UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Genetic and epigenetic factors in a mouse model for multifactorial cleft lip Plamondon, Jenna Ashley 2010

You don't seem to have a PDF reader installed, try download the pdf

Item Metadata

Download

Media
24-ubc_2010_fall_plamondon_jenna.pdf [ 4.37MB ]
Metadata
JSON: 24-1.0071285.json
JSON-LD: 24-1.0071285-ld.json
RDF/XML (Pretty): 24-1.0071285-rdf.xml
RDF/JSON: 24-1.0071285-rdf.json
Turtle: 24-1.0071285-turtle.txt
N-Triples: 24-1.0071285-rdf-ntriples.txt
Original Record: 24-1.0071285-source.json
Full Text
24-1.0071285-fulltext.txt
Citation
24-1.0071285.ris

Full Text

GENETIC AND EPIGENETIC FACTORS IN A MOUSE MODEL FOR MULTIFACTORIAL CLEFT LIP  by  Jenna Ashley Plamondon  B.Sc., Bishop’s University, 2008 A THESIS SUBMITTED IN PARTIAL FULLFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  MASTER OF SCIENCE  in The Faculty of Graduate Studies (Medical Genetics)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  September 2010  ! Jenna Ashley Plamondon, 2010  Abstract Cleft lip (CL/P) is a human birth defect with complex genetic etiology. Two loci are involved in CL/P in the A/WySn mouse model: Wnt9b on Chromosome 11 and clf2 on Chromosome 13. There are two known mutant alleles at Wnt9b; clf1 is a spontaneous recessive mutation caused by an insertion of an IAP transposon near the gene Wnt9b and the second allele is a Wnt9b knock out. Clf2 is the second locus required for the CL/P phenotype. Quantitative PCR was used to investigate expression of other genes that might be affected by the absence of Wnt9b expression. Expression levels of Bmp4, Dkk1, Msx1, Msx2, Raldh3, Sox11, Wnt3, Wnt4 and !-catenin were examined. This experiment detected a significant decrease in expression levels of !-catenin in Wnt9bNull/Null embryos. The clf2 gene has not yet been identified and the function was unknown. Segregants from multi-generation crosses using the Wnt9b knockout, “Cross 1” and “Cross 2”, were examined for CL/P and genotype at polymorphic markers linked to Wnt9b and clf2 to ask whether clf2 modifies the frequency of CL in the Wnt9b null homozygotes. I also used recombinants from a congenic stock, consulted the mouse genome assembly and examined ancestral haplotypes in a strain survey to define the clf2 candidate region and examined mouse genome databases to develop a candidate gene list. My studies have reduced the clf2 candidate region to a 3.0 mb region between Cntnap3 and Ak029746. This region contains 48 genes and a majority of them encode zinc finger proteins. The specific crosses gave unexpected results that interfered with the ability to study the effect of clf2 genotype on penetrance of CL/P; the limited data suggested that clf2 does not affect penetrance of CL/P in the Wnt9b null embryos. However, Cross 2 provided an inverse way to address the role of clf2: to test whether clf2 modifies the methylation of the IAP transposon at Wnt9b in the clf1 mutant allele. Using the COBRA technique we studied the methylation levels !  ""!  of the IAP in Wnt9bclf1/Null embryos segregating at clf2. The results indicated that clf2 modifies the methylation status of the IAP.  !  """!  Table of Contents! Abstract!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!""! Table of Contents!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!"#! List of Tables !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! #""! List of Figures!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! #"""! Glossary !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!"$! %&'()*+,-./,(01 !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! $! 1! Introduction!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!2! 1.1 Cleft Lip with or without Cleft Palate: A Common Human Birth Defect!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 2! 1.2 Morphogenesis of the Upper Lip !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 3! 1.3 Mouse Models with CL/P !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 4! 1.4 A/WySn Mouse Strain!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 5! 1.4.1 History $$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$%! 1.4.2 CL/P in the A/- strain $$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$&! 1.5 Clf1 and Mutations Due to IAP Insertions !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 6! 1.5.1 Wnt9b $$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$'! 1.5.2 Mutations due to IAP insertions $$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ ()! 1.5.3 Methylation at clf1$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ ((! 1.5.4 What are IAPs?$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ ((! 1.6 Clf2 !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!23! 1.7 Major Purposes and Hypotheses !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!27! 1.8 Goals of My Studies !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!28! 1.8.1 Goals for clf1 $$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ (*! 1.8.2 Goals for clf2 $$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ (*! 2! General Materials and Methods !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 24! 2.1 Mouse Husbandry !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!24! 2.2 DNA Preparation and PCR of SSLPs!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!24! 2.3 Specialized Mouse Stocks !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!25! 2.3.1 The WBC stock $$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ (%! 2.3.2 The B.WN stock$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ (%! 2.4 Identification of Recombinants !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!29! 2.5 Observation of CL!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!29! 2.6 Statistical Analysis !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!26! 3! Studies to Test the Function of clf2: Cross 1 and Cross 2 !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 2:! 3.1 Background !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!2:! 3.2 Materials and Methods !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!3;! 3.2.1 Cross 1 $$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ +)! 3.2.2 Cross 2 $$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ +(! 3.2.3 Genotyping of Cross 1 and Cross 2 breeders $$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ ++! 3.2.4 Generation of embryos and observation of CL$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ ++! 3.2.5 Genotyping of embryos $$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ ++! 3.2.6 Test of CL penetrance in B.WN Wnt9bNull/Null embryos $$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ +,! 3.3 Results!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!37! 3.3.1 Cross 1 $$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ +,! 3.3.2 Cross 2 $$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ +-! 3.3.3 Recombinants for the clf2 candidate region $$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ +.! ! "#!  3.3.4 Test of CL penetrance in B.WN Wnt9bNull/Null embryos $$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ +'! 3.4 Discussion!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!3:!  4! Epigenetic Modification of clf2!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 77! 4.1 Background !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!77! 4.2 Materials and Methods !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!78! 4.2.1 Wnt9bNull/clf1 embryos $$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ ,*! 4.2.2 COBRA assay $$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ ,*! 4.4 Discussion!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!7:! 5! Mapping of clf2: W1-U/REC-4 Lines !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 83! 5.1 Background !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!83! 5.2 Materials and Methods !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!87! 5.2.1 Observation of CL $$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ *-! 5.2.2 Genotyping of embryos $$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ *-! 5.3 Results!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!85! 5.3.3 REC-4 test cross $$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ *%! 5.3.4 W1-U test cross $$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ *&! 5.4 Discussion!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!86! 6! Mapping clf2: Haplotype Survey!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 4;! 6.1 Background !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!4;! 6.2 Materials and Methods !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!43! 6.2.1 Haplotype analysis$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ -+! 6.2.2 SNP analysis$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ -+! 6.3 Results!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!47! 6.4 Discussion!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!46! 7! Mapping clf2: Candidate Region Database Search !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 52! 7.1 Background !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!52! 7.2 Materials and Methods !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!53! 7.2.1 UCSC search $$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ %+! 7.2.2 MGI search$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ %,! 7.2.3 SNP query $$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ %,! 7.3 Results!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!57! 7.3.1 Functions of candidate genes $$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ %,! 7.3.2 Expression domains of candidate region $$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ %*! 7.3.3 Potential functional changes in candidate genes. $$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ %-! 7.4 Discussion!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!55! 8 Gene expression studies in Wnt9bNull/Null E10 embryos!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 92! 8.1 Background !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!92! 8.2 Materials and Methods !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!94! 8.2.1 Collection of Wnt9b+/+ and Wnt9bNull/Null E10 embryos$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ &-! 8.2.2 RNA and DNA extraction from embryo heads$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ &&! 8.2.3 cDNA synthesis$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ &&! 8.2.4 Real-time PCR (qPCR) $$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ &.! 8.2.5 Statistical analysis$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ &.! 8.3 Results!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!96! 8.4 Discussion!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!9:! 9 Discussion!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 63! References!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 66! %<<,(-"$!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! ::! ! #!  %<<,(-"$=%>==?@"/,@=A,BC,(&,1!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!::! Appendix B: Cross 2 Data !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 2;;! Appendix C: COBRA data and results!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 2;2! Appendix D: Candidate Genes and their Function!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 2;4! Appendix E: qRT-PCR Primer Sequences !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 2;9! Appendix F: Animal Care Certificates !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 2;6!  !  #"!  List of Tables /0123!,$(!!4566078!9:!;79<<!( $$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ +*! /0123!,$+!!=3>9?8@3<!9:!;79<<!(!361789< $$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ +-! /0123!,$,!!4566078!9:!;79<<!+ $$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ +%! /0123!,$*!!=3>9?8@3<!9:!;79<<!+!361789< $$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ +&! /0123!,$-!!A73B53>C"3<!9:!;D!">!;79<<!+!E3>9?8@3<$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ +.! /0123!,$%!!=3>9?8@3<!9:!73C961">0>?!361789<!">!;79<<!(!0>F!;79<<!+$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ +.! /0123!,$&!!4566078!9:!?3<?!9:!G$HI!;D!@3>3?70>C3$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ +'! /0123!,$.!!=3>9?8@3<!9:!!"#$%&'()**!J!!"#$%&'()**!361789< $$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ +'! /0123!-$(!!4566078!9:!KL;M*!?3<?!C79<<$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ *%! /0123!-$+!!NO3>9?8@3<!0>F!E3>9?8@3<!9:!KL;M*!?3<?C79<<!361789<$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ *%! /0123!-$,!!4566078!9:!H(MP!?3<?MC79<< $$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ *&! /0123!-$*!!NO3>9?8@3<!0>F!E3>9?8@3<!9:!H(MP!?3<?!C79<<!361789<$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ *.! /0123!%$(!!Q0@29?8@3<!9:!#07"95<!695<3!<?70"><!7320?3F!?9!RSH8<>$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ -%! /0123!%$+!!Q0@29?8@3<!9:!;D!361789<!:796!RLT$R!J!RSH84>!?3<?!C79<<$ $$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ -&! /0123!%$,!!Q0@29?8@3!9:!?O3!97"E">02!>9>M;-&GDS%T!CO7969<963!9:!?O3!!"#$%+'U!6023<!0>F! O0@29?8@3<!9:!?O3"7!0>C3<?97<$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ -.! /0123!&$(!!4566078!9:!C0>F"F0?3!E3>3!:5>C?"9>$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ %*! /0123!&$+!!4566078!9:!C0>F"F0?3!E3>3!3V@73<<"9>$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ %-! /0123!&$,!!4566078!9:!4IN!B5378!9:!?O3!,*-./C0>F"F0?3!73E"9>$ $$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ %%! /0123!.$(!!N79@9<3F!E3>3<!?9!<?5F8!">!!"#$%!>522!65?0>?<!0>F!?O3"7!730<9>">E$$$$$$$$$$$$$$$$$$$$$$ &(! /0123!.$+!!R>028W3F!F0?0!:796!K302M/"63!N;K!9:!!"#$%&'&0>F!!"#$%()**'()**!/L()!361789<$&'! /0123!G(!!;79<<!+!F0?0 $$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$())! /0123!;(!!;XGKR!73<52?<$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$(),! /0123!Y(!!;0>F"F0?3!E3>3<!0>F!?O3"7!:5>C?"9>!:796!P;4;!0>F!Z=[!F0?010<3!<307CO$$$$$$$$$$$$()-! /0123!L(!!BK/MN;K!@7"637!<3B53>C3<$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$()&!  !  #""!  List of Figures Figure 1.1 Drawings of bilateral (A) and left unilateral (B) CL/P in humans. .............................. 1 Figure 1.2 Morphogenesis of the human embryonic face............................................................... 4 Figure 1.3 History of the mouse strains related to the A/- strain................................................... 7 Figure 1.4 Mouse genomic region containing the Wnt9b gene and the IAP element in the A/~ strains. ............................................................................................................................................. 9 Figure 3.1 Cross 1 breeding scheme. ........................................................................................... 21 Figure 3.2 Cross 2 breeding scheme ............................................................................................ 22 Figure 4.1 Restriction enzyme cut sites in relation to the Wnt9b-IAP 5’LTR and the Wnt9b gene. ....................................................................................................................................................... 36 Figure 4.2 Example of a COBRA gel after a MboI restriction digest. ........................................ 37 Figure 4.3 Methylation status of the Wnt9b-IAP using MboI. .................................................... 37 Figure 4.4 Methylation status of the Wnt9b-IAP using EcoRI.................................................... 38 Figure 4.5 Phenotype and Genotype correlation of methylation Status of the Wnt9b-IAP using MboI.............................................................................................................................................. 39 Figure 4.6 Phenotype and Genotype correlation of methylation Status of the Wnt9b-IAP using EcoRI. ........................................................................................................................................... 39 Figure 5.1 Clf2 candidate region on Chromosome 13 at the start of my studies.......................... 42 Figure 5.2 Interpretation of location of clf2 based on testcross embryos .................................... 43 Figure 5.3 W1, REC-4 and W1-U breakpoints on Chromosome 13. .......................................... 45 Figure 5.4 The new candidate region for clf2, defined by REC4 and W1-U. .............................. 48 Figure 8.1 Stages of facial prominence fusion by the shape of the nasal pit............................... 72 Figure 8.2 Drawing of dissection of E10 embryos. ..................................................................... 76 Figure C1 MboI gel pictures ...................................................................................................... 101 Figure C2 EcoRI gel pictures..................................................................................................... 102  !  #"""!  Glossary cDNA:  Complementary DNA  CNS:  central nervous system  Bp:  base pair  DNA:  deoxyribonucleic acid  ERV:  Endogenous Retrovirus; retroviruses derived from ancient viral infections  Haplotype:  a combination of alleles at markers that are transmitted together in a contiguous region of a chromosome.  HCl:  hydrochloric acid  ISH:  in situ hybridization  Kb:  Kilobase, 1,000 base pairs  LNP:  Lateral nasal prominence  Mb:  Megabase, 1,000,000 base pairs  Moles:  Pre-implantation mortality up to E11.  MNP:  Medial nasal prominence  MXP:  Maxillary prominence  mRNA:  messenger RNA  PCR:  Polymerase Chain Reaction  qRT-PCR:  quantitative reverse transcriptase PCR  RNA:  Ribonucleic Acid  RT-PCR:  reverse transcriptase PCR  siRNA:  small interfering RNA  SNP:  Single nucleotide polymorphism  Stock:  not fully inbred to strain status  Strain:  inbred the equivalent of greater than 21 generations of brother/sister mating, to full homozygosity at all loci.  UTR:  Untranslated Region  UV:  Ultraviolet  !  "V!  "#$%&'()*+,)%-.! ! First and foremost, I would like to thank my supervisor, Dr. Diana M. Juriloff for providing me with a challenging environment in which to learn and grow into a young scientist in the field of medical genetics. I am appreciative of the one-on-one attention, guidance and support I have received from you; I not only benefitted from your teachings scientifically but also personally as you welcomed me to Vancouver from a small town in Maine. Dr. Muriel Harris, as my committee member and partial supervisor, you have also played a major role in my scientific and personal learning during my studies at UBC. I thank you for all of your time and energy you have invested into my project and personal development. I would also like to extend my gratitude to my committee members, Dr. Dixie Mager and Dr. Joy Richman, for your helpful guidance and comments on my research. A special thank you to Dixie for collaborating with me on parts of my research. My research was aided by the generous support of laboratory technicians. I would like to thank Liane Gagnier, of the Mager Lab, for her technical expertise with COBRA and qRT-PCR. Thank you for the generous amount of time you devoted to my research and being so knowledgeable and patient. I would also like to thank Ron Chan and Matt Trudeau, laboratory technicians in the Juriloff Lab. Thank you for your care and attention to detail while genotyping my samples. I am grateful for the love and support of my family and friends. To my many friends across North America, thank you for the long distance phone calls filled with words of optimism, encouragement and laughter. To my graduate school buddies, thank you for all of our Vancouver adventures, trips and many genetics talks. To Martin, my favorite office buddy, thank you for your computer expertise and afternoon breaks. Thank you to my little brother,  !  V!  Jaime, for your friendship, phone calls and visits to Vancouver. Most of all, I thank my wonderful parents for their endless love, encouragement and support.  !  V"!  1 Introduction 1.1 Cleft Lip with or without Cleft Palate: A Common Human Birth Defect Cleft lip with or without cleft palate (CL/P) is a common human birth defect with complex genetic etiology. Cleft lip is characterized by a unilateral or bilateral gap in the tissue of the upper lip; the gap can extend through the upper lip and jaw and into the nostril (Sperber 2002), figure 1.1. Clefting of the secondary palate, the roof of the oral cavity, sometimes presents as a secondary defect of cleft lip. Cleft palate (CPO) can also occur by itself and is considered developmentally and genetically distinct from CL/P (Fraser 1970). CL/P has an incidence of about 1-2 per thousand births (Fraser 1970, Marazita et al. 2004). This prevalence varies depending on the population; North American Indians in British Columbia and Asian populations have higher frequencies, Caucasians have intermediate frequencies, and African Americans have lower frequencies of CL/P (reviewed by Diewert and Wang 1992). CL/P can be non-syndromic or can present with other defects as a syndrome; approximately 70-80% of CL/P cases are non-syndromic (Saal 2002) and CL/P is a feature in over 200 syndromes, many with single gene etiology (Wong and Hagg 2004). R!  G!  Figure 1.1 Drawings of bilateral (A) and left unilateral (B) CL/P in humans.  !  (!  Non-syndromic CL/P is a multifactorial trait (Fraser 1970) with 3-14 predicted causative loci per person (Schliekelman and Slatkin 2002) all contributing a small effect towards the final clefting phenotype. Most of the factors contributing to CL/P have not yet been identified. Among the genes that have been reported to contribute to the risk of CL/P in some populations are: IRF6 (Chakravarti 2004, Zucchero et al. 2004), a non-gene SNP (rs987525) on Chromosome 8q24.21 (Birnbaum et al. 2009), and linkage studies have identified a factor at the location of WNT9B (Chenevix-Trench et al. 1992, Shaw et al. 1993, Maestri et al. 1997, Peanchitlertkajorn et al. 2003, Marazita et al. 2003, Moreno et al. 2004). A detailed list of human CL/P susceptibility genes can be found in Mossey et al. (2009). To suggest human CL/P candidate genes, three major approaches are being used in humans; linkage studies, genomewide and candidate gene association studies to identify genetic polymorphisms, and searching for variants of genes associated with syndromic forms of CL/P as causative loci in nonsyndromic CL/P. Major signaling pathways involved in the development of the upper lip and pathogenesis of CL/P are the Bmp, Fgf, Shh, and Wnt signaling pathways (Jiang et al. 2006). An environmental component to CL/P also exists adding to the multifactorial nature of this trait (Fraser 1970). Maternal cigarette smoking and maternal folic acid intake are currently the two main environmental factors that interact with genetic factors of CL/P (Vieira 2008). Genetic and environmental factors work together to contribute to defects in facial morphogenesis during embryonic development and these defects lead to CL/P.  1.2 Morphogenesis of the Upper Lip The morphogenesis of the embryonic face has been extensively reviewed in humans and the mouse. Embryonic lip formation and CL/P in the mouse and human is very similar with very few and minor differences (Diewert and Wang 1992). For the purpose of this explanation I will  !  +!  summarize the explanations from Diewert and Wang (1992) on human embryonic face morphogenesis and Trasler (1968) on mouse embryonic face morphogenesis. Embryonic face morphogenesis begins with neural crest cell migration from the neural tube to form the mesenchyme of five facial prominences projecting frontally surrounding the stomodeum, the primitive mouth (figure 1.2a) (Diewert and Wang 1992). Induction of the nasal organs begins by the ectoderm of the frontonasal process thickening on both sides to form the nasal placodes (Diewert and Wang 1992). The frontonasal process continues to grow around the nasal placodes to form oval shaped nasal pits (future nostrils), surrounded by the medial nasal prominence (MNP) and the lateral nasal prominence (LNP) in a horseshoe shape, figure 1.2b (Trasler 1968). As the facial prominences continue to grow the nasal pits are shifted medially from maxillary prominence (MXP) growth and they take on a crescent shape from ventrolateral growth by the MNP, figure 1.2c (Trasler 1968). The LNP and the MNP are in close contact at the bottom edge of the crescent shaped nasal pit and will initiate the fusion process in a posterior to anterior direction to form the nasal fin, an epithelial seam (Diewert and Wang 1992, Trasler 1968). The fusion process gives the nasal pits a comma shape (Trasler 1968). Growth of the MXPs towards the midline brings the MNPs and MXPs into contact and fusion begins (Trasler 1968). As the fusion process continues, growth by the MXPs and MNPs fills in the groove between the right and left MNPs and a smooth upper lip results (Trasler 1968).  !  ,!  R!  G!  ;!  Figure 1.2 Morphogenesis of the human embryonic face. Five facial prominences surrounding the stomodeum (A); Oval stage of face development (B); Crescent stage of face development (C). MXP, Maxillary prominence; MNP, medial nasal prominence; LNP, lateral nasal prominence. The fusion process is a multistep process of cellular transformations that begins with apoptosis of the periderm cells to expose the underlying epithelia (Jiang et al. 2006). Epithelial filopodia bridge the gap between the opposed prominences and a nasal fin will form once they are in contact (Diewert and Wang 1992). The nasal fin is broken down by an unknown mechanism including some combination of apoptosis, epithelial to mesenchymal transformation, or migration of epithelial cells to adjacent epithelia (Diewert and Wang 1992, Jiang et al. 2006). As the nasal fin disintegrates a mesenchymal bridge will form (Diewert and Wang 1992). Failure of the MNP to fuse to the MXP, the LNP, or both results in a lateral cleft lip (Diewert and Wang 1992). Failure of the fusion could be the result of a defect(s) in a number of different mechanisms, explaining the multifactorial nature of the defect. The nasal placodes could be too centrally placed or more medially placed MNPs would enlarge the distance between the MXPs and cause a delay of the fusion process (Juriloff and Harris 2008), delayed or deficient growth of any of the three facial prominences (Diewert and Wang 1992); or an impaired fusion  !  *!  process such as persistence of the nasal fin or retarded mesenchymal proliferation resulting in deficient mesenchymal bridge growth (Diewert and Wang 1992).  1.3 Mouse Models with CL/P Embryonic face development and the CL/P defect in the mouse is very similar to human (Diewert and Wang 1992); mouse models with CL/P direct attention to genes and signaling pathways important for human lip development and to developmental mechanisms that cause CL/P. CL/P is present in fewer than 20 mouse mutants (Juriloff and Harris 2008); a relatively small number in comparison to the over 200 targeted gene knockouts in mice with neural tube defects (NTDs) (Harris and Juriloff 2007). NTDs generally arise in knockouts that were not targeted to generate birth defects whereas finding CL/P in those targeted knockouts is rare. There are a few spontaneous mouse mutants with CL/P, they are the A/- strain, Bent tail, legless, Dancer and Twirler (Juriloff and Harris 2008). Other mouse mutants with CL/P result from null mutations or conditional knockouts, such as: Bmp4 (Liu et al. 2005), Bmpr1a (Liu et al. 2005), Folr1 (Tang and Finnell 2003, Spiegelstein et al. 2004), Sox11 (Sock et al. 2004), Tcfap2a (Nottoli et al. 1998), Wnt9b (Carroll et al. 2005), Lrp6 (Song et al. 2009), Tp63 (Thomason et al. 2008), Alk5 (Li et al. 2008), Rspo2 (Yamada et al. 2009) , and Pax9/Msx1 (Nakatomi et al. 2010). There are also two non-targeted induced mutations with CL/P; the radiation induced XtBph (deletion including Gli3) and the ENU-induced Recessive in “Line 2” (Juriloff and Harris 2008). The known major signaling pathways present in the CL/P mutants are the BMP, WNT, SHH, and TGF" signaling pathways (Juriloff and Harris 2008). The CL/P mouse models that we study are the A/WySn mouse strain, a member of the A/- strains, the Wnt9b null mutant, and the Wnt9bNull/clf1 compound heterozygote.  !  -!  1.4 A/WySn Mouse Strain 1.4.1 History In the late 1800s, scientists discovered that tumors could be transplanted between mice and began using the species to study cancer (Strong 1978). However, the great variability in the results of the studies of cancer genetics in mice made the results hard to decipher and irreproducible. Lionell Strong, a graduate student, argued that an inbred strain of mice was needed to reduce the variability of the data. In the summer of 1920, Strong crossed one of Halsey Bagg’s albino non-inbred (ancestors of the BALB/c strain) females to an albino male from C.C. Little’s commercial stock as the first step in creating an inbred strain of mice through brother/sister mating of the progeny (Strong 1978). It was a rocky, uphill battle marked by lack of funding, co-habitating with the mice in a tent and later a manse and by a faulty coal stove which destroyed the entire colony except one F7 pregnant female, but by the late 1920s Strong had created the first inbred strain of mice, the A strain (A/-) (Strong 1978). The history of the A strain and its descendants and ancestors is diagrammed in figure 1.3.  !  %!  Figure 1.3 History of the mouse strains related to the A/- strain (Juriloff and Harris 2008, Strong 1978). A (+) sign indicates CL/P present in the strain, a (-) sign indicates the strain does not have CL/P.  1.4.2 CL/P in the A/- strain The A/- strain is one of the few mouse models with spontaneous CL/P. CL/P in the A/strain is a digenic trait with a strong maternal effect; the two loci, a recessive clf1, and semidominant clf2, have been mapped to Chromosome 11 and 13, respectively (Juriloff and Mah 1995, Juriloff et al. 2001b, Juriloff et al. 2004). There are at least three substrains that are direct descendants of the original A strain: A/WySn, A/J and A/HeJ (Juriloff and Harris 2008). These substrains all have CL/P but at different frequencies ranging from 2-4% to 25% (Juriloff and Harris 2008). More specifically, the A/WySn mouse model has a CL/P frequency of 15-20% (Juriloff 1982) and a frequency of 1-2% in clf1AA, clf2AB (“A” allele from the A/- strain; “B” allele from C57BL/6J) in backcross animals with an A/WySn mother (Juriloff et al. 2001b). By their shared common ancestry the A/- substrains are assumed to have the same cleft lip susceptibility genes, clf1 and clf2, but they are known to differ for a few polymorphic markers in other locations. A/WySn differs from A/HeJ at the H-12 locus (Juriloff 1982) and A/WySn !  &!  differs from A/J at two Chromosome 13 markers approximately 3.0 mb below the clf2 candidate region, G1831 E/F and G1868 A/B (Juriloff, DM, unpublished data). The differences in frequencies may be attributed to residual differences in genes elsewhere in the genome. The A/WySn-A/J difference appears to be a maternal effect (Juriloff 1982). Studies of CL/P in the A/- strain lineage date back to the late 1920s and include developmental studies and a demonstration of mode of inheritance for CL/P which involves more than one gene locus (reviewed in Juriloff and Harris 2008). Along with clf1 and clf2, the A/- strain has a strong maternal effect that is required for CL/P. Studies of backcrosses following crosses of A/J and C57BL/6J have shown that the frequency of CL/P is extremely low in embryos from an F1 mother compared to reciprocal backcross embryos (A/- strain mother) with genetically comparable litters (Davidson et al. 1969). The maternal effect is of an unknown origin and maternal effect genes have not yet been mapped (Juriloff et al. 2001b).  1.5 Clf1 and Mutations Due to IAP Insertions The recessive clf1 mutation of the A/- strain was mapped to Chromosome 11 and was found to be the result of an Intracisternal A-particle (IAP) element inserted 6.6 kb 3’ of the Wnt9b gene, figure 1.4 (Juriloff et al. 2005a). This finding was confirmed with a complementation test between the clf1 mutation and the Wnt9b null mutation; the study found that clf1 is in fact a mutation of the Wnt9b gene (Juriloff et al. 2006). The IAP inserted is a 5213 bp, type 1 delta 1 (I#1) element (Juriloff et al. 2005a). The frequencies of CL obtained in the complementation test suggested that clf1 is a hypomorphic mutation (Juriloff et al. 2005b). It is unknown exactly how the IAP disrupts the function of the Wnt9b gene; it is hypothesized to cause an antisense mRNA transcript from the bidirectional promoter of the IAP (Juriloff et al. 2006). !  .!  Figure 1.4 Mouse genomic region containing the Wnt9b gene and the IAP element in the A/~ strains. The solid horizontal line represents genomic DNA, the solid vertical bars represent exons, the open vertical bar represents the 3’ UTR of mRNA, the hatched box represents the IAP, the arrows indicate the direction of transcription, and the arrow above the line indicates the hypothetical antisense transcription from the IAP.  1.5.1 Wnt9b Wnt9b is the gene mutated in the A/- strain mouse model for CL/P. Hypomorphic levels of Wnt9b lead to CL/P in the A/- strain and the compound heterozygote (Wnt9bNull/clf1). Wnt9b null mutants (Wnt9bNull/Null) have vestigial kidneys, lack reproductive ducts, and have an incomplete penetrance of CL/P (Carroll et al. 2005).  In the developing kidney, Wnt9b signals  through the Wnt canonical pathway during tubule induction (Carroll et al. 2005), but later during tubule morphogenesis Wnt9b switches to Wnt non-canonical/Polar Cell Polarity signal transduction (Karner et al. 2009). Wnt4 is a direct downstream target of Wnt9b in the kidney; Wnt4 is able to rescue the Wnt9bNull/Null phenotype in the embryonic kidney (Carroll et al. 2005). A detailed spatiotemporal study of Wnt9b expression in the developing mouse face from E9.5- E12.5 was completed by Lan et al. (2006) using in situ hybridization analysis of whole mount embryos and paraffin sections. Details are as follows. Wnt9b is expressed in the surface ectoderm of the head and branchial arches at E9.5. By E10.5, Wnt9b expression is localized to ectoderm of the distal regions of the MNP, LNP and the MXP and also in the mesenchyme of the MXP. At E11.5 Wnt9b is expressed in a highly restricted pattern in the surface ectoderm of the MNP, LNP, and the MXP. Specifically, Wnt9b is expressed in the epithelial seam of the fusion !  '!  points between the LNP and the MNP. Wnt9b mRNA was not detected in the facial mesenchyme at this stage. Expression of Wnt9b is down regulated in the facial ectoderm to levels slightly above background levels during embryogenesis at E12.5. Wnt9b was not detected in the secondary palate during development. The expression domain of Wnt9b in the developing face and the occurrence of CL/P in the A/- strain and in Wnt9bNull/Null embryos demonstrate the importance of the Wnt signaling pathway in embryonic facial development.  1.5.2 Mutations due to IAP insertions Epigenetic modifications of retrotransposons are known to mediate phenotypic variability of traits. Depending on the methylation status of the insertion a range of phenotypes from mutated to wild-type are displayed (Whitelaw and Martin 2001). These alleles are known as metastable epialleles, epialleles that have the ability switch their epigenetic state and once established the state is mitotically inherited (Rakyan et al. 2002). Often the phenotype of an animal can be correlated with the methylation status of the IAP causing the phenotype (Reiss and Mager 2007). There are two mouse models that are examples of metastable epialleles, the agouti viable yellow (Avy) and axin fused (AxinFu) models. Avy is the result of an IAP insertion upstream of the coding region of the Agouti gene. The Avy mutation causes ectopic expression of the gene leading to phenotypically variable phenotype of yellow fur, obesity, diabetes and increased susceptibility to tumors (Morgan et al. 1999). The AxinFu model is the result of an IAP insertion at intron 6 of the Axin gene. The long terminal repeat (LTR) generates aberrant transcripts leading to a gain of function mutation and a variably expressed tail kink phenotype resulting from axial duplications (Vasicek et al. 1997). The variability of the phenotype of both Avy and AxinFu is correlated with the level of methylation of the IAP (Rakyan et al. 2002). Parent-of-  !  ()!  origin effects are also present in the Avy and AxinFu models (Rakyan et al. 2002, Morgan et al. 1999) and are a characteristic of metastable epialleles (Rakyan et al. 2002).  1.5.3 Methylation at clf1 The methylation status of the 5’LTR of the IAP at Wnt9b was assessed by the COBRA assay in A/WySn embryos (Juriloff et al. 2007, Juriloff et al. 2008). They found that A/WySn embryos with normal faces showed a consistent methylation pattern of approximately 50% while CL/P littermates had a consistent methylation pattern of 0-5%. A parent-of-origin effect was also noted; maternally inherited alleles were consistently less methylated than paternally inherited alleles. The hypothesis our lab has been working on, but not as part of my thesis, is that lack of methylation of the IAP leads to lack of Wnt9b through some mechanism like the Avy or AxinFu models.  1.5.4 What are IAPs? IAPs are endogenous retroviral elements that are present in the mouse genome. IAPs belong to the Class II retroviruses and there are two types of IAPs (type I and type II) (Maksakova et al. 2006). Type I elements are 7 kb, contain a protein coding region for gag, pol, and env, and two LTRs (Kuff and Lueders 1988). The IAP at Wnt9b in the A/- strain is a type I#1 element (Juriloff et al. 2005a); the I#1 element has a 1.9 kb deletion resulting in the fusion of the gag and pol regions (Kuff and Lueders 1988). Endogenous retroviruses (ERVs) are highly transcribed in the early zygote and germ cells where new heritable insertions can occur in the germ line (Maksakova et al. 2006). IAP transcripts appear transiently in the pre-implantation embryo (Kuff and Lueders 1988). The most common mutagenic mechanism of ERVs is insertion in the intron of genes (Maksakova et al. 2006). This disrupts gene expression by premature polyadenylation, aberrant splicing, or ectopic transcription driven by an antisense promoter located in the 5’ LTR !  ((!  (Maksakova et al. 2006). LTR driven gene expression results in metastable epialleles. Mutant alleles have variable expressivity in genetically identical individuals due to the variable methylation state of the 5’LTR (Maksakova et al. 2006). Mutagenic insertions can also occur upstream or downstream of genes (Maksakova et al. 2006). Host silencing mechanisms have been developed to silence ERV insertions and protect the host from the harmful consequences of the insertion (Maksakova et al. 2008). Silencing occurs at three levels: transcriptional gene silencing, post-transcriptional gene silencing and host restriction factors (Maksakova et al. 2008). Transcriptional gene silencing involves DNA methylation of promoters and chromatin remodeling (Maksakova et al. 2008). Endogenous retroviruses (ERVs) make up 8-10% of the mouse genome and ERV insertions are responsible for approximately 10-12% of all mouse mutations (Maksakova et al. 2006). Specifically, there are approximately 1,000 IAP elements in the haploid genome and IAPs are responsible for at least 32 mutagenic insertions in the mouse (Maksakova et al. 2006). IAP insertions have a strain bias, the majority of mutagenic insertions occurred on the C3H background (Maksakova et al. 2006), a close relative of the A/- strain. The I#1 elements are the most active in the mouse genome (Maksakova et al. 2006). They are responsible for the majority of IAP insertional mutations (Maksakova et al. 2006).  1.6 Clf2 Recessive clf1 and semi-dominant clf2 have an epistatic relationship; in the context of clf1AA, when clf2 is “AA” there is a high risk of CL/P but when clf2 is “AB” there is only a small but detectable risk of CL/P (Juriloff and Mah 1995, Juriloff et al. 2001b). The clf2B allele is thought to suppress the CL/P phenotype and only a small frequency of CL/P is produced in clf2AB embryos (1-2%) (Juriloff and Mah 1995, Juriloff et al. 2001b). This was confirmed in subsequent studies (Juriloff et al. 2006, Juriloff et al. 2004). !  (+!  Currently, clf2 has not yet been identified and the function is unknown. Clf2 is known to modify the penetrance of CL/P in the Wnt9bNull/clf1 compound heterozygote (Juriloff et al. 2006). That is, in clf2AA embryos the CL/P penetrance is 93%, in clf2AC embryos it is 63%, and in clf2AB embryos it is reduced to 13% (Juriloff et al. 2006). However, it is not known whether clf2 is modifying the Wnt9b signaling pathway, other aspects of facial prominence growth, or the epigenetic modification of the IAP that is clf1. There is also evidence of a third allele at clf2, which confers an intermediate clefting frequency, the “C” allele of unknown origin (Juriloff et al. 2006). Clf2 has been mapped to Chromosome 13 by test-crossing backcross 1 (BC1) segregants, recombinant inbred (RI) strains, and congenic RI strains with A/WySn mothers (Juriloff et al. 2001b, Juriloff et al. 2004). Juriloff et al. (2001) mapped clf2 to a 4 cM region between D13Mit13/54 and D13Mit231 on Chromosome 13 with a method of test crossing BC1 segregants from a cross of A/WySn and C57BL/6J with A/WySn mothers to collect CL/P embryos and RI strains (AXB/Pgn or BXA/Pgn) with high frequencies of CL/P. The candidate region for clf2 was further reduced to a 14.0 mb region between Ntrk2 and Irx2 by test-crossing congenic RI lines (W1, W, R, T) and RI strains (AXB/Pgn or BXA/Pgn) with A/WySn mothers (Juriloff et al. 2004). Prior to my studies the candidate region has been further reduced to a 3.4 mb region using a congenic RI stock, WBC (Juriloff, DM, unpublished data).  1.7 Major Purposes and Hypotheses My studies involve the two loci, clf1 and clf2, involved in CL/P in the A/WySn mouse model. The clf2 gene has not yet been identified and the function was unknown; the major focus of my studies was on identifying the function of clf2 and identifying clf2 candidate genes. The clf1 mutation had already been identified as an IAP insertion downstream of the Wnt9b gene disrupting Wnt9b transcription; however, the signaling pathways affected by the reduced levels !  (,!  of Wnt9b were unknown. As a minor aspect, I addressed the developmental biology of CL/P to identify signaling pathways affected by the loss of Wnt9b in Wnt9bNull/Null embryos.  1.8 Goals of My Studies 1.8.1 Goals for clf1 •  To explore the effect of the lack of Wnt9b on the expression of other craniofacial genes.  1.8.2 Goals for clf2 •  To identify the function of clf2 from among the three major possibilities: modifying Wnt9b signaling; an independent effect on facial prominence development; an effect on the methylation of the IAP.  •  To reduce the list of candidate genes for clf2 by finer definition of the location of clf2.  •  To identify the best candidates for clf2 based on their type of function and expression domain, as listed in genomic databases. This was expected to lead to a platform for the ultimate discovery of the gene that is clf2  and contribute to the understanding of the etiology of CL/P.  !  (*!  2  General Materials and Methods  2.1 Mouse Husbandry The mice used in this study were from the breeding colony of Dr. Diana Juriloff and Dr. Muriel Harris. The mice were housed in the Wesbrook Animal Unit (WAU) at the University of British Columbia (UBC) under standard conditions. This includes a 12-hour light cycle (07:00 to 19:00), and temperature 23 ± 2° Celsius (C). The mice were housed in polycarbonate cages with dried corncob bedding and microisolator filter tops. They were fed Purina Rodent Diet 5001 (Purina, St. Louis, MO) and acidified autoclaved water (pH3.0 with HCl) ad libitum. All procedures involving mice were reviewed and approved by the UBC Animal Care Committee.  2.2 DNA Preparation and PCR of SSLPs DNA was extracted from tissue using the QIAamp DNA mini kit # 51306 (Qiagen, Germantown, MD). Samples were genotyped at clf1 by the WNT15 M/N SSLP marker (at the Wnt9b gene) and by the W9B WT3/WT5/N3 primers for the Wnt9b targeted deletion (Carroll et al. 2005) and at the clf2 candidate region by a variety of flanking and internal SSLP markers. PCR primers for SSLP’s were designed by Drs. Harris and Juriloff and made by Sigma-Aldrich Canada (Oakville, ON). Primer sequences are in Appendix A. The marker genotypes were visualized by PCR methodology. Briefly, 26µl PCR reactions were prepared with Taq DNA Polymerase, Native (Invitrogen, Carlsbad, CA) and run in a Biometra or Perkin Elmer Thermocycler for 35 cycles at a Tanneal of 55° or 53° C. PCR products were separated by electrophoresis in a 4% NuSieve (Lonza, Rockland, ME) agarose gel stained with ethidium bromide. Bands were visualized with UV light and photographed.  !  (-!  2.3 Specialized Mouse Stocks 2.3.1 The WBC stock The WBC stock was used for mapping studies of clf2. This stock was created from crosses between the W1 congenic RI line of mice (Juriloff et al. 2004) and C57BL/6J strain. The WBC stock was a mixture of W1 and C57BL/6J genes, in which by their breeding history, 6265% of genes were of C57BL/6J origin, 34-37% were of A/J origin (from AXB-23/Pgn) and less than 1% were of A/WySn origin, except for Chromosome 13 (Juriloff, personal communication). The clf1 gene in WBC is of A/J origin. The A/WySn-derived clf2-containing region of Chromosome 13 from the W1 stock had been maintained in heterozygous state in WBC, the other chromosome being from C57BL/6J, in order to generate recombinant chromosomes for further mapping of clf2. The progeny in each generation of WBC have been genotyped for SSLP markers within and flanking the entire candidate region on Chromosome 13 to screen for new recombinants. Recombinant chromosomes were indentified by a change in the genotype of the haplotype; that is, down the length of candidate region a recombinant individual may begin as genotype AA and then change to genotype AB.  2.3.2 The B.WN stock The B.WN stock is the stock of heterozygous carriers of the Wnt9b null mutation. These mice lack exon 2 of the Wnt9b gene (Carroll et al. 2005). Two males were imported to the Medical Genetics Mouse Unit at UBC for the complementation test with clf1 (Juriloff et al. 2006). The Wnt9b null mutation is on a heterogeneous background; the ES cells used to create the null mutation were from 129/SvJ and the chimeras were crossed to C57BL/6J, the offspring were out-crossed to Swiss Webster, and their offspring were crossed to C57BL/6J to create the two males sent to UBC (Juriloff et al. 2006). The background of the Wnt9b null mutant males consists of one chromosome being obligate C57BL/6J, and the second a mixture of 129/SvJ, !  (%!  C57BL/6J and Swiss Webster. The markers flanking the Wnt9b null mutation are 129/SvJ, and the clf2 candidate region and flanking markers are obligate C57BL/6J on one chromosome and either 129/SvJ, C57BL/6J or Swiss Webster on the other homolog (Juriloff et al. 2006). Drs. Juriloff and Harris propagated the two males to create the B.WN stock. They crossed the two males to C57BL/6J and propagated their heterozygous carrier offspring. The progeny in each generation of B.WN have been genotyped for SSLP markers at the Wnt9b gene, W9B and WNT15 M/N. Heterozygous carriers were crossed to C57BL/6J six times since their arrival at UBC. From their breeding history, the genetic background of the B.WN stock is calculated to be 97.66% C57BL/6J, 1.56% Swiss Webster and 0.78% 129/SvJ with the exception of the genes closely surrounding the Wnt9b locus, which are 129/SvJ (Juriloff, personal communication).  2.4 Identification of Recombinants Drs. Juriloff and Harris maintained the WBC line by intercrossing mice with the genotypes “AB” and “AA” through the clf2 candidate region. The tissue obtained from the identification of individual progeny by the ear notch method was used for DNA extraction. Various SSLP markers from Chromosome 13 were genotyped on this DNA by lab staff as described above and new recombinants were identified. The Chromosome 13 markers used to identify the new recombinants used in my studies were CNT C/D, CNT E/F, ZF456 A/B, D13M6791 A/B, and MTRR A3/B3. These markers are SSLPs from within or close to the genes as indicated by the names of the markers. The REC4 and W1-U recombinants used in my studies were originally detected by these methods.  2.5 Observation of CL Pregnant females were palpated to identify pregnancies and were euthanized by Carbon Dioxide inhalation according to UBC SOP# 009E4-CO2 at approximately gestation day 14 (E14) !  (&!  with a range of E12-E16. Age of gestation was confirmed by embryonic morphology. The uterus was removed, pinned on black wax and submerged in 0.85% sterile saline solution. The uterus was slit longitudinally and the contents removed under a dissection microscope. Embryos were quickly removed, scored for CL and other obvious external morphological defects and then immediately decapitated. Heads were fixed in buffered formalin; a small portion of the body was saved for immediate genomic DNA preparation and the remainder was archived at minus 20°C. Post-implantation mortality (“moles”) were also recorded.  2.6 Statistical Analysis The Fisher’s Exact Test was used to test associations of genotype and phenotype using SPSS 15.0 for Windows. The $2 tests of goodness-of-fit were used to test the fit of the genotypes to the expected Mendelian proportions. $2 tests of independence were used to test hypothesized relationships. Statistical significance was set at P < .05.  !  (.!  3  Studies to Test the Function of clf2: Cross 1 and Cross 2  3.1 Background Previous work has shown that clf2 modifies the penetrance of CL/P in the compound heterozygote (Wnt9bNull/clf1) (Juriloff et al. 2006) but it is not known whether the effects of deficient Wnt9b on the embryonic face or the effects of the IAP disruption at clf1 are being modified. Future experiments aimed at identifying the clf2 mutation will be expedited if we know what genre of gene we are looking for. If clf2 modifies the functioning of the Wnt9b pathway or affects another aspect of facial prominence development that contributes to risk of CL/P, it is expected to modify the frequency of CL/P in the Wnt9b null mutants. In order to test whether clf2 modifies the frequeny of CL/P in the Wnt9b null mutants I conducted a study involving two separate crosses. Juriloff et al. (2006) have shown that the compound heterozygote (Wnt9bNull/clf1) has a frequency of cleft lip of 93% when clf2 is AA, 63% when clf2 is AC and 13% when clf2 is AB demonstrating that clf2 modifies the clefting frequency in the compound heterozygote. Without isolating the Wnt9bNull allele from the Wnt9bclf1 allele it is impossible to detect which portion of the compound heterozygote clf2 is modifying. Clf2 could have a role in facial prominence development and the loss of Wnt9b in combination with clf2AA could worsen a developmental flaw and contribute to a greater clefting frequency. Conversely, clf2 could modify the mechanism of IAP disruption of Wnt9b activity in the clf1 allele. By examining homozygotes for the Wnt9bNull allele with segregating genotypes of clf2 (clf2AA, clf2AB, or clf2BB) we can detect if clf2AA has an effect on the clefting frequency of Wnt9b deficient embryos. If clf2 is a “face” gene; that is, clf2 is involved in development of at least one of the facial prominences involved in formation of the upper lip, or in signaling pathways that interact with Wnt9b, and thereby modifies the risk of CL/P, we would expect Wnt9bNull/Null CL !  ('!  embryos to have an excess of the clf2AA genotype and embryos with the clf2BB genotype to be more frequently normal. However, if CL is independent of the genotype at clf2, the three genotypes, clf2AA, clf2AB and clf2BB, will have equal frequencies of CL/P. To study the effect of clf2 on the Wnt9bNull/Null CL frequency, we began with Cross 1. This was a cross of B.WN and W1 mice and was aimed to combine the Wnt9bNull/Null segregants with segregating genotypes of clf2 (clf2AA, clf2AB, or clf2BB) to test if clf2 modifies the clefting frequency in Wnt9bNull/Null mice on a genetic background that was mostly C57BL/6J. A few weeks after the Cross 1 breeding pairs were set up they still had not bred. Worried that Cross 1 was not going to breed we began a second study, Cross 2, based on crosses between A/WySn and B.WN. Cross 2 differs from Cross 1 by the larger amount of background genotype with an A/WySn origin and by the presence of clf1, although it also generates the Wnt9bNull/Null segregants. Shortly after Cross 2 began, Cross 1 began breeding and we continued to study both crosses in parallel, each cross serving as a backup for each other in case of further breeding problems.  3.2 Materials and Methods 3.2.1 Cross 1 The details of the Cross 1 breeding scheme can be found in figure 3.1. Briefly, we crossed W1 mice by B.WN mice and selected the offspring with the genotype Wnt9b+/Null, clf2AB. The Wnt9b+/Null, clf2AB animals were then bred with C57BL/6J mice and we selected the Wnt9b+/Null, clf2AB offspring. We crossed to the readily available C57BL/6J to increase the amount of C57BL/6J genes, to increase the homogeneity of the background, and to facilitate a large number of mice in the study. We then intercrossed the Wnt9b+/Null, clf2AB progeny and collected and scored the embryos for CL. A total of four original W1 parents (3 female, 1 male) and four original B.WN parents (1 female, 3 male) were used to originate Cross 1. !  +)!  Figure 3.1 Cross 1 breeding scheme.  3.2.2 Cross 2 The breeding scheme for Cross 2 is detailed in figure 3.2. This cross began by breeding three A/WySn females with three B.WN males. Later a reciprocal cross (A/WySn males X B.WN females) was made; there were two B.WN females used to originate this cross. The progeny with the genotype Wnt9bclf1/Null, clf2AB were selected and intercrossed; embryos from this generation were collected and scored for CL.  !  +(!  A/WySn  X  Wnt9bclf1/clf1 clf2AA Wnt9bclf1/Null clf2AB  Wnt9bclf1/+ clf2AB  X  Wnt9bclf1/Null clf2AB Wnt9bNull/Null clf2AA clf2AB clf2BB  B.WN  Wnt9b+/Null clf2BB  Wnt9bclf1/Null clf2AB  Wnt9bclf1/Null clf2AA clf2AB clf2BB  Wnt9bclf1/clf1 clf2AA clf2AB clf2BB  Figure 3.2 Cross 2 breeding scheme  3.2.3 Genotyping of Cross 1 and Cross 2 breeders Progeny were genotyped at each generation using SSLP markers at Wnt9b and clf2 (flanking and internal), specifically W9B, WNT15 M/N, CNT C/D, ZF456 A/B and MTRR A3/B3 as described in the General Materials and Methods (Chapter 2). DNA extraction and PCR were completed by myself or by lab staff, particularly Ronald Chan.  3.2.4 Generation of embryos and observation of CL Female Wnt9b+/Null, clf2AB (Cross 1) or Wnt9bclf1/Null, clf2AB (Cross 2) mice were placed with singly caged Wnt9b+/Null, clf2AB (Cross 1) or Wnt9bclf1/Null, clf2AB (Cross 2) males and they remained for an average of 5 days. Pregnant females were euthanized and their embryos observed and scored for CL as described in the General Materials and Methods.  3.2.5 Genotyping of embryos Genomic DNA was prepared from a small portion of embryonic tissue and all embryos were genotyped for Wnt9b and clf2 flanking and internal SSLP markers by PCR methodology as described in the General Materials and Methods. All Cross 1 and Cross 2 embryos were genotyped at markers W9B and WNT15 M/N for the genotype at Wnt9b; CNT C/D, PTDS G/H  !  ++!  and D13M6791 A/B for the genotype at clf2; and SMC X1/41 for gender. The genotyping was completed by myself or by lab staff, particularly Ronald Chan or Matt Trudeau.  3.2.6 Test of CL penetrance in B.WN Wnt9bNull/Null embryos To observe the penetrance of CL in B.WN Wnt9bNull/Null embryos in parallel to the Cross 1 and Cross 2 studies we intercrossed Wnt9b+/Null B.WN males and females (described in General Materials and Methods) and collected and observed E12-E14 embryos as described in the General Materials and Methods. The same methods were used as in Cross 1 and Cross 2. To obtain the genotypes for all embryos at Wnt9b DNA extraction and PCR was completed by lab staff, particularly Matt Trudeau. The markers used for Wnt9b were W9B and WNT15 M/N.  3.3 Results 3.3.1 Cross 1 Breeding for Cross 1 began in January 2009 by the setting up of breeding pairs of W1 X B.WN. After two generations of crosses, we obtained Wnt9b+/Null, clf2AB animals to intercross for production of embryos. From 20 Wnt9b+/Null, clf2AB males bred with 49 Wnt9b+/Null, clf2AB females, there were 360 embryos collected on E12-E18 (mean =13.9, standard error = ±.15) (Table 3.1). Of the 360 Cross 1 embryos, 296 were normal, 65 had CL and one embryo had a lip scar. Twenty-nine had a bilateral cleft lip (BCL), 27 had a left unilateral cleft lip (LCL) and nine had a right unilateral cleft (RCL) lip. The frequency of clefting in Wnt9bNull/Null embryos was 82% (65/79). Two of the CL embryos had been dead for one day when the embryos were collected; however, their faces could still be scored and they were genotyped. “Moles” (early post implantation mortality) were also observed; Cross 1 had 32 moles (32/392=8%). All moles were dead prior to lip fusion (E10-11) and could not be scored.  !  +,!  Table 3.1 Summary of Cross 1 Number Number Number Number Number Number % CL in of of Cross 1 of Normal of CL of Litters E12-E16c of Molesa Wnt9bNull/Null males Embryos Embryos Embryos 20 49 360b 32 294 65 65/79=82% a Post-implantation mortality up to E10. b one Wnt9bNull/Null embryo had a scar phenotype. c except one E17 litter of 3 embryos and 1 mole, and one E18 litter of 1 embryo and no moles  The genotypes of the Cross 1 embryos, based on the markers, are shown in Table 3.2. At the Wnt9b locus, the number of each genotype (100 +/+ vs. 181 +/Null vs. 79 Null/Null) fit a Mendelian ratio (Goodness-of-Fit to Mendelian ratio $2=2.46, P = 0.2929). At the clf2 locus, the segregation rations (78 AA vs. 164 AB vs. 117 BB) significantly deviated from a Mendelian ratio (Goodness-of-Fit to Mendelian ratio $2=11.15, P = 0.0038). There is a deficiency of the A gamete and a significant deficiency of the clf2AA and clf2AB genotypes (AA vs. BB Goodness-ofFit to Mendelian ratio $2=7.800, P = 0.0052; AB vs. BB Goodness-of-Fit to Mendelian ratio $2=9.004, P = 0.0027). The sex ratio for all Cross 1 embryos fit a 50:50 ratio, 185 XX : 175 XY (Goodness-of-Fit to Mendelian ratio $2=.278, P = 0.5982).  !  +*!  XX 7 22 2  Table 3.2 Genotypes of Cross 1 embryos clf2AA clf2AB XY Total XX XY Total XX 19 26 24 23 47 17 20 42 41 38 79 30 0 2 4 4 8 2  clf2BB XY Total 10 27 30 60 1 3  XX 0 0 3  clf2AA XY Total 0 0 0 0 5 8  clf2BB XY Total 0 0 0 0 12 27  Normals +/+  Wnt9b Wnt9b+/Null Wnt9bNull/Null Clefts Wnt9b+/+ Wnt9b+/Null Wnt9bNull/Null  XX 0 0 17  clf2AB XY 0 0 12  Total 0 0 29  XX 0 0 15  clf2AA clf2AB XX XY Total XX XY Total XX Wnt9b+/+ 7 19 26 24 23 47 17 +/Null Wnt9b 22 20 42 41 38 79 30 Wnt9bNull/Null 5 5 10 21 16 37 17 *Embryos not in tally: Recombinant C1-147, BCL, Wnt9bNull/Null, XY; Scar: C1-154, Wnt9bNull/Null, clf2AB, XY Total  clf2BB XY Total 10 27 30 60 13 30  All CL embryos had a Wnt9bNull/Null genotype. The genotype at clf2 and the phenotype of the lip in the Wnt9bNull/Null embryos are independent (Fisher’s exact test P = .44), that is clf2 did not modify the frequency of CL in Wnt9b null embryos.  3.3.2 Cross 2 After the first few litters from Cross 2 we could see that the penetrance of CL in Wnt9bNull/Null embryos was high making it difficult to deduce if clf2 modified the rate of CL in Wnt9b null embryos. We hypothesized the high penetrance might be due to an A/WySn cytoplasmic effect. We therefore set up the reciprocal cross (male A/WySn X female B.WN) which removes the A/WySn cytoplasm and hoped to see the penetrance of CL reduced. The reciprocal cross had a high penetrance similar to the original cross. In the A/WySn cytoplasm, Wnt9b and clf2 fit Mendelian ratios (Goodness-of-Fit to Mendelian ratio $2=1.82, P = 0.3996 and Goodness-of-Fit to Mendelian ratio $2=.6, P = 0.7473, respectively). The reciprocal cross with B.WN cytoplasm (C57BL/6J) also fit Mendelian ratios for both loci, Wnt9b and clf2 !  +-!  (Goodness-of-Fit to Mendelian ratio $2=.910, P = 0.6344 and Goodness-of-Fit to Mendelian ratio $2=.545, P = 0.7613, respectively). The penetrance of CL in the Wnt9bNull/Null embryos was not significantly different between the reciprocal crosses (Fisher’s exact test (two tail) P = 0.8510) and therefore the data were pooled. Data from each respective cross can be found in Appendix B. A total of 12 Wnt9bclf1/Null, clf2AB males and 26 Wnt9bclf1/Null, clf2AB females produced 204 embryos (Table 3.3). Embryos were collected from E12-E18 (mean = 13.9; Standard error = ±.24). We observed 137 normal embryos, 66 CL embryos and one embryo with a scar at the fusion site. Of the CL embryos, 61 had a BCL, three had an LCL and two had an RCL. Three of the CL embryos were dead when the embryos were collected on E12 (1) and E14 (2), the embryos were still able to be scored for CL and genotyped. The frequency of clefting in Wnt9bNull/Null embryos was 90% (55/61). 15 “moles” (early (E8-E11) post implantation mortality) were observed (15/219=7%).  Number of Cross 2 males  Table 3.3 Summary of Cross 2 Number Number Number Number of of of of c E12-E16 Normal Litters Molesa Embryos Embryos  Number of CL Embryos  % CL in Wnt9bNull/Null  Female A/WySn X 6 15 115 10 74 41 33/35 = 94% Male B.WN Male A/WySn X 6 11 89b 5 63 25 22/26 = 84% Female B.WN Total 12 26 204 15 137 66 55/61 = 90% a b Null/Null Postimplantation mortality up to E10; one Wnt9b embryo had a scar phenotype; cexcept one E18 litter with 7 embryos and no moles, and one E18 litter with 2 embryos and 1 mole.  The genotypes of Cross 2 are shown in Table 3.4. The Wnt9b locus, (46 +/+ vs. 97 +/Null vs. 61 Null/Null) fit a Mendelian ratio (Goodness-of-Fit to Mendelian ratio $2=2.696, P = !  +%!  0.2597). At the clf2 locus, genotypes (51 AA vs. 100 AB vs. 52 BB) fit a Mendelian ratio (Goodness-of-Fit to Mendelian ratio $2=.054, P = 0.9733). The sex ratio for all Cross 2 embryos fit a 50:50 ratio, 93 XX : 111 XY (Goodness-of-Fit to Mendelian ratio $2=1.588, P = 0.2076).  XX 4 7 0  Table 3.4 Genotypes of Cross 2 embryos clf2AA clf2AB XY Total XX XY Total XX 8 12 9 15 24 5 9 16 22 25 47 14 1 1 2 1 3 1  clf2BB XY Total 5 10 9 22 0 1  XX 0 4 5  clf2AA XY Total 0 0 6 10 7 12  XX 0 0 8  clf2BB XY Total 0 0 0 0 10 18  XX 5 13 9  clf2BB XY Total 5 10 9 22 10 19  Normals clf1/clf1  Wnt9b Wnt9bclf1/Null Wnt9bNull/Null Clefts clf1/clf1  Wnt9b Wnt9bclf1/Null Wnt9bNull/Null  XX 0 0 12  clf2AB XY Total 0 0 1 1 13 25  clf2AA clf2AB XX XY Total XX XY Total Wnt9bclf1/clf1 4 8 12 9 15 24 clf1/Null Wnt9b 11 15 26 22 26 48 Wnt9bNull/Null 5 8 13 14 14 28 Null/Null BB *Embryo not in tally: SCAR: C2-165, Wnt9b , clf2 , XY Total  The CL embryos in Cross 2 were of the genotype Wnt9bNull/Null or Wnt9bclf1/Null; none of the CL embryos were Wnt9bclf1/clf1. The penetrance of CL was high, 90% in Wnt9bNull/Null embryos, only five were normal. The genotype at clf2 and the phenotype of the lip in the Wnt9bNull/Null embryos seem to be independent (Fisher’s exact test P = .84), that is clf2 did not modify the frequency of CL in Wnt9b null embryos. The compound heterozygote (Wnt9bclf1/Null) also provided some CL embryos; 10 of the CL embryos are clf2AA, one was clf2AB and none of the CL embryos were clf2BB. The frequency of clefting in the compound heterozygote varied with the clf2 genotype; 38% of clf2AA embryos were cleft and 2% of clf2AB embryos were cleft. CL was significantly associated with the clf2 genotype in the compound heterozygote (Fisher’s  !  +&!  exact test P < .001). The frequencies of CL in the various Cross 2 genotypes is summarized in Table 3.5. Table 3.5 Frequencies of CL in Cross 2 genotypes. clf2AA clf2AB clf2BB clf1/clf1 Wnt9b 0% 0% 0% Wnt9bclf1/Null 38% 2% 0% Wnt9bNull/Null 92% 89% 95%  3.3.3 Recombinants for the clf2 candidate region Five recombinant embryos were observed in the combined crosses; Cross 1 (C1) had four recombinants and Cross 2 (C2) had one (Table 3.6). They were genotyped for additional SSLP markers throughout and flanking the clf2 candidate region to better define their breakpoints (Table 3.6). Three of the recombinants, C1-33, C1-78 and C2-196, have a breakpoint between ZFP71 A/B2 and D13M6791 A/B, the lower boundary of the clf2 candidate region (see chapter 5). One of the recombinants, C1-161, has a breakpoint between CNT C/D and CNT E/F, the upper boundary of the clf2 candidate region (see chapter 5). The final recombinant, C1-147, has a new breakpoint between the markers ZF456 A/B and 261RIK A/B. C1-147 has a breakpoint in the middle of the candidate region and for this reason has been omitted from Table 3.2 as we cannot distinguish if it is AB or BB at clf2; C1-147 had a BCL, a Null/Null genotype at Wnt9b and was a male. Table 3.6 Genotypes of recombinant embryos in Cross 1 and Cross 2 Marker C1-33 C1-78 C1-147 C1-161 C2-196 CDC14 C/D BB BB AB AB BB CNT C/D BB BB AB AB BB CNT E/F BB BB AB AA BB AK145 A/B BB BB AB AA BB PTDS G/H BB BB AB AA BB ZF456 A/B BB BB AB AA BB 261RIK A/B BB BB BB AA BB ZFP71 A/B2 BB BB BB AA BB D13M6791 A/B AB AB BB AA AB MTRR A3/B3 AB AB BB AA AB Bold font denotes clf2 candidate region.  !  +.!  3.3.4 Test of CL penetrance in B.WN Wnt9bNull/Null embryos We set up breeding pairs of Wnt9b+/Null mice to test the penetrance of CL in Wnt9bNull/Null mice. A total of 34 embryos were collected on E13-15, Table 3.7. There were a total of 29 normal embryos and five CL embryos; three had a right unilateral cleft and two had a left unilateral cleft. 16 “moles” (early (E9-E11) post implantation mortality) were observed (16/50=32%). Table 3.7 Summary of test of B.WN CL penetrance Number Number Number Number Number Number % CL in of of B.WN of Normal of CL of Litters E13-E15 of Molesa Wnt9bNull/Null males Embryos Embryos Embryos 4 6 34 16 29 5 5/9=56% a Post-implantation mortality up to E11. All embryos were genotyped at Wnt9b (W9B and WNT15 M/N) and sexed (SMC X1/41), Table 3.8.  The Wnt9b locus, (11 +/+ vs. 14 +/Null vs. 9 Null/Null) fit a Mendelian  ratio (Goodness-of-Fit to Mendelian ratio $2=1.122, P = 0.5707). The sex ratio fit a normal 50:50 ratio (19:15) (Goodness-of-Fit to Mendelian ratio $2=.471, P = 0.4927). The frequency of clefting in the Wnt9bNull/Null embryos was 56% (5/9). The penetrance of CL/P in the B.WN test cross was significantly lower than Cross 1 and Cross 2 ($2 test of independence Cross 1 vs. B.WN: $2= 3.860, P = 0.0495; Cross 2 vs. B.WN: $2=8.998, P = 0.0027). Table 3.8 Genotypes of Wnt9b+/Null X Wnt9b+/Null embryos Wnt9b+/+ Wnt9b+/Null Wnt9bNull/Null XX XY XX XY XX XY Normal 8 3 7 7 2 2 Cleft 0 0 0 0 2 3 Total 8 3 7 7 4 5  3.4 Discussion Two separate breeding studies were used to test whether clf2 modifies the penetrance of CL in the Wnt9bNull/Null genotype. In other words, we were asking the question: does clf2 have a role in processes such as facial prominence development, the Wnt signaling pathway or a similar ! +'!  signaling pathway important in facial prominence development? If clf2 were a “face” gene we expect an excess of the clf2AA genotype in the CL embryos. Both crosses yielded unexpected results making it difficult to decipher if clf2 had an effect on the frequency of CL in the Wnt9b null homozygotes. Cross 1 had an overall CL frequency of 83% in Wnt9bNull/Null embryos. While the Wnt9b locus and the sex ratio fit Mendelian inheritance patterns, the clf2 locus significantly does not fit Mendelian inheritance patterns. There seems to be a deficiency of the “A” allele producing a severe deficiency of the clf2AA and clf2AB genotypes. Severe deficiency of the “A” allele is unexpected and the reason for this deficiency is unknown. The post-implantation mortality does not seem high enough in this cross to explain the loss of the “A” gamete. Interestingly, Cross 2 follows Mendelian inheritance patterns at clf2. This is a peculiar difference as the same genotypes are present in cross 1 as cross 2, the only difference between the two crosses is the amount of C57BL/6J background. Cross 1 is almost entirely on a C57BL/6J background while Cross 2 is about 50% C57BL/6J suggesting that the factor causing the lack of “A” allele is C57BL/6J-specific; that is, a factor in the C57BL/6J genome is incompatible with the clf2A allele. The shortage of the “A” allele makes it difficult to draw conclusions on the role of clf2 due to the small sample size of clf2AA. A Fisher’s exact test comparing the CL frequency in all three genotypes (clf2AA, clf2AB, or clf2BB) demonstrates that the frequencies are not significantly different. However, the sample size of the clf2AA genotype (10) is small and weakens the confidence in the test. There is no trend to suggest the frequencies are not the same but we cannot say this with great confidence. Cross 2 was also set up to provide Wnt9bNull/Null embryos. The frequency of clefting in the Wnt9bNull/Null embryos was extremely high, 90%. This was an unexpected result as the CL frequency in the B.WN stock is approximately 56%, consistent with personal communication to ! ,)!  Dr. Juriloff from Dr. A.C. Lidral, the source of the knockout mice. The absence of detectable differences between the reciprocal crosses in CL frequency indicates that the cause is not an A/WySn cytoplasmic effect. We hoped that the high CL frequency was a simple statistical sampling problem that would reverse with a larger sample, but this did not happen. The high frequency of CL makes it difficult to determine if clf2 modifies the frequency of CL in Wnt9bNull/Null. In contrast to the Wnt9bNull/Null embryos, there were no CL embryos in the Wnt9bclf1/clf1, clf2AA genotype. This can be explained as the 3rd requirement for CL in the A/WySn mouse model is missing, the strong maternal effect of the A/WySn mother. Previous studies have shown that there is a large difference between the frequency of CL between A/- strain and F1 mothers carrying the same array of genotypes in their embryos (Davidson et al. 1969); therefore our sample may be too small to detect the low frequency of clefting that is expected in an F1 mother. Cross 2 did provide some new information, from the Wnt9bNull/clf1 segregants. Previous research has shown that Wnt9bNull/clf1 compound heterozygotes in an A/WySn mother have a CL frequency of 93% when clf2 is AA and 13% when clf2 is AB (Juriloff et al. 2006). We saw a similar pattern in the compound heterozygotes of Cross 2 but the CL frequency in clf2AA was 38% and 2% in clf2AB in an F1 mother. It is interesting to note that when the A/WySn maternal effect is not present, the frequency of clefting is markedly reduced. Combined with the lack of Wnt9bclf1/clf1, clf2AA clefts this shows the importance of an A/WySn mother on the frequency of CL. Additionally, this study was the first to be able to compare the clefting frequency of the clf2AB versus clf2BB genotype in Wnt9bNull/clf1 embryos. The clf2AB genotype provided an intermediate clefting frequency to clf2AA and clf2BB, as predicted by Juriloff et al. (2006). There were five recombinants: Cross 1 had four and Cross 2 had one. Three of the breakpoints are positioned at the lower boundary of the clf2 candidate region (defined by REC! ,(!  4), one is at the upper boundary (defined by W1-U), and the fifth has a new breakpoint between ZF456 A/B and 261RIK A/B. The multiple recombination events found at the previously observed REC-4 and W1-U sites suggests the possibility of recombination hot spots in these areas. Only one of the recombinants is at a new breakpoint within the candidate region demonstrating an extremely low recombination frequency of approximately 0.2% (1/564). Cross 1 and Cross 2 were designed to test whether clf2 modifies the effects of the Wnt9b null mutation on the developing embryonic face. The unexplained deficiency of the “A” allele in Cross 1 and the high frequency of CL in Cross 2 interfered with our ability to identify the function of clf2. The trends in the data suggest that clf2 does not have an effect on the penetrance of CL in the null mutant but we were not able to conclude this with great confidence. We were able to confirm previous results that clf2 modifies the CL frequency in the Wnt9bNull/clf1 embryos and once again demonstrate the importance of the A/WySn maternal effect on the penetrance of cleft lip. We were also able to observe a lack of CL embryos in Wnt9bNull/clf1, clf2BB embryos, as was expected but never studied previously.  !  ,+!  4  Epigenetic Modification of clf2  4.1 Background Cross 2 was not able to answer our original question, does clf2 modify the penetrance of CL in Wnt9b null embryos, due to a unexpected high penetrance of CL. However, the Wnt9bNull/clf1segregants of Cross 2 provided an inverse way to address the role of clf2: to test whether clf2 modifies the epigenetic modification (methylation) of the Wnt9b-IAP in the clf1 mutant allele using a combined bisulfite restriction analysis (COBRA) assay. Clf2 is known to modify the penetrance of clefting in the compound heterozygote, WntbNull/clf1 (Juriloff et al. 2006). Using the Wnt9bNull/clf1 embryos generated in Cross 2, figure 3.2, we can evaluate the methylation status of the 5’LTR of the Wnt9b-IAP in the clf2 segregating genotypes (clf2AA, clf2AB, clf2BB). If clf2 does modify the epigenetic modification of the Wnt9b-IAP we would expect the clf2AA embryos to have a hypomethylated 5’LTR, clf2AB embryos to have intermediate methylation levels, and clf2BB embryos to have a more methylated 5’LTR. The methylation status of Wnt9b-IAP has been previously correlated with the CL/P phenotype. The 5’LTR of CL/P embryos is usually about 0-5% methylated whereas it has higher levels of methylation in normal littermates (Juriloff et al. 2007, Juriloff et al. 2008). In the present study, we expect the CL embryos to be similarly less methylated than normal embryos. We chose to use Wnt9bNull/clf1 embryos rather than the Wnt9bclf1/clf1 embryos. Compound heterozygotes have an advantage over clf1 homozygotes by only representing one clf1 allele; this allows us to study the methylation status of each IAP independently rather than an average of two alleles, which is the case with clf1 homozygotes. Clf1 homozygotes carry a maternal and paternal allele that may be differentially methylated.  !  ,,!  4.2 Materials and Methods 4.2.1 Wnt9bNull/clf1 embryos  Archived body tissue from Wnt9bNull/clf1 embryos from Cross 2 (section 3.2.4) were used. The methylation statuses of the Wnt9b-IAP in the face and body have been shown to be highly correlated in our lab (Juriloff, DM and Gagnier, L, unpublished data). Embryos with the known genotype of Wnt9bNull/clf1, clf2AA; Wnt9bNull/clf1, clf2AB; or Wnt9bNull/clf1, clf2BB were selected for this study. We selected 16 embryos of each genotype; the Wnt9bNull/clf1, clf2AA genotype had eight randomly selected normal embryos and eight randomly selected CL, the Wnt9bNull/clf1, clf2AB genotype had 15 randomly selected normal embryos and one CL (the one CL embryo in all of Cross 2 with this genotype), and the 16 randomly selected Wnt9bNull/clf1, clf2BB embryos were all normal. A fresh DNA sample was prepared from the archived tissue, as described in General Materials and Methods. The new DNA sample was re-genotyped, as described in General Materials and Methods, at Wnt9b (W9B and WNT15 M/N) and in the middle of the clf2 candidate region (ZF456 A/B) to verify the correct genotype.  4.2.2 COBRA assay The COBRA assay is a sensitive and semi-qualitative methylation assay (Xiong and Laird 1997). The first step of the COBRA assay is a conversion reaction using bisulfite. Methylated DNA is treated with bisulfite; this converts all unmethylated cytosines into uracil while methylated cytosines remain unchanged. The restriction enzymes, MboI and EcoRI, used in later steps have cytosines in their restriction sites; these restriction sites are lost when the cytosines are converted to uracil. The conversion reaction was completed using the EZ DNA Methylation Kit (Zymo Research, Orange, CA) by Liane Gagnier, in the Mager Lab at the BC Cancer Agency. We then amplified the 5’LTR of the Wnt9b-IAP in the converted DNA. We completed two rounds of PCR using semi-nested primers. The semi-nested primers were designed by Liane !  ,*!  Gagnier and made by Invitrogen (Carlsbad, CA). The first round consisted of a 25µl reaction with the primers IAP-BIS1 and IAP-BIS3 and the converted DNA as a template. The second round used the first round product as a template for a 50µl reaction with the primers IAP-BIS1 and IAP-BIS2. Primer sequences are listed in Appendix A. The reactions were prepared with 2x AmpliTaq Gold 360 Master Mix (Applied Biosystems, Foster City, CA) and run in a Biometra Thermocycler for 35 cycles at a Tanneal of 50° C. Two PCR reactions were completed for each sample and combined prior to electrophoresis separation. I completed the PCR reactions in the Juriloff lab at UBC. The PCR products were separated by electrophoresis in a 1.5% UltraPure (Invitrogen, Carlsbad, CA) agarose gel stained for 30 minutes in 1X Gel-Red (Biotium, Hayward, CA). Bands were visualized with the Kodak 1D Gel Documentation System (Kodak, New Haven, CT). PCR products of the 524 bp size were cut from the gel and purified using the Minelute Gel Extraction Kit (Qiagen, Mississauga, ON). The gel and gel purification steps were completed by Liane Gagnier, Mager Lab. Two separate restriction enzyme digests, MboI and EcoRI, were completed for each DNA sample. The cutting site for MboI is located within the 5’LTR of the Wnt9b-IAP just after position 132 of the 524 bp fragment, EcoRI cuts just outside the 5’LTR just after position 167 of the 524 bp fragment, figure 4.1. Approximately 200-500ng of PCR products from each sample were prepared in duplicate, one preparation was incubated with MboI (New England Biolabs, Beverly, MA) and the second with EcoRI (New England Biolabs, Beverly, MA), for 2 hours at 37° C in 20µl reactions. Restriction digest products were separated by electrophoresis in a 1.5% UltraPure (Invitrogen, Carlsbad, CA) agarose gel stained for 30 minutes in 1X Gel-Red (Biotium, Hayward, CA). Bands were visualized with the Kodak 1D Gel Documentation System (Kodak, New Haven, CT). The restriction digest and final gel were completed by Liane Gagnier, !  ,-!  Mager Lab. “Uncut” (unmethylated) bands were 524 bp and “cut” (methylated) bands were 354 bp and 168bp in size. The density of the “cut” and “uncut” bands was used to estimate the methylation status of the 5’LTR in a semi-quantitative, categorical manner; all scoring was done by Liane Gagnier. The percent methylation was scored as 0, 10, 20, 30, 40, 50, or 60.  Figure 4.1 Restriction enzyme cut sites in relation to the Wnt9b-IAP 5’LTR and the Wnt9b gene. The solid horizontal line represents genomic DNA. The solid vertical bars represent exons and the open vertical bar represent the UTRs of mRNA of the Wnt9b gene. The horizontal arrows indicate the direction of transcription. The red horizontal line and arrows represent Wnt9b-IAP and the arrows are the LTRs. The MboI and EcoRI restriction enzymes cut inside and just outside, respectively, of the Wnt9b-IAP 5’LTR.  4.3 Results The two restriction enzymes, MboI and EcoRI, gave very similar COBRA results but are not combined, as their cutting sites are not in the same location. An example of the gel photo is in figure 4.2 and all gel photos are in Appendix C. We have combined the categorical scoring groups for the simplicity of graphs and for adequate group sizes for statistical analysis into the categories of 0, 10-20, 30-40 and 50-60.  !  ,%!  ZZ!  Figure 4.2 Example of a COBRA gel after a MboI restriction digest. The 524 bp product is the unmethylated, uncut band, the 354 bp and 168 bp product is the methylated, cut band. The numbers below are the estimated percent methylation of the Wnt9b-IAP 5’LTR. The MboI restriction digest showed a strong association between methylation status of the 5’LTR and the genotype at clf2, figure 4.3 ($2 Test of Independence = 20.03, P = .002). A similarly strong association between the methylation status of the 5’LTR and the genotype was also observed for the EcoRI digest, figure 4.4 ($2 Test of Independence, $2 = 16.5, P = .011). For both enzymes, the clf2AA embryos were consistently less methylated than clf2AB embryos, which were consistently less methylated than clf2BB embryos. A detailed analysis of the data and the original restriction digest gels are located in Appendix C.  Figure 4.3 Methylation status of the Wnt9b-IAP using MboI. Embryos are all Wnt9bNull/clf1 and segregating at clf2, embryos are grouped by their genotype at clf2.  !  ,&!  Figure 4.4 Methylation status of the Wnt9b-IAP using EcoRI. Embryos are all Wnt9bNull/clf1 and segregating at clf2, embryos are grouped by their genotype at clf2.  Embryos with the CL phenotype were consistently less methylated than embryos with the normal phenotype for both MboI and EcoRI digest, figures 4.5 and 4.6. The CL embryos with the Wnt9bNull/clf1, clf2AA genotype all had less than 20% methylation for both restriction enzymes. With the exception of one unilateral cleft that was 10% methylated, all of the Wnt9bNull/clf1, clf2AA clefts were bilateral clefts. The CL phenotype was found to be associated with the methylation status of the 5’LTR (MboI: $2 Test of Independence, $2 = 9.60, P = .022; EcoRI: $2 Test of Independence, $2 = 7.47, P = .058). The CL embryo with of the Wnt9bNull/clf1, clf2AB genotype had 30% methylation with the MboI digest and 10% methylation with the EcoRI digest, this embryo had a unilateral cleft. There were no CL embryos with the Wnt9bNull/clf1, clf2BB genotype.  !  ,.!  Figure 4.5 Phenotype and Genotype correlation of methylation Status of the Wnt9b-IAP using MboI. Embryos are all Wnt9bNull/clf1 and segregating at clf2, embryos are grouped by their genotype and phenotype at clf2.  Figure 4.6 Phenotype and Genotype correlation of methylation Status of the Wnt9b-IAP using EcoRI. Embryos are all Wnt9bNull/clf1 and segregating at clf2, embryos are grouped by their genotype and phenotype at clf2.  4.4 Discussion  We assessed the methylation status of theWnt9b-IAP 5’LTR in Wnt9bNull/clf1 embryos  segregating at clf2 (clf2AA, clf2AB, clf2BB) using the COBRA assay. We found the methylation status of the 5’LTR to be significantly associated with the genotype at clf2. We noted a similar trend as in previous data (Juriloff et al. 2007) of CL embryos consistently being less methylated than normal embryos with the clf2AA genotype. The strong association between the methylation status of the Wnt9b-IAP and the clf2 genotype indicates that clf2 modifies the epigenetic modification of the Wnt9b-IAP. This finding is consistent with our observation that clf2 does not seem to affect the CL penetrance in Wnt9b null homozygotes (Chapter 3). !  ,'!  Clf2AA embryos were consistently the least methylated, clf2AB embryos had intermediate levels of methylation, and clf2BB embryos were consistently the most methylated. The methylation status of the IAP is responsible for its transcriptional silencing (Rakyan et al. 2002). One hypothesis to explain the data is that a variant in the clf2A allele is interfering with its ability to assist in the process of methylation of the Wnt9b-IAP. Alternatively, the C57BL/6J clf2B allele may provide more than the normal amount of clf2 product, perhaps by gene duplication, and may thereby provide a protective affect by more strongly methylating the IAP. By either mechanism, we hypothesize that the hypomethylated IAP in clf2AA remains active and disrupts the transcription of the Wnt9b gene. When the IAP is active hypomorphic levels of Wnt9b transcript may be produced from the clf1 allele and combined with the null mutation on the other chromosome cause the CL phenotype. The lowest methylation status of the Wnt9b-IAP was found to be associated with the CL phenotype in clf2AA embryos. These results are consistent with previous research on the A/WySn mouse model (Juriloff et al. 2007, Juriloff et al. 2008) and with other mouse models of metastable epialleles (Avy and AxinFu) (Rakyan et al. 2002). All the clf2AA CL embryos clefts were bilateral with the exception of one; this embryo had a right unilateral CL and the 5’LTR was approximately 10% methylated. The one Wnt9bNull/clf1, clf2AB CL embryo was a unilateral right CL and had intermediate levels of methylation for both enzymes. The cells that populate the two sides of the face originate from completely separate populations of neural crest cells (Jiang et al. 2006) and could have differing proportions of silenced IAPs leading to one side of the face becoming cleft and the other fusing normally. The COBRA assay gives an average methylation level for all cells and will not reflect these differences between sides. Therefore, it is not surprising for some embryos to have intermediate levels of methylation and still express the CL phenotype, especially for unilateral CL embryos. !  *)!  The strong association between the methylation status of the Wnt9b-IAP and the genotype of clf2 demonstrates that clf2 modifies the epigenetic modification of the Wnt9b-IAP in the clf1 mutant allele. These results are consistent with the trend noted in Cross 1 and Cross 2 that clf2 does not modify the penetrance of CL in Wnt9bNull/Null embryos. Two mouse models, Avy and AxinFu, have IAPs that are silenced when they are hypermethylated, active when hypomethylated and their phenotypes are dictated by the methylation status of their IAP (Rakyan et al. 2002). Our model of Wnt9bNull/clf1 embryos segregating at clf2 (clf2AA, clf2AB, clf2BB) shows a very similar trend. However, it appears that our study is the first to report the existence and location of a gene responsible for modulating the methylation status of the IAP in a metastable epiallele mouse model.  !  *(!  5  Mapping of clf2: W1-U/REC-4 Lines  5.1 Background Previously, the candidate region for clf2 was defined as being between the genes Ntrk2 and Irx2 on Chromosome 13, a region of approximately 14.0Mb (Juriloff et al. 2004). Subsequent studies using testcrosses of recombinants from the WBC stock (described in General Materials and Methods) were used to reduce the candidate region prior to the beginning of my participation in the study to between CNT C/D and D13M6831 A/B (Figure 5.1).  Figure 5.1 Clf2 candidate region on Chromosome 13 at the start of my studies. Horizontal lines indicate the boundaries of the clf2 candidate region. Chromosome location (Mb) is based on UCSC, July 2007 Assembly. Gene names are in Appendix A. My studies joined and added to this project based on WBC recombinants aimed at minimizing the clf2 candidate region. !  *+!  5.2 Materials and Methods Individual new recombinant mice generally do not produce enough offspring to be testcrossed directly. They therefore were propagated within the WBC stock and several of their male descendents carrying the recombinant chromosome were test-crossed. New recombinants were test-crossed to A/WySn females that provide the necessary maternal effect for CL penetrance. By genotyping the CL embryos from the testcrosses for markers along the candidate region we redefined the location of clf2 (figure 5.2). If the CL embryos are a 50:50 mix of carriers of the recombinant chromosome (AB at markers) and homozygous AA across the length of the haplotype then we can conclude that clf2 is located in the AA region of the recombinant chromosome. If the CL embryos are all homozygous AA haplotype in the segregating region then we can conclude that clf2 is in the region that is AB in the recombinant carriers.  Figure 5.2 Interpretation of location of clf2 based on testcross embryos !  *,!  This method based on test crossing of animals carrying a recombinant chromosome was used in order to obtain “hard” boundaries on the candidate region. Due to the semi-dominant nature of clf2 (1-2% of clf2AB embryos express CL), the common method based on genotyping backcross 1 (BC1) segregant embryos is not useful. The candidate region is now small enough that the probability of a BC1 segregating embryo being a recombinant is approximately equal to the probability of that embryo being a clf2AB penetrant for cleft lip and therefore we can’t distinguish between the CL being due to the residual clf2AA region in a recombinant versus CL being due to clf2AB irrespective of the recombination. Although the method of test crossing new recombinants is more laborious, in the end the candidate region has “hard” boundaries rather than the “soft” probabilistic boundaries that are obtained from segregant panels. This level of certainty is needed for the exclusion of attractive candidate genes near the edges of the candidate region. Around the time of my joining the project, mice with a new WBC recombinant chromosome, “REC4”, began testcrosses and became part of my project. The REC4 breakpoint in the clf2 candidate region on Chromosome 13 is located between the markers ZF456 A/B and D13M6791 A/B, figure 5.3, being haplotype “B” proximal to the breakpoint and haplotype “A” distal to it. REC4 carriers are homozygous for the clf1 mutation at Wnt9b on Chromosome 11. REC4 identified a new breakpoint within the previously defined (unpublished) boundaries. Subsequently mice with another new recombinant chromosome “W1-U” from the WBC stock were ready for test crossing. The W1-U chromosome has a breakpoint between the markers CNT C/D and CNT E/F, figure 5.3, being haplotype “B” proximal to it and haplotype “A” distal to it. The W1-U breakpoint is between the same markers as a previously tested recombinant (“W1-I”), but created the inverse recombinant haplotype. It was used as described below to confirm the boundary previously obtained from W1-I. W1-U carriers are genetically homozygous for the clf1 mutation at Wnt9b on Chromosome 11. !  **!  Figure 5.3 W1, REC-4 and W1-U breakpoints on Chromosome 13. White shading denotes the “A” genotype at marker loci and black shading denotes the “B” genotype at marker loci. REC-4 and W1-U carriers were test-crossed with A/WySn females to further define the clf2 candidate region. Solid horizontal lines indicated the location of the breakpoints. Gene names for markers are listed in Appendix A.  5.2.1 Observation of CL Female A/WySn mice were placed with singly caged W1-U and REC-4 males and they remained until visibly pregnant. Pregnant females were euthanized and their embryos observed and scored for CL as described in the General Materials and Methods. DNA was prepared from all CL embryos and from randomly selected normal littermates. Dr.’s Juriloff and Harris collected all REC-4 testcross embryos and I collected the W1-U testcross embryos.  5.2.2 Genotyping of embryos Genomic DNA was prepared from a small portion of embryonic tissue and embryos were genotyped for the flanking and internal SSLP markers by PCR methodology as described in the General Materials and Methods. REC4 test cross embryos were genotyped at markers CNT E/F, ZF456 A/B, D13M6791 A/B, and MTRR A3/B3. W1-U test cross embryos were genotyped at markers CDC14 C/D, CNT C/D, CNT E/F, and ZF456 A/B.  !  *-!  5.3 Results 5.3.3 REC-4 test cross Five REC-4 carrier males were test crossed with 24 A/WySn females (Table 5.1). The males had an average of 5 litters each (range 2-10). A total of 115 embryos were collected. Of the 115 REC-4 testcross embryos, 104 were normal and 11 had CL. Seven of the CL embryos had a bilateral cleft lip (BCL), three had a left unilateral cleft lip (LCL) and one had a right unilateral cleft lip (RCL). The frequency of clefting in the REC-4 test cross was 9.5% (11/115). “Moles” (early post implantation mortality) were also observed; the REC-4 test cross had 21 moles (21/136=15%). All moles were dead before the lip forms (E10-11) and could not be scored. Table 5.1 Summary of REC-4 test cross Number Number of REC-4 Number Number Number Number of heterozygote males of of of Litters E12-E16 of CL tested Molesa Normal Embryos 5 24 115 21 104 11 a Postimplantation mortality up to E10.  % CL 9.5%  All CL embryos and 28 random normal littermate embryos from 24 litters were genotyped for internal and flanking markers, Table 5.2. All CL embryos had an AA genotype throughout the segregating region (CNT E/F and ZF456 A/B). Of the random normal embryos, 13 had an AA haplotype at the segregating region while 15 were AB (REC-4 carriers). CL was significantly associated with the AA haplotype at CNT E/F and ZF456 A/B (Fisher’s Exact Test P = 0.001). Table 5.2 Phenotypes and genotypes of REC-4 testcross embryos D13M6791 MTRR CNT E/F ZF456 A/B A/B A3/B3 a AA AB AA AB AA AB AA CL-Bilateral 7 0 7 0 7 0 7 CL-Unilateral 4 0 4 0 4 0 4 Normal 13 15 13 15 28 0 28 a A allele is from A/WySn; B allele is from C57BL/6J !  AB 0 0 0 *%!  We attempted to find new SSLP markers between ZF456 A/B and D13M6791 A/B to better define the clf2 region. We designed primers at simple repeat sequences in this region and were able to test two informative markers, 261RIK A/B and ZF71 A/B2. Only the REC-4 males were genotyped for these markers. The new markers proximal to D13M6791 A/B had an AB genotype and therefore the candidate region extends just proximal to D13M6791 A/B, see figure 5.3. We were not able to eliminate any candidate genes with the new markers.  5.3.4 W1-U test cross Four W1-U carrier males were test crossed with 19 A/WySn females (Table 5.3). The males had an average of 5 litters each (range 2-12). A total of 91 embryos were obtained. Of the 91 embryos, 77 were normal and 14 had CL. Four of the CL embryos had a bilateral cleft lip, eight had a left unilateral cleft lip and two had a right unilateral cleft lip. The frequency of clefting in W1-U test cross was 15.4% (14/91). “Moles” were also observed, a total of 18 were noted (18/112=16%) and all were dead before the lip forms at E10-11. Three E12 embryos (2 normal, 1 cleft) had died before the mother was euthanized (based on tissue degradation), the lip was still in good condition to be scored and the embryos were genotyped. Table 5.3 Summary of W1-U test-cross Number Number of W1-U Number Number Number of Number heterozygote males of of of Litters E12-E16 of CL tested Molesa Normal Embryos 4 19 91 17 77 14 a Postimplantation mortality up to E10.  % CL 15.4%  All CL embryos (14) and 14 random normal littermates (one normal littermate per CL embryo) were genotyped for flanking and internal markers, Table 5.4. Seven of the CL embryos were W1-U haplotype carriers (AB) and seven were AA at the segregating haplotype. The normal littermates had eight W1-U haplotype carriers (AB), five with the AA genotype across all markers, and one embryo was a de novo recombinant in the W1-U haplotype with a !  *&!  recombination between the proximal marker outside the candidate region, CDC14 C/D, and CNT C/D so that it was A, not B, at CDC14 C/D while B at CNT C/D, and contributed no new information regarding the location of clf2. Cleft lip was found to be just as likely to occur in the W1-U carriers (AB at CNT C/D) as in the AA embryos (Fisher’s Exact Test P = 0.352). This demonstrates that clf2 is not located in the segregating region at CNT C/D, but is located in the AA region (CNT E/F through MTRR A3/B3) of the W1-U chromosome. Table 5.4 Phenotypes and genotypes of W1-U test cross embryos CDC14 C/D CNT C/D CNT E/F ZF456 A/B AAa AB AA AB AA AB AA AB CL-Bilateral 2 2 2 2 4 0 4 0 CL-Unilateral 5 5 5 5 10 0 10 0 b Normal 6 8 5 9 14 0 14 0 a A allele is from A/WySn; B allele is from C57BL/6J b De novo recombination event.  5.4 Discussion In the embryos from the REC-4 X A/WySn test cross, cleft lip was significantly (P < .001) associated with AA haplotype segregants for markers proximal to the D13M6791 A/B marker. The REC-4 chromosome, therefore, redefines the lower boundary of the candidate region to be between ZFP71 A/B2 and D13M6791 A/B, figure 5.4.  Figure 5.4 The new candidate region for clf2, defined by REC4 and W1-U. White shading denotes the “A” genotype at marker loci and black shading denotes the “B” genotype at marker loci. Solid horizontal lines indicated the location of the breakpoints. Gene names for markers are listed in Appendix A. ! *.!  The frequency of clefting in the REC-4 test cross was 9.5%, about 50% of typical A/WySn levels of 15-20%. As only half of the REC4 testcross progeny were AA between CNT E/F and D13M6791 A/B, this frequency is consistent with the conclusion that the clf2 gene is within this interval. To better define the REC-4 breakpoint we identified new markers (261RIK A/B and ZFP71 A/B2) at simple repeat sequences between ZF456 A/B and D13M6791 A/B. The breakpoint is immediately proximal to D13M6791 A/B. The testcross embryos from the W1-U recombinant, which segregated proximal to the marker CNT E/F, indicated that the cleft lip phenotype is independent of genotype in this region and indicates that the clf2 gene is located distal to CNT C/D, in the “A” region of the W1-U chromosome. The clefting frequency in the W1-U test cross embryos was 14.8%, very similar to the A/WySn clefting frequency. This is consistent with the conclusion that clf2 is located distal to CNT C/D that was not segregating and this confirms a previous (unpublished) candidate region boundary between CNT C/D and CNT E/F from an inverse recombinant haplotype, W1-I.  !  *'!  6  Mapping clf2: Haplotype Survey  6.1 Background A haplotype is a combination of alleles at markers that are transmitted together in a contiguous region of a chromosome. By looking at the haplotype it can be inferred which ancestral strain that region of chromosome was derived from. Using this approach we had three separate goals in mind. The first goal was to investigate if A/WySn had a unique haplotype in the clf2 candidate region. Using this information we could possibly prioritize sub-regions of the candidate region, if there is a small overlapping area of “A” haplotype shared between other closely related CL strains within the clf2 region. The common ancestors of the A/-strains are known to share pieces of chromosome among each other (Juriloff and Harris 2008) and therefore we looked at the clf2 haplotypes of closely related mouse strains (figure 1.3). The strains that we surveyed are A/WySn, A/J, A/HeJ, BALB/cByJ, BALB/cGa, SELH, CL/Fr, DBA/1J, CBA/J, C3H/HeJ, and the L strain. Of these strains, A/WySn, A/J, A/HeJ, CL/Fr, and the L strain have CL/P. A/WySn, A/J, and A/HeJ are direct descendents of the original A inbred strain. FC Fraser created CL/Fr in the 1960s; he out-crossed A/J to a heterogeneous stock (“migratory spot lesion”) and selected for CL/P while inbreeding (Juriloff and Harris 2008). CL/Fr has the clf1 mutation and produces 20-25% CL/P (Juriloff and Harris 2008). The L strain was created from a four way cross between CL/Fr, DBA/1J, C57BL/6Fr, and NS/Fr; L Line heterogeneous breeders were selected for production of CL/P following maternal treatment with 6-aminonicotinamide (6-AN) for several generations (Juriloff and Harris 2008). At F15 the L Line produced 9% CL/P in the absence of treatment. They were then maintained and inbred to form the L strain.  !  -)!  Subsequent analysis of inbred markers indicates that the clf1 gene from A/J and possibly clf2 were selected and fixed in the strain (Juriloff and Harris 2008). The second goal of examining haplotypes was to look for recombination events in the clf2 candidate region in CL embryos from testcrosses of a congenic line, AEJ.A. This congenic strain pair, made up of AEJ.A/Jur and AEJ/RkBc, differs only by chromosomal segments containing the clf1 and clf2 genes from A/WySn (Juriloff and Mah 1995, Juriloff, unpublished data). This congenic strain was originally used to map clf1 (Juriloff and Mah 1995). The clf2 candidate region has been much more narrowly defined since the collection of the AEJ.A testcross embryos and it seemed worthwhile to explore the haplotype of the available archived CL embryos from the testcross of AEJ.A males. Congenic strains reflect many accumulated opportunities for recombination. The CL embryos are from testcrosses of A/WySn females by clf1AB, clf2AB carrier males (AEJ.A 1004, 1115, 1136, 1151, 1157, 1164) (Juriloff and Mah 1995). These males are shown in figure 2 in Juriloff and Mah (1995) or are direct descendents of other males shown (981, 982, 1002, 1004). We hoped that some of these CL embryos might reflect a recombination event within the clf2 candidate region and help us to prioritize our candidate region, with caution that although 15-20% of clf2AA have CL, 1-2% of clf2AB embryos are also penetrant for CL. This would mean that although a clf2AB region in a CL embryo could not be excluded from the clf2 candidate region, it would be much more likely to be excluded than included. The third goal of our haplotype studies was to identify the origin of the “C” allele at loci in the clf2 candidate region in the Wnt9b knockout carrier stock. The two original parental Wnt9b+/Null sires had a BC genotype at D13Wsu50e and the linked marker Dapk1 (one allele is obligate B from C57BL/6J) (Juriloff et al. 2006) and subsequent unpublished data). The ancestors of the parental Wnt9b+/Null sires are 129/SvJ, Swiss Webster and C57BL/6J meaning that the 129/SvJ or the Swiss Webster non-inbred stock donated the “C” allele. The effect on ! -(!  clefting frequency in Wnt9bNull/clf1 compound mutant embryos of the “C” allele at clf2 is intermediate to the effects of clf2A or clf2B (Juriloff et al. 2006). It is possible that clf2 variants are functional polymorphisms (Juriloff et al. 2006) that could be located through haplotype analysis of the ancestral strains. To identify the origin of the clf2C variant, we obtained the haplotypes for SSLP markers of the original Wnt9b+/Null sires (Wnt9b+/Null 9609-2 and Wnt9b+/Null 9609-3), C57BL/6J, 129/SvJ, and SWR/J (an inbred “Swiss” mouse). Swiss Webster are random bred heterogeneous “Swiss” mice and therefore a random Swiss Webster mouse is not expected to provide a representative sample of the SSLP haplotype. The SWR/J strain is equivalent to a random bred Swiss mouse and was included on the possibility that it might provide useful information.  6.2 Materials and Methods 6.2.1 Haplotype analysis Genomic DNA from the various mouse strains was obtained from Juriloff Laboratory archived tissue or DNA or from the Jackson Laboratory (Bar Harbor, ME) (129/SvJ, SWR/J, DBA/1J) and genotyped for flanking and internal SSLP markers throughout the clf2 candidate region by PCR methodology by lab staff, particularly Mandy So and Sarah Kennedy, and myself as described in the General Materials and Methods. The mouse strains were genotyped at markers: CDC14 C/D, CNT C/D, CNT E/F, HIATL C/D, AK145 A/B, MTER A/B, MTER A/B, PTDS A2/B2, PTDS E/F, PTDS G/H, ZF456 A/B, 261RIK A/B, D13M6791 A/B, D13M6831 A/B and MTRR A3/B3 (in order from proximal to distal).  6.2.2 SNP analysis Using the Mouse Genome Informatics Database (Mouse Genome Database [MGD], 2009-2010) we conducted a SNP Query on Chromosome 13 between the genome coordinates 67,029,683-67,928,500 bp (region between markers PTDS A2/B2 and D13M6791 A/B) on May !  -+!  3, 2010. The database contains information for various inbred mouse strains and allows the use of one strain as the reference for comparison with other strains. We did a number of queries; we set C57BL/6J as the reference strain and compared to A/J; we also set A/J as the reference strain and separately selected BALB/cByJ, A/HeJ, DBA/1J, CBA/J or C3H/HeJ for comparison. C57BL/6J is the only strain that has been fully sequenced through this region; the other strains are based on fragments of the region. We searched for any variation type in any function class. We first searched for all SNPs in the region and then narrowed the search to SNPs that are different between the reference strain and the selected strain.  6.3 Results The clf2 candidate region SSLP haplotypes of the various mouse strains related to A/WySn can be found in Table 6.1. The genotypes are labeled “A” if they are the same product size as A/WySn, “B” if they are the same as C57BL/6J and labeled with the estimated product size if they are neither “A” nor “B”. For the current clf2 candidate region A/WySn, A/J, A/HeJ, BALB/cByJ, BALB/cGa and SELH share the same haplotype at the markers tested, and given their shared ancestry it seems reasonable to conclude that they have a haplotype identical by descent. Among these strains, BALB/cByJ, BALB/cGa and SELH differ from the others at a flanking locus, CDC14 C/D. CL/Fr, DBA/1J, CBA/J, C3H/HeJ and the L strain have “chunks” of similarities to A/WySn but are also very different in some areas. CL/Fr and its descendant, the L strain, share the same haplotype through the current clf2 candidate region. They differ at CDC14 C/D. The CL/Fr and L strain haplotype through the current clf2 candidate region seems to be the same as CBA/J and C3H/J. Within the candidate region DBA/1J differs from these at CNT E/F and HIATL C/D. The marker AK145 A/B gives two different products (2 bands) in CL/Fr, DBA/1J, CBA/J, C3H/J and the L strain; this could be due to a duplication of the region. The AK145 A/B SSLP !  -,!  is within the Ak145544 gene, which is a non-coding, hypothetical zinc finger protein (see Chapter 7). The SSLP for which the markers are designed has been duplicated (UCSC genome browser, Mouse, July 2007) in C57BL/6J, where the identical sequence between the two primers (108 bp) is found at two locations on Chromosome 13 (UCSC genome browser, Mouse, July 2007). In C57BL/6J we see only one band because the fragment is perfectly duplicated and the product size is the same. This is probably an ancient duplication and the strains with two product sizes have an altered sequence between the two primers. The strains with only one product size different from C57BL/6J (A/WySn, A/J, A/HeJ, BALB/c, BALB/cGa and SELH) could have been not duplicated in the first place or could have lost a copy. CL/Fr and the L strain share an “A” haplotype with the A/- strains, BALB/cByJ, BALB/cGa and SELH at the distal part of the clf2 candidate region between PTDS A2/B2 and D13M6791 A/B. The shared A/- haplotype in all of the stains surveyed between PTDS A2/B2 and D13M6791 A/B is intriguing. To further investigate this region we looked at the SNP differences between various available strains using the Mouse Genome Database. The sequence data for the mouse strains is not complete but the total number of SNPs probably reflects the amount of SNP data available for each strain in part. We first compared C57BL/6J and A/J and the search yielded a total of 1555 SNPs in this region and they differ at 1270 SNPs. A comparison between A/J and BALB/cByJ yielded a total of 1412 SNPs in the region and A/J and BALB/cByJ differ at 8 SNPs. C3H/HeJ and A/J have a total of 1333 SNPs and they differ at 164 SNPs. The small number of SNPs differing between A/J and BALB/cByJ demonstrates how closely related the two strains actually are and gives confidence that A/J and BALB/cByJ share the same haplotype. A/HeJ, CBA/J and DBA/1J had a total of 0, 3, and 4 total SNPS, respectively compared with C57BL/6J, and differ at none. This demonstrates a lack of sequence information rather than a lack of differing SNPs in the region. !  -*!  The genotypes of the AEJ.A X A/WySn CL testcross embryos throughout the clf2 candidate region are in Table 6.2. AEJ/RkBc is the background that the AEJ.A congenic line is on; they have a “B” haplotype throughout and flanking the clf2 region. All embryos are obligate “A” on one chromosome from their A/WySn mother. The haplotypes of 11 CL embryos were analyzed. Nine of the CL embryos have a standard “AA” genotype across the region. CL 141 reveals a recombination event between 261RIK A/B and D13M6791 A/B, coincidentally this is the same as the REC4 breakpoint but is the inverse relationship. CL 144 has an “AB” haplotype at clf2; as 1-2% of clf2AB embryos express CL, this is not unexpected. None of the CL embryos examined had a breakpoint within the current candidate region. Table 6.3 shows the haplotypes of the original Wnt9b+/Null sires, 129/SvJ, C57BL/6J and SWR/J. The original Wnt9b+/Null sires are obligate “B” on one chromosome from a C57BL/6J parent and they could have inherited the “C” allele from either 129/SvJ or Swiss-Webster. The clf2C chromosome from the original sires is the same as the 129/SvJ haplotype.  !  --!  !  Table 6.1 Haplotypes of various mouse strains related to A/Wysn.  $!%&'!($)*+,!-'++.(/0!$!*1-/.2$3.')4!  567!*8)'38+!+$98!-:'*123!+.;8!$+!6<=0>)?!5@7!*8)'38+!+$98!-:'*123!+.A8!$+!B"C@D<#E! @'/*!F')3!.)*.2$38+!21::8)3!!"#$%2$)*.*$38!:8G.')4!  !  "#!  Table 6.2 Haplotypes of CL embryos from AEJ.A X A/WySn test cross.  %$(/8!#4H!!I$-/'30-8+!'F!BD!89(:0'+!F:'9!6JE46!K!6<=0>)!38+3!2:'++4!  567!*8)'38+!3L8!+$98!-:'*123!+.A8!$+!6<=0>)?!5@7!*8)'38+!3L8!+$98!-:'*123!+.A8!$+!B"C@D<#E4! @'/*!F')3!.)*.2$38+!21::8)3!!"#$!2$)*.*$38!:8G.')4! !  !  "C!  Table 6.3 Haplotype of the original non-C57BL/6J chromosome of the Wnt9b-/+ males and haplotypes of their ancestors. +/Null Wnt9b+/Null 129/SvJ SWR/J Marker C57BL/6J A/WySn Wnt9b CDC14 C/D CNT C/D CNT E/F HIATL C/D AK145 A/B MTER A/B PTDS A2/B2 PTDS E/F PTDS G/H ZF456 A/B 261RIK A/B D13M6791 A/B D13M6831 A/B MTRR A3/B3  B B B B B B B B B B B B B B  A A A A A A A A A A A A A A  9609-2 65 A 155 135 B, 90 135 250 A 110 140 A 100 A A  9609-3 65 A 155 135 B, 90 135 250 A 110 140 A 100 A A  65 A 155 135 B, 90 135 250 A 110 140 A 100 A A  85 145 A 135 100,105 120 340 190 A A 100 95 140  $%&!'()*+(,!,-.(!/0*'12+!,34(!-,!%5678)9!$:&!'()*+(,!,-.(!/0*'12+!,34(! -,!;"<:=5>?@! :*A'!B*)+!3)'32-+(,!2100()+!!"#$%2-)'3'-+(!0(C3*)@! !  6.4 Discussion The haplotypes of mouse strains that are related to A/WySn indicate that A/WySn shares its haplotype at clf2 with other closely related strains. A/J, A/HeJ, BALB/cByJ, BALB/cGa and SELH all have the same haplotype as A/WySn in the clf2 candidate region suggesting that the non A/- strains, BALB/c and SELH, are possibly carriers of the clf2 allele. BALB/cByJ and A/J only differ by 8 SNPs in the region between PTDS A2/B2 and D13M6791 A/B (a 0.9 mb region) demonstrating that this region is highly conserved between A/J and BALB/cByJ. CL/Fr and its descendant the L strain share a haplotype in the clf2 candidate region. CL/Fr is known to carry the clf1 mutation (Juriloff et al. 2005a) and produces 20-25% CL/P in their offspring (Juriloff and Fraser 1980). This suggests that CL/Fr probably has some version of clf2. CL/Fr shares the “A” haplotype with the A/- strains in the distal candidate region. From the above results we can infer two different hypotheses. The first is that clf2 is located in the “A” chunk of chromosome between PTDS A2/B2 and D13M6791 A/B. The second is that clf2  !  "#!  is located in the region between CNT C/D and PTDS E/F where CL/Fr and the L strain differ from A/WySn and that CL/Fr and the L strain have a different clf2 allele. The finding that BALB/cGa and SELH have the same haplotype as A/WySn at clf2 is intriguing, based on previous research on BALB/cGa and SELH. BALB/cGa and SELH have had periods of unusually high mutation rates. A wave of mutations of unknown cause occurred in BALB/cGa of Dr. Allen Gates at Stanford University from 1962-1969 (Juriloff et al. 2005b). Another spontaneous mutation occurred in BALB/cGa in 1982 in the Juriloff colony at UBC (Juriloff et al. 2005b). A second wave of mutations occurred in the SELH strain (descended in part from BALB/cGa, figure 3) in 1978-1994 with 11 total mutations of which two are ETn insertions and three are large deletions; the remainder are not defined at the DNA level (Juriloff et al. 2005a). Juriloff et al. (2005) calculated that the mutation rate in SELH was about 10X higher than normal and hypothesized that a factor causing this was inherited from BALB/cGa and possibly located on mid-chromosome 13 in the Exen1 candidate region, an allele necessary for exencephaly risk in SELH/Bc (Juriloff et al. 2001a). The Exen1 candidate region (Juriloff et al. 2001a) is thought to overlap with the clf2 candidate region. SELH/Bc was found to have a higher level of MusD and Early Transposon (ETn) elements of the ETnII group transcripts compared to C57BL/6J and a young, coding competent subtype of MusD elements that are not found in C57BL/6J; these MusD elements are thought to facilitate retrotransposition of ETnII elements and therefore explain the high mutation rate in the SELH/Bc strain (Baust et al. 2003). As a reminder, the clf1 mutation is a de novo insertion of a transposable IAP element downstream of Wnt9b and clf2 is on Chromosome 13. Therefore, the shared haplotypes of A/WySn, BALB/cGa and SELH/Bc in the clf2 candidate region and the involvement of transposons in both SELH and A/WySn give further evidence that clf2 has a role in transposon activity. !  "D!  C3H/HeJ is known to be highly susceptible to endogenous retrovirus (ERV) insertions, especially to IAP insertions (Maksakova et al. 2006). C3H/HeJ also has 50% more total IAP copies than C57BL/6J; similarly, A/J and BALB/cJ have 20% more total IAP copies than C57BL/6J (Mager DM, personal communication). There may be an undetected sequence difference between the group comprising A/- strains, CL/Fr, L strain, BALB/cByJ, BALB/cGa, C3H/HeJ and SELH, versus DBA and CBA for the shared “A” region between PTDS A2/B2 and D13M6791 A/B. There are 164 SNPs differing between A/J and C3H/HeJ in the region between PTDS A2/B2 and D13M6791 A/B showing that C3H/HeJ does not have a complete A/haplotype in this region and we should use caution when exploring this region. Nonetheless, we may want to use “A” region of the C3H/HeJ haplotype as a starting point for subsequent clf2 studies as C3H/HeJ may harbor a different variant of clf2. Conversely, C3H/HeJ is the same as CL/Fr in the other part of the candidate region. Maybe clf2 is there, and CL/Fr has the same version as C3H/HeJ. We also looked at the haplotype of CL embryos from the AEJ.A congenic line, in search of recombinants. These embryos are from a test cross of A/WySn females to clf2AB males. We discovered a recombination event in CL 141 between 261RIK A/B and D13M6791 A/B; coincidentally, this is the same breakpoint as in REC4 (discussed previously). The other 10 embryos all had a clf2AA haplotype and do not provide any additional information. The final haplotype survey was to find the ancestral origin of the clf2 “C” allele from the original Wnt9b+/Null males. The “C” allele leads to a clefting frequency that is intermediate to the “A” and “B” alleles. The Wnt9b+/Null males have a 129/SvJ ancestor and have the same haplotype as 129/SvJ and therefore the ancestor of the “C” allele is probably 129/SvJ. The intermediate clefting frequency of the “C” allele poses interesting questions about the nature of clf2. !  >E!  7  Mapping clf2: Candidate Region Database Search  7.1 Background The clf2 candidate region is now approximately 3.0 mb and recombinants within it seem too rare for further screens for recombinants for test crossing. The next step in locating clf2 may be to use in vitro studies with a “reporter” system of some type. The candidate region is still too large to include all of the candidate genes in an in vitro study and therefore must be prioritized to test the most likely candidates first. Using a database search we can familiarize ourselves with the genes in the region and prioritize our list of candidate genes. The function, expression domain and potential functional changes of the candidate genes were the basis of prioritizing the genes. The functions of genes that are of particular interest are those that could have an effect on the methylation status of an IAP, such as methylation specific genes or genes with a role in gene silencing. The function of the genes in the region will give us an idea of the types of genes in the region and we can prioritize our list based on those genes most likely to have connections to methylation status of an IAP element. The methylation status of the IAP is set in the very early embryo and from previous research is thought to be similar throughout the body (Juriloff, unpublished data). IAPs are thought to behave like imprinted genes and therefore are speculated to undergo methylation in the testis at E13.5 and in the ovary after birth in the mouse (Sasaki and Matsui 2008). There are waves of global demethylation and remethylation of the genome during embryonic development; the sperm and egg genomes are both methylated, during preimplantation development the genome begins a demethylation process and by the blastocyst stage the genome is not detectably methylated, remethylation begins in gastrulation and continues post gastrulation (Kafri et al. 1992, Monk et al. 1987). IAP transcripts are observed in pre-implantation embryos (Kuff and !  >F!  Lueders 1988); and therefore I hypothesize that IAPs are demethylated and active at this time. I began my search for genes that were expressed in the early embryo from the one-cell stage through gastrulation, the ovary and the testis. The clf2 gene would likely be expressed during the initial stages of methylation the ovary/testis or during the waves of demethylation and remethylation and may have an effect on the remethylation of the IAP. Sequence differences between C57BL/6J and A/J may reveal potential functional changes in candidate genes. Various SNPs could cause non-synonymous changes in the coding sequence, affect splice sites or be in a regulatory region. Candidate genes with SNPs that would have a negative effect on the protein would be our top priority to study; from the haplotype survey and previous data we know that the clf2 allele in C57BL/6J does not have the same function as in the A/- strains so it is possible that a SNP difference between the A/- strains and C57BL/6J could cause the necessary functional change at clf2. The goal of this database search is to prioritize the genes in our candidate region based on their function, expression and potential functional changes. The prioritized candidate gene list will be useful in future in vitro studies where it is impractical to study all of the genes in the clf2 candidate region.  7.2 Materials and Methods 7.2.1 UCSC search Using the UCSC Genome Browser (UCSC genome browser, Mouse, July 2007) I conducted a search of the genes in our candidate region between CNT C/D and D13M6791 A/B. I searched within the clf2 candidate region on Chromsome 13 at 64,893,294 - 67,907,310 in the Mouse, July 2007 assembly (January 2009). From this search I obtained a list of the genes or predicted genes in the region. A search for each of the genes independently gave a gene description, location, function, some expression data and human conservation. !  >G!  7.2.2 MGI search Using the gene names from the UCSC genome browser, I conducted a search in the Mouse Genome Informatics Database (MGI) (Mouse Genome Database [MGD], 2009-2010) (January 2009). MGI and UCSC do not always use the same nomenclature but MGI has a list of synonyms and usually contains the UCSC gene name. For consistency, I will refer to the genes by the UCSC name. The MGI database has information for each gene on function and expression. This information was combined and used in addition to the UCSC information.  7.2.3 SNP query On MGI (Mouse Genome Database [MGD], 2009-2010) I conducted a SNP Query on Chromosome 13 between the genome coordinates 64,893,294-67,907,310 bp (September 2009). The database contains information for various inbred mouse strains and allows you to set one strain as the reference for comparison with other strains. We used C57BL/6J as the reference strain and compared it to A/J for all SNPs that are different between the two strains. I conducted separate searches for non-synonymous and synonymous SNPs in the coding region, SNPs in the introns of candidate genes, SNPs in locus regions (untranscribed region flanking a gene) of candidate genes, SNPs in mRNA-UTR regions, and SNPs at splice sites.  7.3 Results A database survey of the clf2 candidate region revealed 48 genes or predicted genes. The candidate genes were individually inspected for gene function, expression domain and potential functional changes (SNPs) using the UCSC and MGI databases.  7.3.1 Functions of candidate genes The full candidate gene list and their known functions can be found in Appendix D. Many (21) of the candidate genes are transcription factors, all of which are of the zinc finger protein (Zfp) type. With the exception of one gene, Ak019536, all of the Zfps have a Krüppelassociated box (KRAB) domain. Unfortunately, most of these transcription factors have !  >H!  unknown targets. Many (16) of the candidate genes are non-coding or have unknown function and do not relay any information or clues as to their function. Some of the genes are receptors (2), membrane proteins (2) or mitochondrial proteins (3) but none of their specific functions seem to easily correlate with methylation status of an IAP. There are some hypothetical proteins (4) in the candidate region and they also contain zinc fingers and a KRAB domain. See Table 7.1 for a list of candidate genes for each function. Table 7.1 Summary of candidate gene function Function Genes Unknown  4930458L03Rik, 4932441B19Rik, Ak163421  Non-coding or Near coding  Ak145544a, Ak021218, Ak049221, 2410141K09Rik, Ak076976, NR_003702, Bc049692, Bc038328 Ak021377, Ak020476, Ak015755, Ak013687, Ak029746  Receptors  Cntnap3, Olfr466  Transcription Factors  Zfp369, 6D, Ak019536, Zfp708, Zfp759, Rsl1, Zfp455, Zfp458, Zfp457, Zfp595, Zfp456, 2810487A22Rik, Zfp459, C330011K17Rik, 9630025I21Rik, Zfp817, Zfp87, Zfp748, Aa987161, Zfp71-rs1, Zfp85-rs1  Extra-cellular Signaling  0  Membrane Protein  Hiatl1, Ptdss1  Mitochondrial Proteins  Nlrp4f, Uqcrb, Mterfd1  Hypothetical proteins  E130120F12Rik, A530054K11Rik, 3830402I07Rik, Ak140336  a  Ak145544 is duplicated in the C57BL/6J genome and probably in CL/Fr, DBA/1J, CBA/HeJ, C3H/HeJ and the L strain, and has perhaps only 1 copy in A/WySn, A/J, A/HeJ, BALB/cByJ, BALB/cGa, and SELH.  7.3.2 Expression domains of candidate region Using the MGI database, I searched for gene expression domains for each of the candidate genes. The expression domains that I was most interested in are the ovary, testis, and the early embryo from the one cell stage through gastrulation. Much of the expression data is incomplete and the majority of genes do not have data available for the specific time frame or tissue we are looking for. The expression data that was available is outlined in Table 7.2. The  !  >I!  types of expression domain results available on the MGI database are RT-PCR, cDNA source data and RNA in situ hybridization.  Tissue/Stage Ovary Testis  Table 7.2 Summary of candidate gene expression # of genes Genes expressed in designated tissue with no data Nlrp4f, Zfp369, Uqcrb, Rsl1, Zfp748, Zfp85-rs1 42 4930458L03Rik, 4932441B19Rik, Bc049692, Uqcrb, Mterfd1, Ptdss1, Ak019536, Zfp759, Zfp458, Zfp595, 35 Zfp748, Ak015755, Ak029746  1 cell stage  44  Nlrp4f, Mterfd1, Zfp708, Zfp817  2 cell stage  45  Nlrp4f, Ak163421, Ptdss1  4 cell stage  44  Uqcrb, Mterfd1, Ptdss1, Zfp595  8 cell stage  42  Hiatl1, Nlrp4f, Mterfd1, Ptdss1, Zfp459, Zfp817  16+ cell stage  44  Hiatl1, Nlrp4f, Uqcrb, Ptdss1  Blastocyst  37  Hiatl1, Nlrp4f, Ak14544, Uqcrb, Mterfd1, Ptdss1, Zfp759, Zfp459, C330011K17Rik, Zfp87, Zfp71-rs1  Gastrulation  46  Ptdss1, Zfp708  7.3.3 Potential functional changes in candidate genes. An MGI database search yielded 2,275 SNPs that differ between C57BL/6J and A/J in our candidate region (Chr 13: 64,893,294-67,907,310 bp). Table 7.3 outlines genes with SNPs in locations that could potentially cause functional changes. There are 38 non-synonymous SNPs in the coding regions of 18 clf2 candidate genes from UCSC. These genes all have a change in their amino acid coding sequence and are of the most interest to us due to the high likelihood of these SNPs causing a functional change. There are also 17 UCSC clf2 candidate genes with synonymous SNPs in their coding regions. Many genes in the region have SNPs in their introns, locus regions, or the mRNA-UTR; these SNPs could be located in important regulatory regions and are still of interest to us. There are no known SNPs at splice sites in the candidate region.  !  >"!  Table 7.3 Summary of SNP query of the clf2 candidate region. Number Genes Location of SNPs Coding: Non-synonymous  38  Cntnap3, 4930459L03Rik, Olfr466, Nlrp4f, Zfp369, Zfp708, Rsl1, Zfp458, Zfp595, E130120F12Rik, Zfp456, Zfp429, Zfp459, Zfp874, 9630025I21Rik, Zfp748, A530054K11Rik, Zfp85-rs1  Coding: Synonymous  47  Hiatl1, Olfr466, Nlrp4f, Zfp369, Ak019536, Zfp708, Rsl1, Zfp455, Zfp458, Zfp595, Zfp456, 9630025I21Rik, Zfp817, Zfp87, Zfp748, A530054K11Rik, Zfp85-rs1  Intron  942  Cntnap3, 4930459L03Rik, 4932441B19Rik, Hiatl1, Nlrp4f, Zfp369, Mterfd1, Ptdss1, Ak019536, Zfp708, Zfp759, Rsl1, Zfp455, Zfp458, Zfp457, Zfp595, E130120F12Rik, Zfp456, Zfp429, Zfp459, C330011K17Rik, 9630025I21Rik, Zfp817, Zfp87, Zfp748, A530054K11Rik, Zfp85-rs1  Locus Region  141  4932441B19Rik, Hiatl1, Olfr466, Nlrp4f, Zfp369, Uqcrb, Mterfd1, Ptdss1, Ak019536, Zfp708, Zfp759, Rsl1, Zfp455, Zfp458, Zfp595, E130120F12Rik, Zfp429, Zfp459, C330011K17Rik, 9630025I21Rik, Zfp817, Zfp87, Zfp748, A530054K11Rik  mRNA-UTR  93  4932441B19Rik, Hiatl1, Zfp369, Prdss1, Zfp712, Zfp708, Zfp759, Rsl1, Zfp455, Zfp458, Zfp595, E130120F12Rik, Zfp456, Zfp429, Zfp459, 9630025I21Rik, Zfp817, Zfp87, Zfp748, A530054K11Rik, Zfp85-rs1  Splice Site  0  0  7.4 Discussion The clf2 candidate region contains 48 candidate genes with the largest group being zinc finger proteins (Zfps) with KRAB domains. There are about 2,275 SNPs differing between C57BL/6J and A/J, 38 being non-synonymous SNPs in coding regions. This information has !  >>!  enabled us to prioritize our candidate region and select the top candidate genes as a starting point in future studies. The clf2 candidate genes have a variety of different functions and many do not immediately show connections to the proposed function of clf2, to modify the methylation status of an IAP element. The majority of the candidate genes are Zfps with KRAB domains; KRAB domain Zfps are strong transcriptional repressors and make up about 1/3 of 800 different Zfps present in the human genome (Urrutia 2003). KRAB domain Zfps are usually arranged in clusters, the best characterized human cluster being on 19q (Urrutia 2003). KRAB domain Zfps have a DNA binding domain made up of 4-30 zinc fingers and a KRAB domain, they are able to bind directly to the DNA through C2H2 zinc finger domains (Urrutia 2003). One proposed mechanism of gene silencing by Zfps with KRAB domains is by teaming with KAP-1, transcriptional co-repressor for the KRAB-Zfp family. The Zfps bind to regulatory elements and the KRAB domain will recruit and bind to KAP-1(Schultz et al. 2002). KAP-1 will then become a molecular scaffold for coordinating histone methylation (SETDB1), histone deacetylation (HDAC) and the deposition of heterochromatin protein 1 isoforms (HP1) to silence gene expression by forming a facultative heterochromatin environment on the target’s promoter (Schultz et al. 2002). The wide array of KRAB domain Zfps in the human genome provides the opportunity for this gene family to target a variety of different DNA sequences and become a major regulator of epigenetic programs of gene silencing during cellular differentiation and development (Schultz et al. 2002). In order for KRAB domain Zfps to be good candidate genes for clf2 they need to have the ability to silence the IAP element at Wnt9b (Wnt9b-IAP). When the LTR of the Wnt9b-IAP is methylated (50%), or silenced, the face is normal but when the Wnt9b-IAP is unmethylated (010%) cleft lip is present demonstrating that transcription of Wnt9b is affected (Juriloff et al. 2007). Recently there has been evidence that embryonic stem cells (ESCs) use ZFP809 to silence ! ><!  retroviral DNAs (Wolf and Goff 2009). They have shown that ZFP809 recognizes and binds to the primer binding site (PBS) of murine leukemia virus (MLV) and then recruits TRIM28 (KAP1), a transcriptional co-repressor, to mediate retroviral silencing. ZFP809 is also able to repress transcription of human T-cell lymphotrophic virus-1 (HTLV-1) through this same mechanism. MLV and HTLV-1 have the same PBS leading to the hypothesis that different Zfps are able to recognize different PBS elements and recruit TRIM28 (KAP-1) for proper silencing of the integrated retroviruses. KAP-1 deletion leads to an up-regulation of IAP elements in mouse ES cells and early embryos and the loss of histone 3 lysine 9 trimethylation (H3K9me3), which is the repressive mark of KAP-1 (Matsui et al. 2010, Rowe et al. 2010). We can speculate that the Zfps in the clf2 candidate region functions to silence the Wnt9b-IAP making KRAB domain Zfps our top candidate genes. There is a genomic locus on Chromosome 13 called the Regulator of sex-limitation (Rsl) locus; the Rsl locus is 1mb and overlaps with the clf2 candidate region below Ptdss1 and extends outside the candidate region to Adcy2 (Krebs et al. 2005). This region contains over 20 KRAB domain Zfps and from comparative sequence analysis the Zfps are known to arise from simple tandem duplications and inversions (Krebs et al. 2005). There has also been at least one known gene conversion event in which some strains lost one gene and now has a hybrid of two genes (Krebs et al. 2005). The Rsl locus has many coding region polymorphisms and discrete gene conversion haplotypes within several mouse strains demonstrating that this locus is rapidly diversifying within the mouse species (Krebs et al. 2005). The striking differences between the inbred strains that Krebs et al. (2005) studied (C57BL/6J, B.10D2, BALB/cJ, NZB/B1J, 129/SvEvTac, PL/J, B10.D2.PL, B10.D2.FM) leads us to speculate that there could be striking differences in the clf2 candidate region between C57BL/6J and the A/- strain. There are about 21 genes in the candidate region that are KRAB domain Zfps. To further reduce the number of top priority genes to study we looked at the SNPs that are different !  >#!  between C57BL/6J and A/J and found a total of 2,275. C57BL/6J is thought to have the “wildtype” version of clf2 or a protective version. SNPs in the clf2 candidate region that are different between C57BL/6J and A/J could reveal potential functional changes that could cause the clf2 variant, especially non-synonymous SNPs as they cause a change in the amino acid sequence. There are 18 genes with non-synonymous SNPs in the coding regions of clf2 and 14 of these genes are known KRAB domain Zfps (Zfp369, Zfp708, Rsl1, Zfp458, Zfp595, E130120F12Rik, Zfp456, Zfp429, Zfp459, Zfp874, 9630025I21Rik, Zfp748, A530054K11Rik, Zfp85-rs1). There are many other SNPs in the candidate region that are synonymous SNPs in coding regions or located in the introns, locus regions or mRNA-UTRs. These SNPs are less likely to cause functional changes in the protein and therefore will not be used when prioritizing our candidate region at this time. The expression domain information on the MGI database was too limited to be useful. Our best candidate genes would be expressed in the very early embryo, through gastrulation, or in the ovaries or testis. We know that the methylation status of the IAP is similar in the abdomen and the facial prominences and may reflect that the status is set very early in development and that there are waves of demethylation and remethylation throughout embryonic development. Gastrulation is when the major remethylation of the fetal genome occurs (Kafri et al. 1992, Monk et al. 1987), but only two of the candidate genes had expression data available for this stage, Ptdss1 and Zfp708. Based on the gene functions and non-synonymous SNPs in the candidate region, there are 14 genes that are our top candidate genes. They are: Zfp369, Zfp708, Rsl1, Zfp458, Zfp595, E130120F12Rik, Zfp456, Zfp429, Zfp459, Zfp874, 9630025I21Rik, Zfp748, A530054K11Rik and Zfp85-rs1. These genes are all KRAB domain Zfps and have a non-synonymous SNP in their coding region in the A/J sequence of the clf2 candidate region. KRAB domain Zfps have !  >D!  recently been shown to silence IAP elements and therefore could have a role in silencing the IAP at Wnt9b.  !  <E!  8 Gene expression studies in Wnt9bNull/Null E10 embryos 8.1 Background Wnt9b null mutants (Wnt9bNull/Null) have vestigial kidneys, lack reproductive ducts, and have an incomplete penetrance of CLP (Carroll et al. 2005). While the urogenital system of the Wnt9b null mutant has been extensively studied (Carroll et al. 2005), the signaling pathways and downstream targets affected by the loss of Wnt9b in the developing face have not yet been studied. With the goal of exploring a mechanism behind the CL phenotype in the Wnt9b mutants, we have studied the mRNA expression levels of 10 genes in the heads of Wnt9b null mutant embryos. The genes studied were chosen if they were thought to interact with the Wnt signaling pathway and/or have involvement with the etiology of CL/P, Table 8.1. Table 8.1 Proposed genes to study in Wnt9b null mutants and their reasoning. Mouse Wnt Human Gene References Modela Signalingb CL/Pc Wnt9b + + + (Carroll et al. 2005, Marazita et al. (linkage) 2004) !-catenin d + (Komiya and Habas 2008) Sox11 + + + (Sock et al. 2004, Marazita et al. 2004, (linkage) Sinner et al. 2007, Kormish et al. 2010) Dkk1 + (Glinka et al. 1998, Mao et al. 2001, Mukhopadhyay et al. 2001, Zorn 2001) Bmp4 + + + (Jiang et al. 2006, Liu et al. 2005, Lan et al. 2006) Msx1 + + (Juriloff and Harris 2008, Song et al. 2009) Msx2 + + (Juriloff and Harris 2008, Song et al. 2009) Raldh3 + (Song et al. 2009, Dupe et al. 2003) Wnt3 d + + (Lan et al. 2006, Niemann et al. 2004, Chiquet et al. 2008) Wnt4 0 + (Karner et al. 2009) a A plus sign (+) indicates a mouse model with CL/P; a negative sign (-) indicates mouse model known not to have CL/P; a “d” indicates that the null mutant dies prior to lip formation; a “0” indicates no information. b + indicates a published interaction with the Wnt signaling pathway c + indicates known human mutations, association studies or linkage.  !  <F!  The facial prominences fuse to form the upper lip at E10-11 in the mouse (Jiang et al. 2006). The MNPs fuse with the LNPs and the MXPs around the nasal pits (Jiang et al. 2006). Before the fusion process the nasal pits are oblong shaped, during the fusion process the nasal pits are crescent shaped, and after the fusion process they are comma shaped (Trasler 1968), figure 8.1. To capture the mRNA levels at the time of facial prominence fusion we collected the embryos around hour 22-23 on E10 when the lip is just about to fuse, and chose embryos in the oblong and crescent stages.  Figure 8.1 Stages of facial prominence fusion by the shape of the nasal pit We began the study by comparing the expression of Wnt9b in Wnt9b+/+ versus Wnt9bNull/Null embryos. This not only tested the method but also gave us an idea of the kind of variation we could expect to see between embryos. The Wnt canonical pathway accumulates !-catenin, which translocates to the nucleus to mediate transcription of target genes (Komiya and Habas 2008). The canonical pathway (!catenin) is active in the distal regions of the MNP, LNP, and MXP prior to lip fusion (Lan et al. 2006), overlaps with expression of Wnt9b (Lan et al. 2006), and is absent from the facial prominences of the Lrp6 null mutant with cleft lip (Song et al. 2009). Wnt9b is known to signal through the Wnt/!-catenin pathway in urogenital development (Carroll et al. 2005) but later in development switches to the non-canonical pathway (Karner et al. 2009). The quantitative Reverse Transcriptase PCR (qRT-PCR) data from this study was used to test whether Wnt9b is ! <G!  signaling through the canonical pathway in the developing facial prominences. If so, we expect !-catenin expression to be reduced in Wnt9b null mutants. Sox11 was an intriguing candidate for our study for two different reasons. Sox11 null mutants have a 70% penetrance of CL/P among many other defects causing them to die before birth (Sock et al. 2004). Sox transcription factors have also been shown to interact with Wnt signaling in three different ways: repressing Wnt transcriptional responses, enhancing Wnt target gene expression, and by being regulated by Wnt signaling (Sinner et al. 2007, Kormish et al. 2010). Dkk-1, another candidate for our study, is an inhibitor of Wnt signaling. It disrupts Lrp6 from binding to its co-receptor, Frizzled (Glinka et al. 1998, Mao et al. 2001, Zorn 2001). Activation of Wnt canonical signaling induces Dkk-1 transcription leading to the inhibition of the Wnt canonical pathway resulting in a feedback loop (Gonzalez-Sancho et al. 2005), we hypothesize that Dkk-1 expression will be reduced in Wnt9bNull/Null embryos as Wnt signaling is reduced. The next candidate, Bmp4, is thought to be downstream of Wnt/!-catenin signaling in the apical ectodermal ridge of the developing limb (Barrow et al. 2003, Soshnikova et al. 2003) and in the developing lung (Shu et al. 2005). Interestingly, Bmp4 signaling is not thought to be downstream of Wnt/!-catenin signaling in the developing face according to the Lrp6 null mutant (Song et al. 2009). However, the conditional inactivation of Bmp4 in the epithelium and mesenchyme of the MXP and partial inactivation in the epithelium of the LNP and MNP results in 20% penetrance of unilateral cleft lip (Liu et al. 2005). Bmp4 is also known to interact with Wnt signaling in other developing tissues. Msx1 and Msx2 were chosen for study because expression of Msx1 and Msx2 is dramatically reduced in Lrp6 null mutants in the MNP, LNP and the MXP (Song et al. 2009), !  <H!  indicating that Msx1 and Msx2 are direct downstream targets of Wnt signaling. Song et al. (2009) have also found binding sites for Lef and Tcf in Msx1 and Msx2 promoters; Lef and Tcf are transcription factors and the direct downstream targets of !-catenin, they are responsible for regulating the transcription of Wnt canonical target genes (Komiya and Habas 2008). In addition to the connection with the Wnt signaling pathway, the Msx1/Msx2 compound mouse mutant has severe craniofacial defects (Ishii et al. 2005), and MSX1 and MSX2 variants have been implicated in the etiology of human non-syndromic cleft lip (Vieira et al. 2005). We expected to see similar results in our Wnt9b null mutant as was observed in the Lrp6 null mutant, as Lrp6 is the receptor required for Wnt9b signaling in the Wnt canonical pathway (Komiya and Habas 2008). We chose to study Raldh3 for different reasons. Raldh3 has a role in the metabolic pathway of retinoic acid (RA) and RA synthesis is suppressed in the null mutant (Dupe et al. 2003). The Raldh3 null mutant has choanal atresia, the persistence of the nasal fins, which can be thought of as an over-fusion of the lip (Dupe et al. 2003). Raldh3 is also over-expressed in the Lrp6 null mutant (Song et al. 2009) and a cross talk between Wnt/!-catenin signaling and RA signaling has been observed in ocular development (Kumar and Duester 2010). We expected to observe an over-expression of Raldh3 in the Wnt9b null mutants. On Chromosome 11, Wnt9b and Wnt3 are adjacent to each other and Wnt3 was a strong candidate gene for clf1 (Juriloff et al. 2005a). WNT3 has been found to be involved in human syndromic CL/P, causing Tetra Amelia Syndrome (Niemann et al. 2004). SNPs located in WNT3 have been reported to be associated with human non-syndromic CL/P (Chiquet et al. 2008). Due to WNT3’s putative role in human CL/P and the fact that Wnt ligands have been shown to regulate other Wnt ligands (Carroll et al. 2005) we thought Wnt3 would be a worthwhile gene to study. !  <I!  Wnt4 is a direct downstream target of Wnt9b in the developing kidney (Carroll et al. 2005), therefore may be a downstream target of Wnt9b in the developing facial prominences. Wnt4 is expressed at stage 20-25 (prior to lip fusion) in the chick embryo in the external and stomodeal/oral epithelium of the entire chick head (Geetha-Loganathan et al. 2009). Based on the expression domain of Wnt4 and its connection with Wnt9b, Wnt4 expression could be reduced in the developing facial prominences of Wnt9b null mutants and could be a potential new CL/P candidate gene. We used qRT-PCR to evaluate the mRNA levels of each gene listed above in Wnt9b+/+ and Wnt9bNull/Null individual embryos at day 10 of gestation in the head above the mandible. We hoped to find genes that their mRNA levels were altered by the loss of Wnt9b. Our goal was to explore the signaling pathways disrupted in the null mutants.  8.2 Materials and Methods 8.2.1 Collection of Wnt9b+/+ and Wnt9bNull/Null E10 embryos Null mutation carriers were identified by genotyping B.WN mice for SSLP markers at Wnt9b, specifically W9B and WNT 15 M/N. The tissue obtained from the identification of individual progeny by the ear notch method was used for DNA extraction. DNA extraction and PCR were completed by lab staff, particularly Ron Chan. One or two female Wnt9b+/Null mice were placed with singly caged Wnt9b+/Nullmales, a total of eight females and six males were used. Females were checked for vaginal plugs every morning by Dr. Harris, the morning of the vaginal plugs was considered E0.5. Pregnant females were euthanized by Carbon Dioxide inhalation according to UBC SOP# 009E4-CO2 on E10. The uterus was removed, pinned on black wax and submerged in 0.85% sterile saline. The uterus was slit and the contents removed under a dissection microscope. Conceptuses and membranes were dissected out under sterile saline and transferred into a sterile dish of RNAlater (Sigma, Oakville, ON). Embryos were dissected from membranes and face stage (oblong, !  <"!  crescent or comma) scored under RNAlater, figure 8.1. Heads of E10 embryos were cut just above the mandible with care to ensure that maxillary prominences were intact with the head portion of the embryo, figure 8.2. This was the finest level of trimming that could be achieved, as the E10 heads are extremely small. Heads were stored in RNAlater at -20° C. A small portion of the remaining embryo was saved for immediate genomic DNA preparation and the remainder was archived at -20°C. Post-implantation mortality (“moles”) were also recorded. Dr. Diana Juriloff completed all dissections of E10 embryos.  Figure 8.2 Drawing of dissection of E10 embryos. Embryo heads were cleaved just above the mandible but MXP, MNP and LNP were intact with the head portion. Genomic DNA was prepared from a small portion of embryonic tissue and embryos were genotyped at Wnt9b for SSLP markers, specifically W9B and WNT15 M/N, by PCR methodology as described in the General Materials and Methods. I completed all genotyping of E10 embryos. Wnt9b+/+ and Wnt9bNull/Null embryos of the oblong and crescent face stages were selected for RNA extraction.  !  <>!  8.2.2 RNA and DNA extraction from embryo heads Approximately 10mg of RNAlater stabilized tissue was obtained from each E10 embryo head. Single embryos were analyzed, none of the samples were pooled. The tissue was disrupted with RLT buffer (All Prep kit) and three homogenization steps. The first step involved crushing the tissue with a disposable mortar and pestle (Fisher, Whitby. ON), next the tissue was passed through a 22 gage needle 4-6 times, and finally through a Qiashredder spin column (Qiagen, Germantown, MD) two times.  RNA and DNA were extracted from the  homogenized tissue using the All Prep DNA/RNA kit (Qiagen, Germantown, MD). Extracted RNA was treated with Turbo-DNA Free (Applied Biosystems/Ambion, Austin, TX), a DNase digestion, to remove any remaining DNA. Liane Gagnier completed the RNA extraction at the Mager Lab.  8.2.3 cDNA synthesis cDNA synthesis from the RNA samples was carried out by First-Strand cDNA synthesis using random primers. Specifically, 1µg total RNA, 1µl 10mM dNTP mix (Invitrogen, Carlsbad, CA) were combined and sterile distilled water was added to 13µl, the reaction was incubated for 5 minutes at 65° C followed by a minimum of 1 minute ice incubation. The contents of the tube was collected by centrifugation and 4µl 5x First-Strand Buffer (Invitrogen, Carlsbad, CA), 1µl 0.1M Dithiothreitol (DTT) (Invitrogen, Carlsbad, CA), 1µl RNaseOUT (Invitrogen, Carlsbad, CA) and 1µl of Superscript III RT (200 units/µl ) (Invitrogen, Carlsbad, CA) were added, mixed by pipetting gently up and down, incubated at 25° C for 5 minutes, incubated at 50° C for 30-60 minutes, and the reaction was inactivated by heating at 70° C for 15 minutes. The cDNA synthesis was completed by Liane Gagnier, Mager Lab.  !  <<!  8.2.4 Real-time PCR (qPCR) Real-time PCR primers were designed by Liane Gagnier and made by Invitrogen (Carlsbad, CA). Primer sequences are in Appendix E. Real time PCR reactions were prepared in duplicate in a 96-well plate and consisted of 10µl Fast SYBR Green Master Mix 2X, 5µM forward and reverse primers (0.8µl each), 6.4µl RNase-free water, and 2µl cDNA template. An endogenous control reaction for glyceraldehyde-3-phosphate dehydrogenase (Gapdh) was also prepared in parallel. The 96-well plate was sealed and centrifuged briefly to spin down the contents and remove any air bubble. Reactions were run in a 7500 Fast Real-Time PCR System (Applied Biosystems, Austin, TX) for 40 cycles at a Tanneal of 60° C. The target mRNA levels were normalized to the mRNA levels of the endogenous control, Gapdh. An oblong stage Wnt9b+/+ embryo (id# 9-40) was set to “1” for each analysis and all mRNA levels are relative to that embryo. Real-Time PCR evaluated the relative mRNA levels of each individual embryo. The Real-Time PCR reactions were completed by Liane Gagnier, Mager Lab.  8.2.5 Statistical analysis The mean and standard deviation of the mRNA expression levels in the Wnt9b+/+ and Wnt9bNull/Null groups was calculated. A one-way analysis of variance (one-way ANOVA) was calculated to compare the means between the two groups and test if they are significantly different using SPSS 15.0 for Windows. Statistical significance was set at P < .05.  8.3 Results A total of 53 embryos at E10 hour 22-23 were collected and genotyped at Wnt9b (W9B WT3/WT5/N3 and WNT15 M/N). Nine Wnt9b+/+ embryos were of oblong (5) or crescent (4) face stage and analyzed. Seven Wnt9bNull/Null littermate embryos were of the correct face stage and analyzed; they were all oblong stage (there were no crescent stage Wnt9bNull/Null embryos; this reflects the defect in face development in this genotype).  !  <#!  The mean and standard deviation within the Wnt9b+/+ and Wnt9bNull/Null embryos was calculated for mRNA levels for each of the 10 genes and the groups were compared by one-way ANOVA, Table 8.2. Only two of the genes differed significantly (P < .05) between genotype, Wnt9b and !-catenin. None of the other genes had differences that are close to approaching statistical significance. Wnt9bNull/Null embryos had no Wnt9b transcript detectable, not surprising since this is the Wnt9b null mutant. A variance of zero for the Wnt9bNull/Null group violates the structure of a one-way ANOVA test, the two samples must have the same variance; therefore, a non-parametric statistical test has been used to calculate the P value for Wnt9b transcript level differences. We have based a 2x2 contingency table on a “score” (+/+ = 9, Null/Null = 0) or “no score” (+/+ = 0, Null/Null= 7). Table 8.2 Analyzed data from Real-Time PCR of Wnt9b+/+and Wnt9bNull/Null E10 embryos. Wnt9b+/+ Wnt9bNull/Null Wnt9b+/+ Wnt9bNull/Null F-value mean mean Gene P value SD SD df = 1,14 N=9 N=7 Wnt9b .91 .32 0 0 NA <0.0001a !-catenin .99 .11 .82 .12 8.4 0.012 Sox11 .82 .23 .84 .39 .021 .886 Dkk-1 1.6 1.02 1.08 .57 1.458 .247 Bmp4 1.496 .32 1.46 .45 .037 .851 Msx1 1.15 .2 1.25 .39 .447 .515 Msx2 .99 .22 .91 .29 .396 .539 Raldh3 1.09 .29 1.07 .37 .014 .906 Wnt3 .97 .36 .85 .12 .682 .423 Wnt4 .81 .52 1.0 .43 .320 .580 a P value is calculated from non-parametric statistical test, Fisher’s exact test N, sample size; SD, standard deviation; df, degrees of freedom; NA, not applicable  8.4 Discussion We analyzed the mRNA levels of individual embryos just prior to fusion or during lip fusion using qRT-PCR. Comparing the average mRNA levels between the Wnt9b+/+ and Wnt9bNull/Null embryos, significant differences in Wnt9b and !-catenin mRNA levels were found. Wnt9bNull/Null embryos actually had no detectable Wnt9b mRNA; this result was expected as we were analyzing the null mutant. The goal of analyzing Wnt9b mRNA levels was to test the !  <D!  method. The Wnt9b+/+ mRNA levels were quite variable with a mean of .91 and a standard deviation of .32. The mRNA levels of !-catenin were significantly lower in Wnt9bNull/Null embryos. Previous evidence has shown that !-catenin and Wnt9b have overlapping expression domains in the developing facial prominences (Lan et al. 2006) but it was not known for sure if Wnt9b is signaling through the canonical pathway. Based on these results, Wnt9b is signaling through the canonical pathway but is known not to be the only Wnt ligand expressed in the developing head on E10 (Wnt3 is also expressed (Lan et al. 2006)), and there are still some residual !-catenin mRNA levels. Other mouse models with Wnt canonical pathway genes knocked out are known to have CL/P, such as Lrp6 (Song et al. 2009) and Rspo2 (Yamada et al. 2009). This provides the basis to suggest that other genes in the Wnt canonical pathway are worthwhile candidates for CL/P. The other genes analyzed (Sox11, Dkk-1, Bmp4, Msx1, Msx2, Raldh3, Wnt3 and Wnt4) all did not have significant differences between the Wnt9b+/+ and Wnt9bNull/Null embryos in their mRNA levels. This result was surprising and unexpected for some of the genes; for others we didn’t know what to expect. We expected to see significant decreases in Msx1, Msx2 mRNA levels and significant increases in Raldh3 mRNA levels in Wnt9bNull/Null embryos based on previous results in Lrp6 null mutants (Song et al. 2009). We think that some genes in our list of candidates still have the possibility of being altered in expression in Wnt9bNull/Null embryos, but the experiment did not detect the differences. The high level of variation within the groups may be masking the differences in the mRNA levels. The variation could arise from a number of sources. The dissection technique was the finest level of detail we could achieve with the surgical approach available. We were able to cleave off the head above the mandible and below the MXP but the back of the head !  #E!  remained in the test sample. If we could isolate the facial prominences we could reduce the “background noise” from the surrounding tissues. This level of detail would be hard to achieve, as E10 embryos are very small to begin with but could be possible using a Laser Capture Microdissection technique where cells of the facial prominences could be isolated from tissue sections under a microscope. Although in situ hybridization is not a quantitative technique, we would be able to specifically look at the expression levels in the facial prominences and possibly detect expression differences between the two groups. We must also keep in mind that the level of variation may be representative of the actual mRNA levels in the developing facial prominences. The facial prominences are very dynamic and rapidly evolving during this stage in development. The variation may be able to be reduced by narrowing down the face stage even further, i.e. only looking at oblong, or counting tail somites. However, a large sample size would be very hard to obtain. The mRNA levels of Wnt9b and !-catenin were significantly reduced in Wnt9bNull/Null embryo heads at the time of lip fusion. These results demonstrate that Wnt9b is signaling through the Wnt canonical pathway during facial prominence development and is crucial for proper lip development. These results should be confirmed by other read outs such as crossing the Wnt9b+/Null mice with TOPGAL mice, reporter of activation of canonical Wnt signaling. These results could then be compared with wildtype expression of the reporter, which is expressed in the distal regions of the MXP, MNP and LNP prior to lip fusion (Lan et al. 2006). The other genes tested did not provide significant differences.  !  #F!  9 Discussion The major focus of my studies was the clf2 locus – to identify its function and candidate genes. Based on my studies we now know that clf2 modifies the methylation status of clf1 (Wnt9b-IAP) and probably has a role in its silencing. The clf2 candidate region has now been narrowed down to a 3.0 mb region on Chromosome 13, a region consisting mainly of Zfps with a KRAB domain. Zfps with a KRAB domain are thought to have a role in silencing IAP elements in ES cells and early embryos (Wolf and Goff 2009, Rowe et al. 2010, Matsui et al. 2010), and therefore our top candidate genes are Zfps with a KRAB domain. A minor topic of my studies was to address the signaling pathways affected by the loss of Wnt9b in the embryonic head during facial prominence fusion; we have found that !-catenin is significantly reduced in Wnt9bNull/Null embryos and this is the first evidence that Wnt9b is signaling through the Wnt canonical pathway during the lip formation process. Deciphering the type of function of clf2 was not as straightforward a task as originally thought. Cross 1 and Cross 2 both gave unexpected and unexplained results that interfered with detection of a modifying effect of clf2 on the Wnt9bNull/Null mutant. By approaching the question in an inverse way we were able to use Wnt9bclf1/Null Cross 2 embryos to ask whether clf2 modifies the epigenetic modification of clf1 (Wnt9b-IAP). We found statistical evidence that clf2 modifies the methylation status of the 5’LTR of the Wnt9b-IAP. Clf2AA embryos were consistently the least methylated showing that there is some deficiency in the methylation process due to the “A” allele. The activity of the Wnt9b-IAP has been previously correlated with the methylation status of the 5’LTR; the 5’LTR of the Wnt9b-IAP in cleft lip embryos are 0-5% methylated and the IAP element is presumably active while the 5’LTR of the Wnt9b-IAP in normal embryos are approximately 50% methylated and the IAP element is presumably silenced (Juriloff et al. 2007). These results demonstrate that clf2 has a role in the methylation of the Wnt9b-IAP. From other models of metastable epialleles, DNA methylation of the 5’LTR of an !  #G!  IAP silences the IAP and it no longer interferes with transcription of the target gene (Whitelaw and Martin 2001). It appears that the clf2A gene product does not have the ability to methylate the 5’LTR of the Wnt9b-IAP properly and therefore it remains active and somehow interferes with the transcription of Wnt9b. Interference by the IAP leads to hypomorphic levels of Wnt9b and causes CL/P. Other mouse models, Avy and AxinFu operate under a similar mechanism but the factor responsible for the silencing of their IAP has not yet been identified; we are the first to identify a factor responsible for modulating the silencing of an IAP in a mouse model. There is a dactylaplasia mouse model that is the result of MusD insertions and is very similar to the A/- strain. The dactylaplasia mouse model also has two loci, one locus on Chromosome 19 which has two spontaneous alleles caused by insertions of MusD elements (dac) and a second recessive allele at an unlinked locus on Chromosome 13 (mdac) (Johnson et al. 1995, Kano et al. 2007). Like clf2, mdac is required for the dactylaplasia phenotype and is known to modify the epigenetic status of the MusD element (dac) (Kano et al. 2007). Interestingly, mdac is present in BALB/cJ, A/J and 129/J but not present in C57BL/6J, CBA/J and C3H/J and this allele dominantly inactivates dac (Johnson et al. 1995). The mdac candidate region is a 9.4 mb region between D13Mit113 and D13Mit310 (Kano et al. 2007); this region overlaps with the clf2 candidate region by 250 kb (Cntnap3 and 4930458L03Rik). The dactylaplasia mouse model is a similar model to the A/- strain and the possibility exists that mdac and clf2 are the same gene. Not only do the candidate regions overlap but the C57BL/6J allele inactivates the phenotype as is true for clf2B. A simple cross between the dactylaplasia mouse and the A/- strain would be an interesting future study to identify if mdac and clf2 are the same gene. Knowing the function of clf2 we examined the function of the genes in the clf2 candidate region to prioritize the candidate list. There are 48 genes or predicted genes in the clf2 candidate region and the majority are Zfps with KRAB domains, which are thought to transcriptionally ! #H!  silence IAP elements with the specificity of which Zfp based on the primer binding site (Wolf and Goff 2009, Rowe et al. 2010, Matsui et al. 2010). Based on their known function, Zfps with a KRAB domain became our top candidate genes for clf2. Zfps with a KRAB domain mediate silencing by binding to the target DNA and the KRAB domain will recruit a co-repressor, KAP1, which subsequently recruits the silencing machinery (Schultz et al. 2002). From this we hypothesize that a variation in the clf2 gene product leads to a deficiency in binding to the target DNA sequence or in the recruitment of KAP-1, causing the clf2A allele product to be defective in silencing the Wnt9b-IAP element. Possible functional changes (non-synonymous SNPs) and ancestral haplotypes contributed to further prioritizing the clf2 candidate genes for future studies. Fourteen of the Zfp-KRAB domain clf2 candidate genes have non-synonymous SNPS (Zfp369, Zfp708, Rsl1, Zfp458, Zfp595, E130120F12Rik, Zfp456, Zfp429, Zfp459, Zfp874, 9630025I21Rik, Zfp748, A530054K11Rik and Zfp85-rs1). Haplotypes of closely related strains to A/WySn with or without CL/P were examined to look for regions that share an “A” haplotype in the related strains. Regions that share an “A” haplotype in closely related CL/P strains with clf1AA may be a great starting point once the presence of the clf2A allele is confirmed by a cross to A/WySn. All of the closely related strains that we examined share an “A” haplotype between the markers PTDS A2/B2 and D13M6791 A/B. All of the Zfp-KRAB domain clf2 candidate genes with nonsynonymous SNPs are located in this region, with the exception of Zfp369, which is located proximal to PTDS A2/B2. Despite the shared SSLP haplotype we must not forget that A/J and C3H/HeJ differ at 164 SNPs in this region; C3H/HeJ does not have an identical “A” haplotype and therefore a cross to A/WySn would be required before moving forward with this data as a means of reducing the candidate region. An alternative hypothesis is that C57BL/6J provides more clf2 than the normal amount, possibly by gene duplication, and subsequently the clf2B “allele” provides a protective effect. ! #I!  The C57BL/6J genome in the clf2 candidate region contains clusters of segmental duplications (UCSC genome browser, Mouse, February 2006) and this region is prone to major differences in duplications and gene conversion events between inbred strains (Krebs et al. 2005) leading to this hypothesis. An example of a candidate gene by this hypothesis is Ak145544, a hypothetical zinc finger protein with a KRAB domain that is non-coding according to UCSC, July 2007. Unlike A/WySn, which gives one band, the SSLP marker for this gene gives two bands for various mouse strains (CL/Fr, DBA/1J, CBA/HeJ, C3H/HeJ, L strain, 129/SvJ, SWR/J). Further investigation revealed that in the C57BL/6J genome the SSLP for which the markers are designed has been perfectly duplicated (UCSC genome browser, Mouse, July 2007). The strains with two bands may have a differing sequence between the two copies. Coincidentally, 129/SvJ has one copy that is the same size at the C57BL/6J duplication and one altered copy; the clf2C allele (probably from 129/SvJ) has an intermediate clefting frequency between the clf2A and clf2B alleles (Juriloff et al. 2006). CL/Fr, DBA/1J, L strain, CBA/HeJ and C3H/HeJ have two copies that differ in size from C57BL/6J and perhaps have been reduced in functionality. This could explain the high degree of IAP activity in the C3H/HeJ strain (Maksakova et al. 2006) and penetrance of CL/P in CL/Fr and the L strain (Juriloff and Harris 2008). The A/- strain, BALB/c and SELH probably have only one copy and the product is 5 bp less than the C57BL/6J product. The “A” allele may not be able to effectively silence the IAP elements, which may explain the lack of DNA methylation of the 5’LTR of the Wnt9b-IAP (Juriloff et al. 2007) and the high mutation frequency in SELH and BALB/cGa (Juriloff et al. 2005b). Ak145544 could be a likely candidate for a protective clf2 allele due to its potential as a Zfp with a KRAB domain and duplication in C57BL/6J. The REC-4 and W1-U recombinant chromosomes narrowed the clf2 candidate region to between CNT C/B and D13M6791 A/B, an approximately 3.0 mb region. This method of testcrossing recombinant chromosomes with breakpoints within the clf2 candidate region is the !  #"!  necessary method of reducing the candidate region because of the “hard” boundaries that it produces for a semi dominant gene. However, the collection of breakpoints in the REC-4 and W1-U region from Cross 1 and Cross 2 at a frequency of 0.2% recombinant chromosomes between the existing boundaries suggests that finding and test-crossing new recombinant chromosomes in the clf2 candidate region is impractical. Future work will need to narrow the candidate region with other methods. CNT C/D and CNT E/F are located in the Cntnap3 gene approximately 50 kb apart. Given this small distance, the probability of two recombinants (W1-I and W1-U) in the WBC stock and a third in Cross 1 (Chapter 3) all occurring in this region is expected to be extremely small suggesting the possibility of a recombination “hot spot” in this area. Three recombinants in Cross 1 and the REC-4 recombinant chromosome also provides evidence for a second recombination “hot spot” between ZFP71 A/B2 and D13M6791 A/B, a region of approximately 150 kb. The clf2 candidate region is filled with segmental duplications in C57BL/6J, which are not thought to be present in the A/- strain (UCSC genome browser, Mouse, February 2006). The duplications in C57BL/6J and lack of them in the A/- strain would hinder pairing of homologs in the region making a recombination event difficult and may explain the lack of recombinants in this region. The possible recombinant hotspots between the markers CNT C/D and CNT E/F and the markers ZFP71 A/B2 and D13M6791 A/B are consistent with this. Areas surrounding (hundreds of kb) a recombination “hot spot” are usually devoid of crossover events (Arnheim et al. 2007). My accomplishments of a reduced clf2 candidate region, a known function, and a prioritized list of candidate genes will expedite future studies of clf2. At this point mouse crosses no longer seem to be an efficient way to identify clf2; the probability of attaining a new recombinant chromosome within the candidate region is extremely low. To identify the clf2 ! #>!  gene one possible strategy would be to sequence the clf2 candidate region in C57BL/6J, A/Wysn and possibly 129/SvJ (“C” allele); this region is known to be composed of tandem duplications and gene conversion events that are specific to each strain (Krebs et al. 2005) and sequencing could identify possible genes that have been lost in the A/WySn strain that are present in C57BL/6J. A different approach to identify clf2 would be by microarray; a tiling array composed of short fragments covering the entire candidate region that partially overlap would provide an unbiased look at gene expression and possibly lead to the detection of clf2. In vitro systems could also be used to identify clf2, such as by systematically knocking down each gene by siRNA in a reporter system which would look at IAP activity; the clf2 gene when knocked down by siRNA would increase the IAP activity. Future studies assessing the mechanism behind the silencing of the IAP by clf2 would not only be important for this mouse model but also for other mouse models with IAP insertional mutations. Clf2 has the potential to be responsible for globally silencing I"1 IAPs with the same PBS throughout the genome. An interesting study would be to cross Avy with A/WySn to see if clf2 has an effect on the phenotype of Avy and a role in the methylation status of that IAP in addition to the Wnt9b-IAP. Future studies of the signaling pathways affected by the loss of Wnt9b, such as in situ hybridization for candidate genes on Wnt9bNull/Null E10 embryos or an unbiased whole genome screen by microarray, and studies of the maternal effect in my opinion are the next most productive steps towards the understanding of CL/P in the A/WySn mouse model.  !  #<!  References Arnheim N, Calabrese P, Tiemann-Boege I (2007) Mammalian meiotic recombination hot spots. Annu Rev Genet 41, 369-399 Barrow JR, Thomas KR, Boussadia-Zahui O, Moore R, Kemler R, Capecchi MR, McMahon AP (2003) Ectodermal Wnt3/beta-catenin signaling is required for the establishment and maintenance of the apical ectodermal ridge. Genes Dev 17, 394-409 Baust C, Gagnier L, Baillie GJ, Harris MJ, Juriloff DM, Mager DL (2003) Structure and expression of mobile ETnII retroelements and their coding-competent MusD relatives in the mouse. J Virol 77, 11448-11458 Birnbaum S, Ludwig KU, Reutter H, Herms S, Steffens M, Rubini M, Baluardo C, Ferrian M, Almeida de Assis N, Alblas MA, Barth S, Freudenberg J, Lauster C, Schmidt G, Scheer M, Braumann B, Berge SJ, Reich RH, Schiefke F, Hemprich A, Potzsch S, Steegers-Theunissen RP, Potzsch B, Moebus S, Horsthemke B, Kramer FJ, Wienker TF, Mossey PA, Propping P, Cichon S, Hoffmann P, Knapp M, Nothen MM, Mangold E (2009) Key susceptibility locus for nonsyndromic cleft lip with or without cleft palate on Chromosome 8q24. Nat Genet 41, 473-477 Carroll TJ, Park JS, Hayashi S, Majumdar A, McMahon AP (2005) Wnt9b plays a central role in the regulation of mesenchymal to epithelial transitions underlying organogenesis of the mammalian urogenital system. Dev Cell 9, 283-292 Chakravarti A (2004) Finding needles in haystacks – IRF6 gene variants in isolated cleft lip or cleft palate. N Engl J Med 351, 822-824  !  ##!  Chenevix-Trench G, Jones K, Green AC, Duffy DL, Martin NG (1992) Cleft lip with or without cleft palate: associations with transforming growth factor alpha and retinoic acid receptor loci. Am J Hum Genet 51, 1377-1385 Chiquet BT, Blanton SH, Burt A, Ma D, Stal S, Mulliken JB, Hecht JT (2008) Variation in WNT genes is associated with non-syndromic cleft lip with or without cleft palate. Hum Mol Genet 17, 2212-2218 Davidson JG, Fraser FC, Schlager G (1969) A maternal effect on the frequency of spontaneous cleft lip in the A-J mouse. Teratology 2, 371-376 Diewert VM, Wang KY (1992) Recent advances in primary palate and midface morphogenesis research. Crit Rev Oral Biol Med 4, 111-130 Dupe V, Matt N, Garnier JM, Chambon P, Mark M, Ghyselinck NB (2003) A newborn lethal defect due to inactivation of retinaldehyde dehydrogenase type 3 is prevented by maternal retinoic acid treatment. Proc Natl Acad Sci U S A 100, 14036-14041 Fraser FC (1970) The genetics of cleft lip and cleft palate. Am J Hum Genet 22, 336-352 Geetha-Loganathan P, Nimmagadda S, Antoni L, Fu K, Whiting CJ, Francis-West P, Richman JM (2009) Expression of WNT signalling pathway genes during chicken craniofacial development. Dev Dyn 238, 1150-1165 Glinka A, Wu W, Delius H, Monaghan AP, Blumenstock C, Niehrs C (1998) Dickkopf-1 is a member of a new family of secreted proteins and functions in head induction. Nature 391, 357362 Gonzalez-Sancho JM, Aguilera O, Garcia JM, Pendas-Franco N, Pena C, Cal S, Garcia de Herreros A, Bonilla F, Munoz A (2005) The Wnt antagonist DICKKOPF-1 gene is a !  #D!  downstream target of beta-catenin/TCF and is downregulated in human colon cancer. Oncogene 24, 1098-1103 Harris MJ, Juriloff DM (2007) Mouse mutants with neural tube closure defects and their role in understanding human neural tube defects. Birth Defects Res A Clin Mol Teratol 79, 187-210 Ishii M, Han J, Yen HY, Sucov HM, Chai Y, Maxson RE,Jr (2005) Combined deficiencies of Msx1 and Msx2 cause impaired patterning and survival of the cranial neural crest. Development 132, 4937-4950 Jiang R, Bush JO, Lidral AC (2006) Development of the upper lip: morphogenetic and molecular mechanisms. Dev Dyn 235, 1152-1166 Johnson KR, Lane PW, Ward-Bailey P, Davisson MT (1995) Mapping the mouse dactylaplasia mutation, Dac, and the gene that controls its expression, Mdac. Genomics 29, 457-464 Juriloff DM, Harris MJ (2008) Mouse genetic models of cleft lip with or without cleft palate. Birth Defects Res A Clin Mol Teratol 82, 63-77 Juriloff DM, Harris MJ, Gagnier L, Mah DG (2008) A role for DNA methylation in the causation of cleft lip and palate in the A/WySn mouse model. Teratology 82, 300 Juriloff DM, Harris MJ, Gagnier L, Mager DL (2007) The relationship between cleft lip and palate and methylation of an IAP transposon insertion at Wnt9b in the A/WySn mouse model (abstract 691). Presented at the annual meeting of The American Society of Human Genetics, October 25, 2007, San Diego, California. Available at: http://www.ashg.org/genetics/ashg07s/index.shtml  !  DE!  Juriloff DM, Harris MJ, McMahon AP, Carroll TJ, Lidral AC (2006) Wnt9b is the mutated gene involved in multifactorial nonsyndromic cleft lip with or without cleft palate in A/WySn mice, as confirmed by a genetic complementation test. Birth Defects Res A Clin Mol Teratol 76, 574-579 Juriloff DM, Harris MJ, Dewell SL, Brown CJ, Mager DL, Gagnier L, Mah DG (2005a) Investigations of the genomic region that contains the clf1 mutation, a causal gene in multifactorial cleft lip and palate in mice. Birth Defects Res A Clin Mol Teratol 73, 103-113 Juriloff DM, Harris MJ, Mah DG (2005b) The open-eyelid mutation, lidgap-Gates, is an eightexon deletion in the mouse Map3k1 gene. Genomics 85, 139-142 Juriloff DM, Harris MJ, Dewell SL (2004) A digenic cause of cleft lip in A-strain mice and definition of candidate genes for the two loci. Birth Defects Res A Clin Mol Teratol 70, 509-518 Juriloff DM, Gunn TM, Harris MJ, Mah MK, Dewell SL (2001a) Multifactorial genetics of exencephaly in SELH/Bc Mice. Teratology 64, 189-200 Juriloff DM, Harris MJ, Brown CJ (2001b) Unravelling the complex genetics of cleft lip in the mouse model. Mamm Genome 12, 426-435 Juriloff DM, Mah DG (1995) The major locus for multifactorial nonsyndromic cleft lip maps to mouse Chromosome 11. Mamm Genome 6, 63-69 Juriloff DM (1982) Differences in frequency of cleft lip among the A strains of mice. Teratology 25, 361-368 Juriloff DM, Fraser FC (1980) Genetic maternal effects on cleft lip frequency in A/J and CL/Fr mice. Teratology 21, 167-175  !  DF!  Kafri T, Ariel M, Brandeis M, Shemer R, Urven L, McCarrey J, Cedar H, Razin A (1992) Developmental pattern of gene-specific DNA methylation in the mouse embryo and germ line. Genes Dev 6, 705-714 Kano H, Kurahaski H, Toda T (2007) Genetically regulated epigenetic transcriptional activation of retrotransposon insertion confers mouse dactulaplasia phenotype. Proc Natl Acad Sci U S A 104, 19034-19039 Karner CM, Chirumamilla R, Aoki S, Igarashi P, Wallingford JB, Carroll TJ (2009) Wnt9b signaling regulates planar cell polarity and kidney tubule morphogenesis. Nat Genet 41, 793-799 Komiya Y, Habas R (2008) Wnt signal transduction pathways. Organogenesis 4, 68-75 Kormish JD, Sinner D, Zorn AM (2010) Interactions between SOX factors and Wnt/beta-catenin signaling in development and disease. Dev Dyn 239, 56-68 Krebs CJ, Larkins LK, Khan SM, Robins DM (2005) Expansion and diversification of KRAB zinc-finger genes within a cluster including Regulator of sex-limitation 1 and 2. Genomics 85, 752-761 Kuff EL, Lueders KK (1988) The intracisternal A-particle gene family: structure and functional aspects. Adv Cancer Res 51, 183-276 Kumar S, Duester G (2010) Retinoic acid signaling in perioptic mesenchyme represses Wnt signaling via induction of Pitx2 and Dkk2. Dev Biol 340, 67-74 Lan Y, Ryan RC, Zhang Z, Bullard SA, Bush JO, Maltby KM, Lidral AC, Jiang R (2006) Expression of Wnt9b and activation of canonical Wnt signaling during midfacial morphogenesis in mice. Dev Dyn 235, 1448-1454  !  DG!  Li WY, Dudas M, Kaartinen V (2008) Signaling through Tgf-beta type I receptor Alk5 is required for upper lip fusion. Mech Dev 125, 874-882 Liu W, Sun X, Braut A, Mishina Y, Behringer RR, Mina M, Martin JF (2005) Distinct functions for Bmp signaling in lip and palate fusion in mice. Development 132, 1453-1461 Maestri NE, Beaty TH, Hetmanski J, Smith EA, McIntosh I, Wyszynski DF, Liang KY, Duffy DL, VanderKolk C (1997) Application of transmission disequilibrium tests to nonsyndromic oral clefts: including candidate genes and environmental exposures in the models. Am J Med Genet 73, 337-344 Maksakova IA, Mager DL, Reiss D (2008) Keeping active endogenous retroviral-like elements in check: the epigenetic perspective. Cell Mol Life Sci 65, 3329-3347 Maksakova IA, Romanish MT, Gagnier L, Dunn CA, van de Lagemaat LN, Mager DL (2006) Retroviral elements and their hosts: insertional mutagenesis in the mouse germ line. PLoS Genet 2, e2 Mao J, Wang J, Liu B, Pan W, Farr GH,3rd, Flynn C, Yuan H, Takada S, Kimelman D, Li L, Wu D (2001) Low-density lipoprotein receptor-related protein-5 binds to Axin and regulates the canonical Wnt signaling pathway. Mol Cell 7, 801-809 Marazita ML, Murray JC, Lidral AC, Arcos-Burgos M, Cooper ME, Goldstein T, Maher BS, Daack-Hirsch S, Schultz R, Mansilla MA, Field LL, Liu YE, Prescott N, Malcolm S, Winter R, Ray A, Moreno L, Valencia C, Neiswanger K, Wyszynski DF, Bailey-Wilson JE, AlbachaHejazi H, Beaty TH, McIntosh I, Hetmanski JB, Tuncbilek G, Edwards M, Harkin L, Scott R, Roddick LG (2004) Meta-analysis of 13 genome scans reveals multiple cleft lip/palate genes with novel loci on 9q21 and 2q32-35. Am J Hum Genet 75, 161-173 !  DH!  Matsui T, Leung D, Miyashita H, Maksakova IA, Miyachi H, Kimura H, Tachibana M, Lorincz MC, Shinkai Y (2010) Proviral silencing in embryonic stem cells requires the histone methyltransferase ESET. Nature 464, 927-931 Monk M, Boubelik M, Lehnert S (1987) Temporal and regional changes in DNA methylation in the embryonic, extraembryonic and germ cell lineages during mouse embryo development. Development 99, 371-382 Moreno LM, Arcos-Burgos M, Marazita ML, Krahn K, Maher BS, Cooper ME, ValenciaRamirez CR, Lidral AC (2004) Genetic analysis of candidate loci in non-syndromic cleft lip families from Antioquia-Colombia and Ohio. Am J Med Genet A 125A, 135-144 Morgan HD, Sutherland HG, Martin DI, Whitelaw E (1999) Epigenetic inheritance at the agouti locus in the mouse. Nat Genet 23, 314-318 Mossey PA, Little J, Munger RG, Dixon MJ, Shaw WC (2009) Cleft lip and palate. Lancet 374, 1773-1785 Mouse Genome Database [MGD] Mouse Genome Informatics Website, The Jackson Laboratory, Bar Harbor, Maine http://www.informatics.jax.org. Jan. 2009-May 2010 Mukhopadhyay M, Shtrom S, Rodriguez-Esteban C, Chen L, Tsukui T, Gomer L, Dorward DW, Glinka A, Grinberg A, Huang SP, Niehrs C, Izpisua Belmonte JC, Westphal H (2001) Dickkopf1 is required for embryonic head induction and limb morphogenesis in the mouse. Dev Cell 1, 423434 Nakatomi M, Wang XP, Key D, Lund JJ, Turbe-Doan A, Kist R, Aw A, Chen Y, Maas RL, Peters H (2010) Genetic interactions between Pax9 and Msx1 regulate lip development and several stages of tooth morphogenesis. Dev Biol 340, 438-449 !  DI!  Niemann S, Zhao C, Pascu F, Stahl U, Aulepp U, Niswander L, Weber JL, Muller U (2004) Homozygous WNT3 mutation causes tetra-amelia in a large consanguineous family. Am J Hum Genet 74, 558-563 Nottoli T, Hagopian-Donaldson S, Zhang J, Perkins A, Williams T (1998) AP-2-null cells disrupt morphogenesis of the eye, face, and limbs in chimeric mice. Proc Natl Acad Sci U S A 95, 13714-13719 Peanchitlertkajorn S, Cooper ME, Liu YE, Field LL, Marazita ML (2003) Chromosome 17: gene mapping studies of cleft lip with or without cleft palate in Chinese families. Cleft Palate Craniofac J 40, 71-79 Rakyan VK, Blewitt ME, Druker R, Preis JI, Whitelaw E (2002) Metastable epialleles in mammals. Trends Genet 18, 348-351 Reiss D, Mager DL (2007) Stochastic epigenetic silencing of retrotransposons: does stability come with age? Gene 390, 130-135 Rowe HM, Jakobsson J, Mesnard D, Rougemont J, Reynard S, Aktas T, Maillard PV, LayardLiesching H, Verp S, Marquis J, Spitz F, Constam DB, Trono D (2010) KAP1 controls endogenous retroviruses in embryonic stem cells. Nature 463, 237-240 Saal HM (2002) Classification and description of nonsyndromic clefts. In: Wyszynski DF (ed) Cleft Lip and Palate, New York, pp 47 Sasaki H, Matsui Y (2008) Epigenetic events in mammalian germ-cell development: reprogramming and beyond. Nat Rev Genet 9, 129-140 Schliekelman P, Slatkin M (2002) Multiplex relative risk and estimation of the number of loci underlying an inherited disease. Am J Hum Genet 71, 1369-1385 !  D"!  Schultz DC, Ayyanathan K, Negorev D, Maul GG, Rauscher FJ,3rd (2002) SETDB1: a novel KAP-1-associated histone H3, lysine 9-specific methyltransferase that contributes to HP1mediated silencing of euchromatic genes by KRAB zinc-finger proteins. Genes Dev 16, 919-932 Shaw D, Ray A, Marazita M, Field L (1993) Further evidence of a relationship between the retinoic acid receptor alpha locus and nonsyndromic cleft lip with or without cleft palate (CL +/P). Am J Hum Genet 53, 1156-1157 Shu W, Guttentag S, Wang Z, Andl T, Ballard P, Lu MM, Piccolo S, Birchmeier W, Whitsett JA, Millar SE, Morrisey EE (2005) Wnt/beta-catenin signaling acts upstream of N-myc, BMP4, and FGF signaling to regulate proximal-distal patterning in the lung. Dev Biol 283, 226-239 Sinner D, Kordich JJ, Spence JR, Opoka R, Rankin S, Lin SC, Jonatan D, Zorn AM, Wells JM (2007) Sox17 and Sox4 differentially regulate beta-catenin/T-cell factor activity and proliferation of colon carcinoma cells. Mol Cell Biol 27, 7802-7815 Sock E, Rettig SD, Enderich J, Bosl MR, Tamm ER, Wegner M (2004) Gene targeting reveals a widespread role for the high-mobility-group transcription factor Sox11 in tissue remodeling. Mol Cell Biol 24, 6635-6644 Song L, Li Y, Wang K, Wang YZ, Molotkov A, Gao L, Zhao T, Yamagami T, Wang Y, Gan Q, Pleasure DE, Zhou CJ (2009) Lrp6-mediated canonical Wnt signaling is required for lip formation and fusion. Development 136, 3161-3171 Soshnikova N, Zechner D, Huelsken J, Mishina Y, Behringer RR, Taketo MM, Crenshaw EB,3rd, Birchmeier W (2003) Genetic interaction between Wnt/beta-catenin and BMP receptor signaling during formation of the AER and the dorsal-ventral axis in the limb. Genes Dev 17, 1963-1968 !  D>!  Sperber GH (2002) Formation of the primary palate. In: Wyszynski DF (ed) Cleft Lip and Palate, New York, pp 5 Spiegelstein O, Mitchell LE, Merriweather MY, Wicker NJ, Zhang Q, Lammer EJ, Finnell RH (2004) Embryonic development of folate binding protein-1 (Folbp1) knockout mice: Effects of the chemical form, dose, and timing of maternal folate supplementation. Dev Dyn 231, 221-231 Strong LC (1978) Inbred mice in science. In: Morse HC (ed) Origins of Inbred Mice, New York, pp 43 Tang LS, Finnell RH (2003) Neural and orofacial defects in Folp1 knockout mice [corrected]. Birth Defects Res A Clin Mol Teratol 67, 209-218 Thomason HA, Dixon MJ, Dixon J (2008) Facial clefting in Tp63 deficient mice results from altered Bmp4, Fgf8 and Shh signaling. Dev Biol 321, 273-282 Trasler DG (1968) Pathogenesis of cleft lip and its relation to embryonic face shape in A-J and C57BL mice. Teratology 1, 33-49 UCSC genome browser Mouse, July 2007 Freeze. http://genome.ucsc.edu UCSC genome browser Mouse, February 2006 Freeze. http://genome.ucsc.edu Urrutia R (2003) KRAB-containing zinc-finger repressor proteins. Genome Biol 4, 231 Vasicek TJ, Zeng L, Guan XJ, Zhang T, Costantini F, Tilghman SM (1997) Two dominant mutations in the mouse fused gene are the result of transposon insertions. Genetics 147, 777-786 Vieira AR (2008) Unraveling human cleft lip and palate research. J Dent Res 87, 119-125 Vieira AR, Avila JR, Daack-Hirsch S, Dragan E, Felix TM, Rahimov F, Harrington J, Schultz RR, Watanabe Y, Johnson M, Fang J, O'Brien SE, Orioli IM, Castilla EE, Fitzpatrick DR, Jiang ! D<!  R, Marazita ML, Murray JC (2005) Medical sequencing of candidate genes for nonsyndromic cleft lip and palate. PLoS Genet 1, e64 Whitelaw E, Martin DI (2001) Retrotransposons as epigenetic mediators of phenotypic variation in mammals. Nat Genet 27, 361-365 Wolf D, Goff SP (2009) Embryonic stem cells use ZFP809 to silence retroviral DNAs. Nature 458, 1201-1204 Wong FK, Hagg U (2004) An update on the aetiology of orofacial clefts. Hong Kong Med J 10, 331-336 Xiong Z, Laird PW (1997) COBRA: a sensitive and quantitative DNA methylation assay. Nucleic Acids Res 25, 2532-2534 Yamada W, Nagao K, Horikoshi K, Fujikura A, Ikeda E, Inagaki Y, Kakitani M, Tomizuka K, Miyazaki H, Suda T, Takubo K (2009) Craniofacial malformation in R-spondin2 knockout mice. Biochem Biophys Res Commun 381, 453-458 Zorn AM (2001) Wnt signalling: antagonistic Dickkopfs. Curr Biol 11, R592-5 J122K(0*!LMN!;**/(0!MON!M-K(0!:8N!P--2QRS30,K!8N!T(/*.12()*!:N!;-/0-1!PN!;K03,+(),()! UN!8141Q3!VN!M-2K3'-!?N!T-+,1.(!TN!V*,K310-!UN!W3(30-!%XN!Y03*A3!ZMN!M*0()*!=N!%02*,R:10C*,! MN!=3'0-A!%;N![3(A'!==N!=31!VON!X-7!%N!\*A',+(3)!LSN!82K1A+4!XON!8K3!MN!?*K),*)!MUN!U*)'*!8N! 8K1++(!:;N!M-0-43+-!M=N!M100-7!?;!]GEEI^!Z)+(0B(0*)!0(C1A-+*07!B-2+*0!>!]ZX[>^!C()(! _-03-)+,!-)'!+K(!03,Q!*B!3,*A-+('!2A(B+!A3/!*0!/-A-+(@!!T!O)CA!?!M('!H"FN!<>DR<! ! ! ! !  D#!  !  !""#$%&'( !""#$%&'(!)((*+&,#+(-#./#$0#1(  !  ""!  Appendix B: Cross 2 Data  XX 3 5 0  Table B1 Cross 2 data Female A/WySn X Male B.WN clf2AA clf2AB XY Total XX XY Total 4 7 5 9 14 5 10 11 13 24 0 0 1 1 2  XX 0 3 3  clf2AA XY Total 0 0 4 7 5 8  XX 2 6 0  clf2BB XY Total 3 5 6 12 0 0  XX 0 0 7  clf2BB XY Total 0 0 0 0 5 12  Male A/WySn X Female B.WN (Reciprocal Cross) clf2AA clf2AB XX XY Total XX XY Total XX 1 4 5 4 6 10 3 2 4 6 11 12 23 8 0 1 1 1 0 1 1  clf2BB XY Total 2 5 3 11 0 1  Normals Wnt9bclf1/clf1 Wnt9bclf1/Null Wnt9bNull/Null Clefts clf1/clf1  Wnt9b Wnt9bclf1/Null Wnt9bNull/Null Normals Wnt9bclf1/clf1 Wnt9bclf1/Null Wnt9bNull/Null  XX 0 0 9  clf2AB XY Total 0 0 1 1 4 13  clf2AA clf2AB clf2BB XX XY Total XX XY Total XX XY Total Wnt9bclf1/clf1 0 0 0 0 0 0 0 0 0 clf1/Null Wnt9b 1 2 3 0 0 0 0 0 0 Wnt9bNull/Null 2 2 4 3 9 12 1 5 6 Null/Null BB *Embryo omitted from table: Reciprocal Cross, C2-165, SCAR, Wnt9b , clf2 , XY Clefts  !  "##!  Appendix C: COBRA data and results $$!  $$!  $$!  Figure C1 MboI gel pictures. The 524 bp band is the unmethylated, uncut band. The 354 bp and 168 bp bands are the methylated, cut bands.  !  "#"!  $$!  $$!  $$!  Figure C2 EcoRI gel pictures. The 524 bp band is the unmethylated, uncut band. The 392 bp and 132 bp bands are the methylated, cut bands.  !  "#%!  Genotype AA  Table C1 COBRA results MboI EcoRI  Phenotype  Id #  Sex  BCL BCL BCL BCL BCL BCL RCL BCL N N N N N N N N  C2-03 C2-16 C2-19 C2-25 C2-47 C2-49 C2-77 C2-111 C2-13 C2-26 C2-30 C2-32 C2-38 C2-64 C2-85 C2-94  0 0 0 0 10 10 10 10 0 50 30 30 40 20 40 40  0 0 0 0 10 10 10 10 0 50 30 20 30 10 30 40  XY XY XX XX XY XX XY XY XX XY XX XY XX XX XY XY  RCL N N N N N N N N N N N N N N N  C2-55 C2-02 C2-09 C2-22 C2-51 C2-80 C2-89 C2-93 C2-110 C2-117 C2-123 C2-129 C2-137 C2-143 C2-149 C2-163  30 30 20 50 40 50 30 20 40 40 50 20 40 40 10 50  10 30 10 50 30 30 20 20 20 20 30 10 30 30 10 30  XY XX XY XY XY XY XX XY XY XX XX XY XX XY XY XX  N N N N N N N N N N N  C2-14 C2-15 C2-41 C2-42 C2-48 C2-50 C2-61 C2-75 C2-79 C2-88 C2-95  50 60 40 30 30 40 50 40 50 40 40  50 40 20 40 30 30 30 30 30 30 30  XX XX XY XY XY XX XY XX XY XY XX  Cross  RECIP  Age 11.5 14 14 16 14 14 13 14 13 16 16 14 14 13 14 18  AB  RECIP RECIP RECIP RECIP RECIP RECIP RECIP  13 11.5 13 14 14 13 14 18 14 14 14 14 14 15 14 13  BB  !  13 13 14 14 14 14 13 13 13 14 18 "#&!  Genotype BB  !  Phenotype  Embryo ID  MboI  EcoRI  Sex  Cross  Age  N N N N N  C2-98 C2-104 C2-114 C2-173 C2-178  30 30 50 50 50  10 30 30 30 30  XX XX XX XX XY  RECIP RECIP RECIP RECIP  12 12 14 13 14  "#'!  Appendix D: Candidate Genes and their Function Table D1 Candidate genes and their function from UCSC and MGI database search Gene Function Cntnap3 Receptor (CNS), cell recognition molecule 4930458L03Rik  Unknown  4932441B19Rik  Unknown  Hiatl1  Membrane Protein (major facilitator superfamily)  Olfr466  Receptor (Olfactory, initiates neuronal response)  Nlrp4f  Mitochondrial Protein (may be involved in the defense response)  Zfp369  Transcription Factor, zinc finger protein with KRAB domain  Ak145544 (Gm10324)  Non-coding, hypothetical zinc finger protein with KRAB domain  Ak021218 Non-coding, weakly similar to KRUPPLE-RELATED ZINC FINGER (C330022B21Rik) PROTEIN F80-L Ak049221 (Vmn2r-ps105)  Non-coding, weakly similar to KRUPPLE-RELATED ZINC FINGER PROTEIN F80-L  2410141K09Rik  Near coding, weakly similar to KRUPPLE-RELATED ZINC FINGER PROTEIN F80-L  Ak076976  Non-coding  NR_003702  Non-coding  6D BC049692 Uqcrb Ak163421 (Gm10767) Mterfd1 Ptdss1 Ak019536 (Zfp712) Zfp708 BC038328 Zfp759 Rsl1 Zfp455 !  Transcription Factor, zinc finger protein with KRAB domain Non-coding Mitochondrial Protein, ubiquinol-cytochrome C reductase binding protein Unknown, Antisense, hypothetical protein Mitochondrial Protein Membrane Protein, protein involved in phosphatidylserine biosynthesis Transcription Factor, zinc finger protein Transcription Factor, zinc finger protein with KRAB domain Non-coding Transcription Factor, zinc finger protein with KRAB domain Transcription Factor, zinc finger protein with KRAB domain, controls sexually dimorphic liver target genes Transcription Factor, zinc finger protein with KRAB domain "#(!  Gene Zfp458  Transcription Factor, zinc finger protein with KRAB domain  Zfp457  Transcription Factor, zinc finger protein with KRAB domain  Zfp595  Transcription Factor, zinc finger protein with KRAB domain  E130120F12Rik Zfp456 2810487A22Rik (Zfp429)  Hypothetical protein, contains zinc fingers and a KRAB domain Transcription Factor, zinc finger protein with KRAB domain Transcription Factor, Transcription Factor, zinc finger protein with KRAB domain, controls sexually dimorphic liver target genes  Zfp459  Transcription Factor, zinc finger protein with KRAB domain  C330011K17Rik (Zfp874)  Transcription Factor, zinc finger protein with KRAB domain  9630025I21Rik  Transcription Factor, zinc finger protein with KRAB domain  Zfp817 (zfp58)  Transcription Factor, zinc finger protein with KRAB domain  Zfp87  Transcription Factor, zinc finger protein with KRAB domain  Zfp748  Transcription Factor, zinc finger protein with KRAB domain  AK021377 (9430065F17Rik)  Hypothetical protein, Non-coding  AK020476  Hypothetical protein, Non-coding  AA987161  Transcription Factor, zinc finger protein with KRAB domain  A530054K11Rik  Hypothetical protein, zinc finger protein with KRAB domain  3830402I07Rik (Zfp738)  Hypothetical protein, zinc finger protein with KRAB domain  Zfp71-rs1  Transcription Factor, zinc finger protein with KRAB domain  AK015755 (4930441O14Rik)  Non-coding  Zfp85-rs1  Transcription Factor, zinc finger protein with KRAB domain  AK140336 (Zfp493)  Hypothetical protein, zinc finger protein with KRAB domain  AK013687 (Zfp493)  Near coding, no protein, Hypothetical KRAB box containing protein  AK029746 (4930525G20Rik)  !  Function  Non-coding  "#)!  Appendix E: qRT-PCR Primer Sequences Gene !-catenin Msx1 Msx2 Wnt4 Raldh3 Sox11 Wnt9b Bmp4 Wnt3 Dkk-1 Gapdh  !  Table E1 qRT-PCR primer sequences Forward Primer Reverse Primer ATGGAGCCGGACAGAAAAGC  CTTGCCACTCAGGGAAGGA  GAAACTAGATCGGACCCCGT AATTCCGAAGACGGAGCAC CCTGCGACTCCTCGTCTTC TCAGCTGGCTGACCTTGTAG CTGTCGCTGGTGGATAAGGA GGCTGCTGGAATGTCAGTT GAGGGATCTTTACCGGCTCC ACAAAGCCACCCGTGAATC CTCATCAATTCCAACGCGATCA GACTTCAACAGCAACTCCCAC  GTTGGTCTTGTGCTTGCGTA CGGTTGGTCTTGTGTTTCCT GTTTCTCGCACGTCTCCTCT GACGAAAAAGGCATGAAGGA GTGAACACCAGGTCGGAGAA AGGAAGGCCGTCTCCTTAAA GTTGAAGAGGAAACGAAAAGCAG TGGCCCCTTATGATGTGAGTC GCCCTCATAGAGAACTCCCG TCCACCACCCTGTTGCTGTA  "#*!  Appendix F: Animal Care Certificates  !  "#+!  !  "#,!  

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.24.1-0071285/manifest

Comment

Related Items