Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Three chromosomes and a baby : cytogenetic, biological, and clinical aspects of the trisomic placenta Yong, Paul John 2006

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

Item Metadata

Download

Media
831-ubc_2006-199213.pdf [ 10.93MB ]
Metadata
JSON: 831-1.0092891.json
JSON-LD: 831-1.0092891-ld.json
RDF/XML (Pretty): 831-1.0092891-rdf.xml
RDF/JSON: 831-1.0092891-rdf.json
Turtle: 831-1.0092891-turtle.txt
N-Triples: 831-1.0092891-rdf-ntriples.txt
Original Record: 831-1.0092891-source.json
Full Text
831-1.0092891-fulltext.txt
Citation
831-1.0092891.ris

Full Text

THREE CHROMOSOMES A N D A BABY: CYTOGENETIC, BIOLOGICAL, A N D C L I N I C A L A S P E C T S OF T H E TRISOMIC P L A C E N T A  by  PAUL JOHN Y O N G B . S c , The University o f British Columbia, 1999  A THESIS S U B M I T T E D IN P A R T I A L F U L F I L L M E N T OF T H E R E Q U I R E M E N T S FOR T H E D E G R E E OF C O M B I N E D D O C T O R OF P H I L O S O P H Y A N D D O C T O R OF M E D I C I N E in  T H E F A C U L T Y OF G R A D U A T E STUDIES  (Experimental Medicine)  T H E U N I V E R S I T Y OF BRITISH C O L U M B I A A p r i l 2006  ©Paul John Yong, 2006  Abstract Trisomy - the presence of an extra chromosome - is found in one in four miscarriages, while 0.5-1% of ongoing pregnancies assessed by chorionic villus sampling have trisomy mosaicism with the abnormality predominantly or completely confined to the placenta ('confined placental mosaicism' (CPM)). Trisomy of chromosome 16, in particular, may be the most common chromosome abnormality at conception. In this thesis, ongoing trisomy C P M pregnancies, especially those involving trisomy 16, were first investigated to clarify the distribution of maternal-fetal and pediatric outcomes and to delineate the predictors of poorer outcome. C P M of trisomy 16 (CPM 16) was found to be associated with fetal growth restriction and malformation, as well as maternal preeclampsia, although long-term growth and development of the newborns was reassuring. The presence of amniotic fluid trisomy and uniparental disomy (UPD) increased risk of poorer outcome in CPM16 pregnancies. Also, the level of trisomy in the trophoblast, for both C P M 16 and other trisomy C P M , was the key placental lineage important for pregnancy outcome. Second, trisomic spontaneous abortions, especially trisomy 16 miscarriages, were studied to identify biological mechanisms in the pathogenesis of the trisomic placenta, such as trophoblast outgrowth and fibroblast protein kinase expression. Although trisomic trophoblast outgrowth was variable, and apparently normal in trisomy 15, there was a defect in outgrowth in trisomy 16 trophoblast. As well, protein kinase profiling of trisomy 16 fibroblasts showed both dosage-effects and amplified instability when compared to euploid fibroblasts. In conclusion, the trisomic placenta can have significant and varied effects on biological function and the clinical outcome of mother, fetus, and newborn.  Table of Contents Abstract  ii  Table of Contents  iii  List of Tables  vii  List of Figures  ix  List of Abbreviations  xi  Acknowledgements  xiii  Dedication  xiv  Chapter 1 1.1 1.2 1.3 1.4  1.5 1.6 Chapter 2 2.1 2.2 2.3  2.4  Introduction  1  Structure, embryology, and function of the human placenta Cytogenetic terms and definitions Confined placental mosaicism (CPM) Epidemiology of trisomy and C P M during pregnancy 1.4.1 Trisomy in miscarriage 1.4.2 C P M in ongoing pregnancies Research obj ectives References  1 2 4 6 6 8 9 10  Feto-placental growth in trisomy C P M  13  Note Introduction Methods 2.3.1 Trisomy C P M cases 2.3.2 Matched controls 2.3.3 Data analysis 2.2.3.1 Birth weight 2.2.3.2 Placental weight 2.2.3.3 Feto-placental (F-P) weight ratio 2.2.3.4 Determinants of placental weight and birth weight 2.2.4 Statistical analysis Results 2.4.1 Clinical and cytogenetic data 2.4.2 Placental pathology 2.4.3 Birth weight 2.4.4 Placental weight 2.4.5 F-P weight ratio 2.4.6 Determinants of placental weight and birth weight 2.4.6.1 Sex of the fetus  13 13 15 15 17 17 17 18 18 18 19 19 19 20 20 21 21 22 22 iii  2.5 2.6 Chapter 3 3.1 3.2 3.3  3.4  3.5  3.6 Chapter 4 4.1 4.2 4.3  4.4 4.5 4.6  2.4.6.2 Involved chromosome 2.4.6.3 Level of trisomy in trophoblast, mesenchyme, chorion 2.4.6.4 Placental weight Discussion References  22 22 25 26 29  Pathogenesis of C P M 16 pregnancies  40  Note Introduction Methods 3.3.1 CPM16 cases 3.3.2 Statistical analysis Results 3.4.1 Pregnancy outcomes of C P M 16 3.4.2 Clinical outcomes of C P M 16 live births 3.4.3 Trisomy in amniotic fluid in C P M 16 live births 3.4.4 Ascertainment of CPM16 live births 3.4.5 Sex of the fetus in C P M 16 live births 3.4.6 C V S in C P M 16 live births 3.4.7 upd(16)mat in C P M 16 live births 3.4.8 Possible confounding 3.4.9 Gestational age at delivery in CPM16 live births 3.4.10 Intrauterine death and neonatal death in CPM16 Discussion 3.5.1 Clinical outcome of C P M 16 3.5.2 Amniotic fluid trisomy and CPM16 live births 3.5.3 Ascertainment and C P M 16 live births 3.5.4 Sex of the fetus and C P M 16 live births 3.5.5 C P M 1 6 a n d C V S 3.5.6 Intrauterine and neonatal death in CPM16 3.5.7 Evidence for imprinting on chromosome 16 References  40 40 42 42 44 44 44 44 45 46 47 47 48 48 50 50 51 51 51 52 52 53 54 54 57  Preeclampsia and C P M 16  80  Note Introduction Methods 4.3.1 CPM16 cases 4.3.2 Matched controls 4.3.3 Definition of preeclampsia 4.3.4 Statistical analysis Results Discussion References  80 80 82 82 83 83 83 84 85 87  iv  Chapter 5 5.1 5.2 5.3  5.4  5.5 5.6 Chapter 6 6.1 6.2 6.3  6.4 6.5 6.6 Chapter 7 7.1 7.2 7.3  7.4 7.5 7.6 Chapter 8 8.1 8.2 8.3  Postnatal follow-up of newborns from C P M 16 pregnancies  93  Note Introduction Methods 5.3.1 CPM16 cases 5.3.2 Statistical analysis Results 5.4.1 Height 5.4.2 Weight 5.4.3 Development Discussion References  93 93 93 93 95 95 95 96 96 97 100  Cytokeratin staining in villus cultures from miscarriage and C V S  110  Note Introduction Methods 6.3.1 Tissue processing and culture 6.3.2 Immunochemistry 6.3.3 Statistical analysis Results Discussion References  110 110 Ill Ill 112 113 113 113 115  E V T outgrowth in trisomic miscarriage  119  Note Introduction Methods 7.3.1 Miscarriage cases 7.3.2 Tissue processing and culture 7.3.3 Irnmunochemistry 7.3.4 Karyotype confirmation 7.3.5 Statistical analysis Results Discussion References  119 119 121 121 122 122 123 124 124 126 129  Protein kinase profiling in trisomic miscarriage  138  Note Introduction Methods 8.3.1 Miscarriage cases 8.3.2 Tissue processing and culture 8.3.3 Immunochemistry and PCR 8.3.4 Protein kinase protein profiling  138 138 140 140 140 141 142 v  8.3.5 8.3.6  8.3.7 Results 8.4.1 Protein kinase expression at the protein level 8.4.2 Protein kinase expression at the R N A level 8.4.3 Inter-individual variation in expression Discussion References  143 145 145 146 146 147 147 147 148 148 149 152  Conclusion  161  References  164  Appendix A  Ethics approval  167  Appendix B  Extra references  168  Appendix C  Protein kinases profiled in Chapter 8  179  8.4  8.5 8.6 Chapter 9 9.1  Protein kinase R N A profiling Profiling data analysis 8.3.6.1 Confounding 8.3.6.2 Protein kinase expression at protein/RNA levels 8.3.6.3 Inter-individual variation in expression Statistical analysis  vi  List of Tables Table 2.1 Clinical and cytogenetic data for the 69 trisomy C P M cases  31  Table 2.2 Clinical and cytogenetic data for the 69 trisomy C P M cases  34  Table 2.3 Findings for (local) cases where placental pathology was performed  36  Table 2.4 Clinical characteristics of the trisomy C P M cases and the matched controls  38  Table 3.1 Malformations among live births with survival beyond the neonatal period  60  Table 3.2 Association between the presence of trisomy in amniotic fluid and malformation  60  Table 3.3 Association between trisomy in amniotic fluid and trisomy in fetal tissues  60  Table 3.4 Association between upd(16)mat and malformation  60  Table 3.5 Associations between explanatory variables associated with birth weight..  61  Table 3.6 Association between malformation and neonatal death  62  Table 3.7 Association between biased ascertainment and neonatal death  62  Table 3.8 Association between ascertainment by abnormal ultrasound and neonatal death  62  Table 3.9 Association between ascertainment by abnormal SS, and neonatal death  63  Table 3.10 Ascertainment and cardio-pulmonary malformations in neonatal deaths  64  Table 4.1 CPM16 cases meeting inclusion criteria.  89  Table 4.2 CPM16 cases meeting inclusion criteria  90  Table 4.3 Post-partum placenta cytogenetics  91  Table 4.4 Clinical features of the CPM16 cases and controls  92  Table 4.5 Clinical features of the C P M 16 cases with preeclampsia  92  Table 5.1 CPM16 cases meeting inclusion criteria  102  Table 5.2 Clinical data for included C P M 16 cases  103  Table 5.3 Follow-up data for length/height, weight and developmental outcome  104  Table 5.4 Association between trisomy in amniotic fluid and developmental delay  108  Table 5.5 Association between malformation and developmental delay  108  Table 6.1 CK7 and CK18 staining  116  Table 7.1 Association between gestational age and the presence of E V T outgrowths  131  Table 7.2 E V T outgrowth in euploid and trisomy 15 cases <10 weeks gestation  131  Table 7.3 E V T outgrowth in euploid and trisomy 16 cases <10 weeks gestation  131  Table 7.4 E V T outgrowth in euploid and all abnormal cases <10 weeks gestation  131  Table 8.1 Immunochemistry, and protein/RNA profiling of mesenchymal core cultures  .155  Table 8.2 Protein levels of kinases with significant differences in protein expression  156  Table 8.3 R N A levels of kinases with significant differences in protein expression  157  Table 8.4 Coefficients of variation (CV) at the R N A and protein levels  158  List of Figures Figure 2.1 Birth weight means for the chromosomes involved in the trisomy C P M cases  39  Figure 3.1 Distribution of gestational ages for C P M 16 live births  65  Figure 3.2 Distribution of birth weights for CPM16 resulting in live births  66  Figure 3.3 Distribution of the level of trisomy in amniotic fluid among CPM16 live births  67  Figure 3.4 Mean birth weight in the presence or absence of trisomy in amniotic fluid  68  Figure 3.5 Mean birth weight for unbiased and biased ascertainment  69  Figure 3.6 Mean birth weight for ascertainment: unbiased vs. abnormal serum screen...  70  Figure 3.7 Mean birth weight for ascertainment: unbiased vs. abnormal ultrasound  71  Figure 3.8 Mean birth weight for females and males  72  Figure 3.9 Mean birth weight for <100% and 100% trisomy on direct CVS  73  Figure 3.10 Mean birth weight for <100% and 100% trisomy on cultured C V S  74  Figure 3.11 Mean birth weight for bpd(16) and upd(16)mat  75  Figure 3.12 Mean gestational age for unbiased and biased ascertainment  76  Figure 3.13 Mean gestational age ascertainment: unbiased vs. abnormal ultrasound  77  Figure 3.14 Birth weights for live births and intrauterine deaths  78  Figure 3.15 Gestational ages for live births and neonatal deaths  79  Figure 5.1 Association between birth weight and developmental delay  109  Figure 6.1 C K 7 and CK18 staining in CVS cultures  117  Figure 6.2 C K 7 staining in JEG-3 cells (positive control)  118  Figure 7.1 Extravillus trophoblast (EVT) columns deriving from the chorionic villus  132  Figure 7.2 E V T outgrowths  133  Figure 7.3 Fibroblast-like outgrowths  133  Figure 7.4 Cytokeratin-7 staining of E V T outgrowths  134 ix  Figure 7.5 Association between gestational age and E V T outgrowth  135  Figure 7.6 Proportion of explants with E V T outgrowths for euploid and trisomy 15 cases  136  Figure 7.7 Proportion of explants with E V T outgrowths for euploid and abnormal cases  137  Figure 8.1 ERK1 R N A and protein expression  159  Figure 8.2 Coefficient of variation (CV) of R N A and protein expression  160  x  List of Abbreviations AMA  Advanced maternal age  ASD  Atrial septal defect  BPD  Biparental disomy  bpd(16)  Biparental disomy of chromosome 16  BCWH  British Columbia Women's Hospital  BMI  Body mass index  BW  Birth weight  C&W  Children's and Women's Health Centre of British Columbia  CV  Coefficient of variation  DORV  Double outlet right ventricle  Eu  Euploid  EVT  Extravillus trophoblast  ICM  Inner cell mass  CGH  Comparative genomic hybridization  CK7  Cytokeratin 7  CK18  Cytokeratin 18  CPM  Confined placental mosaicism  C P M 16  Confined placental mosaicism involving trisomy 16  CVS  Chorionic villus sampling  GA  Gestational age  F-P  Feto-placental  FISH  Fluorescence in situ hybridization  HELLP  Hemolysis, elevated liver enzymes, and low platelets  IUD  Intrauterine death  IUGR  Intrauterine growth restriction  LB  L i v e birth with survival beyond the neonatal period  MSAFP  Maternal serum alpha-fetoprotein  MShCG  Maternal serum human chorionic gonadotropin  ND  Neonatal death  PDA  Patent ductus arteriosus  PCR  Polymerase chain reaction  PROM  Premature rupture of membranes  RT-PCR  Reverse transcription-polymerase chain reaction  SDS-PAGE  Sodium dodecyl sulfate-polyacrylamide gel electrophoresis  SGA  Small-for-gestational age  SUA  Single umbilical artery  T16  Trisomy 16  T15  Trisomy 15  TA  Termination of pregnancy  UPD  Uniparental disomy  upd(16)mat  Maternal uniparental disomy of chromosome 16  VSD  Ventricular septal defect  xii  Acknowledgements This thesis would not have been possible without the patience, understanding, assistance, and mentorship of the following people: Dr. Wendy Robinson Members of the Robinson laboratory Dr. Deborah McFadden Dr. Sylvie Langlois Dr. Michael Whitlock Irene Barrett Dr. C o l i n MacCalman Dr. Peter von Dadelszen Dr. Anthony Chow Dr. L y n n Raymond Jane Lee Dr. Norman W o n g Dr. Vince Duronio Patrick Carew  Dedication This thesis is dedicated to the following people for their inspiration and support M y parents (Peter and Rosalinda) and siblings (Rachelle, Brian, and Christina) Dr. John Yun Drs. Margaret and Robin Cottle, and the Christian Medical Dental Society (CMDS) Fr. Gregory Smith And all my friends - You have blessed my life.  xiv  1 Introduction Trisomy - the presence of an extra chromosome - has devastating effects on human pregnancy, leading to miscarriage, stillbirth, and children with syndromes of considerable morbidity and mortality. The classic example of the latter is trisomy of chromosome 21, which is associated with Down Syndrome. Although the association between trisomy and Down syndrome has been known for over forty years, much remains to be known about the pathogenesis of trisomy in the human conceptus and its effects on the mother during pregnancy. The purpose of this thesis is to provide insight into how trisomy in the placenta affects maternalfetal and pediatric health from cytogenetic, clinical and biological perspectives using two human models: first-trimester miscarriages with trisomy, and ongoing pregnancies with trisomy confined to the placenta (confined placental mosaicism). This introduction has two goals. First, background information will be provided on the structure, embryology and function of the placenta; some cytogenetic terms and definitions; and confined placental mosaicism (CPM). Second, the epidemiology of trisomic miscarriage and C P M in ongoing pregnancies will be reviewed. Finally, the specific objectives of the thesis research will be described.  1.1 Structure, embryology, and function of the human placenta  The human conceptus is defined as the placenta, the gestational sac membranes, and the embryo/fetus-proper, derived from a single-celled zygote. The membranes consist of the amnion and the chorion laeve. The placenta consists of the chorionic plate, into which the umbilical cord inserts, and a multitude of chorionic villi, which interact with the surrounding maternal tissue of the uterine wall. The chorionic villi consist of an outer layer of trophoblast syncytiotrophoblast and cytotrophoblast - and an inner villus mesenchymal core. The trophoblast is derived from the trophectoderm of the blastocyst, while the villus mesenchymal  1  core, chorionic plate, amnion, and chorion laeve are derived from the inner cell mass (ICM) of the blastocyst. The embryo/fetus-proper also develops from the ICM. The placenta is a vital organ required for maintenance of normal pregnancy and the production of a healthy neonate, being involved in gas and heat transfer; pH and water haemostasis; metabolism and absorption; endocrine and exocrine synthesis and secretion; and in hematopoiesis and immunologic functions (Benirschke and Kaufmann 1995). Most of these functions involve in some way transfer across the trophoblast between maternal blood vessels and glands in the uterus and placental blood vessels in the villus mesenchymal core (Burton et al. 2001; Burton et al. 2002).  1.2 Cytogenetic terms and definitions  Trisomy refers to a karyotype with an extra autosomal chromosome (resulting in 47 chromosomes in total), as opposed to a normal euploid karyotype that has 2 sex chromosomes and 22 pairs of autosomal chromosomes. It can be classified by the timing of its origin relative to fertilization. The extra chromosome may be pre-zygotic or post-zygotic; that is, it may originate pre-fertilization during gametogenesis (oogenesis in the female and spermatogenesis in the male), or post-fertilization during embryonic development, respectively. Moreover, the extra chromosome can arise at different stages of gametogenesis or embryonic development. In gametogenesis, it can arise through an error in mitosis during any of the series of divisions from the primordial germ cell to the oogonium or spermatogonium, or through an error in meiosis I or meiosis II in the process of 'reduction division' from the (diploid) oogonium to the (haploid) secondary oocyte or from the (diploid) spermatogonium to the (haploid) secondary spermatocyte. In embryonic development, it can arise through an error in mitosis during any of the series of divisions from the initial single-celled embryo (i.e. the 'fertilized egg' or 'zygote') to the various stages of the multi-celled embryo such as the morula and blastocyst. If the third 2  copy arises during oogenesis or by duplication of the maternal copy during embryonic development, it is said to have a maternal origin; if the third copy arises during spermatogenesis or by duplication of the paternal copy during embryonic development, it is said to have a paternal origin. It is important to emphasize that cytogenetic terms such as 'trisomy' more precisely describe the karyotype of a particular cell of a multi-celled individual, although often applied to the multi-celled individual itself. This distinction is critical because of the phenomenon of mosaicism, the presence of at least two cell lines with different karyotypes within an individual (i.e. a feto-placental unit) derived from a single zygote. The etiology of mosaicism can be divided into two broad categories: 1) a euploid conceptus with post-zygotic origin of trisomy in a cell at some time during development, resulting in a mixture of trisomic and euploid cells; and 2) pre-zygotic origin of trisomy producing a trisomic conceptus, followed by loss of one of the three chromosomes in a cell at some time during development, also resulting in a mixture of trisomic and euploid cells (Kalousek and Vekemans 1996). The latter is also known as trisomy zygote rescue (Kalousek and Vekemans 1996). Cytogenetic techniques have inherent limitations that can make the detection of mosaicism challenging in practice. For example, traditional (conventional) cytogenetic analysis requires metaphase chromosomes, which are generated by blocking dividing cells at metaphase. For conventional cytogenetic analysis of the villus cytotrophoblast, spontaneous cell divisions (and therefore metaphases) are analyzed either directly or after a short-term (~1 day) incubation (Eiben et al. 1990). Such direct or short-term preparations have the advantage of quicker results, but require very fresh tissue and produce chromosomes of relatively poorer quality (Warburton 2000). For conventional cytogenetic analysis of other tissues (i.e. villus mesenchymal core, chorionic plate, chorion laeve, amnion, and embryonic/fetal tissues), dividing cells are generated by culturing the cells for at least one week. Such long-term culture 3  results in more high-quality metaphases, but has two major problems: 1) chromosomally abnormal cells may arise due to cell division in in vitro conditions ('culture artefact'); and 2) maternal cells from the uterus (in which the conceptus develops) may grow instead of or alongside placental or embryo/fetal cells ('maternal contamination'). Also, since both direct/short-term preparations and long-term culture techniques require cell divisions, conventional cytogenetics may select against chromosomally abnormal cells; that is, although a low level of mosaicism may be present in vivo, the abnormal cells may be unable to grow in vitro and thus may not be detected with conventional cytogenetic techniques. Furthermore, only a practical number of metaphases (e.g. 10-20) are examined for routine cytogenetic analysis; thus, a low level of mosaicism may not be detected simply by chance. Any clinically relevant definition of mosaicism has to take into account the disadvantages of cell culture, while accepting that the probability of detecting mosaicism in a biological sample is dependent on the number of metaphases examined. One definition is the presence of at least two cell lines, with each present in at least two culture flasks set up from a particular tissue sample (Hsu et al. 1992). When only one cell of a given karyotype is found or when more than one cell is present but only in one flask, then the terms single-cell and multiplecell pseudomosaicism, respectively, can be used. A finding of 100% trisomic or euploid cells could be referred to as 'full' or 'non-mosaic' trisomy or euploidy, respectively. In order to circumvent the problems of conventional cytogenetics, non-culture dependent methods of cytogenetic analysis can also be used, including fluorescence in situ hybridization (FISH), comparative genomic hybridization (CGH), and molecular methods such as PCR (Tonnies 2002).  1.3 Confined placental mosaicism (CPM) One special case of mosaicism is confined placental mosaicism (CPM), which 4  describes the phenomenon of chromosomally abnormal cells confined to the placenta. It was first described by Dagmar Kalousek and Fred Dill at the University of British Columbia in 1983 (Kalousek and Dill 1983). C P M pregnancies provide a model to determine, independent of the fetus, the effects of a chromosome abnormality in the placenta on pregnancy outcome. In viable ongoing pregnancies, C P M is usually ascertained by first-trimester chorionic villus sampling (CVS) (10-12 weeks gestation). The sampled chorionic villi are assessed by conventional cytogenetics following a direct/short-term preparation (reflecting the karyotype of villus cytotrophoblast) or long-term culture (reflecting the karyotype of the villus mesenchymal core). The fetal karyotype is then assessed by second-trimester amniocentesis, which samples amniotic fluid cells originate from both the fetus proper (lung, urinary tact, skin) and from the amnion (Hsu 2004). Cytogenetic analysis of the placenta and fetus can also be carried out at the end of pregnancy, whether after miscarriage, termination, or live birth. For live births, investigations of the newborn are limited to only a few cell types, usually peripheral blood lymphocytes or skin fibroblasts. If trisomic cells are found in tissues of the embryo/fetus-proper, then strictly speaking, the trisomic cells are not confined to the placenta. However, some authors have considered cases with a low-level of trisomic cells in the embryo/fetus to still be C P M (Roland et al. 1994; Robinson et al. 1997) because it is thought that trisomic cells had to be at least predominantly confined to the placenta for the conceptus to survive to term. Hence, C P M could be subdivided into those cases in which trisomic cells are completely or only predominantly confined to the placenta. With this definition, C P M could be initially ascertained by low-level trisomy at amniocentesis, followed by investigation of the placenta antenatally or post-partum. In contrast, some authors consider cases with even only low-level trisomy in embryonic/fetal tissues to be cases of generalized or true fetal mosaicism (Kalousek 1994). I use an encompassing definition of C P M that includes both completely and predominantly confined trisomy. However, because  of technical limitations, a case that appears to be completely confined may actually have lowlevel fetal trisomy that went undetected during prenatal or postnatal investigations. Nevertheless, when attempting to estimate the effect of trisomy in the placenta independent of the fetus, it is still more useful to consider apparently 'complete' CPM while accepting that an unspecified proportion in reality are probably only predominantly confined. In another classification, CPM is categorized by tissue distribution (Kalousek 1994): Type I (confined to villus trophoblast); Type II (confined to villus mesenchymal core); Type III (present in both lineages). The effect of CPM on pregnancy outcome depends on several variables, including the chromosome involved in the trisomy, and the timing of chromosome gain or loss (Kalousek and Vekemans 1996). For instance, it would be expected that the larger the chromosome and/or the more genes important for viability are present on the chromosome, the more adverse effects on pregnancy outcome. As well, the earlier the post-zygotic origin of trisomy or the later the trisomy zygote rescue, then the higher level and more extensive distribution of trisomic cells would be expected in the feto-placental unit, which presumably correlates with poorer pregnancy outcome. On the other hand, the level and distribution of trisomy may be also influenced by natural selection against trisomic cells and stochastic events during development, while some tissues may be more 'sensitive' to trisomy than others. Finally, uniparental disomy (where both copies of a chromosome are inherited from the same parent) may be present in the fetus, particularly when the trisomy has a pre-zygotic origin (see Chapter 3).  1.4 Epidemiology of trisomy and C P M during pregnancy 1.4.1  Trisomy in miscarriage  Some pregnancies are 'lost' at the peri-implantation stage before they are detected clinically; that is, they are pre-clinical miscarriages. Although several studies have attempted to 6  quantify the proportion of conceptions that end in these pre-clinical losses, the landmark study was that of Wilcox et al. (1988) (Macklon et al. 2002). Wilcox et al. (1988) used a highly sensitive assay to measure maternal urine human chorionic gonadotropin (hCG), a marker of implantation, in a sample of women attempting to conceive. They also controlled for the detectable background urine hCG level by including a sample of control women. Their results suggest that about half of conceptions are lost, with 22% of conceptions ending as pre-clinical miscarriages and 31% ending as clinical miscarriages. It should be noted that it is likely that many fertilizations are lost even before conception (i.e. implantation), although a good estimate of this number is not available (Macklon et al. 2002). Unfortunately, cytogenetic data are not available for both pre-implantation losses and pre-clinical spontaneous abortions (Banzai et al. 2004). Cytogenetic data are available for those clinically recognized pregnancies that end in spontaneous abortion. Approximately 50% of clinical miscarriages are associated with a chromosome abnormality, of which half are trisomies (Warburton 2000). However, the exact numbers greatly depend on both maternal age and gestational age (Hassold and Hunt 2001; Warburton 1991). At British Columbia Women's Hospital (BCWH), 61% of clinical miscarriages are chromosomally abnormal, with the majority (44% of all miscarriages) associated with trisomy (D.E. McFadden and W.P. Robinson, unpublished data). The distribution of different trisomies among miscarriages is non-random; for example, the most common is trisomy of chromosome 16 (-10% of miscarriages at BCWH), while trisomy 1 miscarriage has only been reported three times in the literature (Hanna et al. 1997; Dunn et al. 2001; Banzai etal. 2004).  7  1.4.2  CPM in ongoing pregnancies  The largest study of C P M to date, the European Collaborative Research on Mosaicism i n C V S ( E U C R O M I C ) (Hahnemann and Vejerslev 1997a), involved 64,053 C V S procedures with 98.1% (62,865) having conventional cytogenetics successfully performed on direct/short-term preparations (cytotrophoblast) and/or long-term cultures (mesenchymal core). 1.04% of all C V S cases involved a mixture of chromosomally abnormal and euploid cells in at least one placental lineage (trophoblast or mesenchyme), or between the lineages. It should be noted that not all of these cases had investigation of fetal karoytype; and in an unspecified proportion, a normal neonatal phenotype at birth was taken to be evidence of euploid karoytype. In an additional 0.15% of all C V S cases, the placental lineage(s) investigated were fully abnormal, while the fetus was found to be euploid. (The authors coined the term 'non-mosaic fetoplacental discrepancy' to describe these latter cases, while I consider them to be C P M . ) Combining both figures gives a C P M incidence of 1.19% at C V S from E U C R O M I C . A second publication from E U C R O M I C provided more cytogenetic details on the C P M cases involving autosomal trisomy, which had an incidence of 0.69% at C V S (Hahnemann and Vejerslev 1997b). A m o n g a subset of these C P M cases that had both direct and cultured preparations, as well as strict exclusion of trisomy from one or more fetal tissues (n = 172), 54% were Type I, 28% Type II, and 18% Type III. Therefore, trisomy in the trophoblast lineage (Type I and Type III) was more frequent than trisomy in the mesenchymal lineage (Type II and Type III). The distribution of different trisomies among ongoing C P M pregnancies was also non-random, and differed from the distribution among miscarriages. Notably, trisomy 7 was the most frequent C P M trisomy (19% of all trisomies), while trisomy 16 was fifth most common (6%). The clinical implications of prenatally diagnosed C P M are controversial. Some studies found an increased risk of subsequent pregnancy loss (Hogge et al. 1986; Johnson et al. 1990; Wapner et al. 1992; Wang et al. 1993), although most have not (Breed et al. 1991; Fryburg et al. 8  1993; Roland et al. 1994; Wolstenholme et al. 1994; Leschot et al. 1996). Similarly, with the exception of one unpublished study (DeLozier-Blanchet 1996), studies have not shown a higher rate of IUGR or small-for-gestational (SGA) newborns (Johnson et al. 1990; Breed et al. 1991; Wapner et al. 1992; Fryburg et al. 1993; Roland et al. 1994; Wolstenholme et al. 1994; Leschot et al. 1996; Goldberg and Wohlferd 1997). A major issue is heterogeneity in methodology, including the definition of C P M (e.g. some studies considered a eumorphic infant as sufficient for C P M , without any fetal cytogenetic analysis) and the placental lineage analyzed (trophoblast and/or mesenchyme), as well as the inclusion of different chromosome abnormalities (e.g. autosomal trisomy versus structural aberrations and sex chromosome aneuploidy). C P M involving trisomy would be more likely to be associated with abnormal pregnancy outcome, given the severe phenotype of most trisomic pregnancies, from early spontaneous abortion to Down syndrome. In particular, case series of trisomy 16 C P M (CPM 16) suggest that such pregnancies are particularly high-risk (Kalousek et al. 1993; Robinson et al. 1997).  1.5 Research objectives The purpose of this thesis is to provide insight into how trisomy in the placenta affects maternal-fetal and pediatric health, through analysis of cytogenetic, biological, and clinical aspects of ongoing trisomy C P M pregnancies and trisomic miscarriages. The objectives are: 1) to use ongoing C P M pregnancies, in particular those involving trisomy 16, to elucidate the role of the chromosome involved in the trisomy, the level and distribution of trisomic cells (in placental lineages and in amniotic fluid), and uniparental disomy (UPD) on maternal-fetal and pediatric outcomes; and 2) to use pregnancies ending in miscarriage, in particular those associated with trisomy 16, to investigate two biological mechanisms in the trisomic placenta: trophoblast outgrowth and mesenchymal core fibroblast protein kinase expression.  9  1.6 References Banzai M , Sato S, Matsuda H, Kanasugi H (2004) Trisomy 1 in a case of a missed abortion. J Hum Genet 49:396-397 Breed AS, Mantingh A , Vosters R, Beekhuis JR, Van Lith J M , Anders GJ (1991) Follow-up and pregnancy outcome after a diagnosis of mosaicism in C V S . Prenat Diagn 11:577-580 Burton GJ, Hempstock J, Jauniaux E (2001) Nutrition of the human fetus during the first trimester—a review. Placenta 22 Suppl A:S70-77 Burton GJ, Watson A L , Hempstock J, Skepper JN, Jauniaux E (2002) Uterine glands provide histiotrophic nutrition for the human fetus during the first trimester of pregnancy. J Clin Endocrinol Metab 87:2954-2959 DeLozier-Blanchet CD, Pellegrini, B., Hahnemann, J.M., Pampallona, S., Vejerslev, L.O. (1996) Birth weight analysis after mosaic/discrepant results on chorionic villus sampling: the E U C R O M I C experience. A m J Hum Genet: A319 Eiben B, Bartels I, Bahr-Porsch S, Borgmann S, Gatz G, Gellert G, Goebel R, Hammans W, Hentemann M , Osmers R, et al. (1990) Cytogenetic analysis of 750 spontaneous abortions with the direct-preparation method of chorionic villi and its implications for studying genetic causes of pregnancy wastage. A m J Hum Genet 47:656-663 Fryburg JS, Dimaio M S , Yang-Feng TL, Mahoney M J (1993) Follow-up of pregnancies complicated by placental mosaicism diagnosed by chorionic villus sampling. Prenat Diagn 13:481-494 Goldberg JD, Wohlferd M M (1997) Incidence and outcome of chromosomal mosaicism found at the time of chorionic villus sampling. A m J Obstet Gynecol 176:1349-1352; discussion 1352-1343 Hahnemann J M , Vejerslev L O (1997a) Accuracy of cytogenetic findings on chorionic villus sampling (CVS)—diagnostic consequences of CVS mosaicism and non-mosaic discrepancy in centres contributing to E U C R O M I C 1986-1992. Prenat Diagn 17:801820 Hahnemann J M , Vejerslev L O (1997b) European collaborative research on mosaicism in CVS (EUCROMIC)—fetal and extrafetal cell lineages in 192 gestations with C V S mosaicism involving single autosomal trisomy. A m J Med Genet 70:179-187 Hassold T, Hunt P (2001) To err (meiotically) is human: the genesis of human aneuploidy. Nat Rev Genet 2:280-291 Hogge W A , Schonberg SA, Golbus M S (1986) Chorionic villus sampling: experience of the first 1000 cases. A m J Obstet Gynecol 154:1249-1252 Hsu L Y (2004) Amniocentesis. In: Milunsky A (ed) Genetic disorders and the fetus: diagnosis, prevention, and treatment. Johns Hopkins University Press, Baltimore, pp 112-115 10  Hsu L Y , Kaffe S, Jenkins EC, Alonso L, Benn PA, David K, Hirschhorn K , Lieber E, Shanske A , Shapiro LR, et al. (1992) Proposed guidelines for diagnosis of chromosome mosaicism in amniocytes based on data derived from chromosome mosaicism and pseudomosaicism studies. PrenatDiagn 12:555-573 Johnson A , Wapner RJ, Davis G H , Jackson L G (1990) Mosaicism in chorionic villus sampling: an association with poor perinatal outcome. Obstet Gynecol 75:573-577 Kalousek D K (1994) Current topic: confined placental mosaicism and intrauterine fetal development. Placenta 15:219-230 Kalousek D K , Dill FJ (1983) Chromosomal mosaicism confined to the placenta in human conceptions. Science 221:665-667 Kalousek D K , Langlois S, Barrett I, Yam I, Wilson DR, Howard-Peebles PN, Johnson M P , Giorgiutti E (1993) Uniparental disomy for chromosome 16 in humans. A m J Hum Genet 52:8-16 Kalousek D K , Vekemans M (1996) Confined placental mosaicism. J Med Genet 33:529-533 Leschot NJ, Schuring-Blom G H , Van Prooijen-Knegt A C , Verjaal M , Hansson K, Wolf H, Kanhai H H , Van Vugt J M , Christiaens G C (1996) The outcome of pregnancies with confined placental chromosome mosaicism in cytotrophoblast cells. Prenat Diagn 16:705-712 Macklon NS, Geraedts JP, Fauser B C (2002) Conception to ongoing pregnancy: the 'black box' of early pregnancy loss. Hum Reprod Update 8:333-343 Regan L, Rai R (2000) Epidemiology and the medical causes of miscarriage. Baillieres Best Pract Res Clin Obstet Gynaecol 14:839-854 Robinson WP, Barrett IJ, Bernard L, Telenius A , Bernasconi F, Wilson RD, Best R G , HowardPeebles PN, Langlois S, Kalousek D K (1997) Meiotic origin of trisomy in confined placental mosaicism is correlated with presence of fetal uniparental disomy, high levels of trisomy in trophoblast, and increased risk of fetal intrauterine growth restriction. A m J Hum Genet 60:917-927 Roland B, Lynch L, Berkowitz G, Zinberg R (1994) Confined placental mosaicism in C V S and pregnancy outcome. Prenat Diagn 14:589-593 Tonnies H (2002) Modern molecular cytogenetic techniques in genetic diagnostics. Trends M o l Med 8:246-250 Wang B B , Rubin C H , Williams J, 3rd (1993) Mosaicism in chorionic villus sampling: an analysis of incidence and chromosomes involved in 2612 consecutive cases. Prenat Diagn 13:179-190  11  Wapner RJ, Simpson JL, Golbus M S , Zachary J M , Ledbetter D H , Desnick RJ, Fowler SE, Jackson L G , Lubs H , Mahony RJ, et al. (1992) Chorionic mosaicism: association with fetal loss but not with adverse perinatal outcome. Prenat Diagn 12:347-355 Warburton D (2000) Cytogenetics of reproductive wastage: from conception to birth. In: Mark H F L (ed) Medical Cytogenetics. Dekker, New York, pp 213-246 Warburton D BJ, Canki N (1991) Chromosome Anomalies and Prenatal Development: A n Atlas. Vol 21. Oxford University Press, New York Wilcox A J , Weinberg CR, O'Connor JF, Baird DD, Schlatterer JP, Canfield RE, Armstrong E G , Nisula B C (1988) Incidence of early loss of pregnancy. N Engl J Med 319:189-194 Wolstenholme J, Rooney DE, Davison E V (1994) Confined placental mosaicism, IUGR, and adverse pregnancy outcome: a controlled retrospective U.K. collaborative survey. Prenat Diagn 14:345-361  12  2 Feto-placental growth in trisomy CPM 2.1  1  Note  I wrote this chapter/manuscript and did the data organization and analysis, with the following clarifications and exceptions. The trisomy C P M cases are from the ongoing U B C study of trisomy mosaicism beginning in 1988. The original primary investigators were Dr. D . Kalousek (Pathology and Laboratory Medicine) and Dr. S. Langlois (Medical Genetics). Clinical data collection (e.g. birth weight, gestational age) and post-partum cytogenetic analysis of the placenta and amnion were done in the Kalousek laboratory. During this period, I. Barrett was the person primarily responsible in the Kalousek lab for the cytogenetic analysis and weighing of the placenta. M y supervisor, Dr. W . Robinson (Medical Genetics), is the current primary investigator, and data from more recent cases have been collected and organized by her laboratory. Matched controls were ascertained by myself from delivery records at B C Women's Hospital under the supervision of Dr. P. von Dadelszen (Obstetrics and Gynaecology). Placental pathology was collected by myself and reviewed with Dr. D . McFadden (Pathology and Laboratory Medicine).  2 . 2 Introduction Studies of outcomes such as fetal loss and R J G R i n prenatally diagnosed C P M have been equivocal (Chapter 1). A confounding factor in these studies is the type of chromosome abnormality; for example, in general, autosomal trisomy is likely to be more high-risk than a structural or sex chromosome abnormality. A l s o , although C V S can serve as an indicator of trisomy in the first-trimester placenta, another strategy is to re-sample placenta post-partum (Kalousek et al. 1991). The advantage of the latter is that multiple sites can be assessed to  A version of this chapter will be submitted for publication. Yong PJ, von Dadelszen P, Barrett IJ, McFadden DE, Kalousek DK, Robinson WP. Effect of the trisomic placenta on feto-placental growth. 1  13  provide a more accurate estimate of the level of trisomy, in cases where the distribution of placental trisomy is heterogeneous. Several studies have studied the degree of trisomy in the post-partum placenta from trisomy C P M pregnancies (Kalousek et al. 1996; Shaffer et al. 1996; Robinson et al. 1997). In a series of 14 cases of C P M involving trisomy 7, Kalousek et al. (1996) noted that the two small-for-gestational age (SGA) cases had higher levels of trisomy in the trophoblast compared to the other cases. In a series of 9 cases of C P M involving trisomy 2, Shaffer et al. (1996) detected high levels of trisomy in the mesenchyme of the post-partum placenta in the two SGA cases compared to the other cases. And in a larger study with statistical analysis, Robinson et al. (1997) found that the level of trisomy in the trophoblast was significantly associated with 'poor outcome' (at least one of SGA, malformation, or intrauterine death) in a series of 44 trisomy C P M cases. Although these studies suggest that lower birth weight is associated with higher levels of trisomy in the placenta, none of them have directly statistically correlated birth weight with the level of trisomy. Furthermore, none of them have examined placental weight. Placental weight is correlated with birth weight (Molteni et al. 1978; Williams et al. 1997; Sanin et al. 2001), although the direction of causation is not clear and is likely bidirectional. Although placental weight has not been investigated in trisomy C P M pregnancies, recent studies strongly support both lower birth weight (Stoll et al. 1998; Myrelid et al. 2002; Frid et al. 2004) and lower placental weight (Stoll et al. 1998; Myrelid et al. 2002) in Down syndrome births. I hypothesized that both placental and birth weight would be decreased in trisomy C P M , and that both weight measures would inversely correlate with the level of trisomy in the placenta. In this study, placental weight and birth weight were investigated in prenatally diagnosed trisomy C P M cases for 1) comparison to matched controls and a reference population; and 2) a statistical analysis of the relationship between placental and birth weight, 14  and the level of trisomy in various lineages of the placenta as well as the sex of the fetus and the involved chromosome.  2.3 Methods 2.3.1  Trisomy C P M cases  The study sample consisted of 69 cases of trisomy C P M singleton pregnancies resulting in a live birth from the ongoing study of trisomy mosaicism at the University of British Columbia (UBC). The study was approved by the ethics committees of U B C and the Children's and Women's Health Centre of British Columbia (C&W) (Appendix A). Twenty-five cases (36%) are local cases from C & W ; the other 44 cases (63%) were referred to the ongoing study from other centres. The cases were collected during a 7-year time period (1988-1994) when C P M placentas were weighed uniformly (see below) and the proportion of trisomic cells in the placenta was determined post-partum by conventional cytogenetics or FISH. Other inclusion criteria included (1) prenatal diagnosis via the detection of trisomic cells on CVS (by conventional cytogenetics); and (2) no trisomy in amniotic fluid cells, infant/fetal blood lymphocytes, and/or the amnion (by conventional cytogenetics, with a few exceptions by FISH or molecular (PCR) methods). The latter criterion was to enrich for cases where the trisomy is more likely to be completely confined to the placenta. Cases with concomitant polyploidy, sex chromosome aneuploidy, or structural chromosome abnormalities were excluded. In 63 of the 69 cases, the malformation status of the child was known: excluding digit and facial dysmorphism (as such data were not consistently reported), 92% (58/63) of the cases did not have malformations . There was one case each of imperforate anus, hypospadias, hip dysplasia, 2  Among the cases classified as normal were one case of familial benign megalencephaly, one case of familial hip dysplasia, and one case of mild hydronephrosis that resolved antenatally. 2  15  hydronephrosis, and 'possible' ventricular septal defect ( V S D ) . In addition, there was one case of placental abruption (case 56). Gestational age and birth weight were collected from U B C Medical Genetics medical records for the local cases and sent from collaborators for referred cases. Placental weights were determined by the C & W Pathology laboratory for the local cases and by I. Barrett in a research laboratory for the referred cases, but in the same manner: excess superficial blood was washed off, the cord and membranes trimmed, and then the 'trimmed' placentas weighed on the same scale. The chorionic plate and villus mesenchyme of the post-partum placenta were cultured and analyzed by conventional cytogenetic analysis, in which 5-15 cells were examined from 1-3 sites. The trophoblast were isolated in a suspension after a variety of short-term enzymatic digests of chorionic v i l l i , followed by F I S H for the chromosome involved, typically on 5001000 nuclei from 1-3 sites (Lomax et al. 1994; Henderson et al. 1996). These digests involve collagenase or trypsin, and produce suspensions consisting of villus cytotrophoblast and syncytiotrophoblast ( D E M , personal communication). Since the cut-off determined by F I S H on control samples (<10%) was determined by different methods during the time period (>3 standard deviations below the mean, or the procedure from Lomax et al. (1994)) and not all cases with F I S H included control samples, all cases with <10% of nuclei with 3 signals were simply all coded as ' 0 % ' ; for cases with >10% trisomic nuclei, the actual percentage was recorded. For a given tissue in a particular placenta, the level of trisomy at all sampled sites were averaged to produce a mean level of trisomy for that tissue. Where available, data from placental pathology were also collected. Some cytogenetic and/or clinical data from 28 of the cases were previously published (Kalousek et al. 1991; Kalousek et al. 1993; Kalousek et al. 1996; Shaffer et al. 1996; Robinson 2  In 3 cases the trophoblast was assessed by direct/short-term culture followed by conventional cytogenetics  et al. 1997; Kuchinka et al. 2001; Penaherrera et al. 2000). A l l data from the other 42 cases are unpublished. The level of trisomy in the chorionic plate, villus mesenchyme, or trophoblast may differ between this and previous studies because (1) data from different methodologies have been used (e.g. only results from conventional cytogenetics were used for the chorionic plate and mesenchyme in this study, while F I S H was also considered in Robinson et al. (1997)); and (2) additional data have become available since the time of previous publications.  2.3.2  Matched controls  Matched controls from routine deliveries were used because adequate 'normal' controls from Pathology were difficult to identify, as they had their own abnormalities that indicated an examination by Pathology. For each trisomy C P M case (both local and referred), 2 controls matched for maternal age (± 5 years) and for parity (0, 1, or > 2) were selected from deliveries on the same date or the consecutive previous or next day by reviewing delivery records at B C Women's Hospital ( B C W H ) . For some trisomy C P M cases, no data for maternal age (n = 5) or parity (n = 20) were available; in these cases, the controls were chosen randomly with respect to maternal age or parity. Gestational age, placental weight, and birth weight were collected from B C W H medical records. The placentas were weighed as per routine practice in Labour and Delivery at B C W H : the placentas were weighed 'untrimmed', and superficial excess blood was not washed clear before weighing.  2.3.3  Data analysis 2.3.3.1 Birth weight  Birth weight was compared directly between the trisomy C P M cases and matched controls. Small-for-gestational age ( S G A ) infants (<10 centile) were identified by comparison th  to published Canadian birth weight standards (Kramer et al. 2001). 17  2.3.3.2 Placental weight  Because the trisomy C P M placentas and matched control placentas were handled differently, the placental weight data were compared to similar reference populations in the literature. For the matched controls, the reference population consisted of gestational-age corrected mean placental weights from 29,902 singleton pregnancies collected from 1984-1991 in Detroit, in which placentas were untrimmed and not washed clear of excess blood (Dombrowski et al. 1994). For the trisomy C P M cases, the reference population consisted of gestational age-corrected mean placental weights from 787 singleton pregnancies from 19931995 in Providence (Rhode Island), in which cord and membranes were trimmed and 'excessive blood from the crevices' was removed (Pinar et al. 1996).  2.3.3.3 Feto-placental (F-P) weight ratio  F-P weight ratio was also calculated for the trisomy C P M cases and matched controls by dividing birth weight by placental weight. For the matched controls, F-P weight ratio was compared to gestational age-corrected mean F-P weight ratios from the reference population of Dombrowski et al. (1994). For the trisomy C P M cases, the reference population was from Heinonen et al. (2001), who produced gestational-age corrected mean F-P weight ratios for 15,047 singleton pregnancies collected from 1990-1999 in Kuopio (Finland). In Heinonen et al. (2001) placentas were washed clear of excess blood, and although cord and membranes were not trimmed before weighing, they used a correction factor (trimmed weight = 0.854 x untrimmed weight) derived from >1500 placentas at their centre.  2.3.3.4 Determinants of placental weight and birth weight  The following factors potentially involved in the determination of placental weight and birth weight in the trisomy C P M cases were investigated: sex of the fetus, the chromosome 18  involved in the trisomy, and the level of trisomy in the placental lineages (trophoblast, villus mesenchyme, and chorion). For these analyses, in order to both control for gestational age and to maintain a continuous outcome variable, placental weight and birth weight were transformed to number of standard deviations from the mean (z-scores) using gestational-age corrected means and standard deviations for placental weight and birth weight measured from a single population of newborns in Denver from Molteni et al. (1974).  2.3.4  Statistical analysis  Statistics were determined using SPSS-10.0 and the Vassar WebSite for Statistical Computation (http://facultv.vassar.edu/~lowrv/VassarStats.html).  Welch's approximate t-test  was used when there was inequality of variances, and the Mann-Whitney test and Spearman's rank correlation were used when the assumption of normality was not met. Tests were 1-tailed due to a priori evidence and rational mechanisms, unless otherwise noted. Means are reported ± standard deviation.  2.4 Results 2.4.1  Clinical and cytogenetic data  Clinical and cytogenetic data for the 69 cases of trisomy C P M in this study are summarized in Table 2.1 and Table 2.2. The following trisomies were present at C V S : trisomy 16 (n = 13), trisomy 7 (n = 10), trisomy 2 (n = 9), trisomy 12 (n = 6), trisomy 8 (n = 3), trisomy 9 (n = 4), trisomy 10 (n = 4), trisomy 15 (n = 3), trisomy 13 (n = 2), trisomy 17 (n = 2), trisomy 18 (n = 2), trisomy 22 (n = 2), trisomy 4 (n = 1), trisomy 11 (n = 1), trisomy 20 (n = 1), trisomy 21 (n = 1), and multiple trisomy (n = 5). In all cases of trisomy 7 where testing for the origin of the chromosomes 7 in the child was performed (n = 5), the result was normal biparental inheritance. None of the trisomy 15 cases (n = 3) had testing of origin. Trisomy was detected in 19  the chorion in 63% (35/56) of informative cases; in the mesenchyme in 69% (42/61); and in trophoblast in 33% (13/40) (Table 2.2).  2.4.2  Placental pathology  O f the 25 local trisomy C P M cases, placental pathology had been performed for 23 cases (92%). O f these, 52% (12/23) had some sort of abnormality (Table 2.3). The frequency and range of abnormalities among these 23 local trisomy C P M cases were considered non-specific by a pediatric pathologist (D. McFadden). Placental pathology was not available for referred trisomy C P M cases or for matched controls.  2.4.3  Birth weight  Birth weight data for the trisomy C P M cases and matched controls are given in Table 2.4. Trisomy C P M birth weight was significantly lower compared to birth weight among matched controls (p = 0.001; Table 2.4). There was also a significantly higher rate of S G A infants i n the trisomy C P M group (23%; 16/69) compared to the expected 10% (Binomial test, z-approximation, p < 0.001) and compared to the matched controls (8%; 11/138) (Fisher Exact test, p = 0.003). In addition, maternal age was higher and sex ratio lower in the trisomy C P M group compared to the matched controls (Table 2.4). The differences in maternal age and sex ratio were expected given the well-established relationship between maternal age and trisomic pregnancy (Hassold and Hunt 2001) and the bias towards females in mosaic trisomies (see Chapter 3). Maternal age was not significantly associated with birth weight among the trisomy C P M cases or matched controls, and sex was not associated with birth weight in the trisomy C P M cases. In contrast, males among the matched controls had heavier birth weights (t = 2.44, df = 136, p = 0.008), as also noted i n surveys of the general population (Kramer et al. 2001). 20  Therefore, multiple linear regression was carried out with birth weight as the outcome variable, and trisomy C P M , maternal age and sex of the fetus as explanatory variables. Neither sex nor maternal age confounded the relationship between trisomy C P M and decreased birth weight (data not shown).  2.4.4  Placental weight  Placental weight data are shown in Table 2.2. A s expected, the matched controls had heavier placentas because they were weighed untrimmed and without removing excess blood (Table 2.4). Thus the placental weights were compared to similar reference populations. For matched controls, 57 cases had placental weights below the means from Dombrowski et al. (1994), while 81 had placental weights above the mean, which was borderline significant (Binomial test, z-approximation, p = 0.05, 2-tailed). In contrast, for the trisomy C P M cases placental weights were clearly lighter compared to the means from Pinar et al. (1996): 53 cases had placental weights below the mean, while 16 had placental weights above the mean (Binomial test, z-approximation, p < 0.001).  2.4.5  F-P weight ratio  F - P weight ratio data are shown in Table 2.2. The F - P weight ratios for the matched controls are lower because their placentas were heavier compared to the trisomy C P M cases (Table 2.4), and thus the F - P weight ratios were compared to reference populations. For matched controls, 66 cases had F - P weight ratios below the means from Dombrowski et al. (1994), while 72 had placental weights above the mean, which was not significantly different from expected (Binomial test, z-approximation, p = 0.67, 2-tailed). Similarly, the distribution of F - P weight ratios among the trisomy C P M cases was as expected when compared to means from Heinonen et al. (2001): 34 cases had F - P weight ratios below and 35 cases above the mean. 21  2.4.6  Determinants of placental weight and birth weight  Possible determinants of placental weight and birth weight in the trisomy C P M cases include: sex of the infant, the chromosome involved in the trisomy, and the level of trisomy in the various placental tissues. To control for gestational age and to maintain a continuous outcome variable, placental and birth weight were converted to standard deviations (SDs) from the mean using the gestational age-corrected reference data of Molteni et al. (1978).  2.4.6.1 Sex of the fetus Sex of the fetus was not associated with placental or birth weight (data not shown).  2.4.6.2 Involved chromosome For the involved chromosome, there were sufficient sample sizes for trisomy 16 (n = 13), trisomy 7 (n = 10), and trisomy 2 (n = 9) to make comparisons between karyotypes; the other trisomies were grouped into 'other' (n = 37).  Means and standard deviations for these 4  trisomic groups are illustrated in Figure 2.1. O n bivariate comparisons (t-test), the only differences were between C P M involving trisomy 16 ( C P M 16) and each of the other 3 groups for birth weight (p < 0.05), with C P M 1 6 significantly lower i n each comparison. Therefore, the C P M 16 cases should be considered separately from all other C P M cases when considering birth weight. There were no significant associations with C P M 16 for placental weight as the outcome variable (data not shown).  2.4.6.3 Level of trisomy in trophoblast, mesenchyme, chorion For bivariate comparisons (Spearman's correlation) with placental weight and birth weight, the level of trisomy i n each lineage was coded to reduce the effect of random sampling error between cases (due to different sites of the placenta sampled, different number of sites 22  sampled, and different number of cells for cytogenetic analysis): 0% trisomy = ' 0 ' ; 0-50% trisomy = T ; 51-100% trisomy = ' 2 ' . The significant associations were between birth weight and the level of trisomic cells in the trophoblast (rho = -0.56, n = 40, p < 0.001), and between birth weight and the level of trisomic cells in the mesenchyme (rho = -0.28, n = 61, p = 0.014). The negative direction of the coefficients indicates that in each instance a higher level of trisomy is associated with a lower birth weight. Since there were no associations with placental weight, it was not surprising that the levels of trisomic cells in the trophoblast and mesenchyme were also significantly negatively correlated with F - P weight ratio (data not shown). The levels of trisomy in the trophoblast and mesenchyme were positively correlated with each other (rho = 0.54, n = 35, p < 0.001), suggesting potential confounding in their associations with birth weight. Furthermore, the level of trisomy in placental lineages may be confounded by the chromosome involved; for example, the C P M 16 cases had a significantly higher level of trisomy in the trophoblast compared to the other C P M cases (Mann-Whitney test, p < 0.005, 2tailed). Ideally, multiple linear regression should be carried out to characterize any independent effects. However, no model could be developed that satisfied the assumptions of linear regression. Instead of multiple regression, associations between birth weight, and the level of trisomy in the trophoblast and in the mesenchyme were analyzed for the C P M 16 cases and the other C P M cases separately. In the C P M 16 group, both the trophoblast and mesenchyme were significantly associated with lower birth weight (rho = -0.75, n = 10, p = 0.006; and rho = -0.79, n = 11, p = 0.002); in the other C P M group, there was a trend towards an association between the trophoblast and lower birth weight (rho = -0.26, n = 30, p = 0.083), but no evidence of an association between the mesenchyme and birth weight (rho = -0.13, n = 50, p = 0.19). When the level of trisomic trophoblast was categorized into an indicator variable (0% = ' 0 ' , >0% = '1'), the trend towards lower birth weight in the other C P M group was statistically significant (t = 23  1.74, df = 28, p = 0.046). Thus, trisomic trophoblast was associated with decreased birth weight in both the C P M 16 and other C P M groups (although the latter association was weaker), while the mesenchyme was associated with decreased birth weight only in the C P M 1 6 group. Furthermore, in the C P M 16 group, the level of trisomic trophoblast was highly significantly associated with the level of trisomy in the mesenchyme (rho = 0.90, n = 9, p < 0.001). However, in the other C P M group, the level of trisomic trophoblast was not associated with the level of trisomic mesenchyme (rho = 0.22, n = 26, p = 0.15). When the levels of trisomy were categorized into indicator variables, there was still no association between trophoblast and mesenchyme in the other C P M group (Fisher Exact test, n -40, p = 0.25). Together, these results suggest that the trophoblast is the key tissue involved in the determination of birth weight in both C P M 16 and other C P M cases; and that the association between birth weight and the level of trisomy in the mesenchyme in the C P M 16 cases was spurious because of its correlation with the level of trisomic trophoblast. To determine whether the presence of trisomy 16 ( C P M 16) and the level of trisomy in the trophoblast had effects on birth weight independent of each other, associations were tested for C P M 16 at different levels of trisomic trophoblast. C P M 16 still showed decreased birth weight both when the level of trisomic trophoblast was ' 0 % ' (means: C P M 1 6 = -0.34 ± 0.43 S D (n = 3) vs. other C P M = 0.65 ± 1.03 S D (n = 24); W e l c h ' s approximate t = 3.08, df = 5.8, p = 0.012), and when the level of trisomic trophoblast was greater than 0% (means: C P M 1 6 = -1.30 ± 0.64 S D (n = 7) vs. other C P M = -0.20 ± 1.25 S D (n = 6); Welch's approximate t = 1.95, df = 7.2, p = 0.046). This suggests that the presence of C P M 1 6 also has an effect on birth weight independent of the level of trisomy trophoblast. In other words, the presence of C P M 16 and the degree of trisomic trophoblast appear to have independent effects on birth weight.  24  2.4.6.4 Placental weight Although sex of the fetus, the involved chromosome, nor the level of trisomy in the three placental lineages were associated with weight of the placenta, placental weight itself may be a determinant of birth weight in trisomy C P M cases. Placental weight and birth weight were found to be positively correlated in the trisomy C P M sample (r = 0.60, n = 69, p < 0.001), which replicates an established finding in the general population (Molteni et al. 1978; Williams et al. 1997; Sanin et al. 2001). Thus, to determine whether C P M 16 and trisomic trophoblast mediate their effects on birth weight independent of placental weight, multiple linear regression was first carried out with birth weight as the outcome variable, and placental weight and the level of trisomic trophoblast as explanatory variables. The level of trisomic trophoblast had to be dichotomized into an indicator variable (0% = ' 0 ' and >0% = '1') because of heteroscedasticity (variance in the residuals decreased as the level of trisomic trophoblast increased). In the resulting model, both trisomic trophoblast (dichotomized) (b = -1.2, p < 0.001) and placental weight (b = 0.38, p = 0.006) had significant independent effects. In other words, the level of trisomic trophoblast had an effect on birth weight independent of placental weight.  Second, multiple linear  regression was carried out with birth weight as the outcome variable, and placental weight and the involved chromosome as explanatory variables ( C P M 1 6 vs. other C P M ) . In the resulting model, both C P M 16 (b = -1.0, p = 0.002) and placental weight (b = 0.42, p < 0.001) had significant independent effects. Thus, the presence of C P M 1 6 reduced birth weight independent of placental weight. In summary, therefore, these results together suggest that the presence of trisomic trophoblast and trisomy 16 reduce birth weight independent of each other and of placental weight.  25  2.5  Discussion  In this study, placental weights and birth weights were lower in trisomy C P M cases compared to matched controls, as hypothesized. Surprisingly, the level of trisomy in the placenta was not associated with placental weight. However, the level of trisomic trophoblast and the presence of trisomy 16 ( C P M 16) decreased birth weight independent of each other and of placental weight, suggesting they affect placental function rather than a simply reducing placental growth. A non-specific range of abnormalities was noted on pathology examination of the C P M placentas; in particular, there were was no pathognomic finding among C P M 16 placentas (Table 2.3). Thus routine placental pathology could not clearly identify the mechanisms by which trisomy affects placental function; instead, it is likely that more in depth cellular and physiological studies are required to determine how placental function is altered by trisomy. For example, trisomic trophoblast could reduce birth weight by a defect in extravillus trophoblast ( E V T ) invasion and remodelling of the spiral arteries resulting in poor uteroplacental perfusion (Wright et al. 2004) or a defect in syncytiotrophoblast formation and function (Frendo et al. 2000). Trisomy 16 in the trophoblast, in particular, appears to have a defect in E V T outgrowth (Chapter 7). In addition, F - P weight ratio was not altered in trisomy C P M , indicating that placental weight and birth weight were reduced to a similar degree. Thus, placental weight was likely decreased in trisomy C P M pregnancies due to the correlation between placental and birth weight seen i n the general population (Molteni et al. 1978; Williams et al. 1997; Sanin et al. 2001). Together, these results imply the following model for the pathogenesis of birth weight in trisomy C P M pregnancies: the presence of trisomy 16 and higher levels of trisomic trophoblast decrease fetal growth through an alteration in placental function; then, because of putative feto-placental signal(s) that regulate placental growth in the general population and produce the correlation between placental and birth weights, placental growth is similarly decreased. 26  One concern is that a number of factors known to be associated with birth weight, placental weight, and F - P weight ratio were not accounted for in this study because the clinical data from referred cases were limited. For instance, F - P weight ratio is decreased in maternal anemia (Williams et al. 1997), and in gestational diabetes (Lao et al. 1997). A s well, Williams et al. (1997) found the following variables had independent effects on birth weight: maternal pre-pregnancy weight, maternal height, maternal body mass index, parity, smoking, anemia, diabetes, weight gain, gestational age, female sex of the fetus, and ethnicity; for placental weight, the following had independent effects: maternal pre-pregnancy weight, maternal height, maternal B M I , parity, anemia, diabetes, weight gain, gestational age, and female sex of the fetus, and ethnicity; and for F - P weight ratio, socio-economic status, maternal pre-pregnancy weight, maternal B M I , smoking, anemia, gestational age, female sex of the fetus, and ethnicity had independent effects. It is possible in theory that these variables - other than parity, which was matched for - may not be equally distributed among the trisomy C P M cases and matched controls, and could account for the relationships seen here. However, in contrast to maternal age and sex of the fetus, which were expected to differ in the mothers of C P M pregnancies, none of the variables discussed above have a theoretical reason to differ in C P M mothers. Another concern is that placental weight is a crude marker, as much of what is measured is not functional placental tissue and variability is induced by differences in handling prior to measurement (Marais 1979; Lao and W o n g 1996; Heinonen et al. 2001). Regardless, placental weight is easy to measure and is weighed routinely after birth, which allows epidemiologic studies of adequate sample size. A l s o , since placental weight is positively correlated with birth weight, it is likely correlated with the 'functional' placental mass that modulates fetal growth. In conclusion, lower placental weight and birth weight were present in trisomy C P M pregnancies. Etiological factors include the level of trisomic trophoblast and the involvement of trisomy 16, which independently affect placental function. Future investigations should include 27  identification of the specific functions that are altered in the trisomic placenta, and the mechanisms causing trisomic placental cells to function abnormally.  28  2.6 References Dombrowski M P , Berry S M , Johnson M P , Saleh A A , Sokol R J (1994) Birth weight-length ratios, ponderal indexes, placental weights, and birth weight-placenta ratios in a large population. A r c h Pediatr Adolesc M e d 148:508-512 Frendo J L , Vidaud M , Guibourdenche J, Luton D , Muller F , Bellet D , Giovagrandi Y , Tarrade A , Porquet D , Blot P, Evain-Brion D (2000) Defect of villous cytotrophoblast differentiation into syncytiotrophoblast in Down's syndrome. J C l i n Endocrinol Metab 85:3700-3707 Frid C , Drott P, Otterblad Olausson P, Sundelin C , Anneren G (2004) Maternal and neonatal factors and mortality in children with D o w n syndrome born in 1973-1980 and 19951998. Acta Paediatr 93:106-112 Hassold T, Hunt P (2001) T o err (meiotically) is human: the genesis of human aneuploidy. Nat Rev Genet 2:280-291 Heinonen S, Taipale P, Saarikoski S (2001) Weights of placentae from small-for-gestational age infants revisited. Placenta 22:399-404 Henderson K G , Shaw T E , Barrett U , Telenius A H , Wilson R D , Kalousek D K (1996) Distribution of mosaicism in human placentae. H u m Genet 97:650-654 Kalousek D K , Howard-Peebles P N , Olson S B , Barrett JJ, Dorfmann A , Black S H , Schulman J D , W i l s o n R D (1991) Confirmation of C V S mosaicism in term placentae and high frequency of intrauterine growth retardation association with confined placental mosaicism. Prenat Diagn 11:743-750 Kalousek D K , Langlois S, Barrett I, Y a m I, W i l s o n D R , Howard-Peebles P N , Johnson M P , Giorgiutti E (1993) Uniparental disomy for chromosome 16 in humans. A m J H u m Genet 52:8-16 Kalousek D K , Langlois S, Robinson W P , Telenius A , Bernard L , Barrett IJ, Howard-Peebles P N , W i l s o n R D (1996) Trisomy 7 C V S mosaicism: pregnancy outcome, placental and D N A analysis in 14 cases. A m J M e d Genet 65:348-352 Kramer M S , Piatt R W , W e n S W , Joseph K S , A l l e n A , Abrahamowicz M , Blondel B , Breart G (2001) A new and improved population-based Canadian reference for birth weight for gestational age. Pediatrics 108:E35 Kuchinka B D , Barrett IJ, M o y a G , Sanchez J M , Langlois S, Y o n g S L , Kalousek D K , Robinson W P (2001) T w o cases of confined placental mosaicism for chromosome 4, including one with maternal uniparental disomy. Prenat Diagn 21:36-39 Lao T T , W o n g W M (1996) Placental ratio and intrauterine growth retardation. B r J Obstet Gynaecol 103:924-926  29  Lomax B L , Kalousek D K , Kuchinka B D , Barrett IJ, Harrison K J , Safavi H (1994) The utilization of interphase cytogenetic analysis for the detection of mosaicism. H u m Genet 93:243-247 Marais W D (1979) Placental size at birth. S A f r M e d J 55:153 Molteni R A , Stys SJ, Battaglia F C (1978) Relationship of fetal and placental weight in human beings: fetal/placental weight ratios at various gestational ages and birth weight distributions. J Reprod M e d 21:327-334 M y r e l i d A , Gustafsson J, Ollars B , Anneren G (2002) Growth charts for Down's syndrome from birth to 18 years of age. A r c h D i s C h i l d 87:97-103 Penaherrera M S , Barrett IJ, B r o w n C J , Langlois S, Y o n g S L , Lewis S, Bruyere H , HowardPeebles P N , Kalousek D K , Robinson W P (2000) A n association between skewed X chromosome inactivation and abnormal outcome i n mosaic trisomy 16 confined predominantly to the placenta. C l i n Genet 58:436-446 Pinar H , Sung C J , Oyer C E , Singer D B (1996) Reference values for singleton and twin placental weights. Pediatr Pathol Lab M e d 16:901-907 Robinson W P , Barrett IJ, Bernard L , Telenius A , Bernasconi F , W i l s o n R D , Best R G , HowardPeebles P N , Langlois S, Kalousek D K (1997) Meiotic origin of trisomy in confined placental mosaicism is correlated with presence of fetal uniparental disomy, high levels of trisomy in trophoblast, and increased risk of fetal intrauterine growth restriction. A m J H u m Genet 60:917-927 Sanin L H , Lopez S R , Olivares E T , Terrazas M C , Silva M A , Carrillo M L (2001) Relation between birth weight and placenta weight. B i o l Neonate 80:113-117 Shaffer L G , Langlois S, M c C a s k i l l C , M a i n D M , Robinson W P , Barrett IJ, Kalousek D K (1996) Analysis of nine pregnancies with confined placental mosaicism for trisomy 2. Prenat Diagn 16:899-905 Stoll C , Alembik Y , Dott B , Roth M P (1998) Study of Down syndrome in 238,942 consecutive births. A n n Genet 41:44-51 Williams L A , Evans S F , Newnham JP (1997) Prospective cohort study of factors influencing the relative weights of the placenta and the newborn infant. Bmj 314:1864-1868 Wright A , Zhou Y , Weier JF, Caceres E , Kapidzic M , Tabata T, Kahn M , Nash C , Fisher SJ (2004) Trisomy 21 is associated with variable defects in cytotrophoblast differentiation along the invasive pathway. A m J M e d Genet A 130:354-364  30  Table 2.1 Clinical and cytogenetic data for the 69 trisomy C P M cases Case  1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17  Previous Publications  Robinson et al. (1997) case 89.20  Kalousek et al. (1996) case 1 Kalousek et al. (1993) case 5 Johnson et al. (1993) caseB Robinson et al. (1997) case 90.90 Penaherrera et al. (2000) case 90.90  18 19 20  Kalousek et al. (1993) case 6 Shaffer et al. (1996) case 7  21 22  23 24 25  26 27 28  Kalousek et al. (1996) case 2 Robinson et al. (1997) case 91.33 Shaffer et al. (1996) case 3 Kalousek et al. (1993) case 7 Robinson et al. (1997) case 91.55 Penaherrera et al. (2000) case 91.55 Robinson et al. (1997) case 91.56 Robinson et al. (1997) case 91.69  Study case #  Local or Referred  Chr  Sex  Anomalies  AF %  Bid %  Am %  88.7 2 3 4 5 6 10 12 89.20  R R R R R R R R R  13 7 7 10 15 16 15 18 8  F M M M F M M F M  None None None None None None None None  0  0  0 0 0 0 0 0  15 16 90.13 90.29 90.75 90.40 90.87  L R R R L L R  12 12 12 12 13 2 7  F M M F F F F  None None None None None None  0 ?  0  90.90  R  16  M  None  0  0  91.7  L  F  None  0  91.10'  L  8/ 8,21 16  F  None  0  0  0  91.43  L  2  F  None  0  0  0  91.24  R  16  F  0  0  ?  91.33  R  7  M  Hip dysplasia None  91.47 91.54  R L  2 2  F F  None None  91.55  R  16  F  None  0  91.56  R  22  F  None  0  91.63 91.69  R L  12 10  F M  None None  ?  0 ? ?  0 0 ? ? ?  0 0 0 0 0 0 0 0 0 0  0 0 0  0 0 0 0  0  0  0  0  0 0  0  0  0 0  0 0  31  Case  Previous Publications  Study case #  Local or Referred  Chr  Sex  Anomalies  AF %  Bid %  Am %  29  Kalousek et al. (1993) case 3 Johnson et al. (1993) caseC Robinson et al. (1997) case 91.71 Penaherrera et al. (2000) case 91.71  91.71  R  16  F  None  0  0  0  91.79 91.80  R R  9 7  F F  None None  ?  0  0  0 0  91.85 91.87 91.88 91.89  L L R R  8 21 15,21 10  F M M F  None None None None  0  91.93 91.94  R L  F M  None None  0  91.96  R  18 2,18,1 8 9  M  None  0  92.11 92.14  R L  7 10  M M  ?  92.20 92.21  R R  20 10,12  F M  None Hydronephrosis None None  92.24 92.25  R R  15 16  F M  None Imp anus  0  92.29  L  2  F  None  0  92.48  L  16  F  Mild hypospadias  0  0  92.49  R  16  F  None  0  0  92.55  L  4  F  None  0  92.56 92.58  L L  13,18 7  F M  None None  0  92.59  R  2  M  92.77  R  17  F  30 31  32 33 34 35  Kalousek et al. (1996) case 5 Robinson et al. (1997) case 91.80  Robinson et al. (1997) case 91.89  36 37 38  Robinson et al. (1997) case 91.96  39 40 41 42 43 44  45  46  47  48 49 50 51 52  Robinson et al. (1997) case 92.21 Kalousek et al. (1993) case 4 Robinson et al. (1997) case 92.25 Shaffer et al. (1996) case 4 Robinson et al. (1997) case 92.29 Kalousek et al. (1993) case 8 Robinson et al. (1997) case 92.48 Kalousek et al. (1993) case 9 Robinson et al. (1997) case 92.49 Kuchinka et al. (2000) case CPM4-2 Kalousek et al. (1996) case 6 Shaffer et al. (1996) case 8  ?  0 0  0 0 0 0  ?  0  0  0  0  0  0 0  0 0  ? ?  0 0  0 0  ?  0 0  0 0  0 0 0  0 None  0  0  0  Case  53 54 55 56 57 58 59 60 61 62 63 64 65 66 67  Previous Publications  Robinson et al. (1997) case 93.1 Robinson et al. (1997) case 93.7 Robinson et al. (1997) case 93.53  Robinson et al. (1997) case 93.134  Kalousek et al. (1996) case 8 Robinson et al. (1997) case 94.85  68 69  Kalousek et al. (1996) case 14 Informative cases Summary statistics  Study case #  Local or Referred  Chr  Sex  Anomalies  AF %  Bid %  Am %  92.81 92.92 93.1  R R R  2 11 2  F F M  None None  0  0 0  0 0 0  93.7  L  22  F  None  0  93.53  L  12  M  None  0  93.54 93.83 93.134  R R R  9 2 16  M F F  None None None  0 0 0  93.177 94.20 94.26 94.35 94.76 94.82 94.85  L L L R L L R  17 8 16 16 9 7 7  F F F M F F M  None None  0  94.105  R  16  F  94.107  L  7  69  69 L=25 R = 44  69  ?  0  0 0 0  0 0 0 0  0 0 0 0 0 0  0 0 0 0  0  0  F  Possible VSD None  0  0  69 M =26  63 None = 58 (92%)  53  None None None  42  58  "Study case #" = case number for the ongoing study of trisomy mosaicism at UBC. "Chr" = chromosome involved in the trisomy. "AF" = % trisomic cells in amniotic fluid, assessed by amniocentesis. "Bid" = % trisomic cells in fetal or infant blood, assessed by cordocentesis or peripheral blood sampling. "Am" = % trisomic cells in the amnion, sampled post-partum. "?" = tissue assessed, but results not available. Full references in Appendix AI.  33  Table 2.2 Clinical and cytogenetic data for the 69 trisomy C P M cases  Case 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49  Gest Age (weeks) 40 38 40 40 38 40 37 41 37 37 41 36 40 40 38 36 41 40 36 38 40 40 35 41 40 38 40 40 38 37 40 40 37 40 40 40 40 40 40 39 38 41 39 35 42 36 31 38 38  Placental weight (g) 550 392 345 513 365 434 587 394 540 425 451 411 509 325 435 411 495 575 405 390 480 652 367 520 446 376 595 470 203 617 511 520 385 512 628 482 570 246 414 500 748 444 485 269 535 430 277 390 450  Birth weight (g) 3370 2556 2784 3238 2414 3689 3481 3481 3120 2530 3518 3377 3831 3420 2700 2336 3008 3445 2660 2970 2681 3482 2700 3555 3320 2010 4284 3940 1960 3717 3717 2985 2620 3774 3859 3944 3650 3292 3348 3210 3575 4120 2638 1800 3377 2280 1045 3175 3200  Feto-placental (F-P) weight ratio 6.13 6.52 8.07 6.31 6.61 8.50 5.93 8.84 5.78 5.95 7.80 8.22 7.53 10.52 6.21 5.68 6.08 5.99 6.57 7.62 5.59 5.34 7.36 6.84 7.44 5.35 7.20 8.38 9.66 6.02 7.27 5.74 6.81 7.37 6.14 8.18 6.40 13.38 8.09 6.42 4.78 9.28 5.44 6.69 6.31 5.30 3.77 8.14 7.11  Chorion  Mesen  Troph  %  %  %  0 4 73 13 0 0  0 50 32 100 10 47 0 0 20 90 0 0 3 18 2 0 28 91 0 0 80 84 0 0 0 75 0 53 100 28 47  36 16 0  -  0 -  100 -  0 33 27 11 0 25 58 20 0 20 40 0 8 0 73 0 0 87 33 67 -  -  11 0 40 0  16 0 20 10 0 40 0 0 23 80 33  -  83 0 -  7 100 0 100 -  100  -  100 100  -  -  60  32  -  0 0 -  0 17 -  63 0 -  0 22 -  0 100 0 0 0 -  0 0 0 0 0 -  0 -  100 0 61 80 -  34  Case  50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 Total* cases Mean {range}  Gest Age (weeks) 39 36 40 40 39 39 33 40 38 39 38 40 40 36 39 41 40 40 36 40  Placental weight  69 38.7±2.0 {31-42}  69 452+106 {203-748}  (g)  365 497 484 526 287 450 280 583 297 576 442 385 572 245 520 360 474 393 480 470  Birth weight (g) 3700 3119 3250 2936 3235 2923 1820 4040 2270 3263 2817 3300 4520 1960 2800 3121 3500 3460 2014 3885  Feto-placental (F-P) weight ratio 10.14 6.28 6.71 5.58 11.27 6.50 6.50 6.93 7.64 5.66 6.37 8.57 7.90 8.00 5.38 8.67 7.38 8.80 4.20 8.27  Chorion %  Mesen %  Troph %  100 0 27 0 18 100 20 7 13 0 0 7 0 27 87 0  7 0 20 0 3 100 50 43 5 0 20 73 50 40 93 13  0 0 0 0 19 0 0 0 35 0 0 14 78 0  69 3118±662 {10454520}  69 7.09±1.63 {3.77-13.38}  56 28.6±35.4 {0-100}  61 31.6+34.8 {0-100}  40 16.0±29.5 {0-100}  'Chorion' = % trisomic cells i n the chorion, sampled post-partum. 'Mesen' = % trisomic cells in the chorionic villus mesenchyme, sampled post-partum. 'Troph' = % trisomic cells i n the trophoblast, sampled post-partum. For the mesenchyme and chorion, "0%" = no trisomic cells on conventional cytogenetics. For the trophoblast, 0-10% on F I S H were all coded as "0%" (see text). " - " = not performed.  35  Table 2.3 Findings for (local) cases where placental pathology was performed Case  Chromosome  Placental pathology  1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19  13 7 7 10 15 16 15 18 8 12 12 12 12 13 2 7 16 8/8,21 16  Normal  20 21 22 23 24 25 26 27 28 29 30 31 32  2 16 7 2 2 16 22 12 10 16 9 7 8  33  21  34 35 36 37 38 39 40  15,21 10 18 2,18,18 9 7 10  41 42 43  20 10,12 15  Normal Normal  Lymphocytic villitis Normal: Minimal focal microscopic findings not considered sufficiently extensive to be of any clinical significance (focally increased perivillus fibrin, slight variation in villus maturity, focal congestion, a few villi which are hypervascular, a suggestion of very focal non-specific vessel villitis) Normal  Early acute chorioamniotis  Focal old mural thrombus  1) Placental infarcts; 2) Trophoblastic cysts; 3) Deciduitis of membranes, focal; 4) Accessory lobe placenta. 1) Decidual necrosis and inflammation - membranes and maternal surface.  Normal  1) Infarct with decidual necrosis and thrombosis (abruption), old, placental margin; 2) Hemosiderin deposition, macrophages, membranes.  36  44 45  16 2  46  16  47 48 49 50 51 52 53 54 55 56  16 4 13,18 7 2 17 2 11 2 22  57 58 59 60 61 62 63  12 9 2 16 17 8 16  64 16 65 9 66 7 7 67 69 16 Total # with abnormalities  1) Focal villitis of undetermined etiology. 2) Solitary intervillus thrombus. 1) Perivillusfibrinosis;2) Hemorrhage, old, extraplacental membranes. Normal Normal Normal  1) Increased intervillus fibrin; 2) Hemorrhage with early organization is consistent with placental abruption. 1) Perivillus fibrin deposition with no evidence of infarction.  Normal Normal 1) Decreased fetal vascularization of chorionic surface; 2) Focal mural thrombosis fetal vessels. 3) Focal villus immaturity. Focally prominent fibrinosis.  Normal 12/23 (52%)  37  Table 2.4 Clinical characteristics of the trisomy C P M cases and the matched controls  Maternal age Parity Gestational age Sex of the fetus Birth weight Placental weight F-P weight ratio  mean 37.8±3.5 years 1.14+1.06  38.7±2.0 weeks 26 M : 43 F 3118±662g  77% < mean 49% < mean  Significance  Matched controls  Trisomy CPM cases  n 64 49 69 69 69 69 69  Mean 35.3±4.4 years 1.06+1.07  38.7+1.8 weeks 79 M : 59 F 3435±586g  41% < mean 48% < mean  n 138 138 138 138 138 138 138  p < 0.001  n.s. n.s.  ,  p = 0.006 p = 0.001 p< 0.001; n.s.  n.s.; n.s.  Placentas were handled differently before weighing for the trisomy C P M cases and matched controls. Therefore, placental weight and F - P weight ratio for the C P M cases and matched controls were instead compared to gestational age-specific means from published reference populations with similar handling procedures (see text). Shown is the proportion of C P M cases and matched controls that were below the gestational age-corrected means from the reference populations, with the p-values when compared to the expected 50% < mean for the C P M cases and controls, respectively. For gestational age and birth weight, the t-test was used; for maternal age, Welch's approximate t-test was used because of unequal variances; for parity, MannWhitney test (z-approximation) was used because of non-normality; and for sex of the fetus, the Fisher Exact test was used. A l l tests were 1-tailed except for parity and gestational age.  3 8  Figure 2.1 Birth weight means for the chromosomes involved in the trisomy C P M cases  1.001  0.50-  0.00-  -0.50"  -1.00"  Other  +2  +7  +16  Trisomy  Mean ± standard deviation. The trisomy 16 group was significantly different from each of the other 3 trisomic groups (see text).  39  3 Pathogenesis of CPM16 pregnancies  4  3.1  Note  I wrote this chapter/manuscript and did the data organization and analysis, with the following clarifications and exceptions. The C P M 1 6 cases are from the ongoing U B C study collected since 1988. Uniparental disomy testing was performed initially by Dr. S. Langlois, and since 1994 i n the lab of Dr. W . Robinson. The results in this Chapter are slightly different from the publications (see footnote) because (1) sample size is increased; (2) live births are separated from neonatal/intrauterine deaths; (3) birth weight is corrected for gestational age using more recent data from (Kramer et al. 2001); and (4) the role o f ascertainment is investigated. However, the major conclusions remain the same.  3.2 Introduction Trisomy 16 may be the most frequent chromosome abnormality at conception (Wolstenholme 1995). A m o n g all clinically recognized pregnancies, it has an incidence of - 1 . 5 % (Hassold and Jacobs 1984). Although most trisomy 16 embryos are spontaneously aborted or are noted to have arrested development between 8-15 weeks gestation, some embryos survive and those pregnancies are candidates for prenatal diagnosis (Benn 1998). It has been estimated that approximately 34 per 100,000 chorionic villus sampling ( C V S ) analyses detect trisomy 16 (Wolstenholme 1995), while Benn (1998) notes that a recent estimate for amniocentesis has not been reported. In these surviving embryos, the trisomy 16 cells are virtually always completely or predominantly confined to the placenta ( C P M ) , with only one case of apparently non-mosaic (full) trisomy 16 i n the fetus diagnosed at C V S (1994).  A version of this chapter has been published. 1) Yong PJ, Marion SA, Barrett IJ, Kalousek DK, Robinson WP. 2002. Evidence for imprinting on chromosome 16: the effect of uniparental disomy on the outcome of mosaic trisomy 16 pregnancies. Am J of Med Genet 112(2): 123-32. 2) Yong PJ, Barrett IJ, Kalousek DK, Robinson WP. 2003. Clinical aspects, prenatal diagnosis and pathogenesis of trisomy 16 mosaicism. J Med Genet 40(3): 175-82.  4  40  Almost all C P M 16 cases originate from a trisomy 16 zygote as a consequence of a maternal meiosis I nondisjunction, followed by trisomy zygote rescue (Robinson et al. 1997). When a trisomy 16 conceptus is rescued, one of the two maternal chromosomes or the paternal chromosome can be lost. If the former occurs, the result is biparental disomy 16 (bpd(16)) or a chromosome 16 inherited from each parent; i f the latter occurs, the result is maternal uniparental disomy 16 (upd(16)mat) or both chromosomes inherited from the mother (Engel 1980; Spence et al. 1988). Uniparental disomy ( U P D ) could have a distinct phenotypic effect i f imprinted genes (i.e. genes whose expression depends on whether they are inherited from the mother or father) exist on chromosome 16. Imprinting on human chromosome 16 has been proposed based on orthology with imprinted regions in the mouse (Searle et al. 1989). Although reports of malformation in cases of upd(16)mat have raised the possibility of imprinting (Ledbetter and Engel 1995), studies of upd(16)mat have neither conclusively supported or excluded imprinting effects (Kotzot 1999). In fact, in an earlier review, Kalousek and Barrett (1994) noted that the degree of trisomy 16 in the placenta seemed to correlate with the growth restriction of a chromosomally normal fetus independent of U P D status. Besides upd(16)mat and the level of trisomy i n the post-partum trophoblast (Chapter 2), other factors potentially contributing to the pathogenesis of C P M 16 are the level of trisomy in amniotic fluid (by amniocentesis) and in the amnion (sampled after delivery), as well as ascertainment bias and sex of the fetus. I hypothesized that C P M 16 pregnancies would tend to be growth restricted with significant rates of malformation, and that the risk of these outcomes w i l l be associated with each of the following predictive factors: upd(16)mat, degree/distribution of trisomy in amniotic fluid (by amniocentesis) and the placenta (by C V S ) , ascertainment, and sex of the fetus.  In this  study, statistical analysis was performed on data from a large series (n = 173) of published and unpublished C P M 16 cases with the purpose of (1) summarizing the prenatal and perinatal outcome of C P M 16 pregnancies; and (2) evaluating the predictive value of the factors for 41  measures of pregnancy outcome. The identification of important predictive factors w i l l aid genetic counseling after prenatal diagnosis and w i l l elucidate factors that are involved in the pathogenesis of C P M 16 during pregnancy.  3.3 Methods 3.3.1  CPM16 cases  The study sample (n = 173 cases) consists of C P M 16 pregnancies diagnosed prenatally by C V S or amniocentesis with or without molecular testing for the U P D status of chromosome 16 (Appendix AI). Rare cases with paternal origin of the trisomy (n = 2), partial trisomy (n = 3), and concomitant aneuploidy (n = 1) were excluded in order not to confound the analysis (Appendix AI).  Some cases are from an ongoing study of C P M at the University of British  Columbia ( U B C ) (n = 69), which consists of cases referred from other centres (n = 63) and cases initially ascertained at the Children's and Women's Health Centre of British Columbia ( C & W ) (n = 6). Some data from most of these cases have been published previously, and there is overlap with cases published by other research groups (Appendix AI).  The study was  approved by the ethics committees of U B C and C & W (Appendix A ) . Informed consent was obtained from parents to provide clinical data and for determination of their child's U P D status. U P D testing was carried out as described (Robinson et al. 1997). Clinical information was sent by the referring physician either directly or by filling out a questionnaire; in those cases where prenatal diagnosis occurred in Vancouver, clinical information was also gathered from medical records at the U B C Department of Medical Genetics. A n additional 104 cases are from other published reports to date. Using the review of C P M 16 cases by Benn (1998) as a starting point, data were verified from the original sources and care was taken to eliminate duplicated cases (i.e. those published in two separate reports).  42  Data were collected on the following variables: (1) pregnancy outcome (i.e. live birth, intrauterine death or termination of pregnancy); (2) sex of the fetus; (3) gestational age at delivery; (4) malformation detected in the fetus/neonate/infant ("malformation" used as general term independent of etiology, including possible disruptions and deformations); (5) fetal/neonatal weight at pregnancy outcome (in standard deviations from the Canadian population mean corrected for gestational age and sex (Kramer et al. 2001)); (6) ascertainment (e.g. advanced maternal age, abnormal serum screen or ultrasound, etc.); (7) U P D status; (8) percent trisomy i n amniotic fluid on amniocentesis (by conventional cytogenetics); and (9) presence or absence of trisomy in tissues of the fetus. For variable (9), data from conventional cytogenetics, F I S H or molecular methods ( P C R ) were taken into account; i f results were contradictory (i.e. one was positive for trisomy, the other negative), then the fetal tissue was coded as positive. Molecular detection of trisomy in fetal tissues was only considered for cases from the U B C study where detailed information on chromosome 16 markers was available, and cases from other published reports where there was an explicit statement that P C R showed or excluded trisomy. It should be emphasized that the study sample may be biased towards cases with poorer outcomes, since such cases are more likely to be ascertained (e.g. due to anomalies observed on ultrasound), referred for research purposes, and/or submitted for publication. Therefore, purely descriptive statistics should not be considered estimates of the actual values in the general population, but are intended as descriptions of the study sample specifically. Statistical associations between variables are more likely to be unbiased. Furthermore, it should be noted that there was variation in the quantity and quality of data available among the cases (e.g. for descriptions of malformations).  43  3.3.2  Statistical analysis  Statistical analysis was carried out using SPSS 10.0 and the VassarStats Web Site for Statistical Computation (http://faculty.vassar.edu/lowry/VassarStats.html). Tests were 1-tailed due to a priori evidence or rational mechanisms, unless otherwise noted. Welch's approximate t-test was utilized when there was inequality of variances. Means are reported ± standard deviation.  3.4 Results 3.4.1  Pregnancy outcomes of CPM16  There were 166 cases informative for pregnancy outcome. Even with an expected bias towards poor outcome, the majority of cases (62%) resulted in live births with survival beyond the neonatal period. Eleven percent of pregnancies ended in intrauterine death (IUD) and 5% in neonatal death, while 22% of the pregnancies were electively terminated. For the descriptive statistics and statistical associations below, only the live births (with survival beyond the neonatal period) were included. Intrauterine and neonatal deaths were analyzed separately below, to allow comparison to the live births with survival beyond the neonatal period.  3.4.2  Clinical outcome of CPM16 live births  Figure 3.1 shows the distribution of gestational ages at delivery for the live births (n = 77 informative cases): the average gestational age was 36.0 ± 3.2 weeks. The distribution of birth weights (number of standard deviations from the mean) for the live births is illustrated in Figure 3.2. Virtually all birth weights (93%) were below the general population mean birth weight (i.e. 0 standard deviations), with an average birth weight of -1.76 ± 1.08 standard deviations from the general population mean. Forty two % of live births had at least one malformation (37 of 89 cases informative for malformation status). W i t h the exception of one case with only a minor 44  anomaly (asymmetric nipples), all cases had at least one major malformation. Some malformations were present in 6 or more live births (i.e. in > 6% of informative cases) and were considered more likely to have a true association with C P M 16: V S D  5  12% (n = 11), A S D  6  10%  (n = 9), and hypospadias 27% (n = 7, of 26 informative male cases). A l l of these malformations were significantly more frequent when compared to the corresponding frequencies among neonates in the general population (Table 3.1, p < 0.0001).  3.4.3  Trisomy in amniotic fluid in CPM16 live births  Fifty-six percent of the live births with amniocentesis (49/87) had 0% trisomy detected in amniotic fluid. For those positive for trisomy (>0%) in amniotic fluid, the distribution was skewed towards low-level mosaicism (Figure 3.3), although the highest level detected was 93% trisomy. For statistical analyses, the level of trisomy in amniotic fluid was coded into a binary variable: 0% trisomy = ' 0 ' ; >0% trisomy = T . The presence of trisomy in amniotic fluid was associated with lower birth weight (t = 3.39, df = 62, p = 0.0005; Figure 3.4) and higher risk of malformation (Fisher Exact test, n = 82, p = 0.017; R R = 1.79; Table 3.2).  Considering only cases where amniotic fluid was positive for  trisomy, the percent trisomy was not associated with either birth weight or malformation (data not shown). Therefore, the simple presence of trisomy in amniotic fluid (as sampled by amniocentesis), but not the actual level above 0%, is predictive of outcome. It should be noted, however, that there was considerable variability in outcomes even for a given result in amniotic fluid. Figure 3.4 shows that there is a large standard deviation for birth weight among both cases negative for trisomy in amniotic fluid (std dev = 1.02) and cases positive for trisomy in amniotic fluid (std dev = 1.00). In addition, Table 3.2 shows that although the risk of  5 6  Excluding 1 case of tetralogy of Fallot and 3 cases of endocardial cushion defect/atrioventricular canal. Excluding 3 cases of endocardial cushion defect/atrioventricular canal. 45  malformation increases when trisomy is present in amniotic fluid, the risk of malformation is still considerable (33%) even when no trisomy is detected. This variability raises the question of how well the level of trisomy in amniotic fluid cells reflect trisomy in 'true' fetal tissues. Fetal tissues (other than umbilical cord and amnion) were examined in 59 live births; most of these were analyzed after birth, although findings from cordocentesis were also included. The most common tissues studied were blood (n = 55) and skin (n = 18). The presence of trisomy i n amniotic fluid was associated with the presence of trisomy in fetal tissues (Fisher exact test, n = 52, p = 0.042; R R = 5.77; Table 3.3). However, for the 28 cases positive for trisomy in amniotic fluid, 75% (21/28) were negative in fetal tissues (Table 3.3), which reflects the patchy distribution of trisomy in fetal tissues and the limited cytogenetic investigations i n live born infants.  3.4.4  Ascertainment o f C P M 1 6 live births  Ascertainment was divided into two types: 1) 'unbiased' (e.g. A M A , past or family history of chromosome abnormality, investigation of Mendelian or biochemical disorder); and 2) 'biased' (abnormal serum screen, and abnormal ultrasound including altered fetal development or amniotic fluid level). Cases with biased ascertainment may have poorer outcome, as an abnormal serum screen or ultrasound indicates underlying pathology. H a l f of live births had an unbiased ascertainment (n = 36); and half had biased ascertainment (n = 35), of which 22 had abnormal serum screen and 13 had abnormal ultrasound findings. Abnormal ultrasound findings included I U G R (n = 4), single umbilical artery (n = 2), I U G R + dolichocephaly (n = 1), I U G R + hydropic placenta (n = 1), I U G R + oligohydramnios (n = 1), I U G R + polydramnios + thick cystic placenta (n = 1), large and thick placenta with echogenic regions (n = 1), large sonolucent and cystic placenta (n = 1), and nuchal edema (n = 1). Biased ascertainment had a significant association with lower birth weight (Welch's approximate t = 46  2.85, df = 35.4, p = 0.0035; Figure 3.5), but not with higher risk of malformation (Fisher Exact test, n = 65, p = 0.13; data not shown). The significant association with lower birth weight still held when biased ascertainment was separated into abnormal serum screen (Welch's approximate t = 2.63, df = 41.9, p = 0.006; Figure 3.6) or abnormal ultrasound finding (Welch's approximate t = 2.80, df = 32.0, p = 0.005; Figure 3.7).  3.4.5  Sex of the fetus in CPM16 live births  The sex ratio of the live births was 0.48 (32 males and 67 females), which is significantly different from the expected ratio of 1.07 calculated from prenatal controls by Huether et al. (1996) (Binomial test, z-approximation, n = 99, p < 0.001). This deviation in sex ratio towards females confirms a previous observation made by Benn (1998). The male sex had a significant association with lower birth weight (t = 2.35, df = 70, p = 0.011; Figure 3.8), but no association with increased risk of malformation (Fisher Exact test, n = 84, p = 0.26; data not shown).  3.4.6  C V S in CPM16 live births  In Chapter 2, the level of trisomy in the trophoblast of the post-partum placenta was associated with birth weight. Here, the level of trisomy on direct C V S (villus cytotrophoblast) and cultured C V S (villus mesenchyme) was investigated because they provide a 'snapshot' of the first-trimester placenta, and a sufficient number of cases were available for a statistical analysis. Since most cases had 100% trisomy on either C V S , these cases were compared to those with <100% trisomy. F u l l (100%) trisomy in direct C V S was significantly associated with lower birth weight (t = 4.08, df = 27, p < 0.001; Figure 3.9), but not with higher risk of malformation (Fisher Exact test, n = 33, p = 0.067; data not shown). Similarly, the level of trisomy in cultured C V S was associated with lower birth weight (t = 1.89, df = 27, p = 0.035; 47  Figure 3.10), but not with risk of malformation (Fisher Exact test, n = 35, p = 0.13; data not shown).  3.4.7  upd(16)mat in C P M 1 6 live births  If chromosome loss occurs randomly during trisomy rescue, one-third of C P M 16 should have upd(16)mat. Forty two-percent (22/53) of live births had upd(16)mat, which is not significantly different from the expected 33% (Binomial test, z-approximation, n = 53, p = 0.13). Upd(16)mat was significantly associated with both lower birth weight (t = 2.10, df = 42, p = 0.021; Figure 3.11), and higher risk of malformation (Fisher Exact test, n = 49, p = 0.020; R R = 1.78; Table 3.4). Again, it is emphasized that there is variability even within categories, with 43% of bpd(16) having a malformation (Table 3.4) and birth weight standard deviations of 1.08 and 1.04 for bpd(16) and upd(16)mat, respectively (Figure 3.11). In addition, a wide range of malformations were present among upd(16)mat cases, with no particular malformation seen significantly more frequently compared to the bpd(16) cases (data not shown). Since cases of upd(16)mat with particularly poor outcome may be more likely to be reported in the literature, the t-test for association between upd(16)mat and birth weight was repeated using only cases from the U B C study. The frequency of upd(16)mat in the U B C study was also 42% (15/36). A m o n g cases in the U B C study, upd(16)mat was still associated with both lower birth weight (t = 1.72, df = 29, p = 0.049; data not shown) and higher risk of malformation (Fisher Exact test, n = 33, p = 0.027; R R = 2.26; data not shown). Thus, any publication bias towards upd(16)mat cases with poor outcome appears to be minimal.  3.4.8  Possible confounding  Thus far, the following variables have shown significant associations with birth weight: trisomy in amniotic fluid, biased ascertainment (both abnormal serum screen and abnormal 48  ultrasound), male sex of the fetus, trisomy on direct and cultured C V S , and upd(16)mat. Therefore, the effect of one of these variables on birth weight may be confounded by any of the other variables. Ideally, multiple linear regression would be utilized to determine i f there are independent effects; however, sample size was too small to accommodate so many explanatory variables. Thus, bivariate associations were sought for all the explanatory variables (Table 3.5). The only significant associations were between the following: (1) trisomy in amniotic fluid and biased ascertainment; (2) trisomy in amniotic fluid and ascertainment by abnormal serum screen, specifically; and (3) trisomy in direct C V S and trisomy in cultured C V S . Therefore, confounding between the variables is unlikely except for between trisomy in amniotic fluid and biased ascertainment (notably, abnormal serum screen), and between direct and cultured C V S . O n the other hand, the effect of placental trisomy ( C V S ) may be independent of fetal trisomy (amniocentesis), and vice versa; and, further, the effect of upd(16)mat may be independent of both placental and fetal trisomy. A multiple linear regression model that satisfied the assumptions for regression - for birth weight as the outcome variable and trisomy on amniocentesis and abnormal ascertainment as explanatory variables - could not be constructed. Stratification of ascertainment - dividing cases with unbiased ascertainment and cases with ascertainment by abnormal serum screening to see i f trisomy amniotic fluid retains its effect on birth weight in both conditions - was limited by small sample size. Similarly, neither multiple linear regression nor stratification could be carried out for direct and cultured C V S . In contrast, only two variables, trisomy in amniotic fluid and upd(16)mat, had significant associations with higher risk of malformation. Therefore, multiple logistic regression modeling was carried out. Both explanatory variables had independent effects on birth weight (data not shown).  49  3.4.9  Gestational age at delivery in CPM16 live births  None of the explanatory variables was significantly associated with gestational age at delivery, except for biased ascertainment (t = 1.92, df = 58, p = 0.030; Figure 3.2), which was associated with lower gestational age. In particular, abnormal ultrasound (t = 2.14, df = 37, p = 0.020; Figure 3.13), but not abnormal serum screen (Welch's approximate t = 1.32, df = 34.6, p = 0.098; data not shown), was significantly associated with lower gestational age at delivery.  3.4.10 Intrauterine death and neonatal death in CPM16 To clarify the nature of the intrauterine and neonatal deaths, associations were determined between each outcome and birth weight, malformation, trisomy in amniotic fluid, ascertainment, upd(16)mat, sex of the fetus, and gestational age at delivery (neonatal deaths only). Risk of intrauterine death was significantly associated only with lower birth weight (t = 2.67, df = 76, p = 0.005; Figure 3.14). Risk of neonatal death was associated with lower gestational age (t = 3.25, df = 83, p = 0.001; Figure 3.15), presence of malformation (Fisher Exact test, n = 97, p = 0.001; R R = infinity; Table 3.6), and biased ascertainment (Fisher Exact test, n = 79, p = 0.043; R R = 6.17; Table 3.7). Abnormal ultrasound had a significant association with neonatal death (Fisher Exact test, n = 54, p = 0.030; R R = 8.71; Table 3-8), while abnormal serum screen had a non-significant trend (Fisher Exact test, n = 62, p = 0.18; R R = 4.44; Table 3-9). Abnormal ultrasound findings included I U G R (n = 1), olighydramnios (n = 1), left kidney agenesis + probable atrioventricular canal (n = 1), I U G R + dilated cerebral ventricles and single umbilical artery + a small placental haematoma that disappeared 2 weeks later (n = 1). The lack of association of outcome with trisomy in amniotic fluid, upd(16)mat, or sex of the fetus indicates that these (potentially) antenatal variables cannot predict intrauterine or neonatal death. Figure 3.15 indicates that gestational age is a poor predictor for neonatal death, 50  with the exception of very pre-term deliveries (< 28 weeks). Similarly, for the prediction of neonatal death, presence of malformation has a positive predictive value of only 18%, but a negative predictive value of 100% (Table 3-6); ascertainment by abnormal ultrasound has a positive predictive value of only 24%, but a good negative predictive value of 97% (Table 3.8). Table 3.10 describes the ascertainment and cardio-pulmonary malformations for each neonatal death, which includes 3 cases with pulmonary hypoplasia (of which 1 was associated with oligohydramnios), 1 case with a rare congenital tracheal narrowing, and 1 case with only one coronary artery that arose from the pulmonary artery trunk. Sample size was insufficient to perform logistic regression modelling or stratification analysis.  3.5 Discussion 3.5.1  Clinical outcome of CPM16  Although this sample is likely biased towards poorer outcome, the majority of prenatally diagnosed C P M 16 pregnancies resulted in live births with survival beyond the neonatal period. The distribution of gestational ages for all the live births suggests that C P M 16 pregnancies may be at higher risk for preterm delivery (Figure 3.1). In addition, consistent with the hypothesis, virtually all birth weights for the live births were below the gestational age-mean (Kramer et al. 2001) (Figure 3.2). This indicates that some level of below-average growth is a nearly universal phenomenon in C P M 1 6 . Malformations that are likely to have a true association with C P M 1 6 are V S D , A S D and hypospadias (Table 3.1). A s hypothesized, birth weight and malformation showed associations with most of the predictive factors.  3.5.2  Amniotic fluid trisomy and CPM16 live births  The presence of trisomy in amniotic fluid as assessed by amniocentesis was associated with lower birth weight (Figure 3.4) and increased risk of malformation (Table 3.2), suggesting 51  that fetal trisomy is an important determinant of outcome. However, trisomy in amniotic fluid was associated with biased ascertainment (namely, by abnormal serum screen), and it was not possible to delineate whether the effect of amniotic fluid trisomy was independent of ascertainment by abnormal serum screen. More abnormal serum screen values may reflect more abnormal placental function (see below).  3.5.3  Ascertainment and CPM16 live births  The association between ascertainment by abnormal serum screen and birth weight suggests that the processes giving rise to increased M S h C G and/or M S A F P in C P M 16 (Benn 1998), such as altered secretion, post-translational modification or clearance of the proteins in the placenta (Frendo et al. 2004), may be associated with processes that cause poorer fetal growth (Morssink et al. 1996). Previous studies have demonstrated associations between idiopathic elevated M S A F P and/or M S h C G and poor pregnancy outcomes such as I U G R (e.g. (Chandra et al. 2003; Lepage et al. 2003). However, as noted, it was not possible to determine whether an abnormal serum screen has an effect independent of trisomy in amniotic fluid. Although ascertainment by abnormal ultrasound was associated with lower gestational age, the difference in magnitude (36.8 vs. 34.8, on average) may not be clinically significant except for the lower bounds of the distributions.  3.5.4  Sex of the fetus and CPM16 live births  The sex ratio in the study sample (0.48) was biased towards females as initially noted in Benn (1998), and was significantly different from the expected sex ratio calculated from prenatal controls (Huether et al. 1996). It was also different from the sex ratio in trisomy 16 spontaneous abortions (1.00) (Hassold et al. 1983), providing evidence for greater frequency of trisomy rescue in female trisomy 16 embryos or for selection against male C P M 16 embryos  post-rescue. This is similar to a report of a significant excess of females in prenatally diagnosed trisomy 21 mosaicism (sex ratio = 0.72) compared to an excess of males i n prenatally diagnosed non-mosaic trisomy 21 (Hook et al. 1999). The same phenomenon for prenatally diagnosed trisomy 21 mosaicism was also reported by Huether et al. (1996), who also found a significant excess of females in prenatally diagnosed trisomy 18 mosaicism (sex ratio = 0.52) and a trend towards excess females in prenatally diagnosed trisomy 13 mosaicism (sex ratio = 0.76) compared to prenatal controls. In addition, it is interesting that males had lower birth weights than females, providing more evidence for selection against C P M 16 males.  3.5.5  C P M 1 6 and C V S  It has previously been shown that the clinical outcome of trisomy apparently confined to the placenta was strongly associated with the level of trisomy in the term trophoblast but not with the level of trisomy on C V S (Robinson et al. 1997). However, the data were confounded by the inclusion of a mixture of trisomies involving different chromosomes and different origins. In this study sample, full (100%) trisomy on direct C V S (cytotrophoblast) was found to be associated with low birth weight. Cytotrophoblast function is important for implantation and for differentiation to the hormone secreting syncytiotrophoblast and invasive extravillus cytotrophoblast (see Chapters 1 and 7). F u l l trisomy on cultured C V S (chorionic villus mesenchyme) was also found to be associated with low birth weight. Given the results in Chapter II, this is likely due to its association with the level of trisomy in direct C V S , thus reflecting the level of trisomic trophoblast. This is supported by the fact that the p-value for the association between direct C V S and birth weight was more statistically significant than that for cultured C V S : p < 0.001 versus p = 0.035, respectively. Interestingly, the level of trisomy in direct C V S and cultured C V S was not associated with the presence of trisomy in amniotic fluid. This observation suggests that the effects of 53  trisomy in the placenta (the trophoblast, specifically) on birth weight is independent of trisomy in fetal tissues, which supports the model of placental trisomy causing an alteration in placental function that negatively affects fetal growth (Chapter 2).  3.5.6  Intrauterine and neonatal death in CPM16  A s only birth weight was significantly associated with intrauterine death, there were no useful variables identified for its prediction. In contrast, risk of neonatal death was associated with lower gestational age, higher risk of malformation, and biased ascertainment (in particular, abnormal ultrasound). Malformation and abnormal ultrasound had poor positive predictive values (18% and 24%) but good negative predictive values (100% and 97%). Thus, while normal ultrasounds are reassuring, malformation or other abnormal ultrasound finding does not predict neonatal death.  3.5.7  Evidence for imprinting on chromosome 16  Maternal uniparental disomy for chromosome 16 (upd(16)mat) had an effect on birth weight and risk of malformation, independent of the presence of trisomy in the placenta and fetus as evaluated by C V S and amniocentesis, respectively. It is therefore hypothesized that upd(16)mat predisposes to growth restriction and malformation as a result of one or more imprinted gene(s) on chromosome 16. Conceptually, these gene(s) may be imprinted in the fetus or placenta, or both. In addition, it is conceivable that individual cases of growth restriction or malformation may be caused by isodisomy, which would result in the fetus being homozygous for a deleterious recessive mutation on chromosome 16 that is heterozygous in the mother. This probably does not account for the growth restriction seen in the study, however, since it is unlikely that deleterious recessive alleles are heterozygous in a sufficient number of mothers to cause the consistent pattern of growth restriction observed in upd(16)mat cases.  It 54  cannot be ruled out that isodisomy accounts for the slightly higher rate of malformation among upd(16)mat cases. However, most cases of upd(16)mat are heterozygous for the majority of chromosome 16 and malformations are observed in some cases even in the absence of any detectable region of isodisomy (W. Robinson, unpublished data). Currently, there are no known imprinted genes on chromosome 16. While Searle et al. (1989) noted that an imprinted region of mouse chromosome 11 has orthology with human chromosome 16, the imprinted genes since identified in this region have orthologues on other human chromosomes (Morison and Reeve 1998). Specifically, the orthologues of G r b l O / M e g l and U 2 a f l - r s l are located on human chromosomes 7 p l 2 and 21q22.2, respectively. The Mouse Genome Database of the Jackson Laboratory lists seven genes that are located within 4 c M of the imprinted mouse chromosome 11 loci and have orthologues on human chromosome 16 (but are not known to be imprinted): (1) N-methylpurine-DNA glycosylase (Mpg); (2) epidermal growth factor receptor, related sequence (Egfr-rs); (3) hemoglobin a, adult chain 1 (Hba-al); (4) hemoglobin a, adult chain 2 (Hba-a2); (5) hemoglobin X , a-like embryonic chain i n Hba complex (Hba-x); (6) proximal locus to the hemoglobin a chain complex (Phg); and (7) a globin regulatory element containing gene (Mare). The human orthologues of murine genes M p g , H b a - a l , Hba-a2, Hba-x, and Mare have been pinpointed to 16pl3.3, while that of murine Egfr-rs has been narrowed tol6pter-pl3. The human orthologue of Egfr-rs (Kielman et al. 1993; Kielman et al. 1996) is of particular interest. Egfr-rs is expressed in the preimplantation and early postimplantation mouse embryo, and may be implicated in the development of both the trophectoderm and I C M of the blastocyst (Hendrey et al. 1995). Additional evidence for imprinting at 16pl3.3 comes from Wyszynski and Panhuysen (1999), who studied recurrent alcoholism in families and found paternal effects at two markers located at 16pl3.3 (D16S475 and D16S2622). The region orthologous to human chromosome 16pl3.3 is only 4 c M from the  55  imprinted gene U 2 a f l - r s l in the mouse. This clustering may indicate a functional interaction that produces imprinting in the orthologous region. For example, two regions in humans where imprinted genes are known to be clustered and have functional interactions are 1 l p l 5 and 1 5 q l l - q l 3 (Morison and Reeve 1998). Tilghman (1999) reviewed models that may account for the short-range (IGF2 and H 1 9 i n l l p l 5 ) and the long-range ( S N R P N and genes up to 1 M b away in 15ql l - q l 3 ) interactions between imprinted genes in these clusters. Thus, it appears that a good candidate region for imprinted gene(s) on chromosome 16 resides in p l 3 . 3 . The results of this study raise the question of whether testing for U P D is indicated after prenatal diagnosis of C P M 16. In this regard it is important to emphasize that although upd(16)mat cases are on average more growth restricted than those with bpd(16), there is substantial overlap in the distribution of birth weights (Figure 3.11). Moreover, although upd(16)mat fetuses may be at higher risk for major malformation, most observed malformations are not life threatening and there is no evidence for higher rates of intrauterine or neonatal death. Knowledge of U P D status is unlikely to change management of C P M 1 6 pregnancies, and thus prenatal testing is not recommended.  56  3.6 References Association of Clinical Cytogeneticists Working Party on Chorionic V i l l i in Prenatal Diagnosis (1994) Cytogenetic analysis of chorionic v i l l i for prenatal diagnosis: an A C C collaborative study of U . K . data. Prenat Diagn 14:363-379 Benn P (1998) Trisomy 16 and trisomy 16 Mosaicism: a review. A m J M e d Genet 79:121-133 Chandra S, Scott H , Dodds L , Watts C , Blight C , V a n Den H o f M (2003) Unexplained elevated maternal serum alpha-fetoprotein and/or human chorionic gonadotropin and the risk of adverse outcomes. A m J Obstet Gynecol 189:775-781 Dragani T A , Peissel B , Zanesi N , A l o i s i A , D a i Y , Kato M , Suzuki H , Nakashima I (2000) Mapping of melanoma modifier loci in R E T transgenic mice. Jpn J Cancer Res 91:11421147 Engel E (1980) A new genetic concept: uniparental disomy and its potential effect, isodisomy. A m J M e d Genet 6:137-143 Frendo J L , Guibourdenche J, Pidoux G , Vidaud M , Luton D , Giovangrandi Y , Porquet D , M u l l e r F , Evain-Brion D (2004) Trophoblast production of a weakly bioactive human chorionic gonadotropin in trisomy 21-affected pregnancy. J C l i n Endocrinol Metab 89:727-732 Hassold T, Quillen S D , Yamane J A (1983) Sex ratio in spontaneous abortions. A n n H u m Genet 47 Pt 1:39-47 Hassold T J , Jacobs P A (1984) Trisomy in man. A n n u Rev Genet 18:69-97 Hendrey J, L i n D , Dziadek M (1995) Developmental analysis of the Hba(th-J) mouse mutation: effects on mouse peri-implantation development and identification of two candidate genes. Dev B i o l 172:253-263 Hook E B , Cross P K , Mutton D E (1999) Female predominance (low sex ratio) i n 47,+21 mosaics. A m J M e d Genet 84:316-319 Huether C A , Martin R L , Stoppelman S M , D'Souza S, Bishop J K , Torfs C P , Lorey F , M a y K M , Hanna JS, Baird P A , K e l l y J C (1996) Sex ratios in fetuses and liveborn infants with autosomal aneuploidy. A m J M e d Genet 63:492-500 Kalousek D K , Barrett I (1994) Confined placental mosaicism and stillbirth. Pediatr Pathol 14:151-159 Karason A , Gudjonsson J E , Upmanyu R , Antonsdottir A A , Hauksson V B , Runasdottir E H , Jonsson H H , Gudbjartsson D F , Frigge M L , K o n g A , Stefansson K , Valdimarsson H , Gulcher J R (2003) A susceptibility gene for psoriatic arthritis maps to chromosome 16q: evidence for imprinting. A m J H u m Genet 72:125-131  57  Kielman M F , Smits R , D e v i T S , Fodde R , Bernini L F (1993) Homology of a 130-kb region enclosing the alpha-globin gene cluster, the alpha-locus controlling region, and two nonglobin genes in human and mouse. M a m m Genome 4:314-323 Kielman M F , Smits R, H o f I, Bernini L F (1996) Characterization and comparison of the human and mouse Distl/alpha-globin complex reveals a tightly packed multiple gene cluster containing differentially expressed transcription units. Genomics 32:341-351 Kotzot D (1999) Abnormal phenotypes in uniparental disomy (UPD): fundamental aspects and a critical review with bibliography of U P D other than 15. A m J M e d Genet 82:265-274 Kramer M S , Piatt R W , W e n S W , Joseph K S , A l l e n A , Abrahamowicz M , Blondel B , Breart G (2001) A new and improved population-based Canadian reference for birth weight for gestational age. Pediatrics 108:E35 Ledbetter D H , Engel E (1995) Uniparental disomy in humans: development of an imprinting map and its implications for prenatal diagnosis. H u m M o l Genet 4 Spec N o : 1757-1764 Lepage N , Chitayat D , Kingdom J, Huang T (2003) Association between second-trimester isolated high maternal serum maternal serum human chorionic gonadotropin levels and obstetric complications in singleton and twin pregnancies. A m J Obstet Gynecol 188:1354-1359 Morison EVI, Reeve A E (1998) A catalogue of imprinted genes and parent-of-origin effects in humans and animals. H u m M o l Genet 7:1599-1609 Morssink L P , Sikkema-Raddatz B , Beekhuis J R , De W o l f B T , Mantingh A (1996) Placental mosaicism is associated with unexplained second-trimester elevation of M S h C G levels, but not with elevation of M S A F P levels. Prenat Diagn 16:845-851 Robinson W P , Barrett IJ, Bernard L , Telenius A , Bernasconi F , Wilson R D , Best R G , HowardPeebles P N , Langlois S, Kalousek D K (1997) Meiotic origin of trisomy in confined placental mosaicism is correlated with presence of fetal uniparental disomy, high levels of trisomy in trophoblast, and increased risk of fetal intrauterine growth restriction. A m J H u m Genet 60:917-927 Searle A G , Peters J, L y o n M F , H a l l J G , Evans E P , Edwards J H , Buckle V J (1989) Chromosome maps of man and mouse. I V . A n n H u m Genet 53 ( Pt 2):89-140 Spence J E , Perciaccante R G , Greig G M , W i l l a r d H F , Ledbetter D H , Hejtmancik JF, Pollack M S , O'Brien W E , Beaudet A L (1988) Uniparental disomy as a mechanism for human genetic disease. A m J H u m Genet 42:217-226 Tilghman S M (1999) The sins of the fathers and mothers: genomic imprinting in mammalian development. C e l l 96:185-193 Wolstenholme J (1995) A n audit of trisomy 16 in man. Prenat Diagn 15:109-121  58  Wyszynski D F , Panhuysen C I (1999) Parental sex effect in families with alcoholism. Genet Epidemiol 17 Suppl 1:S409-413  59  Table 3 . 1 Malformations among live births with survival beyond the neonatal period Malformation  % (frequency) in study sample  % of neonates in general population  Significance  VSD  12%(ll )  ASD  10% (9 )  p< 0.0001; p = 0.005 p< 0.0001  Hypospadias  27% (7 )  0.448%; 5% [*] (Hoffman and Kaplan 2002) 0.106% (Hoffman and Kaplan 2002) 0.8% (Gallentine et al. 2001)  b  b  C  3  p < 0.0001  B i n o m i a l test: n = # of informative cases, k = frequency in study sample, p[expected] = % of neonates in general population. O f 90 live birth cases informative for malformation status. O f 26 male live birth cases informative for malformation status. See Appendix A I for full references. a  b  c  Table 3 . 2 Association between the presence of trisomy in amniotic fluid and malformation  Malformation No Yes Trisomy in amniotic fluid  >0% 0%  15 31  21 15  Total  36 46  Risk  58% 33%  Relative risk 1.79  Presence (>0%) and absence (0%) of trisomy i n amniotic fluid as assessed by amniocentesis. Fisher Exact test, n = 82, p = 0.017.  Table 3 . 3 Association between trisomy in amniotic fluid and trisomy in fetal tissues Trisomy in fetal tissues No Yes Trisomy in amniotic fluid  >0% 0%  21 23  7 1  Total  28 24  Risk  25% 4%  Relative risk 6.00  Presence (>0%) and absence (0%) of trisomy i n amniotic fluid as assessed by amniocentesis. Fisher Exact test, n = 52, p = 0.042.  Table 3 . 4 Association between upd(16)mat and malformation Malformation No Yes upd(16)mat bpd(16)  5 16  16 12  Total  21 28  Risk  76% 43%  Relative risk 1.78  Fisher Exact test, n = 49, p = 0.020.  60  Table 3.5 Associations between explanatory variables associated with birth weight  Amniotic fluid Ascertainment Ascertainment Ascertainment Sex of fetus Direct CVS Cultured CVS  >0% 0% Biased Unbiased Serum screen Unbiased Ultrasound Unbiased M F 100% <100% 100% <100%  bpd(16)  upd(16)mat  11 18 11 10 7 10 4 10 8 22 13 3 11 4  8 13 10 9 4 9 6 9 7 14 10 1 9 0  p = 0.61 p = 0.62 p = 0.42 p = 0.40 p = 0.42 p = 0.46 p = 0.13  Trisomy in amniotic fluid 0% >0% Ascertainment Ascertainment Ascertainment Sex of fetus Direct CVS Cultured CVS  Biased Unbiased Serum screen Unbiased Ultrasound Unbiased M F 100% <100% 100% <100%  13 24 7 24 6 24 12 36 22 3 27 4  17 9 14 9 3 9 13 21 3 1 2 1  p = 0.017 p = 0.005  p = 0.51 p = 0.15 p = 0.47 p = 0.39  Ascertainment Biased Unbiased Sex of fetus Direct CVS Cultured CVS  M F 100% <100% 100% <100%  11 25 9 5 18 5  10 25 12 1 8 1  p = 0.53 p = 0.098 p = 0.45  Ascertainment Serum screen Unbiased Sex of fetus Direct CVS Cultured CVS  M F 100% <100% 100% <100%  11 25 9 5 18 5  8 14 4 0 4 0  p = 0.43 p = 0.23 p = 0.42  Ascertainment Unbiased Ultrasound Sex of fetus Direct CVS Cultured CVS  M F 100% <100% 100% <100%  11 25 9 5 18 5  2 11 8 1 4 1  p = 0.25 p = 0.21 p = 0.72  Sex of the fetus Direct CVS  100% <100% 100% <100%  Cultured CVS  F  M  19  8  7  1  23  6  5  3  p = 0.32 p = 0.29  Direct CVS Cultured CVS  <100%  100%  1  25  5  4  100% <  1  0  0  %  p = 0.002  Significance values from Fisher Exact test, without correction for multiple comparisons.  Table 3.6 Association between malformation and neonatal death Neonatal death Yes No Malformation  Yes No  Risk  Total  0  Relative risk infinity  18% 0%  45 52  8  37 52  Fisher Exact test, n = 97, p = 0.001.  Table 3.7 Association between biased ascertainment and neonatal death Neonatal death Yes No Ascertainment  Biased Unbiased  35 36  7 1  Total 42 37  Risk 16.7% 2.7%  Relative risk 6.17  Fisher Exact test, n = 79, p = 0.043.  Table 3.8 Association between ascertainment by abnormal ultrasound and neonatal death. Neonatal death No Yes Ascertainment  U/S Unbiased  13  4  36  1  Total 17 37  Risk 24%  Relative risk 8.71  3%  " U / S " = ultrasound. Fisher Exact test, n = 97, p = 0.001.  62  Table 3.9 Association between ascertainment by abnormal SS, and neonatal death Neonatal death Yes No Ascertainment  SS Unbiased  22 36  3 1  Total  25 37  Risk  12% 3%  Relative risk 4.44  " S S " = serum screen. Fisher Exact test, n = 62, p = 0.18.  63  Table 3.10 Ascertainment and cardio-pulmonary malformations in neonatal deaths Case  GA  Ascertainment  Hsu et al. (1998) case 3 95.28  29wks  High MSAFP  38wks  AMA  Devi et al. (1992; 1993) Garber et al. (1994) case 2; Hsu et al. (1998) case 2 Sanchez et al. (1997)  33wks  High MSAFP (5.25 MoM) High MShCG  Abu-Amero et al. (1999) case 2; Vaughan et al. (1994) case 2; Moore et al. (1997) case 16  36wks  25wks  28wks  23wks: IUGR, SUA, dilated cerebral ventricles, placental haematoma  Cardio-pulmonary malformations  coarctation of aorta, VSD partial anomalous pulmonary venous return, large ASD, congenital tracheal narrowing pulmonary hypoplasia muscular VSD, large ASD, single coronary artery that arose form pulmonary artery trunk no malformation  24wks: IUGR, SUA, hypokinesia, enlarged cisterna magna, dilation of both cerebral laternal venticles, hypoplasia of corpus callosum no malformation; High MShCG persistent hyaline membrane formation (14-17 MoM) 17wks (following abnormal screen): IUGR, SUA Repeat scans at 20, 23, 25 wks showed appropriate growth  Watson et al. (1988)  35wks  Referred from M  32-33 wks  28wks: no growth for 3 wks AMA Oligohydramnios 26wks: presented with vaginal bleeding and lower abdominal cramping; ultrasound showed agenesis of left kidney and probable atrioventricular canal  pulmonary hypoplasia DORV, posterior subpulmonic VSD, PDA, hypoplastic left ventricle, tubular hypoplasia of proximal aorta ending in a preductal coarctation, aberrant right subclavian artery, retroesophageal vascular ring, hypoplastic left pulmonary artery, ASD secundum, persistent left superior vena cava to coronary sinus, right ventricular hypertrophy, anterior descending artery arising from right coronary total anomalous pulmonary venous return to ductus venosus, pulmonary hypoplasia with left sided diaphragmatic hernia  " A M A " = advanced maternal age; " M S A F P " = maternal serum alpha-fetoprotein; " M S h C G ' maternal serum human chorionic gonadotropin; " S U A " = single umbilical artery.  Figure 3.1 Distribution of gestational ages for CPM16 live births  12-  81  Count  30  32  34  36  38  Gestational age (weeks)  40  42  Figure 3.2 Distribution of birth weights for CPM16 resulting in live births  Birth weight  Figure 3.3 Distribution of the level of trisomy in amniotic fluid among CPM16 live births  Count  20  30  40  50  60  70  r  80  % trisomy in amniotic fluid  Excluding 49 cases with 0% trisomy and 8 cases positive for trisomy but for which the percentage was not known).  67  Figure 3.4 Mean birth weight in the presence or absence of trisomy in amniotic fluid  -1.00'  BW  •2.00 - !  -3.00'  0%  >0%  Level of trisomy on amniocentesis  Mean birth weight ( B W ) ± standard deviation. The standard deviations were 1.02 and 1.00, respectively. The sample sizes were 41 and 23, respectively, t = 3.39, df = 62, p = 0.0005.  68  Figure 3.5 Mean birth weight for unbiased and biased ascertainment.  -1.0CH  TP-45 BW -2.00-  #•22  -3.00"  Biased  Unbiased Ascertainment  Mean birth weight ( B W ) ± standard deviation. The standard deviations were 1.30 and 0.59, respectively. The sample sizes were 27 and 30, respectively. Welch's approximate t = 2.85, df = 35.4, p = 0.0035.  69  Figure 3.6 Mean birth weight for ascertainment: unbiased vs. abnormal serum screen  -1.001  BW -2.00"  -3.00-  Unbiased  Serum screen Ascertainment  Mean birth weight ( B W ) ± 1 standard deviation. The standard deviations were 1.30 and 0.72, respectively. The sample sizes were 27 and 18, respectively. Welch's approximate t = 2.63, df = 41.9, p = 0.006.  70  Figure 3.7 Mean birth weight for ascertainment: unbiased vs. abnormal ultrasound  -1.00H  BW -2.00-  -3.00"  Unbiased  Ultrasound  Mean birth weight ( B W ) ± standard deviation. The standard deviations were 1.30 and 0.32, respectively. The sample sizes were 27 and 12, respectively. Welch's approximate t = 2.80, df = 32.0, p = 0.005.  71  Figure 3.8 Mean birth weight for females and males  -1.00"  B W  -2.00"  -3.00-  female  male  Mean birth weight ( B W ) ± standard deviation. The standard deviations were 0.99 and 1.15, respectively. The sample were 48 and 24, respectively, t = 2.35, df = 70, p = 0.011.  72  Figure 3 . 9  Mean birth weight for < 1 0 0 % and 1 0 0 % trisomy on direct C V S  1.00H  0.00 H  BW  -1.00H  -2.00-  < 1 0 0 %  1 0 0 %  trisomy  trisomy  Direct C V S  Mean birth weight ( B W ) ± standard deviation. The standard deviations were 1.05 and 0.86, respectively. The sample sizes were 3 and 26, respectively, t = 4.08, df = 27, p < 0.001.  73  Figure 3.10 Mean birth weight for <100% and 100% trisomy on cultured C V S  0.00 H  -1.00-i B W  -2.00-1  <100% trisomy  100% trisomy Cultured C V S  Mean birth weight ( B W ) ± standard deviation. The standard deviations were 1.24 and 1.02, respectively. The sample sizes were 4 and 25, respectively, t = 1.89, df = 27, p = 0.035.  74  Figure 3.11 Mean birth weight for bpd(16) and upd(16)mat  -1.00  1  BW -2.00'  -3.00'  bpd(16)  upd(16)mat  Mean birth weight ( B W ) ± standard deviation. The standard deviations were 1.08 and 1.04, respectively. The sample sizes were 25 and 19, respectively, t = 2.10, df = 42, p = 0.021.  75  Figure 3.12 Mean gestational age for unbiased and biased ascertainment  40.0  H  38.0'  Weeks  36.0 •  34.0-1  32.0'  Unbiased  Biased Ascertainment  Mean gestational age ± standard deviation. The standard deviations were 2.8 and 3.4, respectively. The sample sizes were 27 and 33, respectively, t = 1.92, df = 58, p = 0.030.  76  Figure 3.13 Mean gestational age ascertainment: unbiased vs. abnormal ultrasound  38.0'  36.0'  Weeks  34.01  32.0 •  Unbiased  Ultrasound Ascertainment  Mean gestational age ± standard deviation. The standard deviations were 2.8 and 2.1, respectively. The sample sizes were 27 and 12, respectively, t = 2.14, df = 37, p = 0.020.  Figure 3.14  Birth weights for live births and intrauterine deaths  1.00H  o.oo " i  -1.00BW -2.00 H  -3.00-  -4.00 H  Live birth  Intrauterine death  The mean birth weights ( B W ) ± standard deviation were -1.76±1.08 and - 2 . 9 9 ± 1 . 0 5 , respectively. The sample sizes were 72 and 6, respectively, t = 2.67, df = 76, p = 0.005.  Figure 3.15 Gestational ages for live births and neonatal deaths  Weeks  Live birth  Neonatal death  The mean gestational ages ± standard deviation were 36.0±3.2 and 32.0±4.4). The sample sizes were 77 and 8, respectively, t = 3.25, df = 83, p = 0.001.  79  4 Preeclampsia and CPM16  7  4.1  Note  I wrote this chapter/manuscript and did the data organization and analysis, with the following clarifications and exceptions. The C P M 16 cases in this study are from the same ongoing U B C study as previously described. For those C P M 16 cases ascertained locally, I reviewed Medical Genetics and B C Women's Hospital medical records for additional maternal clinical data under the supervision of Dr. S. Langlois (Medical Genetics) and Dr. P. von Dadelszen (Obstetrics and Gynaecology). A s well, I collected matched controls from B C Women's Hospital medical records under the supervision of Dr. P. von Dadelszen. For referred C P M 16 cases, maternal clinical data were collected by D r . D . Kalousek and Dr. W . Robinson.  4.2  Introduction  Preeclampsia is a maternal syndrome during pregnancy characterized by hypertension, decreased systemic organ perfusion related to vasoconstriction due to increased sensitivity to vasopressors, loss of endothelial integrity resulting in fluid leak, and intravascular coagulation (Roberts and Lain 2002). It is the number one cause of maternal mortality in developed countries and increases perinatal mortality five-fold (Roberts and Lain 2002). The only 'cure' is delivery of the placenta, indicating this maternal syndrome has a placental origin. The incidence of preeclampsia is approximately 3-5% (Roberts and Cooper 2001), depending on the definition and the population. There are numerous risk factors for preeclampsia, including nulliparity, advanced maternal age, previous pregnancy with preeclampsia, family history of preeclampsia, and increased placental mass/demand (e.g. hydatidiform mole and multi-fetal pregnancy) (Dekker 1999). Preeclampsia has been classically associated with abnormal placentation; i n A version of this chapter will be submitted for publication. Yong PJ, von Dadelszen P, Langlois S, Barrett IJ, Kalousek D K , Robinson WP. Preeclampsia and confined placental mosaicism for trisomy 16.  7  80  particular, poor invasion and remodeling of the uterine spiral arteries by the extravillus trophoblast ( E V T ) (see Chapter 7). However, similar poor invasion and remodeling have been seen in intrauterine growth restriction (IUGR) without preeclampsia (Khong et al. 1986), placental abruption (Dommisse and Tiltman 1992), and even in pre-term premature rupture of membranes ( p P R O M ) ( K i m et al. 2002) and in pre-term labour with intact membranes ( K i m et al. 2003). Thus, poor placentation does not appear to be sufficient to cause preeclampsia. There are likely other maternal and feto-placental factors that contribute to the risk of developing the syndrome. It may well be that some cases of preeclampsia have a primarily maternal cause, other cases a primarily feto-placental cause, and still others a combination of maternal and fetoplacental causes (Cross 2003). A n i m a l models have shown that a purely maternal and a purely feto-placental initiating event can cause preeclampsia, suggesting that the human syndrome may consist of distinct pathogenic subtypes (Cross 2003). In 1987, a higher risk of preeclampsia was reported in a series of trisomy 13 pregnancies compared to trisomy 18, trisomy 21 and control pregnancies (Boyd et al. 1987). Subsequently, there have been a number of reports of preeclampsia in trisomy 13 pregnancies (Feinberg et al. 1991; B o y d et al. 1995; Heydanus et al. 1995), a series of trisomy 13 pregnancies showing the same association (Bower et al. 1987), and a large case-control study that confirmed the findings of B o y d and colleagues (Tuohy and James 1992). Estimates for the incidence of preeclampsia in trisomy 13 pregnancies (that do not miscarry or are not terminated) range from 22% (Bower et al. 1987) to 44% (Tuohy and James 1992). B o y d et al. (1987) found no association between preeclampsia and trisomies 18 and 21, and Tuohy and James (1992) found no association with trisomy 18. A decreased risk of preeclampsia has been found in trisomy 21 (Zhang et al. 2004). These findings suggest a chromosome 13-specific effect in preeclampsia. I hypothesized that C P M 16 would be significantly associated with a higher risk of preeclampsia, and that the risk would correlate with U P D and placental trisomy. Thus, C P M 16 81  cases were reviewed to determine the frequency of preeclampsia, and to evaluate factors contributing to the risk of preeclampsia. This w i l l improve genetic counseling after prenatal diagnosis of C P M of trisomy 16 ( C P M 16) and add insight into the clinical significance of trisomy 16 confined to the placenta.  4.3  Methods 4.3.1  C P M 1 6 cases  C P M 16 cases from the ongoing study of trisomy mosaicism at the University of British Columbia ( U B C ) , involving cases ascertained locally and referred from other centres, were included in this study i f there was (1) prenatal diagnosis of trisomy 16 v i a C V S or amniocentesis; (2) continuation of the pregnancy past 20 weeks gestation (because preeclampsia rarely occurs before then); (3) sufficient clinical data to include or exclude a diagnosis of preeclampsia; and (4) the availability of D N A for molecular determination of U P D status of chromosome 16 in the fetus (i.e. bpd(16) or upd(16)mat). Exclusion criteria were confirmed paternal origin of the trisomy 16 or concomitant aneuploidy or polyploidy in addition to trisomy 16, since such cases are rare and would confound the analysis. The study was approved by the ethics committees of U B C and the Children's and Women's Health Centre of British Columbia (Appendix A ) . A s research into C P M has generally focused on fetal outcomes, clinical data for preeclampsia varied between cases. For example, in the local cases blood pressure levels and other pertinent findings were available from medical records to independently make or exclude a diagnosis of preeclampsia. In referred cases, data were generally provided by the referring party completing research study forms. These forms asked the referrer to list any pregnancy complications, or to indicate 'yes' or 'no' to a list of pregnancy complications such as hypertension, and then to elaborate on clinical details. For the subset of cases where the post-  82  partum placenta was available, the level of trisomy in the chorionic plate, chorionic villus mesenchyme, and trophoblast was determined as described in Chapter 2.  4.3.2  Matched controls  For each C P M 16 case, 2 matched controls were ascertained from British Columbia's Women's Hospital ( B C W H ) delivery records according to the following criteria: same date of delivery or closest consecutive or previous date of delivery, matched for maternal age (± 5 years) and parity (0, 1 or >2). Gestational age at delivery and sex of the infant were also noted. Then, the corresponding B C W H medical records for each control were reviewed for data regarding blood pressure and other clinical features of preeclampsia.  4.3.3  Definition of preeclampsia  Preeclampsia was defined by the criteria for clinical diagnosis in the 2000 guidelines of the Australasian Society for the Study of Hypertension in Pregnancy (Brown et al. 2000b; Brown et al. 2000a). Briefly, a diagnosis of preeclampsia was made in the presence of hypertension (>140 m m H g systolic or >90 m m H g diastolic), plus one additional feature of proteinuria, low platelets or elevated liver enzymes, but not edema or I U G R . The Australasian guidelines do include I U G R , but it is not useful as a criterion in C P M 16 pregnancies which tend to be growth restricted (Yong et al. 2003).  4.3.4  Statistical analysis  Statistical analyses were performed using SPSS 10.0 and the VassarStats Web Site for Statistical Computation (http://faculty.vassar.edu/lowry/VassarStats.html). Unless otherwise noted, p-values are 1-tailed due to a priori evidence or rational mechanisms.  83  4.4  Results  Nineteen C P M 16 cases met the study criteria (Tables 4.1, 4.2,4.3). Twenty-six percent (5/19) of the C P M 1 6 pregnancies were preeclamptic (Table 4.3), while 5% (2/38) of matched controls were preeclamptic (Fisher Exact test, p = 0.035; R R = 5.00). Table 4.4 compares possible confounders between the C P M 1 6 cases and the matched controls: parity, maternal age, gestational age at delivery, and sex of the infant. Only gestational age was significantly different because pre-term delivery is common in C P M 1 6 pregnancies. However, this factor would result in a difference that is conservative because the longer gestational ages among controls should allow near-term preeclampsia to develop, while C P M 16 pregnancies may be delivered for other indications (e.g. I U G R , fetal distress) before clinical features of preeclampsia appear. The C P M 16 cases were tested for associations between preeclampsia and U P D status of chromosome 16 or level of trisomy in placental lineages post-partum (Table 4.3). The frequency of upd(16)mat was similar among preeclamptic and non-preeclamptic cases 60% (3/5) and 54% (7/13), respectively. The preeclamptic cases all had trends towards higher levels of trisomic cells in the three placental lineages compared to non-preeclamptic cases, although sample size was small and none reached statistical significance: chorionic plate (mean = 98.3% vs. 51.8%; Welch's approximate t = 1.89, df = 3.0, p = 0.08), villus mesenchyme (73.0% vs. 33.3%; t = 1.01, df = 5, p = 0.18), and trophoblast (75.7% vs. 63.4%; t = 0.50, df = 6, p = 0.32). Table 4.3 shows that higher levels of trisomy across the placental lineages were seen in both the preeclamptic C P M 1 6 cases (with the exception of the villus mesenchyme of 92.95) and in some non-preeclamptic C P M 1 6 cases (e.g. 91.71 and 92.25), while lower levels of trisomy across the lineages were only seen i n non-preeclamptic cases (e.g. 91.55 and 91.10). The C P M 1 6 cases with preeclampsia had a wide range of clinical findings (Table 4.5). Three of the 4 preeclamptic male C P M 16 cases had hypospadias, in contrast to 1 of the 3 non84  preeclamptic male C P M 1 6 cases. A m o n g the other male C P M 1 6 live births in the ongoing U B C study of trisomy mosaicism,' which did not meet this study's criteria, 18% (3/19) had hypospadias (Yong et al. 2003). The frequency of hypospadias in the preeclamptic male C P M 16 cases (3/4) was significantly increased when compared to all the other male C P M 16 cases in the U B C study (4/22) (p = 0.047, R R = 4.13).  4.5  Discussion  A s hypothesized, the risk of preeclampsia in C P M 16 pregnancies (26%) was increased compared to controls and should be considered during management of pregnancies after prenatal diagnosis of C P M 16. This association indicates a chromosome 16-specific effect on risk of preeclampsia, in addition to the known chromosome 13-specific effect. One possible cause of the association between C P M 16 and preeclampsia is an increase in placental size, thereby increasing the inflammatory load. However, C P M 16 placentas were smaller when compared to a reference population (Chapter 2). B o y d et al. (1987) also excluded placental weight as an etiological factor in trisomy 13. Consistent with the hypothesis, preeclamptic C P M 1 6 cases showed a trend towards higher levels of trisomic cells in the placenta (chorionic plate, villus mesenchyme and trophoblast). Specifically, both the preeclamptic C P M 16 cases and some non-preeclamptic C P M 16 cases had higher levels of trisomy i n the three placental lineages, while low levels of trisomy were seen only in the non-preeclamptic cases. Therefore, placental trisomy could predispose to preeclampsia, but requires other maternal-fetal factors for development of the full syndrome. A possible mechanism for this predisposition to preeclampsia in C P M 16 pregnancies with high levels of placental trisomy is abnormal extravillus trophoblast ( E V T ) growth and function. E V T outgrowth may be decreased in trisomy 16 (Chapter 7); however, E V T defects have also been identified in trisomy 21 (Wright et al. 2004), which is associated 85  with reduced risk of preeclampsia (Zhang et al. 2004). It should be noted that the risk of preeclampsia was not increased in upd(16)mat, suggesting that no imprinted genes on chromosome 16 are related to preeclampsia in C P M 16 pregnancies. Three of the 4 male C P M 16 cases with preeclampsia had hypospadias. Since hypospadias has consistently been found in male survivors of homozygous oc-thalassemia (Abuelo et al. 1997; Dame et al. 1999; Fung et al. 1999) it has been proposed that a gene associated with hypospadias exists on 16pl3.3 that is affected by the nearby oc-globin gene deletions (Dame et al. 1999). The expression of this putative gene may be abnormal in trisomy 16, predisposing to hypospadias. Alternatively, homozygous a-thalassemia survivors may have hypospadias because of hypoxia secondary to the high affinity of Bart's hemoglobin (Fung et al. 1999). Preeclampsia is associated with utero-placental hypoxia (Roberts and Lain 2002), which could thus also predispose to hypospadias. In a recent study of 8,894 cases in a malformation database, an association between gestational hypertension (including preeclampsia) and presence of malformation was found, which on multivariate analysis proved to be due specifically to hypospadias, other anomalies of the penis, and 'multiple congenital abnormalities' (syndromes affecting the face, limbs, and stature) (Vesce et al. 1997). In conclusion, C P M 16 is at higher-risk for preeclampsia. Future studies should assess the incidence of C P M 16 in a series of idiopathic preeclamptic pregnancies. In particular, trisomy C P M may be implicated in preeclampsia that is pre-term, multi-systemic in presentation, and/or associated with hypospadias.  86  4.6  References  Abuelo D N , Forman E N , Rubin L P (1997) L i m b defects and congenital anomalies of the genitalia in an infant with homozygous alpha-thalassemia. A m J M e d Genet 68:158-161 Bower C , Stanley F , Walters B N (1987) Pre-eclampsia and trisomy 13. Lancet 2:1032 B o y d P A , Lindenbaum R H , Redman C (1987) Pre-eclampsia and trisomy 13: a possible association. Lancet 2:425-427 B o y d P A , Maher E J , Lindenbaum R H , Hoogwerf A M , Redman C , Crocker M (1995) Maternal 3; 13 chromosome insertion, with severe pre-eclampsia. C l i n Genet 47:17-21 Brown M A , Hague W M , Higgins J, Lowe S, M c C o w a n L , Oats J, Peek M J , Rowan J A , Walters B N (2000a) The detection, investigation and management of hypertension in pregnancy: executive summary. Aust N Z J Obstet Gynaecol 40:133-138 Brown M A , Hague W M , Higgins J, L o w e S, M c C o w a n L , Oats J, Peek M J , Rowan J A , Walters B N (2000b) The detection, investigation and management of hypertension in pregnancy: full consensus statement. Aust N Z J Obstet Gynaecol 40:139-155 Cross J C (2003) The genetics of pre-eclampsia: a feto-placental or maternal problem? C l i n Genet 64:96-103 Dame C , Albers N , Hasan C , Bode U , Eigel A , Hansmann M , Brenner R, Bartmann P (1999) Homozygous alpha-thalassaemia and hypospadias—common aetiology or incidental association? Long-term survival of H b Bart's hydrops syndrome leads to new aspects for counselling of alpha-thalassaemic traits. Eur J Pediatr 158:217-220 Dekker G A (1999) Risk factors for preeclampsia. C l i n Obstet Gynecol 42:422-435 Dornmisse J, Tiltman A J (1992) Placental bed biopsies in placental abruption. B r J Obstet Gynaecol 99:651-654 Feinberg R F , Kliman H J , Cohen A W (1991) Preeclampsia, trisomy 13, and the placental bed. Obstet Gynecol 78:505-508 Fung T Y , K i n L T , K o n g L C , Keung L C (1999) Homozygous alpha-thalassemia associated with hypospadias in three survivors. A m J M e d Genet 82:225-227 Heydanus R, Defoort P, Dhont M (1995) Pre-eclampsia and trisomy 13. Eur J Obstet Gynecol Reprod B i o l 60:201-202 K h o n g T Y , De W o l f F, Robertson W B , Brosens I (1986) Inadequate maternal vascular response to placentation in pregnancies complicated by pre-eclampsia and by small-forgestational age infants. B r J Obstet Gynaecol 93:1049-1059  87  K i m Y M , Bujold E , Chaiworapongsa T, Gomez R , Y o o n B H , Thaler H T , Rotmensch S, Romero R (2003) Failure of physiologic transformation of the spiral arteries in patients with preterm labor and intact membranes. A m J Obstet Gynecol 189:1063-1069 K i m Y M , Chaiworapongsa T, Gomez R , Bujold E , Y o o n B H , Rotmensch S, Thaler H T , Romero R (2002) Failure of physiologic transformation of the spiral arteries in the placental bed in preterm premature rupture of membranes. A m J Obstet Gynecol 187:1137-1142 Roberts J M , Cooper D W (2001) Pathogenesis and genetics of pre-eclampsia. Lancet 357:53-56 Roberts J M , Lain K Y (2002) Recent Insights into the pathogenesis of pre-eclampsia. Placenta 23:359-372 ^ Tuohy JF, James D K (1992) Pre-eclampsia and trisomy 13. B r J Obstet Gynaecol 99:891-894 Vesce F , Farina A , Giorgetti M , Jorizzo G , Bianciotto A , Calabrese O, M o l l i c a G (1997) Increased incidence of preeclampsia in pregnancies complicated by fetal malformation. Gynecol Obstet Invest 44:107-111 Wright A , Zhou Y , Weier JF, Caceres E , Kapidzic M , Tabata T, Kahn M , Nash C , Fisher SJ (2004) Trisomy 21 is associated with variable defects in cytotrophoblast differentiation along the invasive pathway. A m J M e d Genet A 130:354-364 Y o n g P J , Barrett IJ, Kalousek D K , Robinson W P (2003) Clinical aspects, prenatal diagnosis, and pathogenesis of trisomy 16 mosaicism. J M e d Genet 40:175-182 Zhang J, Christianson R E , Torfs C P (2004) Fetal trisomy 21 and maternal preeclampsia. Epidemiology 15:195-201  88  Table 4.1 CPM16 cases meeting inclusion criteria  PE Y  Case 91.14  Source Local  Y  92.48  Local  Y  93.94  Referred  Y Y  94.116 92.95  Referred Local  N  91.55  Referred  N  91.71  Referred  N  92.25  Referred  N  93.103  Referred  N  93.73  Referred  N N  93.93 95.28  Referred Referred  N N N N N N N  98.234 99.143 91.10 93.132 16-51 16-55 16-54  Referred Local Local Referred Referred Referred Referred  References Kalousek et al. (1993) case 2 Robinson et al. (1997) Penaherrera et al. (2000) Kalousek et al. (1993) case 8 Robinson et al. (1997) Schneider et al. (1996) Robinson et al. (1997) Robinson et al. (1997) Robinson et al. (1997) Hsu et al. (1997) case XIV-11 Kalousek et al. (1993) case 7 Robinson etal. (1997) Penaherrera et al. (2000) Kalousek et al. (1993) case 3 Johnson et al. (1993) case C Robinson et al. (1997) Penaherrera et al. (2000) Kalousek et al. (1993) case 4 Robinson et al. (1997) Schwinger et al. (1989) case 5 Wolstenholme (1995) Robinson et al. (1997) Verpet al. (1989) Penaherrera et al. (2000) Kennerknecht and Terinde (1990) Robinson et al. (1997) Penaherrera et al. (2000) Penaherrera et al. (2000) Penaherrera et al. (2000) Kalousek et al. (1993) case 6 Woo et al. (1997) Unpublished Unpublished Unpublished  Origin Kalousek et al. (1993) Unpublished Unpublished Schneider et al. (1996) Unpublished Unpublished Unpublished Unpublished  Unpublished Unpublished Unpublished Unpublished Unpublished Unpublished Unpublished Unpublished Woo et al. (1997) Unpublished Unpublished Unpublished  " P E " = preeclampsia. "References" = some data previously published in these references. "Origin" = origin of data regarding preeclampsia. F u l l references in Appendix A I .  Table 4.2 CPM16 cases meeting inclusion criteria PE  Case  Paritv  M a t age  Outcome  Sex  GA  Y Y Y Y Y N N N N N N N N N N N N N N  91.14 92.48 93.94 94.116 92.95 91.55 91.71 92.25 93.103 93.73 93.93 95.28 98.234 99.143 91.10 93.132 16-51 16-55 16-54  1 0 1 0 1 1 1 1 3 1 0 1 0 1 0 1 1 0 0  27 40 34 41 36 35 37 34 39 35 21 41 34 29 44 42 36 31 33  TA LB LB LB LB LB LB LB LB LB LB ND LB TA LB LB LB LB LB  F M M M M F F M M F M F F F F F F F F  25 36 39 33 36 40 38 37 37 37 37 38 38 22 36 33 35 34 37  " P E " = preeclampsia. " T A " = termination of pregnancy; " L B " = live birth with survival beyond the neonatal period; " N D " = neonatal death. " G A " = gestational age in weeks.  90  Table 4 . 3 . Post-partum placenta cytogenetics PE  Case  BW  Malf  UPD  Y Y Y Y Y N N N N N N N N N N N N N N  91.14 92.48 93.94 94.116 92.95 91.55 91.71 92.25 93.103 93.73 93.93 95.28 98.234 99.143 91.10 93.132 16-51 16-55 16-54  410g 2280g 1935g 1498g 1559g 3319g 1960g 1650g 2750g 2315g 2210g 2000g 2578g ? 2660g 1620g 1530g 1278g ?  N Y Y N Y N N Y N N Y Y Y Y N Y Y Y Y  UPD BPD UPD BPD UPD BPD BPD UPD UPD BPD BPD UPD BPD BPD BPD UPD UPD UPD UPD  Post-partum placenta cytogenetics Trophoblast Chorion Mesenchyme 100% 100% 95% 0% 87% 100% 20% -  100% 100% 88% 4% 0% 100% 0% -  70% 61% 96% 0% 100% 85% 54% 78% -  " B W " = birth weight. " M a l f = malformation. " U P D " = uniparental disomy for chromosome 16 (upd(16)mat); " B P D " = biparental disomy for chromosome 16 (bpd(16)). Percentages are averaged over sites sampled. ' ? ' = not available.  91  Table 4.4 Clinical features of the CPM16 cases and controls  Preeclampsia Parity Maternal age Gestational age Sex ratio  C P M (n = 19)  Controls (n = 38)  Sig  26% (5/19) 0.74 + 0.74 35.2 ± 5.6 yrs 35.2 ±4.6 wks 0.58 (7 M; 12 F)  5% (2/38) 0.71 ±0.65 33.9 ±5.1 yrs 39.8 ± 1.3 wks 0.65 (15 M; 23F)  p = 0.035 n.s. n.s. p < 0.001 n.s.  Means ± standard deviations are provided for parity, maternal age, and gestational age. Statistics: Fisher Exact test (pre-eclampsia, sex ratio), 2-sample t-test (maternal age), Welch's approximate t-test (gestational age), Mann-Whitney test (parity).  Table 4.5 Clinical features of the CPM16 cases with preeclampsia CPM16 case 91.14  92.48  93.94  Clinical features of preeclampsia  Severe hypertension and proteinuria at approximately 21 weeks gestation. BP ranged to 180/130 with +4 to +5 proteinuria, but only mild headaches, no hyperreflexia and mild edema. Hypertension did not respond to antihypertensive therapy and decision made to terminate pregnancy at 25 weeks. Eumorphic female fetus. Uncomplicated until 35 weeks, except for peripheral edema 2 months prior. Blood pressure ranged from to 150/100, with low platelets and elevated liver enzymes, but no proteinuria. Male fetus delivered at 36 weeks gestation with hypospadias Intermittent hypertension, proteinuria and edema from 35 to 39 weeks. Male fetus delivered at 39 weeks with hypospadias, left hydronephrosis, and 5 finger clinodactyly. Severe preeclampsia and HELLP syndrome. Date of onset not provided. Eumorphic male fetus delivered at 33 weeks. 'Toxemia' and low platelets. Date of onset not provided. Male fetus delivered at 36 weeks with hypospadias. th  94.116 92.95  5 Postnatal follow-up of newborns from CPM16 pregnancies  1  5.1  Note  I wrote this chapter/manuscript and did the data organization and analysis, with the following clarifications and exceptions. These C P M 16 cases are from the ongoing U B C study of trisomy mosaicism described previously. Long-term follow-up data have been collected by Dr. D . Kalousek, Dr. S. Langlois, and Dr. W . Robinson.  5.2  Introduction  Genetic counselling after prenatal diagnosis of C P M is difficult in part because there has been no systematic study of long-term outcome of neonates from such pregnancies. I hypothesized that neonates from C P M 16 pregnancies w i l l demonstrate good catch-up growth and developmental outcome, since the trisomy is completely or predominantly confined to the placenta. In this Chapter, the postnatal growth and development of C P M 16 cases at least one year after birth were described, and risk factors for abnormal growth or development were identified.  5.3  Methods 5.3.1  CPM16 cases  There were 27 cases of follow-up after prenatal diagnosis of C P M 16. Table 5.1 provides the origin of the postnatal data, either from published studies from other centres, published studies from our centre, and/or unpublished data from the ongoing U B C study of trisomy mosaicism. The inclusion criterion was prenatal diagnosis of trisomy 16 at C V S or A version of this chapter has been accepted for publication. Langlois S, Yong PJ, Yong S-L, Barrett IJ, Kalousek DK, Miny P, Exeler R, Morris K, Robinson WP (2006) Postnatal follow-up of prenatally diagnosed trisomy 16 mosaicism. Prenat Diagn. 8  93  amniocentesis, and the presence of postnatal follow-up data at one year of age or more for length/height, weight, and/or developmental outcome. The study was approved by the ethics committees of U B C and C & W (Appendix A ) . To qualify for analysis of catch-up growth in height or weight, the birth length or weight had to be small-for-gestational age ( S G A ) . S G A has several definitions; the definition of the International S G A Advisory Board was chosen, which is less than 2 standard deviations below the gestational age-corrected mean (~3 percentile) (Lee et al. 2003). Otherwise, the case was rd  denoted as 'not-applicable' ( N / A ) . Catch-up growth in length or weight was defined as a length or weight above 2 standard deviations below the mean at some point of postnatal life. Standard deviations from the mean for length and weight at birth were calculated manually by using gestational age-specific means and standard deviations (Usher and M c L e a n 1969). Standard deviations from the mean for postnatal length and weight were determined using the 2000 growth charts from the Centers for Disease Control ( C D C ) , which represent United States surveys from 1963 to 1994 (the most recent being the joint National Center for Health Statistics ( N C H S ) and C D C Third National Health and Nutrition Examination Survey ( N H A N E S ni)) (Ogden et al. 2002). Standard deviations were calculated using the C D C E p i Info software program, available for download at http://www.cdc.gov/epiinfo, which can calculate percentiles and z-scores (standard deviations from the mean) using the 2000 C D C growth charts. When doing so, corrections in postnatal age were made for the gestational age at delivery. In a few cases, only percentiles (and not raw data for height or weight) were provided by the referring source. For developmental outcome, postnatal age was corrected for gestational age at delivery. Most cases had an explicit statement from the referring source regarding 'normal' versus 'delayed' development. In one case (case 8), no explicit statement was made regarding speech development, but delay was assumed because the child was undergoing speech therapy.  For a 94  few cases no diagnostic statement was available regarding 'normal' or 'delayed' development, although the age at which the child reached certain milestones was provided. These children were evaluated using the Denver II Developmental Assessment. The following additional data were collected on the 27 cases (Table 5.2): 1) % trisomic cells in amniotic fluid at amniocentesis; 2) uniparental disomy ( U P D ) ; 3) malformations; 4) gestational age at delivery; 5) sex of the infant; and 6) ascertainment ('biased' = abnormal ultrasound or serum screen; 'unbiased' = advanced maternal age ( A M A ) , investigation of biochemical or Mendelian disorders, history of aneuploidy pregnancy, etc.).  5.3.2  Statistical analysis  Statistical analyses were performed using SPSS 10.0 and the VassarStats Web Site for Statistical Computation (http://faculty.vassar.edu/lowry/VassarStats.html). Unless otherwise noted, p-values are 1-tailed due to a priori evidence or rational mechanisms.  5 . 4 Results Table 5.2 shows the percent trisomy at amniocentesis, U P D status, malformations, gestational age at delivery, sex of the infant, and ascertainment. Table 5.3 describes the followup data for growth in length/height and weight, and for developmental outcome.  5.4.1  Height  There were 17 cases informative for height (Table 5.2). Ten of the cases (59%) had birth lengths above - 2 standard deviations (SD) from the mean, and therefore were not applicable to assessment of catch-up growth. Seven cases (41%) had birth lengths below -2 S D , of which all (100%) demonstrated catch-up growth in height.  95  5.4.2. Weight There were 18 cases informative for weight (Table 5.2). Nine of the cases (50%) had birth weights above - 2 S D , and therefore were not applicable to assessment of catch-up growth. Nine cases (50%) had birth lengths below - 2 S D , of which 7/9 (78%) demonstrated catch-up growth in weight. The 2 cases without catch-up growth were cases 13 and 16. It should be noted that case 13 had congenital hypothyroidism which may account for the lack of catch-up growth in weight, although the child did exhibit catch-up growth in height and normal development, suggesting adequate treatment.  5.4.3  Development  A l l 27 cases were informative for developmental outcome, of which 21/27 (78%) had normal development.  The 6 cases with evidence of developmental delay were cases 1, 3, 8, 12,  15, and 17. The presence of evidence of developmental delay was significantly associated with trisomy (>0%) in amniotic fluid by amniocentesis (Fisher Exact test, p = 0.038; Table 5.4). The risk of developmental delay when trisomy was present in amniotic fluid was 36% (5/14) and 0% (0/11) when trisomy was absent in amniotic fluid. Risk of developmental delay was significantly associated with presence of malformation (Fisher Exact test, p = 0.035; Table 5.5). The risk of developmental delay in the presence of malformation was 38% (6/16), and 0% (0/10) in the absence of malformation. Risk of developmental delay was significantly associated with lower birth weight (t = 1.76, df = 21, p = 0.047; Figure 5.1). The mean birth weight and standard deviation in cases with developmental delay was -2.66 ± 1.52 (n = 5) and i n cases with normal development -1.60 ± 1.10 (n = 18). Figure 1 shows that the birth weight distributions for cases with normal development and cases with delayed development are overlapping, indicating that birth weight is not a clinically useful predictor of developmental delay. 96  Multiple logistic regression was performed with developmental delay as the outcome variable, and trisomy in amniotic fluid and malformation as explanatory variables. (Birth weight could not be incorporated as an explanatory variable because it was not significantly associated with developmental delay at the 2-tailed level.) The regression results showed that the effects of trisomy in amniotic fluid and malformation on risk of developmental delay were independent of each other (data not shown). In order to assess the possible confounding of birth weight, bivariate associations were sought with trisomy in amniotic fluid and malformation. The presence of trisomy in amniotic fluid was associated with lower birth weight (t = 2.71, df = 19, p = 0.007), but the presence of malformation was not (t = 1.32, df = 20, p = 0.10). This suggests that the association between developmental delay and lower birth weight may be confounded by the presence of trisomy in amniotic fluid (or vice versa). There were no statistically significant associations between developmental delay, and the level of trisomy above 0% at amniocentesis, U P D status, sex of the infant, method of ascertainment, or gestational age at delivery.  5.5  Discussion  The findings in this study support the hypothesis of generally positive long-term outcome after prenatal diagnosis of C P M 1 6 . O f S G A infants from C P M 1 6 pregnancies, the majority showed catch-up growth in height (100%) and weight (78%). Therefore, the prognosis for postnatal growth in infants with severe growth restriction related to C P M 16 appears to be quite good. The majority of infants (78%) also showed normal developmental outcome. The association between developmental delay and the presence of trisomy on amniocentesis suggests that amniotic fluid cells are an indicator of low-level mosaicism in the infant with implications for development. Although this association was confounded by birth weight, it is likely that mosaicism in the fetus/infant accounts for both developmental delay and lower birth weight 97  (Chapter 3); however, it cannot be ruled out that low birth weight itself may contribute to developmental delay, for example secondary to neonatal complications (Gutbrod et al. 2000). The association between developmental delay and malformation suggests that the presence of fetal malformation may also be indicator of low-level mosaicism in the infant. O f the 5/6 infants with developmental delay that had malformations, 3 of the 5 had heart malformations (Tables 5.2 and 5.3): case 1 ( V S D ) , case 8 ( A S D and 'other' cardiac anomalies), and case 15 ( D O R V , P D A , endocardial cushion defect). Although the developmental outcome of infants with congenital heart disease ( C H D ) is usually normal, those requiring open heart surgery may be at increased risk for delay (Limperopoulos et al. 2002). Cases 8 and 15 did undergo surgery for heart anomalies. In case 15, cardiac surgery occurred within the first year of life, and therefore before the onset of speech delay; in case 8, the onset of developmental delay was not available. It should also be emphasized that neither the level of trisomy above 0%, uniparental disomy (UPD), nor gestational age at delivery was associated with developmental delay. Multiple logistic regression results suggested that i f malformation is a marker of lowlevel trisomy in the infant, then it is a different and independent marker compared to the presence of trisomy in amniotic fluid by amniocentesis. Interestingly, Greally et al. (1996) reported that a C P M 16 infant with a hypoplastic aortic isthmus was found to have trisomic cells by F I S H on aortic tissue sampled during surgery. However, another case with A S D , V S D , and single coronary artery arising from the pulmonary trunk was found to have no trisomic cells i n affected heart tissue (case 2 from Garber et al. (1994)). Conversely, a case was reported to have 11% trisomic cells by F I S H in heart tissue, yet the heart was structurally normal (Johnson et al. 2000). In addition, lower birth weight (corrected for gestational age) was associated with developmental delay. The significant association between trisomy in amniotic fluid and birth weight suggests that the birth weight and development delay association may be confounded.  98  The postnatal prognosis of infants after prenatal diagnosis of C P M 16 is generally optimistic. Most demonstrated catch-up growth. The majority (78%) had normal development, and significantly, all 10 cases with no evidence of trisomy on amniocentesis developed normally. These findings are reassuring and w i l l be useful for genetic counselling after C P M 16 prenatal diagnosis.  99  5.6  References  D e v i A S , Velinov M , Kamath M V , Eisenfeld L , Neu R, Ciarleglio L , Greenstein R, Benn P (1993) Variable clinical expression of mosaic trisomy 16 in the newborn infant. A m J M e d Genet 47:294-298 Dorfmann A D , Perszyk J, Robinson P, Black S H , Schulman J D (1992) Rare non-mosaic trisomies in chorionic villus tissue not confirmed at amniocentesis. Prenat Diagn 12:899902 Garber A , Carlson D , Schreck R, Fischel-Ghodsian N , Hsu W T , Oeztas S, Pepkowitz S, Graham J M , Jr. (1994) Prenatal diagnosis and dysmorphic findings in mosaic trisomy 16. Prenat Diagn 14:257-266 Greally J M , Neiswanger K , Cummins J H , Boone L Y , Lenkey S G , Wenger S L , Lewis J L , Fischer D , Paul R A , Steele M W (1996) A molecular anatomical analysis of mosaic trisomy 16. H u m Genet 98:86-90 Gutbrod T, W o l k e D , Soehne B , Ohrt B , Riegel K (2000) Effects of gestation and birth weight on the growth and development of very low birthweight small for gestational age infants: a matched group comparison. A r c h Dis C h i l d Fetal Neonatal E d 82:F208-214 Hajianpour M J (1995) Postnatally confirmed trisomy 16 mosaicism: follow-up on a previously reported patient. Prenat Diagn 15:877-879 Hsu L Y , Y u M T , Neu R L , V a n Dyke D L , Benn P A , Bradshaw C L , Shaffer L G , Higgins R R , Khodr G S , Morton C C , Wang H , Brothman A R , Chadwick D , Disteche C M , Jenkins L S , Kalousek D K , Pantzar T J , Wyatt P (1997) Rare trisomy mosaicism diagnosed in amniocytes, involving an autosome other than chromosomes 13, 18, 20, and 21: karyotype/phenotype correlations. Prenat Diagn 17:201-242 Hsu W T , Shchepin D A , M a o R, Berry-Kravis E , Garber A P , Fischel-Ghodsian N , Falk R E , Carlson D E , Roeder E R , Leeth E A , Hajianpour M J , Wang J C , Rosenblum-Vos L S , Bhatt S D , Karson E M , H u x C H , Trunca C , Bialer M G , L i n n S K , Schreck R R (1998) Mosaic trisomy 16 ascertained through amniocentesis: evaluation of 11 new cases. A m J M e d Genet 80:473-480 Johnson M P , Childs M D , Robichaux A G , 3rd, Isada N B , Pryde P G , Koppitch F C , 3rd, Evans M I (1993) Viable pregnancies after diagnosis of trisomy 16 by C V S : lethal aneuploidy compartmentalized to the trophoblast. Fetal Diagn Ther 8:102-108 Johnson P, Duncan K , Blunt S, B e l l G , A l i Z , C o x P, Moore G E (2000) Apparent confined placental mosaicism of trisomy 16 and multiple fetal anomalies: case report. Prenat Diagn 20:417-421 Lee P A , Chernausek S D , Hokken-Koelega A C , Czernichow P (2003) International Small for Gestational A g e Advisory Board consensus development conference statement: management of short children born small for gestational age, A p r i l 24-October 1, 2001. Pediatrics 111:1253-1261 100  Leung A K , K a o C P (1999) Evaluation and management of the child with speech delay. A m F a m Physician 59:3121-3128, 3135 Limperopoulos C , Majnemer A , Shevell M I , Rohlicek C , Rosenblatt B , Tchervenkov C , Darwish H Z (2002) Predictors of developmental disabilities after open heart surgery in young children with congenital heart defects. J Pediatr 141:51-58 Lindor N M , Jalal S M , Thibodeau S N , Bonde D , Sauser K L , Karnes PS (1993) Mosaic trisomy 16 in a thriving infant: maternal heterodisomy for chromosome 16. C l i n Genet 44:185189 Ogden C L , Kuczmarski R J , Flegal K M , M e i Z , Guo S, W e i R , Grummer-Strawn L M , Curtin L R , Roche A F , Johnson C L (2002) Centers for Disease Control and Prevention 2000 growth charts for the United States: improvements to the 1977 National Center for Health Statistics version. Pediatrics 109:45-60 Penaherrera M S , Barrett IJ, Brown C J , Langlois S, Y o n g S L , Lewis S, Bruyere H , HowardPeebles P N , Kalousek D K , Robinson W P (2000) A n association between skewed X chromosome inactivation and abnormal outcome in mosaic trisomy 16 confined predominantly to the placenta. C l i n Genet 58:436-446 Pletcher B A , Sanz M M , Schlessel JS, Kunaporn S, M c K e n n a C , Bialer M G , Alonso M L , Zaslav A L , Brown W T , Ray J H (1994) Postnatal confirmation of prenatally diagnosed trisomy 16 mosaicism in two phenotypically abnormal liveborns. Prenat Diagn 14:933-940 Schneider A S , Bischoff F Z , M c C a s k i l l C , Coady M L , Stopfer JE, Shaffer L G (1996) Comprehensive 4-year follow-up on a case of maternal heterodisomy for chromosome 16. A m J M e d Genet 66:204-208 Simensen R J , Colby R S , Corning K J (2003) A prenatal counseling conundrum: mosaic trisomy 16. A case study presenting cognitive functioning and adaptive behavior. Genet Couns 14:331-336 Usher R , M c L e a n F (1969) Intrauterine growth of live-born Caucasian infants at sea level: standards obtained from measurements in 7 dimensions of infants born between 25 and 44 weeks of gestation. J Pediatr 74:901-910 Williams J, 3rd, Wang B B , Rubin C H , Clark R D , Mohandas T K (1992) Apparent non-mosaic trisomy 16 in chorionic v i l l i : diagnostic dilemma or clinically significant finding? Prenat Diagn 12:163-168 W o o V , Bridge P J , Bamforth JS (1997) Maternal uniparental heterodisomy for chromosome 16: case report. A m J M e d Genet 70:387-390  101  Table 5.1 CPM16 cases meeting the inclusion criteria Case # 1  Local case #  2 3 4  92.48 93.103 93.73  5 6 7  -  -  8 9 10 11  93.94 94.21 98.92 90.90 91.10 92.95  12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27  -  Origin of postnatal follow-up data  Hsu et al. (1998) case 9 Unpublished data Unpublished data Penaherrera et al. (2000); unpublished data Unpublished data Schneider et al. (1996) Penaherrera et al. (2000); unpublished data Penaherrera et al. (2000); unpublished data Johnson et al. (1993) Unpublished data Hsu et al. (1997) case XIV-11; unpublished data Hajianpour (1995); Hsu et al. (1998) case 5 Unpublished data Lindor et al. (1993) Pletcher et al. (1994) case 1 Unpublished data Unpublished data Williams et al. (1992); Garber et al. (1994) Woo et al. (1997); unpublished data Unpublished data Unpublished data Unpublished data Unpublished data Simensen et al. (2003) Hsu et al. (1997) case XIV-3 Rubin et al. (unpublished case from Devi et al. (1993) Dorfmann et al. (1992)  F u l l references i n Appendix B .  102  Table 5.2 Clinical data for included CPM16 cases Amnio  Case  UPD  Malformations  GA  Sex  Asc  VSD, anterior anus, abnormality of cervical vertebrae, short forearm, hypoplastic left thumb hypospadias none none hypospadias hypospadias, 5th finger clinodactyly none ASD, other cardiac anomalies none none hypospadias microcephaly, dolicocephaly, sacral dimple, high arched palate, mild pectus excavatum, widely spaced, hypoplastic nipples, unilateral hypoplastic labia majoria, tapering fingers, partial cutaneous syndactyly between 2 and 3 toes, mild asymmetry in hand and foot size suspected thyroid agenesis, inguinal hernia dolicocephaly, fingers were minimally tapered distally, 5thfingerclinodactyly DORV, PDA, endocardial cushion defect, brachycephaly, high narrow palate, wide spaced hypoplatic nipples, hyposapdias with cordee, right inguinal hernia, small umbilical hernia, scoliosis, deep sacral dimple, 5th finger clinodactyly, dorsiflexed toes, talipes calcaneovalgus ASD inguinal hernia, scoliosis, assymemmtrical skull, unilateral cryptorchidism ASD, VSD, PDA unilateral talipes equinovarus, left unilateral renal agenesis, left foot larger None  34  F  B  36 37 37 37 39 40 29 41 36 36 35  M M F M M F F F F M F  U U  35 35  F F  U B  40  M  B  37 38  F M  B  37 33  F F  U  30  F  B  42 38 34 32  F F F F  B B B U  ?  F  <37  ?  ? ? u  % 1  10%  BPD  2 3 4  0%  5 6 7 8 9 10 11 12  -  BPD UPD BPD BPD UPD UPD UPD UPD BPD UPD BPD  -  0% 4% 0% 3.7% 0 0 3% 6/20 colonies ; 0% in repeat a  6  nd  13 14  0% 45%  UPD UPD  15  39%  BPD  16 17  0% 3%  UPD BPD  18 19  0% 0%  BPD UPD  20  0%  BPD  21 22  mosaic 25% 21% 50%  BPD  23 24  -  None None  UPD  ?  u  B U B B U U u  B  rd  bicuspid aortic valve, hypoplastic left thumb, small 2 finger, left side body smaller than right None None None  ? ?  nd  9% mosaic 0%  25 26 27  -  F  " G A " = gestational age i n weeks. " A s c " = ascertainment, where U = unbiased and B = biased. "-" = not done; ' ? " = not known. Malformations exclude facial dysmorphism and skin anomalies. 7 % , then 1% i n repeat amniocentesis. 6/20 colonies had at least 1 trisomy 16 cell, of which only 1 colony (with 3 cells) was fully trisomic. a  b  103  Table 5.3 Follow-up data for length/height, weight and developmental outcome Catchup  Weight  Catchup  Case  Length/ Height  1  -  2  Birth = 0.26 7.5 mo = 0.09 11 mo = -0.07 9 yr 9 mo = 0.86  N/A  Birth = -0.66 7.5 mo = 0.20 11 mo = -0.15 9 yr 9 mo = 1.48  N/A  3  Birth = 0.36 6 mo = -1.32 1 yr = -0.50 3yr = 2.51 Birth = -1.31  N/A  Birth = -0.41 6 mo = -0.89 1 yr = -0.98 3yr= 1.50 Birth = -1.25 3 mo = 0.87 6 mo = -0.45 1 yr = -0.78 2 yr = -0.33 3 yr = -0.89 4yr = -1.19 Birth = -1.46 1 mo = -1.52 3 mo = 0.90 6 mo = 0.56 1 yr = 0.30 2 yr = 0.07 4 yr = 0.18 Birth = -3.18 3 mo= 1.69 6 mo = -1.66 8 mo = -1.02 10mo = -1.15 lyr = -1.46 1 yr 6 mo = -2.06 2 yr = -2.24 3yr = -1.68 4yr = -1.60 Report: fussy eater. Birth = -0.67 1 yr 6 mo = 7590  N/A  4  5  6  7  6 mo = -1.32 1 yr = -0.41 2yr = 0.19 3 yr = -0.93 4yr = -1.53 Birth = -1.45 1 mo = -1.87 3 mo = 1.29 6 mo = 0.33 1 yr = 0.03 2 yr = 0.05 4 yr = 0.25 Birth = -1.98 3 mo = -2.36 6 mo = -1.21 8 mo = -0.65 10 mo = 0.00 1 yr = -0.35 1 yr 6 mo = -1.57 2 yr = -0.76 3 yr = -0.95 4yr = -1.36 Report: fussy eater. Birth = -1.20 1 yr 6 mo = 5075  -  N/A  N/A  N/A  N/A  N/A  N/A  Y  N/A  Normal  2 yr = Developmental delay. 9 yr 9 mo = Reached all developmental milestones on time; very advanced vocabulary for age. 6 yr 9 mo = Speech development delayed.  N  Met developmental milestones for rolling over (4.5 mo), sitting (6 mo), walking (1 yr), first words (1 yr), and phrases (1 yr 6 mo). Met developmental milestones for rolling over at (3 mo), sitting (8 mo), walking (1 yr 1 mo), and speech development (mama, papa) (1 yr) 3 yr = Average in all areas of the Hawaii Early Learning Profile (HELP) and Receptive Expressive Emergent Language2 (REEL-2) with strengths in fine motor skills and receptive and expressive language; active and sociable. 1 yr 6 mo = Normal psychomotor.  Y  Parental anecdote: 2 yr = Occupational, physical and speech therapy; close to achieving developmental milestones on time. Cognitively 'right on target'. History of seizures. 8 yr = Math at grade level, above average in other subjects; occupational and  N  Y  N  Y  Y  Y  th  th  8  -  Development  -  -  -  104  Case  Length/ Height  Catchup  Catchup  Weight  9 -  -  -  -  10  -  -  Birth = 0.58 2 y r 8 mo = 0.08 13 yr 6 mo = 10  N/A  Development  Normal  speech therapy to be stopped soon. Evidence of obsessive-compulsive disorder (OCD). 1 yr 4 mo = Bright, sociable, normal developmental milestones. 1 yr 8 mo = Normal development  Y  Y  th  11  Birth = -3.80  Y  1 mo = -5.23 2 mo = -1.58 4 mo = -2.48 5 mo = -3.66 6.5 mo = -0.80 9 mo = -0.62 1 yr 1 mo = -1.35 12  3 mo = 40 2 yr 4.5 mo = 10  Y  Birth = -2.29  Y  2 yr 4.5 mo = Mild developmental and speech delay. 3 yr 9 mo = Developmental and speech delay. 4 yr 3 mo = Normal developmental milestones  N  11 mo = Development of social and communication skills appeared normal.  Y  th  (h  Birth = -4.70  1 yr 1 mo = Normal development  3 mo = 40 2 yr 4.5 mo = 3  ,h  13  Y  1 mo = -1.61 2 mo = -1.26 4 mo = -1.02 5 mo = -1.90 6.5 mo = -0.97 9 mo = -1.56 1 yr 1 mo = -1.40 1 yr 2.5 mo = 1.92 Y  Birth = -7.73  Birth = -3.03  rd  Y  3 mo = -2.91 11 mo = -1.50 2 yr 7 mo = -1.87  Birth = -2.08  N  3 mo = -2.98 11 mo = -2.16 2 yr 7 mo = -3.45  14  -  15  -  Y  Birth = -3.76  2 yr 2 mo = 10  th  -  -  Met developmental milestone for rolling over (6-7 mo), delayed for sitting alone (8 mo), and met milestone for walking (1 yr 2 mo). Delayed in rolling over and sitting unsupported (9 mo), but met developmental milestone for walking (1 yr 2 mo). Delayed for putting 2 words together (2.5 yr). Febrile seizure post-  Y  N  105  Case  16  17  Length/ Height  Birth = -1.45 6 mo = -2.09 1 yr 5 mo = -1.20 Birth = -2.44  Catchup  N/A  Weight  Catchup  Birth = -2.25  N  6 mo = -2.62 1 yr 5 mo = -4.52 Y  5 yr = 0.30  Birth = -2.99  Y  5 yr= 1.30  18  Birth = -1.00 2 yr 2 mo = 40  N/A  th  -  -  19  Birth = -1.45 1 yr = -1.09 2yr = -0.10  N/A  Birth = -1.42 1 yr = -1.84 2 yr = -0.38  N/A  20  Birth = -3.97  Y  Birth = -2.58  Y  -  1 yr 8 mo = -0.44 -  -  Birth = -2.32  Y  21 22  23  24  1 yr 8 mo = -0.10 Birth = -1.50 1 mo = -0.64 2 mo = -0.83 4 mo = -0.83 6 mo = -0.35 9 mo = -0.69 1 yr = -1.15  N/A  Birth = -2.15  Y  6 mo = -1.36 lyr2mo = -1.00 Birth = -0.81 5 yr 6 mo = 0.60  1 mo = -0.93 2 mo = -0.51 4 mo = -0.76 6 mo = -0.72 9 mo = -1.66 1 yr = -1.55 1 yr 4 mo = -0.12  N/A  Birth = -3.43  6 mo = -1.04 1 yr 2 mo = -1.25 Birth = 0.15 5 yr 6 mo = 0.79  Development  operatively at 6 mo. Early intervention services including speech therapy, with resulting good neurodevelopmental progress. 1 yr 5 mo = Normal developmental milestones 8 mo = Central motor disturbance. 11 mo = Mild central motor impairment and statomotoric developmental delay. 5 yr = Developmental delay, central motor impairment, speech and motor therapy. 2 yr 2 mo = Normal psychomotor development 3 yr 6 mo = Appeared developmentally normal. Met developmental milestones for hand eye coordination (appropriate at 1 yr), sitting (at least 7 mo), walking (1 yr). 3 yr = Normal development; full psychomotor assessment showed normal results on the Bayley scales. 1 yr 8 mo = Normal development. 2 yr = Normal development. Parental report: 1 yr 6 mo = 'all developmental milestones at appropriate time'  Normal  Y  N  Y  Y  Y Y Y  Y  1 yr 2 mo = 'Normal baby'  Y  N/A  5 yr 6 mo = Normal cognitive assessment. Early development  Y  106  Case  Length/ Height  Catchup  Weight  Catchup  25 26  -  -  -  -  -  -  -  -  27  Development within normal limits. lyr 5 mo = 'Normal' 2 yr = 'Normal phenotype' 3 yr = 'Clinically normal with no significant complications to date'  Normal Y Y Y  Birth weight and length data only shown for cases with long-term follow-up. Bolded birth weight and lengths were S G A .  107  Table 5.4 Association between trisomy in amniotic fluid and developmental delay Development Trisomy at amniocentesis  Present (>0%) Absent (0%)  Delayed 5 0  Normal 9 11  14 11  Presence (>0%) and absence (0%) of trisomy i n amniotic fluid as assessed by amniocentesis. Fisher Exact test, p = 0.038.  Table 5.5 Association between malformation and developmental delay Development Malformation  Present Absent  Normal 10 10  Delayed 6 0  16 10  Fisher Exact test, p = 0.035.  108  Figure 5.1 Association between birth weight and developmental delay  0.00'  -1.00  1  BW -2.00^  -3.00  -4.00H  Normal development  Developmental delay  The mean birth weight and standard deviation in cases with normal development was - 1 . 6 0 ± 1 . 1 0 and in cases with developmental delay - 2 . 6 6 ± 1 . 5 2 . The sample sizes were 18 and 5, respectively, t = 1.76, df = 21, p = 0.047.  109  6 Cytokeratin staining in villus cultures from miscarriage and CVS  6.1  9  Note  I wrote this chapter/manuscript, and did the experiments, data organization and analysis, with the following clarifications and exceptions. Ascertainment of miscarriage placental samples was coordinated by D r . D . McFadden. C V S placental sample dissection and cultures were done by the Clinical Cytogenetics laboratory at C & W . The immunochemistry was performed by myself with the protocol and equipment of Dr. C . MacCalman (Obstetrics and Gynaecology).  6.2  Introduction  Chorionic villus cultures are routinely used for clinical cytogenetic diagnosis of spontaneous abortions, as well as of ongoing pregnancies via chorionic villus sampling ( C V S ) in the late first-trimester. These cultures are thought to consist of cells from the villus mesenchymal core, which ultimately derive from the inner cell mass (ICM) of the blastocyst. In contrast, the trophoblast derives from the trophectoderm, which is the outer epithelial covering of the blastocyst (Larue et al. 1994). Cytokeratin is an epithelial marker that has been used to assess trophoblast contamination of chorionic villus cultures: using antibody to cytokeratin-8 or - 1 8 , or a pan-cytokeratin antibody, both cytokeratin-negative and -positive cells have been observed by immunochemistry of first-trimester chorionic villus cultures from C V S or terminations of pregnancy (Willers et al. 1990; Zimmer et al. 1993; Haigh et al. 1999). This suggests that trophoblast contamination is common in chorionic villus cultures. Recently, however, Blaschitz et al. (2000) conclusively demonstrated that only cytokeratin-7 ( C K 7 ) is  A version of this chapter will be submitted for publication. Yong PJ, McFadden DE, MacCalman CD, Robinson WP. Developmental origin of chorionic villus cultures from miscarriage and chorionic villus sampling (CVS). 9  110  specific for the trophoblast in histological sections, while other cytokeratins are also expressed in the mesenchymal core. I hypothesized that true trophoblast contamination (CK7-positive cells) of chorionic villus cultures would be rare, while mosaicism for mesenchymal cells expressing other cytokeratins would be present as seen in previous studies. Therefore, in this study chorionic villus cultures from C V S , as well as from miscarriages, were assessed for both C K 7 and C K 1 8 by immunochemistry.  6.3  Methods 6.3.1  Tissue processing and culture  C V S cultures were ascertained from the Clinical Cytogenetics laboratory at the Children's and Women's Health Centre of British Columbia ( C & W ) . C V S placental samples are processed as follows at the Clinical Cytogenetics laboratory: approximately 15-20 mg of dissected chorionic v i l l i are washed in Hanks Balanced Salt Solution ( H B S S ) (with calcium and magnesium) for 10 min at 37 degrees, followed by digestion with 1 m L of collagenase (1 mg/mL in H B S S ) for 20 min at 37 degrees with vortexing at 10 min, 15 min and 20 min. After the last vortex, the supernatant (trophoblast suspension) is removed. The remaining villus mesenchymal cores are then washed once in H B S S , and incubated in l m L of cold diluted collagenase (0.33 mg/mL) overnight at 4 degrees. The next day, the diluted collagenase is removed, and the remaining villus mesenchyme resuspended with culture medium (Amniomax + Amniomax serum supplement + 1 % antibiotic/antimycotic), vigorously disrupted by pipetting once per second for 60 sec (or until the tissue was disaggregated into a suspension), and then placed into a vented tissue culture flask. Leftover backup cultures were sent to our laboratory after approximately 2 weeks in culture. The cells were passaged once, then grown on coverslips for immunochemistry (see below).  Ill  Placental samples from first-trimester miscarriages were ascertained from the Embryopathology laboratory at B C Women's Hospital. Chorionic v i l l i were dissected from chorionic plate, and cleared of maternal decidua and blood clots. A s performed for clinical cytogenetic diagnosis of miscarriages, the v i l l i were simply minced and cultured in culture medium in vented tissue culture flasks. The cells were passaged once, then grown on coverslips for immunochemistry. J E G - 3 choriocarcinoma cells, provided courtesy of Dr. C . MacCalman, were grown on coverslips as a positive control for C K 1 8 and C K 7 immunochemistry. Approval was granted by the ethics committees of the University of British Columbia and C & W (Appendix A ) .  6.3.2  Immunochemistry  The coverslip-grown cells were washed three times in phosphate buffered saline (PBS) and fixed with 4% paraformaldehyde in P B S for 15 min. The cells were then washed three more times in P B S , permeabilized in methanol (with 2% hydrogen peroxide to inhibit endogenous peroxidases) for 20 min, washed in water for 5 min, and then blocked twice for 5 minutes with a blocking solution consisting of 10% Automation Buffer (Biomeda M 3 0 ) and 1% bovine albumin in distilled H2O. Then, the cells were blocked again with 10% normal serum (Vector S2000) in blocking solution for 20 min at 37 degrees. Next, the cells were incubated for 45 min at 37 degrees with mouse anti-human primary antibody to C K 7 (Dako M7018) diluted 1/50 in blocking solution or primary antibody to C K 1 8 (Dako M7010) diluted 1/25 in blocking solution. Negative controls consisted of blocking solution alone in lieu of primary antibody. After blocking twice for 5 min with blocking solution, the cultures were incubated for 30 min at 37 degrees with biotinylated horse anti-mouse IgG secondary antibody (Vector BA-2000) diluted 1/200 i n blocking solution. After blocking twice more for 5 min with blocking solution, the cultures were incubated for 30 min with streptavidin-biotin-horseradish peroxidase. After 112  washing in blocking solution for 5 min twice, the cells were incubated for 5 min in the chromogen, 3,3'-diaminobenzidine ( D A B ) solution, 0.05% i n blocking solution with 1% hydrogen peroxide substrate. The cells were then washed in running water for 5 min, then stained with hematoxylin (30 sec), decolourized in 4% acetic acid solution (20 sec), and then 'blued' in 1% lithium carbonate solution (30 sec), each step separated by washing in running water for 2 min. The coverslips were then dehydrated in ethanol (5 min, 3x) and xylene (5 min, 3x), and then mounted on slides in Permount mounting medium (Fisher SP15-100).  6.3.3  Statistical analysis  Statistical analyses were performed using SPSS 10.0 and the VassarStats Web Site for Statistical Computation (http://facultv.vassar.edu/lowry/VassarStats.htinl).  6.4  Results  Chorionic villus cultures from C V S and miscarriages showed little or no C K 7 staining, while the J E G - 3 cells did stain positive for C K 7 (Figures 6.1 and 6.2; Table 6.1). In contrast, the C V S cultures had variable numbers of CK18-negative and -positive cells, while cultures from miscarriages showed little or no C K 1 8 staining (Table 6.1). The proportion of C K 1 8 positive cells was significantly higher than the proportion of CK7-positive cells in the C V S cultures (paired-sample t = 2.59, df = 4, p = 0.031, 1-tailed). The difference between C V S and miscarriage cultures in CK18-positivity was also statistically significant (Mann-Whitney test, p < 0.02, 2-tailed).  6.5  Discussion  This study confirms the hypothesis that trophoblast contamination in chorionic villus cultures from first-trimester spontaneous abortions and C V S is minimal or non-existent. 113  Therefore, clinical cytogenetic diagnoses from these cultures do reflect the villus mesenchymal core, and ultimately, the inner cell mass ( I C M ) of the blastocyst. A s seen in previous studies (Willers et al. 1990; Zimmer et al. 1993), variable C K 1 8 expression was seen in cultures from C V S . Unexpectedly, little or no C K 1 8 staining was evident in the cultures from miscarriages. This may reflect the fact that the C V S cultures were 2 weeks old before transfer to our laboratory, since culture conditions can induce non-CK7 cytokeratin expression in villus cultures (von K o s k u l l and Virtanen 1987). Alternatively, there may be a true difference in composition of mesenchymal cells i n chorionic villus cultures from C V S and miscarriages, with C K 1 8 expressing mesenchymal cells absent in the latter. This difference may relate to time of sampling: C V S samples placentas from viable ongoing pregnancies, while placentas from miscarriages are affected by tissue degeneration during the period of retention between intrauterine demise and diagnosis. Future studies should include further characterization of these cultures to determine which components of the mesenchymal core are growing in vitro, including fibroblasts, myofibroblasts, endothelial cells, macrophages, and/or smooth muscle cells (Benirschke and Kaufmann 1995). It is possible that chromosome abnormalities diagnosed in different mesenchymal core cell-types may differ in functional and clinical significance.  114  6.6  References  Benirschke K , Kaufmann P (1995) Pathology of the Human Placenta. Springer-Verlag, N e w York Haigh T, Chen C , Jones C J , A p l i n J D (1999) Studies of mesenchymal cells from 1st trimester human placenta: expression of cytokeratin outside the trophoblast lineage. Placenta 20:615-625 Larue L , Ohsugi M , Hirchenhain J, Kemler R (1994) E-cadherin null mutant embryos fail to form a trophectoderm epithelium. Proc Natl Acad Sci U S A 91:8263-8267 von K o s k u l l H , Virtanen I (1987) Induction of cytokeratin expression in human mesenchymal cells. J C e l l Physiol 133:321-329 Willers I, Blankenfeld J, Goedde H W (1990) Characterization of long-term cell cultures of human chorion v i l l i and fibroblasts using antibodies to cytoskeletal proteins. A r c h Gynecol Obstet 248:87-92 Zimmer N , Gottert E , Kraus J, Zang K D , Henn W (1993) Immunophenotyping of mitotic cells from long-term cultures of chorionic v i l l i . H u m Genet 91:317-320  115  Table 6.1 C K 7 and C K 1 8 staining Sample  CK7 staining  CK18 staining  CVS1 CVS2 CVS3 CVS4 CVS5 SAB1 SAB2 SAB3 SAB4 SAB5  0% -1% -1% -1% 0% 0 0 0 0 -1%  31% 18% -1% 5% 22% 0 0 0 0 -1%  116  Figure 6.1 C K 7 and C K 1 8 staining in C V S cultures  lOOx. a) N o C K 7 staining in S A B and C V S cultures; b) C K 1 8 staining i n C V S cultures. See Table 1.  117  Figure 6.2 C K 7 staining in J E G - 3 cells (positive control)  200x.  118  7 E V T outgrowth in trisomic miscarriage 7.1.  Note  I wrote this chapter/manuscript, and did the experiments, data collection and analysis, with the following clarifications and exceptions. Placental samples from miscarriages were ascertained from the Embryopathology laboratory, coordinated by Dr. D . McFadden. Microsatellite P C R to confirm trisomy in the trophoblast was performed by R. Jiang (technician, Robinson laboratory). I performed the immunochemistry using the protocol and equipment of Dr. C . MacCalman.  7.2  Introduction  Extravillus trophoblast ( E V T ) are key to normal placentation. They arise from the villus cytotrophoblast and proliferate to form columns and a shell that adhere the placenta to the uterine tissue (Figure 7.1). From the columns/shell, an invasive subpopulation of E V T migrate further into the decidualized endometrium and myometrium (interstitial E V T ) and invade the spiral arteries (endovascular E V T ) , resulting i n spiral artery remodeling to produce a lowresistance high-flow circuit (Figure 7.1). Poor invasion and spiral artery remodeling is classically associated with preeclampsia (Chapter 4), but has also been found in miscarriage (Khong et al. 1987; Hustin et al. 1990; M i c h e l et al. 1990). In early pregnancy, endovascular E V T are also thought to form plugs that prevent full maternal blood flow from entering the intervillus space (IVS) until the end of the first-trimester; premature high-velocity flow of oxygenated maternal blood into the I V S in the first-trimester may damage the placenta contributing to the pathogenesis of miscarriage (Jauniaux et al. 2003a; Jauniaux et al. 2003b; Jauniaux et al. 2003c). About half of spontaneous abortions have been shown by histology to be A version of this chapter will be submitted for publication. Yong PJ, McFadden D E , MacCalman C D , Robinson WP. Extravillus trophoblast outgrowth in trisomic miscarriage. 1 0  119  associated with an absence or reduction of the E V T columns, shell, interstitial and endovascular subpopulations, and spiral artery remodelling in the basal plate, compared to no such changes in placentas from elective terminations (Hustin et al. 1990). This abnormal placentation in miscarriages may be influenced by the karyotype. There have been several studies of E V T growth and function i n chromosomally abnormal miscarriages. Hustin et al. (1990) found that the frequency of poor E V T column and shell formation and decidual spiral artery remodelling in miscarriage was increased among cases with histopathological findings associated with chromosomal abnormalities; however, actual karyotypes were not available. Other groups did not observe a significant association between poor placentation on histology and the karyotype of the miscarriage (Khong et al. 1987; Khong and Ford 1997; Sebire et al. 2002). In a study of uterine artery resistance (reflecting, in part, poor spiral artery remodelling) at the time of chorionic villus sampling ( C V S ) , Bindra et al. (2001) found no significant difference between euploid pregnancies and individual groups of chromosomally abnormal pregnancies (trisomy 21, trisomy 18, trisomy 13, monosomy X and 'other'). A l s o , Greenwold et al. (2003) showed no difference between euploid and chromosomally abnormal missed abortions in the distribution of intervillus maternal blood flow using Doppler, which depends in part on adequate spiral artery invasion. Together, these findings suggest that E V T are not likely to be significantly altered in all chromosomally abnormal pregnancies relative to euploid pregnancies. However, it is possible that only certain chromosomal abnormalities are associated with abnormal E V T . Autosomal trisomies are more likely to affect E V T growth and function. Trisomy 16, in particular, is the most common trisomy in miscarriage, and when confined to the placenta, is associated with preeclampsia (Chapter 4): these clinical outcomes could be caused by abnormally functioning trisomy 16 E V T . Thus, I hypothesized that consistent with previous studies, there would be no difference on average between euploid and chromosomally abnormal miscarriages in E V T outgrowth in 120  vitro, although chromosome-specific effects may be present. In this Chapter, using a different methodology from the previous studies, E V T outgrowth in vitro was assessed by culturing chorionic v i l l i from miscarriages on Matrigel, with outgrowth compared between different karyotypes.  7.3  Methods 7.3.1  Miscarriage cases  Placental samples were obtained from 52 cases of first-trimester spontaneous abortion examined in the Embryopathology laboratory at the B C Women's Hospital. These include cases with spontaneous expulsion, and cases requiring uterine evacuation following diagnosis of missed or incomplete abortion. Karyotyping was performed by the Clinical Cytogenetics laboratory at the Children's and Women's Health Centre of British Columbia ( C & W ) , which usually involved G-banding of metaphase spreads of cultures of minced chorion or chorionic v i l l i . In cases of culture failure, comparative genomic hybridization ( C G H ) of chorion or chorionic v i l l i was used. The cases had the following karyotypes: euploid (n = 17); trisomy 15 (n = 8); trisomy 16 (n = 4); trisomy 22 (n = 4); trisomy 20 (n = 2); trisomy 21 (n = 2), 47,-15, +i(15) (n = 2) (equivalent in dosage to free trisomy 15); triploidy (n = 2); trisomy 4 (n = 1); trisomy 9 (n = 1); trisomy 10 (n = 1); trisomy 13 (n = 1); trisomy 17 (n = 1); trisomy 18 (n = 1); monosomy X (n = 1); mosaic monosomy X (n = 1); monosomy X and trisomy 22 (n = 1); trisomy 8 and trisomy 20 (n = 1); and an unbalanced translocation (n = 1). Excluding the monosomy X and triploid cases, there were 30 females and 18 males. The average gestational age of the cases (at spontaneous expulsion or uterine evacuation) was 10.2 weeks ± 1 . 6 weeks (± standard deviation), with a range of 6-14 weeks. The experiments were done blind to karyotype. Because there had been no a priori hypothesis regarding gestational age and E V T outgrowth in culture (Aplin 2000), the experimenter (PJY) was aware of gestational age when placental 121  samples were received from Embryopathology. The study was approved by the ethics committees of the University of British Columbia and C & W (Appendix A ) .  7.3.2  Tissue culture and processing  Extravillus trophoblast ( E V T ) outgrowths were propagated from first-trimester chorionic villus explants using a modified protocol based on the methods first described by Yagel et al. (1989) and Genbacev et al. (1992). Chorionic v i l l i were dissected, washed in Dulbeco's M i n i m a l Essential M e d i u m ( D M E M ) , and then minced into 2-5mm pieces. Following further washings in D M E M , the minced v i l l i (explants) were placed in 6-well plates coated with a thinlayer of Matrigel (0.5mm thick) ( B D Biosciences 354603) (compared to the regular 1.0mm thick Matrigel-coated 6-well plates ( B D Biosciences 354432)). The number of explants varied between cases due to different amounts of placental tissue available. The average number of explants was 182 ± 161 explants per case (± standard deviation), with a range of 24 to 742 explants. The explants were allowed to attach i n a minimal amount of D M E M for 1-1.5 hours. Then, 1.5mL of culture medium was placed into the wells and changed every 2 days. The culture medium consisted of D M E M (with high glucose and without L-glutamine and sodium pyruvate) + 10% fetal bovine serum (FBS) + 1% L-glutamine + 1 % antibiotic-antimycotic.  7.3.3  Immunochemistry  After 10 days, the cultures were washed three times in phosphate buffered saline (PBS) and fixed directly on the Matrigel with 4% paraformaldehyde in P B S for 15 min. The cultures were then washed three more times in P B S before standard immunocytochemistry was performed in situ. The cultures were permeabilized in methanol for 20 min, washed in water for 5 min, then blocked twice for 5 min with a blocking solution consisting of 10% Automation Buffer (Biomeda M 3 0 ) and 1% bovine albumin in distilled F L O . Then, they were blocked again 122  with 10% horse normal serum in blocking solution for 20 min at 37 degrees. The cultures were incubated for 45 min at 37 degrees with primary antibody: monoclonal mouse anti-human antibody against cytokeratin-7 ( C K - 7 ) ( D A K O M7018), diluted 1/50 in blocking solution; or monoclonal mouse antibody against vimentin ( D A K O M0725), a mesenchymal marker (Blaschitz et al. 2000), diluted 1/50 in blocking solution. After blocking twice for 5 minutes with blocking solution, the cultures were incubated for 30 min at 37 degrees with secondary antibody: fluoroscein-conjugated (Vector FI-2000) or Texas Red-conjugated (Vector TI-2000) horse anti-mouse antibody against IgG, diluted 1/200 in blocking solution. Following incubation in secondary antibody and washing twice for 5 min with blocking solution, the cultures were incubated for 30 m i n in 5|ig/mL 4',6-diamidino-2-phenylindole (DAPI) in P B S . After washing, the cultures were examined under a fluorescent microscope, and the presence and number of explants with CK7-positive trophoblast outgrowths and vimentin positive villus mesenchymal outgrowths were counted for each case.  7.3.4  K a r y o t y p e confirmation  Griffin et al. (1997) found 6.1% (4/65) of miscarriages had different karyotypes in different placental lineages after cytogenetic analysis of metaphases from cultured mesenchymal cells and direct metaphases from spontaneous divisions of the villus cytotrophoblast. To check for such karyotypic discrepancy, chromosome-specific multiplex microsatellite P C R was performed on trophoblast D N A for 6 cases with a trisomy 15 karyotype found in the chorion or villus mesenchyme by conventional cytogenetics or C G H . Trophoblast was isolated by digesting chorionic v i l l i for 20 min at 37 degrees with l m g / m L collagenase I A diluted in Hank's Balanced Salt Solution, vortexing every 5 minutes. After 20 min, the v i l l i were vigorously vortexed, and the resulting trophoblast suspension removed. This technique has been confirmed histologically (I. Barrett, personal communication), and the resulting trophoblast suspension 123  contains both villus and extravillus trophoblast (D. McFadden, personal communication). Chromosome 15 multiplex P C R was performed using fluorescent-labeled primers for the following microsatellite loci: D15S541, G A B R B 3 , and D15S11. The alleles for each locus were visualized using the A B I Prism 310; an approximate 1:1:1 or 2:1 allelic ratio for each locus was taken as evidence of trisomy in the trophoblast.  7.3.5  Statistical analysis  A l l statistical analysis was done using SPSS 10.0 or the VassarStats Web Site for Statistical Computation (http://faculty.vassar.edu/lowrv/Vassai-Stats.html). A l l p-values are 1tailed due to a priori evidence or rational mechanisms, unless otherwise noted.  7.4  Results  C e l l outgrowths from the explants appeared after ~4 days and exhibited 2 common morphologies under the light microscope: column-like (Figure 7.2) and fibroblast-like (Figure 7.3). The column-like outgrowths resembled the E V T column outgrowths on collagen described by (Aplin et al. 1999). The column-like outgrowths stained positive for C K - 7 (Figure 7.4) and negative for vimentin, confirming that they are E V T . The fibroblast-like outgrowths were negative for C K - 7 and positive for vimentin (data not shown), suggesting a villus mesenchymal origin. The C K - 7 positive E V T column outgrowths were present i n 48% (25/52) of cases. The relationship between E V T outgrowths and karyotype may be confounded by the number of explants, sex of the embryo, or gestational age. There was no association between the number of explants per case and the presence of E V T outgrowths (t = 0.41, df = 50, p = 0.35). Neither was there an association between sex of the embryo and the presence of E V T outgrowths (Fisher Exact test, n = 48, p = 1.00, 2-tailed). However, there was an association between gestational age and E V T outgrowths (Figure 7.5). Figure 7.5 illustrates that there 124  appears to be a threshold at 10 weeks gestation, with outgrowths more likely to occur in cases <10 weeks compared to cases >10 weeks. Analyzed statistically, miscarriages <10 weeks were significantly more likely to have the presence of E V T outgrowths (Fisher Exact test, n = 49, p = 0.019, 2-tailed; R R = 1.88; Table 7.1). To eliminate the confounding effect of gestational age, only the cases <10 weeks gestation (n = 28) were considered in the following analyses. A m o n g these cases, there was a sufficient number for comparison only for euploids (n = 9), trisomy 15 (n = 6), and trisomy 16 (n = 3). The distribution for the presence or absence of E V T outgrowths was similar between the euploid and trisomy 15 cases (Fisher Exact test, n = 15, p = 0.54; Table 7.2). When the proportion of explants with E V T outgrowths was compared between the euploid and trisomy 15 cases, the distribution was again similar between the 2 groups (t = 0.12, df = 13, p = 0.45; Figure 7.6). Trophoblast suspensions were available for 4/6 trisomy 15 cases <10 weeks: chromosome 15-specific multiplex microsatellite P C R confirmed trisomy 15 in the suspensions for each case. In contrast, no outgrowths were observed among the trisomy 16 cases, which compared to the euploid cases was statistically significant (Fisher Exact test, n = 12, p = 0.045; R R = 0; Table 7.3). For the remaining cases <10 weeks, the cases with outgrowths (n = 6) had the following karyotypes: trisomy 10; trisomy 13; trisomy 21; trisomy 8 and trisomy 20; 4 5 , X and trisomy 22; and 4 5 , X mosaicism. The cases that did not have any outgrowths (n = 3) had the following karyotypes: trisomy 17, trisomy 22, and triploidy. When E V T outgrowth was compared between the euploid cases and the chromosomally abnormal cases as a total group (n = 19), there was no significant difference in the presence of outgrowths (Fisher Exact test, n = 28, p = 0.20; Table 7.4) or in the proportion of explants with outgrowths (t = 0.12, df = 26, p = 0.45; Figure 7.7). The number of explants was not significantly associated with the presence or absence of E V T outgrowths among cases <10 weeks, with the trend in the opposite direction (cases with 125  E V T outgrowths had fewer explants on average) (Welch's approximate t = 1.62, df = 12.8, p = 0.07). Similarly, the number of explants was not significantly different between the euploid and trisomy 16 cases <10 weeks, again with the trend in the opposite direction (trisomy 16 cases had more explants on average) (t = 1.02, df = 10, p = 0.17). In addition, the sex of the embryo was not significantly associated with the presence of E V T outgrowths among cases <10 weeks (Fisher Exact test, n = 25, p = 0.43, 2-tailed).  7.5  Discussion  There were no differences on average between euploid and chromosomally abnormal miscarriages in E V T in vitro outgrowth, which was the hypothesized finding based on previous studies. However there was some evidence of a chromosome-specific effect; specifically, trisomy 16 E V T may have poor outgrowth ability compared to euploid E V T . It should be noted, however, that only cases <10 weeks gestation could be analyzed, as older miscarriages demonstrated few or no E V T outgrowths. Considering all chromosomally abnormal cases, E V T outgrowth was quite variable. This supports the observations of previous authors that E V T structure and spiral artery remodeling of chromosomally abnormal pregnancies in vivo are quite variable and not significantly different as a total group from that of euploid pregnancies (Khong et al. 1986; Khong et al. 1987; Hustin et al. 1990; Khong and Ford 1997; Bindra et al. 2001; Sebire et al. 2002; Greenwold et al. 2003). Only Khong and Ford (1997) studied trisomy 16 specifically; they found that 40% (2/5) of trisomy 16 miscarriages showed poor placentation at histology, which was higher compared to euploid miscarriages (21%; 4/19), although not significantly so (Fisher Exact test, n = 24, p = 0.37). Considering all cases, a reduction in E V T outgrowth was observed after 10 weeks gestation. Previous studies using tissue from terminations have reported E V T cultures in specimens up to 12 weeks (Irving et al. 1995) and 13 weeks (Aplin et al. 1999) gestation, with 126  no mention of altered growth at later gestational ages. Thus, A p l i n (2000) concluded that late first-trimester chorionic v i l l i continue to be able to form new anchoring sites. The reduction in outgrowth in this study is likely due to the period of intrauterine retention and tissue degeneration in miscarriages before expulsion or diagnosis. Cases retained for longer periods (and hence with later gestational ages at expulsion or diagnosis) would therefore have undergone more tissue degeneration, which presumably affects tissue culture. The difficulty with culturing human miscarriage material due to tissue retention and degeneration also limited sample size in this study, as did the relatively small probability of ascertaining any one trisomy. The E V T morphology on a thin-layer of Matrigel resembled the anchoring columns seen on collagen (Aplin et al. 1999), rather than the invasive outgrowths previously reported on Matrigel (Caniggia et al. 1997; A p l i n et al. 1999). The reasons for this are not clear, although it could be related to the tissue source (miscarriage vs. termination) or to the fact that wells precoated with a thin-layer of Matrigel was used in this study. A l s o , the onset of the E V T outgrowths (~4 days) represented a significant delay compared to previous studies using Matrigel or collagen (~1 day) (e.g. Caniggia et al., 1997; A p l i n et al., 1999). In previous studies using collagen, fibroblasts have been previously shown to grow only after weeks of culture; therefore, the fibroblast-like outgrowths observed at ~4 days in this study represent a significant acceleration in growth. These differences are likely also due to the tissue degeneration in miscarriages, which may affect trophoblast more than the mesenchyme. The abnormal trisomy 16 E V T outgrowth may explain in part the common occurrence of trisomy 16 in miscarriages, and the high rate of pregnancy complication ( I U G R and preeclampsia) seen in ongoing pregnancies with trisomy 16 confined to the placenta (Chapters 2-5). The altered E V T outgrowth may be related to the possible over-expression of the chromosome 16-encoded E-cadherin (E-cad). E-cadherin is well expressed in villus cytotrophoblast, but is progressively downregulated proximal to distal from the villus tip to the 127  invasive E V T (Zhou et al. 1997; Floridon et al. 2000; Shih le et al. 2002). Therefore, overexpression of E-cadherin in trisomy 16 (which has not yet been investigated) may inhibit E V T growth and differentiation in trisomy 16 pregnancies. There is also evidence for a different distribution of maternal immune cells in the decidua between euploid and chromosomally abnormal (in particular, trisomy 16) spontaneous abortions (Yamamoto et al. 1999; Quack et al. 2001). Therefore, trisomy 16 miscarriages may have unique pathogenic mechanisms compared to euploid miscarriages. The similar distribution of E V T outgrowth between trisomy 15 and euploid cases suggests that the aspects of E V T growth assessed in vitro with this methodology do not differ between trisomy 15 and euploid E V T . Trisomy 15 is common in miscarriages, present in - 5 % of spontaneous abortions at B C Women's Hospital (D. McFadden and W . Robinson, unpublished data). Thus mechanisms other than poor E V T outgrowth may be involved in trisomy 15 miscarriage. It is also notable that the one trisomy 21 case (<10 weeks gestation) did exhibit E V T outgrowths. Khong and Ford (1997) found normal placentation at histology in a single case of trisomy 21 miscarriage. However in a recent study of 4 second-trimester terminations of D o w n syndrome pregnancies, there was a clear defect in E V T outgrowth on Matrigel (Wright et al. 2004). Additional studies with larger sample size are needed to clarify these apparently contradictory findings in trisomy 21. In conclusion, E V T outgrowth in vitro was similar to previous in vivo histological findings: chromosomally abnormal miscarriages show variable degrees of placentation, with no significant difference, on average, compared to euploid miscarriages. Trisomy 16 E V T may be abnormal, while trisomy 15 E V T , at least for initial outgrowth on Matrigel, was similar to E V T from euploid miscarriages. Further investigations should include assessment of E-cadherin expression in the trisomy 16 placenta.  128  7.6  References  A p l i n J (2000) Maternal influences on placental development. Semin C e l l Dev B i o l 11:115-125 A p l i n J D , Haigh T, Jones C J , Church H J , Vicovac L (1999) Development of cytotrophoblast columns from explanted first-trimester human placental v i l l i : role of fibronectin and integrin alpha5betal. B i o l Reprod 60:828-838 Bindra R, Curcio P, Cicero S, Martin A , Nicolaides K H (2001) Uterine artery Doppler at 11-14 weeks of gestation in chromosomally abnormal fetuses. Ultrasound Obstet Gynecol 18:587-589 Blaschitz A , Weiss U , Dohr G , Desoye G (2000) Antibody reaction patterns in first trimester placenta: implications for trophoblast isolation and purity screening. Placenta 21:733741 Caniggia I, Taylor C V , Ritchie J W , L y e S J , Letarte M (1997) Endoglin regulates trophoblast differentiation along the invasive pathway in human placental villous explants. Endocrinology 138:4977-4988 Floridon C , Nielsen O, Holund B , Sunde L , Westergaard J G , Thomsen S G , Teisner B (2000) Localization of E-cadherin in villous, extravillous and vascular trophoblasts during intrauterine, ectopic and molar pregnancy. M o l H u m Reprod 6:943-950 Genbacev O, Schubach S A , M i l l e r R K (1992) Villous culture of first trimester human placentamodel to study extravillous trophoblast ( E V T ) differentiation. Placenta 13:439-461 Greenwold N , Jauniaux E , Gulbis B , Hempstock J, Gervy C , Burton G J (2003) Relationship among maternal serum endocrinology, placental karyotype, and intervillous circulation in early pregnancy failure. Fertil Steril 79:1373-1379 Griffin D K , M i l l i e E A , Redline R W , Hassold T J , Zaragoza M V (1997) Cytogenetic analysis of spontaneous abortions: comparison of techniques and assessment of the incidence of confined placental mosaicism. A m J M e d Genet 72:297-301 Hustin J, Jauniaux E , Schaaps JP (1990) Histological study of the materno-embryonic interface in spontaneous abortion. Placenta 11:477-486 Irving J A , Lysiak JJ, Graham C H , Hearn S, Han V K , Lala P K (1995) Characteristics of trophoblast cells migrating from first trimester chorionic villus explants and propagated in culture. Placenta 16:413-433 Jauniaux E , Greenwold N , Hempstock J, Burton G J (2003a) Comparison of ultrasonographic and Doppler mapping of the intervillous circulation in normal and abnormal early pregnancies. Fertil Steril 79:100-106 Jauniaux E , Gulbis B , Burton G J (2003b) The human first trimester gestational sac limits rather than facilitates oxygen transfer to the foetus—a review. Placenta 24 Suppl A:S86-93  129  Jauniaux E , Hempstock J, Greenwold N , Burton G J (2003c) Trophoblastic oxidative stress in relation to temporal and regional differences in maternal placental blood flow i n normal and abnormal early pregnancies. A m J Pathol 162:115-125 Khong T Y , De W o l f F , Robertson W B , Brosens I (1986) Inadequate maternal vascular response to placentation in pregnancies complicated by pre-eclampsia and by small-forgestational age infants. B r J Obstet Gynaecol 93:1049-1059 Khong T Y , Ford J H (1997) Lack of correlation between conceptual karyotype and maternal response to placentation. Reprod Fertil Dev 9:271-274 Khong T Y , Liddell H S , Robertson W B (1987) Defective haemochorial placentation as a cause of miscarriage: a preliminary study. B r J Obstet Gynaecol 94:649-655 M i c h e l M Z , Khong T Y , Clark D A , Beard R W (1990) A morphological and immunological study of human placental bed biopsies in miscarriage. B r J Obstet Gynaecol 97:984-988 Quack K C , Vassiliadou N , Pudney J, Anderson D J , H i l l J A (2001) Leukocyte activation in the decidua of chromosomally normal and abnormal fetuses from women with recurrent abortion. H u m Reprod 16:949-955 Sebire N J , Fox H , Backos M , R a i R , Paterson C , Regan L (2002) Defective endovascular trophoblast invasion in primary antiphospholipid antibody syndrome-associated early pregnancy failure. H u m Reprod 17:1067-1071 Shih le M , Hsu M Y , Oldt R J , 3rd, Herlyn M , Gearhart J D , Kurman R J (2002) The Role of E cadherin in the Motility and Invasion of Implantation Site Intermediate Trophoblast. Placenta 23:706-715 Wright A , Zhou Y , Weier JF, Caceres E , Kapidzic M , Tabata T, Kahn M , Nash C , Fisher SJ (2004) Trisomy 21 is associated with variable defects in cytotrophoblast differentiation along the invasive pathway. A m J M e d Genet A 130:354-364 Yagel S, Casper R F , Powell W , Parhar R S , Lala P K (1989) Characterization of pure human first-trimester cytotrophoblast cells in long-term culture: growth pattern, markers, and hormone production. A m J Obstet Gynecol 160:938-945 Yamamoto T, Takahashi Y , Kase N , M o r i H (1999) Role of decidual natural killer ( N K ) cells in patients with missed abortion: differences between cases with normal and abnormal chromosome. C l i n E x p Immunol 116:449-452 Zhou Y , Fisher SJ, Janatpour M , Genbacev O, Dejana E , Wheelock M , Damsky C H (1997) Human cytotrophoblasts adopt a vascular phenotype as they differentiate. A strategy for successful endovascular invasion? J C l i n Invest 99:2139-2151  130  Table 7.1 Asssociation between gestational age and the presence of E V T outgrowths  E V T outgrowths Present Absent Gestational age  <10 weeks >10 weeks  11 16  17 5  Total 38 21  E V T outgrowths were less frequent i n miscarriages <10 weeks (Fisher Exact test, p = 0.011).  Table 7.2 E V T outgrowth in euploid and trisomy 15 cases <10 weeks gestation E V T outgrowths Absent Present Karyotype  Trisomy 15 Euploid  2 2  4 7  Total 6 9  Fisher Exact test, p = 0.54.  Table 7.3 E V T outgrowth in euploid and trisomy 16 cases <10 weeks gestation E V T outgrowths Present Absent Karyotype  Trisomy 16 Euploid  3 2  0 7  Total 3 9  Fisher Exact test, p = 0.045.  Table 7.4 E V T outgrowth in euploid and chromosomally abnormal cases <10 weeks gestation E V T outgrowths Present Absent Karyotype  Abnormal Euploid  9 2  10 7  Total 19 9  Fisher Exact test, p = 0.20.  131  Figure 7.1. Extravillus trophoblast (EVT) columns deriving from the chorionic villus  Chorionic villus  The columns join to form a shell, while invasive E V T migrate to form endovascular plugs in maternal arteries ( B V ) in early pregnancy, and then remodel the arteries to create a lowresistance circuit.  132  Figure 7.2 EVT outgrowths  lOOx.  Figure 7.3 Fibroblast-Iike outgrowths  lOOx.  133  Figure 7.4 Cytokeratin-7 staining of E V T outgrowths  lOOx.  134  Figure 7.5 Association between gestational age and E V T outgrowth  i  i  i  i  6-9  10  11  12-14  Gestational age (weeks)  The black bars represent the number of cases where E V T outgrowths were observed; the white bars represent the number of cases where no E V T outgrowths were observed. The proportion of cases with E V T outgrowths is as follows: 6-9 weeks = 67% (10/15), 10 weeks = 54% (7/13), 11 weeks = 33% (3/9), 12-14 weeks = 17% (2/12).  135  Figure 7.6 Proportion of explants with EVT outgrowths for euploid and trisomy 15 cases  0.40  0.30 H  0.20 H  0.10i  o.oo H  Euploid  Trisomy 15  For cases with gestational age <10 weeks. Means and standard deviations for the euploid and trisomy 15 cases were 0.07±0.11 and 0.08±0.08, respectively. The sample sizes were 9 and 6, respectively, t = 0.12, df = 13, p = 0.45.  136  Figure 7.7 Proportion of explants with EVT outgrowths for euploid and abnormal cases  0.60 H  0.40 H  0.20'  0.00-  Euploid  1  Abnormal  For cases with gestational age <10 weeks. Means and standard deviations for euploid and all chromosomally abnormal cases were 0.07±0.11 and 0.08±0.15, respectively. The sample sizes were 9 and 19, respectively, t = 0.12, df = 26, p = 0.45.  137  8 Protein kinase profiling in trisomic miscarriage  11  8.1  Note  I wrote this chapter/manuscript, and did the experiments, data collection and analysis, with the following clarifications and exceptions. Placental samples from miscarriages were ascertained from the Embryopathology laboratory, coordinated by D r . D . McFadden. Microsatellite P C R to check for maternal contamination and to confirm trisomy was done by myself and R. Jiang. I performed the immunochemistry using the protocol and equipment of Dr. C. MacCalman. R N A extractions were done by myself and Dr. C . Anderson (post-doc, Robinson lab); R T - P C R to check for D N A contamination was done by Dr. C . Anderson. Protein expression profiling was performed by staff at Kinexus Bioinformatics Corporation by commercial agreement. R N A expression profiling was performed by A . Haegert at the Prostate Centre Array Facility by commercial agreement.  8.2  Introduction  The phenotype of trisomy is ultimately due to developmental perturbation from altered protein expression secondary to the extra chromosome; however, the pathways are not well elucidated. T w o hypotheses have been proposed for the pathogenesis of trisomy: the gene dosage hypothesis and the amplified instability hypothesis (Reeves et al. 2001). In the gene dosage hypothesis, the trisomic phenotype is considered primarily due to over-expression of specific genes on the particular trisomic chromosome and their corresponding functional effects at the protein level (Epstein 1990). This would account for the specific phenotypic features of the different trisomies. In the amplified developmental instability hypothesis, the trisomic phenotype is thought to result because the expression of hundreds of trisomic genes disrupts A version of this chapter will be submitted for publication. Yong PJ, McFadden DE, MacCalman CD, Robinson WP. Protein kinase profiling in trisomic miscarriage. 11  138  cellular homeostasis non-specifically (i.e. regardless of chromosome) and causes the cell (and organism) to have a non-specific amplified sensitivity to environmental effects and genomic mutations and polymorphisms during development (Shapiro 1983; Shapiro 1989). This hypothesis is supported by the observation that D o w n syndrome phenotypic traits (e.g. malformations) are also present at a low rate in the general population (Pritchard and K o l a 1999), and that certain bilateral traits (e.g. dermatoglyphics) show increased fluctuating asymmetry ( F A ) in D o w n syndrome (Katznelson et al. 1999). If environmental and genetic factors are randomly distributed in the population regardless of karyotype, then the hypothesis can account for the many overlapping features of different trisomies. A corollary is that environmental effects and genome-wide mutations and polymorphisms can act as 'modifiers' of the phenotype of an individual with a given trisomy. Reeves et al. (2001) contend that the two hypotheses are not mutually exclusive and could be integrated at the molecular level. Furthermore, Shapiro (2001) argues that when comparing gene expression between a particular trisomy and a normal karyotype, another trisomy must be used as a control to demonstrate that the differences in gene expression are either specific to the particular trisomy (gene dosage hypothesis) or common to both trisomies (amplified instability hypothesis). N o studies have assessed gene dosage effects (using another trisomy as a control) or amplified instability using protein expression profiling in trisomic individuals. Protein kinases were chosen for profiling in the present study as they act as fundamental effectors of signal transduction by catalyzing reversible protein phosphorylation (Manning et al. 2002), and so can be sensitive indicators of functional perturbation i n cell function. Further, both protein and R N A levels were examined, as they do not necessarily correlate (Greenbaum et al. 2003). I hypothesized that both amplified instability and gene dosage effects would operate simultaneously at the molecular level in trisomic cells. In this study, the expression of seventyfive protein kinases were assessed by 2 D Western blots and oligonucleotide microarrays in  chorionic villus fibroblast cultures from trisomy 16 miscarriages, and compared to euploid miscarriages using trisomy 15 miscarriages as control. Expression data were analyzed to determine any differences in level of expression (gene dosage hypothesis) or differences in variability in expression (amplified instability hypothesis) between the karyotypes.  8.3  Methods 8.3.1  Miscarriage cases  Chorionic v i l l i , chorion or chorionic plate, and decidua were obtained from a series of miscarriages (n = 73) from the Embryopathology laboratory at B C Women's Hospital. Routine karyotype information for each case was provided by the Clinical Cytogenetics laboratory at the Children's and Women's Health Centre of British Columbia ( C & W ) after the experiments were performed. The study was approved by ethics committees of the University of British Columbia and C & W (Appendix A )  8.3.2  Tissue processing and culture  A s karyotype was not known ahead of time, all samples were initially processed in the same manner. The tissues were dissected and cleaned of blood, and samples of the maternal decidua, placental chorionic v i l l i , and chorion (or chorionic plate) were taken. The chorionic v i l l i were cultured, while maternal decidua, and leftover chorionic v i l l i and/or chorion, were frozen for future D N A extraction. The chorionic villus cultures were set up as described for C V S samples in Chapter 6. Culture medium was changed every 2 days. Appearance of outgrowths from attached explants (and/or single cells) was highly variable, sometimes evident after 2 days, but in other cases taking about 1 week. A t the second passage, cells were transferred to another culture flask and also to coverslips for immunochemistry. In addition, some cells at the second passage were 140  frozen for future D N A extraction. A t the third passage, the cells from the flask were transferred to 5 other flasks, from which the cells were harvested for protein and R N A extraction at - 7 5 % confluence.  8.3.3  Immunochemistry and PCR  Immunochemistry was carried out to characterize the cultures from a subset of the cases: the cultures appeared to consist primarily of fibroblasts (Table 8.1). Immunochemistry was carried out as described i n Chapter 7, with a human skin fibroblast cell line (provided by Dr. D . Speert, U B C Department of Paediatrics) used as a positive control. Primary antibodies to the following antigens were used: cytokeratin-7 ( C K - 7 ) (Dako M7018), vimentin (mouse monoclonal IgG) (Dako M0725), CD45/Leucocyte C o m m o n Antigen ( L C A ) (mouse monoclonal IgG) (Dako M0701), CD31/platelet endothelial cell adhesion molecule-1 ( P E C A M 1) (mouse monoclonal IgG) (Dako M0823), desmin (rabbit polyclonal IgG) (Dako A0611), and fibroblast surface protein (FSP) (mouse monoclonal I g M ) (Sigma F4771). Vimentin is specific for the chorionic villus mesenchymal core (Blaschitz et al. 2000), while the other antibodies specify components of the mesenchymal core: L C A for hematopoietic cells (Blaschitz et al. 2000); P E C A M - 1 for endothelium (Blaschitz et al. 2000) and megakaryocytes, platelets, myeloid cells, natural killer cells, some T-cell subsets, and B - c e l l precursors (Dako); desmin for muscle (Blaschitz et al. 2000); and F S P for fibroblasts (Blaschitz et al. 2000) and tissue macrophages and 95% of peripheral blood monocytes (Sigma). The normal serum utilized was normal horse serum (Vector S2000), except for desmin and F S P , where normal goat serum (Vector S-1000) was used. The following dilutions in blocking solution were used for the primary antibodies: C K - 7 (1/50), vimentin (1/50), P E C A M - 1 (1/20), L C A (1/50), desmin (1/20), and F S P (1/500). The secondary antibody was fluoroscein-conjugated (Vector FI-2000) or Texas Red-conjugated (Vector TI-2000) horse anti-mouse I g G antibody, with the exception 141  of FITC-conjugated goat anti-rabbit IgG antibody (Vector FI-1000) for desmin and Texas Redconjugated goat anti-mouse I g M antibody (Vector TI-2020) for F S P . In addition to immunochemistry, genotyping of microsatellites by P C R of the maternal decidua, chorionic v i l l i and/or chorion, and chorionic villus cultures was utilized to rule out maternal contamination and trisomic/euploid mosaicism in the cultures used for profiling experiments.  8.3.4  Protein kinase protein profiling  Protein kinases were profiled at the protein level in fibroblast cultures for each of the following cases (Table 8.1): Eu-4, Eu-5, Eu-6, Eu-7 (n = 4, euploid); T16-4, T16-5, T16-6 (n = 3, trisomy 16); and T15-1, T15-2, T15-3 (n = 3, trisomy 15). Profiling was done using the Kinetworks K P K S - 1 . 2 screen for 75 protein kinases developed and performed by Kinexus Bioinformatics (Appendix C ) . Details for protein extraction are provided on the Kinexus website (http://www.kinexus.ca). Briefly, cultured cells at the third passage were harvested using a brief treatment of warm 0.2% trypsin-EDTA, and the cell pellet sonicated in lysis buffer (containing protease and phosphatase inhibitors and 0.5% Nonidet P-40 detergent in a 20 m M M O P S buffer at p H 7.2). The homogenate was centrifuged at 1000,000 g for 30 min, the supernatant was removed, and protein concentration determined by a Bradford assay (Bio-Rad) using bovine serum albumin as a concentration standard. Then, 750 p:L of protein was diluted to 1 p:g/uL in S D S sample buffer, and boiled for 4 min before freezing at - 8 0 degrees. The Kinetworks K P K S - 1 . 2 screen involves S D S - P A G E , Western transfer, and immunoblotting using a 20-lane multiblotter apparatus from B i o - R a d (allowing a different cocktail of antibodies in each well/lane). Further details are available (Pelech et al. 2003). Isoforms for a particular kinase are detectable by shifts in molecular weight. Protein expression levels were determined by using Quantity One software (Bio-Rad) to quantify band signal  142  intensity, arising from chemoluminescence detected with a FluorS M a x Multi-Imager (BioRad). The raw band signal intensity for each kinase was used as the 'expression level' without any statistical normalization. Appendix A H , which lists the protein kinases in this study, differs from the corresponding table on the Kinexus website in two ways. First, J N K 2 is not included because it is very close to another band, and its accuracy has been called into question. Second, glycogen synthase kinase 3-a and 3-(3 were considered separate kinases because the isoforms are encoded on different genes. Third, the expression levels of the 3 isoforms for C K 2 are added together as a composite measure, because although they are encoded on different genes, they function as a tetramer and are therefore non-independent due to stochiometric relationships. In addition, K P K S - 1 . 2 has one lane with an antibody specific to E R K 2 , and a second lane with an antibody to both E R K 1 and E R K 2 . The E R K 2 (37) isoform was taken to be the average from the corresponding bands in the two lanes. In the second lane, the E R K 2 (39) isoform is difficult to distinguish from the E R K 1 (40) isoform (due to overlapping of the bands), and in our analyses, only the E R K 1 (40) band was called. Thus, the E R K 2 (39) isoform was taken to be its band in the first lane, while the 'true' E R K 1 (40) isoform was calculated by subtracting the band intensity of the E R K 2 (39) isoform in the first lane from the band intensity of the E R K 1 (40) isoform in the second lane. Finally, the E R K 1 (41) isoform was taken to be its band in the second lane.  8.3.5  Protein kinase R N A profiling  Protein kinases were profiled at the R N A level in fibroblast cultures for each of the following cases (Table 8.1): Eu-6, Eu-7, Eu-8, Eu-9 (n = 4); T16-5, T16-6 (n = 2); and T15-2, T15-3, T15-4 (n = 3). A human 14,000 oligonucleotide microarray developed and performed by the Gene Array Facility of the Prostate Centre at the Vancouver Hospital and Health Sciences 143  Centre ( V H H S C ) (http://www.prostatecentrexonVresearch/genearray.html) was utilized for R N A profiling. A l l K P K S - 1 . 2 kinase genes were present on the microarray except for calcium/calmodulin-dependent kinase kinase a. Cultured cells at the third passage were harvested using a brief treatment of warm 0.2% trypsin-EDTA. R N A was extracted immediately (Eu-6, T16-5, T15-3), or the cell pellet was first placed in R N A l a t e r (Ambion) using the provided protocol (Eu-7, Eu-8, Eu-9, T16-6, T15-1, T15-2). R N A was extracted using a GenElute Mammalian Total R N A Miniprep kit (Sigma) according to the provided protocol. Isolated R N A was precipitated with absolute ethanol overnight at 4 degrees. R T - P C R using primers upstream and downstream of the transcribed region of the X I S T gene was used to exclude D N A contamination of the R N A . Next, the R N A was run on an Agilent 2100 Bioanalyzer, and its quality assessed in two ways (Auer et al. 2003): 1) degradation peak signals between the small R N A s and the 18S peak (evidence of degradation); and 2) decreased 28S/18S peak ratio and 28S breakdown peaks above and below the 18S peak (evidence of apoptosis). Using Degradometer software (Auer et al. 2003; http://www.dnaarrays.org), the cases showed minimal degradation (degradation factor less than 1.5%), but the average 28S/18S ratio was decreased in the samples stored in RNAlater compared to the samples extracted immediately (mean peak area ratio: 1.07±0.08 (n = 6) vs. 1.31±0.16 (n = 3), respectively; t = 2.50, df = 7, p = 0.041). However, RNAlater can cause a specific reduction in the 28S-specific peak by changing 28S secondary structure (Ambion) and none of the cases demonstrated 28S breakdown peaks. Hence, apoptosis was not considered significant. The Prostate Centre Gene Array Facility uses the 3 D N A Array 350 kit (Genisphere) for R N A profiling. The detailed protocol is available at http://www.genisphere.com. Briefly, case (test) R N A was matched to an equal amount of universal reference human (control) R N A : either  144  1.5 jag (Eu-7, T16-6, T15-2) or 5 [ig (Eu-6, Eu-8, Eu-9, T16-5, T15-1, T15-3) was used. The test R N A was reverse transcribed with poly(T) primers with a specific capture sequence extension, while the control R N A was reverse transcribed with poly(T) primers with a different specific capture sequence extension. The test and control c D N A were then each amplified by P C R to produce probes. The test and control probes were co-hybridized onto the microarray chip overnight. After washing, the microarray chips were co-hybridized with Cy3-labelled 3 D N A capture reagent (-375 C y 3 per molecule) that binds the test specific capture sequence and Cy5-labelled 3 D N A capture reagent that binds the control specific capture sequence. The microarray chips were then washed, imaged, and C y 3 and C y 5 signals quantified using Imagene 5.6 software. After quantification, the C y 3 and C y 5 signal data were normalized with GeneSpring software. A standard intensity-dependent (non-linear or Lowess) normalization was utilized to account for dye-related artefacts, which works by adjustment of control C y 5 signals. The natural logarithm of the ratio between the test C y 3 signal and adjusted control C y 5 signal at each microarray spot was used as the 'expression level' for that spot (i.e. oligonucleotide for a particular gene).  8.3.6  Profiling data analysis 8.3.6.1 Confounding  The euploid and trisomic cases were not perfectly matched for sex of the conceptus or gestational age because of the difficulty in ascertaining any one particular trisomy, and the prevalence of culture failure and maternal contamination in cultures from miscarriage. To check for confounding, associations between the protein expression level of each of the kinases, and sex or gestational age, were carried out using the two-sample t-test and Pearson correlation coefficient, respectively. Five kinases were significantly associated with sex ( C A M K 1 ) or  145  gestational age ( R A F B , PKC-cc, P K C - ( i , R O K - a ) and were therefore excluded from the study, leaving seventy kinases for the rest of the analyses. Twenty-three kinases were undetectable at the protein level in all cases, and also removed from the study. R N A expression was detected for all kinases. However, inspection of R N A expression data showed two outliers with extreme levels (both > 8 standard deviations above the global mean) that were considered methodologyrelated (poor control probe hybridization) and so were excluded.  8.3.6.2 Protein kinase expression at the protein/RNA levels Pair-wise comparisons of protein expression levels of each kinase were made between karyotypes (euploid vs. trisomy 16; and euploid vs. trisomy 15). Because of the difference in variance between the groups (see text), the non-parametric 2-sample t-test of ranked data (Zar 1996) was utilized for the pair-wise comparisons. For kinases encoded by genes on the trisomic chromosome, a one-tailed test was used. A correction for multiple comparisons was not made because the small number of samples limited statistical power. For each of the kinases with statistically significant differences when compared between karyotypes at the protein level, the corresponding analysis was performed at the R N A level also using a 2-sample t-test rank test because of differences in variance.  8.3.6.3 Inter-individual variation in expression Inter-individual variation in protein and R N A expression was assessed within each karyotype.  For the cases within a given karyotype, the mean and standard deviation for protein  or R N A level was calculated for each kinase. Then the standard deviation was divided by the mean to give the coefficient of variation ( C V ) in protein or R N A level for each kinase within each karyotype. The coefficients were then compared between karyotypes using the nonparametric W i l c o x o n paired-sample test. Because C V is a dimensionless parameter, 146  comparisons can be made between samples with difference measurement scales (Zar 1996). Hence, the C V for protein and C V for R N A were compared within each karyotype using the W i l c o x o n test.  8.3.7  Statistical analysis  Statistical analyses were performed using SPSS 10.0 and the VassarStats W e b Site for Statistical Computation (http://facultv.vassar.edu/lowry/VassarStats.html). In the text, X ± Y represents mean ± standard deviation unless otherwise noted.  8.4  Results 8.4.1  Protein kinase expression at the protein level  Seven kinases showed significantly different expression in trisomy 16 compared to the euploid group ( C D K 1 , C D K 7 , P K C - E , P K G 1 , E R K 1 , S 6 K p70, I K K - a ) (Table 8.2), two of which were also significantly different in trisomy 15 compared to the euploid group ( S 6 K p70, I K K - a ) . Seven other kinases were significantly altered only in trisomy 15 ( C K l - e , S R C , C D K 9 , D N A P K , M E K 2 , P K C - z , PKC-(3) (Table 8.2). For those kinases with genes on chromosome 16 ( C K 2 [isoform-oc], PKC-(3, E R K 1 ) , only E R K 1 was significantly overexpressed in trisomy 16. The larger of the two E R K 1 isoforms (with apparent M W 41kDa) was 2.8x more expressed in trisomy 16 (mean = 3675 ± 1938) than in euploids (mean = 1297 ± 389) (p = 0.034). In a separate experiment, Kinexus phospho-antibody to T202/Y204 on E R K 1 was used for 4 of the cases, which showed a 2.9x increase in trisomy 16 (2075 and 6949; mean = 4512) compared to euploids (1738 and 1370; mean = 1554). Hence the activated phosphorylated form of E R K 1 was also increased in trisomy 16. The upstream E R K 1 activators, M E K 1 and M E K 2  147  (Roux and Blenis 2004), were not differentially expressed in trisomy 16 (data not shown), which supports a specific gene dosage-related increase in E R K 1 expression. Interestingly, there were 13 significant decreases and 3 significant increases in kinase expression i n the trisomic groups (combined) compared to euploid; i f one assumes a null hypothesis that the direction of (significant) expression change should be random with respect to whether it is increased or decreased in the trisomic groups, then the trend of decreased expression in trisomy was statistically significant (Sign test, p = 0.021).  8.4.2  Protein kinase expression at the R N A level  R N A levels were compared between the trisomic groups and the euploid group for the kinases whose comparisons were statistically significant at the protein level (Table 8.3). T w o kinases, E R K 1 ( 1 6 p l 2 - p l 1.2) and C D K 1 (10q21.1), had significantly altered expression in the same direction at both the R N A and protein levels: E R K 1 (higher) and C D K 1 (lower) in trisomy 16 compared to euploid (Tables 8.2 and 8.3). Figure 8.1 illustrates how the chromosome 16encoded E R K 1 is significantly increased at both the R N A and protein levels in trisomy 16 compared to the euploid group. In contrast, P K G 1 and D N A P K showed inverse relationships between R N A and protein expression (Tables 8.2 and 8.3).  8.4.3  Inter-individual variation in expression  The coefficients of variation ( C V ) for R N A and protein expression for each karyotype are illustrated in Figure 8.2 and compared in Table 8.4. The coefficient ( C V ) was used as a measure of inter-individual variability in expression. A t the protein level, the coefficient of variation was significantly higher in both trisomic groups compared to the euploid group. A t the R N A level, the coefficient of variation was significantly lower in both trisomic groups compared to the euploid group. Figure 8.2 and Table 8.4 illustrate that for the euploid group, 148  the coefficients at the R N A and protein levels were equal, while for both trisomic groups, the coefficient was significantly lower at the R N A level compared to the protein level.  8.5 Discussion Consistent with the hypothesis, this study demonstrated the simultaneous operation of both gene dosage effects and amplified instability on protein kinase expression in trisomy 16 placental fibroblasts in vitro. The mostly unique pattern of differentially expressed protein kinases in trisomy 16 versus euploid, using trisomy 15 as a control, supports the gene dosage hypothesis. The gene dosage hypothesis is also supported by the E R K 1 over-expression at both the R N A and protein levels in trisomy 16. The two other trisomy 16 genes, however, did not exhibit protein over-expression, which is consistent with the finding of no change in protein expression for most chromosome 21 genes i n D o w n syndrome fetal cerebral cortex (Cheon et al. 2003a; Cheon et al. 2003b; Cheon et al. 2003c; Cheon et al. 2003d; Ferrando-Miguel et al. 2004). These protein results are in contrast to the increased R N A expression on average of genes on the trisomic chromosome (Chrast et al. 2000; FitzPatrick et al. 2002; Gross et al. 2002; M a o et al. 2003; Saran et al. 2003; Amano et al. 2004; Giannone et al. 2004; Kahlem et al. 2004; L y l e et al. 2004; Tang et al. 2004; Dauphinot et al. 2005). The trisomic groups combined showed a significant trend toward decreased protein kinase expression, an observation also seen in studies of signalling proteins in Down syndrome (Engidawork and Lubec 2003). This trend may be related to the undetectable levels of S 6 K p70 in both trisomic groups (Table 8.2), since S 6 K p70 promotes translation of ribosomal proteins and translational factors (Shah et al. 2000). Similarly, of 9 translation factors that were examined in D o w n syndrome fetal cerebral cortex, 3 were down-regulated and 5 were unchanged (Freidl et al. 2001; Engidawork and Lubec 2003). Thus, a translational 'deficiency' could account for decreased expression of protein kinases in trisomy. 149  Increased inter-individual variation (quantified by the coefficient of variation) in protein kinase expression at the protein level provides support for the amplified instability hypothesis in both trisomy 16 and trisomy 15. The finding of decreased inter-individual variation at the R N A level, however, is contradictory to previous studies. Saran et al. (2003) observed increased R N A expression variation between trisomic mice on inspection of graphs produced by principal components analysis ( P C A ) , while M a o et al. (2003) observed no difference for human fetal D o w n syndrome brain on inspection of graphs of the coefficient of variation. The difference between previous studies and this study may be related to differences in methodology: Saran et al. (2003) and M a o et al. (2003) used genome-wide Affymetrix microarrays, as opposed to only 75 protein kinases i n this study; and the biological interpretations of P C A and the coefficient of variation are not identical. A l s o , the coefficients of variation were statistically compared only in this study. Regardless, the increased inter-individual variation in trisomy observed at only the protein level in this study suggests amplified instability manifests specifically on protein expression for these protein kinases. This may be due to amplified sensitivity to environmental differences (e.g. slight variation between cell cultures) or, more likely, to genome-wide genetic variation between trisomic conceptuses. Since variability at the R N A level was not increased, trisomic protein expression may be particularly sensitive to genetic variation specific to the translational machinery (e.g. regulatory factors binding cis elements on R N A transcripts) or to protein degradation (e.g. proteases). A Western blot approach was utilized in this study instead of 2-D gel electrophoresis techniques because the 2-D gel positions of fewer than two-dozen kinases have been elucidated (Pelech et al. 2003). Protein kinases are difficult to identify on 2-D gels because their intracellular concentrations are much lower (100-1000x) than those of proteins involved in metabolism or cell structure (Pelech et al. 2003). Our study was limited by small sample size, in part because of the difficulty with culturing of human miscarriage material and relatively 150  small probability of obtaining any one trisomy; this limited the power of the analyses, and as a consequence, a correction was not made for multiple comparisons of kinase expression levels between karyotypes.  Furthermore, the conclusions may be only applicable to this set o f protein  kinases in trisomy 16. It is possible that different patterns of gene expression may be present with other protein kinases or other types of signalling proteins and in other trisomies. A l s o , cell cultures are always artificial models of complex in vivo systems, where gene expression is affected by complex cell-cell and cell-matrix interactions within a finely tuned hormonal and growth factor milieu. The chorionic villus mesenchymal fibroblasts are also only one cell type in the placenta, and different expression patterns may be present in the functionally important trophoblast subpopulations. Despite the limitations o f this study, the findings add some insight into the pathogenesis of trisomy 16 in miscarriages, as well as in ongoing pregnancies complicated by confined placental mosaicism ( C P M ) . It is interesting that E R K 1 , a major intracellular regulator of cell proliferation (Roux and Blenis 2004), showed gene dosage-related over-expression in trisomy 16. It remains to be seen whether a small increase (~2x) in E R K 1 expression can alter signal transduction pathways enough to perturb trisomy 16 embryological development. The presence of amplified instability in vitro raises the possibility that genetic variation (e.g. single gene mutations) or environmental variation (e.g. maternal effects during pregnancy) may play a role in the trisomy 16 pregnancies in vivo. The presence o f both dosage effects and amplified instability supports chromosome 16-specific, as well as genome-wide and multifactorial factors, in the pathogenesis of trisomy 16.  151  8.6  References  Amano K , Sago H , Uchikawa C , Suzuki T, Kotliarova S E , Nukina N , Epstein C J , Yamakawa K (2004) Dosage-dependent over-expression of genes in the trisomic region of T s l C j e mouse model for D o w n syndrome. H u m M o l Genet 13:1333-1340 Auer H , Lyianarachchi S, N e w s o m D , Klisovic M I , Marcucci G , Kornacker K (2003) Chipping away at the chip bias: R N A degradation in microarray analysis. Nat Genet 35:292-293 Blaschitz A , Weiss U , Dohr G , Desoye G (2000) Antibody reaction patterns in first trimester placenta: implications for trophoblast isolation and purity screening. Placenta 21:733741 Cheon M S , Bajo M , K i m S H , Claudio J O , Stewart A K , Patterson D , Kruger W D , Kondoh H , Lubec G (2003a) Protein levels of genes encoded on chromosome 21 in fetal D o w n syndrome brain: challenging the gene dosage effect hypothesis (Part II). A m i n o Acids 24:119-125 Cheon M S , K i m S H , Ovod V , Kopitar Jerala N , Morgan JI, Hatefi Y , Ijuin T, Takenawa T, Lubec G (2003b) Protein levels of genes encoded on chromosome 21 in fetal D o w n syndrome brain: challenging the gene dosage effect hypothesis (Part III). A m i n o Acids 24:127-134 Cheon M S , K i m S H , Yaspo M L , Blasi F , A o k i Y , Melen K , Lubec G (2003c) Protein levels of genes encoded on chromosome 21 in fetal D o w n syndrome brain: challenging the gene dosage effect hypothesis (Part I). A m i n o Acids 24:111-117 Cheon M S , Shim K S , K i m S H , Hara A , Lubec G (2003d) Protein levels of genes encoded on chromosome 21 in fetal D o w n syndrome brain: Challenging the gene dosage effect hypothesis (Part IV). A m i n o Acids 25:41-47 Chrast R , Scott H S , Papasavvas M P , Rossier C , Antonarakis E S , Barras C , Davisson M T , Schmidt C , Estivill X , Dierssen M , Pritchard M , Antonarakis S E (2000) The mouse brain transcriptome by S A G E : differences in gene expression between P30 brains of the partial trisomy 16 mouse model of Down syndrome (Ts65Dn) and normals. Genome Res 10:2006-2021 Dauphinot L , L y l e R , Rivals I, Dang M T , M o l d r i c h R X , Golfier G , Ettwiller L , Toyama K , Rossier J, Personnaz L , Antonarakis S E , Epstein C J , Sinet P M , Potier M C (2005) The cerebellar transcriptome during postnatal development of the T s l C j e mouse, a segmental trisomy model for D o w n syndrome. H u m M o l Genet 14:373-384 Engidawork E , Gulesserian T, Fountoulakis M , Lubec G (2003) Aberrant protein expression in cerebral cortex of fetus with D o w n syndrome. Neuroscience 122:145-154 Engidawork E , Lubec G (2003) Molecular changes in fetal D o w n syndrome brain. J Neurochem 84:895-904  152  Epstein C J (1990) The consequences of chromosome imbalance. A m J M e d Genet Suppl 7:3137 Ferrando-Miguel R, Cheon M S , Lubec G (2004) Protein levels of genes encoded on chromosome 21 in fetal D o w n Syndrome brain (Part V ) : overexpression of phosphatidyl-inositol-glycan class P protein ( D S C R 5 ) . A m i n o Acids 26:255-261 FitzPatrick D R , Ramsay J, M c G i l l N I , Shade M , Carothers A D , Hastie N D (2002) Transcriptome analysis of human autosomal trisomy. H u m M o l Genet 11:3249-3256 Freidl M , Gulesserian T, Lubec G , Fountoulakis M , Lubec B (2001) Deterioration of the transcriptional, splicing and elongation machinery in brain of fetal D o w n syndrome. J Neural Transm Suppl:47-57 Giannone S, Strippoli P, Vitale L , Casadei R, Canaider S, Lenzi L , D'Addabbo P, Frabetti F , Facchin F , Farina A , Carinci P, Zannotti M (2004) Gene expression profile analysis in human T lymphocytes from patients with D o w n Syndrome. A n n H u m Genet 68:546-554 Greenbaum D , Colangelo C , Williams K , Gerstein M (2003) Comparing protein abundance and m R N A expression levels on a genomic scale. Genome B i o l 4:117 Gross SJ, Ferreira J C , Morrow B , Dar P, Funke B , Khabele D , Merkatz I (2002) Gene expression profile of trisomy 21 placentas: a potential approach for designing noninvasive techniques of prenatal diagnosis. A m J Obstet Gynecol 187:457-462 Kahlem P, Sultan M , Herwig R, Steinfath M , Balzereit D , Eppens B , Saran N G , Pletcher M T , South ST, Stetten G , Lehrach H , Reeves R H , Yaspo M L (2004) Transcript level alterations reflect gene dosage effects across multiple tissues in a mouse model of down syndrome. Genome Res 14:1258-1267 Katznelson M B , Bejerano M , Yakovenko K , Kobyliansky E (1999) Relationship between genetic anomalies of different levels and deviations in dermatoglyphic traits. Part 4: Dermatoglyphic peculiarities of males and females with D o w n syndrome. Family study. Anthropol A n z 57:193-255 L y l e R, Gehrig C , Neergaard-Henrichsen C , Deutsch S, Antonarakis S E (2004) Gene expression from the aneuploid chromosome in a trisomy mouse model of down syndrome. Genome Res 14:1268-1274 Manning G , Whyte D B , Martinez R, Hunter T, Sudarsanam S (2002) The protein kinase complement of the human genome. Science 298:1912-1934 M a o R, Zielke C L , Zielke H R , Pevsner J (2003) Global up-regulation of chromosome 21 gene expression in the developing D o w n syndrome brain. Genomics 81:457-467 Pelech S, Sutter C , Zhang H (2003) Kinetworks protein kinase multiblot analysis. Methods M o l B i o l 218:99-111  153  Pritchard M A , K o l a I (1999) The "gene dosage effect" hypothesis versus the "amplified developmental instability" hypothesis in D o w n syndrome. J Neural Transm Suppl 57:293-303 Reeves R H , Baxter L L , Richtsmeier JT (2001) Too much of a good thing: mechanisms of gene action in D o w n syndrome. Trends Genet 17:83-88 Roux P P , Blenis J (2004) E R K and p38 M A P K - a c t i v a t e d protein kinases: a family of protein kinases with diverse biological functions. Microbiol M o l B i o l Rev 68:320-344 Saran N G , Pletcher M T , Natale J E , Cheng Y , Reeves R H (2003) Global disruption of the cerebellar transcriptome in a D o w n syndrome mouse model. H u m M o l Genet 12:20132019 Shah O J , K i m b a l l SR, Jefferson L S (2000) A m o n g translational effectors, p70S6k is uniquely sensitive to inhibition by glucocorticoids. Biochem J 347:389-397 Shapiro B L (1983) D o w n syndrome—a disruption of homeostasis. A m J M e d Genet 14:241-269 Shapiro B L (1989) The pathogenesis o f aneuploid phenotypes: the fallacy o f explanatory reductionism. A m J M e d Genet 33:146-151 Shapiro B L (2001) Developmental instability of the cerebellum and its relevance to Down syndrome. J Neural Transm Suppl: 11-34 Tang Y , Schapiro M B , Franz D N , Patterson B J , Hickey F J , Schorry E K , Hopkin R J , W y l i e M , Narayan T, Glauser T A , Gilbert D L , Hershey A D , Sharp F R (2004) Blood expression profiles for tuberous sclerosis complex 2, neurofibromatosis type 1, and Down's syndrome. A n n Neurol 56:808-814 Zar J H (1996) Biostatistical Analysis. Prentice H a l l , Upper Saddle River, N e w Jersey  154  Table 8.1 Immunochemistry, and protein/RNA profiling of mesenchymal core cultures Case  Karyotype  GA  CK-7  Eu-1 Eu-2 Eu-3 Eu-4 Eu-5 Eu-6 Eu-7 Eu-8 Eu-9 T16-1 T16-3 T16-4 T16-5 T16-6 T15-1 T15-2 T15-3 T15-4  46,XY 47,XY 47,XY 46.XX 46,XY 46,XY 46,XX 46,XY 46,XX 47,XX,+ 16 47,XY,+16 47,XY,+ 16 47,XX,+16 47,XX+16 47,XY,+15 Male,+15' 47,XX,+15 47,XY,+15  12 10 10 1" 13 6 1 10 12 8 10 12 8 10 10 11 10 8  -  Vimen  +  LCA  PECAM  Desmin  FSP  -  -  -  +  -  -  + + + + +  -  -  -  R  +  * * * *  -  st  P  -  +  -  ±  * * * *  +  ±  +  * * * * * *  * * * * *  " G A " = gestational age; " I " = first-trimester, but exact gestational age unknown. " V i m e n " = Vimentin. " - " = <1% cells stained positive. " ± " = most cells were F S P positive, but relatively less (both i n proportion of cells and intensity) compared to vimentin staining. " P " = protein profiling was done. " R " = R N A profiling was done. ' C G H failed for this sample; the sex and extra chromosome 15 was determined by microsatellite P C R . s  155  Table 8.2 Protein levels of kinases with significant differences in protein expression  Kinase  Chr  CDK1 CDK7 PKC-e PKG1 ERK1  10q21.1 5ql2.1 2p21 10qll.2 16pl2pl 1.2 17q23.2  S6K p70 IKK-a CKl-e SRC CDK9 DNAPK MEK2 PKC-z PKC-P  14ql3 22ql3.1 20ql2ql3 9q34.1 8qll 19pl3.3 lp36.33p36.2 16pll.2  Euploid vs. Trisomy 15 -  3401 ±743 6115 ±1120  12148 ±4326  180 ± 80  0+-0  0±0  p = 0.012  p = 0.012  1484 ± 254 192 ± 99 59 ± 17  743 ± 35 36 ±62 14 ±24  1019±185 0±0 0±0  p = 0.012 -  p = 0.012 p = 0.012 p = 0.012  5234 ± 805  4728 + 419  3401 ±785  -  p = 0.012  577 + 401  0±0 1015 ±441 9968 ± 3248  -  15778 ±1968  1340±703 13839± 1817  -  p = 0.012 p = 0.012 p = 0.012  5994 ± 583  5294 ± 1267  9882 ±2371  -  p = 0.012  755 ± 270 1754 ± 220  Trisomy 16 Mean expression  Euploid vs. Trisomy 16 p = 0.010 p = 0.012 p = 0.012 p = 0.012 p = 0.006  Trisomy 15 Mean expression 362 ±319 729 ± 73 1408±966 3970± 1874 6985± 1559  Euploid Mean expression 520 ± 202 887 ±114 2708 ± 301  101 ± 174 569 ±100 1271 ±621 5642 ±419  -  Bolded mean expression levels are the higher of each statistically significant pair-wise comparison (2-sample rank t-test; see text). Italicized mean expression levels are o f the other trisomy not involved i n the statistically significant pair-wise comparison.  156  Table 8.3 R N A levels of kinases with significant differences in protein expression  Kinase  Chr  CDK1 CDK7 PKC-e PKG1 ERK1  10q21.1 5ql2.1 2p21 10qll.2 16pl2pll.2 17q23.2 14ql3 22ql3.1 20ql2-ql3 9q34.1 8ql 1 19pl3.3 lp36.33p36.2 16pll.2  S6K p70 IKK-oc CKl-e SRC CDK9 DNAPK MEK2 PKC-z PKC-p  Euploid Mean expression 1.03 ± 0.25  1.15 + 0.46 1.23 ±0.16 1.37 ± 0.44 1.12 ± 0.12  Trisomy 16 Mean expression  0.47 ± 0.36 1.54 ±0.14 0.98 ±0.12 1.00 ±0.008 1.54 ± 0.07  Trisomy 15 Mean expression -  Euploid vs. Trisomy 16  p = 0.41 p= 1.00  -  p = 0.012  -  p= 1.00  0.88 ± 0.23 0.78 ± 0.22 0.49 ± 0.07 1.07 ±0.53 0.98 ±0.18 0.68 ± 0.04 1.20 ±0.16 0.41 ±0.07  0.92 +- 0.009 0.74 ± 0.004 -  0.88 ± 0.25 0.79 ±0.10 0.45 ±0.21 0.89 ± 0.26 1.04 ±0.15 0.87 ±0.18 1.15 ±0.24 0.52 ±0.11  1.27 ±0.36  -  1.10 ±0.01  p = 0.042  p = 0.41 p = 0.31 p = 0.042 p = 0.021  -  -  Euploid vs. Trisomy 15 -  p = 0.76 p = 0.53 p = 0.53 p = 0.76 p = 0.76 p= 1.00 p = 0.33  Bolded mean expression levels are the higher of a statistically significant pair-wise comparison (2-sample rank t-test; see text).  157  Table 8 . 4  Coefficients of variation (CV) at the R N A and protein levels  Karyotype [Level]  Mean  Karyotype [Level]  Mean  n  Comparison  Euploid [RNA] Trisomy 16 [RNA] Trisomy 15 [RNA] Euploid [RNA]  0.32 + 0.19 0.19 ± 0.18  0.31 ±0.24 0.49 ± 0.50  47 44  p < 0.001  0.52 ± 0.42  42  p < 0.001  0.19±0.18  47  p < 0.001  Euploid [RNA]  0.32 ±0.19  0.25 ±0.17  47  p = 0.02  Euploid [Protein] Euploid [Protein]  0.27 ±0.14  Euploid [Protein] Trisomy 16 [Protein] Trisomy 15 [Protein] Trisomy 16 [RNA] Trisomy 15 [RNA] Trisomy 16 [Protein] Trisomy 15 [Protein]  0.49 ± 0.50  44  p < 0.01  0.52 ± 0.42  42  p < 0.001  0.24 ±0.18 0.32 ±0.19  0.29 ± 0.24  p = 0.50  Bolded mean expression levels are the higher of each statistically significant pair-wise comparison (Wilcoxon test; see text). There are different numbers of kinases (n) i n these paired-sample comparisons because for the trisomic karyotypes, some kinases were not expressed at the protein level such that the coefficient of variation was undefined.  158  Figure 8.1 E R K 1 R N A and protein expression  1.50  15000  H  Protein  RNA  1.00  J  U  IOOOO  -  5000  E u T16 T15  0.50  J  0  0 RNA  Protein  Mean ± standard error. E R K 1 R N A and protein expression in trisomy 16 (T16), but not in trisomy 15 (T15), are significantly increased compared to euploid (Eu) expression (see Tables 8.2 and 8.3, and Figure 8-1.  159  Figure 8.2 Coefficient of variation (CV) of R N A and protein expression  0.60'  0 . 5 0 1  CV  0.40-  0.30 •  I  0.201  Eu  Eu  T16 T15  T16 T15  0.10-L  RNA  Protein  Mean ± standard error. For statistical comparisons, see Table 8.4 and the text.  9 Conclusion This thesis research has focused on cytogenetic, biological, and clinical aspects of trisomy in the placenta. Trisomy C P M pregnancies have reduced placental weight and birth weight, and birth weight was determined in part by the level of trisomic trophoblast, through an alteration in placental function independent of placental weight (Chapter 2). Pregnancies with trisomy 16 C P M ( C P M 16) were at particularly high-risk for abnormal outcomes such as I U G R and malformation, and the risk was modulated by factors such as the presence of trisomy 16 cells in amniotic fluid, full trisomy on chorionic villus sampling ( C V S ) , sex of the fetus, mode of ascertainment, and uniparental disomy for chromosome 16 (upd(16)mat) in the fetus (Chapter 3). C P M 16 pregnancies were also at increased risk for preeclampsia (Chapter 4). However, infants from such pregnancies had a good postnatal prognosis in terms of catch-up growth and developmental progress (Chapter 5). Predictors of developmental delay included trisomy in amniotic fluid, presence of malformation, and low birth weight for gestational age, which may all be markers of low-level trisomy mosaicism in tissues of the infant. The higher risk of abnormal outcome in C P M 16 pregnancies and the high incidence of trisomy 16 in miscarriages may be related in part to the observed poorer outgrowth of trisomy 16 extravillus trophoblast ( E V T ) in vitro (Chapter 7). Other than trisomy 16, E V T outgrowth in vitro was found to be variable among other chromosomally abnormal miscarriages as a group; trisomy 15 in particular showed a similar distribution in E V T outgrowth compared to euploid miscarriages. Although gene expression in trisomy 16 trophoblast could not be examined (due to no outgrowths), it is apparent that there is an altered distribution of protein kinase expression in trisomy 16 placental fibroblasts compared to euploid placental fibroblasts, and in a manner that differs from trisomy 15 placental fibroblasts (Chapter 8). A dose-related increase in E R K 1 protein expression was present in trisomy 16. There was also an increase in the variance in  161  protein kinase expression in the trisomic groups, which interestingly, manifested only at the protein level. These observations indicate that both dosage-effects and amplified instability due to the effects of genomic and environmental variation, can operate simultaneously in trisomy. There have been a number of studies involving post-partum cytogenetic analysis of the placenta in pregnancies with small-for-gestational age ( S G A ) newborns, with widely divergent results in the frequency of C P M (Kalousek and D i l l 1983; Verp and Unger 1990; Kennerknecht et al. 1993; Wolstenholme et al. 1994; Artan et al. 1995; Krishnamoorthy et al. 1995; WilkinsHaug et al. 1995; Stipoljev et al. 2001; Masuzaki et al. 2004; Grati et al. 2005). The variable results are likely related to the heterogeneity i n methodology, including the type of cytogenetic analysis (conventional vs. F I S H vs. molecular/PCR), the number of placental sites sampled, and the tissue-types analyzed (e.g. trophoblast, mesenchymal core, and/or chorionic plate). N o systematic study of post-partum placental sampling has been published for pre-eclampsia, though such a study is in-progress in the laboratory of my supervisor (W. Robinson). For idiopathic I U G R or preeclamptic pregnancies, it is possible that identification of more 'highrisk' cases w i l l yield higher rates of C P M at post-partum cytogenetic investigation. Examples include severe S G A newborns (<3 percentile), with or without malformation, from pregnancies rd  with pre-term labour, P R O M and/or other complications such as oligohydramnios. In addition, most of the studies on S G A pregnancies depended on tissue culture for cytogenetic analysis. It is reasonable to expect that chromosomally abnormal cells may be less likely to divide in vitro and produce adequate metaphases, resulting in some cases of C P M being missed with such methods. A s an alternative, methods that not depend on cell divisions can be used, such as comparative genomic hybridization ( C G H ) . For example, A m i e l et al. (2002) used C G H to look for C P M in I U G R pregnancies with other features 'suspicious' of chromosome abnormality as determined by a perinatal pathologist. The majority (61%; 14/24) of the I U G R cases showed chromosome abnormalities, compared to 0/6 controls (Fisher Exact test, p = 0.013); however, in 162  only a few cases was there any attempt to confirm the findings by F I S H . It is possible that selected cases of complicated pregnancy may harbour undiagnosed chromosome abnormalities in the placenta. The mostly normal developmental outcome of C P M 16 suggests that trisomy 16 completely confined to the placenta has minimal postnatal consequences. However, there is evidence from mouse models that an abnormality confined to the placenta can contribute to neurologic affects in the fetus. In two fascinating studies, de Bruin et al. (2003) and W u et al. (2003) showed that retinoblastoma (Rb) knockout mice have extensive placental abnormalities, and that when Rb knockout embryos are 'rescued' by aggregating with a tetraploid embryo to produce a tetraploid placenta (James and West 1994), much of the Rb knockout fetal phenotype, including some o f the neurologic abnormalities, were corrected. This suggests that an abnormal placenta, in this case lacking the R b gene, can contribute to neurologic outcome. There are well-developed mouse models of D o w n Syndrome involving partial trisomy for the region of mouse chromosome 16 (orthologous to human chromosome 21) carried on a marker chromosome, originally produced as a translocation from chromosome 16 (Dierssen et al. 2001). To further address the role of the placenta in neurodevelopment, trisomic embryos that are produced from a balanced carrier parent could be rescued with a tetraploid placenta to determine whether some aspect of the mental retardation could be rectified. In conclusion, this thesis research has further clarified the clinical implications of trisomy confined to the placenta, in particular trisomy 16 C P M , as well as cytogenetic and biological aspects of the pathogenesis of placental mosaicism. It is likely that with properly conducted post-partum studies of the placenta, confined placental trisomy w i l l be more widely recognized as an etiological factor in idiopathic pregnancy complications. In the future, clinical features that make a postnatal diagnosis of C P M more likely should be identified, and in such pregnancies, C P M listed in the differential diagnosis.  9.1  References  A m i e l A , Bouaron N , Kidron D , Sharony R , Gaber E , Fejgin M D (2002) C G H in the detection of confined placental mosaicism ( C P M ) in placentas of abnormal pregnancies. Prenat Diagn 22:752-758 Artan S, Basaran N , Hassa H , Ozalp S, Sener T, Sayli B S , Cengiz C , Ozdemir M , Durak T, Dolen I, et al. (1995) Confined placental mosaicism in term placentae: analysis of 125 cases. Prenat Diagn 15:1135-1142 de Bruin A , W u L , Saavedra H I , W i l s o n P, Yang Y , Rosol T J , Weinstein M , Robinson M L , Leone G (2003) Rb function in extraembryonic lineages suppresses apoptosis in the C N S of Rb-deficient mice. Proc Natl A c a d Sci U S A 100:6546-6551 Dierssen M , Fillat C , Crnic L , Arbones M , Florez J, Estivill X (2001) Murine models for Down syndrome. Physiol Behav 73:859-871 Galdzicki Z , Siarey R , Pearce R , Stoll J, Rapoport SI (2001) O n the cause of mental retardation in D o w n syndrome: extrapolation from full and segmental trisomy 16 mouse models. Brain Res Brain Res Rev 35:115-145 Grati F R , M i o z z o M , Cassani B , Rossella F , Antonazzo P, Gentilin B , Sirchia S M , M o r i L , Rigano S, Bulfamante G , Cetin I, Simoni G (2005) Fetal and placental chromosomal mosaicism revealed by Q F - P C R in severe I U G R pregnancies. Placenta 26:10-18 Heller J H , Spiridigliozzi G A , Doraiswamy P M , Sullivan J A , Crissman B G , Kishnani PS (2004) Donepezil effects on language in children with D o w n syndrome: results of the first 22week pilot clinical trial. A m J M e d Genet A 130:325-326 Heller J H , Spiridigliozzi G A , Sullivan J A , Doraiswamy P M , Krishnan R R , Kishnani PS (2003) Donepezil for the treatment of language deficits in adults with D o w n syndrome: a preliminary 24-week open trial. A m J M e d Genet A 116:111-116 Hunter P, Smith N , Fernandez J, Tawhai M (2005) Integration from proteins to organs: the IUPS Physiome Project. M e c h Ageing Dev 126:187-192 Hunter P J , Borg T K (2003) Integration from proteins to organs: the Physiome Project. Nat Rev M o l C e l l B i o l 4:237-243 James R M , West J D (1994) A chimaeric animal model for confined placental mosaicism. H u m Genet 93:603-604 Johnson N , Fahey C , Chicoine B , Chong G , Gitelman D (2003) Effects of donepezil on cognitive functioning in D o w n syndrome. A m J Ment Retard 108:367-372 Kalousek D K , D i l l F J (1983) Chromosomal mosaicism confined to the placenta in human conceptions. Science 221:665-667  164  Kennerknecht I, Kramer S, Grab D , Terinde R , Vogel W (1993) A prospective cytogenetic study of third-trimester placentae in small-for-date but otherwise normal newborns. Prenat Diagn 13:257-269 Kishnani P S , Sullivan J A , Walter B K , Spiridigliozzi G A , Doraiswamy P M , Krishnan K R (1999) Cholinergic therapy for Down's syndrome. Lancet 353:1064-1065 Kondoh T, Amamoto N , D o i T, Hamada H , Ogawa Y , Nakashima M , Sasaki H , A i k a w a K , Tanaka T, A o k i M , Harada J, Moriuchi H (2005) Dramatic improvement in Down syndrome-associated cognitive impairment with donepezil. A n n Pharmacother 39:563566 Krishnamoorthy A , Gowen L C , B o l l K E , Knuppel R A , Sciorra L J (1995) Chromosome and interphase analysis of placental mosaicism in intrauterine growth retardation. J Perinatol 15:47-50 Lott IT, Osann K , Doran E , Nelson L (2002) D o w n syndrome and Alzheimer disease: response to donepezil. A r c h Neurol 59:1133-1136 Masuzaki H , M i u r a K , Yoshiura K I , Yoshimura S, N i i k a w a N , Ishimaru T (2004) Detection of cell free placental D N A in maternal plasma: direct evidence from three cases of confined placental mosaicism. J M e d Genet 41:289-292 Prasher V P (2004) Review of donepezil, rivastigmine, galantamine and memantine for the treatment of dementia in Alzheimer's disease in adults with D o w n syndrome: implications for the intellectual disability population. Int J Geriatr Psychiatry 19:509-515 Prasher V P , Adams C , Holder R (2003) L o n g term safety and efficacy of donepezil in the treatment of dementia in Alzheimer's disease in adults with D o w n syndrome: open label study. Int J Geriatr Psychiatry 18:549-551 Prasher V P , Huxley A , Haque M S (2002) A 24-week, double-blind, placebo-controlled trial of donepezil in patients with D o w n syndrome and Alzheimer's disease—pilot study. Int J Geriatr Psychiatry 17:270-278 Stipoljev F , Latin V , K o s M , M i s k o v i c B , Kurjak A (2001) Correlation of confined placental mosaicism with fetal intrauterine growth retardation. A case control study of placentas at delivery. Fetal Diagn Ther 16:4-9 Verp M S , Unger N L (1990) Chorionic chromosome abnormalities and intrauterine growth retardation. J Perinatol 10:52-54 Wilkins-Haug L , Roberts D J , Morton C C (1995) Confined placental mosaicism and intrauterine growth retardation: a case-control analysis of placentas at delivery. A m J Obstet Gynecol 172:44-50 Wolstenholme J, Rooney D E , Davison E V (1994) Confined placental mosaicism, I U G R , and adverse pregnancy outcome: a controlled retrospective U . K . collaborative survey. Prenat Diagn 14:345-361 165  W u L , de Bruin A , Saavedra H I , Starovic M , Trimboli A , Yang Y , Opavska J, W i l s o n P, Thompson J C , Ostrowski M C , Rosol T J , Woollett L A , Weinstein M , Cross J C , Robinson M L , Leone G (2003) Extra-embryonic function of Rb is essential for embryonic development and viability. Nature 421:942-947  Appendix B Extra references References for C P M cases in Table 2.1 in Chapter 2 Johnson M P , Childs M D , Robichaux III A G , Isada N B , Pryde P G , Koppitch III F C , Evans M I (1993) Viable pregnancies after diagnosis of trisomy 16 by C V S : lethal aneuploidy compartmentalized to the trophoblast. Fetal Diagn Ther 8:102-108. Kalousek D K , Langlois S, Barrett IJ, Y a m I, W i l s o n D R , Howard-Peebles P N , Johnson M P , Giorgiutti E (1993) Uniparental disomy for chromosome 16 in humans. A m J H u m Genet 52:8-16. Kalousek D K , Langlois S, Robinson W P , Telenius A , Bernard L , Barrett IJ, HowardPeebles P N , W i l s o n R D (1996) Trisomy 7 C V S mosaicism: pregnancy outcome, placental and D N A analysis in 14 cases. A m J M e d Genet 65:348-352. Kuchinka B D , Barrett IJ, M o y a G , Sanchez J M , Langlois S, Y o n g S L , Kalousek D K , Robinson W P (2001) T w o cases of confined placental mosaicism for chromosome 4, including one with maternal uniparental disomy. Prenat Diagn 21:36-39 Penaherrera M S , Barrett IJ, Brown C J , Langlois S, Y o n g S - L , Lewis S, Bruyere H , Howard-Peebles P N , Kalousek D K , Robinson W P (2000) A n association between skewed X-chromosome inactivation and abnormal outcome in mosaic trisomy 16 confined predominantly to the placenta. C l i n Genet 58:436-446. Robinson W P , Barrett IJ, Bernard L , Telenius A , Bernasconi F , W i l s o n R D , Best R G , Howard-Peebles P N , Langlois S, Kalousek D K (1997) Meiotic origin of trisomy in confined placental mosaicism is correlated with presence of fetal uniparental disomy, high levels of trisomy in trophoblast, and increased risk of fetal intrauterine growth restriction. A m J H u m Genet 60:917-927. Shaffer L G , Langlois S, M c C a s k i l l C , M a i n D M , Robinson W P , Barrett IJ, Kalousek D K (1996) Analysis of nine pregnancies with confined placental mosaicism for trisomy 2. Prenat Diagn 16:899-905  References for CPM16 cases in Methods in Chapter 3 Excluded CPM16 cases of paternal origin of the trisomy Kohlhase J, Janssen B , Weidenauer K , Harms K , Bartels I (2000) First confirmed case with paternal uniparental disomy of chromosome 16. A m J M e d Genet 91:190-191. Paulyson K J , Sherer D M , Christian S L , Lewis K M , Ledbetter D M , Salafia C M , Meek J M (1996) Prenatal diagnosis of an infant with mosaic trisomy 16 of paternal origin. Prenat Diagn 16:1021-1026.  168  Excluded CPM16 cases of partial trisomy Devi A S , Egan J F X , Campbell W , Ingardia C , Rosengren S, Tezcan K , Weiser J , Benn P A (1997) Poor pregnancy outcome and the presence of trisomy 16 cells in amniotic fluid. A m J H u m Genet 61:A151. Hsu W T , Shchepin D A , M a o R, Berry-Kravis E , Garber A P , Fischel-Ghodsian N , Falk R E , Carlson D E , Roeder E R , Leeth E A , Hajianpour M J , Wang J-C C , Rosenblum-Vos L S , Bhatt S D , Karson E M , H u x C H , Trunca C , Bialer M G , L i n n S K , Schreck R R (1998) Mosaic trisomy 16 ascertained through amniocentesis: evaluation of 11 new cases. A m J M e d Genet 80:473-480. Schinzel A , Kotzot D , Brecevic L , Robinson W P , Dutly F , Dauwerse H , Binkert F , Baumer A , Ausserer B (1997) Trisomy first, translocation second, uniparental disomy and partial trisomy third: a new mechanism for complex chromosomal aneuploidy. Eur J H u m Genet 5:308-314. Excluded CPM16 case of concomitant aneuploidy Robinson W P , Barrett IJ, Bernard L , Telenius A , Bernasconi F , W i l s o n R D , Best R G , Howard-Peebles P N , Langlois S, Kalousek D K (1997) Meiotic origin of trisomy i n confined placental mosaicism is correlated with presence of fetal uniparental disomy, high levels of trisomy in trophoblast, and increased risk of fetal intrauterine growth restriction. A m J H u m Genet 60:917-927. CPM16 cases from U B C study with some data previously published Kalousek D K , Howard-Peebles P N , Olson S B , Barrett IJ, Dorfmann A , Black S H , Schulman J D , Wilson R D (1991) Confirmation of C V S mosaicism in term placentae and high frequency of intrauterine growth retardation association with confined placental mosaicism. Prenat Diagn 11:743-450. Kalousek D K , Langlois S, Barrett IJ, Y a m I, Wilson D R , Howard-Peebles P N , Johnson M P , Giorgiutti E (1993) Uniparental disomy for chromosome 16 in humans. A m J H u m Genet 52:8-16. Robinson W P , Barrett IJ, Bernard L , Telenius A , Bernasconi F , Wilson R D , Best R G , Howard-Peebles P N , Langlois S, Kalousek D K (1997) Meiotic origin of trisomy in confined placental mosaicism is correlated with presence of fetal uniparental disomy, high levels of trisomy in trophoblast, and increased risk of fetal intrauterine growth restriction. A m J H u m Genet 60:917-927. Penaherrera M S , Barrett IJ, Brown C J , Langlois S, Y o n g S - L , Lewis S, Bruyere H , Howard-Peebles P N , Kalousek D K , Robinson W P (2000) A n association between skewed X-chromosome inactivation and abnormal outcome in mosaic trisomy 16 confined predominantly to the placenta. C l i n Genet 58:436-446. Stavropoulos D J , B i c k D , Kalousek D K (1998) Molecular cytogenetic detection of  169  confined gonadal mosaicism in a conceptus with trisomy 16 placental mosaicism. A m J H u m Genet 63:1912-1914. CPM16 cases from U B C study overlapping with papers by other research groups Chan Y , Silverman N , Jackson L , Wapner R, Wallerstein R (2000) Maternal uniparental disomy of chromosome 16 and body stalk anomaly. A m J M e d Genet 94:284-286. Dworniczak B , Koppers B , Kurlemann G , Holzgreve W , Horst J, M i n y P (1992) Uniparental disomy with normal phenotype. Lancet 340:1285. Holzgreve R, Exeler R, Holzgreve W , Wittwer B , M i n y P (1992) Non-viable trisomies confined to the placenta leading to poor pregnancy outcome. Prenat Diagn 12 (Suppl):S95. Hsu L Y F , Y u M - T , Neu R L , V a n Dyke D L , Benn P A , Bradshaw C L , Shaffer L G , Higgins R R , Khodr G S , Morton C C , Wang H , Brothman A R , Chadwick D , Disteche C M , Jenkins L S , Kalousek D K , Pantzer T J , Wyatt P (1997) Rare trisomy mosaicism diagnosed in amniocytes, involving an autosome other than chromosomes 13, 18, 20, and 21: karyotype/phenotype correlations. Prenat Diagn 17:201-242. Hsu W T , Shchepin D A , M a o R, Berry-Kravis E , Garber A P , Fischel-Ghodsian N , Falk R E , Carlson D E , Roeder E R , Leeth E A , Hajianpour M J , Wang J-C C , Rosenblum-Vos L S , Bhatt S D , Karson E M , H u x C H , Trunca C , Bialer M G , L i n n S K , Schreck R R (1998) Mosaic trisomy 16 ascertained through amniocentesis: evaluation of 11 new cases. A m J M e d Genet 80:473-480. Johnson M P , Childs M D , Robichaux U l A G , Isada N B , Pryde P G , Koppitch III F C , Evans M I (1993) Viable pregnancies after diagnosis o f trisomy 16 by C V S : lethal aneuploidy compartmentalized to the trophoblast. Fetal Diagn Ther 8:102-108. Kennerknecht I, Terinde R (1990) Intrauterine growth retardation associated with chromosomal aneuploidy confined to the placenta. Three observations: triple trisomy 6, 21, 22; trisomy 16; and trisomy 18. Prenat Diagn 10:539-544. Schneider A S , Bischoff F Z , M c C a s k i l l C , Coady M L , Stopfer J E , Shaffer L G (1996) Comprehensive 4-year follow-up on a case o f maternal heterodisomy for chromosome 16. A m J M e d Genet 66:204-208. Schwinger E , Seidl E , K l i n k F , Rehdar H (1989) Chromosome mosaicism of the placenta: a cause of developmental failure of the fetus? Prenat Diagn 9:639-647. Verp M S , Rosinsky B , Sheikh Z , Amarose A P (1989) Non-mosaic trisomy 16 confined to v i l l i . Lancet 2:915-916. Wolstenholme J (1995) A n audit of trisomy 16 in man. Prenat Diagn 15:109-121. W o o V , Bridge P J , Bamforth JS (1997) Maternal uniparental disomy for chromosome 16: case report. A m J M e d Genet 70:387-390. 170  Zimmerman R , Lauper U , Streichier A , Huch R , Hugh A (1995) Elevated alphafetoprotein and human chorionic gonadotropin as a marker for placental trisomy 16 in the second trimester? Prenat Diagn 15:1121-1124. CPM16 cases from other published reports to date Abu-Amero S N , A l i Z , Abu-Amero K K , Stanier P, Moore G E (1999) A n analysis of common isodisomic regions in five m U P D 16 probands. J M e d Genet 36:204-207. Association of Clinical Cytogenetics Working Party on Chorionic V i l l i in Prenatal Diagnosis (1994) Cytogenetic analysis of chorionic v i l l i for prenatal diagnosis: A n A C C collaborative study of U . K . data. Prenat Diagn 14:363-379. Astner A , Schwinger E , Caliebe A , Jonat W , Gembruch U (1998) Sonographically detected fetal and placental abnormalities associated with trisomy 16 confined to the placenta. A case report and review of the literature. Prenat Diagn 18:1308-1315. Benn P (1998) Trisomy 16 and trisomy 16 mosaicism: A review. A m J M e d Genet 79:121-133. Benn P, Craffey A , H o m e D , Cusick W , Smeltzer J (1995) A n association between trisomy 16 (and other fetal aneuploidy) in women with grossly elevated second trimester maternal serum human chorionic gonadotropin ( M S H C G ) . A m J H u m Genet 57:A275. Bennett P, Vaughan J, Henderson D , Loughna S, Moore G (1992) Association between confined placental trisomy, fetal uniparental disomy, and early intrauterine growth retardation. Lancet 340:1284-1285. Brandenburg H , L o s F J , In't V e l d P (1996) Clinical significance of placenta-confined nonmosaic trisomy 16. A m J Obstet Gynecol 174(5): 1663-1664. Callen D F , Korban G , Dawson G , Gugasyan L , Krumins E J M , Eichbaum S, Petrass J, Purvis-Smith S, Smith A , Dendulk G , Martin N (1988) Extraembryonic/fetal karyotypic discordance during diagnostic chorionic villus sampling. Prenat Diagn 8:453-460. Caspari D , Bartels I, Rauskolb R , Prange G , Osmers R , Eiben B (1994) Discrepant karyotypes after second- and third-trimester combined placentesis/amniocentesis. Diagn 14:569-576.  Prenat  Chen C P , Shih J C , Chern S R , Lee C C , Wang W (2004) Prenatal diagnosis of mosaic trisomy 16 associated with congenital diaphragmatic hernia and elevated maternal serum alpha-fetoprotein and human chorionic gonadotrophin. Prenat Diagn Jan;24(l):63-6. Davies G A L , G a d I K , Diamond T, Papenhausen P (1995) Discordant maternal serum and amniotic fluid alpha-fetoprotein results i n mosaic trisomy 16 pregnancies. A m J H u m Genet 57:A278. D e v i A S , Egan J F X , Campbell W , Ingardia C , Rosengren S, Tezcan K , Weiser J, Benn 171  P A (1997) Poor pregnancy outcome and the presence of trisomy 16 cells in amniotic fluid. A m J H u m Genet 61:A151. Devi A S , Kamath M V , Eisenfield L , Neu R, Ciarleglio L , Greenstein R, Benn P (1992) Mosaic trisomy 16 in the newborn: A recognizable syndrome. A m J H u m Genet 51:A1193. Devi A S , Velinov M , Kamath M V , Eisenfield L , Neu R, Ciarleglio L , Greenstein R, Benn P (1993) Variable clinical expression of mosaic trisomy 16 in the newborn infant. A m J M e d Genet 47:294-298. Dorfmann A D , Perszyk J, Robinson P, Black S H , Schuman J D (1992) Rare non-mosaic trisomies in chorionic villus tissue not confirmed at amniocentesis. Prenat Diagn 12:899-902. Farra C , Giudicelli B , Pellissier M C , Philip N , Piquet C (2000) Fetoplacental chromosomal discrepancy. Prenat Diagn 20:190-193 Fryburg JS, Dimaio M S , Mahoney M J (1992) Postnatal placental confirmation of trisomy 2 and trisomy 16 detected at chorionic villus sampling: a possible association with intrauterine growth retardation and elevated maternal serum alpha-fetoprotein. Prenat Diagn 12:157-162 Fryburg JS, Dimaio M S , Yang-Feng T L , Mahoney M J (1993) Follow-up of pregnancies complicated by placental mosaicism diagnosed by chorionic villus sampling. Prenat Diagn 13:481-494 Garber A , Carlson D , Schreck R, Fischel-Ghodsian N , Hsu W T , Oeztas S, Pepkowitz S, Graham J M , Jr. (1994) Prenatal diagnosis and dysmorphic findings in mosaic trisomy 16. Prenat Diagn 14:257-266 Gollop T R , Piere P De C , Naccache N F , Bittencourt E A (1990) Brazilian chorionic villus sampling ( C V S ) : Experience with 900 cases. A m J H u m Genet 47:A1087. Groli C , Cerri V , Tarantini M , Bellotti D , Jacobello C , Gianello R, Zanini R, Lancetti S, Zaglio S (1996) Maternal serum screening and trisomy 16 confined to the placenta. Prenat Diagn 16:685-689. Hajianpour M J (1995) Postnatally confirmed trisomy 16 mosaicism: follow-up on a previously reported patient. Prenat Diagn 15:877-879. Hajianpour M J , Randolph L M , Parvizpour D , Habibian R (1992) Trisomy 16 mosaicism in amniotic fluid cells and poor pregnancy outcome associated with unexplained elevated maternal serum alpha-fetoprotein. A m J H u m Genet 51:A409. Hashish A F , M o n k N A , Lovell-Smith M P F , Bardwell L M , Fiddes T M , Gardner R J M (1989) Trisomy 16 detected at chorion villus sampling. Prenat Diagn 9:427-432. Hogge W A , Schonberg S A , Golbus M S (1986) Chorionic villus sampling: Experience of 172  the first 1,000 cases. A m J Obstet Gynecol 154:1249-1252. Holzgreve R, Exeler R, Holzgreve W , Wittwer B , M i n y P (1992) Non-viable trisomies confined to the placenta leading to poor pregnancy outcome. Prenat Diagn 12 (Suppl):S95. Hsu L Y F , Y u M - T , Neu R L , V a n D y k e D L , Benn P A , Bradshaw C L , Shaffer L G , Higgins R R , Khodr G S , Morton C C , Wang H , Brothman A R , Chadwick D , Disteche C M , Jenkins L S , Kalousek D K , Pantzer T J , Wyatt P (1997) Rare trisomy mosaicism diagnosed in amniocytes, involving an autosome other than chromosomes 13, 18, 20, and 21: karyotype/phenotype correlations. Prenat Diagn 17:201-242. Hsu W T , Shchepin D A , M a o R, Berry-Kravis E , Garber A P , Fischel-Ghodsian N , Falk R E , Carlson D E , Roeder E R , Leeth E A , Hajianpour M J , Wang J-C C , Rosenblum-Vos L S , Bhatt S D , Karson E M , H u x C H , Trunca C , Bialer M G , L i n n S K , Schreck R R (1998) Mosaic trisomy 16 ascertained through amniocentesis: evaluation of 11 new cases. A m J M e d Genet 80:473-480. Huff D S , Watkins C , Davis G , Wallerstein D , Lee M , Dyer K , M c M o r r o w L E (1991) Mosaic trisomy 16 detected by mid-trimester amniocentesis. A m J H u m Genet Suppl 49:174. Jalal S, Lindor N M , Bonde D , Karnes P (1992) Trisomy 16 mosaicism in a six month old infant. A m J H u m Genet 51: A290. Johnson A , Wapner R J , Davis G H , Jackson L G (1990) Mosaicism i n chorionic villus sampling: an association with poor perinatal outcome. Obstet Gynecol 75:573-577. Johnson M P , Childs M D , Robichaux HI A G , Isada N B , Pryde P G , Koppitch III F C , Evans M I (1993) Viable pregnancies after diagnosis of trisomy 16 by C V S : lethal aneuploidy compartmentalized to the trophoblast. Fetal Diagn Ther 8:102-108. Johnson P, Duncan K , Blunt S, B e l l G , A l i Z , C o x P, Moore G E (2000) Apparent confined placental mosaicm of trisomy 16 and multiple fetal anomalies: case report. Prenat Diagn 20:417-421. Kalousek D K , Howard-Peebles P N , Olson S B , Barrett U , Dorfmann A , Black S H , Schulman J D , Wilson R D (1991) Confirmation of C V S mosaicism in term placentae and high frequency of intrauterine growth retardation association with confined placental mosaicism. Prenat Diagn 11:743-450. Leschot N J , W o l f H (1991) Is placental mosaicism associated with poor perinatal outcome? Prenat Diagn 11:403-404. Leschot N J , W o l f H , Verjaal M , V a n Prooijen-Knegt L C , De Boer E G , Kanhai H H H , Christiaens G C M L (1987) Chorionic villi sampling: cytogenetic findings in 500 pregnancies. B M J 295:407-410. Lindor N M , Jalal S M , Thibodeau S N , Bonde D , Sauser K L , Karnes PS (1993) Mosaic 173  trisomy 16 in a thriving infant; maternal heterodisomy for chromsome 16. C l i n Genet 44:185-189. Moore, G E , A l i Z , Khan R U , Blunt S, Bennett P R , Vaughan JI (1997) The incidence of uniparental disomy associated with intrauterine growth retardation in a cohort of thirtyfive severely affected babies. A m J Obstet Gynecol 176(2):294-299. Morssink L P , Sikkema-Raddatz B , Beekhuis JR, Dewolf B T H M , Mantingh A (1996) Placental mosaicism is associated with unexplained second trimester elevation of M S h C G levels but not with elevation of M S A F P levels. Prenat Diagn 16:845-851. O'Riordan S, Greenough A , Moore G E , Bennett P, Nicolaides K H (1996) Case report: uniparental disomy 16 in association with congenital heart disease. Prenat Diagn 16:963-965. Pletcher B A , Sanz M M , Schlessel JS, Kunaporn S, Mckenna C , Bialer M G , Alonso M L , Zaslav A - L , B r o w n W T , Ray J H (1994) Postnatal confirmation of prenatally diagnosed trisomy 16 mosaicism in two phenotypically abnormal liveborns. Prenat Diagn 14:933940. Post J G , Nijhuis J G (1992) Trisomy 16 confined to the placenta. Prenat Diagn 12:10011007. Roeder E , Immken L , Bansal V , Favou A , Hersney D , Tabors E , Curry C J R (1994) N e w clinical findings in mosaic trisomy 16. A m J M e d Genet 52:383. Roland B , L y n c h L , Berkowitz G , Zinberg R (1994) Confined placental mosaicism in C V S and pregnancy outcome. Prenat Diagn 14:589-593. Rosenblum-Vos L S , Roberson A E , Meyers C M , Cohen M M (1993) Trisomy 16 mosaicism identified in mid-trimester amniocentesis and confirmed in fetal tissues. A m J H u m Genet 53:1808. Sanchez J M , Lopez De Diaz S, Panal M J , M o y a G , Kenny A , Iglesias D , Wolstenholme J (1997) Severe fetal malformations associated with trisomy 16 confined to the placenta. Prenat Diagn 17:777-779. Simoni G , Brambati B , M a g g i F , Jackson L (1992) Trisomy 16 confined to chorionic v i l l i and unfavourable outcome of pregnancy. A n n Genet 35:110-112. Simoni G , G i m e l l i G , Cuoco C , Romitti L , Terzoli G , Guerneri S, Rossella F , Pescetto L , Pezzolo A , Porta S, Brambat B , Porro E , Fraccaro M (1986) First trimester fetal karyotyping: one thousand diagnoses. H u m Genet 72:203-209. Simoni G , G i m e l l i G , Cuoco C , Terzoli G L , Rossella F , Romitti L , Dalpra L , Nocera G , Tibiletti M G , Tenti P, Fraccaro M (1985) Discordance between prenatal cytogenetic diagnosis after chorionic v i l l i sampling and chromosomal constitution of the fetus. In: Fraccaro M , Simoni G , Brambati B (eds) First Trimester Fetal Diagnosis. Berlin: Springer-Verlag 137-143. 174  Sirchia S M , Garagiola I, Colucii G , Guerneri S, Lalatta F , Grimoldi M G , Simoni G (1998) Trisomic zygote rescue revealed by D N A polymorphism analysis in confined placental mosaicism. Prenat Diagn 18:201-206. Smith R, Zackai E H , Donnenfeld A E (1997) Prenatally diagnosed trisomy 16 mosaicism which escapes postnatal detection in an infant with congenital anomalies. A m J H u m Genet 61 :A140. Sundberg K , Smidt-Jensen S (1991) Non-mosaic trisomy 16 on chorionic villus sampling but normal placenta and fetus after termination. Lancet 337:1233-1234. Sutcliffe M J , Mueller O T , Gallardo L A , Papenhausen P R , Tedesco T A (1993) Maternal isodisomy in a normal 4 6 , X X following trisomic conception. A m J H u m Genet 53:A1464. Tantravahi U , Matsumoto C , Delach J, Craffey A , Smeltzer J, Benn P (1996) Trisomy 16 mosaicism in amniotic fluid cell cultures. Prenat Diagn 16:749-754. Tharapel A T , Elias S, Shulman L P , Seely L , Emerson D S , Simpson J L (1989) Resorbed co-twin as an explanation for discrepant chorionic villus results: non-mosaic 47,XX,+16 in v i l l i (direct and culture) with normal ( 4 6 , X X ) amniotic fluid and neonatal blood. Prenat Diagn 9:467-472. V a n Opstal D , van den Berg C , Galjaard R J H , L o s F J (2001) Follow-up investigations in uncultured amniotic fluid cells after uncertain cytogenetic results. Prenat Diagn 21:7580. V a n Opstal D , van den Berg C , Deelen W H , Brandenburg H , Cohen-Overbeek T E , Halley D J J , V a n Den Ouweland A M W , In't V e l d P A , Los F J (1998) Prospective prenatal investigations on potential uniparental disomy in cases of confined placental mosaicism. Prenat Diagn 18:35-44. Vaughan J, A l i Z , Bower S, Bennett P, Chard T, Moore G (1994) Human maternal uniparental disomy for chromosome 16 and fetal development. Prenat Diagn 14:751756. Wang B T , Peng W , Cheng K - T , Chiu S-F, H o W , Khan Y , Wittman M , Williams J (1994) Chorionic villus sampling: laboratory experience with 4,000 consecutive cases. A m J M e d Genet 53:307-316. Wang J - C , Mamunes P, K o u S - Y , M a o R, Schmidt J, Habibian R, Hsu W T (1997) Centromeric D N A break in a 10; 16 whole arm translocation associated with trisomy 16 confined placental mosaicism and maternal uniparental disomy for chromosome 16. A m J H u m Genet 61 :A142. Wang J - C C , Mamunes P, K o u S - Y , Schmidt J, M a o R, Hsu W - T (1998) Centromeric  175  D N A break in a 10; 16 reciprocal translocation associated with trisomy 16 confined placental mosaicsm and maternal uniparental disomy for chromosome 16. A m J M e d Genet 80:418-422. Watson J D , W a r d B E , Peakman D , Henry G (1988) Trisomy 16 and 12 confined chorionic mosaicism i n liveborn infants with multiple anomalies. A m J H u m Genet 43:A252. Whiteford M L , Coutts J , A l - R o o m i L , Mather A , Lowther G , Cooke A , Vaughan JI, Moore G E , Tolmie J L (1995) Uniparental disomy for chromosome 16 in a growthretarded infant with congenital heart disease. Prenat Diagn 15:579-584. Williams III J , Wang B B T , Rubin C H , Clark R D , Mohandas T K (1992) Apparent nonmosaic trisomy 16 i n chorionic v i l l i : diagnostic dilemma or clinically significant finding? Prenat Diagn 12:163-168. Wolstenholme J , Rooney D E , Davison E V (1994) Confined placntal mosaicism, I U G R , and adverse pregnancy outcome: a controlled retrospective U . K . collaborative study. Prenat Diagn 14:345-361. Zimmerman R , Lauper U , Streichier A , Huch R , Hugh A (1995) Elevated alphafetoprotein and human chorionic gonadotropin as a marker for placental trisomy 16 i n the second trimester? Prenat Diagn 15:1121-1124.  References for CPM16 cases in Table 3.1 from Chapter 3 Gallentine M L , Morey A F , Thompson, I M , Jr (2001) Hypospadias: A contemporary epidemiologic assessment. Urology 57:788-790. Hoffman JI, Kaplan S (2002) The incidence of congenital heart disease. J A m C o l l Cardiol 39(12): 1890-900.  References for CPM16 cases in Table 4.1 in Chapter 4 Hsu L Y F , Y u M - T , Neu R L , V a n D y k e D L , Benn P A , Bradshaw C L , Shaffer L G , Higgins R R , Khodr G S , Morton C C , Wang H , Brothman A R , Chadwick D , Disteche C M , Jenkins L S , Kalousek D K , Pantzer T J , Wyatt P (1997) Rare trisomy mosaicism diagnosed i n amniocytes, involving an autosome other than chromosomes 13, 18, 20, and 21: karyotype/phenotype correlations. Prenat Diagn 17:201-242. Johnson M P , Childs M D , Robichaux U l A G , Isada N B , Pryde P G , Koppitch III F C , Evans M I (1993) Viable pregnancies after diagnosis of trisomy 16 by C V S : lethal aneuploidy compartmentalized to the trophoblast. Fetal Diagn Ther 8:102-108. Kalousek D K , Langlois S, Barrett IJ, Y a m I, W i l s o n D R , Howard-Peebles P N , Johnson M P , Giorgiutti E (1993) Uniparental disomy for chromosome 16 in humans. A m J H u m Genet 52:8-16. 176  Kennerknecht I, Terinde R (1990) Intrauterine growth retardation associated with chromosomal aneuploidy confined to the placenta. Three observations: triple trisomy 6, 21, 22; trisomy 16; and trisomy 18. Prenat Diagn 10:539-544. Penaherrera M S , Barrett IJ, B r o w n C J , Langlois S, Y o n g S - L , Lewis S, Bruyere H , Howard-Peebles P N , Kalousek D K , Robinson W P (2000) A n association between skewed X-chromosome inactivation and abnormal outcome in mosaic trisomy 16 confined predominantly to the placenta. C l i n Genet 58:436-446. Robinson W P , Barrett IJ, Bernard L , Telenius A , Bernasconi F , Wilson R D , Best R G , Howard-Peebles P N , Langlois S, Kalousek D K (1997) Meiotic origin of trisomy in confined placental mosaicism is correlated with presence of fetal uniparental disomy, high levels of trisomy in trophoblast, and increased risk of fetal intrauterine growth restriction. A m J H u m Genet 60:917-927. Schneider A S , Bischoff F Z , M c C a s k i l l C , Coady M L , Stopfer J E , Shaffer L G (1996) Comprehensive 4-year follow-up on a case of maternal heterodisomy for chromosome 16. A m J M e d Genet 66:204-208. Schwinger E , Seidl E , K l i n k F , Rehdar H (1989) Chromosome mosaicism of the placenta: a cause of developmental failure of the fetus? Prenat Diagn 9:639-647. Verp M S , Rosinsky B , Sheikh Z , Amarose A P (1989) Non-mosaic trisomy 16 confined to v i l l i . Lancet 2:915-916. Wolstenholme J (1995) A n audit of trisomy 16 in man. Prenat Diagn 15:109-121. W o o V , Bridge P J , Bamforth JS (1997) Maternal uniparental disomy for chromosome 16: case report. A m J M e d Genet 70:387-390.  References for CPM16 cases in Table 5.1 in Chapter 5 D e v i A S , Velinov M , Kamath M V , Eisenfield L , Neu R , Ciarleglio L , Greenstein R, Benn P (1993) Variable clinical expression of mosaic trisomy 16 in the newborn infant. A m J M e d Genet 47:294-298. Dorfmann A D , Perszyk J, Robinson P, Black S H , Schuman J D (1992) Rare non-mosaic trisomies in chorionic villus tissue not confirmed at amniocentesis. Prenat Diagn 12:899-902. Johnson M P , Childs M D , Robichaux III A G , Isada N B , Pryde P G , Koppitch III F C , Evans M I (1993) Viable pregnancies after diagnosis of trisomy 16 by C V S : lethal aneuploidy compartmentalized to the trophoblast. Fetal Diagn Ther 8:102-108. Garber A , Carlson D , Schreck R, Fischel-Ghodsian N , Hsu W T , Oeztas S, Pepkowitz S, Graham J M , Jr. (1994) Prenatal diagnosis and dysmorphic findings in mosaic trisomy 16. Prenat Diagn 14:257-266 177  Hajianpour M J (1995) Postnatally confirmed trisomy 16 mosaicism: follow-up on a previously reported patient. Prenat Diagn 15:877-879. Hsu L Y F , Y u M - T , Neu R L , V a n D y k e D L , Benn P A , Bradshaw C L , Shaffer L G , Higgins R R , Khodr G S , Morton C C , Wang H , Brothman A R , Chadwick D , Disteche C M , Jenkins L S , Kalousek D K , Pantzer T J , Wyatt P (1997) Rare trisomy mosaicism diagnosed in amniocytes, involving an autosome other than chromosomes 13, 18, 20, and 21: karyotype/phenotype correlations. Prenat Diagn 17:201-242. Hsu W T , Shchepin D A , M a o R, Berry-Kravis E , Garber A P , Fischel-Ghodsian N , Falk R E , Carlson D E , Roeder E R , Leeth E A , Hajianpour M J , Wang J-C C , Rosenblum-Vos L S , Bhatt S D , Karson E M , H u x C H , Trunca C , Bialer M G , L i n n S K , Schreck R R (1998) Mosaic trisomy 16 ascertained through amniocentesis: evaluation of 11 new cases. A m J M e d Genet 80:473-480. Lindor N M , Jalal S M , Thibodeau S N , Bonde D , Sauser K L , Karnes PS (1993) Mosaic trisomy 16 in a thriving infant; maternal heterodisomy for chromsome 16. C l i n Genet 44:185-189. Penaherrera M S , Barrett IJ, Brown C J , Langlois S, Y o n g S - L , Lewis S, Bruyere H , Howard-Peebles P N , Kalousek D K , Robinson W P (2000) A n association between skewed X-chromosome inactivation and abnormal outcome in mosaic trisomy 16 confined predominantly to the placenta. C l i n Genet 58:436-446. Pletcher B A , Sanz M M , Schlessel JS, Kunaporn S, Mckenna C , Bialer M G , Alonso M L , Zaslav A - L , Brown W T , Ray J H (1994) Postnatal confirmation of prenatally diagnosed trisomy 16 mosaicism in two phenotypically abnormal liveborns. Prenat Diagn 14:933940. Schneider A S , Bischoff F Z , M c C a s k i l l C , Coady M L , Stopfer J E , Shaffer L G (1996) Comprehensive 4-year follow-up on a case of maternal heterodisomy for chromosome 16. A m J M e d Genet 66:204-208. Simensen R J , Colby R S , Corning K J (2003) A prenatal counseling conundrum: mosaic trisomy 16. A case study presenting cognitive functioning and adaptive behavior. Genet Couns 14:331-336. Williams U l J, Wang B B T , Rubin C H , Clark R D , Mohandas T K (1992) Apparent nonmosaic trisomy 16 in chorionic v i l l i : diagnostic dilemma or clinically significant finding? Prenat Diagn 12:163-168. W o o V , Bridge P J , Bamforth JS (1997) Maternal uniparental disomy for chromosome 16: case report. A m J M e d Genet 70:387-390.  178  Appendix C Protein kinases profiled in Chapter 8 Kinase 3-phosphoinositide dependent protein kinase 1 (PKB kinase) Aurora2 Cancer Osaka thyroid oncogene (Tpl2) Casein kinase 2 Cyclin-dependent kinase 1 (cdc2) Cyclin-dependent kinase 2 Cyclin-dependent kinase 4 Cyclin-dependent kinase 5 Cyclin-dependent kinase 6 Cyclin-dependent kinase 7 Cyclin-dependent kinase 9 DNA-activated protein kinase dsRNA dependent kinase eEF2k Extracellular regulated kinase 1 Extracellular regulated kinase 2 Extracellular regulated kinase 3 Glycogen synthase kinase 3 alpha Glycogen synthase kinase 3 beta IKKbeta MAP Kinase Kinase 1 (MKK1) MAP kinase kinase 2 (MKK2) MAP kinase kinase 4 (MEK4) MAP Kinase Kinase 6 (MEK6) p21 activated kinase 1 (PAK alpha) p21 activated Kinase 3 (PAK beta) p38 Hog MAP kinase Protein kinase B alpha Protein kinase C alpha Protein kinase C Beta] Protein kinase C delta Protein kinase C epsilon Protein kinase C gamma Protein kinase C lambda Protein kinase C mu Protein kinase C theta Protein kinase C zeta RhoA kinase Ribosomal S6 kinase 1 Ribosomal S6 kinase 2 S6 Kinase p70 v-mos Moloney murine sarcoma viral oncogene homolog 1 v-raf murine sarcoma viral oncogene homolog B1 Bone marrow X kinase Bruton agammaglobulinemia tyrosine kinase Calmodulin-dependent kinase 1 Calmodulin-dependent kinase 4 Calmodulin-dependent kinase kinase Casein kinase 1 delta Casein kinase 1 epsilon c-SRC tyrosine kinase Death associated protein kinase 1 Focal adhesion kinase Fyn oncogene related to SRC G protein-coupled receptor kinase 2 (BARK2)  Kinase PDK1 Aurora2 COT CK2 CDK1 CDK2 CDK4 CDK5 CDK6 CDK7 CDK9 DNAPK PKR eEF2k ERK1 ERK2 ERK3 GSK3a GSK3b IKKbeta MEK1 MEK2 MEK4 MEK6 PAK1 PAK3 p38 MAPK PKBa PKCa PKCb PKCd PKCe PKCg PKC1 PKCm PKCt PKCz ROKa RSK1 RSK2 S6K p70 MOS1 RAFB BMX (Etk) BTK CaMKl CaMK4 CaMKK CKld CKle CSK DAPK FAK FYN GRK2  Germinal centre kinase Hematopoietic progenitor kinase 1 Inhibitor NF kB kinase alpha/beta Janus kinase 1 Janus kinase 2 Kinase suppressor of Ras 1 Lymphocyte-specific protein tyrosine kinase Mammalian sterile 20-like 1 MAP kinase interacting kinase 2 Oncogene Lyn Oncogene Raf 1 Oncogene SRC Protein kinase A (cAMP-dependent protein kinase) Protein kinase Gl (cGMP-dependent protein kinase) Protein tyrosine kinase 2 Spleen tyrosine kinase Yamaguchi sarcoma viral oncogene homolog 1 Zeta-chain (TCR) associated protein kinase ZIP kinase (death associated protein kinase 3)  GCK HPK1 IKKa JAK1 JAK2 KSR1 LCK MST1 MNK2 LYN RAF1 SRC PKA PKG1 PYK2 SYK YES1 ZAP70 ZIP  180  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0092891/manifest

Comment

Related Items