JOURNAL TRANSCRIPT
THE ROLE OF PTTG AND PBF IN GENOMIC INSTABILITY AND DNA REPAIR IN THYROID CANCER
By Jim, Chi Wai Fong
A thesis presented to the College of Medical and Dental Sciences at the University of Birmingham for the Degree of Doctor of Philosophy
Centre for Endocrinology, Diabetes and Metabolism, School of Clinical and Experimental Medicine
March 2015
University of Birmingham Research Archive e-theses repository This unpublished thesis/dissertation is copyright of the author and/or third parties. The intellectual property rights of the author or third parties in respect of this work are as defined by The Copyright Designs and Patents Act 1988 or as modified by any successor legislation. Any use made of information contained in this thesis/dissertation must be in accordance with that legislation and must be properly acknowledged. Further distribution or reproduction in any format is prohibited without the permission of the copyright holder.
SUMMARY Thyroid cancer, the most common endocrine malignancy has a rising incidence worldwide. Radiation damage is a known aetiological factor in thyroid tumourigenesis, particularly in children. Pituitary tumor transforming gene (PTTG) and its binding partner (PTTG binding factor; PBF) are overexpressed in thyroid cancers. Critically, PTTG and PBF have been shown to be independent markers of poor prognosis in thyroid cancer. Both PBF and PTTG have been shown to be tumourigenic in vivo. PTTG, a human securin, has multifunctional roles in mitotic control, DNA repair, apoptosis, cell transformation and genomic instability. PTTG null murine embryogenic fibroblasts demonstrate prolonged G2/M transition and aneuploidy. Overexpression of PTTG in MG63 osteosarcoma, H1299 lung cancer and HeLa cell lines induces aneuploidy. Additionally, PTTG binds and inhibits Ku70, a DNA repair protein involved in double-strand DNA breaks. PBF, which has independent tumourigenic and transforming actions, binds to PTTG, transports it to the nucleus and facilitates its actions within the nucleus. Taken together, the above implicate the functional role of PTTG and PBF in genetic instability. This thesis describes the generation of a transgenic murine model of thyroid cancer which overexpressed both human PTTG and human PBF in the thyroid gland (BI-Trans). The BI-Trans murine model developed goitres from a young age in both genders. Additionally, they develop thyroid adenomas at a later age, which had a female preponderance. Unexpectedly, this BI-Trans murine model did not develop cancers. We measured the index of genetic instability (GI) in the thyroids of our transgenic murine models with fluorescent inter-simple sequence repeat-PCR (FISSR-PCR). This technique was refined to measure GI with small quantities of DNA obtained from murine thyroids. Additionally, we identified target gene for further evaluation through microarray analysis to elucidate the mechanism by which PBF and PTTG induced genetic instability.
DEDICATION
To my parents and sisters for your continual love and support. And my research supervisor, Professor Christopher McCabe for your unwavering support, patience and kindness.
ACKNOWLEDGEMENTS
I would like to express my gratitude to Professor John Watkinson for initiating this research project and the Get Ahead charity for the priming grant, without whose support this work would not have been possible. A special thank you to Professors Christopher McCabe and Jayne Franklyn and Kristien Boelaert whose mentoring and support as research supervisors have made completing this thesis possible. Dr. Martin Read has been a great source inspiration for the countless hours of intellectual discussion. Additionally, Drs. Martin Read, Vicki Smith and Rachel Watkins have been excellent teachers in the multitude of techniques used in experimental molecular biology. Vital collaborations with Drs. Margaret Eggo, Martin Read, Gregory Lewy, Adrian Warfield and Andrea Bacon have made this line of investigation complete. The company of Robert Seed, Neil Sharma, Gavin Ryan and Perkin Kwan has made my time in the laboratory both enjoyable and memorable. Finally, the generous financial support of the Wellcome Trust has made this piece of research a reality.
ABBREVIATIONS ATM
Ataxia telangiectasia mutated
ATR pathway
Ataxia telangiectasia and Rad3-related pathway
BI-Trans
Homozygote for both PBF and PTTG
qRT-PCR
Quantitative real-time polymerase chain reaction
BSA
Bovine serum albumin
CGH
Comparative Genomic Hybridization
DAPI
4',6-diamidino-2-phenylindole
DSB
Double strand break
DTC
Differentiated thyroid cancer
FGF-2
Fibroblast growth factor 2
FISH
Fluorescent in-situ hybridization
FISSR-PCR
Fluorescent inter-simple sequence repeat PCR
FTC
Follicular thyroid cancer
GI index
Genetic instability index
GI
Genetic instability
HBSS
Hank's Balanced Salt Solution
hPBF
Human PTTG binding factor
hPTTG
Human pituitary tumor transforming gene 1
IR-
No radiation treatment
IR +
Ionising radiation treatment
PBF
human PTTG binding factor
PBFHomo-PTTGHet
PBF homozygote – PTTG1 heterozygote
PBFHET-PTTGHET
PBF heterozygote – PTTG1 heterozygote
PBS
Phosphate buffered saline
PBF-Tg
Homozygote murine model of PBF expression in the thyroid
PCR
Polymerase chain reaction
PMTC
Primary murine thyroid culture
PTC
Papillary thyroid cancer
PTTG +/-
Heterozygote for PTTG1
PTTG +/+
Homozygote for PTTG1
PTTG
Human pituitary tumor transforming gene 1
Cdk
Cyclin dependent kinase
MPF
Maturation promoting factor
SDS
Sodium Dodecyl Sulfate
SSB
Single strand break
Trp53
p53
TSH
Thyroid stimulating hormone
MSI
Microsatellite instability
VEGF
Vascular endothelial growth factor
Table of contents
TABLE OF CONTENTS 1
GENERAL INTRODUCTION
1
1.1
THE THYROID GLAND
2
1.1.1
Anatomy and physiology
2
1.1.2
Thyroid hormone biosynthesis
3
1.1.3
Thyroid hormone at the end-organ
4
1.2
THYROID CANCER
6
1.2.1
Classification
6
1.2.2
Epidemiology
7
1.2.3
Risk factors for non-medullary thyroid cancer
8
1.2.4
Management of non-medullary thyroid cancer
12
1.2.5
Prognosis of differentiated thyroid cancer
15
1.2.6
Molecular genetics
15
1.2.7
Other genes involved in thyroid cancer
27
PITUITARY TUMOR TRANSFORMING GENE (PTTG)
30
1.3
1.3.1
Identification
30
1.3.2
Structure and function of PTTG
30
1.3.3
Regulation of PTTG expression
31
1.3.4
Subcellular localisation of PTTG
33
1.3.5
Expression in normal human tissue and cancers
34
1.3.6
Expression in thyroid cancer
34
1.3.7
Murine models of PTTG overexpression
35
1.3.8
PTTG and genetic instability
35
1.3.9
PTTG, Trp53 and apoptosis
36
1.3.10
PTTG, DNA damage and repair
37
1.3.11
Gene interactions with PTTG
38
Table of contents
1.4
PTTG BINDING FACTOR (PBF)
41
1.4.1
Structure, function and localisation of PBF
41
1.4.2
PBF and NIS
42
1.4.3
Expression in human tissue
42
1.4.4
Expression of PBF in thyroid cancer
43
1.4.5
PBF overexpression in murine thyroids
43
1.4.6
PBF and genetic instability
45
1.5
GENETIC INSTABILITY
46
1.5.1
Hypotheses of tumourigenesis
46
1.5.2
Genetic instability and cancer
47
1.5.3
Measuring genetic instability
48
1.6
THE CELL CYCLE
50
1.6.1
CDK and DNA damage
50
1.6.2
CDK and chromosomal instability
51
1.7
ATM/ATR signalling pathway
52
1.8
HYPOTHESIS AND AIMS
55
2
MATERIALS AND METHODS
2.1
MURINE THYROID DISSECTION
58
2.2
PRIMARY MURINE THYROID CULTURE (PMTC)
58
2.3
DNA EXTRACTION
59
2.4
RNA EXTRACTION
60
2.5
REVERSE TRANSCRIPTION
60
57
Table of contents
2.6
QUANTITATIVE REAL-TIME POLYMERASE CHAIN REACTION (qRT-PCR)
61
2.7
PROTEIN EXTRACTION
63
2.8
WESTERN BLOT
63
2.9
IMMUNOFLUORESCENCE
64
STATISTICAL ANALYSIS
64
2.10
3
BI-TRANSGENIC MURINE MODEL OF HUMAN PTTG AND PBF
OVEREXPRESSION IN THE THYROID GLAND
65
3.1
INTRODUCTION
66
3.2
MATERIALS AND METHODS
68
3.2.1
Generation of murine model
68
3.2.2
Ageing colony
71
3.2.3
Tissue DNA extraction
71
3.2.4
Zygosity screening
71
3.2.5
Murine organ dissection
71
3.2.6
Western blot
72
3.2.7
Cardiac puncture
73
3.2.8
Thyroid function test
73
3.2.9
Histology
73
3.3
RESULTS
74
3.3.1
Validation of the murine model
74
3.3.2
Reduced mortality in founder BI-Trans models
75
3.3.3
Mice overexpressing both PBF and PTTG in the thyroid gland have reduced fertility
77
3.3.4
Bi-Trans and PBFHomo-PTTGHet murine models develop goitres
81
3.3.5
Thyroid hormone levels in transgenic murine models
88
Table of contents
3.3.6
3.4
PBF Homo-PTTG Het mice develop adenomas
DISCUSSION
96
100
3.4.1
Creation of bi-transgenic model colony
100
3.4.2
Reduced mortality
101
3.4.3
Fertility issues
101
3.4.4
Goitre formation in the PBFHomo-PTTGHet murine model
103
3.4.5
Thyroid function tests
103
3.4.6
Adenoma formation in PBFHomo-PTTGHet
104
3.5
CONCLUSION
104
4
GENETIC INSTABILITY IN TRANSGENIC MURINE THYROIDS
4.1
INTRODUCTION
106
4.2
MATERIAL AND METHODS
108
105
4.2.1
Primary murine thyroid cultures (PMTC)
108
4.2.2
Protein extraction and Western blot
108
4.2.3
Immunofluorescence
108
4.2.4
Radioiodine uptake assays
108
4.2.5
DNA extraction
109
4.2.6
Polymerase chain reaction (PCR)
110
4.2.7
Agarose gel electrophoresis
110
4.2.8
Fluorescent inter-simple sequence repeat-PCR (FISSR-PCR)
110
4.3
RESULTS
112
4.3.1
Primary murine thyroid culture
112
4.3.2
Immunofluorescence confirmed the presence of thyroid cells
113
4.3.3
PMTCs overexpress PBF and/or PTTG
114
4.3.4
Primary murine thyroid cultures were functional
115
Table of contents
4.3.5
PCR optimisation
116
4.3.6
Modified FISSR-PCR
117
4.3.7
BI-Trans thyroids demonstrated significant genetic instability
119
4.3.8
PBF, PTTG and BI-Trans PMTCs exhibited genetic instability
120
4.3.9
PBF is associated with reduced genetic instability in cultured thyroid cells following exposure to
ionising radiation
121
4.3.10
123
4.4
GI index before and after ionising radiation by genotype
DISCUSSION
124
4.4.1
Primary murine thyroid culture
124
4.4.2
FISSR-PCR
124
4.4.3
Index of genetic instability
125
126
4.5
CONCLUSION
5
THE EFFECT OF GENOTYPE ON DNA DAMAGE AND REPAIR GENES
5.1
INTRODUCTION
5.1.1
5.2
127 128
Background
128
MATERIALS AND METHODS
129
5.2.1
Primary murine thyroid cultures
129
5.2.2
RNA extraction
129
5.2.3
Reverse transcription
130
5.2.4
RT Profiler
5.3
2
TM
PCR Arrays
RESULTS
130
131
5.3.1
Expression of DNA damage signalling genes in PBF PMTCs
131
5.3.2
Expression of DNA damage signalling genes in PTTG PMTCs
133
5.3.3
Expression of DNA damage signalling genes in BI-Trans PMTCs
135
5.3.4
Affect upon apoptotic genes
137
Table of contents
5.3.5
Affect upon genes involved in cell cycle arrest
138
5.3.6
Affect upon cell cycle checkpoint genes
139
5.3.7
Affect upon other genes related to the cell cycle
140
5.3.8
Affect upon genes involved in damaged DNA binding
141
5.3.9
Affect upon genes involved in base excision repair
142
5.3.10
Affect upon genes involved in nucleotide excision repair
143
5.3.11
Affect upon genes involved in double strand break repair
144
5.3.12
Affect upon genes involved in mismatch repair
145
5.3.13
Affect on other genes related to DNA repair
146
5.4
DISCUSSION
148
5.4.1
Murine DNA Damage Signalling Microarray studies
148
5.4.2
The effect of PTTG on DNA damage/DNA repair genes
148
5.4.3
The effect of PBF on DNA damage/DNA repair genes
150
5.4.4
The effect of both PBF and PTTG on DNA damage/DNA repair genes
150
5.5
CONCLUSION
151
6
THE EFFECT OF RADIATION ON GENOTYPE IN DNA DAMAGE AND REPAIR
GENES 6.1
INTRODUCTION
6.1.1
6.2
152 153
Background
153
MATERIALS AND METHODS
154
6.2.1
Primary murine thyroid cultures
154
6.2.2
RNA extraction
154
6.2.3
Reverse transcription
154
6.2.4
RT Profiler
6.3
2
RESULTS
TM
PCR Arrays
154
155
Table of contents
6.3.1
The effect of radiation on WT PMTCs
155
6.3.2
The effect of radiation on PBF PMTCs
157
6.3.3
The effect of radiation on PTTG+/+ IR+ PMTCs
159
6.3.4
The effect of radiation on BI-Trans PMTCs
161
6.3.5
PBF, PTTG and radiation reduced the expression of Brca1 and Mbd4, genes involved with
apoptosis.
163
6.3.6
Radiation and PTTG reduced the expression of Chek1, a gene involved in cell cycle arrest.
165
6.3.7
Genes involved in cell cycle checkpoint remained unchanged following irradiation regardless of
genotype
167
6.3.8
Reduced Chaf1a expression in PBF and PTTG genotypes was further suppressed by radiation
169
6.3.9
Genes involved in damaged DNA binding
171
6.3.10
Genes involved in base excision repair
173
6.3.11
Genes involved in nucleotide excision repair
175
6.3.12
Genes involved in double strand break repair
177
6.3.13
Genes involved in mismatch repair
178
6.3.14
Expression of other genes related to DNA repair
180
6.4
DISCUSSION
184
6.4.1
Study on gene expression changes following radiation damage in PMTCs
184
6.4.2
PBF IR+ PMTCs and DNA damage / DNA repair gene changes
184
6.4.3
PTTG IR+ PMTCs and DNA damage / DNA repair gene changes
185
6.4.4
The effect of both PBF and PTTG on DNA damage and DNA repair genes in PMTCs
185
6.5
CONCLUSION
7
EVALUATION OF SHORTLISTED GENES WITHIN THE DNA DAMAGE / DNA
REPAIR PATHWAY 7.1
INTRODUCTION
7.1.1
Background
186
187 188 188
Table of contents
7.2
MATERIALS AND METHODS
189
7.2.1
Primary murine thyroid cultures
189
7.2.2
RNA extraction and reverse transcription
189
7.2.3
Real-time quantitative polymerase chain reaction (qRT-PCR)
189
7.2.4
Western blotting
189
7.3
RESULTS
7.3.1
191
The presence of γ-H2AX in PMTCs confirmed the presence of DNA damage following ionising
radiation
191
7.3.2
The mRNA expression of Brca1 was reduced by PTTG expression and ionising radiation
193
7.3.3
The mRNA expression of Chek1 was inhibited by PTTG and ionising radiation
195
7.3.4
The mRNA expression of Exo1 was repressed by PTTG and radiation
197
7.3.5
The mRNA expression of Mgmt was increased by radiation
199
7.3.6
The mRNA expression of Rad51 was reduced by PTTG and radiation
201
7.3.7
The mRNA expression of Tdg remained unchanged
203
7.3.8
The mRNA expression of Trp53 was unchanged by PBF, PTTG or radiation
205
7.3.9
Rad 51 expression was not significantly altered on Western blot.
207
7.3.10
Trp53 appeared to be stabilised following radiation
208
7.4
DISCUSSION
210
7.4.1
Validation of cDNA array gene changes
210
7.4.2
Taqman qRT-PCR validated the gene expression pattern on the microarray
210
7.4.3
DNA damage was present in PMTCs following irradiation
211
7.4.4
The relationship between γ-H2AX and Brca1
211
7.4.5
Rad51
212
7.4.6
The relationship between Chek1 and Rad51
212
7.4.7
The relationship between Trp53 and PBF/PTTG
213
7.4.8
Exo1, Tdg and Mgmt
213
Table of contents
7.5
CONCLUSION
213
8
FINAL CONCLUSIONS AND FUTURE STUDIES
8.1
The interaction between PBF and PTTG overexpression in transgenic murine thyroid gland
216
8.2
Use of FISSR-PCR for measuring genetic instability
218
8.3
The expression of genes associated with PBF, PTTG and ionising radiation in the thyroid gland
219
215
Table of figures
TABLE OF FIGURES FIGURE 1-1 NEGATIVE FEEDBACK LOOP OF THYROID HORMONE REGULATION. .................... 3 FIGURE 1-2 THE REACTIONS IN THYROID HORMONE BIOSYNTHESIS CATALYSED BY THYROID PEROXIDISE (TPO). ..................................................................................................... 4 FIGURE 1-3 THE RATE OF THYROID CANCER DETECTED IN MALES AND FEMALES BY YEAR IN THE UK (CANCER RESEARCH UK, 2014). ............................................................................. 7 FIGURE 1-4 NUMBER OF CASES OF THYROID CANCER DIAGNOSED PER YEAR BY AGE AND GENDER. ....................................................................................................................................... 9 FIGURE 1-5 SCHEMATIC DIAGRAM OF THE UPSTREAM INTERACTION BETWEEN PI3K/AKT AND MAPK PATHWAY IN THYROID CANCER, REPRODUCED FROM (XING, 2010). THE PI3K/AKT PATHWAY IS ILLUSTRATED ON THE LEFT AND THE MAPK PATHWAY ON THE RIGHT. THE COMMON POINTS OF INTERACTION INCLUDE THE TYROSINE KINASE RECEPTOR (RTK) AND RAS. PTEN INHIBITS THE PI3K/AKT SIGNALLING PATHWAY. ...... 16 FIGURE 1-6 THE HYPOTHESIZED PATHWAYS INVOLVED IN THE PATHOPHYSIOLOGY OF PTC (PAPILLARY THYROID CANCER), FTC (FOLLICULAR THYROID CANCER) AND ATC (ANAPLASTIC THYROID CANCER). REPRODUCED FROM (XING, 2010). ............................. 17 FIGURE 1-7 PROFILE OF ABNORMAL GENES FOUND IN FOLLICULAR THYROID CANCER, REPRODUCED FROM (BHAIJEE AND NIKIFOROV, 2011). ..................................................... 18 FIGURE 1-8 PROFILE OF ABNORMAL GENES FOUND IN PAPILLARY THYROID CARCINOMA ADAPTED FROM (CANCER GENOME ATLAS RESEARCH, 2014). THE BAR CHARTS IN RED DENOTE FUSIONS AND IN BLUE, MUTATIONS. ...................................................................... 19 FIGURE 1-9 SCHEMATIC OVERVIEW OF MAPK SIGNALLING PATHWAY SHOWN ON THE EXTREME LEFT. VARIOUS KNOWN GENE INTERACTIONS ARE ILLUSTRATED TO THE RIGHT OF THE LINE. REPRODUCED FROM (DHILLON ET AL., 2007). ................................. 20 FIGURE 1-10 PI3K/AKT SIGNALLING PATHWAY ADAPTED FROM (VIVANCO AND SAWYERS, 2002). ACTIVATION OF CLASS IA PHOSPHATIDYLINOSITOL 3-KINASES (PI3KS) OCCURS THROUGH STIMULATION OF RECEPTOR TYROSINE KINASES (RTKS) AND THE CONCOMITANT ASSEMBLY OF RECEPTOR–PI3K COMPLEXES. THESE COMPLEXES LOCALISE AT THE MEMBRANE WHERE RECEPTOR-PI3K CATALYSES THE CONVERSION
Table of figures
OF PTDINS(4,5)P2 (PIP2) TO PTDINS(3,4,5)P3 (PIP3). PIP3 SERVES AS A SECOND MESSENGER THAT HELPS ACTIVATE AKT. THROUGH PHOSPHORYLATION, ACTIVATED AKT MEDIATES THE ACTIVATION AND INHIBITION OF SEVERAL TARGETS (GSK3Β, GLYCOGEN SYNTHASE KINASE-3Β; NF-ΚB, NUCLEAR FACTOR OF ΚB; MDM2; MTOR; FKHR; BAD) RESULTING IN CELLULAR GROWTH, SURVIVAL AND PROLIFERATION THROUGH VARIOUS MECHANISMS. ........................................................................................ 22 FIGURE 1-11 SCHEMATIC REPRESENTATION OF HUMAN PTTG PROTEIN REPRODUCED FROM (SMITH ET AL., 2010). THE REGULATORY N-TERMINAL CONTAINS THE KEN AND DESTRUCTION BOXES. THE SH3 INTERACTING DOMAIN AND PHOSPHORYLATION SITE LIES WITHIN THE FUNCTIONAL C-TERMINUS. ....................................................................... 31 FIGURE 1-12 SCHEMATIC REPRESENTATION OF PBF REPRODUCED FROM (SMITH ET AL., 2010). PBF CONTAINS A BIPARTITE NUCLEAR LOCALISATION SIGNAL SEQUENCE AND PHOSPHORYLATION SITE (RESIDUE Y174) AT ITS C-TERMINUS. THE N-TERMINAL CONTAINS THE SIGNAL SEQUENCE AND GLYCOSYLATION SITES DENOTED BY GLY. .. 42 FIGURE 1-13 AN OVERVIEW OF THE ATM MOLECULAR PATHWAY DEMONSTRATING ITS ROLE IN DOUBLE-STRAND DNA DAMAGE (ADAPTED FROM SABIOSCIENCES.COM). THE DIAGRAM DEPICTS KNOWN GENES THAT INTERACT WITH ATM AND ITS DOWNSTREAM FUNCTIONS. ................................................................................................................................ 53 FIGURE 1-14 ATR SIGNALLING PATHWAY ADAPTED FROM (SHIOTANI AND ZOU, 2009).THE DIAGRAM SHOWS THE ACTIVATION AND FUNCTION OF ATR IN RESPONSE TO SINGLESTRAND BREAKS. ...................................................................................................................... 54 FIGURE 2-1 VIEW OF MURINE THYROID UNDER 10X MAGNIFICATION DURING DISSECTION. THIS VIEW OF THE THYROID GLAND WAS OBTAINED FOLLOWING THE REMOVAL OF FUR AND STRAP MUSCLES IN THE NECK. ............................................................................. 58 FIGURE 2-2 LINEAR PLOTS OF LOGARITHMIC CDNA AGAINST CT VALUES (A) AND Δ CT VALUES (B). ................................................................................................................................. 62 FIGURE 3-1 MATING GENETICS OF PBF AND PTTG HOMOZYGOTES. MATING A PBF HOMOZYGOTE WITH A PTTG HOMOZYGOTE CREATED A COLONY OF PBF HETEROZYGOTE – PTTG HETEROZYGOTE (PBFHET-PTTGHET). ...................................... 69
Table of figures
FIGURE 3-2 MATING GENETICS OF PBFHET-PTTGHET. MATING PBFHET-PTTGHET WITH PBFHET-PTTGHET YIELDED A COLONY OF MIXED GENOTYPES ILLUSTRATED ABOVE. THE OFFSPRING GENOTYPE PATTERN IS CAUSED BY RANDOM PBF / PTTG GENE INSERTION IN DIFFERENT CHROMOSOMES DURING THE CREATION OF THE MURINE MODELS. ...................................................................................................................................... 69 FIGURE 3-3 MATING GENETICS OF BI-TRANS MURINE MODELS. THE MATING OF BI-TRANS MODELS CREATES A COLONY COMPRISING EXCLUSIVELY OF BI-TRANS OFFSPRING. 70 FIGURE 3-4 WESTERN BLOT SHOWING THE EXPRESSION OF HPBF AND HPTTG IN WT, PBF, PTTG+/+ AND BI-TRANS THYROIDS (N=4 FOR EACH GENOTYPE; FIGURE SHOWS N=2 FOR EACH GENOTYPE). ............................................................................................................ 74 FIGURE 3-5 WESTERN BLOT SHOWING THE EXPRESSION OF HPBF AND HPTTG IN PBFHOMO-PTTGHET (*) IN VARIOUS ORGANS (N=1) COMPARED TO WILD TYPE (WT). . 75 FIGURE 3-6 KAPLAN-MEIER SURVIVAL PLOT OF VARIOUS MURINE GENOTYPES ACCORDING TO GENOTYPE. IF 0.01 < WILCOXON P-VALUE ≤ 0.05 (*) AND WILCOXON P-VALUE ≤ 0.0001 (****). N=53 (BI-TRANS), N=89 (PBFHOMO-PTTGHET), N=82 (PBF), N=108 (PTTG+/+), N=129 (PTTG+/- ) AND N=30 (WT). ............................................................................................ 76 FIGURE 3-7 AVERAGE MATING PERIOD BY GENOTYPE PAIRING. THE BREEDING TIME 2
BETWEEN BI-TRANS AND BI-TRANS, (BI-TRANS ) WAS STATISTICALLY SIGNIFICANT (P=0.003; N=2 PAIRS). THE NUMBERS FOR EACH GENOTYPE PAIRING WAS AS 2
2
2
FOLLOWS; N=9 PAIRS (WT ), N=12 PAIRS (PBF ), N=10 PAIRS (PBFHOMO-PTTGHET ) AND N=41 PAIRS (PBFHOMO-PTTGHET X BI-TRANS). ................................................................... 77 2
FIGURE 3-8 AVERAGE NUMBER OF LITTERS BY GENOTYPE PAIRING. THE WT (N=9) AND 2
PBF (N=12) PAIRING PRODUCED SIMILAR NUMBER OF LITTERS OVER THEIR BREEDING 2
PERIOD. THE PBFHOMO-PTTGHET PAIRING (N=23) HAD FEWER NUMBER OF LITTERS 2
COMPARED WT (P=0.0002) BUT MORE THAN THE PBFHOMO-PTTGHET X BI-TRANS PAIRING (N=46; P0.05). ............... 85 FIGURE 3-15 THYROID WEIGHT BY GENDER AND GENOTYPE AT 12 MONTHS. P-VALUES ARE DENOTED WITH ** (0.001 < P ≤ 0.01), **** (P ≤ 0.0001) AND NS (P>0.05). ............................. 86 FIGURE 3-16 THYROID WEIGHTS AT 18 MONTHS BY GENDER AND GENOTYPE. P-VALUES ARE DENOTED WITH *** IF 0.0001 < P ≤ 0.001 AND **** IF P ≤ 0.0001. ................................ 87 FIGURE 3-17 T3 LEVELS IN PBFHOMO-PTTGHET MICE BY GENDER AND AGE. P-VALUES ARE DENOTED WITH *** FOR 0.0001 < P ≤ 0.001 AND **** FOR P ≤ 0.0001. ............................... 88 FIGURE 3-18 T4 LEVELS PBFHOMO-PTTGHET BY AGE AND GENDER. P-VALUES ARE DENOTED WITH ** FOR 0.001 < P ≤ 0.01 AND *** FOR 0.0001 < P ≤ 0.001. ........................ 89 FIGURE 3-19 T3 LEVELS BY GENOTYPE AND GENDER AT 6 WEEKS. P-VALUES ARE DENOTED WITH ** (0.001 < P ≤ 0.01) AND NS (P > 0.05). ..................................................... 90
Table of figures
FIGURE 3-20 T4 LEVELS AT 6 WEEKS BY GENOTYPE AND GENDER. P-VALUES ARE DENOTED BY NS (P>0.05) AND * (0.01>P≤ 0.05). ..................................................................................... 91 FIGURE 3-21 TSH LEVELS AT 6 WEEKS BY GENOTYPE AND GENDER. P>0.05 DENOTED BY NS. ................................................................................................................................................ 92 FIGURE 3-22 T3 LEVELS AT 12 MONTHS BY GENOTYPE AND GENDER. P-VALUES ARE DENOTED WITH * (0.01 < P ≤ 0.05) AND NS (P>0.05). .......................................................... 93 FIGURE 3-23 T4 LEVELS AT 12 MONTHS BY GENOTYPE AND GENDER. P-VALUE >0.05 IS DENOTED BY “NS”. ..................................................................................................................... 94 FIGURE 3-24 TSH LEVELS AT 12 MONTHS BY GENOTYPE AND GENDER. P-VALUES ARE DENOTED WITH * (0.01 < P ≤ 0.05) AND ** (0.001 < P ≤ 0.01). ............................................ 95 FIGURE 3-25 PBFHOMO-PTTGHET THYROID AT 5X MAGNIFICATION. ........................................ 96 FIGURE 3-26 PBFHOMO-PTTGHET THYROID SHOWING HYPERCELLULARITY AT 10X MAGNIFICATION. ........................................................................................................................ 97 FIGURE 3-27 PBFHOMO-PTTGHET DISCRETE THYROID LESION AT 10X MAGNIFICATION...... 97 FIGURE 3-28 LESION WITH HYPERCELLULARITY IN PBFHOMO-PTTGHET THYROID AT 10X MAGNIFICATION. ........................................................................................................................ 98 FIGURE 3-29 GENETICS OF PBFHOMO-PTTGHET X PBFHOMO-PTTGHET BREEDING. ......... 102 FIGURE 3-30 GENETICS OF PBFHOMO-PTTGHET AND BI-TRANS BREEDING. ....................... 102 FIGURE 4-1 PRIMARY MURINE THYROID CULTURE UNDER LIGHT MICROSCOPY DEPICTING WHOLE THYROID FOLLICLES IMMEDIATELY FOLLOWING PROCESSING AT DAY 0 (10X MAGNIFICATION). ..................................................................................................................... 112 FIGURE 4-2 LIGHT MICROSCOPY OF PRIMARY MURINE THYROID CULTURE (10X MAGNIFICATION) SHOWING THE FORMATION OF CLUSTERS IN A 12-WELL PLATE ON DAY 10........................................................................................................................................ 113 FIGURE 4-3 IMMUNOFLUORESCENCE IN PBF-HA PRIMARY MURINE THYROID CULTURE (PMTC) AT 10X MAGNIFICATION. THE NUCLEUS FLUORESCED IN BLUE (DAPI) AND THE HA TAG FLUORESCED IN RED. IMAGE A DEPICTS THE PATTERN OF CELLULAR GROWTH IN PMTCS AND IMAGE B CONFIRMED THE PRESENCE OF PBF-HA THYROID
Table of figures
CELLS. THE COMPOSITE IMAGE C DEPICTS THE GROWTH OF THYROID CELLS IN CLUSTERS. ................................................................................................................................ 114 FIGURE 4-4 WESTERN BLOT OF PMTCS PROBED WITH PTTG (TOP) AND PBF (BOTTOM) ANTIBODIES. WT, PBF, PTTG AND BI-TRANS REFER TO WILD-TYPE, PBF, PTTG AND BITRANS PMTCS (N=4 FOR EACH GENOTYPE). ...................................................................... 114 FIGURE 4-5 RELATIVE THE UPTAKE OF
125
I UPTAKE IN WT AND PBF PRIMARY MURINE THYROID CULTURES.
125
I IS SIGNIFICANTLY REPRESSED (P=0.0004) IN PBF (N=6)
COMPARED TO WT (N=5). ....................................................................................................... 115 FIGURE 4-6 DNA PCR PRODUCTS RESOLVED ON 1% AGAROSE GEL, EXTRACTED FROM WILD TYPE (WT)(N=3) AND PBF (N=3) PRIMARY MURINE THYROID CULTURES. THE PCR CONDITIONS FOR THIS RUN WERE DEEMED OPTIMAL AND USED FOR SUBSEQUENT FISSR EXPERIMENTS. PCR PRODUCTS RANGED BETWEEN 200 AND 1000 BASE PAIRS. .................................................................................................................................................... 116 FIGURE 4-7 AN EXAMPLE OF PCR PRODUCTS RESOLVED ON ABI GENEMAPPER SEQUENCER. ARROWS INDICATE INDIVIDUAL SEQUENCING CAPILLARIES. YELLOW/ORANGE COLOUR REPRESENT THE GENESCAN LIZ SIZE STANDARD (THERMOFISHER SCIENTIFIC). THE BLUE COLOUR ARISES FROM THE FAM FLUOROPHORE ATTACHED TO THE PRIMER USED IN THE POLYMERASE CHAIN REACTION (PCR). ..................................................................................................................... 117 FIGURE 4-8 RESOLVED PCR PRODUCTS IN GENEMAPPER (APPLIED BIOSYSTEMS) REPRESENTED IN NUMERICAL FORM. THE PRODUCT SIZE FOR EACH CAPILLARY RUN IS IDENTIFIED. THE QUANTITY, DEFINED BY HEIGHT AND AREA, IS ALSO PROVIDED AS ILLUSTRATED ABOVE. THE FAM MARKER REFERS TO THE FLUOROPHORE ATTACHED TO THE (CA)8RG PRIMER USED IN THE PCR. ....................................................................... 118 FIGURE 4-9 GRAPHICAL REPRESENTATION OF PCR PRODUCTS RESOLVED BY SEQUENCING ON GENEMAPPER (APPLIED BIOSYSTEMS) SHOWING GROSS DIFFERENCES (ENCIRCLED) BETWEEN WT AND PBF GENOTYPES, USED IN CALCULATING THE INDEX OF GENETIC INSTABILITY. ...................................................................................................... 118
Table of figures
FIGURE 4-10 GI INDEX COMPARISON ACCORDING TO GENOTYPE IN VIVO. PBF (N=5) AND PTTG+/+ (N=5) THYROIDS APPEAR TO HAVE NO STATISTICALLY SIGNIFICANT GENETIC INSTABILITY COMPARED TO WT (N=5). HOWEVER, BI-TRANS (N=4) THYROIDS HAD A GI INDEX WHICH WAS STATISTICALLY SIGNIFICANT COMPARED TO WT (P=0.01, DENOTED WITH *). ...................................................................................................................................... 119 FIGURE 4-11 GI INDEX IN PRIMARY MURINE THYROID CULTURES BY GENOTYPE. ALL 3 GENOTYPES OF PBF (N=4), PTTG+/+ (N=5) AND BI-TRANS (N=5) EXHIBIT GENETIC INSTABILITY COMPARED TO WT. P-VALUES ARE DENOTED WITH ** (0.001 < P ≤ 0.01) AND **** (P ≤ 0.0001). ................................................................................................................ 120 FIGURE 4-12 GENETIC INSTABILITY (GI) INDEX OF WT, PBF, PTTG+/+ AND BI-TRANS FOLLOWING IONISING RADIATION. WT IR+ (N=2), PBF IR+ (N=5), PTTG+/+ IR+ (N=5) AND BI-TRANS IR+ (N=5). P-VALUES ARE DENOTED WITH NS FOR P>0.05, * (0.01 < P ≤ 0.05), ** (0.001 < P ≤ 0.01), *** (0.0001 < P ≤ 0.001) AND **** (P ≤ 0.0001). .................................... 122 FIGURE 4-13 GENETIC INSTABILITY (GI) INDEX WITH AND WITHOUT IONISING RADIATION BY GENOTYPE. PMTCS WERE EXPOSED TO 25 GY OF IONISING RADIATION AND THE GI INDEX MEASURED USING FISSR-PCR. P-VALUES ARE DENOTED WITH ** (0.001 < P ≤ 0.01), *** (0.0001 < P ≤ 0.001) AND **** (P ≤ 0.0001). ............................................................. 123 FIGURE 5-1 DNA DAMAGE GENE MRNA EXPRESSION CHANGES IN PBF PMTC COMPARED 2
WITH WILD TYPE (N=3) ON THE RT PROFILER
TM
PCR ARRAYS THAT IS LESS THAN
HALF. NO GENES ON THE ARRAY WAS EXPRESSED MORE THAN TWICE. STATISTICAL P-VALUES ARE REPRESENTED BY ** IF 0.001 < P ≤ 0.01. IR- REFERS TO WITHOUT IONISING RADIATION. .............................................................................................................. 132 FIGURE 5-2 DNA DAMAGE / DNA REPAIR GENE MRNA EXPRESSION IN PTTG+/+ PMTC. THE GENES THAT WERE EXPRESSED BY MORE THAN TWICE (B) OR LESS THAN HALF (A) COMPARED TO WT IR- PMTC (N=3) WERE CONSIDERED SIGNIFICANT. HOWEVER, STATISTICAL P-VALUES ARE ALSO SHOWN. IF 0.01 < P ≤ 0.05 (*) AND 0.001 < P ≤ 0.01 (**). IR- REFERS TO WITHOUT IONISING RADIATION. ......................................................... 134 FIGURE 5-3 DNA DAMAGE / DNA REPAIR GENE EXPRESSION IN BI-TRANS PMTC WITHOUT IRRADIATION (N=3) COMPARED WITH WILD TYPE WITHOUT IRRADIATION (WT IR-). IF
Table of figures
0.01 < P ≤ 0.05 (*) AND 0.001 < P ≤ 0.01 (**). FOLD CHANGES WERE CONSIDERED SIGNIFICANT IF EXPRESSED BY LESS THAN HALF. NO GENES WERE EXPRESSED BY MORE THAN TWICE. IR- REFERS TO WITHOUT IONISING RADIATION. ............................ 136 FIGURE 5-4 EXPRESSION OF GENES INVOLVED IN APOPTOSIS ACCORDING TO GENOTYPE (N=3). STATISTICAL SIGNIFICANCE IS REPRESENTED BY * IF 0.01 < P ≤ 0.05 AND ** IF 0.001 < P ≤ 0.01.IR- DENOTES WITHOUT IONISING RADIATION. ........................................ 137 FIGURE 5-5 CELL CYCLE ARREST GENE EXPRESSION BY GENOTYPE (N=3). IF 0.01 < P ≤ 0.05 (*), 0.001 < P ≤ 0.01 (**) AND 0.0001 < P ≤ 0.001 (***). IR- DENOTES WITHOUT IONISING RADIATION. .............................................................................................................. 138 FIGURE 5-6 EXPRESSION OF GENES INVOLVED IN CELL CYCLE CHECKPOINT BY GENOTYPE (N=3). THE P-VALUE IS REPRESENTED BY * IF 0.01 < P ≤ 0.05. IR- DENOTES WITHOUT IONISING RADIATION. .............................................................................................................. 139 FIGURE 5-7 EXPRESSION OF FURTHER GENES RELATED TO THE CELL CYCLE (N=3). PVALUES ARE DENOTED BY * IF 0.01 < P ≤ 0.05 (*) AND ** IF 0.001 < P ≤ 0.01 (**). IRDENOTES WITHOUT IONISING RADIATION........................................................................... 140 FIGURE 5-8 DAMAGED DNA BINDING GENE EXPRESSION BY GENOTYPE. P-VALUES ARE DENOTED BY * IF 0.01 < P ≤ 0.05 (*) AND ** IF 0.001 < P ≤ 0.01 (**). IR- DENOTES WITHOUT IONISING RADIATION. ............................................................................................ 141 FIGURE 5-9 BASE EXCISION REPAIR GENE EXPRESSION BY GENOTYPE (N=3). P-VALUES ARE DENOTED BY * IF 0.01 < P ≤ 0.05 (*) AND ** IF 0.001 < P ≤ 0.01 (**). IR- DENOTES WITHOUT IONISING RADIATION. ............................................................................................ 142 FIGURE 5-10 GENES INVOLVED WITH NUCLEOTIDE EXCISION REPAIR BY GENOTYPE. PVALUES ARE DENOTED BY * IF 0.01 < P ≤ 0.05 (*) AND ** IF 0.001 < P ≤ 0.01 (**). IRDENOTES WITHOUT IONISING RADIATION........................................................................... 143 FIGURE 5-11 DOUBLE STRAND BREAK REPAIR GENE EXPRESSION BY GENOTYPE. PVALUES ARE DENOTED BY * IF 0.01 < P ≤ 0.05 (*) AND ** IF 0.001 < P ≤ 0.01 (**). IRDENOTES WITHOUT IONISING RADIATION........................................................................... 144
Table of figures
FIGURE 5-12 THE MRNA EXPRESSION OF GENES INVOLVED IN MISMATCH REPAIR. PVALUES ARE DENOTED BY * IF 0.01 < P ≤ 0.05 (*) AND ** IF 0.001 < P ≤ 0.01 (**). IRDENOTES WITHOUT IONISING RADIATION........................................................................... 145 FIGURE 5-13 EXPRESSION OF OTHER GENES RELATED TO DNA REPAIR BY GENOTYPE (N=3). IF 0.01 < P ≤ 0.05 (*), 0.001 < P ≤ 0.01 (**) AND 0.0001 < P ≤ 0.001 (***). IRDENOTES WITHOUT IONISING RADIATION........................................................................... 147 FIGURE 6-1 GENE EXPRESSION CHANGES IN WT PMTC FOLLOWING IRRADIATION (WT IR+) THAT WERE LESS THAN 0.5 FOLD (A) AND MORE THAN 2 FOLD (B). FOLD CHANGES ARE EXPRESSED RELATIVE TO WT PMTC WITHOUT IRRADIATION (WT IR-). P-VALUES ARE DENOTED WITH * (0.01 < P ≤ 0.05) AND ** (0.001 < P ≤ 0.01). .......................................... 156 FIGURE 6-2 DNA DAMAGE / DNA REPAIR GENE EXPRESSION CHANGES IN PBF PMTC FOLLOWING EXPOSURE TO RADIATION. ALL VALUES ARE RELATIVE TO WT IR-. PVALUES ARE DENOTED WITH * (0.01 < P ≤ 0.05), ** (0.001 < P ≤ 0.01) AND *** (0.0001 < P ≤ 0.001). THERE WERE NO GENES ON THE RT2 PROFILER™ PCR ARRAY THAT WERE EXPRESSED MORE THAN TWICE IN PBF IR+ PMTCS. ........................................................ 158 FIGURE 6-3 DNA DAMAGE / DNA REPAIR GENE EXPRESSION CHANGES IN PTTG+/+ IR+ PMTC LESS THAN 0.5 (A) AND MORE THAN 2 (B) FOLD. P-VALUES ARE DENOTED WITH * (0.01 < P ≤ 0.05) AND ** (0.001 < P ≤ 0.01). .......................................................................... 160 FIGURE 6-4 DNA DAMAGE AND DNA REPAIR GENE EXPRESSION CHANGES IN BI-TRANS IR+ PMTCS OF LESS THAN 0.5 (A) AND MORE THAN 2 (B) FOLD. FOLD EXPRESSION IS RELATIVE TO WT IR-. P-VALUES ARE DENOTED WITH * (0.01 < P ≤ 0.05), ** (0.001 < P ≤ 0.01) AND *** (0.0001 < P ≤ 0.001). .......................................................................................... 162 FIGURE 6-5 EXPRESSION OF GENES INVOLVED IN APOPTOSIS BY GENOTYPE. TOP FIGURE (A) DENOTES PMTC GENOTYPE WITH IRRADIATION (IR+) AND BOTTOM FIGURE (B) DENOTES PMTC GENOTYPE WITHOUT IRRADIATION (IR-). FOLD EXPRESSION IS RELATIVE TO WT PMTC WITHOUT IRRADIATION (WT IR-). P-VALUES ARE DENOTED WITH * (0.01 < P ≤ 0.05) AND ** (0.001 < P ≤ 0.01). ............................................................. 164 FIGURE 6-6 CELL CYCLE ARREST GENE EXPRESSION BY GENOTYPE. TOP FIGURE (A) DENOTES PMTC GENOTYPE WITH IRRADIATION (IR+) AND BOTTOM FIGURE (B)
Table of figures
DENOTES PMTC GENOTYPE WITHOUT IRRADIATION (IR-). FOLD EXPRESSION IS RELATIVE TO WT PMTC WITHOUT IRRADIATION (WT IR-). P-VALUES ARE DENOTED WITH * (0.01 < P ≤ 0.05), ** (0.001 < P ≤ 0.01), *** (0.0001 < P ≤ 0.001) AND **** (P≤0.0001). .................................................................................................................................................... 166 FIGURE 6-7 EXPRESSION OF CELL CYCLE CHECKPOINT GENES BY GENOTYPE. CHART A DENOTES PMTC GENOTYPE WITH IRRADIATION (IR+) AND CHART B DENOTES PMTC GENOTYPE WITHOUT IRRADIATION (IR-). FOLD EXPRESSION IS RELATIVE TO WT PMTC WITHOUT IRRADIATION (WT IR-). P-VALUES ARE DENOTED WITH * IF 0.01 < P ≤ 0.05. .................................................................................................................................................... 168 FIGURE 6-8 OTHER GENES RELATED TO CELL CYCLE FOLLOWING DNA DAMAGE, GENE EXPRESSION BY GENOTYPE. TOP FIGURE (A) DENOTES PMTC GENOTYPE WITH IRRADIATION (IR+) AND BOTTOM FIGURE (B) DENOTES PMTC GENOTYPE WITHOUT IRRADIATION (IR-). FOLD EXPRESSION IS RELATIVE TO WT PMTC WITHOUT IRRADIATION (WT IR-). P-VALUES ARE DENOTED WITH * (0.01 < P ≤ 0.05) AND ** (0.001 < P ≤ 0.01). ................................................................................................................................. 170 FIGURE 6-9 EXPRESSION OF GENES INVOLVED IN DNA BINDING BY GENOTYPE. THE TOP CHART (A) DENOTES PMTC GENOTYPE WITH IRRADIATION (IR+) AND BOTTOM CHART (B) DENOTES PMTC GENOTYPE WITHOUT IRRADIATION (IR-). FOLD EXPRESSION IS RELATIVE TO WT PMTC WITHOUT IRRADIATION (WT IR-). P-VALUES ARE DENOTED WITH * (0.01 < P ≤ 0.05), ** (0.001 < P ≤ 0.01) AND *** (0.0001 < P ≤ 0.001). ................... 172 FIGURE 6-10 EXPRESSION OF GENES INVOLVED IN BASE EXCISION REPAIR BY GENOTYPE. TOP FIGURE DENOTES PMTC GENOTYPE WITH IRRADIATION (IR+) AND BOTTOM FIGURE DENOTES PMTC GENOTYPE WITHOUT IRRADIATION (IR-). FOLD EXPRESSION IS RELATIVE TO WT PMTC WITHOUT IRRADIATION (WT IR-). P-VALUES ARE DENOTED WITH * (0.01 < P ≤ 0.05) AND ** (0.001 < P ≤ 0.01). ............................................................. 174 FIGURE 6-11 EXPRESSION OF GENES INVOLVED IN NUCLEOTIDE EXCISION REPAIR BY GENOTYPE. TOP CHART (A) DENOTES PMTC GENOTYPE WITH IRRADIATION (IR+) AND BOTTOM CHART (B) DENOTES PMTC GENOTYPE WITHOUT IRRADIATION (IR-). FOLD
Table of figures
EXPRESSION IS RELATIVE TO WT PMTC WITHOUT IRRADIATION (WT IR-). P-VALUES ARE DENOTED WITH * (0.01 < P ≤ 0.05) AND ** (0.001 < P ≤ 0.01). ................................. 176 FIGURE 6-12 EXPRESSION OF GENES INVOLVED IN DOUBLE BREAK STRAND REPAIR BY GENOTYPE. TOP FIGURE (A) DENOTES PMTC GENOTYPE WITH IRRADIATION (IR+) AND BOTTOM FIGURE (B) DENOTES PMTC GENOTYPE WITHOUT IRRADIATION (IR-). FOLD EXPRESSION IS RELATIVE TO WT PMTC WITHOUT IRRADIATION (WT IR-). P-VALUES ARE DENOTED WITH * (0.01 < P ≤ 0.05) AND ** (0.001 < P ≤ 0.01). ................................. 177 FIGURE 6-13 EXPRESSION OF GENES INVOLVED IN MISMATCH REPAIR BY GENOTYPE. TOP CHART (A) DENOTES PMTC GENOTYPE WITH IRRADIATION (IR+) AND BOTTOM CHART (B) DENOTES PMTC GENOTYPE WITHOUT IRRADIATION (IR-). FOLD EXPRESSION IS RELATIVE TO WT PMTC WITHOUT IRRADIATION (WT IR-). P-VALUES ARE DENOTED WITH * (0.01 < P ≤ 0.05), ** (0.001 < P ≤ 0.01), AND *** (0.0001 < P ≤ 0.001). .................. 179 FIGURE 6-14 EXPRESSION OF GENES RELATED TO DNA REPAIR BY GENOTYPE. TOP FIGURE (A) DENOTES PMTC GENOTYPE WITH IRRADIATION (IR+) AND BOTTOM FIGURE (B) DENOTES PMTC GENOTYPE WITHOUT IRRADIATION (IR-). FOLD EXPRESSION IS RELATIVE TO WT PMTC WITHOUT IRRADIATION (WT IR-). P-VALUES ARE DENOTED WITH * (0.01 < P ≤ 0.05), ** (0.001 < P ≤ 0.01) AND *** (0.0001 < P ≤ 0.001). ................... 183 FIGURE 7-1 WESTERN BLOT AND DENSITOMETRY STUDY OF Γ-H2AX IN PMTCS BY GENOTYPE WITH AND WITHOUT IRRADIATION.WT IR- (N=6), WT IR+ (N=7), PBF IR- (N=7), PBF IR+ (N=8), PTTG+/+ IR-(N=7), PTTG+/+ IR+ (N=7), BI-TRANS IR- (N=8) AND BI-TRANS IR+ (N=8). P-VALUES ARE DENOTED WITH * (0.01 < P ≤ 0.05). ........................................ 192 FIGURE 7-2 BRCA1 MRNA EXPRESSION IN PMTCS BY GENOTYPE BEFORE AND AFTER IRRADIATION ON THE MICROARRAY (A) AND TAQMAN QRT-PCR (B). N=4 PMTCS ALL GENOTYPES. P-VALUES OBTAINED BY COMPARING WITH WT IR- ARE DENOTED WITH * (0.01 < P ≤ 0.05), ** (0.001 < P ≤ 0.01) AND *** (0.0001 < P ≤ 0.001). ................................ 194 FIGURE 7-3 CHEK1 MRNA EXPRESSION IN PMTCS BEFORE AND AFTER IRRADIATION ON THE MICROARRAY (A) AND TAQMAN QRT-PCR (B). N=4 FOR PMTCS ALL GENOTYPES. PVALUES SHOWN, COMPARED TO WT IR-, ARE DENOTED WITH * (0.01 < P ≤ 0.05), ** (0.001 < P ≤ 0.01), *** (0.0001 < P ≤ 0.001) AND **** (P ≤ 0.0001). ........................................ 196
Table of figures
FIGURE 7-4 EXO1 MRNA EXPRESSION IN PMTCS BEFORE AND AFTER IRRADIATION ON THE MICROARRAY (A) AND TAQMAN QRT-PCR (B). N=4 FOR PMTCS ALL GENOTYPES. PVALUES OBTAINED BY COMPARING TO WT IR- ARE DENOTED WITH * (0.01 < P ≤ 0.05), ** (0.001 < P ≤ 0.01) AND *** (0.0001 < P ≤ 0.001). ................................................................. 198 FIGURE 7-5 MGMT MRNA EXPRESSION IN PMTCS BEFORE AND AFTER IRRADIATION ON THE MICROARRAY (A) AND TAQMAN QRT-PCR (B). N=4 FOR PMTCS ALL GENOTYPES. PVALUES OBTAINED BY COMPARING TO WT IR- ARE DENOTED WITH * (0.01 < P ≤ 0.05) AND ** (0.001 < P ≤ 0.01). ........................................................................................................ 200 FIGURE 7-6 RAD51 MRNA EXPRESSION IN PMTCS BEFORE AND AFTER IRRADIATION ON THE MICROARRAY (A) AND TAQMAN QRT-PCR (B). N=4 FOR PMTCS ALL GENOTYPES. PVALUES OBTAINED BY COMPARING TO WT IR- ARE DENOTED WITH * (0.01 < P ≤ 0.05) AND *** (0.0001 < P ≤ 0.001). ................................................................................................... 202 FIGURE 7-7 TDG MRNA EXPRESSION IN PMTCS BEFORE AND AFTER IRRADIATION ON THE MICROARRAY (A) AND TAQMAN QRT-PCR (B). N=4 FOR PMTCS ALL GENOTYPES. PVALUES OBTAINED BY COMPARING TO WT IR- ARE NOT DENOTED BECAUSE ALL VALUES WERE STATISTICALLY INSIGNIFICANT IE. P>0.05. ............................................... 204 FIGURE 7-8 TRP53 MRNA EXPRESSION IN PMTCS BEFORE AND AFTER IRRADIATION ON THE MICROARRAY (A) AND TAQMAN QRT-PCR (B). N=4 FOR PMTCS ALL GENOTYPES. PVALUES OBTAINED BY COMPARING TO WT IR- ARE NOT DENOTED BECAUSE ALL VALUES WERE STATISTICALLY INSIGNIFICANT IE. P>0.05. ............................................... 206 FIGURE 7-9 THE EXPRESSION OF RAD51 ON WESTERN BLOT BY GENOTYPE, WITH AND WITHOUT IRRADIATION IN 4 SEPARATE EXPERIMENTS. ................................................... 207 FIGURE 7-10 WESTERN BLOT AND DENSITOMETRY STUDY OF TRP53 EXPRESSION BY GENOTYPE, BEFORE AND AFTER IRRADIATION (N=3). P-VALUES OBTAINED BY COMPARING TO WT IR- ARE DENOTED WITH * (0.01 < P ≤ 0.05) AND *** (0.0001 < P ≤ 0.001). ......................................................................................................................................... 209 FIGURE 8-1 THYROID CHANGES IN THE VARIOUS MURINE MODELS AND POTENTIAL FOR FUTURE STUDIES. .................................................................................................................... 217 FIGURE 8-2 DIAGRAM OF KEY FINDINGS. .................................................................................... 219
1 General Introduction
General Introduction
1.1
THE THYROID GLAND
1.1.1 Anatomy and physiology The thyroid gland, located in the neck is the largest endocrine organ. Embryologically, the thyroid gland originates in the foramen caecum at the junction of the anterior and posterior tongue and descends to its adult position by 7 weeks in the embryo (Kay and Goldsmith, 2010). The thyroid gland measures approximately 10 -30 g in an adult and increases in size during pregnancy (Dorion and Lemaire, 2008). Histologically, the thyroid gland consists of numerous follicles, each of which is comprised of thyroid epithelial cells surrounding a globule of colloid (Krause, 2005). The main function of the thyroid gland is the production of thyroid hormone which is essential for energy metabolism, growth and maturation of tissue. The secretion of thyroid hormone from the thyroid gland is under the influence of thyroid stimulating hormone (TSH) which is produced in the anterior pituitary gland and secreted into the bloodstream. Circulating levels of TSH is in turn controlled by thyrotropin-releasing hormone (TRH) which is produced in the hypothalamus and transported via the superior hypophyseal artery to the anterior pituitary gland. The secretion of TRH and TSH is negatively regulated by high levels of circulating thyroid hormone (Yen, 2001). See Figure 1-1. At a molecular level, TSH binds to the TSH receptor on the membrane of thyroid epithelial cells which leads to an increase in the stimulation of several thyroid genes including the sodium-iodide symporter (NIS), thyroglobulin and thyroid peroxidise which promote the biosynthesis of thyroid hormone (Yen, 2001).
2
General Introduction
Figure 1-1 Negative feedback loop of thyroid hormone regulation.
1.1.2 Thyroid hormone biosynthesis Iodine, an essential component of thyroid hormone biosynthesis, is concentrated from the bloodstream by the thyroid gland. The ability of thyroid epithelial cells to concentrate iodine lies within the sodium-iodide symporter (NIS) which sits on the basolateral membrane of the thyroid cell (Dohan et al., 2003). NIS is coupled to the Na-ATPase pump which generates the energy required to concentrate iodine against its natural gradient. At the apical part of the thyroid epithelial cell, the transporter pendrin drives iodine from the cytoplasm to the colloid. Iodine subsequently binds to the tyrosine residue of thyroglobulin, a protein found within the colloid, in a reaction catalysed by the enzyme
3
General Introduction
thyroid peroxidise (TPO). This reaction produces mono-iodothyrosine (MIT) or di-iodothyrosine (DIT). The combination of MIT with DIT results in the active triiodothyronine (T 3) whilst DIT with DIT results in thyroxine (T4). See Figure 1.2. The products of the reaction catalysed by TPO, MIT, DIT, T3 and T4 are stored in the colloid of the thyroid follicle. During the secretion of thyroid hormone, MIT, DIT, T3 and T4 are internalised into the thyroid epithelial cell, undergoes proteolytic digestion to recapture MIT and DIT with the release of T3 and T4 into the bloodstream (Yen, 2001).
Figure 1-2 The reactions in thyroid hormone biosynthesis catalysed by thyroid peroxidise (TPO).
1.1.3 Thyroid hormone at the end-organ Thyroid hormone in the bloodstream can exist in a free active state or bound to protein such as thyroxine binding globulin (TBG) or albumin. It is the free, unbound T3 and T4 that enters target cells and generates a biological response. Overall, the ratio of T3 to T4 in the bloodstream is approximately
4
General Introduction
1:20. The effect of T3 is about four times more potent at the end-organ compared to T4. End-organs convert T4 to the more active T3 by the enzyme iodothyronine deiodinase and is actively transported across the cell membrane by OATP 1C1, monocarboxylate transporter (MCT) 8 and MCT10 (Visser et al., 2011). T3 has a short half-life of 2.5 days compared to the half-life of T4 which is about 6.5 days.
5
General Introduction
1.2
THYROID CANCER
1.2.1 Classification The vast majority of thyroid cancers are of the well-differentiated variety which comprises mainly papillary carcinoma (80 %) and follicular carcinoma (10 - 15 %). These well-differentiated tumour types arise from thyroid follicular cells. Poorly differentiated thyroid carcinomas such as anaplastic (ATC) tumours are rare (1 - 3 %) (Nagaiah et al., 2011) and are thought to arise de-novo or be a dedifferentiation progression of well-differentiated thyroid carcinoma (Smallridge et al., 2009). Hurtle cell carcinoma, characterized by the presence of oncocytes rich in mitochondria, is considered a subtype of follicular carcinoma of the thyroid gland. Medullary carcinoma (5%) of the thyroid, a completely separate aetiological entity, arises from the parafollicular C-cells of the thyroid gland. Parafollicular C-cells of the thyroid are involved in the secretion of calcitonin, a hormone related to calcium homeostasis. Calcitonin reduces calcium levels by reducing resorption of calcium in the kidneys, inhibiting the activity of osteoclasts, stimulating osteoblastic activity and inhibits calcium resorption in the intestines. Another type of thyroid cancer, primary lymhoma of the thyroid gland, is not common and is associated with Hashimoto’s disease of the thyroid (Kato et al., 1985). The molecular mechanism by which primary lymphoma of the thyroid gland develops in Hashimoto’s disease of the thyroid is not understood. Hashimoto’s disease is an autoimmune condition causing chronic lymphocytic thyroiditis leading to an underactive thyroid gland. Patients with this condition may have high autoantibody titres to thyroid peroxidise, thyroglobulin or TSH receptor. For the purpose of this thesis, DTC refers to the common papillary and follicular carcinomas of the thyroid.
6
General Introduction
1.2.2 Epidemiology Thyroid cancer has an incidence of 3.23 cases per 100,000 per year in the the UK. This incidence rate is increasing as shown in Figure 1-3 (Cancer Research UK, 2014). The increasing rate of thyroid cancer, documented over 30 years across 5 continents, showed an average increase of 67% in females and 48% in males (Peterson et al., 2012). This increase has been attributed in part to the early detection of small tumours (“incidentalomas”) on neck scanning for other non-thyroid related conditions (Leenhardt et al., 2004, American Thyroid Association Guidelines Taskforce on Thyroid et al., 2009). The increasing trend in the detection of large advanced thyroid tumours (≥4cm; distant spread) does not suggest that the increasing rate of thyroid cancer is solely related to incidentolomas and suggests a true increase in the incidence of thyroid cancer (Chen et al., 2009, Sipos and Mazzaferri, 2010). The main type of thyroid cancer which is on the increase is papillary carcinoma followed by a small increase in follicular carcinoma (Reynolds et al., 2005, Smailyte et al., 2006, Mitchell et al., 2007).
Figure 1-3 The rate of thyroid cancer detected in males and females by year in the UK (Cancer Research UK, 2014).
7
General Introduction
1.2.3 Risk factors for non-medullary thyroid cancer 1.2.3.1 Female gender There is a predilection for females to develop thyroid cancer compared to males, seen in Figure 1-4. The ratio of thyroid cancer between male to female is approximately 1:2.5 (Cancer Research UK, 2014). The epidemiological data suggest a link between oestrogen and thyroid cancer. This has been supported by further molecular studies demonstrating the stimulatory effect of oestrogen on thyroid cell proliferation (Santin and Furlanetto, 2011). However, the mortality rates in males is significantly higher compared to females and is largely due to late diagnosis and more advanced disease at the time of initial diagnosis (Sipos and Mazzaferri, 2010).
1.2.3.2 Age The detection of thyroid cancer peaks in the 4th decade for females and the 6th decade for males in the UK as shown in Figure 1-4 (Cancer Research UK, 2014). The sharp increase in the incidence of thyroid cancer in females from puberty to the end of the 4th decade closely mirrors the fertility period of a female and circulating oestrogen levels.
8
General Introduction
Figure 1-4 Number of cases of thyroid cancer diagnosed per year by age and gender.
1.2.3.3 Ionising radiation Previous exposure to radiation is a well recognised risk factor for the pathogenesis of thyroid cancer. This association was reported in 1950 where 10 out of 28 children with thyroid cancer had received radiation to their thymus gland between the fourth and sixteenth months of life (Duffy and Fitzgerald, 1950). Kaplan et al recently described the historic account between thyroid cancer and medical radiation (Kaplan et al., 2009). The most convincing evidence for thyroid cancer and radiation came following the 1945 nuclear explosion in Hiroshima and Nagasaki. The long term report from the atomic bomb survivors described the risk of thyroid cancer being greatest before the age of 20 years. The magnitude of excess risk decreased with increasing age or time since exposure. The majority of thyroid cancers from the cohort were mainly papillary carcinomas (Furukawa et al., 2013). The Chernobyl nuclear fallout in 1986 exposed a large population in Russia, Ukraine and Belarus to radiation. It has been estimated that individuals most exposed, including recovery workers, each averaged an effective dose equivalent to 50 years of typical background radiation. This led to a sharp
9
General Introduction
increase in well differentiated thyroid carcinoma (Stsjazhko et al., 1995). Post-Chernobyl thyroid cancers were characterised by a predominance of papillary carcinomas (93 – 98 %). Aggressive tumour histological characteristics, lymphatic invasion, intrathyroid infiltration and multifocality, were described to be dose-dependent (Zablotska et al., 2015). Known and novel gene rearrangements are thought to be the oncogenic drive for radiation-induced thyroid cancer (Ricarte-Filho et al., 2013). Also, early thyroid cancers that are radiation induced are more likely to have gene rearrangements compared to thyroid tumours that occur after a long latency (9 – 12 years) (Nikiforov, 2006).
1.2.3.4 Family history of thyroid cancer The cumulative lifetime risk of developing non-medullary thyroid cancer in females and males is 3 fold and 1.5 fold respectively compared to the normal population if they had a first-degree relative with thyroid cancer (Fallah et al., 2013).
1.2.3.5 Other proposed risk factors A higher thyroid stimulating hormone (TSH) level, even within the normal range, is associated with thyroid cancer (Boelaert et al., 2006, Haymart et al., 2008, Jonklaas et al., 2008, Polyzos et al., 2008, Boelaert, 2009, Haymart et al., 2009). The thyroid hormone receptor is dependent on transcription factors that regulate apoptosis, cell proliferation and differentiation (Cheng, 2000). The receptor, consisting of four binding components (TRα1, TRα2, TRβ1 and TRβ2) is encoded by two genes, THRA (chromosome 3) and THRB (chromosome 17). Mutation of TRβ1, the PV mutation, results in the inability of the receptor to bind the thyroid hormone T3, causing thyroid hormone resistance syndrome in humans (Kamiya et al., 2003, Kaneshige et al., 2000, Suzuki et al., 2003). The transgenic murine model harbouring this mutation, known as the TRbetaPV/PV mouse, has high levels of thyroid
10
General Introduction
stimulating hormone and develop follicular thyroid carcinoma (Suzuki et al., 2002) . This murine model supports the observation that TSH has a role in thyroid tumourigenesis. The association between iodine intake and thyroid cancer in humans is unclear (Blomberg et al., 2012). However, populations with comparative higher iodine intake have a higher proportion of papillary compared to follicular carcinomas of the thyroid. Conversely, there is a higher incidence of follicular carcinoma in populations with low iodine intake (Lind et al., 1998, Feldt-Rasmussen, 2001). Weight and obesity have been explored by numerous groups as a risk factor for thyroid cancer. Systematic reviews did not however prove definitive causation (Peterson et al., 2012, Zhao et al., 2012). Female reproductive factors such as pregnancy, number of children, use of prescription hormones, menstrual cycle patterns and menopause were not associated with thyroid cancer in a systematic review (Peterson et al., 2012).
1.2.3.6 Association with other hereditary disorders Gardner syndrome, a subtype of familial adenomatous polyposis (FAP), is associated with a high risk of colon cancer and papillary thyroid cancer. Both conditions are caused by defects in the APC gene and is autosomal dominant. Cowden disease, associated with hamartomas, have an increased risk of thyroid, uterine and breast cancer. This syndrome is commonly caused by defects in the PTEN gene. The Carney complex (Type 1) caused by defects in the PRKAR1A gene have an increased risk of papillary and follicular thyroid cancers. Familial non-medullary thyroid carcinoma, with suspect defective genes on chromosome 19 and 1 is associated with well-differentiated thyroid cancer. Thus, in rare instances, thyroid cancer is associated with other hereditary disorders.
11
General Introduction
1.2.4 Management of non-medullary thyroid cancer 1.2.4.1 Investigations The most common presentation of thyroid cancer to a thyroid specialist is the presence of a lump in the thyroid. The investigations for the assessment of the neck lump include neck ultrasonography (USS) and fine needle aspiration cytology (FNAC). Features suggestive of cancer on ultrasonography include microcalcification, taller rather than wide shape, irregular margins and absence of elasticity of the thyroid lump (Remonti et al., 2015). FNAC is now commonly performed under ultrasound guidance to accurately sample the area in question. The FNAC result is reported using the 5 main categories of the “Thy” classification as shown in Table 1 below. Thyroid cytology reporting can be subjective and thus the decision for further management is dependent on the overall estimated risk of cancer by the clinician based on thyroid USS, FNAC and clinical history. Patient choice is also taken into consideration when the decision is made as to whether further management include repeat USS and FNAC or diagnostic surgery. Presently, studies are underway using biomarkers to improve the diagnostic accuracy of FNAC to aid the clinician’s decision-making process.
12
General Introduction
DIAGNOSTIC CATEGORY
RISK OF MALIGNANCY (%)
Non-diagnostic for cytological diagnosis
0 - 10
Thy 1 / Thy 1c Benign
0 -3
Thy 2 Neoplasm possible / Atypia
5 - 15
Thy 3a Neoplasm possible / Follicular neoplasm
15 - 30
Thy 3f Suspicious for malignancy
60 - 75
Thy 4 Malignant
97 - 100
Thy 5
Table 1 The Thy classification of thyroid cytology and the associated risk of malignancy (reproduced from the Royal Society of Pathologists, 2009).
13
General Introduction
1.2.4.2 Staging Differentiated thyroid cancer is staged according to the Tumour, Node and Metastasis (TNM) classification shown in Table 2 below. STAGE
DESCRIPTION
Tx
Primary tumour cannot be assessed
T0
No evidence of primary tumour
T1a
Tumour ≤ 1 cm in greatest dimension, limited to the thyroid
T1b
Tumour > 1 cm and ≤ 2 cm in greatest dimension, limited to the thyroid
T2
Tumour > 2 cm and ≤ 4 cm in greatest dimension, limited to the thyroid
T3
Tumour > 4 cm in greatest dimension, limited to the thyroid or any tumour with minimal extrathyroid extension
T4a
Tumour of any size extending beyond the thyroid capsule and invades any of the following: subcutaneous soft tissues, larynx, trachea, oesophagus, recurrent laryngeal nerve
T4b
Tumour invades prevertebral fascia, mediastinal vessels, or encases the carotid artery
Nx
Regional nodes cannot be assessed
N0
No regional lymph node metastases
N1
Regional lymph node metastasis
N1a
Metastasis in Level VI (pretracheal and paratracheal, including prelaryngeal and Delphian lymph nodes)
N1b
Metastasis in other unilateral, bilateral or contralateral cervical or upper/superior mediastinal lymph nodes
cM0
Clinically no distant metastasis
cM1
Distant metastasis clinically
Table 2 TNM staging for differentiated thyroid cancer (reproduced from AJCC Cancer Staging Manual, 7th edition.
14
General Introduction
1.2.4.3 Treatment The most widely offered treatment options for differentiated thyroid cancer include surgery, radioiodine and TSH suppression with thyroxine. External beam radiotherapy and chemotherapy treatment modalities have largely been reserved for the palliation of thyroid cancer (Perros et al., 2014). The chemotherapy drug Paclitaxel has been used in the treatment of undifferentiated anaplastic thyroid carcinoma (Gomez Saez et al., 2015). Futhermore, targeted therapy with the use of tyrosine kinase inhibitors such as vemurafenib, selumetinib, dabrafenib, sorafenib, sunitinib, pazopanib, cabozantinib, motesanib, axitinib and vandetanib are currently being evaluated for use in patients with advanced or recurrent thyroid cancer refractory to conventional treatments (Ferrari et al., 2015).
1.2.5 Prognosis of differentiated thyroid cancer The prognosis of well managed well differentiated thyroid cancer is excellent, with the majority of patients diagnosed at an early stage of the disease and having a 5 year survival rate in excess of 95 % (Cancer Research UK 2014). Interestingly, mortality rates from thyroid cancer have remained largely unchanged despite the improved detection (Chen et al., 2009). The worst prognosis arises from anaplastic thyroid tumours. The mortality from anaplastic thyroid tumours contributes 14 – 50 % of the overall mortality rates of thyroid cancer despite being 1- 2 % of all thyroid cancers. The median survival of anaplastic tumours is between 3 to 5 months (Nagaiah et al., 2011).
1.2.6 Molecular genetics The main oncogenic pathways investigated and implicated in the aetiology of differentiated thyroid cancer are the mitogen-activated protein kinase (MAPK) and phosphatidylinositide 3-kinase/Akt pathways (PI3K/Akt) which are frequently dysregulated in thyroid cancer. The MAPK and PI3K
15
General Introduction
pathways have upstream common points such as Ras and tyrosine kinase receptor (RTK) illustrated in Figure 1-5 below.
Figure 1-5 Schematic diagram of the upstream interaction between PI3K/Akt and MAPK pathway in thyroid cancer, reproduced from (Xing, 2010). The PI3K/Akt pathway is illustrated on the left and the MAPK pathway on the right. The common points of interaction include the tyrosine kinase receptor (RTK) and Ras. PTEN inhibits the PI3K/Akt signalling pathway.
The most common form of differentiated thyroid cancer, papillary thyroid cancer, is postulated to be largely attributed to dysfunctional MAPK signalling. The PI3K/Akt pathway appears to be mainly involved in the formation of follicular thyroid cancers and adenomas. However, both pathways can act
16
General Introduction
synergistically in the transformation or dedifferentiation of papillary carcinoma to anaplastic carcinoma, as shown in Figure 1-6 (Xing, 2010).
Figure 1-6 The hypothesized pathways involved in the pathophysiology of PTC (papillary thyroid cancer), FTC (follicular thyroid cancer) and ATC (anaplastic thyroid cancer). Reproduced from (Xing, 2010).
The molecular profile of oncogenes causing differentiated thyroid cancer has implications in tumour classification as well as prognosis. For example, currently classified papillary thyroid cancers driven by BRAFV600E or RAS oncogenes have different appearances on histology and response to mitogenactivated protein kinase (MEK) inhibitors (Cancer Genome Atlas Research, 2014, Ho et al., 2013). In PTC, the most frequently encountered gene aberrations include point mutations of BRAF and Ras, and the RET/PTC rearrangement (discussed individually in subsequent sections). The profile of abnormal genes found in FTC is different, with the PAX8/PPARγ featuring in 30 % of follicular carcinomas but not being apparent in papillary thyroid carcinoma (Bhaijee and Nikiforov, 2011). The constituent
17
General Introduction
mutation proportions found in follicular thyroid cancer is shown in Figure 1-7 and papillary thyroid cancer, Figure 1-8.
Follicular thyroid carcinoma
NONE 30%
RAS 40%
PAX8/PPARγ 30%
Figure 1-7 Profile of abnormal genes found in follicular thyroid cancer, reproduced from (Bhaijee and Nikiforov, 2011).
18
Gene aberration
General Introduction
THADA FGFR2 MET LTK ALK NTRK3 NTRK1 PPARG RET BRAF
1.50% 0.50% 0.30% 0.30% 0.80% 1.30% 1% 1% 6.30% 2.30%
NF1 APC MLL3 SPOP CDH4 MEN1 EZH1 TSHR RB1 ATM ZFHX3 TG PTEN BDP1 MLL ARID1B TRP53 CHEK2 KRAS PPM1D EIF1AX HRAS NRAS BRAF
0.50% 0.50% 1% 0.20% 0.20% 0.20% 0.50% 0.50% 0.50% 1.20% 1.70% 2.70% 0.50% 1.20% 1.70% 1.20% 0.70% 1.20% 1% 1.20% 1.50% 3.50%
0.00%
8.50% 59.70% 10.00%
20.00%
30.00%
40.00%
50.00%
60.00%
70.00%
Percentage
Figure 1-8 Profile of abnormal genes found in papillary thyroid carcinoma adapted from (Cancer Genome Atlas Research, 2014). The bar charts in red denote fusions and in blue, mutations.
19
General Introduction
1.2.6.1 Mitogen-activated Protein Kinase Pathway The mitogen-activated protein kinase (MAPK) pathway regulates cellular functions of proliferation, differentiation, migration, growth and apoptosis. MAPKs are highly conserved proteins which belong to the serine/threonine family of protein kinases. Conventional MAPK signalling follows a three-tiered kinase cascade of MAPKKK → MAPKK → MAPK. MAPKKK is activated via small GTPases, belonging to the Ras/Rho family. MAPKKK phosphorylates and activates MAPKK which in turn phosphorylates and activates MAPK. Phosphorylated MAPK exerts its effect directly on the nucleus which translates to cellular action(s) (Figure 1-9). Dysregulation of this pathway is critical in the development and progression of thyroid cancer. (Pearson et al., 2001, Dhillon et al., 2007).
Figure 1-9 Schematic overview of MAPK signalling pathway shown on the extreme left. Various known gene interactions are illustrated to the right of the line. Reproduced from (Dhillon et al., 2007).
20
General Introduction
1.2.6.2 Phosphatidylinositide 3-kinase/Akt Pathway The lipid kinase phosphoinositide 3-kinase/Akt (PI3K/Akt) pathway is involved in cell growth and metabolism. The central gene Akt is activated by the PI3K complex which in itself interacts with the upstream tyrosine kinase receptor. Activated Akt phosphorylates several downstream targets which lead to cellular actions such as apoptosis, proliferation, DNA repair, growth and survival (Monsalves et al., 2014, Vivanco and Sawyers, 2002).
Figure 1-10 describes the PI3K/Akt pathway.
Dysregulation of this pathway has been implicated in the development of follicular adenomas and thyroid carcinoma as described above in section 1.2.6.
21
General Introduction
Figure 1-10 PI3K/AKT signalling pathway adapted from (Vivanco and Sawyers, 2002). Activation of class IA phosphatidylinositol 3-kinases (PI3Ks) occurs through stimulation of receptor tyrosine kinases (RTKs) and the concomitant assembly of receptor–PI3K complexes. These complexes localise at the membrane where receptor-PI3K catalyses the conversion of PtdIns(4,5)P2 (PIP2) to PtdIns(3,4,5)P3 (PIP3). PIP3 serves as a second messenger that helps activate AKT. Through phosphorylation, activated AKT mediates the activation and inhibition of several targets (GSK3β, glycogen synthase kinase-3β; NF-κB, nuclear factor of κB; MDM2; mTOR; FKHR; BAD) resulting in cellular growth, survival and proliferation through various mechanisms.
22
General Introduction
1.2.6.3 Point mutations of genes involved in differentiated thyroid cancer 1.2.6.3.1 BRAF The RAF family of protein kinases is encoded by the BRAF gene and has 3 isoforms; ARAF, BRAF and CRAF. Although all three isoforms play a role in the RAS-RAF cascade, the main activator of the MAPK pathway is BRAF. The most prevalent thyroid oncogene found in papillary thyroid carcinoma, BRAF, is activated by the V600E substitution (BRAFV600E) (Krause, 2005). In this mutation the valine (V) amino acid is replaced by glutamic acid (E) at codon 600 and this gene remains active, independent of upstream cues (Davies et al., 2002). Consequently there is excessive downstream signalling of the MAPK pathway, resulting in tumour formation (Hilger et al., 2002, Peyssonnaux and Eychene, 2001). Papillary thyroid carcinomas with the BRAF V600E mutation are more aggressive tumours as demonstrated by extrathyroidal invasion, lymph node metastases and advanced tumour stage at initial surgery (Xing et al., 2005, Lee et al., 2007). In a multicentre study involving 16 centres and 2099 patients, the hazards ratio of developing recurrence in BRAFV600E positive compared to BRAFV600E negative thyroid cancers is 1.82, after adjusting for clinicopathologic factors (Xing et al., 2015). Murine models of BRAFV600E thyroid knock-in develop goitres and invasive papillary thyroid carcinoma which later transition to poorly differentiated carcinoma (Knauf et al., 2005). Thyroid cancer initiation via the oncogene BRAFV600E is dependent on TSH signalling (Franco et al., 2011).
1.2.6.3.2 Ras The Ras gene encodes for 3 isoforms of the protein; H-, K- and N-Ras. The Ras gene belongs to a family of GTP binding proteins that are located in the plasma membrane and has a central role in the transduction of signals from tyrosine kinase and G-protein-coupled receptors to effector genes in the
23
General Introduction
MAPK and PI3K/Atk pathways (Dhillon et al., 2007, Xing, 2010). Ras possesses intrinsic GTPase activity and in normal cells is mostly in an inactive GDP-bound state. In point mutations of Ras, the gene is active by either having an increased affinity for GTP or has losing its intrinsic GTPase activity (Howell et al., 2013). Consequently, the MAPK or the PI3K/Atk pathways are aberrantly activated, resulting in tumourigenesis. Mutation of the Ras gene is interesting because it is found in both papillary and follicular thyroid cancers. This probably reflects the ability of mutated Ras to activate both MAPK and PI3K/Akt signalling pathways. Numerous Ras point mutations have been identified but the most common Ras point mutation in follicular neoplasms of the thyroid involves N-Ras at codon 61 (Vasko et al., 2003, Liu et al., 2004). Ras mutations are also found in follicular adenomas of the thyroid in up to 48 % of cases. This observation suggests that Ras mutations may not be a key oncogene in the malignant transformation of thyroid cells to cancer and that additional mutation or epigenetic insult is required for tumourigenesis.
1.2.6.4 Tyrosine kinase receptor rearrangements 1.2.6.4.1 RET/PTC rearrangement The RET proto-oncogene, implicated in 15 % of papillary thyroid carcinomas, is located on the long arm of chromosome 10 (10q11.2) and encodes a tyrosine kinase receptor (Bhaijee and Nikiforov, 2011, Ishizaka et al., 1989). The RET receptor has extracellular, transmembrane and intracellular domains (Ishizaka et al., 1989, Takahashi et al., 1993). Whilst RET is essential in the development of the urogenital and nervous system, its expression is at very low levels in normal thyroid tissue and is not required in the development of the thyroid gland (Pachnis et al., 1993, Schuchardt et al., 1994, Takaya et al., 1996).
24
General Introduction
The specific type of RET mutation dictates the phenotype. For example, point mutations in the RET gene produce medullary thyroid cancer, a tumour arising from the parafollicular c-cells of the thyroid. Medullary thyroid cancers with the methionine to threonine substitution at position 918 of the RET gene are most aggressive and the RET mutation with the replacement of serine by alanine at position 891 produces disease less likely to metastasise (Arighi et al., 2005). Papillary thyroid cancers are associated with the RET/PTC translocation. To date, 13 different oncogenic RET/PTC fusions have been identified. In RET-associated differentiated thyroid cancer, RET/PTC1 accounts for 60 %, RET/PTC3 for 30 % and RET/PTC2 10 % of cases. The other RET/PTC translocations are very rare in papillary carcinoma of the thyroid (Prescott and Zeiger, 2015). In support of the oncogenic nature of RET/PTC fusions, thyroid-specific expression of RET/PTC1 and of RET/PTC3 in transgenic mice results in the development of papillary thyroid cancer (Jhiang et al., 1996, Powell et al., 1998). However, not all RET/PTC fusions result in papillary thyroid carcinoma. RET/PTC fusions are also expressed in benign thyroid adenomas and Hashimotos thyroiditis (Ishizaka et al., 1991, Rhoden et al., 2006). Thyroid cancers arising from radiation exposure in children in the Chernobyl disaster have a high prevalence of RET/PTC rearrangements which results in the activation of RET kinase, causing papillary carcinomas of the thyroid (Bounacer et al., 1997, Nikiforov et al., 1997, Fugazzola et al., 1995, Mizuno et al., 2000). The gene loci of RET and PTC lie in close proximity to one another within the thyroid nucleus during interphase and are predisposed to recombination after DNA damage from ionising radiation (Nikiforova et al., 2000, Roccato et al., 2005). The finding that 20 % of papillary thyroid carcinomas have the RET/PTC rearrangement in the general non-irradiated population and up to 72 % of papillary thyroid cancers post Chernobyl harboured the RET/PTC fusion suggested that exposure to radiation increases the risk of RET/PTC rearrangement (Unger et al., 2004). The exact molecular pathway from activation of the RET tyrosine kinase receptor to papillary carcinoma tumourigenesis is yet to be fully defined.
25
General Introduction
1.2.6.4.2 PAX8/PPARγ rearrangement The PAX8 gene belongs to a family of genes that encode for transcription factors, crucial in the development of the thyroid gland and the production of thyroid hormone. The peroxisome proliferatoractivated receptor gamma (PPARγ)
encodes for a Type 2 nuclear receptor. Translocational
rearrangement of these genes occur between chromosomes 2 and 3 (Wang et al., 2007). The PAX8/PPARγ fusion results in an increased expression of the chimeric protein which inhibits the tumour suppressor activity of PPAR and deregulates PAX8 function (Kroll et al., 2000, Tallini, 2002, Gregory Powell et al., 2004). The PAX8/PPARγ rearrangement is found in 30 to 35 % of follicular thyroid carcinomas (French et al., 2003, Nikiforova et al., 2003). 2 – 13 % of follicular adenomas and 1 – 5 % of follicular variant of papillary carcinomas have this rearrangement (Marques et al., 2002, Nikiforova et al., 2002, Dwight et al., 2003). Follicular carcinomas with the PAX8/PPARγ translocation affect younger patients, are smaller in size and more likely to exhibit vascular invasion (French et al., 2003, Nikiforova et al., 2003). The detection of the PAX8/PPARγ fusion in fine needle aspiration cytology for distinguishing between benign and malignant follicular lesions has not been particularly helpful (Ferraz et al., 2012, Pauzar et al., 2012).
1.2.6.4.3 TRK rearrangement The neurotrophic tyrosine kinase receptor Type 1 (NTRK1) gene encodes a protein that is found on the cell surface and has a role in activating other genes or itself through phosphorylation. NTRK1 is located on the long arm of chromosome 1 (1q21 – q22). Rearrangement of the NTRK1 gene with various activating genes such as TPM3, TPR and TFG in the thyroid results in constitutive activation of tyrosine kinase activity leading to the transformation of thyroid cells. The NTRK1 rearrangements, also known as TRK, commonly found in papillary thyroid cancers are TRK (NTRK1/TPM3 fusion) (Wilton et al., 1995), TRK-T1, TRK-T2 (both
26
General Introduction
NTRK1/different portions of TPR gene fusion) (Greco et al., 1992, Miranda et al., 1994, Greco et al., 1997) and TRK-T3 (NTRK1/TFG fusion) (Greco et al., 1995). The frequency of TRK rearrangements in papillary thyroid cancers is approximately 3 % (Cancer Genome Atlas Research, 2014). The frequency of TRK rearrangements in papillary thyroid carcinoma in patients previously exposed to radiation was the same as patients who were not exposed to ionising radiation (Rabes et al., 2000). The prognostic significance of the TRK rearrangement in papillary thyroid carcinomas remains unclear, with the numbers involved in various studies being small (Bongarzone et al., 1998, Brzezianska et al., 2006, Musholt et al., 2000). Transgenic murine models expressing TRK-T1 in the thyroid gland develop follicular hyperplasia and papillary carcinoma from 7 months onwards. This provides in vivo evidence that TRK-T1 supports oncogenesis in the thyroid gland. This oncogenic effect on thyroid cells is likely to be downstream in nature because of the incomplete penetrance observed within the cohort with an abnormal phenotype (Russell et al., 2000).
1.2.7 Other genes involved in thyroid cancer 1.2.7.1 Trp53 The Trp53 gene, guardian of the genome, regulates multiple vital cellular functions such as growth, proliferation, apoptosis, cell cycle progression and DNA repair. Mutated Trp53, and hence dysfunctional Trp53, is found in a variety of human cancers. In the Li-Fraumeni syndrome where there is a germline autosomal dominant mutation of Trp53, patients are at high risk of developing sarcomas, breast cancer, glioblastomas, adrenocortical carcinoma and haematopoietic malignancies at a young age (Li et al., 1988, Malkin et al., 1990). Trp53 mutations are uncommon in differentiated thyroid cancer but found in as many as 83 % of undifferentiated thyroid carcinoma. It is thought that p53 initiates the progression of well
27
General Introduction
differentiated thyroid cancer to anaplastic cancer (Ito et al., 1992, Dobashi et al., 1993, Donghi et al., 1993, Fagin et al., 1993, Dobashi et al., 1994).
Murine models of papillary thyroid cancer,
overexpressing RET/PTC1 in the thyroid gland, developed anaplastic tumours of the thyroid gland when crossed with Trp53 null mice suggesting that the lack of functional Trp53 promotes dedifferentiation of papillary thyroid cancer (La Perle et al., 2000).
1.2.7.2 PTEN PTEN is a tumour suppressor that preferentially dephosphorylates phosphoinositide substrates and negatively regulates the PI3K/Akt pathway. In human germline mutations of PTEN, also known as Cowden disease, there is an increased risk of thyroid cancer. In a study of 146 patients with Cowden disease, 74 different mutations were observed. Approximately 70 % of patients with Cowden disease have benign lesions of the thyroid and 17 % develop thyroid cancer. The thyroid cancers observed were papillary carcinomas (59 %), follicular carcinomas (27.5 %), hybrid between follicular and papillary carcinoma (4.5 %), medullary carcinoma (4.5 %) and oxyphil cell tumour (4.5 %) (Bubien et al., 2013). The loss of PTEN alone in the thyroid gland in murine models induces goitre and follicular adenomas in the presence of normal TSH levels. None of the models developed thyroid cancer by the age of 11 months, strongly suggesting that the persistent activation of the PI3K/Akt pathway alone is not sufficient for malignant transformation within the thyroid gland (Yeager et al., 2007).
1.2.7.3 Mutations to the thyroid hormone receptor The murine model harbouring the PV mutation which results in an abnormality of the TRβ1 component of the thyroid hormone receptor develop follicular thyroid carcinoma with vascular invasion and metastases (Suzuki et al., 2002).
28
General Introduction
Interestingly, two further novel genes have been discovered to be associated with human thyroid cancer. Both Pituitary Tumor Transforming Gene (PTTG) and PTTG binding factor (PBF) are dysregulated in thyroid cancer (Boelaert et al., 2003, Heaney et al., 2001, Saez et al., 2006, Hsueh et al., 2013, Stratford et al., 2005). The next section describes these two genes, PTTG and PBF, in further detail.
29
General Introduction
1.3
PITUITARY TUMOR TRANSFORMING GENE (PTTG)
1.3.1 Identification Pituitary tumor transforming gene (PTTG) was first identified and found to be exclusively highly expressed in rat pituitary tumour cells but not in normal pituitary cells. Subsequently, it was shown that PTTG had transforming abilities where overexpression of PTTG in murine 3T3 fibroblasts injected in athymic nude mice resulted in tumourigenesis (Pei and Melmed, 1997). Human PTTG shares 89 % sequence homology with rat PTTG and lies on chromosome 5q33 (Zhang et al., 1999). Although two further PTTG homologues have been identified (PTTG2 on chromosome 4p12 (Kakar and Jennes, 1999, Prezant et al., 1999) and PTTG3 on chromosome 8q13 (Chen et al., 2000)), PTTG (PTTG1) is most widely expressed in tumours. Hence, it is the most extensively studied and for the purpose of this thesis, PTTG refers to PTTG1.
1.3.2 Structure and function of PTTG PTTG, shown schematically in Figure 1-11, has an open reading frame containing 609 bp which encodes for 203 amino acids (Zhang et al., 1999). The structure of PTTG can be divided to the basic N-terminal regulatory domain and acidic C-terminal functional domain (Dominguez et al., 1998, Kakar and Jennes, 1999, Zhang et al., 1999). The regulatory N-terminal domain contain the KEN box (amino acid 9 to 11) and destruction box (amino acid 61 to 68) which are targets for the anaphase promoting complex (APC). APC functions by binding to PTTG, resulting in the degradation of PTTG and consequently allowing the progression from metaphase to anaphase during mitosis. (Zou et al., 1999, Zur and Brandeis, 2001). PTTG has transcriptional abilities as shown by the presence of a DNA binding domain, from amino acid 61 to 118, and a transactivation domain further downstream, from amino acids 119 to 164 within its structure (Zhang et al., 1999, Pei, 2000, Wang and Melmed, 2000). Although the subcellular
30
General Introduction
localisation of PTTG, either cytoplasmic or nuclear, is dependent on cell type (Mu et al., 2003), in order for this gene to be functional, it needs to be within the nucleus (Zou et al., 1999). Interestingly, PTTG does not contain a nuclear localisation signal sequence, essential for the ability of a gene to cross the nuclear membrane. This led to the discovery of PTTG binding factor (PBF) which binds to PTTG and translocate it to the nucleus (Chien and Pei, 2000). The transactivational and transforming capabilities of PTTG is attributed to the proline-rich region in the C-terminal (amino acids 163 to 173). This region contains the highly conserved and putative Srchomology-3 (SH3) interacting domain (Zhang et al., 1999). The only reported phosphorylation site of PTTG, thought to play a role during mitosis, lies at serine residue 165 within the SH3 interacting domain (Pei, 2000, Ramos-Morales et al., 2000, Boelaert et al., 2004).
Figure 1-11 Schematic representation of human PTTG protein reproduced from (Smith et al., 2010). The regulatory N-terminal contains the KEN and destruction boxes. The SH3 interacting domain and phosphorylation site lies within the functional C-terminus.
1.3.3 Regulation of PTTG expression The first regulator of PTTG to be identified was the hormone oestrogen. Oestrogen was found to upregulate PTTG expression in pituitary lactotroph tumours in rats (Heaney et al., 1999). Thyroid
31
General Introduction
stimulating hormone was also found to induce PTTG expression (Heaney et al., 2001). Insulin has been demonstrated to upregulate PTTG expresssion but the effect of insulin differed between malignant and non-malignant astrocytes suggesting that the response of PTTG to insulin is variable depending on tissue type (Chamaon et al., 2005, Thompson and Kakar, 2005). The basal levels of PTTG transcription is regulated by the transcription factor SP1 (Clem et al., 2003). Oct-1 expression activates the PTTG promoter and increases the expression of PTTG (Zhou et al., 2008). β-catenin was found to increase PTTG expression in oesophageal cancers (Zhou et al., 2005) and colorectal cancers (Hlubek et al., 2006). γ-catenin which exists in conjunction with β-catenin has been shown to increase the transcription of PTTG (Pan et al., 2007). The binding of NF-Y to Trp53 is thought to inhibit the transcription of PTTG (Zhou et al., 2003). The relationship between PTTG expression and growth factors has been extensively studied. The epidermal growth factor (EGF) receptor, itself regulated by EGF and transforming growth factor alpha (TGF-α), has been demonstrated to upregulate PTTG expression. The promalignant hepatocyte growth factor (HGF) increases the expression of PTTG via the c-Met membrane receptor in astrocytic cells (Tfelt-Hansen et al., 2004). Insulin-like growth factor (IGF-1) has also been shown to enhance transcription of PTTG (Chamaon et al., 2005, Thompson and Kakar, 2005). The relationship between the induction of PTTG by fibroblast growth factor-2 (FGF-2) is well established (Chamaon et al., 2010, Heaney et al., 1999, Tsai et al., 2005). A small number of studies have investigated the epigenetic mechanisms involved in the control of PTTG expression since no mutations of PTTG and its promoter sequence were found in pituitary neoplasms (Zhang et al., 1999, Kanakis et al., 2003). The overexpression of histone acetyltransferase (HAT) p300, a transcription co-activator, increased PTTG promoter activity, ultimately resulting in increased PTTG mRNA and protein. It was also found that increased PTTG promoter activity with p300 was mediated by NF-YA and NF-YB. (Li et al., 2009).
32
General Introduction
1.3.4 Subcellular localisation of PTTG The expression of PTTG is cell cycle dependent and increases from low in S-phase, higher in G2phase and peaks at M-phase in HeLa cells (Zou et al., 1999, Ramos-Morales et al., 2000, Yu et al., 2000b). PTTG was largely localised to the nucleus during interphase, with small amounts in the cytoplasm in JEG-3 cells overexpressing PTTG (Yu et al., 2000b). In contrast, earlier studies have indicated that PTTG was located primarily in the cytoplasm with only partial nuclear expression in Jurkat T lymphoma cells, pituitary adenomas, lung and breast adenocarcinomas (Dominguez et al., 1998, Saez et al., 1999). It is possible that the difference in localisation of PTTG is cell-specific. The function of PTTG requires its presence in the nucleus but it has no nuclear localisation signal within its structure. It is thought that PTTG being a small protein of low molecular weight, has the ability to diffuse freely across the nuclear membrane. Alternatively, PTTG requires a transport protein to facilitate its entry into the nucleus and the discovery of PTTG binding factor (PBF) has supported this hypothesis (Chien and Pei, 2000). One study observed PTTG expression in the Golgi apparatus and vesicles. Analysis of the culture medium detected the presence of PTTG. It was thus suggested that PTTG is a secretory protein that may have an autocrine and/or paracrine function in murine pituitary tumour cells and human pituitary adenomas (Minematsu et al., 2007). PTTG has been observed to co-localise with mitotic spindles during mitosis when its concentration within the cell was highest (Yu et al., 2003). These observations are consistent with the role of PTTG as the human securin involved in the regulation of mitosis. Furthermore, the observation that PTTG phosphorylates during mitosis suggests that PTTG is involved in the regulatory pathway controlling cell proliferation (Ramos-Morales et al., 2000).
33
General Introduction
1.3.5 Expression in normal human tissue and cancers PTTG is expressed at low levels within the spleen, prostate, ovary, heart, brain, liver, skeletal muscle, kidney and pancreas. High levels of PTTG expression are found in the testes, thymus and placenta (Dominguez et al., 1998). It is thought that PTTG plays a role in spermatogenesis (Pei, 1999). High PTTG expression has been reported in various human cancers including astrocytomas (TfeltHansen et al., 2004), breast (Puri et al., 2001), oesophagus (Shibata et al., 2002), lung (Kakar and Malik, 2006), stomach (Wen et al., 2004), liver (Cho-Rok et al., 2006), colon (Heaney et al., 2000), ovary (Puri et al., 2001) and haematopoietic (Dominguez et al., 1998). Interestingly, promoter mutation was not found to play an important role in PTTG overexpression in pituitary adenomas (Kanakis et al., 2003). Further investigations indicate that neither methylation nor loss of heterozygosity are involved in PTTG overexpression, suggesting that the misregulation of PTTG is a post-transcriptional event (Hidalgo et al., 2008).
1.3.6 Expression in thyroid cancer Of interest to this study is the overexpression of PTTG in thyroid neoplasms (Heaney et al., 2001, Boelaert et al., 2003). Heaney et al initially reported increased PTTG mRNA expression in follicular thyroid neoplasms. Our own group has confirmed that PTTG expression is increased at both an mRNA and protein level in well differentiated thyroid cancers with no difference between papillary or follicular lesions. In addition, thyroids from multinodular goitres or Graves disease did not show a significant increase in PTTG. More importantly, high PTTG expression was an independent prognostic indicator of early recurrence (Boelaert et al., 2003). Saez et al analysed PTTG expression in differentiated thyroid cancers using immunohistochemistry and found a positive association of high PTTG expression with nodal and distant metastases, disease persistence, advanced TNM stage and reduced radioiodine uptake (Saez et al., 2006).
34
General Introduction
Reduced iodine uptake was observed by Heaney et al in FRTL5 cells overexpressing PTTG (Heaney et al., 2001). Subsequently, PTTG has been shown to repress the sodium-iodide symporter (NIS) responsible for active radioiodine transport into thyroid cells via FGF-2 (Boelaert et al., 2007), discussed in section 1.3.11.4. This observation may account for a poorer prognosis in patients with thyroid cancer overexpressing PTTG.
1.3.7 Murine models of PTTG overexpression To date, there has been a few studies investigating the effect of PTTG overexpression in specific organs. In vivo overexpression of PTTG in murine pituitary cells induced abnormal cell proliferation and adenomas (Abbud et al., 2005). The overexpression of PTTG in the ovarian surface epithelium of transgenic murine models resulted in the formation of pre-cancerous lesions but did not result in ovarian tumourigenesis (El-Naggar et al., 2007). Our group has created a transgenic murine model of targeted PTTG overexpression in the thyroid gland demonstrating small thyroids with reduced cellular proliferation and did not develop thyroid cancer (Lewy et al., 2013).
1.3.8 PTTG and genetic instability PTTG, also known as the human securin, inhibits the separation of sister chromatids (Zou et al., 1999). The ubiquitination of PTTG by APC/CCdc20 is inhibited by Cdk1 which phosphorylates PTTG (Holt et al., 2008). The activation of separase at the metaphase-anaphase transition triggers the release of Cdc14 phosphatase which dephosphorylates PTTG, resulting in its ubiquitination, ultimately allowing the separation of sister chromatids (Queralt et al., 2006, Stegmeier et al., 2002). Defects in chromatid separation result in aneuploidy (Nasmyth, 2002). Hence, it is not surprising that overexpression of PTTG results in aneuploidy, a histological observation associated with
35
General Introduction
chromosomal instability (Winnepenninckx et al., 2006, Yu et al., 2000a, Yu et al., 2003, Wang et al., 2001). The measurement of intrachromosomal instability by fluoresecent intersimple sequence repeat PCR in thyroid cancer demonstrated a positive correlation between PTTG expression and genetic instability (Kim et al., 2005). In keeping with this, PTTG knockout (PTTG -/-) murine embryo fibroblasts had a prolonged G2/M phase and exhibited aneuploidy (Wang et al., 2001). PTTG -/- knockout murine models are surprisingly viable and fertile, with testicular and splenic hypoplasia, thymic hyperplasia and thrombocytopenia (Wang et al., 2001). In addition, PTTG deficient mice developed non autoimmune insulin deficiency with pleiotropic beta islet cells (Wang et al., 2003). This was attributed to a combination of apoptosis and senescence in beta islet cells secondary to DNA damage (Chesnokova et al., 2009). Taken together, the observed PTTG -/- mouse phenotype could be explained by the varying importance of PTTG in different organ development. Also, it appears PTTG plays an important role in cell cycle progression, albeit not being the only mechanism that regulates sister chromatid separation (Mei et al., 2001, Wang et al., 2001).
Overall, murine models of both PTTG knockout and
overexpression demonstrate that PTTG has a role in genomic stability.
1.3.9 PTTG, Trp53 and apoptosis The functions of Trp53 in apoptosis, cell cycle regulation and DNA repair (Levine, 1997, Vousden, 2000) in response to DNA damage alludes to its importance in the maintenance of genomic stability. The importance of Trp53 in tumourigenesis is demonstrated by the fact that it is mutated in over 50 % of human cancers (Hollstein et al., 1994). The loss of function of or inability to illicit a normal Trp53 response to DNA damage can cause malignant transformation (Vousden, 2006). PTTG has been demonstrated to cause apoptosis both in a Trp53 dependent and independent manner. The same group found that Trp53 was capable of preventing aneuploidy caused by an overexpression
36
General Introduction
of PTTG (Yu et al., 2000a, Yu et al., 2000b). These observations suggests that Trp53 exerts a protective effect via apoptosis to ameliorate the chromosomal instability induced by PTTG. PTTG upregulates the transcription of Trp53 through the expression of c-myc. Bax, a pro-apoptotic downstream gene of the Trp53 signalling pathway is upregulated in the presence of PTTG overexpression only when normal functioning Trp53 is present (Hamid and Kakar, 2004). This suggests that PTTG-induction of Trp53-dependent apoptosis is at least partially mediated via Bax. In vivo and in vitro studies have determined that Trp53 and PTTG interact directly (Bernal et al., 2002). In contrast to the above findings by other groups, Bernal et al reported that the transcriptional ability of Trp53 was inhibited by its interaction with PTTG. In addition, they reported that PTTG inhibited the ability of Trp53 to induce apoptosis (Bernal et al., 2002). Other groups have since also reported that PTTG negatively regulates the apoptotic function of Trp53 (Cho-Rok et al., 2006, Lai et al., 2007). The DNA damaging drugs doxorubicin and bleomycin suppressed PTTG only in the presence of functional Trp53 with one study proposing that Trp53 suppressed PTTG via the transcription factor NF-Y (Zhou et al., 2003). In the study of apoptosis induced by 5-fluorouracil in HCT116 cells, Trp53 was found to bind directly with the PTTG promoter and repress its expression (Kho et al., 2004). PTTG is involved in the DNA damage pathway. In eukaryotic cells, ultraviolet (UV) radiation has been found to reduce PTTG protein levels by reducing its synthesis and increasing its degradation, independent of Trp53. PTTG was also found to be required for cell cycle arrest following UV radiation, with its loss resulting in premature entry into mitosis and increased apoptosis (Romero et al., 2004).
1.3.10 PTTG, DNA damage and repair Further evidence for the role of PTTG in the DNA damage response comes from the interaction between PTTG and Ku70 (Romero et al., 2001). Ku70 forms a heterodimer with Ku80 and together
37
General Introduction
with DNA-PK, this complex facilitates non-homologous DNA end-joining repair of double strand breaks (Smith and Jackson, 1999, Kharbanda et al., 1998). In the presence of double strand breaks, PTTG is phosphorylated by DNA-PK and dissociates from Ku70 allowing for DNA repair to occur (Romero et al., 2001). It has been observed that PTTG overexpression inhibits Ku70 DNA binding and represses double strand break repair (Kim et al., 2007b). Thus, in the presence of high PTTG expression Ku proteins are bound, consequently adversely affecting DNA repair which results in genetic instability. Overall, the above studies with Trp53 and Ku70 have demonstrated the multifaceted role PTTG plays in the DNA damage response which is complex and not fully understood. Certainly, the role of PTTG in the context of genetic instability and tumourigenesis in the thyroid remains to be elucidated.
1.3.11 Gene interactions with PTTG The observation that PTTG has a domain for DNA binding, acidic transactivation and partial localisation in the nucleus, suggests PTTG may transactivate other genes. Deletion of amino acids 124 to 202 abolished the transactivational abilities of PTTG, suggesting that the C-terminal end of PTTG is responsible for this function (Dominguez et al., 1998). The acidic C-terminus of PTTG contains multiple glutamic and proline residues which are characteristic of transactivation domains (Pei, 2000, Wang and Melmed, 2000). Within this region are two PXXP motifs that are SH3-binding sites which are highly conserved regions that govern transduction of intracellular signalling pathways. Point mutations in the PXXP motifs resulted in the abrogation of its transactivational ability and more importantly, in its in vitro transforming and in vivo tumour-inducing activity (Zhang et al., 1999). Furthermore, a point mutation at a key prolene residue disrupted the transactivation and transforming ability of PTTG, suggesting an intrinsic link between its transactivation and transforming functions (Wang and Melmed, 2000).
38
General Introduction
1.3.11.1 FGF2, VEGF and PTTG PTTG has been shown to interact with growth factors and cytokines that could promote the progression of thyroid neoplasms. The established relationship between PTTG, fibroblast growth factor 2 (FGF-2) and vascular endothelial growth factor (VEGF) appear to be critical in tumour progression through the promotion of growth and angiogenesis. Angiogenesis is an integral part of tumour growth and survival; a blood supply is critical in maintaining tumour viability (Folkman, 1972, Folkman, 1990, Folkman, 1992, Folkman and Shing, 1992, Hanahan and Folkman, 1996). FGF-2 and VEGF act synergistically to modulate tumour angiogenesis and invasion (Bikfalvi et al., 1997, Ferrara and Davis-Smyth, 1997, Gospodarowicz et al., 1987, Fujii et al., 2006, Minematsu et al., 2007, Wang et al., 2004). Stable overexpression of PTTG in NIH3T3 cells resulted in increased FGF-2 mRNA expression (Zhang et al., 1999). Interestingly, FGF-2 has been shown to induce PTTG expression in NIH3T3 cells (Heaney et al., 1999), suggesting the existence of a regulatory autocrine pathway between FGF-2 and PTTG. The expression of VEGF by PTTG is FGF-2 independent (McCabe et al., 2002).
1.3.11.2 c-Myc and PTTG The transcription factor c-Myc was found to bind to PTTG. c-Myc is involved with the regulation of cellular proliferation and is dysregulated in tumours. PTTG was found to bind to the c-Myc promoter with the upstream stimulatory factor (USF1) (Pei, 2001). Furthermore, the modulation of p53 function by PTTG is mediated through the expression of c-Myc (Hamid and Kakar, 2004).
39
General Introduction
1.3.11.3 SP1 and PTTG The global transcriptional effect of PTTG was assessed using chromatin immunoprecipitation (ChIP)on-Chip in JEG-3 cells. PTTG was found to interact with the transcription factor SP1 to induce cyclin D3 expression to promote G1- to S-phase transition, independent of p21 (Tong et al., 2007).
1.3.11.4 NIS and PTTG The sodium-iodide symporter (NIS) is involved in the active transport of iodine into thyroid cells. This has clinical significance in the treatment of thyroid cancers where poor radioiodine uptake correlates to a poor prognosis. Heaney et al demonstrated reduced iodine uptake in rat thyroid FRTL5 cells transfected with PTTG (Heaney et al., 2001). Subsequently, it was shown that PTTG overexpression in thyroid cancers was associated with reduced radioiodine uptake (Saez et al., 2006). Boelaert et al investigated the mechanism by which PTTG repressed NIS. PTTG was found to repress NIS via FGF2. In addition, it was found that PTTG repressed NIS by binding to the hNUE element within the NIS promoter (Boelaert et al., 2007).
1.3.11.5 Other interactions Pei et al discovered the interaction between the ribosomal protein S10 and the molecular chaperone HSJ2 with PTTG in testicular cells (Pei, 1999). PTTG was also found to inhibit the phosphorylation of Aurora-A and histone H3 Aurora-A substrate which resulted in abnormally condensed chromatin in HCT116 cells (Tong et al., 2008).
40
General Introduction
1.4
PTTG BINDING FACTOR (PBF)
PTTG binding factor (PBF), mapped to chromosome 21q22.3, contains an open reading frame of 179 amino acids with a predicted molecular mass of 22 kDa and is also known as C21orf3 (Yaspo et al., 1998) or PTTG1IP. As a protein, PBF shares no significant homology with other human proteins but is highly conserved between species, suggesting unique function and evolutionary importance.
1.4.1 Structure, function and localisation of PBF PBF interacts with PTTG at its transactivating domain between amino acids 123 to 154 (Chien and Pei, 2000). Based on its structure it was speculated that PBF was a protein involved in cell signalling. Initially it was thought PBF was a cell surface protein because of the presence of a N-terminal signal peptide, transmembrane domain, endocytosis motif and N-glycosylation sites (Yaspo et al., 1998). PBF contains a bipartite nuclear localisation signal between amino acids 149 and 166 at the Cterminus suggesting it may be a nuclear protein (Chien and Pei, 2000). Subsequently, the nuclear role of PBF was confirmed when it was demonstrated to facilitate the entry of PTTG into the nucleus. Furthermore, the transcription and hence expression of FGF-2 by PTTG requires the presence of PBF (Chien and Pei, 2000). PBF has also been identified in intracellular vesicles where it co-localised with the late endosomal marker CD63 (Smith et al., 2009). The structure of PBF contains phosphorylation sites for cyclic AMP- and GMP- protein kinase and casein kinase II (Chien and Pei, 2000). Recently, our group has reported phosphorylation of PBF at residue Y174 by Src, a proto-oncogene tyrosine protein kinase. Phosphorylation of PBF leads to reduced ability to bind NIS (Smith et al., 2013). Also, glycosylation sites for N-linked and O-linked oligosaccharides were identified. Finally, a cleavable N-terminal signal sequence was detected (Chien and Pei, 2000).
41
General Introduction
Recently, PBF has also been found to be a secretory protein in MCF-7 cells and this function is encoded in the PBF amino acid region of 29 to 93 (Watkins et al., 2010). The schematic representation of PBF is shown in Figure 1-12 below.
Figure 1-12 Schematic representation of PBF reproduced from (Smith et al., 2010). PBF contains a bipartite nuclear localisation signal sequence and phosphorylation site (residue Y174) at its C-terminus. The N-terminal contains the signal sequence and glycosylation sites denoted by Gly.
1.4.2 PBF and NIS PBF has been shown to repress the expression of the NIS promoter via the upstream enhancer hNUE (Boelaert et al., 2007). Subsequently, it was found that PBF colocalises with NIS, altering its subcellular localisation and affecting the efficacy of radioiodine uptake. In the presence of PBF, NIS translocates from the membrane of COS-7 cells to the cytoplasm. Since the uptake of iodine is dependent on normal NIS function at the membrane, PBF alters the ability of the cell to uptake radioiodine (Smith et al., 2009). Hence, PBF impairs the ability of a thyroid epithelial cell to absorb iodine by directly repressing the expression of NIS and by altering the subcellular localisation of NIS from the membrane to the cytoplasm.
1.4.3 Expression in human tissue PBF is a ubiquitious protein that is expressed in more than one hundred different human tissues with the highest level of expression in the placenta. PBF was also shown to be overexpressed in colon
42
General Introduction
carcinoma, Wilm’s tumour, parathyroid tumours (Chien and Pei, 2000, Yaspo et al., 1998). The finding that PTTG is overexpressed in breast cancer (Ogbagabriel et al., 2005) led to the investigation of PBF in breast cancer. PBF was found to correlate with ER-positive breast cancers (Watkins et al., 2010). Recently PBF was reported to be implicated in the repression of angiogenesis in a mouse model of Down’s syndrome (Reynolds et al., 2010).
1.4.4 Expression of PBF in thyroid cancer Stable PBF overexpression in NIH3T3 murine fibroblast cells induced significant colony formation and the injection of nude mice with NIH3T3 cells stably overexpressing PBF led to tumour formation. These observations support the ability of PBF to transform cells in vitro and induce tumours in vivo (Stratford et al., 2005). Unsurprisingly, PBF expression was found to be increased in differentiated thyroid cancers and PBF mRNA expression was independently associated with tumour recurrence (Stratford et al., 2005). A cohort of 153 papillary carcinomas of the thyroid with a follow-up of 11.2 years was studied in the context of PBF expression. It was found that increased PBF expression was an independent predictor of worse prognosis and shorter disease-specific survival. In addition, PBF expression was associated with distant metastases at diagnosis, locoregional recurrence and tumour multicentricity (Hsueh et al., 2013).
1.4.5 PBF overexpression in murine thyroids The human homolog of PBF shares 80 % homology with murine PBF (Chien and Pei, 2000). Our group has generated a transgenic murine model of PBF overexpression in the thyroid gland (PBF-Tg) to study the effects of PBF in thyroid tumourigenesis. The human PBF gene was driven by the bovine thyroglobulin promoter and tagged with haemagglutinin (HA) (Read et al., 2011).
43
General Introduction
1.4.5.1 PBF and goitre formation The PBF-Tg mouse developed goitres by 6 weeks and more than 65 % of mice developed hyperplastic lesions by 52 weeks of age. These in vivo thyroid changes were not TSH driven and expression of growth factors VEGF, TGFβ, EGF and IGF-1 was unaltered compared to wild-type mice (Read et al., 2011). Further investigation into the aetiology of thyroid pathology revealed dysregulation of TSHR mRNA expression, which was significantly induced by PBF. The mRNA expression of PTTG, throglobulin, transcription factor 1 and paired box gene 8 mRNA was found to be unaltered in the presence of PBF overexpression in the thyroid of PBF-Tg mice (Read et al., 2011). Interestingly, no cancers were induced although close inspection of the nuclei of the hyperplastic lesions revealed large nuclei consistent with a proliferating cell. Although Akt mRNA expression was normal in PBF-Tg mice, the phosphorylation of Akt was found to be significantly induced by PBF. Furthermore, the expression of cyclin D1, a marker of proliferation and downstream target of activated Akt, showed an upregulation in PBF goitres. Thus, Akt activation, involved in the PI3K/Akt signalling pathway (described in 1.2.6.2), might be involved in inducing goitre and hyperplastic thyroid lesion formation. (Read et al., 2011). The thyroid phenotype observed in the PBF-Tg murine model was very similar to the PtenL/L;TPO-Cre mouse in which there is constitutive activation of the PI3K/Akt pathway (Yeager et al., 2007). The PtenL/L;TPO-Cre mouse did not develop thyroid cancers but developed metastatic follicular carcinomas when the Kras oncogenic mutation was introduced. On its own, the Kras mutation did not cause thyroid cancer (Miller et al., 2009). Further to the above finding in PBF-Tg mice thyroids, the expression of PBF in human thyroid goitres was examined. PBF mRNA expression was increased in human multinodular goitres (MNG) but protein expression of PBF in MNGs were similar to normal thyoid tissue. TSHR protein expression was also found to be upregulated in MNGs compared to normal thyroids (Read et al., 2011). The above findings implicate PBF, dysregulated TSHR and the PI3K/Akt pathway in goitrogenesis.
44
General Introduction
1.4.5.2 PBF and thyroid hormone The absorption of iodine is impaired in the presence of PBF due to reduced expression of the sodiumiodide symporter (NIS) and altered localisation of NIS, discussed in section 1.4.3 above (Boelaert, 2007, Smith, 2009). This observation was confirmed in vivo, in the PBF-Tg murine thyroids where radioiodide uptake was represssed (Read et al., 2011). Subsequently, Smith et al demonstrated that MCT8, a monocarboxylase transporter which mediates thyroid hormone secretion, binds and co-localises with PBF, resulting in reduced thyroid hormone secretion and an accumulation of thyroid hormone in the thyroid gland of the PBF-Tg mouse (Smith et al., 2012).
1.4.6 PBF and genetic instability PBF has been shown to be overexpressed in differentiated thyroid cancer and is an independent prognostic marker for recurrent disease (Hsueh et al., 2013, Stratford et al., 2005). Genetic instability is a hallmark feature of human cancers and a critical protein that governs genetic instability, Trp53, mediates crucial cellular responses to DNA damage (Petitjean et al., 2007). Although Trp53 is frequently mutated in human cancer, it is rarely mutated in differentiated thyroid cancer. PBF was found to increase the ubiquitination of Trp53 via the E3 ligase, Mdm2 (Read et al., 2014a, Read et al., 2014b). PBF was also found to interfere with the expression of DNA repair genes which may explain the genetic instability induced in thyroid cells overexpressing PBF (Read et al., 2014a).
45
General Introduction
1.5
GENETIC INSTABILITY
1.5.1 Hypotheses of tumourigenesis In the 1950s, researchers came to the hypothesis that tumourigenesis is a multistep process (Foulds, 1958). Calculations by Ashley based on gastric cancers estimated that three to eight mutations are required for tumourigenesis (Ashley, 1969). However, a study on retinoblastoma has shown that as few as 2 mutations was required for tumour formation (Knudson, 1971). It was initially proposed that most neoplasms are monoclonal and that tumour cells are more genetically unstable compared to normal cells (Nowell, 1976). Mutations are a necessary prerequisite for cancer formation and progression because they confer a selective growth or survival advantage. Mutations in genes that are critical in cell cycle control and DNA repair, that play an important role in maintaining genomic stability, play an important role in tumourigenesis, and led to the development of the mutator phenotype hypothesis (Tomlinson and Bodmer, 1999, Loeb, 2001). Negrini et al argues that the low frequency of mutations in caretaker genes in high-throughput sequencing studies of tumours in sporadic cancer does not support the mutator phenotype hypothesis. Furthermore, comparing untreated with treated glioblastomas, the mutations of caretaker genes was higher following treatment, suggesting that caretaker gene mutation is a late event rather than being essential in the initial development of the tumour (Negrini et al., 2010). It was therefore argued that whilst the mutator phenotype hypothesis explained the cause of hereditary tumours, it did not necessarily explain sporadic tumour genesis. The oncogene-induced DNA replication stress model put forward by Halazonetis et al. attempts to address the issue of genetic instability in tumourigenesis (Halazonetis et al., 2008). In this model, activated oncogenes induce replicative stress resulting in genetic instability and early cancer formation.
46
General Introduction
Overall, tumourigenesis from a normal to a cancerous cell is a complex process that is not fully understood. However, genetic instability is a consistent feature within cancers and has close association with genes involved in DNA damage and repair.
1.5.2 Genetic instability and cancer The hallmark feature of solid tumours is an unstable genome (Rajagopalan et al., 2002, Loeb and Monnat, 2008). Genetic instability in cancer reflects an increased rate of mutation which may arise either from increased rates of damage or defects in repair ability resulting in reduced genomic integrity within cells (Rajagopalan et al., 2002, Loeb and Monnat, 2008). Loss of genetic stability facilitates tumour development by generating mutants that can undergo clonal selection (Storchova and Pellman, 2004). At a molecular level, the accumulation of extra copies of DNA or chromosomes, chromosomal translocations, chromosomal inversions, chromosomal deletions, single-strand breaks in DNA, doublestrand breaks in DNA, the intercalation of foreign substances into the DNA double helix or any abnormal changes in the DNA tertiary structure resulting in the gain or loss of DNA or the misexpression of genes, cause genetic instability. Genetic instability can arise at a nucleotide (microsatellite instability) or chromosomal level (chromosomal instability). Microsatellite instability, characterised by nucleotide base changes that occur preferentially in repeat sequences, is typically caused by a defective repair process whilst chromosomal instability is characterised by aneuploidy caused by deletions or amplification of whole / part chromosomes (Fishel et al., 1993, Ionov et al., 1993, Lengauer et al., 1998). It has been reported that microsatellite instability and chromosomal instability are mutually exclusive, suggesting that instability of one variety alone is sufficient for driving tumourigenesis (Loeb and Loeb, 1999). Previous studies by Lengauer et al (Lengauer et al., 1997) and Willenbucher (Willenbucher et al., 1999) suggested that chromosomal instability is more dominant in driving tumourigenesis than microsatellite instability, at least in the case of colorectal tumours.
47
General Introduction
The role of genetic instability in thyroid cancer has been described previously (Kim et al., 2005, Mitmaker et al., 2008, Vaish et al., 2004, Lohrer et al., 2001). In addition, benign thyroid lesions have been shown to demonstrate microsatellite instability albeit at a lower level compared to thyroid cancer (Mitmaker et al., 2008). However, other studies did not show this (Bauer et al., 2002).
1.5.3 Measuring genetic instability Chromosomal instability, characterised by abnormal segregation of chromosomes and aneploidy (Rajagopalan et al., 2003), is caused by dysregulation of checkpoint proteins involved in mitosis, resulting in the failure to halt cell cycle progression to allow DNA repair to occur (Roschke et al., 2008). Additionally, apoptosis has an important role to play in removing cells that cannot be repaired following DNA damage. Consequently, apoptotic genes when dysregulated can induce chromosomal instability. This is confirmed by the observation that loss of p53 and p73 is associated with increased aneuploidy in mouse embryonic fibroblasts (Talos et al., 2007, Tomasini et al., 2008). Chromosomal instability is predominantly quantified using comparative genomic hybridization (CGH) or fluorescent in-situ hybridization (FISH). The FISH technique uses fluorescent probes to detect and localise the presence or absence of specific DNA sequences on chromosomes. In CGH, differences in hybridisation between normal and tumour human DNA can be compared using fluoresecent labelled fragments of normal and tumour on a scaffold of human metaphase chromosomes. This technique is sensitive for deletions or amplifications between 1 – 10 megabases (Bentz et al., 1998). CGH is an improvement of the FISH technique for quantifying chromosomal instability. Microsatellite (or simple sequence repeats) are repeat sequences scattered throughout the genome and are highly variable from individual to individual. These repeating units consists of 1 to 6 base pairs in length (but can be longer) and most commonly in humans, consist of repeat C and A nucleotides. The technique of measuring microsatellite instability arose when random oligonucleotide primers used to DNA fingerprint colorectal tumours and its adjacent tissue found tumour specific PCR products of
48
General Introduction
altered length encompassing segments with repetitive nucleotide sequences (Peinado et al., 1992). Basik et al used the (CA)8RY primer to measure the index of genetic instability in colorectal carcinoma and coined the phrase “Inter-simple sequence repeat PCR” for this technique (Basik et al., 1997). Subsequently, Stoler et al used the same technique of inter-simple sequence repeat PCR to determine genetic instability in colorectal polyps and found that genetic instability is an early event in colorectal tumour progression and is not a consequence of malignancy (Stoler et al., 1999). Initially, microsatellite instability was shown to be caused by a defective Msh2 gene, involved in mismatch repair, in colorectal carcinoma (Fishel et al., 1993). The mismatch repair family members repair base-base mispairs and larger insertions / deletions (Hakem, 2008). Defective mismatch repair results in increased rates of DNA replication errors within the genome causing microsatellite instability (MSI). Areas within the genome preferentially affected are genes such as TGFβRII, IGF-2R and BAX which contain microsatellites within their coding regions. As such, mismatch repair family defects result in incorrect DNA replication which may consequently cause microsatellite instability preferentially in the above genes (Loeb and Monnat, 2008). Microsatellite instability can also be caused by base excision repair pathway defects (Hakem, 2008, Guo and Loeb, 2003).
49
General Introduction
1.6
THE CELL CYCLE
Cell cycle dysregulation and consequently genetic instability is found widely in human cancer. Cell duplication begins with synthesis of DNA in the S phase and division in the mitosis (M) phase. In between these two phases are the G1 and G2 gaps where the cell “rests” to recover from the preceding phase. During G1 the cell may enter a special G0 phase which can last indefinitely. If the cell progresses beyond the end of G1, known as the restriction point, the cell is committed to DNA replication (Alberts et al., 2014). Cyclins and cyclin-dependent kinases (CDK) determine a cell’s progress through the cell cycle. The CDK-cyclin complexes directly involved in driving the cell cycle include the three interphase CDKs (CDK2, CDK4, and CDK6), CDK1 and ten cyclins belonging to 4 different classes (A, B, D and E cyclins) (Malumbres and Barbacid, 2009). DNA damage induces cell cycle arrest via the inhibition of CDK (Bartek et al., 2004) to allow for DNA repair. Unsuccessful DNA repair results in senescence or apoptosis, but if the cell cycle is allowed progress, genetic instability results (Kastan and Bartek, 2004).
1.6.1 CDK and DNA damage The genes ataxia-telangiectasia mutated (ATM) and ataxia telangiectasia and Rad3 related (ATR) are activated together with checkpoint kinases Chk1 and Chk2 in DNA damage, resulting in increased levels of the CDK inhibitor p21 or inhibition of CDK activators such as Cdc25 phosphatases (Bartek et al., 2004). This results in cell cycle arrest at G1/S and G2/M allowing for DNA repair to occur. Mutations in ATM result in ataxia-telangiectasia and ATR in Seckel syndrome (Cimprich and Cortez, 2008).
50
General Introduction
There is increasing evidence for the role of CDKs in DNA repair (Yata and Esashi, 2009). Homologous recombination repair during S phase and G2 is mediated by CDK1 and CDK2 phosphorylating Brca2 to modulate its interaction with Rad51 (Esashi et al., 2005).
1.6.2 CDK and chromosomal instability The CDK1-cyclin complex regulates the centrosome cycle and the onset of mitosis. The phosphorylation of downstream targets to the CDK1-cyclin complex during G2 results in centrosome separation, nuclear envelope breakdown and chromosome condensation (Malumbres and Barbacid, 2005). CDK1 activity is repressed at the metaphase-anaphase interface through the activation of separase, to allow sister chromatid separation (Musacchio and Salmon, 2007). Hence, in the dysregulation of CDK1, aneuploidy and chromosomal instability results.
51
General Introduction
1.7
ATM/ATR signalling pathway
The cell is under constant stress and requires coping mechanisms to respond and maintain its integrity. Genetic instability arises when the appropriate response mechanism fails. As discussed above, genetic instability is related to progressive degeneration, cancer predisposition and sensitivity to various DNA damaging agents (Bensimon et al., 2010, Bhatti et al., 2011, Nam and Cortez, 2011, O'Driscoll et al., 2007, Ruzankina et al., 2007). The ATM / ATR signalling pathway, which plays a central role in response to DNA damage is shown in Figure 1-13. The ATM (ataxia-telangiectasia mutated) gene belongs to a family of serine/threonine kinases which was first discovered in the ataxia-telangiectasia disorder. This autosomal recessive condition is characterized by progressive cerebellar neurodegeneration and is predisposed to lymphoreticular maliganancies (Lavin, 2008, Perlman et al., 2012). Strikingly, patients with the condition of ataxia-telangiectasia are particularly sensitive to ionising radiation (Perlman et al., 2012). ATM plays a central role in the coordination of cellular functions such as cell cycle arrest, DNA repair and apoptosis in response to DNA damage (Sancar et al., 2004). The general consensus is that the ATM response occurs in double strand breaks whilst ATR (shown in Figure 1-14) is more pertinent in single strand breaks or stalled replication forks. However there is interplay between the 2 genes where ATM can activate ATR (Shiotani and Zou, 2009) or ATR can phosphorylate ATM, rendering it active (Dodson and Tibbetts, 2006, Stiff et al., 2006, Yajima et al., 2009, Sirbu et al., 2011). Murine ATM knockout models develop leukaemia, in particular thymic lymphoma and sporadic intestinal cancer (Ejima et al., 2000). ATM has been shown to phosphorylate and activate Trp53, the “guardian of the genome”, in response to DNA damage (Banin et al., 1998, Nakagawa et al., 1999).
52
General Introduction
Figure 1-13 An overview of the ATM molecular pathway demonstrating its role in doublestrand DNA damage (Adapted from SABiosciences.com). The diagram depicts known genes that interact with ATM and its downstream functions.
53
General Introduction
Figure 1-14 ATR signalling pathway adapted from (Shiotani and Zou, 2009).The diagram shows the activation and function of ATR in response to single-strand breaks.
54
General Introduction
1.8
HYPOTHESIS AND AIMS
PTTG has been shown to have transforming and tumourigenic effects in vitro and in vivo. Additionally, PTTG causes aneuploidy and genetic instability. PTTG is implicated in DNA damage and DNA repair through its known interaction with Trp53 and Ku70. The binding partner of PTTG, PBF, transports PTTG to the nucleus for its transcription and securin function. PBF is independently transforming in vitro and tumourigenic in vivo. PBF is also known to interact with Trp53 which is intimately involved in the response to DNA damage. Important to this investigation is the finding that PBF and PTTG are both overexpressed in human differentiated thyroid cancer. As both PTTG and PBF are transforming genes which show functional interaction and overexpression in thyroid cancer, we hypothesised that PBF and PTTG have dependent and independent roles in maintaining genomic stability in the thyroid gland. The aims of this investigation were: 1) The development of a murine model of simultaneous PBF and PTTG overexpression in the thyroid gland. Our group had murine models of PTTG or PBF overexpression in the thyroid gland. The observation that murine models of PBF overexpression in the thyroid developed hyperplastic nodules but not cancer and the murine model of PTTG overexpression in the thyroid had small thyroid glands led us to hypothesise that two “hits” might be required for thyroid tumourigenesis. Hence, I aimed to create a bi-transgenic murine model of both PBF and PTTG overexpression in the thyroid gland. 2) The use of fluorescent inter-simple sequence repeat PCR (FISSR-PCR) for determining genetic instability in our murine models. FISSR-PCR was modified for measuring genetic instability with small quantities of DNA obtained from our transgenic models and from primary murine thyroid culture. 3) Establishing key genes involved in the DNA damage / DNA repair pathway in the context of PBF and PTTG expression with ionising radiation. This part of the study involved the use of a
55
General Introduction
murine DNA damage PCR microarray to screen for genetic alterations within the DNA damage / DNA repair pathway.
56
2 Materials and Methods
Materials and Methods
2.1
MURINE THYROID DISSECTION
Murine models used in the following experiments were euthanized with an intra-peritoneal injection of sodium pentobarbital (0.2 mls of Euthatal; 200mg in 1 ml) and confirmed to be dead prior to harvesting the thyroid gland. The thyroid gland was removed with the aid of a dissecting microscope at 10x magnification. Initially, the mouse was secured to a cork board, exposing the neck. Subsequently, the fur and skin were excised and the strap muscles divided inferiorly. Superior retraction of the strap muscles revealed the thyroid gland sitting on the trachea as shown in Figure 2-1. The thyroid gland was removed with micro forceps and scissors and stored in formalin for histology, RNAlater for RNA analysis, liquid nitrogen for protein analysis or phosphate buffer solution (PBS) for primary cell culture.
THYROID GLAND
TRACHEA Figure 2-1 View of murine thyroid under 10x magnification during dissection. This view of the thyroid gland was obtained following the removal of fur and strap muscles in the neck.
2.2
PRIMARY MURINE THYROID CULTURE (PMTC)
Murine thyroids were mechanically disrupted in PBS prior to being digested in 0.2 % Type II collagenase (Worthington Biochemicals) at 37 °C for 45 minutes on a rotator. Subsequently, thyroid culture medium described previously by Ambesi-Impiombato et al (Ambesi-Impiombato et al.,
58
Materials and Methods
1980), was added to inactivate collagenase. The mixture was centrifuged at 700 G for 10 minutes to obtain a pellet comprising of single thyroid follicles. The supernatant was discarded and the pellet resuspended in 1ml of thyroid culture medium. The resuspended thyroid cells were seeded into 12well plates (2 wells for each hPTTG or WT thyroid; 4 wells for each PBF and Bi-Transgenic thyroid) in thyroid culture medium, supplemented with thyrotrophin (300 mU/l), insulin (10 mg/ml),
transferrin (5 mg/ml) [Sigma], hydrocortisone (3.5 ng/ml) [Sigma], somatostatin (10 ng/ml) [Sigma], glycyl-L-histidyl-L-lysine acetate (2 ng/ml) [Sigma], penicillin (105 U/l), streptomycin (100 mg/l) and 5 % fetal calf serum. After 3 days, the thyroid culture medium was changed with calf serum omitted. Experiments were performed at day 10. In experiments involving radiation, primary murine cultures were exposed to 15 Gy of radiation on day 9 and harvested 24 hours later.
2.3
DNA EXTRACTION
DNA was extracted as per the DNA extraction kit protocol (DNeasy 96 Blood & Tissue Kit; Qiagen). Briefly, tissue was dissolved in ATL buffer and Protein K at 56°C for 3 hours. RNAse A was added to the mixture and left to stand for 2 minutes at room temperature to remove RNA contaminants. AL buffer and 100% ethanol were added to the mixture and filtered in a DNeasy Mini spin column. The DNA binds to the filter whilst contaminants wash through. The filter was washed twice to ensure a clean DNA sample. Finally, the DNA was extracted by elution through adding AE buffer to the filter. The amount of DNA was quantified using a spectrophotometer (Nanodrop ND-1000; Thermoscientific).
59
Materials and Methods
2.4
RNA EXTRACTION
RNA was extracted using the RNAeasy micro extraction kit protocol (RNeasy Micro; Qiagen). Thyroid tissue was processed using β-Mercaptoethanol to denature RNAses and buffer RLT (proprietary buffer; Qiagen RNAeasy micro extraction kit) to promote RNA binding to the filter. RNA was filtered onto the proprietary filter within the RNeasy MinElute spin column and washed several times to remove contaminants. RNase-free water to dissolve the purified RNA was used in the final step to collect purified RNA. RNA concentration was quantified using spectrophotometry (Nanodrop ND-1000; Thermoscientific). To ensure consistent and good RNA quality, the spectral value of A260:A230 ratio was greater than 1.7 and A260:A280 ratio, between 1.8 and 2.0. The average amount of total RNA obtained in this manner ranged between 220 to 800 ng.
2.5
REVERSE TRANSCRIPTION
Reverse transcription was carried out using the Promega Reverse Transcription System. Briefly, each reaction consisted of 0.5 µg RNA, 2 µl 25 mM MgCl2, 1 µl 10X reverse transcriptase buffer, 1µl 10mM deoxynucleotide triphosphate, 5 pmol random hexamers, 10 units ribonuclease inhibitor (RNAsin) and 7.5 units avian myeloblastosis virus (AMV) reverse transcriptase, in a total reaction volume of 10 µl. The reaction was incubated for 10 minutes at room temperature followed by 60 minutes at 42 °C to allow primer fixation and extension. The reaction was terminated by heating the sample for 5 minutes at 95 °C.
60
Materials and Methods
2.6
QUANTITATIVE REAL-TIME POLYMERASE CHAIN REACTION (qRTPCR)
Quantitative real-time polymerase chain reaction (qRT-PCR) was performed using TaqMan™ chemistry. In this technique, the target gene of interest is amplified through polymerase chain reaction (PCR). The probe used within the reaction was specially labelled with a 5’ reporter dye (FAM 6carboxy-fluorescein or VIC) and a 3’ quencher dye (TAMRA 6-carboxy-tetramethyl-rhodamine). PCR amplification of the target gene released the reporter dye for each cycle of the PCR reaction which was read by a laser and CCD camera. The quantity of gene within the known amount of cDNA or DNA was thus determined. Pre-optimised specific gene-expression assays for qRT-PCR were purchased from Applied Biosystems, UK. All target gene probes were labelled with FAM and housekeeping genes (DS-CAM, 18S or β-actin) labelled with VIC. These reactions were carried out in 25 µl volumes on 96 well plates. Each reaction comprised of 5 ng DNA, 450 nM each target gene forward and reverse primers, 175 nM target gene probe, 450 nM housekeeping gene forward and reverse primers, 175 nM housekeeping gene probe and 12.5 µl of 1x TaqMan Universal PCR Master Mix (Applied Biosystems). 1x TaqMan Universal PCR Master Mix consisted of 3 mM Magnesium Acetate, 200 µM dNTPs, 1.25 units Ampli-Taq Gold polymerase and 1.25 units AmpErase UNG. The reaction cycle in the ABI 7500 Sequence Detection System was as follows: 50 °C for 2 minutes, 95 °C for 10 minutes followed by 44 cycles of 95 °C for 15 seconds and 60 °C for 1 minute.
The cycle number at which a calculated threshold line bisects the logarithmic PCR plot determined the Ct value. ΔCt values were calculated by subtracting the Ct of the housekeeping gene from Ct of the target gene of interest. The fold change was derived from the following formula: 2 –(ΔCt of experimental group – ΔCt of control group).
61
Materials and Methods
Reactions were multiplexed where possible. To determine if the reaction could be multiplexed, qRTPCR was performed with cDNA amounts of 1 ng, 3 ng, 10 ng, 30 ng and 100 ng in a multiplexing experiment using the target gene of interest and housekeeping gene in question. The logarithmic value of cDNA quantity was plotted against the Ct value of each gene in the qRT-PCR experiment (Figure 2-2A). The ΔCt value, obtained by subtracting the target gene Ct value from the β-actin Ct value was plotted in a separate graph shown in Figure 2-2B. The gradient of the best fit linear plot of the logarithmic cDNA quantity against the ΔCt value should be between -0.1 and 0.1 if multiplexing was possible. Figure 2-2 shows the multiplexing experiment between Trp53 and β-actin, used as an example illustrating that these two genes cannot be multiplexed.
Ct
A
Trp53 and β-actin 30 25 20 15 10 5 0
y = -3.3576x + 27.798 β-actin Trp53
y = -3.0686x + 22.233
Linear (β-actin) 0
0.5
1
1.5
2
2.5
Linear (Trp53)
Log total cDNA amount (ng)
B
∆Ct Trp53 and β-actin
ΔCt
6 y = -0.289x + 5.5655
4
Series1 Linear (Series1)
2
0 0
0.5
1
1.5
2
2.5
Log total cDNA amount (ng)
Figure 2-2 Linear plots of logarithmic cDNA against Ct values (A) and Δ Ct values (B).
62
Materials and Methods
2.7
PROTEIN EXTRACTION
Tissue and cell culture harvested for protein extraction were homogenized in a tube containing a cocktail of Radio Immuno Precipitation Assay (RIPA) buffer (329 µl) and protease inhibitor (21 µl). Subsequently, the crude lysate was centrifuged at 13,200 rpm for 10 mins at 4 °C. The supernatant was removed and protein concentration quantified using spectrophotometry (Nanodrop ND-1000; Thermoscientific).
2.8
WESTERN BLOT
12 % separating gels were made using 1.875 ml 1.5 M Tris HCl (pH 8.8), 3 mls acrylamide, 75 µl 10 % SDS, 75 µl 10 % APS and 7.5 µl TEMED. Stacking gels were made using 0.625 ml 0.8 M Tris HCl (pH 6.8), 0.3125 ml acrylamide, 50 µl 10 % SDS, 50 µl 10 % APS and 5 µl TEMED. 20 µg of protein in 4X Laemmli buffer (0.1075 g DTT: 1 ml Laemmli buffer) was loaded into each well. Proteins separated by electrophoresis were transferred to polyvinylidene fluoride (PVDF) membranes, incubated in 5 % non-fat milk in Tris-Bufferred Saline Tween-20 (TBST) and subsequently incubated overnight at 4 °C with a primary monoclonal antibody. After washing in TBST, membranes were incubated in the appropriate secondary antibodies conjugated to horseradish peroxidise. After additional washes, antigen-antibody complexes were visualized by the ECL chemiluminescence detection system on Kodak BioMax Light film. Actin expression (Anti-βactin; 1:10,000; AC-15; Sigma-Aldrich) was also determined to control for differences in protein loading between wells. Western blot films were scanned using the CanoScan Flatbed Photo and Document scanner and relative protein quantification was determined using Image J software.
63
Materials and Methods
2.9
IMMUNOFLUORESCENCE
The cells in each well were fixed with 800 μl fixing solution (20 ml of 0.2 M PBS, 0.8g of paraformaldehyde, 0.8 g glucose, 8 ml of 0.1 % sodium azide and 12 ml water) for 20 minutes at room temperature. Following rinsing with PBS, the cells were permeabilised in 800 μl of chilled methanol for 10 minutes. The cells were rinsed again and blocked with 5 % blocking serum for 20 minutes at room temperature diluted in PBS containing 1 % BSA (bovine serum albumin). Subsequently, excess blocking serum is aspirated and cells incubated in primary HA antibody (mouse monoclonal antiHA.11 (16B12); 1:1000; Covance Research Products) diluted in PBS containing 1% BSA for 1 hour. The cells are rinsed again and incubated for 1 hour in a red fluorescent labelled secondary antibody diluted in PBS containing 1 % BSA and 1.5 % blocking serum. The cells are then visualised using an inverting microscope. DAPI (4',6-diamidino-2-phenylindole; 300 µl of diluted stock solution as per manufacturer’s recommendation; ThermoFisher Scientific) staining was used to identify nuclei by staining DNA blue.
2.10 STATISTICAL ANALYSIS Data was analysed using Sigma Stat and SPSS (SPSS Science Software UK Ltd). The numerical data was determined if it had a normal distribution. Student’s t-test was used for comparison between two groups of parametric data. Kaplan-Meier survival analysis was determined using XLSTAT add-on to Microsoft Excel 2007. Two different statistical tests were used to determine the p-value for survival. The p-value derived from the Cox-Mantel or log-rank test is more powerful for detecting late, whilst the Wilcoxon test is more powerful for detecting early differences in probabilities for survival. A p value ≤ 0.05 was taken to be statistically significant.
64
3 Bi-transgenic murine model of human PTTG and PBF overexpression in the thyroid gland
Bi-transgenic murine model of human PTTG and PBF overexpression in the thyroid gland
3.1
INTRODUCTION
The human securin PTTG was first described in rat pituitary tumours (Pei and Melmed, 1997). The role of PTTG in cell transformation and tumourigenesis was subsequently confirmed when murine fibroblast 3T3 cells overexpressing PTTG, injected into nude mice developed tumours (Pei and Melmed, 1997). Further experiments in vitro showed aneuploidy or chromosomal instability when overexpressed (Wang et al., 2001, Winnepenninckx et al., 2006, Yu et al., 2000a, Yu et al., 2003). Further evidence for the role of PTTG in tumourigenesis came from the observation that PTTG was overexpressed in a variety of tumour types (Cho-Rok et al., 2006, Dominguez et al., 1998, Heaney et al., 2000, Kakar and Malik, 2006, Puri et al., 2001, Shibata et al., 2002, Tfelt-Hansen et al., 2004, Wen et al., 2004). More importantly, the link between thyroid cancer and PTTG has been well-established (Liang et al., 2011, Kim et al., 2007a, Kim et al., 2005, Boelaert et al., 2003, Zatelli et al., 2010). The binding factor for PTTG (PBF) was initially discovered by Yaspo et al and its function as a transport protein for PTTG was described later on (Chien and Pei, 2000, Yaspo et al., 1998). PTTG does not have a nuclear localising signal of its own and yet needs to be within the nucleus for its function. The finding that PTTG requires the presence of PBF to upregulate fibroblast growth factor 2 (FGF-2), a downstream target of PTTG, provides compelling evidence for the need for PBF for its actions (Chien and Pei, 2000). Hence, it was not surprising that the presence of PTTG upregulated the expression of PBF (Stratford et al., 2005). Interestingly, our group has shown that PBF has transforming and tumourigenic actions independent of PTTG (Stratford et al., 2005). Additionally, PBF has been shown to cause genetic instability, is overexpressed in well differentiated thyroid cancer and an independent prognostic marker for tumour recurrence (Hsueh et al., 2013, Stratford et al., 2005). PBF and PTTG are widely conserved in the eukaryote species. Human PTTG and PBF share 89 % and 93 % homology respectively with murine PBF and PTTG (Zhang et al., 1999, Chien and Pei, 2000). Murine models were generated by our group to study the effects of PTTG or PBF
66
Bi-transgenic murine model of human PTTG and PBF overexpression in the thyroid gland
overexpression in the thyroid gland. Murine models with thyroglobulin driven human PTTG overexpression in the thyroid gland (PTTG +/+) did not demonstrate tumour formation but instead developed small thyroids with reduced cellular proliferation (Lewy et al., 2013). Murine models overexpressing human PBF (PBF-Tg) in the thyroid exhibited hyperplastic and macrofollicular lesions but with no evidence of thyroid tumours (Read et al., 2011). Taken together, PTTG or PBF overexpression individually is inadequate for thyroid tumour formation. Overall, PBF and PTTG are proto-oncogenes which are overexpressed in human thyroid cancers. PBF binds to PTTG, facilitating its entry into the nucleus, and has independent tumourigenic actions. The observation that murine models overexpressing PBF or PTTG alone in the thyroid do not develop thyroid cancers suggests that PTTG and PBF independently do not induce thyroid tumours in vivo. However, given the strong tumourigenic potential of both PBF and PTTG, and that the genes bind to each other, a murine model overexpressing both PTTG and PBF in the thyroid gland was hypothesised to develop thyroid cancer. Thus, the aim of this chapter is to describe and characterise a transgenic murine model overexpressing both PBF and PTTG in the thyroid gland.
67
Bi-transgenic murine model of human PTTG and PBF overexpression in the thyroid gland
3.2
MATERIALS AND METHODS
3.2.1 Generation of murine model Two FVB/N murine models of targeted human PBF (hPBF) and PTTG (hPTTG) overexpression in the thyroid gland were generated by Dr. Martin Read and Dr. Gregory Lewy at the University of Birmingham. Very briefly, plasmids containing the gene of interest attached to a bovine thyroglobulin promoter and tag were injected into FVB/N embryos. The tags used for the hPBF gene and hPTTG gene were HA and FLAG respectively. These embryos were implanted into pseudo-pregnant mice and founder lines were obtained from these litters. The murine model expressing both human PBF and PTTG in the thyroid gland was created by mating the two lines described above. The first pairing between the two different models created a mouse that was heterozygous for both PBF and PTTG (PBFHet-PTTGHet) as depicted in Figure 3-1. The subsequent mating genetics was predictably complicated because of random gene insertion(s) within the genome for the PBF and PTTG murine models. Consequently, mating of the PBF HET-PTTG HET lines had created a colony of transgenic mice with different genotypes, as seen in Figure 3-2. Our plan was to perform all our experiments on the murine model homozygous for both PBF and PTTG in the thyroid gland (BI-Trans). To achieve the objective of producing BI-Trans mice, we aimed to mate BI-Trans with BI-Trans mice, to ensure a continual supply of BI-Trans murine models shown in Figure 3-3.
68
Bi-transgenic murine model of human PTTG and PBF overexpression in the thyroid gland
Figure 3-1 Mating genetics of PBF and PTTG homozygotes. Mating a PBF homozygote with a PTTG homozygote created a colony of PBF heterozygote – PTTG heterozygote (PBFHetPTTGHet).
Figure 3-2 Mating genetics of PBFHet-PTTGHet. Mating PBFHet-PTTGHet with PBFHetPTTGHet yielded a colony of mixed genotypes illustrated above. The offspring genotype pattern is caused by random PBF / PTTG gene insertion in different chromosomes during the creation of the murine models.
69
Bi-transgenic murine model of human PTTG and PBF overexpression in the thyroid gland
Figure 3-3 Mating genetics of BI-Trans murine models. The mating of BI-Trans models creates a colony comprising exclusively of BI-Trans offspring.
A total of 1271 mice were generate for the above study. Comparison data for the individual hPTTG and hPBF murine models were provided by Dr. Gregory Lewy and Dr. Martin Read respectively (University of Birmingham).
70
Bi-transgenic murine model of human PTTG and PBF overexpression in the thyroid gland
3.2.2 Ageing colony To study the long term effects of PBF and PTTG overexpression in the thyroid gland, we had a separate colony of mice homozygous for PBF and heterozygous for PTTG (PBFHomo-PTTGHet). The rationale for choosing this model for this aspect of our investigation is discussed in section 3.3.2 and 3.4.2.
3.2.3 Tissue DNA extraction DNA from ear clippings was obtained from the murine colony and extracted as per protocol described in section 2.3.
3.2.4 Zygosity screening Zygosity determination was performed using qRT-PCR on the ABI 7500 Sequence Detection System which employs TaqMan™ chemistry for quantification of DNA levels, described previously in section 2.6. In this chapter, the target gene of interest is PBF and PTTG and hence, the appropriate primers and probes were used. The housekeeping gene for the qRT-PCR experiments in this chapter was DSCAM. The zygosity was determined based on the fold change value of the mice of unknown genotype compared to the control mice samples of known genotype.
3.2.5 Murine organ dissection Following euthanasia of the murine model, the animal was weighed and organs dissected. Harvested organs including the thyroid gland, lungs, heart, kidney, spleen and liver were initially weighed and
71
Bi-transgenic murine model of human PTTG and PBF overexpression in the thyroid gland
stored appropriately. Organs for protein experiments were stored in liquid nitrogen and RNA experiments, in RNAlater. Microdissection for the thyroid gland was described in section 2.1.
3.2.6 Western blot Protein was extracted from harvested murine organs as described in section 2.7 and used for Western blot described in section 2.8. The primary antibodies used were against human PTTG at 2 μg/ml (Invitrogen, UK) and human PBF at 1:1000 (rabbit polyclonal; manufactured in-house Dr. Turnell, University of Birmingham). Subsequent incubation with the appropriate secondary antibody conjugated to horseradish peroxidise (mouse anti-rabbit; Dakocytomation, UK) was performed for 1 hour at room temperature before being developed.
72
Bi-transgenic murine model of human PTTG and PBF overexpression in the thyroid gland
3.2.7 Cardiac puncture The animal was rendered unconscious by an overdose of Sevofluorane. Cardiac puncture was carried out using a 23G needle and approximately 1ml of blood was aspirated. The sample was stored in a 1.5 ml Eppendorf at 4 °C for 24 hours. The sample was then spun at 13,200 rpm for 20 minutes. The extracted supernatant (serum) was transferred to a second 1.5 ml Eppendorf and spun at 13,200 rpm for 10 minutes. The serum was extracted again and transferred to the final Eppendorf which was stored at -80 °C prior to usage.
3.2.8 Thyroid function test Murine thyroid stimulating hormone (TSH) levels were determined by the laboratory of Professor Samuel Refetoff at the University of Chicago. Technical details of this assay has been published previously (Pohlenz et al., 1999). Total T3 and T4 levels were measured using a radioimmunoassay technique (MP Biomedicals thyroid radioimmunoassay kit). Briefly, murine serum and radiolabelled tracer solution were added to tubes coated with either T3 or T4 antibodies. The analyte and radiolabelled tracer compete for limited antibody binding sites. Radioactivity level, inversely proportional to analyte concentration, was quantified using a gamma counter. Finally, the concentration of T 3 and T4 was determined by interpolation from a standard curve of percentage of trace level versus µg/dL T3 or T4.
3.2.9 Histology Murine thyroids were fixed in formalin and sent to the Pathology Department at University Hospital Birmingham NHS Trust for processing (courtesy of Dr. Adrian Warfield). Tissue samples were embedded in paraffin and sectioned to produce slides. Slides were stained with Haematoxylin and Eosin (H&E).
73
Bi-transgenic murine model of human PTTG and PBF overexpression in the thyroid gland
3.3
RESULTS
3.3.1 Validation of the murine model We generated a murine model which was homozygous for both PBF and PTTG in the thyroid gland only (BI-Trans). We confirmed the overexpression of PBF and PTTG in the thyroid gland of the BITrans model. We compared the expression of PBF and PTTG in the thyroid gland of the various murine models used in our experiments; wild-type (WT), PBF (homozygous for PBF), PTTG+/+ (homozygous for PTTG) and BI-Trans. The Western blot below (Figure 3-4) confirmed the overexpression of hPBF in the PBF and BI-Trans models and hPTTG in the PTTG+/+ and BI-Trans models (n=4). PBF appeared as a band between 25kDa and 37kDa and PTTG was expressed as a band around 29 kDa. The overexpression of hPBF did not appear to increase PTTG protein. Conversely, the overexpression of hPTTG did not upregulate the protein expression of endogenous PBF as seen in Figure 3.4.
Figure 3-4 Western blot showing the expression of hPBF and hPTTG in WT, PBF, PTTG+/+ and BI-Trans thyroids (n=4 for each genotype; figure shows n=2 for each genotype).
Next, we used the PBFHomo-PTTGHet model (homozygous for PBF and heterozygous for PTTG) to determine the expression of our genes of interest, PBF and PTTG, in the thyroid, heart, lung, liver, spleen and kidney. The reasons for this is discussed in section 3.3.2 and 3.4.2 below. The murine PBFHomo-PTTGHet thyroids overexpressed human PBF and human PTTG in the thyroid gland only and not in the lung, liver, spleen and kidney. A β-actin result in the heart was difficult to obtain as
74
Bi-transgenic murine model of human PTTG and PBF overexpression in the thyroid gland
demonstrated in Figure 3.5 because of its fibrous nature. This was consistent with our previous experience with the heart organ on Western blot (unpublished).
Figure 3-5 Western blot showing the expression of hPBF and hPTTG in PBFHomo-PTTGHet (*) in various organs (n=1) compared to wild type (WT).
3.3.2 Reduced mortality in founder BI-Trans models Seven PBFHet-PTTGHet breeding pairs were set up to create founder lines. Genotype was determined from ear-clippings using qRT-PCR, described above. Offspring genotype followed Mendelian genetic principles as shown in Figure 3-2, in expected ratios. Early bi-transgenic (BI-Trans) founder members homozygous for both PBF and PTTG, were put in breeding pairs. Interestingly, early BI-Trans founder members had reduced mortality compared to PBFHomoPTTGHet. In a cohort of breeding pairs over a period of 200 days, 64.3% of Bi-Trans mice died or were culled due to illness (n=27). In comparison, the mortality rate in the PBFHomo-PTTGHet genotype was 17.9 % over the same period of time (n=5). The Bi-Trans murine model had an average lifespan of 150.9 ± 1.0 days (n=53). The average lifespan of a PBFHomo-PTTGHet model was 477.9 ± 2.9 days (n=89). There was a statistically significant difference between the mortality rates of BITrans (Wilcoxon and log-rank p PBFHomoPTTGHet > PTTG+/+ > BI-Trans. The reduced mortality in PBFHomo-PTTGHet compared to PTTG+/- cannot be explained by its PTTG zygosity. Hence, PBF appeared to have a synergistic effect with PTTG in reducing mortality in our model. On its own, however, PBF did not impact on survival rates.
3.4.3 Fertility issues Overall, the PBFHomo-PTTGHet and BI-Trans models had reduced fertility. Breeding pairs consisting of two BI-Transgenic mice did not reproduce. The cause for infertility may lie in the BITransgenic male. Pairings between BI-transgenic females and male PBFHomo-PTTGHets led to successful breeding. By contrast, BI-Trans male and PBFHomo-PTTGHet females yielded no offspring. The PBFHomo-PTTGHet and PBFHomo-PTTGHet pairing yielded more offspring than PBFHomoPTTGHet and BI-Transgenic. To maximize yield of the PBFHomo-PTTGHet or BI-Trans model, a new breeding strategy was adopted. The pairings and Mendelian genetics shown in Figures 3-29 and 3-30 were adopted to maintain perpetuity of the breeding colony.
101
Bi-transgenic murine model of human PTTG and PBF overexpression in the thyroid gland
Figure 3-29 Genetics of PBFHomo-PTTGHet x PBFHomo-PTTGHet breeding.
Figure 3-30 Genetics of PBFHomo-PTTGHet and BI-Trans breeding. 102
Bi-transgenic murine model of human PTTG and PBF overexpression in the thyroid gland
3.4.4 Goitre formation in the PBFHomo-PTTGHet murine model Goitres found in the PBFHomo-PTTGHet model were comparable in size to the PBF model. This suggests that PTTG in the presence of PBF has no additional effect on thyroid size. PBFHomo-PTTGHet thyroid weights increased over time in a similar fashion to the PBF model. WT and PTTG+/- thyroid weights remained constant over the same period of time. Goitre formation appeared to be PBF dependent, even in the presence of PTTG expression. There was no gender difference in the corrected thyroid weight of WT and PTTG+/- mice (data not shown). Female thyroid weight adjusted for body weight in both PBGHomo-PTTGHet and PBF murine models was consistently higher than for male counterparts over the study period (data not shown). Hence, it appeared that goitre formation caused by the overexpression of PBF in the thyroid had a greater effect in females than males.
3.4.5 Thyroid function tests Thyroid function tests are usually interpreted in conjunction with each other. Thyroid hormone biosynthesis operates on a negative feedback loop. Thyroid hormone is produced by the thyroid gland in the form of T4 and T3. The active component T3 is secreted in small amounts relative to T4. The proportion of T4 to T3 in humans is approximately 20:1. High blood thyroid hormone levels inhibit the secretion of thyroid stimulating hormone (TSH) from the anterior pituitary gland. Low TSH results in reduced thyroid hormone secretion from the thyroid gland. Prolonged and high levels of TSH can cause thyroid cell growth in addition to stimulating thyroid production. Hence, the concentration of T3 and T4 should have an inverse relationship to TSH levels. We performed thyroid function tests on the various genotypes at different time points. There was no observed correlation between TSH, T3 and T4 levels within each genotype suggesting that the study had insufficient power to demonstrate a statistical difference. However, given that we were dealing
103
Bi-transgenic murine model of human PTTG and PBF overexpression in the thyroid gland
with a mutant model where changes should be gross, any effect caused by the overexpression of our genes of interest should be apparent, even in low numbers. Since there was no gross increase in TSH levels, goitre formation in the PBFHomo-PTTGHet and PBF models is not likely to be TSH driven.
3.4.6 Adenoma formation in PBFHomo-PTTGHet PBFHomo-PTTGHet thyroids demonstrated areas of hypercellularity and lesion formation microscopically. These histological changes were found predominantly in female PBFHomoPTTGHet from 12 months onwards. Lesions appeared well-circumscribed with no invasive features suggestive of a cancer. The likelihood is that these lesions represent adenomas. Thyroid function tests did not reveal hyperthyroidism in the mice with adenomas suggesting these lesions are non-functional. It is not possible at this stage to determine if these histological changes are separate entities or progression of the same event. Since the above histological changes were similar to PBF thyroids microscopically, adenoma formation and hypercellularity are likely an effect of PBF overexpression in the thyroid.
3.5
CONCLUSION
In terms of thyroid changes, the PBFHomo-PTTGHet model appeared phenotypically to be a PBF model. The survival characteristics of the PBFHomo-PTTGHet model mirrors the PTTG model. Reduced fertility of the PBFHomo-PTTGHet model is likely a PTTG phenotype because the PBF murine model had normal fertiility. Unfortunately, we did not have any breeding data for the PTTG model to make a direct comparison. The PBFHomo-PTTGHet model did not produce any cancers as hypothesized. Hence, we next examined whether the PBFHomo-PTTGHet thyroids exhibit genetic instability.
104
4 Genetic instability in transgenic murine thyroids
Genetic instability in transgenic murine thyroids
4.1
INTRODUCTION
The ability to replicate is a hallmark feature of a living cell. Occasionally, errors of replication result in a mutation. Gene mutations that involve cell cycle regulation, DNA repair and the repair of DNA damage, which is integral to maintaining genomic stability, can result in mutations that lead to the development of tumours. Hence, genetic instability is an integral feature of cancers (Loeb and Monnat, 2008, Rajagopalan et al., 2002). Furthermore, genetic instability has been demonstrated previously in thyroid cancers by multiple groups (Kim et al., 2005, Mitsutake et al., 2005, Saavedra et al., 2000, Ward et al., 1998, Wreesmann et al., Tung et al., 1997). A technique described by Basik et al, inter-simple sequence repeat PCR (ISSR-PCR), was used for measuring genetic instability in colorectal cancers (Basik et al., 1997). In this technique, a primer to microsatellite sequences was used in the polymerase chain reaction (PCR) to compare normal and tumour samples. Subsequently, Kim et al demonstrated genetic instability in thyroid cancers using the modified technique of fluorescent ISSR-PCR (FISSR-PCR) (Kim et al., 2005). PBF and PTTG have been shown previously to have independent and dependent effects on tumourigenesis (Stratford et al., 2005). PBF expression is increased in thyroid cancer both at a protein and mRNA level (Stratford et al., 2005). A high level of PBF expression has been shown to be an independent prognostic marker for thyroid cancer recurrence in well-differentiated thyroid cancer (Hsueh et al., 2013). Additionally, PTTG expression has been found to be increased in thyroid cancer (Boelaert et al., 2003) and has been described to cause genetic instability (Kim et al., 2005). Consequently, we aimed to determine if the simultaneous overexpression of PBF and PTTG caused genetic instability in murine thyroids. Ionising radiation induces DNA damage by causing single and double strand breaks within the DNA. Genetic instability results if the surviving cell cannot be repaired adequately. Additionally, ionising radiation is a recognised cause of thyroid cancer (Nikiforov, 2006, Nikiforov et al., 1996, Nikiforov et al., 1997, Richardson, 2009, Stsjazhko et al., 1995). Since PBF and PTTG overexpression in the
106
Genetic instability in transgenic murine thyroids
thyroid gland have not been shown to cause thyroid cancers, we aimed to observe the effects of ionising radiation with a background of PBF and PTTG overexpression independently and dependently in the thyroid gland to determine whether our genes of interest affected genomic stability of the thyroid cell following radiation.
107
Genetic instability in transgenic murine thyroids
4.2
MATERIAL AND METHODS
4.2.1 Primary murine thyroid cultures (PMTC) Primary murine thyroid culture (PMTC) was developed as a platform for experiments involving radiation. Microdissection on 6 week old murine thyroids is described in section 2.2.
4.2.2 Protein extraction and Western blot Protein extraction and western blot was performed as described in section 2.7 and 2.8. Primary antibodies used were rabbit anti-human PBF-8 polyclonal and anti-human PTTG (Invitrogen, UK) used in concentrations of 1:1000 and 2 µg/ml respectively. The secondary antibody used was monoclonal anti-β-actin clone AC-15 (Sigma-Aldrich, Poole, UK) in a concentration of 1:10,000.
4.2.3 Immunofluorescence PMTC was performed in 6-well plates for immunofluorescence. On Day 10, medium was removed from the wells and washed with phosphate buffer solution (PBS). Subsequently, the protocol for immunofluorescence was followed as described in section 2.9.
4.2.4 Radioiodine uptake assays Murine primary thyrocytes were grown in 12-well plates in 0.5 ml medium. Thyroid function was determined by the ability of the cell to actively absorb 0.05 μCi
125
125
I from the medium. A solution containing
I (Hartmann Analytic, Germany) and 10-9 NaI was added to each well and incubated for 2
hours as previously described (Eggo et al., 1996, Boelaert et al., 2007). Subsequently, the cell layer
108
Genetic instability in transgenic murine thyroids
was washed with HBSS to remove unincorporated iodide and the cells lysed in 2 % SDS protein lysis buffer. Incorporated radioactivity was estimated in a gamma counter. Relative iodide uptake was corrected for protein concentration measured by the Bradford assay.
4.2.5 DNA extraction DNA was extracted from murine thyroids and primary murine cultures using the same technique described for ear clippings in Chapter 2 (DNeasy 96 Blood & Tissue Kit; Qiagen).
109
Genetic instability in transgenic murine thyroids
4.2.6 Polymerase chain reaction (PCR) Genomic DNA was subjected to PCR in a Mastercyler® thermal cycler (Eppendorf, Hamburg, Germany). Each reaction comprised 50 ng DNA, 1 µl 10 mM dNTP, 2 µl 50 mM MgCl2, 3 µl of 200 nmol (CA)8RG (R = a 50:50 mix of the purines adenine and guanine; Y = a 50:50 mix of the pyrimidines cytosine and thymine) primers labelled with FAM fluorophore (Eurogentec), 1x TaqMaster PCR Enhancer (5Prime, Hamburg, Germany), 0.5x NH2 reaction buffer (Bioline), 2.5 U Biotaq™ DNA polymerase (Bioline) and nuclease-free water to make up a total volume of 50 µl. Initial denaturation of the DNA template took place at 95 °C for 5 minutes before holding at 74 °C. Subsequently, DNA polymerase and NH2 reaction buffer was added and the reaction subjected to 35 cycles of 94°C for 30 seconds, 50 °C for 30 seconds and 72 °C for 60s. The final extension step for 10 minutes took place at 72 °C.
4.2.7 Agarose gel electrophoresis PCR products were initially electrophoresed in 1 % Agarose (Bioline) gel in 1X TAE (Tris-AcetateEDTA) buffer (Eppendorf) before being visualized to optimize the PCR reaction for FISSR-PCR experiments.
4.2.8 Fluorescent inter-simple sequence repeat-PCR (FISSR-PCR) This technique involved a polymerase chain reaction (PCR) using primers labelled with a FAM fluorophore targetted to (CA)n microsatellites or simple sequence repeats within the genome. Traditionally, the PCR products or fragments were resolved on a PAGE gel and visualised by autoradiography. The genetic instability index developed by Basik et al and validated by others (Basik, Stoler et al 1997; Viswanathan, Sangiliyandi et al 2003) was determined by dividing the number of
110
Genetic instability in transgenic murine thyroids
altered fragments in the target tissue by the total number of bands in the corresponding tissue and multiplied by 100 to form a percentage. The modified technique of FISSR-PCR involved performing the fragment analysis in a DNA sequencer. PCR reaction products were diluted 20 times in nuclease-free water. 1 µl of diluted PCR reaction was added to 8.5 µl Hi-Di Formamide (Applied Biosystems) and 0.5 µl of Genescan™ 1200 LIZ® Size Standard (Applied Biosystems) and aliquoted into a 96-well plate. The whole mixture was denatured at 95ºC for 5 minutes. The sample was analysed using the ABI 3730 automated DNA sequencer (Applied Biosystems) with the microsatellite fragment analysis setting. The “fsa” file obtained was subsequently read using Genemapper v3.5 software (Applied Biosystems). The index of genetic instability was calculated as follows:GENETIC INSTABILITY =
No. of bands lost/gained + No. of different bands No. of expected bands
INDEX
111
x 100
Genetic instability in transgenic murine thyroids
4.3
RESULTS
4.3.1 Primary murine thyroid culture Primary murine thyroid culture (PMTC) was successfully perfected following several attempts at optimisation. Whole thyroid follicles were observed to be settling on the floor of a 12-well plate under light microscopy, seen in Figure 4-1. Subsequently, the follicles dispersed over several days, creating a sheet of cells with various clusters. Finally, on Day 10 when the cells were ready for experimentation, multiple clusters could be observed, shown in Figure 4-2. The cluster pattern tended to be variable, comprising between 60 – 70 % of the surface area of the well. Overall, the cultures were viable for up to 12 days but were used for experimentation on Day 10.
Thyroid follicle
100 µm
Figure 4-1 Primary murine thyroid culture under light microscopy depicting whole thyroid follicles immediately following processing at Day 0 (10x magnification).
112
Genetic instability in transgenic murine thyroids
Thyroid follicle
100 µm Figure 4-2 Light microscopy of primary murine thyroid culture (10x magnification) showing the formation of clusters in a 12-well plate on Day 10.
4.3.2 Immunofluorescence confirmed the presence of thyroid cells Having established the technique of PMTC, we next confirmed the purity of our primary cultures. The PBF-Tg murine model overexpressing human PBF tagged to haemagglutinin (PBF-HA) was used for immunofluorescence. DAPI, used to stain nuclei which fluoresced in blue, confirmed the presence of cells in culture. The presence of thyroid cells was confirmed by red fluorescence, detected using a primary HA antibody and a secondary antibody tagged with a red fluorescent dye (Alexa Fluor® 594; Life Technologies). The images obtained from the inverting microscope is shown in Figure 4-3. Interestingly, the presence of HA and hence, thyroid cells, appeared in clusters.
113
Genetic instability in transgenic murine thyroids
A
B
C
Figure 4-3 Immunofluorescence in PBF-HA primary murine thyroid culture (PMTC) at 10x magnification. The nucleus fluoresced in blue (DAPI) and the HA tag fluoresced in red. Image A depicts the pattern of cellular growth in PMTCs and image B confirmed the presence of PBF-HA thyroid cells. The composite image C depicts the growth of thyroid cells in clusters.
4.3.3 PMTCs overexpress PBF and/or PTTG Next, we examined the expression of our genes of interest in each of the PMTC genotypes; wild-type (WT; n=4), PBF (n=4), PTTG+/+ (n=4) and BI-Trans (n=4) shown in Figure 4-4 below. The Western blot confirms the overexpression of PBF in both PBF and BI-Trans PMTCs and PTTG in both PTTG+/+ and BI-Trans PMTCs.
Figure 4-4 Western blot of PMTCs probed with PTTG (top) and PBF (bottom) antibodies. WT, PBF, PTTG and BI-Trans refer to wild-type, PBF, PTTG and BI-Trans PMTCs (n=4 for each genotype). 114
Genetic instability in transgenic murine thyroids
4.3.4 Primary murine thyroid cultures were functional We next confirmed that the thyroid cells in PMTCs were functional. Previous publications by our group have shown that thyroid cells overexpressing PBF have reduced uptake of
125
I (Boelaert et al.,
2007, Smith et al., 2009, Read et al., 2011). A total of 5 WT and 6 PBF thyroids which were age- and sex-matched were used for this study. The average adjusted CPM (counts per minute) for WT was 25,210.4 ± 4,819.99 and for PBF 7,400.29 ± 1,680.58. We observed a 70% reduction of radioiodine uptake in PBF derived cultures compared to WT cultures at 10 days (p=0.0004), as shown in Figure 4-5. Taken together, these data confirmed that primary murine thyroid culture techniques were appropriate, consistent with previous data and functional in that significant reduction in 125I was demonstrated.
1.4
*** 1.2
RELATIVE
125I
UPTAKE
1 0.8 0.6 0.4 0.2 0
WT
GENOTYPE
PBF
Figure 4-5 Relative 125I uptake in WT and PBF primary murine thyroid cultures. The uptake of 125I is significantly repressed (p=0.0004) in PBF (n=6) compared to WT (n=5).
115
Genetic instability in transgenic murine thyroids
4.3.5 PCR optimisation The technique of FISSR-PCR used in measuring the index of genetic instability (GI) is dependent on PCR fragments obtained using the (CA)8RY primer which binds to microsatellite segments within the genome. The quantity of DNA obtained from PMTC is small and hence, it was important to ensure optimal PCR conditions to obtain maximal PCR products that can be subsequently resolved successfully. Several optimisation experiments were performed to determine the optimal conditions for the PCR reaction for FISSR. The aim of optimisation was to obtain clear fragments from the PCR reaction and to determine the length of products obtained. Initially, we resolved the PCR products on 1 % Agarose gel to compare various PCR reactions. The amount of MgCl2 and NH4SO4 buffer in each reaction was altered and the reaction which yielded the optimal number of clear fragments is shown in Figure 4.6 below. Interestingly, the PCR reaction yielded fragments mainly between 200 to 1000 base pairs.
Figure 4-6 DNA PCR products resolved on 1% Agarose gel, extracted from wild type (WT)(n=3) and PBF (n=3) primary murine thyroid cultures. The PCR conditions for this run were deemed optimal and used for subsequent FISSR experiments. PCR products ranged between 200 and 1000 base pairs. 116
Genetic instability in transgenic murine thyroids
4.3.6 Modified FISSR-PCR The technique of FISSR-PCR described by Kim et al involved the use of PAGE-gel and autoradiography (Kim et al., 2005). This method of resolving PCR products is cumbersome and would require a moderate amount of DNA which our primary murine thyroid cultures do not yield. The DNA sequencer (ABI 3730; Applied Biosystems) is a sensitive and efficient way of resolving PCR products. Hence, the FISSR-PCR technique was modified using the ABI 3730 DNA sequencing machine to resolve PCR products. Figure 4-7 shows PCR products being resolved within each capillary tube in the ABI 3730 automated DNA sequencer. The files obtained from the sequencer are read using Genemapper software which express data in numerical (shown in Figure 4-8) or graphical format (shown in Figure 4-9). Interpretation of results was challenging because of the huge volume of data present for each sample. A difference is defined as a change of more than 500 units in peak area and a band height difference of more than 3 fold. The estimated GI index from the same species within the same PCR run had a variation of less than 10 % using these criteria.
Figure 4-7 An example of PCR products resolved on ABI Genemapper sequencer. Arrows indicate individual sequencing capillaries. Yellow/Orange colour represent the GeneScan LIZ Size Standard (ThermoFisher Scientific). The blue colour arises from the FAM fluorophore attached to the primer used in the polymerase chain reaction (PCR).
117
Genetic instability in transgenic murine thyroids
Figure 4-8 Resolved PCR products in Genemapper (Applied Biosystems) represented in numerical form. The product size for each capillary run is identified. The quantity, defined by height and area, is also provided as illustrated above. The FAM marker refers to the fluorophore attached to the (CA)8RG primer used in the PCR.
Figure 4-9 Graphical representation of PCR products resolved by sequencing on Genemapper (Applied Biosystems) showing gross differences (encircled) between WT and PBF genotypes, used in calculating the index of genetic instability.
118
Genetic instability in transgenic murine thyroids
4.3.7 BI-Trans thyroids demonstrated significant genetic instability We compared the 4 different murine thyroid genotypes; wild type (WT), PBF, PTTG+/+ and BI-Trans (homozygous for both PBF and PTTG) using FISSR-PCR. A mean total of 154 unique fragments were found in the WT sample (n=5). PBF (n=5) did not have any unique fragments compared to WT resulting in a GI index of 0.00. PTTG+/+ (n=5) had an average of 3 unique fragments resulting in a genetic instability (GI) index of 1.95 ± 0.96 (p=0.78). Bi-trans (n=4) had a GI index of 3.73 ± 1.28 (p=0.01) as shown in Figure 4-10. These results suggest that in vivo, PBF or PTTG+/+ overexpression in the thyroid gland independently do not cause genetic instability. However, when PBF and PTTG are overexpressed together within the
GENETIC INSTABILITY INDEX
thyroid gland, the effect is one of increased genetic instability.
*
6
ns
5 4 3 2 1
0 WT
PBF
PTTG +/+
BI-Trans
GENOTYPE
Figure 4-10 GI index comparison according to genotype in vivo. PBF (n=5) and PTTG+/+ (n=5) thyroids appear to have no statistically significant genetic instability compared to WT (n=5). However, BI-Trans (n=4) thyroids had a GI index which was statistically significant compared to WT (p=0.01, denoted with *).
119
Genetic instability in transgenic murine thyroids
4.3.8 PBF, PTTG and BI-Trans PMTCs exhibited genetic instability We next quantified the index of genetic instability (GI) PMTCs for all four genotypes as a control for our experiments with ionising radiation. DNA was extracted from PMTCs on day 10. As shown in Figure 4-11 PBF (n=4), PTTG+/+ (n=5) and BI-Trans (n=5) PMTCs exhibited detectable genetic instability compared to wild type (WT) (n=5). The mean total number of unique fragments in this PCR run for WT was 128. The GI index for PBF, PTTG+/+ and BI-Trans was 18.75 ± 1.78 (p