposure has specific effects on bone development, in addition to its effects on general growth, in the fetal rat. The first study was designed to test the hypotheses that: (1) prenatal ethanol exposure affects skeletal development at doses of ethanol lower than those required to affect general growth (assessed by fetal body weight and length) (2) skeletal sites differ in their sensitivity to the effects of prenatal ethanol exposure If the effects of ethanol that have been previously observed in the fetal skeleton are simply the result of general growth retardation, then we would expect the degree of the effects on body weight, length and ossification at different skeletal sites to be the same (i.e., that each of these parameters would be delayed to the relative same extent). In contrast, if the effects are not merely a reflection of general growth retardation, we would expect these hypotheses to be supported. To test these hypotheses, rat dams were fed ethanol at 15%, 25% and 36% EDC, doses designed to approximate low, moderate and high levels of maternal drinking, respectively. The normal pattern of development of body weight and ossification at various skeletal sites was determined across the final week of gestation in control fetuses, and then compared with that of ethanol-exposed (and pair fed) fetuses at the end of gestation. The results of the first study provide evidence that prenatal ethanol exposure has effects on endochondral ossification 41 that are independent of ethanol's effects on general growth. Furthermore, the results suggest that ethanol may be affecting the later, rather than the earlier stages of endochondral ossification. The second study was therefore designed to extend these observations and provide further evidence that ethanol has specific effects on bone development by testing the hypothesis that: (1) prenatal ethanol exposure disrupts the morphology of the growth plate of the fetal tibia. Since the purpose of this study was to take a first step in examining the effect of ethanol on the morphology of the fetal growth plate, the tibia was chosen as it was found to be highly affected in both Experiment #1 and in previous studies in our lab (Keiver et al, 1996; Keiver & Weinberg, 2004). If ethanol's effects on the skeleton are simply a reflection of its effects on general growth, we would not expect to see a disruption in the morphology of the growth plate. Specifically, we would not expect alterations in the later stages of bone development. To test this hypothesis, bone from fetuses exposed to a maternal intake of 36% EDC was examined histologically for evidence of disruptions in the histological zones of the growth plate. 42 C H A P T E R 2: M E T H O D S 2.1 Experiment #1 2.1.1 Breeding and Feeding of Animals Three-month-old virgin female Sprague Dawley rats (Charles River, St. Constant, Quebec) were weighed (initial body weights 238-283 g, mean 264 g) and assigned to their respective diet groups. Rats were maintained in temperature-controlled rooms with lights on between 0600 and 1800 h. The diets (BioServ Inc., Frenchtown, NJ) were administered for 6 wk: 3 wk prior to breeding and throughout 21 d of gestation. Diet was presented daily at 1700 h. Food intake was measured daily and rats were weighed weekly throughout the experiment. Mean time for rats to breed was 3 d (range = 1-5 d). Rats received lab chow ad libitum while in the breeding cages. Providing the ethanol-containing diet in the breeding cage would allow the male rats to consume ethanol and paternal ethanol intake has been shown to affect fetal growth (Bielawski et al, 2002). The appearance of a vaginal plug was considered day 1 of gestation. The University of British Columbia Animal Care Committee approved the experimental procedures (Appendix D). Rats were placed into one of seven weight-matched groups: three Ethanol (E) groups, which were fed a liquid diet containing ethanol at a level of 15%, 25% or 36% EDC (E15 (n=15), E25 (n=15) and E36 (n=26), respectively); three Pair-Fed (PF) groups, which received an isocaloric liquid control diet (with the ethanol calories replaced by maltose-dextrin) in amounts equivalent to those consumed by rats in the ethanol groups (PF15 (n=15), PF25 (n=15) and PF36 (n=25), respectively); and one 43 Control (C) group (n=25), which received liquid control diet ad libitum. The PF rats were matched to their respective E partners by day of gestation: the same amount of diet (g diet/kg body weight) consumed by an E rat on day 1 of gestation was fed to its PF partner on day 1 of its gestation. As such, the pregnancy of a PF rat was always behind its E partner by at least one calendar day. Figure 2.1 shows the experimental feeding regimen for Experiment #1. The doses of ethanol were chosen to approximate low (15% EDC), moderate (25% EDC), and high (36% EDC) levels of maternal drinking. The 36% EDC is the dose of ethanol that has been previously used in our lab and thus allows for comparisons with our previous work. As previously discussed, in order to place animal models in the context of the human situation, the dose of ethanol per se is less relevant than the BEC achieved. The 36% EDC dose typically yields peak BECs in pregnant rats of 100-200 mg/dL (Keiver & Weinberg, 2003; Weinberg, 1985; Weinberg et al, 1990), which represents a level that would correspond to higher than the legal BEC limits for drunk driving (50-100 mg/dL) in Canada and the United States. Also of significance with regards to the method of ethanol administration is the pattern of ethanol intake and how well this pattern mirrors the human situation. Previous research with the liquid diet method of ethanol administration in rats has demonstrated a circadian feeding pattern consisting of two major periods of feeding, one when the dams are first presented with their daily ration (just prior to lights out) and another several hours later, around 0300-0700 h, with low levels of feeding throughout the rest of the day and night (Weiner et al, 44 E15 PF15 3 weeks prior to breeding 15% EDC diet ad lib control diet - pair-fed to E15 breed 3 weeks gestation 15% EDC diet ad lib control diet-pair fed to E15 =3 E25 25% EDC diet ad lib PF25 control diet - pair-fed to E25 E36 36% EDC diet ad lib PF36 control diet - pair-fed to E36 25% EDC diet ad lib control diet-pair fed to E25 36% EDC diet ad lib control diet-pair fed to E36 control diet ad lib control diet ad lib Figure 2.1: Experimental feeding regimen for the rat dams for Experiment #1. Dams were fed a liquid diet containing ethanol at a level of 15%, 25% or 36% ethanol-derived calories (E15, E25 and E36, respectively) or a liquid control diet (Pair Fed, PF, and Control, C, groups) for 6 weeks. 45 1981). This level and pattern of ethanol intake models a chronic, steady intake rather than a binge pattern of intake. 2.1.2 Determination of Blood Ethanol Concentrations Peak BECs were determined in E rats just prior to breeding (termed week 0 of gestation) and at approximately 1, 2 and 3 wk of gestation. To ensure that the sampling procedure did not have adverse effects on food intake of the pregnant rats, no more than two samples were obtained from each individual rat (Martin et al, 1978). Thus, samples were obtained from approximately half of the rats from each E group at 0 and 2 wk of gestation and from the other half at 1 and 3 wk of gestation. Blood samples were also taken from PF and C rats to ensure consistent treatment among groups, and a subset of these blood samples were analyzed for BECs as negative controls. Blood samples for determination of peak BEC were obtained by tail nick from the dams and collected into heparinized micro-capillary tubes 2-4 h after lights were turned off at 1800 h. This sampling time was just after the main period of feeding and thus corresponded approximately to the peak BEC achieved in dams in this model (Weiner et al, 1981). The BECs were also measured in blood taken from dams at termination, between 0800 and 1200 h, a few hours after the early morning period of feeding. This blood was collected by cardiac puncture into syringes without anticoagulant. All blood was centrifuged at 4QC and serum or plasma stored at -70QC until analyzed for BEC (kit #A7504-150, Pointe Scientific, Lincoln Park, Ml). In this assay, serum and plasma from PF and C rats (negative controls) had BEC values of 0-22 mg/dL; therefore values <25 mg/dL were considered to be negative 46 for ethanol (or undetectable). Details of the procedure used to assess BEC are given in Appendix A. 2.1.3 Termination of Dams and Fetuses Fetuses were collected on d 21 of gestation (approximately 1.5 d prior to birth), with the exception of those collected on d17-20 of gestation (discussed below), between 0800 and 1200 h. Dams were lightly anesthetized with ether and maternal blood was collected by cardiac puncture for terminal BEC analysis. Dams were then terminated by decapitation and fetuses euthanized by cervical dislocation. Fetuses were counted and the litter weight determined. 2.1.4 Fetal Growth and Skeletal Development Fetuses were randomly selected (without reference to sex) from each litter and measured for length (nose-to-rump, n=3/litter) or stained (n=2/litter) with alcian blue and alizarin red for cartilage and bone, respectively. Random selection of fetuses consisted of blindly picking fetuses from the group of fetuses contained in the weigh boat after weighing. The process of clearing and staining fetuses is very time-consuming and laborious and so length and ossification measures were not conducted on all the fetuses within each litter. Previous research on the effect of prenatal ethanol exposure on growth has shown that ethanol affects all fetuses in a litter equally (i.e., the variability among ethanol-exposed fetuses is no larger than 47 among control fetuses) (Hannigan et al, 1993), and this was confirmed by measuring the coefficients of variation (CVs) among littermates in this study. The staining procedure used has been previously described (Weinberg et al, 1990) and is a modification of the method of MacLeod (1980). Briefly, fetuses were fixed in 95% ethanol for 2-3 h, skinned and eviscerated, further fixed in 95% ethanol for 5 d, de-fatted in acetone, stained with alcian blue and alazarin red, and then cleared with potassium hydroxide. Fetuses were then taken through a series of potassium hydroxide/glycerine mixtures to replace the potassium hydroxide with glycerine in order to stop the clearing and prepare the fetuses for storage in 100% glycerol until analysis. More detailed methodology for the alcian blue/alazarin red staining and clearing procedure is provided in Appendix B. Ponderal index (weight/length3) was calculated to examine the effect of ethanol on the relationship between fetal body weight and length (Han, 1999). The percent ossification (defined as the length of mineralized portion total bone length x 100) of the fetal bones (scapula, humerus, ulna, radius, femur and tibia) was measured by a single investigator blind to treatment group using a Zeiss dissecting microscope with a linear eyepiece raticule. The length of the mineralized portion of the bone was that part of the diaphysis which had taken up the alizarin red stain. Alizarin red stains for calcium, so this portion of the bone consists of both the bone tissue and the mineralized portion of the cartilage matrix. The number of ossification centres was determined for the sternum, sacrum and metatarsals by counting individual ossification centres, which had taken up the alizarin red stain, 48 within each of these skeletal elements. The value for each litter represents the mean of two (ossification measures) or three (body length) fetuses. In addition to the d 21 C rats, the normal pattern of development for fetal body weight and skeletal ossification was determined in fetuses (n=2-5/litter) collected from additional C dams (n = 11) on days 17, 18, 19 or 20 of gestation (n=2-4 litters/day). Fetuses were counted, weighed and stained for analysis of skeletal ossification as described above. Day 21 E and PF fetuses were compared with the d 17-21 (for fetal weight) or d 18-21 (for ossification, as there was no measurable ossification in any bones for d 17 fetuses) C fetuses. Normal development curves were generated and the equations of those curves were determined. The equation of each curve was used to determine if the measured parameters differed in their sensitivity to ethanol's effects, or if ethanol caused a uniform delay in development. 2 . 1 . 5 Statistical Analyses The numbers of rats per group were based on the number of rats required in previous studies in our lab to obtain significant results for measures of fetal growth and ossification (Keiver et al, 1996, 1997; Keiver & Weinberg, 2004). Group differences, with the exception of number of ossification centres, were determined by analysis of variance (ANOVAs), followed by 2-tailed Newman-Keuls post-hoc tests where warranted. According to the convention outlined by Zar (1984), post-hoc tests were conducted only for significant main effects or interactions. Thus, for 2-way ANOVAs, comparisons between individual group means were only made if the interaction term between factors was significant. If the interaction term was not 49 significant, post-hoc tests were conducted only for the significant main effects (collapsed across the other factor). The effects of diet and dose of ethanol were analyzed by 2-way ANOVAs for the factors of diet (E and PF only) and dose (15%, 25%, 36% EDC). Because there was only a single C group, comparisons among E, PF and C diets were analyzed by separate 1-way ANOVAs for each of the 15%, 25% and 36% EDC exposure levels. Since the number of ossification centres in the sternum, sacrum and metatarsals were derived from nominal data (i.e., presence or absence of ossification centre), nonparametric tests were used to analyze these data. Group differences for the number of ossification centres were determined by the Kruskal-Wallis test, followed by a two-tailed Tukey-like nonparametric multiple comparison test (Zar, 1984). The effects of diet and dose of ethanol consumption on number of ossification centres were analyzed by 2-way Kruskal-Wallis tests for the factors of diet (E and PF only) and dose (15%, 25%, 36% EDC). Comparisons among E, PF and C diets were analyzed by separate 1-way Kruskal-Wallis tests for each of the 15%, 25% and 36% EDC exposure levels. Repeated measures ANOVAs were conducted to determine the effects of week of gestation on maternal ethanol intake (wk 0,1,2 and 3) and on peak BEC (0 vs. 2, 1 vs. 3 wk) because these measures were obtained from the same individual rats across time. Maternal food intake was assessed using a repeated measures ANOVA with factors of time (repeated measure; wk 1, 2, 3, 4, 5, and 6 of feeding), diet (E and PF), and dose (15%, 25%, 36% EDC). Post-hoc tests on individual group 50 differences were conducted for maternal food intake in the absence of a time X diet X dose interaction because the hypothesis that each E group would not differ significantly from its respective PF group for any given week was made a priori (Zar, 1984). Because there was only a single C group, comparisons among E, PF and C diets were analyzed by separate repeated measures ANOVAs for each of the 15%, 25% and 36% EDC exposure levels. Terminal BECs were undetectable for the E15 group, and so an unpaired, two-tailed t-test was used to compare terminal BEC samples between E25 and E36 dams. Relationships between fetal body weight or ossification and day of gestation in C fetuses were analyzed by regression analyses. Coefficients of determination are presented as r2 values. Outliers (defined as values exceeding \u00C2\u00B1 2 SD from the mean) were eliminated from data sets. Analysis without the elimination of outliers did not substantially change our results. Level of significance was p < 0.05. Values presented are the mean \u00C2\u00B1 SE. 2.2 Experiment #2 2.2.1 Breeding and Feeding of Animals Housing and feeding procedures for rats in Experiment #2 were as described for Experiment #1, with the exception that only the high dose of ethanol (36% EDC) was used. Rats were assigned to one of three weight-matched groups (initial body weights 235-286 g, mean 252 g): Ethanol (E, n = 13), which were fed a liquid diet containing 36% EDC; Pair-Fed (PF, n = 12), which were pair fed as described for 51 Experiment #1; and Control (C, n = 12), which received liquid control diet ad libitum. Figure 2.2 shows the experimental feeding regimen for Experiment #2. Mean time for rats to breed was 3 d (range = 1-7 d). Ethanol's effects on growth and ossification are subtle, and the preparation and histological analysis of fetal bone is very time-consuming and laborious, so only the highest dose was chosen to maximize the possibility of detecting effects, as this study was intended to examine if ethanol disrupts the morphology of the tibial growth plate. It was decided that assessing the bones at the lower doses without first examining effects at the highest dose was not warranted. 2.2.2 Determination of Peak Blood Ethanol Concentrations Peak BECs were determined as described for Experiment #1 in a subset of E rats (n = 5) on d 10-11 of gestation. Serum was analyzed for BEC (kit# 333-B, Sigma Diagnostics, St. Louis, MO). In this assay, serum from PF and C rats (negative controls) have BEC values of 0-22 mg/dL; therefore values <25 mg/dL were considered to be negative for ethanol (or undetectable). Details of the procedure used to measure BEC are given in Appendix A. 2.2.3 Histology Two fetuses from each litter were randomly selected (without reference to sex) and the left tibiae dissected free from soft tissue under a Zeiss dissecting microscope. As in Experiment #1, random selection of fetuses consisted of picking fetuses, without looking at them, from the litter of fetuses that had been placed in a weigh boat for weighing. 52 E PF C 3 weeks prior to breeding 36% EDC diet ad lib control diet - pair-fed to E control diet ad lib breed 3 weeks gestation 36% EDC diet ad lib control diet-pair fed to E control diet ad lib Figure 2.2: Experimental feeding regimen for the rat dams for Experiment #2. Dams were fed a diet containing 36% ethanol-derived calories (EDC) (E) or a control diet (Pair Fed, PF, and Control, C, groups) for 6 wk. 53 Since the purpose of this study was to take a first step in examining the morphology of ethanol-exposed fetal bone, the tibia was chosen as it was found in Experiment #1, as well as previous studies in our lab (Keiver et al, 1996; Keiver & Weinberg, 2004) to be highly affected by prenatal ethanol exposure. In addition, the tibia was assessed in the only other study to examine the effect of prenatal ethanol exposure on growth plate morphology (Miralles-Flores & Delgado-Baeza, 1992) and so examining the tibia in our study allowed us to compare our results with that study. As well, fetal bones at d 21 of gestation are extremely fragile and it is difficult and time-consuming to dissect out and prepare these bones for histological examination; tibia is more robust than other highly affected bones (such as the ulna or radius) and thus was chosen due to these practical considerations. Tibiae were fixed in formalin for 3 d, decalcified in 10% w/v ethylenediaminetetraacetic acid (EDTA, pH 7.4) for 5 d, rinsed in four 1 h washes of 70% ethanol, and embedded in paraffin. Care was taken to ensure that all bones were oriented in the same direction during embedding to minimize differences in the angle at which the bones were sectioned and a single investigator processed all bones to maintain consistency. Tibiae were sectioned (5 um thickness) longitudinally through the entire bone and sections were stained with hematoxylin and eosin for histological examination. The longest section of each tibia was selected for analysis. Photomicrographs (Retiga EX Digital CCD monochrome camera) were taken under light microscopy (Axioscop 2 Microscope) at 31.5X magnification for assessment of total bone and diaphyseal length, diaphyseal width, and distal epiphyseal length, and at 63X magnification for assessment of the lengths and areas of the different 54 chondrocyte zones in the proximal epiphysis. Photomicrographs were imported into Adobe Photoshop, in which the parameters to be measured were delineated based on cell morphology. Figure 2.3 shows the boundaries used for the morphometric analyses. Figure 2.3A illustrates the boundaries for the lengths of the total bone, diaphysis and distal epiphysis, and the width of the diaphysis. Total bone length and diaphyseal length were measured centrally along the long axis of the bone, from end to end for total bone length, and from the distal to the proximal epiphyses for diaphyseal length. The length of the distal epiphysis was determined from the mean of five measurements made parallel to the long axis of the bone, taken at equal intervals across the width of the bone, from the end of the epiphysis to the bottom of the hypertrophic zone. The width of the diaphysis was measured perpendicularly to the long axis of the bone at its longitudinal midpoint. Figure 2.3B shows the boundaries for the measurements of the lengths and areas of the resting, proliferative, and hypertrophic zones of the proximal epiphysis. Standard morphometric analyses based on Howell & Dean (1992) were employed to measure the lengths of the resting, proliferative and hypertrophic zones. Specifically, the resting zone was defined as the top of the epiphysis to the top of the proliferative zone, the proliferative zone was defined as the first flattened chondrocyte to the top of the hypertrophic zone, and the hypertrophic zone was defined as the first rounded hypertrophic chondrocyte to the last rounded hypertrophic chondrocyte (Howell & Dean, 1992; Miralles-Flores & Delgado-Baeza, 1992). The length of each zone was determined from the mean of five measurements made parallel to the long axis of the bone, taken at equal intervals 55 Figure 2.3: Morphometric parameters analyzed in fetal rat tibia stained with hematoxylin and eosin. (A) Total bone length, width, diaphyseal length and distal epiphyseal plate height were measured under 31.5X magnification. (B) The boundaries of the resting (RZ), proliferative (PZ) and hypertrophic (HZ) zones were delineated based on cell morphology under 63X magnification. The height of each zone was determined from the mean of five measurements per zone. (RZ, dotted line; PZ, dashed line; HZ, solid line). Scale bar in A and B represent 400 um and 200 pm, respectively. 56 across the width of the bone (Miralles-Flores and Delgado-Baeza,J992). The total length of the proximal epiphysis was calculated from the sum of the lengths of the resting, proliferative and hypertrophic zones. The area of each zone was also determined using the boundaries delineated as described above (shown in Figure 2.3B). All parameters were measured using Image J software (Rasband, 2005). The value for each litter represents the mean of two fetuses and the n per group represents the number of litters. A single investigator, blinded to treatment group, performed all measurements. To determine intra- and inter-rater reliability of the measures, measurements of total bone length, bone width and diaphysis length (n = 9) and resting, proliferative and hypertrophic zone length (n = 22) were performed on a subset of the tibiae by the same investigator at a later date, and by a different investigator, respectively. 2.2.4 Statistical Analyses Data analyses followed the same approach as that described for Experiment #1. Treatment (diet) differences were determined by 1-way ANOVAs, followed by 2-tailed Newman-Keuls post-hoc tests where warranted. Intra- and inter-rater reliability was assessed using the intraclass correlation coefficient (ICC), models 2 and 3, respectively (Portney & Watkins, 2000; Shrout & Fleiss, 1979). Values for ICC greater than 0.75 were considered to represent high reliability (Portney & Watkins, 2000). Maternal food intake for Experiment #2 was assessed using a repeated measures ANOVA with factors of time (repeated measure; wk 1, 2, 3, 4, 5, and 6 of feeding) and treatment (E, PF and C). 57 CHAPTER 3: RESULTS 3.1 Experiment #1 3.1.1 Fetal Growth and Skeletal Development There were no significant differences in the number of fetuses per litter among groups (Table 3.1). The effect of prenatal ethanol exposure on fetal body weight and length is shown in Figure 3.1. Overall, maternal ethanol intake resulted in marginally lower mean fetal body weights (Fig. 3.1 A) compared with PF fetuses (two-way ANOVA, factor diet; p = 0.052). Moreover, fetal body weight varied significantly with dose (two-way ANOVA, factor dose; p < 0.001). Fetuses from the 36% dose groups were significantly lighter than fetuses from the 25% and 15% dose groups (p's < 0.001), and there was no significant difference between the 25% and 15% doses. When E, PF and C fetuses were compared within each dose, a significant effect of diet on weight was found only for the 36% dose (one-way ANOVA, diet p < 0.001). Post hoc tests indicated that E36 fetuses were significantly lighter than PF36 (p = 0.003) and C (p < 0.001) fetuses, whereas there was no significant difference between PF36 and C fetuses. The effects of prenatal ethanol exposure on fetal body length (Fig. 3.1 B) were similar to those for fetal body weight. Overall, E fetuses were significantly shorter than PF fetuses (two-way ANOVA, factor diet; p = 0.044). In addition, there was a significant main effect of dose (two-way ANOVA, factor dose; p = 0.004). Fetuses from the 36% dose groups were significantly shorter than those from the 25% and 58 Treatment Group Average Number of Fetuses Per Litter E15 15\u00C2\u00B10.8(n = 12) E25 14 \u00C2\u00B10.6 (n = 14) E36 14 \u00C2\u00B10.9 (n = 15) PF15 15 \u00C2\u00B10.4 (n = 14) PF25 15 \u00C2\u00B10.6 (n = 13) PF36 13\u00C2\u00B10.8 (n = 11) C 13\u00C2\u00B1 1.1 (n = 14) Table 3.1: The average number of fetuses per litter for Experiment #1. Effects of maternal ethanol intake at 15%, 25% and 36% ethanol-derived calories (E15, E25 and E36, respectively), pair-feeding (PF) or ad lib control (C) diet on the average number of fetuses per litter are shown. Values are mean \u00C2\u00B1 SE. Values for n are given in the parentheses. There were no significant differences among the groups. 59 3.40 E15 E25 E36 PF15PF25PF36 Treatment 56 55 H 54 53 52 cn c T J O n To 5 1 <*\u00E2\u0080\u00A2\u00C2\u00AB CD U- 50 49 ab B 10 ift \u00E2\u0080\u0094 I Trr-W t-v* , ^ . , ssr\u00E2\u0080\u0094 \u00E2\u0080\u0094 n t . i , n r m E15 E25 E36 PF15 PF25 PF36 Treatment Figure 3.1: Fetal body weight (A) and length (B) in Experiment #1. Effects of maternal ethanol intake at 15%, 25% and 36% ethanol-derived calories (E15, E25 and E36, respectively), pair-feeding (PF) or ad lib control (C) diet on fetuses are shown. Values are mean \u00C2\u00B1 SE. Values for n are given in the bars. Groups sharing the same letter are significantly different by Newman-Keuls post hoc test: weight, a c p < 0.001, bp = 0.003, ^p = 0.001; length, ap < 0.001, bp = 0.017, \u00C2\u00B0p = 0.036, d p = 0.031. 60 15% dose groups (p = 0.004 and p = 0.002, respectively), and there were no significant differences between fetuses from the 25% and 15% dose groups. Further analyses within each dose again showed significant differences at the 36% dose only (one-way ANOVA, diet p < 0.001). Fetuses from E36 dams were significantly shorter than C fetuses (p < 0.001), and marginally shorter than PF36 fetuses (p = 0.080), while PF36 fetuses were, in turn, significantly shorter than C fetuses (p = 0.017). The ponderal index was not significantly affected by diet or dose, indicating that the ethanol-induced restriction in growth did not alter the relationship between body weight and length. Values for the ponderal index (g/cm3) were 2.25 + 0.07, 2.23 + 0.06 and 2.22 + 0.06 for E15, E25 and E36 fetuses, 2.12 + 0.04, 2.18 + 0.06 and 2.23 + 0.06 for PF15, PF25 and PF36 fetuses, and 2.11 + 0.08 for C fetuses, respectively. Fetal bone ossification appeared to be more sensitive to ethanol's effects than body weight or length (Fig. 3.2). Moreover, the effects of ethanol differed among skeletal sites (Figs 3.2 to 3.4). The bones most sensitive to ethanol's effects were the ulna (Fig. 3.2A), radius (Fig. 3.2B), tibia (Fig. 3.2C) and sacrum (Fig. 3.2D). For each of these four bones, prenatal ethanol exposure resulted in decreased ossification compared with that in the PF groups (two-way ANOVAs, factor diet; p's < 0.001 for ulna, radius and tibia, p < 0.01 for sacrum). Moreover, ethanol effects differed according to dose. ANOVAs revealed a significant effect of dose for sacrum (p<0.01), a dose effect for radius that approached significance (p=0.060), and a dose by diet interaction for ulna (p=0.021) and tibia (p<0.001). These dose effects were due primarily to the dose-dependent decreases in ossification in E fetuses, 61 Figure 3.2: Ossification of the fetal ulna (A), radius (B), tibia (C) and sacrum (D). Effects of maternal ethanol intake at 15%, 25% and 36% ethanol-derived calories (E15, E25 and E36, respectively), pair-feeding (PF) or ad lib control (C) diet on fetuses are shown. Values are mean \u00C2\u00B1 SE. Values for n are given in the bars. Groups sharing the same letter were significantly different by Newman-Keuls post hoc test or Tukey-like nonparametric multiple comparison test: ulna, ap = 0.002, b ' c ' d p < 0.001, ep = 0.052, fp = 0.004; radius, a p = 0.010, b ' cp < 0.001, d p = 0.058, ep = 0.035, fp = 0.004,9p = 0.014; tibia, a b d p < 0.001, cp = 0.054, ep = 0.047, fp = 0.006, g p = 0.005, hp = 0.051; sacrum, a p < 0.005, bp = 0.004. 62 c o y \u00C2\u00ABE '(A (A O M s E15 E25 E36 PF15 PF25 PF36 Treatment E15 E25 E36 PF'tS P F 2 5 P F 3 6 Treatment E15 E25 E36 PF ! I5PF25 PF36 C Treatment Figure 3.3: Ossification of the fetal femur (A), humerus (B), and scapula (C). Effects of maternal ethanol intake at 15%, 25% and 36% ethanol-derived calories (E15, E25 and E36, respectively), pair-feeding (PF) or ad lib (C) control diet on fetuses are shown. Values are mean \u00C2\u00B1 SE. Values for n are given in the bars. Groups sharing the same letter are significantly different by Newman-Keuls post hoc test: femur, ap = 0.054, bp = 0.035; scapula, ap = 0.040, bp = 0.025, cp = 0.050. 63 VI 01 J= 5 c y (0 u c w o 0) E 3 z 1 1 Sternum E15 E25 E36 P?15PF25 PF36 Treatment 14 E15 E25 E36 p i l 5 PF25 PF36 Treatment Figure 3.4: Ossification of the fetal sternum (A) and metatarsals (B). Effects of maternal ethanol intake at 15%, 25% and 36% ethanol-derived calories (E15, E25 and E36, respectively), pair-feeding (PF) or ad lib control (C) diet on fetuses are shown. Values are mean \u00C2\u00B1 SE. Values for n are given in the bars. There were no significant differences among the groups for either bone. 64 rather than to differences among PF groups (see Fig. 3.2A-D). For ulna, post hoc tests showed that ethanol exposure at both 36% and 25% significantly reduced ossification compared to that in PF36 (p<0.001) and PF25 (p=0.021) fetuses, respectively, whereas for tibia, a significant decrease in ossification occurred only at the 36% dose (E36 wk 6 = wk 2 = wk 4 > wk 3 > wk 1, p's < 0.05). There were no significant differences between any E group and its respective PF group for any week. E36 and PF36 consumed significantly less food than C rats on wk 1, 2 and 4 (see Table for p values). There were no significant differences in food intake between C rats and E15, PF15, E25 or PF25 rats. There were no significant differences in initial body weights of the dams among groups. Maternal weight gain throughout gestation is shown in Figure 3.7. Overall, there was no significant difference in weight gain between E and PF dams. There was, however, a significant difference in weight gain with dose (two-way ANOVA, factor dose, p < 0.001). Maternal weight gain was significantly decreased at the 36% compared to the 25% and 15% doses (p's < 0.001). Further analysis within each dose showed significant effects of diet at each dose (one-way ANOVAs, 36%: p < 0.001, 25%: p = 0.005, 15%: p = 0.048). That is, C dams gained significantly more weight throughout gestation than E and PF dams at the 36% (p's < 0.001) and 25% (C>E, p=0.005, C>PF, p=0.020) doses. The post hoc test for the 15% dose was not significant. Weight gain did not differ significantly between E and PF dams at any dose. Maternal ethanol intake and peak BECs throughout gestation are shown in Figure 3.8. Altering the concentration of ethanol in the diet of the dams resulted in 77 Figure 3.7: Maternal weight gain throughout gestation. Effects of maternal ethanol intake at 15%, 25% and 36% ethanol-derived calories (E15, E25 and E36, respectively), pair-feeding (PF) or ad lib control diet (C) on weight gain are shown. Values are mean \u00C2\u00B1 SE. Values for n are given in the bars. Values sharing the same letter were significantly different by Newman-Keuls post hoc test ( a b p < 0.001, cp = 0.005, d p = 0.020). 78 14.00 11.80 9.60 \u00C2\u00A3 7.40 ro 5.20 h 3.00 1 2 Week of gestation 5 O LU m CU QL 140 120 100 80 60 40 20 0 1 2 Week of gestation \ / T T i t \u00E2\u0080\u0094 \u00E2\u0080\u0094 ' Figure 3.8: Maternal ethanol intake and blood ethanol concentrations. Weekly ethanol intakes (A) and peak blood ethanol concentrations (B) prior to (wk 0) and throughout (wk 1, 2 and 3) gestation in the dams consuming 15% (\u00E2\u0080\u00A2 ) , 25% (A) and 36% (\u00E2\u0080\u00A2) ethanol-derived calories. Rats consuming the 15% dose did not have detectable levels of blood ethanol. Values are mean \u00C2\u00B1 SE. 79 average ethanol intakes (mean \u00C2\u00B1 SE) throughout gestation of 5.2 \u00C2\u00B1 0.1, 8.2 \u00C2\u00B1 0.2 and 10.4 \u00C2\u00B1 0.2 g/kg body wt/d for the E15, E25 and E36groups, respectively. ANOVAs indicated that ethanol intake over the course of gestation (Fig. 3.8A) varied significantly with dose (two-way ANOVA, factor dose, p < 0.001; E36>E25>E15, p's < 0.001) and with week of gestation (two-way ANOVA, factor time, p < 0.001; week 2 > week 0 = week 1 = week 3, p's<0.05). As expected, peak BECs in the dams also increased with dietary ethanol concentration (Fig. 3.8B). Blood ethanol levels were not detectable in E15 dams, but varied significantly with dietary ethanol concentration in E25 and E36 dams (two-way ANOVAs, factor dose, p < 0.001 for week 0 vs. 2 and week 1 vs. 3). There was a significant main effect of week of gestation in the week 1 and 3 analysis (two-way ANOVA, factor time, p = 0.022), with peak BEC lower at week 1 compared to week 3, whereas peak BECs did not differ between week 0 and 2. The low peak BEC at week 1 of gestation was clearly due to a decrease in the E36 dams (Fig. 3.8B), and was surprising given that ethanol intake was not decreased, as seen in Figure 3.8A. Further analysis showed that this difference was not due to differences between the subset of dams used for BEC determination and for ethanol intake (ethanol intake was measured in all dams, whereas peak BEC at wk 1 was only measured in a subset of the dams). Analysis of BECs in terminal blood samples confirmed the trends seen in peak BEC with level of ethanol intake. Although the major period of feeding activity occurs just after lights out, a second period of feeding occurs in the early morning (Weiner et al, 1981), a few hours prior to termination. Ethanol was detectable in terminal blood samples from only seven of thirteen E15 dams (range 26-36 mg/dL) and so this 80 group was not included in the analysis. Terminal BECs in the dams were 38 \u00C2\u00B1 4 and 69 \u00C2\u00B1 10 mg/dL in E25 and E36 dams, respectively and this difference was significant (t-test, p = 0.010). 3.2 Experiment #2 3.2.1 Fetal Parameters There were no significant differences in the number of fetuses per litter among the diet groups (Table 3.4). Fetal body weights are shown in Figure 3.9 and were significantly affected by diet (ANOVA p<0.001). Maternal ethanol intake resulted in significantly lower mean fetal body weights compared with PF and C fetuses (p's<0.001). There was no significant difference in the body weights of PF compared with C fetuses. Visual examination of the tibia under 31 x magnification showed no obvious abnormalities in the ethanol-exposed, compared with PF and C, fetuses other than a decrease in total bone length (Fig. 3.10). The effect of ethanol on total length of the fetal tibia is shown in Figure 3.11. Tibial length was significantly affected by diet (ANOVA p<0.001), and ethanol exposure resulted in significantly shorter tibiae compared with PF and C fetuses (p's<0.001). There was no significant difference in tibial length between PF and C fetuses. Figure 3.12 shows the total length of the diaphysis, expressed in absolute terms (Figure 3.12A) and relative to total tibial length (Figure 3.12B). The shortening of the tibia by ethanol exposure appeared to be the result of a shorter diaphysis (Figure 3.12A; ANOVA p=0.002), with E fetuses 81 Treatment Group Average Number of Fetuses Per Litter E PF C 16 \u00C2\u00B10.7 (n = 13) 15 \u00C2\u00B10.5 (n = 12) 14 \u00C2\u00B10.9 (n = 12) Table 3.4: The average number of fetuses per litter for Experiment #2. Effects of maternal ethanol intake at 36% ethanol-derived calories (E), pair-feeding (PF) or ad lib control (C) diet on the average number of fetuses per litter are shown. Values are mean \u00C2\u00B1 SE. Values for n are given in the parentheses. There were no significant differences among the groups. 82 3.50 Figure 3.9: Fetal body weight from Experiment #2. The effects of maternal ethanol intake (E), pair-feeding (PF) or ad lib control (C) diet on fetal body weight. Values are mean \u00C2\u00B1 SE. Values for n are given in the bars. Groups sharing the same letter are significantly different by Newman-Keuls post hoc test: a> bp < 0.001. 83 E PF C Figure 3.10: Typical tibiae from ethanol (E), pair-fed (PF) or ad lib control (C) fetuses. No differences among the groups could be detected by visual examination of the tibias, except for the differences in total bone length, among the tibia upon histological examination. 8 4 4600 4380 C 4160 CD la =\u00E2\u0080\u00A2= 3940 CO O 3720 3500 PF Figure 3 .11 : Total length of the fetal tibia. The effect of maternal ethanol intake (E), pair-feeding (PF) or ad lib control (C) diet on the total length of the fetal tibia. Values are mean \u00C2\u00B1 SE. Values for n are given in the bars. Groups sharing the same letter are significantly different by Newman-Keuls post hoc test:a' bp < 0.001 85 Figure 3.12: Diaphysis length of the fetal tibia. The effect of maternal ethanol intake (E), pair-feeding (PF) or ad lib control (C) diet on (A) absolute diaphysis length and (B) diaphysis length relative to total tibial length. Values are mean \u00C2\u00B1 SE. Values for n are given in the bars. Groups sharing the same letter are significantly different by Newman-Keuls post hoc test: (A) a p = 0.009, b p = 0.002, (B) a p = 0.048, b p = 0.051. 86 having significantly shorter diaphyses compared with PF and C (p=0.009 and 0.002, respectively) fetuses, which did not differ from each other. When expressed relative to total tibial length (Figure 3.12B), there were still significant differences among the diet groups (ANOVA p=0.040), indicating that the decrease in diaphysis length in E fetuses was not just proportional to the decrease in total tibial length. The relative diaphysis length of E fetuses was significantly shorter than that of PF fetuses (p=0.048), while the difference between E and C fetuses did not quite reach significance (p=0.051). There was no significant difference between PF and C fetuses with respect to diaphysis length (relative to total tibial length). It should be noted that when data analysis was conducted without removing outliers from the dataset, the absolute diaphysis length was significantly shorter in E fetuses, but the effect of treatment on relative diaphysis length did not reach significance (ANOVA p=0.108). The outlier removed in this instance was a value from the C group that appeared to represent a litter of small C fetuses. In contrast with ethanol's effect on the diaphysis, there was no significant effect of ethanol on the lengths of the proximal or distal epiphyses (Figure 3.13) when expressed as an absolute measure (Figure 3.13A) or relative to total tibial length (Figure 3.13B). There was also no significant effect of diet on the width of the diaphysis (E: 492 \u00C2\u00B1 15 pm, PF: 518 \u00C2\u00B1 12 um, C: 526 \u00C2\u00B1 12 pm). While the total length of the epiphyses was not significantly affected by ethanol, the organization of the different zones within the epiphyses was disrupted (Figure 3.14). In the proximal epiphysis, significant diet effects were found on the 87 1500 \" E P F C E P F C Proximal Epiphysis Distal Epiphysis Figure 3.13: Proximal and distal epiphysis length. The effect of maternal ethanol intake (E), pair-feeding (PF) or ad lib control (C) diet on (A) absolute proximal and distal epiphysis length and (B) proximal and distal epiphysis length relative to total tibial length. Values are mean \u00C2\u00B1 SE. Values for n are given in the bars. There were no significant differences among the groups. 88 650 \u00C2\u00A3 570 7= 490 h CJ) c 0 CD C o N 410 330 h 250 Resting Zone Proliferative Zone Hypertrophic Zone CD c Q> C o N > CP DC 15 12 PF 1 C Resting Zone I E Proliferative Zone Hypertrophic Zone PF lH C Figure 3.14: Resting, proliferative and hypertrophic zone lengths in the proximal epiphysis. The effects of maternal ethanol intake (E), pair-feeding (PF) or ad lib control (C) diet on (A) absolute length of the resting, proliferative and hypertrophic zones in the proximal epiphysis and (B) length of the resting, proliferative and hypertrophic zones in the proximal epiphysis relative to total tibial length. Values are mean \u00C2\u00B1 SE. Values for n are given in the bars. Groups sharing the same letter are significantly different by Newman-Keuls post hoc test: (A) a p = 0.008, b p = 0.014, c p < 0.001, dp < 0.001, e p = 0.041 (B) a p < 0.001, b p = 0.004 89 lengths of both the resting (ANOVA p=0.005) and hypertrophic (ANOVA p<0.001) zones (Figure 3.14A). Fetuses exposed to ethanol had shorter resting zones compared with PF (p=0.008) and C (p=0.014) fetuses, and longer hypertrophic zones than PF (p<0.001) and C (p=0.020) fetuses (Figure 3.14A). As well, the hypertrophic zones of PF fetuses were significantly shorter than those of C fetuses (p = 0.041). When expressed relative to total tibial length (Figure 3.14B), the effect of diet on the hypertrophic zones was still apparent (ANOVA p<0.001), with lengths in E fetuses remaining longer than those of PF and C fetuses (p's<0.001), but the difference between PF and C did not reach significance (p = 0.113). The differences in resting zone lengths disappeared, however, indicating that the decrease in resting zone length was proportional to the decrease in total tibial length. There was no significant effect of ethanol exposure on the proliferative zone length, expressed either as an absolute measure or relative to total tibial length. Figure 3.15 shows the effect of ethanol on the areas of the different histological zones. There was a significant effect of diet on the absolute areas of the resting zone (ANOVA p<0.001), but no significant effect on the proliferative or hypertrophic zones (Figure 13.5A). The resting zone area of E tibiae was significantly smaller than that of PF (p<0.001) and C (p=0.018) tibiae and that of C fetuses were significantly smaller than that of PF fetuses (p=0.026). When expressed relative to total tibial length (Figure 13.5B), the effect on RZ area remained (ANOVA p=0.005) and there was a significant effect on HZ area (ANOVA p=0.032). The resting zone area relative to total tibial length was significantly smaller in E compared with PF fetuses (p=0.004). The difference between E and C 90 Figure 3.15: Area of the resting, proliferative and hypertrophic zones in the proximal epiphysis. The effects of maternal ethanol intake (E), pair-feeding (PF) or ad lib control (C) diet on (A) absolute area of the resting, proliferative and hypertrophic zones in the proximal epiphysis and (B) area of the resting, proliferative and hypertrophic zones in the proximal epiphysis relative to total tibial length. Values are mean \u00C2\u00B1 SE. Values for n are given in the bars. Groups sharing the same letter are significantly different by Newman-Keuls post hoc test: (A) a p < 0.001, b p = 0.018,c p = 0.026; (B) a p = 0.004, b p = 0.064. 91 fetuses was not significant (p=0.131). Relative to total tibial length, E fetuses have significantly larger HZ areas compared with PF fetuses (p=0.035), which were significantly smaller than C fetuses (p=0.044). There was no significant difference between the E and C fetuses in the HZ area relative to total tibial length. Intra- and inter-rater reliability, assessed using the ICC, was high for all parameters assessed. The ICC values are shown in Table 3.5 and were greater than 0.75 for all measures. 3.2.2 Maternal Parameters Maternal ethanol intake throughout gestation in the E group averaged 11.3 \u00C2\u00B1 0.2 g/kg body weight per day. Mean peak BEC on d 10-11 of gestation was 114 \u00C2\u00B1 16 mg/dL. Food intake values over the 6 wk of feeding (3 wk prior to pregnancy and 3 wk during gestation) are presented in Table 3.6. Maternal food intake was significantly affected by group (ANOVA p<0.001) and time (ANOVA p<0.001), and there was a significant group by time interaction (ANOVA p<0.001). Overall, there was no significant difference in the food intake of E and PF dams, which both consumed less than C dams (p's < 0.001). As well, food intake, collapsed over groups, varied by week (wk 5 > wk 4 > wk 6; wk 2 > wk 4 > wk 6; wk 3 > wk 6; p's < 0.05). The post-hoc test of the individual groups by week demonstrated that there were no significant differences between the E and PF groups for any week. E dams consumed significantly less food than C dams on weeks 1-5, and PF dams 92 ] Measure Intra-Rater Reliability (ICC) Inter-Rater Reliability (ICC) Bone Length 0.999 0.997 Bone Width 0.956 0.917 Diaphysis Length 0.999 0.987 Resting Zone Length 0.982 0.911 Proliferative Zone Length 0.762 0.756 Hypertrophic Zone Length 0.806 0.966 Table 3.5: Intra- and inter-rater reliability for the histological morphometric measures assessed using the intraclass correlation coefficient (ICC). 93 week 1 week 2 week 3 week 4 week 5 week 6 1 0.206 \u00C2\u00B1 0.007a 0.222 \u00C2\u00B1 0.004\u00C2\u00B0 0.227 \u00C2\u00B1 0.004c 0.212 \u00C2\u00B10.008\u00C2\u00B0 0.230 \u00C2\u00B1 0.006e 0.203 \u00C2\u00B1 0.003 PF 0.192 \u00C2\u00B1 0.008 f 0.208 \u00C2\u00B1 0.0059 0.218 \u00C2\u00B10.005h 0.204 \u00C2\u00B10.005' 0.243 \u00C2\u00B10.005 0.227 \u00C2\u00B10.004 C 0.297 \u00C2\u00B1 0.009af 0.296 \u00C2\u00B1 0.008b g 0.270 \u00C2\u00B1 0.010ch 0.266 \u00C2\u00B1 0.010di 0.263 \u00C2\u00B1 0.010e 0.201 \u00C2\u00B1 0.008 Table 3.6: Maternal food intake (g food per g body weight) by week in Experiment #2. Weeks 1 to 3 represent the 3 wk prior to gestation and weeks 4 to 6 represent the 3 wk during gestation. There were no significant differences between any E group and its respective PF group for any wk. Values are mean \u00C2\u00B1 SE. The number of dams per group were: E: n=13, PF: n=12, C: n=12. Values sharing the same letter were significantly different for a given week by Newman-Keuls post hoc test (a' bp<0.001, cp=0.024, dp=0.005, ep=0.04, 9p<0.001, hp=0.007, 'p=0.002). CO consumed significantly less food than C dams on weeks 1-4. (see Table for p values). There were no significant differences in food intake between PF and C dams on week 5 or among E, PF and C dams on week 6. Similarly, maternal weight gain during gestation varied with diet (ANOVA p<0.001), and was not significantly different between E and PF dams (Figure 13.6). The C dams gained significantly more weight during gestation than E and PF dams (p's<0.001). 95 140 E P F C Figure 3.16: Maternal weight gain during gestation in Experiment #2. Effects of maternal ethanol intake at 36% ethanol-derived calories, pair-feeding (PF) or ad lib control diet (C) on weight gain during gestation are shown. Values are mean \u00C2\u00B1 SE. Values for n are given in the bars. Values sharing the same letter were significantly different by Newman-Keuls post hoc test ( a b p < 0.001). r \ 96 C H A P T E R 4: G E N E R A L DISCUSSION 4.1 S U M M A R Y A N D DISCUSSION The objective of this thesis was to determine if prenatal ethanol exposure has specific effects on bone development, in addition to its effects on general growth, in the fetal rat. To meet this objective, the following hypotheses were tested: (1) prenatal ethanol exposure affects skeletal development (specifically ossification) at doses of ethanol lower than those required to affect general growth (assessed by fetal body weight and length), (2) prenatal ethanol exposure has a differential effect on skeletal development (specifically ossification) at different skeletal sites, and (3) prenatal ethanol exposure disrupts the morphology of the fetal tibial growth plate. We conducted two studies to address these hypotheses. The first study (addressing hypotheses 1 & 2) examined the effect of different doses of ethanol (designed to approximate low, moderate and high levels of exposure) on fetal growth and skeletal development. Our results provide two lines of evidence that support the hypothesis that ethanol has specific effects on skeletal development, independent of its effects on general growth. First, effects on bone were seen at levels of ethanol exposure below those required to affect fetal body weight and length. Furthermore, there was a dose-dependent decrease in ossification in ethanol-exposed, but not pair fed, fetuses, suggesting that this effect on ossification can be attributed to ethanol exposure. Second, prenatal ethanol exposure affected some bones more than others, with some bones being fairly resistant to ethanol's effects. As all bones 97 examined underwent significant development at the end of gestation, the differences cannot be explained by a lack of development in the unaffected bones. The second study, designed to address hypothesis 3, examined the effect of a high level of ethanol exposure (36% EDC) on the morphology of the fetal tibial growth plate. Ethanol exposure resulted in decreased tibial length, which was due to a decrease in the length of the diaphysis rather than the epiphyses. However, although there was no effect on the total length of the epiphyses, the organization of the zones within the epiphyses was subtly altered by ethanol. Specifically, ethanol exposure resulted in a decrease in the resting zone (which was proportional to the decrease in total tibial length) and an enlargement of the hypertrophic zone. This disruption in the normal morphology of the tibial growth plate resulting from prenatal ethanol exposure also supports the hypothesis that ethanol exerts specific effects on skeletal development. The results of this thesis clearly demonstrate that prenatal ethanol exposure affects fetal skeletal ossification at levels lower than those required to affect fetal body weight and length. The differential responses of fetal body weight and length (which appeared to vary together) and skeletal ossification to the different levels of ethanol exposure suggest that ethanol's effects on skeletal development do not merely reflect ethanol's effects on general growth. Maternal drinking at high levels is well known to result in decreased body weight in the fetus or neonate (Abel, 1996; Breese et al, 1994; Hannigan et al, 1993; Keiver & Weinberg, 2004; Mauceri et al, 1994; Pullen et al, 1988; Reyes et al, 1985; Sanchis & Guerri, 1986; Weinberg et al, 1990), and, consistent with the present studies, significant effects on body weight 98 tend to be associated with maternal ethanol intakes which result in peak BECs >100 mg/dL (Gallo & Weinberg, 1986; Savage et al, 2002). Less is known about the effect of prenatal ethanol exposure on body length. Studies in the rat suggest that significant decreases in length also occur at high (Detering et al, 1979; Keiver & Weinberg, 2004; Lee, 1987, Lee & Leichter, 1980, Leichter & Lee, 1979, Lochry et al, 1980) but not lower (27% EDC; Samson, 1981) levels of exposure, although Lochry et al (1980) found a significant linear trend between body length at birth and prenatal ethanol exposure (0, 12, 23 and 35% EDC, group differences not reported). Studies in humans show similar trends; Jacobson et al (1994) found effects on body length at birth only at high levels of maternal drinking, whereas Day et al (1989) found a dose-response relationship between ethanol exposure and body length at birth. It should be noted, however, that potential confounding effects of maternal nutrition are not completely controlled for in these human studies. As there was a trend for body length to decrease with increasing dose in the PF rats in this thesis, effects of ethanol in the human studies may be partially due to ethanol's effects on maternal nutrition. The ponderal index was not affected by maternal ethanol intake, indicating that the ethanol-induced growth retardation did not alter the relationship between body weight and length. This is consistent with the understanding that teratogens, such as ethanol, typically result in a symmetric pattern of growth retardation (Han, 1999; Sparks et al, 1998), and, at first glance, suggests that ethanol may have uniform effects on general growth and skeletal development. However, when skeletal development was examined, uniform effects between general growth and 99 ossification were not seen and at least some effects occurred without a concomitant decrease in bone size. This study is the first to examine the effects of different levels of prenatal ethanol exposure on skeletal development. With exposure to 25% EDC (moderate exposure), no growth retardation was seen, but there was a decrease in the ossification of some bones (ulna, radius and tibia). The fact that we were able to demonstrate significant effects of ethanol at levels that approximate moderate as well as high levels of exposure is important, because most women who consume ethanol during pregnancy do so at low to moderate levels (Ebrahim et al, 1998; Flynn et al, 2003). Consistent with the findings of this thesis, prenatal ethanol exposure at high levels has previously been shown to result in decreased skeletal ossification (Keiver et al, 1996, 1997; Keiver and Weinberg, 2004; Lee and Leichter, 1983; Weinberg et al, 1990), although the severity of ethanol's effects has varied. Our lab has previously hypothesized that this variation in ethanol's effects among studies on fetal skeletal ossification are related to differences in maternal/fetal BEC (Keiver & Weinberg, 2004). The results of the present thesis are consistent with this hypothesis, as the severity of ethanol's effects and maternal BECs varied with dose of ethanol. The most reliable measure of exposure of the fetus to ethanol is fetal BEC. Alternatively, maternal BECs can be used because ethanol crosses the placenta, and fetal BEC is highly correlated with maternal BEC (Brien et al, 1985; Guerri & Sanchis, 1985, Keiver & Weinberg, 2004). Maternal or fetal BEC, however, is rarely known in human studies and is often not measured in animal studies, making 100 comparisons among studies difficult. As many factors modify the relationship between concentration of ethanol in the diet (or drinking water), ethanol intake and BEC (Maier et al, 1995; Pierce & West, 1986; Sankaran et al, 1991; Wiener et al, 1981), the level of exposure of the fetus cannot always be determined from the concentration of ethanol in the diet or the amount consumed by the dam. Dams in the present study were fed ethanol at doses that approximate low, moderate and heavy (or high) levels of maternal drinking. These terms, however, are not strictly or uniformly defined. For human females, the generally accepted definition of \"moderate\" drinking is <1 drink (or <0.5 oz absolute ethanol) per day (USDA/DHHS, 1990), but \"low\" and \"heavy\" levels of drinking are not well defined. Examination of the BECs achieved in animal models facilitates some comparison with the human situation. The peak BECs in the dams consuming 36% EDC in the present study were slightly higher than the legal BEC limits for drunk driving (50-100 mg/dL) in Canada and the United States, and would generally be considered to correspond to heavy drinking. Peak BECs in the dams consuming 25% EDC were appreciably lower, approximating the level accepted to represent moderate drinking (up to ~ 50 mg/dL) (Hingson et al, 1999; Savage et al, 2002). Interestingly, the pattern of peak BECs across the weeks of gestation at the 36% dose did not mirror the pattern of ethanol intake. Notably, there was a decrease in peak BEC at week 1 of gestation, whereas there was not a decrease in ethanol intake. While it is not immediately clear why this is so, it is possible that the time point at which BEC was taken did not correspond to peak BECs in week 1 gestation dams in our study. Blood samples for analysis of BEC were taken at two 101 hours after the lights were turned off based on a study by Weiner et al (1981) which demonstrated that this time point was just after the main period of feeding and thus approximately corresponded to the peak BEC achieved. However, in that study, the food intake was assessed on day 13 of gestation (Weiner et al, 1981). The dams in our study were fed ethanol prior to breeding, but were not given ethanol while in the breeding rack, so the first week of gestation represents a time at which the dams are adjusting to the reintroduction of ethanol in the diet. Thus, it is possible that if there was a disruption in the circadian pattern of food intake in week 1 of gestation, as the dams readjust to ethanol, compared with the other weeks of gestation, the BEC measured did not correspond to the peak BEC achieved. Alternatively, as a variety of physiological changes occur at the beginning of pregnancy, it is possible that the metabolism of ethanol is altered, resulting in a significantly lower peak BEC at week 1 of gestation compared with the other weeks. It is clear from this thesis that the skeletal sites measured differed in their sensitivity to ethanol, and this also supports the hypothesis that ethanol's effects on fetal skeletal development are not due to effects on general growth. At the 36% dose of ethanol, fetal body weight and length were delayed by approximately 0.4 d. If ethanol's effects were solely due to a generalized growth effect, then the degree of delay would have been expected to be uniform for all measures, and all bones would have been delayed by 0.4 d. This was not the case, however, and the delay in ossification varied from no delay (sternum and metatarsals) to 0.5 d (scapula and radius), depending on the skeletal site. Moreover, the lack of effect of ethanol in some bones cannot be explained by a lack of significant development between d 20 102 and d 21 gestation, as the normal development curves showed that ossification increased significantly in all bones between d 20 and d 21. Thus, a 0.4 d delay in ossification would have been detectable in all bones measured. The reason for the difference among skeletal sites in terms of sensitivity to ethanol's effects is not known. The relative sensitivity of the different sites, however, appears to be consistent among studies. Effects of exposure to high levels of ethanol can generally be demonstrated in ulna, radius and tibia (Keiver et al, 1996; Keiver & Weinberg, 2004; Lee & Leichter, 1983; Weinberg et al, 1990), and these were also the bones affected by the lower dose (25% EDC) of ethanol in the present study. Effects on humerus, femur and scapula are variable (Keiver et al, 1996, 1997; Keiver & Weinberg, 2004; Lee & Leichter, 1983; Weinberg, 1990). When examined, metatarsals have generally shown no effect (Keiver et al, 1996, 1997; Keiver & Weinberg, 2004; Weinberg, 1990). Interestingly, sensitivity to ethanol appears to be greatest in the bones that start to ossify earlier in gestation (Strong, 1925) and undergo a greater proportion of their development in utero. Ulna, radius and tibia were approximately 60-80% ossified by day 21 gestation in control fetuses, whereas the femur, humerus and scapula, which were less affected by ethanol, were approximately 50% ossified. The metatarsals and sternum, which were unaffected by even the highest dose of ethanol, were the least developed of all the bones examined, with ossification centres only just appearing. This differential sensitivity may provide insight into the stages of bone formation affected by ethanol. Endochondral bone formation begins with the formation of a cartilage template, which is eventually replaced by bone. 103 Chondrocytes regulate the formation of the ossification centres. They control the calcification of cartilage and produce growth factors that stimulate the invasion of blood vessels, attract chondroclasts to resorb calcified cartilage and direct the formation of osteoblasts from the perichondrium (Kronenberg, 2003). As the calcified cartilage is resorbed, the osteoblasts secrete matrix, which eventually calcifies, and cartilage is gradually replaced with bone. As ethanol appeared to have little effect on ossification in bones in which the ossification centres were just appearing, and the most effect in bones in which ossification was relatively advanced, this may suggest that ethanol's effects on ossification result from effects on the stages of development that occur after the calcification of cartilage. The histological examination of fetal tibia exposed to ethanol support this hypothesis. The increased length of the hypertrophic zone seen in the ethanol fetuses, combined with the lack of an effect of ethanol on the length of the proliferative zone, suggests that ethanol delays the exit of chondrocytes from the hypertrophic stage, possibly due to interference with the initiation of cartilage calcification, chondrocyte apoptosis, angiogenesis or osteoblast function. Enlargement of the hypertrophic zone could underlie the decrease in diaphysis length (and thus have contributed to the decrease in total bone length) seen in this study, as well as the decrease in ossification (measured as a decrease in length of mineralized portion) seen with ethanol exposure in previous studies. Hypertrophic zone enlargement has been shown to occur with disruptions in angiogenesis or resorption of calcified cartilage, and is often associated with a decrease in total bone length (Gerber et al, 1999; Haigh et al, 2000; Stickens et al, 2004,. Vu et al, 1998; 104 Zelzer et al, 2004). Whether ethanol exerts its effects in these ways in the developing bone, however, remains to be determined. For example, ethanol exposure has been shown to either impair (Radek et al, 2005) or stimulate angiogenesis (Gu et al, 2001), depending on the tissue being examined, and so the effect of ethanol exposure on angiogenesis in the developing bone would need to be examined. Importantly, however, effects of ethanol on the transition from hypertrophic cartilage to formation of bone tissue are consistent with the observation that ethanol can delay ossification at doses below those required to inhibit longitudinal growth. Effects of ethanol on earlier stages of bone development, however, cannot be excluded. The effect of prenatal ethanol exposure on resting zone length suggests that ethanol may have effects on early stages of long bone development, which may contribute to the decrease in total bone length seen in this thesis and other studies with high levels of ethanol exposure. Development of the fetal skeleton begins with the condensation of mesenchymal cells in the embryo, which subsequently differentiate into chondrocytes and form the cartilage anlagen of the skeleton. The initial size of a condensation is a reflection of the growth potential of the bones that will result from that condensation (Hall & Miyake, 2000). Thus, it is possible that ethanol exposure reduces the initial size of the mesenchymal cell condensation, resulting in a smaller anlagen, and this reduction in size is reflected in the smaller resting zone in the ethanol-exposed fetus. Ethanol has been shown to enhance the differentiation of mesenchymal cells into chondrocytes in chick limb buds in vitro (Kulyk & Hoffman, 1996), but whether this precocious chondrogenesis results in a 105 smaller cartilage anlagen or influences the fate of the developing limb in some other way has not been investigated. The only other study to histologically examine the effects of prenatal ethanol exposure on developing bone is that of Miralles-Flores and Delgado-Baeza (1992). In that study, dams were fed ethanol for 30 d prior to and during gestation, and the tibia of offspring examined at birth or postnatal d 15. Ethanol exposure resulted in decreased tibial length at both ages, but only affected the organization of the growth plate at postnatal d 15. Ethanol decreased the length of the hypertrophic zone, which resulted in a decrease in total growth plate length. The length of the resting zone was not measured. The differences between our results and those of Miralles-Flores and Delgado-Baeza (1992) may be due to differences in the ages of the rats (gestation d 21 vs. at birth vs. postnatal d 15), as the regulation of bone development differs between the pre- and postnatal periods. Thus, it is possible that the effect of prenatal ethanol exposure on the expression of local growth factors and/or systemic hormones that affect the regulation of bone development differs during the prenatal and postnatal periods. Alternatively, it is possible that differences in level of ethanol exposure could account for the differences between the studies. It is common in many labs to remove dams from ethanol-containing diets for 1-2 d prior to birth as ethanol can cause complications during parturition. It is unknown if this was the case in the Miralles-Flores and Delgado-Baeza study (1992), but if so, then it is possible that the removal of ethanol during this period could have allowed the bone to partially recover, resulting in no significant effects on the growth plates of those pups at birth. 106 It should be noted that, in a few instances, there were significant effects of the pair-feeding regimen compared with ad lib controls. Pair-feeding is used as a control for the decrease in food intake that commonly occurs with ethanol intake; thus, when PF values are intermediate between E and C rats, it indicates that decreased nutrition is partially responsible for the effect being observed. Although pair-feeding did not affect ossification of most bones measured, it did decrease the ossification of the radius and tibia. In the case of the radius, the ethanol-induced decrease in ossification observed was likely due at least partly to decreased nutrition. However in the tibia, ossification was decreased in the PF15 and PF25, compared with the C, groups, but was not decreased in the PF36 group. As the food intakes of the PF15 and PF25 groups were not significantly different from that of the C group throughout the experiment, whereas the intake of the PF36 group was significantly decreased, the decrease in ossification in the PF15 and PF25 groups does not appear to be nutritionally mediated. The effects of reduced nutrition also did not appear to contribute to ethanol's effects on the histology measurements of the tibia. Although some pair-feeding effects on tibia histology were seen, they were not intermediate between the E and C values. Specifically, the resting zone area was significantly larger, and the hypertrophic zone length was significantly decreased, in PF compared with C tibia, and these effects were in the opposite direction from the ethanol-induced changes. While it is not clear why pair feeding had these effects in these instances, it does highlight the fact that the effect of ethanol was not due to the decreased food intake seen in E rats. 107 Furthermore, it should be noted that, in addition to serving as a nutritional control for E rats, pair feeding is, in itself, an experimental treatment (Weinberg, 1984). The stress resulting from food deprivation that can occur with the pair feeding regimen can potentially affect the developing fetus, resulting in different effects than those caused by ethanol exposure. The mechanisms by which ethanol affects fetal skeletal development are unknown. Ethanol readily crosses the placenta (Guerri & Sanchis, 1985) and in vitro studies have shown that ethanol inhibits osteoblast differentiation, proliferation and function (Chavassieux et al, 1993; Farley et al, 1985; Gong and Wezeman, 2004; Klein, 1997); thus ethanol may have direct effects on the cells involved in bone development. Moreover, chronic ethanol exposure in the adult results in increased total bone and bone marrow fat content in both humans and animals (Gong & Wezeman, 2004; Sampson et al, 1996, 1998; Wezeman & Gong 2001, 2004), and ethanol has been shown to inhibit osteogenic differentiation (Gong & Wezeman, 2004) and increase adipogenesis (Wezeman & Gong, 2004) in mesenchymal stem cells. Interestingly, preliminary data indicate that the tibia of fetal guinea pigs prenatally exposed to ethanol also has an increased fat content (Keiver and Brien, unpublished data), thereby suggesting that ethanol also alters the balance between adipogenesis and osteogenesis in the fetus. Alternatively, or in addition, ethanol may exert its effects indirectly through perturbations in maternal and/or fetal physiology. We have previously shown that maternal ethanol intake results in maternal and fetal hypocalcemia (Keiver et al, 1996; Keiver and Weinberg, 2003, 2004), and the severity of the hypocalcemia 108 varies with BEC (Keiver & Weinberg, 2003, 2004). Maternal and fetal hypocalcemia are associated with adverse effects on skeletal development, including reduced mineralization (Chalon & Garel, 1985; Lima et al, 1993; Loughead et al, 1990; Rebut-Bonneton et al, 1983a, 1983b; Sinclair 1942), thus altered maternal/fetal calcium homeostasis is a potential mechanism for ethanol's effects on bone development. Effects of ethanol on osteoblasts or on maternal/fetal calcium homeostasis are both consistent with an effect of ethanol on the later stages of bone formation. Ethanol intake is also known to affect other maternal/fetal systems that could potentially affect bone development. For example, insulin signaling is impaired by ethanol (Wan et al, 2005); however, as insulin affects chondrocyte proliferation (Fowden, 1995), the results of this thesis suggest that ethanol is not exerting its effects on bone via its effects on insulin signaling. Similarly, prenatal ethanol exposure is known to affect the HPA axis (Zhang et al, 2005) and GH/IGF (Breese et al, 1993; Gabriel et al, 1998) systems.. However, like insulin, glucocorticoids and GH/IGF appear to exert their effects on chondrocyte proliferation (Mushtaq et al, 2004; Robson et al, 2002) and thus disturbances in these hormonal systems seem unlikely to underlie ethanol's effects on fetal bone. Given the effect of ethanol on multiple systems relevant to bone development, determining the mechanism(s) by which ethanol exerts its effects on the developing skeleton will be difficult. Moreover, it is likely that ethanol is acting by more than one mechanism. In addition, the complexity and interrelationships among the regulatory cascades involved in 109 bone development will make it difficult to pinpoint the mechanism(s) underlying ethanol's effects. As ethanol's effects on the morphology of the growth plate in this study were subtle, and the mechanism(s) underlying them will likely be difficult to elucidate, it may be prudent to first investigate the functional significance of ethanol's effects. Previous work has shown that ethanol's effects on skeletal development do not resolve after birth but extend into postnatal life in both humans (Habbick et al, 1998) and animals (Leichter & Lee, 1979; Ludeha et al, 1983; Miralles-Flores & Delgado-Baeza, 1992). Moreover, prenatal ethanol exposure may affect bone function, as ethanol-induced alterations in fetal bone strength have been reported in the sheep model (Given et al., 2004). However, the implications of the effects of prenatal ethanol exposure on the long-term bone health of the offspring have not been studied. Although research in the area is still in its infancy, some recent research supports the hypothesis that changes in the prenatal environment may alter the programming of bone cells during fetal life and result in long-term effects on bone. Epidemiological studies provide correlative evidence of a link between body weight at birth or during infancy and later skeletal health (Cooper et al, 1995, 1997; Dennison et al, 2005; Gale et al, 2001). In the rat, moderate maternal protein restriction during pregnancy resulted in decreased bone mineral content and increased epiphyseal growth plate length in the offspring in adulthood (Mehta et al, 2002). Similarly, administration of IL-1, at a dose known to stimulate the HPA axis, to pregnant dams resulted in shorter tibia in all offspring, and decreased height and 110 bone mineral density and content in the vertebrae of male offspring at 10-12 wk of age (Swolin-Eide et al, 2004). Moreover, prenatal exposure to a synthetic estrogen increased bone mass in adulthood, which resulted from a decrease in osteoclast number and an increase in mineral apposition rate (Migliaccio et al, 1996, 2000). Circulating estrogen levels were not altered in the adult mice, indicating that in utero estrogen exposure resulted in a permanent effect on bone cells. The effects were only partially ameliorated by prepubertal ovariectomy (Migliaccio et al, 1996, 2000), suggesting that the response of bone cells to estrogen deficiency was modified by prenatal estrogen exposure. These data are compatible with the concept that the prenatal environment may alter the programming of bone cells that occurs during fetal life, and thus result in long-term effects on bone. It has been hypothesized that the prenatal environment may alter how the growth plate responds to glucocorticoids or GH postnatally (Harvey & Cooper, 2004). Thus, although ethanol-induced alterations in maternal/fetal HPA or GH/IGF systems may not underlie ethanol's effects on fetal bone morphology, they could affect bone health of the offspring later in life. In addition, the stimulation of PTH/PTHrP activity in the fetus in response to altered calcium homeostasis has been hypothesized to program bone cells, thus affecting bone health later in life (Tobias & Cooper, 2004). Whether the subtle effects of prenatal ethanol exposure on the epiphysis seen in this thesis will result in significant functional effects postnatally is an important area for investigation. In addition to specific effects on the growth plate and ossification of fetal bone, the effects of prenatal ethanol exposure on body weight and length may also 111 influence the offspring's risk of osteoporosis later in life. Ethanol's effects on weight and length persist into the postnatal period (Breese et al, 1994; Day et al, 1994; Lee, 1987; Lee & Leichter, 1980; Leichter & Lee, 1979; Lochry et al, 1980; Pullen et al, 1988), and effects on body length appear to be permanent (Day 2002; Habbick et al, 1998; Streissguth et al, 1991). Significant associations have been found between birth weight and length, and neonatal bone mineral content (BMC) and bone mineral density (Godfrey et al, 2001), as well as between weight in infancy and BMC in adulthood (Cooper et al, 1995, 1997). Furthermore, a low rate of childhood growth has been shown to be associated with hip fracture later in life (Cooper et al, 2001). These studies suggest that the poor growth during fetal and postnatal life that results from ethanol exposure may affect peak bone mass and contribute to the risk of developing osteoporosis. Our studies may have implications for the health of infants and children who are exposed to ethanol in utero. The results presented in this thesis demonstrate that prenatal ethanol exposure exerts effects on the developing skeleton at exposure levels that approximate moderate drinking (i.e., BECs of -50 mg/dL), which are lower than those which affect body weight and length, or those that cause FAS. As FASD is estimated to occur in approximately 1% of live births (May & Gossage, 2001; Sampson et al, 1997), and this disorder is believed to be underdiagnosed (May & Gossage, 2001; Chudley et al, 2005), a significant number of children may be affected by this effect of prenatal ethanol exposure. 112 4.2 S T R E N G T H S A N D LIMITATIONS The experiments presented in this thesis provide novel and important findings on the effects of prenatal ethanol exposure on endochondral bone development. These studies are the first to examine the dose response effects of prenatal ethanol exposure on endochondral ossification and provide evidence that ethanol's effects on the developing skeleton occur not only at high levels of exposure, but also with more moderate exposure levels. Moreover, they support the hypothesis that ethanol specifically affects bone (independent of its effects on general growth). The use of animal models provides several advantages in the investigation of ethanol's effects on fetal development. Human studies are often complicated by a number of confounding variables (e.g., polydrug use, nutrition, socioeconomic status) and the use of animal models eliminates these confounders. In addition, interpretation of human studies is complicated by the difficulties associated with estimating the level of ethanol exposure (e.g., pattern of drinking varies among different women and within a single woman at different times during a pregnancy, self-report of alcohol intake used in human studies is often inaccurate) (Streissguth et al, 1980), whereas animal studies allow for control over the dose of ethanol. On the other hand, a limitation of animal models is that there are biological differences between species and these may prevent generalization of what is found in animal models to the human. Therefore, it is important that the animal chosen adequately models the human condition under investigation in a given study. Rats are born with less developed skeletons than humans (equivalent to approximately a second trimester human fetus) and thus this species models the effects of ethanol exposure 113 during the first two trimesters of human pregnancy. Similarly, it is important to know that the effects being modeled by the chosen species also occur in humans. Only one study has examined the effect of prenatal alcohol exposure on bone in humans (Habbick et al, 1998), so clearly more study of this phenomenon in humans is required. One limitation of this thesis is the use of areal measurements of the different chondrocyte zones as a proxy for bone volume measurements. Areal measurements are not always equivalent to volume measurements (e.g., composition of zones may not be uniform throughout; bone geometry may differ between bones) and so confirmation of the effects on chondrocyte zone height and area observed in this thesis using volumetric measurements is warranted. In addition, we did not perform any measurements to assess the functional significance of ethanol's effects on fetal bone. Such measures might include bone mineral density and mechanical strength, as well as an examination of the microarchitecture by 3D microcomputed tomography (microCT). A recent study in sheep demonstrated that maternal binge alcohol exposure altered fetal bone strength (Given et al, 2004), but assessment of bone strength postnatally, as well as investigation of other measures of functional significance, are lacking. Moreover, the effect of prenatal ethanol exposure on bone health later in life remains unknown. Little is known about effects of prenatal ethanol exposure on the development of chronic disease later in life, including the long-term effects of ethanol on bone health. As FAS was first described in the medical literature in the early 1970s, the earliest diagnosed cohorts available for study are now only approaching 114 middle age, precluding studies of bone health in elderly individuals with FAS at this time. This thesis demonstrates that prenatal ethanol exposure specifically affects developing bone, and other studies have shown that the skeleton is subject to fetal programming (Mehta et al, 2002; Migliaccio et al, 1996, 2000; Swolin-Eide et al, 2004). Thus, it will be important to investigate if prenatal exposure to ethanol increases the risk of osteoporosis in later life as these cohorts age. 4.3 FUTURE DIRECTIONS 4.3.1 Elucidating Mechanisms Further work is needed to elucidate the mechanisms by which prenatal ethanol exposure exerts its effects on endochondral bone development. The results of this thesis suggest that the later stages of endochondral bone development, such as cartilage calcification, cartilage resorption, angiogenesis or bone formation, are disrupted by prenatal ethanol exposure. In addition, it appears that earlier stages may also be affected. Further studies to investigate these hypotheses should be conducted. For example, immunohistochemistry could be used to compare protein expression levels of factors (e.g., vascular endothelial growth factor [VEGF]) that control the various steps of endochondral bone development (e.g., angiogenesis) between ethanol-exposed and non-exposed fetal bone. As the results of this thesis suggest that the later stages of endochondral bone development are adversely affected by ethanol, it will be important to investigate these later stages more thoroughly. One of the drawbacks of the rat as a 115 model of fetal bone development is that, relative to humans, the fetal skeleton is less mature at birth (i.e., the fetal rat skeleton is at a developmental stage similar to a second trimester human fetus). Early postnatal exposure to ethanol has been used to study the effects of ethanol exposure on brain development during the third-trimester equivalent, by artificial rearing methods (West, 1993) or by intubation of the neonate (Tran et al, 2005) and these models have the potential to allow for similar studies in bone. Alternatively, an animal model such as the guinea pig, which undergoes more of its bone development in utero than the rat, may be useful in this regard. However, no studies investigating ethanol's effects on fetal bone development in the guinea pig have been published. Thus, the validation of the guinea pig as a model for these effects would be necessary before attempting to study the mechanisms of ethanol's effects in this species. Examining the effect of prenatal ethanol exposure on the process of intramembranous ossification is another important area for future research. Lee & Leichter (1983) demonstrated that prenatal ethanol exposure results in a delay in ossification of bones of the skull, which form by intramembranous ossification. Intramembranous and endochondral ossification share some features (e.g., vascular regression followed by mesenchymal cell condensation during early development and angiogenesis and osteoblastogenesis later in the process) and differ in other features (e.g., transition through a cartilage stage occurs during endochondral, but not intramembranous, ossification). Comparisons of the effect of ethanol on intramembranous and endochondral ossification could shed light into the mechanisms by which ethanol exerts its effects on developing bone. For example, 116 the results of this thesis suggest that prenatal ethanol exposure may be affecting later stages of endochondral bone development, such as resorption of calcified matrix, angiogenesis and ossification. If ethanol is affecting resorption of calcified matrix, one might not expect to observe effects on intramembranous ossification, whereas if ethanol is affecting angiogenesis or osteoblastogenesis, effects on intramembranous ossification would be expected. In addition, it is possible that the effects of ethanol on calcium regulation mediate at least some of ethanol's effects on developing bone. Prenatal ethanol exposure results in maternal and fetal hypocalcemia (Keiver et al, 1996; Keiver & Weinberg, 2003, 2004), with the severity of the fetal hypocalcemia being related to fetal BEC level, and thus, with the dose of ethanol (Keiver et al, 2004). Maternal and fetal hypocalcemia have been shown to be associated with adverse effects on skeletal development, including reduced bone mineralization (Chalon & Garel, 1985, Lima et al, 1993, Loughead et al, 1990, Rebut-Bonneton et al, 1983a, 1983b, Sinclair 1942), suggesting that altered maternal/fetal calcium homeostasis caused by maternal ethanol intake is a potential mechanism for ethanol's effects on fetal bone development. Moreover, disruptions in calcium homeostasis result in similar effects on fetal bone to those caused by prenatal ethanol-exposure (Miao et al, 2002, Rummens et al, 2002). To test this hypothesis, studies in which maternal blood iCa levels are normalized in ethanol-consuming dams could be employed. Potential methods for normalizing blood iCa could include feeding dams a calcium supplemented diet, or through the use of osmotic minipumps to deliver calcium directly into the extracellular fluid. 117 4.3.2 Fetal Programming (Potential for Long Term Effects) It will be important to determine the effects of prenatal ethanol exposure on the long-term bone health of the offspring. Studies could be conducted first using animal models to provide direction for research on human subjects. Determination of bone density and mechanical strength of newborn, adolescent and adult rats that have been exposed to ethanol in utero would shed light onto the potential long-term effects of prenatal ethanol exposure on bone health. In addition, studies to investigate the response of prenatally exposed bone to perturbation (e.g., ovariectomy to simulate the estrogen deficiency experienced during menopause) are warranted. Furthermore, as only one study has examined the bone of children with FAS (Habbick et al, 1998), it will be important to confirm any results of the animal models in human studies. For example, studies in which the bone density and fracture rates among infants, children.and adults with FASD are compared with age-matched controls are warranted to determine if prenatal ethanol exposure has short and/or long-term implications on bone health. In addition, obtaining accurate information on the level of maternal ethanol intake during pregnancy can be problematic, and thus predicting the level of damage to the fetus (and the corresponding level and nature of support and services the child will require) is difficult. The effects on the fetal skeleton appear to be dose dependent and if skeletal effects could be correlated with other effects (e.g., cognitive effects), bone may be useful as a marker for ethanol-induced damage. 118 REFERENCES Abad V, Meyers JL, Weise M, Gafni RI, Barnes KM, Nilsson 0, Bacher JD, Baron J (2002) The role of the resting zone in growth plate chondrogenesis. Endocrinology 143:1851-1857. Abel EL (1980) Procedural considerations in evaluating prenatal effects of alcohol in animals. Neurobehav Toxicol 2:167-174. Abel EL (1995) An update on incidence of FAS: FAS is not an equal opportunity birth defect. Neurotoxicol Teratol 17:437-443. Abel EL (1996) Effects of prenatal alcohol exposure on birth weight in rats: is there an inverted U-shaped function? Alcohol 13:99-102. Abel EL (2000) The role of dietary fat in alcohol's prenatal effects. Alcohol 20:83-86. Abel EL, Dintcheff BA (1978) Effects of prenatal alcohol exposure on growth and development in rats. J Pharmacol Exp Ther 207:916-921. Baran DR, Bryant C, Robson D (1982) Alcohol-induced alterations in calcium metabolism in the pregnant rat. Am J Clin Nutr 36:41-45. Barker DJP (1998) Mothers, babies and health in later life, 2nd ed. Edinburgh: Churchill Livingstone. Barker DJ (2004) The developmental origins of chronic adult disease. Acta Paediatr Suppl 93:26-33. 119 Bernstein, I, Gabbe SG, Reed KL (2002) Chapter 25 - Intrauterine Growth Restriction, in Obstetrics - Normal and Problem Pregnancies, 4th ed. (Gabbe SG, Niebyl JR, Simpson JL eds), pp. 869-892. Churchill Livingstone, New York. Available at http://home.mdconsult.com/das/book/53343329-2/view/1007/1 .html/top. Accessed Dec 21, 2005. Bertram C, Trowern AR, Copin N, Jackson AA, Whorwood CB (2001) The maternal diet during pregnancy programs altered expression of the glucocorticoid receptor and type 2 11 beta-hydroxysteroid dehydrogenase: potential molecular mechanisms underlying the programming of hypertension in utero. Endocrinology 142:2841-2853. Bielawski DM, Zaher FM, Svinarich DM, Abel EL (2002) Paternal alcohol exposure affects sperm cytosine methyltransferase messenger RNA levels. Alcohol Clin Exp Res 26:347-351. Braddock R, Siman CM, Hamilton K, Garland HO, Sibley CP (2002) Gamma-linoleic acid and ascorbate improves skeletal ossification in offspring of diabetic rats. Pediatr Res 51:647-652. Breese CR, D'Costa A, Sonntag WE (1994) Effect of in utero ethanol exposure on the postnatal ontogeny of insulin-like growth factor-1, and type-1 and type-2 insulin-like growth factor receptors in the rat brain. Neuroscience 63:579-589. Brien JF, Clarke DW, Richardson B, Patrick J (1985) Disposition of ethanol in maternal blood, fetal blood, and amniotic fluid of third-trimester pregnant ewes. Am J Obstet Gynecol 152:583-590. 120 Brodsky D, Christou H (2004) Current concepts in intrauterine growth restriction. J Intensive Care Med 19:307-319. Bucher T, Redetzki H (1951) Eine Spezifische Photometrische Bestimmung von thylakohoul auf Fermentativem Wege. Klin Wochenschr 29:615. Carlson BM (1996) Patten's Foundations of Embryology. 6 t h ed. McGraw Hill, New York. Chalon S, Garel JM (1985) Plasma calcium control in the rat fetus. I. Influence of maternal hormones. Biol Neonate 48:313-322. Chavassieux P, Serre CM, Vergnaud P, Delmas PD, Meunier PJ (1993) In vitro evaluation of dose-effects of ethanol on human osteoblastic cells. Bone Miner 22:95-103. Chen YH, Xu DX, Zhao L, Wang H, Wang JP, Wei W (2006) Ascorbic acid protects against lipopolysaccharide-induced intra-uterine fetal death and intra-uterine growth retardation in mice. Toxicology 217:39-45. Chudley AE, Conry J, Cook JL, Loock C, Rosales T, LeBlanc N (2005) Fetal alcohol spectrum disorder: Canadian guidelines for diagnosis. CMAJ 172:S1-S21. Coles C (1994) Critical periods for prenatal alcohol exposure: Evidence from animal and human studies. Alcohol Health Res World 18:22-29. Colnot C (2005) Cellular and molecular interactions regulating skeletogenesis. J Cell Biochem 95:688-697. Colnot C, Lu C, Hu D, Helms JA (2004) Distinguishing the contributions of the perichondrium, cartilage, and vascular endothelium to skeletal development. 269:55-69. 121 Cooper C, Cawley M, Bhalla A, Egger P, Ring F, Morton L, Barker D (1995) Childhood growth, physical activity, and peak bone mass in women. J Bone Miner Res 10:940-947. Cooper C, Erikson JG, Forsen T, Osmond C, Tuomilehto J, Barker DJP (2001) Maternal height, childhood growth and risk of hip fracture in later life: A longitudinal study. Osteoporos Int 12:623-629. Cooper C, Fall C, Egger P, Hobbs R, Eastell R, Barker D (1997) Growth in infancy and bone mass in later life. Ann Rheum Dis 56:17-21. Cooper C, Javaid MK, Taylor P, Walker-Bone K, Dennison E, Arden N (2002) The fetal origins of osteoporotic fracture. Calcif Tissue Int 70:391-394. Corstius HB, Zimanyi MA, Maka N, Herath T, Thomas W, van der LA, Wreford NG, Black MJ (2005) Effect of intrauterine growth restriction on the number of cardiomyocytes in rat hearts. Pediatr Res 57:796-800. Cudd TA (2005) Animal model systems for the study of alcohol teratology. Exp Biol Med (Maywood) 230:389-393. Das UG, Sysyn GD (2004) Abnormal fetal growth: intrauterine growth retardation, small for gestational age, large for gestational age. Pediatr Clin North Am 51:639-654. Day N, Richardson G, Geva D, Robles, N (1994) Alcohol, marijuana and tobacco: the effects of prenatal exposure on offspring growth and morphology of age six. Alcohol Clin Exp Res 18:786-794. 122 Day NL, Jasperse D, Richardson G, Robles N, Sambamoorthi U, Taylor P, Scher M, Staffer D, Cornelius M (1989) Prenatal exposure to alcohol: Effect on infant growth and morphologic characteristics. Pediatrics 84:536-541. Day NL, Leech SL, Richardson GA, Cornelius MD, Robles N, Larkby C (2002) Prenatal alcohol exposure predicts continued deficits in offspring size at 14 years of age. Alcohol Clin Exp Res 26:1584-1591. Dennison EM, Hindmarsh PC, Kellingray S, Fall CH, Cooper C (2003) Growth hormone predicts bone density in elderly women. Bone 32:434-440. Dennison EM, Syddall HE, Sayer AA, Gilbody HJ, Cooper C (2005) Birth weight and weight at 1 year are independent determinants of bone mass in the seventh decade: the Hertfordshire cohort study. Pediatr Res 57:582-586. Detering N, Reed WD, Ozand PT, Karahasan A (1979) The effects of maternal ethanol consumption in the rat on the development of their offspring. J Nutr 109:999-1009. Eastell R (1999) Pathogenesis of Postmenopausal Osteoporosis. In: Favus MJ (ed) Primer on the Metabolic Bone Diseases and Disorders of Mineral Metabolism. Lippincott Williams & Wilkins, Philadelphia, PA, p 260-262. Ebrahim SH, Luman ET, Floyd RL, Murphy CC, Bennett EM, Boyle CA (1998) Alcohol consumption by pregnant women in the United States during 1988-1995. Obstet Gynecol 92:187-192. Ergaz Z, Avgil M, Ornoy A (2005) Intrauterine growth restriction-etiology and consequences: what do we know about the human situation and experimental animal models? Reprod Toxicol 20:301-322. 123 Fall C, Hindmarsh P, Dennison E, Kellingray S, Barker D, Cooper C (1998) Programming of growth hormone secretion and bone mineral density in elderly men: a hypothesis. J Clin Endocrinol Metab 83:135-139. Farley JR, Fitzsimmons R, Taylor AK, Jorch UM, Lau KHW (1985) Direct effects of ethanol on bone resorption and formation in vitro. Arch Biochem Biophys 238:305-314. Floyd RL, Sidhu JS (2004) Monitoring prenatal alcohol exposure. Am J Med Genet Pt C (Semin Med Genet) 127C:3-9. Flynn HA, Marcus SM, Barry KL, Blow FC (2003) Rates and correlates of alcohol use among pregnant women in obstetrics clinics. Alcohol Clin Exp Res 27:81-87. Forrest JD (1994) Epidemiology of unintended pregnancy and contraceptive use. Am J Obstet Gynecol 170:1485-1489 Fowden AL (1995) Endocrine regulation of fetal growth. Reprod Fertil Dev 7:351-363. Fowden AL, Forhead AJ (2004) Endocrine mechanisms of intrauterine programming. Reproduction 127:515-526. Gabriel K, Hofmann C, Glavas M, Weinberg J (1998) The hormonal effects of alcohol use on the mother and fetus. Alcohol Health Res World 22: 170-177. Gaiser KG, Maddox BK, Bann JG, Boswell BA, Keene DR, Garofalo S, Horton WA (2002) Y-position collagen II mutation disrupts cartilage formation and skeletal development in a transgenic mouse model of spondyloepiphyseal dysplasia. J Bone Miner Res 17:39-47. 124 Gale CR, Martyn CN, Kellingray S, Eastell R, Cooper C (2001) Intrauterine programming of adult body composition. J Clin Endocrinol Metab 86:267-272. Gallo PV, Weinberg J (1982) Neuromotor development and response inhibition following prenatal ethanol exposure. Neurobehav Toxicol Teratol 4:505-513. Gallo PV, Weinberg J (1986) Organ growth and cellular development in ethanol-exposed rats. Alcohol 3:261-267. Genant HK, Cooper C, Poor G, Reid I, Ehrlich G, Kanis J, Nordin BE, Barrett-Connor E, Black D, Bonjour JP, Dawson-Hughes B, Delmas PD, Dequeker J, Ragi ES, Gennari C, Johnell O, Johnston CC, Jr., Lau EM, Liberman UA, Lindsay R, Martin TJ, Masri B, Mautalen CA, Meunier PJ, Khaltaev N. (1999) Interim report and recommendations of the World Health Organization Task-Force for Osteoporosis. Osteoporos Int 10:259-264. Gerber HP, Vu TH, Ryan AM, Kowalski J, Werb Z, Ferrara N (1999) VEGF couples hypertrophic cartilage remodeling, ossification and angiogenesis during endochondral bone formation. Nat Med 5:623-628. Geva D, Goldschmidt L, Staffer D, Day NL (1993) A longitudinal analysis of the effect of prenatal alcohol exposure on growth. Alcohol Clin Exp Res 17:1124-1129. Given J, Hogan HA, West JR, Cudd TA (2004) Binge alcohol exposure during all 3 trimesters alters fetal bone strength in sheep. Alcohol Clin Exp Res 28:42A. 125 Godfrey K, Walker-Bone K, Robinson S, Taylor P, Shore S, Wheeler T, Cooper C (2001) Neonatal bone mass: influence of parental birthweight, maternal smoking, body composition, and activity during pregnancy. J Bone Miner Res 16:1694-1703. Goldring MB, Tsuchimochi K, Ijiri K (2006) The control of chondrogenesis. J Cell Biochem 97:33-44. Gong Z, Wezeman FH (2004) Inhibitory effect of alcohol on osteogenic differentiation in human bone marrow-derived mesenchymal stem cells. Alcohol Clin Exp Res 28:468-479. Goodlett CR, Horn KH, Zhou FC (2005) Alcohol teratogenesis: mechanisms of damage and strategies for intervention. Exp Biol Med (Maywood) 230:394-406. Gress CJ, Jacenko O (2000) Growth plate compressions and altered hematopoiesis in collagen X null mice. J Cell Bio 149:983-993. Gu JW, Elam J, Sartin A, Li W, Roach R, Adair TH (2001) Moderate levels of ethanol induce expression of vascular endothelial growth factor and stimulate angiogenesis. Am J Physiol Regul IntegrComp Physiol 281 :R365-R372 Guerri C, Sanchis R (1985) Acetaldehyde and alcohol levels in pregnant rats and their fetuses. Alcohol 2:267-270. Habbick BF, Blakely PM, Houston CS, Snyder RE, Senthilselvan A, Nanson JL (1998) Bone age and growth in fetal alcohol syndrome. Alcohol Clin Exp Res 22:1312-1316. 126 Haigh JJ, Gerber HP, Ferrara N, Wagner EF (2000) Conditional inactivation of VEGF-A in areas of collagen2a1 expression results in embryonic lethality in the heterozygous state. Development 127:1445-1453. Hall BK, Miyake T (2000) All for one and one for all: condensations and the initiation of skeletal development. Bioessays 22:138-147. Han VKM (1999) Intrauterine growth restriction and mechanisms of fetal growth. In: Knobil E, Neill JD (eds) Encyclopedia of Reproduction, Volume 2. Academic Press, San Diego, p 877-892. Hannigan JH, Abel EL, Kruger ML (1993) \"Population\" characteristics of birthweight in an animal model of alcohol-related developmental effects. Neurotoxicol Teratol 15:97-105. Hanson M, Gluckman P, Bier D, Challis J, Fleming T, Forrester T, Godfrey K, Nestel P, Yajnik C (2004) Report on the 2nd World Congress on Fetal Origins of Adult Disease, Brighton, U.K., June 7-10, 2003. Pediatr Res 55:894-897. Harding JE (2001) The nutritional basis of the fetal origins of adult disease. Int J Epidemiol 30: 15-23. Harvey N, Cooper C (2004) The developmental origins of osteoporotic fracture. J Br Menopause Soc 10:14-5, 29. Health Canada. Joint Statement: Prevention of Fetal Alcohol Syndrome (FAS) Fetal Alcohol Effects (FAE) in Canada. Cat.: H39-348/1996E, October 1996. Accessed 22 July 2005, http://www.phac-aspc.gc.ca/dca-dea/publications/pdf/fasjoint_statement_e.pdf. 127 Hingson RW, Heeren T, Winter MR (1999) Preventing impaired driving. Alcohol Res Health 23:31-39. Howell DS, Dean DD. (1992) The biology, chemistry, and biochemistry of the mammalian growth plate. In Coe FL, Favis MJ (eds) Disorders of Bone and Mineral Metabolism, New York: Raven Press, pp. 313-353. Hughes PCR, Tanner JM (1970) The assessment of skeletal maturity in the growing rat. JAnat 106:371-402. Hunziker EB (1994) Mechanism of longitudinal bone growth and its regulation by growth plate chondrocytes. Microsc Res Tech 28:505-519. Institute pf Medicine (1996) Fetal Alcohol Syndrome. Diagnosis, Epidemiology, Prevention, and Treatment. National Academy Press, Washington, DC. Jacobson JL, Jacobson SW (1999) Drinking moderately and pregnancy. Effects on child development. Alcohol Res Health 23:25-30. Jacobson JL, Jacobson SW, Sokol RJ (1994) Effects of prenatal exposure to alcohol, smoking, and illicit drugs on postpartum somatic growth. Alcohol Clin Exp Res 18:317-323. Jones G, Dwyer T (2000) Birth weight, birth length, and bone density in prepubertal children: evidence for an association that may be mediated by genetic factors. Calcif Tissue Int 67:304-308. Jones G, Riley M, Dwyer T (1999) Maternal smoking during pregnancy, growth, and bone mass in prepubertal children. J Bone Miner Res 14:146-151. Jones KL, Smith DW (197,3) Recognition of the fetal alcohol syndrome in early infancy. Lancet 2:999-1001. 128 Jones KL, Smith DW, Ulleland CN, Streissguth P (1973) Pattern of malformation in offspring of chronic alcoholic mothers. Lancet 1:1267-1271. Kapoor A, Matthews SG (2005) Short periods of prenatal stress affect growth, behaviour and hypothalamo-pituitary-adrenal axis activity in male guinea pig offspring. J Physiol 566:967-977. Karsenty G, Wagner EF (2002) Reaching a genetic and molecular understanding of skeletal development. Dev Cell 2:389-406. Keiver K, ENis L, Anzarut A, Weinberg J (1997) Effect of prenatal ethanol exposure on fetal calcium metabolism. Alcohol Clin Exp Res 21:1612-1618. Keiver K, Herbert L, Weinberg J (1996) Effect of maternal ethanol consumption on maternal and fetal calcium metabolism. Alcohol Clin Exp Res 20:1305-1312. Keiver K, Weinberg J (2003) Effect of duration of alcohol consumption on calcium and bone metabolism during pregnancy in the rat. Alcohol Clin Exp Res 27:1507-1519. Keiver K, Weinberg J (2004) Effect of duration of maternal alcohol consumption on calcium metabolism and bone in the fetal rat. Alcohol Clin Exp Res 28:456-467. Klein RF (1997) Alcohol-induced bone disease: impact of ethanol on osteoblast proliferation. Alcohol Clin Exp Res 21:392-399. Klug MG, Burd L, Martsolf JT, Ebertowski M (2003) Body mass index in fetal alcohol syndrome. Neurotox Teratol 25:689-696. 129 Kovacs CS, Chafe LL, Fudge NJ, Friel JK, Manley NR (2001a) PTH regulates fetal blood calcium and skeletal mineralization independently of PTHrP. Endocrinology 142:4983-4993. Kovacs CS, Lanske B, Hunzelman JL, Guo J, Karaplis AC, Kronenberg HM (1996) Parathyroid hormone-related peptide (PTHrP) regulates fetal-placental calcium transport through a receptor distinct from the PTH/PTHrP receptor. Proc Natl Acad Sci USA 93:15233-15238. Kovacs CS, Manley NR, Moseley JM, Martin TJ, Kronenberg HM (2001b) Fetal parathyroids are not required to maintain placental calcium transport. J Clin Invest 107:1007-1015. Koyama E, Leatherman JL, Shimazu A, Nah HD, Pacifici M (1995) Syndecan-3, tenascin-C, and the development of cartilaginous skeletal elements and joints in chick limbs. Dev Dyn 203:152-62. Kronenberg HM (2003) Developmental regulation of the growth plate. Nature 423:332-336. Kulyk WM, Hoffman LM (1996) Ethanol exposure stimulates cartilage differentiation by embryonic limb mesenchyme cells. Exp Cell Res 223:290-300. Lee M (1987) Early fetal growth and development in mice chronically exposed to ethanol during gestation. Growth 51:146-153. Lee M, Leichter J (1980) Effect of litter size on the physical growth and maturation of the offspring of rats given alcohol during gestation. Growth 44:327-335. Lee M, Leichter J (1983) Skeletal development in fetuses of rats consuming alcohol during gestation. Growth 47:254-262. 130 Leichter J, Lee M (1979) Effect of maternal ethanol administration on physical growth of the offspring in rats. Growth 43:288-293. Lemoine P, Harousseau H, Borteyru JP, Menuet JC (1968) Children of alcoholic parents: abnormalities observed in 127 cases. Ouest Medical 8:476-482. Leon DA (1998) Fetal growth and adult disease. Eur J Clin Nutr 52 Suppl 1:S72-S78. Lian JB, Stein SG, Canalis E, Robey PG, Boskey AL (1999) Chapter 3: Bone Formation: Osteoblast Lineage Cells, Growth Factors, Matrix Proteins and the Mineralization Process, in Primer on the Metabolic Bone Diseases and Disorders of Mineral Metabolism. 4 th ed. (Favus MJ ed.), pp. 14-29. Lippincott Williams & Wilkins: Philadelphia, PA. Lima MS, Kallfelz F, Krook L, Nathanielsz PW (1993) Humeral skeletal development and plasma constituent changes in fetuses of ewes maintained on a low calcium diet from 60 days of gestation. Calcif Tissue Int 52:283-290. Livy DJ, Parnell SE, West JR (2003) Blood ethanol concentration profiles: a comparison between rats and mice. Alcohol 29:165-171. Lochry EA, Shapiro NR, Riley EP (1980) Growth deficits in rats exposed to alcohol in utero. J Stud Alcohol 41:1031-1039. Loughead JL, Mughal Z, Mimouni F, Tsang RC, Oestreich AE (1990) Spectrum and natural history of congenital hyperparathyroidism secondary to maternal hypocalcemia. Am J Perinatol 7:350-355. 131 Ludefia MC, Mena MA, Salinas M, Herrera E (1983) Effects of alcohol ingestion in the pregnant rat on daily food intake, offspring growth and metabolic parameters. Gen Pharmacol 14:327-332. MacLeod MJ (1980) Differential staining of cartilage and bone in whole mouse fetuses by Alcian blue and Alizarin red S. Teratology 22:299-301. Maier SE, Strittmatter MA, Chen WJ, West JR (1995) Changes in blood alcohol levels as a function of alcohol concentration and repeated alcohol exposure in adult female rats: Potential risk factors for alcohol-induced fetal brain injury. Alcohol Clin Exp Res 19:923-927. Maier SE, West JR (2001) Drinking patterns and alcohol-related birth defects. Alcohol Res Health 25:168-174. Martin JC, Martin DC, Radow B, Sigman G (1978) Blood alcohol level and caloric intake in the gravid rat as a function of diurnal period, trimester, and vehicle. Pharmacol Biochem Behav 8:421-427. May PA, Gossage JP (2001) Estimating the prevalence of fetal alcohol syndrome. A summary. Alcohol Res Health. 25:159-167. Mauceri HJ, Lee WH, Conway S (1994) Effect of ethanol on insulin-like growth factor-ll release from fetal organs. Alcohol Clin Exp Res 18:35-41. Mehta G, Roach HI, Langley-Evans S, Taylor P, Reading I, Oreffo RO, Aihie-Sayer A, Clarke NM, Cooper C (2002) Intrauterine exposure to a maternal low protein diet reduces adult bone mass and alters growth plate morphology in rats. Calcif Tissue Int 71:493-498. 132 Miao D, He B, Karaplis AC, Goltzman D (2002) Parathyroid hormone is essential for normal fetal bone formation. J Clin Invest 109:1173-1182. Migliaccio S, Newbold RR, Bullock BC Jefferson WJ, Sutton FG Jr, McLachlan JA, Korach KS (1996) Alterations of maternal estrogen levels during gestation affect the skeleton of female offspring. Endocrinology 137:2118-2125. Migliaccio S, Newbold RR, Teti A, Jefferson WJ, Toverud SU, Taranta A, Bullock BC, Suggs CA, Spera G, Korach KS (2000) Transient estrogen exposure of female mice during early development permanently affects osteoclastogenesis in adulthood. Bone 27:47-52. Miralles-Flores C, Delgado-Baeza E (1992) Histomorphometric analysis of the epiphyseal growth plate in rats after prenatal alcohol exposure. J Orthop Res 10:325-336. Moore KL, Persaud TVN (1998) The Skeletal System. The Developing Human: Clinically Oriented Embryology. W. B. Saunders, Philadelphia, PA, p 405-424. Mushtaq T, Bijman P, Ahmed SF, Farquharson C (2004) Insulin-like growth factor-l augments chondrocyte hypertrophy and reverses glucocorticoid-mediated growth retardation in fetal mice metatarsal cultures. Endocrinology 145:2478-2486. Olszynski WP, Shawn Davison K, Adachi JD, Brown JP, Cummings SR, Hanley DA, Harris SP, Hodsman AB, Kendler D, McClung MR, Miller PD, Yuen CK (2004) Osteoporosis in men: epidemiology, diagnosis, prevention, and treatment. Clin Ther 26:15-28. 133 Osteoporosis Canada (2005) About osteoporosis [Osteoporosis Canada website] Available at: http://www.osteoporosis.ca/english/About%200steoporosis/ default.asp?s=1. Accessed October 7, 2005. Pham TD, MacLennan NK, Chiu CT, Laksana GS, Hsu JL, Lane RH (2003) Uteroplacental insufficiency increases apoptosis and alters p53 gene methylation in the full-term IUGR rat kidney. Am J Physiol Regul Integr Comp Physiol 285:R962-R970. Pierce DR, West JR (1986) Blood alcohol concentration: a critical factor for producing fetal alcohol effects. Alcohol 3:269-272. Portney LG, Watkins MP (2000) Foundations of Clinical Research: Applications to Practice, 2 n d ed. Upper Saddle River: Prentice Hall, 2000 Pullen GL, Singh SP, Snyder AK (1988) Growth patterns of the offspring of alcohol-fed rats. Growth Dev Aging 52:85-89. Rasband, W.S., Image J (1997-2005) U.S. National Institutes of Health, Bethesda, Maryland, USA, http://rsb.info.nih.gov/ij/ Radek KA , Matties AM, Burns AL, Heinrich SA, Kovacs EJ, Dipietro LA (2005) Acute ethanol exposure impairs angiogenesis and the proliferative phase of wound healing. Am J Physiol Heart Circ Physiol 289:H1084-H1090. Rebut-Bonneton C, Demignon J, Amor B, Miravet L (1983a) Effect of calcitonin in pregnant rats on bone resorption in fetuses. J Endocrinol 99:347-353. Rebut-Bonneton C, GarelJM, Delbarre F (1983b) Parathyroid hormone, calcitonin, 1,25-dihydroxycholecalciferol, and basal bone resorption in the rat fetus. Calcif Tissue Int 35:183-189. 134 Reyes E, Garcia KD, Jones BC (1985) Effects of the maternal consumption of alcohol on alcohol selection in rats. Alcohol 2:323-326. Riley EP, Meyer LS (1984) Considerations for the design, implementation, and interpretation of animal models of fetal alcohol effects. Neurobehav Toxicol Teratol 6:97-101. Roberts G, Nanson J (2000) Best practices, fetal alcohol syndrome/fetal alcohol effects and the effects of other substance use during pregnancy. Health Canada, Ottawa. Robinson GC, Conry JL, Conry RF (1987) Clinical profile and prevalence of fetal alcohol syndrome in an isolated community in British Columbia. CMAJ 137:203-207. Robson H, SieblerT, Shalet SM, Williams GR (2002) Interactions between GH, IGF-I, glucocorticoids, and thyroid hormones during skeletal growth. Pediatr Res 52:137-147. Rummens K, van Bree R, Van Herck E, Zaman Z, Bouillon R, Van Assche FA, Verhaeghe J (2002) Vitamin D deficiency in guinea pigs: exacerbation of bone phenotype during pregnancy and disturbed fetal mineralization, with recovery by 1,25(OH)2D3 infusion or dietary calcium-phosphate supplementation. Calcif Tissue Int 71:364-375. Sampson HW (1998) Effect of alcohol consumption on adult and aged bone: a histomorphometric study of the rat animal model. Alcohol Clin Exp Res 22:2029-2034. 135 Sampson HW, Hebert VA, Booe HL, Champney TH (1998) Effect of alcohol consumption on adult and aged bone: composition, morphology, and hormone levels of a rat animal model. Alcohol Clin Exp Res 22:1746-1753. Sampson HW, Perks N, Champney TH, DeFee B (1996) Alcohol consumption inhibits bone growth and development in young actively growing rats. Alcohol Clin Exp Res 20:375-384. Sampson PD (1997) Incidence of fetal alcohol syndrome and prevalence of alcohol-related neurodevelopmental disorder. Teratology 56:317-326. Samson HH (1981) Maternal ethanol consumption and fetal development in the rat: a comparison of ethanol exposure techniques. Alcohol Clin Exp Res 5:67-74. Sanchis R, Guerri C (1986) Alcohol-metabolizing enzymes in placenta and fetal liver: effect of chronic ethanol intake. Alcohol Clin Exp Res 10:39-44. Sanchez CP, He YZ (2002) Alterations in the growth plate cartilage of rats with renal failure receiving corticosteroid therapy. 30:692-698. Sankaran H, Baba GC, Deveney CW, Rao GA (1991) Enteral macronutrients abolish high blood alcohols levels in chronic alcoholic rats. Nutr Res 11: 217-222. Savage DD, Becher M, de la Torre AJ, Sutherland RJ (2002) Dose-dependenteffects of prenatal ethanol exposure on synaptic plasticity and learning in mature offspring. Alcohol Clin Exp Res 26:1752-1758. Shrout PE, Fleiss JL. Intraclass correlations: uses in assign rater reliability (1979) Psychol Bull 86:420-428. 136 Sinclair JG (1942) Fetal rat parathyroids as affected by changes in maternal serum calcium and phosphorus through parathyroidectomy and dietary control. J Nutr 23:141-152. Singh SP, Snyder AK (1982) Ethanol ingestion during pregnancy: effects on pregnant rats and their offspring. J Nutr 112:98-103. Smink JJ, Buchholz IM, Hamers N, van Tilburg CM, Christis C, Sakkers RJ, de Meer K, van Buul-Offers SC, Koedam JA (2003a) Short-term glucocorticoid reatment of piglets causes changes in growth plate morphology and angiogenesis. Osteoarthritis Cartilage 11:864-871 Smink JJ, Gresnigt MG, Hamers N, Koedam JA, Berger R, Van Buul-Offers SC (2003b) Short-term glucocorticoid treatment of prepubertal mice decreases growth and IGF-I expression in the growth plate. J Endocrinol 177:381-388. Smith EJ, Little DG, Briody JN, McEvoy A, Smith NC, Eisman JA, Gardiner EM (2005) Transient disturbance in physeal morphology is associated with long-term effects of nitrogen-containing bisphosphonates in growing rabbits. J Bone Miner Res 20:1731 -1741. Sokol RJ, Delaney-Black V, Nordstrom B (2003) Fetal alcohol spectrum disorder. JAMA 290:2996-2999. Sokol RJ, Ager J, Martier S, Debanne S, Ernhart C, Kuzma J, Miller SI (1986) Significant determinants of susceptibility to alcohol teratogenicity. Ann NY Acad Sci 477:87-102. 137 Sood B, Delaney-Black V, Covington C, Nordstrom-Klee B, Ager J, Templin T, Janisse J, Martier S, Sokol RJ. (2001) Prenatal alcohol exposure and childhood behavior at age 6 to 7 years: I. dose-response effect. Pediatrics 108(2):E34. Sparks JW, Ross JC, Cetin I (1998) Intrauterine Growth and Nutrition. In: Polin R, Fox W (eds) Fetal and Neonatal Physiology. W. B. Saunders, Philadelphia, PA, p 267-289. St Jacques B, Hammerschmidt M, McMahon AP (1999) Indian hedgehog signaling regulates proliferation and differentiation of chondrocytes and is essential for bone formation. Genes Dev 13:2072-2086. Stickens D, Behonick DJ, Ortega N, Heyer B, Hartenstein B, Yu Y, Fosang AJ, Schorpp-Kistner M, Angel P, Werb Z (2004) Altered endochondral bone development in matrix metalloproteinase 13-deficient mice. Development 131:5883-5895. Streissguth AP, Aase JM, Clarren SK, Randels SP, LaDue RA, Smith DF (1991) Fetal alcohol syndrome in adolescents and adults. JAMA 265:1961-1967. Streissguth AP, Landesman-Dwyer S, Martin JC, Smith DW (1980) Teratogenic effects of alcohol in humans and laboratory animals. Science 209:353-361. Strong RM (1925) The order, time and rate of ossification of the albino rat (Mus Norvegicus Albinus) skeleton. Am J Anat 36:313-355. Swolin-Eide D, Nilsson C, Holmang A, Ohlsson C (2004) Prenatal exposure to IL-1beta results in disturbed skeletal growth in adult rat offspring. Pediatr Res 55:598-603. 138 Tobias, JH, Cooper C (2004) PTH/PTHrP activity and the programming of skeletal development in utero. J Bone Miner Res 19:177-182. Tran TD, Jackson HD, Horn KH, Goodlett CR (2005) Vitamin E does not protect against neonatal ethanol-induced cerebellar damage or deficits in eyeblink classical conditioning in rats. Alcohol Clin Exp Res 29:117-129. Tuan RS (2004) Biology of developmental and regenerative skeletogenesis. Clin Orthop Relat Res S105-S117. Turner RT, Aloia RC, Segel LD, Hannon KS, Bell NH (1988) Chronic alcohol treatment results in disturbed vitamin D metabolism and skeletal abnormalities in rats. Alcohol Clin Exp Res 12:159-162. U.S. Department of Agriculture/U.S. Department of Health and Human Services (1990) Home and Garden Bulletin No. 232. Nutrition and Your Health: Dietary Guidelines for Americans. 3rd ed.: Supt. of Docs., U.S. Govt. Print. Off., Washington. Urso T, Gavaler JS, Van Thiel DH (1981) Blood ethanol levels in sober alcohol users seen in an emergency room. Life Sci 28:1053-1056. US Surgeon General (2005) U.S. Surgeon General Releases Advisory on Alcohol Use in Pregnancy [US Surgeon General website] Available at http://www.surgeongeneral.gov/pressreleases/sg02222005.html. Accessed July 27, 2005. van Buul-Offers SC, Smink JJ, Gresnigt R, Hamers N, Koedam J, Karperien M (2005) Thyroid hormone, but not parathyroid hormone, partially restores glucocorticoid-induced growth retardation. Pediatr Nephrol 20:335-341. 139 Vavrousek-Jakuba EM, Baker RA, Shoemaker WJ (1991) Effect of ethanol on maternal and offspring characteristics: comparison of three liquid diet formulations fed during gestation. Alcohol Clin Exp Res 15:129-135. Verhaeghe J, van Bree R, Van Herck E, Rummens K, Vercruysse L, Bouillon R, Pijnenborg R (1999) Pathogenesis of fetal hypomineralization in diabetic rats: evidence for delayed bone maturation. Pediatr Res 45:209-217. Vu TH, Shipley JM, Bergers G, Berger JE, Helms JA, Hanahan D, Shapiro SD, Senior RM, Werb Z (1998) MMP-9/gelatinase B is a key regulator of growth plate angiogenesis and apoptosis of hypertrophic chondrocytes. Cell 93:411-422. Wan Q, Liu Y, Guan Q, Gao L, Lee KO, Zhao J (2005) Ethanol feeding impairs insulin-stimulated glucose uptake in isolated rat skeletal muscle: role of Gs alpha and cAMP. Alcohol Clin Exp Res 29:1450-1456. Warren KR, Foudin LL (2001) Alcohol-related birth defects\u00E2\u0080\u0094The past, present, and future. Alcohol Res Health 25:153-158. Weinberg J (1984) Nutritional issues in perinatal alcohol exposure. Neurobehav Toxicol Teratol 6:261-269. Weinberg J (1996) New directions in Fetal Alcohol Syndrome research. Alcohol Clin Exp Res 20:72A-77A. Weinberg J (1985) Effects of ethanol and maternal nutritional status on fetal development. Alcohol Clin Exp Res 9:49-55. 140 Weinberg J, D'Alquen G, Bezio S (1990) Interactive effects of ethanol intake and maternal nutritional status on skeletal development of fetal rats. Alcohol 7:383-388. West JR (1993) Use of pup in a cup model to study brain development. J Nutr 123:382-385. Wezeman FH, Gong Z (2001) Bone marrow triglyceride accumulation and hormonal changes during long-term alcohol intake in male and female rats. Alcohol Clin Exp Res 25:1515-1522. Wezeman FH, Gong Z (2004) Adipogenic effect of alcohol on human bone marrow-derived mesenchymal stem cells. Alcohol Clin Exp Res 28:1091-1101. Wiener SG, Shoemaker WJ, Koda LY, Bloom FE (1981) Interaction of ethanol and nutrition during gestation: influence on maternal and offspring development in the rat. J Pharmacol Exp Ther 216:572-579. Zar, JH (1984) Biostatistical Analysis. 2nd ed. Prentice-Hall, Englewood Cliffs. Zelzer E, Mamluk R, Ferrara N, Johnson RS, Schipani E, Olsen BR (2004) VEGFA is necessary for chondrocyte survival during bone development. Development 131:2161-2171. Zhang X, Sliwowska JH, Weinberg J (2005) Prenatal alcohol exposure and fetal programming: effects on neuroendocrine and immune function. Exp Biol Med (Maywood) 230:376-388. 141 APPENDICES Appendix A Assessment of Blood Ethanol Concentration Alcohol assessment kits from Pointe Scientific Inc. (Lincoln Park, Ml) and Sigma Diagnostics (St. Louis, MO) were used to assess BECs for Experiments #1 and #2, respectively. Different suppliers were used based on availability of the kits at the time the analyses were conducted. Both kits use the same procedure (outlined below), based on a modification of the method developed by Bucher & Redetzki (1951). In this assay, serum from PF and C rats (negative controls) have BEC values of 0-22 mg/dL; therefore values <25 mg/dL were considered to be negative for ethanol (or undetectable). Principle Alcohol dehydrogenase catalyzes the oxidation of ethanol to acetaldehyde. This reaction (shown below) converts NAD to NADH. The increase in absorbance at 340 nm, measured by the spectrophotometer, is directly proportional to the concentration of ethanol. ADH ethanol + NAD \u00E2\u0080\u00A2 acetaldehyde + NADH + H + Materials 1. alcohol reagent (contains yeast alcohol dehydrogenase (ADH), nicotinamide adenine dinucleotide (NAD), buffer and stabilizers). 142 2. distilled water 3. ethanol standard solution (80 mg/dL) Methods 1. Serum samples were thawed to room temperature. 2. Alcohol reagent was reconstituted with distilled water, gently inverted the vial to dissolve the reagent. 3. 3 mL of reagent solution was added to a cuvet for each of the blank, the standard and the serum samples. 4. 10 uL of distilled water was added to the blank cuvet. 5. 10 uL of ethanol standard solution was added to the standard cuvet. 6. 10 uL of each serum sample was added to their respective cuvets. 7. The cuvets were covered with Parafilm and mixed by gentle inversion. 8. Cuvets were incubated at room temperature for 10 min. 9. The spectrophotometer was set to an absorbance of 340 nm. 10. The absorbance at 340 nm of the ethanol standard and the serum samples, compared to the blank solution, was measured. 11. The following calculation was used to determine ethanol concentrations: Ethanol (mg/dL) = absorbance of sample X 80 mg/dL absorbance of standard where 80 mg/dL is the concentration of the ethanol standard solution 143 Appendix B Alcian Blue and Alizarin Red Staining Procedure The staining procedure used has been previously described (Weinberg et al, 1990) and is a modification of the method of MacLeod (1980). Principle Alcian blue stains cartilage tissue, while alizarin red stains for calcium. Thus, after using this staining procedure, the ends of the bone, consisting of non-mineralized cartilage, appear blue, while the diaphysis, consisting of both bone tissue and mineralized cartilage matrix, appears red. Materials 1. 95% ethanol 2. 70% ethanol 3. acetone 4. staining solution (1 part 0.3% filtered alcian blue in 70% ethanol, 1 part 0.1% filtered alizarin red S in 95% ethanol, 1 part glacial acetic acid, 17 parts 70% ethanol) 5. distilled water 6. 1% potassium hydroxide 7. glycerine 144 Methods 1. After decapitation, fetuses were fixed in 95% ethanol for 2-3 h. 2. Fetuses were skinned and eviscerated and then returned to jars containing 95% ethanol for 3 d. 3. Fetuses were transferred to jars containing 70% ethanol for storage until the remainder of the procedure could be conducted. 4. Fetuses were transferred to jars containing 95% ethanol for 2 d. 5. Fetuses were transferred to jars containing acetone, which removes fat and keeps the specimen firm, for 3 d. 6. Fetuses were placed in jars containing the staining solution for 3 d in a water bath at 37QC, with the jars containing the fetuses and solution being gently agitated by hand daily during this time. The staining solution was prepared fresh on the day in which the fetuses are placed in staining solution. 7. Fetuses were rinsed with distilled water. 8. Fetuses were placed in jars containing 1% potassium hydroxide for 1-7 d until soft tissue was cleared. Fetuses were checked daily to assess the level of clearing in order to determine when to proceed to the next step. 9. Fetuses were transferred to jars containing a 1:4 solution of glycerine: 1% potassium hydroxide for 1 wk, to replace the potassium hydroxide solution with glycerine in order to stop the clearing process and prepare the fetuses for storage in glycerine. 145 10. Fetuses were transferred to jars containing a 1:1 solution of glycerine: 1% potassium hydroxide for 1 wk. 11. Fetuses were transferred to jars containing a 4:1 solution of glycerine:1% potassium hydroxide for 1 wk. 12. Fetuses were stored in 100% glycerine. 146 Appendix C Assessment of Fetal Skeletal Ossification After staining (as per the procedure outlined in Appendix B), fetal bones were assessed for the degree of ossification, which is defined as the portion stained with Alizarin red. Alizarin red binds to calcium, and so stains both the bone tissue and calcified cartilage. The scapula, humerus, ulna, radius, femur and tibia were assessed by comparing the length of the ossified portion to the total bone length. Fetal sacrum, sternum and metatarsals were assessed for the number of ossification centres present. The atlas compiled by Menegola et al (2001) was used as a guide. A single investigator, blinded to treatment groups, made all assessments. Methods 1. Fetuses stored in glycerine were examined using a Zeiss dissecting microscope with a linear eyepiece reticule. 2. The number of sternabrae in which ossification centres were present was determined with the fetus placed on its dorsal surface (i.e. ventral side facing upwards). 3. The fetus was next placed on its ventral surface (i.e., dorsal side facing upwards) to assess the sacrum, which consists of five fused vertebrae. The number of sacral vertebral bodies and processes in which ossification centres was present were recorded. 147 4. The forelimb was disarticulated by gentle pulling with forceps, keeping the scapula, humerus, radius, ulna and paw attached together. 5. The forelimb was placed on its ventral surface (i.e., with the palm facing upwards and the coracoid process of the scapula facing down) to assess ossification of the scapula. The total length of the scapula was measured from the tip of the scapular spine (i.e., the acromion) to the far end of the scapula. The ossified portion of the scapula was measured from one end of the red portion to the other, along the same axis as total length was measured. 6. The forelimb was placed on its dorsal surface (i.e., ventral side up, with palm facing upwards and scapular spine facing down) to assess the humerus, ulna and radius. The total length of the humerus was measured along the long scapula, to the trochlea, where it articulates with the olecranon of the ulna. The ossified portion of the humerus was measured from one end of the red portion to the other, along the same axis as total length was measured. axis of the bone, from the head of the humerus, where it articulates with the 148 7. The total length of the ulna was measured along the long axis of the bone, from the olecranon to the styloid process. The ossified portion of the ulna was measured from one end of the red portion to the other, along the same axis as total length was measured. 8. The total length of the radius was measured along the long axis of the bone, from the head of the radius to the styloid process. The ossified portion of the radius was measured from one end of the red portion to the other, along the same axis as total length was measured. 9. The hindlimb was disarticulated by detaching the head of the femur from the acetabulum of the pelvic girdle, keeping the bones of the hindlimb attached together and taking care to avoid breaking off the head of the femur, as the bones are delicate and this joint tends to be tightly attached. This is best achieved by securing the ischium by placing the forceps through the obturator foramen using the left hand and gently, using forceps in the right hand, guiding the forceps up and over the head of the femur to separate it from the hip joint. 10. The hindlimb was placed medial side facing upwards to assess the femur and tibia. The total length of the femur was measured along the long axis of the bone, from the head of the femur to the far end of the femur, along the ventral 149 surface of the bone. The ossified portion of the femur was measured from one end of the red portion to the other, along the same axis as total length was measured. 11 .The total length of the tibia was measured along the long axis of the bone from the head of the tibia to the far end of the tibia, where it articulates with the talus, along the ventral surface of the bone. The ossified portion of the tibia was measured from one end of the red portion to the other, along the same axis as total length was measured. 12.The hindpaw was placed on its plantar surface (i.e., dorsal side facing upwards) to assess the metatarsals. The number of metatarsals in which ossification centres were present was recorded. patella 150 Appendix D Animal Care Certificates The University of British Columbia ANIMAL CARE CERTIFICATE PROTOCOL NUMBER: A00-0148 INVESTIGATOR OR COURSE DIRECTOR: Keiver, K . M . DEPARTMENT: Agricultural Sciences PROJECT OR COURSE TITLE: Mechanisms of Alcohol's Kffecls on Bone Metabolism ANIMALS. Rats 156 START DATE: 99-05-01 APPROVAL DATE: July 26, 2001 FUNDING AGENCY: .National Institute's of Health (US) The Animal Care Committee has examined and approved the use of animals for the above experimental project or teaching course, and have been given an assurance that the animals involved will be cared for in accordance with the principles contained in Care of Experimental Animals - A Guide for Canada, published by the Canadian Council on Animal Care. Approval of the UBC Committee on Animal Care by one of: Dr, D W. Rurak, Chairman Dr. J. Love, Director, Animal Care Centre Ms. L. Macdonald, Manager, Animal Care Committee This certificate is valid for one year from Ihe above start or approval date (whichever is later) provided there is no change in the experimental procedures. Annual review is required by the CCAC and some granting agencies. A copy of thl* certificato must tw displayed in your animal facility. Office of Research ServicoG and Administration 323-2134 Health Sciences Mali, Vancouver, V6T 1Z3 Phone. 604-822-8155 FAX: 604-822-5093 The University of British Columbia ANIMAL C A R E CERTIFICATE PROTOCOL NUMBER: A02-0022 INVESTIGATOR OR COURSE DIRECTOR: Keiver, K.M. DEPARTMENT: Agricultural Sciences PROJECT OR COURSE TITLE: Effects of alcohol on calcium regulation and bone ANIMALS: Rats 1558 START DATE: 02-07-01 APPROVAL DATE; 04-09-17 FUNDING AGENCY: Canadian institutes of Health Research The Animal Care Committee has examined and approved the use of animals for the above experimental project or teaching course, and have been given an assurance that the animals involved will be cared for in accordance with the principles contained in Care of Experimental Animals - A Guide for Canada, published by the Canadian Council on Animal Care. f. v,-^.^, Approval of the UBC Committee on Animal Care by one of: Dr, W.K. Milsom, Chair Or. J. Love, Director, Animal Care Centre Ms, L. Macdonald, Manager, Animal Care Committee This certificate is valid for one year from the above start or approval date (whichever is later) provided there is no change in the experimental procedures. Annual review is required by the CCAC and some granting agencies. A copy of this certificate must be displayed in your animal facility. Office of Research Services and Administration 102, Agronomy Road, Vancouver. V9T 1Z3 Phone; 804-827-8111 FAX: 804-822-S003 "@en . "Thesis/Dissertation"@en . "10.14288/1.0093015"@en . "eng"@en . "Human Nutrition"@en . "Vancouver : University of British Columbia Library"@en . "University of British Columbia"@en . "For non-commercial purposes only, such as research, private study and education. Additional conditions apply, see Terms of Use https://open.library.ubc.ca/terms_of_use."@en . "Graduate"@en . "The effects of prenatal alcohol exposure on endochondral bone development in the fetal rat"@en . "Text"@en . "http://hdl.handle.net/2429/18569"@en .