Molecular Genetics of Corneal
Dystrophy
A THESIS SUBMITTED FOR THE M.D. TO THE UNIVERSITY OF
LONDON
MOHAMED EL-ASHRY, MB CHB FRCS (Ed)
CLINICAL RESEARCH FELLOW
DEPARTMENT OF MOLECULAR GENETICS
INSTITUTE OF OPHTHALMOLOGY
UNIVERSITY COLLEGE LONDON
BATH STREET
LONDON
AND
MOORFIELDS EYE HOSPITAL
CITY ROAD
LONDON
2001 ProQuest Number: 10013866
All rights reserved
INFORMATION TO ALL USERS The quality of this reproduction is dependent upon the quality of the copy submitted.
In the unlikely event that the author did not send a complete manuscript and there are missing pages, these will be noted. Also, if material had to be removed, a note will indicate the deletion. uest.
ProQuest 10013866
Published by ProQuest LLC(2016). Copyright of the Dissertation is held by the Author.
All rights reserved. This work is protected against unauthorized copying under Title 17, United States Code. Microform Edition © ProQuest LLC.
ProQuest LLC 789 East Eisenhower Parkway P.O. Box 1346 Ann Arbor, Ml 48106-1346 Abstract
Abstract
Comeal dystrophies are inherited disorders characterised by progressive accumulation of deposits in the cornea causing visual impairment. They occur in either an autosomal dominant or recessive form and are usually manifested in the first few decades of life. The present classification is solely based on the layer or layers of the cornea involved. This study aimed at better understanding of the underlying molecular basis of such disorders via linkage to specific chromosomal loci and then mutation screening of the disease genes by means of amplification of the genomic DNA using polymerase chain reaction and then sequencing and restriction enzyme digest analysis. A family with a combined cornea plana and microphthalmia phenotype was mapped to the autosomal recessive cornea plana (CNA2) locus on 12q21. A novel mutation in KERA gene (CNA2) has also been identified. The phenotype described is wider than that previously reported for KERA mutations and represents the first description of a stmctural protein causing microphthalmia. Mutation screening of a new carbohydrate sulfotransferase 6 gene {CHST6) in 5 families with autosomal recessive macular comeal dystrophy (MCD) type I has revealed 6 novel missense mutations, 4 homozygous and 2 compound heterozygous. This study provides further evidence that CHST6 is the MCD gene and the loss of gene function by these mutations may result in production of abnormal keratan sulfate, which would account for the MCD phenotype. Analysis of a human transforming growth factor-6 -induced gene (BIGH3) in 36 patients with lattice comeal dystrophy type I (LCDI), Avellino comeal dystrophy (ACD), granular comeal dystrophy (GCD), and Reis-Bucklers’ comeal dystrophy (RBCD) has revealed 4 heterozygous missense mutations. In conclusion, molecular genetics may provide a more accurate basis for the diagnosis and classification of comeal dystrophies. This work contributes to our understanding of the role of specific genes in comeal transparency and may help to identify novel therapeutic approaches for these dystrophies that cause significant visual impairment. Table o f contents
Table of Contents Page
Abstract 2
Table of contents 3
List of tables 12
List of figures 14
Declaration 18
Acknowledgements 19
List of abbreviations 20
Key to pedigree symbols 22
Chapter 1 23
Introduction 23
1.1 General introduction 23
1.2 Cornea 24
1.2.1 Gross anatomy 24
1.2.2 Microscopic anatomy 24
1.2.2.1 Epithelium 26
1.2.2.2 Bowman’s membrane 27
1.2.2.3 Stroma 27
1.2.2.4 Descemet’s membrane 28
1.2.2.5 Endothelium 28
1.2.3 Comeal transparency 29
1.2.4 Development of the cornea 29
1.2.5 Metabolism and physiology of the cornea 33
1.3 Corneal dystrophy: A phenotypic review 35 Table o f contents
1.3.1 Epithelial dystrophies 36
1.3.1.1 Meesmann's dystrophy 36
1.3.1.2 Epithelial basement membrane dystrophies 37
1.3.1.3 Band-shaped dystrophy 38
1.3.2 Bowman's layer dystrophies 3 8
1.3.2.1 Reis-Bücklers' comeal dystrophy (type I 38
Bowman’s layer, CDBI dystrophy, RBCD)
1.3.2.2 Honeycomb dystrophy of Thiel and Behnke 40
(type II Bowman’s layer, CDBII dystrophy)
1.3.2.3 Grayson-Wilbrandt dystrophy 41
1.3.3 Stromal dystrophies 43
1.3.3.1 Granular comeal dystrophy (GCD) 43
1.3.3.2 Lattice comeal dystrophy (LCD) 44
1.3.3.2.1 Lattice comeal dystrophy type I (LCD type I) 44
1.3.3.2.2 Lattice comeal dystrophy type II (LCD type II) 45
1.3.3.2.3 Lattice comeal dystrophy type HI (LCD type HI) 45
1.3.3.2.4 Lattice comeal dystrophy type IIIA (LCD type 46
IIIA)
1.3.3.2.5 Lattice comeal dystrophy type IV (LCD type IV) 46
1.3.3.3 Avellino comeal dystrophy (ACD) 48
1.3.3.4 Macular comeal dystrophy (MCD) 48
1.3.3.5 Fleck comeal dystrophy (Mouchetée or
speckled comeal dystrophy) 49
1.3.3.6 Central crystalline dystrophy of Schnyder's 50 Table o f contents
1.3.4 Endothelial dystrophy 52
1.3.4.1 Fuch's dystrophy 52
1.3.4.2 congenital hereditary endothelial dystrophy (CHED) 52
1.3.4.3 Posterior polymorphous dystrophy (PPD) 53
1.3.5 Ectatic dystrophies 56
1.3.5.1 Keratoconus 5 6
1.3.5.2 Keratoglobus 56
1.3.5.3 Macrocomea 57
1.4 Anomalous corneal development 57
1.4.1 Cornea plana 57
1.4.2 Microphthalmia 59
1.4.3 Sclerocomea 60
1.4.4 Genetics of cornea plana and microphthalmia 61
1.5 Genetics of comeal dystrophy 63
1.6 Mapping single gene disorders by linkage analysis 65
1.6.1 History of linkage analysis 65
1.6.2 Linkage analysis 65
1.6.3 Genetic and physical map distances 66
1.6.4 Genetic markers 67
1.6.4.1 Requirements for genetic mapping 67
1.6.4.2 Restriction fragment length polymorphisms (RFLPs) 68
1.6.4.3 Variable number of tandem repeats (VNTRs) or 68
minisatellites
1.6.4.4 Microsatellite markers 69
(short tandem repeat polymorphisms [STRPs]) Table o f contents
1.6.4.3 Single nucleotide polymorphisms 69
(biallelic markers) or SNPs
1.6.5 Linkage analysis for autosomal dominant diseases 70
1.6.6 Linkage analysis for autosomal recessive diseases 70
1.6.7 Two point mapping to locate a disease locus 71
1.6.8 Evidence of genetic heterogeneity 72
1.7 The human genome project 72
1.7.1 Microsatellite maps 73
1.7.2 Identification of genes 74
1.7.3 Mutation screening 75
1.7.3.1 Identifying disease-causing mutations 76
1.8 Aims of the study 76
Chapter 2 78
Materials and methods 78
2.1 Extraction of DNA 78
2.1.1 Extraction of DNA from peripheral blood lymphocytes 78
2.1.2 Extraction of DNA from buccal smears 79
2.2 DNA amplification by polymerase chain reaction (PCR) 80
2.2.1 The PCR reaction 80
2.2.2 Standard parameters for PCR 82
2.2.3. Primer design 82
2.3 Fractionation of DNA by gel electrophoresis 83
2.3.1 Agarose gel electrophoresis 83
2.3.2 Non-denaturing polyacrylamide gel electrophoresis 84
2.4 Purification of DNA 86 Table o f contents
2.4 Purification of DNA 86
2.4.1 Phenol chloroform extraction and ethanol precipitation 86
2.4.2 Use of Sephacryl microspin columns (Sephacryl-S400 HR 87
columns, Pharmacia, UK)
2.4.3 Use of Qiagen gel extraction/ PCR purification kits 88
2.5 Restriction enzyme digests of DNA 89
2.6 Mutation detection techniques 89
2.6.1 Heteroduplex analysis method 89
2.6.2 Single strand conformation polymorphism analysis (SSCP) 91
2.6.3 DNA sequencing 92
2.7 Assay of sulphated keratan sulfate (KS) in serum 94
2.7.1 Specimens 94
2.7.2 Antibodies 95
2.7.3 Inhibition—Enzyme-Linked Immunosorbent Assay 95
(ELISA) for keratan sulfate detection
2.8 Computer aided analysis 96
2.8.1 Computational analysis of DNA sequence 96
2.8.2 Linkage computer programs (Cyrillic and MLINK) 96
2.9 Buffers and solutions 97
2.10 Electronic database information 98
Chapter 3 99
A novel mutation in keratocan causes autosomal recessive cornea plana and 99 microphthalmia.
3.1 Introduction 99
3.1.1 Cornea plana 99 Table o f contents
3.1.2.1 Autosomal dominant inheritance 100
3.1.2.2 Autosomal recessive inheritance 101
3.1.2.3 Identification of the gene for cornea plana 101
3.1.3 Microphthalmia 102
3.1.3.1 Phenotype of microphthalmia 102
3.1.3.2 Genetics of microphthalmia 103
3.1.4 Small leucine rich proteoglycans (SLRPs) 103
3.1.4.1 Class I SLRPs 104
3.1.4.2 Class II SLRPs 105
3.1.4.3 Class III SLRPs 105
2iA.5 Kera 107
3.2 Aim of the study 109
3.3 Patients and methods 110
3.3.1 Family and clinical data 110
3.3.2 Methods 114
3.3.2.1 Genotyping 114
3.3.2.2 Linkage analysis 114
3.3.2.3 Mutation screening of KERA by sequencing 116
3.4 Results 117
3.4.1 Linkage of the family to the known locus for cornea plana 117
3.4.2 Exclusion of linkage to the known loci of microphthalmia 118
3.4.3 Mutation screening of KERA 124
3.4.4 Functional analysis of the T215K mutation by protein modelling 126
3.5 Discussion 129 Table o f contents
Chapter 4 133
Spectrum of BIGH3 mutations in comeal dystrophies 133
4.1 Introduction 133
4.1.1 Chromosome 5q linked dystrophies 133
4A2BIG H 3 gene 134
4.1.3 Kerato-epithelin 134
4.1.4 Comeal dystrophies caused by BIGH3 gene mutations 136
4.1.4.1 Granular comeal dystrophy (GCD) or 136
(Groenouw type I) (CDGGI) (OMIM NO 121900)
4.1.4.2 Reis- Bucklers' comeal dystrophy (RBCD) 137
(OMIM N0121900)
4.1.4.3 Avellino comeal dystrophy (ACD) 138
(OMIM NO 121900)
4.1.4.4 Lattice comeal dystrophies (LCD) 139
4.2 Aim of the study 140
4.3 Patients and methods 140
4.3.1 Patients 140
4.3.1.1 RBCD patients 141
4.3.1.2 GCD patients 145
4.3.1.3 ACD patients 150
4.3.1.4 LCDI patients 151
4.3.1.5 A Bangladeshi family with RBCD and LCDI phenotyp 156
4.3.2 Methods 158
4.3.2.1 DNA extraction 158
4.3.2.2 PCR amplification of BIGH3 gene 158 Table o f contents
4.3.2.3 Mutation detection 158
4.4 Results 160
4.4.1 Identification of R555Q mutation in RBCD Patients 160
4.4.2 Identification of R555W mutation in GCD patients 160
4.4.3 British families with ACD 163
4.4.3.1 Histopathological study 163
4.4.3.2 RI24H is the underlying genetic defect in ACD patients 163
4.4.4 R124C is the causative molecular defect in LCDI patients 166
4.4.5 Single point mutation in the BIGH3 gene responsible for
intrafamilial phenotypic heterogeneity in a Bangladeshi family 166
4.4.5.1 Mutation analysis 166
4.4.5.2 Cosegregation study and novel polymorphism appraisal 167
4.4.6 polymorphisms identified 171
4.5 Discussion 173
Chapter 5 181
Identification of novel mutations in the carbohydrate sulfotransferase 6 gene
(CHST6) causing MCD
5.1 Introduction 181
5.1.1 Macular comeal dystrophy (MCD) 181
5.1.2 Keratan sulfate (KS) 183
5.1.3 Immunophenotypic classification of MCD 184
5.1.4 Pathogenesis of MCD 185
5.1.5 Histopathology of MCD 186
5.1.6 Genetics of MCD 186
5.2 Aim of the study 188
10 Table o f contents
5.3 Patients and methods 189
5.3.1 Patients 190
5.3.2 Methods 194
5.3.2.1 DNA extraction 194
5.3.2.2 Mutation detection 194
5.3.2.3 Assay of sulfated KS in serum 195
5.4 Results 196
5.4.1 Identification of novel mutations in the coding region of CHST6X96
5.4.2 Restriction enzyme digest of the mutations identified 199
5.4.3 Haplotype analysis 203
5.4.4 Diagnosis of MCD type I by assay of sulfated KS in serum 203
5.4.5 Assessment of the significance of the mutations identified 205
5.5 Discussion 209
Chapter 6 212
General Discussion 212
6.1 Molecular genetics and corneal dystrophy 212
6.2 overview of the work presented 215
6.3 Therapeutics of corneal dystrophy 216
6.3.1 Advances in the surgical therapy for comeal dystrophy 216
6.3.2 Moelcular genetics and its therapeutic role 218
6.4 Future work 220
References 222
Publications and presentations arising from this work 250
11 List o f Tables
LIST OF TABLES
Page Table 1.1: Genetics of major types of comeal dystrophy 64
Table 1.2: Markers map of the human genome 74
Table 2.1: Agarose gel concentrations for separating
DNA fragments 83
Table 2.2: The acrylamide concentrations prepared to provide
maximum resolution of single stranded DNA fragment 85
Table 3.1: Ocular features of affected individuals in the family 111
Table 3.2: Markers on chromosomes 11, 12, 14, and 15 used for
the linkage study 115
Table 3.3: Primers used for amplification of KERA 116
Table 3.4: Two-Point Lod Scores between the
CNA2/microphthalmia locus and 6 markers on 12q21 123
Table 4.1: Clinical criteria of RBCD, GCD, and LCDI patients 155
Table 4.2: Primers designed for amplification of BIGH3 gene 159
Table 4.3: Restriction enzymes used for confirmation of
mutations and polymorphisms identified 162
Table 4.4: Polymorphisms identified in the BIGH3 gene 171
Table 4.5: BIGH3 mutations reported in different phenotypes of
comeal dystrophy 173
Table 4.6: Mutations identified in the BIGH3 gene 174
Table 5.1: Clinical features of the probands with MCD from all
families participated in the study 192
Table 5.2: Primers used for amplification and
12 List o f Tables
sequencing of the coding region of CHST6 195
Table 5.3: Mutations identified within the CHST6 gene in MCD
type I families 197
Table 5.4: Restriction enzymes used for confirmation of the mutations
identified 199
13 List o f Figures
LIST OF FIGURES Page Figure .1: Light micrograph of normal human cornea 25
Figure .2: Diagrammatic representation of the three successive
waves of ingrowth of neural crest cells 31
Figure .3: Diagrammatic representation of the
development of the cornea 32
Figure .4: Factors and forces involved in the control
of comeal stromal hydration 34
Figure .5: Meesmann's comeal dystrophy 42
Figure .6: Finger print dystrophy 42
Figure .7: CDBI 42
Figure .8: CDGGI 47
Figure .9: LCDI 47
Figure .10: LCDII 47
Figure .11: LCDIIIA 47
Figure .12: ACD 51
Figure .13: MCD 51
Figure .14: Fleck comeal dystrophy 51
Figure .15: Central crystalline dystrophy 51
Figure .16: Fuch’s endothelial dystrophy 55
Figure .17: Congenital hereditary endothelial dystrophy 55
Figure .18: Posterior polymorphous dystrophy 55
Figure .19: Keratoconus 62
Figure .20: Keratoglobus 62
Figure .21: Megalocomea 62
14 List o f Figures
Figure 1.22: Microphthalmos with cyst 62
Figure 1.23: Sclerocomea 62
Figure 3.1: A dendogram representing the three
classes of SLRP genes 105
Figure 3.2: Schematic diagram of the stmcture
of human keratocan gene 109
Figure 3.3a: Slit lamp photograph of an affected sibling (case IV-1 ) 112
Figure 3.3b: Slit lamp photograph of an affected sibling (case IV-3) 113
Figure 3.4: Pedigree of a four-generation consanguineous family
from Bangladesh 118
Figure 3.5: Schematic diagram of chromosome 12 119
Figure 3.6a: Cosegeregation of the disease alleles with markers
D12S95andPAH 120
Figure 3.6b: Cosegeregation of the disease alleles with markers
D12S351 andD12S322 121
Figure 3.7: Linkage analysis to chromosome 14 122
Figure 3.8: Electropherogram of exon 2 of the KERA gene 125
Figure 3.9: 3D modelling of ALS, wild and mutated Keratocan 127
Figure 3.10: A 3D modelling of wild type and T215K mutant
Keratocan 128
Figure 4.1: Schematic diagram of kerato-epithelin stmcture 135
Figure 4.2: Pedigrees of CDBll families (A1 -A3) 142
Figure 4.3: Pedigrees of CDBll families (A4 and A5) 143
Figure 4.4: Slit lamp photographs of CDBll in the proband of
family A1 caused by R555Q mutation 144
15 List o f Figures
Figure 4.5: Pedigrees of CDGGI families (B1 and B2) 147
Figure 4.6: Slit lamp photographs of CDGGI phenotype in three
generations of the same family caused by R555W 148
Figure 4.7: Slit lamp photographs of a sporadic patient (case 1) with
CDGGI 149
Figure 4.8: Pedigrees of ACD families (Cl and C2) 152
Figure 4.9: Pedigrees of the LCDI (family D) and the Bangladeshi
family (family E) 153
Figure 4.10: Slit lamp photograph of ACD in the right (a) and
left (b) corneas of the proband of family Cl 154
Figure 4.11: Slit lamp photograph (retroillumination) of LCDI
in the proband of family D 154
Figure 4.12: Slit lamp photograph of affected individuals
in the Bangladeshi family 157
Figure 4.13: Electrophereogram of the BIGH3 gene 161
Figure 4.14: Cosegregation study of members of family Blby
BstXl 162
Figure 4.15: Light microscopic study of comeal specimen
from the proband of family Cl 164
Figure 4.16: Cosegregation study wdth CSPl restriction enzyme 165
Figure 4.17: Ethidium bromide stained gel in BIGH3 exon 4. 168
Figure 4.18: Cosegregation study using Ava II restriction digest 169
Figure 4.19: Ddel restriction analysis of intron 7 polymorphism. 170
Figure 4.20: Electropherograms depicting polymorphisms identified in the
BIGH3 gene. 172
16 List o f Figures
Figure 5.1: Genomic structures of CHST5 and CHST6. 188
Figure 5.2: Pedigrees structure of MCD families 191
Figure 5.3: Slit lamp photographs of MCD before and after
keratoplasty 193
Figure 5.4: Electrophereogram of the coding region of CHST6 198
Figure 5.5a: An example of restriction analysis, Apa I
restriction enzyme digest on 3% agarose gels. 200
Figure 5.5b: ^5/N I restriction digest on 6% polyacrylamide gels 201
Figure 5.5c: Drd I restriction digest on 3% agarose gels 202
Figure 5.6a: Inhibition-ELISA of sera from MCD families 204
Figure 5.6b: Inhibition ELISA for KS 204
Figure 5.7: Alignment of C-GlcNAc6ST and other carbohydrate
sulfotransferase protein sequences 207
Figure 5.8: Prediction of secondary structure of the 3’ phosphate
binding domain in normal and A206V CHST6 mutation 208
17 Declaration
Declaration
I, Mohamed El-Ashry, declare that, unless otherwise stated, this thesis, entitled "Molecular genetics of comeal dystrophy" and submitted to The University of London for the M.D. degree is entirely my own original work and composition and has not been submitted for a degree at any other university.
Signed:
Mohamed F W^Uhry MB CHB FRCS (Ed)
Date:
18 A cknowledgements
Acknowledgements
First of all I would like to thank ALLAH to who I owe everything in my life. I am particularly grateful to my supervisors, Professor Shomi Bhattacharya and Dr Alison Hardcastle for their support and guidance throughout the project. I am deeply grateful for all your support, love and understanding, Alison. And to Mr Stephen Tuft and Miss Linda Picker for their help in facilitating patient ascertainment or for identifying probands. I would like to thank Neil Ebenezer, the dearest friend, whom I worked closely alongside. I would also like to thank all the families and patients without whom the project would not have been possible.
I also acknowledge the following consultants and colleagues for their help Mr J Dart, F Larkin, D Gartery, A Webster, B Clarke, Prof A Bird, Simon Wilkins, Bart Leroy, Annette Payne, Ordan Lehmann, Catherine Wills and Catherine Plant. There are many other people whom I wish to thank, but are too numerous to name individually. Every one in the molecular biology department at Institute of ophthalmology has always been totally supportive. The whole of the lab has been absolutely brilliant to work with. Sue, Eranga, Stephanie and Tin deserve extra credits for painstakingly reviewing this thesis.
Particular thanks to Sue Wilkie and Reshma Patel for helping with the protein modelling. I extend my gratitude to my family for their support and encouragement throughout my MD, especially over the last months. And last but not least to my wife, Mai and my son, Ahmed. I can honestly say that without you this thesis would not have been completed. And thank you to every one who has supported and encouraged me throughout my MD, especially over the last months.
19 Abbreviations
Abbreviations
A: Adenine ABI: Applied Biosystems Incorporated AD: Autosomal dominant AR: Autosomal recessive ALS: Acid-labile sub-unit of a serum insulin-like growth factor ATP: Adenosine triphosphate ACD: Avellino comeal dystrophy BAG: Bacterial artificial chromosome Bp: Base pair C: Cytosine cDNA: Complementary deoxyribonucleic acid CEPH: The Centre d'Etudes du Polymorphisme Humaine (Paris) C-GlcNAc6ST: Comeal GlcNAc-6-sulfotransferase CHED: Congenital hereditary endothelial dystrophy CHLC: Co-operative Human Linkage Centre Chon6ST: Human chondroitin-6-sulfotransferase CM: CentiMorgan CNAl : Autosomal dominant comea plana CNA2: Autosomal recessive comea plana DNA: Deoxyribonucleic Acid dNTP: Deoxynucleotide triphosphate EDTA: Ethylenediamine- tetra acetic Acid ELISA: Enzyme linked immunosorbent assay EST: Expressed sequence tag G: Guanine GDB: Genome data base GCD: Granular comeal dystrophy HecGINAc6ST: Human high- endothelial-cell GlcNAc-6-sulfotransferase GIcNAc6ST: Human GlcNAc-6-sulfotransferase HGMP Human Genome Mapping Project I-GlcNAc6ST: Intestinal GlcNAc-6-sulfotransferase lOP: Intraocular pressure
20 Abbreviations kb: Kilobase KS: Keratan Sulfate KSG6ST: human keratan sulfate Gal-6-sulfotransferase KSPG: Keratan sulfate proteoglycans LCD: Lattice comeal dystrophy LOD: Likelihood of the Odds LK: Lamellar keratoplasty Mb: Megabase MestST: Mouse oestrogen sulfotransferase MCD: Macular comeal dystrophy mRNA: Messenger ribonucleic acid NCBI: National Centre for Biotechnology Information NIH: National Institute of Health OMIM: Online Mendelian Inheritance in Man PCR: Polymerase Chain Reaction PIC: Polymorphism information content PK: Penetrating keratoplasty PTK: Phototherapeutic keratectomy PPD: Posterior Polymorphous Dystrophy RBCD: Reis-Bücklers’ comeal dystrophy RFLP: Restriction fragment length polymorphism RNA: Ribonucleic Acid SNP: Single nucleotide polymorphisms STRPs: Short tandem repeat polymorphisms 0: Theta, recombination fraction T: Thymine TBE: Tris-borate-EDTA Buffer UTRs: Untranslated regions VNTR: variable number of tandem repeats YAC: Yeast Artificial Chromosome
21 Key to pedigree symbols
Key to pedigree symbols
o □ e ■ affected female and male respectively / / 0 deceased individuals O-n non-consangumeous umon ( ^ consanguineous umon / proband cases examined
22 Introduction
Chapter 1
1.1 General introduction
Molecular genetics is revolutionising our understanding of the classification and pathophysiology of ophthalmic diseases, particularly comeal disorders. One of the most
important blinding comeal disorders is comeal dystrophy by virtue of its interference with comeal transparency. The present classification of comeal dystrophy, based largely on the clinical description of the phenotype, is being replaced with a more rational classification based on genetic background.
Advances in molecular knowledge of comeal dystrophy via the recognition of comeal disease genes and of fundamental defects in their DNA will undoubtedly lead to a better understanding of normal comeal physiology and pathobiology. Once the genes for specific comeal dystrophy are identified, it will be possible to develop cell culture and animal models to study the function of the relevant genes. This will ultimately lead to the design of better, innovative methods for the treatment of comeal dystrophy using
logical strategies with chemical agents or methods of gene therapy.
The work in this thesis concems mapping the gene for a family that presented with a combined phenotype with comea plana and microphthalmia. This is followed by a case report of an intrafamilial phenotypic variation in a family presenting with two types of comeal dystrophies. Then Phenotypic and genotypic characterisation of families and sporadic patients with autosomal dominant stromal comeal dystrophy via slit lamp biomicroscopic examination and mutational screening of BIGH3 gene, the disease gene for a number of stromal comeal dystrophies, is presented. And finally mutation
23 Introduction screening of the carbohydrate sulfotransferase 6 gene {CHST6), the gene implicated in the pathogenesis of macular comeal dystrophy, an example of autosomal recessive stromal dystrophy.
Therefore in the following sections an overview of the genetic concepts behind the project will be described with more emphasis given to the phenotype of the patients involved in this study and the strategies employed for the identification of the disease gene. The reader will also be introduced to the structure and function of the comea, clinical, pathological and genetic background of comeal dystrophy.
1.2 Cornea
1.2.1 Gross Anatomy
The comea is a transparent avascular tissue with a smooth convex outer surface and concave inner surface, which resembles a small watch glass. It covers the anterior one sixth of the outer coat of the eye and is continuous with the white opaque sclera at the limbus. It is slightly elliptical, measures about 12.5x11.5 mm with an average thickness of 0.52 mm in the centre and 1 mm in the periphery.
The main function of the comea is optical as it accounts for 70% of the total refractive power of the eye. The regular anterior comeal curvature and the optically smooth quality of the overlying tear film are the main factors that enable the comea to maintain its strength, transparency and refractive power.
1.2.2 Microscopic anatomy
The comea consists of five layers: epithelium. Bowman's membrane, stroma, Descemet's membrane and endothelium (Figure 1.1).
24 Introduction
EPITHELIUM
BOWMAN'S_____ LAYER
SUBSTANTIA, PROPRIA
DESCEMET'S MEMBRANE
ENDOTHELIUM - - ' '
CORNEA
Figure 1.1 Light micrograph of normal human cornea. Five layers are classically described: mutilayered comeal epithelium with its basement membrane, Bowman's membrane, comeal stroma with its keratocytes (substantia properia), Descemet's membrane and the single layered comeal endothelium (reproduced from Krachmer et al, 1 9 9 7 ).
25 Introduction
1.2.2.1 Epithelium
The surface of the cornea is covered by stratified squamous non-keratinising epithelium that constitutes 10% of the comeal thickness. It is 50-52 pm thick and is formed of five to seven layers which can be divided morphologically into three types of cells: columnar basal, polygonal wing and squamous superficial. Normally the basal cells are mitotically active, so they produce the daughter cells that move forward to become the wing cells that migrate anteriorly to become the flattened squamous cells that then slough into the tear film. The superficial cells represent the highest level of differentiation that graduate into transitional epithelium at the limbus and then into conjunctival epithelium.
For the comeal epithelium to act as a mechanical barrier to foreign materials and to create a smooth transparent optical surface, the comeal epithelial cells must be tightly adherent to each other and to the underlying stmctures. This is facilitated by several types of specialised junctions that can be divided into three classes: occluding, anchoring and communicating (Randall et al., 1998). The occluding junctions are found between the optical portion of basal cells, between basal and wing cells and between the wing and the squamous cells. Anchoring junctions include zonula adherence and desmosomes. Communicating junctions include synapses and gap junctions. The cytoskeleton in the cytoplasm of the comeal epithelium is made up of three types of protein filaments: intermediate filaments, microtubules and actin filaments (Gipson and
Joyce, 2000). The most important is the intermediate filaments as they provide the stmctural fi-amework of the cytoplasm of the epithelial cells and they are also a component of desmosome and hemidesmosomes. These intermediate filaments are
26 Introduction formed by the pairing of two specific keratin proteins, type I or acidic and type II or basic. These keratins provide a structural framework for the comeal epithelium.
1.2.2.2 Bowman's membrane
Bowman's layer is an acellular zone 8-10 pm thick that forms an interface between the basal lamina of the epithelium and the cellular stroma. Both epithelial cells and stromal keratocytes are believed to have a role in the laying down and maintenance of Bowman's layer (Rodrigues et al., 1982; Tisdale et al., 1988). Bowman's membrane contains collagen type I, V, and VII and chondroitin sulfate proteoglycans. This layer along with the stroma may help the cornea to maintain a fixed shape, thereby enabling it to perform its optical function.
1.2.2.3 Stroma
The stroma constitutes about 90% of comeal thickness. It is about 500 pm thick, formed of regularly arranged lamellae of collagen bundles, stromal cells and ground substance.
The collagen bundles account for 80% of the dry weight of the comea. Collagen type I and V are the predominant fibrillar collagens in the stroma, although small amounts of other fibrillar collagens such as type III, VI and XII may be present (Trelstad et al.,
1972; Zimmermann et al., 1986). The ground substance surrounding the collagen fibrils is composed of proteoglycans. Keratan sulfate with its core protein lumican and chondroitin sulfate with its core protein decorin are the predominant proteoglycans within the comeal stroma in a ratio of approximately 3:1 (Hassell et al., 1992). These proteoglycans play a role in maintaining the ordered spacing of collagen fibrils that is thought to be necessary for the maintenance of comeal transparency. The predominant
27 Introduction cell type in the comeal stoma are the keratocytes which are flat cells, derived from neural crest, that lie between collagen bundles. In response to stromal injury, the keratocytes migrate into the wound area and undergo transformation into fibroblasts, which contribute to scar formation by proliferation and collagen production (Randall et al., 1998). In addition to keratocytes, polymorphonuclear leucocytes, plasma cells and macrophages in small numbers are seen in the normal stroma.
1.2.2.4 Descemet's membrane
Descemet's membrane is approximately 10 pm thick in adults, consists of thick extracellular matrix, which is synthesised and secreted by the comeal endothelium. Its peripheral terminus is known as Schwalbe's line. Normal Descemet's membrane has two morphologic components an anterior banded zone secreted in utero at about 4 months of gestation and is 3 pm thick and a posterior homogenous zone, which is produced after birth and thickens progressively with age. It contains type IV and type VIII collagen and fibronectin. The fibronectin may play a role in maintaining the endothelium in a uniform monolayer and in attaching it to the Descemet's membrane (Maurice, 1987). In contrast to Bowman's layer, Descemet's membrane is easily detached from the stroma and regenerate readily after injury.
1.2.2.5 Endothelium
The comeal endothelium forms a single layer of cuboidal cells on the posterior comeal surface. The endothelial cells are about 10 pm height at birth; flatten with age to about 4 pm in adults. The cell density decreases from approximately 3500-4000 cells/mm^ at birth to 2500-3000 cells/mm^ in the adult comea. This increase in cell area and decrease
28 Introduction in cell density is due to gradual decrease in the endothelial cell population with age or trauma so, the adjacent cells spread out to cover the vacant areas (Cox et aL, 1972; Grass and Robinson, 1988). The numerous mitochondria within the cytoplasm of the endothelial cells reflect its main function for maintaining the relatively low level of stromal hydration necessary for comeal transparency.
1.2.3 Corneal transparency
The comea transmits approximately 90% of the visible spectrum of light waves that enter it. To perform such function, the comea must be transparent through interaction of anatomical and physiological factors. First, the presence of a continuous sheet of epithelium on its anterior surface inhibits diffusion of water and solutions from tears into the stroma. Secondly, the regular arrangement of stromal collagen bundles which is maintained by the relatively low level of stromal hydration. Also, organisation of tissue proteoglycans in the collagen matrix that maintain the collagen fibrils in an orderly lamellar array plays an important role. Lastly the endothelial pump, which retards the flow of aqueous into the stroma, determines the level of hydration of the glycosaminoglycans and thus transparency.
1.2.4 Development of the cornea
The heterogeneous tissues that constitute the human eye are derived embryologically from surface ectoderm, neural ectoderm, neural crest, and mesodermal mesenchyme. For successful comeal development, the lens must be present; its loss before mesenchymal invasion results in microphthalmos and opaque comea (Zinn and Mockel, 1975).
29 Introduction
Comeal development begins at approximately 5 to 6 weeks of gestation by invagination
of the surface ectoderm into the optic cup producing the lens vesicle, which
subsequently detaches from the overlying surface ectoderm (O’Rahilly, 1983). Then a
basal lamina is formed from a faint amorphous line beneath the surface ectoderm, by this
time the surface ectoderm can be referred to as epithelium. The detachment of the lens
vesicle induces the basal layer of the epithelium to secrete collagen fibrils to fill the
space between the developing epithelium and the anterior surface of the lens forming the
incipient comeal stroma (Millin et al., 1986). At approximately 6 weeks of gestation, mesenchymal cells derived from the neural crest, which are situated at the margins of the rim of the optic cup, migrate into the developing eye beneath the basal lamina of the comeal epithelium to form the comeal endothelium. This event marks the first of the three successive waves of ingrowth of the neural crest cells into the developing eye
(Figure 1.2). The second wave forms the iris and the papillary membrane. So by 39 days, the comea consists of two-layered epithelium, an acellular matrix of collagen and an endothelial layer consisting of two or three layers of cells (Figure 1.3).
The acellular stroma expands possibly as a result of hydration of the increasing amounts of extracellular matrix (Figure 1.3). This setting prepares the way for the migration of mesenchymal (future keratocytes) from the rim of the optic cup marking the third wave of migration (Figure 1.2) (Flay, 1980). By the 714 weeks the mesenchymal cells are arranged as four to five incomplete layers and a few collagen fibrils are present. By 3 months of gestation, the epithelium consists of two to three layers of cells, the stroma has 25-30 layers of keratocytes that are arranged more regularly in the posterior half, and a thin uneven Descemet's membrane is present between the most posterior keratocyte and the now monolayer endothelium. Initially the endothelium is formed of a strand of
30 Introduction cells with contiguous and overlapping processes. By 4 months of gestation the endothelial cells begin to lay down their basement membrane and the comeal epithelial cells differentiate into three types: small, medium sized and large cells. By the end of the fourth month of gestation the Bowman’s membrane is formed by the epithelial basal lamina, which secretes the collagenous tissue. By 7 months of gestation the Descemet’s membrane is demarcated clearly and the stroma consists of 15 layers of cells containing numerous active keratocytes (Figure 1.3).
Ep:*,helium
'/V.
R e tin a
Figure 1.2 Diagrammatic representation of the three successive waves of ingrowth of neural crest cells. 1, First wave forms comeal endothelium; II, second wave forms iris and pupillary membrane; III, third wave forms keratocytes (reproduced from Krachmer et al., 1997).
31 D 3 months Introduction A 39 B 7VI«€j« Stnxre —mm * C 7)» weeks D esccm el'8 ------ " Endothelium E 4X m o n th s Eprthefium F 7 months ÔBb]ÉElovo2e,œ„ B ow m ans —_ loyer ^ ?=-z?2S^ -jI . —Stroms - Doscemets fî sni^os^rss membrane EndothcHum Figure 1.3 Diagrammatic representation of the development of the comea. (a) At 39 days' gestation a two layered epithelium rests on its basal lamina and is separated from the endothelium, which consists of two or three layers of cells, by a narrow space, (b) at 7 weeks mesenchymal cells at the periphery migrate between the epithelium and the endothelium; (c) by 71/2 weeks the mesenchymal cells (future keratocytes) are arranged as four to five incomplete layers and a few collagen fibrils are present; (d) by 3 months of gestation, the epithelium consists of two to three layers of cells, the stroma has 25-30 layers of keratocytes that are arranged more regularly in the posterior half, and a thin uneven Descemet's membrane is present between the most posterior keratocyte and the now monolayer endothelium; (e) by 3 !4 months Bowman’s layer is formed by the superficial keratocytes and the stroma consists of immature extracellular matrix that includes collagen fibrils and active keratocytes; (f) by 7 months the Descemet’s membrane is demarcated clearly and the stroma consists of 15 layers of cells and scant collagen fibrils (reproduced from Bron et al.^ 1997). 32 Introduction 1.2.5 Metabolism and physiology of the cornea Comeal avascularity is an important factor for maintenance of its transparency. Despite this avascularity, the comea can obtain its nutritious requirements through its metabolically active epithelial and endothelial cells (Nishida, 1997). Energy is required by such cells in order to supply their metabolic needs. Therefore, the supply of glucose and oxygen is essential to maintain the normal functions of the comea. All layers of the comea obtain oxygen through the anterior comeal surface, primarily by diffusion from tear fluid and from the limbal vessels. The direct exposure of the tear fluid to the atmosphere is essential for the supply of oxygen to the comea (Nishida, 1997). Glucose and most other nutritious substances are obtained through the posterior comeal surface by diffusion through the aqueous humor. An exception is retinol, which is supplied by tears to superficial epithelial cells. Without the retinol the epithelial cells become keratinised and dysfunctional (Ubels and MacRae, 1984). The surface of the comea is covered by the tear fluid which protects the comea from dehydration and maintains the smooth epithelial surface. The tear film is about 0.7 mm thick and consists of three layers: a superficial lipid layer, a middle aqueous layer and an inner mucin rich glycocalyx layer (Van Setten and Schultz, 1994). It contains many biologically important factors and active substances. Therefore, it serves not only as a lubricant and nutritional source for the comeal epithelium but also as a source of the regulatory factors for the maintenance and repair of the comeal epithelium (Nishida, 1997). Comeal stroma normally is kept in a relatively deturgesced and therefore, optically transparent state. It naturally imbibes water because of two forces: first, the glycosaminoglycans exert an osmotic pressure that pulls water into the stroma; second, 33 Introduction the intraocular pressure forces aqueous humor into the stroma. The endothelium counteracts this hydrophilic tendency and maintains comeal transparency in two ways: first, its barrier function decreases the flow of water into the stroma; second, its metabolic pump transports ions from the stroma to the aqueous humor, and water follows by diffusion. So stromal hydration is kept in balance by opposing forces: the stromal swelling pressure reinforced by the intraocular pressure and the endothelial pump on one side and the epithelial barrier on the other side (Figure 1.4). EPITHELIAL BARRIER ft" ENDOTHELIAL BARRIER STROMAL IMBIBITION PRESSURE INTRAOCULAR PRESSURE TEAR FILM EVAPORATION ENDOTHELIAL TRANSPORT EPITHELIAL TRANSPORT \N Figure 1.4 Factors and forces involved in the control of corneal stromal hydration. Intraocular pressure and stromal imbibition pressure are forces that promote water accumulation in the stroma. Epithelial and endothelial barriers are factors that reduce oedema (reproduced from Kaufman et al, 1998). 34 Introduction 1.3 Corneal dystrophy: A phenotypic review Comeal dystrophies are a genetically and clinically heterogeneous group of disorders. Most comeal dystrophies are inherited as an autosomal dominant trait. The autosomal dominant pattem of comeal dystrophies accounts for the variable expressivity, not only among unrelated affected individuals but also within a single family and even between the two eyes of the same patient (Waring III and Mbekeani, 1998). By conventional definition, comeal dystrophies are primary comeal diseases unassociated with prior inflammation or trauma, or with systemic diseases. This definition distinguishes readily between dystrophy and degeneration. A dystrophy has been defined as a condition of deficiency of tissue nutrition, whereas degeneration indicates degradation, or passage from a higher structure to a lower one (Alexander and Messenger, 1957). Clinically, comeal dystrophy occurs most frequently bilaterally and is manifested around the time of puberty, progressing slowly throughout life (Mc-Tigue, 1967). The prevalence of comeal dystrophy among different population cannot be predicted because there are few studies documenting the frequency with which comeal dystrophy occur. However the reported data are considered as a measure of the prevalence of those cases severe enough to need keratoplasty. Comeal dystrophies are classified according to the layer or layers of the comea involved. So, the disease phenotype can be categorised according to anatomy of the comea. 35 Introduction 1.3.1 Epithelial dystrophies Epithelial dystrophies are characterised by intraepithélial cysts and multiple layers of subepithelial basement membrane and fibrillar tissue. They include Meesmann's dystrophy, Cogan's map-dot-fingerprint, and band-shaped dystrophy. 1.3.1.1 Meesmann's dystrophy Meesmann's comeal dystrophy is a rare, juvenile, autosomal dominant disorder, first described clinically by Pameijer in 1935. It has a strong penetrance but variable expressivity (Wittebol et al., 1987a). It appears during the first year of life, manifesting as tiny epithelial vesicles that gradually spread throughout the comeal epithelium, which can be visualised by retro-illumination and slit lamp examination (Figure 1.5). Visual acuity remains good in most of cases, with some blurring as recurrent punctate erosions may lead to lacrimation, blepharospasm and photophobia (Corden et al., 2000). Meesmann's dystrophy has been found to be caused by mutations in two keratin genes, keratin 3 (I2q31) and keratin 12 (17q2I) (Table l.I) (Irvine et al., 1997; Nishida et al., 1997). Both keratins 3 and 12 encode cytoskeletal proteins and they are specifically expressed within the comeal keratocytes, which are regarded as markers of advanced comeal epithelial differentiation. Mutations result in focal aggregation of keratin within the cytoplasm of the comeal epithelium that form the characteristic "peculiar substance" detected by transmission electron microscopy (Rodrigues et al., 1987; Schermer et al., 1986). 36 Introduction 1.3.1.2 Epithelial basement membrane dystrophies Epithelial basement membrane dystrophy (also known as Cogan microcystic dystrophy or Map-dot-finger-print dystrophy) is the most common of all anterior comeal dystrophies. The term map-dot-finger-print dystrophy is descriptive of the biomicroscopically visible features of intraepithélial microcysts (dots), subepithelial ridges (finger-prints) (Figure 1.6), and geographic opacities (maps) (Rodrigues et al., 1974). The disorder occurs bilaterally with autosomal dominant pattem of inheritance showing variable penetrance (Laibson and Krachmer, 1975). Clinically most patients are asymptomatic, and the map changes are faint. The second most common clinical presentation is transient blurring of vision when the epithelium dries as in hot dry and windy conditions. The least common form of this disorder occurs in patients between the ages of 20 and 40 years where they develop painful recurrent epithelial erosions, most commonly on awakening (Waring et al., 1978). Half of the patients who develop recurrent erosions may have map dot fingerprint dystrophy; conversely 10% of patients with map dot fingerprint dystrophy have recurrent erosions (Weblin et al., 1981). The pathogenesis of epithelial basement membrane dystrophy is probably the synthesis of abnormal basement membrane and adhesion complexes by the dystrophic epithelium. Epithelial cells adjacent to the aberrant basement membrane have intracellular junctions but do not form hemidesmosomes. Consequently, epithelial erosions result, as the epithelial cells are poorly adherent to their basement membrane (Fogle et al., 1975). Familial cases of map-dot-fingerprint dystrophy with early onset have been described, but no definite information is currently available about genetic linkage (Gupta and Hodge, 1999). 37 Introduction 1.3.1.3 Band-shaped dystrophy Band-shaped or microcystic dystrophy or Lisch dystrophy is an epithelial dystrophy characterised by bilateral or unilateral gray opacities in the form of a band, a feather and a whorl (Lisch et al., 1992). In retro-illumination, the opacities appear as an intra- épithélial and densely crowded clear microcysts (Robin et al., 1994). The condition was reported in five family members and three unrelated patients from Germany. Visual acuity was severely reduced in three of the reported patients. Microcystic dystrophy is clinically distinct from Meesmann's and epithelial basement membrane dystrophies (Waring III and Mbekeani, 1998). Recently Lisch et al. (2000) reported a family with a total of 48 members, excluded linkage for the keratin K3 and K12 genes and revealed linkage with a maximum LOD score of 2.93 at Xp22.3. 1.3.2 Bowman's layer dystrophies Dystrophies that primarily affect Bowman's layer include Reis-Bücklers' dystrophy. Honeycomb dystrophy of Theil and Behnke and Grayson-Wilbrandt dystrophy. 1.3.2.1 Reis-Bucklers' corneal dystrophy (type I Bowman's layer, CDBI) Reis and Bucklers first described Reis-Bücklers’ comeal dystrophy (CDBI) as an autosomal dominant disorder with strong penetrance. Its initial manifestations are in early childhood in which the patient has recurrent comeal erosions but retains relatively good vision. The erosions occur approximately three to four times a year and gradually 38 Introduction decrease in frequency over 5 to 20 years of age but vision usually begins to deteriorate when patients are in their twenties (Yamaguchi et al., 1982; Mol 1er, 1989a). Clinically the comeal appearance in early stages is characterised by bilaterally symmetric fine reticular opacification at the level of Bowman's membrane. As the disease progresses Bowman's layer becomes replaced by scar tissue by mid life. Slit lamp examination usually discloses focal grey-white subepithelial opacities with a geographic configuration that occupy the central comea and spread paracentral ly (Figure 1.7) (Kanai etal., 1973; Yamaguchi et al., 1980). Histopathologically the lesions are characterised by a marked pink stain with Masson's trichrome and the presence of rod shaped bodies on transmission electron microscopy. The ultrastmctural findings of rod-shaped bodies are similar to those observed in the superficial variant of granular comeal dystrophy (CDGGI). As a result, CDBI may be confused with a superficial variant of CDGGI (Bron and Rabinowitz, 1996). The nature and location of the primary abnormality in Reis-Bücklers' dystrophy is unclear. It has been suggested that the superficial keratocytes produce abnormal collagen fibrils, damaging the Bowman's layer, with secondary epithelial injury. Another possibility is that a primary epithelial cell abnormality leads to recurrent epithelial erosions, with secondary activation of stromal cells, leading to deposition of fibrocellular material and absorption of Bowman's layer (Cogan et al., 1964; Gurrey, 1965). Kuchle et al. (1995) described two forms of Bowman layer dystrophies, CDBI (geographic or "tme" Reis-Bücklers' dystrophy) and CDBII (honeycomb-shaped or Theil-Behnke dystrophy) The chromosomal locus for CDBII has been mapped to 5q31 by linkage analysis (Small et al., 1996). Subsequently, it has been reported that the original or the geographic form 39 Introduction of CDB is caused by heterozygous R124L point mutation of the BIGH3 gene (Okada et ai, 1998a) (Table 1.1). 1.3.2.2 Honeycomb dystrophy of Thiel and Behnke (type II Bowman’s layer, CDBII) Honeycomb dystrophy was first described by Thiel and Behnke in 1967 as a bilateral, subepithelial dystrophy transmitted as an autosomal dominant trait. It begins in childhood and runs a progressive course of recurrent erosions and decrease in vision. Slit lamp examination reveals bilaterally symmetrical, central, confluent honeycomb opacities with grey interlacing lines punctuated by clearer zones. Anterior extension of the subepithelial grey material leads to thinning of the comeal epithelium (Stocker and Holt, 1955). Histopathologically, the most characteristic finding is the subepithelial accumulation of atypical, fine, curly, collagen filaments. These curly fibres are considered pathognomonic of type II Bowman's layer dystrophy (Kuchle et al., 1995). The pathogenesis of CDBII is unknown however, the primary lesion may be due to fragmentation of the collagen fibrils of Bowman's layer and the epithelial lesion may occur secondarily (Yamaguchi et al., 1980). CDBII has been reported to be caused by heterozygous R555Q point mutation of the BIGH3 gene (Munier et al., 1997). Recently, a new locus for Thiel-Behnke has been mapped to a small region on chromosome lGq23-q24 (Yee et al., 1997) (Table 1.1). 40 Introduction 1.3.2.3 Grayson-Wilbrandt dystrophy Grayson and Wilbrandt in 1966 described an anterior membrane dystrophy in two generations family with autosomal dominant inheritance. The onset of the disease does not occur until 10 or 11 years of age, and erosions are infrequent. Slit lamp examination reveals polymorphic opacities at the level of Bowman's layer and clear comea between the opacities. The disorder is considered as an attenuated form of CDB. However, it differs from most cases of CDB in its later onset, infrequency of erosions, variable effect on vision, and normal comeal sensation (Grayson and Wilbrandt, 1966) 41 Introduction Figure 1.5 Meesmann's comeal dystrophy. In retroiliumination multiple eysts within the comeal epithelium are shown (arrows). Figure 1.6 Epithelial basement membrane dystrophy, finger-print pattem. In retroiliumination the finger-print lines form a roughly parallel configuration of refractile ridges (rectangle). Figure 1.7 CDBI. A narrow slit view, shows the sub-epithelial location of the grey deposits. Photographs are reproduced from Krachmer et al., 1997 42 Introduction 1.3.3 Stromal dystrophies Stromal comeal dystrophies are hereditary disorders characterised by deposition of abnormal substance within the keratocytes or among the collagen fibrils. The nature of this substance varies according to each type of dystrophy. 1.3.3.1 Granular corneal dystrophy (CDGGI) Granular comeal dystrophy, also known as Groenouw dystrophy type I (CDGGI) or breadcmmb dystrophy is an autosomal dominant disorder with 100% penetrance (Moller, 1990). It was originally described by Groenouw as a noduli comeae owing to its characteristic sugar granule-like opacities located centrally in the comea. Bilateral, symmetrical, sharply demarcated, grey white opacities develop in the first and second decades (Figure 1.8). As time progresses the opacities progress, coalesce, and involve the deeper stromal layers. Histologically, Masson trichrome positive materials and rod shaped bodies are present between stromal keratocytes (Jones and Zimmerman, 1961). Granular dystrophy has been identified as another type of comeal dystrophy caused by mutations in the BIGH3 gene. Recently, it has been reported that the typical form of CDGGI is caused by mutation at codon 555 (arginine to tryptophan) in the BIGH3 gene (Munier et al., 1997). Okada et al. (1998b) reported a homozygous mutation in the same codon resulted in a severe placoid type of the disease and Stewart et al. (1999a) described a form of CDGGI with R124S mutation in the BIGH3 gene and an atypical form of CDGGI with R124H mutation reported previously for ACD but with no lattice features as determined by histopathological examination. 43 Introduction 1.3.3.2 Lattice corneal dystrophy (LCD) Lattice comeal dystrophy is the most common stromal dystrophy. It has been subdivided into five types. 1.3.3.2.1 Lattice corneal dystrophy type I (LCD type I) LCD type I has an autosomal dominant mode of inheritance with variable expressivity. The disease presents toward the end of the first or second decade of life and slowly progresses to cause discomfort and visual impairment, usually before the sixth decade. It is characterised by thin greyish, linear branching deposits of amyloid material that progressively accumulate in the subepithelial and stromal layers of the comea (Durand et al., 1985) (Figure 1.9). The specific cause of the amyloid deposits is unclear, however, they may be secondary to collagen degeneration or produced by abnormal keratocytes. Histopathologically the white dots and lattice lines consist of amyloid in the comeal stroma, which is confirmed histochemically by the pink to orange staining of the deposits by Congo red (Waring III and Mbekeani, 1998). Recent studies have shown that LCD results from BIGH3 gene mutations and that an R124C mutation was detected in LCDI in both Caucasian and Japanese families (Munier et el., 1997; Hotta et al., 1998; Korvatska et al., 1998; Gupta et al., 1998; Afshariet al., 2001). In addition, a different mutation, L518P, in the BIGH3 gene has been reported to cause LCDI (Endo et al., 1999; Hirano et al., 2000). 44 Introduction 1.3.3.2.2 Lattice corneal dystrophy type II (LCD type II) Lattice comeal dystrophy type II was originally described by Meretoja (1969) and has been called familial amyloid polyneuropathy type IV or Finnish-type familial amyloidosis. Tlie disease has an autosomal dominant pattem of inheritance and is characterised by accumulation of amyloid material in the comea and other parts of the body giving rise to systemic manifestation including cranial and peripheral neuropathies, skin changes such as lichen amyloidosis, cutis laxa, blepharochalasis, protmding lips and mask facies (Starck and Kenyon, 1991). The condition begins in the third decade of life with randomly scattered short fine glassy lines, which are sparse, more delicate and more radially oriented than those in LCD type I (Waring III and Mbekeani, 1998) (Figure 1.10). The condition is caused by mutations in gelsolin gene {GSN) on the long arm of chromosome 9 (9q34) (Maury et al., 1990; dela-Chapelle et al., 1992). 1.3.3.2.3 Lattice corneal dystrophy type III (LCD type III) It is inherited as an autosomal recessive trait and has a late adult onset (70-90 years). It was recently reported in two Japanese families with no evidence of systemic amyloidosis (Waring III and Mbekeani, 1998). The disorder occurs unilaterally or bilaterally in the form of thick, ropy lattice lines extending from limbus to limbus. Visual acuity is not greatly affected until the sixth decade and comeal erosions are rare (Hida et al., 1987). The nature of the amyloid and the gene responsible for this disorder have not been determined (Kawasaki et al., 1999). 45 Introduction 1.3.3.2.4 Lattice corneal dystrophy type IIIA (LCD type IIIA) This type of dystrophy is different from LCD type III as the pattem of inheritance is autosomal dominant, comeal erosions represent a substantial component of the disease and the occurrence is mainly in Caucasian (Stock et al., 1991). Thick lattice lines that traverse the comea from limbus to limbus characterise this form of dystrophy (Figure 1.11). Lattice type ITT A is associated with P501T, N622H, H626R, A546T mutations in the BIGH3 gene (Yamamoto et al., 1998; Stewart et al., 1999b; Kawasaki et al., 1999; Schmitt-Bemard et al., 2000; Dighiero et al., 2000a). 1.3.3.2.5 Lattice corneal dystrophy type IV (LCD type IV) This is another type of late onset atypical LCD characterised by large nodular and lattice shaped deposits in the deep comeal stroma of the pupillary zone (Fujiki et al., 1998a). It has an autosomal dominant pattem of inheritance and is associated with L527R mutation in the BIGH3 gene (Fujiki et al., 1998a; Hirano et al., 2001). 46 Introduction Figure 1.8 CDGGI. Amorphous subepithelial 1 deposits with clear stroma inbetween are shown (arrows). H H H k Figure 1.9 LCDI. Slit lamp photomicrograph shows refractile lines (arrows) in retroillumunation. Figure 1.10 LCDII. The lattice lines (arrows) are delicate and extend to the peripheral comea Hv r n Figure 1.11 LCD IIIA.The arrows show thick, ropy and branching lattice lines. Photographs are reproduced from Krachmer et at., 1997 47 Introduction 1.3.3.3 Avellino corneal dystrophy (ACD) This is also called granular comeal dystrophy type II or combined lattice granular comeal dystrophy. The disorder has been called ACD after the Italian province near Naples where the first affected families to be identified originated (Folberg et al., 1988). It has an autosomal dominant pattem with high penetrance. Patients present, in their first or second decade of life, with well circumscribed central anterior stromal opacities and deeper lattice like lines (Figure 1.12). In tissue sections rod-shaped crystalloid bodies accumulate in the comeal stroma together with amyloid deposits (Holland et at., 1992; Rosenwasser er a/., 1993). To date, all cases studied with molecular genetic techniques have a mutation in codon 124 of BIGH3 in which histidine replaces arginine (Munier et at., 1997; Korvatska et al., 1998; Afshariet al., 2001). 1.3.3.4 Macular cornea! dystrophy (MCD) Macular comeal dystrophy (MCD; OMIM 217800) is also known as Groenouw dystrophy type II. It is the least common of the classic stromal dystrophies. The comeal changes are usually noted between 3 and 9 years of age, when a diffuse clouding and aggregations of greyish-white, irregularly shaped spots, are seen in the central superficial stroma (Figure 1.13). With time the comea becomes increasingly cloudy and vision is severely impaired by the second decade of life (Jones and Zimmerman, 1961). Three distinct immunophenotypes (MCD types I, LA, and II) have been identified based on the immunohistochemical reactivity of the comeal tissue to an anti-keratan sulfate antibody and keratan sulfate (KS) levels in comea and semm (Yang et al., 1988; Jonasson et al., 1996; Klintworth et al., 1997). In MCD type I neither the comea nor the 48 Introduction serum contain appreciable levels of KS, while in MCD type II KS is present in comea and serum. MCD type IA has no detectable KS in serum or in the comeal stroma but the accumulations within the keratocytes react with the anti-keratan sulfate antibody. MCD type I has been mapped to human chromosme 16 (16q21) and type II has been suggested to be allelic with MCD type I (Vance et al., 1996; Liu et al., 1998a; Liu et al., 2000a). Recently, Akama et al. (2000) identified a carbohydrate sulfotransferase 6 gene (CHST6) within the MCD critical region. Several mutations have been identified within the coding region of CHST6 in patients with MCD type I (Akama et al., 2000; Liu et al., 2000b; Bao et al., 2001) and two DNA rearrangements in the upstream region of CHST6 were found in MCD type II (Akama et al., 2000). 1.3.3.5 Fleck corneal dystrophy (Mouchetée or speckled corneal dystrophy) Fleck dystrophy is a rare autosomal dominant dystrophy which was first described by Francois and Neetens in 1957 in 31 members of a single pedigree. The condition is usually noted as an incidental finding on routine examination as vision is unaffected and comeal sensitivity is normal. Fleck like dystrophic lesions was found throughout all layers of the central and peripheral comeal stroma but not affecting the epithelium. Bowman's membrane, Descemet's membrane or the endothelium (Laibson, 1997) (Figure 1.14). Histopathological examination reveals abnormal keratocytes that on transmission electron microscopy appear to contain a fibrillogranular substance within intracytoplasmic vacuoles (Laibson, 1997). Histochemical staining shows glycosaminoglycans and lipids within these vacuoles. Based on these histological 49 Introduction findings, fleck dystrophy appears to be a storage disorder that involves glycosaminoglycans and complex lipids which is limited to the cornea. 1.3.3.6 Central crystalline dystrophy of Schnyder’s Central crystalline dystrophy is a dominantly inherited disorder and is one of the most rare and least severe type of stromal comeal dystrophies. Characteristic findings were initially described by VanWent and Wibout in 1924 and later by Schnyder in 1939 who clarified the entity and documented its stable and asymptomatic clinical course (Laibson, 1997). It is characterised by bilateral clouding of the central cornea, arcus lipoides and/or visible crystalline deposits of cholesterol in the stroma and suggested to be caused by an accumulation of phospholipids, unesterified cholesterol and cholesterol ester in the comeal stroma (Laibson, 1997) (Figure 1.15). Recently, the dystrophy has been linked to chromosome 1 but the disease gene has yet to be identified (Shearman et al, 1996). 50 Introduction Figure 1.12 ACD. Slit lamp photograph of large, well circumscribed anterior stromal and sub-epithelial fusiform lesions in later stages of granular-lattice dystrophy. Figure 1.13 MCD. Slit lamp biomicroscopy shows focal grey-white stromal opacities \ (arrows) throughout stroma with intervening stromal haze. Figure 1. 14 Fleck comeal dystrophy. Slit lamp examination shows individual stromal opacities located at various stromal levels. Figure 1.15 Central crystalline dystrophy. A frontal view shows central disc like opacity (arrow) with anterior stromal cholesterol crystals and peripheral comeal A arcus (asterisk). Photographs are reproduced from Krachmer et al, 1997 51 Introduction 1.3.4 Endothelial dystrophy The importance of comeal endothelial dystrophy arises because disturbance of the endothelial function can reduce vision markedly as a result of loss of comeal clarity. Three types of comeal dystrophy have been identified, Fuchs’ comeal dystrophy, congenital hereditary endothelial dystrophy (CHED) and posterior polymorphous comeal dystrophy (PPD). 1.3.4.1 Fuchs' dystrophy Fuchs’ endothelial dystrophy is a bilateral condition that is usually observed in the fifth or sixth decade and is more common in females. The clinical and pathologic hallmarks of Fuchs’ comeal dystrophy are focal thickenings of Descemet's membrane known as comeal guttatae that are seen as wart-like endothelial excrescences (Vogt, 1981). As comeal endothelial function decreases, stromal oedema manifests resulting in subepithelial and epithelial bullae (Figure 1.16), which mpture and cause pain. In the late stages of the disease, subepithelial scar formation lessens the formation of bullae but also decreases visual acuity (Miller and Krachmer, 1998). The condition is thought to be inherited in an autosomal dominant fashion. Genetic linkage information is not yet available however, mitochondrial defects might underlie the disorder (Albin, 1998). 1.3.4.2 Congenital hereditary endothelial dystrophy (CHED) CHED is characterised clinically by bilateral, diffuse comeal oedema in which the comeal thickness is two or three times greater than normal, but comeal diameter is not enlarged with grey white ground glass appearance of the comea that extends to the 52 Introduction periphery (Waring et al., 1978) (Figure 1.17). It can be inherited in both autosomal dominant and autosomal recessive forms. The autosomal dominant form (CHEDl) presents during the first two years with photophobia and lacrimation. However, sensory nystagmus does not develop as the corneas are clear at birth. In contrast, the recessive form (CHED2) is more common and more severe and is associated with nystagmus (Judisch and Maumenee, 1978). Histopathologically CHED is characterised by profound thickening of Descemet's membrane and degeneration of the endothelium (Stainer et al., 1982; Sedundo et al., 1994). CHEDl has been mapped to pericentromeric region of chromosome 20 (20pll- qll.2) in an area overlapping a gene for autosomal dominant posterior polymorphous dystrophy (Toma et al., 1995). 1.3.4.3 Posterior polymorphous dystrophy (PPD) PPD was first reported by Koeppe in 1916 as an uncommon, congenital disorder of Descemet's membrane and comeal endothelium (Laibson, 1997). It is transmitted in an autosomal dominant fashion but recessive pattern of inheritance has been demonstrated. In some cases, presentation occurs as cloudy comea at birth (Liakos and Casey, 1978). However visual impairment is minimal and slowly progressive, so the age of onset is difficult to determine. The comeal lesions are of various shapes, some of which are vesicular or annular with surrounding halos, at the level of Descemet's membrane (Figure 1.18). PPD patients usually demonstrate normal vision, which rarely requires keratoplasty. It has been suggested that abnormal differentiation of the embryonal precursors of the endothelial cells or endothelial cell transformation into epithelial like cells might underlie the pathogenesis of PPD (Grayson, 1974; Bahn et al., 1984). 53 Introduction Recently PPD has been mapped to the pericentromeric region of human chromosome 20 (20 qll) which is the same locus for CHEDl (Table 1.1). This raises the possibility that PPD may be allelic to CHEDl. 54 Introduction Figure 1.16 Fuchs’ endothelial dystrophy. Slit lamp view shows focal thickening of Descemet’s membrane (arrow) and stromal haziness. Figure 1.17 Congenital hereditary endothelial dystrophy. Severe comeal haziness together with diffuse, homogeneous, grey-white, ground-glass appearance of the central comea that extends to the periphery. Figure 1.18 Posterior polymorphous dystrophy. Slit lamp illumination shows central stromal haze and grey vesicle like opacities at the level of the Descemet’s membrane. Photographs are reproduced from Krachmer et al., 1997 55 Introduction 1.3.5 Ëctatic dystrophies 1.3.5.1 Keratoconus Keratoconus is a non inflammatory corneal thinning disorder which is characterised in its most advanced stages by a localised conical protrusion of the comea. The cone may be small or large, round or oval, and is located either near the visual axis or inferior to it (Figure 1.19) (Perry et aL, 1980; Wilson et aL, 1991). The prevalence of keratoconus in the general population has been estimated to be between 4 and 600 per 100,000 (Kennedy et aL, 1986). Irregular astigmatism and high myopia are among the characteristics of keratoconus. Falls and Allen (1969) reported a family with irregular dominant inheritance through two generations. Also, keratoconus has been reported in seven sets of monozygotic twins. In five sets, both twins are affected (Bourne and Michels, 1982; Harrison e/a/., 1989). 1.3.5.2 Keratoglobus A bilateral comeal ecstatic disorder characterised by a globoid protmsion of a clear, diffusely thin comea of normal to moderately increased diameter. The comeal stroma is one-third to one-fifth normal thickness and is most attenuated near the limbus (Figure 1.20). Keratoconus and keratoglobus may be genetically linked in some cases (Biglan et aL, 1977). Cavara, (1950) reported a father with keratoglobus and a son with keratoconus. Both keratoconus and keratoglobus may be associated with leber congenital amaurosis or with blue sclera syndrome (Biglan et aL, 1977). 56 Introduction 1.3.5.3 Megalocornea A comeal disorder in which the comeal diameter is greater than or equal to 13 mm (Figure 1.21). It is most commonly transmitted as an X-1 inked recessive disorder, and for that reason, 90% of affected patients are males (Meire, 1994). It is associated with numerous ocular and systemic disorders. The entire anterior segment is disproportionately larger than the rest of the eye (Vail, 1931). Megalocornea is usually bilateral, non progressive and the comea remains clear except over sites of tears in Descemet’s membrane. The locus for megalocomea has been identified on chromosome Xql2 (Mackey e/a/., 1991). 1.4 Anomalous corneal development 1.4.1 Cornea plana Comea plana is a rare congenital anomaly in which the comea is flat with a comeal curvature of less than 43 diopters, the radius of curvature may reach levels as low as 20 or 30 diopters. The limbal landmarks are also obscured, simulating microcomea. Peripheral sclerisation of the comea is almost always present and the condition is indistinguishable clinically from peripheral sclerocomea. The former terminology is used more frequently in Europe while the latter is more common in the United States (Forsius et aL, 1998). In comea plana, the anterior chamber is shallow by virtue of the low comeal dome. Comea plana occurs in two forms: a severe autosomal recessive form (CNA2, OMIM 217300) and a milder autosomal dominant form (CNAl, OMIM 121400) (McKusick, 1994). Both types share many clinical features including reduction of the comeal power, 57 Introduction broadening of the limbus, frequent hyperopia and arcus lipoides at an early age (Tahvanainen, 1996). In the autosomal dominant type, visual acuity may be normal, the comeal parenchyma is clear and the diopteric power of the comea is nearly 3-7 diopters less than the normal value (normal diopteric power of the comea is 43 diopters) (Forsius et aL, 1998). On the other hand, in the recessive form a markedly reduced refractive power to about 20-30 diopters, widened limbal zone, smaller comeal diameter, strong hyperopia, shallow anterior chamber and rounded opaque central thickening nearly 5 mm in diameter are observed (Forsius et aL, 1998). The autosomal recessive form also features concurrent anterior segment abnormalities including iris coloboma, congenital cataract and occasional posterior segment coloboma. However the most distinctive feature of the recessive form is the presence of central comeal opacity (Forsius et aL, 1998). Histological findings in comea plana include acanthosis, kératinisation of the comeal epithelium, defects in Bowman's membrane and vascularisation of the superficial stroma (Tahvanainen er a/., 1995a). A flattened comea has been reported to occur secondary to other malformation of the eye, for example in patients with blue sclera, or as a part of Ehlers-Danlos syndrome (Villanueva et aL, 1997). It also features concurrent anterior segment abnormalities, including iris colobomas, congenital cataract, and occasional posterior segment colobomas. 58 Introduction 1.4.2 Microphthalmia Congenital microphthalmia is an ocular malformation characterised by reduced axial length of the globe (Figure 1.22) (Weiss et aL, 1989a; Weiss et aL, 1989b; Warburg, 1993) with a reported prevalence at birth of 1.5 per 10,000 (Kallen et aL, 1996). The condition is presumed to develop as a result of incomplete invagination of the optic vesicle into the optic cup, or defective closure of the optic vesicle, occurring between the second and the sixth weeks of gestation (Warburg, 1993). However not all cases of microphthalmia are genetic since environmental teratogens have been implicated in its development such as viral infection (rubella, cytomegalovirus, and toxoplasma) and drugs (for example alcohol, thialodomide and retinoic acid) (Warburg, 1993). The severity of the condition often ranges from mild to severe microphhalmia (anophthalmia) within a single family, suggesting variable gene expression (Warburg, 1993). Also, microphthalmia is often asymmetric or unilateral, suggesting that non genetic factors may also contribute to the phenotype (Warburg, 1993). Up to 80% of microphthalmia occur as part of syndromes that include systemic malformations, especially cardiac defects, microcephaly, and hydrocephaly. Chromosomal abnormalities of virtually all the chromosomes may result in syndromic microphthalmia, trisomy 13 being the most frequently observed (Warburg, 1993). The remaining percentage of microphthalmia cases represents the isolated form which is frequently associated with other ocular abnormalities-namely, cataracts, microcomea, sclerocomea, retinal coloboma, and optic nerve coloboma (Warburg, 1993). Most cases of isolated microphthalmia are sporadic however, a few families with autosomal dominât (Vingolo et aL, 1994; Othman et aL, 1998), autosomal recessive (Zlotogara et 59 Introduction aL, 1994; Bessant et aL, 1998) or more seldom, X-linked trait (Hoefhagle et aL, 1963) have been reported. 1.4.3 Sclerocomea Sclerocomea is an uncommon non-progressive congenital malformation of the anterior segment of the eye. Usually it applies to a peripheral, white, vascularised 1 to 2 mm comeal rim that blends with the sclera, obliterating the limbus and the scleral sulcus. The central comea generally is normal in appearance, and there is a fairly distinct transition between the two zones (Elliot et aL, 1985). The opacification affects full-thickness stroma and commonly severely limits visualisation of the posterior comeal surface and the intra ocular stmctures. Sclerocomea falls into four broad groups: isolated sclerocomea, sclerocomea plana, peripheral sclerocomea with anterior chamber cleavage abnormality and total sclerocomea in which the entire comea is opaque and vascularised (Figure 1.23). It has been reported as either an autosomal dominant (OMIM 181700) or recessive (OMIM 269400) (Elliot et aL, 1985). It has been described in association with isolated microphthalmia in three pedigrees (Pearce, 1991) and as a part of the MIDAS (microphthalmia, dermal aplasia, and sclerocomea) syndrome, which may be seen in females with an Xp22 deletion (Happle et aL, 1993). Histopathologically, absence of the Bowman's layer and greater diameter of the collagen fibrils in the deeper than the superficial stroma is a distinctive criterion between the sclerocomea and comea plana or normal comea (Kanai et aL, 1971; Sharkey et aL, 1992; Vingolo et aL, 1994). 60 Introduction 1.4.4 Genetics of cornea plana and microphthalmia The gene for autosomal recessive form of comea plana in Finnish population was assigned by linkage analysis to the approximately 3 cM interval between markers D12S82 and D12S327 on chromosome 12 (Tahvanainen et aL, 1995a, 1996) Based on linkage disequilibrium analysis, the critical region for CNA2 was further narrowed defining a critical interval 0.08-0.6 cM in length. Although linkage to this locus was initially excluded in two other families with autosomal dominant form, a Cuban family with dominantly inherited comea plana was mapped to the same small region to which CNA2 had been assigned (Tahvanainen et aL, 1996). Recently keratocan has been described as the disease gene for CNA2. No mutations have been identified in CNAI (Pellagata et aL, 2000). By linkage analysis the first locus for autosomal dominant nanophthalmos was identified on chromosome 11 (Othman et aL, 1998) and that for autosomal recessive on chromosome 14q32 (Bessant et aL, 1998). Another locus for autosomal dominant colobomatous microphthalmia has been mapped on chromosome 15ql2-ql5 (Morle et aL, 2000). Recently mutations in the retinal homeobox gene CHXIO that disturb its function have been identified in two families presenting with non syndromic microphthalmia and mutations in both alleles of the RX gene were identified in a patient with severe congenital microphthalmia (Voronina and Mathers, 2000). 61 , Introduction Figure 1.19 Keratoconus. a) Conical protrusion of the comea is manifest in the lateral view of the eye. A thin slit beam view of keratoconus. b) The comea is thinnest at the region of maximal protmsion. Figure 1.20 Keratoglobus. A diffuse thinning of the comea to 1/3 -1/5 normal thickness. The thinning is more pronounced in the periphery. Figure 1.21 Megalocomea. The figure shows large size of the comea. The comeal diameter is greater than normal range (14 mm). .5m m =1.0m m nitlTiTffîfîiiiirn Figure 1.22 Microphthalmos with cyst. The eye is small and malformed, and a cyst is contiguous with the globe. Figure 1.23 Sclerocomea. The comea may be totally opaque, as in the right eye of this patient, or there may be a central relatively clearer area, as seen in the left eye. Photographs are reproduced from Krachmer et aL, 1997 62 Introduction 1.5 Genetics of corneal dystrophy The genetic classification of comeal dystrophies via recognition of comeal disease gene will lead to a better understanding of the pathogenesis of those dystrophies that cause significant visual disturbances. The following table lists the main types of comeal dystrophies according to their gene loci, OMIM number, the disease gene and the mutation detected (Table 1.1). 63 Introduction Table 1.1 Genetics of major types of comeal dystrophy, their OMIM numbers, gene loci and the reported mutations Corneal dystrophy phenotype OMIM Gene locus Disease gene Mutations Number Epithelial dystrophies Meesmann’s dystrophy 122100 12ql3 Keratin 3 E509K 17ql2 Keratin 12 R135T,R135G, R1351,V143L, M129T, Y429D, L140R Map dot finger print dystrophy 121820 Bowman’s laver dystrophies Reis Bucklers’ dystrophy 121900 5q31 B1GH3 R124L, AF540, G623D Thiel-Behnke dystrophy 602082 5q3I, and R555Q 10q23-q24 Stromal dystrophies Granular comeal dystrophy type I 121900 5q31 B1GH3 R555W, R124S Lattice comeal dystrophy Lattice dystrophy type I 122200 5q31 BIGH3 R124C, L518P Lattice dystrophy type II 105120 9q34 Gelsolin D187N,D187Y, G654A Lattice dystrophy type III Lattice dystrophy type IDA 122200 5q31 BIGH3 P501T, A622H, H626A, A546T Lattice dystrophy type IV 5q31 BIGH3 L527R Avellino comeal dystrophy 121900 5q31 BIGH3 R124H Macular comeal dystrophy MCD type I frame shift after 217800 16q21 137A,K174R, D203E, E274K, CHST6 R211W, MCD type LA A128V,R166P, 707-708ins C165S, V66D, F67V R127C MCD type II R50C, replacement of 5' region, deletion of 5' region Fleck comeal dystrophy I2I850 Schnyder’s Crystalline dystrophy 121800 Ip34.1-p36 Endothelial dystrophy Fuchs’ dystrophy 136800 Congenital hereditary endothelial dystrophy Type I 121700 20pll.2-qIL2 Type n 217700 20pll.2-qlL2 Posterior polymorphous dystrophy 122000 20pll.2-qlL2 64 Introduction 1.6 Mapping single gene disorder by linkage analysis 1.6.1 History of linkage analysis Genetic linkage can be defined as the tendency for alleles close together on the same chromosome to be transmitted together, as an intact unit through meiosis. It was not recognised until nearly half a century after Mendel's pioneering work on the inheritance of seven characteristics in pea plants. In 1905 Bateson et a/., were the first to publish data on the inheritance of two characteristics that did not assort independently during gamete formation, but it was Thomas Hunt Morgan who interpreted such phenomenon in terms of exchanges between homologous chromosomes for which he used the term "crossing over" (Morgan and Catell, 1912). A year later, Sturtavent (1913) produced the first genetic map, using data on recombination frequencies on Dorsophila. The genetic recombination events or the chiasmata were interpreted by Janseens (1909) (McConkey, 1993) and then modified by Robin Holliday (1964). The first human autosomal linkage groups were identified by pedigree analysis in the 1950s and all of them involved one or another of the blood groups because they were the most polymorphic loci known at this time. 1.6.2 Linkage analysis The principal value in genetic linkage in human and medical genetics is to aid in identifying, mapping, and diagnosing the genes responsible for inherited diseases. The method rests on the concept that two genes located near to one another on the same chromosome will be inherited together, unless they are separated at meiosis by a recombination event. During germ cell formation, specifically during prophase of the first meiotic cell division, homologous pairs of chromosomes undergo synapsis. At this time, each chromosome consists of two chromatids joined only at their centromeres; thus 65 Introduction there are four chromatides, in each pair of chromosomes. While the chromosomes are synapsed, crossing over may occur between any two non-sister chromatids. Thus one crossover creates two recombinant chromatids and leaves two non-recombinant, giving 50% recombinants. The overall frequency of recombinant chromosomes in one individual or in a population is called recombination fraction (RF) or (0). If two genes on the same chromosome are close enough so that the RF is less than 50%, the two genes are genetically linked. However, if two loci are so far apart on the same chromosome that crossovers between them produces an RF value of 50%, such genes are genetically unlinked (assort independently). We cannot distinguish genes that assort independently because they are on different chromosomes from those that assort independently because they are far apart on the same chromosome. 1.6.3 Genetic and physical map distances The strength of linkage can be used as a unit of measurement to distinguish how close genetically different loci are to each other. This unit of map distance was suggested as a reflection of physical distance. However, the frequency of recombination is not constant along the length of a chromosome or throughout the genome. Thus genetic distance (measured as percentage recombination) and physical distance (measured in base pairs or chromosome bands) are different map parameters. When linkage is estimated between two syntenic loci the frequency of recombination expressed as a percentage over a given number of meioses defines the recombination fraction (0) between the two loci. For example, if 0 value is 0.1, it translates into ten recombination events in 100 meioses. The observed 0 value is converted into genetic distance (cM) by using a mapping function. Therefore a 0 value of 0.01 corresponds to a 66 Introduction genetic distance of IcM. In a physical term, a genetic distance of 1 cM corresponds to approximately 10^ bp of DNA. On average it is estimated that the genetic length of a chromosome will be approximately 100-300cM. Finally as a general rule the recombination events are expected to be 1-3 per chromosome per meiosis such that a chromosome inherited by a child from his father or mother is usually different from the two copies of that chromosome in the paternal or maternal genome. 1.6.4 Genetic markers 1.6.4.1 Requirements for genetic mapping There are two major requirements for carrying out a linkage analysis, (i) a family must be informative for the loci being considered and (ii) the particular alignment of alleles in the parent (the phase) must be known or able to be determined. Eleven meiosis are sufficient to give evidence of linkage if there are no recombinants, but 85 meiosis would be needed to give equally strong evidence of linkage if the recombination fraction was 0.3. Thus, mapping requires markers spaced at intervals no greater than about 20 cM across the genome. Useful marker must be highly polymorphic and can be scored easily and cheaply using readily available materials. The informativeness of a marker is indicated by its polymorphic information content (PIC). The PIC value of a marker is calculated from the number of alleles and their frequencies in the population and is related to the mean repeat length of the marker. To be of use in linkage analysis the PIC value of a genetic marker needs to be 0.7-0.8 (70-80% heterozygosity) (Botstein et a/., 1980). Thus, the PIC of a marker is given by the formula: PIC= 1- E"i=] Pp - E"i=i Z" 2Pi^Pj^ where Pi and Pj are population frequencies of the i^*^ and alleles. 67 Introduction 1.6.4.2 Restriction fragment length polymorphism (RFLPs) The first generations of DNA markers were restriction fragment length polymorphisms or RFLPs (Botstein et aL, 1980). They depend on DNA polymorphisms between individuals that can be detected as variations in the length of DNA fragments after digestion with sequence specific restriction endonucleases. The polymorphisms in a restriction enzyme cleavage site generated codominant alleles. These codominant alleles are a very useful property of marker loci, because heterozygotes can be distinguished from homozygotes. However, RFLPs suffer from limited informativeness. 1.6.4.3 Variable number of tandem repeats (VNTRs) or minisatellites Much higher levels of heterozygosity are associated with a different type of polymorphism that occurs in repeated sequences known as minisatellites (Jeffreys et aL, 1985). The term, variable number of tandem repeats, or VNTRs (Nakamura et aL, 1987) describes these polymorphisms. These minisatellites consist of tandem repeats of oligonucleotide sequences that are usually 10-60 bp long; the number of repeats varies from less than ten to several dozen. So, the occurrence of these markers is caused by the number of repetitions of the unit sequence between two invariant cleavage sites. However these markers suffer from the fact that they are difficult to handle by standard PCR protocols because large alleles may fail to amplify and they must be detected by hybridisation which is time consuming analytical process. Additionally they tend to be clustered in the subtelomeric regions of chromosomes. This leads to the discovery of microsatellites to overcome the limitations by VNTRs. 68 Introduction 1.6.4.4 Microsatellite markers (short tandem repeat polymorphisms [STRPs]) Microsatellites are another class of tandem repeats which are simple in sequence (1-4) bp and are randomly distributed throughout the genome. They may be mono, di, tri, or tetra nucleotide repeats. Of the mononucleotide repeats, runs of A and of T are very common and they account for 0.3% of the nuclear genome. By contrast, runs of C and of G are very much rarer. In the case of nucleotide repeats, arrays of the CA or TO repeats are very common, accounting for 0.5% of the genome and are highly polymorphic. CT/AG are also common occurring once every 50 kb and accounting of 0.2% of the genome, but CG/GC repeats are very rare. These dinucleatide repeat sequences are liable to replication slippage during PCR amplification, which make results difficult to read. So the tri and tetranucleotide repeats which are highly polymorphic are considered the markers of choice because they usually give clearer results with a single band from each allele. 1.6.4.5 Single nucleotide polymorphisms (biallelic markers) or SNPs SNPs are genetic markers which have two alleles hence the term biallelic, therefore a plus/minus assay is required for their genotyping. They vary in their polymorphism rates which is based on a single nucleotide change at the sequence level, that could range in frequency from 50% to 100% (Kruglyak, 1997). To overcome the lower polymorphism rates of SNPs the maps of biallelic markers need to be about 2.25-2.5 times the density of microsatellites to provide a comparable information content. The main advantages of SNPs over microsatellites is the amenability of their assay to automation and their great abundance, which is estimated on average to be one every kilobase pair. 69 Introduction With improvements in technology, screening of denser maps may prove to be useful for homozygosity mapping and to be cheaper than the use of microsatellites. 1.6.5 Linkage analysis for autosomal dominant diseases In dealing with an autosomal dominant disease particularly in a family with only one affected parent, the genotype of each child at the disease locus will be evident from the child’s phenotype. For each informative family in a linkage study, it is necessary to calculate the likelihood that the observed progeny genotypes have arisen from linked genes, with a given frequency of recombinants, L (RF) and the likelihood that the observed progeny genotypes could have arisen from unlinked genes (i.e. by independent assortment), L (0.5). If the ratio L (RF)/L (0.5) is more than 1, it indicates that linkage is more likely; if it is less than 1, then non linkage is more likely. For each informative family this odds ratio can be calculated and the overall odds can be obtained by multiplying all the individual odds together. However the method of lods by the use of logarithms rather than the raw odds ratio has more advantages (Morton, 1955). The lod score “Z (RF)” is equal to log [L (RF)/L (0.5)]. Conventionally a lod score of 3 (odds= 1000:1) is considered to be “proof’ of linkage, and a lod score of 2 (odds= 100:1) is “strong evidence” and a score of -2 is taken as proof that two loci are not linked at the RF value that leads to that score. 1.6.6 Linkage analysis for autosomal recessive diseases In autosomal recessive diseases the linkage analysis calculations are more complicated. This is due to the fact that dealing with an autosomal recessive disease, the genotype for the unaffected children at the disease locus is not known as it may be either homozygous 70 Introduction for the normal allele or heterozygous. This means that for double heterozygous parents, the unaffected offspring cannot be used to deduce the linkage phase. However, the presence of unaffected offspring in a pedigree involving a recessive genetic disease is not without value but provides less information than in dominant disease pedigrees. So, to demonstrate linkage between a marker locus and an autosomal recessive disease more individuals are required than for an autosomal dominant disease. In rare recessive diseases which do not provide enough material for standard linkage analysis homozygosity mapping can be used to demonstrate linkage (Lander and Botstein, 1987). The method requires DNA of affected individuals and polymorphic markers of high heterozygosity. Such individuals have a high probability of being homozygous for any given allele by descent and the rarer the genetic disease, the higher the proportion of affected individuals who are homozygous by descent. The detection of homozygosity by descent, by evaluating regions in which a contiguous stretch of markers are homozygous is now facilitated by the fact that dense maps of highly polymorphic markers are available. 1.6.7 Two point mapping to locate a disease locus Lod scores are the statistical measure of the evidence for linkage. The lod score (Z) was introduced by Morton (1955) as the logarithm of the odds that the loci are linked (with recombination fraction 0) rather than unlinked (recombination fraction 0.5). Lod scores are calculated by looking at each meiosis in turn and comparing the likelihood of the observed genotypes on the alternative hypothesis of linkage (with recombination fraction 0) or no linkage. Human linkage analysis, except in the very simplest cases, is 71 Introduction entirely dependant on computer programs (the Elston-Stewart algorithm) for handling branching trees of genotype probabilities given a pedigree structure and a table of gene frequencies. The best known programs for data handling are Cyrillic and MLINK (see section 2.8.2). 1.6.8 Evidence of genetic heterogeneity At the present time the most frequent use of linkage analysis is the mapping of genes responsible for diseases. Another important use is the detection of heterogeneity resulting from a clinical phenotype being caused by defects at two or more genetic loci. Certain diseases are caused by abnormalities in two or more unlinked or very loosely linked genes, so the observed RF in different families will sometimes be very different from the overall RF calculated for the rest of the families that have been analysed (Morton, 1956). The recognition of heterogeneity is essential in elucidating the molecular basis of a disease. 1.7 The Human Genome Project The Human Genome Project is an international project whose ultimate aim is to obtain a complete description of the human genome by DNA sequencing. The main goals of the Human Genome Project are generation of high resolution genetic maps, production of a variety of physical maps, determination of complete sequence of the human DNA and of the DNA of selected living organisms, development of support technologies for human genome research and expansion of communication networks and database capacities. The Human Genome Project represents a great source of information on the structure of the genome including microsatellite maps, physical maps and gene maps. Individual 72 Introduction gene function and regulation can often be postulated through information on gene structure and database comparison with gene families. With the development of mutation screening techniques, an expected benefit of the Human Genome Project would be to alter radically the current approach to medical care, from one of treating advanced disease to preventing disease based on the identification of individuals risk (Guyer and Collins, 1993). The draft sequence of the human genome including the full sequencing of chromosome 21 and 22 has been achieved (Hattori et al, 2000). Two publications of the human genome draft sequence were generated by the HGP (International Human Genome Sequencing Consortium 2001) and Celera Genomics (Venter et aL, 2001). The achievement so far has already produced a vast amount of information, at least there appears to be about 30,000-40,000 genes in the human genome, much less than the predicted 100,000 (Antequera and Bird 1994; Liang et aL, 2000). Moreover, the recent draft of the genome provided an estimate of 65,000-75,000 transcriptional units, with sequences containing 4% (Wright et aL, 2001). As part of the Human Genome Project, partial and complete gene sequences from multiple tissue types have been sequenced and deposited in databases such as dbest and UniGene (see http://www.ncbi.nlm.nih.gov/). These data, together with the draft human genome sequence can often place a candidate gene within a disease interval. This bioinformatic approach has now superseded. 1.7.1 Microsatellite maps The identification of a large number of genetic markers has led several authors to compile marker maps that span the genome. For example Généthon have produced 73 Introduction human genetic map based on linked (CA)n repeat markers which can be accessed through WWW-URL:http://www.genethon.fr/; FTP:Ap.genethon.fr. Another genetic map is CHLC (Co-operative Human Linkage Centre) in which collaborative human genetic map database containing genotypes, marker data and linkage server, are co ordinated. A number of these maps are presented in table 1.2. Table 1.2 Markers map of the human genome Research grouping Year of publication Number of markers Co-operative human linkage center 1994 1123 (CHLC) Genethon 1994 2066 CHLC/Genethon/Centre d’Etude du 1994 5840 Polymorphisme Humain (CEPH) Genethon 1996 5264 1.7.2 Identification of genes Four general strategies can be described for identifying disease genes, these are functional cloning, positional cloning, position-independant candidate gene approaches, and positional candidate gene approaches. The choice of the method for identifying the disease gene depends on the available resources and the information known about the pathogenesis of the disease. Initially candidate genes have to be identified and then tested individually for evidence that implicates them as the disease locus. In this study positional candidate gene approaches were used as a strategy for identifying the disease 74 Introduction gene. The method relies on the fact that confidence in a particular candidate disease gene is increased if it can be shown to map to the same subchromosomal region as the disease gene. With more and more human genes being assigned to subchromosomal locations, there is high chance that database searches will reveal one or more possible candidate genes in the appropriate location. 1.7.3 Mutation screening Mutation screening techniques are based on testing DNA samples from a panel of patients and control subjects. After amplifying the coding DNA obtained from genomic or cDNA, the products are then subjected to any of the mutation screening procedures. If the correct gene is tested, samples from patients will usually show a variety of different mutations and consequently deleterious effect on gene expression will result. Then the conclusion that the gene being tested really is the locus for the disease becomes inescapable, if the identified mutations are absent from the control samples. Methods for general mutation screening, other than simple sequencing, usually test for differences between the sequence under test and some standard sequence. Heteroduplex analysis is based on the fact that the electrophoretic mobility of heteroduplexes in polyacrylamide gels is less than that of homoduplexes, and they can be detected as extra slow moving bands. Insertions, deletions, single base substitutions can be detected by this method if fragments of under 200bp are tested (Keen et al\ 1991). Single-strand conformational polymorphism analysis (SSCP) is another method of mutation detection which is based on the fact that there is a tendency of the single stranded DNA to fold up and form complex structures stabilised by weak hydrogen bonds. The electrophoretic mobilities of such structures on nondenaturing gels will depend not only on their chain 75 Introduction lengths but also on their conformations (Scheffield et al; 1993). In some single gene disorders the responsible mutation eliminates a restriction enzyme recognition site. Although this approach is of paramount importance to prove the mutation identified by other methods, it is of limited use clinically as it is relatively rare for mutations to occur within the recognition sequence of restriction endonucleases. 1.7.3.1 Identifying disease-causing mutations A number of confirmatory steps must be taken to conclude that a sequence alteration in a gene is a disease causing. The change must cosegregate with the disease within the pedigree and must not be present in a panel of unrelated control individuals ascertained from an ethnically similar background. Convenient methods to demonstrate cosegregation include direct DNA sequencing, single strand conformational polymorphism (SSCP), denaturing gradient gel electrophoresis (DGGE) or heteroduplex analysis (Keen et al., 1991). In addition, mutations can result in the introduction or loss of a restriction enzyme binding site. Thus a mutation may be detected by digesting the relevant DNA sequence in the presence of enzyme. Finally, gene expression studies must confirm disturbed physiology of the protein of interest. 1.8 Aims of the study The work presented in this study is aimed at better understanding of comeal dystrophy, one of the blinding conditions in ophthalmology, which very often necessitates a graft. Unfortunately some grafts are unsuccessful and the patients require a second or a third and even a fourth surgery. Consequently a reasonable number of patients are registered blind and most of them are disabled by this ocular problem. 76 Introduction Identifying the disease gene is an important step and is required prior to the development of effective genotherapy, which could be the ultimate hope of the generations of many patients who inherited such a condition. Moreover, when the gene is known genetic testing of at-risk individuals will be possible; allowing early detection and treatment. This study included, (i) linkage analysis of a family with microphthalmia and cornea plana phenotype and identification of the underlying genetic defect, (ii) mutation screening of BIGH3 gene in a number of patients and families with 5q linked comeal dystrophies and (iii) CHST6 gene screening in 5 families with MCD. 77 Materials and Methods Chapter 2 Materials and Methods 2.1 Extraction of DNA 2.1.1 Extraction of DNA from peripheral blood lymphocytes DNA was extracted from venous blood in EDTA, based on a standard protocol (Sambrook et al., 1989). Blood samples were collected in 10 ml sterile tubes containing 1 ml, 0.5 M EDTA (ethylene diamine tetra-acetic acid) then either extracted immediately upon arrival in the laboratory or stored at -80 °C until required. A Nucleon extraction kit (Scotlab, Scotland) was used to carry out all extractions. Before DNA extraction, samples were kept at room temperature until completely thawed. For cell preparation from the whole blood the samples were transferred to 50 ml sterile falcon tubes and 40 ml of reagent A (0.05 M Tris-Hcl pH 8.0, 1.6 M Sucrose, 0.025 M MgCli, 5 % Triton X-100) was added. After mixing by inversion for 4 minutes at room temperature, the tubes were centrifuged at 1300 x g for 4 minutes and the resulting supernatant was discarded carefully leaving a lymphocyte cell pellet. For cell lysis 2 ml of reagent B (0.4 M Tris HCL pH 8.0, 0.06 M EDTA, 0.15 M NaCl, 1% SDS) was added to resuspend the pellet. The pellet was broken using a clean plastic Pasteur pipette and the lysate was transferred to 5 ml screw-capped polypropylene tubes. For deproteinization of the cells 500 pi of 5 M sodium perchlorate was added and mixed manually by inverting for at least 7 times. For DNA extraction 2 ml of chloroform was added and mixed for at least 5 times by manual inversion and then 300 pi of Nucleon resin was added and without re-mixing the phases the tubes were centrifuged at 1300 x g 78 Materials and Methods for 3 minutes. The upper phases containing the DNA were then transferred carefully to 7.5 ml fresh tubes and two volumes of cold absolute ethanol (100%) were added to precipitate the DNA. After mixing several times the DNA strands became visible and then picked out with a sterile needle to be transferred to 1.5 ml eppendorf tubes with the patients identification and number. They were then washed with 70% ethanol, allowed to dry, resuspended in a suitable amount of sterile distilled water (200-400 pi) depending on the size of the pellet and allowed to dissolve at 4 °C. These DNA stocks were stored at- 80 °C while 1/5 dilution was kept at 4 °C for routine usage. The concentration after resuspension was determined by electrophoresis of an aliquot in an agarose gel and comparing the resultant band intensity with aliquots of known concentration. 2.1.2 Extraction of DNA from buccal smears Buccal smears were collected by means of cotton swabs. The persons providing the samples should not consume any food or drink for at least 30 minutes prior to sample collection. The swabs were firmly scraped against the inside of each check 6 times and air-dried for at least 2 hours after collection. A QIAamp DNA Mini Kit (Qiagen, Ltd, UK) was used to carry out all extraction steps. The swabs were separated from their sticks with scissors and placed in 2 ml microcentrifuge tubes, then 400 pi PBS (Phosphate Buffered Saline), 20 ul Qiagen Protease stock solution and 400 pi Buffer AL were added. The samples were mixed immediately by vortexing for 15 seconds and incubated at 56 for 10 minutes, then briefly centrifuged to remove drops from inside the lid. 400 pi absolute ethanol was added to the previous mixture and followed by brief centrifuging at 6000 x g. 700 ul of the mixture was carefully applied to QIAamp spin columns with their 2 ml collection tubes then centrifuged at 6000 x g for 1 minute. The 79 Materials and Methods flow-through was discarded and the columns were placed in a clean 2 ml collection tubes. 700 pi of the remaining mixture from the last step, and 500 pi Buffer AWl were applied to the columns and centrifuged at 6000 x g for 1 minute. The flow-through was discarded and 500 pi Buffer AW2 was added to the columns and then centrifuged at 20,000 X g for 3 minutes. The flow-through was discarded and QIAamp spin columns were placed in a clean 1.5 ml microcentrifuge tubes. Prior to centrifuging at 6000 x g for 1 minute 150 pi Buffer AE was added to the columns to elute DNA. The eluted DNA was stored at -20 °C until required. Typically, one buccal swab yields 0.5- 3.5 pg of DNA in 150 pi of buffer (3-23 ng/pl). (PBS, AL, AWl, AW2, and AE Buffers provided by Qiagen, no formulation available). 2.2 DNA amplification by polymerase chain reaction (PCR) 2.2.1 The PCR reaction In 1985 Saiki et al. described an in vitro DNA enzymatic amplification technique known as polymerase chain reaction. In only a few hours this technique can isolate and amplify a specific segment of DNA by as much as 10^ fold. Its ability to amplify small amounts of DNA even from single cells has revolutionised the analysis of clinical and biological material. PCR has been applied to the high-efficiency cloning of genomic sequences, the direct sequencing of mitochondrial and genomic DNA, and the analysis of nucleotide sequence and the rapid detection of viral pathogens. PCR amplification involves two oligonucleotide primers that flank the DNA segment of interest and repeat cycles of heat dénaturation of the DNA, annealing of the primers to their complementary sequences and extension of the annealed primers with DNA 80 Materials and Methods polymerase I. The primers are designed to hybridise to opposite strands of the target sequence and are oriented so that DNA synthesis by the polymerase proceeds across the region between the primers, effectively doubling the amount of that DNA segment. Since the extension products are also complementary to and capable of binding primers, successive cycles of amplification continue to double the amount of DNA synthesised in the previous cycle. This results in an exponential accumulation of the DNA segment, approximately 2" where n is the number of cycles of amplification. Because the primers are physically incorporated into the ends of the extension product, they define the primary product of the reaction, a discrete fragment whose length is the distance between the 5' termini of the primers on the target sequence. The technological development of thermostable DNA polymerase isolated from the bacterium Thermus aquations (Saiki et at,, 1988) has substantially improved the performance of the procedure. Unlike the thermolabile Klenow fragment of E.coli DNA polymerase, Taq polymerase retains its activity after heat dénaturation of the DNA and does not need to be replaced during each cycle. In addition to amplifying the reaction the higher temperature optimum of this enzyme (70-75 °C) significantly increases the specificity, yield and length of DNA fragments that can be amplified. The technique is limited by the fact that the sequence flanking the DNA section of interest has to be known and be unique and the methods only amplifies efficiently up to 3 kb. PCR amplification of microsatellite repeat sequences is now the method of choice in linkage analysis. Several microsatellite products can be amplified together in one reaction and visualised as distinct allele systems on one polyacrylamide gel allowing for rapid linkage analysis using a number of systems simultaneously. 81 Materials and Methods 2.2.2 Standard parameters for PCR PCR was carried out using the standard parameters and any alterations were stated whenever necessary. PCR was carried out in a 50 pi reaction volume consisting of IX buffer (lOX NH4 buffer, Bioline), 1.5 mM MgCb (Bioline), 0.2 mM dNTPs (dATP, dCTP, dGTP and dTTP, Promega), 25 pmoles of each primer, 0.5 unit of Taq polymerase (Bioline) and -100 ng of DNA. One drop of mineral oil was added to the reaction mixture if using a thermal cycler without a heated lid to prevent evaporation. The thermal cycling steps were carried out using a Hybaid Omnigene Thermal Cycler. Incubation was carried out in three cyclical steps: initial dénaturation step at 94 °C for 3 minutes, followed by 30-35 cycles of dénaturation at 94 °C for 1 minute, annealing for 1 minute at the optimum temperature and extension at 72 °C for 1 minute and final extension step at 72 °C for 5 minutes. The optimum annealing temperature depends on the length of the primer and its GC content and it was determined to be 1 °C or 2 °C less than the lower melting temperature (Tm) of either of the two primers or can be calculated depending on the nucleotide sequence using the following equation Tm= 4(C+G)+2(A+T). 2.2.3 Primer design Primer3 Output (www-genome.wi.mit.edu/cgi-bin/primer) program was used for designing all primers in this thesis. PCR primers should be generally between 18-30 nucleotides long and have a 40-60% GC content in order to optimise the chances of successful PCR. All primers designed in this thesis conformed to these criteria. Primer pairs were designed to have similar Tm values. An estimate of annealing temperature of 5 °C below the estimated Tm was used as starting point for PCR 82 Materials and Methods optimisation. Base complementarity of the two bases at the extreme 3’ end of the two primers should be avoided. Otherwise primer dimers can result, reducing amplification efficiency. Primers were synthesised commercially by Cruachem or GenoSys. 2.3 Fractionation of DNA by gel electrophoresis 2.3.1 Agarose gel electrophoresis This method was used to visualise PCR products, restriction enzyme digests and to estimate DNA concentrations. PCR products were first checked by electrophoresis on a 1-3% agarose gel. The appropriate agarose concentrations used for separating DNA fragments of various sizes are shown in table 2.1. Table 2.1 Agarose gel concentrations for separating DNA fragments Agarose [%(w/v)] Range of resolution of linear DNA (kb) 0.3 5.0-60 0.6 1.0-20 1.0 0 .5-10 1.5 0.2 - 6.0 2.0 0.1 -2.0 3.0 0.05-<0.1 2% agarose gel electrophoresis was prepared according to the following protocol. 2 gm of agarose (Biorad) was weighted in a clean bottle and 100 ml of Ix TAB buffer (tris acetate EDTA) was added. The mixture was heated in a microwave oven on full power for 2 minutes and then cooled to about 50 °C before adding ethidium bromide to a final 83 Materials and Methods concentration of 0.5 ug/ml. The cooled mixture was poured into a sealed loading tray with a comb in place and allowed to set for 30 minutes to polymerise. Then the seals and the comb were removed from the gel and it was placed in an electrophoresis tank (Gel electrophoresis apparatus GNA-200, Pharmacia, Sweden) containing sufficient Ix TAE buffer, with the wells positioned near the cathode. DNA samples were prepared by adding an appropriate amount of lOX loading dye (2 pi of dye for 20 pi samples) and were loaded into wells, with the addition of an appropriate size marker (0xl74/HaeIII, advanced Biotechnology's Ltd) into one well. Electrophoresis was carried out at 100 V for 45 minutes or until the required resolution had been achieved. Gels were photographed on an UV transilluminator using a Polaroid MP4 Instant Camera System with an orange filter and Kodak plus-X film. 2.3.2 Non denaturing polyacrylamide gel electrophoresis The non-denaturing polyacrylamide gel electrophoresis is used for resolution of non radioactive microsatellite PCR products in genetic linkage analysis. The concentration of acrylamide that should be prepared to provide maximum resolution of single stranded DNA fragments is shown in table 2.2. 84 Materials and Methods Table 2.2 The Acrylamide concentrations prepared to provide maximum resolution of single stranded DNA fragments. Acrylamide (%) Size range resolved Comigration of Comigration of (bp) Xylene cyanol bromophenol Blue marker (bp) marker (bp) 3.5 500-1000 230 50 5.0 40-250 130 32 8.0 30-200 80 22 12.0 20-100 35 10 15.0 12- 75 30 8 20.0 3- 50 22 6 The two gel plates (40x60 cm) were washed with detergent prior to use and their inner aspects were wiped with 100% ethanol. The back plate containing the buffer reservoir was then coated with a seleconising agent - dichlorodimethyl - sialine (sigmacoat, Sigma). The spacers (2x0.4 mm) were placed appropriately before assembling the gel plates. Then, the assembled gel plates were clamped together and placed horizontally ensuring that they were well balanced. 6% gels were used routinely, 200 ml of 6% gel solution was prepared with 130 ml distilled water, 50 ml acrylamide solution (Protogel, EC890), 20 ml lOX THE, 700 ul 25% ammonium persulphate (APS), and 80 pi of tetramethylethylenediamine (TEMED, Sigma). Once the TEMED was added the mixture was poured immediately between the two glass plates using a 60 ml syringe avoiding the formation of air bubbles. A 60 well comb was inserted into the top of the gel to form the 85 Materials and Methods sample lanes. The gel was left to set for 45-60 minutes. Then, the polymerised gel was placed vertically and fixed in the buffer tank after removing the casting tray. The buffer reservoir of the apparatus was filled with 2 litres of IX TBE and the apparatus connected to the power supply. The gel was pre-run for 30 minutes at 100 W before removing the comb and flushing the wells. 5 ul of the PCR products was added to 3 ul of 15% ficoll dye (15% ficoll, 0.15% Xylene cyanol, and 0.15% bromophenol blue) and then loaded into the wells together with ladder (equal volumes of 0x174 Haelll and the ficoll dye) to mark sample order. Electrophoresis was carried out at 100 W for the required length of time. The duration of run varied from 3-4 hours depending on the size of DNA fragment to be resolved. Bromophenol blue and Xylene cyanol both of which are present within the loading dye act as indicators of resolution (see table 2.2). On run completion the plates were separated and the gel was cut appropriately and transferred very carefully into a tank with 500 ml IX TBE and 3-6 drops of ethidium bromide were added. Staining was allowed to continue for 20-30 minutes for better resolution of bands. Gels were photographed on an UV transilluminator using a Polaroid MP4 camera with an orange film and Kodak plus X film. 2.4 Purification of DNA 2.4.1 Phenol chloroform extraction and ethanol precipitation (a) Phenol chloroform extraction This method removes impurities such as proteins and salts from DNA solutions. For small volumes of DNA (such as restriction enzyme digests) the extracted volume was increased to 200 pi by the addition of sterile IX TE to prevent DNA loss. Large volumes 86 Materials and Methods of DNA were extracted directly. An equal volume of phenol chloroform isoamyl (25: 24: 1) was added and mixed by inversion prior to centrifugation at 6000 rpm for 3 minutes. The top aqueous layer was then carefully removed to a clean 1.5 ml eppendorf and an equal volume of chloroform was added, mixed and centrifuged as before. The phenol extraction was repeated 2-3 times prior to the chloroform stage if the protein content of the sample was high. After the chloroform extraction the top DNA containing aqueous layer was removed into a clean 1.5 ml eppendorf and subjected to ethanol precipitation. (b) Ethanol precipitation Two volumes of cold absolute ethanol and 1/10 th volume of 3 M sodium acetate was added and the DNA solution was placed at -80 for 30 minutes to allow DNA precipitation. (The addition of salt and freezing at -80 ^ C to facilitate DNA precipitation was only performed when purifying small quantities of DNA such as restriction digests of DNA, and PCR products). After freezing, the DNA precipitate was centrifuged at 10,000 X g for 15 minutes to pellet the DNA. The pellet was rinsed in 70% ethanol then air-dried and dissolved in the required volume of sterile distilled water for subsequent use. 2.4.2 Use of Sephacryl microspin columns (Sephacryl-S400 HR columns, Pharmacia, UK) These columns were routinely used for purification of PCR products by removal of excess primers, unincorporated dNTPs, and “primer dimers” prior to sequencing. The Sephacryl ® HR resin (sephacryl equilibrated in TE buffer, pH 7.6) was first resuspended in the column by tapping. The screw cap of the column was then loosened 87 Materials and Methods and the base snapped before placing the column inside an open 1.5 ml eppendorf and centrifuging at 3000 rpm for 1 minute. The column was removed to a clean open eppendorf and 30 pi of DNA sample was applied to the top centre of the compacted bed before centrifuging for 1 minute to elute out the DNA. 2.4.3 Use of Qiagen gel extraction/PCR purification Kits The QlAquick spin columns (Qiagen, Ltd, UK) were used to purify single or double stranded PCR products ranging from 100 bp to 10 kb from primers, nucleotides, polymerases and salts prior to sequencing. For purification of DNA form the PCR reaction, 5 volumes of Buffer PB, which provide the correct buffer conditions for DNA binding to the membrane, was added to 1 volume of the PCR reaction and mixed. The mix was then applied to the QlAquick spin columns and centrifuged at 10,000 x g for 30-60 seconds. The flow-through was discarded and 750 pi Buffer PE was added and centrifuged at 10,000 x g for 30-60 seconds. Additional centrifugation at maximum speed for 1 minute is recommended after the flow-through was discarded to get rid of the residual ethanol from the Buffer PE. The columns were then placed in a clean 1.5 ml eppendorf and prior to centrifuging for 1 minute at 10.000 X g 30 pi elution buffer was added to the centre of each column to elute the DNA (the average elution volume was 28 pi). The eluted DNA was kept at -20 °C to avoid DNA degradation in the absence of a buffering reagent until required. (PB, PE and elution buffers provided by Qiagen, no formulation available) 88 Materials and Methods 2.5 Restriction enzyme digests of DNA Restriction digests were performed on genomic DNA and on PCR products. For PCR product digests, a total volume of 20 pi containing 2 pi lOX appropriate buffer, 0.2 pi IX Bovine Serum Albumin (BSA) to enhance the enzyme activity, 10 pi DNA (10 pg), 0.5- 1.0 pi of the appropriate enzyme (5- 10 units) and 7.3 pi distilled water was used. The reactions were then incubated for at least 2 hours or overnight at the specified temperature for the enzyme (usually 37 °C). After heat inactivation the products of the restriction digest were visualised by agarose gel electrophoresis (section 2.3.1) or by non denaturing polyacrylamide gel electrophoresis (section 2.3.2). Enzymes were obtained from Promega (UK), and New England Biolabs (UK). 2.6 Mutation detection techniques 2.6.1 Heteroduplex analysis method Mutation detection by heteroduplex using denaturing gel electrophoresis is based upon conformational differences that occur in the DNA molecule as a result of insertions, deletions, or single base pair mismatches. After standard PCR amplification, of both the normal and mutated allele, the PCR products were denatured and allowed to reanneal by leaving them to cool down slowly at room temperature. Hetroduplexes are formed when a normal DNA strand anneals with a mutated complementary strand. These heteroduplexes migrate at a lower rate on acrylamide than the corresponding homoduplexes. Heteroduplexes optimum resolution is achieved with 100-400 bp PCR fragments. First the glass plates (24x18 cm, Hoefer, UK) were washed with soap and water, dried and 89 Materials and Methods wiped with 100% ethanol. They were vertically assembled in sets of two, consisting of two outside plates and one inner plate set and then clamped within the casting tray with the spacers 1.00 mm thick in-between. 100 ml of gel solution was prepared by the addition of 50 ml MDE gel solution (J.T Baker, USA), 6 ml lOX TBE, and 40 ml of sterile distilled water to 15 gm of urea. Prior to the addition of APS and TEMED, 1.5 ml of the gel solution was transferred to 1.5 ml eppendorf, to which 10 pi of 10 % APS and 5 pi of TEMED were added then, poured into each plate to form a plug at the base. 440 pi of 10% APS and 44 pi of TEMED were added to the remaining gel solution mixed and poured between the gel plates avoiding air bubble formation. An appropriate comb (20 well 1mm Hoefe) was inserted and clamped into place and the gel was allowed to set for 1 hour. The comb was removed after gel polymerisation and the wells rinsed thoroughly in IX TBE. The upper and lower reservoir were filled with the buffer (IX TBE) and 10 pi of DNA sample containing 100-200 ng of PCR product in 5 pi of green loading dye (40% sucrose, 0.25% Orange G, 0.25% Xylene cyanol and 0.25% bromophenol blue dye) was loaded into each well. Two lanes were reserved for the size marker 0X174 Haelll and positive control containing a known heteroduplex DNA. The gel cassette was mounted on the electrophoresis apparatus and allowed to run for 16-20 hours at 150 V (10-15 mA), Xylene cyanol and bromophenol blue that comigrates with a 230 bp fragment were used as an indicator. When the run was completed, the power was switched off and the plates were separated. The gels were stained with ethidium bromide (0.5 pg/ml) for 10-15 minutes and UV photographed. 90 Materials and Methods 2.6.2 Single strand conformation polymorphism analysis (SSCP) SSCP analysis enables the detection of DNA mutations in PCR products based upon the observation of mobility shifts caused by mutation induced changes of tertiary structure of the single stranded DNA. The sensitivity of SSCP depends on the temperature and the ionic environment as these physical conditions are the most important for the DNA single stranded tertiary structure to show the appearance of new bands in autoradiograms. The plates were prepared as above, then 100 ml non denaturing polyacrylamide gel was prepared by the addition of 10 ml, 50% glycerol, 33 ml protogel, 6 ml lOX TBE and 51 ml of sterile distilled water. Prior to the addition of APS and TEMED 1 ml of the gel solution was mixed with 10 pi 10% APS and 5 pi TEMED and poured between the plates to form a plug at the base. 400 pi of 10% APS and 40 pi of TEMED were added to the remaining gel solution, mixed and poured between the gel plates avoiding the formation of air bubbles. After 20 well comb was inserted and clamped into place the gel was allowed to polymerise for 2 hours. The comb was then carefully removed fi'om the gel and the wells were washed thoroughly with IX TBE using a syringe. The gel cassette was mounted on the electrophoresis apparatus and sufficient IX TBE was added to the upper and lower buffer compartments. 1.5 pi of the PCR products was added to an equal volume of SSCP dye (800 pi of (95% deionized formamide, 0.05% bromophenol blue, 0.05% Xylene cyanol, 20 mM EDTA}+200 pi 100 mM NaOH). 3 pi of the sample dye mix was heated up to 95 °C for 5 minutes to denature them and suddenly cooled down by putting on ice to prevent re-annealing of the DNA strands and then 2-3 pi of the denatured samples was loaded into the gel. The gel was allowed to run at 100 V (5 mA) in cold room for 20-24 hours for the best resolution. When the run was completed the power was switched off and the plates were 91 Materials and Methods separated. For silver staining the gel was transferred very carefully into a tank containing 500 ml (10% ethanol, 1% acetic acid) to be fixed for 10 minutes and then washed with distilled water. The gel was then transferred to silver staining solution (1 gm silver nitrate/500 ml distilled water) to be stained for exactly 30 minutes then washed with distilled water. Lastly, the gel was transferred into a tank with the developing solution (15 gm NaOH, 5 ml formaldehyde, and 995 ml distilled water) to be stained until the banding pattern was resolved and then washed with water. The fixed gel was then transferred onto a 3 MM Whatmann paper, wrapped in cling film and dried under a vacuum of 80 °C for 1-2 hours and then stored to be scanned. 2.6.3 DNA sequencing This method of mutation detection is based on a laser detection system of flourescently tagged dideoxyterminators in a cycle sequencing reaction. The detected florescence is presented as a graphical image on the computer. The ABI 373a DNA sequencer (Perkin Elmer, USA) was used routinely in this study. PCR was performed and the products cleaned up using phenol chloroform extraction, 8400 microspin columns, or Qiaquick spin columns as described in sections 2.4.1, 2.4.2, and 2.4.3. Cycle sequencing was performed in a Perkin - Elmer Cetus 2400 or 9600 PCR machine. The reaction was carried out in 0.2 ml microfuge tubes. 10 ul reaction volume was prepared containing 4 pi ABI PRISM Dye Terminator Cycle Sequencing Ready Reaction Mix (Big dye), 1 pi 4 pM forward or reverse primer, 4 pi purified PCR product and 1 pi distilled sterile water. The tubes were well capped and placed in the Perkin-Elmer machine and the PCR was carried out on a specific ABI program. The temperature cycling profile consisted of 25 cycles of dénaturation at 96 ®C for 10 seconds, 50 °C for 5 seconds, and 60 °C for 4 92 Materials and Methods minutes. Upon completion of the program the resultant products were transferred to a sterile 0.5 pi eppendorf to remove residual dye terminators by the standard ethanol precipitation method. The final reaction volume was completed to 20 pi with deionised water, then 16 pi deionised water and 64 ul non denatured 95% ethanol were added to the 20 pi reaction volume at room temperature. The tubes were capped, vortexed briefly and left at room temperature for 15 minutes to precipitate the extension products. The capped tubes were centrifuged at the maximum speed for 20 minutes after marking their orientation. Immediately after centrifugation the supernatants were carefully aspirated and discarded using a pipettor without disturbing the pellet. 250 pi 70% ethanol was added to the pellet then, the tubes were capped, vortexed and centrifuged at the maximum speed for 5- 10 minutes. The supernatants were carefully aspirated and discarded and the pellets were dried in a vacuum centrifuge for 10-15 minutes or placed in a heat block at 90 for 1 minute. At this stage the samples were used immediately or kept at -20 °C for a maximum period of seven days prior to analysis. Immediately prior to loading on the ABI gel, the pellets were resuspended in 3 ul of ABI loading buffer (deionised formamide and 25 mM EDTA pH 8.0 containing 50 pg/ml blue dextran in a ratio of 5:1). These samples were then denatured at 95 °C for 5 minutes and snap frozen on ice prior to loading. A polyacrylamide gel was prepared. First the plates (35x24 cm) were washed with hot water only, rinsed with distilled deionised water, dried with one wipe o f a white kimwipe and assembled horizontally with 0.4 mm spacers. In a clean plastic container 50 ml of the 6% acrylamide gel was made as follows: 40 ml sequagel 6, 10 ml sequagel complete (National diagnostics) and 0.04 gm ammonium persulphate. The solution was poured between the gel plates with the aid of a syringe. The comb was placed and 93 Materials and Methods clamped and the gel was allowed to set for 2 hours. Once the gel had polymerised, the comb was removed very carefully to avoid injury of the gel front. The plates were cleaned with tap water, then distilled water and left to dry prior to placing them within the electrophoresis tank. A plate check was done, if this was satisfactory the plate heater and acrylic bar were connected and the buffer tank fixed. 1000 ml IX TBE was added to fill top and bottom buffer tanks then the electrodes were connected. A pre-run of 15 minutes was carried out. Prior to loading, a 48 or 64 well sharks toothcomb was appropriately positioned into the gel and the wells were rinsed by a syringe containing IX TBE to clean them from urea. The necessary settings for the run (12 hours, 2500 V, 40 mA and 25 W) and software for data collection were set up and the run was started. The ready ice kept samples were then loaded into alternate wells, run in for 5 minutes and then the remaining wells loaded. Analysis of the data was performed using an Apple Macintosh computer with software linked to the sequencer (ABI Prism). The output consisted of two files, a text file containing the read sequence and an analysis file containing the raw data, analysed data and detailed sequence information. The sequence quality could be judged by electropheogram data present within this file. 2.7 Assay of sulfated keratan sulfate (KS) in serum 2.7.1 Specimens 10 ml venous blood was obtained fi-om the probands of all families with macular comeal dystrophy (MCD) participated in this study. Serum was obtained by centrifugation and stored at -20 °C. 94 Materials and Methods 2.7.2 Antibodies The anti-keratan sulfate antibody, 5-D-4, was obtained from ICN Biomedicals, Ltd. A secondary antibody, goat antimouse IgG 5 nm gold conjugated (British Biocell International, UK) was used to visualise the primary antibody. 2.7.3 Inhibition—Enzyme-Linked Immunosorbent Assay (ELISA) for keratan sulfate detection Inhibition enzyme-linked immunosorbent assay (ELISA) was carried out on serum samples to determine whether keratan sulfate was present. A 96-well ELISA plate was coated with Aggrecan (Sigma Chemical Co.) and was left to incubate at room temperature overnight. The inhibition mixture, consisting of duplicates of serial dilutions of each of the samples, plus the primary antibody to keratan sulfate (5-D-4), was incubated at room temperature for I hour and then overnight at 4 °C. 5-D-4 was used at a concentration of 1:5000 in a diluent phosphate buffered saline (PBS) buffer containing 1% bovine serum albumin (BSA), 0.05% Tween 20, and 0.05 M EDTA, pH 5.3. A standard also was prepared consisting of serial dilutions of a keratan sulfate fraction extracted from bovine articular cartilage. The next day the plate was washed automatically with PBS buffer containing 0.05% Tween 20 pH 6, and coated with the diluent buffer for 1 hour at room temperature to block non specific binding. The inhibition mixtures were allowed to reach room temperature and were added to the washed plate after the blocking step. The plate was incubated at room temperature for 10 minutes before washing. The secondary antibody, goat anti-mouse peroxidase IgG (Sigma), in the diluent buffer, was incubated in the plate for 1 hour at room temperature. The plate was again washed and the enzyme substrate, 95 Materials and Methods ABTS (2, 2’-Azino-di-[3-ethylbenzthiazoline sulphonate 6] Roche Molecular Biochemicals), was added to the plate and left for 10 minutes at room temperature to produce a measurable colour change. The reaction was killed by the addition of sulfuric acid and the plates were read automatically at 410 nm with a reference wavelength of 490 nm on a ICN Flow plate reader. 2.8 Computer aided analysis 2.8.1 Computational analysis of DNA sequence Analysis of nucleic acid sequences was mainly performed using Geneworks (version 4.45). This software developed for Apple Macintosh computer, was used to align sequences of different PCR products in order to identify overlapping regions and also to identify restriction sites in sequences. BLAST/BLASTN/FASTA programs (Altschul et al., 1990) accessible through internet site HGMP, were consistently used to compare sequences with entries in the GeneBank database for homologies. This was of marked importance when designing primers, since primers with homologies to repetitive sequences had to be avoided, primer sequences that did not have any homology to sequences present in the database most often proved to bind only at the desired target region. 2.8.2 Linkage computer programs (Cyrillic and MLINK) Markers used for linkage analysis in this study were analysed using Cyrillic software. Each pedigree and its disease allocation was entered and copies were made of these family files. Genotype information for each marker was then entered on each individual sampled. Locuslib and phenolib files containing the information on the phenotype and 96 Materials and Methods microsatellite marker characteristics were made (LINKSYS) (Attwood and Bryant, 1988). FED and DAT files were created and exported to the LINKAGE subdirectory. 2.9 Buffers and solutions TAE (Tris acetate EDTA) 5OX TAE stock solution 1210 Tris base 285.5 ml acetic acid 500 ml 0.5 M EDTA, pH 8.0 TBE (Tris borate EDTA) lOX TBE stock solution 108.0 gm Tris base 55.0 gm boric acid 8.3 gm Na: EDTA TE buffer 10 mM Tris-HCl (required pH) 1 mM EDTA (pH 8.0) PCR buffer (lOX) 100 mM Tris-HCl (pH 8.4) 500 mM KCl 15 mM MgCL Phenol chloroform isoamyl 25 ml equilibrated phenol 97 Materials and Methods 24 ml chloroform 1 ml isoamyl alcohol 2.10 Electronic data base information GenBank: http//www.ncbi.nlm.nih.gov/rentrz Genethon: http://www.genethon.fr/ Human Genome Mapping Project (HGMP): http://www.hgmp.mrc.ac.uk Human Genome Database (GDB): http://www.gdb.org Online Mendelian inheritance in Man (OMIM): http://www.ncbi.nlm.nih.gov/Gmim/ Research Genetics (http://www.resgen.com) Whitehead (http://www.ncbi.nlm.nih.gov) www-genome.wi.mit.edu/cgi-bin/primer UniBLAST: http://gcg.tigem.it/INlBLAST/uniblast.html 98 Results Chapter 3 A novel mutation in keratoean causes autosomal recessive cornea plana and microphthalmia 3.1 Introduction 3.1.1 Cornea plana Cornea plana is a rare condition in which there is a degree of flattening of the normally convex comeal surface (Forsius et aL, 1998). It is inherited either as an autosomal dominant (CNAl, OMIM 121400) or autosomal recessive trait (CNA2, OMIM 217300) (McKusick 1994; Tahvanainen et al., 1995a). The autosomal recessive form of the condition exhibits the more severe phenotype and is accompanied by a degree of comeal opacity and sclerocomea (OMIM 181700) (Elliot et al., 1985). Both forms of comea plana share many clinical features including reduced comeal curvature leading to hypermetropia, indistinct comeal limbus and arcus lipoides at an early age. However, in addition to the mode of inheritance, the diseases may be distinguished by the more severe symptoms of the recessive form, particularly a rounded and opaque central comeal thickening with a diameter of up to 5 mm (see section 1.4.1). The recessive form may also occur in conjunction with other malformations in the eye or elsewhere in the body (see section 1.4.1). Although recessively inherited comea plana is a rare disease in most countries, Finland forms an exception, with the majority of all reported cases and the largest pedigrees originated from. (Eriksson et al., 1973; Tahvanainen et al., 1995a; Tahvanainen et al., 1995b; Tahvanainen et al., 1996). The reason why comea plana is relatively common. 99 Results particularly in northern Finland, is that it was populated about 2000 years ago by small groups of settlers who interbred rapidly, resulting in unmasking of genetic mutations, including those leading to cornea plana (Forsius et al., 1998). As a result more than 30 hereditary diseases including cornea plana, which are rare in other countries, are relatively common in Finland (Norio et ah, 1973; de la Chapelle, 1993). The phenotype of the affected people is classical (see section 1.4.1), with extremely low comeal refraction due to flattening of the ocular curvature as the most prominent sign. The pathogenesis of cornea plana is unknown, but the embryological development of the cornea may give some clues (section 1.2.4). At the sixth embryonic week neural crest cells differentiate to form a large portion of the anterior segment of the eye. Crest cells and mesoderm migrate into the developing eye in three waves after the basement membrane of the surface ectoderm and the lens vesicle separate. The first wave of crest cells differentiates to form the comeal endothelium. The cells of the second wave contribute to the development of the iris and the third wave gives rise to the comeal keratocytes (Bahn et ah, 1984). Abnormalities arising from either faulty migration or function of the first wave of neural crest cells have been hypothesised as a cause of comea plana (Hittner et ah, 1982; Bahnet ah, 1984; Salmon et ah, 1988). 3.1.2 Genetics of cornea plana 3.1.2.1 Autosomal dominant inheritance In all described autosomal dominant (CNAl) cases the disease is mild compared to recessive forms. The comeal power varied from 34 to 40 diopters (emmetropic range of comeal power is from 38 to 48 diopters). The comea is clear, but the limbal zone is broad and the horizontal comeal diameter varies from 10 to 11.5 mm (normal horizontal 100 Results comeal diameter 11.75 mm, normal vertical 10.6 mm). The condition was first described in a woman and her two daughters by Barkan and Borley (1936). Later, two large pedigrees with a total of 13 affected members of both sexes, were reported (Larsen and Eriksen, 1948). 3.1.2.2 Autosomal recessive inheritance This form of inheritance produces a more severe form of the disease (CNA2). The first report was in 1909 by Broekema in his dissertation on hyperopia. A family was presented with three affected brothers whose parents had normal corneas. The mode of inheritance thus appeared to be autosomal recessive (Eriksson et al., 1973). 3.1.2.3 Identification of the gene for cornea plana At the start of this study the gene for comea plana had not been identified, however, both the autosomal dominant and recessive forms of the disease in the Finnish population had been linked to chromosome 12. The autosomal recessive form of the disease was first assigned by linkage analysis to an approximately 10 cM interval between markers D12S82 and D12S327 on chromosome 12q21 (Tahvanainen et al., 1995a). Further analysis using the current Genethon map, has refined this interval to approximately 3 cM (Tahvanainen et al., 1995b). Subsequently, linkage disequilibrium studies further narrowed the critical region for CNA2 to within 0.04-0.3 cM of marker D12S351, defining a critical interval of 0.08-0.6 cM (Tahvanainen et al., 1995b). Linkage to this locus was initially excluded in two Finnish families with the autosomal dominant form of the disease (Tahvanainen et al., 1996). Later, a Cuban family with dominantly inherited comea plana was linked to the same small region on chromosome 101 Results 12q21 to which CNA2 had been assigned. From these studies, it was concluded that there must be at least two genes causing autosomal dominant comea plana (Tahvanainen et al., 1996). Recently KERA, encoding keratocan, on chromosome 12 has been identified as the gene for CNA2. Two different mutations (N247S and Q147X) have been identified in a Finish family and an American patient respectively with CNA2 (Pellegata et al., 2000). To date no gene has been identified for CNAl. 3.1.3 Microphthalmia The work described includes the study of a Bangladeshi family with a combination phenotype of shortened axial length and small globe in addition to comea plana. The phenotype and genetic characteristics of microphthalmia are thus described. 3.1.3.1 Phenotype of microphthalmia Microphthalmia is a rare heterogeneous group of congenital disorders in which the axial diameter of the eye is reduced to less than the age-adjusted 95^ percentile (Warburg, 1993). The condition occurs either as an isolated ocular form (also known as nanophthalmos) or in conjunction with a diverse range of systemic malformations examples of which include mental retardation (OMIM 251500), oesophageal atresia (OMIM 600992) and dwarfism (OMIM 309700). The isolated form is frequently associated with other ocular abnormalities (see section 1.4.2). The prevalence of all types of microphthalmia in the Caucasian population is estimated to be between 1.2 - 1.8 cases per 10,000 births (De Wals and Lechat, 1987; Stoll et al., 1992). 102 Results 3.1.3.2 Genetics of microphthalmia Autosomal dominant, autosomal recessive and X-linked inheritance patterns have been reported for microphthalmia (Duke-Elder, 1964; Kimbrough et al., 1979; Singh et al., 1982; Vingolo et al., 1994). The first locus to be identified was for autosomal dominant microphthalmia/nanophthalmia (NNOl, OMIM 600165) on chromosome lip (Othman et al., 1998). Subsequently, the locus for autosomal recessive microphthalmia (arCMIC, OMIM 251600) has been localised to chromosome 14q32 (Bessant et al., 1998). Another form of autosomal dominantly inherited microphthalmia, colobomatous microphthalmia (adCMIC, OMIM 600165), has been mapped to chromosome 15ql2- ql5 (Mode et al., 2000). Recently mutations in the retinal homeobox gene (CHXIO, OMIM 142993) have been identified in two families presenting with non-syndromic severe microphthalmia. (Percin et al., 2000). 3.1.4 Small leucine-rich proteoglycans (SLRPs) Small leucine-rich proteoglycans (SLRPs) gene family comprises at least nine members which, although structurally related have evolved from different genes, acquired unique functions and have undergone a significant degree of structural sophistication. The proteoglycans encoded by these genes represent a number of structurally related proteins, SLRPs, that are typically components of the interstitial matrix of the comea. Members of this family share a series of 24-amino acid leucine rich repeats (LRRs) that make up the central portion of each protein. They perform functions as diverse as maintaining the mineralised matrix of bones and teeth, the transparency of the comea, the tensile strength of the skin and tendon and the viscoelasticity of the blood vessels. 103 Results Transparency of the comea, according to the lattice model of Maurice (1957), is based on the presence of collagen fibrils of small and regular diameter (22.5-35 mm) spaced at regular intervals (42-44 nm). The SLRPs have long been thought to be key regulators of both fibril diameter and interfibrillar spacing during collagen fibrillogenesis (Rada et al., 1993). It has been proposed that this results from the bifunctional character of SLRPs, whereby the core protein binds to the collagen fibrils and the highly-charged glycosaminoglycan side-chains (GAGs) regulate interfibrillar spacing, thus contributing to the precise topology of the fibrils in the comea. Three classes of SLRPs can be identified based on several criteria including their sequence conservation, the presence of a distinct cysteine-rich cluster in the N-terminal region, the number of the LRRs and genomic organization (lozzo, 1997)(Figure 3.1). 3.1.4.1 Class I SLRPs This group includes decorin (Krusius and Ruoslahti, 1986) and biglycan (Fisher et al., 1989), which shows the highest degree of sequence conservation (-57% identity to each other). All members have a pro-peptide, which is highly conserved across species and functions as a recognition signal for xylotransferase, the first enzyme involved in the synthesis of GAG side chains. The presence of 10 LRRs flanked by cysteine rich regions and an N-terminal Cys consensus sequence that is unique (CX3CXCX6C) and different from the other two classes are the most distinctive features of this group. Another notable feature is that they are encoded by genes composed of 8 exons with the exon/intron junctions in highly conserved positions. 104 Results 3.1.4.2 Class II SLRPs This group comprises five members that can be further divided into three distinct subfamilies. Fibromoduiin and lumican constitute the first subfamily and exhibit -48% protein sequence identity; keratocan and the primary structure of the basic leucine-rich repeat protein (PRELP) (Bengtsson et al., 1995) constitute the second subfamily with -55% primary sequence identity, whilst osteoadherin (Sommarin et al., 1998) represents a distinct subfamily with 37-42% primary sequence identity to the other class II members. Members have a cysteine-rich consensus (CX3CXCX9C) just in front of the LRRs. Class II members are primarily substituted with KS chains, and polylactosamine, essentially an unsulfated KS can be found in both fibromoduiin and keratocan. Finally, class II members are encoded by only 3 exons, with a large central exon encoding nearly all the 10 LRRs (Oldberg et al., 1989; Bengtsson et al., 1995). 3.1.4.3 Class III SLRPs Epiphycan/PG-Lb (Kurita et al., 1996) and mimecan/osteoglycin (Johnson et al., 1997), which exhibit only -40% protein sequence identity to each other, are the two members of this class. These proteoglycans can be distinguished from the other two classes by a different cysteine- rich consensus (CX 2CXCX6 C) and by the presence of only 6 LRRs. In addition they are encoded by a gene containing 7 exons and the LRRs are encoded by only 3 exons. 105 Results Decorin gene — Decorin Class I I la II III IV V VI VII VIII CicGcCxfiC ■ ■ — Biglycan LRR — Fibomodulin Fibromodulin gene — Lundcan Class II I II III Keratocan Cx3Cx Cx9C — PRELP Osteoadherin LRR Epiphycan gene I II III IV V VI VII Epiphycan Cx2 Cx Cxfi C □ B □□ __ Mimecan LRR Figure 3.1 A dendogram representing the three classes of SLRP genes. Roman numerals indicate exon numbers. The 5'- and 3’-exons encoding untranslated regions are represented by black rectangles. The consensus sequences for the N-terminal cysteine-rich regions are given inside rectangles . Oval circles represent the LRR motifs. 106 Results 3.1.5 JŒÆ4 KERA, previously been cloned in bovine and mouse (Corpuz et al., 1996; Liu et al., 1998), has been assigned to human chromosome 12q22 by using fluorescence in situ hybridisation (FISH) (Tasheva et al., 2000a). The two other proteoglycan genes important to comeal function, decorin {DCN) and lumican (LUM), also reside on human chromosome 12 at bands q22 and q23 (Tasheva et al., 1999; Tashevaet al., 2000b; Tasheva et al, 2000c; Long et al., 2000). The gene has been isolated by establishing a BAC contig spanning the interval between D12S82 and D12S351. The complete genomic sequence and exon organisation of KERA were obtained by sequencing the BAC clone RPCIl 1-91705. KERA is expressed in several ocular tissues including the comea and sclera as well as in the trachea. Also, it is present in low levels in the intestine, skeletal muscle, ovary and lung. Human KERA spans 6.6 kb and is composed of 3 exons. Exon 1 is untranslated and exon 2 contains the start codon and encodes both the N-terminal peptide sequence and the central leucine-rich domain comprising 10 LRR motifs (Figure 3.2). The predicted protein is 352 amino acids (Pellegata et al., 2000). Mutations in KERA have recently been shown to be responsible for CNA2. A single nucleotide substitution was found in exon 2 comprising an A ^ G transition at codon 247, resulting in an amino acid change from Asn (N) to Ser (S). This amino acid substitution affects the N residue in the LRR consensus motif (LXXLXLXXNL) a highly conserved amino acid. A second homozygous missense mutation was found in exon 2 comprising a C—>T transition at codon 174 resulting in a premature stop and a truncated protein of only 173 amino acids. No potentially disease-causing mutations 107 Results have been found in the coding region, intron-exon boundaries or the 5’ and 3' untranslated regions (UTRs) of KERA in subjects with CNAl (Pellegata et al, 2000). Keratocan, the protein product of KERA, is an SLRP that interacts with specialised collagens to produce transparent comeal structure. Keratocan together with lumican and mimecan represent the major KS-containing proteoglycans (KSPGs) of the vertebrate comea. They share many common structural motifs like the central LRRs and conserved N-linked glycosylation. They diversify in the N- and C-terminal domains and in their sites of KS-substitution. The differences between keratocan, lumican and mimecan imply that each individual KSPG member may have a distinct role in regulating comeal stromal collagen fibrillogenesis, and therefore, comeal transparency. 108 Results Intron 1:1453 bp intron 2: 3878 bp 240 bp bp \ 172 bp893 cDNA:2l60bp ATGt TAA f Figure 3.2 Schematic diagram of the structure of human keratocan gene. The 3 exons are represented by boxes and are numbered. The translated regions are represented by filled boxes, untranslated regions by hatched boxes. [Adapted from (Pellegata et al., 2 0 0 0 )] 3.2 Aim of the study The aim of this study was to evaluate a family from Bangladesh with a combined phenotype, comea plana and microphthalmia, and to identify the genetic defect causing this novel phenotype via linkage mapping of the family and candidate gene mutation sc re e n in g . 109 Results 3.3 Patients and methods 3.3.1 Family and clinical data The family studied is a consanguineous pedigree originating from Bangladesh which displays an autosomal recessive pattern of inheritance (see figure 3.4). The parents are first cousins and neither they nor their relatives (apart from four of their offspring) are affected. The other three children are unaffected. The family members were examined in the UK and ophthalmic examination included slit-lamp biomicroscopy, gonioscopy, measurement of intraocular pressure, optic disc assessment and B-scan ultrasonography to determine axial length. Axial length measurement in addition to the clinical appearance of reduced globe size, comea plana and sclerocomea, were the criteria to assign an individual as affected (Figure 3.3a and b). The clinical characteristics of the four affected individuals are as follows: their mean age was 7 years (range 3-12) and all exhibited bilateral comea plana with a varying degree of comeal opacity centrally and peripherally, sclerocomea and microphthalmia (mean axial length 19.3 mm, range 18.6 - 22.3 mm, normal 24.2 mm in male and 23.9 mm in female) (Larsen, 1971). All four also had reduced visual acuity, ranging from Snellen 6/24 to 6/60 in the better eye, hypermetropia and intraocular pressure within normal limits (Table 3.1). None of them had undergone ocular surgery and there was no evidence of posterior segment changes. No systemic features associated with microphthalmia, such as mental retardation, were present in this pedigree. Examination of other siblings revealed healthy anterior and posterior segment and no systemic abnormalities. 110 Results Table 3.1 Ocular features of affected individuals in the family Patient Age Visual Corneal Corneal Anterior Refractive Axial No (Yrs)/ acuity diameter opacity chamber state length Sex depth IV-1 6/M R4/60 9.5 mm Central and 1.3 mm 18.2 mm L6/24 8.5 mm peripheral 2.2 mm 18.7 mm IV-2 7/F R3/60 9.0 mm Peripheral 1.2 mm 20.3 mm L6/36 8.5 mm 2.0 mm t 20.3 mm IV-3 3/F R6/48 9.0 mm Peripheral 1.4 mm 1 18.3 mm L6/60 9.5 mm 2.1 mm 1 18.7 mm IV-4 12/F R6/36 10.25 mm Central and 3.5 mm 22.3 mm L6/36 10.25 mm peripheral 3.0 mm 21.7 mm 111 Results Æ Figure 3.3a Slit lamp photograph of an affected sibling (case IV-1) showing flattened small comea, indistinct limbus, irregular eccentric pupil and shallow anterior chamber. 12 Results Figure 3.3b Slit lamp photographs of an affected sibling (case IV-I) showing the right (a) and the left (b) eyes with a central diffuse corneal opacification and broad limbus. 113 Results 3.3.2 Methods 3.3.2.1 Genotyping Genomic DNA was extracted from venous blood or buccal mucosa samples (see sections 2.1.1 and 2.1.2), and amplified using the following tri/tetranucleotide repeat markers spanning known genetic loci for arCMIC, NNOl, CNA2, and adCMIC. Chromosome 14: D14S987, D14S65, D14S267, D14SI025, D14S77 and D14S78 (arCMIC); chromosome 11: D11S1313, D11S903, D11S4191, and D11S987 (NNOl); chromosome 12 D12S92, D12S351, D12S322, D12S95, D12S327, and PAH (CNA2); and chromosome 15 AFMb293ygl, and AFM312zd9 (adCMIC) (all from Research Genetics, Huntsville, USA) (see table 3.2). PCR reactions were performed as described (section 2.2) and the annealing temperatures ranged from 50°C -60°C. The amplified PCR products were separated on 6% non-denaturing polyacrylamide gels by electrophoresis and were visualised by staining with ethidium bromide (see section 2.3.2). Subsequent genotyping was performed manually. 3.3.2.2 Linkage analysis Pedigree data was collated with Cyrillic and two point linkage analysis performed using the MLINK component of the LINKAGE program, version 5.1 (Lathrop and Lalouel 1984) using an autosomal recessive model, equal allele frequencies, a gene freqygncy of 0.0001 and a mutation rate of 0.000001. m Results Table 3.2 Markers on chromosomes 11, 12, 14 and 15 used for the linkage study Marker Annealing Tm Location Heterozygosity Sequence length (bp) D11S1313 55 °C Ilpl3-llpl2 0.8520 184 D11S903 55 °C Ilp l3 -llq l3 0.7500 99 D11S4191 55 °C 1 lpter-1 Iqter 0.8800 111 D11S987 55 °C 1 lpter-1 Iqter 0.8200 82 D12S322 55 “C 12pter-12qter 0.6741 169 D12S327 50 °C 12q21-12q22 0.8501 182 D12S351 55 °C 12pter-12qter 0.75 164 PAH 50 “C 12q22-12q24.2 0.5 200 D12S95 58 °C 12pl3.2-12q24.33 0.7667 146 D12S92 50 °C 12pl3.2-12q24.1 0.7702 188 D14S987 55 “C 14pter-14qter 0.7633 308 D14S65 55 “C 14q32.1-14q32.1 0.8035 125 D14S267 55 °C 14q32-14q32 0.9005 224 D14S77 57 “C 14q24-14q244 0.9390 203 D14S1025 57 “C 14qter-14qter 0. 8700 135 D14S78 55 “C 14q32.1-14q32.2 0.6793 211 AFMb293ygl 55 °C 15pter-15qter 0.8700 165 AFM312zd9 53 °C 15pter-15qter 0.7500 299 115 Results 3.3.2.3 Mutation screening of KERA by sequencing PCR amplification of the KERA gene sequence was performed using published primers (Pellegata et ah, 2000). The primer sequences, genomic locations, annealing temperatures, and predicted product lengths (bp) are listed in Table 3.3. After purification samples were sequenced bi-directionally with fluorescent dideoxynucleotides on an ABI 373 automated sequencer using standard conditions (see section 2.6.3). Table 3.3 Primers used for amplification of KERA Primer Primer Genomic Sequence (5 ->3 ) Tm Predicted name No. locations product lengths (bp) KERA Ifw 1 5' region fw T ggtgactgggacgagtagg 61.12 273 KERA Irev 2 Intron Irev Ctttttcagaatagggttttgg 57.04 KERA 2Efw 3 Intron Ifw T gttgacatatttcacctcttcc 59.42 477 KERA 2Erev 4 Exon 2rev Agggctcctttttcaattcc 59.53 KERAlGfsN 5 Exon2fw Ctgtaggtgctataatggcagg 58.81 522 KERAIAvqy 6 Exon2rev Gggtcaggttctccagattg 59.51 KERA 2Cfw 7 Exon2fw Ggaggtaccttctccattgc 57 576 KERA2BrQ\ 8 Jntron2 rev Gggcaacacatttgctcttc 60.65 ÆEÆ4 2Ffw 9 Exon2fw Ctcatgcagctaaacatggc 59.45 292 KERAIV xqw 10 Exon2rev Gatcaaggtgaaggtgctgc 60.8 KERA 3fw 11 Intron2fw Ttgggggaaacagatagg 59.9 463 KERA 3r 12 3'UTRrev Gaaaatggtggccgagagc 66.9 116 Results 3.4 Results 3.4.1 Linkage of the family to the known locus for cornea plana Haplotype analysis based on six polymorphic markers around the known locus for CNA2 in the following order (Centro-D 12S92-D12S351 -D12S322-D12S95-D12S327- PAH-Telo) showed segregation with disease for CNA2 locus (Figure 3.4). A region of homozygosity of approximately 1 cM for markers D12S351, D12S322, and D12S95 that includes the CNA2 locus (Figure 3.5) was noted. Markers PAH, D12S95, D12S351, and D12S322 segregate with the disease in the family (Figure 3.6a and b). Positive LOD scores were obtained with markers in a region on chromosome 12q21 spanning approximately 3 cM. The maximum two-point LOD scores were obtained with markers D12S95 and D12S327 (2.18 at 0 = 0) as these markers were the most informative. Markers and the LOD scores are summarised in table 3.4. 117 0 II-1 II-2 'II-3 6 6 6 III-l III-8 III-3 III-4 HI-5 III-6 III-l 1 III-10 III-12 III-13 Marker order IV-2 IV-3 IV-4 IV-5 IV-6 IV-7 D 12S92 3 2 3 2 3 2 3 1 3 1 2 1 D12S351 2 1 2 1 1 D12S322 1 2 1 2 1 2 D12S95 2 2 3 1 3 1 1 D I2S327 3 3 2 3 2 3 3 PAH 1 4 1 4 2 3 2 3 1 4 Figure 3.4 Pedigree of a four-generation consanguineous family from Bangladesh with i autosomal recessive comea plana and microphthalmia, showing haplotypes for the polymorphic markers on chromosome 12q21. 13.3 Results - DI2S92 11.2 14 cM - DI2S3I9 IcM 14 D12S82 15 IcM DI2S351 OcM 21.3 D12S322 I IcM D12S95 3cM - DI2S327 20cM L PAH Q CNA2 I CNAl [] Area of homozygosity Figure 3.5 Schematic diagram of chromosome 12 indicating the genetic distances between markers used in the study and the locations of CNAl, CNA2, and the disease interval defined in the Bangladeshi family presented. 19 Results □ - o III-2 III-l 1 i i H66 6 IV-I IV-2 IV-3 IV-4 IV-5 IV-6 IV-7 PAH 1/2 1/4 1/4 4/3 1/4 1/4 2/3 2/3 1/4 m 0 » # # » # # % # m Ml 234 56789 DI2S95 — ^ • • # 2/i 2/2 2/2 2/1 2/2 2/2 .VI .VI 2/1 M 12 3 4 5 6 7 8 9 Figure 3.6a Cosegregation of disease alleles with markers PAH and D12S95. Affected individuals (IV-I, IV-2, IV-3, and IV-4) share the same disease alleles (lanes 2, 3, 5, and 6). Heterozygous parents (III-2, and III-l I) and heterozygous sibling (IV-7) share heterozygous disease allele (lanes 1, 4, and 9). Homozygous unaffected siblings (IV-5, and IV-6) share similar normal alleles (lanes 7, and 8). Allele assignments are shown. M= (|)XI74 RF DNA//ac III. 120 Results D12S351 VI VI la V\ 2Æ M 1/1 2/1 in ^ 4* ###*##* * * D12S322 in 1/1 in 1/2 1/1 1/1 1/2 1/2 1/2 M 12345«789 Figure 3.6 b Cosegeregation of the disease alleles with markers D12S351 and D12S322. Affected individuals (IV-1, IV-2, IV-3, and IV-4) share the same disease alleles (lanes 2, 3, 5, and 6). Unaffected parents (III-2, and III-l 1, lanes 1 and 4) and unaffected siblings (IV-5, IV-6 and IV-7, lanes 7, 8, and 9) share similar normal alleles. Allele assignments are shown. M= (f)X174 RF DNA Hae III. 12 Results 3.4.2 linkage analysis to the known loci for microphthalmia Linkage analysis with an array of highly polymorphic microsatellite markers closely linked to the arCM lC locus on chromosome 14q32 was performed and excluded linkage of the family to this locus (Figure 3.7). Similarly the family was not linked to the other loci for NNOl on chromosomes 11, and 15. 1>I4S'J«7 1/2 1/2 1/2 1/2 1/2 1/2 1/1 1/2 2/2 W fi w M *» * .M Figure 3.7 Linkage analysis to chromosome 14. Marker D14S987, closely linked to arCM lC, is not linked with the disease in this family since affected individuals (lanes 2, 3, 5 and 6) share the same alleles as the unaffected individual (lane 8). Allele assignments are shown. M= (j)X174 RF DNA Hae III m a rk e r. 122 Results Table 3.4 Two-Point Lod Scores between the CNA2/microphthalmia locus and 6 markers on 12q21 ordered from the centromere Lod score at 0 Locus 0.00 0.05 0.10 0.2 0.3 0.4 D12S92 -00 0.15 0.31 0.32 0.21 0.07 D12S351 1.28 1.16 1.03 0.75 0.45 0.15 D12S322 1.28 1.16 1.03 0.75 0.45 0.15 D12S95 2.18 1.98 1.77 1.3 0.78 0.24 D12S327 2.18 1.98 1.77 1.3 0.78 0.24 PAH -00 0.99 1.09 0.93 0.6 0.21 123 Results 3.4.3 Mutation screening of KERA Sequencing of KERA revealed a single-nucleotide substitution in exon 2 that segregated with the disease phenotype (Figure 3.8). The sequence change (ACA^AAA at codon 215) results in an amino acid substitution from threonine (T) to lysine (K). All 4 affected siblings were homozygous for the mutated allele whilst both parents and one of the unaffected siblings (IV-7) were heterozygous. The other two siblings were homozygous for the normal ACA allele. The mutated sequence does not generate or alter a restriction enzyme site. Primary sequence analysis indicates that the T215K substitution occurs at the start of an LRR motif, the 7th of 10 such motifs present in the sequence. Analysis of the predicted secondary structure indicated that the 7th LRR motif is associated with a short element of p-strand that may be altered by the T215K substitution (University of Wisconsin GCG software package, Peptide Structure; Wolf et al. 1988). 124 111:2 (heterozygote) III: 11 (heterozygote) T/K T/K C A A T A C7A A A C A A I A C/A A A IV: 1 affected IV:2 affected IV:3 affected IV:4 affected IV:5 unaffected IV:6 unaffected IV:7 unaffected K K K K T T T/K CAATAAAA CAATAAAA CAATAAAA CAATACAA CAATA c/a AA 1 Figure 3.8 Electropherogram of exon 2 of the KERA gene. Heterozygous parents (I1I-2, and 111-11) and heterozygous unaffected I sibling (IV-7) show double peaks (heterozygous mutation, T/K); homozygous affected individuals (IV-1 to IV-4) reveal the mutant I allele only (homozygous mutation, K); and homozygous unaffected siblings (lV-5, and lV-6) with the wild type allele, T. Arrows w indicate the mutation in codon 215 with the translation of this codon shown above (T= threonine, K= lysine). LA Results 3.4.4 Functional analysis of the T215K mutation by protein modelling A 3D structure of keratocan was modeled (with Dr. Patel and Dr. Wilkie) based on a theoretical structure of the acid-labile sub-unit of a serum insulin-like growth factor (ALS) (Janosi and Ramsland, 1999) using the Swiss model server program (Peitsch, 1996; Guex et al., 1999) and the model was analysed using Swiss Pdb Viewer v3.7b2 (Guex and Peitsch, 1997) and Rasmol v2.6 (Roger Sayle, Glaxo Wellcome, Stevenage, UK). A modest degree of sequence identity (30.5%) was found with the protein ALS, which also contains a series of LRR motifs, but this was sufficient to allow modelling of keratocan based on the structure of ALS (Figure 3.9a). The availability of a theoretical structural model for ALS facilitated the generation of a partial 3D model for keratocan (Figure 3.9b and c). This model is incomplete at the N- and the C-termini, where homology between keratocan and ALS is low, but it does include the region of the protein comprising LRR 1-9. Thus it was possible to examine the structural implication of the mutation at site 215, which is located at the start of LRR7. The model indicates that in the protein the LRR motifs form a series of parallel p-strands, which stack into an arched p-sheet array. The mutation reduces the effective length of the p-strand region of LRR7 and causes the loop connecting LRR6 to LRR7 to be laterally displaced (Figure 3.10). 126 Results Figure 3.9 3D modelling of wild type and T2I5K mutant of Keratocan. (a) 3D model of ALS is represented as ribbon b-sheets (the first is depicted in light blue) and a-helices (the first in red), (b) 3D mutant type keratocan in ribbon is shown on the left (in yellow) and the wild type (in pink) on the right, (c) Superimposed wild and mutated types showing LRRs numbered from 1-9 with displacement of the loop connecting LRR6 and 7. 127 Results a u. u (51 ■: V2 A / V I C term 1234 56 789 N-term T2I5 ^ K215 il~J Figure 3.10 a) 3D modelling of wild type and T215K mutant keratocan. Partial structure of the wild type represented in ribbon form with LRR motifs 1-9 numbered and highlighted in darker grey. The model is represented from two perpendicular view points, b) An enlargement of the region boxed in A showing the superimposition of the wild type (light grey) and T215K mutant (dark grey). The side chains of residues 215 are shown projecting out from the protein backbone and the lateral displacement of the loop connecting LRRs 6 and 7 is clearly visible 128 Results 3.5 Discussion Mutations in keratocan have recently been shown to cause autosomal recessive cornea plana in a large cohort of Finnish and American patients (Pellegata et al., 2000). The mutations identified to date (an AAC-> AGC transition at codon 247 and a CAG^TAG transition causing a premature stop) both occur in a conserved domain and result in a typical cornea plana phenotype. The data presented here from a family of different ethnicity indicate that a novel ACA—> AAA mutation at codon 215 causes a more complex phenotype than one solely involving the cornea. The presence of the combination of microphthalmia and cornea plana, suggests that KERA may play a role in the development of the eye as well as structural role in the cornea, a view supported by the observation of KERA expression in both murine (Liu et al., 1998) and bovine sclera (Corpuz et al., 1996; Tasheva et al., 1998). In common with other members of the keratan sulfate proteoglycan sub-group of the SLRP family, keratocan includes a series of LRR motifs with a consensus sequence of LXXLXLXXNXL (Liu et al., 1998). Similar LRR repeats are found in the structure of porcine ribonuclease inhibitor whose crystallographic structure indicates that the LRR repeats form a series of p-strands which stack into a parallel p-sheet array (Kobe and Deisenhofer, 1993). This defined protein structure has been used to model the structures of other proteins containing analogous LRR motifs, such as the protein ALS. It had previously been suggested that the structure of keratocan includes a similar p-sheet array and the model presented here supports this (Dunlevy er al., 1998). The modelling predicts that the 129 Results structural consequences of the mutation are to cause a substantial displacement of one of the loops holding together the p-sheet structure. However, it should be noted that this conclusion is based on theoretical considerations only. The mutation at codon 215 substitutes the positively charged amino acid lysine for the polar residue threonine at the start of the 7th LRR motif and the modelling described herein indicates that this may result in a slight modification of the motif structure. Since threonine is conserved at this position in all published keratocan proteins (human, bovine, murine and chicken) the substitution of this highly conserved amino acid may explain the different phenotype observed in this family. The motif arrangement has been proposed to be important in the spacing of collagen fibrils on which comeal transparency depends and this is compatible with the observation that mutations in KERA result in comeal opacity (Pressman et al., 2000). The actual effect of the change in the 7th LRR motif on protein fimction and the sclera are however, at present unknown. Similarly the other mutations in CNA2 have been reported to have structural implication on the protein. N247S reported in the Finnish population affects LRR5 and Q147X reported in an American patient affects LRR8. Although 100 Caucasian control chromosomes did not reveal this mutation, it would be important to screen an ethnically matched control panel of 250 individuals (Bangladeshi) to fully exclude the possibility that this change is a polymorphism. SLRPs are important mediators of normal connective tissue assembly which influence, via collagen binding, the rate of assembly and diameter of collagen fibrils (lozzo, 1999). In decorin, one of several comeal-expressed SLRPs, this collagen-regulating activity is believed to be mediated by the central LRR region (Schonherr et al., 1995; Kresse et al.. 130 Results 1997) and it is likely that the evolutionarily conserved structure of SLRPs reflects a common mechanism of collagen interaction (lozzo et al., 1999). The T215K substitution in keratocan is predicted to induce a conformational change in the LRR domain and may also affect collagen binding. This would provide a potential mechanism by which the observed comeal and scleral phenotypes could be caused either by alterations in the diameter or the spacing of collagen fibrils. A particularly interesting feature of this consanguineous pedigree is the co-segregation of cornea plana with reduced axial length. This may indicate that a single KERA mutation causes both phenotypes. The possibility that a mutation in a second gene, co- segregating with KERA, is responsible for microphthalmia in this consanguineous pedigree cannot be excluded, although the probability of this occurring by chance is low (0.16%, (0.25 "^x 0.75 ^), equating to a lod score of 2.8 that the two traits segregate independently. This low figure, together with the nature of the mutation and the transient expression of KERA in embryonic (El 5.5) murine sclera (Liu et al., 1998), suggest that a mutation in KERA could also be responsible for the microphthalmia. However, the possibility that the reduced axial length in this family is an additional phenotypic feature for cornea plana can not be excluded. Future diagnosis of cornea plana should include measurement of the posterior axial length to either confirm or exclude this observation as a feature of the disease. Microphthalmia has a wide spectrum of disease severity with anophthalmia and minor degrees of congenital microphthalmia co-existing in the same pedigree (Warburg, 1993). Although the microphthalmia in this family lies at the milder end of the disease spectrum, the axial lengths overlap with those found in the arCMIC and NNOl pedigrees (Othman et al., 1998; Bessant et al., 1999). The clinical phenotype is also 131 Results compatible with reports of altered scleral thickness and abnormal collagen bundles in histological specimens from other microphthalmia cases (Yue et at., 1988). The milder degree of microphthalmia observed may reflect KERA's presumed structural role, albeit one modulated by the LIM transcription factor Imxlb (Pressman et al., 2000), and contrasts with the more severe phenotype observed with mutations in the developmental gene CHXIO (Percin et al., 2000). In order to confirm that the microphthlamia is attributable to the KERA mutation may necessitate screening a large panel of isolated cases due to the known genetic heterogeneity and the low prevalence (2%) of mutations reported for the first gene (Percin et al., 2000). Such a study would also clarify whether comeal pathology is an invariable feature of ÆE7L4-induced microphthalmia. 132 Results Chapter 4 Spectrum of BIGH3 mutations in corneal dystrophies 4.1 Introduction 4.1.1 Chromosome 5q linked dystrophies Several autosomal dominant comeal dystrophies have been mapped to chromosome 5q. CDGGI was first mapped by Eiberg and colleagues in 1993 using linkage analysis to chromosome 5q31. Stone and colleagues (1994) subsequently showed that ACD and LCDI dystrophies were also linked to a similar interval on 5q and genetic refinement of LCDI to within a 2 cM interval was reported (Gregory et al., 1995). CDBI was also mapped to the same common interval on 5q31 in the vicinity of interlukine (IL) 9 (Small et al., 1996). Thus CDBI, LCDI, ACD, and CDGGI all mapped to the same genetic locus suggesting that (i) comeal dystrophies represent allelic heterogeneity (that is, different mutations within the same gene manifest as different phenotypes), (ii) they are all the same disease or (iii) that a comeal disease gene family exists in this region. Korvatska et al. (1996) refined the genetic localisation of CDGGI and LCDI to a 1 cM region between markers D5S393 and D5S399 which equates to approximately 1 megabase (Mb). Subsequently the BIGH3 gene was identified as the causative gene for a number of comeal dystrophies (Munier et al., 1997). 133 Results 4.1.2 BIGH3 gene The BIGH3 gene or transforming growth factor-6 (TGF-6)-induced gene was first isolated from human adenocarcinoma cell lines derived from the lung after treatment with TGF-6 (Skonier et al., 1992). It was considered as a good candidate gene for comeal dystrophies that mapped to chromosome 5q31 (Stone et al., 1994; Gregory et al., 1995; Small et al., 1996) because it displayed a distinct expression pattern in human eye, mainly in the comeal epithelium and stromal keratocytes (Skonier et al., 1994; Escribano et al., 1994). Munier et al. (1997) identified the BIGH3 gene by constructing a yeast artificial chromosome (YAC) contig covering the region of interest followed by several rounds of cDNA selection to identify genes within the candidate region. Potential clones were identified and one clone, when aligned with sequences from Genbank, was 100% homologous to a portion of the BIGH3 gene identified by Skonier et al. (1992). The gene coding sequence is 2052 nucleotides long, divided into 17 exons. The size and structure of BIGH3 is well conserved between human and mouse (Schorderet et al., 2000). It spans approximately 30 kb of genomic sequence on mouse chromosome 13 while the human gene is approximately 35 kb. The characterisation of the murine structure of the BIGH3 gene may provide the tools necessary to create mouse models, allowing better understanding of the physiological role of the gene. 4.1.3 Kerato-epithelin The product of the BIGH3 gene is a highly conserved 68 kDa secretory protein known as piG-H3 protein or kerato-epithelin (KE) that is found in the comea. It contains 683 amino acids, with an amino terminal secretory sequence, a carboxyl-terminal RGD (Arg- 134 Results Gly-Asp) sequence located from codon 642 to codon 644, and four 140-amino-acid repeats with internal homology (Figure 4.1). D1 D2 D3 D4 RGD Î' R124C R555W R124H R555Q Figure 4.1 Schematic diagram of kerato-epithelin structure. D1 to D4 represents homologous domains that contain 2 highly conserved repeats designated R and r. S= secretory sequence. Arg-Gly-Asp is a recognition sequence for integrins. The location of the mutations associated with ACD, CDGGI, CDBI and LCD are marked in the diagram (arrows), (modified from Munier et at., 1997) The RGD sequence, found in many extracellular matrix proteins, modulates cell adhesion and acts as a ligand recognition site for integrins. The integrins that specifically bind to corneal KE protein are unknown but it has been suggested that the four domains of the protein can be folded into a potential bivalent tetrameric structure that may act as a bridge between cells expressing the appropriate ligand (Skonier et al., 1992). Thus, KE may be involved with cell adhesion in a similar manner to other homologous cell adhesion proteins (fascilin I, {Dorsophila neuronal cell adhesion molecule}, OSF-2, (human osteoblast specific factor 2} and MPB70 (mycobacterium bovis adhesion molecule}) (Zinn et al., 1988). KE has a wide distribution in connective tissue and its message is up-regulated during normal development and wound healing suggesting a role in morphogenesis of extracellular matrices (Skonier et al., 1992; Klintworth et al., 1994). Immunohistochemically the protein is present in human foetal eyes in many extracellular 135 Results tissues including the comeal stroma, conjunctival connective tissue, iris, ciliary body, and retinal pigment epithelial basement membrane (Kublin and Cintron, 1996). It has been reported that KE is associated with type VI collagen in comeal stroma proposing a role for KE in maintaining proper fibril spacing important for comeal transparency (Rawe gr aA, 1995). 4.1.4 Corneal dystrophies caused by BIGH3 gene mutations 4.1.4.1 Granular corneal dystrophy (GCD) or (Groenouw type I) (CDGGI) (OMIM NO 121900) GCD is an autosomal dominant condition that usually becomes apparent in the first or second decade of life. It is characterised by opacities which are seen in the subepithelial and stromal layers of the comea in the form of small, discrete, greyish white and sharply demarcated deposits (section 1.3.3.1). At this stage vision is not impaired with little discomfort reported. As the condition advances, the opacities enlarge, coalesce and multiply to invade the deeper stroma (Moller, 1990). In the fourth or fifth decade, visual impairment usually occurs secondary to light scatter by the opacities but most patients do not require keratoplasty. Slit lamp examination at the start of the disease shows fine dots and radial lines in the superficial stroma. Later, focal white opacities in the form of chains, rings or branching pattems develop in the anterior stroma. Histopathologically, the deposits stain intense red with Masson's trichrome stain and are weakly positive with PAS (Jones and Zimmerman 1961; Gamer, 1969a). The exact nature and source of the deposits are unknown however, histochemical studies disclosed that they contain tyrosine, tryptophan, arginine, sulphur containing amino acids (Gamer, 1969) and phospholipids (Kanai et al., 1977; Rodrigues et al., 1983). 136 Results 4.1.4.2 Corneal dystrophy of Bowman’s layer (CDB) CDB is a relatively rare autosomal dominant disorder originating in the anterior limiting membrane (Bowman's layer) which causes moderate visual impairment (section 1.3.2.1). It was first described by Reis in 1917 when he diagnosed a 20-year old patient who had history of intermittent attacks of ocular irritation since age 5 years as herpetic keratitis, but he noted that several family members had experienced similar attacks. In 1949, Bucklers examined the same pedigree in more detail and showed that the disease was characterised by comeal opacities in a geographic pattern at the level of Bowman's layer and a progressive decrease of visual acuity (Laibson, 1997). Subsequently, Kuchle et al. (1995) classified comeal dystrophy of Bowman's layer and the superficial stroma (CDB) into CDBI (geographic or "tme" Reis- Bucklers' dystrophy) and CDBII (honeycomb-shaped or Theil-Behnke dystrophy). Both forms share the autosomal dominant pattem of inheritance, early start in childhood, painful recurrent erosions and opacities at the level of Bowman's layer. CDBI is clinically characterised by confluent geographic opacities at the level of Bowman's layer, histopathologically by band shaped granular Masson's trichrome positive subepithelial deposits, and ultrastmcturally by the presence of rod shaped bodies. However, slit lamp examination in CDBII revealed bilaterally symmetric, central, confluent honeycomb opacities, with grey interlacing lines punctuated by clearer zones (Kuchle et al. 1995). Histopathological examination revealed that the Bowman's layer is absent and replaced by an avascular fibrocellular tissue which stains mildly positive with eosin and PAS stain, and transmission electron microscopy showed subepithelial 137 Results accumulation of atypical, fine, curly, collagen filaments with an arcuate to round configuration. 4.1.4.3 Avellino corneal dystrophy (ACD) (OMIM NO 121900) ACD is inherited as an autosomal dominant trait with high penetrance (section 1.3.3.3). The condition has been named Avellino, after the Italian province near Naples where the first affected families to be identified originated (Folberg et al., 1988). However the condition has been reported in Germany (Rosenwasser et al., 1993), Ireland (Kennedy et al., 1996), Europe (Munier et al., 1997), Japan (Santo et al., 1995, Konishi et al., 1997, Akimune et al., 2000), and France (Dighiero et al., 2000b) and in this study two British families were described for the first time. The condition appears in the first decade of life as granular lesions that resembles the sharply demarcated round, focal deposits of CDGGI. Later in life, lattice lesions begin to appear but, they are larger, denser, whiter and more polygonal than those of LCD type I (Lucarelli and Adamis, 1994). However, the last clinical sign to emerge is the stromal haze so, all patients with the stromal haze has both granular and lattice lesions and they represent the most advanced form of the disease (Holland et al., 1992). Histologically, superficial granular deposits that appear bright red with Masson's trichrome together with superficial deposits that have the staining characteristic of amyloid are present (Folberg et al., 1994). Patients with ACD unlike those with LCD, achieve better visual rehabilitation with penetrating keratoplasty, with recurrence as late as 9 years (Cennamo et al., 1994). 138 Results 4.1.4.4 Lattice corneal dystrophies (LCD) LCD is a heterogeneous group of inherited comeal dystrophies (section 1.3.3.2) that have been classified into five distinct subtypes I, II (Klintworth, 1977), III (Hida et al., 1987), IIIA (Stock et al., 1991) and type IV (Fujiki et al., 1998). Types I, IIIA, and IV have been reported to be caused by mutations in the BIGH3 gene (Munier et al., 1997; Yamamoto et al., 1998; Kawasaki et al., 1999; Stewart et al., 1999b; Fujiki et al., 1998). LCD type I (OMIM 122200) is an autosomal dominant bilaterally symmetric comeal disorder with complete penetrance (Frayer and Blodi, 1959). It is characterised by thin greyish linear, branching deposits of amyloid material that progressively accumulate in the subepithelial and stromal layers of the comea. The symptoms appear during the first or second decade of life with recurrent comeal erosions leading to comeal scarring (Mannis et al., 1997; Waring and Mbekeani, 1998). The natural course of the disorder involves progressive comeal opacification and eventual blindness. LCD type IIIA is a late onset form of LCD characterised by recurrent comeal erosions and by an autosomal dominant inheritance pattem (Stock et al., 1991). Markedly thick, ropy lattice lines throughout the stroma characterise this form of LCD and on histologic examination, the amyloid deposits vary in size and ribbons of amyloid between the stroma and Bowman’s layer are typical of this form (Dighiero et al., 2000a). LCD type IV is a late-onset atypical LCD with deep stromal opacities which has been reported to be caused by BIGH3 gene mutation (Fujiki et al., 1998a). 139 Results 4.2 Aim of the study To identify and characterise novel underlying mutations in families and sporadic patients with CDB, CDGGI, ACD, and LCDI from different ethnic origins. To achieve this goal mutation screening of the BIGH3 gene together with careful study of the clinical phenotype of the studied patients were undertaken. 4.3 Patients and methods 4.3.1 Patients 10 families were included in this study, 5 with CDBII, 2 with CDGGI, 2 with ACD, and 1 with LCD together with 10 sporadic patients, 4 with CDBII, 3 with CDGGI, and 3 with LCD. So, in total 36 patients, 10 sporadic subjects and 26 patients who are members of families together with 27 normal relatives and 50 Caucasian controls were studied. A Bangladeshi family with both CDBI and LCDI phenotypes was included, with 3 affected and 2 unaffected members. Informed consent was obtained for clinical and molecular investigations (according to the approval of Moorfields Eye Hospital Ethics Committee). The disease showed an autosomal dominant inheritance pattem in all the studied families as determined from the pedigrees (Figures 4.2, 4.3,4.5,4.8, and 4.9). All individuals in the study underwent clinical examination including best-corrected visual acuity measurement, slit lamp examination, and fundus examination before molecular investigations. Age, sex, time of onset of the disease, and of visual deterioration, best corrected visual acuity, treatment received, and recurrence of the 140 Results disease (3 years) after keratoplasty in CDBII, CDGGI, and LCDI patients are listed in table 4.1. 4.3.1.1 CDBII patients 17 patients with CDBII were studied, 4 sporadic and 13 patients who are members of 5 families. All families and sporadic patients are British except one of the sporadic subjects of Turkish origin. 5 patients and 3 unaffected relatives from family Al (Figure 4.2), 3 patients and 7 unaffected relatives from family A2 (Figure 4.2), 2 patients and 1 unaffected relative from family A3 (Figure 4.2), 2 affected and one normal subjects from family A4 (Figure 4.3), and 1 affected and 2 unaffected individuals from family A5 (Figure 4.3). All patients in this group were presenting with one or more of the following symptoms: photophobia, lacrimation, comeal erosions, and /or visual deterioration. Phenotypes of CDBII patients Slit lamp examination of the probands, all affected members from families Al to A5, and the sporadic cases revealed bilateral fine honeycomb-shaped opacities in the subepithelial region of the comea. The opacities were symmetrical, regularly homogeneous, and more prominent in the central comea (Figure 4.4a and b). This phenotype was consistent with CDBII form of CDB that is sometimes called honeycomb or Thiel-Behnke dystrophy. 141 Family Al Results II-5 III Ô / II]-4 IV fIV -1 i' IV-2 Family A2 66àà>r à ^ III 1 III-3 ^ m -6 IV boàic5~~ % Family A3 6 4 ' è O II-3 / m -1 Figure 4.2 Pedigrees of CDBII families (A I-A3) participated in the study. Arrows indicate the probands in each family. Asterisks indicate members who were examined clinically and genetically. 142 Results Fam ily A4 III IV V Family AS 00 Figure 4.3 Pedigrees of CDBII families (A4 and A5) participated in the study. Arrows indicate the probands in each family. Asterisks indicate members who were examined clinically and genetically. 143 Results Figure 4.4 Slit lamp photographs of CDBII in the proband of family Al caused by R555Q B1GH3 mutation. Direct focal illumination with a broad beam (a) and broad tangential illumination (b) showing typical fine honeycomb-shaped opacities in the central comea. These opacities are symmetrical and homogeneous. Yellow deposits in the centre represent iron deposits. 144 Results 4.3.1.2 CDGGI patients In total 9 patients with CDGGI were studied, 3 sporadic, and 6 who are members of two families. The first family (Bl) is British and consists of four generations from which 5 affected and 8 unaffected members (Figure 4.5) were included. The second family (B2) is of Jamaican origin and consisted of four generations from which one affected and 2 unaffected individuals (Figure 4.5) were included. The main presenting symptoms in this group were intermittent ocular irritation, photophobia, and late onset visual deterioration. Phenotypes of CDGGI patients Familv Bl (see figure 4.5) Case II-7 (the proband): Slit lamp examination of his corneas disclosed bilateral, confluent greyish white opacities located in the anterior central stroma (Figure 4.6a). Case II-3 (the proband’s sister): is the eldest member in the family, examination of her corneas by slit lamp biomicroscopy showed a similar phenotype to the proband. Case IÏI-4 Examination of his corneas by slit lamp biomicroscopy revealed bilateral sharply demarcated opacities with clear stroma in between. Variable shapes of opacities were detected; white breadcrumb deposits and annular figures with clear centre (Figure 4.6b). Case III-6 has a similar phenotype to case III-4 revealed by Slit lamp biomicroscopic examination of her corneas. Case IV 5 is the youngest in the pedigree, examination of her corneas showed fine, very small, drop shaped, whitish granules with clear stroma inbetween (Figure 4.6 c). 145 Results Familv B2 (see figure 4.5) Slit lamp examination of the proband (case HI-3) revealed bilateral focal greyish white placoid, dense opacities and fine, linear branching deposits in the anterior comeal stroma. This picture was suggestive of ACD but histopathological examination was unavailable because the patient had not undergone keratoplasty. Sporadic cases Examination of the sporadic case 1 showed bilateral comeal opacities of different configurations, ring shaped with clear centre, fine dots, and breadcrumb deposits (Figure 4.7a and b). The patient underwent penetrating keratoplasty (PK) in his right eye that showed recurrence with fine superficial deposits in the centre and at the host graft (Figure 4.7 c). Examination of sporadic cases 2 and 3 revealed a similar phenotype to cases HI-4 and HI-6 in family Bl. 146 Results Family B l / I O é r ü Or II-3 n-7 / IV rv-5 Family B-2 o / III-3 IV Figure 4.5 Pedigrees of CDGGI families (Bl and 82) participated in the study. Arrows indicate the probands in each family. Asterisks indicate members who were examined clinically and genetically. 147 Results Figure 4.6 Slit lamp photographs of CDGGI phenotype in three generations of family Bl caused by R555W mutation, (a) Advanced confluent grey white opacities, (b) stromal opacities with variable shapes, annular figures, clear centred and white bread crumb deposits, (c) Small discrete, well defined, rounded and oval shapes with clear stroma between the lesions. 148 Results Figure 4.7 Slit lamp photographs of a sporadic patient (case 1) with CDGGI phenotype caused by R555W mutation, (a) and (b) right and left corneas of the patient before keratoplasty showing variation in the shape of the lesion with different configurations of the deposits; ring-shaped with clear centre, fine dots, and breadcrumb deposits (arrows) (c) right cornea after keratoplasty showing fine superficial deposits in the centre and at the host graft junction (arrows). 149 Results 4.3.1.3 ACD patients Two British families, family Cl (Figure 4.8), and family C2 (Figure 4.8) were studied. Clinical data and phenotypes of ACD families Family Cl (see figure 4.8) Case II I (the proband): was first examined in Cornea and External Disease clinic in 1990 at the age of 72 years, when she complained of visual disturbances. The patient's best corrected visual acuity at the initial examination was 6/60 in the right eye and 6/18 in the left eye. Slit lamp biomicroscopic examination showed discrete dot opacities in the centre of both corneas with thick lattice lines inbetween and the condition was diagnosed as atypical form of granular comeal dystrophy (Figure 4.10). The patient also had cataract but no other ocular abnormalities were present. Extra-capsular cataract extraction and intra-ocular lens implantation were performed in the right eye in 1990, where right eye visual acuity improved to 6/36 unaided. Penetrating keratoplasty in the right eye was then performed in 1991 which further improved her visual acuity to 6/18 unaided. Case III-3 (the proband’s son): was referred to the clinic in 1996 at the age of 38 years, when he complained of bilateral decrease in visual acuity. He had been diagnosed as having a possible comeal dystrophy when he attended casualty for removal of a foreign body in 1980. The patient's unaided visual acuity was 6/9 in either eye improving to 6/6 with pinhole. Slit lamp examination revealed discrete granular deposits in the anterior stromal layer and star like deposits which were typical of ACD. In 1994 the patient’s visual acuity has deteriorated and right Eximer laser phototheraputic keratectomy was performed which improved his vision. 150 Results Case III-l (the proband’s daughter): 25-year-old with large number of central granules but no lattice changes revealed by slit lamp examination of her cornea. The patient’s unaided visual acuity was 6/9 in both eyes and she had no keratoplasty. Case II 5 (the proband’s sister): clinical examination by slit lamp biomicroscopy revealed no changes in her corneas. Familv C2 (see figure 4.8) The second family consisted of 90-year-old proband (case 1-2) and 63-year-old son (case II-1). Slit lamp examination of their corneas showed a typical picture of ACD. They had no history of keratoplasty. Examination of the proband’s daughter (case II-2) revealed bilateral normal corneas. All members of both families were bom in England and both families’ origin has been traced to the UK. They have no relatives of Italian origin, making them the first English families to be reported with ACD. 4.3.1.4 LCDI patients Three sporadic subjects and 2 subjects who are members of a three generation British family were studied (family D, Figure 4.9). Phenotypes of LCD patients Slit lamp examination of the proband from the studied family and the sporadic subjects revealed bilaterally symmetrical comeal opacities that were characterised by numerous translucent fine lattice lines and associated with white dots and faint haze in the superficial and middle layers of the stroma (Figure 4.11). Full medical history was obtained and no clinically apparent systemic manifestations of amyloidosis were detected. These criteria were diagnostic of LCDI. 151 Results Fam ily C-1 T r O r t O t O III-3 IV Family C2 / / I - 2 \ Ô II-1 Figure 4.8 Pedigrees of ACD families (C1 and C2) participated in the study. 152 Results F am ily D o ■ 0 - T - — III-7 III-9 IV Ô O V Family E I - l 1-2 II-1 II-2 II-3 Figure 4.9 Pedigrees of LCD I (family D) and the Bangladeshi (family E) participated in the study. 153 Results Figure 4.10 Slit lamp photographs of ACD in the right (a) and left (b) corneas of the proband from family Cl with R124H mutation showing fusiform, branching and white bread crumb deposits. Figure 4.11 Slit lamp photograph (retroillumination) of LCDI in the proband of family D showing the delicate branching refractile lines occupying the central part of the cornea (arrows). 154 Results Table 4.1 Clinical criteria of CDBII, CDGGI, and LCDI patients participated in the study Patients Age Sex Age of Best Treatment M Comments (Yrs) corrected VA onset Visual OD OS OD OS Deterio ration i CDBII patients Family A1 Case II-5 55 F 12 39 6/18 6/36 LK, PK LK,PK +ve Case II-7 50 M 9 35 6/36 6/60 LK, PTK LK, PTK +ve Cataract and glaucoma Case III 4 45 F 6 18 6/9 6/9 PK PK -ve Case IV-1 26 M 12 22 6/18 6/18 LK -ve Case rV-2 23 FII 15 6/9 6/18 LK -ve Keratoconus Familv A2 Case III-l 76 F 7 20 6/36 6/60 PTK PTK -ve Case III-3 72 F 10 23 6/36 6/36 PK PK -ve Case III-6 61 F 9 23 6/12 6/24 LK -ve Familv A3 Case II-3 70 F 6 17 6/24 6/24 PTK -ve Case III-l 52 F 8 19 6/9 6/18 LK -ve Familv A4 Case IV-1 55 F 19 25 6/60 6/60 LK,PK LK,PK +ve Case rV-5 57 F 13 22 6/18 6/36 PTK -ve Familv A5 Case II-8 55 M 15 18 6/24 6/36 LK LK -ve Snoradic natients Case 1 52 M 8 20 6/12 6/12 LK -ve Case 2 37 M 7 12 6/9 6/9 Case 3 44 M 10 17 6/9 6/12 LK -ve Case 4 39 M 4 16 6/9 6/9 CDGGI patients Familv B1 Case II-3 75 F 35 50 6/60 6/60 PK -ve Case II-7 65 M 38 55 6/36 6/60 PKPK +ve Case III-4 48 M 34 42 6/9 6/6 Case III-6 46 F 35 44 6/18 6/6 PK -ve Case IV-5 16 F 6/6 6/6 Familv B2 Case ni-3 45 F 30 42 6/24 6/24 Sporadic patients Case 1 44 M 38 43 6/12 6/9 PK +ve Case 2 45 M 33 6/6 6/6 Case 3 50 F 40 46 6/12 6/24 PK -ve LCD patients Familv D Case III-6 65 M 5 15 6/12 6/9 PK Case III-8 40 M 10 20 6/18 6/24 PK PK Snoradic patients Case 1 31 F 3 12 6/24 6/24 PK PK +ve Case 2 65 F 7 16 6/24 6/36 PK -ve Case 3 39 F 14 23 6/12 6/9 155 Results 4.3.1.5 A Bangladeshi Family with CDBI and LCDI The pedigree (family E) is shown in (Figure 4.9) Clinical data and phenotypes offamily members Case ÏÏ-2 (the proband): A 10 year old boy who was referred to Moorfields Eye Hospital complaining of intermittent ocular irritation and mild visual impairment. The episodes of ocular irritation began at the age of 5. The patient's best-corrected visual acuity was 6/18 in both eyes. His intra-ocular pressure (lOP) was 12 mm Hg in the right eye and 11 mm Hg in the left eye. He had no history of eye surgery. Slit lamp biomicroscopic examination revealed confluent dense geographic subepithelial comeal opacities. The opacities were bilaterally asymmetrical and heterogeneous, and more prominent in the peripheral cornea as shown in (Figure 4.12a). Case II-3 (the proband's brother): Affected with the same phenotypic presentation as the proband, as determined by slit lamp examination (Figure 4.12b). Case I-l (the proband's mother): A 42 year old female had been examined at the Cornea and External disease clinic at the age of 40 when she complained of visual disturbance. She had a past history of photophobia, lacrimation, and chronic ocular irritation since childhood. Her visual acuity gradually decreased and on her first visit to the clinic, visual acuity was 6/24 in both eyes. Her lOP was 14 mm Hg in the right eye and 12 mm Hg in the left eye. She had no history of keratoplasty. Slit lamp examination of both eyes showed dense subepithelial opacity with fine lattice lines in the superficial and midstromal layers of both corneas (Figure 4.12c). Full medical history was obtained and no systemic manifestations of amyloidosis were detected. Case 1-2 (the proband's father) and Case II-l (the proband's brother): were unaffected as determined by the slit lamp biomicroscopic examination. 156 Results / V Figure 4.12 Slit lamp photographs of the affected idividuals in the Bangladeshi family, (a) the proband’s right eye with CDBI showing geographic subepithelial deposits, (b) right eye of the proband’s brother showing diffuse superficial comeal haze with superimposed amorphous deposits, (c) left eye of the proband’s mother showing the lattice lines typical of LCDI. 157 Results 4.3.2 Methods 4.3.2.1 DNA extraction After obtaining a blood sample from each participant, the genomic DNA was extracted from the peripheral blood leukocytes (see section 2.1.1). 4.3.2.2 PCR amplification OÎBIGH3 gene BIGH3 exons were amplified by PCR according to the standard protocol (section 2.2). The primers used were designed using the PrimerS Output (www- genome.wi.mit.edu/cgi-bin/primer) program. Primers length, product size, sequence and PCR conditions are listed in table 4.2. 4.3.2 3 Mutation detection The PCR-heteroduplex technique (section 2.6.1) was used to assess the coding sequences and the splice acceptor and donor sites in the introns immediately adjacent to the exons of the BIGH3 gene. Exons that displayed double bands on heteroduplex were sequenced using a direct sequencing method. Amplified DNA was purified using a PCR purification kit (section 2.4.2) and both sense and antisense strands were sequenced (section 2.6.3). Nucleotide sequences for coding regions were compared with the amino acid sequence of the BIGH3 human cDNA (GenBank accession number AC005219 for exons 12, 13, 14, 15, 16 and number AC004503 for the remaining exons). 158 Results Table 4.2 Primers designed for amplification of BIGH3 gene Exon No Length Tm Product M gC l; Gc% Sequence size Exon 4 forward 20 63.50 310 bp 1.5 mM 50.00 tccctccttctgtcttctgc Exon 4 reverse 20 63.90 50.00 agactcccattcatcatgcc Exon 5 forward 20 65.10 330 bp 1. 5 mM 50.00 tgtttcccagagttgcaagg Exon 5 reverse 21 65.10 4?.62 ccacacatggaacagaaatgg Exon 6 forward 18 64.30 310 bp 1.5 mM 61.11 ggggctttgggactatgc Exon 6 reverse 21 64.30 5?.14 gcagaagagttcctgctaggc Exon? forward 20 64.10 315 bp 2 mM 55.00 accagtgaagctgtgtgtgc Exon? reverse 20 63.40 50.00 tggcaggtggtatgttcatc Exon 8 forward 20 59.50 313 bp 1.5 mM 60.00 ggaccctgacttgacctgag Exon 8 Reverse 20 58.50 50.00 cacaaaggatggcagaagag Exon 9 forward 20 61.36 320 bp 1.5 mM 55.00 cctgctgatgtgtgtcatgc Exon9 reverse 20 56.69 55.00 gggtgctgtaaatcggagag Exon 10 forward 26 61.50 310 bp 1.5 mM 34.62 tttatctctcatcactctcttcattg Exon 10 reverse 26 61.00 34.62 tttttacactaatacaagtcccacag Exon 11 forward 22 64.30 330 bp 1.5 mM 50.00 ggataatgaccctgctacatgc Exon 11 reverse 20 62.00 50.00 tccccaaggtagaagaaagc Exon 12 forward 20 63.50 330 bp 1.5 mM 45.00 tcaatccttgatgtgccaac Exon 12 reverse 21 62.90 42.80 aaaatacctctcagcgtggtg Exon 13 forward 20 64.10 290 bp 1.5 mM 55.00 catcctgggggtgagatatg Exon 13 reverse 18 64.90 66.6? ctctgggccctccttgac Exon 14 forward 26 61.?0 31? bp 2 mM 34.62 tttagaacagtttttcttctctccac Exon 14 reverse 24 63.00 33.33 tccatctcaaaaacaaagaacaag Exon 15 forward 20 63.00 250 bp 1.5 mM 60.00 gctggagaggctcctctatg Exon 15 reverse 21 64.20 52.38 ccctcagtcacggttgttatg Exon 16 forward 18 66.60 200 bp 1.5 mM 61.11 agcagatggcaggcttgg Exon 16 reverse 19 65.90 5?.89 tgagtaggggtggcaatgg 159 Results 4.4 Results 4.4.1 Identification of R555Q in CDBII patients Sequence analysis of exon 12 from the affected individuals of all families and the sporadic patients participating in the study revealed a heterozygous G to A transition at the second nucleotide position of codon 555, which resulted in an R555Q substitution (Figure 4.13 A and B). None of the 14 unaffected family members or 50 control Caucasian individuals possessed the mutation as determined by sequencing. 4.4.2 Identification of R555W mutation in CDGGI patients Sequence analysis of the probands from the two CDGGI families (B1 and B2) and the sporadic subjects in the study showed a single heterozygous base pair transition (C—>T) at nucleotide position 1710. This substitution converts an arginine at codon 555 into tryptophan (R555W, Figure 4.13C and D). The mutation creates a BstX I restriction site (Table 4.3) and was used for cosegregation study in all members of both families. All affected members showed undigested product of the expected size (330 bp) for the normal allele plus two digest products (210 bp and 120 bp) from the mutated allele. Unaffected members gave only one band at the expected size for the wild allele (330 bp) because the enzyme recognised only the mutant allele (Figure 4.14). The mutation specific BstX 1 digestion pattern was not detected in 50 control individuals. 160 A A A A A A A CGG G A A C A A A A A A A A CGGG A AC A ZIÊ <) 190 i) 190 C ACG G ACUULACG G AG A A R C G G fl C H G C R C G G R G R CACG G AC CGCACGG AG A A 10( Figure 4.13 Electropherograms of the B1GH3 gene. (A) the proband of family AI with CDBII showing a heterozygous G to A transition at codon 555, (CGG^CAG, R555Q).(B) the proband's brother from family AI showing the wild type allele (C) the proband of family B1 with CDGGI displaying a heterozygous C to T transition at codon 555,(CGG^TGG, R555W). (D) the proband's unaffected son from family BI displaying the wild type allele. (E) the proband of family Cl with ACD showing a heterozygous G to A transition at ON codon 124, (CGC->CAC, RI24H). (F) the proband's daughter from family Cl showing normal sequence. (G) the proband of family D with LCDI depicting heterozygous single base pair transition (CGC^TGC, RI24C). (H) the proband’s unaffected relative from family D depicting the wild type allele. Results Table 4.3 Restriction enzymes used for confirmation of mutations and polymorphism identified Enzyme name Recognition site Mutation and swquence alteration BstX I 5’...CCANNNNN^NTGG...3’ R555W Ava II 5’...G^G(A/T)CC...3’ RI24C 5’...CG"G(A/T)CGG...3’ RI24H Dde I 5 ... C-TNAG...3' IVS7+46C^T S g | M 1 2 Î 4 5 6 7 8 9 10 II 12 13 14 15 M Figure 4.14 Cosegregation study of members of family BI using BstX I restriction enzyme. Lane I is undigested PCR product (330 bp). Unaffected individuals (lanes 2, 3, 4, 5, 10, 12, 13, 14) show undigested product (one product, 330 bp) representing the wild type allele. Affected individuals (lanes 6, 7, 8, 9, II) show three digest products, one for the wild type allele and two (210 and 120 bp) for the mutant allele. M is the (()XI74 RF DNA Hae\\\ marker. Lane 15 contains no sample. 162 Results 4.4.3 British families with ACD 4.4.3.1 Histopathological study Histopathological examination of the probands’ comeal specimen from family BI after keratoplasty showed an amorphous eosinophilic deposit in the anterior stroma which stains positively with Masson’s trichrome stain and occasional areas of amyloid deposition in the deep stroma which stains positively with Congo red stain (Figure 4.15). These criteria were classical for ACD. 4.4 3.2 R124H is the underlying genetic defect in ACD patients Direct sequence analysis of BIGH3 exon 4 in the probands from both families in this study revealed a single heterozygous base pair transition at nucleotide position 418 (G to A) that converts an arginine at codon 124 into a histidine (Figure 4.13E and F). The unaffected members of both families did not have the mutation as determined by restriction analysis in which the mutation abolishes the recognition site of CSPl (Table 4.3). The three affected members from family Cl and the two affected individuals from family C2 produced one undigested product of the expected size (310 bp) for the mutant allele plus two restriction digest products (190 bp and 110 bp) from the normal allele. Unaffected members from both families gave two restriction digest fragments (190 bp and 110 bp) (Figure 4.16). The mutation was not detected in 50 control individuals. BIGH3 exon 12 sequencing revealed a polymorphism with T to C transition at nucleotide position 1620 in the proband from family Cl (Table 4.4). This polymorphism does not change the amino acid sequence of the encoded protein, and it had been reported previously as a polymorphism (Korvatska et a/., 1998, Schmitt-Bemard et a/., 2000). 163 Results Figure 4.15 Light microscopic study of corneal specimen from the proband of family Cl. (a) subepithelial fusiform and anterior stromal deposits stained pink with Congo red (XIO). (b) focal accumulation of hyaline stained bright red with Massone’s trichrome; subepithelial hyaline material and connective tissue (X20). (c) characteristic birefringence seen with polarised light in amyloid deposits (X20). 164 Results O Ml 2 3 4 Figure 4.16 Cosegregation study with CSP\ restriction enzyme. Lane I is undigested PCR product (310 bp). The two affected individuals (Lanes 2 and 3) from family C2 show one undigested product of the expected size (310 bp) for the mutant allele plus two restriction digest products (190 and 110 bp) from the normal allele. Unaffected member (Lane 4) show two restriction digest fragments (190 and 110 bp). M is the (|)XI74 RF D N A Hae III marker. 165 Results 4.4.4 R124C is the causative molecular defect in LCDI patients BIGH3 exon 4 direct sequencing analysis in the proband from the family and the sporadic individuals showed a single heterozygous base pair transition at nucleotide position 417 (C to T) which converts an arginine at codon 124 into a cysteine (Figure 4.13G and H). This mutation abolishes the Ava II recognition site. Thus, on analysis the affected members DNA produced one undigested product of the expected size (310 bp) for the mutant allele plus two restriction digest products (190 bp and 110 bp) from the normal allele in unaffected individuals DNA from the studied family and 50 control. 4.4.5 Single point mutation in the BIGH3 gene responsible for intrafamilial phenotypic heterogeneity in a Bangladeshi family 4.4.5.1 Mutation analysis Mutation screening of the BIGH3 gene by heteroduplex analysis gave positive results for exons 4 (Figure 4.17), 11, and 12 and for introns 4 and 7 in all affected members with the exception of intron 7 which was positive for the mother only. No changes were detected in other exons. Direct sequencing of the proband (II-2) and his mother (I-l) for exon 4 revealed a single heterozygous base pair transition at nucleotide position 417 (C to T), which converts an arginine at codon 124 into cysteine. This mutation has previously been identified in LCDI (Munier et al, 1997). Sequence analysis of the other four PCR products that were aberrantly migrating as identified by heteroduplex analysis showed polymorphisms which did not change the amino acid sequence of the protein. The polymorphisms were identified in intron 4 (IVS4-16C->T), in intron 7 (IVS7+46T^C), exon 11 at nucleotide 1416 with C ^ T 166 Results change and in exon 12 at nucleotide 1620 with T-»C change. Three of these polymorphisms have previously been identified (Korvatska et al., 1998; Schmitt- Bemard et al., 2000), however the polymorphism in intron 7 was novel (Table 4.4). 4 4.5.2 Cosegregation study and novel polymorphism appraisal Cosegregation study by restriction analysis showed that the unaffected members of the family did not have the mutation as it abolishes the recognition site of Ava II (Table 4.3). The three affected members gave one undigested product of the expected size (31 Obp) for the mutant allele plus two restriction digest products (190bp and 11 Obp) from the normal allele. Unaffected members gave two restriction digest fragments (190bp and 11 Obp) as shown in (Figure 4.18). The novel polymorphism was detected in the control population at a frequency of 92 out of 100 chromosomes screened using Dde I restriction enzyme (Figure 4.19). 167 Results --wt* W# ## Ml 2 34567 89 10 Figure 4.17 Ethidium bromide stained gel depicting heteroduplex band patterns produced by codon 124 mutation in BIGH3 exon 4. Lanes 2, 3, and 6 show heteroduplexes of the proband of family D and the proband and the proband’s mother of family E. Lane 1 is exon 4 PCR product. The remaining lanes represent unaffected individuals. 168 Results I-l 1-2 11-2 11-3 1 ■ # i a m W f ## m m ■mu Mr w M 1 2 3 4 5 6 7 M Figure 4.18 Co segregation study using Ava 11 restriction digest. Lane 1: undigested 31 Obp PCR product; lanes 2 and 5: Ava 11 digested product from unaffected members (11-1) and (1-2) respectively showing the digested wild type alleles (190bp and 1 lObp); lanes 3, 4, and 6: Ava 11 digested product from affected members (1-1), (11-2), (11-3) respectively carrying R124C mutation showing the undigested mutant allele (31 Obp). Lane 7 contains no sample. M is (j)X174 RF DNA Hae\\\ m a rk e r. 169 Results M î 2 Figure 4.19 Dde\ restriction analysis of intron 7 polymorphism. Lane 1 shows undigested PCR product (315bp). Lanes 2 and 3 represent digested PCR products (215 bp and lOObp). Lanes 4, 5, and 6 show one digested product (215 bp). M is the (j)X174 R F D N A Hae\\\ m ark er. 170 Results 4.4.6 Polymorphisms identified Seven nucleotide substitutions were identified (Table 4.4) that did not change the amino acid sequence of the encoded protein, and did not cosegregate with the disease. Three of these polymorphisms were identical with those that other investigators had reported (Korvatska et aL, 1998; Schmitt-Bemard et al., 2000). However the G ^A change in intron 4, and the changes in introns 7, 13, and 14 were novel ([Figure 4.20). Table 4.4 Polymorphisms identified in B1GH3 g e n e Location Nucleotide Codon Intron 4 IVS4-16C->T Intron 4 IV S4+I9G ^A Intron 7 IV S7+46C ^T Exon 11 1416C/T L472L Exon 12 1620T/C F540F Intron 13 IV SI3-9G ^T Intron 14 IVSI4+55A^T 171 GGGÛACTNTTATGGC CA T TTAT T TA TA CA TTTAT T AGCTAÛANiGAGCCCAA 220 !30 1 3 0 220 230 ü L L : N Figure 4.20 Electropherograms depicting polymorphisms identified in the BIGH3 gene, (a) intron 4 polymorphism with heterozygous C ^ T change, (b) exon 12 polymorphism with homozygous T ^C transition at nucleotide position 1620. (c) intron 14 polymorphism with homozygous A ^ T change, (d) intron 7 polymorphism with heterozygous C-^T change, (e) Exon 11 polymorphism with heterozygous C ^ T I K) transition at nucleotide position 1416 (f) Intron 4 polymorphism with heterozygous G-^A change, (g) Intron 13 polymorphism with heterozygous G->T change. Results 4.5 Discussion Each form of 5q31-linked autosomal dominant comeal dystrophy has been associated with a different mutation (Table 4.5). Current molecular knowledge of comeal dystrophies suggests the existence of a genotype-phenotype correlation, because each phenotype of comeal dystrophy seems to be linked to specific mutations. Table 4 .5 BIGH3 mutations reported in different phenotypes o f comeal dystrophy Phenotype Mutation References CDB R555Q Munier et at., 1997; Korvatska et al., 1998; Okada et al., 1998a RI24L Okada et al., 1998a; Mashima et al., 1999 AF540 Rozzo et al., 1998 G623D Afshari et al., 2001 CDGGI R555W Munier et al., 1997; Korvatska et al., 1998; Okada et al., 1998b R124S Stewart et al., 1999a LCDI R124C Munier et al., 1997; Korvatska et al., 1998; Gupta et al., 1998 L518P Endo et al., 1999 LCDIIIA P501T Yamamoto et al., 1998 N622H Stewart et al., 1999b H626A Stewart et al., 1999b A546T Dighiero et al., 2000 LCDI/IIIA H626R Stewart et al., 1999b; Schmitt-Bemard et al., 2000 NVP 629-630 ins Schmitt-Bemard et al., 2000 LCDIV L527R Fujiki et al., 1998a, Hirano et al., 2001 ACD R124H Munier et al., 1997; Fujiki et al., 1998b; Korvatska et al., 1998 In this study, the BIGH3 gene was analysed in 36 patients affected by four types of comeal dystrophy from different ethnic origins. Four different missense mutations were detected, R124C for LCDI, R124H for ACD, R555Q for CDBII, and R555W for CDGGI (Table 4.6) which were the same as those reported previously (Munier et al; 1997; Mashima et al; 1997; Gupta et al; 1998). In addition an R124C mutation was identified as the causative BIGH3 mutation in a Bangladeshi family that displayed intrafamilial phenotypic heterogeneity of LCDI and CDBI. 173 Results Table 4.6 Mutations identified in the BIGH3 gene Clinical phenotype Mutation CDBII R555Q CDGGI R555W ACD R555W ACD R124H LCDI RI24C CDBI and LCDI R124C Initially the BIGH3 gene was analysed in CDBII patients with the detection of an R555Q heterozygous mutation. Thus these findings confirm that the honeycomb type of CDB or CDBII is caused by the R555Q BIGH3 mutation. The R124L BIGH3 mutation has been reported to cause the geographic type of CDB or CDBI (Okada et al., 1998a; Mashima et al., 1999) which was not diagnosed in the patients of this study. However, recent reports suggest that the R555Q and RI24L BIGH3 mutations do not account for all forms of Bowman’s layer dystrophies. Other BIGH3 related mutations may cause similar phenotypes, including a AF540 mutation described in a Sardinian family (Rozzo et al., 1998), and G623D reported in a geographic form of CDB (Afshari et ah, 2001). In addition other phenotype of CDB which has also been designated as CDBII or Thiel- Behnke dystrophy, show genetic linkage to chromosome 10q24 (Yee et al., 1997). One family with CDGGI (family BI) was a four generation British kindred in which slit lamp biomicroscopy of the five affected members revealed comeal changes which were progressive with age. These results are consistent with that of Moller (1989b) who reported, in a study of 140 patients, that all affected members in a given family showed the same distribution of comeal granules and the comeal appearances of patients at age groups parallel to those in the present study. The development of CDGGI was suggested to be caused by the mutated KE which is secreted by stromal keratocytes (Akiya and Brown, 1970; Witschel and sundmacher, 1979; Wittebol et al., 1987b). However several 174 Results lines of evidence now implicate an epithelial genesis. For example, the early stages of CDGGI show fine epithelial and subepithelial deposits. A superficial varaiant of CDGGI has also been described with subepithelial and superficial stromal deposits (Haddad et al, 1977; Rodrigues et al, 1983). In addition graft recurrences following PK for CDGGI occur initially in the subepithelial region and only later the stroma is affected (Akhtar et al., 1999). Also it has been reported that KE is of such size that it could diffuse through the comeal stroma from a more superficial source, thereby causing a stromal dystrophy but of epithelial origin (Akhtar et al, 1999). These lines of evidence suggest that the stromal deposits of CDGGI and perhaps the other stromal dystrophies related to the BIGH3 gene are derived from comeal epithelial cells. Molecular analysis of the BIGH3 gene revealed heterozygous R555W point mutation in all affected members which was previously reported with the typical form of CDGGI (Munier et al, 1997; Korvatska et al., 1998; Okada et al, 1998b). The same mutation was detected in the proband of family B2 of Jamaican origin although the clinical phenotype of this patient was similar to that of ACD. Two pedigrees of ACD were identified in the UK, thus the condition may occur in any population. The younger individual (case III-1 in family Cl) demonstrates predominantly granular stromal opacities and the appearance of lattice changes occurs gradually, starting later in life, increasing with age. Thus the dystrophic process progresses with age. This observation supports and extends those of previous reports (Folberg et al, 1988; Sassani et al, 1992; Holland et al., 1992; Rosenwasser et al, 1993; Konishi et al, 1997). 175 Results Molecular genetic analysis showed that all five ACD patients in both families had a heterozygous R124H BIGH3 gene mutation, which is identical to the mutation reported previously for ACD (Munier et al., 1997). It appears that the unique phenotype of ACD is caused by this particular amino acid change. This fact is supported by the finding that different amino acid changes in the same position result in different phenotypes [R124C results in LCD! (Munier et al., 1997), R124L results in the geographic form of CDB (Okada et al., 1998a), and R124S results in atypical form of CDGGI (Stewart et al., 1999a) and even the homozygous form of the same change (R124H) results in a severe variant of CDGGI which is characterised by juvenile-onset and confluent superficial discrete opacity (Okada et al., 1998b). Thus, the heterozygous R124H mutation of the B1GH3 gene is particularly linked to ACD phenotype however, in this study an ACD family originating from Jamaica presents a BIGH3 mutation in a different codon (R555W). Moreover, Stewart et al. (1999a) reported atypical form of CDGGI caused by heterozygous R124H mutation. In conclusion, gene mutations in ACD families may reside in codons outside 124. Lastly, a family from Bangladesh with two different phenotypes (LCDI and CDBI) is described. The 2 affected offspring have the geographic form of CDB, or CDBI whereas the mother has typical LCDI phenotype. All affected members of the family have an R124C mutation in the BIGH3 gene and not R555Q mutation previously found in Caucasians with CDBII (Munier et al., 1997) or the R124L mutation found in Japanese individuals with CDBI (Mashima et al., 1999). The same mutation was found in Caucasian (Munier et al., 1997) and Japanese patients with LCDI (Mashima et al.. 176 Results 1997). This suggests that this mutation occurs independently in several ethnic groups (Korvatska g/ a/., 1998). It has been postulated that CDBI may be a variant of LCD, with a mild to non-existent deep stromal pathology (King and Geeraets, 1969). These findings, based on clinical and genetic studies of the family support this theory. It is possible that the mother's phenotype originally resembled that of CDBI and that the development of lattice lines in her cornea is secondary to chronic irritation produced by numerous episodes of superficial erosions. A previous report on a four generation family, in which histopathological examination of the affected members confirmed the presence of both CDBI and LCDI comeal dystrophy within the same family, strongly supports the findings in this study (King and Geeraets, 1969). Another possible explanation for the observed variant phenotype is the effect of a modifier locus in which other genes affect the phenotypic expression of the mutated BIGH3 gene. For example, there are well documented examples of intrafamilial variable expressivity for deafiiess in humans, which have been attributed to a modifier gene (Friedman et al., 2000). Also, the partial penetrance phenotype seen in retinitis pigmentosa was attributed to a modifier locus in close proximity to the RP11 gene on 19ql3.4 (Mc-Gee et al., 1997). Such modifier genes can act to suppress or enhance the mutant phenotype (Marin et al., 2000). An alternative hypothesis is that single nucleotide polymorphisms (SNPs) associated with diseases could explain intra-familial clinical heterogeneity. The polymorphism in intron 7 is seen in the mother with LCD type I and is not seen in her offspring with CDBI. 177 Results The possibility of CDBI being a variant of LCDI is reinforced by the study of this family. This is the first genetic study to show that a single mutation could result in CDBI and LCDI phenotypes within a single family. In general identical mutations can result in different phenotypic expression. This may be due to the genetic background of the individual within a family and between families. Thus modifier loci may have a role in disease presentation. All the mutations detected occurred in the CpG dinucleotide of two arginine codons, RI24 and R555. It is interesting to note that both amyloid-related phenotypes (LCDI and ACD) involve an R124 mutation and that both nonamyloid-related phenotypes (CDBII and CDGGI) involve an R555 mutation. Hence, codons 124 and 555 are mutation hot spots and are associated with different types of deposition, amyloid for 124 and non amyloid for 555 (Korvatska et al., 1998). The mutations described for LCDI and ACD, which are characterised by comeal amyloid deposits, alter the same codon R124 which is conserved in human, mouse, and rabbit and is located just before the first of the four internal repeats. In addition, these mutations abolish a putative phosphorylation site that could be implicated in the tertiary structure of KE leading to amyloid conversion. Meanwhile, the mutations identified for CDBII and CDGGI are located in the fourth internal repeat. They alter codon 555, which is not conserved between species, but belongs to a predicted coiled-coil domain that is highly conserved in mouse and rabbit KE, as well as in mouse and human OSF-2 (Gregory er a/., 1995). The role of KE protein in the pathogenesis of amyloid and non amyloid deposits has been proved by the fact that non amyloid deposits, as in CDGGI, the R555W mutation 178 Results results in misfolding of the whole KE protein followed by its aggregation in deposits (Korvatska et al; 1999). The CDGGI corneas have been shown to contain an excess of full length KE protein (Klintworth et al., 1998). Thus, KE is the major, if not the sole, component of the non amyloid deposits in CDGGI. Again, R555Q in CDBII could interrupt a predicted coiled-coil domain allowing precipitation of dimers and rod forms (Blake er a/., 1995). However in the amyloid deposits of LCDI and ACD corneas the KE protein was found in a truncated form, which means that proteolysis proceeds or accompanies amyloid conversion (Korvatska et al., 1999) and this supports the biochemical study which showed the presence of KE in a degraded form in LCD corneas (Takacs et al., 1998). Further explanation is required for the associated granular changes in ACD. Other comeal components may contribute to the aggregates such as lectin-positive carbohydrate in LCDI, CDGGI and ACD (Folberg et al., 1988), amyloid P protein in LCDI (Rodrigues and Krachmer, 1988) and phospholipids in CDGGI (Rosenwasser et al., 1993). Mutant KE may bind these carbohydrate and lipid moieties aberrantly so they become incorporated in the deposits. Thus the aggregate structure could be related to both the specific mutation and availability of other components in the matrix and this could explain the wide variation in clinical phenotype in some ACD families. Recently it has been reported that transient over-expression of BIGH3 LCD and CDGGI mutations in HeLa cells induces apoptosis. Based on these results, comeal deposits could be induced or increased by apoptosis and inability to clean up the mutated KE (Schorderet et al., 2000b). Analysis of the predicted secondary stmcture of the native and mutated KE (http://ibcp.fr/NPSA) did not reveal any particular stmcture to enable differentiation of 179 Results amyloid from non amyloid-inducing mutations. Thus the BIGH3 gene mutation may not be the only factor responsible for amyloid fibrils precipitation. Other factors such as misfloding of the protein, increased hydrophobicity, and ionic interactions may be involved (Glennner, 1980, Argiles, 1996). In summary, the 5q31-linked comeal dystrophies in patients from different ethnic origins are due to recurrent mutations at two specific codons of the BIGH3 gene. The identification of mutations has enabled the morphological manifestations to be correlated to some extent with the underlying genetic defect. The phenotype-genotype correlation with mutations in the BIGH3 gene is strong, however exceptions do arise for example through genetic heterogeneity, familial phenotypic heterogeneity and diagnosis or other factors. The genetic classification, therefore, is much more accurate than the anatomical or the clinical description for the diagnosis of such disorders. This information will be useful for future genetic counselling, as well as functional studies to investigate how the mutations result in different phenotypes of comeal dystrophy. 180 Results Chapter 5 Identification of novel mutations in the carbohydrate sulfotransferase 6 gene (CHST6) causing MCD 5.1 Introduction 5.1.1 Macular corneal dystrophy (MCD) Macular comeal dystrophy (MCD, OMIM 217800) is an autosomal recessive hereditary disorder characterised by abnormal deposits in the comeal stroma, keratocytes and the endothelium (Jones and Zimmerman, 1961; Klintworth and Vogel, 1964; Klintworth, 1982) with progressive stromal clouding and central comeal thinning in both eyes (Klintworth, 1980; Donnenfeld et al., 1986). Like other recessive disorders, it is clinically more severe and occurs often in pedigrees with consanguinity. It may appear to occur as sporadic cases because the heterozygous carriers do not manifest comeal changes (section 1.3.3.4). Although MCD is the least common of the classical stromal dystrophy it represents 10% to 75% of the comeal dystrophies requiring penetrating keratoplasty, depending on the population (Jonasson et al., 1996; Santo et al., 1995). In Europe it accounts for about one third of all stromal comeal dystrophies requiring keratoplasty (Vilchis et al, 1996), also in Saudi Arabia it is the most frequently recorded type of comeal dystrophy (Al- Faran and Tabbara, 1991). In Iceland MCD is very common, being the most frequent condition necessitating penetrating keratoplasty and accounting for almost one third of all comeal grafts performed (Jonasson et al., 1996). 181 Results MCD is characterised by symmetrical changes which present in the first decade of life starting with a diffuse, fine, superficial clouding in the central stroma extending to the periphery and usually involving the entire thickness of the cornea by the second decade of life. These opacities are more superficial and prominent in the central cornea and are deeper and more discrete in the periphery. Multiple irregular, grey-white nodules develop within the haze and may project into the anterior chamber posteriorly or protrude anteriorly, causing surface irregularity (Francois, 1966). Clinical distinction between the early stages of MCD and CDGGI may be difficult because focal, rather than discrete opacities in the superficial stroma appear early in both conditions. However, the very early intervening stromal haze, involvement of the peripheral and deep stroma, decreased comeal thickness, with-the-rule astigmatism, and recessive family history help to distinguish MCD. The autosomal dominant inheritance and sparing of the peripheral 2 or 3 mm of the cornea help to distinguish the late changes of GCD. Similarly, late opacification in LCD may occur, but typical filaments are usually seen peripheral to the opacification, and the peripheral stroma is spared except in the most severe cases (Miller and Krachmer, 1998). MCD can also be distinguished from other disorders involving glycosaminoglycans, such as the systemic mucopolysaccaridosis. In the latter there is a generalized abnormality in the breakdown of the glycosaminoglycan portion of different proteoglycans, resulting in their accumulation and deposition in a variety of tissues. In the systemic mucopolysaccaridosis, the abnormal material accumulates in lysosomal vacuoles, whereas in MCD it accumulates in endoplasmic reticulum. Clinically, in the systemic mucopolysacaridosis, epithelial involvement is prominent, and Descemet’s membrane is usually not affected, in contrast to the findings in MCD (Quigley and Goldberg, 1971). 182 Results 5.1.2 Keratan sulfate (KS) KS is a sulfated glycosaminoglycan that is a prominent constituent of the proteoglycans found in the extracellular matrix of cartilage and comeal stroma. It was first isolated from bovine comeal stroma by Meyer et al. (1953) who subsequently classified KS into comeal KS or KSI which is N-linked to aspargine residues in the core protein and cartilage KS or KSII which is 0-1 inked to Serine or Threonine residues (Meyer, 1970). A third type of KS (Man-O-Ser) has been identified as KSIII (Krusius et a/., 1986). The amount of KS in the comea is more than 10-fold that in cartilage and is 2-4 orders of magnitude greater than KS found in other tissues (Funderburgh et al., 1987). KSPGs are of paramount importance for the maintenance of comeal transparency. Current evidence suggests that the regular collagen lattice of the comea is sustained by several proteoglycans, binding to specific sites on the collagen fibrils to maintain their crucial size and ordered structure (Scott and Haigh, 1988). They also affect comeal transparency by influencing comeal hydration (Scott, 1991). KSPGs are SLPGs which are glycosylated with KS or chodroitin/Dermatan sulfate (CS/DS) to form glycosaminolglycans (GAGs). The side chains of the GAGs are thought to be responsible for the correct inter-fibrillar spacing of the collagen fibrils (Lewis et al., 2000). The structure of KS chains present in the comeal stroma have been shown to consist of the repeating disaccharide N- acetyllactosamine which may be sulfated on C-6 of either or both of the galactose and N-acetylglucosamine residues (Oeben et al., 1987; Tai et al., 1996; Tai et al., 1997). The highly anionic nature of the sulfate moiety of this molecule confers a water-holding ability that is thought to contribute in comeal 183 Results transparency (Hedbys, 1961). Low sulfation has been suggested in scarred comea, wound comea and keratoconus-afifected comea (Funderburgh et a/., 1988; Funderburgh et al., 1989; Funderburgh et al., 1991) and increasing sulfation of KSPGs occurs as the comea acquires transparency during development (Hart, 1976; Nakazawa et al., 1995). In normal comea, high levels of sulfated KS were detected in the stroma. Bowman’s layer, and Descemet’s membrane and low levels in the keratocytes, epithelium and endothelium. On the other hand the abnormal or the unsulfated form of KS (N- acetyllactosamine) was present in small amounts or not present at all in normal comea (Thonar et al., 1985 ; Klintworth et al., 1986). 5.1.3 Immunophenotypic classification of MCD Monoclonal antibodies have been raised against proteoglycan core protein isolated after chondroitinase digestion of human articular cartilage proteoglycan monomer. Characterisation of one of these monoclonal antibodies, (1/20/5-D-4), indicated that it specifically recognised an antigenic determinant in the polysaccharide moiety of both comeal and skeletal KS (Caterson et al., 1993). Enzyme-linked immunosorbent assay (ELISA) using this antibody has been used to measure the concentration of normal sulfated KS in MCD. ELISA showed that this sulfated epitope was present at less than 1 % of the normal concentration (normal individuals without eye or joint disease have semm KS levels of 112-617 ng/ml with mean level of 241 ng/ml) (Klintworth et al., 1986; Thonar et al., 1986; Jonasson et al., 1996; Klintworth et al., 1997). Based on measurement of sulfated KS with ELISA (Donnenfeld et al., 1986) and an immunohistochemical evaluation of the comeal tissue (Hassell et al., 1985), MCD has been subdivided into three immunophenotypes I, lA, and II. In type I, neither the comea 184 Results nor the serum contains appreciable levels of sulfated KS while in MCD type n, the less prevalent type, patients have detectable levels of KS in the comea and semm (Yang et al., 1988). In type lA, in Saudi Arabian cases, sulfated KS is absent in the comea and the semm but can be detected in the keratocytes (Klintworth et al., 1997). A case of macular comeal dystrophy with immunophenotype LA has recently been reported in Germany (Cursiefen et al., 2000). This immunohistochemical distinction is not evident in the patient’s clinical presentation. 5.1.4 Pathogenesis of MCD Abnormalities in the metabolism of KS have been implicated in the pathogenesis of MCD (Klintworth and Smith, 1977). A number of studies have shown that MCD comeas fail to synthesise normal KSPGs (LLassell et al., 1980; Klintworth and Smith, 1983) and that the condition is attributed to an error in the synthesis of KS, possibly involving a specific sulfotransferase required for sulfation of the lactosaminoglycan backbone (Hassell etal., 1984). It has been reported that MCD type I comeas synthesise normal dermatan sulfate PG (DSPG) and in place of KSPG, they synthesise lactosaminoglycan-glycoprotein (L-GP). This L-GP has a core protein of similar hydrophobicity, identical glycoconjugates, and nearly similar mass (42 kDa) to the core protein of the KSPG except that it lacks sulfate. This suggested that patients with type I MCD fail to synthesise KS as a result of a defect in a sulfotransferase specific for sulfating lactosaminoglycans (Midura et al., 1990). Absence of sulfate groups, which normally provide almost all the negative charges on the KSPG will produce a dramatic decrease in hydrophilicity and would thereby markedly affect the swelling properties of the PG molecule. This PG will probably also 185 Results be much less soluble and more likely to precipitate (Edward et al., 1990). In MCD type n the primary biochemical defect was attributed to the fact that the DSPG molecule is shorter than that seen in normal individuals, resulting in abnormal packing of collagen fibrils (Midura et al., 1990). KSPGs, derived predominantly fi-om cartilage, are also found in the serum of MCD patients in an unsulfated form, suggesting that MCD might be the only clinical manifestation of a systemic metabolic disorder (Klintworth et al., 1986; Edward et al., 1990). 5.1.5 Histopathology of MCD Histologically, macular dystrophy is characterised by the accumulation of glycosaminoglycans between the stromal lamellae, underneath the epithelium and within the keratocytes and endothelial cells (Jones and Zimmerman, 1961; Snip, 1973). These deposits stain positively with PAS reagent and acid mucopolysacharide stain (alcian blue and colloidal iron). Fragmented Bowman’s condensations are also found in histological sections. Electron microscopy has shown extensive fibrillogranular deposits with a fingerprint pattern characteristic of MCD, in the comeal keratocytes as well as among the extrastromal lamellae (Klintworth and Vogel, 1964; Gamer, 1969b). 5.1.6 Genetics of MCD MCD type I has been linked to an interval of 7 cM on the long arm of chromosome 16 (16q22.3), flanked by the markers D12S512 and D12S518 (Vance etal., 1996). Analysis of several MCD type II families with the same markers as the original study for MCD type I provided a maximum LOD score of 2.5, indicative of linkage, on 16q22.3. Given 186 Results the prior finding that MCD types I and II are linked to this interval, the co-existence of MCD types I and II in a single sibship suggested that these diseases are caused by mutations in the same gene (Liu et al., 1998a). Subsequently, haplotype analysis of 10 Icelandic families led to refinement of the interval for the MCD type I to a very small region, <1 cM and several MCD type II individuals were found to share one of the disease haplotypes (Liu etal., 1998b). Through bioinformatic approach, a new carbohydrate sulfotransferase 6 gene {CHST6) encoding a protein of 395 amino acids was identified (Akama et al., 2000). Multiple alignment analysis disclosed the fact that it is highly homologous to other carbohydrate sulfotransferases, particularly to intestinal N-acetylglucosamine-6-sulfortransferase (I- GlcNAc6ST) or CHST5 (90.6 % identity at the nucleotide level and 89.2% identity at the amino acid level) (Figure 5.1). CHST6 and CHST5 genes are highly homologous to each other both in the coding region and in the 5’ and 3’ UTRs suggesting that these genes have arisen through gene duplication event (Akama et al., 2000). CHST6 encoding comeal GlcNAc6 sulfotransferase (C-GlcNAc6ST) has been identified as the gene responsible for MCD. A number of missense mutations in MCD type I (Akama et al., 2000; Liu et al., 2000; Bao et al, 2001) and type LA patients (Bao et al, 2001) have been reported while in MCD type II patients either large deletions and/or replacements in the upstream region of CHST6 have been found (Akama et al., 2000). Recently a missense mutation in CHST6 has been identified in MCD type II patients (Bao etal., 2001). 187 Results T % II I5I ill i III % :: X. 5 kb CHS75 CHSJÎ (l-GfcNAc6ST| |0GlcNAc5ST| Figure 5.1 Genomic structures of CHST5 and CHST6. Directions of transcription of each gene are indicated (arrows). Exons and coding regions are shown in open and grey boxes respectively. Filled boxes represent homologous regions located upstream of CHST5 and CHST6. Restriction digest map is also shown on the horizontal bar. (Adapted from Akama et al., 2000) 5.2 Aim of the study The aim of this study was to determine the immunophenotype of the probands from 5 families with autosomal recessive MCD and to identify the underlying genetic defect by screening the CHST6 gene. 188 Results 5.3 Patients and methods 5.3.1 Patients Autosomal recessive MCD in all patients was diagnosed on the basis of pedigree structure and the following clinical features: bilateral symmetrical superficial stromal cloudiness studded by small irregular rounded grey-white anterior stromal patches. The diagnosis was confirmed by histopathological examination of comeal buttons Ifom all patients after keratoplasty. Five British families designated A, B, C, D, and E were included in this study (Figure 5.2). The probands from all families were referred to Moorfields Eye Hospital presenting with one or more of the following symptoms: photophobia, blurring of vision, and /or marked diminution of visual acuity. The age of onset, visual deterioration, and keratoplasty, recurrence (3 years after keratoplasty), hi laterality, consanguinity, and family history for the probands from all families are listed in table 5.1. Slit lamp examination of the probands from all families revealed a central, faint, ground- glass-like haze in the superficial stroma, which was best seen by oblique illumination. Within the hazy matrix were multiple, small, pleomorphic, and grey-white opacities with irregular borders. These opacities were more prominent and superficial in the central comea and deeper and more discrete in the peripheral comea. Later in life, the whole comea became opacified and needed keratoplasty. The opaque host comea could be seen after grafting (Figure 5.3a, b, c and d). Histopathological examination of comeal buttons from all probands confirmed the diagnosis of MCD. The comeal epithelium was regular, with Bowman’s layer locally 189 Results disrupted by deposits of pale, amorphous slightly amphophilic materials. Similar materials were also evident in the stoma, where they stained positively with Alcian blue and not with Masson’s trichrome or Congo red. This phenotype was constant in the probands from all families with the exception of one patient (family B). The comeal buttons from this patient showed, unusually prominent anteriorly situated extracellular deposits, an almost totally absent Bowman’s layer, and considerable thinning of the overlying epithelium. 190 Fam ily A Fam ily B Results ^1:3 I 1:4 ^ I 0 I-I 1:1 1:2 II-r II-2 II-3 1 1 -4 ^ II-5 I II-6 II-4 II-5 II-6 II-7 II-3 II-l II-2 III-l o t III-5 III-4 IV-1 IV-2 IV-3 Family C Family D 1-2 1-5 1-6 I-l 1-2 II-2II-1 II-3II-2 II-1 III-2 III-3 III-4III-5 III-l III-2 III-3 IV -1 IV-2 IV-3 IV-4 Family E 1-2 1-3 1-4 II-1 II-2 III-l III-2 III-3 IV -1 rV-2 IV-3 IV-4IV-5 V-4 V-3 V-1 V-2 Figure 5.2 Pedigree structure of MCD familles. 191 Table 5.1 Clinical features of the probands with MCD from all families participated in the study Age (yrs) Family Sex Onset Visual Keratoplasty Recurrence Unilateral Consanguinity Family deterioration or bilateral history A M 12 20 32 Positive Bilateral Positive Negative B F 15 23 38 Negative Bilateral Negative One affected brother CM 20 25 33 Negative Bilateral Negative Negative D F 17 21 31 Negative Bilateral Negative Negative E F 10 16 24 Positive Bilateral Negative One affected brother 192 Results Figure 5.3 Slit lamp photographs of MCD before and after keratoplasty, a an d b: L o w and high magnification respectively; show the right comea of the proband from family E with diffuse opacification of the stroma and multiple irregular grey-white opacities extending to the limbus. c: Early after keratoplasty; d: The proband’s cornea from family C showing clear central graft and totally opacified peripheral host comea in the later stages of the disease. 193 Results 5.3.2 Methods 5.3.2.1 DNA extraction Blood samples were obtained from all subjects involved in the study, and genomic DNA was extracted from peripheral leukocytes as described in section 2.1.1. 5.3.2.2 Mutation detection The coding region of CHST6 was amplified by PCR as described in section 2.2. Primer pairs used are listed in table 5.2. Primers have been designed using published cDNA sequence of CHST6 (GenBank accession number AF219990) amplifying the 5’- coding, the middle coding, the 3’-coding region of the cDNA of CHST6. A mismatch primer pair was also designed, to confirm one of the homozygous mutations (in family A, F107S), which does not alter any known restriction enzyme site. The mismatch primer created a Hinf I site in conjunction with the mutation (Table 5.2). The SSCP has been used for mutation screening (see section 2.6.2) but the results were inconclusive. For direct sequencing analysis, PCR products were purified (section 2.4.2) and sequenced using an automatic fluorescent DNA sequencer (section 2.6.3). Nucleotide sequences were compared with the published cDNA sequence of CHST6. For amino acid numbering, the codon for the initiator methionine starting at nucleotide 693 of the CHST6 cDNA is numbered as codon 1. Southern blot analysis has been performed to screen for rearrangements or deletions. DNA from affected individuals was digested with HindRl and Spel and a Southern blot was prepared. A radioactive probe was generated as described by Akama et al. (2000). Following hybridisation and exposure to X-ray film, no gross rearrangements or deletions were detected. 194 Results 5.3.23 Assay of sulfated KS in serum Inhibition ELISA for KS detection using the monoclonal antibody 5-D-4 (ICN Biochemicals Ltd.) was carried out to assess the presence of KS in serum obtained from the probands of the families participating in the study (section 2.7). Table 5.2 Primers used for amplification and sequencing the coding region of CHST6 CH ST6 region Primer name Primer sequence Tm GC Annealing fragment % temp. size 5 ’-coding region CK71h-intm 5’ gcccctaaccgctgcgctctc ‘3 71 % 68 °C 500 bp CK71h-R1180 5’ ggcttgcacacggcctcgct ‘3 66 °C 70% Middle coding CK71M-F1 5’ gacatggacgtgtttgatgc ‘3 60 °C 50% 62 T 400 bp region 1 CK71M-RI 5’ gcacgatgccgttgtcac ‘3 62 °C 61 % Middle coding CK71M-F2 5’ gctcaacctacgcatcgtg ‘3 60 °C 58% 64 °C 360 bp region 2 CK71M-R2 5’ atccgtgggtgatgttatgg ‘3 60 °C 50% 3 ’-coding region CK71L-F 5 ’ gagccgctggcagaaatc ‘3 62 °C 61 % 62 °C 450 bp CK71L-R 5’ tgcaccatgcactctcctc ‘3 61 °C 58% Mismatch primer Sulphohin-F 5’ ctgtgcgacatggacgagt ‘3 61 °C 58% 60 °C 260 bp Sulphohin-R 5’ caccacgtggctgtaggag ‘3 60 °C 63% 195 Results 5.4 Results 5.4.1 Identification of novel mutations in the coding region of CHST6 Six novel missense mutations were identified within the coding region of CHST6 gene in families with MCD type I (section 5.4.4) (Table 5.3). None of these changes were detected in the control population as determined by restriction enzyme digest on 100 chromosomes confirming that they are likely to be pathogenic mutations. In addition they have been shown to segregate with the disease using sequencing (Figure 5.4) and restriction enzyme analyses (Figure 5.5) In families A, D and E the mutations were homozygous. In femily A, a T1012C transition at the second nucleotide position of codon 107 has detected resulting in phenylalanine (F) to serine (S) substitution (Figure 5.4a and b). In family D, a C1309T transition at the second nucleotide position of codon 206 has been found substituting alanine (A) with valine (V) (Figure 5.4g and h). In family E, two consecutive homozygous changes have been identified. The first is G905T transversion, located at the third nucleotide position of codon 71 changing the amino acid fi*om glutamic acid (E) to aspartic acid (D). The second change, a C906T transition located at the first nucleotide position of codon 72 leads to a proline (P) to serine (S) substitution in the same patient (Figure 5.4i and j). Although the first mutation predicts a change in the amino acid within the same group, the second results in a change of a non-polar to a polar residue. In families B and C two heterozygous changes were identified with a different mutation on each allele (i.e. compound heterozygotic mutations), which could also account for the autosomal recessive inheritance. The first heterozygous change, a C783T transition 196 Results occurred at the first nucleotide position of codon 31 predicting a change of amino acid from proline (P) to serine (S) (Figure 5.4c and d). The other change, a T1291G transversion, has been identified at the second nucleotide position of codon 200 changing the amino acid from leucine (L) to arginine (R)(Figure 5.4e and Q. Table 5.3 Mutations identified within the CHST6 gene in MCD type I families. Family Nucleotide change Type of change Amino acid change Type of substitution A 1012T>C Homozygous F107S Non-polar-> polar B [7830T+ 129n>G] Heterozygous P31S Non-polar—>polar Heterozygous L200R Non-polar->basic C [7830T+ 1291T>G] Heterozygous P31S Non-polar->polar Heterozygous L200R Non-polar->basic D 1309O T Homozygous A206V N on-polar->non-polar E [905G>T; 906OT] Homozygous E71D Acidic—>acidic Homozygous P72S Non-polar-^polar 197 A GGC CAGGGC CCTCGTC C j GC CAGGG CCCTCG TCC 360 C CCGCGGG TCGTGCTG A A T G G A T TC C GC G TG G GCACCNGGTÛCGCQACCCGCG 310 320 — Z50 Figure 5.4 Electrophereogram of the coding region of CHST6. (a) unaffected member of family A (IV-1) shows the wild type allele, (b) The proband of family A (III-3) shows a homozygous single base pair transition (TTT—>TCT, F107S). (c and e) unaffected members of families B and C (lU-l and IV-3) show the wild type allele, (d) the probands of families B and C (II-5 and III-3) show a heterozygous single base pair transition (CCC->TCC, P31S), (f) the probands of families B and C (11-5 and III-3) show heterozygous single base pair transition (CTG->CGG, L200R). (g) unaffected member of family D (1-2) shows the wild type allele, (h) the proband of family D (II-2) shows a homozygous single base pair transition (GCC->GTC, A206V). (i) unaffected member of family E (IV-3) shows the wild type vO I 00 allele, (J) The proband of family E (III-2) shows two homozygous base pair transversion and transition (GAG->GAT, E71D and CCC->TCC, P72S). Results 5.4.2 Restriction enzyme digest of the mutations identified All the mutations identified were confirmed by testing the probands, the unaffected relatives, and 100 control chromosomes of Caucasian origin by restriction enzyme digestion (section 2.5, Table 5.4 and Figures 5.5a, b, c and d). The PCR products digested by Hini I, BstN I, and Ban II restriction enzymes were analysed on 6% non denaturing polyacrylamide gel electrophoresis and were visualised by staining with ethidium bromide (section 2.3.3). Products digested by Apa I and Drd I were analysed on 3% agarose gels (section 2.3.1). Table 5.4 Restriction enzymes used for confirmation of the mutations identified Enzym e Recc^nition site Mutation identified nam e Hinn 5 ...G' ANTC...3' homozygous F I07S in femily A and homo 2ygous P72S in femilyE A pa I 5 ...GGGCC* C...3 heterozygous P31S in families B and C £ s tN \ 5 ...CC (AT)GG...3' heterozygous L200R in families B and C D rd \ 5\..GACNNNN* NNGTC...3’ homozygous A206V in femily D Ban n 5’...G(A/G)GC(T/C)‘ C...3’ homozygous D 7IE in family E 199 Results First row Second row Third row Figure 5.5a An example of restriction analysis, Apa I restriction enzyme digest on 3% agarose gels. Lane 1 in all rows is a 500 bp undigested PCR product. Lanes 12, 14, 19 (arrows) showing PCR products from the affected individuals of families B and C with one undigested 500bp size product and two digested products (130 and 370 bp). The remaining lanes in all rows show digested PCR products of 56 controls with two bands of 130 and 370 bp. 200 Results M 12345678 9 10 111213141516 Figure 5.5b Analysis of B st^ I restriction digest on 6% polyacrylamide gels. Lanes (1- 3) showing digested PCR products from the affected of families B and C with two bands (mutated and wild alleles). The remaining lanes show digested PCR products from unaffected individuals with wild type allele (one product). M= (j)XI74 RF DNA Hae III m a rk e r. 201 Results M 1 2 3 4678 9 10 11121314151617181920 Figure 5.5c Drd I restriction digest on 3% agarose gels. Lane I in the first row is the undigested PCR product (400 bp). Lane 2 in the first row and lane 1 in the second row represent the digested mutant allele (360bp) from the proband of family D. Lane 3 in the first row shows the digested mutant and the undigested wild type alleles from unaffected individuals (heterozygous). The remaining lanes represent the control individuals with undigested wild type band. M= (j)X174 RF DNA Hae III m a rk e r. 202 Results 5.4.3 Haplotype analysis Haplotype analysis was performed for families B and C who shared the same compound heterozygotic mutations. The following microsatellite markers were used: cen-D16S512- D16S3115-D16S3083-D16S518-tel, based on the Généthon genetic map (Dib et a/., 1996) and the Whitehead physical map. The markers were amplified by PCR (section 2.2), except that the annealing temperature was 56 for D16S512, 58 for D16S3115, 54 °C for D16S3083, and 58 for D16S518. The amplified PCR products were separated on 6% non-denaturing polyacrylamide gel electrophoresis and were visualised by staining with ethidium bromide and genotyped (see section 2.3.3). The results revealed different haplotypes for all used markers (data not shown). This indicates that although the two families share the same mutations that they are probably not related and there is no founder effect. 5.4.4 Diagnosis of MCD type I by assay of sulfa ted KS in serum Inhibition ELISA carried out on serum fi-om the MCD patients studied showed little or no inhibition of the monoclonal 5-D-4. Control experiments using increasing KS concentrations effectively inhibited 5-D-4 binding (Figure 5.6 a and b). These data show that the patients are of MCD type I. 203 Results 0.6 0.5 0.4 0.3 0.2 H 0 1 r # i Negative Positive A B Serum sample Figure 5.6a Inhibition-ELISA of sera from MCD families. A negative control with no antibody and a positive one with no serum are shown. No inhibition of antibody 5-D-4 binding is observed with any of the five sera tested (families A, B, C, D, and E). This indicates that no detectable levels of KS are present and the MCD is of type I. 0.7 - C 0.6 - 1 0 5 - o ^ 0.4 - ♦ * 5 O 0.3 - ■2 0.2 - * g 0.1 - 1 10 100 1000 10000 KS concentration (ng/ml) Figure 5.6b Inhibition ELISA for KS. Assay was performed with solutions containing varying concentrations of KS (ng/ ml) as a control for ELISA. The curve shows a dose- dependant inhibition of antibody 5-D-4 binding. 204 Results 5.4.5 Assessment of the significance of the mutations identified In order to assess the significance of the mutations detected, an alignment of the primary sequence of CHST6 with other published carbohydrate sulfotransferases (Lee et al., 1999; Fukuta et al., 1997; Uchimura et al., 1998; Fukuta et al., 1998; Bistrup et al., 1999), using Clustal W was performed and is presented in Figure 5.7 (Thompson et al., 1994). With the exception of the P3 IS substitution, all the mutations occur at positions where the residue is highly conserved across all the carbohydrate sulfotransferases. Furthermore, mutations F107S (family A), L200R (families B and C) and P72S (family E) are all non-conservative changes involving substitution of a non-polar residue for polar residue and are likely to affect CHST6 enzyme activity. Two of the substitutions are conservative changes, E71D (acidic^acidic, family E) and A206V (nonpolar->nonpolar, family D). Of these, E71D occurs in tandem with the P72S substitution (which is more likely to be disease causing) and thus may not itself have a major deleterious effect. It is not immediately apparent why the A206V substitution should be disease causing, but conservative substitutions elsewhere in CHST6 have been reported to underlie MCD type 1 in Icelandic families (Liu et al., 2000b). In the alignment, 5’-and 3’-phosphate-binding domains (5’-PB and 3’-PB) that interact with adenosine 3’-phosphate 5’-phosphosulfate (PAPS), a sulfate donor for sulfotransferases (Kakuta et al., 1997; Kakuta et al., 1998), are boxed (Figure 5.7). These domains form part of the vital active site region of the enzyme. Two of the new mutations reported in this study, L200R and the conservative substitution A206V occur within the 3 -PB domain. 205 Results The crystal structure of a more distantly related sulfotransferase, mouse oestrogen sulfotamsferase (mestST) (Kakuta et aL, 1998), which contains the same 5'-and 3’-PB domains, provides a model for examining the protein structure in this important region of the protein (Figure 5.8). The 3’PB domain in mestST forms a (3-strand ((38) joined at 90^ to an a-helix (a6). Although the sequence homolgy between the domains in mestPB and CHST6 is quite low, secondary structure prediction of the 3 PB domain in CHST6 using both the Chou-Fasman (Chou and Fasman, 1978) and Gamier-Osguthrope-Robson (Gamier et a\., 1978) methods indicate that the domain forms a similar p-strand-tum- helix structure in the wild type protein (analysis was performed in collaboration with Dr. Wilkie). An A206V substitution in CHST6, however, is predicted to disrupt this structure by substituting another region of p-strand for the start of the a-helix. Whilst it must be emphasised that this is merely a prediction of a structural change, nonetheless any slight alteration in structure in such a sensitive region of the protein is likely to have dire consequences on protein function. It is likely, therefore that both substitutions found in the 3’PB domain of the protein in these patients, L200R and A206V, result in a non functioning enzyme. 206 Results C-GlcNAc6ST I-GlcNAc6ST ------MGMRARVPKVAHSTRRPPAARHWLPR HecGlcNAc6ST ------MLLPK G1CNAC6ST — — — — — — — — — — — — — — — — — — — — — —— — — — — — — — — — — — — — — — — — — — — — — ——MKGRRRRRREYCK C h o n 6 S T MSRSPQRALPPGALPRLLQAAPAAQPRALLPQWPRRPGRRWPASPLGMKVFRRKALVLCA K S G 6 S T ------MQCSWKAVLLLA C-GlcNAc6ST VSSTAVTALLLAQT------FLLLFLVSRPGPSSPAGGE------I-GlcNAc6ST FSSKTVTVLLLAQTT------CLLLFIISRPGPSSPAGGE ------HecGlcNAc6ST KMKLLLFLVSQMAIL------ALFFHMYSHNISSLSMKAQP------G1CNAC6ST -FALLLVLYTLVLL------LVPSVLDGGRDGDKGAEHCPG------LQRSLGVWSLEAAAA C h o n 6 S T GYALLLVLTMLNLLDYKWHKEPLQQCNPDGPLGAAAGAAGGKLGAPRAASGRAAPCSCPF K S G 6 S T LASIAIQYTAIRTFT ------AKSFHTCPGLABAGLAERLCE------ 5 ' P B C-GlcNAc6ST ------ARVHVLVLSS j ' r 5 5 5 5 f I-GlcNAc6ST ------DRVHVLVLSS i R S G S S F HecGlcNAc6ST } R S G S S F G1CNAC6ST GEREQGAEARAAEEGGANQSP------RFPSNLSGAVGEAVSREKQHIYVHAT J R T G S S F C h o n 6 S T GPPHSLPPSRCRRRGDTLQPRQGWRGLRPLQAMALGAPEGVGDKRHWMYVFTT J R S G S S F K S G 6 S T ------E ------SPTFAYNLSRKTHILILAT r R S G S S F C-GlcNAc6ST QLFNQHPDVFYLMEPAWHVWTTLSQ------GSAATLHMAVRDLVRSVFLCDMDVP I-GlcNAc6ST 3QLFSQHPDVFYLMEPAWHVWTTLSQ------GSAATLHMAVRDLMRSIFLCDMDVF HecGlcNAc6ST QLFGQHPDVFYLMEPAWHVWMTFKQ------STAWMLHMAVRDLIRAVFLCDMSVF G lc N A c 6 S T ELFNQHPDVFYLYEPMWHLWQALYP------GDAESLQGALRDMLRSLFRCDFSVL C h o n 6 S T 3ELFNQNPEVFFLYEPVWHVWQKLYP------GDAVSLQGAARDMLSALYRCDLSVF K S G 6 S T QLFNQHLDVFYLFEPLYHVQNTLIPRFTQGKSPADRRVMLGASRDLLRSLYDCDLYFL C-GlcNAc6ST DAYL-PWRRNL------SDLFQWAVSRALCSPPACS------AFPRGAISSEAVCKPL I-GlcNAc6ST DAYM-PQSRNL------SAFFNWATSRALCSPPACS------AFPRGTISKQDVCKTL HecGlcNAc6ST DAYMEPGPRRQ------SSLFQWENSRALCSAPACD------IIPQDEIIPRABCRLL G1CNAC6ST RLYAPPGDPAARAPDTANLTTAALFRWRTNKVICSPPLCPGAPRARAEVGLVEDTACERS C h o n 6 S T QLYSPAGSGGR------NLTTLGIFGAATNKVVCSSPLCP--AYRKEVVGLVDDRVCKK- K S G 6 S T ENYIKPPPVNH ------TTDRIFRRGASRVLCSRPVCD--P-PGPADLVLEEGDCVRK 3 ’ P B C-GlcNAc6ST CARQSFTLAREACRSYSHVVLKEVRFFNLQVLYPLLSDPALNLR i K^h L v RDPRAVLRSRE I-GlcNAc6ST CTRQPF------HecGlcNAc6ST CSQQPFEVVEKACRSYSHVVLKEVRFFNLQSLYPLLKDPSLNLHI VHLVRDPRAVFRSRE G lc N A c 6 S T CPPVAIRALEAECRKYPVVVIKDVRLLDLGVLVPLLRDPGLNLKV VQLFRDPRAVHNSRL C h o n 6 S T CPPQRLARFEEECRKYRTLVIKGVRVFDVAVLAPLLRDPALDLKV IHLVRDPRAVASSRI K S G 6 S T CGLLNLTVAAEACRERSHVAIKTVRVPEVNDLRALVEDPRLNLKV IQLVRDPRGILASRS C-GlcNAc6ST ÜTr AKALARDNGIVLGTNG------TWVEADPGLRVVREV I-GlcNAc6ST ------ARDNGIVLGTNG------KWVEADPHLRLIREV HecGlcNAc6ST rKGDLMIDSRIVMGQHEQ------KLKKEDQPYYVMQVI G1CNAC6ST SRQGLLRESIQVLRTRQRGDRFHRVLLAHGVGARPGGQSRALPAAPRADFFLTGALEVI C h o n 6 S T SRHGLIRESLQVVRSRD--PRAHRMPFLEAAGHKLGAKKEGVGGP--ADYHALGAMEVI K S G 6 S T TFRDTYRLWRLWYGTGR------K PYNLDVTQLTTV C-GlcNAc6ST CRSHVRIAEAATLKPPPFLRGRYRLVRFEDLAREPLAEIRALYAFTGLSLTPQLEAWIHN I-GlcNAc6ST CRSHVRIAEAATLKPPPFLRGRYRLVRFEDLAREPLAEIRALYAFTGLTLTPQLEAWIHN HecGlcNAc6ST CQSQLEIYKTIQSLP-KALQERYLLVRYEDLARAPVACJTSRMYEFVGLEFLPHLQTWVHN G1CNAC6ST CEAWLRDLLFARGAP-AWLRRRYLRLRYEDLVRQPRAQLRRLLRFSGLRALAALDAFALN C h o n 6 S T CNSMAKTLQTALQPP-DWLQGHYLVVRYEDLVGDPVKTLRRVYDFVGLLVSPEMEQFALN K S G 6 S T CEDFSNSVSTGLMRP-PWLKGKYMLVRYEDLARNPMKKTEEIYGFLGIPLDSHVARWIQN C-GlcNAc6ST ITHGSGPGARREAFKTSSRNALNVSQAWRHALPFAKIRRVQELCAGALQLLGYRPVY------I-GlcNAc6ST ITHGSGIGKPIEAFHTSSRNARNVSQAWRHALPFTKILRVQEVCAGALQLLGYRPVY------HecGlcNAc6ST ITRGKGMGD--HAFHTNARDALNVSQAWRWSLPYEKVSRLQKACGDAMNLLGYRHVR------G lc N A c 6 S T MTRGAAYGAD-RPFHLSARDAREAVHAWRERLSREQVRQVEVACAPAMRLLAYPRSGEEG C h o n 6 S T MTSGSGS-SS-KPFVVSARNATQAANAWRTALTFQQIKQVEEFCYQPMAVLGYERVN------K S G 6 S T NTRGDPTLG--KHKYGTVRNSAATAEKWRFRLSYDIVAFAQNACQQVLAQLGYKIAA------ C-G1CNAC6ST SEDEQRNLALDLVLPRGLNGFTWASSTASHPRN I-GlcNAc6ST SADQQRDLTLDLVLPRGPDHFSWASPD------HecGlcNAc6ST SEQEQRNLLLDLLS------TWTVPEQIH------G lc N A c 6 S T DAEQPREGETPLEM------DADGAT------C h o n 6 S T SPEEVKDLSKTLLR------KPRL------K S G 6 S T SEEELKNPSVSLVE------ERDFRPFS- Figure 5.7 Alignment of C-GlcNAc6ST and other carbohydrate sulfotransferase protein sequences. Clustal W version 1.7 was used for multiple alignment of amino acid sequences. The positions of mutated amino acids in MCD type I patients reported in this study are shaded in grey. The 5’-and 3’-phosphtae binding site motifs (5’ PB and 3’ PB) are boxed. Note that two of the mutations occur at highly conserved position within the 3’ PB. The protein sequences compared are l-GlcNAc6ST (intestinal GlcNAc-6- sulfotransferase), HecGlcNAcbST (human high-endothelial-cell GlcNAc-6- sulfotransferase), GlcNAcbST (human GlcNAc-6-sulfotransferase), ChonbST (human chondroitin-6-sulfotransferase), KSG6ST (human keratan sulfate Gal-6- sulfotransferase). 207 Results P8 a6 mestST CKM lYLCRNAKDVAVSYYY FLLM turn a HCHST6 (normal) L R I VHLVRDPRAVLRSREQ TAKA turn a hCHST6 (A206) LRI IV H L V R D P R V V L R S R E Ol TAKA * Figure 5.8 Prediction of secondary structure of the 3’ phosphate binding domains in normal and A206V CHST6 mutation. The actual structure of the domain in mouse oestrogen sulfotransferase (mestST) derived from the crystal structure is given above, with the structure predictions for normal and A206V human carbohydrate sulfotransferase (hCHST6) shown underneath. The sequences of the 3’PB motifs are boxed and the position of the A206V mutation is marked with an asterisk. 208 Results 5.5 Discussion In this study five unrelated families with clinically and histopathologically diagnosed MCD were screened for mutations in the CHST6 gene, which is reported to be involved in the pathogenesis of this disorder by affecting the synthesis of KS. KSPGs (lumican, keratocan, and mimecan) are present in the cornea as the major class of proteoglycans and are thought to play an important role in comeal transparency (Funderburgh et al., 1991; Blochberger et al. 1992; Corpuz et al., 1996). The sulfate group of KS appears to be crucial for its biological hmction since the degree of sulfation of KS increases during comeal development (Hart, 1976; Nakazawa et al., 1995) and the synthesis of unsulfated KS is observed in the comeas of MCD patients (Nakazawa et al., 1984). In view of the autosomal recessive inheritance of the condition, MCD probably results from a deficiency in sulfotransferase specific for proper sulfation of KS (Francois et al., 1975). Six novel missense mutations were identified in the probands from the 5 families participated in this study. The mutations were detected by direct sequencing of CHST6 PCR products. Using restriction enzyme analysis, segregation was confirmed and the changes were not found in unaffected family members and 50 controls. To fully exclude the possibility that these changes are polymorphisms, a statistically significant number of chromosomes should be tested (approximately 500). In order to measure levels of sulfated KS in the semm of our MCD patients ELISA was performed and revealed no detectable levels in all cases confirming that the patients are of MCD type 1. 209 Results All mutations detected occur at positions in the protein where residue type is highly conserved across carbohydrate sulfotransferases and/or involve non-conservative substitutions of non-polar residues for polar or charged residue. One possible exception is E71D (family E) but this occurs as part of a double, homozygous substitution with P72S. It seems unlikely that E71D on its own would result in absence of protein function. Two mutations occur in the 3’PB domain, an essential part of the active site responsible for PAPS binding. Of these, L200R involves a non-conservative substitution of a non-polar for a basic residue at a site, which is completely conserved across carbohydrate sulfotransferases. The A206V substitution, although representing a conservative change at a position which is not so highly conserved across carbohydrate sulfotransferases, is nonetheless predicted to result in a structural change at a highly sensitive region of the protein when compared to the crystal structure of mestST (Figure 5.8). It has recently been shown that there is a decrease in the activity of C-GlcNAc6ST in the cornea of MCD patients and this results in the formation of poorly or nonsulfated KS causing comeal opacity (Hasegawa et al., 2000). More recently it has been reported that mouse intestinal GlcNAc6-0-ST (mIGnbST) has similar activity as human C- GlcNAc6ST suggesting that mIGnbST is the orthologue of C-GlcNAc6ST and functions as a sulfotransferase to produce KS in the cornea. Moreover, the amino acid substitutions in C-GlcNAc6ST resulting from missense mutations in CHST6 found in MCD patients abolished the sulfotransferase activity by functional inactivation rather than protein degradation or mislocalisation (Akama et al., 2001). The homozygous mutations detected in this study infer an essential role of C- GlcNAc6ST, the CHST6 protein product, in the production of normally functioning KS 210 Results whilst its inactivation is responsible for the MCD phenotype. Furthermore the heterozygous changes identified in two families affected both alleles suggesting that compound heterozygous mutations can also result in MCD phenotype. Mutated carbohydrate sulfotransferase could therefore result in loss of function of sulfotransferase enzyme required for proper sulfation of KS, an essential element for comeal transparency, leading to MCD type I phenotype. 211 General discussion Chapter 6 General Discussion 6.1 Molecular genetics and corneal dystrophy The recent application of the novel techniques of molecular genetics to problems in clinical medicine have led to dramatic new insights into the pathophysiology of human disease. Molecular genetic technology has become the dominant approach to the study of many basic biological questions, particularly those questions concerning the nature of genes and how they work in eukaryotic cells (Wiggs, 1993). The genes responsible for many of the 3000 known hereditary disorders that affect humans have been mapped in the human genome and a number of these genes have been cloned (Stephens et al., 1990). Recent genetic studies into comeal dystrophies, which cause visual impairment by affecting comeal transparency, provide insight into some of these disorders at a basic molecular level. The major benefit fi-om the application of molecular genetics to such hereditary eye disease have been improved methods of diagnosis based on genes and specific DNA mutations that are responsible for these conditions. Comeal dystrophies are a group of diseases which are sometimes difficult to distinguish, difficult to control and treat and often difficult to diagnose. They do not fall into a strict disease category. They are currently classified according to the layer or layers of the comea involved. This allows the ophthalmologists to conveniently categorise disease phenotype according to anatomy. The study by Jones and Zimmerman in 1961 provided a framework for the classification of granular, lattice, and macular dystrophies of the comeal stroma according to their clinical appearance and, in particular, their histopathologic staining pattems. Perhaps, the most 212 General discussion convenient classification was the one proposed by Franceschetti (1954) in which the dystrophies were listed as (1) those affecting the anterior limiting membrane, (2) those affecting the comeal stroma, (3) those affecting the posterior limiting membrane, and (4) those affecting some combination of two or more of these layers. This method of classification does not relate disorders to molecular defects and does not group dystrophies that are biochemically related. So, the diagnosis of such hereditary disorders is customarily based on slit lamp and histopathologic findings. The characteristic biomicroscopic appearance and histopathologic features of each dystrophy generally provide a clue to the diagnosis. However, altered expression of the genetic defect may result in variations in the natural history of the dystrophy, leading to a diagnostic challenge (Santo et al., 1995). Most of the classifications of comeal dystrophies include only those hereditary disorders with major effect on the comeal topography rather than the size or shape. Nonetheless, recent classifications include abnormalities of the comeal size and shape like keratoconus, keratoglobus and megalocomea among the broad terminology of comeal dystrophy (Klintworth, 1999). Recently some authors describe comea plana as one of the comeal dystrophies. This is based on the hypothesis that lattice, granular, macular comeal dystrophy and comea plana all could be due to specific stmctural and developmental anomalies of the stroma (Chakravarti, 2001). Some comeal dystrophies that were previously believed to be distinct clinicopathologic entities are closely related at the molecular level with the different phenotypes resulting from distinct mutations in the same gene. Until the identification of the responsible genes, accuracy of the clinicopathologic diagnosis could not be tested. Particularly disconcerting were patients with equivocal or ambiguous findings. Examples were cases with clinical and pathologic features of 213 General discussion both CDGGl and LCD (now categorised as ACD) and cases with features of both CDGGl and CDB. Also, cases with Bowman’s layer dystrophy due to the emergence of numerous variants of type I Bowman’s layer dystrophy which is now known as the true or geographic type and Type II, the honeycomb type, which is categorised as Thiel-Behnke. This confusion made it difficult to construct a clear-cut classification of these entities. This situation is compounded by further confusion of these dystrophies with other disorders, such as advanced epithelial basement membrane dystrophy, Fuch’s dystrophy with marked subepithelial fibrosis, LCD with marked recurrent epithelial erosions and subepithelial fibrosis, and simple superficial comeal scarring. The identification of the gene defects responsible for most of the comeal dystrophies provide precision to the diagnosis. Recently a preliminary classification of comeal dystrophies based on molecular aetiology by Gupta and Hodge (1999) proved to be useful in understanding the aetiology and pathogenesis of comeal dystrophy. Comeal dystrophies have now been classified according to gene mutations into one of four categories: first, dystrophies due to abnormal deposition of materials in the extracellular matrix such as LCD II caused by Gelsolin gene mutations; second, dystrophies due to dismption of cell-cell interaction or cell-surface interactions such as LCDI, IIIA, CDGGl, CDB, and ACD caused by mutations in BIGH3 gene; third, dystrophies related to abnormalities in the cytoskeletal proteins as in Meesmann caused by mutations in keratin 3 and 12; and fourth, dystrophies associated with abnormal metabolic function such as MCD. 214 General discussion Hence, the new knowledge gained by the recent genetic studies of comeal dystrophies contributed to the revision of our understanding of such ocular disorders and may lead to improved innovative tools for therapy. 6,2 Overview of the work presented The molecular genetics of comeal dystrophies described herein gives some idea of the phenomenal progress that has been made in the field of comeal genetics. Linkage analysis and the consecutive molecular procedures have revealed the diagnostic value of molecular genetic analysis in patients vvdth comeal dystrophies. Early in this work a consanguineous pedigree in which comea plana cosegregated with microphthalmia was investigated and a novel mutation in KERA, which alters a highly conserved motif and is predicted to affect the tertiary stmcture of the molecule leading to the phenotype, is described. The combined phenotype of comea plana and microphthalmia in the family has not been previously reported. The phenotype- genotype correlation in this family sheds light on the importance of molecular genetic analysis to assist in the diagnosis and to explain hereditary ophthalmic diseases presenting with unusual clinical phenotype. In addition, the mutation identified reflects the possibility of KERA being involved in ocular development. A description of a series of families with various types of autosomal dominant stromal dystrophies has also been presented in this work. The diagnosis of CDBII, sometimes called curly fibre comeal dystrophy, in all families has been strongly suggested by revealing the underlying genetic defect linked to the disease. The variable phenotype of CDGGl described in different generations of the same family, which ranged from scattered subepithelial deposits to the confluent opacities at the anterior and deep stromal level simulating either other types of dystrophies or 215 General discussion different comeal disorders, reflects the difficulty in relying solely on the slit lamp for categorisation of comeal dystrophies. In addition, ACD pedigrees have been identified and studied in the U.K. Mutation screening of BIGH3 gene revealed a single point mutation as the underlying genetic defect in all cases of ACD described in other populations. However, the detection of ACD in a family with underlying genetic defect linked to the CDGGl either raises the possibility of the two phenotypes being different stages of the same disease or a specific point mutation could lead to variable phenotype. Intrafamilial phenotypic variation is described in a Bangladeshi family displaying two types of comeal dystrophy, CDBI and LCDI, with underlying defined mutation. The probability of the two phenotypes being different stages of the same disease entity cannot be excluded. Modifier genes, whose expression that can influence a phenotype resulting from a mutation at another locus, must also be considered. The molecular investigation of five British families with MCD using the conventional tools for muation screening of CHST6 revealed novel mutations in MCD which further supports the role of CHST6 in the pathogensis of MCD. The work presented here is not only a contribution to ophthalmic genetics but also it is a step forward in the study of the pathogenesis of comeal dystrophy. In particular, it highlights the significance of genetics in ophthalmic diagnosis and treatment. 6.3 Therapeutics in comeal dystrophy 6.3.1 Advances in the surgical therapy for corneal dystrophy Penetrating keratoplasty (PK) is currently the surgical method of choice for treating comeal dystrophies with stromal and endothelial involvement. Another surgical procedure is lamellar keratoplasty (LK) in which the anterior comeal layers, to a 216 General discussion variable stromal depth, are replaced with donor tissue. In contrast, all layers are replaced in PK (Barraquer, 1972; Anwar, 1972; Ehrlich, 1988). In the past few years, new LK procedures have become available for the management of various comeal disorders including comeal dystrophy. Those limited to the superficial stromal layers may be managed with automated lamellar keratoplasty. For dystrophies extending to the deeper stromal layers deep anterior LK may be employed (Chau et al., 1992; Sujita and Kondo, 1997). Comeal endothelial dystrophies may be approached by using posterior LK (Melles et al., 1998; Van Doreen et al., 1999). Excimer laser superficial keratectomy, commonly known as PTK, is an excellent modality for treatment of selected dystrophies localised to the comeal epithelium. Bowman’s layer, and anterior stroma (Hersh et al., 1996; Hersh and Wagoner, 1998). The role of PTK in the management of comeal dystrophies that present with anterior involvement, but have progressive involvement of the deep stroma, has not been clearly defined (Wagoner et al., 1999). PTK can be useful to treat anterior comeal dystrophies both before and after PK. Indeed, it can be a useful therapeutic altemative to LK or PK in these patients. Even though comeal dystrophies are likely to recur eventually after PTK, successful retreatment with PTK is possible. Based on the recent concept that the deposits in stromal dystrophies (and perhaps the other dystrophies related to the BIGH3 gene) are of epithelial origin, limbal stem cell transplant combined with PK might be the treatment of choice to reduce the frequency or the severity of recurrence of the disorder in the graft. Although these procedures broaden the array of treatments for comeal dystrophy, the option of an ideal curative line is yet to be developed. 217 General discussion 6.3.2 Molecular genetics and its therapeutic role Finding DNA mutation(s) responsible for a disorder can form the basis of new, DNA-based diagnostic tests. Such tests may allow better care of affected patients, because individuals at risk for the disease who are identified and treated at the earliest stages of the disease will have the best chance of a good visual outcome. Carriers of mutant genes can also be identified, making it possible not only to identify offspring at risk for specific comeal dystrophies but also to make precise diagnoses before the development of typical, clinically recognisable phenotypes. So it will be possible to offer prenatal and postnatal diagnosis of presymptomatic individuals. This will enable genetic counselling and any possible therapeutic interventions to be made at an early stage. The next step will be to realise new methods of treatment based on the specific molecular pathophysiology that has been discovered as a result of the characterisation of the abnormal gene(s) and resultant gene product(s). Once the genes for specific comeal dystrophies are identified, it will be possible to develop cell culture and animal models to study the function of the relevant genes. The characterisation of gene(s) product will perhaps lead to novel therapy for the disease based on the specific activity of the normal protein. Ultimately, the characterisation of the gene responsible for these conditions will be a first step toward gene therapy, which theoretically can involve the specific replacement, correction, or augmentation of a dysfunctional gene. In addition to understanding the pathogenesis of comeal dystrophies, the identification of the disease gene may lead to possibility of treatment at a molecular level by the direct transfer of therapeutic genes into the relevant tissue. Such genes may be transferred by a variety of vectors, including vimses, liposomes and 218 General discussion protoplasts. Viral vectors that are being evaluated include Herpes simplex virus (type I), adenovirus and adeno-associated virus (Ali et al., 1997). The majority of the work in this field has been carried out on retinal degenerations (Chong and Bird, 1999). The retina is accessible, but poses the problem of being post-mitotic. However, there are blood-retinal and blood aqueous barriers that may concentrate vectors in the target area (Miller, 1992). Recently, it has been demonstrated that the use of in vivo gene transfer was able to correct the complex ultrastructural photoreceptor cell defect in the retinal degeneration slow mouse (Ali et al, 2000). This approach is potentially much easier in a disease, which is recessive or semi dominant, as the transfer of a single copy of the gene may be sufficient to correct the loss of function that is causing the disease phenotype. However the situation is more difficult in the dominant diseases, where merely inserting a functional copy of the gene will not fully correct the defect or negate the deleterious effects of the mutant allele. Another problem with the design of gene therapies for autosomal recessive disease is the large intragenic heterogeneity seen in these conditions. Strategies in vitro for gene therapies directed to dominant mutations are being developed (Millington-Ward et al., 1997). These include methods that avoid the requirement to target individual mutations for genetic suppression. General methods for suppression may be directed towards the primary effect (such as using hammerhead ribozymes designed to target transcript) or a secondary effect associated with disease process, such as apoptosis. Also the treatment may be combined, such as the simultaneous introduction of a gene to replace a loss of function and a gene encoding an anti-apoptotic factor (Bessant et al, 2001). Another potential strategy of gene therapy is antisense oligonuceotides 219 General discussion (OGN) which by binding to the mRNA molecules block the synthesis of the protein (Cordeiro et al., 2000). The eye is an informative and easily examined structure. It is an attractive target for gene therapy because of this accessibility and immune privilege (Wright, 1997). Research studies are working towards retinal therapy (Hangai et al., 1996), and hopefully advances will ultimately allow therapy for a larger range of ocular disorders. Also, the similarities between some ocular and systemic disorders like LCD and Alzheimer’s disease (stone et al., 1994) may allow the eye to be used as a model system for the development of effective therapy for systemic disorders. In summary, the novel techniques of molecular genetics are becoming increasingly important to our understanding of comeal dystrophy. In the next decade, ophthalmology as well as other disciplines in medicine should witness an explosion of information concerning the molecular basis of a variety of disorders. The knowledge gained from this study and others may contribute enormously in refinement of molecular pathophysiology of comeal dystrophy and could lead to new methods of prevention, diagnosis, and treatment. 6.4 Future work More recently a large panel of MCD patients has been collected. Screening of CHST6 for more novel mutations in this panel will establish strong phenotype- genotype correlation in the British population. Further analysis of the fimctional implications of the novel mutations identified in keratocan and CHST6 and of the mutations previously reported in BIGH3 gene will be an important research direction towards therapies. 220 General discussion With the collection of families containing reasonable numbers of affected individuals one would be able to identify genetic loci of other comeal dystrophies, such as fleck and map-dot comeal dystrophy which are yet unmapped. The research at the Institute of Ophthalmology is now focused on identification of the gene for Schnyder’s comeal dystrophy through screening of candidates that map to the same chromosomal region as the disease. Collections of more patients and families with various comeal dystrophies will help the research to continue in this important area of ophthalmology. 221 References References Afshari, N.A., Mullally, J.E., Afshari, M.A., Steinert, R.F., Adamis, A.P., Azar, D.T., Talamo, J.H., Dohlman, G.H., and Dry)a, T.P. (2001) Survey of patients with granular, lattice, Avellino, and Reis-Bucklers comeal dystrophies for mutations in the BIGH3 and gelsolin gene. Arch. Ophthalmol. 119, 16-22. Akama, T.O., Nishida, K., Nakayama, J., Watanabe, H., Ozaki, K., Nakamura, T., Dota, A., Kawasaki, S., Inoue, Y., Maeda, N., Yamamoto, S., Fujiwara, T., Thonar, E.J., Shimomura, Y., Kinoshita, S., Tanigami, A., and Fukuda, M.N. (2000) Macular comeal dystrophy type I and type II are caused by distinct mutations in a new sulfotransferase gene. Nat. Genet. 26, 237-241. Akama, T.O., Nakayama, J., Nishida, K., Hiraoka, N., Suzuki, M., McAauliffe, J., Hindsgaul, O., Fukuda, M., and Fukuda, M.N. (2001) Human comeal GlcNac 6-0- sulfotransferase and mouse intestinal GlcNac 6-0 sulfotransferase both produce keratan sulfate. J. Biol. Chem. 276 (19), 16271-16278. Akhtar, S., Meek, K.M., Ridgway, E.A., Bonshek, R.E., and Bron, A.J. (1999) Deposits and proteoglycan changes in primary and recurrent granular dystrophy of the comea. Arch. Ophthalmol. 117,310-321. Akimune, C., Watanabe, H., Maeda, N., Okada, M., Yamamoto, S., Kiritoshi, A., Inoue, Y., Shimomura, Y., and Tano, Y. (2000) Comeal guttata associated with the comeal dystrophy resulting from a Big-h3 R124H mutation. Br. J. Ophthalmol. 84, 67-71. Akiya, S., and Brown, S.I. (1970) Granular dystrophy of the comea: characteristic electron microscopic \Qs\on. Arch. Ophthalmol. 84, 179-192. Albin, R.L. (1998) Fuchs’ comeal dystrophy in a patient with mitochondrial DNA mutations. J. Med. Genet. 35, 258-259. Alexander, P.J., and Messenger, H.K. (1957) Dorland’s Medical Dictionary, 23"^^ ed., Philadelphia, Saunders. Al Faran, M.F., and Tabbara, K.F. (1991) Comeal dystrophies among patients undergoing keratoplasty in Saudia Arabia. Cornea 10, 13. Ali, R.R., Reichel, M B., Hunt, D.M., and Bhattacharya, S.S. (1997) Gene therapy for inherited retinal degeneration. Br. J. Ophthalmol. 81, 795-801. Ali, R.R., MarcoSarra, G., Stephens, C., de Alwis, M., Bainbridge, J.W., Munro, P.M., Fauser, S., Reichell, M.B., Kinnon, C., Hunt, D M., Bhattacharya, S.S., and Thrasher, A.J. (2000) Restoration of photoreceptor ultrastructure and function in retinal degeneration slow mice by gene therapy. Nat. Genet. 25, 306-310. 222 References Altschul, S.F., Gish, W., Miller, W., Meyers, E.W., and Lipman, D.J.(1990). Basic local alignment search tool. J. Mol Biol 215, 403-410. Antequera, P., and Bird, A. (1994) Predicting the total number of human genes. Nature Genet. 25, 232-234. Anwar, M. (1972) Dissection techniques in lamellar keratoplasty. Br. J. Ophthalmol. 56, 711-713. Arffa, A.C. (1997) Dystrophies of the epithelium. Bowman’s layer, and stroma. In: Arffa, A.C., and Grayson, M. eds. Graysons diseases of the comea, 4^^ ed. St Louis: Mosby, p413-463. Argiles, A. (1996)p2-Microglobulin amyloidosis. Nephrology. 2, 373-386. Asaoka, T., Amano, S., Sunada, Y., and Sawa, M. (1993) Lattice comeal dystrophy type II with familial amyloid polyneuropathy type IV Jpn. J. Ophthalmol. 37, 426. Attwood, J., and Bryant, S. (1988) A computer program to make analysis with LIPID and LINKAGE easier and less prone to errors. Ann. Hum. Genet. 52, 259. Bahn, C F., Falls, H.F., Varley, G.A., Meyer, R.F., Edelhouse, H.F., and Boume, W.M. (1984) Classification of comeal disorders based on neural crest origin. Ophthalmology 91, 558-563. Bao, W., Smith, C F., Al-Rajhi, A., Chandler, J.W., Karcioglu, Z.A., Akama, T.O., Fukuda, M.N., and Klintworth, G.K. (2001) Novel mutations in the CHST6 gene in Saudi Arabic patients with macular comeal dystrophy. Invest Ophthalmol Vis Soi. 42(Suppl), S483. Barkan, H., and Borley, W.E. (1936) Familial comea plana complicated by cataracta nigra and glaucoma. Am. J. Ophthalmol. 19, 307-310. Barraquer, J.I. (1972) Lamellar keratoplasty (special techniques). Ann. Ophthalmol. 4, 437-469. Bateson, W., Saunders, E.R., and Punnett, R.C. (1905) Experimental studies in the physiology of heredity. Reb. Evol. Comm. R. Soc. 2, 1-55, 80-99. Bengtsson, E., Neame, P.J., Heinegrad, D., and Sommarin, Y. (1995) The primary stmcture of basic leucine-rich repeat protein, PRELP, found in connective tissues. J. Biol Chem. 270, 25639-25644. Bessant, D.A., Khaliq, S., Hameed, A., Anwar, K., Qasim Mehdi, S., Payne, A.M., and Bahttacharya, S.S. (1998) A locus for autosomal recessive congenital microphthalmia maps to chromosome 14q32. Am. J. Hum. Genet. 62 (5), 1113-1116. 223 References Bessant, D.A., Anwar, K., Khaliq, S., Hameed, A., Ismail, M., Payne, A.M., Qasim Mehdi, S., and Bahttacharya, S.S. (1999) Phenotype of autosomal recessive congenital microphthalmia mapping to chromosome 14q32. Br. J. Ophthalmol 83 (8), 919-922. Bessant, D.A., Ali, R.R., and Bhattacharya, S.S. (2001) Molecular genetics and prospects for therapy of the inherited retinal dystrophies. Current Opinion in Genetics and Development 11, 307-316. Biglan, A.W., Brown, S.I., and Jhonson, B.C. (1977) Keratoglobus and blue sclera. Am. J. Ophthalmol. 83, 225. Bistrup, A., Bhakta, S., Lee, J.K., Belvo, Y.Y., Gunn, M.D., Zuo, F.R., Huang, C.C., Kannagi, R., Rosen, S.D., and Hemmerich, S. (1999) Sulfotransferases of two specificities function in the reconstruction of high endothelial cell ligands for L-selectin. J. Cell Biol 145, 899-910. Blake, D.J., Tinsley, J.M., Davies, K.E., Knight, A.E., Winder, S.J., and Kendrick-Jones, J. (1995) Coiled-coil regions in the carboxy-terminal domains of dystrophia and related proteins: potentials for protein-protein interaction. Trends Biochem. Sci. 20, 133-135. Blochberger, T.C., Vergnes, J.P., Hempel, J., and Hassell, J R. (1992) cDNA to chick lumican (comeal keratan sulfate proteoglycan) reveals homology to the small interstitial proteoglycan gene family and expression in muscle and intestine. J. Biol. chem. 267, 347-352. Botstein, D., White R.L., Skolnick, M., and Davis, R.W. (1980) Construction of a genetic linkage map in man using restriction fragment length polymorphisms. Am. J. Hum. Genet. 32, 314-331. Boume, W.M., and Michels, V.V. (1982) keratoconus in one identical twin. Cornea 1, 35. Bron, A.J., and Rabinowitz, Y.S. (1996) Comeal dystrophies and keratoconus. Curr. Opinion Ophthalmol. 7, 71-82. Bron, A.J., Tripathi, R.C., and Tripathi, B.J. (1997) Wolff’s anatomy of the eye and orbit. Eighth edition Chapman and Hall medical London. Caterson, B., Christner, J.E., and Baker, J.R. (1983) Identification of a monoclonal antibody that specifically recognises comeal and skeletal keratan sulfate. Monoclonal antibodies to cartilage proteoglycan. J. Biol. Chem. 258 (14), 8848-8854. Cavalli-Sforza, L.L., wilson, A.C., Cantor, C.R., Cook-Deegan, R.M., and King, M.C. (1991) Call for a world wide survey of human genetic diversity: a vanishing opportunity for the human genome project Genomics 11, 490-491. 224 References Cavara, V. (1950) Keratoglobus and keratoconus. Br. J. Ophthalmol. 34, 621. Cennamo, G., Rosa, N., Rosenwasser, G.O., and Sebastian!, A. (1994) Phototherapeutic keratectomy in the treatment of Avellino dystrophy. Ophthalmologica 208, 198-200. Chakravarti, S. (2001) Stromal structure and lumican. Paper presented at the Cardiff Annual Conference, Cardiff, UK, July 5-6. Chau, G.K., Dilly, S.A., Sheard, C.E., and Rostron, C.K. (1992) Deep lamellar keratoplasty on air with lyophilised tissue. Br. J. Ophthalmol. 76, 646-650. Chong, N.H., and Bird, A.C. (1995) Management of inherited retinal dystrophies: present and future. Br. J. Ophthalmol. 83, 120-122. Chou, P.Y., and Fasman, G.D. (1978) Prediction of the secondary structure of proteins from their amino acid sequence. Adv. in Enzymology 47, 45-148. Cogan, D.G., Donaldson, D.D., Kuwabara, T., and Marshall, D. (1964) Microcystic dystrophy of the comeal epithelium. Trans. Am. Ophthalmol. Soc. 63,213. Corden, L.D., Swensson, O, Swensson B., Smith, F.J.D., Rochels, R., Uitto, J., and McLean, I. (2000) Molecular genetics of Meesmann’s comeal dystrophy: Ancestral and novel mutations in keratin 12 and complete sequence of human KRT12 gene. Exp. Eye Res. 70, 41-49. Cordeiro, M.F., Chang, L., Lim, S., Daniels, J.T., Pleass, R.D., Siriwardena, D., and Khaw, P.T.(2000) Modulating conjunctival wound healing. Eye 14, 536-547. Corpuz, L.M., Funderburgh, J.L., Funderburgh, M.L., Bottom ley, G.S., Prakash, S., and Conrad, G.W. (1996) Molecular cloning and tissue distribution of keratocan. Bovine comeal keratan sulfate proteoglycan 37A. J. Biol. Chem. 271 (16), 9759-9763. Cox, W.V., Kupeferman, A., and Leibowitz, H.M. (1972) Topically applied steroids in comeal disease. II: the role of drug vehicle in stromal absorption of dexamethasone. Arch. Ophthalmol. 88, 549. Cursiefen, C., Hofmann, R.C., Schlotzer, S.U., Fischer, D C., and Kuchle, M. (2000) Immunophenotype classification of MCD: First case report of immunophenotype lA outside of Saudi Arabia. A clinical histopathological correlation with immunohistochemistry and electron microscopy. Klin Monatsbl Augenheilkd 217 (2), 118-126. de la Chapelle, A., Kere, J., Sack, G.H.Jr., Tolvanen, R., and Mauiy, C.P. (1992) Familial amyloidosis, Finnish type: G654A mutation of the gelsolin gene in Finnish families and an unrelated American family. Genomics 13, 898-901. 225 References de la Chapelle, A. (1993) Disease gene mapping in isolated human populations: the example of Finland. J. Med. Genet. 30, 857-865. DeWals, P., and Lechat, M.F. (1987) EUROCAT report 2. Surveillance of congenital anomalies. Years 1980-1984. Department of Epidemiology, Catholic University of Louvain, Brussels. Dib, C., Faure, S., Fizames, C., Samson, D., Drouot, N., Vignal, A., Millasseau, P., Marc, S., Hazan, J., Seboun, E., Lathrop, M., Gyapay, G., Morissette, J., and Weissenbach, J. (1996) A comprehensive genetic map of the human genome based on 5,264 microsattelites. Nature 380, 152-154. Dighiero, P., Dumat, S., Ellies, P., D’Hermies, F, Savoldelli, M., Legeais, J.M., Renard, G., Delpech, M., Grateau, G., and Valleix, S. (2000a) A new mutation (A546T) of the Big-h3 gene responsible for a French lattice comeal dystrophy type III A. Am. J. Ophthalmol. 129 (2), 248-251. Dighiero, P., Ellies, P., Valleix, S., D'hermies, F., Savoldelli, M., Poulique, Legeais, J., and Renard, G. (2000b) Avellino comeal dystrophy. J. Fr. Ophthalmol. 23 (7), 735-737. Donnenfeld, E.D., Cohen, E.J., Ingraham, H.J., Poleski, S.A., Goldsmith, E., and Laibson, P R. (1986) Comeal thinning in macular comeal dystrophy. Am. J. Ophthalmol. 101, 112-113. Duke-Elder, S. (1964) Anomalies in the size of the eye. In Duke-Elder S: System of Ophthalmology, vol 3, pt 2. St. Louis, CV Mosby, p 488. Dunlevy, J.R., Neame, P. J., Vergnes, J.P., and Hassell, J.R. (1998) Identification of the N-linked oligosaccharide sites in chick comeal lumican and keratocan that receive keratan sulfate. J. Biol. Chem. 273 (16), 9615-9621. Durand, L., Resal, R., and Burillon, C. (1985) Focus on an antomoclinical entity: Biber- Haab-Dimmer lattice dystrophy. J. Fr. Ophthalmol. 8, 729. Edward, D.P., Thonar, E.J., Srinivasan, M., Yue, B.J., and Tso, M.O. (1990) Macular comeal dystrophy of the comea. A systemic disorder of keratan sulfate metabolism. Ophthalmology 97, 1194-1200. Ehrlich, M.I., Phinney, R.B., Mondino, B.J., and Pettit, T.H. (1988) Techniques of lamellar keratoplasty. Int. Ophthalmol. Clin. 28, 24-29. Eiberg, H., Moller, H.U., Berendt, I., and Mohr, J. (1993) Assignment of granular comeal dystrophy Groneuw type I (CDGGI) to chromosome 5q. Eur. J. Hum. Genet. 2, SI 32-138 Elliot, J.H., Feman, S.S., O’Day, D.M., and Garber, M. (1985) Hereditary sclerocomea. Arch. Ophthalmol. 103, 676-679. 226 References Endo, S., Ha, N.T., Fujiko, K., et al. (1999) Leu518Pro mutation of the Big-h3 gene causes lattice comeal dystrophy type 1. Am. J. Ophthalmol. 128, 104-106. Eriksson, A.W., Lehmann, W., and Forsius, H. (1973) Congenital comea plana in Finland. Clin. Genet. 4 (4), 301-310. Escribano, J., Hemando, N., Gohosh, S., Crabb, J., and Coca-Prados, M. (1994) cDNA from human ocular ciliary epithelium homologous to Big-h3 is preferentially expressed in the comeal epithelium. J. Cell Physiol. 160, 511-521. Falls, H.F., and Allen, A.W. (1969) Dominantly inherited keratoconus. Report of a family. J. Genet. Hum. 17, 317. Fisher, L.W., Termine, J.D., and Young, M.F. (1989) Deduced protein sequence of bone small proteoglycan I (biglycan) shows homology with proteoglycan II (decorin) and several nonconnective tissue proteins in a variety of species. J. Biol. Chem. 264 (8), 4571-4576. Fogle, J.A., Kenyon, K.R., Stark, W.J., and Green, W.R. (1975) Defective epithelial adhesion in anterior comeal dystrophies. Am. J. Ophthalmol. 79, 925. Folberg, R., Alfonso, E., Croxatto, O., Driezen, N.G., Panjwani, N., Laibson, P R., Bomchoff, S.A., Baum, J., Malbran, E.S., Femandez-Meijide, R., Morrison, J.A., Bemardino, V.B., Arbizo, V.V., and Albert, D.M. (1988) Clinically atypical granular comeal dystrophy with pathologic features of lattice-like amyloid deposits. Ophthalmology 95, 46:51. Folberg, R., Stone, E.M., Sheffield, V.C., and Mathers, W.D. (1994) The relationship between granular, lattice type I, and Avellino comeal dystrophies. Arch. Ophthalmol. 112, 1080-1085. Forsius, H., Damsten, M., Eriksson, A., Fellman, J., Lindh, S., and Tahvanainen, E. (1998) Autosomal recessive comea plana. A clinical and genetic study of 78 cases in Finland, Ophthalmol. Scand. 76 (2), 196-203. Franceschetti, A. (1954) Classification and treatment of hereditary comeal dystrophies. Arch. O p h th a lm o l.1. Francois, J. (1966) Heredo-familial comeal dystrophies. Trans Ophthalmol Soc. UK 86, 367. Francois, J., Hansenns, M., Teuchy, H., and Sebruyns, M. (1975) Ultrastructural findings in comeal macular dystrophy (Groenouw type II). Ophthalmic Res. 7, 80-98. Friedman, T.B., Battey, J., Kachar, B., et al. (2000) Modifiers genes of hereditary hearing loss. Curr. Opin. Neurobiol. 10, 487-493. 227 References Frayer, W.C., and Blodi, P C. (1959) The lattice type of familial comeal degeneration: a histopathologic study. Arch. Ophthalmol. 61, 712-719. Fujiki, K., Hotta, Y., Nakayasu, K., Yokoyama, T., Tkano, T., Yamaguchi, T., and Kanai, A .A (1998a) A new L527R mutation of the Big-h3 gene in patients with lattice comeal dystrophy with deep stromal opacities. Hum. Genet. 103, 286-289. Fujiki, K., Hotta, Y., Nakayasu, K., and Kanai, A. (1998b) Homozygotic patient with Bigh-3 gene mutation in granular dystrophy. Comea 17, 288-290. Fukuta, M., Inazaw, J., Torri, T., Tsuzuki, K., Shimada, E., and Habuchi, O. (1997) Molecular cloning and charaterization of human keratan sulfate Gal-6-sulfotransferase. J. Biol. Chem. 272, 32321-32328. Fukuta, M., Kobayashi, Y., Uchimura, K., Kimata, K., and Habuchi, O. (1998) Molecular cloning and expression of human chondroitin 6-sulfotransferase. Biochem. Biophys. Act. 1399, 57-61. Fullwood, N.J., Davies, Y., Nieduszynski, I.A., Marcyniuk, B., Ridgeway, A.E.A., and Quantock, A.J. (1996) Cell surface-associated karatan sulphate on normal and migrating comeal endothelium. Invest. Ophthalmol. Vis. Sci. 37, 1256-1270. Funderburgh, J.L., Caterson, B., Conrad, G.W. (1987) Distribution of proteoglycans antigenically related to comeal keratan sulfate proteoglycan. J. Biol. Chem. 262, 11634- 11640. Funderburgh, J.L., Cintron, C., Cvington, H I., and Conrad, G.W. (1988) Immuno- analysis of keratan sulfate proteoglycans from comeal scars. Invest. Ophthalmol. Vis. Sci. 29, 1116-1124. Funderburgh, J.L., and Chandler, D.W. (1989) Proteoglycans of rabbit corneas with non perforating wounds. Invest. Ophthalmol. Vis. Sci. 30, 435-442. Funderburgh, J.L., Funderburgh, M.L., Mann, M.M., and Conrad, G.W. (1991) Physical and biological properties of keratan sulfate proteoglycans. Biochem. Soc. Trans 19, 871- 876. Funderburgh, J.L., Funderburgh, M.L., Brown, S.J., Vergnes, J.P., Hassell, J.R., Mann, M.M., and Conrad, G.W. (1993) Sequence and stmctural implications of a bovine comeal keratan sulfate proteoglycan core protein. J. Biol. Chem. 268, 11874-11880. Funderburgh, J.L., Funderburgh, M.L., Hevelon, N.D., Stech, M.E., Justice, M.J., Liu, C.Y., Kao, W.W.Y., and Conrad, G.W. (1995) Sequence, molecular properties, and chromosomal mapping of mouse lumican. Invest. Ophthalmol. Vis. Sci. 36 (11), 2296- 2303. 228 References Funderburgh, J.L., Corpuz, LM., Roth, M R., Funderburgh, M.L., Tasheva, E.S., and Conrad, G.W. (1997) Mimecan, the 25-Kda comeal keratan sulfate proteoglycan, is a product of the gene producing osteoglycin. J. Biol Chem. 272, 28089-280095. Gamer, A. (1969a) Histochemistry of comeal granular dystrophy. Br. J. Ophthalmol. 53, 799-807. Gamer, A. (1969b) Histochemistry of comeal macular dystrophy. Invest. Ophthalmol. Vis. Sci. 8, 473. Gamier, J., Osguthorpe, D.J., and Robson, B. (1978) Analysis of the implications of simple methods for predicting the secondary structure of globular proteins. J. Mol. Biol. 120, 97-120. Ghose, F.M., and Mehta, U. (1978) Microcomea with correctopia and macular hypoplasia in a family. Jpn. J. Ophthalmol. 28, 126. Gipson, I.K., and Joyce, N.C. (2000) Anatomy and cell biology of the comea, superficial limbus and conjunctiva. In: Albert, D M., Jakobiec, F.A., Azar, D.T., and Gragouds, E.S. eds. Principles and practice of ophthalmology, 2"^ ed. volume 2 Philadelphia, WB Saunders p614. Glenner, G.G. (1980) Amyloid deposits and amyloidosis: the beta-fibrilloses. N. Engl. J. Med. 302, 1283-1292. Grass, G.M., and Rabinson, J.R. (1988) Mechanisms of comeal dmg penetration I: In vivo and invitro kinetics, J. Pharmacol Sci. 77, 3. Grayson, M., and Wilbrandt, H. (1966) Dystrophy of the anterior limiting membrane of the comea (Reis-Bucklers’ type). Am. J. Ophthalmol 63, 345. Grayson, M. (1974) The nature of hereditary deep polymorphous dystrophy of the comea: its association with iris and anterior chamber dysgenesis. Trans. Am. Ophthalmol. Soc. 72, 516-559. Gregory, C.Y., Evans, K., and Bhattacharya, S.S. (1995) Genetic refinement of the chromosome 5q lattice comeal dystrophy type I locus to within a 2 cM interval. J. Med. Genet. 32, 224-226. Guerry, D. (1965) Observations on Cogan's microcystic dystrophy of comeal epithelium. Trans. Am. Ophthalmol. Soc. 63, 320. Gupta, S.K., Hodge, W.G., Damji, K.F., Guemsey, D.L., and Nuemann, P.E. (1998) Lattice comeal dystrophy type I in a Canadian kindred is associated with the Argl24->Cys mutation in the kerato-epithelin gene. Am. J. Ophthalmol. 125, 547-549. 229 References Gupta, S.K., and Hodge, W.G. (1999) A new clinical perspective of comeal dystrophies through molecular genetics. Current opinion in Ophthalmology 10, 234-241. Guex, N., and Peitsch, M.C. (1997) SWISS-MODEL and the Swiss-PdbViewer: an environment for comparative protein modelling. Electrophoresis 18 (15), 2714-2723. Guex, N., Diemand, A., and Peitsch, M.C. (1999). Protein modelling for all. Trends Biochem. Sci. 24 (9), 364-367. Guyer, M.S., and Collins, F.S. (1993) The human genome project and the future of medicine. Am. J. Dis. Child 147, 1145-1152. Haddad, R., Font, R.L., and Fine, B.S. (1977) Unusual superficial variant of granular dystrophy of the comea. Am. J. Ophthalmol. 83, 213-218. Hangai, M., Kaneda, Y., Tanihara, H., and Honda, Y. (1996) In vivo gene transfer into the retina mediated by a novel liposome system. Invest. Ophthalmol. Vis. Sci. 37 (13), 2678-2685. Happle, R., Daniels, O., and Koopman, R.J. (1993) MIDAS syndrome (microphthalmia, dermal aplasia, and sclerocomea): an X linked phenotype distinct from Goltz syndrome. Am. J. Med. Genet. 47, 710-713. Harrison, R.J., Klouda, P.T., Easty, D.L., et al. (1989) Association of keratoconus and atopy. Br. J. Ophthalmol. 73, 816. Hart, G.W. (1976) Biosynthesis of glycosaminoglycans during comeal development. J. Biol. Chem. 251, 6513-6521. Hasegawa, N., Takayoshi, T., Kato, T., Miyajima, H., Furuhata, A., Nakayasu, K., Kanai, A., and Habuchi, O. (2000) Decreased GlcNAc 6-0-sulfotransferase activity in the comea with macular comeal dystrophy. Invest. Ophthalmol. Vis. Sci. 41(12), 3670- 3677. Hassell, J R., Newsome, D.A., Krachmer, J.H., and Rodrigues, M. (1980) Macular comeal dystrophy. Failure to synthesis a mature keratan sulfate proteoglycan. Proc. Natl. Acad. Sci. USA 77, 3705. Hassell, J.R., Hascall, V.C., Ledbetter, S., Caterson, B., Thonar, E., Nakazawa, K., and Krachmer, J. (1984) Comeal proteoglycan biosynthesis and macular comeal dystrophy. In Scehffield, JB, and Hilffer SR. (eds.): Cell and Developmental biology of the Eye. Hereditary and Visual Development. Agw York, Springer-Verlagp.\0\. Hassell, J.R., Hascall, V.C., Ledbetter, S., et al. (1985) Comeal proteoglycan biosynthesis and macular comeal dystrophy. In: Sheffield, J.B., and Hilfer, S.R. eds. Hereditary and visual development. New York: Springer Verlag^\0\- \ 14. 230 References Hassell, J., Schrecengost, R., Rada, J., et ah (1992) Biosynthesis of stromal matrix proteoglycans and basement membrane components by human comeal fibroblasts. Invest. Ophthalmol. Vis. Sci. 33, 547. Hattori, M., Fujiyama A., Taylor, T.D., Watanabe, H., Yada, T., Park, H.S., Toyoda, Ishii, K., Totoki, Y., Choi, D.K., Soeda, E., Ohki, M., Takagi, T., Sakaki, Y., Taudien, S., Blechschmidt, K., et al. (2000) The DNA sequence of human chromosome 21. The chromosome 21 mapping and sequencing consortium. Nature 405 (6784), 311-319. Hay, E.D. (1980) Development of the vertebrate comea Int. Rev. Cytol. 61, 263. Hedbys, B.O. (1961) The role of polysaccharides in comeal swelling. Exp. Eye Res. 1, 81-91. Hersh, P.S., and Wagoner, M.D. (1998) Excimer Laser surgery for comeal disorders. New York, NY: Thieme Medical Publishers. Hida, T., Tsubota, K., Murata, H., Ogata, T., and Akiya, S. (1987a) Clinical features of a newly recognised type of lattice comeal dystrophy. Am. J. Ophthalmol. 104, 241-248. Hirano, K., Hotta, Y., Fujiko, K., et al. (2000) Comeal amyloidosis caused by Leu518Pro mutation of Bigh-3 gene. Br. J. Ophthalmol. 84, 583-585. Hirano, K., Hotta, Y., Nakamura, M., Fujiko, K., Kanai, A., and Yamamoto, N. (2001) Late onset form of lattice comeal dystrophy caused by Leu527Arg mutation of the TGFBI gene. Cornea 20, 525-529. Hittner, H.M., Kretzer, F.L., Antoszyk, J.H., Ferrell, R.E., and Mehta, R.S. (1982) Variable expressivity of autosomal dominant mesenchymal dysgenesis in six generations. Am. J. Ophthalmol. 93, 57-70. Hoefnagle, D., Keenan, M.E., and Allen, F.H. (1963) Heredofamilial bilateral anophthalmia. Arch. Ophthalmol. 69, 760-764. Holland, E.J., Daya, S.M., Stone, E.M., Folberg, R., Dobler, A.A., Cameron, J.D., and Doughman, D.J. (1992) Avellino comeal dystrophy. Clinical manifestations and natural history. Ophthalmolgy 1564-1568. Holliday, R. (1964) A mechanism for gene conversion in fungi. Genet. Res. 5, 282-304. Hotta, Y., Fuijiki, K., and Ono, K. (1998) Argl24Cys mutation of the Big-h3 gene in a Japanese family with lattice comeal dystrophy type 1. Jpn. J. Ophthalmol. 42, 450-455. Intemational Human Genome Sequencing Consortium (2001) Initial sequencing and analysis of the human genome. Nature 409, 860-921. 231 References Irvine, A.D., Corden, L.D., Swensson, O., Swensson, B., Moore, J.E., Frazer, D.G., Smith, F.J.D., Knowlton, R.G., Chrestophers, E., Rochels, R., Uitto, J., and McLean, W.H.I. (1997) Mutations in comea specific keratin K3 or K12 genes cause Meesman’s comeal dystrophy. A a / 16, 184-187. lozzo, R.V. (1997) The family of the small leucine-rich proteoglycans: key regulators of matrix assembly and cellular growth. Crit. Rev. Biochem. Mol. Biol. 32 (2), 141-174. lozzo RV (1999) The biology of the small leucine-rich proteoglycans (review). J. Biol. Chem 274,18843-46. Janosi, J.B., and Ramsland, P.A., (1999) The acid-labile subunit of the serum insulin like growth factor-binding protein complexes. Structural determination by molecular modeling and electron microscopy. J. Biol. Chem. 274 (33), 23328-23332. Jeffreys, A.J., Wilson, V., and Thein S.L. (1985) Hypervariable “minisatellites” regions in human DNA. Nature 314, 67-73. Johnson, H.J., Rosenberg, L., Chois, H.U., Garza, S., Hook, M., and Neame, P.J. (1997) Characterization of epiphycan, a small proteoglycan with a leucine-rich repeat core protein. J. Biol. Chem. 272 (30), 18709-18717. Jonasson, F., Oshima, E., and Thonar, E.J.M.A. (1996) Macular comeal dystrophy in Iceland: a clinical, geneologic, and immunohistochemical study of 28 patients. Ophthalmology 103, 1111-1117. Jones, S.T., and Zimmerman, S.E. (1961) Histopathologic differentiation of granular, macular, lattice dystrophies of the comea. Am. J. Ophthalmol. 51, 394-410. Judisch, G.F. and Maumenee, l.H. (1978) Clinical differentiation of recessive congenital hereditary endothelial dystrophy and dominant hereditary dystrophy. Am. J. Ophthalmol. 85, 606. Kakuta, Y., Pedersen, L.G., Carter, C.W., Negishi, M., and Pedersen, L.C. (1997) Crystal structure of oestrogen sulfotransferase. Nature Struct. Biol. 4, 904-908. Kakuta, Y., Pedersen, E.G., Pedrsen, L.C., and Negishi, M. (1998) Conserved structural motifs in the sulfotransferase family. Trends Biochem. Sci. 23, 129-130. Kallen, B., Robert, E., and Marris, J. (1996) The descriptive epidemiology of anophthalmia and microphthalmia. In. J. Epidemilo. 25, 1009-1116. Kanai, A., Waltman, S., Polack, F.M., and Kaufman, H E.. (1971) Electron microscopic study of hereditary comeal oedema. Invest. Ophthalmol. 10, 89. Kanai, A., Kaufman, H.E., and Polack, P.M. (1973) Electron microscpic study of Reis- Bucklers’ dystrophy. Ann. Ophthalmol. 5, 953-962. 232 References Kanai, A., Yamaguchi, T., and Nakajima, A. (1977) The histochemical and analytical electron microscopy studies of the comeal granular dystrophy [Japanese]. Nippon Ganka Gakkai Zasshi 81, 145-154. Kaufman, H.E., Barron, B.A., and McDonald, M B. (1998) The comea. 2^^ edition Butterworth-Heinemann. Kawasaki, S., Nishida, K., Quantock, A.J., Dota, A., Bennett, K., and Kinoshita, S. (1999) Amyloid and Pro501Thr mutated beta-ig-h3 gene product colocalize in lattice comeal dystrophy type IIIA. Am. J. Ophthalmol. 127, 456-458. Keen, J., Lester, D., Ingleheam, C., Curtis, A., and Bhattacharya S.S. (1991) Rapid detection of single base mismatches as heteroduplexes on hydrolink gels. Trends Genet. 7,5. Kennedy, R.H., Boume, W.M., and Dyer, D.A. (1986) A 48-year clinical and epidemiologic study of keratoconus. Am. J. Ophthalmol. 101, 267. Kennedy, S.M., McNamara, M., Hillery, M., et al. (1996) Combined granular lattice dystrophy (Avellino comeal dystrophy). Br. J. Ophthalmol. 80, 489-490. Kimbrough, R.L., Trempe, C.S., Brockhurst, R.J., and Simmons, R.J. (1979) Angle- closure glaucoma in nanophthalmos. Am. J. Ophthalmol. 88, 572-579. King, R.G., and Geeraets, W.J. (1969) Lattice or Reis-Bucklers' comeal dystrophy; A question of stromal pathology. Southern Medical Journal 62, 1163-1169. Klintworth, G.K., and Vogel, F.S. (1964) Macular comeal dystrophy. An inherited acid mucopolysaccaride storage disease of the comeal fibroblast. Am. J. Pathol. 45, 565-586. Klintworth, G.K. (1977) The comea-stmcture and macromolecules in health and disease. A review. Am. J. Pathol. 89, 719-808. Klintworth, G.K., and Smith, C.P. (1977) Macular comeal dystrophy. Am. J. Pathol. 89, 167. Klintworth, G.K. (1980) Research into the pathogenesis of macular comeal dystrophy. Trans. Ophthalmol. Soc. t/K 100, 186-194. Klintworth, G.K. (1982) Current concept of macular comeal dystrophy. In: Cotlier, E., Maumenee, I.H., and Berman, E.R. eds. Genetic eye disease: Retinitis Pigmentosa and Other Inherited Eye Disorders. New York: Alan R. Liss 463-477 (Birth Defects: Original article series; Vol. 18, no. 6). 233 References Klintworth, G.K., and Smith, C.F. (1983) Abnormalities of proteoglycans synthesised by comeal organ cultures derived from patients with macular comeal dystrophy. Lab. Invest. 48, 603. Klintworth, G.K., Meyer, R., Dennis, R. et al. (1986) Macular comeal dystrophy: lack of keratan sulfate in semm and comea. Paediatr. Genet. 7, 139-143. Klintworth, G.K., Enghild, J., and Valnickova, Z. (1994) Discovery of a novel protein (Big-h3) in normal human comea. ARVO abstracts. Invest. Ophthalmol. Vis. Sci. 35, SI 938. Klintworth, G.K., Oshima, E., Al-Rajhi, A., Al-Saif, A., Thonar, E.J., and Karcioglu, Z.A. (1997) Macular comeal dystrophy in Saudi Arabia: a review of 56 cases and recognition of a new immunophenotype. Am. J. Ophthalmol. 124, 9-18. Klintworth, G.K., Valinskova, Z., and Enghild, J.J. (1998) Accumulation of beta ig-h3 gene product in comeas with granular dystrophy. Am. J. Pathol. 152, 743-748. Klintworth, G.K. (1999) Advances in the molecular genetics of comeal dystrophies. Am. J. Ophthalmol. 128, 747-754. Kobe, B., and Deisenhofer, J. (1993) Crystal structure of porcine ribonuclease inhibitor, a protein with leucine-rich repeats. Nature 366 (6457), 751-756. Konishi, M., Yamada, Y., Nakamura, Y., Yamada, M., and Sugiura, H. (1997) Granular- lattice (Avellino) comeal dystrophy in Japanese patients. Cornea 16, 635-638. Korvatska, E., Munier, F.L., Zografos, L., Ahmad, F., et al. (1996) Delineation of a 1 cM region on distal 5q containing the locus for comeal dystrophies Grenouw type I and lattice type I and exclusion of the candidate genes SPARC and LOX. Eur. J. Hum. Genet. 4, 214-218. Korvatska, E., Munier, F.L., Djemai, A., Wang, M.X., Fmeh, B., Chiou, A.G.Y., Uffer, S., Ballestrazzi, E., Braunestein, R.E., Forster, R.K., Culbertson, W.W., Boman, H., Zografos, L., and Schorderet, D.F. (1998) Mutation hot spots in 5q31-linked comeal dystrophies. Am. J. Hum. Genet. 62, 320-324. Korvatska, E., Munier, F.L., Chaubert, P., Wang, M X., Mashima, Y., Yamada, M., Uffer, S., Zografos, S., and Schorderet, D.F. (2001) On the role of kerato-epithelin in the pathogenesis of 5q31-linked comeal dystrophies. Invest. Ophthalmol. Vis. Sci. 40 (10), 2213-2222. Krachmer, J.H., Mannis, M.J., and Holland, E.J. (1997) Comea, comea and extemal diseases, clinical diagnosis and management. Vol.2 Mosby St. Louis. 234 References Kresse, H., Liszio, C., Schonherr, E., and Fisher, L.W. (1997) Critical role of glutamate in a central leucine rich repeat of decorin for interaction with type I collagen. J. Biol. Chem. 272 (29), 18404-18410. Kruglyak, L. (1997) The use of genetic map of bialleleic markers in linkage studies. Nat. Genet. 17, 21-24. Krusius, T., and Ruoslahti, E. (1986) Primary structure of an extracellular matrix proteoglycan core protein deduced from cloned cDNA. Proc. Natl. Acad. Sci. USA 83 (20), 7683-7687. Krusius, T., Finne, J., Margolis, R.K., and Margolis, R.U. (1986) Identification of an O- glycosidic mannose-1 inked sialylated tetrasaccharide and keratan sulfate oligosaccharides in the chondroitin sulfate proteoglycan of brain J. Biol. Chem. 261, 8237-8242. Kublin, C.L., and Cintron, C. (1996) Big-h3 is associated with normal development of extra-cellular matrices in human foetal eyes. Invest. Ophthalmol. Vis. Sci. 37, S327. Kuchle, M., Green, W.R., Volker, H E., and Barraquer, J. (1995) Re-evaluation of comeal dystrophies of Bowman’s layer and the anterior stroma (Reis-Bucklers’ and the Theil-Behnnke types): a light and electron microscopic studies of eight comeas and a review of literature [review]. Cornea 14, 333. Kurita, K.T., Shinomura, T., Ujita, M., Zako, M., Kida, D., Iwata, H., and Kimata, K. (1996) Occurrence of PG-Lb, a leucine-rich small chondroitin/dermatan sulphate proteoglycan in mammalian epiphyseal cartilage: molecular cloning and sequence analysis of the mouse cDNA. Biochem. J . 318 (Pt 3) (12), 909-914. Laibson, P R., and Krachmer, J.H. (1975) Familial occurrence of dot (microcystic), map and fingerprint dystrophy of the comea. Invest. Ophthalmol Vis. Sci. 14, 397-399. Laibson, P R. (1997) Anterior comeal dystrophies. In: Krachmer, J.H., Mannis, M.J., and Holland, E.J., eds. Comea St. Louis: Mosby, vol. 2, Chap. 88. Lander, E.S., and Botstein, D. (1987) Homozygosity mapping: a way to map human recessive traits with the DNA of inbred children. Science 236, 1567-1570. Larsen, V., and Eriksen, A. (1948) Comea plana. Acta Ophthalmol. (Copenh) 27, 287- 286. Larsen, J.S. (1971) The sagittal growth of the eye. IV. Ultrasonic measurement of the axial length of the eye from birth to puberty. Acta Ophthalmol. 49 (6), 873-886. Lathrop, G.M., and Lalouel, J.M. (1984) Easy calculations of lod scores and genetic risks on small computers. Am. J. Hum. Genet. 36 (2), 460-465. 235 References Lee, J.K., Bhakta, S., Rosen, S.D., and Hammerich, S. (1999) Cloning and characterization of a mammalian N-acetylglucosamine-6-sulfotransferase that is highly restricted to intestinal tissue. Biochem, Biophys. Res. Commun. 263, 543-549. Lewis, D., Davies, Y., Nieduszynski, LA. Lawrence, F., Quantock, A.J., Bonshek, R., and Fullwood, N.J. (2000) Ultrastructural localisation of sulfated and unsulfated keratan sulfate in normal and macular comeal dystrophy type I. Glycobiol. 10 (3), 305-312. Liakos, G.M., and Casey, T.A. (1978) Posterior polymorphus keratopathy. Br. J. Ophthalmol. 62, 39-45. Liang, F., Holt, L, Pertea, G., Karamycheva, S., Salzberg, S.L., and Quackenbush, J. (2000) Gene index analysis of the human genome estimates approximately 120,000 genes. Nature Genet. 2, 239-240. Lisch, W., Steuhl, K.P., Lisch, C.,et al. (1992) A new, band shaped and whorled microcystic dystrophy of the comeal epithelium. Am. J. Ophthalmol. 114, 35-44. Lisch, W., Buttner, A., Oeffner, F., Boddeker, I., Engel, H., Lisch, C., Ziegler, A., and Grzeschik, K.H. (2000) Lisch comeal dystrophy is genetically distinct from Meesmann comeal dystrophy and maps to Xp22.3. Am. J. Ophthalmol. 130, 461-468. Liu, C.Y., Shiraishi, A., Kao, C.W., Converse, R.L., Funderburgh, J.L., Corpuz, L.M., Conrad, G.W., and Kao, W.W. (1998) The cloning of mouse keratocan cDNA and genomic DNA and the characterization of its expression during eye development. J. Biol. Chem. 273 (35), 22584-22588. Liu, N.P, Baldwin, J., Jonasson, F., Dew-Knight, S., Stajich, J.M., Lennon, F., Pericak- Vance, M.A., and Klintworth, G.K., and Vance, J.M. (1998a) Haplotype analysis in Icelandic families defines a minimal interval for macular comeal dystrophy type I gene. Am. J. Hum. Genet. 63, 912-917. Liu, N.P., Baldwin, J., Lennon, F., Stajich, J.M., Thonar, E.J., Pericak-Vance, M.A., Klintworth, G.K., and Vance, J.M. (1998b) Coexistence of macular comeal dystrophy types I and II in a single sibship Br. J. Ophthalmol. 82, 241-244. Liu, N.P., Dew-Knight, S., Jonasson, F., Gilbert, J R., Klintworth, G.K., and Vance, J.M. (2000a) Physical and genetic mapping of the macular comeal dystrophy locus on chromosome 16q and exclusion of TAT and LCAT as candidate genes. Mol. Vis. 6, 95- 100. Liu, N.P., Dew-knight, S., Rayner, M., Jonasson, F., Akama, T.O., Fukuda, M.N., Bao, W., Gilbert, J.R., Vance, J.M. and Klintworth, G.K. (2000b) Mutations in comeal carbohydrate sulfotransferase 6 gene (CHST6) cause macular comeal dystrophy in Iceland. Mol. Vis. 6, 261-264. 236 References Long, C.J., Roth, M.R., Tasheva, E.S., Funderburgh, M., Smit, R., Conrad, G.W., and Funderburgh, J.L. (2000) Fibroblast growth factor-2 promotes keratan sulfate proteoglycan expression by keratocytes in vitro. J. Biol. Chem. 275 (18), 13918-13923. Lucarelli, M.J., and Adamis, A.P. (1994) Avellino comeal dystrophy. Arch. Ophthalmol. 112,418. Mackey, D.A., Buttery, R.G., Wise, G.M., and Denton, M.J. (1991) Description of X- 1 inked megalocomea with identification of the gene locus. Arch, ophthalmol. 109, 829. Mannis, M.J., De Sousa, L.B., and Gross, R.H. (1997) The stromal dystrophies. In: Krachmer JH, Mannis MJ, Holland EJ, eds. Comea: diagnosis and management. Vol II. St. Louis: Mosby, 1043-1062. Mashima, Y., Imamura, Y., Konishi, M., et al. (1997) Homogeneity of kerato-epithelin codon 124mutations in Japanese patients with either of two types of comeal stromal dystrophies, J. Hum. Genet. 61, 1448-1450. Mashima, Y., Nakamura, Y., Noda, K., Konishi, M., Yamada, M., Kudoh, J., and Shimizu, N. (1999) A novel mutation at codon 124 (R124L) in the BIGH3 gene is associated with a superficial variant of granular comeal dystrophy. Arch. Ophthalmol. 117, 90-93. Marin, M.C., Jost, C.A., Brooks, L.A., Irwin, M.S., O’Nions, J., Tidy, J.A., James, N., McGregor, J.M., Harwood, C.A., Yulug, I.G., Vousden, K.H., Allday, M.J., Gusterson, B., Ikawa, S., Hinds, P.W., Crook, T., and Kaelin, W.G. (2000) A common polymorphism acts as intragenic modifier of mutant p53 behaviour. Nat. Genet. 25, 47- 54. Maurice, D.M. (1957) The structure and transparency of the comea. J. Physiol. 136, 263. Maurice, D.M. (1987) Kinetics of topically applied ophthalmic drugs. In Sacttone, M.F., Bucci, M., Speiser, P. (eds): Ophthalmic dmg delivery: biopharmaceutical, technological and clinical aspects. New York Springer-Verlag. Maury, C.P.J, Kere, J., Tolvanen, R. et al. (1990) Finnish hereditary amyloidosis is caused by a single nucleotide substitution in the gelsolin gene. FEBS Lett. 276, 75-77. McConkey, E.H. (1993) Gene mapping II Linkage analysis and genetic maps. In Human genetics, the molecular revolution (E.H. McConkey, ed) Jones and Bartlett, Boston London p68. Mc-Gee, T.L., Devoto, M., Ott, J., Berson, E.L., and Dry)a, T.P. (1997) Evidence that the penetrance of mutations at the RP11 locus causing dominant retinitis pigmentosa is influenced by a gene linked to the homologous RPll allele. Am. J. Hum. Genet. 61(5), 1059-66. 237 References McKusick, V.A. (1994). “Mandelian Inheritance in Man,” 1 ed., Johns Hopkins Univ. Press, Baltimore. McTigue, J.W. (1967) The human eomea: a light and electron microscopic study of the normal comea and its alterations in various dystrophies . Trans. Am. Ophthalmol. Soc. 65, 591. Meire, P.M. (1994) Megalcomea. Clinical and genetic aspects. Doc. Ophthalmol. 87, 1. Melles, G.R.J., Lander, P., Pels, E., Rietveld, P.J.R., Beekhuis, W.H., and Binder, B.S. (1998) A surgical technique for posterior lamellar keratoplasty. Cornea 17, 618-626. Meretoja, J. (1969) Pamilial systemic paramyloidosis with lattice dystrophy of the comea, progressive cranial neuropathy, skin changes and various intemal symptoms. Ann. Clin. Res. 1, 314. Meyer, K., Linker, A., Davidson, E.A., and Weissman, B. (1953) The mucopolysaccarides of the comea. J. Biol. Chem. 205, 611-616. Meyer, K. (1970) in Chemistry and Molecular Biology of the Intercellular matrix. (Balazs, EA, ed). Academic press New York p. 15 Midura, R.J., Hascall, V.C., MacCallum, O.K., Meyer, R.P., Thonar, E.J., Hassell, J R., Smith, C.P., and Klintworth, G.K. (1990) Proteoglycan biosynthesis by human comeas from patients with types I and II macular comeal dystrophy. J. Biol. Chem. 15, 256 (26), 15947-15955. Miller, A.D. (1992) Human gene therapy comes of age. Nature 357, 455-460. Miller, C.A., and Krachmer, J.H. (1998) Endothelial dytsrophies In: Kaufman, H.E., Barron, B.A., and McDonald, M.B. eds. The comea. Butterworth-Heinemann p472. Millin, G.A., Golub, B.M., Poster, C.S. (1986) Human basement mebrane components of keratoconus and normal comeas. Invst. Ophthalmol. Vis. Sci. 27, 604. Millington Ward, S., O’Neill, B., Tuohy, G., Al-Janda, N., Kiang, A., Kenna, P.P., Palfi, A., Hayden, P., Mansergh, P., Kennan, A., Humpheries, P., and Parrar, G.J. (1997) Stratigies in vitro for gene therapies directed to dominant mutations. Hum. Mol. Genet. 6, 1415-1426. Moller, H.U. (1989a) Granular comeal dystrophy Groenouw type 1 and Reis-Bucklers comeal dystrophy: one entity Ophthalmol. 67, 678-684. Moller, H.U. (1989b) Intrafamilial variability and interfamilial similarities of granular comeal dystrophy Grenouw type I with respect to biomicroscopical appearance and symptomatology. Acta Ophthalmol. (Cophen). 67, 669-677. 238 References Moller, H.U. (1990) Granular comeal dystrophy Groneuw type 1: Clinical aspects and treatment, Ophthalmol 68, 384-389. Morgan, T.H., and Cattell, E. (1912) Data for the study of sex linked inheritance in Dorsophila. J! Exp. Zool 13, 79-101. Morle, L.M., Bozon, M., Zech, J.C., Alloisio, N., Raas-Rothschild, A., Philippe, C., Lambert, J.C., Godet, J., Plauchu, H., and Edery, P. (2000) A locus for autosomal dominant colobomatous microphthalmia maps to chromosome 15ql2-ql5. Am. J. Hum. Genet. 67 (6), 1592-1597. Morton, N.E. (1955) Sequential tests for the detection of linkage. Am. J. Hum. Genet. 7, 277-318. Morton, N.E. (1956) The detection and estimation of linkage between the genes for elliptocytosis and the Rh blood type. Am. J. Hum. Genet. 8, 80-96. Munier, F.L., Korvatska, E., Djemai, A., Le Paslier, D., Zografos, L., Pescia, G., and Schorderet, D.F. (1997) Kerato-epithelin mutations in four 5q31 -linked comeal dystrophies. Aar. Genet. 15, 247-251. Nakamura, Y., Leppert, M., O’Connell, P., et al. (1987) Variable number of tandem repeat (VNTR) markers for human gene mapping. Science 235, 1616-1622. Nakazawa, K., Hassell, J R., Hascall, V.C., Lohmander, L.S., Newsome, D.A., and Krachmer, J. (1984) Defective processing of keratan sulfate in macular comeal dystrophy./. Biol. Chem. 259, 13751-13757. Nakazawa, K., Suzuki, S., Wada, K., and Nakazawa, K. (1995) Proteoglycan synthesis by comeal explants from developing embryonic chicken . 117, 707-718. Nishida, K., Honma, Y., Dota, A., Kawasaki, S., Adashi, W., Nakamura, T., Quantock, A.J., Hosotani, H., Yamamoto, S., Okada, M., Shimomura, Y., and Kinoshita, S. (1997) Isolation and chromosomal localisation of a comea specific human keratin 12 gene and detection of four mutations in Meesamnn’s comeal epithelial dystrophy. Am. J. Hum. Genet. 61, 1268-1275. Nishida, T. (1997) Basic Science: Comea, sclera, and ocular adnexa anatomy, biochemistry, physiology and biomechanics. In: Krachmer, J.H., Mannis, M.J., and Holland, E.J. eds. Comea: Fundementals of comea and extemal disease vol. 1 Mosby St. Louise p5-6. Norio, R., Nevanlinna, H R., and Perheentupa, J. (1973) Hereditary diseases in Finland; rare flora in rare soul. Ann. Clin. Res. 5 (3), 109-141. 239 References Oeben, M., Keller, R., Stuhlsatz, H.W., and Greiling, H. (1987) Constant and variable domains of different disaccharide structure in comeal keratan sulfate chains. Biochem. J. 248, 85-93. Okada, M., Yamamoto, S., Tsujikawa, M., Watanabe, H., Inoue, Y., Maeda, N., Shimomura, Y., Nishida, K., Quantock, A.J., Kinoshita, S., and Tano, Y. (1998a) Two distinct kerato-epithelin mutations in Reis-Bucklers’ comeal dystrophy. Am. J. Ophthalmol. 126, 535-542. Okada, M., Yamamoto, S., Inoue, Y., Watanabe, H., Maeda, N., Shimomura, Y., Ishii, Y., and Tano, Y. (1998b) Severe comeal dystrophy phenotype caused by homozygous R124H keratoepithelin mutations. Invest. Ophthalmol. Vis. Sci. 39, 1947-1953. Oldberg, A., Antonsson, P., Lindblom, K., and HeinegSrd, D. (1989) A collagen-binding 59-kd protein (fibromodulin) is structurally related to the small interstitial proteoglycans PG-Sl and PG-S2 (decorin). Embo. J . 8 (9), 2601-2604. O’Rahilly, R. (1983) The timing and sequence of events in the development of the human eye and ear during the embryonic period proper. Anat. Embryol. 168, 87. Othman, M l., Sullivan, S.A., Skuta, G.L., Cockrell, D.A., Stringham, H.M., Downs, C.A., Pomes, A., Mick, A., Boehnke, M., Vollrath, D., and Richards, J.E. (1998) Autosomal dominant nanophthalmos (NNOl) with high hyperopia and angle-closure glaucoma maps to chromosome 11, J. Hum. Genet. 63 (5), 1411-1418. Pearce, W.G. (1991) Autosomal dominant megalocomea with congenital glucoma: evidence of germ line mosaicism. Can. J. Ophthalmol. 26, 21. Peitsch, M.C. (1996) ProMod and Swiss-Model: Intemet-based tools for automated comparative protein modelling. Biochem. Soc. Trans. 24 (1), 274-279. Pellegata, N.S., Dieguez-Lucena, J.L., Joensuu, T., Lau, S., Montgomery, K.T., Krahe, R., Kivela, T., Kucherlapati, R., Forsius, H., and de la Chapelle, A. (2000) Mutations in KERA, encoding keratocan, cause comea plana. Nat. Genet. 25 (1), 91-95. Percin, F.E., Ploder, L.A., Yu, J.J., Arid, K., Horsford, J.D., Rutherford, A., Bapat, B., Cox, D.W., Duncan, A.M., Kalnins, V.I, Kocak-Altintas, A., Sowden, J.C., Traboulsi, E., Sarfarazi, M., and Mclnnes, R.P. (2000) Human microphthalmia associated with mutations in the retinal homeobox gene CHXIO. Nat. Genet. 25 (4), 397-401. Perry, H.D., Buxton, J.N., and fine, B.S. (1980) Round and oval cones in keratoconus. Ophthalmology 87, 905. Pressman, C.L., Chen, H., and Johnson, R.L. (2000) LMXIB, a LIM homeodomain class transcription factor, is necessary for normal development of multiple tissues in the anterior segment of the murine eye. Genesis 26 (1), 15-25. 240 References Quigley, H.A., and Goldberg, M.F. (1971) Scheie syndrome and macular comeal dystrophy. Ophthalmol 85, 553. Rada, J.A., Comuet, P.K., and Hassell, J.H. (1993) Regulation of comeal collagen fibrillogenesis in vitro by comeal proteoglycan (lumican and decorin) core proteins. Exp. Eye Res. 56 (6), 635-648. Randall, V.T., Edelhauser, H.F., Leibowitz, H.M., and Fredddo, T.F. (1998) Comeal structure and function. In: Leibowitz, H.M., and Waringlll, G.O. eds Comeal disorders clinical diagnosis and management 2"^ edition Philadelphia WB Saunders plO. Rawe, I.M., Zhan, Q., Komai, Y., Burrows, B., Kublin, C.L., Bennett, K., and Cintron, C. (1995) Localization and characterization of a novel extracellular matrix protein Big- h3.Invest. Ophthalmol. Vis. Sci. 36,S27. Robin, S B., Epstein, R.J., and Kommehl, E.W. (1994) Band-shaped, whorled, microcystic comeal dystrophy. yJ/w. J. Ophthalmol. 117, 543. Rodrigues, M.M., Fine, B.S., Laibson, P.R., and Zimmerman, L.E. (1974) Disorders of the comeal epithelium: A clinicopathologic study of dot, geographic, and fingerprint pattems. Arch. Ophthalmol. 92, 475. Rodrigues, M.M., Waringlll, G.O., Mackett, J., et al. (1982) Comea In Jakobiec F A. ed. Ocular Anatomy, Embryology, and Teratology. Philadelphia, Harper and Row. Rodrigues, M.M., Streeten, B.W., Krachmer, J.H., Laibson, PR., Salem, N.Jr., Passonnoue, J., and Chock, S. (1983) Microfibrillar protein and phospholipid in granular comeal dystrophy. Ophthalmol 101, 802-810. Rodrigues, M.M., Ben-Zvi, A., Krachmer, A., and Sun, T.T. (1987) Suprabasal expression of a 65-Kda keratin (No.3) in developing human comeal epithelium. Differentiation 34, 60-67. Rodrigues, M.M., and Krachmer, M.D. (1988) Recent advances in comeal stromal dystrophies. Cornea 7, 19-29. Rosenwasser, G.O., Sucheski, B.M., Rosa, N., Pastena, B., Sebastiani, A., Sassani, J.W., and Perry, H.D. (1993) Phenotypic variation in combined granular-lattice (Avellino) comeal dystrophy, Ophthalmol 111, 1546-1552. Rozzo, C., Fosarello, M., Galleri, G., Sole, G., Serru, A., Orzalesi N, Serra A, and Pirastu M. (1998) A common Big-h3 gene mutation (AF540) in a large cohort of Sardinian Reis-Bucklers’ comeal dystrophy patients. Hum. Mut. 12, 215-216. Saiki, P.K., Schorf, S., Faloona, F., Mullis, K.B., Hom, G.T., Ehrlich, H.A., and Amheim, N., (1985) Enzymatic amplification of beta globin sequences and restriction site analysis for diagnosis of sickle cell anaemia. Science 230, 1350-1354. 241 References Saiki, R.K., Gefland, D.H., Stoffel, S., Scharf, S., Higuchi R., Hom, G.T., Mullis, K.B., and Ehlrich, H.A. (1988) Primer directed enzymatic amplification of DNA with thermostable DNA polymerase. Science 239, 487-491. Salmon, J.F., Wallis, G.E., and Murray, A.D. (1988) Variable expressivity of autosomal dominant microcomea with cataract. Arch. Ophthalmol 105, 506-510. Sambrook, J., Fritsch, E.F., and Maniatis, T. (1989) Molecular cloning: a laboratory manual. Cold Spring Harbor Laboratory Press. Cold Spring Harbor, NY. Santo, R.M., Yamagushi, T., Kanai, A., Okisaka, S., and Nkajima, A. (1995) Clinical and hidtopathological features of comeal dystrophies in Japan. Ophthalmology 102, 557- 567. Sassani, J., Smith, S.G., and Rabinowitz, Y.S. (1992) Keratoconus and bilateral lattice- granular comeal dystrophies. Cornea 11, 343-350. Schermer, A., Galvin, S., and Sun, T.T. (1986) Differentiation related expression of a major 64K comeal keratin in vivo and in culture suggests limbal location of comeal epithelial stem cells. J. Cell Biol 103, 49-62. Schmitt-Bemard, C.F., Guittard, C., Amaud, B., Démaillé, J., Argiles, A., Claustres, M., and Tuffery-Giraud, S. (2000) BIGH3 exon 14 mutations lead to intermediate type I/IIIA of lattice comeal d y s t r o p h i e s . Ophthalmol Vis. Sci. 41, 1302-1308. Schonherr, E., Hausser, H., Beavan, L., and Kresse, H. (1995) Decorin type I collagen interaction. J. Biol. Chem. 270, 8877-8883. Schorderet, D.F., Menasche, M., Morand, S., Bonnel, S., Buchillier, V., Marchant, D., Auderset, K., Bonny, C., Abitbol, M., and Munier, F. (2000) Genomic characterization and embryonic expression of the mouse BIGH3 (Tgfbi) gene. Biochem. and Biophys. Res. Commun. 274, 267-274. Schorderet, D.F., Morand, S., Bonny, C., and Munier, F.L. (2000) Overexpression of mutated BIGH3 induces apoptosis in an in vitro system. (ARVO abstract no. 1405- B780). Scott, J.E., and Haigh, M. (1988) Identification of specific binding sites for keratan sulfate proteoglycans and chondroitin-dermatan sulfate proteoglycans on collagen fibrils in comea by the use of cupromeronic blue in ‘critical-electrolyte-concentration’ techniques. Biochem. J. 253, 607-610. Scott, J.E. (1991) Proteoglycan: Collagen interactions and comeal ultrastmcture. Biochem. Soc. Trans. 19, 877-881. 242 References Scott, J.E. (1996) Proteodermatan and proteokeratan sulfate (decorin, lumican/fibromodulin) proteins are horseshoe shaped. Implications for their interactions with collagen. Biochemistry 35 (27), 8795-8799. Sedundo, W., Marshall, G.E., Lee, W.R., and Kirkness, C.M. (1994) Immunoelectron labelling of matrix components in congenital hereditary endithelial dystrophy. Graefes Arch. Clin. Exp. Ophth(ümol. 232, 237. Sekundo, W., et al. (1994) An ultrastructural investigation of an early manifestation of posteriopr polymorphus dystrophy of the comea. Ophthalmology 101, 1422-1431. Sharkey, J.A., Kervick, G.N., Jackson, A.J., and Johnston, P.B. (1992) Comea plana and sclerocomea in association with recessive epidermolysis bullosa dystrophica. Case report. Cornea 11, 83 Shearman, A.M., Hudson, T.J., Andresen, J.M. Wu, Z., Sohn, R.L., Haluska, F., Housman, DE., and Weiss, J.S. (1996) The gene for Schnyder’s crystalline comeal dystrophy maps to human chromosome Ip34.1-p36. Hum. Mol. Genet. 5, 1667-1672. Sheffield, V.C., Beck, J.S., Kwitek, A.E., Sandstrom, D.W., and Stone, E.M. (1993) The sensitivity of single strand conformational polymorphism analysis for the detection of single base substitutions. Genomics 16, 325-332. Singh, O.S., Simmons, R.J., Brockhurst, R.J., Singh, O.S., et al. (1982) Nanophthalmos: a perspective on identification and therapy. Ophthalmology 89 (9), 1006-1012. Skonier, J., Neubauer, M., Madisen, L., Bennett, K., Plowman, G.D., and Purchio, A.F. (1992) cDNA cloning and sequence analysis of Big-h3, a novel gene induced in a human adenocarcinoma cell line after treatment with a transforming growth factor-)). DNA Cell Biol. 7,511-522. Skonier, J., Bennett, K., Rothwell, V., et al. (1994) Big-h3: A transforming growth factor-P- responsive gene encoding a secreted protein that inhibits cell attachment in vitro and suppresses the growth of CHO cells in nude mice. DNA Cell Biol. 13, 571-584. Small, K.W., Mullen, L., Barletta, J., Graham, K., Glasgow, B., Stem, G., and Yee, R. (1996) Mapping of Reis-Bucklers comeal dystrophy to chromosome 5q. Am. J. Ophthalmol. 121, 382-390. Snip, R.C., Kenyon, K.R., and Green, W.R. (1973) Macular comeal dystrophy. Ultrastructural pathology of comeal endothelium and Descemet’s membrane. Invest. Ophthalmol. Vis. Sci. 12, 88-97. Sommarin, Y., Wendel, M., Shen, Z., Heilman, U., and Heinegard, D. (1998) Osteoadherin, a cell-binding keratan sulfate proteoglycan in bone, belongs to the family of leucine-rich repeat proteins of the extracellular matrix. J. Biol. Chem. 273 (27), 16723-16729. 243 References Stainer, G.A., Akers, P.H., Binder, P.S., and Zavala, E.V. (1982) Correlative microscopy and tissue culture of congenital hereditary endothlial dystrophy. Am. J. Ophthalmol. 93, 456. Stam, L.F., and Laurie, C.C. (1996) Molecular dissection of a major gene effect on a qualitative trait: The level of alcohol dehydrogenase expression in Drosophila melanogaster. Genetics 144, 1559-1564. Starck, T., and Kenyon, K. (1997) Clinical and histopathlogical studies of two families with lattice comeal dystrophy and familial systemic amyloidosis (Meretoja syndrome). Ophthalmology 9% 1197. Stephens, F.E. (1947) A case of sex linked microphthalmia. J. Hered. 38, 307-310. Stephens, J.C., Cavanaugh, M.L., Gradie, M.I., et al. (1990) Mapping the human genome current status. Science 250,237-244. Stewart, H.S., Ridgway, A.E., Dixon, M.J., Bonshek, R., Parveen, R., and Black, G. (1999a) Heterogeneity in granular comeal dystrophy: Identification of three causative mutations in the TGFB (BIGH3) gene-Lessons for comeal amyloidogenesis. Hum. Mut. 14, 126-132. Stewart, H., Black, G.C.M., Donnai, D., Bonshek, R.E., McCarthy, J., Morgan, S., Dixon, M.J., and Ridgway, A.E.A. (1999b) A mutation within exon 14 of the TGFBI (BIGH3) gene on chromosome 5q31 causes an asymmetric late onset form of lattice comeal dystrophy. Ophthalmology 106, 964-970. Stock, E.L., Feder, R.S., O'Grady, R.B., Sugar, J., and Roth, S.L (1991) Lattice comeal dystrophy type IIIA. Arch. Ophthalmol. 109, 354-358. Stocker, F.W., and Holt, L.B. (1955) Rare form of hereditary epithelial dystrophy. Arch. Ophthalmol. 536. Stoll, C., Alembik, Y., Dott, B., and Roth, M.P. (1992) Epidemiology of congenital eye malformations in 131, 760 consecutive births. Ophthalmic pediatr. Genet. 13, 179-816. Stone, E.M., Mathers, W.D., Rosenwasser, G.O.D., Holland, E.J., Folberg, R., Krachmer, J.H., Nicols, B.E., Gorevic, B.D., Taylor, C.M., Streb, L.M., Fishbaugh, J.A., Daley, T.E., Sucheski, B.M., and Sheffield, V.C. (1994) Three autosomal dominant comeal dystrophies map to chromosome 5q. Nat. Genet. 6, 47-51. Sturetevant, A.H. (1913) The linear arrangement of sex linked factors in Dorsophila, as shown by their mode of association. J. Exp. Zool. 14, 43-59. Sugita, J., and Kondo, J. (1997) Deep lamellar keratoplasty with complete removal of pathological stroma for vision improvement. Br. J. Ophthalmol. 81, 184-188. 244 References Tai, G.H., Huckerby, T.N., and Nieduszynski, I.A. (1996) Multiple non reducing chain termini isolated from bovine comeal keratan sulfates. J. Biol. Chem. 271, 23535-23546. Tai, G.H., Nieduszynski, I.A., Fullwood, N.J., and Huckerby, T.N. (1997) Human comeal keratan sulfates. J. Biol. Chem. 272, 28227-28231. Takacs, L., Boross, P., Tozser, J., Modis, L., Toth, G., and Berta, A. (1998) Transforming growth factor-beta induced protein. BetaIG-H3, is present in degraded form and altered localization in lattice comeal dystrophy type I. Exp. Eye Res. 66, 739- 745. Tahvanainen, E., Forsius, H., Karila, E., Ranta, S., Eerola, M., Weissenbach, J., Sistonen, P., and de la Chapelle, A. (1995a) Comea plana congenita gene assigned to the long arm of chromosome 12 by linkage analysis. Genomics 26 (2), 290-293. Tahvanainen, E., Forsius, H., Damsten, M., Karila, E., Kolehmainen, J., Weissenbach, J., Sistonen, P., and de la Chapelle, A. (1995b) Linkage disequilibrium mapping of the comea plana congenita gene CNA2. Genomics 30 (3), 409-414. Tahvanainen, E., Villanueva, A.S., Forsius, H., Salo, P., and de la Chapelle, A. (1996) Dominantly and recessively inherited comea plana congenita map to the same small region of chromosome 12. Genome Res. 6 (4), 249-254. Tasheva, E.S., Funderburgh, J.L., Corpuz, L.M., and Conrad, G.W. (1998) Cloning, characterization and tissue-specific expression of the gene encoding bovine keratocan, a comeal keratan sulfate proteoglycan. Gene 218 (1-2), 63-68. Tasheva, E.S., Funderburgh, J.L., Funderburgh, M.L., and Conrad, G.W. (1999) Structure and sequence of the gene encoding human keratocan. DNA Seq. 10 (1), 67-74. Tasheva, E.S., Pettenati, M., and Von Kap-Her, C. (2000a). Assignment of keratocan gene (KERA) to human chromosome band 12q22 by in situ hybridization. Cytogenet. Cell Genet. 88 (3-4), 244-245. Tasheva, E.S., Conrad, A.H., and Conrad, G.W. (2000b) Identification and characterization of conserved cis-regulatory elements in the human keratocan gene promoter. Biochim. Biophys. Acta 1492 (2-3), 452-459. Tasheva, E.S., Pettenati, M., et al. (2000c). Assignment of mimecan gene (OGN) to human chromosome band 9q22 by in situ hybridization. Cytogenet. Cell Genet. 88 (3-4), 326-327. Thompson, J.D., Higgins, D.G., and Gibson, T.J. (1994) Clustal W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Research 22, 4673-4680. 245 References Thonar, E.J., Lenz, M.E., Klintworth, G.K., Caterson, B., Pachman, L.M., Glickman, P., Katz, R., Huff, J. and Kuettner, K.E. (1985) Quantification of keratan sulfate in blood as a marker of cartilage metabolism. Arthritis Rheum. 28, 1367-1376. Thonar, E.J., Meyer, R.F., Dennis, R.F., Lenz, M.E., Maldonado, B., Hassell, J.R., Hewitt, A.T., Stark, W.J., Stock, E.L., Kuettner, K.E., and Klintworth, G.K. (1986) Absence of normal keratan sulfate in the blood of patients with macular comeal dystrophy, J. Ophthalmol. 102, 561-569. Tisdale, A.S., Spurr-Michud, S.J., Rodrigues, M., et al. (1988) Development of the anchoring structures of the epithelium in rabbit and human fetal comeas. Invest. Ophthalmol. Vis. Sci. 29, 727. Toma, N.M.G., Ebenezer, N.D., Ingleheam, C.F., Plant, C., Ficker, L.A., and Bhattacharya, S.S. (1995) Linkage of congenital hereditary endothelial dystrophy to chromosome 20. Hum. Mol. Genet. 4, 2395-2398. Trelstad, R.L., Hayashi, K., and Toole, B.P. (1972) Epithelial collagen and glycosaminoglycans in the embryonic comea: Macromolecular order and morphogenesis in the embryonic basement membrane. J. Cell Biol. 62, 815. Ubles, J.L., and McRae, S.M. (1984) Vitamin A is present as retinol in the tears of humans and rabbits. Curr. Eye Res. 3, 815. Uchimura, K., Muramatsu, H., Kaname, T., Ogawa, H., Yamakawa, T., Fan, Q.W., Mitsuoka, C., Kannagi, R., Habuchi, O., Yokoyama, L, Yamamura, K., Ozaki, T., Nakagawara, A., Kadomasto, K., and Muramastu, T. (1998) Human N- acetylglucosamine-6-O-sulfotransferase involved in the biosynthesis of 6-sulfo sialyl Lewis X: molecular cloning, chromosomal mapping, and expression in various organs and tumour cells. J. Biochem. 124, 670-678. Vail, D.T. (1931) Adult hereditary anterior megalophthalmos in glaucoma: a definite disease entity. Arch. Ophthalmol. 6, 39. Vance, J.M., Jonasson, F., Sarrica, J., Damji, K.F., Stauffer, J., Pericak-Vance, M.A., and Klintworth, G.K. (1996) Linkage of a gene for macular comeal dystrophy to chromosome 16. J. Hum. Genet. 58, 757-762. Van Doreen, B.T.H., Melles, G.R.J., Lander, F., Beekhuis, W.H., and Binder, B.S. (1999) Preliminary clinical results of posterior lamellar keratoplasty. Invest. Ophthalmol. Vis. Sci. 40, 3324. Van Setten, G., and Schultz, G. (1994) Transforming growth factor alpha is a constant component of human tear fluid, Graefe's yfrc/z. Clin. Exp. Ophthalmol. 232, 523-526. Venter, J.C., et al. (2001) The sequence of the human genome. Science 291, 1304-1351. 246 References Vilchis, E., Kuchle, M., Handel, A., Seitz, B., Rummelt, V., and Naumann, G.O.H. (1996) Incidence of recurrent comeal stromal dystrophies after penetrating keratoplasty. Invest. Ophthalmol. Vis. Sci. Suppl. 37, 556. Villanueva, A S., Tahvanainen, E., Lindh, S., Lucena, J.D., and Forsius, H. (1997) Autosomal dominant comea plana: clincal findings in a Cuban family and a review of the literature. Ophthalmic. Genetics 18 (2), 55-62. Vingolo, E.M., Steindl, K., and Forte, R. (1994) Autosomal dominant simple microphthalmos. J. Med. Genet. 31 (9), 721-725. Vogt, A. (1981) Textbook and atla sof slit lamp. In: Microscopy of the living eye. Vol. 1 Springer-Verlag, Bonn. Voronina, V.A., and Mathers, P.H.(2000) Mutations in the RX gene in a patient with anophthalmia. Paper presented at the ARVO Annual Meeting, Fort Luderdale, FL, April 30-May 5. Warburg, M. (1993) Classification of microphthalmos and coloboma. J. Med. Genet. 30 (8), 664-669. Waring 111, G O., Rodrigues, M.M., and Laibson, P R. (1978) Comeal dystrophies. In: Dystrophies of the epithelium, Bowman’s layer and stroma. Surv. Ophthalmol. 23, 71- 122. Waring 111, G O., and Mbekeani, J.N. (1998) Comeal dystrophies. In: Leibowitz, H.W., and Waring 111, G.O. eds. Comeal disorders. Clinical diagnosis and management 2nd ed. Philadelphia: WB Saunders p245-251. Weblin, T.P., Hirst, L.W., Stark, W.J., and Maumenee, l.H. (1981) Prevalence of map dot finger print change in the comea. Br. J. Ophthalmol. 65, 401-409. Weiss, A.H., Kousseff, B.G., Ross, E.A., and Long bottom, J. (1989a) Simple microphthalmos Ophthalmol. 107,1625-1630. Weiss, A.H., Kousseff, B.G., Ross, E.A., and Long bottom, J. (1989b) Complex microphthalmos, Ophthalmol. 107,1619-1624. Wiggs, J.L. (1993) Advances in ophthalmic genetics and heritable eye diseases. In: Jakobeic, F.A., and Lamcin, J.C. eds Intemational ophthalmology clinics. Little, Brown and Company, Boston p i-30. Wilson, S.E., Lin, D.T., and Klyce, S.D. (1991) Comeal topography of keratoconus. Cornea 10, 2. 247 References Witschel, H., and Sundmacher, R. (1979) Bilateral recurrence of granular comeal dystrophy in the grafts: a clinico-pathologic study. Albrecht Von Graefes Arch Klin. Exp. ophthalmol. 209, 179-188. Wittebol, P.D., Von Bijsterveld, O.P., and Delleman, J.W. (1987a) Meesmann’s epithelial dystrophy of the comea. Biometrics and a hypothesis Ophthalmologica 194, 44. Wittebol,P.D., Vanderwant, J.J., and Van Bijsterveld, O.P. (1987b) Granular dytsrophy of the comea (Groenouw’s type I): Is the keratocyte the primary source after all ? Ophthalmologica 195, 169-177. Wolf, H., Modrow, S., Motz, M., Jameson, B.A., Hermann, G., and Fortsch, B. (1988) An integrated family of amino acid sequence analysis programs. Comput. Appl. Biosci. 4, 187-191. Wright, F.A. (1997) Gene therapy for the eye. Br. J. Ophthalmol. 81 (8), 620-623. Wright, F.A., Lemon, W.J., Zhao, W.D., Sears, R., Zhuo, D., Wand, J.P., Yang, H.Y., Baer, T., Stredney, D., Spitzner, J., Stutz, A., Krahe, R., and Yuan, B. (2001) Adraft annotation and overview of the human geneome. Genome Biol. 2(7), research0025.1- 0025.18. Yamagushi, T., Polack, F.M., and Valenti, J. (1980) Electron microscopic study of recurrent Reis-Bucklers’ comeal dystrophy, yl/». J. Ophthalmol. 90, 95-101. Yamagushi, T., Polack, F.M., and Rowsey, J.J. (1982) Honeycomb-shaped comeal dystrophy: a variation of Reis-Bucklers dystrophy. Cornea 1, 71-83. Yamamoto, S., Okada, M., Tsujikawa, M., Shimomura, Y., Nishida, K., Inoue, Y., Watanabe, H., Maeda, N., Kurahashi, H., Kinoshita, S., Nakamura, Y., and Tano, Y. (1998) A kerato-epithelin (Big-h3) mutation in lattice comeal dystrophy type IIIA. Am. J. Hum. Genet. 62, 719-722. Yang, C.J., SandarRaj, N., Thonar, E.J., and Klintworth, G.K. (1988) Immunohistochemical evidence of heterogeneity in macular comeal dystrophy. Am. J. Ophthalmol. 106, 65-71. Yee, R.W., Sullivan, L.S., Lai, H.T., Stock, E.L., Lu, Y., Khan, M.N., Blanton., S.H., and Daiger, S.P. (1997) Linkage mapping of Thiel Behnke comeal dystrophy (CDBII) to chromosome 1 Gq23-q24.Genomics 46, 152-154. Yue, B.Y., Kurosawa, A., Duvall, J., Goldberg, M.F., Tso, M.F., and Sugar, J. (1988) Nanophthalmic sclera. Fibronectin studies. Ophthalmology 95 (1), 56-60. Zimmerman, D., Tmeb, B., Winterhalter, K., et al. (1986) Type III collagen is a major component of the human comea. FEBS Lett 197, 55. 248 References Zinn, K.M., and Mockel-Pohl S. (1975) Fine structure of the developing comea. Int. Ophthalmol. Clin. 15, 19. Zinn, K., McAllister, L., and Goodman, C.S. (1988) Sequence analysis and neuronal expression of fasciclin I in grasshopper and Dorsophila. Cell 53, 577-587. Zlotogara, J., Legum, C., Raz, J., Merin, S., and BenEzra, D. (1994) Autosomal recessive colobomatous microphthalmia. Am. J. Med. Genet. 49, 261-262. 249 Presentations and publications PRESENTATIONS AND PUBLICATIONS ARISING FROM THIS WORK Presentations: Oral Presentation: 1 .Identification of novel mutations in a carbohydrate sulfotransferase gene CHST6 causing macular comeal dystrophy (2001). lO VS 42 (4) 622 2. A novel mutation in keratocan causes autosomal recessive comea plana and microphthalmia (2001). lO VS 42 (4) 623 3. Identification of novel mutations in CHST6 gene responsible for macular comeal dystrophy (2001). The Comeal Conference, Cardiff, UK 4. Positional cloning of the gene for CHED 1: A Yac/Pac and transcript map spanning the locus on 20pl 1.2 (2000). lO VS 41(4) 2867 Poster Presentations: Homozygozity mapping of a Bangladeshi family with comea plana (CNA2) to chromosome 12. (2000)/OE5 41(4) 1400 Publications: 1.Lehmann OJ*, El-Ashry MF*, Ebenezer N, Khaw PT, Francis PJ, Ficker L, Bhattacharya SS (2001): A mutation in keratocan causes autosomal recessive comea plana and microphthalmia. lOVS (in press). * Joint first authorship 2. Mohamed F. EL-Ashry, Mai M. Abd El-Aziz, Simon Wilkins, Michael E. Cheetham, Susan E. Wilkie, Alison J. Hardcastle, Stephanie Halford, Ahmed Y.Bayoumi, Linda A. Ficker, Stephen Tuft, Shomi S. Bhattacharya, and Neil D. Ebenezer (2001). Identification of novel mutations in a carbohydrate sulfotransferase gene (CHST6). lO VS (submitted). 3. El-Ashry MF, Abd Elaziz MM, Ebenezer ND, Ficker LA, Bhattacharya SS: Kerato- epithelin mutation in a Bangladeshi family with Reis-Bucklers' and lattice comeal dystroph (2001) (Manuscript in preparation) 4. El-Ashry MF, Abd Elaziz MM, Ebenezer ND, Hardcastle A, Bhattacharya SS. The spectrum of B1GH3 mutations and polymorphisms in the British population. (2001) (Manuscript in preparation) 250