UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Role of Nf1 signaling in regulating pigmentation in the mouse Deo, Mugdha Anand 2012

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

Item Metadata

Download

Media
24-ubc_2012_spring_deo_mugdha.pdf [ 2.54MB ]
Metadata
JSON: 24-1.0072689.json
JSON-LD: 24-1.0072689-ld.json
RDF/XML (Pretty): 24-1.0072689-rdf.xml
RDF/JSON: 24-1.0072689-rdf.json
Turtle: 24-1.0072689-turtle.txt
N-Triples: 24-1.0072689-rdf-ntriples.txt
Original Record: 24-1.0072689-source.json
Full Text
24-1.0072689-fulltext.txt
Citation
24-1.0072689.ris

Full Text

Role of Nf1 signaling in regulating pigmentation in the mouse  by  Mugdha Deo  B.Sc., University of Mumbai 2004 M.Sc., University of Pune, 2006  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Medical Genetics)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) April 2012  © Mugdha Deo, 2012  ABSTRACT Neurofibromatosis type 1 is caused by mutations in neurofibromin (NF1). Neurofibromas, which are Schwann cell based tumors, and skin hyper-pigmentation are characteristic of NF1 loss. Melanocytes differentiate from Schwann cell precursors (SCPs) during development, suggesting that there may be a mechanistic link between these NF1-related manifestations. In this thesis, we use Cre-LoxP technology to test the cell types that require Nf1 for normal pigmentation.  We discovered that an Nf1 targeted knockout (Nf1tm1Par) and an ENU-generated N1453K substitution in Nf1 (Nf1Dsk9) are associated with darker skin in mice. Nf1Dsk9/Nf1Dsk9 embryos exhibit increased numbers of melanoblasts at E10.5, and Nf1 -/- knock out in already committed melanoblasts causes dermal and epidermal hyper-pigmentation. In contrast, Nf1Dsk9/+ embryos exhibit an increase in the number of melanoblasts beginning at E12.5, and Nf1 +/- knockout in already committed melanoblasts has no effect on skin pigmentation. Nf1 haploinsufficiency in SCPs causes hyper-pigmentation of the dermis (but not the epidermis) in tamoxifen-inducible Plp1-CreER/+; Nf1tm1Par/+ mice when tamoxifen is administered at E11.5. Consistent with a lack of epidermal darkening in these mice, we found that cells expressing Plp1-CreER at E11.5 do not persist in the adult epidermis.  We found that Nf1 regulates skin pigmentation by an endothelin-dependent, as well as an independent mechanism. Nf1Dsk9/+;Ednrbs-l/Ednrbs-l  mice lack tail skin pigmentation, like  +/+;Ednrbs-l/Ednrbs-l mice. However, Nf1Dsk9/+;Ednrbs-l/Ednrbs-l mice exhibit an increased percentage of pigmented coat compared to +/+;Ednrbs-l/Ednrbs-l mice.  ii  Our data suggests that there are at least two mechanisms by which Nf1 regulates pigmentation in mice. Two copies of Nf1 are required to determine the appropriate number of melanoblasts that differentiate from bipotent melanoctye-SCPs. Subsequently, at least one copy of Nf1 is required to restrain the number of melanoblasts in the epidermis and dermis after they have committed to the melanocyte fate. These findings suggest that neurofibromin plays an important role in the specification of melanocytes within the glial lineage and may help design therapeutic options for treating NF1-related hyper-pigmentation in the future.  iii  PREFACE The research work presented in this thesis was originally identified and designed by Dr. Catherine Van Raamsdonk. The Dsk9 mouse mutants were recovered in a ENU-mutagenesis screen at National Research Centre for Environment and Health (GSF, Germany) by Dr. Helmut Fuchs and Dr. Martin Hrabé de Angelis. Positional cloning of the Dsk9 mutation was performed Dr. Catherine Van Raamsdonk, in the lab of Dr. Gregory Barsh. Experimental results presented in Figures 4.1c-e, 4.2, 4.3f,g 4.4 were performed by Dr. Van Raamsdonk. All other experimental results and data analyses presented in this thesis were performed by me. The Animal Care Committee provided ethical approval for the current research. Protocol numbers are A09-0898 and A08-0721. Our mouse lines were maintained by technicians at the Transgenic facility at the Centre for Molecular Medicine and Therapeutics, UBC. They also performed ear notching and plug checking and weaning of our mice. The Pathologists at the UBC Animal Care Centre and Dr. Nick Nation, University of Alberta for performed necropsy and histopathological analysis of our mice. Manuscript submitted for publishing the data presented in this thesis includes: Neurofibromin restrains melanocyte production within the glial lineage. Mugdha A Deo *, Jenny Huang, Helmut Fuchs, Martin Hrabé de Angelis, Catherine D Van Raamsdonk; Journal of Investigative Dermatology. (Chapters 1-7). Work performed for the current thesis was presented as the following abstracts: Effects of loss of Nf1 on pigmentation. Mugdha A Deo *, Helmut Fuchs, Martin Hrabé de Angelis, Gregory S Barsh, Valerie A White, Catherine D Van Raamsdonk. (2010) Pan American Pigment Cell Research Conference, Vancouver, Canada. (Chapters 1,3 and 4)  iv  Effects of loss of Nf1 on pigmentation. Mugdha A Deo *, Helmut Fuchs, Martin Hrabé de Angelis, Gregory S Barsh, Valerie A White, Catherine D Van Raamsdonk. (2010) NF Conference, Baltimore, MD, USA. (Chapters 1, and 3) Mugdha A Deo *, Helmut Fuchs, Martin Hrabé de Angelis, Gregory S Barsh, Valerie A White, Catherine D Van Raamsdonk. Role of Nf1 signaling in regulating pigmentation in mice. (2009) Frontiers in Basic Cancer Research Conference, Boston, MA, USA. (Chapters 1 and 6)  v  TABLE OF CONTENTS ABSTRACT ................................................................................................................................... ii PREFACE .......................................................................................................................................iv TABLE OF CONTENTS ...............................................................................................................vi LIST OF TABLES ........................................................................................................................... x LIST OF FIGURES ........................................................................................................................xi LIST OF ABBREVIATIONS ...................................................................................................... xii ACKNOWLEDGEMENTS..........................................................................................................xvi DEDICATION ............................................................................................................................ xvii CHAPTER 1 INTRODUCTION ..................................................................................................... 1 1.1 THE SKIN ................................................................................................................................. 1 1.2 SKIN PIGMENTATION ........................................................................................................... 3 1.2.1 Melanocyte location in the skin …………………………………………………………..3 1.2.2 Pigment synthesis ............................................................................................................... 3 1.2.3 Melanosome biogenesis ..................................................................................................... 5 1.2.4 Determinants of pigmentary variation in humans .............................................................. 7 1.3 MELANOCYTE ORIGIN AND DEVELOPMENT................................................................. 8 1.3.1 Origin of melanocytes in the neural crest .......................................................................... 8 1.3.2 Glial cells of the PNS ......................................................................................................... 9 1.3.3 Differentiation of melanoblasts from Schwann cell precursors ....................................... 12 1.3.4 Transcription factors important for melanocyte development ........................................ 13 1.3.5 Signaling pathways important for melanocyte development ........................................... 15 1.4 THE MOUSE AS A MODEL FOR STUDYING SKIN PIGMENTATION . ………………16 1.4.1 Dark skin mouse mutants ................................................................................................. 16 1.4.2 Epidermal dark skin mutants ........................................................................................... 17 vi  1.4.3 Dermal dark skin mutants ................................................................................................ 18 1.4.4 Dark skin 9 ....................................................................................................................... 19 1.5 NEUROFIBROMIN ………………………………………………………………………....25 1.6 MOUSE MODELS FOR STUDYING EFFECTS OF LOSS OF Nf1 .................................... 26 1.6.1 Cre/loxP technology......................................................................................................... 27 1.6.2 Targeted/knockout alleles of Nf1 ..................................................................................... 30 1.7 NEUROFIBROMATOSIS 1 AND SKIN PIGMENTATION ................................................ 32 1.7.1 Effects of loss of Nf1 in humans...................................................................................... 32 1.7.2 Pigmentary manifestations associated with NF1 ............................................................. 35 1.7.2.1 Café-au-lait macules (CALMs) ................................................................................. 35 1.7.2.2 Axillary/intertriginous freckling................................................................................ 36 1.7.2.3 Generalized hyper-pigmentation ............................................................................... 36 1.7.2.4 Pigmented plexiform neurofibromas ......................................................................... 37 1.7.2.5 No association of neurofibromatosis 1 with melanoma ............................................ 37 1.8 PROJECT SUMMARY AND RESEARCH QUESTIONS .................................................... 37 CHAPTER 2 MATERIALS AND METHODS ............................................................................ 40 2.1 MOUSE HUSBANDRY ......................................................................................................... 40 2.2 GENOTYPING........................................................................................................................ 40 2.3 POSITIONAL CLONING OF Nf1Dsk9 .................................................................................... 41 2.4 S100a4-Cre PRODUCTION ................................................................................................... 42 2.5 β-GALACTOSIDASE (LacZ) DETECTION ......................................................................... 43 2.6 DERMAL-EPIDERMAL SEPARATION AND PIXEL INTENSITY OF TAIL SKIN ........ 43 2.7 EPIDERMAL SCALE DARKNESS VERSUS MELANOCYTE DENSITY ........................ 43 2.8 TAMOXIFEN INDUCTION OF CreER ACTIVITY ............................................................. 44 2.9 TISSUE EMBEDDING ........................................................................................................... 44 vii  2.10 HEMATOXYLIN AND EOSIN STAINING ....................................................................... 44 2.11 IMMUNOFLUORESCENCE ............................................................................................... 44 2.12 FACS ANALYSIS ................................................................................................................ 45 2.13 PERCENT OF COAT PIGMENTATION ............................................................................ 45 2.14 SEQUENCING ...................................................................................................................... 45 2.15 STATISTICS ......................................................................................................................... 46 CHAPTER 3: EFFECT OF LOSS OF Nf1 IN THE HOMOZYGOTE ........................................ 47 3.1 INTRODUCTION ................................................................................................................... 47 3.2 RESULTS ................................................................................................................................ 49 3.2.1 Nf1Dsk9/Nf1Dsk9 embryos exhibit increased melanoblast numbers during development ... 49 3.2.2 Skin hyper-pigmentation in Mitf-Cre/+; Nf1tm1Par/Nf1tm1Par mice ................................... 50 3.2.3 FACS sorting EYFP-positive cells in Mitf-Cre/+; Nf1tm1Par/Nf1tm1Par embryos .............. 52 3.2.4 No change in melanosome structure in Mitf-Cre/+; Nf1tm1Par/Nf1tm1Par dermal melanocytes............................................................................................................................... 53 3.2.5 No obvious role for Nf1 in the formation of melanocytic lesions .................................... 53 3.3 DISCUSSION .......................................................................................................................... 62 CHAPTER 4: EFFECT OF Nf1 HAPLOINSUFFICIENCY ON DERMAL AND EPIDERMAL PIGMENTATION .................................................................................................................... 65 4.1 INTRODUCTION ................................................................................................................... 65 4.2 RESULTS ................................................................................................................................ 67 4.2.1 Nf1Dsk9/+ mice exhibit an increase in the number of melanoblasts beginning at E12.5 .. 67 4.2.2 No skin hyper-pigmentation in Mitf-Cre/+; Nf1tm1Par/+ mice .......................................... 68 4.2.3 No skin hyper-pigmentation in S100a4-cre/+; Nf1tm1Par/+ mice .................................... 69 4.2.4 No skin hyper-pigmentation in Vav1-cre/+; Nf1tm1Par/+ mice ........................................ 69 4.2.5 Skin hyper-pigmentation in Plp1-creER/+; Nf1tm1Par/+ mice ......................................... 70 4.2.6 Fate mapping of melanoblasts expressing Plp1-CreER at E11.5 .................................... 71 viii  4.3 DISCUSSION .......................................................................................................................... 79 CHAPTER 5: EPISTATIC INTERACTIONS BETWEEN Nf1 AND Ednrb .............................. 81 5.1 INTRODUCTION ................................................................................................................... 81 5.2 RESULTS ................................................................................................................................ 83 5.2.1 Determination of the Ednrbs-l deletion breakpoints ......................................................... 83 5.2.2 Nf1Dsk9/+ melanocytes require Ednrb for survival in the tail dermis ............................... 84 5.3 DISCUSSION .......................................................................................................................... 88 CHAPTER 6: GENERAL DISCUSSION ..................................................................................... 89 6.1 SUMMARY OF RESULTS AND CONCLUSIONS.............................................................. 89 6.1.1 Evidence supporting the Nf1N1453K mutation as causative in Dsk9 mice ......................... 89 6.1.2 Evidence suggesting release of Ras inhibition causes darker skin in Nf1 mutants .......... 90 6.1.3 Nf1 haploinsufficiency affects melanocytes of the glial lineage..................................... 93 6.1.4 Comparison of effect of loss of neurofibromin on pigmentation in mice and humans ... 96 6.1.5 Comparison of different alleles of Nf1 normalized to wildtype ...................................... 98 6.1.6 Treatment options for pigmentary manifestations in NF1 patients................................ 100 6.2 FUTURE DIRECTIONS ....................................................................................................... 101 6.2.1 Determine whether Nf1 haploinsufficiency alters Schwann cell development.............. 102 6.2.2 Determine whether SCP-derived melanoblasts in the head have a reduced requirement for Ednrb ................................................................................................................................. 102 6.2.3 Determine when during development Nf1 -/- loss can cause skin hyper-pigmentation 102 6.3 CONCLUSIONS ................................................................................................................... 103 References.................................................................................................................................... 104  ix  LIST OF TABLES Table 1. 1 Effect of loss of Nf1 on various cell types. ................................................................ 28 Table 1. 2 Diagnostic criteria for Neurofibromatosis 1 ............................................................... 33 Table 6. 1 Comparison of the effects of neurofibromin mutations on pigmentation in humans and mice………………………………………………………………………………………………97 Table 6. 2 The percent increase in dermal and epidermal average pixel intensity over wildtype for various Nf1 mutants……………………………………………………………………….…99  x  LIST OF FIGURES Figure 1. 1 Structure of the skin .................................................................................................... 2 Figure 1. 2 Schematic representation of normal sites of pigmentation in human skin .................. 4 Figure 1. 3 Pigment synthesis pathway ......................................................................................... 6 Figure 1. 4 Mechanisms melanocyte development...................................................................... 10 Figure 1. 5 Development of melanocytes and glia from the neural crest .................................... 11 Figure 1. 6 Transcriptional regulation of Mitf by FoxD3............................................................ 14 Figure 1. 7 Positional cloning of the Dsk9 mutation. .................................................................. 20 Figure 1. 8 Mutations in Nf1 cause darker skin in mice.............................................................. 22 Figure 1. 9 Targeted alleles of Nf1 .............................................................................................. 31 Figure 3.1 Increased number of LacZ-positive cells in E10.5 Nf1Dsk9/Nf1Dsk9;Dct-LacZ/+ embryos…………………………………………………………………………………………..55 Figure  3.2  Quantification  of  Tuj1-positive,  LacZ-positive  cells  in  Nf1Dsk9/Nf1Dsk9  embryos…………………………………………………………………………………………..56 Figure 3.3 Quantification of Tuj1-negative, LacZ-positive cells in Nf1Dsk9/Nf1Dsk9 embryos…...57 Figure 3.4 Skin hyper-pigmentation in Mitf-Cre/+; Nf1tm1Par/Nf1tm1Par mice…………………...58 Figure 3.5 Histological analysis of melanin accumulation in Mitf-Cre/+; Nf1tm1Par/Nf1tm1Par animals…………………………………………………………………………………………...60 Figure 3.6 Trunk skin hyper-pigmentation in Mitf-Cre/+; Nf1tm1Par/Nf1Dsk9 mice……………...61 Figure 4.1: Increase in melanoblasts in E12.5 Nf1Dsk9/+ embryos……………………………....73 Figure 4.2 Increase in epidermal melanocyte numbers in Nf1Dsk9/+ epidermal scales…………..74 Figure 4.3: Effect of Nf1tm1par +/- knockout using various Cre lines……………………………75 Figure 4.4 Fate mapping of cells expressing Plp1-CreER at E11.5……………………………..77 Figure 4.5: S100a4-Cre is widely expressed in the skin………………………………………...78 Figure 5. 1 The Ednrbs-l deletion is 97.63 kb, encompassing the entire Ednrb gene……………85 Figure 5.2 Genetic interactions between Ednrb and Nf1……………………………………………..86 Figure 6. 1 Signaling pathways causing dermal and epidermal hyper-pigmentation…………...92 Figure 6.2 Haploinsufficiency of Nf1 causes an expansion of Schwann cell precursor derived melanoblasts during embryogenesis……………………………………………...…………….95  xi  LIST OF ABBREVIATIONS α Melanocyte stimulating hormone (αMSH) Adenylyl cyclase (AC) Adenosine monophosphate (AMP) Adenosine triphosphate (ATP) Adrenocorticotropic hormone (ACTH) Agouti signaling protein (ASIP) Base pair (bp) Beta galactosidase (LacZ) Bone morphogenic protein (BMP) Brain lipid binding protein (Blbp) Bromo-chloro-indolyl-galactopyranoside (X-gal) Café-au-lait macules (CALMs) Calcium chloride (CaCl2) Cyclization recombinase (Cre) Cyclic AMP (cAMP) Day post coitum (d.p.c.) Deoxyribonucleic acid (DNA) Deoxy ribonucleic acid triphosphates (dNTPs) 5,6-dihydroxyindole (DHI) 5,6-dihydroxyindole-2-carboxylic acid (DHICA) Diaminopyridine imidazole (DAPI) Dopachrome tautomerase (Dct) Embryonic developmental day (E) xii  Endothelin 3 (Edn3) Endothelin receptor (Ednrb) Enhanced yellow fluorescence protein (EYFP) Epidermal growth factor receptor (EGFR) Epithelial mesenchymal transition (EMT) Exrtacellular related kinase (Erk) Forkhead box D3 (FoxD3) Giant congenital melanocytic nevi (GCMN) Glial fibrillary acidic protein (GFAP) Guanosine triphosphate (GTP) GTPase activating protein (GAP) GAP related domain (GRD) Hepatocyte growth factor (HGF) Homeobox 1 (Hmx1) Human Neurofibromatosis 1 gene (NF1) Insulin growth factor (IGF) Jun N-terminal kinases (JNK) kit ligand (kitl [SCF/MGF/Steel]) kit receptor (c-kit) Locus of crossing [x] over P (loxP) Loss of heterozygosity (LoH) Magnesium chloride (MgCl2) Mitogen activated protein kinase (MAPK) Mast cell growth factor (MGF [kitl/SCF/Steel]) xiii  Micromoles (µM) Microphthalmia associated transcription factor gene (Mitf) Milimoles (mM) Moles (M) Mouse endothelin receptor B gene (Ednrb) Mouse endothelin receptor B protein (Ednrb) Mouse Neurofibromatosis 1 gene (Nf1) Nanograms (ng) Nano moles (nM) Neural crest cells (NCCs) Neuregulin 1(Nrg1) Neurofibromatosis 1 disease (NF1) Optimal cutting temperature (O.C.T.) Paired box gene 3 (Pax3) Peripheral nervous system (PNS) Phosphate buffered saline (PBS) Phosphatidylcholine (PtdCho) Phosphatidylinositol (PtdIN) Pico moles (pM) Piebald lethal (Ednrbs-l) Platelet derived growth factor (PDGF) Plexiform neurofibromas (PNF) Polymerase chain reaction (PCR) Post natal day (P) xiv  Proopiomelanocortin (POMC) Proteolipid protein (Plp1) Saccharomyces cerevisiae phosphatidyl transfer protein (Sec14p) Schwann cell (SC) Schwann cell precursors (SCPs) Solute carrier family 24 (sodium/potassium/calcium exchanger) member 4 (SLC24A4) Solute carrier family 24 (sodium/potassium/calcium exchanger) member 5 (SLC24A5) SRY (Sex determining region on Y chromosome)-related transcription factor 10 (Sox10) Standard error of mean (SEM) Stem cell factor (SCF [kitl/MGF/Steel]) Transgenic (tg) Two pore segment channel 2 protein (TPCN2) Tyrosinase (Tyr) Tyrosinase related protein 1 (Trp1) Tyrosinase related protein 2 (Trp2/Dct) Ultra violet radiations (UV radiations) Wildtype (WT)  xv  ACKNOWLEDGEMENTS I would like to extend my heartfelt gratitude to my supervisor, Dr. Catherine Van Raamsdonk for providing me the opportunity to work on one of the most exciting and a beautifully designed project for my Ph.D. I would like to specially thank her for challenging my research and critical thinking potentials at every step during the course of learning and helping me push my boundaries. I would like to thank my supervisory committee Dr. Jan Friedman, Dr. Rob Kay and Dr. Kelly McNagny for guiding me through my project by giving excellent suggestions during my committee meetings and on my Ph.D. research proposal and for helping me critically analyze my data. I would specially like to thank Dr. Kelly McNagny for providing me with the Vav-Cre/+ transgenic mice. I am extremely grateful to the members of Dr. Jane Roskams lab for sharing their immunofluorescence protocols, antibody aliquots and training me with immunofluorescence staining technique. I would also like to thank Dr. Louis Lefebvre for being on my committee, although for a short period and showing interest in my research everytime we interacted. I am grateful to Megan Jones and Aaron Bogutz from the Lefebvre lab for training me with techniques including immunofluorescence, fluorescence microscopy, qPCR and mouse embryonic fibroblast derivation. Last but not the least, I would like to thank my husband for his strong desire to see me have my own identity and for standing by me. I am also very much grateful to my parents and my in-laws for their love, understanding and encouragement during my Ph.D.  xvi  DEDICATION  To my husband and my family for their love and faith in me.  xvii  CHAPTER 1 INTRODUCTION  1.1 THE SKIN The skin is the largest organ of our body, and it acts as the main protective barrier against external environmental damages including ultraviolet radiation, injury and pathogens. The skin consists of three main layers: the epidermis, the dermis and the hypodermis (Figure 1.1).  The epidermis consists of keratinocytes, melanocytes and Langerhans cells. Keratinocytes in the basal layer proliferate to replace cells in the upper layers, which are lost through shedding. Hair follicles are composed of keratinocytes that produce hair keratins. Hair follicles also maintain tissue homeostasis, and contribute to regeneration and repair after injury (Fuchs 2007). Melanocytes, pigment producing cells, are present in the basal layer of the epidermis and in hair follicles. Melanocytes package melanin into organelles called melanosomes, which are distributed to neighboring keratinocytes. The specific characteristics and distribution of melanosomes in keratinocytes is responsible for natural variation in skin and hair color. Langerhans cells recognize and process pathogens and act as antigen presenting cells in the epidermis.  The dermis forms the connective tissue of the skin. Fibroblasts in the dermis produce collagen and fibronectin, components of the extracellular matrix. Melanocytes are also present in the dermis in some species. The dermis harbors mast cells, which play a role in immune modulation in response to pathogens. It is vascular and contains nerves, Schwann cells, sweat glands and sebaceous glands. 1  EPIDERMIS  MELANOCYTES  FIBROBLASTS  KERATINOCYTES  MAST CELLS  ENDOTHELIAL CELLS NERVES  DERMIS  BLOOD VESSELS  SWEAT GLANDS  SEBACEOUS GLANDS  ADIPOSE TISSUE  LANGERHANS CELLS  HYPODERMIS  SCHWANN CELLS DERMAL / SKIN DERIVED PRECURSORS FOLLICULAR MELANOCYTES MELANOCYTE STEM CELLS  Figure 1. 1 Structure of the skin The skin consists of three main layers: the epidermis, the dermis and the hypodermis. The epidermis consists of a basal layer of keratinocytes interspersed with melanocytes. The hair follicles are also composed of keratinocytes. The dermis contains mast cells, blood vessels (endothelial cells), Schwann cells, nerves, sebaceous glands and sweat glands, and sometimes melanocytes. The hypodermis contains adipose tissue, blood vessels and nerves.  2  The third layer of the skin is hypodermis, which consists of adipose tissue, sweat glands and blood vessels. It provides protective cushioning and maintains thermoregulation (Tortora and Derrickson 2009).  1.2 SKIN PIGMENTATION 1.2.1 Melanocyte location in the skin The pigmentary system in mammals consists of melanocytes. The normal location of melanocytes in the skin varies among different species. In mice, melanocytes are located in the dermis, inter-follicular epidermis, and hair follicles of the glabrous skin (on the ears, nose, footpad, and tail). In the rest of the body, melanocytes are located in hair follicles and are occasionally observed in the dermis, but they are absent from the inter-follicular epidermis. In humans, melanocytes are located in the inter-follicular epidermis of the entire body, as well as in hair follicles. Melanocytes are not normally found in the human dermis, but there are several types of dermal melanocytoses, such as blue nevi, nevus of Ota or nevus of Ito (Gleason et al. 2008) (Figure 1.2).  1.2.2 Pigment synthesis Melanocytes synthesize two types of melanin, eumelanin (brown/black) and pheomelanin (red/yellow). Melanin synthesis is regulated by the melanocortin signaling pathway. The melanocortin 1 receptor (MC1R) is a G-protein coupled receptor activated by αMSH (α melanocyte stimulating hormone) and ACTH (adenocorticotropic hormone), downstream products of the precursor, proopiomelanocortin (POMC). POMC production is stimulated by UV radiation. Activation of MC1R by αMSH results in an increase in adenylyl cyclase activity 3  Human  Mouse  Glabrous skin  Non-glabrous (Hair bearing) skin  Epidermis Dermis  Figure 1. 2 Schematic representation of normal sites of pigmentation in human skin Shaded regions in the boxes indicate presence of pigment. Glabrous skin in mice consists of skin on the tail, foot pad, ears and nose and is homologous to the palm skin in humans. The hair bearing skin in mice consists of the trunk skin under the coat and is homologous to the trunk skin and scalp skin in humans.  4  through Gαs. This in turn results in an increase in the levels of cAMP, which, through downstream pathways, stimulates eumelanin production. ASIP (Agouti signaling protein) is an antagonist of MC1R, leading to decreased levels of cAMP and synthesis of pheomelanin.  Melanin synthesis occurs through a series of biochemical reactions (Figure 1.3). The enzyme tyrosinase (tyr) catalyzes oxidation of amino acid L-tyrosine to dihydroxyphenylalanine (LDOPA), which is further oxidized to glutathione/cysteine/thioredoxine,  L-Dopaquinone.  L-Dopaquinone  In the presence of thiols such as  is converted to cysteinyl-DOPA and  ultimately results in the synthesis of pheomelanin. Thus, intracellular levels of the amino acid cysteine play an important role in determining the type of pigment that will be synthesized, pheomelanin or eumelanin. In the absence of thiols, L-Dopaquinone undergoes cyclization to yield leuocoDOPA and DOPAchrome as a default mechanism. Tyrosinase related protein 2 (Tyrp2/ DCT [Dopachrome tautomerase]) catalyzes tautomerization of DOPAchrome to a more stable 5,6-dihydroxyindole,2-carboxylic acid (DHICA) intermediate, or DOPAchrome can be spontaneously decarboxylated to 5,6-dihydroxyindole (DHI). Tyrosinase related protein 1 (TRP1) and tyrosinase catalyze the synthesis of DHICAeumelanin or DHIeumelanin. A basal level of tyrosinase activity is sufficient for pheomelanogenesis while eumelanogenesis requires greater tyrosinase activity. The melanosomal pH, which is regulated by cAMP and αMSH, helps to determine the level of tyrosinase activity (Scherer and Kumar 2010).  1.2.3 Melanosome biogenesis Melanin is synthesized in melanosomes, specialized organelles produced from endosomes. During the production process, melanosome ultrastructure is characterized by four sequential  5  Figure 1. 3 Pigment synthesis pathway Melanin synthesis occurs inside melanosomes through a series of biochemical reactions.  6  stages. Melanosome formation begins with the pinching off of a vesicle from the endoplasmic reticulum. Longitudinal striatures form within the vesicle, created by an internal matrix of proteins. Unpigmented striatures within a vesicle mark stage I melanosomes. During stage II, tyrosinase (TYR), tyrosinase related protein 1 (TRP1), and tyrosinase related protein 2 (DCT) are sorted into the melanosomes, along with other proteins. The first deposition of pigment on the matrix begins. In stage III and stage IV melanosomes, pigment deposition continues, such that in stage IV melanosomes, the internal matrix is completely obscured from view (Kushimoto et al. 2001). Stage IV melanosomes also show enhanced mobility within the melanoctye. Stage IV melanosomes are taken up by keratinocytes via endocytosis or phagocytosis of melanocyte filopodia.  The shape of the melanosome depends on the type of pigment it carries.  Eumelanosomes are large and ellipsoidal, while pheomelanosomes are small and spherical (Scherer and Kumar 2010).  1.2.4 Determinants of pigmentary variation in humans In humans, a number of factors determine the variation observed in skin pigmentation among different population groups. These variations in skin color are most often due to difference in the type of pigment synthesized by the melanocytes rather than a difference in the total number of melanocytes in the skin (Scherer and Kumar 2010).  The MC1R gene is known to be highly polymorphic, and loss-of-function mutations in MC1R are associated with red hair, fair skin and freckling through inefficient activation of cAMP and subsequently decreased levels of tyrosinase activity (Scherer and Kumar 2010). Variants of the TYR and TRP1 gene are associated with differences in skin, hair and eye color, although not all the associations identified have been confirmed. Variants of several genes encoding 7  melanosomal transport proteins show associations with eye and hair color. These include OCA2, which encodes the P protein that transports small molecules into melanosomes, SLC24A4 and SLC24A5, which are solute carrier proteins, and TPCN2, a two pore segment channel 2 protein (Rebbeck et al. 2002; Sturm 2006; Han et al. 2008; Sulem et al. 2008; Cook et al. 2009).  1.3 MELANOCYTE ORIGIN AND DEVELOPMENT 1.3.1 Origin of melanocytes in the neural crest Melanocytes arise from the neural crest. During gastrulation, the neural crest arises at the edge of the neural plate on the border between neural and non-neural ectoderm (White and Zon 2008). Initial induction of the neural crest cells (NCCs) is regulated by bone morphogenic protein (BMP) signaling. Neural crest cells become migratory following epithelial-to-mesenchymal transition (EMT). Snail/slug, two genes that are expressed by the NCCs, repress cell adhesion molecules such as E-cadherin, thus leading to EMT (White and Zon 2008). At least some of the NCCs are multipotent (Woodhoo and Sommer 2008), and, depending on their anatomic location and the local environment, these cells become fate restricted over time to generate neural cells of sensory, autonomic and enteric nervous system or non-neural cells, including smooth muscle cells in the outflow tract of the heart, melanocytes, connective tissue, craniofacial bones and cartilage (Motohashi et al. 2007; White and Zon 2008; Woodhoo and Sommer 2008).  NCCs migration proceeds along two different migratory pathways. Initially, a wave of migration proceeds along the dorsoventral pathway between the neural tube and sclerotome of the somites. These NCCs give rise to neurons and glia of the peripheral nervous system. A second wave of migration proceeds between the dermamyotome of the somites and the ectodermal epithelium, 8  and these crest cells give rise to melanocytes in the skin (Dupin and Le Douarin 2003) (Figure 1.4, 1.5).  In addition to the NCC migratory pathway described in the previous section, melanocytes also arise at a later stage from bipotential Schwann cell-melanoblast precursors lining peripheral nerves (Dupin et al. 2000; Rizvi et al. 2002; Dupin and Le Douarin 2003; Ignatius et al. 2008; Thomas and Erickson 2009) (Figure 1.4, 1.5). The next section will describe the different types of glial cells, such as Schwann cells, and factors known to regulate their development.  1.3.2 Glial cells of the PNS There are a number of glial cells in the PNS, including satellite cells that form a covering around sympathetic and parasympathetic ganglia, myelinating and non-myelinating Schwann cells, olfactory ensheathing cells that ensheath the olfactory nerve axons, teloglia that cover the axon terminals at skeletal neuromuscular junctions and enteric glia that are present in the gut. Glial cells of the PNS are derived from migrating NCCs by a series of signaling events. As the NCCs begin to delaminate from the neural tube, the migrating cells rest in the migration staging area. This is followed by a wave of migration ventrally between the neural tube and the sclerotome of the somites. At this point, the cells can give rise to either neurons or glia of the PNS. Glial cell development in the PNS may occur through suppression of neuronal development or due to down regulation of molecules that normally suppress gliogenesis, rather than as a default  9  Figure 1. 4 Mechanisms melanocyte development (1) The first wave of NCCs migrate along the dorsoventral pathway that generates neurons and glia of the PNS. A second wave of NCCs (2) migrates along the dorsolateral pathway and gives rise to melanoblasts. The boundary cap cells of the dorsal root ganglia (3a) and SCPs associated with the peripheral nerves (3b) also give rise to melanoblasts in the skin.  10  Figure 1. 5 Development of melanocytes and glia from the neural crest Melanoblasts can differentiate from three known progenitors cell types: Schwann cell precursors, boundary cap cells and neural crest cells. The genes expressed by these cells during various stages of development are given.  11  Glial cell development in the PNS may occur through suppression of neuronal development or due to down regulation of molecules that normally suppress gliogenesis, rather than as a default mechanism (Jessen and Mirsky 2005). Molecules known to be important for gliogenesis include SOX10, Neuregulin 1, Notch, and Hmx1.  Sox10 (SRY-related transcription factor 10) is expressed by all migrating NCCs. Sox10 mutant mice lack satellite glia and SCPs without any effect on neuronal development. This suggests that Sox10 is specifically required for glial cell development (Britsch et al. 2001). Nrg1 (Neuregulin 1) is essential for the migration of NCCs and promotes Schwann cell proliferation, migration and myelination (Newbern and Birchmeier 2010). Nrg1 mutant mice exhibit hypoplasia of the sympathetic ganglia (Britsch et al. 1998) and a loss or severe depletion of SCPs (Morris et al. 1999; Wolpowitz et al. 2000). Neurogenins also stimulate the Notch signaling pathway, reducing neurogenesis and increasing the number of Schwann cells (Morrison et al. 2000). Similarly, expression of the homeobox transcription factor, Hmx1, leads to complete loss of neurogenesis while increasing the number of glial cells and melanoblasts (Adameyko et al. 2009; Ernfors 2010).  1.3.3 Differentiation of melanoblasts from Schwann cell precursors Lineage-tracing experiments in chicken and mouse embryos showed that Sox10+ and Plp1+ (Proteolipid protein 1) cells migrating in close association with the ventral nerves are SCPs that eventually develop into Schwann cells. It was observed that some SCPs at the nerve terminals migrate away from the nerve, express melanocyte specific transcription factors and eventually develop into melanocytes. It is possible that the switch between Schwann cell versus melanocyte 12  cell fate occurs due to reduced Neuregulin/ErbB3 signaling as a consequence of loss of contact of some SCP’s with the nerves (Adameyko et al. 2009; Ernfors 2010). ErbB3-/- embryos exhibit a 178% increase in the number of melanoblasts around the distal ends of the dorsal spinal nerves at embryonic day (E)12, despite a decrease in SCP’s along those nerves. In cultures of embryonic dorsal root ganglia, Nrg1 treatment reduced the number of melanocytes formed. Interestingly, this effect was ameliorated by treatment with IGF1 (Insulin-like growth factor 1) and PDGF (Platelet derived growth factor 1), which are also produced by developing nerves, suggesting that a balance of these factors might be required for normal melanocyte development.  1.3.4 Transcription factors important for melanocyte development Several transcription factors play a role in the process of determining melanocyte cell fate. Amongst these, Mitf (Microphthalmia associated transcription factor), FoxD3 (Forkhead box D3), and Pax3 (Paired box gene 3) play important roles.  Mitf expression marks the specification of immature melanocytes, called melanoblasts, from precursor cells (Levy et al. 2006). Neural crest cells express Pax3 and FoxD3 prior to melanoblast specification (Figure 1.6). FoxD3 represses transcription of Mitf by inhibiting the binding of Pax3 to the Mitf promoter (Ignatius et al. 2008; Thomas and Erickson 2009). Neural crest cells on the dorsal-lateral migration pathway down-regulate FoxD3 in order to express Mitf. Mitf then upregulates the expression of several other downstream melanogenic, cell cycle progression and differentiation genes.  13  Figure 1. 6 Transcriptional regulation of Mitf by FoxD3 FoxD3 is expressed in neural crest cells, where it prevents Pax3 from interacting with the Mitf promoter. As NCC’s begin to migrate, some cells down-regulate FoxD3, initiate Mitf expression, and differentiate into melanoblasts, which migrate on the dorsal-lateral pathway. (Ignatius et al. 2008; Thomas and Erickson 2009).  14  1.3.5 Signaling pathways important for melanocyte development Melanoblasts require different growth factors as they migrate from the neural crest to the skin during development.  Several mouse mutants exhibit changes in pigmentation due to the  mutation or over-expression of signaling molecules. These include Kitl/c-Kit, Edn3/Ednrb and HGF/c-Met.  The receptor tyrosine kinase c-kit and its ligand, kitl, (also known as Steel or MGF) play intermittent roles in melanoblast survival. The soluble form of the kitl ligand is required for the initial survival and dispersion of melanoblasts on the lateral migratory pathway (Wehrle-Haller and Weston 1995). This is followed by a kit-independent stage before the melanoblasts enter the epidermis, beginning at E12.5. Once in the epidermis, melanoblasts again require c-kit for survival and/or proliferation. Interestingly, melanocyte stem cells in the hair follicle bulge region are c-kit-independent, while the differentiated melanocytes in hair follicles are c-kit-dependent (Yoshida et al. 1996; Kawaguchi et al. 2008).  Endothelin 3 (Edn3) is the ligand for the G protein coupled receptor, Endothelin receptor B (Ednrb). Edn3 is required for early migration of melanoblasts from the neural crest (Baynash et al. 1994; Yoshida et al. 1996). Similarly, Ednrb is required between E10.5-E12.5 for survival/differentiation of melanoblasts (Shin et al. 1999). Ednrb knockout after E12.5 does not affect coat pigmentation. Over-expression of Edn3 under a keratinocyte specific promoter causes dermal, but not epidermal, melanocytosis of the skin, the opposite effect of transgenic kitl expression. .  15  Similarly, overexpression of Hepatocyte growth factor (HGF) by keratinocytes causes dermal, but not epidermal, melanocytosis. HGF activates c-Met, a tyrosine kinase receptor. C-Met is not required for melanoblast development prior to E14 and its exact role in melanoblast development remains to be determined.  1.4 THE MOUSE AS A MODEL FOR STUDYING SKIN PIGMENTATION The mouse serves as an excellent model for studying the mechanisms regulating skin pigmentation in humans. In both humans and mice, melanin is produced by melanocytes (Quevedo 1969). Mutations affecting the survival, proliferation and pigment producing ability of melanocytes cause similar defects, for example piebaldism (decreased melanocyte survival and/or proliferation during development) (Rice et al. 2000), melanoma (Pollock et al. 2003) and Hermansky Pudlak syndrome (hypo-pigmentation of skin due to decreased melanocyte numbers and immaturity of melanosomes) (Nguyen and Wei 2007).  1.4.1 Dark skin mouse mutants In the last decade, several large scale mutagenesis screens have produced a number of new mouse mutants. Using N-ethyl-N-nitrosourea (ENU) as a chemical mutagen, a group in Germany led by Martin Hrabe de Angelis screened ~32,000 F1 mice on a C3HeB/FeJ inbred genetic background, and recovered mutants at a frequency of 1.3% (Hrabe de Angelis et al. 2000b). Twelve mice were discovered with darker skin on the tail and footpads, and were named Dark skin (Dsk) 1-12. Histologically, the Dsk mutants can be grouped by whether the epidermis or the dermis is darker (Fitch et al. 2003). In most Dsk’s, the coat color is unaffected. Molecular identification of some of the genes mutated in the Dsk mice has already helped to identify 16  mechanisms causing melanoma (Van Raamsdonk et al. 2009), hyper-keratosis (Fitch et al. 2003; McGowan et al. 2006) and Diamond-Blackfan anemia (Mason and Bessler 2008).  1.4.2 Epidermal dark skin mutants The Dsk mutants that darken the epidermis specifically are Dsk 2,3,4,5,6 and 8. These mutants are characterized by an onset of hyper-pigmentation around 2-3 weeks of age. Of these, Dsk2 and Dsk5 cause thickening of the epidermis. Hyper-pigmentation in these two mutants is due to an increase in the number of epidermal melanocytes as a secondary effect of kertatinocyte hyperproliferation. The Dsk2 mutation is a threonine to proline substitution in the rod2B domain of Keratin 2e. The same mutation was identified in a patient with ichthyosis bullosa of Siemens, which causes blistering at birth and flexural hyper keratosis. The Dsk5 mutation is a leucine to glycine substitution in the epidermal growth factor receptor, which may alter catalytic interactions or ATP binding ability of the EGFR (Epidermal growth factor receptor) protein. These two mutants will be used as tools to study how melanocyte and keratinocyte proliferation are co-regulated in the skin.  Dsk3 and Dsk4 exhibit epidermal hyper-pigmentation due to an increase in the number of melanoblasts during development. Dsk3 and Dsk4 are caused by mutations in 40S ribosomal protein subunits, Rsp19 and Rsp20 (McGowan et al. 2008). An imbalance in ribosomal protein components causes accumulation of p53 in keratinocytes, which increases the expression of kitl, leading to melanocyte hyper-proliferation through paracrine signaling. Mutations in RPS19 cause Diamond-Blackfan syndrome in humans, which is characterized by low birth weight, complete red cell aplasia, and isolated macrocytosis. Rps19Dsk3/+ mice exhibit low birth weight, reduced red blood cell counts and bone marrow progenitor cell aplasia. Dsk3 and Dsk4 illustrate 17  how the study of pigmentation can sometimes help unravel the etiopathogenesis of other disorders, when different cells types share pathways with melanocytes.  1.4.3 Dermal dark skin mutants The dermal dark skin mutations, Dsk1, Dsk7, and Dsk10, cause an increase in the number of melanoblasts in the dermis beginning at E10.5, and they share identical phenotypes. Dsk1 and Dsk10 are two different substitutions in the heterotrimeric G protein alpha subunit Gnaq, while Dsk7 is a substitution in the closely related family member, Gna11. These three mutations cause alpha subunit hyperactivity. They alter either the intrinsic GTPase activity of the respective alpha subunits, or the exchange of GTP for GDP.  Motivated by the identification of Gnaq and Gna11 in Dsk mice, Van Raamsdonk et al investigated the mutational status of GNAQ and GNA11 in human melanocytic lesions. They found somatic, constitutive active mutations in GNAQ in 55% of blue nevi (benign dermal melanocytic neoplasias), 45% of primary uveal melanomas (melanomas of the uveal tract in the eye) and 22% of metastatic uveal melanomas. Similarly, GNA11 was mutated in 7% of blue nevi, 32% of primary uveal melanomas and 57% of metastatic uveal melanomas (Van Raamsdonk et al. 2010). Interestingly, no mutations were found in GNAQ or GNA11 in melanomas located in the epidermis, showing a strong conservation in the role of these two genes between mice and humans. Melanocytes transduced with constitutively active forms of GNAQ and GNA11 exhibit an increased in mitogen activated protein kinase (MAPK) signaling and are tumorigenic in nude mice. Thus, Dsk1, Dsk7, and Dsk10 led to the identification of the  18  first oncogenes discovered in uveal melanoma, opening possible avenues for new therapeutic options.  1.4.4 Dark skin 9 Dsk9 is unique in that it is the only dark skin mutant that darkens both the dermis and the epidermis. The footpad, tail and ear skin are darkened, with no change in coat color. (Figure 1.7 a-c) In an outcross-backcross mapping strategy with C57BL/6J, SSLP markers were used to link Dsk9 to chromosome 11 (Fitch et al. 2003). Subsequently, an outcross-intercross mapping strategy was employed when it was determined that homozygosity of the dark skin-linked region of chromosome 11 bearing Dsk9 is homozygous lethal (46 Dsk9/Dsk9 expected, 0 observed).  From these mapping crosses, Dsk9 was localized to a 4.3 megabase physical interval on chromosome 11, which contains 83 candidate genes, including neurofibromin (Nf1) (Figure 1.7d). Mutations in human NF1 cause neurofibromatosis type 1, which is characterized by several forms of skin hyper-pigmentation. All exons of Nf1 were sequenced in DNA extracted from E11.5 Dsk9/Dsk9 embryos and a single nucleotide substitution in exon 33 (ENSMUST00000071325, NCBI m37), was discovered that was not present in the strain of origin (Figure 1.8a). The nucleotide change predicts that asparagine 1453 in the C-terminal end of 7c of the GTPase activating protein related domain (GRD) is replaced with lysine (ENSMUSP00000071289, NCBI m37) (Figure 1.8 c). This particular amino acid is conserved through Drosophila (Figure 1.8 b) and mutational analysis screening has shown that it is  19  +/+  Dsk9/+  a  d  b  c  e  g  f +/+  Dsk9/+  Dsk9/Dsk9  E12.5  Figure 1. 7 Positional cloning of the Dsk9 mutation. Dsk9/+ (right) mutants exhibit hyper-pigmentation on the hind foot (a), tail (b) and ear (c) compared to wild type littermates (left). (d) Genetic and physical map of the Dsk9 interval on mouse chromosome 11. The number of recombinants and number of mapping progeny are recorded for each of the SSLP markers. The location of the mutation and the SSLP markers on the current Ensembl mouse genomic assembly (NCBI m37) are indicated below. Representative 20  picture of wild type (e), Nf1Dsk9/+ (f) and Nf1Dsk9/Dsk9 embryos at E12.5 (g). Nf1Dsk9/Dsk9 embryos show enlargement of the heart, peripheral edema, small hypo-pigmented eyes and hemorrhage.  21  Cystein and serine rich domain  a  GAP-related domain Neurofibromin  150 amino acids  Sec14p domain Pleckstrin-homology domain  C3H (wild type) Dsk9  …GCC AAT CAT… …GCC AAA CAT…  b  c Mus musculus Nf1-Dsk9 Mus musculus Nf1 Homo sapiens NF1 Gallus gallus NF1_CHICK Danio rerio NF1 Drosophilla melanogaster Nf1  LQSIAKHVLF LQSIANHVLF LQSIANHVLF LQSIANHVLF LQSIANHVLF LQSIANHVEF N1453  d  200  e  dermis  h  g  f  epidermis  i  175 150 125 100 +/+  Nf1Dsk9/+  Nf1tm1Par/+  ACTB-Cre/+;  +/+  Nf1Dsk9/+  Nf1tm1Par/+  ACTB-Cre/+;  n=6  n=4  n=3  Nf1tm1Par/+  n=6  n=4  n=3  Nf1tm1Par/+ n=5  n=5  j  +/+  Nf1tm1Tyj/+  Figure 1. 8 Mutations in Nf1 cause darker skin in mice. (a) Graphic representation showing position and sequence of the Dsk9 mutation within neurofibromin. (b) The amino acid sequence flanking the Dsk9 substition is shown, aligned with the orthologous sequences in humans (Homo Sapiens), chicken (Galus galus), zebra fish (Danio 22  rerio) and fruit fly (Drosophilla melanogaster). (c) The position of the asparagine analogous to the one mutated in Dsk9 (Asn1453) is highlighted on the crystal structure of the GRD of human neurofibromin (Scheffzek et al. 1998). Tail dermis (d) and epidermis (f) samples from 3 week old wildtype and Nf1Dsk9/+ littermates representing at least 3 mice of each genotype. Tail dermis (e) and epidermis (g) samples from wildtype and ACTB-Cre/+;Nf1tm1Par/+ littermates representing at least 3 mice of each genotype. The average pixel intensity of photographed tail dermis (h) and epidermis (i) of the respective genotypes is represented graphically. (j) Nf1tm1Tyj/+ animals (Mean pixel intensity =152.8 ± 8.3) show significant hyper-pigmentation of the tail dermis compared to respective wildtype animals (Mean pixel intensity = 123.4 ± 2.1). Statistical analysis: p=0.000165, +/+ v/s Nf1Dsk9/+ dermis; p= 0.0139, Nf1tm1Par/+ v/s ACTBCre/+;Nf1tm1Par/+ dermis; p=0.0279, +/+ v/s Nf1Dsk9/+ epidermis; p= 0.0032, Nf1tm1Par/+ v/s ACTB-Cre/+;Nf1tm1Par/+ epidermis; p=0.000182, Nf1tm1Tyj/+v/s wildtype dermis.  Note: Epidermal hyper-pigmentation phenotype of Nf1Dsk9/+ was not identified at the time of analysis of the Nf1tm1Tyj/+ tail skin and hence this analysis the epidermal pigmentation analysis of the Nf1tm1Tyj/+ was missed.  23  important for interactions with Ras (Morcos et al. 1996). Missense mutations in the human NF1 GRD have been previously reported in neurofibromatosis type 1 patients (Scheffzek et al. 1998). The most frequently altered human residue in the GRD is also located on 7c (Scheffzek et al. 1998).  E12.5 Nf1Dsk9/Nf1Dsk9 embryos show enlargement of the heart, hemorrhage, small hypopigmented eyes and edema (Figure 1.7 e-g), phenotypes previously described for Nf1tm1Tyj/Nf1tm1Tyj embryos, which have a targeted knockout of Nf1 (Jacks et al. 1994). We obtained tails or mice of two targeted Nf1 alleles, Nf1tm1Tyj and Nf1tm1Par to determine if they too have darker skin. To quantify the darkness, we group photographed split tail skin samples and used ImageJ to determine the average pixel intensity of each sample. Nf1Dsk9/+ mice, Nf1tm1Tyj/+ mice (data not shown), and ACTB-Cre/+;Nf1tm1Par/+* mice all had significantly darker skin (Figure 1.8d-i). This data shows that mutations in Nf1 cause skin hyper-pigmentation in mice, as they do in humans (De Schepper et al. 2005). The main goal of this thesis is to clarify the role of neurofibromin in melanocytes and determine how its loss leads to skin hyper-pigmentation.  *Footnote: The Nf1tm1Par allele contains loxP sites flanking exons 40 and 41 (Figure 1.9). Upon expression of Cre recombinase, the loxP sites are recombined, leading to the deletion of exons 40 and 41. In this experiment, the Cre expressing line used was ACTB-cre, which expresses Cre in every cell type. Cre-Lox technology is described in greater detail in section 1.6.1.  24  1.5 NEUROFIBROMIN The Nf1 gene encodes the neurofibromin protein which consists of 2818 amino acids. The neurofibromin protein is highly conserved beween mice and humans, with 96% identity at the protein level (Trovo-Marqui and Tajara 2006). The currently identified domains of neurofibromin include the GTPase activating protein related domain [GRD] (Ballester et al. 1990), a cysteine/serine-rich domain (CSRD) (Tokuo et al. 2001), a sec14p domain, and a pleckstrin homology-like domain (Bonneau et al. 2004; D'Angelo et al. 2006).  The GRD of neurofibromin shows homology with IRA1 and IRA2 of Saccharomyces Cerevesia and GAP1 of Drosophila (Ballester et al. 1990), and is able to down regulate Ras p21 activity in vivo (Xu et al. 1990). The switch 1 and switch 2 regions of Ras bind to an arginine finger motif on neurofibromin, stimulating the hydrolysis of GTP bound to Ras (Bos et al. 2007). This interaction converts Ras from an active to an inactive state. Nf1-deficient cells show increased survival and/or proliferation (Kim et al. 1995; Yang et al. 2006) which can be reversed by pharmacological inhibition of Ras or the addition of the GRD domain. (Kim et al. 1997; Hiatt et al. 2001; Yang et al. 2006). This suggests that loss of neurofibromin leads to hyper-activation of Ras, which signals through the MAPK (mitogen activated protein kinase) and PI3K (phosphatidylinositol 3-phosphate) pathways (Lau et al. 2000; Trovo-Marqui and Tajara 2006).  The C-terminus of neurofibromin has been shown to increase activity of adenylyl cyclase (AC) in Drosophila and in mouse astrocytes and neurons. Adenylyl cyclase is an enzyme that converts adenosine triphosphate (ATP) into cyclic adenosyl monophosphate (cAMP). The C-terminus of neurofibromin is required for immediate memory formation in Drosophila (Ho et al. 2007). In 25  mice, the complete loss of neurofibromin in Blbp-expressing neurons leads to disrupted cortical development, which can be rescued by rolipram treatment, which raises cAMP levels (Hegedus et al. 2007). NG,NG-dimethylarginine dimethylaminohydrolase (DDAH) binds to neurofibromin at the CSRD and the C-terminus. DDAH mediates phosphorylation of neurofibromin via cAMPdependent protein kinase (Tokuo et al. 2001).  Two domains implicate lipids in the regulation of neurofibromin, the Sec14p and pleckstrin homology like domains. Saccharomyces cerevisiae phosphatidylinositol (PtdIN) transfer protein (Sec14p) plays a role in lipid metabolism, membrane trafficking and secretory function of the Golgi by mediating exchange of PtdIN for phosphatidylcholine (PtdCho) across the lipid bilayer of cell membranes. Pleckstrin homology domains play a role in signal transduction by binding to phosphoinositides (Zhang et al. 2009). The GTPase activating activity of neurofibromin can be  inhibited  by binding of  lipids  such  as  arachidonic  acid,  phosphatidate  and  phosphatidylinositol-4,5-bisphosphate (Bollag and McCormick 1991).  Neurofibromin is uniformly expressed throughout embryonic days 10 and 12 in rat embryos. Its expression is enhanced in neural tissues by E16 and decreases in non-neural tissues, including the skin between E16 and P6 (Daston and Ratner 1992).  1.6 MOUSE MODELS FOR STUDYING EFFECTS OF LOSS OF Nf1 Since Nf1 is homozygous lethal at E12.5 due to cardiac defects, alternative methods must be employed to study the effects of homozygous Nf1 loss after this time point. Several studies have  26  employed Cre/loxP technology in order to bring about tissue-specific deletion of Nf1 in mice (Table 1.1).  1.6.1 Cre/loxP technology The Cre recombinase (cyclization recombinase) is derived from bacteriphage P1. It recognizes a 34bp target sequence in the bacteriophage genome called loxP (locus of crossing over [x] over P) and brings about recombination between two loxP sites. The loxP sequence consists of two 13bp palindromic sequences that flank an 8bp asymmetric linker sequence. The linker assigns directionality to the loxP site. Recombination between two loxP sites oriented in the same direction brings about deletion of the region that they flank; if the two loxP sites are oppositely oriented, then a recombination between the two sites results in an inversion of the sandwiched region, whereas presence of two loxP sites on two different chromosomes results in translocation (Feil 2007). Therefore, to use this technology, two mouse strains are needed: a strain expressing Cre in the cell type of interest and a strain in which the region to be deleted is flanked by loxP sites.  First, I will describe the production of Cre expressing mice. The cell type of interest is determined and a tissue-specific promoter is obtained by screening a genomic DNA library with a probe for a gene known to be expressed specifically in the cell type of interest. The promoter is subcloned and ligated upstream of a Cre expression cassette. The construct is then injected into mouse pronuclei where it can insert itself into the genome in a random location. The integration process is successful in only a subset of mice and the number of copies of that  27  Nf1 KNOCKOUT MOUSE REFERENCE MODEL  Nf1  flox/flox  ; Syn1-Cre  (Zhu, Y., et al. 2001, Genes Dev 15(7): 859876)  Cre EXPRESSION TIME  CELLS EXPRESSING Cre  ORIGIN  E12.5  Differentiated neurons  NEURAL CREST  Nf1  flox/flox  ; Dhh-Cre  1. 2. 3.  E12.5  Nf1flox/- ; Krox20-Cre  EFFECT OF LOSS OF Nf1-/-  Schwann cells  1. 2. 3. 4.  (Zhu, Y., et al. 2002, Science 296(5569): 920922) (Yang, F.C. et al., 2008, Cell 135(3): 437-448)  (Wu, J. et al., 2008 E12.5 Cancer Cell 13(2): 105116)  Developing cells  E14.5  Astrocytes  (Bajenaru, M.L., et al., 2002, Mol Cell Biol 22(14): 5100-5113)  Nf1flox/flox; Tek-Cre  (Gitler, A.D. et al., 2003, E11.5 Nat Genet 33(1): 75-79)  Endothelial cells  Nf1flox/flox ; Prx1-Cre  ( Kossler, N. et al., 2011, E13.5 Hum Mol Genet 20(14):2697-709)  Undifferentiated mesenchymal cells  Abnormal development of cerebral cortex. Extensive astrogliosis No tumors Plexiform neurofibromas Functionally WT No neurofibromas Nf1+/- bone marrow (mast cells) required for plexiform neurofibroma development.  Plexiform neurofibromas  glial  Nf1flox/flox ; HGFAP-Cre Nf1flox/- ; HGFAP-Cre  COMPLETE  1. 2. 3. 4.  Developmental defects Astrogliosis Enlarged optic nerves Optic pathway gliomas  Cardiac defects MESENCHYME Skeletal abnormalities  Table 1. 1 Effect of loss of Nf1 on various cell types.  28  transgene that integrate varies with each event. Sometimes, unexpected cell types will also express the Cre molecule, confounding the use of the line.  A slightly different Cre molecule that is sometimes used is CreER, which is a fusion protein between Cre and the estrogen receptor. In the absence of the drug tamoxifen, the CreER fusion protein localizes to the cytoplasm. When tamoxifen is administered, it binds to CreER and the complex translocates to the nucleus, where it can bind to and recombine loxP sites (Feil 2007; Feil et al. 2009). Tamoxifen can be administered through drinking water, breast milk, skin contact, or by intraperitoneal injection. With CreER technology, gene deletion can be initiated at a particular time point during development or adulthood, whenever tamoxifen is first administered.  LoxP sites are engineered in the genome in the following way. First, a genomic DNA library is screened for the region where the loxP sites will be introduced. Using restriction enzymes and ligases, a construct is made containing ~4 kb of genomic DNA, with loxP sites introduced at the desired sites. This construct is then electroporated into mouse embryonic stem cells, where homologous recombination replaces the endogenous sequence with the targeted version. ES cells must be screened to identify the ones that underwent homologous recombination correctly, since this is an uncommon event. The positive ES cells are expanded and single cells are injected into 3 day old blastocysts to produce chimeras. Some of the chimeras will contain germline cells that arose from the injected ES cell and these mice are able to pass the targeted allele to their offspring. Crossing a Cre expressing mouse with these mice produces progeny with certain exons deleted in the desired type of cells.  29  Typically, Cre/loxP technology does not induce recombination in 100% of the cells that would be expected to express the Cre transgene. Therefore, it is important to assess the rate of recombination at a marker locus in each Cre line that is made. Reporter lines to assess Cre efficiency use a “stop” sequence to prevent reporter gene transcription until Cre removes the stop cassette, which is flanked by loxP sites. Downstream of the stop cassette is a minigene encoding a marker, either beta-galactosidase or a fluorescent protein (Srinivas et al. 2001). The Rosa26 locus is one site for the placement of such markers, because this locus is expressed in every cell type, when it is not being blocked by a Stop cassette. For example, a mouse line expressing Cre in glial cells could be crossed to Rosa26-loxP flanked “Stop”-EYFP (enhanced yellow fluorescence protein), and all glial cells which underwent Cre mediated recombination would fluoresce with EYFP.  1.6.2 Targeted/knockout alleles of Nf1 Several targeted alleles of Nf1 have been generated and studied in mice (Brannan et al. 1994; Jacks et al. 1994; Zhu et al. 2001) (Figure 1.9). The Nf1tm1Fcr allele was generated by insertion of neo cassette in exon 40 (based on transcript ENSMUST00000071325, NCBI m37). No protein was detected in the Nf1tm1Fcr/Nf1tm1Fcr embryonic extracts. Reduced protein levels were present in the Nf1tm1Fcr/+ embryonic extracts compared to those of the wild type embryos. Similarly protein levels were reduced in the brain of Nf1tm1Fcr/+ adults (Brannan et al. 1994).  30  Figure 1. 9 Targeted alleles of Nf1 The Nf1tm1Fcr and Nf1tm1Tyj alleles possess neo cassette inserted into exon 40 and result in constitutive inactivation of Nf1. The Nf1tm1Par allele possesses loxP sites flanking exon 40 and 41 and Cre-mediated deletion of the floxed region results in conditional inactivation of Nf1.  31  Jacks et al. generated another Nf1 knockout allele, Nf1tm1Tyj, by insertion of a neo cassette into exon 40 of Nf1. During translation of the Nf1tm1Tyj transcript, splicing occurs from the splice acceptor site of exon 39 into splice donor site of the neo cassette and subsequently from the  splice acceptor site of the neo cassette into the splice donor site of exon 41, thus replacing exon 40 with the neo cassette. No stop codon is produced by this insertion and thus a stable mRNA is transcribed, which could produce 10kDa larger neurofibromin protein than the wild type neurofibromin. However, two different antibodies used in western blot analysis did not detect neurofibromin protein in the Nf1tm1Tyj/Nf1tm1Tyj homozygotes (Jacks et al. 1994).  The Nf1tm1Par allele is a loxP flanked (or “floxed”) allele of the Nf1 gene in which exons 40 and 41 are flanked by loxP sites. Recombination of the Nf1tm1Par allele using a Cre line expressed in the germline (Krox-20-cre) creates a null allele. The Krox20-cre/+; Nf1tm1Par/Nf1tm1Fcr embryos show characteristic features of Nf1tm1Tyj/Nf1tm1Tyj embryos including enlargement of the heart, peripheral edema, small hypo-pigmented eyes, hemorrhage and embryonic lethality at E12.5, thus suggesting that the Nf1tm1Par allele is a phenocopy of the Nf1-null allele (Zhu et al. 2001).  1.7 NEUROFIBROMATOSIS 1 AND SKIN PIGMENTATION 1.7.1 Effects of loss of Nf1 in humans Neurofibromatosis 1 (NF1) is an autosomal dominantly inherited neurocutaneous disorder that affects about 1 in 3500 individuals. Neurofibromatosis 1 shows 100% penetrance, but a variable expressivity. The clinical features can be broadly classified as cutaneous and non-cutaneous manifestations. Cutaneous manifestations can be further classified as cutaneous tumors or 32  Diagnostic criteria for neurofibromatosis 1 Six or more café-au-lair macules >5mm or >15mm in greatest diameter in prepubertal or post pubertal individuals, respectively. Two or more neurofibromas of any type (cutaneous or subcutaneous) or one plexiform neurofibroma. Freckling in axillary or other intertriginous regions such as neck-folds or under the breasts in women. Optic glioma Two or more Lisch nodules A distinctive osseous lesion such as sphenoid dysplasia or tibial pseudoarthrosis A first degree relative meeting the diagnostic criteria for Neurofibromatosis 1  Table 1. 2 Diagnostic criteria for Neurofibromatosis 1 Individuals who meet two or more of the diagnostic criteria for Neurofibromatosis 1 have the disease. These criteria are met by 97% of the cases (NIH 1988).  33  pigmentary manifestations. The cutaneous tumors include cutaneous neurofibromas and plexiform neurofibromas (PNF). Neurofibromas are composed of a heterogeneous population of cells, including Schwann cells, mast cells, fibroblasts, blood vessels. In addition, plexiform neurofibromas are occasionally pigmented by melanocytes. Cutaneous neurofibromas arise from peripheral nerves and develop in about 95% of the NF1 patients beginning in late childhood and continuing throughout adulthood. Cutaneous neurofibromas may continue to grow slowly over many years, but rarely (if ever) undergo malignant transformation. Plexiform neurofibromas on the other hand can affect any nerve of the body and are observed in about 50% of Neurofibromatosis 1 patients (Mautner et al. 2008). These tumors can cause severe lifelong disfigurement, disability and have the potential to transform into malignant peripheral nerve sheath tumors (MPNSTs) (Staser et al. 2010).  The non-cutaneous manifestations of Neurofibromatosis 1 may also include bone dysplasias, neurological/psychological abnormalities, ophthalmologic manifestations including optic gliomas and iris hamartomas (i.e. Lisch nodules), cardiovascular abnormalities including hypertension and vasculopathies, endocrine dysfunction, gastrointestinal tumors or constipation, and juvenile myelomonocytic leukemias. (Boyd et al. 2009).  The diagnostic criteria for NF1 are summarized in Table 1.2. Many of the clinical manifestations are exhibited at 1 year of age and about 97% of the patients meet all the diagnostic criteria by 8 years of age (Boyd et al. 2009). The effect of NF1 heterozygosity is pleiotropic and suggests the importance of NF1 in regulating a number of cell types such as melanocytes, glial cells, nerves, endothelial cells, osteoclasts and osteoblasts, hematopoietic cells and mesenchymal cells. Loss of heterozygosity is observed in Schwann cells leading to neurofibroma development (Zhu et al. 34  2002; Le and Parada 2007), in chromaffin cells causing pheochromocytomas (Benn et al. 2000; Cichowski and Jacks 2001), in melanocytes causing CALMs (De Schepper et al. 2008), in skin derived precursors (SKPs) causing cutaneous neurofibromas (Le et al. 2009), in myeloid cells causing myelomonocytic leukemias (Le et al. 2004) and in glial cells causing astrocytomas (Bajenaru et al. 2002) .  The pigmentary manifestations are in the form of generalized hyper-pigmentation of the skin, benign hyper-pigmentary spots called café-au-lait macules (CALMs), and axillary and other intertriginous freckling. Having six or more CALMs (>5mm pre-puberty and >15mm postpuberty) and axillary freckling, is strongly suggestive for neurofibromatosis. However, CALMs may be found in normal individuals and in other disorders (Alper and Holmes 1983; De Schepper et al. 2005), so additional criteria have to be considered for the diagnosis of NF1, as listed in Table 1.2.  1.7.2 Pigmentary manifestations associated with NF1 The association of skin hyper-pigmentation with neurofibromatosis 1 is strong. However, much remains to be understood about the mechanisms by which neurofibromin loss causes different types of skin hyper-pigmentation.  1.7.2.1 Café-au-lait macules (CALMs) CALMs are benign epidermal lesions that are typically ovoid in shape, have a sharp boundary, are uniformly colored and can range from tan to dark brown in color. CALMs are stochastically distributed along the body and appear in ~95% of neurofibromatosis type 1 cases prior to 1 year  35  of age. Melanocytes cultured from CALM’s exhibit biallelic mutation of NF1 (Maertens et al. 2007; De Schepper et al. 2008).  1.7.2.2 Axillary/intertriginous freckling Freckles are light brown colored, oval-shaped macules that can be normally seen in the sunexposed epidermis of fair-skinned individuals. However, neurofibromatosis type 1 patients have freckles in unusual locations such as the axillae, inguinal regions, skin surface below the breast in women and on the neck folds (De Schepper et al. 2005). Importantly, the freckles can be observed at birth. Amer et al observed that the freckled skin in neurofibromatosis type 1 patients exhibit an increase in the number of epidermal melanocytes, giant macromelanosomes, elongation of rete ridges of epidermal melanocytes along with inflammatory cell infiltration and presence of melanophages in the dermis (Amer et al. 2001). Because the NF1-specific freckles localize to skin folds, it might be that the melanocytes in this region are responding to the elevated temperature, increased sweat secretion and friction that occurs in these locations.  1.7.2.3 Generalized hyper-pigmentation It has been noted anecdotally that neurofibromatosis type 1 patients have slightly darker skin than their family members. To address this, Maertens et al studied individuals with mosaic neurofibromatosis type 1, in which there are clearly demarcated dark and light regions, with CALMs appearing on top of the darker areas. Melanocytes cultured from the darker areas exhibited the loss of one copy of NF1, while melanocytes cultured from the CALMs had biallelic mutation (Maertens et al. 2007). This suggests that haplo-insufficiency of NF1 causes subtle, generalized epidermal hyper-pigmentation.  36  1.7.2.4 Pigmented plexiform neurofibromas The skin overlying plexiform neurofibromas (PNF) in people with neurofibromatosis may be hyper-pigmented and hairy, a finding that is sometimes confused with giant congenital melanocytic nevi (GCMN) (Anderson and Robertson 1979; Mahe et al. 2001; Schaffer et al. 2007). Often these PNF’s form on the spinal nerves and nerve roots (Riccardi 1992). The major distinction between PNF and GCMN is that the former have Schwann cells as the major component and are associated with neurofibromatosis 1 while the latter have melanocytes as the major component (Fetsch et al. 2000) and have no reported association with neurofibromatosis 1.  1.7.2.5 No association of neurofibromatosis 1 with melanoma Neurofibromin a negative regulator of RAS, which is frequently activated in melanoma (GraySchopfer et al. 2007). Over the years, there have been several reported cases of melanoma in neurofibromatosis type 1 patients, however, the incidence of melanoma is not significantly higher in neurofibromatosis type 1 patients (Gutzmer et al. 2000; Foster et al. 2003; Guillot et al. 2004; Rubben et al. 2006). To date, there is only one case report describing the development of a melanoma within a CALM (Perkinson 1957). Therefore, NF1 loss in melanocytes does not appear to be sufficient for melanoma formation.  1.8 PROJECT SUMMARY AND RESEARCH QUESTIONS Loss of NF1 has a pleiotropic effect and affects a number of cells including melanocytes. Patients with neurofibromatosis type 1 develop cutaneous and subcutaneous neurofibromas as well as hyper-pigmentary manifestations in the form of café-au-lait macules, intertriginous or 37  axillary freckling and iris hamartomas. Several studies that investigated the role of neurofibromin in melanocytes observed that:  (a) Hyper-pigmented skin of neurofibromatosis 1 patients shows haploinsufficiency of NF1 in melanocytes (Maertens et al. 2007);  (b) biallelic loss of NF1 is observed in melanocytes in CALM’s (De Schepper et al. 2008);  (c) haploinsufficient fibroblasts underlying CALMs secrete elevated levels of KITL and HGF (Okazaki et al. 2003);  (d) Skin hyper-pigmentation is observed overlying plexiform neurofibromas (Schaffer et al. 2007);  (e) Nf1 heterozygosity partially rescues coat color spotting and mast cell hypoplasia in KitW41/W41 mice (Ingram et al. 2000);  (f) Nf1-/- mice show an increase in the number of melanoblasts during embryogenesis (WehrleHaller et al. 2001) and  (g) biallelic loss of Nf1 in early glial cell precursors results in hyper-pigmentation around plexiform neurofibromas in mice (Wu et al. 2008).  38  We still do not understand exactly how or when the loss of Nf1 increases melanocyte numbers. The overall goal of my thesis is to understand the mechanisms by which loss of Nf1 affects melanocyte development and causes skin hyper-pigmentation, using mouse models.  Objective 1: To investigate the effects of homozygous loss of Nf1 in melanocytes using Cre-loxP technology.  Objective 2: To determine whether fibroblasts, mast cells, Schwann cell precursors and/or melanocytes are sensitive to Nf1 haploinsufficiency.  Objective 3: To determine when during embryogenesis Nf1 haploinsufficiency disrupts melanocyte development. .  Objective 4: To determine whether Nf1Dsk9 depends on Ednrb to cause dark skin.  39  CHAPTER 2 MATERIALS AND METHODS  2.1 MOUSE HUSBANDRY Experiments were preformed under the approval of the CACC at UBC. Nf1tm1Tyj /+ tail samples were provided by Karen Cichowski(Jacks et al. 1994). Tg(Mitf-cre)7114Gsb, Tg(Vav1cre)A2Kio/J, and Tg(Dct-LacZ)A12Jkn mice were provided by Greg Barsh(Mackenzie et al. 1997; Alizadeh et al. 2008) and Kelly NcNagny(de Boer et al. 2003). Tg(Plp1-cre/ERT)3Pop, Ednrbs-l, Gt(ROSA)26Sortm1(EYFP)Cos/J , and Gt(Rosa)26Sortm1Sor/J mice were obtained from the Jackson Laboratories(Lane 1966; Soriano 1999; Srinivas et al. 2001; Doerflinger et al. 2003). Nf1tm1Par mice were obtained from the NCI mouse repository(Zhu et al. 2001). Nf1Dsk9, S100A4cre, Dct-LacZ, Mitf-cre, and Ednrbs-l mice were backcrossed and/or maintained on a C3HeB/FeJ background.  Plp1-cre,  Nf1tm1Tyj,  Tg(Vav1-cre)A2Kio/J,  Gt(ROSA)26Sortm1(EYFP)Cos/J,  Gt(Rosa)26Sortm1Sor/J , and Nf1tm1Par mice were obtained on an inbred C57BL6/J background. For embryo collection and tamoxifen injection, noon on the day the copulatory plug was found was marked as day 0.5 of gestation.  2.2 GENOTYPING Genomic DNA from ear skin biopsies or embryonic membranes was extracted using Qiagen DNeasy Blood and Tissue Kit (Invitrogen). 50 ng of DNA was used for each PCR reaction including 1X PCR buffer containing 1.5mM MgCl2, 0.5μM of each primer, 1 unit of HotStar Taq polymerase (Qiagen) and 0.5mM dNTP’s. 2.5mM of extra MgCl2 was added for genotyping Nf1Dsk9,  Nf1tm1Par,  Ednrbs-l  and  Vav1-Cre.  Primers:  Nf1Dsk9  (forward  5’-  GCCAGTAGAAATATCAATGGAAAA-3’, reverse 5’-GGGTGGGGAATCACATACAG-3’, 40  followed by digestion with AflIII), Ednrbs-l (S-L forward 5’-CCCTACCCTTCTCACCCACT-3’, S-L  reverse  5’-  GCATTACCTCAGGCTCCAC-3’,  WT  forward  5’-  CATTTGTCCCAGGGATAGGA-3’, WT reverse 5’-CAGCTTTTGCTAATGGCTGA-3’), and Cre  (forward  5’-GCGGTCTGGCAGTAAAAACTA-3’,  reverse  5’-  GTGAAACAGCATTGCTGTCAC-3’), When Mitf-cre was used in combination, Mitf-cre was genotyped using a phenotypic marker(Alizadeh et al. 2008), while the following PCR reactions genotyped the other Cre’s: S100a4-cre (forward 5’-CAACAGATGGCTGGCAACTA-3’, reverse  5’-CCACCAGCCAGCTATCAACT-3’),  Vav1-cre  (forward  5’-  CCATGGCACCCAAGAAGAAG -3’, reverse 5’- GCTTAGTTTTCCTGCAGCGG -3’). Nf1tm1Par and Dct-LacZ mice were genotyped as previously described(Zhu et al. 2001; Takemoto et al. 2006). The breakpoints of the Ednrbs-l deletion were identified using a series of sequential PCR reactions to map the un-amplifiable region in DNA from Ednrbs-l/Ednrbs-l mice, span the un-amplifiable region with a set of primers (S-L, see above), and sequence the resulting PCR product.  2.3 POSITIONAL CLONING OF Nf1Dsk9 Nf1Dsk9 mice were recovered from a dominant, N-ethyl-N-nitrosourea mutagenesis screen of inbred C3HeB/FeJ mice at the Institute of Experimental Genetics, Neuherberg, Germany(Hrabe de Angelis et al. 2000a; Fitch et al. 2003). To map the Dsk mutations(Fitch et al. 2003), Dsk/+ mice were outcrossed to C57BL6/J. In the Dsk9 outcross, mice with the darkest tail and foot skin were selected from the progeny and backcrossed to C3HeB/FeJ. Of the backcross progeny, 8 dark skinned mice and 8 light skinned mice were selected for DNA extraction. The DNA from the dark or light mice were pooled and scanned with SSLP markers spaced 1 every 30 41  centimorgans(Fitch et al. 2003). Chromosome 11 was identified when two markers showed only the C3HeB/FeJ allele in the dark pool DNA, and both the C3HeB/FeJ and C57BL6J alleles in the light pool DNA. These 16 mice plus an additional 38 of the darkest animals from the backcross were genotyped with additional markers on Chromosome 11, with 1 recombination event found between dark skin and the C3HeB/FeJ allele of D11Mit365, with only the C3HeB/FeJ allele present at markers to the left (see fig.1f). To identify the homozygous phenotype of Dsk9 and use it for mapping, an outcross-intercross mapping strategy was employed. Nf1Dsk9/+ F1 animals were intercrossed and genotyped for markers along chromosome 11. No animals with only the C3HeB/FeJ at both D11Mit365 and D11Mit245 were recovered out of 184 intercross progeny, showing that the dark skinned linked region of Chromosome 11 in Dsk9 mice is homozygous lethal. One animal out of the 184 had only the C3HeB/FeJ allele at D11Mit245, with both the C3HeB/FeJ and C57BL6 alleles at D11Mit365. Putting the two sets of data together, the Dsk9 interval is defined by D11Mit245 and D11Mit365. ENSEMBL was used to list the genes located within the interval.  2.4 S100a4-Cre PRODUCTION 3.44 kb of the S100a4 promoter was subcloned from a positively identified C57BL/6 BAC clone (RP23: 284L9) using homologous recombination into pSP72 (Promega). A BamHI restriction enzyme site was engineered to the 3’ end of the S100a4 promoter. The Cre cDNA sequence was cloned downstream using BamHI and BglII sites. The 300 bp BGH polyA signal sequence was cloned into the vector by XhoI/SalI ligation. The transgenic vector was confirmed by restriction analysis and by sequencing after each modification step. Purified targeting vector was digested by NotI and SphI and microinjected into hybrid (129; C57 F1) pronuclear stage embryos to 42  generate two founder lines (A and B). Rosa26-floxed stop-LacZ mice were used to assess Cre activity in the two lines. Line A was chosen for the cross to Nf1tm1Par.  2.5 β-GALACTOSIDASE (LacZ) DETECTION Embryos, trunk skin and tail skin samples were incubated with X-gal staining solution (10% NP40, 1% Deoxycholate, 0.5M Potassium ferricyanide, 0.5M Potassium ferrocyanide trihydrate, 1M MgCl2, 10X PBS and 40mg X-gal/1mL of dimethylformamide per 40mL of staining solution) for 16-36 hours at room temperature. In tail skin samples, X-gal staining was followed by a 2M sodium bromide incubation to separate the dermis and epidermis.  2.6 DERMAL-EPIDERMAL SEPARATION AND PIXEL INTENSITY OF TAIL SKIN 1 cm piece of skin from the middle of the tail was removed from the tail bones of 3 week old mice, incubated in 2M sodium bromide for 2 hours at 37oC, and separated using fine forceps. Skin samples to be compared to one another were photographed within a single picture. Dermal and epidermal darkness was quantified using ImageJ in terms of average pixel intensity and are not directly comparable between different experiments.  2.7 EPIDERMAL SCALE DARKNESS VERSUS MELANOCYTE DENSITY Two 1 cm pieces of tail skin from the middle of the tail were removed from each 3 week old animal, with one piece stained in X-gal and the other left unstained. In each of the X-gal stained epidermal sheets, the density of the blue stained cells was determined in 10 scales located at the  43  very dorsum of the tail. In the unstained sheets, the pixel intensity of 10 scales from the very dorsum of each tail was measured. The two data points for each animal were plotted.  2.8 TAMOXIFEN INDUCTION OF CreER ACTIVITY Tamoxifen (Sigma) was dissolved in sunflower oil:ethanol (10:1) mixture at a concentration of 10 mg/mL. Pregnant females were injected intraperitoneally with 1mg of tamoxifen at either E9.5 or E11.5.  2.9 TISSUE EMBEDDING For paraffin embedding, mouse trunk and tail skin samples were dissected, fixed overnight in 4% paraformaldehyde at 4oC, washed 4 times with 1X PBS for 10 minutes, then washed in 0.85% sodium chloride (saline), 1:1 saline:100% ethanol, a series of ethanol washes (70%, 70%, 85%, 90%, 95%, 100%, 100%) and xylene twice, for 30 minutes each at room temperature. Samples were then incubated in 65oC paraffin for 3 hours and embedded. For cyroembedding, mouse embryos were dissected from pregnant females and incubated with 4% paraformaldehyde for 4 hours at 4oC. The embryos were incubated in 10% sucrose and 30% sucrose, each for 24 hours, and then embedded in O.C.T.  2.10 HEMATOXYLIN AND EOSIN STAINING 8 μM sections of skin from 6 week old mice were obtained from paraffin blocks, dewaxed, rehydrated, incubated in hematoxylin for 3 minutes, dehydrated, incubated in eosin for 15 seconds, coverslipped and photographed using a Zeiss Axioplan bright-field microscope.  2.11 IMMUNOFLUORESCENCE  44  8 μM frozen embryo sections were washed at room temperature three times in 1X PBS for 5 minutes each, 0.1% Triton-X-100 for 30 minutes, 1X PBS for 5 minutes, 4% donkey serum for 30 minutes, and incubated with rabbit anti-Tuj1 antibody (Covance, 1:1000) and mouse anti-βgalactosidase antibody (Promega, 1:500) overnight at 4oC. The following day, the sections were washed in PBS and incubated with Alexa594 conjugated anti-rabbit antibody and Alexa488 conjugated anti-mouse antibody (Invitrogen Molecular Probes), for 1 hour in 2% serum. The slides were washed in PBS, incubated with 0.5μg/mL of diaminopyridine imidazole (DAPI), washed in PBS, and mounted. Immunofluorescence was detected using Leica DMI 6000B epifluorescence microscope.  2.12 FACS ANALYSIS Embryos were dissected at E11.5 and the trunk skin of the embryos was removed and dissociated by pipetting in trypsin-EDTA for 15 minutes. Cells were sorted for EYFP expression and counted using a FACS LSRII cell sorter. Subsequently, embryonic membrane DNA from the embryos was used for genotyping.  2.13 PERCENT OF COAT PIGMENTATION The regions of pigmented coat and total coat for mice positioned ventrum side down were traced in ImageJ and the areas were measured for each mouse.  2.14 SEQUENCING 5μL of diluted PCR product (35ng) was mixed with 2μL of Exo-SAP-IT PCR product clean-up mix (Affymetrix). 3μL of Exo-SAPed PCR product, 1pM of primer, and 3μL of Big dye mix (v3.1, Applied Biosystems) were used for each sequencing reaction, and run on an Applied Biosystems 3730 DNA Analyzer. 45  2.15 STATISTICS Normal distribution of the data was determined using Shapiro-Wilk W test using JMP software. Values were means ± SEM. Statistical significance was calculated using either Student’s t-test (Figure 1.8h,i, 4.1e, 4.2c, 4.3d and5.2d) or ANOVA (Figure 3.4h, 5.2b)  46  CHAPTER 3: EFFECT OF LOSS OF Nf1 IN THE HOMOZYGOTE  3.1 INTRODUCTION We have found that heterozygous mutations in neurofibromin (Nf1) cause darker skin in mice (Figure 1.7, 1.8), as they do in neurofibromatosis type 1 patients. To further investigate the role of neurofibromin in pigmentation, we decided to study the effects of homozygous loss of Nf1 in embryos and adult mice. Nf1-/- mouse embryos die at E13.5. To be able to study the effects of homozygous loss of Nf1 after E13.5, we made use of Cre-loxP technology (see Section 1.6.1) to delete Nf1 in melanocytes.  In humans, biallelic mutation of NF1 in melanocytes produces Cafe au lait macules (CALMs) (De Schepper et al. 2008). CALMs are benign, epidermal, tan-brown, clearly demarcated macules with smooth boarders. In neurofibromatosis type 1, the CALMs are darker than the surrounding mildly hyper-pigmented skin. CALMs have an increased melanocyte density, and they usually develop during early childhood. In one study, the melanosomes within CALM skin from neurofibromatosis type 1 patients were reported to be greatly enlarged (Jimbow et al. 1973). We were interested in determining whether mouse skin would exhibit hyper-pigmentation and/or changes in melanosome structure upon the biallelic deletion of Nf1 in melanocytes beginning during development.  We were also curious whether homozygous loss of neurofibromin in melanocytes would alter coat colour. Hair follicle melanocytes use adenylyl cyclase (AC) activity to regulate the switch 47  between eumelanin (black pigment) and pheomelanin (yellow pigment) production. Pheomelanin synthesis predominates when AC produces less cAMP. The C-terminus of neurofibromin has been shown to increase AC activity in a Gαs-dependent manner in mouse neurons (Tong et al. 2002). If neurofibromin is required in melanocytes for AC activity, then loss of both copies of Nf1 in melanocytes might generate yellow fur in the mutant mice.  The experiments in this chapter show that, in mice, deleting both copies of Nf1 in melanocytes causes skin hyper-pigmentation. Biallelic loss of Nf1 affects both dermal and epidermal melanocytes, beginning during development.  48  3.2 RESULTS  3.2.1 Nf1Dsk9/Nf1Dsk9 embryos exhibit increased melanoblast numbers during development In order to analyse the effects of Nf1Dsk9 homozygosity on melanoblast development, we obtained a transgenic mouse line that uses the Dct (i.e. Tyrp2) promoter to express LacZ in melanoblasts beginning at E10.5 (Mackenzie et al. 1997). We crossed Nf1Dsk9/+; Dct-LacZ/DctLacZ mice to Nf1Dsk9/+ mice expressing LacZ and dissected the resulting embryos at E10.5. In whole mount preparations, we observed a dramatic increase in the number of LacZ-positive cells in the head and neck region in the Nf1Dsk9/Nf1Dsk9 embryos compared to wildtype (Figure 3.1a-c).  Wehrle-Haller et al. (Wehrle-Haller et al. 2001) previously reported that E11.5 embryos homozygous for the targeted Nf1tm1Fcr allele exhibit increased numbers of Dct-expressing cells, with some cells ectopically localized at an extreme dorsal position along the trunk. We also noticed LacZ-positive cells in our E10.5 Nf1Dsk9/Nf1Dsk9 embryos in this same location (Figure 3.1a-c). Because Wehrle-Haller et al. showed that the ectopic cells in Nf1tm1Fcr/Nf1tm1Fcr embryos are c-kit independent, we considered whether they might not be melanoblasts, despite being migratory cells expressing a known melanogenic enzyme (Dopachrome tautomerase).  We decided to examine the expression of neuronal (Tuj1) and glial (S100β, Blbp and Gfap) cell markers, using immunofluorescence specific for these proteins, in addition to LacZ. We were not able to detect colocalization of S100β, Blbp and Gfap with LacZ, but we did observe colocalization of LacZ and Tuj1 in cells located at the dorsal-most position in Nf1Dsk9/Nf1Dsk9; Dct49  LacZ/+ embryos (Figure 3.2 a,b). Dct is expressed in the telencephalon (Mackenzie et al. 1997) and may play a role in neuronal progenitor cell proliferation (Jiao et al. 2006). It is possible that the ectopic cells in Nf1Dsk9/Nf1Dsk9 embryos are neuronal progenitors.  We also used immunofluorescence to quantify the number of Tuj-negative, LacZ-positive cells in Nf1Dsk9/Nf1Dsk9 and wildtype E11.5 embryos. These cells are presumably melanoblasts. We observed a 3.5-fold increase the number of Tuj-negative, LacZ positive cells in Nf1Dsk9/Nf1Dsk9 (37.5 ± 10.48) embryos compared to wildtype (10.5 ± 2.13) (Figure 3.3). Altogether, these findings suggest that germline biallelic mutation of Nf1 increases melanoblast numbers immediately after the differentiation of these cells from the neural crest and may also disrupt the behavior of neural crest neuronal progenitor cells.  3.2.2 Skin hyper-pigmentation in Mitf-Cre/+; Nf1tm1Par/Nf1tm1Par mice In order to assess the effect of homozygous loss of Nf1 on adult melanocytes, we used two strains of transgenic mice: Mitf-Cre and Nf1tm1Par. Mitf-Cre expresses Cre under the control of melanocyte-specific 1M promoter of Microphthalmia. Microphthalmia encodes a Myc supergene family basic helix-loop-helix zipper transcription factor and is one of the first genes expressed in committed melanoblasts. In Mitf-Cre mice, Cre expression is first detectable beginning at E11.5 and labels dermal, epidermal and follicular melanocytes in adult tail skin (Alizadeh et al. 2008). Nf1tm1Par is an engineered Nf1 allele, in which exons 40 and 41 are flanked by loxP sites (Zhu et al. 2001).  50  We crossed Mitf-Cre/+; Nf1tm1Par/+ mice to Nf1tm1Par/+ mice to obtain Mitf-Cre/+; +/+ and MitfCre/+; Nf1tm1Par/Nf1tm1Par animals. We genotyped Nf1 using three separate PCR assays specific for (1) the wildtype Nf1 allele, and (2) the non-recombined Nf1tm1Par allele, and (3) the recombined Nf1tm1Par allele (Zhu et al. 2001). We also genotyped for the presence of Mitf-Cre. Absence of a PCR product in the Nf1 wildtype allele assay, in combination with a positive reaction for Mitf-Cre and the recombined Nf1tm1Par allele identified the Mitf-Cre/+; Nf1tm1Par/ Nf1tm1Par animals. “Wildtype” mice either did not inherit Mitf-Cre, or did inherit Mitf-Cre, but did not give a positive reaction for either the recombined or non-recombined Nf1tm1Par allele assays.  We examined the Mitf-Cre/+; Nf1tm1Par/Nf1tm1Par animals by gross inspection and found that the skin of these mice is even darker than Nf1 +/- mutant mice (Figure 3.4). Both the dermis and the epidermis are hyper-pigmented (Figure 3.4g). The coat colour of the non-agouti (a/a) Mitf-Cre/+; Nf1tm1Par/Nf1tm1Par mice did not differ from wildtype under visual inspection. Thus, we found no evidence that neurofibromin regulates AC activity in melanocytes (Figure 3.4d).  Histological sections of 3 week old tail skin reveal increased melanin in the dermis and the epidermis of Mitf-Cre/+; Nf1tm1Par/Nf1tm1Par mice (Figure 3.5 b). Melanin was also frequently visible in the trunk skin dermis of the Mitf-Cre/+; Nf1tm1Par/Nf1tm1Par mice (Figure 3.5 d). Usually, the trunk skin dermis is only sparsely pigmented (Figure 3.5 c). Like wildtype mice, the trunk skin epidermis of Mitf-Cre/+; Nf1tm1Par/Nf1tm1Par mice is unpigmented (Figure 3.5c, d). 51  These findings demonstrate that Mitf-Cre is sufficient to induce skin hyper-pigmentation of the dermis and epidermis in mice bearing two Nf1tm1Par alleles.  3.2.3 FACS sorting EYFP-positive cells in Mitf-Cre/+; Nf1tm1Par/Nf1tm1Par embryos In an effort to study Nf1 -/- melanoblasts further, we decided to try to isolate the Mitf-Cre expressing cells in Mitf-Cre/+; Nf1tm1Par/Nf1tm1Par embryos. We crossed Mitf-Cre/+; Nf1tm1Par/+ mice to Nf1tm1Par/+; Rosa26-floxed stop-EYFP/Rosa26-floxed stop-EYFP mice and dissected the resulting embryos at E15.5. We removed the trunk skin from each embryo, dissociated the cells with trypsin, and sorted the EYFP-positive cells from each embryo using FACS. Subsequently, the amniotic membrane from each embryo was used for genotyping.  Although we have not yet been able to culture the sorted EYFP-positive cells, we found that there is a 2.4-fold increase in the percent of EYFP-positive cells in Mitf-Cre/+; Nf1tm1Par/Nf1tm1Par trunk skin compared to Mitf-Cre/+; +/+ trunk skin (Figure 3.4h). We tried to sort the dissociated cells for c-kit-positive (marker for melanocytes and mast cells); CD45negative cells (marker for hematopoietic cells), but failed because the lengthy anti-c-kit/ antiCD45 staining procedure lead to a loss of a substantial number of cells. However, since there is no evidence indicating that Mitf-Cre is expressed in any other cell type besides melanocytes in the skin (Alizadeh et al. 2008), it is likely that Mitf-Cre/+; Nf1tm1Par/Nf1tm1Par have an increased number of melanoblasts at E15.5.  52  3.2.4 No change in melanosome structure in Mitf-Cre/+; Nf1tm1Par/Nf1tm1Par dermal melanocytes We next examined the appearance of melanosomes in dermal melanocytes from Mitf-Cre/+; Nf1tm1Par/Nf1tm1Par and wildtype tail skin using transmission electron microscopy (TEM) in collaboration with Dr. Douglas Keene at the Shriner’s Hospital in Portland, Oregon. No aberrantly large melanosome similar to the previously described NF1 “macromelanosomes” (Jimbow et al. 1973) were observed. ImageJ software was used to measure the average area occupied by melanosomes in each photograph (3.1i-k). No difference was found between MitfCre/+; Nf1tm1Par/Nf1tm1Par and wildtype samples, despite the fact that dermis is 70% darker in the Mitf-Cre/+; Nf1tm1Par/Nf1tm1Par animals. Although melanosome area was not measured in the epidermis, Dr. Keene did not notice anything unusual in the ultrastructure of the epidermal melanocytes. Thus, our data does not support a role for neurofibromin in regulating melanosome structure in mice.  3.2.5 No obvious role for Nf1 in the formation of melanocytic lesions Biallelic loss of Nf1 in Schwann cells triggers the development of neurofibromas, tumours composed of spindle shaped Schwann cells, mast cells and perineural cells, combined with collagen and extracellular matrix. In order to assess any possible neoplasia development due to biallelic loss of Nf1 in melanocytes, we aged Mitf-Cre/+; Nf1tm1Par/Nf1tm1Par and Mitf-Cre/+; Nf1tm1Par/Nf1Dsk9 mice for up to 18 months.  53  No  obvious  tumours were  observed  in  Mitf-Cre/+;  Nf1tm1Par/Nf1tm1Par,  Mitf-Cre/+;  Nf1tm1Par/Nf1Dsk9, or wildtype mice. On a HairlessHr/HairlessHr mutant background, we noticed that some of the Mitf-Cre/+; Nf1tm1Par/Nf1tm1Par and Mitf-Cre/+; Nf1tm1Par/Nf1Dsk9 mice developed unusual patches of trunk skin hyper-pigmentation over time (Figure 3.6). No irregular mitoses, conspicuous nucleoli, pagetoid cells or spindle-shaped cells were observed in the trunk skin, including regions of hyper-pigmentation, during cytological analysis (performed by Dr. Nick Nation at Per Animal Pathology Services, Alberta). These findings suggest that Nf1 -/melanocytes have an enhanced ability to survive/proliferate in the adult trunk dermis, but, by itself, this does not lead to a transformed phenotype.  54  a  +/+;Dct-LacZ/+  b  +/+;Dct-LacZ/+  c Nf1Dsk9/Nf1Dsk9;Dct-LacZ/+  10.5  Figure 3. 1 Increased number of LacZ-positive cells in E10.5 Nf1Dsk9/Nf1Dsk9;Dct-LacZ/+ embryos  (a-c) E10.5 Dct-Lacz/+ embryos stained with X-gal. (a) The arrowhead indicates the otic vesicle region, where many LacZ-positive cells are located at E10.5. Bracketed region indicates location of the neural tube at E10.5. There are more LacZ-positive cells in Nf1Dsk9/Nf1Dsk9 (c) embryos compared to wildtype (b). There are also LacZ-positive cells located at more dorsal positions in Nf1Dsk9/Nf1Dsk9 embryos compared to wildtype (i.e. to the left of the dashed line in c.). Arrows in (b) and (c) represent the first dorsal root ganglion. Scale bars: b,c = 500 µm.  55  Nf1Dsk9/Nf1Dsk9;Dct-LacZ/+  a  b  LacZ  Average number of Dct-LacZ-positive; Tuj-positive cells per section at the extreme dorsum  NT  Tuj  NT  LacZ/Tuj dorsum  NT  16 14 12 10 8 6 4 2 0 +/+; Dct-LacZ/+ n=6  Nf1Dsk9/Nf1Dsk9; Dct-LacZ/+ n=5  Figure 3.2 Quantification of Tuj1-positive, LacZ-positive cells in Nf1Dsk9/Nf1Dsk9 embryos. (a)Immunofluorescence was used to label Tuj (red) and LacZ (green)-expressing cells in 10 sections, taken from the otic vesicle region at E11.5, from 6 +/+;Dct-LacZ/+ and 5 Nf1Dsk9/Nf1Dsk9;Dct-LacZ/+ embryos. Tuj-positive; LacZ-positive cells (yellow) located dorsal to the middle of the neural tube (blue arrow) is shown. (b) Graph represents quantification of Tujpositive, LacZ-positive cells in the Nf1Dsk9/Nf1Dsk9;Dct-LacZ/+ embryos. No Tuj-positive, LacZpositive cells were found dorsal to this cutoff line in +/+ embryos. Scale bar in a = 30µm.  56  Average number of Tuj-negative; Dct-LacZ-positive cells per section  100 80 60 40 20 0 +/+ n=6  Nf1Dsk9/Nf1Dsk9 n=5  Figure 3.3 Quantification of Tuj1-negative, LacZ-positive cells in Nf1Dsk9/Nf1Dsk9 embryos. Immunofluorescence was used to label Tuj and LacZ-expressing cells in 10 sections, taken from the otic vesicle region at E11.5 of 6 +/+;Dct-LacZ/+ and 5 Nf1Dsk9/Nf1Dsk9;Dct-LacZ/+ embryos. The number of Tuj-negative, LacZ-positive cells is shown (Mean ± S.E.M.). Individual dots represent the value from each section. There were significantly more cells per section in Nf1Dsk9/Nf1Dsk9 embryos (p=0.0257).  57  a  b Ear  Tail  d Coat color  c a Footpad  +/+  Mitf-Cre/+; Nf1tm1Par/+ Mitf-Cre/+; Nf1tm1Par/Nf1tm1Par  +/+  e  f  WT  g  250  Mitf-Cre/+; Nf1tm1Par/Nf1tm1Par  dermis  WT  Mitf-Cre/+; Nf1tm1Par/Nf1tm1Par  Mitf-Cre/+; Nf1tm1Par/Nf1tm1Par epidermis  200  150  100  WT Mean % of EYPF positive cells at E15.5  h  0.5  n=4  Mitf-Cre/+; Nf1tm1Par/Nf1tm1Par n=6  0.4  i  0.3  WT  WT n=4  j  Mitf-Cre/+; Nf1tm1Par/Nf1tm1Par n=6  Mitf-Cre/+;Nf1tm1Par/Nf1tm1Par  0.2  0.1  0 Mitf-Cre/+; Mitf-Cre/+; Mitf-Cre/+; +/+ Nf1tm1Par/+ Nf1tm1Par/Nf1tm1Par n=6 n=3 n=15  k Melanosomes Mean Area  WT 0.55 ± 0.01  Mitf-Cre/+;Nf1flox/flox 0.58 ± 0.01  58  Figure 3.4 Skin hyper-pigmentation in Mitf-Cre/+; Nf1tm1Par/Nf1tm1Par mice Mitf-Cre/+; Nf1tm1Par/Nf1tm1Par mice exhibit darkening of tail (a), ear (b) and footpad skin (c). There is no visible change in coat color in Mitf-Cre/+;Nf1tm1Par/Nf1tm1Par  mice (d).  Representative tail dermis (e) and epidermis (g) of 3 week old wildtype (left) and Mitf-Cre/+; Nf1tm1Par/Nf1tm1Par (right) littermates. Average pixel intensity of photographed adult tail dermis and epidermis (± S.E.M.) (g). The percent of EYFP-positive cells sorted from Rosa26-floxed stop-EYFP E15.5 embryos expressing Mitf-Cre/+ (h). Individual dots represent the value from each sample. Representative transmission electron micrographs of tail skin dermis from wildtype (i) and Mitf-Cre/+; Nf1tm1Par/Nf1tm1Par (j) adult animals. No difference is observed in the average area (± S.E.M.) of stage II-IV melanosomes in Mitf-Cre/+; Nf1tm1Par/Nf1tm1Par animals (k). Statistical analysis: Mean pixel intensity of Mitf-Cre/+;Nf1tm1Par/Nf1tm1Par dermis, p<0.0001; epidermis, p=0.0148. Mean percent of EYFP-positive cells from Mitf-Cre/+ +/+ v/s MitfCre/+;Nf1tm1Par/Nf1tm1Par p=0.0238  59  +/+ a  Mitf-Cre/+;Nf1tm1Par/Nf1tm1Par b  Tail skin  E  E D  c Dorsal skin  D  d  trunk E  E D  D  Figure 3.5 Histological analysis of melanin accumulation in Mitf-Cre/+; Nf1tm1Par/Nf1tm1Par animals Tail and trunk skin sections from 3 week old wildtype (a,c), and Mitf-Cre; Nf1tm1Par/Nf1tm1Par (b,d) mice, respectively, were stained with eosin. Black arrows show dermal melanin and white arrowheads show epidermal melanin. Dotted line demarcates the boundary between dermis and the epidermis. Scale bars: 50μm. D: dermis; E: epidermis  60  WT  Mitf-Cre/+; Nf1tm1Par/Nf1Dsk9  Figure 3.6 Trunk skin hyper-pigmentation in Mitf-Cre/+; Nf1tm1Par/Nf1Dsk9 mice Dark patches of trunk skin pigmentation are visible on the neck of a 1 year old MitfCre/+;Nf1tm1Par/Nf1Dsk9 mouse (right) (see arrows). These mice also show hyper-pigmentation of the ears, footpads and tail skin. Mice are genetically hairless (HairlessHr/HairlessHr).  61  3.3 DISCUSSION  Multiple CALMs are one of the cardinal diagnostic criteria of neurofibromatosis type 1 (De Schepper et al. 2005). We knocked out both copies of Nf1 in melanoblasts expressing Mitf-Cre and found that this causes epidermal and dermal hyper-pigmentation. Furthermore, Nf1-/embryos exhibit increased numbers of melanoblasts during development. As with CALMs, which are typically benign, biallelic mutation of Nf1 in melanocytes expressing Mitf-Cre was not found to be tumorigenic in mice.  The skin surrounding CALMs has been reported to differ in several ways. For example, Amer et al. showed that there are increased numbers of inflammatory cells and melanophages under CALMs (Amer et al. 2001), while Okazaki et al. showed that there are increased levels of HGF and KITL secreted by fibroblasts under the CALMs (Okazaki et al. 2003). These differences could result from amplification of normal processes due to increased melanocyte density or from qualitative changes in the behavior of NF1-/- melanocytes. FACS sorting melanocytes from Mitf-Cre/+; Nf1tm1Par/Nf1tm1Par; Rosa26-floxed stop-EYFP postnatal tail dermis might produce enough cells to study gene expression by qRT-PCR and address these possibilities.  Several lines of evidence suggest that the darker skin caused by the loss of neurofibromin is a result of increased signaling through the Map kinase pathway. In addition to neurofibromatosis type 1, multiple CALM’s are a feature of Legius syndrome, which is caused by loss of function mutations in SPRED1, another negative regulator of MAPK signalling (Brems et al. 2007). 62  Transgenic mice expressing activated HA-RAS in melanocytes exhibit both epidermal and dermal hyper-pigmentation, the only darker skinned transgenic mice to do so (Powell et al. 1995). The Nf1Dsk9 mutation is in the GRD, which is known to negatively regulate RAS activity (Ballester et al. 1990) and the altered residue, Asn1453, has been shown to be important for RAS interaction (Morcos et al. 1996). It is important to note that the MAPK pathway plays a central role in melanoma, with somatic BRAF and RAS mutations occurring in ~ 80% of cutaneous melanomas.  While the dermis of humans is normally unpigmented, melanin is frequently observed in the dermis overlying plexiform neurofibromas in neurofibromatosis type 1 patients (Fetsch et al. 2000; Schaffer et al. 2007). Similarly, Wu et al. knocked out both copies of Nf1 in Schwann cell precursors using a Desert hedgehog-Cre expressing transgene, and found that 39% of the animals exhibited hyper-pigmentation overlying the plexiform neurofibromas/spinal cord (Wu et al. 2008). Dhh-Cre is expressed in Schwann cell precursors that can give rise to both melanocytes and glial cells during development. Other forms of dermal melanocytoses (i.e. blue nevus, nevus of Ota) have not been reported to have increased frequency in neurofibromatosis type 1 patients, so it is likely that the neurofibroma microenvironment supports melanocytes in the dermis. No one has reported whether the dermal melanocytes overlying plexiform neurofibromas have one or two mutant alleles of NF1, so their source remains a mystery. We found that homozygous knockout of Nf1 in committed melanocytes causes patches of dermal trunk skin pigmentation, indicating that Nf1-null melanocytes have an enhanced ability to survive/proliferate in the mouse trunk dermis.  63  In conclusion, these experiments show that biallelic mutation of Nf1 in melanocytes causes skin hyper-pigmentation in mice, as it does in humans. It is intriguing that melanoma is not more common in neurofibromatosis type 1 patients, given the role of neurofibromin in MAPK signalling, the developmental connection between Schwann cells and melanocytes, and the increased numbers of melanocytes in the skin of Nf1 mutants. What controls the switch between benign hyper-proliferation and malignancy in the MAPK pathway? To explore this, it might be interesting to examine the effects of Nf1 loss in melanocytes on Erk1/2, p38 and JNK phosphorylation, downstream effectors of MAPK signalling.  64  CHAPTER 4: EFFECT OF Nf1 HAPLOINSUFFICIENCY ON DERMAL AND EPIDERMAL PIGMENTATION  4.1 INTRODUCTION  In humans, Nf1 haploinsufficiency causes generalized epidermal hyper-pigmentation of the skin. We have found that Nf1Dsk9/+ and Nf1 +/- knockout mice exhibit hyper-pigmentation of the dermis and the epidermis of the tail (Figure 1.8). These Nf1 mouse mutants provide an opportunity to better understand the pigmentary manifestations of neurofibromatosis type 1. For example, we can address whether the skin of Nf1 +/- mice is darker due to an increased number of melanocytes or due to increased pigment production. We can also use Cre/loxP technology (see Section 1.6.1) to determine which cell type(s) are participating in the dark skin phenotype.  Nf1 haploinsufficiency is known to affect a number of cell types in addition to melanocytes, such as Schwann cells (Kim et al. 1995), neurons (Zhu et al. 2001), endothelial cells (Gitler et al. 2003), mesenchymal cells (Le et al. 2004; Kolanczyk et al. 2007), fibroblasts (Rosenbaum et al. 1995), and hematopoietic cells (Le et al. 2004; Yang et al. 2006).  In the skin, Nf1 haploinsufficiency could be required solely in melanocytes, solely in another cell type, or in multiple cells types, in order to cause hyper-pigmentation. If heterozygous knockout of Nf1 in melanocytes causes an identical phenotype as heterozygous knockout in all cells, then it would be suspected that no other cell type is required for the dark skin phenotype. Alternatively, neurofibromin haploinsufficiency might affect the behavior of one type of cell, 65  which then secondarily changes the behavior of melanocytes. For example, neurofibromin haploinsufficiency in fibroblasts might cause the fibroblasts to increase secretion of a cytokine that stimulates melanocyte proliferation.  In the dermis, mast cells and fibroblasts might interact with melanocytes. Nf1 haploinsufficiency increases cutaneous mast cell numbers (Ingram et al. 2000). Nf1+/- mast cells secrete elevated levels of TGF-β, which increase the survival and/or proliferation of Nf1 +/- fibroblasts (Yang et al. 2006). Fibroblasts secrete HGF and Kitl, factors which stimulate melanocytes (Shishido et al. 2001; Okazaki et al. 2003; De Schepper et al. 2006). In the epidermis, keratinocytes also secrete Kitl.  In this chapter, we examine the effects of Nf1 haploinsufficiency on fibroblasts, keratinocytes, melanocytes and Schwann cell precursors, using various Cre lines and Cre/loxP technology. We did not find skin hyper-pigmentation when we induced Nf1 haploinsufficiency in already committed melanocytes, in fibroblasts + keratinocytes + melanocytes, or in mast cells + melanocytes. Although we did not address every possible scenario, we found that haploinsufficiency of Nf1 in bipotential Schwann cell-melanoblast precursors induces skin hyper-pigmentation, suggesting that neurofibromin plays a role in melanoblast specification within the glial lineage.  66  4.2 RESULTS  4.2.1 Nf1Dsk9/+ mice exhibit an increase in the number of melanoblasts beginning at E12.5 To determine whether darker skin in Nf1 +/- mice is due to an increased number of melanocytes, we crossed Nf1Dsk9/+ mice with homozygous Dct-LacZ transgenic mice (Mackenzie et al. 1997) to obtain Nf1Dsk9/+ and +/+ mice expressing one copy of the Dct-LacZ transgene. We dissected embryos at E10.5 and E12.5 and performed whole mount staining using X-gal. We found no difference in the number of LacZ-positive cells in Nf1Dsk9/+ and +/+ embryos at E10.5 (Figure 4.1 a, b). However, at E12.5, Nf1Dsk9/+ embryos exhibit an increase in the number of LacZpositive cells in the skin overlying the dorsal root ganglia (Figure 4.1c-e).  We next quantified the number of LacZ-positive cells in the tail epidermis of Nf1Dsk9/+ and wildtype adult mice. We removed the tail skin, stained it with X-gal, and then split the dermis and epidermis using sodium bromide. We observed a 24% ± 8% increase in melanocytes per mm2 in Nf1Dsk9/+ mice as compared to wildtype littermates (Figure 4.2 a, b). In addition, we measured the pixel intensity of individual epidermis samples and correlated this to melanocyte density using a scatter plot of 10 mice. We found a general correlation between pixel intensity and melanocyte density in the scales (Figure 4.2 c).  We conclude from these experiments that Nf1 haploinsufficiency causes skin hyperpigmentation, at least in part through an increased number of melanoblasts produced during development. 67  4.2.2 No skin hyper-pigmentation in Mitf-Cre/+; Nf1tm1Par/+ mice In order to assess whether haploinsufficiency of Nf1 in melanocytes causes dermal hyperpigmentation, we measured the pixel intensity of group photographed tail dermis and epidermis from Mitf-Cre/+; Nf1tm1Par/+ and Mitf-Cre/+; +/+ animals. We observed no difference between the Mitf-Cre/+; Nf1tm1Par/+ and Mitf-Cre/+; +/+ samples in either the epidermis or the dermis (Figure 4.1 a, b). This was surprising, given the strong phenotype observed in Mitf-Cre/+; Nf1tm1Par/ Nf1tm1Par animals, and opens the possibility that neurofibromin haploinsufficiency and complete neurofibromin loss cause hyper-pigmentation through different ways.  As described in section 1.6.1, Cre mouse lines vary in recombination efficiency. To assess the efficiency of Mitf-Cre during embryogenesis, we crossed Mitf-Cre to Rosa26-floxed stop-EYFP and used FACS to sort the EYFP-positive cells from E15.5 trunk skin. We observed that 0.11% ± 0.01% (± S.E.M.) of cells in the skin express EYFP. We then dissected the trunk skin from E15.5 Dct-LacZ/+ embryos, stained in X-gal and counted both the number of LacZ-positive cells per gram of skin and the total number of cells per gram of skin, determining that 0.46% ± 0.15% (± S.E.M.) of skin cells express Dct-LacZ at E15.5. From this we estimate that Mitf-Cre brings about recombination at the Rosa26 locus in 24 ± 8% of melanoblasts by E15.5. Although this is sufficient to induce a phenotype in Mitf-Cre/+; Nf1tm1Par/ Nf1tm1Par embryos at E15.5, it may be too little/too late to induce a phenotype in Mitf-Cre/+; Nf1tm1Par/ + animals if neurofibomin haploinsufficiency only affects melanoblast specification or very early development.  68  4.2.3 No skin hyper-pigmentation in S100a4-cre/+; Nf1tm1Par/+ mice In order to address whether haploinsufficiency of Nf1 in fibroblasts and/or keratinocytes leads to epidermal skin hyper-pigmentation through a paracrine mechanism, we replicated a previously published Cre allele (Shrestha et al. 1998; Bhowmick et al. 2004) reported to be expressed in fibroblasts and keratinocytes using the services of Ingenious Targeting Laboratories. In order to confirm the expression of our S100a4-Cre allele in the skin, we crossed S100a4-Cre/+ mice to Rosa26-floxed stop-LacZ reporter mice and performed X-gal staining on tail skin. As expected, the S100a4-Cre/+; Rosa26-floxed stop-LacZ/+ mice exhibited strong β-galactosidase expression in fibroblasts as well as keratinocytes in the inter-follicular epidermis and hair follicles (Figure 4.5b).  We then crossed S100a4-cre to Nf1tm1Par and analyzed the pixel intensity of S100a4-Cre/+; Nf1tm1Par/+and wildtype adult mouse tails. We found no difference in the dermis or epidermis of S100a4-Cre/+; Nf1tm1Par/+ mice compared to wild type littermates. Furthermore, there was no difference in the dermis pigmentation of Mitf-Cre/+; S100a4-Cre/+; Nf1tm1Par/+ mice compared to wildtype (Figure 4.3 b). These findings do not support a role for neurofibromin gene dosage in regulating paracrine signalling between fibroblasts and/or keratinocytes and melanocytes.  4.2.4 No skin hyper-pigmentation in Vav1-cre/+; Nf1tm1Par/+ mice We next assessed whether Nf1 haploinsufficiency in mast cells of the skin causes dermal hyperpigmentation. (At the time of this experiment, the epidermal hyper-pigmentation of Nf1 +/- mice had not yet been discovered, so the epidermis was not examined.) We obtained a commercially 69  available Cre line, Vav1-Cre, which is expressed in all hematopoietic cells, including mast cells. Gan et al. showed that Vav1-Cre has a deletion efficiency of 100% at E13.5 (Gan et al. 2010).  We crossed Vav1-Cre/+ mice to Mitf-Cre/+; Nf1tm1Par/+ mice and measured the pixel intensity of the adult tail dermis of the progeny. We found no difference between Vav1-Cre/+; Mitf-cre/+; Nf1tm1Par/+ and wildtype mice (Figure 4.3c). Thus, these findings also do not support a role for neurofibromin gene dosage in regulating paracrine signalling between mast cells and melanocytes.  4.2.5 Skin hyper-pigmentation in Plp1-creER/+; Nf1tm1Par/+ mice A lack of phenotype in the above mouse knockouts could be due to incomplete deletion of Nf1 or due to a need for Nf1 knockout in more than two cell types. However, a third possibility is that neurofibromin haploinsufficiency might only affect melanoblast specification or very early melanoblast development. Therefore, we decided to generate Nf1 haploinsufficiency in melanoblast precursor cells.  Recently Adameyko et al. demonstrated a commercially available Plp1-creER transgenic line to be expressed in Schwann cell precursors that give rise to melanoblasts. A single injection of tamoxifen at E11.5 of embryogenesis resulted in 66% of adult hair follicle melanocytes exhibiting recombination. We note that tamoxifen is only transiently present in this experimental system. Before injection, and at some point after injection, CreER will not be active in cells expressing it. Therefore, Plp1-creER does not target all cells in the embryo that express Plp1. 70  We crossed Nf1tm1par/+ mice to Plp1-CreER/+ mice and treated pregnant females with tamoxifen using a single intraperitoneal injection. When tamoxifen was injected at E9.5, there was no difference in the pixel intensity of the resulting Plp1-CreER/+; Nf1tm1par/+ tail dermis compared to wildtype. However, when tamoxifen was injected at E11.5, the Plp1-CreER/+; Nf1tm1Par/+ mice exhibited hyper-pigmentation of the tail skin dermis (Figure 4.3d). Since Mitf-Cre mice do not exhibit dermal hyper-pigmentation, we conclude that neurofibromin haploinsufficiency must either regulate melanoblast specification within the glial lineage or regulate the survival and/or proliferation of early melanoblasts. We favor the former hypothesis because (1) only the E11.5 injection causes hyper-pigmentation and (2) increased numbers of melanoblasts first appear in Nf1Dsk9/+ embryos at E12.5, not at E10.5. We speculate that neurofibromin protein levels regulate the number of melanoblasts that are produced from Schwann cell precursors.  4.2.6 Fate mapping of melanoblasts expressing Plp1-CreER at E11.5 When we examined the tails of Plp1-CreER/+; Nf1tm1Par/+ animals produced by injecting 1 mg tamoxifen at E11.5, we found that, despite the darker dermis, the epidermis is unaffected. To examine this intriguing finding further, we used Rosa26-floxed stop-LacZ reporter mice (Soriano 1999) to fate map melanocytes expressing Plp1-creER at E11.5.  We crossed Plp1-CreER/+ mice to homozygous Rosa26-floxed stop-LacZ mice, and injected pregnant females with a single dose of tamoxifen at E11.5. We removed the tail skin at P5, stained with X-gal, and then split the dermis and the epidermis. We found many LacZ-positive 71  cells in the hair follicles of the tail, but almost none in the inter-follicular epidermis, even though melanoblasts expressing Dct-LacZ are abundant in the epidermis at P5.  We conclude that it was not possible for the tail epidermis to be darker in the Plp1-CreER/+; Nf1tm1Par/+ mice injected at E11.5, since the inter-follicular epidermis mostly lacks melanocytes that expressed Plp1-creER at E11.5. Inter-follicular epidermal melanocytes might arise from Schwann cell precursors either before or after E11.5 and, if so, they might not have received any tamoxifen. Alternatively, inter-follicular epidermal melanocytes might never express Plp1creER. The cause of epidermal hyper-pigmentation in Nf1 +/- mice remains to be determined. Our next step will be to see whether a traditional Cre line driven by Plp1 causes epidermal hyper-pigmentation when crossed to Nf1tm1Par.  72  +/+;Dct-LacZ/+  +/+;Dct-LacZ/+ a  Nf1-Dsk9/+;Dct-LacZ/+ b  c  d  E10.5  4 5 6 4 5 6  E12.5  Figure 4.1: Increase in melanoblasts in E12.5 Nf1Dsk9/+ embryos.  Number of Dct-LacZ positive cells at E12.5  e 300  200  100  0  +/+ n=5  Nf1Dsk9/+ n=5  X-gal staining pattern showing melanoblasts in wild type and Nf1Dsk9/+ embryos at E10.5 (a,b) and E12.5 (cod). The LacZ positive cells overlying dorsal root ganglia 4-6 (in box) were quantified (e). Graphs in e represent mean ± S.E.M and p = 0.0052 for mean number of DctLacZ-positive cells in +/+ v/s Nf1Dsk9/+ embryos at E12.5. Scale bar in a,b = 100µm.  73  LacZ positive cells in the epidermis a  c  b  Nf1Dsk9/+  Mean pixel intensity of epidermal scales  +/+  Mean number of LacZ positive cells per mm2 of epidermis  Figure 4.2 Increase in epidermal melanocyte numbers in Nf1Dsk9/+ epidermal scales LacZ positive cells (blue) in the epidermal tail scales of adult wild type (a) and Nf1Dsk9/+ (b) mice. (c) Nf1Dsk9/+ mice show a significant (pvalue = 0.02;t-test) increase in the number of LacZ positive cells (88.63 ± 44.31) compared to wild type mice (44.28 ± 18.08). The number of LacZ-positive cells per mm2 is generally correlated to the pixel intensity of the epidermal scales (d).  74  Mean pixel intensity (Dermal pigmentation)  a  200  b  150 100 50 0 Mitf-Cre/+; Mitf-Cre/+; +/+ Nf1tm1Par/+ n=6 n=5  WT n=17  d  200  Mitf-Cre/+; S100a4-Cre/+; Nf1tm1Par/+ n=5 n=6  WT n=6  Mitf-Cre/+; Vav1-Cre/+; Nf1tm1Par/+ n=5  e  Tamoxifen injection at E11.5  200  Tamoxifen injection at E9.5  180  Mean pixel intensity (Dermal pigmentation)  Mean pixel intensity (Dermal pigmentation)  c  160 140 120  100  50  0  100 WT n=6  Mean pixel intensity (Epidermal pigmentation)  150  160  Plp1-CreERT2; Nf1tm1Par/+ n=6  WT  h Tamoxifen injection at E11.5  g  f  Plp1-CreERT2; Nf1tm1Par/+ n=7  n=6  140 120 100 WT n=6  WT  Mitf-Cre/+; Nf1tm1Par/+ n=7  n=3  S100a4-Cre/+; Nf1tm1Par/+ n=8  WT n=5  Plp1-CreERT2; Nf1tm1Par/+ n=6  Figure 4.3: Effect of Nf1tm1par +/- knockout using various Cre lines. No increase in dermal pigmentation is observed in Mitf-Cre/+;Nf1tm1Par/+ (a), MitfCre/+;S100a4-Cre/+;Nf1tm1Par/+ CreERT2/+;Nf1tm1Par/+  mice  (b), injected  Mitf-Cre/+;Vav1-Cre/+;Nf1tm1Par/+ with  tamoxifen  at  E9.5  (c) (d).  or  Plp1-  The  Plp1-  CreERT2/+;Nf1tm1Par/+ mice injected with tamoxifen at E11.5 show a significant hyper75  pigmentation of the dermis (e). No increase in epidermal pigmentation is observed in the tail epidermis  of  Mitf-Cre/+;Nf1tm1Par/+  (f),  S100a4-Cre/+;Nf1tm1Par/+  (g),  and  Plp1-  CreERT2/+;Nf1tm1Par/+ mice injected with tamoxifen at E11.5 (h). The graphs represent mean pixel intensity ± S.E.M. Statistical analysis: p = 0.0218 for wildtype v/s Plp1CreER/+;Nf1tm1Par/+.  76  Epidermis  Dermis  a  Hair follicles  b  c  e  f  Dct-LacZ/+  Plp1-CreER/+;  d  Rosa26-floxed stop-LacZ  Figure 4.4 Fate mapping of cells expressing Plp1-CreER at E11.5 LacZ expressing cells in P5 dermis (a,d), epidermis (b,e) and hair follicles (c,f) of Dct-LacZ/+ mice (top) or Plp1-CreERT2/+; Rosa26-floxed stop-LacZ mice injected with tamoxifen at E11.5 (bottom). Arrows in e show faintly blue stained cells, which were only rarely seen in the epidermis of Plp1-CreER/+;Rosa26-floxed stop-LacZ mice.  77  a  +/+;Rosa26-floxed stop-LacZ/+  b  S100a4-Cre/+;Rosa26-floxed stop-LacZ/+ E D  Tail E skin D  *  Figure 4.5: S100a4-Cre is widely expressed in the skin Expression of the S100a4-Cre in the skin was assessed by crossing the S100a4-Cre/+ mice with Rosa26-floxed stop-LacZ mice. X-gal staining of tail skin showed expression of LacZ (b) in fibroblasts (arrow) in the dermis, keratinocytes (arrowhead) in the epidermis and hair follicles (asterisk) of S100a4-Cre/+; Rosa26-floxed stop-LacZ/+ animals. No LacZ-positive cells were observed in mice without S100a4-Cre (a). E: epidermis; D: dermis.  78  4.3 DISCUSSION We found a surprising difference in the effects of heterozygous and homozygous knockout of the floxed Nf1 allele using the Mitf-Cre transgene. Mitf is a transcription factor that is critical for melanocyte cell-fate choice during cell commitment. In our quantitative assays, Mitf-Cre induced a strong phenotype in homozygotes, but no detectable phenotype in heterozygotes. Because we found that neurofibromin haploinsufficiency in bipotential Schwann cellmelanoblast  precursors  causes  dermal  hyper-pigmentation,  we  hypothesize  that  haploinsufficiency must occur prior to melanocyte cell fate commitment to cause darker skin. Adameyko et al. suggest that melanoblasts differentiate from the Schwann cell precursor cells with reduced nerve contact and thus reduced Neuregulin/ErbB3 signaling. Possibly, neurofibromin protein levels modulate cell fate choices or proliferation of Schwann cell precursor cells. In the future, it would be interesting to investigate whether Nf1 haploinsufficiency alters the number of Schwann cell precursors or the number of differentiated Schwann cells during development.  Experimentally cutting adult mouse sciatic nerve induces pigmentation “streaks” in the dermis, a phenomenon that is enhanced in Nf1 heterozygous mutant mice, which also sometimes develop neurofibroma-like tumors at the lesion (Rizvi et al. 2002). Labeled Nf1 +/- Schwann cells, but not +/+ Schwann cells, grafted into a Nf1 +/- cut nerve environment produced pigmented cells after 1 month, suggesting that Nf1 haploinsufficient Schwann cells might more easily dedifferentiate into bipotential Schwann cell-melanoblast precursors (Rizvi et al. 2002; Adameyko et al. 2009).  79  Melanoblasts migrate first through the dermis and then in the epidermis to reach all areas of the body to pigment the skin and hair. Melanoblasts are first seen in the epidermis at E12.5 and in hair follicles (epidermal appendages) at E16.5. We used LacZ to fate map the cells in the postnatal tail skin that expressed Plp1-creER at E11.5, when tamoxifen was injected. We observed many cells expressing LacZ in the dermis and hair follicles, but almost no cells expressing LacZ in the inter-follicular epidermis. Since migrating melanoblasts seem attracted to hair follicles, it could be that the first wave of epidermal melanoblasts mostly end up in hair follicles, leaving later waves of melanoblasts to populate the inter-follicular epidermis. To address this hypothesis, we could fate map Plp1-creER expressing cells following tamoxifen injection on day E9.5, E10.5, E12.5, E13.5, E14.5, E15.5 or E16.5. It also might be interesting to investigate whether Schwann cells (or another Plp1-creER expressing cell) could give rise to epidermal melanocytes in adults, by providing Plp1-CreER/+ ; Rosa26-floxed stop-LacZ/+ mice with tamoxifen in their drinking water and, after several months, examining whether there are any LacZ-positive cells in the inter-follicular tail epidermis.  Neurofibromatosis is one of the most common genetic diseases, affecting ~1:3500 individuals. While the pigmentary alterations of neurofibromatosis are tolerable, the developmental connection between melanocytes and Schwann cells makes it important to understand the role of neurofibromin in both. We find that neurofibromin protein levels play a specific role in Schwann cell-melanoblast precursors, different from the effects of complete neurofibromin loss.  80  CHAPTER 5: EPISTATIC INTERACTIONS BETWEEN Nf1 AND Ednrb  5.1 INTRODUCTION We hypothesized in chapter 4 that Nf1 haploinsufficiency increases the number of melanoblasts that differentiate from Schwann cell precursors. If this is the case, Nf1 haploinsufficiency might offset mutations that reduce melanoblast survival or proliferation during development. To test this, we decided to study the genetic interactions between Nf1 and Ednrb. Ednrb is a seven transmembrane G protein coupled receptor. It is expressed in melanocytes, where it is activated by the Endothelin 3 ligand (Reid et al. 1996). Mice with mutations in Edn3 or Ednrb have a very similar phenotype, but Ednrb loss has a greater effect (Baynash et al. 1994). Ednrb -/- mice exhibit completely unpigmented tail skin and a predominantly white (unpigmented) coat. Occasionally, Ednrb -/- mice exhibit pigmented coat spots on the head and rump.  E10.5 Ednrb -/- embryos exhibit greatly reduced numbers of Dct-expressing cells, suggesting that there is an early requirement for Ednrb in melanoblast development (Pavan and Tilghman 1994). Over-expression of Endothelin 3 in epidermal keratinocytes has the interesting effect of causing dermal hyper-pigmentation, but not epidermal hyper-pigmentation (Garcia et al. 2008). In this inducible mouse model, dermal hyper-pigmentation dissipates when the Keratin 5 promoter driven Endothelin 3 expression is shut off. This suggests that endothelin signaling might regulate the density of melanocytes in the adult dermis. In another report, it was shown that Ednrb knockout after E12.5, when melanoblasts enter the epidermis, does not produce white fur (Shin et al. 1999). These studies suggest that endothelin signaling has little effect on melanocytes once they are located in the epidermis.  81  In E10.5 Ednrb -/- embryos, some Dct-positive cells remain in the head and rump regions, but they are fewer in number than in wildtype embryos (Pavan and Tilghman 1994). Presumably, Ednrb -/- mice exhibit pigmented spots in the head and rump because there is a reduced need for Ednrb in those areas. Melanoblasts have a very high proliferative potential (Wilkie et al. 2002). Once the rare survivors in Ednrb -/- embryos reached the epidermis, they would be unfettered and could expand.  82  5.2 RESULTS  5.2.1 Determination of the Ednrbs-l deletion breakpoints The Ednrbs-l allele arose spontaneously in 1959 (Lane 1966). According to southern blot analysis, the entire coding region of the Ednrb gene is deleted in the Ednrbs-l allele (Hosoda et al. 1994). We obtained a commercially available strain of Ednrbs-l mice (SSL/Le) from Jackson Laboratories. To simplify the genotyping of these mice, and to verify that no other known genes are included in the deletion, we first located the Ednrbs-l deletion breakpoints.  Our strategy was to determine where PCR either did, or did not, produce a product in DNA from Ednrbs-l/Ednrbs-l mice, using DNA from wildtype mice as a positive control. We designed our PCR primer pairs using sequence from the Ensembl mouse genome database (m37). Beginning 20 Mb away from Ednrb on either side, we sequentially designed primer pairs at denser intervals, targeting the region between the last known primer pair that produced a product and first known primer pair that failed. After several rounds, we were able to narrow this deletion flanking region to ~1 kb on either side of the deletion. Next, we designed several primer pairs in which the left primer was positioned in one deletion flanking region and the right primer was positioned in the other. Some of these primer pairs produced a product in Ednrbs-l/Ednrbs-l DNA, but not in wildtype DNA. We sequenced these PCR products, which contained the breakpoint.  We found that the Ednrbs-l deletion encompasses 97.63 kb of chromosome 14 (Figure 5.1). There is an overlapping GC dinucleotide where the two breakpoints are brought together in the Ednrbs-l allele. According to Ensembl, no other genes are located in the region that is deleted. In our  83  subsequent experiments, we used a deletion spanning PCR assay to detect the Ednrbs-l allele, while a primer pair located within the deletion was used to detect the wildtype allele.  5.2.2 Nf1Dsk9/+ melanocytes require Ednrb for survival in the tail dermis To study the genetic interactions between Nf1 and Ednrb, we crossed Ednrbs-l/+; Nf1Dsk9/+ mice to Ednrbs-l/+ mice. In the resulting progeny, we examined the coat and tail pigmentation.  First, we measured the mean pixel intensity of split tail dermis at 2-3 weeks of age. Interestingly, we found that Ednrbs-l/+ mice have reduced tail dermis pigmentation (Figure 5.2a,b). This hypopigmentation is offset by the addition of Nf1Dsk9/+, which restores pigmentation to wildtype levels. This result shows that heterozygous mutations in these two genes can act additively.  As expected, the tail dermis of Ednrbs-l/Ednrbs-l mice is completely unpigmented (Figure 5.2a,b). The tail dermis of Ednrbs-l/Ednrbs-l; Nf1Dsk9/+ mice is also unpigmented. This indicates that Nf1Dsk9/+ melanocytes in the tail dermis continue to require Ednrb for survival.  Surprisingly, we discovered that Ednrbs-l/Ednrbs-l; Nf1Dsk9/+ mice exhibit pigmentation of a greater area of the coat (Figure 5.2c,d). The coat pigmentation was restricted to the head and rump, where spots of coat pigmentation are sometimes observed in Ednrbs-l/Ednrbs-l mice. Thus, we conclude that the melanoblasts with a decreased need for Ednrb are among those affected by Nf1Dsk9.  84  Figure 5. 1 The Ednrbs-l deletion is 97.63 kb, encompassing the entire Ednrb gene The Ednrbs-l deletion breakpoints are 15.72 kb upstream and 54.98 kb downstream of the Ednrb gene. Primer sets 63 and 49 fall within the deletion, and fail to amplify in Ednrbs-l/Ednrbs-l DNA. Primer pairs 64 and 40 lie in the deletion flanking region on either side. Primer set 89 produces a product only in Ednrbs-l/Ednrbs-l DNA, while primer set 59 produces a product only in wildtype DNA. These were used for genotyping, as shown in the table.  85  a  +/+  Dermal pigmentation (Mean pixel intensity)  b  Nf1Dsk9/+  Nf1Dsk9/+; Ednrbs-l/+  +/+; Ednrbs-l/+  +/+; Ednrbs-l/Ednrbs-l  Nf1Dsk9/+; Ednrbs-l/Ednrbs-l  200  180  160  140  120  100 +/+ n=3  Nf1Dsk9/+ n=4  Nf1Dsk9/+; Ednrbs-l/+ n=4  +/+; Ednrbs-l/+ n=4  c +/+; Ednrbs-l/Ednrbs-l  Nf1Dsk9/+; Ednrbs-l/Ednrbs-l  Fraction of coat that is pigmented (Mean % area)  d  +/+; Ednrbs-l/Ednrbs-l n=5  Nf1Dsk9/+; Ednrbs-l/Ednrbs-l n=6  20  15  10  5  0 +/+; Ednrbs-l/Ednrbs-l n=5  Nf1Dsk9/+; Ednrbs-l/Ednrbs-l n=6  Figure 5.2 Genetic interactions between Ednrb and Nf1 (a) Tail dermis of representative animals and (b) mean pixel intensity of the tail dermis. (c) Nf1Dsk9/+; Ednrbs-l/Ednrbs-l mice (bottom row) exhibit increased amounts of coat pigmentation 86  compared to Ednrbs-l/Ednrbs-l mice (top row). (d) Mean percent area of the coat that is pigmented. Graphs represent mean value ± S.E.M. Statistical analysis: p = 0.003 for +/+;Ednrbsl  /+ v/s Nf1Dsk9/+;Ednrbs-l/+ in b, p = 0.0022 for percent coat pigmentation of +/+;Ednrbs-  l  /Ednrbs-l v/s Nf1Dsk9/+;Ednrbs-l/Ednrbs-l.  87  5.3 DISCUSSION We have found that Nf1Dsk9/+ requires Ednrb to cause a darker dermis. Ednrb might be required for the specification of melanoblasts from Schwann cell precursors, or for the ongoing survival of melanocytes in the tail dermis. Adameyko et al. (2009) hypothesized that melanoblasts in the trunk differentiate from Schwann cell precursor cells with reduced nerve contact and thus reduced Neuregulin/ErbB3 signaling. Interestingly, endothelin blocks the effects of Neuregulin in vitro (Brennan et al. 2000). The experiments of Garcia et al. (2008) suggest that endothelin signaling regulates the density of melanocytes in the adult dermis.  Melanocytes in the head and rump seem to have a reduced requirement for Ednrb compared to melanocytes in the trunk. This could be due to environmental differences, and/or to differences in the origin of the melanoblasts themselves. The trunk dermis is derived from the somatic dermamyotome, while the head dermis is derived from the cranial neural crest. In addition, the origins of melanoblasts in the head and rump have not yet been investigated. We have noticed that at E12.5, many Dct-LacZ positive cells seem to lie along the path of cranial nerves in the head, so these might have arisen from Schwann cell precursors. If so, one copy of Nf1Dsk9 may increase their numbers, allowing a greater number to enter the epidermis and give rise to spots.  In the future, we will generate Plp1-CreER/+; Nf1tm1Par/+; Ednrbs-l/Ednrbs-l mice and examine them to determine whether Nf1 haploinsufficiency in Schwann cell precursors increases the area of the coat that is pigmented. If it does, we can also use this tamoxifen inducible system to determine when during development these melanocytes arise.  88  CHAPTER 6: GENERAL DISCUSSION  6.1 SUMMARY OF RESULTS AND CONCLUSIONS This research project began with the finding that Dark skin 9 mice have a point mutation in the neurofibromin gene, which also causes darker skin and neurofibromas in humans with neurofibromatosis type 1. In this thesis, we sought to better understand the mechanisms by which Nf1 regulates pigment cells and causes skin hyper-pigmentation when mutated in mice.  6.1.1 Evidence supporting the Nf1N1453K mutation as causative in Dsk9 mice Dsk9 is unique in that it is the only Dsk mutation that causes both dermal and epidermal hyperpigmentation (Fitch et al. 2003). We have found that Nf1Dsk9 mutant mice are very similar to mice with targeted Nf1 mutant alleles: Nf1tm1Tyj, Nf1tm1Par, and Nf1tm1Fcr. First, like Nf1Dsk9/+ mice, ACTB-cre/+; Nf1tm1Par/+ mice have a darker dermis and epidermis, while Nf1tmTyi/+ mice have a darker dermis (the epidermis was not examined in Nf1tm1Tyj/+ mice). Nf1Dsk9/Nf1Dsk9, Nf1tm1Tyj / Nf1tm1Par (with knockout in all cells), and Nf1tm1Tyj /Nf1tm1Tyj mice all have the same gross morphology at E12.5 (edema, hemorrhage and small, hypo-pigmented eyes). Nf1Dsk9/Nf1Dsk9 and Nf1tm1Fcr/ Nf1tm1Fcr E11.5 embryos each exhibit increased numbers of melanoblasts and ectopic localization of Dct-positive cells adjacent to the dorsal neural tube.  From these similarities, we operated on the assumption that the Nf1N1453K mutation is causative for darker skin in Dsk9 mice. However, even though it would be a great coincidence, there could be a second mutation linked to Nf1 in Dsk9 mice that causes the darker skin instead of Nf1N1453K. One further test we could do would be to cross ACTB-Cre/+; Nf1tm1Par/+ mice to Nf1Dsk9/+ mice 89  and look to see if the resulting Nf1recombined  tm1Par  /Nf1Dsk9 embryos die at E12.5 (i.e. a  complementation test). Similar to this approach, we found that Mitf-Cre/+; Nf1tm1Par/Nf1Dsk9 animals exhibit hyper-pigmentation in excess of Nf1Dsk9/+ mice. Since Mitf-Cre/+; Nf1tm1Par/+ mice exhibit no skin darkening, this suggests that Nf1Dsk9 and Nf1tm1Par cause darker skin through mutation of the same gene.  6.1.2 Evidence suggesting release of Ras inhibition causes darker skin in Nf1 mutants Several lines of evidence suggest that the darker skin caused by the loss of neurofibromin is a result of increased signaling through the Map kinase pathway.  In addition to neurofibromatosis type 1, multiple CALM’s are a feature of Legius syndrome, which is caused by loss of function mutations in SPRED1, another negative regulator of MAPK signaling (Brems et al. 2007). Transgenic mice expressing activated HA-RAS in melanocytes exhibit both epidermal and dermal hyper-pigmentation, the only darker skinned transgenic mice to do so (Powell et al. 1995). The Nf1Dsk9 mutation is in the GRD, which is known to negatively regulate RAS activity (Ballester et al. 1990) and the highly conserved residue altered in Dsk9, Asn1453, was isolated in mutagenic screening as important for RAS interaction (Morcos et al. 1996).  Hyperactivity of the MAPK pathway is associated with increased proliferation, increased cell survival and anchorage independent growth in melanocytes.  The MAPK pathway can be 90  activated downstream of endothelin and constitutively active Gq/11 signaling (Imokawa et al. 1996), HGF/c-Met signaling (Ye et al. 2008) and c-kit signaling (Imokawa et al. 2000) (see Figure 6.1), and a growing number of individual components of the MAPK pathway have been found to be mutated in benign nevi and malignant melanoma (Stark et al. 2011).  The mechanisms by which excessive endothelin and c-kit signaling cause only dermal and epidermal hyper-pigmentation, respectively, are not understood, and are a topic of investigation in our lab. Although endothelin may signal through more than one kind of G protein alpha subunit, we found that forced expression of a constitutively active GNAQQ209L transgene in mouse melanocytes causes only dermal hyper-pigmentation, as does Edn3 over-expression by keratinocytes. This shows that the outcome of GNAQ activity is somehow limited within the epidermis compared to the dermis. Similarly, in humans, somatic GNAQ and GNA11 constitutively active mutations have only been found in melanocytic lesions of the dermis and eye, not in lesions located in the epidermis. In the future, we will use our GNAQQ209L transgenic mice to compare MAPK activity in the tail dermis versus the epidermis.  In light of the dermal and epidermal specific effects of c-kit and endothelin signaling, it is interesting that RAS hyper-activation, through direct mutation of the GTPase domain, or perhaps through the loss of neurofibromin, causes both dermal and epidermal hyper-pigmentation. It is possible that Ras itself is normally activated downstream of c-kit and endothelin signaling, however, the reported connection between endothelin signaling and MAPK pathway is through phosphorylation of Raf by PKC. Since Raf is downstream of RAS in 91  Figure 6. 2 Signaling pathways causing dermal and epidermal hyper-pigmentation. Mitf-Cre/+; Nf1tm1Par/Nf1tm1Par mice exhibit both dermal and epidermal hyper-pigmentation. Dermal hyper-pigmentation is observed after over-expression of endothelin 3 (Edn3) or hepatocyte growth factor (HGF), while epidermal hyper-pigmentation is observed after overexpression of kit ligand (SCF). Edn3, HGF and SCF signaling can feed into the Map kinase pathway.  92  the MAPK pathway, it could be that a gain of function in mutant RAS leads to neomorphic MAPK activation, by-passing activation by endothelin or c-kit signaling.  6.1.3 Nf1 haploinsufficiency affects melanocytes of the glial lineage We found a surprising difference in the effects of heterozygous and homozygous knockout of the floxed Nf1 allele using the Mitf-Cre transgene. Mitf is a transcription factor that is critical for melanocyte cell-fate choice during cell commitment (Widlund and Fisher 2003). In our assays, Mitf-cre induced a strong phenotype in homozygotes (in excess of germline Nf1 +/- mutants), but no detectable phenotype in heterozygotes. This suggested to us that homozygous and heterozygous loss of Nf1 is different, even while they both cause hyper-pigmentation.  Adameyko et al, following the same tamoxifen injection dose and schedule as we did, reported that 66% of hair follicle melanocytes express Plp1-CreER at E11.5. At E11.5, Plp1-CreER expressing Schwann cell precursors cluster around the dorsal root ganglia in the trunk, while Cre-negative, Dct-positive, Mitf-positive melanoblasts are located close to, but not in direct contact with these peripheral nerves. Nf1Dsk9 heterozygotes have an increased number of Dctpositive cells adjacent to the dorsal root ganglia at E12.5, which is just a day after tamoxifen injection was able to cause a darker dermis in Plp1-CreER/+; Nf1tm1Par/+ mice. Tamoxifen injection at E9.5 did not cause skin darkening and Nf1Dsk9/+ embryos do not exhibit an increase in Dct-positive cells at E10.5. Furthermore, Nf1Dsk9/+ is able to increase the percent of the coat that is pigmented in Ednrbs-l/Ednrbs-l animals, suggesting that Nf1 haploinsufficiency increases the number of melanoblasts during development. Altogether, this data has led us to hypothesize 93  that Nf1 haploinsufficiency causes skin hyper-pigmentation by increasing the number of melanoblasts that are produced from Schwann cell precursors (Figure 6.2).  By labeling Plp1-CreER expressing cells and following their progeny in vivo, Delaunay et al. demonstrated that neurogenic progenitor cells express Plp1-CreER at E9.5 are distinct from gliogenic Plp1-CreER-positive progenitor cells at E13.5 (Delaunay et al. 2008). The fact that we did not observe skin darkening with tamoxifen injection at E9.5 indicates that the cells expressing Plp1-CreER at E9.5 do not give rise to melanocyte precursors affected by Nf1 loss.  Experimentally cutting adult mouse sciatic nerve induces pigmentation “streaks” in the trunk dermis. More streaks form in Nf1 +/- mice compared to wildtype mice (Rizvi et al. 2002). Labeled Nf1 +/- Schwann cells, but not +/+ Schwann cells, grafted into a Nf1 +/- cut nerve environment produced pigmented cells after 1 month, suggesting that Nf1 haploinsufficient Schwann cells might more easily dedifferentiate into bipotential Schwann cell-melanoblast precursors (Rizvi et al. 2002; Adameyko et al. 2009). Labeled Nf1 +/- Schwann cells were not able to form pigmented cells in an Nf1 +/+ cut nerve environment, suggesting that Nf1 haploinsufficiency in another cell type might enhance melanocyte survival. In the future, we could compare “streaking” after sciatic nerve cutting in S100a4-Cre/+; Krox20-Cre/+; Nf1tm1Par/+ and Krox20-Cre/+; Nf1tm1Par/+ mice, to search further for a role for neurofibromin in paracrine signaling between fibroblasts and melanocytes.  94  Figure 6.2 Haploinsufficiency of Nf1 causes an expansion of Schwann cell precursor derived melanoblasts during embryogenesis 95  6.1.4 Comparison of effect of loss of neurofibromin on pigmentation in mice and humans While humans normally lack dermal pigmentation, there are still several parallels between the Nf1 mouse phenotypes we have observed and the pigmentation of individuals with neurofibromatosis type 1 (Table 6.1). First, individuals with neurofibromatosis type 1 have mild, generalized epidermal hyperpigmentation (Boyd et al. 2009). We have found that Nf1 heterozygous mice have epidermal hyper-pigmentation of the tail skin. In mosaic neurofibromatosis type 1, melanocytes cultured from regions of generalized skin hyper-pigmentation have a single hit in NF1, while the dermal fibroblasts are wildtype (Maertens et al. 2007). Similarly, S100a4-Cre/+; Nf1tm1Par/+ mice do not exhibit a darker epidermis. Multiple Café au lait macules (CALMs) are one of the cardinal diagnostic criteria of neurofibromatosis (De Schepper et al. 2005). CALMs are epidermal, tan-brown, clearly demarcated macules with smooth boarders and an increased melanocyte density, and they usually develop during early childhood (De Schepper et al. 2006). In neurofibromatosis type 1, the CALMs are darker than the surrounding hyper-pigmented skin. Melanocytes cultured from CALMs have been shown to have a second hit in NF1 (Maertens et al. 2007; De Schepper et al. 2008). Thus, the epidermal hyper-pigmentation of Mitf-Cre/+; Nf1tm1Par/Nf1tm1Par mice may serve as a mouse model for CALMs in the future. As with CALMs, which are typically benign, the Mitf-Cre/+; Nf1tm1Par/Nf1tm1Par mice did not exhibit melanocytic tumors. In neurofibromatosis type 1, skin hyper-pigmentation can overlie superficial plexiform neurofibromas, and can appear similar to giant congenital melanocytic nevi (Boyd et al. 2009). Rarely, neurofibromas are also laden with melanin containing cells (Fetsch et al. 2000). Wu et al  96  DERMIS  Nf1 +/- effects  Humans   ? Pigmentation overlying plexiform neurofibromas  Mice (trunk)  No hyper-pigmentation  Nf1 -/- effects ?  Hyper-pigmentation in the neck region (Mitf-Cre) and hyper-pigmentation overlying plexiform neurofibromas (Dhh-Cre)  Mice (ear, tail,  Generalized hyper-pigmentation  (Mitf-Cre)  feet)  EPIDERMIS  Generalized hyper-pigmentation  Nf1 +/- effects  Nf1 -/- effects  Humans  Generalized hyper-pigmentation  CALMs  Mice (trunk)  No hyper-pigmentation  No hyper-pigmentation  Mice (ear, tail,  Generalized hyper-pigmentation  Generalized hyper-pigmentation  feet)  (Mitf-Cre)  Table 6. 1 Comparison of the effects of neurofibromin mutations on pigmentation in humans and mice.  97  knocked out both copies of Nf1 in Schwann cell precursors using a Desert hedgehog-Cre (DhhCre) expressing transgene, and found that 39% of the mice exhibited hyper-pigmentation overlying the plexiform neurofibromas/spinal cord (Wu et al. 2008). Similar pigmentation was not reported for neurofibromas in Plp1-CreER homozygous knockout mice, when tamoxifen was administered either perinatally or in adulthood, suggesting that perhaps Plp1-CreER expressing cells in adults do not produce melanocytes (Mayes et al.; Mayes et al. 2010). We found that homozygous knockout of Nf1 in committed melanocytes increases trunk skin pigmentation, indicating that Nf1-null melanocytes have an enhanced ability to survive/proliferate in the trunk dermis, which normally has few melanocytes.  6.1.5 Comparison of different alleles of Nf1 normalized to wildtype Since our measurements of average pixel intensity of the tail dermis and epidermis are relative values, we decided to calculate the percent increase of the mutants over wildtype in each experiment to be able to compare the different Nf1 mutants to each other (Table 6.2). The caveat to bear in mind with this analysis is that the wildtypes differ from cross to cross with regards to the exact genetic makeup and the age at which the mice were sacrificed, so the percent increase of the mutants may not be directly comparable between crosses. However, the analysis reveals a couple of interesting observations.  98  Genotype  Cell type(s) targeted  Dermis  Epidermis  Nf1Dsk9/+  All  13% (+/-2%)  9% (+/-4%)  Nf1tm1Tyj /+  All  24% (+/-7%)  *  ACTB-Cre/+; Nf1tm1Par/+  All  22% (+/-8%)  13% (+/-3%)  Mitf-Cre/+; Nf1tm1Par/Nf1tm1Par  Melanocytes  70% (+/-8%)  14% (+/-6%)  Plp1-CreER/+; Nf1tm1Par/+  Schwann cell precursors 8% (+/-3%) expressing Plp1-CreER at E11.5  None  Melanocytes  None  None  keratinocytes None S100a4-Cre/+; Mitf- Melanocytes, tm1Par and fibroblasts Cre/+; Nf1 /+  None  Vavl-Cre/+; Mitf-Cre/+; Melanocytes hematopoietic cells Nf1tm1Par/+  None  (E11.5 injection)  tamoxifen  Mitf-Cre/+; Nf1tm1Par/+  and None  Table 6. 2 The percent increase in dermal and epidermal average pixel intensity over wildtype for various Nf1 mutants. *Not determined.  99  First, considering the dermis, we see that Nf1Dsk9/+ mice have a 13% increase in average pixel intensity, while Nf1tm1Tyj /+ and ACTB-Cre/+; Nf1tm1Par/+ mice have a 24% and 22% increase, respectively. This suggests that the Nf1Dsk9 mutation might be a hypomorph, which is not surprising given that it is a missense mutation. However, it is embryonic lethal, so it presumably has a significant impact on neurofibromin activity. Second, Plp1-CreER/+; Nf1tm1Par/+ mice have an 8% increase in average pixel intensity of the dermis, compared to a 22% increase in ACTB-Cre/+; Nf1tm1Par/+mice. This suggests that either some dermal melanocytes were not targeted with a single tamoxifen injection at E11.5 or that another cell type besides melanocytes contributes to the Nf1 +/- dark skin phenotype. A Plp1Cre line (not CreER) is needed to distinguish between these two possibilities.  Third, we see that Mitf-Cre/+; Nf1tm1Par/Nf1tm1Par and ACTB-Cre/+; Nf1tm1Par/+ mice have about the same increase in epidermal pixel intensity (14% versus 13%), but Mitf-Cre/+; Nf1tm1Par/Nf1tm1Par mice have a much greater increase in dermal pixel intensity (70% versus 22%). This could mean that there is an upper limit to the ratio of keratinocytes to melanocytes in the mouse tail epidermis. It would be interesting to see whether Nf1 +/- knockout in all cells plus Nf1 -/- knockout in melanocytes would result in even greater epidermal hyper-pigmentation. If so, this could indicate a role for other cell types in the etiology of CALMs.  6.1.6 Treatment options for pigmentary manifestations in NF1 patients Non-uniform skin pigmentation, such as birthmarks, can cause significant emotional distress and adverse psychological outcomes due to discrimination, particularly in certain cultures. Short100  pulsed lasers can be used to selectively destroy melanosomes, which can achieve cosmetic improvement for CALMs. Total clearing of CALMs is obtained in approximately 50% of patients undergoing laser therapy. Re-pigmentation and patchy re-pigmentation occur in the remaining patients. Laser therapy for CALMs is generally safe, however, there is a risk of purpura and postoperative abradement of the treated area. Laser therapy is also not advised for darker skin types, because post-inflammatory hypo-pigmentation is a risk. Thus, a more effective treatment option is desirable (Stratigos et al. 2000).  The goal of treatment for neurofibromatosis type 1 CALMs is not to eliminate all melanocytes in the macule, since re-pigmentation of skin can be slow and irregular. Selective removal of the Nf1 -/- melanocytes in the CALMs would be ideal. If we could devise a method to culture Nf1 -/melanocytes, then we could test whether any of the currently available MAPK inhibitors leads to apoptosis of Nf1 -/- melanocytes, but not Nf1 +/- melanocytes. For example, AZD 6244 and temozolomide block the MAPK/ERK pathway and are currently in a randomized phase II trial for melanoma (Board et al. 2009). A topical application of any potential therapeutic agent might have fewer side effects than systemic treatment.  6.2 FUTURE DIRECTIONS In addition to the experiments described in the discussion sections of chapters 3, 4, 5 and 6, we also propose the following experiments to further investigate the role of neurofibromin in pigmentation.  101  6.2.1 Determine whether Nf1 haploinsufficiency alters Schwann cell development SCP’s give rise to both Schwann cells and melanocytes. Haploinsufficiency Nf1 in Schwann cell precursors causes dermal hyper-pigmentation in adult mice. Alteration in certain signaling pathways such as the Wnt/β-catenin pathway (Ikeya et al. 1997; Dorsky et al. 1998) or ErbB3 signaling pathway (Adameyko et al. 2009) can cause an expansion in either melanocytes or Schwann cells, with a concurrent reduction in the other cell type. Thus, it would be interesting to determine whether Nf1 haploinsuffiency causes an expansion of melanoblasts at the expense of Schwann cells. To test this, we could examine the expression of Sox10, Plp1, Krox20 and/or Dhh at the dorsal root ganglia at E11.5 and E12.5 in Nf1 +/- embryos.  6.2.2 Determine whether SCP-derived melanoblasts in the head have a reduced requirement for Ednrb We observed that Nf1Dsk9/+; Ednrbs-l/Ednrbs-l mice exhibit an increased percentage of pigmented coat compared to Ednrbs-l/Ednrbs-l mice. We are curious about the source of the pigmented spots. It appears that melanoblasts in the head region have a reduced requirement for Ednrb. Because they are able to respond to Nf1Dsk9 haploinsufficiency, they may be derived from Schwann cell precursors. To test this, we will measure the percent of pigmented coat in Plp1-CreER/+; Nf1tm1Par/+; Ednrbs-l/Ednrbs-l embryos treated with tamoxfien at E11.5.  6.2.3 Determine when during development Nf1 -/- loss can cause skin hyper-pigmentation To determine when homozygous Nf1 loss can cause skin hyper-pigmentation, we will cross Nf1tm1Par mice to Tyr-CreER mice, which express CreER under the control of the tyrosinase promoter. We will induce CreER activity at 3 weeks of age in Tyr-CreER/+; Nf1tm1Par/Nf1tm1Par mice by providing the mice with tamoxifen containing drinking water for one month after 102  weaning. We will sacrifice the mice after 4 more months to examine the pixel intensity of the tail dermis and epidermis and the histopathology of the trunk skin.  6.3 CONCLUSIONS Neurofibromatosis is one of the most common genetic diseases, affecting ~1:3500 individuals. While the pigmentary alterations of neurofibromatosis are not life threatening, the developmental connection between melanocytes and Schwann cells requires that we understand the role of neurofibromin in both. We find that neurofibromin protein levels play an early role in the glial lineage of melanoblasts, different from the effects of complete neurofibromin loss. We now have mouse resources with which to study the etiology of CALMs. We have identified a new mouse mutant, Nf1Dsk9, which has a point mutation in the GRD, and could be used to distinguish between phenotypes that result from a decrease in GAP activity versus the other activities of neurofibromin. We have begun to tease apart the differences between melanoblasts in the head versus the trunk, and the epidermis versus the dermis. Melanocytes have a fascinating and complex developmental biology that has intrigued geneticists for many years. Neurofibromin plays a central role in this process, as evidenced by skin-wide hyper-pigmentation in Nf1 mouse mutants. Much remains to be discovered about how the Map kinase pathway, a central melanoma axis, is regulated in melanocytes.  103  References Adameyko, I., Lallemend, F., Aquino, J.B., Pereira, J.A., Topilko, P., Muller, T., Fritz, N., Beljajeva, A., Mochii, M., Liste, I., Usoskin, D., Suter, U., Birchmeier, C., and Ernfors, P. 2009. Schwann cell precursors from nerve innervation are a cellular origin of melanocytes in skin. Cell 139(2): 366-379. Alizadeh, A., Fitch, K.R., Niswender, C.M., McKnight, G.S., and Barsh, G.S. 2008. Melanocyte-lineage expression of Cre recombinase using Mitf regulatory elements. Pigment Cell Melanoma Res 21(1): 63-69. Alper, J.C. and Holmes, L.B. 1983. The incidence and significance of birthmarks in a cohort of 4,641 newborns. Pediatr Dermatol 1(1): 58-68. Amer, M., Mostafa, F.F., and Nasr, A.N. 2001. Lentiginous macules and patches of neurofibromatosis (an approach to better terminology). J Eur Acad Dermatol Venereol 15(1): 39-42. Anderson, B. and Robertson, D.M. 1979. Melanin containing neurofibroma: case report with evidence of Schwann cell origin of melanin. Can J Neurol Sci 6(2): 139-143. Bajenaru, M.L., Zhu, Y., Hedrick, N.M., Donahoe, J., Parada, L.F., and Gutmann, D.H. 2002. Astrocyte-specific inactivation of the neurofibromatosis 1 gene (NF1) is insufficient for astrocytoma formation. Mol Cell Biol 22(14): 5100-5113. Ballester, R., Marchuk, D., Boguski, M., Saulino, A., Letcher, R., Wigler, M., and Collins, F. 1990. The NF1 locus encodes a protein functionally related to mammalian GAP and yeast IRA proteins. Cell 63(4): 851-859. Baynash, A.G., Hosoda, K., Giaid, A., Richardson, J.A., Emoto, N., Hammer, R.E., and Yanagisawa, M. 1994. Interaction of endothelin-3 with endothelin-B receptor is essential for development of epidermal melanocytes and enteric neurons. Cell 79(7): 1277-1285. Benn, D.E., Dwight, T., Richardson, A.L., Delbridge, L., Bambach, C.P., Stowasser, M., Gordon, R.D., Marsh, D.J., and Robinson, B.G. 2000. Sporadic and familial pheochromocytomas are associated with loss of at least two discrete intervals on chromosome 1p. Cancer Res 60(24): 7048-7051.  104  Bhowmick, N.A., Chytil, A., Plieth, D., Gorska, A.E., Dumont, N., Shappell, S., Washington, M.K., Neilson, E.G., and Moses, H.L. 2004. TGF-beta signaling in fibroblasts modulates the oncogenic potential of adjacent epithelia. Science 303(5659): 848-851. Board, R.E., Ellison, G., Orr, M.C., Kemsley, K.R., McWalter, G., Blockley, L.Y., Dearden, S.P., Morris, C., Ranson, M., Cantarini, M.V., Dive, C., and Hughes, A. 2009. Detection of BRAF mutations in the tumour and serum of patients enrolled in the AZD6244 (ARRY142886) advanced melanoma phase II study. Br J Cancer 101(10): 1724-1730. Bollag, G. and McCormick, F. 1991. Differential regulation of rasGAP and neurofibromatosis gene product activities. Nature 351(6327): 576-579. Bonneau, F., D'Angelo, I., Welti, S., Stier, G., Ylanne, J., and Scheffzek, K. 2004. Expression, purification and preliminary crystallographic characterization of a novel segment from the neurofibromatosis type 1 protein. Acta Crystallogr D Biol Crystallogr 60(Pt 12 Pt 2): 2364-2367. Bos, J.L., Rehmann, H., and Wittinghofer, A. 2007. GEFs and GAPs: critical elements in the control of small G proteins. Cell 129(5): 865-877. Boyd, K.P., Korf, B.R., and Theos, A. 2009. Neurofibromatosis type 1. J Am Acad Dermatol 61(1): 1-14; quiz 15-16. Brannan, C.I., Perkins, A.S., Vogel, K.S., Ratner, N., Nordlund, M.L., Reid, S.W., Buchberg, A.M., Jenkins, N.A., Parada, L.F., and Copeland, N.G. 1994. Targeted disruption of the neurofibromatosis type-1 gene leads to developmental abnormalities in heart and various neural crest-derived tissues. Genes Dev 8(9): 1019-1029. Brems, H., Chmara, M., Sahbatou, M., Denayer, E., Taniguchi, K., Kato, R., Somers, R., Messiaen, L., De Schepper, S., Fryns, J.P., Cools, J., Marynen, P., Thomas, G., Yoshimura, A., and Legius, E. 2007. Germline loss-of-function mutations in SPRED1 cause a neurofibromatosis 1-like phenotype. Nat Genet 39(9): 1120-1126. Brennan, A., Dean, C.H., Zhang, A.L., Cass, D.T., Mirsky, R., and Jessen, K.R. 2000. Endothelins control the timing of Schwann cell generation in vitro and in vivo. Dev Biol 227(2): 545-557. Britsch, S., Goerich, D.E., Riethmacher, D., Peirano, R.I., Rossner, M., Nave, K.A., Birchmeier, C., and Wegner, M. 2001. The transcription factor Sox10 is a key regulator of peripheral glial development. Genes Dev 15(1): 66-78. 105  Britsch, S., Li, L., Kirchhoff, S., Theuring, F., Brinkmann, V., Birchmeier, C., and Riethmacher, D. 1998. The ErbB2 and ErbB3 receptors and their ligand, neuregulin-1, are essential for development of the sympathetic nervous system. Genes Dev 12(12): 1825-1836. Cichowski, K. and Jacks, T. 2001. NF1 tumor suppressor gene function: narrowing the GAP. Cell 104(4): 593-604. Cook, A.L., Chen, W., Thurber, A.E., Smit, D.J., Smith, A.G., Bladen, T.G., Brown, D.L., Duffy, D.L., Pastorino, L., Bianchi-Scarra, G., Leonard, J.H., Stow, J.L., and Sturm, R.A. 2009. Analysis of cultured human melanocytes based on polymorphisms within the SLC45A2/MATP, SLC24A5/NCKX5, and OCA2/P loci. J Invest Dermatol 129(2): 392-405. D'Angelo, I., Welti, S., Bonneau, F., and Scheffzek, K. 2006. A novel bipartite phospholipid-binding module in the neurofibromatosis type 1 protein. EMBO Rep 7(2): 174-179. Daston, M.M. and Ratner, N. 1992. Neurofibromin, a predominantly neuronal GTPase activating protein in the adult, is ubiquitously expressed during development. Dev Dyn 195(3): 216-226. de Boer, J., Williams, A., Skavdis, G., Harker, N., Coles, M., Tolaini, M., Norton, T., Williams, K., Roderick, K., Potocnik, A.J., and Kioussis, D. 2003. Transgenic mice with hematopoietic and lymphoid specific expression of Cre. Eur J Immunol 33(2): 314-325. De Schepper, S., Boucneau, J., Lambert, J., Messiaen, L., and Naeyaert, J.M. 2005. Pigment cell-related manifestations in neurofibromatosis type 1: an overview. Pigment Cell Res 18(1): 13-24. De Schepper, S., Boucneau, J., Vander Haeghen, Y., Messiaen, L., Naeyaert, J.M., and Lambert, J. 2006. Cafe-au-lait spots in neurofibromatosis type 1 and in healthy control individuals: hyperpigmentation of a different kind? Arch Dermatol Res 297(10): 439-449. De Schepper, S., Maertens, O., Callens, T., Naeyaert, J.M., Lambert, J., and Messiaen, L. 2008. Somatic mutation analysis in NF1 cafe au lait spots reveals two NF1 hits in the melanocytes. J Invest Dermatol 128(4): 1050-1053. Delaunay, D., Heydon, K., Cumano, A., Schwab, M.H., Thomas, J.L., Suter, U., Nave, K.A., Zalc, B., and Spassky, N. 2008. Early neuronal and glial fate restriction of embryonic neural stem cells. J Neurosci 28(10): 2551-2562. 106  Doerflinger, N.H., Macklin, W.B., and Popko, B. 2003. Inducible site-specific recombination in myelinating cells. Genesis 35(1): 63-72. Dorsky, R.I., Moon, R.T., and Raible, D.W. 1998. Control of neural crest cell fate by the Wnt signalling pathway. Nature 396(6709): 370-373. Dupin, E., Glavieux, C., Vaigot, P., and Le Douarin, N.M. 2000. Endothelin 3 induces the reversion of melanocytes to glia through a neural crest-derived glial-melanocytic progenitor. Proc Natl Acad Sci U S A 97(14): 7882-7887. Dupin, E. and Le Douarin, N.M. 2003. Development of melanocyte precursors from the vertebrate neural crest. Oncogene 22(20): 3016-3023. Ernfors, P. 2010. Cellular origin and developmental mechanisms during the formation of skin melanocytes. Exp Cell Res 316(8): 1397-1407. Feil, R. 2007. Conditional somatic mutagenesis in the mouse using site-specific recombinases. Handb Exp Pharmacol(178): 3-28. Feil, S., Valtcheva, N., and Feil, R. 2009. Inducible Cre mice. Methods Mol Biol 530: 343363. Fetsch, J.F., Michal, M., and Miettinen, M. 2000. Pigmented (melanotic) neurofibroma: a clinicopathologic and immunohistochemical analysis of 19 lesions from 17 patients. Am J Surg Pathol 24(3): 331-343. Fitch, K.R., McGowan, K.A., van Raamsdonk, C.D., Fuchs, H., Lee, D., Puech, A., Herault, Y., Threadgill, D.W., Hrabe de Angelis, M., and Barsh, G.S. 2003. Genetics of dark skin in mice. Genes Dev 17(2): 214-228. Foster, W.J., Fuller, C.E., Perry, A., and Harbour, J.W. 2003. Status of the NF1 tumor suppressor locus in uveal melanoma. Arch Ophthalmol 121(9): 1311-1315. Fuchs, E. 2007. Scratching the surface of skin development. Nature 445(7130): 834-842. Gan, T., Jude, C.D., Zaffuto, K., and Ernst, P. 2010. Developmentally induced Mll1 loss reveals defects in postnatal haematopoiesis. Leukemia 24(10): 1732-1741.  107  Garcia, R.J., Ittah, A., Mirabal, S., Figueroa, J., Lopez, L., Glick, A.B., and Kos, L. 2008. Endothelin 3 induces skin pigmentation in a keratin-driven inducible mouse model. J Invest Dermatol 128(1): 131-142. Gitler, A.D., Zhu, Y., Ismat, F.A., Lu, M.M., Yamauchi, Y., Parada, L.F., and Epstein, J.A. 2003. Nf1 has an essential role in endothelial cells. Nat Genet 33(1): 75-79. Gleason, B.C., Crum, C.P., and Murphy, G.F. 2008. Expression patterns of MITF during human cutaneous embryogenesis: evidence for bulge epithelial expression and persistence of dermal melanoblasts. J Cutan Pathol 35(7): 615-622. Gray-Schopfer, V., Wellbrock, C., and Marais, R. 2007. Melanoma biology and new targeted therapy. Nature 445(7130): 851-857. Guillot, B., Dalac, S., Delaunay, M., Baccard, M., Chevrant-Breton, J., Dereure, O., Machet, L., Sassolas, B., Zeller, J., Bernard, P., Bedane, C., and Wolkenstein, P. 2004. Cutaneous malignant melanoma and neurofibromatosis type 1. Melanoma Res 14(2): 159163. Gutzmer, R., Herbst, R.A., Mommert, S., Kiehl, P., Matiaske, F., Rutten, A., Kapp, A., and Weiss, J. 2000. Allelic loss at the neurofibromatosis type 1 (NF1) gene locus is frequent in desmoplastic neurotropic melanoma. Hum Genet 107(4): 357-361. Han, J., Kraft, P., Nan, H., Guo, Q., Chen, C., Qureshi, A., Hankinson, S.E., Hu, F.B., Duffy, D.L., Zhao, Z.Z., Martin, N.G., Montgomery, G.W., Hayward, N.K., Thomas, G., Hoover, R.N., Chanock, S., and Hunter, D.J. 2008. A genome-wide association study identifies novel alleles associated with hair color and skin pigmentation. PLoS Genet 4(5): e1000074. Hegedus, B., Dasgupta, B., Shin, J.E., Emnett, R.J., Hart-Mahon, E.K., Elghazi, L., BernalMizrachi, E., and Gutmann, D.H. 2007. Neurofibromatosis-1 regulates neuronal and glial cell differentiation from neuroglial progenitors in vivo by both cAMP- and Ras-dependent mechanisms. Cell Stem Cell 1(4): 443-457. Hiatt, K.K., Ingram, D.A., Zhang, Y., Bollag, G., and Clapp, D.W. 2001. Neurofibromin GTPase-activating protein-related domains restore normal growth in Nf1-/- cells. J Biol Chem 276(10): 7240-7245.  108  Ho, I.S., Hannan, F., Guo, H.F., Hakker, I., and Zhong, Y. 2007. Distinct functional domains of neurofibromatosis type 1 regulate immediate versus long-term memory formation. J Neurosci 27(25): 6852-6857. Hosoda, K., Hammer, R.E., Richardson, J.A., Baynash, A.G., Cheung, J.C., Giaid, A., and Yanagisawa, M. 1994. Targeted and natural (piebald-lethal) mutations of endothelin-B receptor gene produce megacolon associated with spotted coat color in mice. Cell 79(7): 1267-1276. Hrabe de Angelis, M., Flaswinkel, H., Fuchs, H., Rathkolb, B., Soewarto, D., Marschall, S., Heffner, S., Pargent, W., Wuensch, K., Jung, M., Reis, A., Richter, T., Alessandrini, F., Jakob, T., Fuchs, E., Kolb, H., Kremmer, E., Schaeble, K., Rollinski, B., Roscher, A., Peters, C., Meitinger, T., Strom, T., Steckler, T., Holsboer, F., Klopstock, T., Gekeler, F., Schindewolf, C., Jung, T., Avraham, K., Behrendt, H., Ring, J., Zimmer, A., Schughart, K., Pfeffer, K., Wolf, E., and Balling, R. 2000a. Genome-wide, large-scale production of mutant mice by ENU mutagenesis. Nat Genet 25(4): 444-447. Hrabe de Angelis, M.H., Flaswinkel, H., Fuchs, H., Rathkolb, B., Soewarto, D., Marschall, S., Heffner, S., Pargent, W., Wuensch, K., Jung, M., Reis, A., Richter, T., Alessandrini, F., Jakob, T., Fuchs, E., Kolb, H., Kremmer, E., Schaeble, K., Rollinski, B., Roscher, A., Peters, C., Meitinger, T., Strom, T., Steckler, T., Holsboer, F., Klopstock, T., Gekeler, F., Schindewolf, C., Jung, T., Avraham, K., Behrendt, H., Ring, J., Zimmer, A., Schughart, K., Pfeffer, K., Wolf, E., and Balling, R. 2000b. Genome-wide, large-scale production of mutant mice by ENU mutagenesis. Nat Genet 25(4): 444-447. Ignatius, M.S., Moose, H.E., El-Hodiri, H.M., and Henion, P.D. 2008. colgate/hdac1 Repression of foxd3 expression is required to permit mitfa-dependent melanogenesis. Dev Biol 313(2): 568-583. Ikeya, M., Lee, S.M., Johnson, J.E., McMahon, A.P., and Takada, S. 1997. Wnt signalling required for expansion of neural crest and CNS progenitors. Nature 389(6654): 966-970. Imokawa, G., Kobayasi, T., and Miyagishi, M. 2000. Intracellular signaling mechanisms leading to synergistic effects of endothelin-1 and stem cell factor on proliferation of cultured human melanocytes. Cross-talk via trans-activation of the tyrosine kinase c-kit receptor. J Biol Chem 275(43): 33321-33328. Imokawa, G., Yada, Y., and Kimura, M. 1996. Signalling mechanisms of endothelininduced mitogenesis and melanogenesis in human melanocytes. Biochem J 314 ( Pt 1): 305312.  109  Ingram, D.A., Yang, F.C., Travers, J.B., Wenning, M.J., Hiatt, K., New, S., Hood, A., Shannon, K., Williams, D.A., and Clapp, D.W. 2000. Genetic and biochemical evidence that haploinsufficiency of the Nf1 tumor suppressor gene modulates melanocyte and mast cell fates in vivo. J Exp Med 191(1): 181-188. Jacks, T., Shih, T.S., Schmitt, E.M., Bronson, R.T., Bernards, A., and Weinberg, R.A. 1994. Tumour predisposition in mice heterozygous for a targeted mutation in Nf1. Nat Genet 7(3): 353-361. Jessen, K.R. and Mirsky, R. 2005. The origin and development of glial cells in peripheral nerves. Nat Rev Neurosci 6(9): 671-682. Jiao, Z., Zhang, Z.G., Hornyak, T.J., Hozeska, A., Zhang, R.L., Wang, Y., Wang, L., Roberts, C., Strickland, F.M., and Chopp, M. 2006. Dopachrome tautomerase (Dct) regulates neural progenitor cell proliferation. Dev Biol 296(2): 396-408. Jimbow, K., Szabo, G., and Fitzpatrick, T.B. 1973. Ultrastructure of giant pigment granules (macromelanosomes) in the cutaneous pigmented macules of neurofibromatosis. J Invest Dermatol 61(5): 300-309. Kawaguchi, A., Chiba, K., Tanimura, Y., Motohashi, T., Aoki, H., Takeda, T., Hayashi, S., Shimizu, K., and Kunisada, T. 2008. Isolation and characterization of Kit-independent melanocyte precursors induced in the skin of Steel factor transgenic mice. Dev Growth Differ 50(2): 63-69. Kim, H.A., Ling, B., and Ratner, N. 1997. Nf1-deficient mouse Schwann cells are angiogenic and invasive and can be induced to hyperproliferate: reversion of some phenotypes by an inhibitor of farnesyl protein transferase. Mol Cell Biol 17(2): 862-872. Kim, H.A., Rosenbaum, T., Marchionni, M.A., Ratner, N., and DeClue, J.E. 1995. Schwann cells from neurofibromin deficient mice exhibit activation of p21ras, inhibition of cell proliferation and morphological changes. Oncogene 11(2): 325-335. Kolanczyk, M., Kossler, N., Kuhnisch, J., Lavitas, L., Stricker, S., Wilkening, U., Manjubala, I., Fratzl, P., Sporle, R., Herrmann, B.G., Parada, L.F., Kornak, U., and Mundlos, S. 2007. Multiple roles for neurofibromin in skeletal development and growth. Hum Mol Genet 16(8): 874-886. Kushimoto, T., Basrur, V., Valencia, J., Matsunaga, J., Vieira, W.D., Ferrans, V.J., Muller, J., Appella, E., and Hearing, V.J. 2001. A model for melanosome biogenesis based on the 110  purification and analysis of early melanosomes. Proc Natl Acad Sci U S A 98(19): 1069810703. Lane, P.W. 1966. Association of megacolon with two recessive spotting genes in the mouse. J Hered 57(1): 29-31. Lau, N., Feldkamp, M.M., Roncari, L., Loehr, A.H., Shannon, P., Gutmann, D.H., and Guha, A. 2000. Loss of neurofibromin is associated with activation of RAS/MAPK and PI3K/AKT signaling in a neurofibromatosis 1 astrocytoma. J Neuropathol Exp Neurol 59(9): 759-767. Le, D.T., Kong, N., Zhu, Y., Lauchle, J.O., Aiyigari, A., Braun, B.S., Wang, E., Kogan, S.C., Le Beau, M.M., Parada, L., and Shannon, K.M. 2004. Somatic inactivation of Nf1 in hematopoietic cells results in a progressive myeloproliferative disorder. Blood 103(11): 4243-4250. Le, L.Q. and Parada, L.F. 2007. Tumor microenvironment and neurofibromatosis type I: connecting the GAPs. Oncogene 26(32): 4609-4616. Le, L.Q., Shipman, T., Burns, D.K., and Parada, L.F. 2009. Cell of origin and microenvironment contribution for NF1-associated dermal neurofibromas. Cell Stem Cell 4(5): 453-463. Levy, C., Khaled, M., and Fisher, D.E. 2006. MITF: master regulator of melanocyte development and melanoma oncogene. Trends Mol Med 12(9): 406-414. Mackenzie, M.A., Jordan, S.A., Budd, P.S., and Jackson, I.J. 1997. Activation of the receptor tyrosine kinase Kit is required for the proliferation of melanoblasts in the mouse embryo. Dev Biol 192(1): 99-107. Maertens, O., De Schepper, S., Vandesompele, J., Brems, H., Heyns, I., Janssens, S., Speleman, F., Legius, E., and Messiaen, L. 2007. Molecular dissection of isolated disease features in mosaic neurofibromatosis type 1. Am J Hum Genet 81(2): 243-251. Mahe, E., Zeller, J., Wechsler, J., Wolkenstein, P., and Revuz, J. 2001. [Large hairy pigmented spots in neurofibromatosis type 1: an atypical form of neurofibromas]. Ann Dermatol Venereol 128(5): 619-621.  111  Mason, P.J. and Bessler, M. 2008. Dark skin mutations shed light on inherited anemia. Nat Genet 40(8): 931-932. Mautner, V.F., Asuagbor, F.A., Dombi, E., Funsterer, C., Kluwe, L., Wenzel, R., Widemann, B.C., and Friedman, J.M. 2008. Assessment of benign tumor burden by wholebody MRI in patients with neurofibromatosis 1. Neuro Oncol 10(4): 593-598. Mayes, D.A., Rizvi, T.A., Cancelas, J.A., Kolasinski, N.T., Ciraolo, G.M., StemmerRachamimov, A.O., and Ratner, N. Perinatal or adult Nf1 inactivation using tamoxifeninducible PlpCre each cause neurofibroma formation. Cancer Res 71(13): 4675-4685. -. 2010. Perinatal or Adult Nf1 Inactivation Using Tamoxifen-Inducible PlpCre Each Cause Neurofibroma Formation. Cancer Res 71(13): 4675-4685. McGowan, K.A., Aradhya, S., Fuchs, H., de Angelis, M.H., and Barsh, G.S. 2006. A mouse keratin 1 mutation causes dark skin and epidermolytic hyperkeratosis. J Invest Dermatol 126(5): 1013-1016. McGowan, K.A., Li, J.Z., Park, C.Y., Beaudry, V., Tabor, H.K., Sabnis, A.J., Zhang, W., Fuchs, H., de Angelis, M.H., Myers, R.M., Attardi, L.D., and Barsh, G.S. 2008. Ribosomal mutations cause p53-mediated dark skin and pleiotropic effects. Nat Genet 40(8): 963-970. Morcos, P., Thapar, N., Tusneem, N., Stacey, D., and Tamanoi, F. 1996. Identification of neurofibromin mutants that exhibit allele specificity or increased Ras affinity resulting in suppression of activated ras alleles. Mol Cell Biol 16(5): 2496-2503. Morris, J.K., Lin, W., Hauser, C., Marchuk, Y., Getman, D., and Lee, K.F. 1999. Rescue of the cardiac defect in ErbB2 mutant mice reveals essential roles of ErbB2 in peripheral nervous system development. Neuron 23(2): 273-283. Morrison, S.J., Perez, S.E., Qiao, Z., Verdi, J.M., Hicks, C., Weinmaster, G., and Anderson, D.J. 2000. Transient Notch activation initiates an irreversible switch from neurogenesis to gliogenesis by neural crest stem cells. Cell 101(5): 499-510. Motohashi, T., Aoki, H., Chiba, K., Yoshimura, N., and Kunisada, T. 2007. Multipotent cell fate of neural crest-like cells derived from embryonic stem cells. Stem Cells 25(2): 402-410. Newbern, J. and Birchmeier, C. 2010. Nrg1/ErbB signaling networks in Schwann cell development and myelination. Semin Cell Dev Biol 21(9): 922-928. 112  Nguyen, T. and Wei, M.L. 2007. Hermansky-Pudlak HPS1/pale ear gene regulates epidermal and dermal melanocyte development. J Invest Dermatol 127(2): 421-428. NIH. 1988. Neurofibromatosis. Conference statement. National Institutes of Health Consensus Development Conference. Arch Neurol 45(5): 575-578. Okazaki, M., Yoshimura, K., Suzuki, Y., Uchida, G., Kitano, Y., Harii, K., and Imokawa, G. 2003. The mechanism of epidermal hyperpigmentation in cafe-au-lait macules of neurofibromatosis type 1 (von Recklinghausen's disease) may be associated with dermal fibroblast-derived stem cell factor and hepatocyte growth factor. Br J Dermatol 148(4): 689-697. Pavan, W.J. and Tilghman, S.M. 1994. Piebald lethal (sl) acts early to disrupt the development of neural crest-derived melanocytes. Proc Natl Acad Sci U S A 91(15): 71597163. Perkinson, N.G. 1957. Melanoma arising in a cafe au lait spot of neurofibromatosis. Am J Surg 93(6): 1018-1020. Pollock, P.M., Cohen-Solal, K., Sood, R., Namkoong, J., Martino, J.J., Koganti, A., Zhu, H., Robbins, C., Makalowska, I., Shin, S.S., Marin, Y., Roberts, K.G., Yudt, L.M., Chen, A., Cheng, J., Incao, A., Pinkett, H.W., Graham, C.L., Dunn, K., Crespo-Carbone, S.M., Mackason, K.R., Ryan, K.B., Sinsimer, D., Goydos, J., Reuhl, K.R., Eckhaus, M., Meltzer, P.S., Pavan, W.J., Trent, J.M., and Chen, S. 2003. Melanoma mouse model implicates metabotropic glutamate signaling in melanocytic neoplasia. Nat Genet 34(1): 108-112. Powell, M.B., Hyman, P., Bell, O.D., Balmain, A., Brown, K., Alberts, D., and Bowden, G.T. 1995. Hyperpigmentation and melanocytic hyperplasia in transgenic mice expressing the human T24 Ha-ras gene regulated by a mouse tyrosinase promoter. Mol Carcinog 12(2): 82-90. Quevedo, W.C., Jr. 1969. The control of color in mammals. Am Zool 9(2): 531-540. Rebbeck, T.R., Kanetsky, P.A., Walker, A.H., Holmes, R., Halpern, A.C., Schuchter, L.M., Elder, D.E., and Guerry, D. 2002. P gene as an inherited biomarker of human eye color. Cancer Epidemiol Biomarkers Prev 11(8): 782-784. Reid, K., Turnley, A.M., Maxwell, G.D., Kurihara, Y., Kurihara, H., Bartlett, P.F., and Murphy, M. 1996. Multiple roles for endothelin in melanocyte development: regulation of progenitor number and stimulation of differentiation. Development 122(12): 3911-3919. 113  Riccardi, V.M. 1992. Neurofibromatosis : phenotype, natural history, and pathogenesis. Johns Hopkins University Press, Baltimore. Rice, J., Doggett, B., Sweetser, D.A., Yanagisawa, H., Yanagisawa, M., and Kapur, R.P. 2000. Transgenic rescue of aganglionosis and piebaldism in lethal spotted mice. Dev Dyn 217(1): 120-132. Rizvi, T.A., Huang, Y., Sidani, A., Atit, R., Largaespada, D.A., Boissy, R.E., and Ratner, N. 2002. A novel cytokine pathway suppresses glial cell melanogenesis after injury to adult nerve. J Neurosci 22(22): 9831-9840. Rosenbaum, T., Boissy, Y.L., Kombrinck, K., Brannan, C.I., Jenkins, N.A., Copeland, N.G., and Ratner, N. 1995. Neurofibromin-deficient fibroblasts fail to form perineurium in vitro. Development 121(11): 3583-3592. Rubben, A., Bausch, B., and Nikkels, A. 2006. Somatic deletion of the NF1 gene in a neurofibromatosis type 1-associated malignant melanoma demonstrated by digital PCR. Mol Cancer 5: 36. Schaffer, J.V., Chang, M.W., Kovich, O.I., Kamino, H., and Orlow, S.J. 2007. Pigmented plexiform neurofibroma: Distinction from a large congenital melanocytic nevus. J Am Acad Dermatol 56(5): 862-868. Scheffzek, K., Ahmadian, M.R., Wiesmuller, L., Kabsch, W., Stege, P., Schmitz, F., and Wittinghofer, A. 1998. Structural analysis of the GAP-related domain from neurofibromin and its implications. EMBO J 17(15): 4313-4327. Scherer, D. and Kumar, R. 2010. Genetics of pigmentation in skin cancer--a review. Mutat Res 705(2): 141-153. Shin, M.K., Levorse, J.M., Ingram, R.S., and Tilghman, S.M. 1999. The temporal requirement for endothelin receptor-B signalling during neural crest development. Nature 402(6761): 496-501. Shishido, E., Kadono, S., Manaka, I., Kawashima, M., and Imokawa, G. 2001. The mechanism of epidermal hyperpigmentation in dermatofibroma is associated with stem cell factor and hepatocyte growth factor expression. J Invest Dermatol 117(3): 627-633.  114  Shrestha, P., Muramatsu, Y., Kudeken, W., Mori, M., Takai, Y., Ilg, E.C., Schafer, B.W., and Heizmann, C.W. 1998. Localization of Ca(2+)-binding S100 proteins in epithelial tumours of the skin. Virchows Arch 432(1): 53-59. Soriano, P. 1999. Generalized lacZ expression with the ROSA26 Cre reporter strain. Nat Genet 21(1): 70-71. Srinivas, S., Watanabe, T., Lin, C.S., William, C.M., Tanabe, Y., Jessell, T.M., and Costantini, F. 2001. Cre reporter strains produced by targeted insertion of EYFP and ECFP into the ROSA26 locus. BMC Dev Biol 1: 4. Stark, M.S., Woods, S.L., Gartside, M.G., Bonazzi, V.F., Dutton-Regester, K., Aoude, L.G., Chow, D., Sereduk, C., Niemi, N.M., Tang, N., Ellis, J.J., Reid, J., Zismann, V., Tyagi, S., Muzny, D., Newsham, I., Wu, Y., Palmer, J.M., Pollak, T., Youngkin, D., Brooks, B.R., Lanagan, C., Schmidt, C.W., Kobe, B., MacKeigan, J.P., Yin, H., Brown, K.M., Gibbs, R., Trent, J., and Hayward, N.K. 2011. Frequent somatic mutations in MAP3K5 and MAP3K9 in metastatic melanoma identified by exome sequencing. In Nat Genet, pp. 165-169. Staser, K., Yang, F.C., and Clapp, D.W. 2010. Mast cells and the neurofibroma microenvironment. Blood 116(2): 157-164. Stratigos, A.J., Dover, J.S., and Arndt, K.A. 2000. Laser treatment of pigmented lesions-2000: how far have we gone? Arch Dermatol 136(7): 915-921. Sturm, R.A. 2006. A golden age of human pigmentation genetics. Trends Genet 22(9): 464468. Sulem, P., Gudbjartsson, D.F., Stacey, S.N., Helgason, A., Rafnar, T., Jakobsdottir, M., Steinberg, S., Gudjonsson, S.A., Palsson, A., Thorleifsson, G., Palsson, S., Sigurgeirsson, B., Thorisdottir, K., Ragnarsson, R., Benediktsdottir, K.R., Aben, K.K., Vermeulen, S.H., Goldstein, A.M., Tucker, M.A., Kiemeney, L.A., Olafsson, J.H., Gulcher, J., Kong, A., Thorsteinsdottir, U., and Stefansson, K. 2008. Two newly identified genetic determinants of pigmentation in Europeans. Nat Genet 40(7): 835-837. Takemoto, Y., Keighren, M., Jackson, I.J., and Yamamoto, H. 2006. Genomic localization of a Dct-LacZ transgene locus: a simple assay for transgene status. Pigment Cell Res 19(6): 644-645.  115  Thomas, A.J. and Erickson, C.A. 2009. FOXD3 regulates the lineage switch between neural crest-derived glial cells and pigment cells by repressing MITF through a non-canonical mechanism. Development 136(11): 1849-1858. Tokuo, H., Yunoue, S., Feng, L., Kimoto, M., Tsuji, H., Ono, T., Saya, H., and Araki, N. 2001. Phosphorylation of neurofibromin by cAMP-dependent protein kinase is regulated via a cellular association of N(G),N(G)-dimethylarginine dimethylaminohydrolase. FEBS Lett 494(1-2): 48-53. Tong, J., Hannan, F., Zhu, Y., Bernards, A., and Zhong, Y. 2002. Neurofibromin regulates G protein-stimulated adenylyl cyclase activity. Nat Neurosci 5(2): 95-96. Tortora, G.J. and Derrickson, B. 2009. Principles of anatomy and physiology. John Wiley & Sons, Hoboken, N.J. Trovo-Marqui, A.B. and Tajara, E.H. 2006. Neurofibromin: a general outlook. Clin Genet 70(1): 1-13. Van Raamsdonk, C.D., Bezrookove, V., Green, G., Bauer, J., Gaugler, L., O'Brien, J.M., Simpson, E.M., Barsh, G.S., and Bastian, B.C. 2009. Frequent somatic mutations of GNAQ in uveal melanoma and blue naevi. Nature 457(7229): 599-602. Van Raamsdonk, C.D., Griewank, K.G., Crosby, M.B., Garrido, M.C., Vemula, S., Wiesner, T., Obenauf, A.C., Wackernagel, W., Green, G., Bouvier, N., Sozen, M.M., Baimukanova, G., Roy, R., Heguy, A., Dolgalev, I., Khanin, R., Busam, K., Speicher, M.R., O'Brien, J., and Bastian, B.C. 2010. Mutations in GNA11 in uveal melanoma. N Engl J Med 363(23): 2191-2199. Wehrle-Haller, B., Meller, M., and Weston, J.A. 2001. Analysis of melanocyte precursors in Nf1 mutants reveals that MGF/KIT signaling promotes directed cell migration independent of its function in cell survival. Dev Biol 232(2): 471-483. Wehrle-Haller, B. and Weston, J.A. 1995. Soluble and cell-bound forms of steel factor activity play distinct roles in melanocyte precursor dispersal and survival on the lateral neural crest migration pathway. Development 121(3): 731-742. White, R.M. and Zon, L.I. 2008. Melanocytes in development, regeneration, and cancer. Cell Stem Cell 3(3): 242-252.  116  Widlund, H.R. and Fisher, D.E. 2003. Microphthalamia-associated transcription factor: a critical regulator of pigment cell development and survival. Oncogene 22(20): 3035-3041. Wilkie, A.L., Jordan, S.A., and Jackson, I.J. 2002. Neural crest progenitors of the melanocyte lineage: coat colour patterns revisited. Development 129(14): 3349-3357. Wolpowitz, D., Mason, T.B., Dietrich, P., Mendelsohn, M., Talmage, D.A., and Role, L.W. 2000. Cysteine-rich domain isoforms of the neuregulin-1 gene are required for maintenance of peripheral synapses. Neuron 25(1): 79-91. Woodhoo, A. and Sommer, L. 2008. Development of the Schwann cell lineage: from the neural crest to the myelinated nerve. Glia 56(14): 1481-1490. Wu, J., Williams, J.P., Rizvi, T.A., Kordich, J.J., Witte, D., Meijer, D., StemmerRachamimov, A.O., Cancelas, J.A., and Ratner, N. 2008. Plexiform and dermal neurofibromas and pigmentation are caused by Nf1 loss in desert hedgehog-expressing cells. Cancer Cell 13(2): 105-116. Xu, G.F., O'Connell, P., Viskochil, D., Cawthon, R., Robertson, M., Culver, M., Dunn, D., Stevens, J., Gesteland, R., White, R., and et al. 1990. The neurofibromatosis type 1 gene encodes a protein related to GAP. Cell 62(3): 599-608. Yang, F.C., Chen, S., Clegg, T., Li, X., Morgan, T., Estwick, S.A., Yuan, J., Khalaf, W., Burgin, S., Travers, J., Parada, L.F., Ingram, D.A., and Clapp, D.W. 2006. Nf1+/- mast cells induce neurofibroma like phenotypes through secreted TGF-beta signaling. Hum Mol Genet 15(16): 2421-2437. Ye, M., Hu, D., Tu, L., Zhou, X., Lu, F., Wen, B., Wu, W., Lin, Y., Zhou, Z., and Qu, J. 2008. Involvement of PI3K/Akt signaling pathway in hepatocyte growth factor-induced migration of uveal melanoma cells. Invest Ophthalmol Vis Sci 49(2): 497-504. Yoshida, H., Kunisada, T., Kusakabe, M., Nishikawa, S., and Nishikawa, S.I. 1996. Distinct stages of melanocyte differentiation revealed by anlaysis of nonuniform pigmentation patterns. Development 122(4): 1207-1214. Zhang, T.T., Li, H., Cheung, S.M., Costantini, J.L., Hou, S., Al-Alwan, M., and Marshall, A.J. 2009. Phosphoinositide 3-kinase-regulated adapters in lymphocyte activation. Immunol Rev 232(1): 255-272.  117  Zhu, Y., Ghosh, P., Charnay, P., Burns, D.K., and Parada, L.F. 2002. Neurofibromas in NF1: Schwann cell origin and role of tumor environment. Science 296(5569): 920-922. Zhu, Y., Romero, M.I., Ghosh, P., Ye, Z., Charnay, P., Rushing, E.J., Marth, J.D., and Parada, L.F. 2001. Ablation of NF1 function in neurons induces abnormal development of cerebral cortex and reactive gliosis in the brain. Genes Dev 15(7): 859-876.  118  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0072689/manifest

Comment

Related Items