3 3 3 I II III 3 I II." I. III LI '32 3333 IIIIIIIIII' I333 :33 '3~.-__’-|°II 33"} ' 42': 3 . I... ~ .. . . .4.- . ." ~‘ 313:?!» If '.. ' r 1 '.- - .. «III-I . I3 _I'I ~ I3I<,-:IIU}I III:I'II"-I. . II' .- I31. . . - 3: 'I’ . ,7 v. . '4. ‘ ‘ , ‘s.‘c ‘3i3 I.'. III‘Izib‘IIICI '.' 3 II: I . ‘3 .. I .I I IIII 1.; 13:31," 5“: ;;:,1_.I:I‘gw .3ng» ‘,’.\ .1; 3.1..» . ’.I..53- 'Ib JIII h3g1: —- ‘_ ,, 214; {:3 3'3—3‘33Ifl33‘3 4-:- II-FII ' 5-H ~ 1 It - I. 11.1””... - I mun-#34 3.? 3.33.3.3 ,0 33,333.43 ’3 . ”£3333 IIIIII3I.II 313",” 3'33I333 . IIILIIKJ I333. I'I'IIII‘I ,W 33I333I3III IhI3330I3333I'I3333I3 IIIIIIII ,I,3.3,I'I"I3I III 33333I33333II33'33I3'I'I“ "33; ,. I. 3.33.3.3, . I333II IIIIII 33.3I.II.II3 I3. 33 “III 33333333333333 ‘3‘} ,,, II..I3I33 .- I333 Ilrffi, ILIIIIII .II .33II\I. 3,3 ',[-| II‘III 335. 33 IPIIII II‘III3'IIQIII3 IIIII, . II.33-3 3 IIIIIII ,I, I '3 33333 ’IL33‘3IIII3II IIIIIIIIIIIIIIIII '3 §.II’I,,"N-} 5.. I .III I ,. ‘ I II: _ ~ . ‘ ' ‘ ... .32 2 . “a. I III”, '9" I3I3I3I3IIIIIII I3 I“ [LIIC WWII (III, , 3 {.- Ilr,rr3?':3.-3~.;-Iiz IIIKIII‘II {”33 ' 53: II I -. I ~ “I :3 .3” "J I”? III‘IL ,. ,. I, ”Amt 3.0: .2, IIII33-[I‘3 Ira .525: 332,; :v. I" ' I I3 - ,‘III'II .I I - 3r: 3 2 fig!" 3 M3333 ’ 331i:3:I_\IIII3I:a_1\B7; 33.333 31333” I33”. ' , 2-4.3- 1' ff I .1' l :8?" III% 333 Q3 I.I3£l‘:I‘I‘.I (9‘3. 7:"1, II I 3133 II 'II'IIIIs‘IIIliIv .3 {fly-'- #3:: :‘l‘IIIééq ‘3'“ 3' ' 3&0 “I: :. I ' . 3. 4:031, II . . |I(_ . 2. I:‘I:I.V.L {kink : ; ‘ 3‘ g ‘ I ‘ ‘W‘IL II II 3 ‘1”“wL . . H. (.3 .. . I IN?- Ic- «31- , 3 3 3| , .. ‘5'. . ‘ I 433” 3.3 fl I I. IIII'E'I I‘33I3 'II3I3 III' II'I, 3'3 III II 333 3333.3»; II‘II I3. 33333333 3333I|IIIII IIInIHJ , '3Il3kHIIII _ . I3I33I I'IIIII'I'I IIIII’I 3333 3333,.II 333333333. I '3',I'III I333 ‘33 33:33:23.. . IIIIII33II33 IIII 33333: 333 333,3 I3333I3I I333). 33.3 3m I 3 ' 33 I333 I 3 _ _. ,II3333IIIIIIIII III-IIIIII3I‘ ,3I3I-III3I323 ”33133333333 “335' ”33333533213391: 3.33133 3 HI ' I .33 .,, ..I. .33 3:] 7.- 39,"? 33 w I3333.3I EIIII IIII‘II I.“ I333I3‘IIII I ”II“ ‘VI [2'31] 3 Q333 ’E )3: I353. 33333,! [C633]. ‘3 I3‘II. III I3I' 3 III'IIII II3II3 I I,IIIII l {8:33,} nu. IIIIII 33.335 1.13;; 5?}:- III, ,II III333I “I. III3I3I.‘ III',‘ IIII I . 3' 333.153 II'IIrImsIIIk rim III, IIIIIIII IN”, 333333, I‘; 2,5 '3_I' flu”?! I3“, '3, III ,IIIIIIII 33,I .llII -, I-"3333 I333 I I334, WI .33??? girl’s-I333 -.... 3 3123,33 II I 3 I )3 . III I. MI I " II .. 353‘ I 14:33 . ' 3 .3 1m: \ LII‘ 3 I! . I33 I .I‘.'. _ IIIIIIIIIII'III II33I3II‘I3 ‘I.3.3,- IIIII_-~.Ir3:;‘VT. 3 3 I" I l' ‘ I . III II] r_ 3.3““ 3 III '3 3 I 33 .3 33 .I-I', I]. 4333?,” ”33333 4131”.” ‘_ If 3.."3 - . 3:653 34.3 II3 3333.3”. I. 33,3 3.3.2 3 3, ~ .I'I ‘vI.r.3;I\I3:I_(I{::T;I'} Maggi“, my??? I. (Z; I “42333 If, IIIIIII%- ‘IIv- '3 33 . , % Is-IIIIII73-3if.-I‘—’,~§’II£ITIIIIIf ~ ' ‘. III II I , 7-3. ‘~ .‘Z‘L- " 'I3 . '~'. I "I3 3' "73,2“ .13. ._ _. ,- 3.233 -:"u _: 137331313. 13:53 " . - - 3 n)! 3173353" I I '4 I). 3,Iv .I,I_ All}: ,.‘EITI:‘I , .III‘I I'Ifi‘ I.“ 33III .33" IIII' ,I A. 33 ' 33I|.3 3 ,i‘II‘_ |,I INS-III 33‘ ' II" . “‘1, *3: REIIIII‘ I333; 1:“? H 'II I I 3 I . III .I 'I' 3 ..3 1'. {(33433}; .-—333 ”III” , 133331IinmIII3I‘III33" . I. IIIIIIIIWI I III I IIIIII'b' dl‘lI| 3 Iii-3,13“, , 2" .333", 'II’II “II ,2. leI‘?tI~-I‘II'I"J'I"$:" I“ 3'?“ #f.’ Ikl. 31% I “I; I' ‘ . I", ‘I _I .3 7,5} _: IIIUI: III 3" II'IHII- <3»? ._- .“3‘ I tar}. ,.II 33333 I 'I III a I I" W I L . III II II 3. I . h 'I I 3.33I' 333 ,3 IQ'II II 3’1”" In... .. . u. . 3. U“. . .- 3» .3'3I . I-I II . I . "3333' 'II: III I 3 II. 33 , .I' 33 V.‘l '3'. I. “II ”I I 3| i . ‘ 2' I, " I g“. 6 IIIIIIH 3 I, .II,:I,IIIII ‘. I3I..III|, .II'II «33.3 I". “.I‘I III I III IFfHI“ .Il 33" I ‘II .,I I 3}|‘II‘I’ II.§?I\**‘II’I IITWI I. I, '3'). Iy’ntl’gul" ”2”; ‘& I I]?! ‘1 “ling,” 3,34?” I" ‘ 'IIlLIIrI' I I . III IIII. II”, 3'33 1 I II H3 , I III," ‘ I. 33 I“ II II I, II “.I' II III. I II'3III III III II IIIII‘I ‘I'IIIII33II3 3I'I 33 3" IIII.II"I'I II’II'III'I 3"II'I3I3I33333‘I33 'IIII -I‘I III‘ n 33333333 U.» 3 I I IIIIII33I "MINI“ l I,.,3II3 '3 -r.... VIN-I I ”III ,I H III. I ('13 3 333 III II 333': ".II IIIII'II 3.I LLJMI “3:1,." ‘JItI 3&3 3' 3 I M03“ IIIIIIIUILIIIII41.3I-“33'333333133 ‘ Wag 5 .. This is to certify that the thesis entitled AN ANALYSIS OF THE PHENOTYPE AND GENE ACTION OF MICROPHTHALMIC WHITE (MIWH) IN THE HOUSE MOUSE presentedby Bruce Montford Pratt has been accepted towards fulfillment of the requirements for .D_Q.c_£.9_r.al_ degree in 219.122.!— ' / , 7* / Major professor / Date /€/// 77 / 0-7 639 OVERDUE PINES ARE 25¢ PER DAY PER ITEM Return to book drop to remove this checkout from your record. AN ANALYSIS OF THE PHENOTYPE AND GENE ACTION OF MICROPHTHALMIC WHITE (MIWH) IN THE HOUSE MOUSE By Bruce Montford Pratt A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Zoology 1979 ABSTRACT AN ANALYSIS OF THE PHENOTYPE AND GENE ACTION OF MICROPHTHALMIC WHITE (MIWH) IN THE HOUSE MOUSE BY BRUCE MONTFORD PRATT At the present time more than 50 loci and 130 mutant alleles are known to affect the quality, quantity and spatial distribution of pigmentation in the skin and hair of the house mouse (Mus musculus). The work described in this dissertation is focused on the mechanisms of action of one h). Non- of these mutant alleles, microphthalmic white (giw h agouti black mice, heterozygous for Ni" have a uniform gray coat color and show incomplete penetrance and variable h wh /Mi expressivity for white spotting. Homozygous mice (giw ) have completely white fur, and are deaf and microphthalmic. Single locus mutations affecting pigmentation, inner ear function and eye development, have been described in several species including man (Waardenburg, 1951), dog (Sorsby and Davey, 1954), cat (Bergsma and Brown, 1971) and hamster (Knapp and Polivanov, 1958; Robinson, 1964; Yoon, 1975). Robinson (1964) and Searle (19688) have SUSSeStEd that the mutant alleles of giWh in the mouse and anophthalmic white (Eh) in the Syrian hamster (Mesocricetus auratus) may represent mutations at homologous loci in these two species. Specifically, the work in this dissertation has been focused on three main questions concerning the mutant allele E1Wh: 1) why are the Nivhl+ mice gray, 2) why are h/giWh mice white, and 3) are ngh in the mouse and the H1“ Eh in the hamster mutations at a homologous locus? To answer the first question, melanocyte morphology, distribution and function were examined on one or more of the following genetic backgrounds: nonagouti black (g/g;§/§;2/25§/§), dilute black (§/§;§/§;d/d;§/§) and re- cessive yellow (é/a;§/§;Q/2;g/g). To address the second question, dermal-epidermal recombination grafts were made utilizing embryonic skin of normal (+/+), gray (§$Wh/+), h/fliWh) and dominant spotting (fl/fl) mice. To white (Elw address the third question several comparisons were made. First, the distribution and function of the oculocutaneous melanocytes in the two species were compared. Second, sev- eral measures of growth and metabolism in the mouse were compared with data previously collected by Asher (1979). Third, the cell density in the sex zone of the adenohypophysis was compared with data previously collected in the hamster (James, 1979). The general conclusions of this study are listed below. hivhl+ causes a change in the shape and a reduction in the number and size of melanosomes of the hair. EEWh causes a reduction in the number and size of the melanocytes of the zigzag hair follicles. EEWhl+ causes a reduction in the number of epidermal melanocytes of the tail and ear. In the presence of recessive yellow (g/g), EEWh /+ increases the quantity of phaeomelanin deposited in the hair. E1Wh/+ does not change the electrophoretic mobility of the tyrosinase isozymes. The dermal-epidermal recombination grafts yielded the following information. 6. h wh Epidermis from 13 day Elw lEl embryos is inimical to proper melanocyte differentiation. h/Elvh embryos does not con- Epidermis from 13 day Eiw tain cells which can differentiate into functional melanocytes. h/E1Wh embryos does not adversely Dermis from 13 day Eiw affect normal (+/+) or EEWhl+ follicular melanocyte differentiation. Melanoblasts from 13 day Eiyhl+ dermis are subnormal in their ability to differentiate into functional melano- cytes. The comparative study between EEWh in the mouse and Eh in the hamster gave the following information. 10. 11. The distribution and function of oculocutaneous melano- cYtes differ between hivh and Eh. Several parameters of growth and metabolism are differ- entially affected by EEWh and K2- 12. EEWh and Eh have different effects on the cellular density of the sex zone of the adenohypophysis. To Tricia ii ACKNOWLEDGMENTS I would like to acknowledge the assistance of my ad- visor, Dr. J.H. Asher, Jr., and the members of my commit— tee, Dr. R.N. Band and Dr. S.C. Bromley, Department of Zoology, and Dr. W.W. Wells, Department of Biochemistry, in the successful completion of this endeavour. I would also like to thank the following people and organizations for their contributions of time, equipment, materials, expertise and financial support: Dr. M. Balaban, Dr. J. Butcher, Dr. L. Clemens, Dr. J. Edwards, Dr. T. Friedman, Dr. R. Hill, Dr. J. King and Lynn Harper, and the office staff of the Department of Zoology; Dr. J. Wilson, Department of Biochemistry; Dr. C. Tweedle, and Dr. E. Retzlaff, Department of Bio- mechanics; Dr. T. Adams, Department of Physiology; C. Jones, Science Library; Dr. T. Mayer, Ryder College, New Jersey; Dr. G. Wolfe, University of Kansas; the Department of Zoology, Michigan State University; the Jackson Laboratory, Bar Harbor, Maine; and the National Science Foundation, Graduate Fellowship Program. I also gratefully acknowledge the considerable finan- cial and moral support of my wife, Patricia, Mr. and Mrs. D.S. Pratt and Mr. and Mrs. H.B. LaFrance. iii TABLE OF CONTENTS LIST OF TABLES . . . . . . . LIST OF FIGURES . . . . . . INTRODUCTION . . . . . . . . . . . Melanocytes as a Model System . . . Outline of Literature Review . . . . Outline of Research Problem . . . . . . LITERATURE REVIEW . . . . . . . . . . . Embryological Origin of Melanocytes . . Migration of Melanoblasts . . . . . . Differentiation . . . . . . . . . . Early events . . . . . . . . . . . . Melanogenesis . . . . . . . . . . . Melanosome development . Melanocyte-keratinocyte interaction Genetic Control of Pigmentation . . . . Uses of genetic variation . . . Classification of mutations Intrisic, cell-specific loci . . . . Intrinsic, tissue-specific loci Intrinsic, generalized loci . . . Extrinsic, generalized loci Other loci iv viii xi 12 12 13 17 19 24 24 27 31 41 43 S9 69 v Developmental Mechanisms of White Spotting . . . . . 7S Mode of gene action . . . . . . . . . . . . . . . 75 Hypotheses of gene action . . . . . . . . . . . . 78 Homology of EEWh and Eh . . . . . . . . . . . . . . . 82 REPRISE . . . . . . . . . . . . . . . . . . . . . . . . 87 METHODS . . . . . . . . . . . . . ; . . . . . . . . . . 89 Mice . . . . . . . . . . . . . . . . . . . . . . . . 89 Hamsters . . . . . . . . . . . . . . . . . . . . . . 89 Follicular Melanocytes . . . . . . . . . . . . . . . 89 Tail Skin Epidermal Melanocytes . . . . . . . . . . . 93 Dermal and Epidermal Melanocytes of the Ear . . . . . 95 Staining properties of melanocytes and mast cells . . . . . . . . . . . . . . . . . . . . . . 95 Effects of EEWh on melanocyte distribution . . . . 96 Eyes . . . . . . . . . . . . . . . . . . . . . . . . 98 Hamster Skin . . . . . . . . . . . . . . . . . . . . 98 Melanosomes ... . . . . . . . . . . . . . . . . . . . 99 Phaeomelanin . . . . . . . . . . . . . . . . . . . . 100 Tyrosinase Isozymes . . . . . . . . . . . . . . . . . 101 Dermal-Epidermal Recombination Grafts . . . . . . . . 102 In Vitro Epidermal Melanocyte Culture . . . . . . . . 107 Hematology . . . . . . . . . . . . . . . . . . . . . 108 Growth Curves . . . . . . . . . . . . . . . . . . . . 109 Metabolic Activity . . . . . . . . . . . . . . . . . 110 Pituitary Cytology . . . . . . . . . . . . . . . . . 110 RESULTS . . . . . . . . . . . . . . . . . . . . . . . . 112 Pigmentary Phenotype of hi . . . . . . . . . . . . 112 vi Follicular melanocytes . . . Tail Melanocytes Ear melanocytes Ocular melanocytes Melanosomes . . . . . . . Phaeomelanin - . . . . . Tyrosinase isozymes Non-Pigmentary Phenotype of EEWh Hematology . . . . . . . . Growth curves . . . . . . . Metabolic measurements - - Pituitary cell density - - Dermal-Epidermal Crafts and £2 Vitro Epidermal wh Melanocyte Cultures Involving E; Experiment I - - . . . . Experiment II . - . . Experiment III . . . . . . . . Experiment IV - . . - . . . . 13 ElEEE epidermal culture . - Pigmentary Phenotype of Eh Ear melanocytes . . . . Skin melanocytes . . . . . . . Ocular melanocytes . . . . . . Dermal-Epidermal Grafts Involving DISCUSSION . . . . . . . . . . . . . Phenotype and Mode of Gene Action h Site of Gene Action of Eiw /+ . . 112 118 122 127 129 137 137 137 137 141 145 145 150 150 157 162 164 165 166 166 168 168 171 179 179 185 Melanocyte 185 O O O O O 0 O O Dermis . . . . . . . . . . . . . . . . . . . . . . 185 Epidermis . . . . . . . . . . . . . . . . . . . . 187 h wh Phenotype of Eiw IE3 . . . . . . . . . . . . . . . 187 Site of Gene Action of EEWh/EEWh . . . . . . . . . . 188 Epidermis . . . . . . . . . . . . . . . . . . . . 188 Dermis . . . . . . . . . . . . . . . . . . . . . . 189 Melanocyte . . . . . . . . . . . . . . . . . . . . 190 Comparison of EEWh to Other Mutations Affecting the Mouse Pigmentation Phenotype . . . . . . . . . . 192 Phenotype . . . . . . . . . . . . . . . . . . . . 192 Specificity of gene action . . . . . . . . . . . . 197 Site of gene action . . . . . . . . . . . . . . . 198 Comparison of EEWh and Eh . . . . . . . . . . . . . . 199 Pigmentation phenotype . . . . . . . . . . . . . . 200 Site of gene action of follicular melanocytes , , 201 Eye development and pigmentation , , . , . , , , , 201 Ear development . . . . . . . . . . . . . . . . . 202 PhysiolOgy . . . . . . . . . . . . . . . . . . . . 202 Microscopic anatomy of adrenal and pituitary , , , 203 Homology . . . . . . . . . . . . . . . . . . . . . 204 SUMMARY, CONCLUSIONS AND FURTHER RESEARCH . . . , . , . 208 APPENDIX A: Analysis of Pigment Volume Data of E. Russell (1948) . . . . . . . . . . . . . 213 APPENDIX B: Elliptical Geometry and Eccentricity , , , 215 APPENDIX C: Formulations and Procedures . . . . . . . . 218 BIBLIOGRAPHY . . . . . . . . . . . . . . . . . . . . . . 225 LIST OF TABLES Melanocyte Pigment in Melanocyte-Follicular Environment Recombination Grafts . . . . . . . Classification of Pigmentation Mutations . . . Effects of c-locus Alleles on Fur Pigmentation Phenotype of Dominant Spotting Alleles . . . . Dermal-Epidermal Grafts from Mayer (1973a) . . Phenotype of Steel Alleles . . . . . . . Cell Processes in the Development of Follicular Pigmentation . . . . . . . . . . . . . . . . . Mouse Inbred Strains . . . . . . . . . . . . . Matings to Create Mice of Desired Genotypes . . Ear Skin Staining Treatments . . . . . . . . . Matings for Embryos for Dermal-Epidermal Grafts Diameter of Zigzag Follicles . . . . . . . . . Anova of Follicular Diameters . . . . . . . . Number of Melanocytes per Follicle . .1. . . . Diameter of Follicular Melanocytes . . . . . Melanocyte Density in Tail Skin Epidermis . . »Contingency Table Results of Intra-Scale Melanocyte Densities . . . . . . . . . . . . viii 26 28 39 51 54 65 76 90 91 97 103 113 114 116 119 121 123 ix Ear Melanocyte Distribution Staining Characteristics . . . . . . . . . . . Enzyme Localization . . . . . . . . . . . . Distribution of Ocular Melanocytes . . . . Melanosome Number . . . . . . . . . . . . . . . . Effects of 97g and EEWh on Melanosome Number . Melanosomal Dimensions . . . . . . . . . . . . . . Simultaneous Confidence Intervals of the Melanosomal Axial Dimensions . . . . . . . . . . Melanosomal Eccentricity . . . . . . . . . . . . Phaeomelanin Content of Hair '. . . . . . . . . . . Hematology of Male Mice . . . . . . . . . . . . . Anova of Erythrocyte Diameters and Reticulocyte Counts Erythrocyte Diameters . . . . . . . . . . . . . Reticulocyte Counts . . . . . . . . . . . . . . Rectal Temperatures of Male Mice . . . . . . . . . 24 Hour Metabolic Measurements of Male Mice . Cell Density in the Sex Zone of the Adenohypophysis Anova of Sex Zone Cell Density Measurements . . . . Dermal-Epidermal Recombination Grafts- Experiment I . . . . . . . . . . . . . . . . . . . Dermal-Epidermal Recombination Grafts- Experiment II . . . . . . . . . . . . . . . . . . . Dermal-Epidermal Recombination Grafts- Experiment III . . . . . . . . . . . . . . . . . . Hamster Ear Melanocyte Distribution . . .'. . . . . 125 126 128 130 132 133 135 136 138 140 142 143 146 147 148 149 151 158 163 167 x Cutaneous Melanocyte Distribution in Hamsters . . . . . 169 Distribution of Hamster Ocular Melanocytes . . . . . . 170 Hamster Dermal-Epidermal Recombination Grafts . . . . . 172 Proportion of Hamster Dermal-Epidermal Grafts Without Pigmented Hair . . . . . . . . . . . . . . . . 174 LIST OF FIGURES Development of Cutaneous Melanocytes . . Melanogenesis . . . . . . . Melanosome Synthesis . . . Follicular Melanocyte Number Follicular Melanocyte Diameter Intra-Scale Melanocyte Density Tyrosinase Isozymes . . . . Growth Curves of Male Mice Prolate Spheroid . . . Ellipse O O O O O O O O I O 0 xi 14 18 117 120 124 139 144 216 216 INTRODUCTION Melanocytes as a Model Developmental System There is an immense diversity of model systems which can be used to investigate mammalian development. One of these model systems, the oculocutaneous pigmentation system, has a number of features of both theoretical and pragmatic importance that make it a particularly attractive system with which to investigate some aspects of cell differentiation. First, the production of oculocutaneous pigmentation is the biochemical prerogative of a single class of cells, the melanocytes. Second, the intracellular production of pigment, melanogenesis, involves an enzyme, chemical pathways and final macromolecular products which are unique to melanocytes. Third, the morphology and function of melanocytes are modulated by both the genome of the melanocyte and the cellular environment in which the melanocytes have developed. Thus, an investigation of mammalian oculocutaneous pigmentation involves the analysis of a single cell type, which is characterized by a unique biochemical phenotype, whose morphology and function are regulated by both intrinsic genetic information and ex- trinsic environmental cues. 2 In addition to these theoretical considerations, several pragmatic aspects of this pigmentation system also recommend its use as a model system. First, there is a tremendous diversity of identified mutations which affect the mammalian pigmentation phenotype. In the house mouse alone, 63 loci and more than 150 mutant alleles that alter oculocutaneous pigmentation have been identified (Silvers, 1979). Such a quantity of mutations provides the investi- gator with a wealth of genetic probes with which to analyze melanocyte differentiation. Second, many aspects of pigmentation phenotypes can be investigated without perma- nent damage or major trauma to the experimental animal. Finally, oculocutaneous pigmentation, in the laboratory at least, is not essential for normal viability and fecundity of the animal. Therefore, the experimental or genetic manipulation of the pigmentation system does not, in most cases, present the investigator with any inordinate diffi- culties in either the breeding or maintenance of the animals under investigation. In short, oculocutaneous pigmentation is a model system well suited for the investi- gation of the genetic and environmental control of cell differentiation. The analysis of any developmental system is usually divided into two major phases. These phases are first, an examination of the model system without genetic variability, 3 and second, a re-examination of the model system with the deliberate introduction of controlled genetic variability. The first phase is comprised of two parts: 1) description of the morphology and biochemistry of the model system, and 2) experimental manipulation of the system, g.g., the addition of morphogenetic or biochemical inhibitors, the extirpation of a portion of the system, or heterotopic and heterochronic transplantations. The second phase consists of the modification of the model system by the introduction of controlled genetic variability, followed by a repetition of the precedures described for the first phase, l.g., description and experimental manipulation of the geneti- cally modified model system. The data derived from the first phase of analysis, only allows the investigator to make correlations between different portions of the model system and to formulate hypotheses of developmental causality. The data derived from the genetically manipulated system allows the investigator to formulate and test hypotheses about the genetic control of the differentiation and function of the model system, which is, after all, one of the main ob— jectives of an investigation of a developmental system. Outline of Literature Review The literature review of this dissertation will follow the same two-part sequence described in the previous paragraph. The first portion of the literature review will 4 describe the basic developmental processes leading to the production of oculocutaneous pigmentation including: 1) the origin, migration and localization of the presump- tive melanocytes (= melanoblasts), 2) melanogenesis, and 3) melanocyte-keratinocyte interactions. The second portion of the literature review will examine the morpho- logical and functional effects of several mouse mutations, including microphthalmic white (EEWh), which alter normal melanocyte differentiation and function. The third portion of the literature review will examine several of the current hypotheses of melanocyte differentiation. The final section of the literature review will summarize: 1) the kinds of data used to establish interspecific gene homologies, and 2) the aspects of the anophthalmic white (Eh) phenotype in the Syrian hamster which are to be compared with the EEWh phenotype in the mouse. Outline of Research Problem The original research to be discussed in this disser- tation is focused on three aspects of the gene action of the mutation microphthalmic white (EEWh). Specifically these areas are: 1) a determination of the factors of the pigmentation phenotype altered by EEWh which cause the hivhl+ mice to have a diluted pigmentation phenotype, 2) a determination of the site of the gene action of EEWh h wh which leads to the unpigmented phenotype of the Eiw lEi' mice, and 3) a comparative study of the phenotype and site 5 of gene action of EEWh in the mouse and anophthalmic white (Eh) in the Syrian hamster, to attempt to establish or refute the homology of these two loci as proposed by Robinson (1964) and Searle (1968a). To investigate the causes of the diluted fur color- h ation of thehiw /+ mice, several aspects of the pigmen- tation phenotype, which are listed below, were examined. 1. The melanocyte density of the tail and ear epidermis, and the dorsal zigzag hair follicles 2. The size and shape of the follicular melanosomes 3. The eumelanosome output of the follicular melanocytes 4. The phaeomelanin output of the follicular melanocytes 5. The isozyme pattern of follicular tyrosinase To investigate the site of gene action of gth on the pigmentation phenotype, a series of dermal-epidermal re— combination grafts and $2 XLEEQ epidermal cultures, which are outlined below, were performed. 1. Dermal-epidermal recombination grafts of skin from 13-14 day +/+, EEWh/+, Eth/MIWh , and E/E embryos 2. Dermal-epidermal recombination grafts of epidermis from 11 day EjE;+/+ embryos with dermis from 13-14 day £j£;+/+, slssflth/+» and g/g;§3"h/§;“h embryos 3. Dermal-epidermal recombination grafts of skin from 11 day E/E and 5/3 embryos h 4. £3 vitro culture of epidermis from 14 day +/+, E3? /+, and EEWh/EEWh embryos Based on the external phenotypic similarities of the h/EEWh mouse, specifically the ab- Eh/Eh hamster and Eiw normal eye development and the lack of cutaneous pigmen- tation, Robinson (1964) and Searle (1968a) have suggested that the Eh and EEWh may represent mutations at homologous loci. To further examine this hypothesis, several pheno- typic comparisons, which are listed below, were made between the mice and hamsters carrying comparable genotypes at these two loci. 1. The distribution and function of the oculocutaneous melanocytes 2. The site of gene action as determined by dermal-epider- mal recombination grafts 3. The epistatic interactions of Eh and EEWh with their respective recessive yellow (5) loci 4. The metabolic activity and growth rate 5. The cell densities of the sex zone of the adenohypo- physis LITERATURE REVIEW Embryological Origin of Melanocytes In mammals, melanocytes are the cellular source of pigmentation for the eyes, hair and skin. The melanocytes of the eye that form the pigmented epithelium of the retina ciliary body and iris are all bona fide cuboidal or colum- nar epithelial cells with subjacent basement lamina (Rhodin, 1974). These melanocytes are derived from both the outer and inner walls of the optic cup which are in turn derived from the ventrolateral walls of the prosen— cephalon. The rest of the melanocytes, l.g., those found in the uveal stroma, skin and various internal tissues, have a dendritic morphology and are all derived from the neural crest ( Rawles, 1947; Brihaye-Van Geertruyden, 1963; Markert and Silvers, 1956; Nichols and Reams, 1960). The cutaneous pigmentation of the mouse is the result of the interaction of neural crest-derived melanocytes with their immediate environment, the skin. The first portion of this literature review will discuss the origin and differentiation of the cutaneous melanocytes and the interactions of these melanocytes with their skin environ- ments. For a diagramatic summary of the information 8 discussed in this section, see Figure 1. The neural crest is an ephemeral aggregation of cells located between the neural tube and the dorsal ectoderm of the vertebrate neurula. As the result of primary embryonic induction, the dorsal ectoderm forms the neural plate or groove, whose margins are morphologically defined by the neural ridge. As the neural groove deepens to form the neural tube, the neural ridges containing the pre- sumptive neural crest elevate into the neural folds. During neurulation, fusion of the two neural folds at the dorsal midline of the embryo unites the paired neural crest primordia into a wedge shaped cell mass located between the dorsal midline of the neural tube and the overlying dorsal ectoderm (DiVirgilio E£ 31., 1967; Schulte and Tilney, 1915) Migration of Melanoblasts Shortly after fusion of the neural tube, the neural crest cells disperse from their dorsomedial position (see Figure 1). The cranial neural crest disperses along two major pathways. These cells either move laterally along the mesodermal-ectodermal boundary or disperse medially into various regions of the cephalic mesenchyme and visceral arches (Johnston, 1966; Johnston and Listgarten, 1972; Noden, 1975). The trunk neural crest also disperses in two well defined groups of cells. One group of cells moves ventromedially between the lateral walls of the gastrula ectoderm other neural neural crest —""—"crest derivatives DERMIS dermal dermal melanoblast melanocyte EPIDERMIS epidermal melanoblast EMU follicular non-follicular melanocyte melanocyte follicular keratinocyte keratinocyte melanogenesis melanosome transfer‘ Ids~ ‘ :- pigmented pigmented keratinocyte hair Figure 1. Development of Cutaneous Melanocytes 10 neural tube and somites. Some of these cells enter the somitic mesenchyme and somatopleure, while others continue ventrally into the splanchnopleure and endodermal tisSues. The other group of cells migrates ventrolaterally and ultimately disperses throughout the dermis and stratum germinativum of the epidermis (Rawles 1947; Weston, 1963; Tosney, 1978; Bronner and Cohen, 1979). Melanocytes develop from all four groups of migrating neural crest cellsl. Melanocytes in internal organs and the dendritic uveal melanocytes are derived from portions of the cranial and trunk ventromedially migrating groups of neural crest cells. The cutaneous melanocytes are derived from the ventrolaterally migrating cranial and trunk neural crest cells (Rawles, 1947; Teillet, 1971; Johnston, 1966). Although there is a consensus as to the general direction of migration of the cutaneous melanoblasts, there is some debate as to the precise route of migration of this group of cells. Using 3H-thymidine labeled chick neural tube grafts, Weston (1963) observed that the dis- persing neural crest cells did not migrate within the meso- dermal component of the skin, but rather became located within the epidermis almost immediately after leaving the 1Neural crest cells also differentiate into a diversity of other cell types including, but probably not limited to neural, glial, skeletal, connective and endocrine cellular elements (Horstadius, 1950; Weston, 1970; LeDouarin, 1975; LeLeivre and LeDouarin, 1975). 11 neural tube. In corroboration of this observation of the epidermal localization of the neural crest cells, epidermal grafts from Rhode Island Red chicken embryos ( a pigmented breed) to White Leghorn chicken embryos (an unpigmented breed), revealed the early presence of melanoblasts within the Rhode Island Red embryonic epidermis. Conversely, descriptive histological studies in the chicken (Ris, 1941; Watterson, 1942) and fetal Negroid skin (Zimmerman and Becker, 1959), chick-quail neural tube recombination grafts (Teillet and LeDouarin, 1970; Teillet, 1971), and mouse dermal-epidermal recombination grafts (Mayer, 1973b), have indicated that the initial pathway of cutaneous melano- blast migration is though the dermal mesenchyme, and that the melanoblasts only secondarily migrate into the epi- dermal stratum germinativum. Mayer (1973b) suggested that the precocious presence of the tritiated chicken neural crest cells in the epiderms, as reported by Weston (1963), may have been, in part, an artifact of the particular surgical technique used. However, the absence of tritiated migrating cells in the dermis and the early presence of melanoblasts in the Rhode Island Red embryonic epidermis is not so easily explained. Notwithstanding these unre- solved differences between Weston's observations and those of LeDouarin and Teillet (both in the chicken), the avail- able mammalian evidence is consistent with the hypothesis of the dermal route of migration of the cutaneous melano- blasts. 12 Differentiation Early events At some time shortly before or during their ventro- lateral migration, the trunk neural crest cells become committed to the developmental pathway of melanocyte differentiation. The precise timing and control of this commitment step are not well understood. There is evidence from $2 vitro chick neural tube culture that some neural crest cells may be committed to becoming melanocytes be- fore they migrate away from the neural tube (Cohenand Konigsberg, 1975). However, this same work and that of others (see reviews: Weston, 1970; Morriss and Thorogood, 1978), present evidence that the environments through which a neural crest cell migrates, and the cell's final tissue localization, play a major role in the differentiation of the pluripotential neural crest cell into a specific cell type, 3.3., a melanocyte. Part of the difficulty in determining the timing and mechanisms of melanocyte differentiation is that, in con- trast to the cephalic neural crest and the dorsal root spinal ganglia neuroblasts, presumptive melanoblasts are not known to possess any intrinsic histochemical, antigenic or morphological markers which can serve to distinguish them from the surrounding mesenchyme though which they migrate (Chiquoine, 1954; Miliare, 1974; Tosney, 1978; Fujita £3 1., 1971). Thus, from the time the trunk 13 ~neural crest cells disperse from their dorsomedial location (9-11 days of development in the mouse) until differenti- ated melanocytes can first be detected in the skin by the presence of tyrosinase activity (18 days of development in the mouse), the location and behavior of the melano- blasts cannot be directly observed. Lacking the possibil- ity for immediate and direct observation of the differen- tiative process, the events leading to melanocyte differ- entiation can only be inferred by examination of the color, intensity and pattern of the fully differentiated pigmen- tation phenotype, g,g., the fur color pattern of the adult mouse. Melanogenesis The fur color of the adult mouse is determined by the type of pigment produced by the follicular melanocytes. A single melanocyte can produce two kinds of pigment, either the black-brown eumelanin or the yellow-orange phaeo- melanin. The general pathway for eumelanin synthesis was first proposed by Raper (1926) and was confirmed and ex- tended by Mason (1948) (see Figure 2). While the general outline of this pathway is still accepted, the precise structure of the melanin polymer remains unknown. Early investigators (Bu'Lock and Harley-Mason, 1951; Mason, 1959) suggested that melanin was a homopolymer of indole-5,6- quinone monomers. More recently, several authors have suggested that melanin may be a complex poikilopolymer 14 Tyrosine O2 tyrosinase dihydroxy- phenylalanine O2 tyrosinase Dopa quinone uy-cysteine-——qp v l 2-S-cysteiny1- 5—S-cysteinyl- dopachrome dopa dopa 5,6,dihy- ¢ ‘ droxyindole Eumelanin °2 indole-5,6- Phaeomelanin quinone Figure 2. Melanogenesis 15 formed from the random polymerization of all of the inter- mediates of the Raper-Mason pathway (Blois 23 Elk, 1964; Blois, 1969; Swan, 1973). The pathway for phaeomelanin synthesis consists of: 1) the formation of 2-S-cysteinyl- dihydroxyphenylalanine (2-S-cysteinyl-dopa) and 5-S- cysteinyl-dopa from cysteine and dopa quinone, and 2) the oxidation, cyclization and polymerization of the cys- teinyl-dopa into phaeomelanin (see Figure 2) (Prota, 1972; Prota and Thomson, 1976). Only the first two steps of eumelanin and phaeo— melanin synthesis, i.g., the hydroxylation of tyrosine to dopa and the oxidation of dopa to dopa quinone, are known to be enzymatically mediated. The formation of the rest of the intermediate compounds of the Raper-Mason pathway and the polymerization of eumelanin can proceed non-enzy- matically $2 zihgg, and are thought to do so $2.2lX2 as well. Traditionally, both of these enzymatic steps in mammalian melanocytes have been considered to be catalyzed by a single enzyme, tyrosinase (Lerner, 1953). Recently, however, Okun and his associates have revived the hypoth- esis of Hogeboom and Adams (1942) that, within the mam— malian melanocyte, tyrosine hydroxylation and dopa oxidation are catalyzed by two separate enzymes. Okun maintains that peroxidase mediated hydroxylation of tyro- sine to dopa is the normal and only source of dopa for melanogenesis, and that mammalian "tyrosinase" has only dopa oxidase activity (Edelstein E£ 31., 1975). However, 16 the more generally accepted hypothesis that mammalian tyrosinase is a bifunctional enzyme that can, and does, catalyze both tyrosine hydroxylation and dopa oxidation is strongly supported by the work of a variety of other investigators (Holstein ££.él-a 1973; Hearing, 1973a; Hearing and Ekel, 1975; Mufson, 1975; Abramowitz and Chavin, 1978). Isozymes of tyrosinase have been isolated from melanoma tissue and the melanocytes of a variety of mam- mals (Burnett sh él" 1967; Burnett and Seiler, 1969; Holstein E£.£l°: 1971; Chen and Singh, 1974). The normal mouse follicular melanocyte tyrosinase electrophoretic pattern consists of a fast migrating band, T and a 1, doublet of slower migration forms, T2 and T3 (Burnett SE élr: 1969; Holstein E£,£l-a 1971). Some of the isozymes clearly differ in amino acid composition, while others have very similar amino acid composition and show partial or complete immunological cross reactivity (Burnett, 1971; Ohtaki and Miyazaki, 1973, 1976). It has been suggested that some of the observed differences in electrophoretic mobility may be due to a difference in other proteins which may be tightly associated with the various tyrosinase isozymes (Holstein sh 2l°: 1971). However, among the two cross reactive tyrosinase isozymes in the Harding-Passy murine melanoma, the differences in electrophoretic mobil- .ity are due to neuraminidase-sensitive differences in glycosylation (Miyazaki and Ohtaki, 1976). 17 Melanosome development Within the melanocyte, tyrosinase activity and con- sequently, melanogenesis, is restricted to a discrete sub- cellular organelle, the melanosome (see Figure 3). The stage I melanosome originates as a membrane bound vesicle, devoid of obvious internal structure and tyrosinase acti- vity, derived from a dilation of the smooth endoplasmic reticulum (Moyer, 1961). The stage I melanosome gradu— ally develops an internal matrix of longitudinal helical protein (?) fibrils. As these fibrils are linked together by thinner cross fibrils, the stage I melanosome becomes a stage II melanosome (Moyer, 1961; Weiss, 1970). As the stage II melanosome matures, the longitudinal fibrils come into register so that longitudinal sections of stage II melsnosomes show transverse striations and cross sections show either a stippled or laminar pattern (Rittenhouse, 1968a; Hearing 2E _l., 1973). Stage III melanosomes have active tyrosinase (- tyrosinase-positive) and are distin- guished by the synthesis and deposition of melanin on the fibrillar matrix. Melanin deposition continues until the mature (stage IV) melanosome appears as an ovoid electron- dense body (Moyer, 1961, 1963). Phaeomelanosomes undergo a similar developmental sequence except that the internal matrix never shows the organized structure seen in the eumelanosome. The mature phaeomelanosome is a moderately electron-lucent spherical 18 Follicular Melanocyte Eumelano- Phaeomelano- genesis genesis stage I stage I vesicle vesicle matrix tyrosinase matrix assembly vesicle assembly stage II stage II stage III stage III tyrosine eumelano- phaeomelano- ‘L ) genesis genesis cysteine V stage IV stage IV Eumelanosome Phaeomelanosome Figure 3. Melanosome Synthesis 19 body whose internal structure is characterized by 20-503 vesicles embedded in a tangled fibrous matrix (Moyer, 1966; Sakurai EL él-9 1975). As mentioned above, tyrosinase activity is not present in the developing melanosome until the stage II to stage III transition. Although the stage II melanosomes are tyrosinase-negative, small vesicles (400-5003) derived from the Golgi-smooth endoplasmic reticulum show strong tyrosinase activity. These vesicles apparently penetrate the stage II melanosomal membrane and contribute their tyrosinase to the now melanogenic stage III melanosome (Hunter E£,él-’ 1970; Maul and Brumbaugh, 1971; Turner et 1., 1975). These vesicles remain as an integral part of the developing melanosome and have been identified in the stage IV melanosomes as small electron-lucent holes in the otherwise electron-dense, melanized melanosome (Jimbow and Fitzpatrick, 1973). Melanocyte-keratinocyte interaction As mentioned at the beginning of this section, the cutaneous pigmentation is the result of the interaction of the melanocyte with the non-follicular and follicular keratinocytes. 20 Non-follicular epidermis: In the non-follicular skin, the anatomical basis for this interaction is the epidermal melanin unit (EMU), which consists of one dendritic epidermal melanocyte and a surrounding population of keratinocytes which receive melanosomes from the melanocyte (see Figure 1) (Hadley and Quevedo, 1966). The actual mechanism of melanosome trans- fer can be separated into three major stages: dendrite insertion, apocopation and melanosome dispersion (Klaus, 1969; Okazaki sh Ei-: 1976). $2 X£££29 when a dendrite of a melanocyte moves adjacent to a keratinocyte, the membrane of the keratinocyte exhibits increased ruffling and gradu- ally surrounds the distal portion of the melanosome laden dendrite. In apocopation, the keratinocyte membrane grad- ually constricts around the intruded dendrite tip, squeezing off the distal tip of the dendrite as a double membrane vesicle containing 20-100 melanosomes (Mottaz and Zellickson, 1967). $2.Xl££2 studies also indicate that the keratinocyte appears to be the more active of the two cell types involved in the melanosome transfer, l.g., the keratinocyte may control the overall rate of melanosome dendrite apocopation and melanosome transfer (Breathnach, 1969; Cruishank and Harcourt, 1964). ' Within the keratinocyte, the fate of the melanosome- containing vesicle seems to depend on the size of the individual melanosomes within the vesicle (Wolff, 1973). 21 When the melanosomes exceed a critical size threshold (0.4-1.0um, depending on the system examined), the melano- somes are dispersed singly within membrane-limited vesicles of the keratinocyte. Malanosomes smaller than the critical size threshold are distributed in groups of two or more within membrane-limited vesicles (Toda 2h al., 1972; Okazaki EE 2l°: 1976). Follicular epidermis: In the follicular epidermis, the melanocytes also transfer melanosomes to the adjacent keratinocytes, which make up the cortical and medullary cells of the growing hair shaft. Before describing the kinetics of follicular melanosome transfer, it is necessary to briefly describe the morphology and growth cycle of the murine hair follicle. The fur of the mouse is comprised of four hair types: the zigzag, auchene, awl, and guard or monotrich (Dry, 1926). The zigzag hairs, while the smallest in size, make up the majority of the coat (71-83%). The rest of the coat is primarily awl hairs (24%), while the auchenes and guard hairs each represent approximately two percent (Dry, 1926). All hair follicles contain an epidermal and dermal component (see Figure 1). The epidermis is the sole source of the cells which make up the growing hair shaft, and is the immediate cellular environment of the follicular melanocytes. The dermis has developmental and trophic influences on the type of hair and pigment produced and 22 the initiation of hair growth (Sengal, 1974; Poole and Silvers, 1976a). All of the hair types display a cyclic growth pattern which can be divided into three stages: 1) anagen, the morphogenetic and proliferation phase which results in the formation of a complete hair, 2) catagen, a transitional phase, and 3) telogen, the resting phase in which the overtly inactive follicle contains the mature club hair (Chase E£,£l-9 1951). The melanocytes within the follicle also display a parallel cycle of proliferation and function. In the telogen follicle, amelanogenic melanoblasts can be found resting on the basement membrane just above the dermal papilla (Silver E£.2i°’ 1969; Potten and Chase, 1970). Telogen melanoblasts can be distinguished from adjacent keratinocytes by the lack of desmosomes and coarse clumps of microfilaments, sparsity of ribosomes, unusually few numbers of nuclear pores and the presence of dendrites (Silver 2£.él°9 1977). Histological observations and radiation induced depigmentation studies have established that the telogen zigzag follicles, contain only one or two melanoblasts, while the telogen follicles of the larger awls, auchenes and guard hairs may contain three to five melanoblasts (Chase, 1951; Potten, 1968; Potten and Howard, 1968; Potten and Chase, 1970; Silver 2£.2l-’ 1969). In a hair follicle, anagen can be initiated either spontaneously or by plucking the club hair from the follicle. 23 As the follicle enters a new anagen phase, the melanoblasts begin to proliferate and resume melanogenesis. In anagen I to anagen II (1-3 days post-pluck), the melanoblasts increase in cell size and show massive increases in rough and smooth endoplasmic reticulum, ribosomes and Golgi zones (Silver E£.£l-9 1977). By anagen III (3-4 days post- pluck) the melanocytes become tyrosinase-positive and the first stage IV melanosomes become visible (Chase SE hh., 1951; Chase, 1954; Kukita, 1957). By anagen V (8 days post-pluck) the follicle has attained its maximum diameter and maximum hair growth rate, which is sustained for the next eight to nine days (anagen VI). Beginning in anagen V and throughout anagen VI, the follicular melanocytes transfer melanosomes to the cortical and medullary kera— tinocytes of the growing hair (Chase EE hl,, 1951). ‘In addition to the initiation of melanogenesis and melanosome transfer, the follicular melanocytes have also undergone considerable proliferation between anagen I and anagen VI. Estimates for the number of melanocytes in anagen VI zigzag follicles vary from 4-8 (Chase, 1951), to 11-12 (Galbraith and Arceci, 1974), and 17-22 (Potten, 1968). The variability in the estimates of the number of follicular melanocytes is probably due to differences in the methods used to estimate the follicular melanocyte population size and strain differences between the mice used by the different investigators.’ The anagen VI awl and auchene follicles contain 20-40 melanocytes, while the T______I 24 anagen VI guard hair follicles contain 40 or more melano- cytes (Potten, 1968; Potten and Chase, 1970). By catagen (16-17 days post-pluck), the follicular melanocytes have ceased melanogenesis and appear moribund (Chase 33 él°’ 1951). When the follicle enters telogen (19 days post-pluck) melanotic melanocytes are no longer visible, but as noted above, melanoblasts can be located on the basement membrane above the dermal papilla. From the available data, it is not clear whether these telogen melanoblasts are part of a permanent non-melanogenic stem cell population or whether, during catagen, melanogenic melanocytes can revert to a non-melanogenic stem cell (melanoblast) condition (Silvers, 1979). Genetic Control of Pigmentation Uses 2; genetic variation As previously described, the second major phase of the analysis of a developmental model system involves the use of genetic variation to dissect the control mechanisms of the model system. Specifically, the genetic variation is used in two ways. First, a careful documentation of the phenotypic effects of a variety of mutations will allow the investigator to determine the number and kinds of genetic and environmental factors which affect the development of a model system. As an illustration of this use of genetic variation, consider the following example. In a nonagouti 25 b1adk (h/h;h[h) mouse, the cutaneous melanocytes in both the hair follicles and non-follicular ear and tail skin produce eumelanin. In congenic mice that carry the mutant allele, agouti yellow (hy/h), the follicular melanocytes produce phaeomelanin, while the non-follicular melanocytes of the ear and tail skin continue to produce eumelanin. The differential effect of hy on the two classes of cu- taneous melanocytes, follicular and non-follicular, demon- strates that although all cutaneous melanocytes are de— rived from the neural crest, their final melanogenic phenotype depends in part on: 1) the alleles present at the agouti locus, and 2) the anatomical location of the melanocytes within the skin. The second use of genetic variation involves the use of recombination or reconstitution experiments. In these types of experiments, normal and mutant components of the model system are combined in such a fashion so as to reveal the site of gene action of the mutant (and normal) allele. Again for the purposes of illustration, consider the following example. The two major functional components which give rise to the fur pigmentation phenotype are the follicular melanocytes and the follicular environment. Referring again to the agouti locus, it is possible, by the appropriate experimental techniques, to combine both normal (273) and mutant (hy/h) melanocytes and follicular environment in all four pairwise combinations. The results of such combinations are summarized in Table 1. From these 26 Table 1. Melanocyte Pigment in Melanocyte-Follicular Environment Recombination Grafts Melanocyte Follicular Environment Genotype Genotype fl/é éy/é h/h eumelanin phaeomelanin hy/h. eumelanin phaeomelanin 27 results, it is clear that the pigment phenotype expressed within the follicular melanocytes is controlled not by the agouti locus genotype within the melanocytes but rather by the agouti locus genotype of cells of the follicular environment in which the melanocytes reside. Thus the site of gene action of the agouti locus is within cells of the follicular environment, and a further analysis of the mechanism of gene action of the agouti locus should be focused on the follicular environment as well as the melanocytes themselves. Classification 3; mutations Within the mouse, 63 loci and more than 150 mutant alleles have been identified which affect the oculocu- taneous pigmentation phenotype. Of these 63 loci, approx- imately one quarter of them have been utilized in either of the methods outlined above to investigate the genetic control of the pigmentation phenotype. The following sec- tion of the literature review will discuss the phenotypic effects, and the site and specificity of gene action of some of these loci. The general classification scheme on which this discussion of the pigmentation loci is based is given in Table 2. This classification scheme based on the specifi- city, and the site of gene action of the various loci is adapted from a classification system described by Hadorn (1961). The subheadings under "specificity of gene 28 Table 2. Classification of Pigmentation Mutations Specificity of Gene Action 22:: Zition Cell Specific Tissue Specific Generalized (melanocyte) (neural crest) Intrinsic extension piebald dilute b-locus mottled pink eye beige dilute albino dominant spotting Extrinsic agouti pallid steel Unknown splotch? microphthalmia? 1 based in part of the classification scheme of Hadorn (1961; chapters 12 and 13) 29 action", X$£°: cell-specific, tissue-specific and general- ized, indicate the categories of cell types which are affected by a mutation at a specific locus. In this discussion, a cell-specific mutation is one whose pheno— typic effects are primarily restricted to the melanocyte. Similarly, a tissue-specific mutation is one whose pheno- typic effects can be ascribed to changes in a single tissue type, in this case, the neural crest. Finally, a generalized mutation is one which has widespread effects on both neural crest and non-neural-crest-derived cell types. According to this classification scheme, the site of gene action with respect to the melanocyte can be either intrinsic or extrinsicl. In the former case, the mutant genotype within the melanocyte itself is responsible for the changes in the pigmentation.phenotype. In the later case, the mutant genotype of the cells surrounding the melanocytes, and not the melanocytes' own genotype, is responsible for the changes in the pigmentation phenotype, 3.3., the case of the hy/h follicular melanocytes. 1The terms originally used by Hadorn (1961, pp. 185- 186) to describe the site of gene action were Pautophenic" and "allophenic" respectively. Since, however, the term "allophenic" has been redefined by Mintz (1967) to des- cribe mice which have been created by the fusion of two or more genetically distinct blastocysts, the terms "intrin- sic" and "extrinsic" will be used in this discussion to characterize the sites of gene action. 30 ‘In the category "unknown" site of gene action, the splotch and microphthalmia loci are specifically included for two reasons. Although the site of gene action of the splotch locus has not been identified, this locus seems to be involved in the differentiation of the neural crest from the general embryonic ectoderm (see Figure 1). It seems appropriate therefore, to include splotch in a discussion of the genetic control of melanocyte differen- tiation. The microphthalmia locus, being the subject of the experimental portion of this dissertation, is, of course, also included. Before describing the phenotypic effects of the mutations listed in Table 2, the distribution and func— tional state of the neural crest-derived melanocytes will be briefly described for the normal, nonagouti black (h[h;h[h), mousel. In this genotype, the follicular melanocytes of the general body produce eumelanin through- out anagen. The follicular melanocytes of the perineum and those surrounding the mammae and the base of the exter- nal ear produce phaeomelanin throughout anagen. The skin of neonate mice contains large numbers of eumelanogenic melanocytes (413 cells/mm2 in the epidermis of the two day old mouse). By 30 days of age, the epidermal melanocyte density has dropped to less than 5 cells/mm2 (Quevedo BE 3h., 1Unless otherwise stated, all other loci affecting the pigmentation phenotype will be assumed to be wild type. 31 1966). In the adult mouse, neither the epidermis nor the dermis of the interfollicular skin contain melanotic melanocytes. However, after treatment of the skin with a variety of irritants, 3.3., chemical mitogens or ultra- violet radiation, eumelanogenic melanocytes can be observed in the epidermis (Reynolds, 1954; Markert and Silvers, 1956; Quevedo and McTague, 1963). In the tail and external ear, both dermis and epidermis contain eumelanogenic melanocytes (Gerson and Szabo, 1969). The uveal stroma and a variety of internal organs also contain eumelano- genic neural crest-derived melanocytes (Markert and Silvers, 1956). Intrinsic, cell-specific loci Extension: The extension locus is an intrinsic, cell-specific locus which controls in part the type and quantity of pigment synthesized within the follicular melanocytes, and number of melanocytes present in non-follicular skin. Of the four alleles which have been identified at this locus, the’following discussion will focus on only one allele, recessive yellow (3). Within the hair follicles, the 3/3 melanocytes synthe- size phaeomelanin throughout anagen. In the juvenile coat, and to a lesser degree in subsequent molts, 3/3 also allows the synthesis and deposition of small amounts of 32 eumalanin in the distal portion of the hair. This distal eumelanin gives the 3/3 fur a dusky yellow appearance which is particularly prominent in the juvenile coat. Within the hair, the quantity of phaeomelanin synthesized by 3/3 is much less than the quantity synthesized by another phenotypically yellow mutation, agouti yellow (3y/3) (Hauschka 33,33., 1968). The dermis of the non-follicular 3/3 skin, choroid, hardarian gland and meninges all contain approximately normal numbers of eumelanogenic melanocytes. In contrast, the 373 ear epidermis contains fewer than normal numbers of eumelanogenic melanocytes (Poole and Silvers, unpublished data in Silvers, 1979). Two kinds of non-allelic interactions have been repor- ted between 3/3 and other pigmentation loci. First, 3/3 reduces the amount of white spotting caused by piebald, lethal spotting, belted, and microphthalmic white, but does not affect the degree of white spotting caused by splotch (Hauschka 33‘33., 1968; Lamoreux and E. Russell, 1979). Second, leaden (33), a recessive mutation which causes a dilution of the normally black fur color in E/E;33/3_ mice to a uniform gray color, is epistatic to recessive yellow, 3.3., E/E3lE/lfl mice are a uniform gray color and are phenotypically indistinguishable from E/E;33/3_ mice (Hauschka 33 33., 1968). Results from dermal-epidermal recombination grafts strongly suggest that 3/3 controls follicular melanocyte 33 phaeomelanogenesis from within the melanocytes themselves and that the extension locus genotype of the cells of the follicular environment does not affect follicular melano- cyte pigment synthesis (Lamoreux and Mayer, 1975; Poole and Silvers, 1976b). h-locus: The h-locus in an intrinsic, cell-specific locus which controls in part: 1) the number, size, structure and melanization of the melanosome, 2) the quantity and type of tyrosinase present in the melanocyte, and 3) the morphology of follicular melanocytes and the number of uveal melanocytes. Four alleles have been identified at the h-locus. This discussion will focus on the effects of two of these alleles, light (31:) and brown (3). The fur color of Eltlglt mice is pale brown in color, It while the fur of Elt/E and h /h mice is darker brown, intermediate between Elt/hlt and B/B mice. Within indi- vidual hairs of the EItIEIt mice, the melanosomes are pre- sent in large clumps which disrupt the normal morphology of the medullary cells. Proximally, the Elt/Elt hairs are nearly or completely devoid of pigmentation. Elt/E and filt/h hair has similar but less severely clumped melano- somes in the distal portion of the hair, and the pigmenta- lt/Elt hairs. tion extends more proximally than in the E Blt/_ has no visible effects on phaeomelanosome distribu- tion in either agouti (3/3) or agouti yellow (3y/3) 34 genotypes (MacDowell, 1950; Quevedo and Chase, 1958). The distal clumping of the melanosomes within the medulla of the hair appears to be caused by the lack of It normal cytocrine activity of the E /_ follicular melano- cytes. In early anagen, the Elt/ _ melanocytes have normal morphology with fully extended, melanosome-laden dendrites (- nucleofugal morphology). By anagen V, the melanocytes have retracted or stubby dendrites (- nucleopetal morphol- osy). Normal cytocrine activity of these nucleopetal melanocytes is inhibited and whole dendrites or entire melanocytes become embedded in the medullary cells of the growing hair. By anagen VI, the follicle has been nearly or completely stripped of its melanogenic melanocytes and the proximal portion of the hair remains almost or entire- ly unpigmented (McGrath and Quevedo, 1965; Sweet and Quevedo, 1968). Within the follicular Elt/_ melanocytes, the melano- somes are reduced in number and clumped in their distri- bution. The Elt/_ melanosomes are smaller and more round in shape than normal (E/h) melanosomes, but the intensity of melanosomal pigmentation appears similar to that of 3/3 melanosomes (Quevedo and Chase, 1958). Elt/_ does not alter the electrophoretic pattern of follicular tyrosinase lt/ isozymes (Holstein et al., 1967). The E _ non-follicular cutaneous melanocytes have a normal nucleofugal morphology, but within the uveal stroma, the melanocytes are reduced in number. The Elt/_ melanosomes of the neural crest-derived 35 uveal melanocytes show changes in size and shape parallel to those seen in the._B_1t /_ follicular melanosomes (Sweet and Quevedo, 1968; Pierro, 1963a). Brown (h/h) changes the fur and eye color of the mouse to a uniform brown color. All h/h neural crest- derived melanocytes have a nucleofugal morphology. Brown has no visible effects on agouti locus-directed phaeo- melanin pigmentation. Brown melanosomes are smaller and more round in shape than black (h/E) melanosomes (E. Rus- sell,1946, 1948, 1949a,b). Ultrastructurally, h/h dis- organizes the matrix of the developing melanosome and causes a pattern of reduced and granular melanization in the stage IV melanosomes (Moyer, 1961, 1963; Rittenhouse, 1968a; 33. Hearing 33_33., 1973). Brown (h/h) skin appears to have greater than normal (E/h) levels of tyrosinase activity (L. Russell and W. Russell, 1948; Fitzpatrick 33 33., 1958; Coleman, 1962). Under normal 33 vivo conditions, however, the total quan- tity of melanin synthesized by h/h skin is less than that of normal E/E skin (L. Russell and W. Russell, 1948; E. Russell, 1948; Foster, 1963a, 1965; Foster 33 33,, 1972). The dopa oxidase activity of the T1 tyrosinase isozyme is unaffected by h/h but both the T2 and T3 isozymes have greatly reduced dopa oxidase activity (Holstein 33 33., 1967, 1971; Quevedo, 1971). Rittenhouse (1965, 1968a) suggests that the primary effect of h/h is to alter the internal matrix of the 36 melanosome, thus altering the 3343333,levels of tyrosinase activity and melanosome morphology. It is difficult to imagine, however, how such an effect on the melanosome matrix could be related to the intermittent nucleopetal lt/ morphology of the E _ melanocytes. Pink eye dilute: The 3—locus is an intrinsic, cell-specific locus which affects melanosome morphology and melanization. The following discussion will focus on the 3 allele as representative of the pigmentary phenotypic effects of most of the 3—locus alleles. anhy/3 background, the follicular 3/3 melanosomes are smaller and more round than normal (37/3;E/E) melano- somes. On most 3(3 backgrounds, the follicular 3/3 melanosomes are even smaller than the 3y/3;3/3 melanosomes. The 3(3;3[3 melanosomes have a "shred-like" appearance and tend to form flocculent clumps within the medullary cells of the hair ( E. Russell, 1946, 1948, 1949a,b). Ultra- structurally, the condition of the 3/3 melanosomal matrix appears to vary with the anatomical location of the melano- cytes. The retinal and follicular melanosomes have a dis- organized internal matrix, possibly due to a reduction in the number of crosslinking fibers in the stage II melano- somes (Moyer, 1961, 1963; Sidman and Pearlstein, 1965; RittenhouSe, 1968b). In contrast, the choroidal melanosomes have no obvious matrix abnormalities (Hearing 33 33., 1973). 37 Pink eye dilute reduces tyrosinase activity in whole skin or skin homogenates (L. Russell and W. Russell, 1948; Foster, 1963a,1965; Coleman, 1962) and decreases the dopa oxidase activity of the T2 and T3 tyrosinase isozymes (Holstein 33 33., 1971). This decreased tyrosinase activ- ity is reflected in the reduced and delayed melanization of the stage III 3/3 melanosomes (Moyer, 1963; Sidman and Pearlstein,1965; Rittenhouse, 1968b). Under 33.33333 culture, 3/3;3/3 melanosomes from retinal epithelial melanocytes retain their abnormal matrix and reduced melanization. However, with tyrosine supple- mentation of the media, the melanization of the 3/3;3/3 melanosomes approaches normal levels, but the melanosomal matrix remains abnormal (Sidman and Pearlstein, 1965). Sidman and Pearlstein (1965) have suggested that the ef- fects of the h-locus alleles on melanogenesis may result from a failure of the mutant melanocytes to utilize the existing melanogenic enzyme system at a rate sufficient to achieve normal levels of pigmentation. Albino: The albino locus is an intrinsic, cell-specific locus which alters: 1) the size and number of the melanosomes, and 2) the enzymatic activity of tyrosinase. Six non- radiation-induced mutations have been described at the 3— locus. This discussion willfocus on four of these alleles: albino (3), extreme dilution (3e), chinchilla (cah), and 38 himalayan (3h). The phenotype of homozygous albino (3/3) on any genetic background in characterized by a complete lack of pigmentation in all melanocytes regardless of embryologic origin or anatomical location. This unpigmen- ted phenotype of 3/3 mice is not due to a lack of melano- cytes, but is rather due to a lack of functional tyrosinase within the melanocytes (Silvers, 1956, 1958c; Quevedo, 1957). The phenotypic effects of 3Ch and 3e are summarized in Table 3. In general, 3e and 3Ch have the strongest dilut- ing effect on 3y/_ phaeomelanin pigment, with a less severe effect on h/E eumalanin pigment, and an even slighter effect on h/h eumelanin pigment (Detlefsen, 1921; Feldman, 1922; Gruneberg, 1952). The fourth allele, 3h, is acro- melanistic, allowing pigmentation to develop only on the nose, ears and tail (Green, 1961). The effects of 3Ch and 3e on the melanin content, size and number of the follicular melanosomes parallel the effects of these alleles on overall fur color. Chinchilla causes a slight to moderate reduction in the melanin con- tent, size and number of the melanosomes, while 3e causes a much greater reduction in these same parameters. Like- h and 3e on the relative dilution wise, the effects of 3c of yellow, black and brown pigmentation, as described in Table 3, is reflected in similar relative reductions of melanin content, size and number of 3y/_, 3/3;3/h and a/ ;h/h melanosomes (E. Russell, 1946, 1948; Dunn and 39 Table 3. Effects of 3-locus Alleles on Fur Pigmentation Genetic Background c-locus Ay/_ e/esfi/E slash/P. E/E yellow black brown 3Ch/3Ch ivory sepia brown 3e/3 white very pale pale brown brown 3/3 white white white 40 Einsele, 1938; Gruneberg, 1952; Moyer, 1966). Ultra- structurally, the matrix of all 3Ch/_,l3h/ , 3e[_, and 3/3 melanosomes is normal, but melanin deposition in stage III melanosomes is reduced or completely absent, depending on the alleles present (Moyer, 1963, 1966: Hearing 33 33., 1973). The overall levels of tyrosinase activity are affected by alleles of the 3-locus. The rank order of tyrosinase activity of the 3-locus genotypes corresponds to the rank order of the intensity of fur pigmentation of the adult mice, 3,3., 3/323Ch/3Ch>3e/3e23/3 (L. Russell and W. Rus- sel, 1946; Coleman, 1962; Foster 33 33., 1972). Tyrosinase activity in 3h is temperature sensitive. Under both 33 33333 and 3_ vivo conditions, lowered temperatures promote h 3 /ch melanogenesis while higher temperatures inhibit 3h/3 h melanogenesis (Green, 1961; Coleman, 1962; Moyer, 1966). The tyrosinase isozyme pattern is also affected by the Ch/gCh) has a single isozyme with a 3-locus. Chinchilla (3 greater electrophoretic mobility than T , while 3h/3h has 1 two isozymes, one with a mobility equal to T1 and the other 2 or T3. Albino (3/3) has no detectable tyrosinase activity (Wolfe and Coleman, with a mobility less than either T 1966; Holstein 33 33., 1971). It is generally accepted that the 3-locus is the structural locus for tyrosinase. Evidence cited to support this hypothesis includes: 1) the parallel between 33 33333 3-locus tyrosinase activity and the 33 vivo levels of fur 41 pigmentation, 2) the i vivo and in vitro effects of 3h, 3Ch, and 3e on tyrosinase thermal stability (Bulfield, 1974), 3) the effects of cCh on the K and V of the - m max tyrosine hydroxylase activity of tyrosinase (Bulfield, 1974), and 4) the effects of 3Ch, 3h and 3 on the electro- phoretic mobility and dopa oxidase activity of the tyros- inase:isoaymesu Considering that epigenetic modifications are known to affect the electrophoretic mobility and enzymatic activiy of proteins, and that tyrosinase is a glycoprotein whose electrophoretic mobility can be altered by neuraminidase treatment, all of the observations noted above are only consistent with, but do not prove the hy- pothesis that the 3-locus is the structural locus for tyrosinase (Hearing, 1973b; Foster, 1967; Pomerantz and Li, 1974). Furthermore, the effects of the 3-locus alleles on melanosome size and number are difficult to reconcile with the tyrosinase structural locus hypothesis. Intrinsic, tissue-specific loci Piebald: The piebald locus (3) appears to be an intrinsic, neural crest specific locus which affects the number and distribution of the melanocytes and the number of enteric autonomic neurons. Piebald (3/3) causes white spotting of the fur without dilution of the surrounding pigmentation. The extent and distribution of the spotting pattern of the fur is largely but not completely dependent on the genetic 42 background of the mouse (Dunn and Charles, 1937; Charles, 1938; Dunn, 1942; Schaible, 1969). The internal organs also show partial or complete white spotting (Markert and Silvers, 1956; Mayer, 1965; Deol, 1970a, 1971, 1973). Again, depending on the genetic background, 3/3 mice may develop megacolon (Bielschowsky and Schofield, 1962). A second allele, piebald lethal (31), when homozygous, has complete penetrance for megacolon as well.as white spotting (Lane, 1966). The megacolonic mice have a marked decrease in the vagal parasympathetic innervation, and a total lack of the enteric sympathetic innervation of the distal por- tion of the colon (Lane, 1966; Webster, 1974). Concerning the etiology of the white spotting of 3/3, Mayer (1965, 1967a,b), utilizing neural tube-skin recombi- nation grafts, has shown that factors in both the embryonic neural crest and the skin environment contribute to the failure of the 3/3 melanocyte differentiation. Mayer con- cluded that g/g melanoblasts undergo apparently normal migration, but are defective in their ability to differen- tiate into functional melanocytes. Mayer also suggests that the skin environmental factors inimical to 3/3 melano- cyte differentiation are not directly related to the pie- bald locus, but rather are associated with modifying loci which affect the expressivity of piebald (3/3) white spotting, 3.3., the "k complex" described by Charles (1938). More recently, Mayer (1977a) has demonstrated that the en- tire 3/3 trunk neural crest, even those portions adjacent 43 to regions of the skin that are routinely white spotted in the adult, possesses presumptive melanoblasts, albeit de- fective, which can, in the appropriate skin environment, differentiate into functional melanocytes (33. Mintz, 1971a,b). Alternatively, based on the results of selection experiments for increased white spotting in 3/3 mice, Schaible (1969) has argued that 3/3 does not block melano- cyte differentiation, but rather inhibits melanoblast migration. This hypothesis of inhibition of migration is supported by the observation of reduced migration rates 1 embryos (Webster, 1974). in the vagal neuroblasts of 31/3 Considering that: 1) the observations of Mayer and Schaible were made on mice which differed substantially in their genetic background, and 2) the expressivity of 3/3, which reflects in some measure the severity of its effects on the melanoblast-melanocyte population, is highly dependent on the genetic background, it is most probable that 3/3 affects both the migration and differ- entiation of the neural crest melanocytes. Intrinsic, generalized loci Dilute: Dilute (3) is a locus with generalized phenotypic effects whose mutant alleles affect melanocyte number and morphology, central nervous system function and may also 44 alter phenylalanine metabolism.‘ The site of 3/3 gene action on melanocyte morphology is intrinsic to the melano- cytes. The site of the 3f3 gene action which causes the other aspects of the phenotype has not been conclusively determined. The fur of 3/3 mice is characterized by a dilution of both eumelanin and phaeomelanin pigmentation. Within the hair, the melanosomes in the medullary cells are distribu- ted in a clumped and irregular fashion (E. Russell, 1946, 1948; McGrath and Quevedo, 1965). According to E. Russell (1946, 1948, 1949b), dilute (3/3;3/E;3/3) melanosomes are slightly more round than normal (3/3;E/E;E/E) melanosomes, but the average volume of the dilute melanosome is the same as that of the normal melanosome. E. Russell (1948) also observed that dilute causes a 47% reduction in the total number of melanosomes deposited in the hair. Russell nevertheless concludes that the total volume of pigment deposited in the hair of dilute (3f3) mice is equal to or greater than the quantity of pigment deposited in the hair of normal (th) mice1 (33. Foster, 1963b). Ultrastructur- ally, the 3/3 melanosomes have a normal internal matrix and normal melanization (Moyer, 1963; Rittenhouse, 1968a). The 3/3 T2 and-T tyrosinase isozymes have reduced dopa 3 oxidase activity, but the activity of the T isozyme is 1 normal (Holstein 33 33., 1971). 1See appendix A for further comments on this conclusion. 45 All neural crest-derived 3/3 melanocytes have, to varying degrees depending on anatomical location, a nu- cleopetal morphology. This abnormal morphology disrupts the normal melanosome transfer between melanocyte and keratinocyte. The abnormal melanosome transfer leads to the irregular and clumped distribution of melanosomes in the medullary cells of the hair, and a reduction in the quantity of melanosomes in the non-follicular keratino- cytes (Markert and Silvers, 1956, 1959; Quevedo and Mc- Tague, 1963; Gerson and Szabo, 1969). Although the nucleopetal morphology of the 3/3 melanocytes is similar to that of gilt/g1t melanocytes, 3/3 pigmented melanocytes are able to remain in place within the hair follicle until late anagen VI, at which time some of the 3/3 melanocytes may become dislodged from the follicle and become embedded in the base of the club hair (E.Russell, 1949b; McGrath and Quevedo, 1965). The 3/3 genotype does not affect the number of cuta- neous melanocytes in the dermis or epidermis of adult non- hairy skin (Gerson and Szabo, 1969). However, 3/3 does increase the number and possibily the rate of proliferation of melanogenic melanocytes in the interfollicular skin of both neonate and adult mice (Quevedo 33 33., 1966). A series of whole skin transplantation studies and histological observations suggest that while 3/3 acts directly within the melanocyte to cause the nucleopetal morphology, the anatomical location of the melanocyte 46 determines to what degree the melanocyte will display the nucleopetal morphology (Markert and Silvers, 1956, 1959; Gerson and Szabo, 1969). A second allele of the dilute locus, dilute lethal (31), causes, not only a dilution of the pigmentation identical to 373, but also causes changes in behavior and phenylalanine metabolism. Juvenile 31/3l mice display clonic and tonic convulsions with opisthotonus and ataxia. The 3;/31 mice die shortly after weaning, presumably due to starvation (Searle, 1952). The 31/3l central nervous system has abnormalities in myelination and lipid syn- thesis, and the 31/31 adrenal has elevated adrenalin and noradrenalin levels. These morphological and biochemical changes are thought to be subordinate pleiotropic effects of the 34/31 behaviorally-induced malnutrition and stress (Hamburgh and Bornstein, 1970; Winterbourn 33 33., 1971; Doolittle and Rauch, 1965). Several authors have reported that 31/3l also alters hepatic phenylalanine metabolism (Coleman, 1960; Rauch and Yost, 1963; Zannoni and Moraru, 1970; Treiman and Tourain, 1973; Simler 33 33., 1977). Conversely, other authors have reported that 31/31 does not alter either hepatic phenylalanine metabolism or blood phenylalanine concen- trations (Zannoni 33 33., 1966; Mauer and Sidman, 1967; Woolf 33 33., 1970; Seller, 1972). The reasons for these contradictgry conclusions are not completely clear. How- ever, the differences may due in part to differences in 47 experimental design and technique. Even if 31/3l does actually affect hepatic.pheny1alanine metabolism, it is not clear how this effect of 31 is related to either the 1 effects of 31 on behavior, or the effects of 3 and 3 on melanocyte morphology. Mottled: Mottled (E3) is an X-linked locus with generalized phenotypic effects, whose mutant alleles disrupt the nor- mal function of a variety of copper-requiring enzymes. Six alleles have been identified at the mottled locus. In heterozygous females, the mutant mottled alleles cause patchy dilution of the hair. The hemizygous males, when viable, and the homozygous females have a uniform extreme dilution of the fur color (Silvers, 1979). The non- follicular melanocytes of the ear and tail skin of hemi- zygous brindle (E3br) males appear to be unaffected (Falconer, 1953). The various mottled alleles also cause a variety of other phenotypic effects including malformed hair and vibrissae, skeletal abnormalities, reduced skin tensile strength, aortic aneurysms and emphysema-like changes in lung structure (Phillips, 1961; Andrews 33_al., 1975; Fisk and Kuhn, 1976; Rowe 33 _3., 1974). Hunt (1974) has demonstrated that intestinal uptake of copper by mottled mice is normal but that the copper levels in the brain and liver are depressed. Hunt suggests that the mutant alleles of the mottled locus may block the 48 normal transport of copper within the cells of the intes- tinal wall or across the serosal surface of these cells. According to this hypothesis, the effects of mutant mottled alleles can be explained as the consequence of the depressed activity of‘a variety of copper-requiring enzymes, including tyrosinase (Hunt and Johnson, 1972; Hunt, 1974). In support of this hypothesis, Hunt and Skinner (1976) have reported that: 1) the activity of two copper-requiring enzymes, cytochrome 3 oxidase and superoxide dismutase, is reduced in E3br brain tissue, and 2) the pigment dilution in hemizygous brindle males could be partially corrected with copper chloride supple- mentation. Since the patchy dilution of fur pigment of the heterozygous mottled females is inconsistent with a sys- temic deficiency of copper, Rowe 33 33. (1974) have sug- gested that the copper transport defect of mottled must not act soley within the cells of the small intestine, but must act at the cellular level throughout the body. Like- wise, the presence of normally functioning non-follicular melanocytes in hemizygous brindle males is inconsistent with the hypothesis that mottled causes a systemic defi- ciency of copper. 49 Beige: Beige (33) is a locus with generalized phenotypic effects which disrupt the normal development of Golgi- derived storage vesicles in a variety of cell types. In the homozygous state, beige causes a uniform dilution of both retinal, epithelial and neural crest-derived pigmen— tation. The development and melanization of individual premelanosomes is fairly normal, but the stage III melano- 'somes show a tendency to fuse into large (3-11um) clumps which interfere with normal cytocrine activity and cause the nonuniform distribution of the fused melanosomal clumps within the hair. The melanocytes of the retina and uveal tract, although they do not participate in any melanosome cytocrine activity, also form the characteristic irregular clumps of melanosomes (Pierro, 1963b; Robison _3 33., 1975; Lutzner and Lowerie, 1972). Beige has no effect on the tyrosinase isozymes (Holstein 33 33., 1973). The non-pigmentary phenotypic effects of beige are manifold. Every cell type which produces large quantities of Golgi-derived storage vesicles, 3.3., pancreas, liver, gastric mucosa, adrenal pituitary, spleen, kidney, leuco- cytes and megakaryocytes, show a similar clumping of the Golgi-derived vesicles (Lutzner 33 33., 1967; Essner and Oliver, 1973; Oliver and Essner, 1975; Brandt 33 33., 1975; Holland, 1976). Most authors concur that the pri- mary defect caused by 33/33 is an alteration of the Golgi 50 vesicle membrane. The defective component of the membrane has not been identified, but the presence of lipid-like deposits in the lysosomes of kidney proximal tubule cells, hepatocytes, pancreatic acinar cells and lymphocytes may indicate a defective lipid component (Oliver and Essner, 1973; Essner and Oliver, 1974). Dominant spotting: Dominant spotting is a locus with intrinsic gene action in a variety of cell types including melanocytes, primordial germ cells, and hematopoietic stem cells. More than a dozen different alleles have been identified at the dominant spotting locus. This discussion will focus on the effects of two of the alleles, dominant spotting (E) and viable dominant spotting (Ev) as representative of the dominant spotting lacus. The general phenotypic effects of E and E? are summarized in Table 4. As can be seen from Table 4, E and Ev have three major phenotypic effects, anemia, sterility, and defects of pigmentation. More specifically, E and Ev cause a macro- cytic, hypoplastic anemia characterized by an increased diameter and mean cell volume of the erythrocytes (E. Russell,1949c; Attfield, 1951) and an elevation of the proportion of reticulocytes in the blood (deAberle, 1927; Niece 33 33., 1963). E and Ev cause this anemia by acting directly within the hematopoietic stem cells to reduce their capacity for proliferation and differentiation 51 Table 4. Phenotype of Dominant Spotting Alleles Genotype Phenotypic Description1 E/+ ventral white spot, otherwise normal pigmentation; normal viability and fertility E/E white fur with black eyes; severe anemia, usually early post-natal lethal; severe germ cell deficiency Evl+ ventral white spot, otherwise diluted fur pigmenta- tion; slight anemia; normal viability and fertility Ev/Ev white fur with black eyes, occasionally some ear skin pigmentation; anemia, less severe than E/E, may survive to maturity; usually severe germ cell deficiency 1summarized from Gruneberg (1942), Mintz and E. Russell (1957), Little and Cloudman (1937) and E. Russell (1949c) 52 ( E. Russell 33 33., 1956, 1959; Bernstein and E. Russell, 1959; Bennett 33 33., 1968).. Likewise, the effects of E and Ev on gametogenesis are characterized by a reduction in the number and rate of maturation of germ cells in the adult gonad (Coulombre and E. Russell, 1954; E. Russell and Fetke, 1958). The deficiency in germ cell number arises from reduced rates of migration and proliferation of the primordial germ cells during their migration from the yolk sac splanchno- pleure to the germinal ridges (Mintz and E. Russell, 1957). The general dilution of pigmentation of Evl+ mice is caused by a reduction in the size and number of melanosomes in the hair (E. Russell, 1949c). In non-cutaneous tissues, Ev reduces the number, size and distribution of the melano- cytes in several internal organs, while E/E and Ev/Ev cause a complete absence of neural crest-derived melano- cytes within the hair follicles, non-follicular skin and the internal organs (Markert and Silvers, 1956; Silvers, 1956 ; Deol, 1971, 1973; Mayer and Green, 1968). The lack of follicular melanocytes particularly in E/E and Ev/Ev mice appears to be the result of the inviability or fail- ure of differentiation of the dominant spotting melano- blasts. The gene action of dominant spotting that results in a lack of melanocytes appears to be intrinsic to the melanoblasts, 3.e., dominant spotting appears to act within the melanoblasts themselves and does not operate 53 through the cellular environment of the melanoblasts. The evidence to support this conclusion comes from two series of investigations: one utilizing skin-neural tube and dermal-epidermal recombination grafts (Mayer and Green, 1968; Mayer, 1970, 1973a) and the other using chimeric mice (Mintz, 1970). Using grafts of neural tube from 9 day embryos and skin from 11-18 day embryos, Mayer and Green (1969) and Mayer (1970) concluded that EV/Ev skin does not contain any melanoblasts, but thatEv/Ev skin can support the differentiation of melanoblasts derived from the 9 day +/+ neural tubes (neural crest). Furthermore, they con- cluded that Ev/Ev neural tubes do not produce any melano- blasts which can migrate into the skin (either Ev/Ev or +/+) and differentiate into melanocytes. In a further analysis of the effects of the dominant spotting locus on melanocyte differentiation, Mayer (1973a) reported the results of dermal-epidermal recombination grafts between Ev/Ev and normal skin. The results of these grafts are summarized in Table 5. In this series of experiments all of the (+/+ // E/Ev)1 grafts were pigmen- ted in both the hair and the dermis. Mayer concluded that 1This notation (+/+ // E/Ev), which will be used throughout this dissertation, indicates a grafts with +/+ epidermis combined with a E/Ev dermis. 54 Table 5. Dermal-Epidermal Grafts from Mayer (1973a) Graft Pigmented Pigmented Genotype Hair Dermis yes no yes no v 1 ' E/E epidermis 8 8 8 8 +/+ dermis +/+ epidermis 20 0 20 0 my" dermis 1both epidermis and dermis from 13 day embryos 55 this pigmentation was from +/+ melanocytes which were already in the +/+ epidermis before the time of the graft, 3.3., before 13 days of development. Subsequent work by Mayer (1973b) has confirmed this conclusion. In the reverse skin combination of (E/EY // +/+) only one half (8 of 16) grafts were pigmented in the hair and dermis. In the eight pigmented grafts Mayer (1973a) concluded that +/+ melanoblasts present in the dermis had migrated into the E/Ev epidermis where they became functional follicular melanocytes. The other eight (E/Ev // +/+) grafts had normal hair, but were completely unpigmented. Since hair morphogenesis requires the inter- action of both dermis and epidermis, the presence of normal hair in these unpigmented grafts indicates that both the dermis and epidermis of these grafts survived. However, the lack of pigmentation in these grafts indicates that either the +/+ dermis of these grafts did not possess a population of melanoblasts, or that theE/Ev epidermis of these grafts blocked the normal differentiation of the +/+ melanoblasts in both the dermis and the epidermis. Focusing on the first of these two possibilities, Mayer (1973a) considers two possible explanations for these results. First, Mayer suggests that theseresults may indicate that, similar to the chick embryo, mouse melano- blasts migrate within the epidermis and only secondarily invade the dermis. According to this hypothesis, the I variable results of the (E/Ev // +/+) grafts indicates 56 that the process of dermal invasion begins at 13 days of development, and that in only eight of sixteen grafts had melanoblasts migrated from the +/+ epidermis into the +/+ dermis before the grafts were made. Alternatively, Mayer (1973a) suggests that if melanoblasts migrate pri- marily through the dermis and only secondarily invade the epidermis, then the variable results of the (E/Ev // ‘+/+) grafts indicate that, by 13 days of development, melanoblast migration from the dermis into the epidermis is essen- tially completed and that few or no melanoblasts remain within the 13 day +/+ dermis. Subsequent work by Mayer (1973b) has disproved both of these alternative hypotheses. Using dermal-epidermal recombination grafts between albino (3/3) and black (3/3) skin from 11-14 day embryos, Mayer demonstrated that: 1) the normal route of migration of melanoblasts is through the dermis, 2) the melanoblasts begin to migrate into the epidermis on day 12 and by day 13, melanoblasts are present in considerable numbers in the epidermis, and 3) dermis from 13-14 day embryos retains a population of melanoblasts capable of normal differentiation in either dermis or epidermis. Having disproved both of these hypotheses which were based on the assumption that the +/+ dermis lacked a population of melanoblasts, one is left with the other possible explan— ation previously alluded to, namely, that the E/Ev 57 epidermis is capable, under some circumstances, of blocking the normal differentiation or survival of the +/+ melanoblasts in both the epidermis and the dermis of the QUE" // +/+) grafts. As mentioned above, a second line of evidence, this one using chimeric mice, is also consistent with the hypothesis that the unpigmented phenotype of the dominant spotting mice is the result of an intrinsic defect within the melanoblasts. Mintz (1970) has observed that most E/E’++ +/+1 chimeric mice are completely pigmented, with relatively few of the genetically mosaic animals showing any white spotting at all. .This lack of coat color mosaicism is in sharp contrast to most other coat color chimeric combinations, 3.3., c/c ++ C/C, in which pheno- typic mosaicism for coat color is the rule rather than the exception. Mintz (1969, 1970, 1971a,b, 1974) suggests that the relative dearth of white spotting in the W/W ++ +fi+ chimeras is due to the "preprogrammed clonal cell death" of the E/E melanoblast clones during embryogenesis. According to this hypothesis, following such E/E clonal cell death, the surrounding +/+ melanoblast clones would proliferate to fill in these "ghost clone" regions. In the mouse, late in embryonic development, melanoblasts 1The notation E/E ++ +/+ indicates a chimeric mouse derived from the fusion of two 8 cell embryos, one E/E and the other +/+. 58 normally become restricted in their ability to migrate through the skin. Thus, if the E/E melanoblast clonal cell death and +/+ melanoblast migration and repopulation were to occur sufficiently early in development, the result could be a completely pigmented E/E ++ +/+ chimeric mouse. If, however, the E/E melanoblast clonal cell death occurred late in development, the surrounding +/+ melanoblast clones would have insufficient time to fill in the regions of the skin_formerly occupied by the E/E melanoblasts before all melanoblast migration became restricted. In these chimeras, the regions of the skin which were not repopulated by the invading +/+ melanoblast clones would become white spots. To further support her contention that "preprogrammed clonal cell death", and not skin environmental factors is responsible for the white spotting seen in E/E, Mintz (1970) created E/E ++ +/+ chimeric mice in which each of the contributing genotypes were marked with a different H-Z histocompatibility allele. Grafts of skin, both pigmented and unpigmented, from these doubly marked b k k ;K/W ++ H-Z lH"; chimeras, 3.3., ErEb/EfiE _ _3_ _ ;+/+, were transplanted back to host mice homozygous for each of the H-2 alleles. In all cases, semi-rejection of the grafts ocurred, indicating that both the pigmented and unpigmented regions of the chimeric skin contained both E/E and +/+ cells. Such an admixture of both E/E and +/+ cells in both pigmented and unpigmented skin, Mintz concludes, argues against environmentally mediated white spotting and 59 strongly supports the hypothesis that E/E white spotting is intrinsically controlled within the melanoblast. In summary, the majority of the available evidence supports the conclusion that the major aspects of the phenotype of the dominant spotting locus, anemia, sterility and white spotting, are caused by an intrinsic failure of the proliferation and differentiation of three stem cell ' populations, the hematopoietic stem cells, the primordial germ cells and the melanoblasts. Extrinsic, generalized loci Agouti: Agouti is a locus with generalized phenotypic effects whose site of gene action with respect to cutaneous pig- mentation, is extrinsic to the follicular melanocyte. Seventeen alleles have been identified at the agouti locus. All of these alleles affect the synthesis of phaeomelanin within the follicular melanocytes. One allele, agouti yellow (3y) causes a drastic reduction in the number of epidermal melanocytes of the ear (Markert and Silvers, 1956) and plantar skin (Quevedo and Smith, 1963), and also decreases the ability of the non- follicular epidermal melanocytes to respond -by increased melanogenesis and mitosis—- to ultraviolet irradiation (Quevedo and Smith, 1963; Quevedo and McTague, 1963; Quevedo 33 33., 1967). 3y also drastically alters the 60 electrophoretic pattern of the tyrosinase isozymes. During 3y-directed phaeomelanogenesis, the dopa oxidase activity of T is greatly reduced while the T and T iso- 1 2 3 zymes are totally absent (Holstein 33 33., 1967, 1971). Finally, 3y also decreases the amount of white spotting caused by the mutant alleles at several of the white spot- ting loci, 333,, piebald, splotch, dominant spotting, and microphthalmia (Lamoreux and E. Russell, 1979, 33. Gruneberg, 1952). The various alleles at the agouti locus also cause a variety of non-pigmentary pleiotropic effects including obesity (Plocher and Powle, 1976), pre-implantation embryonic mortality (Cizaldo and Granholm, 1978), and changes in tumor susceptibility (Heston and Vlahakis, 1968b tooth and bone dimensions (Leamy and Sustarsic, 1978), and liver enzyme levels (Wolff and Pitot, 1972). It has been determined by a variety of grafting methods that the agouti alleles direct follicular melano- cyte phaeomelanogenesis by acting through the follicular environment, and not by acting directly within the follic- ular melanocytes (Reed and Henderson, 1940; Silvers and E. Russell, 1955; Silvers, 1957, 1958a,b). More specifically, results from dermal-epidermal recombination grafts have shown that the various alleles of the agouti locus, with the exception of 3y, exert their effects soley through the dermal component of the follicle, 3.e., the pattern of phaeomelanin synthesis by the 61~ follicular melanocytes is determined only by the agouti genotype of the dermal papilla of the follicle and is altogether unaffected by the agouti genotype of the melanoé cytes themselves (Mayer and Fishbane, 1972; Poole, 1974, 1975; Poole and Silvers, 1976a). In the case of 3y, this allele directs follicular melanocyte phaeomelanogenesis when it is present in either the dermis or the epidermis (Poole, 1964, 1975). Poole does note, however, that since the follicular melanocytes are located exclusively within the epidermal portion of the follicle, it is reasonable to assume that the dermally- directed influences of the agouti locus alleles on the follicular melanocytes are mediated by the follicular epidermis. Therefore, due to the experimental procedure used, Poole's results with 3y do not exclude the possibility that 3? dermis has irrevocably committed 3y epidermis to direct the follicular synthesis of phaeomelanin at some time in development before the dermal—epidermal recombina- tion grafts were made, 3.3., before 13 days of development. Attempts to test this hypothesis using epidermis from 11 day agouti-yellow (3y/3) embryos and dermis from 13 day non agouti (3/3) embryos were unsuccessful since no grafts of this age combination could be recovered (Poole, 1975). Concerning the mechanisms by which the agouti locus alleles exert their effects, Cleffman (1963, 1964) reported that agouti locus-directed phaeomelanogenesis could be influenced by the intracellular concentrations of 62 sulfhydral compounds. More recently, however, these ob- servations could not be confirmed under similar experimen- tal conditions (Knisley 33 33., 1975; Galbraith and Patrignani, 1976). Galbraith has made an extensive study of the factors affecting agouti locus phaeomelanogenesis and has con- cluded that while follicular mitotic rate and follicular melanocyte number are not influenced by the agouti locus alleles, the follicular diameter, and thus presumably the follicular cell.umss,affects the penetrance of agouti locus directed phaeomelanogenesis in individual follicles (Galbraith, 1971; Galbraith and Arceci, 1974). Thus for any given agouti locus allele, the smaller the follicle diameter, the more likely that follicle is to synthesize phaeomelanin (Galbraith, 1969). In summary, it is clear that the agouti alleles act at least through the dermal papilla to control follicular phaeomelanin synthesis. The causes of the action of 3y on epidermal melanocyte number, the tyrosinase isozyme pattern, expressivity of white spotting mutations, as well as the other non-pigmentary effects of the various agouti locus alleles remains unknown. Pallid: Pallid is a locus with generalized phenotypic effects, whose mutant allele (33) causes: 1) a dilution of oculo- cutaneous pigment, and 2) ataxia due to malformed or 63 absent otoliths (Lyon, 1951, 1953; Shrader 33 33., 1973). Pallid melanosomes (BE/22): regardless of anatomical location, are smaller than normal and have an ultrastruc- turally normal internal matrix, but fail to become com- pletely melanized (Theriault and Hurley, 1970; Hearing 33 33., 1973). Based on: 1) the therapeutic effects of manganese on otolith development in 33(33 embryos (Erway 33 33., 1971; Lim and Erway, 1974), 2) the effects of manganese depriva- tion on otolith development in normal (E3/Eg) embryos (Erway 33 33., 1966; Purichia and Erway, 1972; Shrader 33 33., 1973), and 3) the essential role of manganese in the incorporation of carbohydrates into mucopolysaccharides (Leach, 1971), Erway has suggested that the defect in 33/33 may be related to a systemic decrease in manganese- dependent mucopolysaccharide synthesis. In support of this hypothesis, Cotzias 33 33. (1972) have demonstrated that although manganese levels in the liver and intestinal tract of 33/33 mice are normal, the bone and brain manganese levels are depressed in 33/33 mice. These authors suggest that 33/33 may block the transport of manganese from the liver, thereby causing manganese deprivation in the rest of the body. Normal melanosomes contain large quantities of cations including manganese (Cotzias 33 33., 1964). The melano- somal matrix is Periodic Acid Schiffs positive and tyrosi- nase itself is a glycoprotein (Sidman and Pearlstein, 1965; 64 Miyazaki and Ohtaki, 1976). According to Erway's hypothe- sis, it seems possible that the 33/33-induced reduction in melanogenesis could be due to either a defect in the melanosomal matrix or a reduction of tyrosinase enzymatic activity due to a decrease in manganese-dependent gly- cosylation. Steel: Steel is a locus whose mutant alleles affect the development of several cell types including melanocytes, primordial germ cells and hematopoietic stem cells. More than thirty mutations have been described which behave as alleles of the steel locus. The following discussion will be restricted to two of the better studied alleles, steel (33), and steel-Dickie (S d ). The general phenotypic des- cription of these two alleles is summarized in Table 6. As can be seen from Table 6, the overall phenotypes of the steel alleles are quite similar to those of the dominant spotting alleles, 3,3., anemia, sterility and reduced pig- mentation. However, in contrast to dominant spotting, the site of gene action of steel is extrinsic to the stem cell populations, 3,3., steel acts through the microenvironments in which the stem cell populations develop. More specifically, the mutant steel alleles cause macrocytic, hypoplastic anemia with persistently elevated reticulocyte counts, reduced numbers of megakaryocytes and neutrophils, and abnormal granulocytopoiesis (Bennett, 65 Table 6. Phenotype of Steel Alleles Genotype Phenotypic Description1 §3/+ dilution of fur pigmentation, white spotting on extremities, ventrum and head; moderate anemia, normal viability; germ cell deficiency but fertile 33/3_ white furz, black eyes; severe anemia, usually embryonic lethal; no germ cells Sldl+ similar to 33/+ ‘33d/Sld white fur, black eyes; severe anemia, may survive up to one year; sterile 1from Sarvella and L. Russell (1956), Bennett (1956), E. Russell and Bernstein (1966) and Deol (1970a) 2Since 33/33 is an embryonic lethal, the fur color was indirectly determined. See text for details. 66 1956; E. Russell and Bernstein, 1966; Ruscetti 33 33., 1976). It has been repeatedly demonstrated in a wide variety of experimental systems that the defect in steel hematopoiesis is in the cellular microenvironment of the stem cell populations. Specifically, 33/33d or _3d[3_d hematopoietic stem cells will, when placed in a non-steel environment, undergo normal proliferation and differentia- tion. Conversely, non-steel hematopoietic stem cells, when placed in a steel environment, fail to proliferate or differentiate normally (Cole 33 33., 1975; Adler and Trobaugh, 1978; Mintz and Cronmiller, 1978). As noted in Table 6, 33 and 3_d also cause reduced fertility or complete sterility. The embryonic gonads of ,fil/fil are completely deficient of germ cells, while the gonads of‘33/33d and_S_3d/33d embryos and adults contain only very few germ cells (Bennett, 1956; McCoshen and McCallion, 1975). This steel-induced deficiency in germ cells is the consequence of severely reduced rates of proliferation (and possibly migration) of the primordial germ cells during their migration from the yolk sac splanch- nopleure to the germinal ridges (Bennett, 1956; McCoshen and McCallion, 1975). The tissue site of action of steel on the primordial germ cell proliferation has not been determined. The mutant steel alleles also affect cutaneous pig- mentation. 33/+ causes a general dilution of the fur pigmentation and can also cause white spotting on the 67 ventrum, extremities and top of the head. 33/+ does not affect the number or distribution of the non-cutaneous melanocytes (Mayer and Green, 1968). Since 33L33 is a late embryonic lethal, the effects of 33/33 on fur pig- mentation cannot be directly determined. To surmount this difficulty, skin from 14-15 day embryos was transplanted to subcutaneous sites in histocompatible adult hosts. Skin grafts derived from +/+ and 33l+ embryos formed pigmented skin, while grafts from 33/33 embryos developed only unpigmented hair (Bennett, 1956). ld/+ mice also have slightly diluted fur pigmentation and occasionally have a white spot on the ventrum. S d/_S_3d and S / ld mice have completely white fur and black eyes. In §_j33d, the optic pigment is confined to the retinal epithelium (Mayer and Green, 1968). Although not specifically documented in the literature, it is generally accepted that the follicles of the unpigmented regions of fur of steel heterozygotes and homozygotes lack amelanotic melanocytes. 33/33d mice are completely devoid of neural crest-derived pigmentation in all non- cutaneous tissues which have been examined (Mayer and Green, 1968; Deol, 1970s). As to the tissue site of action of steel on fur pigmentation, Mayer and Green (1968) and Mayer (1970), using neural tube-skin grafts, have demonstrated; 1) skin from 13-18 day S d/ 1C1 or S /_§_l_d embryos will not support the differentiation of +/+ melanoblasts into functional 68 melanocytes, and 2) all neural tubes from 9 day embryos (45 of 45) from 33l+ x 33dl+ matings produced pigmented melanocytes when combined with melanoblast-free skin of 11 day +/+ embryos. Mayer and Green concluded that the neural crest from steel embryos is capable of producing functional melanoblasts, but that normal differentiation of the steel (33(33d or 33d/33d) melanoblasts is inhibited in some fashion by steel skin. To further investigate the effects of steel skin on melanocyte development, Mayer (1973a) made a series of dermal-epidermal recombination grafts between +/+ and 33/33d skin. From the results of these grafts, Mayer concluded that the presence of S /S d in either the dermis or epidermis blocks the development of' normal 'pigmentation in that portion of the skin which is mutant. Mayer does caution, however, that these results cannot differentiate between possible mechanisms by which steel may interfere with pigment development, 3,3,, inhibition of migration, survival or differentiation. While embryonic steel skin blocks the development of embryonic melanocytes, adult steel skin is not inimical to adult melanocyte function. When small black non-steel ear skin grafts are placed in the center of well established d S /_S_3d trunk skin grafts, some 33/8 follicles immediately adjacent to the +/+ ear skin grafts become pigmented (Poole and Silvers, in Silvers, 1979). From these results, Silvers (1979) concludes that either the effect of steel skin on pigment formation is a transitory one, acting only 69 during embryogenesis, or that melanocytes are susceptible to the effects of steel skin only at some specific early stage in their differentiation, and that melanocytes past this critical stage are unaffected by a steel skin envi- ronment . Other loci Splotch: The splotch heterozygote (33/33) is characterized by non-diluted fur coloration with white spotting on the ventrum, feet and tail (Auerbach, 1954; Silvers, 1956). Homozygous splotch (33/33) is an embryonic lethal which kills the fetus at approximately 14 days of devel- opment. Retinal epithelial pigmentation of the 33/33 embryos is normal. However, 33/33 embryonic skin does not develop any pigmentation under culture conditions in which normal (33/33) embryonic skin routinely produces pigmen- tation (Auerbach, 1954). The non-pigmentary effects of 33/33 include excessive growth of the rhombencephalic and lumbosacral region of the neural tube, and disruption of the normal development of the dorsal root spinal ganglia and sympathetic ganglia. Notwithstanding these apparent deficiencies of some neural crest-derived cell types, the pre-migratory neural crest appears histologically normal (Auerbach, 1954). Auerbach has suggested that 33/33 may affect the 70 general differentiation of the trunk neural crest. Considering the apparent normality of the early neural crest, and the subsequent hyperproliferation of the era-v nial and spinal neural tube, Auerbach favors the possi- bility that 33/33 may affect the intensity of the primary embryonic induction. Raven and Kloos (1945) demonstrated that the intensity of a neuralizing stimulus affected the relative proportions of neural and neural crest derivatives which were induced. Based on the conclusions of Raven and Kloos (1945), Auerbach proposed that the primary effect of 33/33 may be to increase the intensity or duration of the neuralizing stimulus, thus causing an excessive production of neural tissue at the expense of the adjacent neural crest and its derivatives. Microphthalmia: Ten alleles have been described at the microphthalmia locus. The following discussion will focus on one of these alleles, white (E3Wh)1. Since the experimental portion of this dissertation is concerned with the gene action of E3Vh, the phenotype of this allele will be 1To avoid confusion of this allele name with the use of "white" as an adjective indicating the lack of color, E3Wh will be referred to either by its symbolic designation or the term "microphthalmic white". 71 described in some detail. ‘E3Wh/+ causes a uniform dilu- tion of eumelanin fur color and white spotting on the ven- trum and flanks. The overall intensity of pigmentation of the E3Whl+ fur is slightly lighter than that of dilute (3/3) mice, but is darker than the double mutant h/+). The ear and tail skin of E3Whl+ mice also (g/gngf' have less pigmentation than the normal mice ((Grobman and Charles 1947; Gruneberg, 1953). The melanocytes of the internal organs are reduced in number or are completely absent (Markert and Silvers, 1956, Deol, 1970a, 1971, 1973). In the eye, the choroid is partially (Deol, 1970a, 1971, 1973) or completely unpigmented (Markert and Silvers 1956). In some animals the intensity of pigmentation of the outer layer of the iris is also reduced. The retinal epithelium and inner layer of the iris have approximately the same amount of pigmentation as that of normal mice (Deol, 1971, 1973). E3Whl+ appears to reduce, to some degree, the number of melanosomes in the cortex and medulla of the hair. In the cortex, E3Whl+ delays the onset of deposition of melanosomes so that the distal unpigmented portion of the hair cortex is longer than normal (Wolfe and Coleman, 1964} Within the medulla, E3Whl+ restricts the distribution (or quantity) of melanosomes in'a more regular fashion than does 3/3, 3.e., E3Wh/+ does not cause a clumping of the medullary melanosomes (Grobman and Charles, 1947). h wh The homozygous microphthalmic white mice (E3w IE3 ) 72 lack any trace of neural crest-derived pigmentation and have only slight and irregular pigmentation in the inner layer of the iris and little or no pigmentation in the retinal epithelium (Grobman and Charles, 1947; Gruneberg, 1953; Markert and Silvers, 1956; Packer, 1967; Deal, 1970a, 1973). Neural crest-derived melanocytes have not been h wh detected in the follicles or skin of either the E3? /E3 mice or the unpigmented regions of E3Vhl+ mice (Silvers, 1956; Rittenhouse, 1965). However, in a situation similar h wh IEI hostile to the terminal phases of non-(E3?h to that seen in steel skin, neonateE3w skin is not /E3Vh) melano- cyte differentiation. When normal (+/+) neonate skin is h transplanted to E3w /E3Wh neonate hosts, melanoblasts from the edges of the +/+ graft migrate into the adjacent wh wh E3 /E3 follicles and produce pigmented hairs (Silvers and E. Russell, 1955). At the biochemical level, Wolfe and Coleman (1964) measured whole skin-homogenate tyrosinase activity and reported that E3Whl+ reduced by 452 the rate of 14C-tyro— h wh sine incorporation in neonate mouse skin, while E3w /E3 neonate skin incorporated 14C-tyrosine at 2% of the normal rate. Since these tyrosine incorporation assays were not controlled for peroxidase activity, it is quite probable h wh that the "tyrosinase activity" of E3? /E3 skin is actual- ly peroxidase activity. ‘E3Wh also has several phenotypic effects which are not directly related to the pigmentation phenotype. E3Wh/E3Wh 73 causes a failure of closure of the choroid fissure which in turn leads to variable but severe microphthalmia. The failure of choroid fissure closure is caused by either eversion of the nervous layer of the retina (Packer, 1967), hyperplasia and inversion of the retinal pigment epithelium (Konyukhov and Osipov, 1968) or persistence of the base- ment membrane at the apposed edges of the choroid fissure h/E3Wh mice (Berman and Pierro, 1969). The sclera of E3w is thinner and less well developed than the sclera of normal mice (Konyukhov 33 33., 1965). In both E3Whl+ and E3Wh/E3Wh mice, the inner ear is consistently abnormal. Within the cochlea, the neural epithelium of the organ of Corti and the stria vascularis show a failure of normal development and degeneration (Deol, 1970a). The dorsal root spinal ganglia and adrenal medulla of E3Wh/E3Wh mice are reduced in volume and cell number. The size of the adrenal cortex is unaffected by E3Wh (Konyu- khov 33 33., 1965; Konyukhov and Osipov, 1968). The growth h curves and adult body weights are also unaffected by E3w (Gruneberg 33 al., 1965). It has been suggested that the primary action of E3Wh is to disrupt the normal differentiation of the neural crest (Konyukhov 33 33., 1965; Konyukhov and Osipov, 1968). This hypothesis rests on the observation that E3Wh disrupts the normal development of a variety of tissues and cell types, viz., melanocytes, adrenal medulla, dorsal root _ - M.-- all...” 74 spinal ganglia, meninges, and sclera, which are all derived from the neural crest. Likewise, the failure of closure of the choroid fissure and lack of retinal epithelial pig- mentation except in the iris, can be explained as a failure of proper retinal epithelial development due to a lack of normal inductive stimuli from the neural crest-derived sclera to the developing optic cup (Stroeva, 1960, 1967; Lopashov, 1963; Lopashov and Stroeva, 1961; Genis-Galvez, 1965; Giroud, 1957). The development of the retinal epi- thelium (and its subsequent pigmentation) from the poste- rior outer wall of the optic cup is absolutely dependent on the presence and close apposition of cranial mesenchyme, 3.3., neural crest-derived tissue which will become the sclera (Stroeva, 1960; Lopashov, 1963; Lopashov and Stroeva 1961). The differentiation of the iris, including its pigmentation is dependent on the presence of a lens (Lo- pashov, 1963; Genis-Galvez, 1965; Giroud, 1957). In wh/ h E3 E3? mice, the sclera is underdeveloped (Konyukhov 33 33., 1965), but a lens, albeit defective, is nearly always induced (Horch 33 33., 1978). The inner ear malformations can also possibly be explained by the hypothesis of abnormal neural crest devel- opment. Within the inner ear, the acoustic ganglion is partially derived from the cranial neural crest (Bartelmez, 1922; Adelmann, 1925; 33. Halley, 1955: Batten, 1958). Based on the widespread occurence of mutations which simul- taneously affect inner ear morphogenesis and oculocutaneous 75 pigmentation (Searle, 1968a), Deol (1967, 1970a,b) has sug- gested that the neural crest, possibly through the acoustic ganglion, may have a trophic or inductive role in the mor- phogenesis or maintenance of the inner ear. Developmental Mechanisms of White Spotting Mode 33 3ene action Before discussing the several hypotheses of the devel- opmental mechanisms of the spotting mutations, it is neces- ary to make a few further comments on the general method of analysis of developmental systems using genetic varia- tion. When investigating the genetic control of a develop- mental system, in this case, neural crest-derived pigmen- tation, three aspects of the genetic control of the system must be considered, 333., specificity of gene action, site of gene action, and mode of gene action. The first two of these aspects have been previously discussed, but the third, the mode of gene action, has not yet been considered During the development of the normal pigmentation phenotype of the mouse, melanocytes undergo a variety of cellular processes, 3.3., maintemmuuametabolism, prolifera- tion, migration and differentiation. For purposes of illustration, a partial list of the cellular processes in- volved in the production of the normal follicular pigmen- tation phenotype are summarized in Table 7. An alteration of one or more of these cellular processes by a specific 76 Table 7. Cell Processes in the Development of Follicular Pigmentation 1. Neural crest determination from embryonic ectoderm 2. Melanoblast determination from the neural crest 3. Melanoblast migration into dermal mesenchyme 4. Melanoblast survival in dermal mesenchyme 5. Melanoblast proliferation in dermis 6. Melanoblast migration from dermis into epidermis 7. Melanoblast survival in epidermis 8. Melanoblast proliferation in epidermis 9. Melanoblast migration into follicle 10. Melanoblast survival in follicle 11. Melanoblast to melanocyte differentiation in follicle 12. Melanocyte proliferation in follicle 13. Melanosome morphogenesis 14. Melanin synthesis 15. Melanosome transfer to follicular keratinocytes 77 mutation (which results in a change in the pigmentation phenotype) will be, for the purposes of this discussion, referred to as the "mode of gene action" of that specific mutation. The methods by which the specificity and site of gene action can be determined, 3,3., phenotypic observation and recombination experiments, have previously been discussed and are fairly straightforward in their application. How- ever, the methods of analysis which are appropriate to use for the investigation of the mode of gene action of pigmen- tation mutations present some difficulties. For example, an investigation of the mode of gene action of a mutation that affects follicular pigmentation necessitates an exam- ination of the normal and mutant cellular processes of development that are outlined in Table 7. However, as pre- viously mentioned, melanoblasts cannot readily be identi- fied, much less have their cellular processes investigated, during most of their development. Specifically, the melano- blasts are, by current methods, inaccessible to direct ob- servation from the time they migrate from their dorsomedial position until they begin melanogenesis. For loci whose mutant alleles affect the pigmentation phenotype primarily by altering melanogenesis, 3,3., albino, pink eye dilute or extension, this inability to detect and examine the melano- blasts is not a serious difficulty in the analysis of the mode of gene action. However, for those mutations that cause white spotting, this inability to detect and examine 78 melanoblasts during their development presents serious obstacles to the investigation of the mode of gene action of these white spotting mutations. If the melanoblasts cannot be directly detected, it is difficult to determine why they fail to undergo a normal developmental sequence to melanogenic melanocytes. Hypotheses 33 33ne action Notwithstanding the difficulties in obtaining direct information on the mode of gene action of white spotting mutations, several authors have made some pertinent obser- vations. In a histological study, Schumann (1960) observed that the appearance of differentiated melanocytes in the skin of piebald (3/3) embryos was delayed by several days with respect to the time of melanocyte appearance in normal (3/3) embryonic skin. Schumann concluded that the primary mode of gene action of 3/3 was to retard the rate of migra- tion of the cutaneous melanoblasts. While Schumann's data are consistent with a hypothesis of delayed migration, some observations of Rawles (1947) and Mayer (1973b) have clearly demonstrated the presence of competent melanoblasts in these embryonic regions by 12-13 days of development. Given this four to five day time lag between the time of arrival of the melanoblasts and the ability to detect the melanocytes, it is entirely possible that the 3/3 melano- blasts have normal rates of migration and the delay in melanocyte appearance detected by Schumann is caused by a 79 delay in the differentiation of 3/3 melanoblasts into melanogenic melanocytes. The observations of two other authors, Schaible (1969) and Mintz (1969, 1970, 1971a,b, 1974) are of par- ticular interest since both of these investigators have worked with E3Wh and have suggested a specific mode of gene action for this mutation. Based primarily on the phenotypic results of selection experiments designed to increase the expressivity of the white spotting phenotype of E3Whl+ mice, Schaible (1969) concluded that: 1) all of the cutaneous melanocytes are clonally derived from 14-16 primordial melanoblasts, and 2) the white spotting pheno- type of the E3th+ mice is the result of a decrease in the proliferation and migration of the clonal descendents of these primordial melanoblasts. Due to this reduction h/+ in the proliferation and migration rates of theE3w melanoblasts, and the normal inhibition of extensive melanoblast migration after 20-21 days of gestation, h/+ Schaible considers that the white spots of the E3w mice represent regions of skin which were never invaded by melanoblasts. Schaible's hyopthesis of reduced proli- feration of neural crest-derived melanoblasts is consistent with the observations of Konyukhov _3 33. (1965) and Konyukhov and Osipov (1968) on the decreased cell number of other neural crest-derived cell types. However, in the absence of direct evidence concerning the proliferation kinetics or migration rates of E3Wh/+ melanoblasts, other 80 modes of gene action, 3,3., increased cell death, or failure of melanoblast differentiation, that could also result in white spotting cannot be eliminated from consi- deration. Based primarily on the examination of the cutaneous pigmentation phenotypes of chimeric and non-chimeric mice, Mintz (1969, 1970, 1971a,b, 1974) has concluded that the mode of gene action of all white spotting mutations is either: 1) the intrinsic "preprogrammed clonal cell death" of some or all of the cutaneous melanoblasts, or 2) a failure of migration or proliferation of some of the viable melanoblast clones. A summary of the evidence supporting these conclusions is given below. Chimeric mice can be created by the fusion of two genetically distinct blastocyst stage embryos, followed by a short 33 33333 culture and reimplantation into a foster pseudopregnant female (Mintz, 1967). When colored- albino (3/3 ++ 3/3) chimeras are created, the fur pigmen- tation phenotype of these mice can be considered to be a series of more or less discrete transverse stripes of colored and white fur extending laterally from the dorsal midline of the mouse toward the flanks and ventrum. By the examination of a large number of these chimeras, Mintz (1967) has concluded that the cutaneous pigmentation of the mouse is derived from 34 primordial melanoblasts which are arranged pairwise along the dorsal midline of the 5-7 day mouse embryo. According to Mintz, in the normal 81 (non-chimeric) mouse, the clonal descendents of these primordial melanoblasts migrate in a predominantly lateral direction to fill in the entire skin of the mouse with viable melanoblasts. In contrast to the stripped pigmentation pattern seen in most all EIE ++ 3I3 chimeras, chimeric mice created with white spotting mutations, 3,3., E3Vh, show a markedly different pigmentation pattern. The frequency of wh wh ‘E3 /E3 ++ +/+ chimeras which had any white spots at all was considerably less than the frequency of 3/3 ++ 3/3 chimeras which had unpigmented regions (Mintz, 1971a,b). h wh Furthermore, of those E3w IE3 ++ +/+ chimeras which did possess white spots, ...the location and shape of a white area often convinvingly corresponded to a single [melano- blast] clone..." (Mintz, 1970). Based on the relative h wh +3 +/+ infrequency of white spotting in the E3w IE3 chimeras, and the pattern of the unpigmented regions in those chimeras which did show white spots, Mintz (1971b) h wh 3+ +/+ concludes that "...the white areas [of the E3? IE3 chimeras] are apparently left by [melanoblast] clones that formed and proliferated, but subsequently died prenatally at a predetermined time.", 3.3., intrinsic preprogrammed clonal cell death. Mintz (1971a) further contends that the white spotting pigmentation phenotype seen in single genotype mice 3.3., E3WhI+, is the result of the prepro- grammed clonal cell death of some but not all of the melano- blast clones of the single genotype mouse. 82 ’This hypothesis of intrinsic preprogrammed clonal cell death may be the correct interpretation of the mode of gene action for some white spotting mutations. However, in the absence of direct evidence of clonal cell death, other modes of gene action 3,3., clonal failure of migra- tion or differentiation, as the proximate cause for white spotting cannot be excluded. Furthermore, the observa- tions of Mayer (1970, 1977a) and Mayer and Green (1968) on: 1) the viability of 3/3 melanoblasts derived from regions of the neural crest that are adjacent to white spotted regions of the adult fur, 2) the ability of steel melanoblasts to undergo normal differentiation when trans- planted into a non-steel environment, and 3) the inability of the steel skin environment to support the differentia- tion of non-steel melanoblasts, all strongly suggest that intrinsic preprogrammed clonal cell death of melanoblasts is not the correct interpretation for the white spotting phenotype in all white spotting mutations. wh Homology of E3 and Eh Searle (1968a) has summarized the methods which can be used to establish gene homology between species. These methods involve the comparisons of: 1) the nucleic acid sequences of the chromosomal loci or their RNA, 2) the amino acid sequence of the polypeptide gene products of the loci (provided that the loci in question direct the production of polypeptides), 3) the mechanism (site, 83 specificity and mode) of gene action, 4) the phenotypic effects of the mutant alleles of the loci, and 5) the epi- static interactions between loci within each species. Of the methods listed above, clearly the first two types of comparisions, nucleic and amino acid sequences, provide the most direct information on genetic homology. However, these kinds of information are not available for E3Wh or Eh, or any other mammalian pigmentation locus. Informa- tion on the mechanism of gene action and phenotypic effects, while clearly not as definitive as the first two kinds of information, are the kinds of information most commonly used to "establish" gene homologies. However, the strength of any conclusions based on comparisons of gene action and phenotype must be tempered by an appreciation of the radical differences in genetic background in which the two loci must be operating. The last method, epistatic inter- actions, can be best illustrated by the following example. Suppose that two loci, 3 and 3' in the mouse and hamster respectively, have already been established to be homol- ogous loci, and that two other loci, 3 and 3, in the mouse and hamster, which appear phenotypically similar, are being investigated to determine their possible homology. If the epistatic interactions between the 3 and 3 loci in the mouse are phenotypically similar to the interactions of the 3 and 3' loci in the hamster, then these phenotypic similarities suggest that the 3 and 3 loci in the mouse and hamster are homologous. 84 Having considered the methods by which gene homology can be investigated, the following section will describe the phenotype of the hamster mutation anophthalmic white (Eh), and a second locus cream (3) whose epistatic interactions with Eh are relevant to the consideration of homology between E3 and E3Wh. The normal hamster (EIE;33I33) has agouti coloration on the dorsal fur and pale yellow fur on the ventrum. The ears, perineum and costovertebral gland are all black by virtue of eumelanogenic dermal melanocytes. Dermal melanocytes are also found in the hairy skin, around the upper hair canal of some, but not all, of the follicles. The eyes of EIE;3h/33 hamsters are dark black (Robinson, 1958; Ghadially and Barker, 1960; Rappaport 33 33., 1963). Hamsters carrying homozygous cream (3/3;33I3h) have dark yellow fur on the dorsum and pale yellow fur on the ventrum with ventral white spotting. The ears, perineum and costovertebral gland are black from eumelanogenic melanocytes. However, 3I3;EhIEh hamsters do not have the dermal perifollicular melanocytes found in the EI_;_hI_h hamsters (Robinson, 1955, 1958; Illman and Ghadially, 1960). By virtue of their similar phenotypic effects on follicular and non-follicular pigmentation, and the lack of other pleiotropic effects, cream (3) in the hamster and recessive yellow (3) in the mouse, are considered to represent homologous loci (Robinson, 1955, 1958; Hauschka t 1., 1968; Searle, 1968a,b). While phenotypic 85 similarity is not definitive proof, the putative homology of these two loci is also supported by the observation that in 13 mammalian species from 5 orders, only a single auto- somal locus per species has a recessive mutation that curects the production of phaeomelanin in the follicular melanocytes (Searle, 1968a). In the absence of more than one locus per species with similar phenotypic effects, it is considered probable that all of these "recessive yellow" loci, including those of the mouse and hamster, are homologous (Searle, 1968a,b). Considering next the 3h-locus, hamsters heterozygous for anophthalmic white (EI_;EhIEh) are similar to EI_;33/33 hamsters in most respects except that the ventral fur is white, there is a sprinkling of white guard hairs on the dorsum, and the eyes are lighter in color than the normal hamster (Robinson, 1962, 1964). Depending on the genetic background, the compound mutant hamsters, 3/3;Eh/33 have almost or completely white fur. Pigmentation, when it is present in the fur, is an extremely pale yellow, and is usually restricted to the dorsum and flanks. The ears have only patchy black pigmentation. The eyes are lighter in color than the eyes of 3/3;33I33 or EI_;33I3h hamsters and are slightly microphthalmic (Robinson, 1962, 1964; Asher, 1979). The 3I3;Eh/Eh and EI_;Eh/Eh hamsters are completely lacking in neural crest-derived pigmentation and are anophthalmic and deaf (Robinson, 1962, 1964; Knapp and 86 Polivanov, 1958; Yoon, 1975; Asher, 1979). The anophthal- mia is degenerative, in that early development of the optic cup is normal, but later development, beginning with the failure of proper induction of the retinal epithel- ium, is abnormal (Asher, 1979; _3. Yoon, 1975). Eh/Eh also causes a variety of other non-pigmentary phenotypic effects including: 1) reduced growth rates and adult body weight, 2) increased basal metabolic rate with a concomitant increase in food and water consumption, 3) reduced fertility or complete sterility in both sexes, 4) decreased volume of the fasciculata+reticularis layer of the adrenal cortex, and S) abnormal histogenesis of the adenohypophysis including a 40% decrease in cell number, increased numbers of tight and occluding junctions and the presence of abnormally ciliated cells in the sex zone of the adenohypophysis (Asher, 1979; James, 1979). Asher (1979) has suggested that the first four of these pheno- typic effects are subordinate pleiotropic effects of both the lack of eyes and a more fundamental defect in the pituitary gland of the Eh/Eh hamsters. REPRISE The original research to be discussed in the follow- ing sections is focused on three aspects of the gene action of the mouse mutation microphthalmic white, 33?“. Specifically, these three aspects are: 1) a determina- tion of the factors of the pigmentation phenotype altered by E3Wh which cause the E3Whl+ mice to have a diluted fur pigmentation phenotype, 2) a determination of the site of gene action of E3"h which leads to the unpigmented fur h/Eéyh mice, and 3) a comparative study phenotype of the E3? of the phenotype and site of gene action of E3wh in the mouse and anophthalmic white (Eh) in the Syrian hamster, to attempt to establish or refute the proposed homology of these two loci. To investigate the causes of the diluted fur colora- tion of the E3Whl+ mice, several aspects of the pigmentation phenotype, including: 1) melanosome number, size and shape, 2) melanocyte number and size, and 3) tyrosinase isozyme pattern, were examined. To investigate the site of gene action ofE3Wh on the pigmentation phenotype, a series of dermal-epidermal recombination grafts and 33 vitro epidermal cultures were performed. To examine the proposed homology of E3Wh in the mouse and Eh in the 87 88 hamster, several phenotypic comparisons, including: 1) the distribution of the oculocutaneous melanocytes, 2) the site of gene action on follicular melanocytes, 3) the epistatic interactions of E3?h and Eh with their respective recessive yellow loci, 4) the metabolic activity and growth rate, and 5) the cell density of the sex zone of the adenohy- pophysis, were made between mice and hamsters carrying comparable genotypes at the two loci. METHODS Mice The mouse strains used in this study are summarized in Table 8. Mice with specific genotypes other than those of the inbred strains, were created by the matings listed in Table 8 (see also Table 10). The 3/3 mice used in the dopa staining of the ear skin were obtained from Dr. P. Fraker, Department of Biochemistry, Michigan State University. The mice were housed in stainless steel or polycar- bonate cages containing wood chip bedding. The animals were fed either Wayne Lab Blox or Wayne Breeder Blox (Allied Mills). Water and food were available 33 lib. Hamsters The hamsters used in this study were obtained from a colony maintained by Dr. J. Asher, Department of Zoology, Michigan State University. Follicular Melanocytes Two methods for determining the number of melanocytes per follicle have been reported in the literature, X-ray 89 90 Table 8a. Mouse Inbred Strains Strain Genotype1 Source Generations of Designation Inbreeding when received C57B1I6J Jackson F115 Laboratory C57Bl/6J-E3Wh E3Whl+ Jackson N102F1 Laboratory C57Bl/6J-E EI+ Jackson N118 Laboratory C57Bl/6J-3 3/3 Jackson N16F2 Laboratory NAW-CS7Bl/6J2 3 seI3 s 3 c. Wolfe F6 U. of Kansas 1alleles which differ from the genotype of C57Bl/6J 2Strain originally created by R. Schaible in the early 1960's. More precise information on the pedigree of this strain is not available. 3Short ear (33) is closely linked to dilute and causes the 33/33 mice to have short ears. This short ear phenotype was used to confirm the genotypes of the 3 33/3 33;+/+ and 3 33/3 33;E3Wh/+ mice used in the follicular melano- cyte studies. 91 Table 8b. Matings to Create Mice of Desired Genotypes Desired Relationship to Parental Genotypes Inbred Strains Inbred Strains £1iwh/lewh F1 BQELWh 3/3;+/+ F2 B63, alsoyth 3622;.“ g/g._3wh/MiWh 3 33/3 33;+/+ F2 NAW, 9. 9.2/9. £3flWh/+ 86E!“ 36mWh - C57Bl/6J-E3Wh; B63 - CS7Bl/6J-3; NAW . NAW- 057B1/6J 92 inactivation (Chase, 1951; Potten and Howard, 1968; Potten and Chase, 1970), and a follicle squash technique (Potten, 1968; Galbraith and Arceci, 1974). The X-ray inactivation method was not used for two reasons: 1) the method provides only indirect estimates of the numbers of follicular melanocytes, and 2) the variance estimates for this method are quite large. The second method, the follicular squash technique, was attempted, but in preliminary studies, this author was not able to achieve satisfactory preparations. Since neither of the published methods for counting follicular melanocytes was usable, a third method, using serial sections of paraffin embedded skin was employed. In preliminary studies, using2[2;+/+ and 2/2;E;Wh/+ mice, the dendritic morphology and intense melanization of the follicular melanocytes prohibited the identification of cell boundaries or nuclei, thus making an accurate count of the melanocytes in the sectioned material unfeasible. However, with the addition of homozygous dilute (d/d) to the genetic background, it became possible to count the follicular melanocytes in the sectioned material. To synchronize the follicular growth cycles, the dorsal fur of five d/d;+/+ and five d/drgiWh/+ adult mice was removed with depilatory waxl. Seven days after depi- lation, the mice were killed by cervical dislocation. A 1See Appendix C for formulation and composition. 93 strip of dorsal skin was removed, pinned out in a paraffin- filled dish and fixed for 48hr in either Bouin's fluid1 or Vandegrift's fixativel. After dehydration, the tissues were embedded in either Tissue Prep (Fisher Scientific Co.) or Paraplast (Sherwood Medical Industries). The skin sam- ples were sectioned at 411m, mounted on slides with Mayer's albumin (Harleco) and stained with Harris's hematoxylin1 and alcoholic eosinl. The sectioned material was examined at low magnification and five zigzag follicles per animal were randomly chosen for melanocyte counts. The serial sections of these follicles and their melanocytes were traced with the use of a camera lucida. From these tracings, three types of data were extracted: 1) the number of melanocytes per follicle, 2) the maximum dia- meter of the follicle, and 3) the number of serial sections spanned by each melanocyte. These data were analyzed with either a one way analysis of variance, Student's t-test or a x2 contingency table. Tail Skin Epidermal Melanocytes The tail skin of five +/+ and five gth/+ adult mice was prepared for counts of the epidermal melanocytes according to the following protocol. After killing the mice by cervical dislocation, the hair was removed from the base of the tail with depilatory 1See Appendix C for formulation and composition. 94 wax. The use of a commercial depilatory agent containing thioglycolates, Nair (Carter-Wallace Inc.), was not accept- able since the cellular integrity of the epidermal melano- cytes of the tail skin was disrupted by this treatment. After depilation, a ring of skin (8-10mm in width) was removed from the base of the tail, affixed to a coverslip with stopcock grease and incubated in 2M NaBr at 37°C for 28hr to separate the dermis and the epidermis (Starrico and Pinkus, 1957). Following the NaBr incubation, the dermis and epidermis were manually separated, the epider— mis was re-affixed to the coverslip (dermal side upward), incubated in 0.075% DL-dopa, 0.01% catalase in 33mM phos- phate buffer pH 7.4 (25% Sorensen's buffer) at 37°C for 2-28hr, and fixed in neutral buffered formalin for 24-48hr. After fixation, the sebacous glands and follicles were trimmed from the lower surface of the epidermis with sharpened jeweler's forceps. The fixed and trimmed epider- mis was dehydrated, cleared in xylene and mounted dermal side upward in Permount (Fisher Scientific Co.). Within the tail epidermis, the melanocytes are located within rectangular scales which are arranged in a more or less regular pattern around the circumference of the tail. Within each skin, three continuous rows of scales were measured for the area of each scale and the number of mel- anocytes within each scale. These data were compared with a x2 contingency table. 95 Dermal and Epidermal Melanocytes of the Ear Staining properties of melanocytes and mast cells In most tissues, melanocytes can be histochemically characterized by their unique ability to produce melanin from an exogenously provided substrate, dopa. In the dermis, however, two cell types, melanocytes and mast cells are capable of converting dopa into melanin (Okun gt al., 1970). In the melanocytes, the initial step of melanin production, i.g., the oxidation of dopa into dopa quinone, is catalyzed by the copper-dependent enzyme, tyrosinase. In mast cells, dopa oxidation is catalyzed by the heme- dependent enzyme, peroxidase. The dopa oxidase activity of these two enzymes can be distinguished by the differential effects of diethyldithiocarbamate (DDC) and catalase on their enzymatic activities. DDC is a potent chelator of divalent cations. At a concentration of 10mM, DDC will chelate the Cu++ from tyrosinase, blocking the enzyme's dopa oxidase activity. However, 10mM DDC will not chelate Fe++ from the porphyrin ring of the peroxidase and consequently does not affect the dopa oxidase activity of the peroxidase. Conversely, catalase, which converts hydrogen peroxide to water and molecular oxygen, will block the dopa oxidase activity of peroxidase but has no effect on the dopa oxidase activity of tyrosinase. Peroxidase and tyrosinase can also be distinguished 96 by the ability of peroxidase to catalyze the oxidative polymerization of 3,3'diaminobenzidine (DAB). This DAB oxidase activity of peroxidase, like the dopa oxidase activity, can be inhibited by catalase (Okun £5 al., 1970) Effects of fliWh 23 melanocyte distribution The original observations of the effects of giWh on the distribution of cutaneous melanocytes were made on either skin sections stained with hematoxylin and eosin (Silvers, 1956) or intact skin samples incubated in phos- phate buffered dopa (Markert and Silvers, 1956). To con- firm and extend these observations, the ear dermis and epi- Wh/§$Wh: and 3/51) dermis from adult mice (+/+, giyh/+, El were examined according to the following protocol. After killing the animal and removing the ears, the dorsal and ventral ear skin were manually separated, affixed to coverslips (epidermis downward) and incubated in 2M NaBr at 37°C for 28hr. After NaBr incubation and manual separation of the dermis from the epidermis, both compo- nents of the +/+, EEWh Igth, and 3/3 ear skin were incubated with various substrates and inhibitors according to the treatments listed in Table 9. The ELWhl+ ear epidermis was incubated only in dopa with catalase according to 1The c/c skin was used as a control for peroxidase activity. The melanocytes in albino skin do not contain functional tyrosinase, but the albino dermis contains mast cells with normal peroxidase activity. 97 Table 9. Ear Skin Staining Treatments 1. Dopa: 0.0752 DL-dopa in $81, 2hr, 37°C. 0 2. Dope-catalase: 0.01% catalase in SB, 2hr, 37 C; 0.075% DL-dopa + 0.01% catalase in SB, 2hr, 37°C. 3. Dopa-DDC: 10mM DDC in SB, 90min, 37°C; 4X'(SB,.5min, 37°C)2; 0.0752 DL-dopa in SB, 2hr, 37°C. 4. DAB+H - 0.3mg/ml DAB in SB, 10min, 25° 2 2' 0.06% H /ml DAB solution, 20min, 25°C. C; add 40ul 2°2 5. DAB-catalase: 0.01z catalase in SB, 2hr, 37°C; 0.3mg/ml DAB + 0.01% catalase in SB, 30min, 25°C. 6. DAB+H202-DDC: 10mM DDC in SB, 90min, 37° H202 in SB, 5min, 25°C); treatment 4. C; 4X (0.06% 1SB - 252 Sorensen's phosphate buffer (33mM phosphate, pH 7.4) 2DDC must be thoroughly rinsed from the tissue since DDC can act as an antioxidant and will suppress peroxidase aetivity (Randall, 1946; Okun gt al,, 1970, 1972; Brumbaugh 33 gl,, 1973). 3Modification of peroxidase staining procedure of Graham and Karnofsky (1966). DDC 8 diethyldithiocarbamate DAB ' 3,3' diaminobenzidine 98 treatment 2 of Table 9. After staining, the skin samples were fixed in neutral buffered formalin for 24-48hr. After fixation, the tissues were dehydrated, cleared in xylene, and whole mounted in Permount. The mounted tissues were examined for the presence or absence of dopa- and DAB-stained cells, and the morphology of these stained cells. In the +/+ and ELWhI+ epidermis, the spatial dis- tribution of the melanocytes was sufficiently irregular so as to make quantitative estimates of melanocyte number impractical. However, the rank order of the genotypes, based on the relative numbers of melanocytes, could be determined. Eyes The eyes of both adult mice (g/g;+/+, g/g;giyh/+ and h 2 23M“ Uzi“) and hamsters (s/ssfl/zflz. yams/2:1. and Wh/Eh) were fixed intact in Bouin's fluid for 48-72hr, / 51/2. dehydrated and embedded in paraffin. The embedded eyes were sagittally or cross sectioned at 10pm, stained with Harris's hematoxylin and alcoholic eosin and examined for the presence and distribution of melanocytes in the retinal epithelium and uveal tract. Hamster Skin Twelve to fourteen days after wax depilation of the fur (when the skin appeared to be in anagen VI), the dorsal skin and the costovertebral gland were excised from the 99 killed hamsters (g/g;gh/!h,‘g[g;Eh/wh, and g/g;flh/flh), pinned out on paraffin-filled dishes and fixed 24-48hr with Bouin's fluid. The tissues were dehydrated, embedded in paraffin, sectioned at 10pm, stained with Harris's hema- ,toxylin and alcoholic eosin, and examined for the presence and distribution of melanocytes. Melanosomes Melanosomes were obtained from adult mouse hair (2/2;+/+,_D_/2;y_i_Wh/+, g/g;+/+ and g/g;_M_i_°h/+) by incubation of the hair in 4M KOH (10mg hair/1.0m1 KOH) at 70°C for 8hr. Following solubilization of the keratin, the melano- somes in KOH were centrifuged at 3,000 xg, the supernatant was decanted and the melanosomes were resuspended in 1.0ml of distilled water. This procedure was repeated for a total of three rinses or until the supernatant reached neutrality. Following the last resuspension, the volume of the melanosomal suspension was measured. To count the number of melanosomes, an aliquot of the melanosomal suspension was diluted (1/200 for the +/+ geno- types and 1/100 for the g$Wh/+ genotypes) and counted in a hemacytometer. For each melanosomal sample two replicate counts were made. These data were compared with proper orthogonal contrasts. For a determination of the size, shape and degree of melanization of the melanosomes, samples of the melanosomal suspensions from QID;+/+ and 2/2;giWh/+ hair were placed 100 on Formvar-coated grids1 and evaporated to dryness. The unstained melanosomes were examined and photographed on a Philips 201 electron microscope at 7,000X. (The melano- somes were not stained because preliminary studies estab- lished that negative staining of the grids obliterated the edges of the individual melanosomes, making accurate measurements of the melanosomal dimensions impossible.) From prints of these photographs the dimensions of the ma- ‘jor and minor axes of the melanosomes were measured. Fol— lowing exponential transformation (X0'4) to achieve a bi- variate normal distribution, these data were compared with a two sample Hotellings T-squared test (Morrison, 1967). To analyze the melanosomal shape, a univariate metric of shape, eccentricityz, was computed for each melanosome and the resulting data were compared with a z-statistic. Phaeomelanin For the determination of the phaeomelanin production of the follicular melanocytes, a procedure modified from Heidenthal (1940) was used. Dorsal fur was clipped from five g/g;+/+, five g/g;§iWh/+ and five g/g;§th/§iWh adult mice, rinsedtwice in cold ether to remove surface lipids and allowed to air dry. From each mouse, 50.0mg of hair 1See Appendix C for formulation 2A derivation of this metric and the reasons for its use are discussed in Appendix B. 101 were added to 5.0ml of 0.5M NaOH. After initial vigorous agitation to remove trapped air from between the hairs, each hair sample was incubated at room temperature for 48hr with moderate agitation every 12hr. After incubation, the 0.5M NaOH containing the dissolved phaeomelanin was decanted from each hair sample and the optical density of each sample was measured at 580nm. Tyrosinase Isozymes The tyrosinase isozyme patterns were analyzed by poly- acrylamine electrophoresis with modifications of the tech- niques of Burnett E£.££° (1969), Holstein gt fii- (1973), and Hearing and Ekel (1975). Ten days after wax depilation h/ElWh of the hair, three +/+, three gth/+ and three giw adult male mice were killed by cervical dislocation and the dorsal skin of each mouse was used as the source of the follicular tyrosinase. The skin was pinned out, the pannic- ulus carnosus was removed and the panniculus adiposus was scraped to remove all of the anagen VI hair follicles. The follicles scraped from each mouse skin were rinsed in phosphate buffered salinel, pelleted by centrifugation at 4°C and frozen at -70°C until electrophoresis. The electrophoresis was done according to the method of Clarke (1964) with the following modifications: 1) the gels were 7&2 acrylamidel, 2) the gel buffer was 400mM 1See Appendix C for formulation and composition 102 . glycine, 50mM TRIS, pH 8.3, and the upper and lower buffers were 40mM glycine, 50mM TRIS, pH 8.3, 3) each sample con- taining a drop of glycerol and Sul of 0.5% bromophenol blue (tracking dye was loaded onto the top of the running gel under 1-2cm of upper buffer, and 4) the electrophoresis was run at room temperature with 2-3mA/tube for 40min. After electrophoresis, the gels were neutralized in 0.1M sodium phosphate buffer, pH 6.8, for 15min, stained in 0.22 DL-dopa, 0.01% catalase in 0.1M sodium phosphate buffer, pH 6.8, for 3-4hr, and fixed in 782 acetic acid. The stained gels were scanned at 435nm for comparisons of the tyrosinase patterns. Dermal-epidermal Recombination Grafts The dermal-epidermal recombination grafts were pre- pared according to the following protocol modified from Mayer (1977b). Embryos of the desired genotypes were ob- tained from the matings described in Table 10. To obtain these embryos from two or more matings at the correct stage of embryonic development, females known to be in proestrous (by examination of vaginal smears (Allen, 1922) stained with toluidine blue 01) were given overnight access to a male of the appropriate genotype. The presence of a vaginal plug the next morning was taken as evidence of successful copulation. The morning after copulation was 1See Appendix C-for formulation and composition 103 Table 10. Matings for Embryos for Dermal-Epidermal Grafts Embryo Mating Genotype 1. +/+ +/+ x +/+1 2 giWh/+ +/+ x MiWh/MiWh wh wh wh2 wh wh wh 3. Mi /+, g; /gi Mi /+ x g; /g; 4 MiWh/MiWh M$Wh/MiWh x gym/MiWh 5. 3/23 g/+ x g/+ 1matings 1-4 were done on both §/§ and 2/2 backgrounds 2 h 11 and 13 day Miw IMiWh embryos can be distinguished from MiWhl+ embryos by the lack of visible pigmentation in the optic cup. 313 day K/fl embryos can be distinguished from E/+ and +/+ embryos by the pale color of the fl/fl liver caused by fetal anemia. 104 considered to be day 0 of gestation. If no vaginal plug was evident, the male was removed and the estrous cycle of the female was followed by vaginal smears until the next proestrous when a mating was re-attempted. At the appro- priate day of embryonic development, the female was killed by cervical dislocation and the embryos were dissected out of the uterus into sterile, room temperature Hanks' Balanced Salt Solution1(HBSS). From the 13-14 day embryos, a piece of skinlmm2 from each lateral flank between the limb buds was dissected out with sharpened iridectomy scissors, peeled off the flank with fine forceps and transferred by Pasteur pipette to cold sterile HBSS. When using 11 day embryos, the flank skin and the underlying mesoderm between the limb buds were removed in a similar fashion. After all embryos had been dissected, the skins were incubated in 1.02 trypsin1 at 4°C to separate the epidermis from the dermis, or in the case of the 11 day embryos, the epidermis from the dermis+ mesoderm. For 13-14 day skin, trypsin incubation was 2-28hr, for 11 day skin 1-18hr was sufficient. Following the trypsin incubation the skins were transferred to Minimal Essential Medium-Eagle (MEM)2 with 10% fetal calf serum2 or 10% horse serumz, or MEM with 25% fresh egg albumin. 1See Appendix C for formulation and composition 2Purchased from Grand Island Biological Company 105 While in the MEM rinse, the skins were manually separated into epidermis and dermis, or epidermis and dermis+mesoderm. After separation, the tissues were left in the MEM rinse and stored on ice. To form the recombination grafts, a dermis of the appropriate genotype was transferred by pipette into a small drop of HBSS on the surface of an agar1 filled culture dish. While in this drop of HBSS, the dermis was oriented so that the epidermal surface of the dermis was facing upward. If the dermis was incorrectly oriented, the graft would fail to produce hair or pigmentation. Next, an epidermis of the appropriate genotype was also trans- ferred to the drop of HBSS and arranged over the dermis. The orientation of the epidermis was not critical to the normal morphogenesis of the hair and formation of pigment. Excess HBSS was removed from around the recom- bined skin and the process was repeated until the plate contained four skin explants. The agar plate with explants was incubated at 37°C at 52 CD for 48hr to allow reattach- 2 ment of the epidermis to the dermis. After incubation, the skin explants were transferred to sterile, room temp- erature HBSS and examined for satisfactory dermal-epidermal adhesion. Any explants with poor adhesion were discarded. The explants were then implanted and cultured in the testes of adult C57Bl/6J mice (two grafts/testis) for 2-4 1See Appendix C for formulation and composition 106 weeks to allow hair and pigment morphogenesis. In a few cases (+/+ // +/+) grafts were cultured in the testes of ELWh/+ mice. For graft implantation, the host males were anesthetized with intraperitoneal injections of Avertin1 (0.2-0.3ml/10g body weight), and the lower ventrum was shaved with electric clippers. A 1cm midventral longitu- dinal incision, beginning just above the preputial glands, was made through the skin and muscle wall. Using blunt forceps, one testis was exteriorized. A small incision was made in the testis capsule, avoiding the major blood vessels. A channel into the testis was made with sharp- ened forceps. The skin explant was picked up with the forceps and placed within the testis. This process was repeated with a second graft of the same genotypic com— bination, and the testis was returned to its normal po- sition. The procedure was repeated with the other testis. The abdominal wall was sutured, the incision was swabbed with Chloromycetin (Parke Davis) and the skin was secured with Michel wound clips. After surgery, the host male was placed under a 100W lamp until it had regained consciousness and normal mobility. Two to four weeks after surgery, the host males were killed, the testes removed and the grafts dissected free of the adhering semeniferous tubules. The grafts were examined for hair growth and pigmentation phenotype, dehydrated and cleared in a-terpinol (Eastman 1See Appendix C for formulation and composition 107 Kodak). For the hamster dermal-epidermal recombination grafts, both components of the skin were from 118 day embryos. At this stage of development, just before the appearance of the epidermal anlage of the follicles at 12 days (Boyer, 1948), the hamster skin appears to be at a stage of devel- opment comparable to the 13 day mouse skin. The hamster skin explants were prepared with the same procedure used for the 13-14 day mouse embryos. The implantation proce- dures were also the same as those used for the mice with the following modifications: 1) the Avertin dosage was 0.2-0.3m1 of 2X-Avertin1/10g body weight, 2) up to four grafts were placed in each testis, 3) the incision in the testis capsule was sutured closed, 4) both the abdominal muscle wall and the overlying fascia were separately sutured, and 5) the grafts were cultured in the testes for 3-4 weeks. In Vitro Epidermal Melanocyte Culture h wh The epidermis of +[tigiWh/+ and Miw /Mi 14 day embryos was isolated and cultured according to Mayer and Oddis (1977). The epidermis was isolated from the embryos according to the procedure described for the dermal-epider- mal grafts. The epidermis from several embryos of the same genotype was mechanically dissociated by gently flushing 1See Appendix C for formulation and composition 108 through a Pasteur pipette. The dissociated epidermal frag- ments were seeded onto plastic Petri dishes (Corning 10x35mm) containing MEM with 10% fetal calf serum. The cultures were grown at 37°C, 20% 002 for 12-21 days, fixed in neutral buffered formalin, stained with toluidine blue 0, air dried and examined for the presence of melanocytes. Hematology Three parameters of mouse hematology which might be expected to vary in an anemia, were examined in male mice h wh of the genotypes +/+, Miyhl+ and Mi? lg; . The parameters examined were: 1) the leukocyte differential count, i,g., the kinds and numbers of leukocytes present, 2) the retic- ulocyte count, and 3) the erythrocyte diameter. Since the leukocyte differential count shows a marked diurnal rhythm, all blood samples were taken between 11 am and 2 pm (Halberg gt _l., 1953). The blood samples were collected from the lateral tail vein of each mouse into an unheparinized capillary tube. For the leukocyte differential counts, a drop of blood was placed on a glass slide, smeared, air-dried and stained with Wright's stain1 (Harleco) according to Wil- liams gt El- (1972). For the reticulocyte counts, the blood was stained with new methylene blue1 (Brecher, 1949). The erythrocyte diameter measurements were made on the 1See Appendix C for formulation and composition 109 reticulocyte smears. The leukocyte differential counts were done in the third quarter of the smear. One hundred leukocytes were counted and classified as either eosinophil, segmented neutrophil, lymphocyte or monocyte. Two replicate counts were made for each mouse. The reticulocyte counts were made in the outer one third of the smear. The number of reticulocytes per 1000 total erythrocytes was determined in two replicate counts for each mouse. The erythrocyte diameters were measured in regions of the smear where the erythrocytes showed normal central pallor. The diameter of 100 erythrocytes per mouse were measured to the nearest 0.5um. The reticulocyte counts and erythrocyte diameters were analyzed with one way analyses of variance. Growth Curves h Post natal growth was measured in 10 +/+, 10 Miw /+ h/§;°h male mice. The body weights of these mice and 10 Eiw were measured weekly between the fourth and seventeenth weeks of life. The growth curves were compared with a series of Student's t-tests. Hotelling's two sample T2 was not used since the sample size (n - 10) from which each curve was derived was too small with respect to the length of the vectors (k - 14) which were compared. 110 Metabolic Activity The metabolic activity was examined in the same groups of mice which were used to collect the growth curve data. Two methods were used to collect information on metabolic activity: 1) the mice were individually housed for 24hr in cages in which the food and water consumption and urine and fecal output could be measured, and 2) the rectal temp- erature of the mice was determined with the use of a ther- mistor probe calibrated to 0.1°C. A11 rectal temperatures were non-replicate determinations since preliminary studies had revealed that the mice were adversely affected and sometimes killed by this procedure. Pituitary Cytology h wh The pituitary glands of g/g;+/+ and g/E;Miw IMi adult male mice (three per genotype) were fixed 24hr in Bouin's solution, dehydrated, embedded in paraffin, cut in serial cross sections at 2pm and stained with Harris's hematoxylin and alcoholic eosin. Within the ventral para- medial region of the sex zone of each pituitary (Baker and Gross, 1978), the cell density was determined in 6400mm2 fields of four non-consecutive sections. The cell densities h/E$Wh pituitary glands were of the g/g;+/+ and g/E;Miw compared with Student's t-test. While the number of pit- uitary glands in this analysis is small (n = 3), since no differences had been found between the somatic physiology 111 h wh of the +/+ and E1? lg; mice, and the difference in cell density found in the sex zone of the hamster pituitary glands was quite large (a 40% reduction in cell number in the g/g;Wh/Wh hamsters), this sample size was considered adequate. RESULTS The data in the Results section are divided into five main sections: 1) the pigmentary phenotype of hivh, 2) the non-pigmentary phenotype of hivh, 3) the dermal-epidermal grafts and $2 vitro epidermal cultures involving hhyh, 4) the pigmentary phenotype of Eh, and 5) the dermal-epidermal grafts involving Eh. Pigmentary Phenotype of h;Wh The data in this section were collected to: 1) deter- mine the proximate causes of the pigmentary phenotypes of h wh [hi and 2) allow a comparison of the pig- mentary phenotypic effects of hhyh and Eh to investigate iiw'Vi- and 14.1." the possible homology of these mutations. Follicular melanocytes The follicular diameter measurements of h/h;+/+ and h/h;h$Wh/+ zigzag follicles are summarized in Table 111. These measurements were compared with a one way analysis of variance (see Table 12). The major proportion of the 1Throughout the tables in the Results section, the mean values of the data sets are listed with their standard errors as i r S.E. 112 113 Table 11. Diameter of Zigzag Follicles Genotype Follicular Diameter (um) g/g;+/+ 51.12 i' 1.861 (25)1 wh h/h;Mi /+ 54.02 i 1.077 (25) 1number of follicles measured for each genotype; 5 mice/ genotype, 5 follicles/mouse 114 Table 12. Anova of Follicular Diameters Sources of Variation d.f. SS MS f Genotypes 1 104.84 104.84 0.61 n.s.1 Mice/genotype 8 1379.61 172.42 4.95** Follicles/mouse 40 1395.07 34.88 1: =532 .05,1,8 ° ** f = 2.99 .01,8,40 115 variation in follicle diameter is caused by the sampling error between mice, and the average zigzag follicle diam- eters do not significantly differ between the h/h;+/+ and i/éleWh/i' mice. Since the genotype has no significant effect on the zigzag follicle size, the number of melanocytes per follicle in the two genotypes was directly compared without any normalization based on follicle size. The results are summarized in Table 13 and Figure 4. Since hiyh/hiyh mice have no follicular melanocytes, it was considered probable that hivhl+ could affect the follicular melanocyte number in one of two ways. Either hivh could behave as a simple recessive allele, i.e., the highl+ follicles would contain the same number of melanocytes as the +/+ follicles, or .hiWh could behave as a partially dominant allele, i.e., hivhl+ follicles would contain fewer melanocytes than the h wh +/+ follicles, but more than the hi? /hi follicles. Given these h priori hypotheses (HO:X+/+ - xm1Wh/+ or HI:X+/+ > XMiwh/+), the mean follicular values of the h7h;+/+ and i/i;hiWh/+ mice were compared with a one-tailed t-test (see Table 13). Based on the results of this t-test, it can be concluded that hivhl+ has a significant effect on the numbers of melanocytes found in the zigzag follicles such that h/h;hiWh/+ zigzag follicles contain, on the average, 85% the number of melanocytes which are found in h/h;+/+ zigzag follicles. 116 Table 13. Number of Melanocytes per Follicle Genotype Melanocytes/Follicle 2745+” 8.84 t 0.38 (25)1 g/g; iWh/+ 7.52 i 0.30 (25) Comparison of means: t = 2.737** 1number of follicles/genotype *7: t.01,48 = 2.407 117 39:52 258222 8.32.6... .v 8:3“. 83:52 238282 N. _ o. m m AS; x\\\\\\\3 \\\\\\\\\\\N t \\\\\\\\\ onohmmsum {2.336% _ _ mmonvmoum 5.23.... S \ W“ w” 1 V Kouanb 81:] N 0 Q Q 118 The sizes of the individual follicular melanocytes were estimated by counting the number of serial sections in which each melanocyte appeared. These data are sum- marized in Table 14 and Figure 5. Since the melanocyte size data were not normally distributed, the size distri- butions from the two genotypes were compared in a 2 x 7 contingency table. Based on the apparent difference between the mean values (see Table 14) and the results of x.2 test of the contingency table (see Table 14), it appears that hiyhl+ affects the linear dimensions of the h/+ zigzag follicular melanocytes such that i/h;hiw follicular melanocytes are, in a linear dimension, 81% the size of the h/h;+/+ follicular melanocytes. Tail melanocytes In the tail epidermis, the melanocytes are arranged in a more or less regular pattern of rectangular scales which run in circumferential rows around the surface of the tail. The area of individual scales and the number of melanocytes per scale were determined in tail skins of adult +/+ and hiyhl+ mice. These data are summarized in Table 15. All of the scales of the +/+ epidermis contained three or more melanocytes. In the hiWhl+ epidermis, how- ever, 33 scales contained no melanocytes at all (= O-class scales). If these O-class scales represent small white spots similar to the white spots seen in the general body 119 Table 14. Diameter of Follicular Melanocytes Genotype Melanocyte Diameter (um) g/g;+/+ 19.76 i 0.49 (221)1 g/g; iWh/+ 15.96 i 0.47 (188) Comparison of melanocyte diameter distributions: x2 - 44.739** 1number of melanocytes/genotype ** 2 x .01,6 3 16.812 120 4:21... 46 8 s ‘ 3 o 8 +1 +1 CD 8 8 "° ' e: e b". o". I 3; . a: ‘_° , i s \\\\\ '° ‘Oi 1: ‘° U \\\\\\\ N K\\\\\\\\\\\\3i a: I L\\\\\\\\\\\\\\\\\§] 2 L\\\\\\\\\\\\\\\\\\\\\\\\\\\\\ 3 t \\\\\\\\\\\\\\\\j 9' lo i , . , L L m 30 no o 2 o no N N - (o/o) Kouenbaid Melanocyie Diomeier (pm) Figure 5. Follicular Melanocyte Diameter 121 Table 15. Melanocyte Density in Tail Skin Epidermis Genotype Area1 (sq. mm) Melanocytes Melanocyte2 per Scale Density +/+ 0.225 0.002 65.41 t 1.52 287.24 (257)3 hiWh/+ 0.250 0.003 30.96 i 1.94 125.1 (261) hiWhl+5 0.250 0.003 35.44 i 2.05 143.6 (excl. (228) O-class) 1area of individual epidermal scales 2melanocytes/mm2 3number of epidermal scales measured 4standard errors were not calculated for these means since the hiyhl+ data were non-normally distributed 533 of the epidermal scales in the hth/+ skin contained no melanocytes (O-class scales) and are excluded from these mean values 122 fur, then it could be argued that the data from these white spots should be excluded from the comparisons of the melanocyte density between the two genotypes. Therefore, Tables 15 and 16 and Figure 6 show the data for hivh /+ both with and without the 0-class scales. The distribution of the intra-scale melanocyte densities for the +/+ and hivhl+ tail epidermis are shown in Figure 6. Since the hivhl+ intra-scale melanocyte density data are clearly non-normal in their distribution, the +/+ and hivhl+ melanocyte density distributions were compared in a 2 x 11 contingency table (see Table 16). Based on the apparent differences between mean intra- scale melanocyte densities (Table 15) and the highly significant x2 values from the contingency table (Table 16), it can be concluded that hivh affects the tail epidermal melanocyte density such that hiyhl+ tail epidermal scales have, on the average, 43% (50% excluding the 0-class scales) the number of melanocytes found in the epidermal scales of +/+ tail skin. Ear melanocytes The results of the dopa- and DAB-staining of the +/+, h”iiWh and £/£ ear skins are summarized in 14.1.“°/+. iii“ Table 17a,b. From these results, three conclusions can be made. First, consistent with the observations of Markert wh wh and Silvers (1956), hi /hi ear skin, both dermis and epidermis, does not possess cells with tyrosinase 123 Table 16. Contingency Table Results of Intra-Scale Melanocyte Densities Comparison X2 +/+ vs. gym/.9 199.1** +/+ vs. hiWh/+ 166.77** (excl. 0-class scales) ** 2 X .005,10 ‘ 25°2 124 Estosqmnzo... .o bison 232.222 295-32. .m 8:3... one 08 8.. cos. 8m N 00m 00 553232202 com: 8. ..nSfiEr _ . I“ \\\\\\\\§ may I . 0 was... .6 2: .x 60.3» 329.0 +£3.22 Ghana. {eta—D sauna .5 § \\\\\\\\\ ._ t\\\\\\\\\\ \\ \\\\\\\4 CO. on m. C CD O N 8 saunas 10 ieqmnN C m OD. 125 Table 17. Ear Melanocyte Distribution a. Staining Characteristics Genotype +/+ Lim“ Paw/11.1.“ 2/9. Staining E1 D E D E D E D Procedure 1. Dopa2 ' +3 + + + - + - + 2. Dopa- + + + + - - - - catalase 3. Dopa-DDC - + - + - + - + 4. DAB+H202 - + - + - + - + 5. DAB- - - - - - - - - catalase 6. DAB+H202 - + - + - + - + catalase 1E = epidermis; D = dermis 2staining procedures listed are the same as those described in detail in Table 9 of the Methods section 3+ = presence of stained cells; - = absence of stained cells 126 Table 17 (cont'd.). b. Enzyme Localization Genotype wh wh wh Enzyme +{+ Ei /+ Hi /§$ S/E Activity E D E D E D E D Tyrosinase ++2 + + + - - _ _ Peroxidase - + - + - + - + 1E = epidermis; D . dermis 2++ and + - presence of cells with specified enzyme activity, the +/+ mice had more melanocytes in the ear epidermis than did the hivhl+ mice; - = absence of cells with specified enzyme activity 127 activity, i,h., MiWh/hiwh ear skin does not appear to have any functional melanocytes. Second, the ear dermis of all four genotypes contains approximately the same number of peroxidase positive cells. Although not con- firmed by histochemical staining, 5.3., the astrablau technique of Bloom and Kelley (1960), based on their mor- phology and location, it is probable that these peroxidase positive cells are mast cells (Okun 3h hi., 1970). Third, although +/+ and hiWh/+ ear epidermis have qualitatively the same staining pattern (Table 17a), from comparisons of dopa stained +/+ and hiWh/+ epidermis (Table 17b), it was clear that the hiWh/+ epidermis contained fewer melano- cytes than the +/+ epidermis. Ocular melanogytes The distribution of melanocytes in the eyes of 2/23+/+a h/hiWh mice is summarized in Table s/ssflwhli' and slszliiw 18. Based on a comparison of the observations of Markert and Silvers (1956) and Deol (1973), it is apparent that h/+ the distribution of melanocytes in the eyes of +/+, hiw wh wh and hi /hi mice is essentially similar for both 3/3 and h/h genetic backgrounds. The only point of difference between the observations of Markert and Silvers, Deol and the data in Table 18 concerns the presence or absence of melanocytes in the choroid of the hiWh/+ eyes. Markert and Silvers did not find any pigmentation in the choroid of the h .hiWh/+ eyes. Deol noted that the choroids of the hiw /+ 128 Table 18. Distribution of Ocular Melanocytes Retinal Layers Uveal Tract Retinal and ciliary ciliary epi- iris choroid body and Genotype thelium iris £/£,+/+ + + + + g/g.§_1;Wh/+ + + -(+)1 + h/h,MiWh/MiWh _ +2 _ _ Choroidal pigmentation was absent from most eyes, however in one of twelve eyes examined, the choroid was completely pigmented. 2Pigmentation is present in the iris, but is greatly re- duced in quantity and is present only in a narrow ring adjacent to the pupillary margin of the iris. 129 eyes were "...generally very unevenly pigmented...." As noted in Table 18, of the 12 hiWh/+ eyes examined, the choroidal layers of 11 of the eyes were completely unpig- mented, while the choroid of one eye was completely pig- mented. Since the expressivity of white spotting on the ventral body fur of hivhl+ mice is variable, even on a uniform genetic background, it is probable that the ob- served differences in hiWhl+ choroidal pigmentation are another reflection of the variable expressivity of the spatial distribution of melanocytes (or white spotting) in the hiwh/+ mice. Melanosomes Number: The data on the number of melanosomes in the hair of wh wh h7h;+/+, h/h;hi /+, h/h;+/+ and h/h;hi /+ mice are summarized in Table 19. Based on the overall intensity of fur pigmentation, the genotypes listed in Table 19 can be ranked in the following order: h/h;+/+ > h/h;+/+ > h h 19./M1" /+ > 91/3111" /+. As can be seen from the data in Table 19, the relative quantities of melanosomes per milligram of hair are entirely parallel with the rank order of the genotypes based on overall visual intensity of fur pigmentation. To analyze the data in Table 19, three proper ortho- gonal contrasts were made: 1) h/h;+/+ and h/h;hiWh/+ vs. 130 Table 19. Melanosome Number Melanosomes Relative Genotype (x10-8)/mg hair Quantity 2/23+/+ 3.073 t 0.151 (3)1 1.00 'g/g;+/+ 1.176 t 0.037 (6) 0.38 wh h/h;_i /+ 0.874 t 0.056 (3) 0.28 wh h/h;Mi /+ 0.643 t 0.015 (6) 0.21 1number of mice/genotype, 2 replicate counts/mouse 131 h h/h;+/+ and h/h;hiy /+, 2) h/h;+/+ vs. h/h;hiwh/+ and 3) g/g;+/+ v.s. yyyf" /+. The results of these contrasts are summarized in Table 20. Based on the h priori obser- vations of the rank order of the overall intensity of fur pigmentation, the t values for the orthogonal contrasts were compared against critical values for a one-tailed t-test. From the results of these contrasts, two conclusions can be made. First, from contrast (1), it can be conclu- ded that h/i causes a significant decrease in the number of melanosomes in the hair. This conclusion is consistent with the data, but not the conclusions of E. Russell (1948) (see Appendix A). Second, based on the results of contrasts (2) and (3), it can be concluded that on either a 2/2 or h/i background, hiyhl+ significantly reduces the number h/+ of melanosomes in the hair. On a h/h background, hiw reduces the number of melanosomes in the hair to 28% of the normal number. On a h/h background, hiWh/+ only reduces the number of melanosomes in the hair to 55% of the normal (h/h;+/+) number. Size: The dimensions of the major and minor axes of h/h;+/+ and h/h;hiWh/+ melanosomes are listed in Table 21. To acheive a bivariate normal distribution of the axial dimensions, the data were transformed such that X = transform X0.4 After transformation, the data were analyzed with a 132 Table 20. Effects of h/h and hiWh on Melanosome Number Contrast t 1. h/h;+/+ and h/h;hiWh/+ vs. 16.013** g/g;+/+ and h/h;hth/+ 2. 2/2;+/+ vs. g/g;hiwh/+ 20.342** 3. g/g;+/+ vs. g/g;y_i“’"/+ 3.632** ** for a one-tailed t-test, t.01,36 = 2.432 133 Nme.N.~o. m m u make consumes mmaomosmHmE mo Hon—Es:H «ho.ocm u NH “mamas mo comaumnaoo A83 eoo.o H who.o moo.o H -m.o moo.o H mmm.o ooo.o H moo.o +\:3flm ~A~w~v coo.o H mmm.o coo.o H omw.o oco.o H ewq.o moo.o H ~w~.o +\+ Aanv A83v Asnv Aanv mnzuoooo me< yoga: me< Home: me< nost me< Homo: mum: voEHOMmsmuH mcofimsmsfio HmSomosmHoz .~N manna 134 two-sample T2 test (see Table 21). The covariance matricies for the melanosomal axial dimensions of the two genotypes were non-homogeneous. Therefore, the T2 statistic listed in Table 21 was computed using the larger (+/+) covariance matrix. Even using this robust estimate for the covariance matrix, the T2 statistic is highly significant, i.h., hivhl+ has a highly signi- ficant effect on the axial dimensions of the melanosomes. To determine which of the melanosomal axial dimensions are significantly affected by hivh/+, 99% Roy-Bose simultaneous confidence intervals (Morrison, 1967) were constructed for both the major and minor axes (see Table 22). Since zero is not included in either confidence interval, it can be concluded that hiWh/+ significantly reduces the length of both the major and minor axes of the melanosomes. Shape: To determine the effect of hiWh/+ on melanosomal shape, the eccentricity was computed for the untransformed axial dimensions of the +/+ and hivhl+ melanosomes (see h Table 23). The mean eccentricities of the +/+ and hi? /+ melanosomes were compared with a z-statistic. Based on the results of this test, it can be concluded thathiWh affects the shape of the melanosomes such that the eccen- tricity of the hiyhl+ melanosomes is significantly less h than that of the +/+ melanosomes, i.e., the hiw /+ melanosomes are rounder in profile (more spherical in shape) 135 Table 22. Simultaneous Confidence Intervals of the Melanosomal Axial Dimensions - - 1 Major Axis: 0.015 < ( +/+ - ) < 0.100 Minor Axis: 0.016 < (X+/+ - X wh ) < 0.078 2 1 confidence intervals were constructed using T .01’2,1000 136 Table 23. Melanosomal Eccentricity Genotype Eccentricity +/+ 0.722 r 0.007 (781)1 MiWh/+ 0.692 t 0.008 (803) Comparison of means: 2 = 2.633** 1 number of melanosomes ** 2.01 a 2.575 137 than the +/+ melanosomes. Phaeomelanin The quantity of phaeomelanin derived from-equal weights of E/E3+/+ and h/h;hiWh/+ hair is summarized in Table 24. By inspection of the data in Table 24, it is clear that hiWh significantly increases the quantity of phaeomelanin deposited in the hair, such that g/h;hiWh/+ hair contains approximately 202 more phaeomelanin than E/£;+/+ hair. Tyrosinase isozymes The isozyme patterns of +/+ and hiWh/+ anagen VI follicular tyrosinase are shown in Figure 7. hiWh does not seem to affect the mobility of the soluble tyrosinase isozymes. Non-Pigmentary Phenotype of hiWh The data of this section were collected to: 1) de- h termine the specificity of gene action of hiw and 2) allow a comparison of the non-pigmentary phenotypic effects of hiWh and hh to investigate the possible homology of these two mutations. Hematology The results of the erythrocyte diameter measurements, reticulocyte counts and the leukocyte differential counts are summarized in Table 25. The data from the erythrocyte 138 Table 24. Phaeomelanin Content of Hair Genotype Phaeomelanin (A580) 1 2 2/23+/+ 0.092 0.003 (6) 5/3;Mi°h/+ 0.109 0.003 (6) 1optical density at 580nm of a 1:100 dilution of phaeo- melanin solubilized from 50mg of hair incubated in 0.5M NaOH for 48hr 2number of hair samples examined 139 A435 f," +/+ + to L migration —e- , —'>'i LO cm 1“- Figure 7. Tyrosinase Isozymes 140 omsoa\mou>oox=oa on: mo muasoo «mmazuoaoQooHE mm omsoa\mouhoousu>uo Good No mucsou N 6930:0300:— mm moses—\mouzoounumuo cod no muasoo N dakuoaom\oofi=m~ m.o H ~.~ m.o H m.~ o.~ H N.sm s.~ H e.o~ ~m.~ H m.- ~o.o H Ho.m ssmm\sswm m.o H o.~ «.0 H ~.~ m.~ H ~.¢m m.~ H o.- me.~ H N.o~ ~0.o H no.0 +\:3flm m.o H ~.~ 0.0 H ~.N o.N H w.mw o.~ H a.- NHw.H H w.mm flNo.o H o~.c +\+ mawnnochom mouhoosoz oouhoonaahq mfiwnnouusoz assoc AB v Houoamaa onhuocoo ou>ooH=oHuom ouzoousuzum mmouhooxaoa 00H: oaoz mo xwofiouoaom .mm mHnoB 141 measurements and reticulocyte counts were analyzed with one way analyses of variance (see Table 26a,b). From the results of these analyses, it is clear that hiWh has ,no significant effects on either the erythrocyte dia- meter or reticulocyte count. Therefore, hivh, even when homozygous, does not disturb the normal hematology of the adult mouse. It is also clear from inspection of the data of Table 25 that hiwh also has no effect on the relative proportions of the different classes of leuko- cytes present in the peripheral blood. In summary, hiWh has no discernable effects on the hematology of either 11th” or Evil/fl“ male mice. Growth curves h wh The growth curve data for male +/+ and hi? /hi mice are shown in Figure 8. Each pair of weekly weights of the +/+ and hiWh Wh lhi miCe were compared by a Student's t-test. None of the weekly mean weights of the two geno- types were significantly different from each other. This author is aware that the significance levels of the indi- vidual t-tests is compromised by the non-independence of the data from which these multiple t-tests were generated. Nevertheless, since none of the tests showed significant differences, i.h., at no point did the growth curves consistently diverge from each other, it seems reasonable h Ito conclude that hiw had no effect on the growth rates of these male mice. This conclusion is entirely 142 Table 26. Anova of Erythrocyte Diameters and Reticulocyte Counts a. Erythrocyte Diameters Source of Variation d.f. SS MS f Genotype 2 4.822 2.411 0.618 n.s.' Mice/genotype 12 46.805 3.900 32.667** Erythrocytes/ 1485 177.339 0.199 mouse 1f =389 .05,2,12 ° ** f = 2.18 .01,12,1485 143 Table 26 (cont'd.). b. Reticulocyte Counts Source of Variation d.f. SS MS Genotypes 2 69.07 34.533 1.184 n.s.1 Mice/genotype 12 351.80 29.166 1.00 n.s.2 Reticulocytes/ 15 437.00 29.133 mouse 1f a 3 89 .05,2,12 ' 2f . 2.02 .05,12,15 144 8:2 gas. eo 32:0 5320 .m 952... 9.33 b. m. 9.1m. N. = o. m .o h m n v q . q q q! u a q . - u . - d zzgxzsqz xillx . +\+ oiio \ XXX \\H\\\\\H\ lWWliie IO N .1 o. 19. O N (tub) iubiaM on 145 consistent with the earlier observations of Gruneberg EE 1., (1965). Metabolic measurements The data on the rectal temperatures of the +/+, h/hiWh male mice is summarized in Table 27. By inspection, it is clear that hiWh does not seem to 81“”. and 81" affect the rectal temperatures of the male mice. The results of the 24hr metabolic measurements are summarized in Table 28. By inspection of the data, it is evident that the four variables measured (food and water h consumption, urine and feces production) for thehiw /+ h/h_i_Wh mice do not significantly differ from the and hi" respective mean values of the +/+ mice. Therefore, it can be concluded that hiWh does not affect either the food and water consumption or the urine and feces pro- duction rates of male mice. Pituitary cell density The cell density measurements within the sex zone h/hiwh male of the adenohypophysis of E/E3+/+ and h/h;hiw mice are summarized in Table 29. These data were compared using a one way analysis of variance (see Table 30). From the results of the analysis of variance, it can be con- cluded that hiwh does not affect the cell density within h wh the sex zone of the adenohypophysis of male _e_/§._;hiw [hi mice. 146 Table 27. Rectal Temperatures of Male Mice Genotype Temperature (00)1 +/+ 37.2 i 0.33 (1072 hiWh/+ 37.1 t 0.26 (7) Miwh/MiWh 37.1 t 0.34 (10) 1all temperatures were taken between 11:00 a.m. and 2:00 p.m. 2 number of mice measured 147 mahuocow\ooas Cd so moms mums musoaousmmoe Ham AuanoB xvon mo va\Ha :H venomous oaHu: mam Houmsw AuanoB upon mo va\aw cH venomous mooow mam moowfl ~oo.o H Hmo.o moo.o H «No.0 moo.o H om~.o -o.o H no~.o :3flm\nsflm ~oo.o H m~o.o moo.o H w~o.o ¢o0.o H qm~.o woo.o H coo.o +\:3flm ~oo.o H m~o.0 moo.o H amo.o o~o.o H mq~.o mmao.o H oo~.o +\+ flmooom NocHuD Nuouoz Hpooh onzuoaoo mow: mam: mo musmsousmooz oHHonouoz mac: «N .mN manmh 148 Table 29. Cell Density in the Sex Zone of the Adenohypophysis- Cell Density2 Genotype (cells/(80pm) ) £/£;+/+ 95.83 t 2.78 (12)1 h/h;MiWh/hiWh 93.42 t 3.10 (12) 1number of measurements/genotype Table 30. Anova of Sex Zone Cell Density Measurements 149 Source of Variation d.f. SS MS f Genotypes 1 35.04 35.04 0.147 n.s.1 Mice/genotype 4 953.34 238.36 2.827 n.s.2 Cell counts/ 18 1517.25 84.29 mouse 1 f.05,1’4 7.71 2f -2.93 .05,4,18 150 Dermal-Epidermal Crafts and ih Vitro Epidermal Cultures Involving hiWh The data in this section were collected to: 1) deter- mine the site of gene action of hiwh on cutaneous melano- cytes and 2) allow a comparison of the site of gene action ofhiWh and Eh, to investigate the possible homology of these two mutations. Experiment i The results of the first series of dermal-epidermal recombination grafts are summarized in Table 31. From the results of this series of grafts, several conclusions can be made. First, the isogeneic recombinations, i.h., h h <+/+ // +/+). . and (h/h // h/h) which were all cultured in the testes of +/+ hosts, all produced grafts whose pigmentation pheno- type was the same as the pigmentation phenotype of a mouse of the comparable genotype. These results demonstrate that the recombination and i vivo culture conditions are per- missive to melanocyte differentiation. Several (+/+ // +/+) grafts were also cultured in the testes of hivh/+ hosts. In these culture conditions, the (+/+ // +/+) grafts produced black pigmentation. These results, combined with the gray pigmentation phenotype of the 041““ // yiwh/H grafts cultured in the +/+ 151 Table 31. Dermal-Epidermal Recombination Grafts-1 Experiment I Eoidermal Genotype Dermal Genotype +/+ hiWh/+ hiyh/hiyh h/h +/+ 32(26)3 -4 B (16) B (15) B (4) B (5) i i w (10) w (0) hi““/+ —‘ c (24) c (3) G (2) w (31) w (18) hiWh/hiWh B (28) G (31) w (27) w (16) EU! B (22) G (21) w (14) w (18) 1The dermis and epidermis of these grafts were from 13-14 day embryos. 2B - grafts with black hair, Bi - grafts with white hair and black interfollicular melanocytes, G - grafts with gray hair, W - grafts with only white hair 3number of grafts of each phenotype 4This genetic combination was not extensively made since preliminary studies indicated that hiwhl+ and +/+ melanocytes could not be reliably distinguished in the same graft. 152 testes, indicate that the reduced follicular pigmentation of the hiWhl+ mouse is probably not due to a systemic deficiency in the substrates or cofactors necessary for melanogenesis within the follicular hiWh/+ melanocytes. These observations do not, however, rule out the possibility that the reduced follicular pigmentation of the hiWh/+ mice could be caused by a defect at the level of the individual EEYh/+ melanocytes which reduced the rate or efficiency of melanogenesis. Second, the pigmentation phenotypes of the (+/+ // h wh wh /fl$ hi lhiyh) and (hivh/+ // fliw ) grafts are the same as those of the (+/+ // +/+) and (hiWh/+ // hiYh/+) grafts, i.e., black and gray respectively. These results indicate h[hiWh dermis from 13 day embryos does not that thehiw adversely affect the differentiation of either +/+ or hivhl+ melanoblasts which are already present in the 13 day epidermis. Likewise, the black pigmentation phenotype of the (+/+ // h/h) grafts and the gray pigmentation phenotype of the (hth/+ // h/h) grafts indicate that the h[h dermis derived from 13 day embryos also does not adversely affect the differentiation of either +/+ or hiWh/+ melanoblasts already present in the epidermis of 13 day embryos. This latter observation of the permissive nature of the h/h dermis with respect to the normal dif- ferentiation of epidermal melanoblasts is consistent with the observations of Mayer (1973a). Third, in a comparison of the pigmentation phenotypes 153 h wh of the (+/+ // +/+) grafts and the (hiw /Hi h wh // +/+) grafts, the presence of (hi? /hi // +/+) grafts with white hair indicates that the epidermis derived from 13 h day hiw /hiWh embryos adversely affects the development of functional +/+ melanocytes (derived from the 13 day +/+ “lgth epidermal follicular environ- dermis) within the hiw ment. Likewise, in a comparison of the pigmentation wh wh phenotypes of the (hi /+ // hi /+) grafts and the wh wh h (hi lg; /MiWh // // hiWh/+) grafts, the presence of (hiw hivh/+) grafts with only white hair is also consistent with the conclusion that epidermis derived from 13 day hiWh/hiWh embryos adversely affects the development of highl+ melanocytes derived from the 13 day hiyhl+ dermis. Fourth, in a comparison of the pigmentation phenotypes of the (+/+ // +/+) and (hivh/+ // hth/+) grafts with the (h/h // +/+) and (h/h // hiWh/+) grafts, the presence of white haired grafts in the latter two genotypic com- binations suggests that the epidermis derived from 13 day h/h embryos adversely affects the develpment of functional melanocytes derived from either +/+ or hiWhl+ dermis. This conclusion is consistent with the data reported by Mayer (1973a), 312-: of 16 (13 day fl/K // 13 day +/+) grafts, eight of the grafts had pigmented hair, and eight of the grafts had only unpigmented hair. However, the con- clusion that epidermis from 13 day h/h embryos adversely affects melanocyte differentiation differs from Mayer's (1970) conclusions, based on 9 day +/+ neural tube-13 day 154 h[h skin recombination grafts, that h[h skin is completely permissive to normal +/+ melanocyte differentiation. Returning to the data in Table 31, in a comparison of the percentage of non-pigmented grafts recovered from h the (hiw lhiWh // +/+) and (h/h // +/+) grafts (33% vs. OZ), it appears that the epidermis derived from 13 day hth/hiWh embryos is more hostile to the normal develop- ment of +/+ melanocytes than is the epidermis derived from 13 day h/h embryos. Fifth, from a comparison of the percentage of pig- h wh mented grafts recovered from the (hiw [hi // +/+) and wh (hi [hiWh //‘hiWh/+) recombinations, (100% vs. 102), it can be concluded that hiWh/+ dermal melanoblasts are, when challenged with the same epidermal environment as +/+ dermal melanoblasts, less capable of undergoing normal development resulting in the production of pigmented hair. This subnormal developmental potential of the melano- blasts derived from 13 day hth/+ dermis suggests that the gene action of hiwh is in part intrinsic to the post-13 day melanoblasts. h wh Sixth, the skin combination of (h[h //hiw /hi ) was designed to test for the presence of functional h melanoblasts in thehiw /hiWh dermis. h/h epidermis does not possess any melanoblasts (Mayer, 1970, 1973a). Thus, if these grafts had become pigmented, then the pigmentation h wh would have been derived from hiw lhi melanocytes which had migrated into and differentiated within the "permissive" 155 h/h epidermal environment. Conversely, if the grafts had remained unpigmented, then it could have been concluded h wh that the 13 day hi? /hi dermis lacked melanoblasts capable of development into functional melanocytes. How- ever, the recovery of unpigmented (h/h // +/+) grafts demonstrated that the h[h epidermis is not a completely normal environment for melanocyte development. Furthermore, ‘the subnormal developmental potential h wh of the hivhl+ melanoblasts suggests that the hi? [hi melanoblasts, if they exist at all, would probably be extremely sensitive to changes in the epidermal environ- ment. Therefore, the failure to recover pigmented (h/h // hiWh/hiWh) grafts could have been caused by either the action of a non-permissive h/h epidermal environment on h extremely sensitive hiw lhiWh melanoblasts, or by the h complete absence of hi? [hiwh melanoblasts in the 13 day hiWh/hiWh dermis. Based on the available data, these two possible interpretations cannot be resolved. Seventh, a comparison of the pigmentation phenotypes h h h wh new" // my) and (9.1.” /_M_i"h // 8.1.” /1_1_i_ > grafts of (if was designed to test for the possible inhibitory effects h wh of the 13 day Hiw [Hi dermis on the differentiation of hIEiWh melanoblasts which might already be located h/hiyh epidermis. If the (hiw°/hi°° any is" within the 13 day hi" // /W) grafts had become pigmented, then it could have been h wh concluded that the epidermis from the 13 day hi? /hi h embryos was capable of supporting hiw /hiWh melanocyte 156 h wh lhi dermis was the h wh differentiation and that the hi? cause of the failure of pigmentation in the hi3 lhi mice. However, the 1002 recovery of pigmented grafts from h wh h wh the (+/+ // hi“ /hi ). (hfhn // hi“ llfi >. (+/'+ // 91/3) and (hiWh/+ // h/h) recombinations demonstrates that the dermis derived from both the MiWh/hiWh and h/h embryos is permissive to the differentiation of melanoblasts already within the 13 day epidermis. Therefore, the recovery of h/Eiyh // fl/h) grafts indicates that h only unpigmented (hiw either the epidermis from the 13 day hi? lhiwh embryos did not contain any melanoblasts at all, or that the high/high epidermal environment is sufficiently hostile to h wh prevent the differentiation of hi? /hi melanoblasts into functional melanocytes. Again, the available data do not permit a resolution of these two possible interpretations. In summary, the results of the (13 day epidermis // 13 day dermis) recombination grafts lead to the following conclusions. 1. The reduced pigmentation of the hiWh/+ mouse is probably not caused by a systemic defect in the availability of sub- strates or cofactors necessary for normal melanogenesis in the follicular hiWh/+ melanocytes. The data for this conclusion do not, however, exclude the possibility that ‘hiWh/+ may cause, at the level of the individual melano- cytes, a defect in the availability of substrates or cofactors necessary for normal levels of melanogenesis. h wh 2. The dermis from 13 day hiw /hi and 13 day h/h embryos 157 h is permissive to the differentiation of +/+ and hiw /+ melanoblasts present in the 13 day epidermis. h/hi_wh or 13 day 3. a) Epidermis from either 13 day hi? h/h embryos adversely affects the normal development of melanoblasts derived from either +/+ or hiWh/+ 13 day dermis. b) The deleterious effects of epidermis from 13 day hivh/hiWh embryos (on +/+ melanocyte development) are more severe than the effects of epidermis from 13 day h[h embryos. 4. The hivhl+ melanoblasts from 13 day hiyhl+ dermis are subnormal in their ability to develop into functional melanocytes within either a hiwh Wh h [hi or h/h epidermis. 5. Since both the hiw /hiwh and h/h epidermis are inimical to the development of functional melanocytes, it is not possible from the data in experiment I to decide if the h/MiWh embryos contains melanoblasts. skin from 13 day Lilw Experiment ii To address the problem raised in point five of the summary in the preceeding paragraph, i.e., the demonstration of the presence or absence of melanoblasts in the skin of hKhiWh embryos, the recombination grafts listed 13 day hiw in Table 32 were made. The rationale for this series of grafts is discussed in the following paragraphs. Mayer (1973b) had demonstrated that: 1) flank skin epidermis from 11 day +/+ embryos is essentially devoid of 158 Table 32. Dermal-Epidermal Recombination Grafts-1 Experiment II Phenotype of Recovered Grafts 2 Genotypic Black Dusky Clear White Combinations Yellow Yellow E/E-+/+3 - -’ 2 1 0 18 g/g:+/+ 5’5”” 7 0 0 6 s/smiwhfi 5/1'3- 4” 7 0 0 19 2/23 !_ wh/ iwh 1the epidermis is from 11 day embryos, the dermis from 13 day embryos 2Black 8 hair color produced by h/h;+/+ melanocytes, Dusky yellow - hair color produced by E/E3+/+ melanocytes, Clear yellow - hair color produced by h/h;hiWh/+ melanocytes, White - unpigmented hair 3skin combinations are listed as: epidermis dermis 159 melanoblasts and 2) 11 day epidermis, when combined with 11 day +/+ dermis, will permit the migration of melanoblasts from the dermis into the epidermis and the normal differ- entiation of these melanoblasts into functional melanocytes Therefore, 11 day embryos were used as the source of epi- dermis for the grafts listed in Table 32. To allow positive identification of the genotype of any melanocytes recovered in these grafts, the dermis and epidermis were differentially marked, using the h and g alleles of the extension locus. The extension locus was chosen as a melanocyte marker since Lamoreux and Mayer (1975) and Poole and Silvers (1976b) had previously demon- strated that the gene action of the alleles of the exten- sion locus is intrinsic to the melanocytes and that the alleles of this locus do not affect melanocyte function through either the dermal or epidermal environment. As can be seen in Table 32, the epidermis of all grafts carried hfh while the dermis of all grafts carried h/g. Therefore, if any of the recovered grafts contained black hairs, then the melanocytes of these grafts would have come from the h/h melanoblasts of the epidermis. Like- wise, if any of the recovered grafts contained yellow hairs then the melanocytes of these grafts would have come from the 3/5 dermis. In each graft, the dermis also carried either +/+, h/ wh hiWh/+ or hi? hi . It was expected that as controls, the (11 day h/h;+/+ // 13 day g/g;+/+) grafts would 160 produce yellow hairs tipped with black, i.e., the dusky yellow phenotype characteristic of the fur of h[h;+/+ mice. Likewise, it was expected that the (11 day h/h;+/+ // 13 day h/h;hiWh/+) grafts would produce completely yellow hair characteristic of the yellow fur of h/h;hth/+ mice. h wh) Finally, if the (11 day §[§;+/+ // 13 day g/g:hiy /§i grafts produced any yellow pigmentation, then it could be wh/ concluded that the 13 day h/h;hi hiWh dermis contained melanoblasts capable of differentiation in the permissive §[§;+/+ environment. Conversely, if the (11 day hfh;+/+ // h/hiWh) grafts were all unpigmented, then it hlhiwh dermis 13 day alsyf could be concluded that the 13 day g[h;hiy did not contain melanoblasts capable of developing into functional melanoctyes within the permissive h/h;+/+ environment. However, as can be seen by the results summarized in Table 32, the predicted results outlined in the previous paragraph were not obtained. First, 16 grafts of all three genotypic combinations contained black hairs. In all cases, this pigmentation was restricted to one or two small clusters of pigmented hairs in grafts that otherwise con- tained large numbers of unpigmented hairs. Based on the black color and restricted distribution of the pigmentation in these grafts, it is evident that the 11 day h/h;+/+ epidermis used in these grafts contained a few melanoblasts which had already migrated from the dermis into the epi- dermis before 11 days of development. 161 Second, the majority of the (11 day h/h;+/+ // 13 day E/£3+/+) grafts (20 of 21) and the (11 day h/h;+f+// 13 day h/h;hiWh/+) grafts (13 of 13) failed to produce any yellow pigmentation at all. Based on the failure of these control grafts to consistently produce pigmentation, the failure of the (11 day h/h;+/+ // 13 day 5/35hiWh/hiWh) grafts to produce any yellow pigmentation cannot be re- garded as definitive proof of either the presence or absence of melanoblasts in the 13 day h/h;hiyh/hivh dermis. The failure of the (11 day §/§;+/+ // 13 day g/g;+/+) and (11 day h[§;+/+ // 13 day g[h;hiyh/+) grafts to pro- duce yellow pigmentation can be explained by at least two possible hypotheses. Heterochrony: It is possible that the melanoblasts from 13 day dermis cannot develop into melanocytes within the epidermis derived from 11 day embryos due to the differences in age between the dermis and the epidermis. The 48 hour dif- ference in the ages of these two tissues may be sufficient to disrupt the interactions between the dermis and epi- dermis that permit the melanoblasts to migrate into the epidermis, survive and proliferate within the epidermis, and differentiate into functional melanocytes within the follicular epidermis. 162 h-locus: It is also possible that the 3 allele of the exten- sion locus may have some previously undetected effects on melanoblasts such that the 3/5 melanoblasts within the 13 day dermis are unable to either migrate into the h/h epidermis, or differentiate into functional melanocytes within the follicular epidermis. Alternatively, the E42 dermis may act on the h/h epidermis to prevent the migra- tion or differentiation of the 2/2 melanoblasts within the h/h epidermis. Experiment III In an attempt to investigate the hypotheses discussed above, a preliminary series of (11 day epidermis // 11 day dermis) grafts were made (see Table 33). As expected, the recovery rate of these grafts was poor (Mayer, 1970). However, of the 24 grafts recovered, only one of the re- combinant grafts, and three of the intact skins produced pigmented hairs. Clearly, if either heterochrony or the 3 allele are affecting the ability of the dermal melano- blasts to develop into functional melanocytes, other fac- tors must also be involved. With the preliminary results summarized in Table 33, it was clearly unprofitable to attempt to assay for the h wh presence of melanoblasts in the 11 day hiw /hi dermis using this experimental system. 163 Table 33. Dermal-Epidermal Recombination Grafts---1 Experiment III Recovered Grafts Genotypic Implanted White Black Dusky Combinations Grafts Yellow 2 1°13 26 7 - 1 0 iii/E yfi 18 2 0 0 2/2 5’3 20 1 0 0 E/E e/e3 - - 14 10 0 3 13./s (intact) 1dermis and epidermis from 11 day embryos 2skin combinations are listed as epidermis dermis 3these skins were treated identical to all other grafts except they were not trypsinized and recombined 164 Experiment i! To inwestigate the mode of gene action by which the h wh 13 day hi? [hi epidermis blocked the normal development of the +/+ and highl+ melanocytes derived from the dermis of 13 day embryos, a final series of grafts was attempted. Mouse melanocytes can be immunologically identified by cell surface antigens produced by the he; locus (Mintz and Silvers, 1970). A series of (13 day hiwh 13 day +/+) and (13 day hiwh h /MiWh // /hiy // 13 day hiyh/+) grafts were made in which the epidermis and dermis carried dif- ferent h-i alleles. Specifically,the epidermis carried fish/9:2" and the dermis carried hrib/hfiid. It was in- tended that the dermally-derived heib/hfiid b/hfiib epidermis by means melanoblasts could be detected within the £52 of indirect immunofluorescent staining using mouse anti- (mouse h-id) and FITC-conjugated rabbit anti-(mouseIgG) antisera. These experiments were unsuccessful for two reasons. First, based on the condition of the grafts recovered from h—ib[hfib host males, it appeared that the testes were not an immunologically privileged site for the survival of the grafts containing the h-ib/h-id dermis. This tentative conclusion is based on two observations. 1) The hair in the (h-ib/h-ib // hfib/h-id) grafts was consistently less well developed than in comparably aged (hegblgegb // hfiiblh-gb) grafts. 2) (h-ib/heib // h-ib/h-id) grafts grown 165 in H-2 lH-ib hosts for 4-5 weeks consisted of only pigmen- ted epidermal cysts with no visible traces of either hairs or follicular development; a situation never seen in com- b // g-gblg-gb) grafts. parably aged (h-ib/h-i Second, and more serious, despite the use of a variety of techniques designed to reduce or supress non-speicfic fluorescence (Johnston and Beinenstock, 1974; Schenk and Churukian, 1974; Fey, 1972; Karlsson SE _i., 1975; Burtin and Sabine, 1972; Burtin and Gendron, 1971, Grossi and von- Mayersbach, 1964; vonMayersbach, 1967), both the dermis and epidermis of the recombination skin grafts had extremely high non-specific flourescence. In the face of this high non-specific fluorescence, it was impossible to determine d whether or not the hfib/h-i melanoblasts were present in the h-ib/h-ib epidermis. ih vitro epidermal cultures Markert (1948) and Mayer and Oddis (1977) have demonstrated that genetically normal dermis can in some situations delay or completely inhibit the normal develop- ment of melanoblasts into functional melanocytes. To de- dermine whether the presence of any dermis, regardless of genotype, was sufficient to block the development of [hiWh/hiwh melanoblasts into functional melanocytes, ih xihhh primary cultures of epidermis from 14 day +/+, Wh/§$Wh hivhl+ and hi embryos were established. Ten to twelve days after explantation, pigmented melanocytes could 166 readily be identified in both the +/+ and hiWh/+ cultures. h wh However, the hi? /hi cultures did not contain any pig- mented melanocytes. wh h/lfl. As with the (13 day hi? // 13 day E/E) grafts, this apparent lack of melanocytes could be due to a real absence of melanoblasts in the 14 day hiwh/hiWh epidermis, h wh or a failure of the hi? [hi h/hiVb epidermal environment. In any case, melanoblasts to develop within the hi? the absence of the dermis did not stimulate the develop- h wh ment of hi? /hi melanocytes. Pigmentary Phenotype of Eh The data in this section were collected to allow a comparison of the pigmentation phenotypes of high and Eh to investigate the possible homology of these two mutations. Ear melanocytes The distribution of melanocytes in the ear skin of in Table 34. Two conclusions can be drawn from these data. First, as previously noted by Rappaport gh_hi. (1963), the melanocytes of the hamster ear are restricted to the der- 'mis. Second, within the ear, Eh appears to act as a typical white spotting mutation, causing, in the homozygous condition, a total absence of dopa-positive melanocytes. 167 Table 34. Hamster Ear Melanocyte Distribution Genotype Dermis Epidermis 1 s/ssze/xh + - 2 E/Eifl/V—h. +/- - s/safl/Eb. - - + a presence of melanocytes, - a absence of melanocytes 2g/h;Eh/Eh ear pigmentation is patchy, however, dopa staining does not reveal any melanocytes outside of the macroscopically visible pigmented patches 168 Skin melanocytes The distribution of melanocytes in the skin of male h[h;gh[hh, h[g;Eh[Eh and h[h;Eh/Eh hamsters is summarized in Table 35. In the.£4E3EE/EE hamsters, the absence of dorsal skin dermal melanocytes is consistent with the pre- vious observations of Illman and Ghadially (1960). The presence of melanocytes in the dermis of the costovertebral gland is also consistent with the observations of Illman and Ghadially (1960), and parallels the presence of melano- cytes in the ear skin dermis. In the h[h;Eh[Eh hamsters, the lack of melanocytes in the dorsal skin and the epidermis of the costovertebral gland is consistent with the genotype (£42): and the overall pigmentation phenotype of these hamsters. However, the absence of melanocytes in the dermis of the costo- vertebral gland does not parallel the patchy presence of melanocytes in the ear skin dermis of these hamsters. The absence of all cutaneous melanocytes in the E/E3 Eh/Eh hamsters is consistent with the overall white spot phenotype of these hamsters. Ocular melanocytes The distribution of melanocytes in the eyes of g/h;gh/_h, h/g;Eh/Eh and h/h;Eh/Eh hamsters is summarized in Table 36. Like hiWh/+ in the mouse, Eh[Eh causes irregular pigmentation of the choroid. However, in 169 Table 35. Cutaneous Melanocyte Distribution in Hamsters Dorsal Skin Costovertebral Gland Follicular Follicular Genotype Dermis Epidermis Dermis Epidermis 1 algae/21.13 - P E P e/ssEh/zh - - - - flea/fl - - - - - - no identifiable melanocytes present, P - phaeomelano- genic melanocytes present, E = eumelanogenic melanocytes present 170 Table 36. Distribution of Hamster Ocular Melanocytes Retinal Layers Uveal Tract Retinal and Ciliary Ciliary Epi- Iris Choroid Body and Genotype thelium Iris s/ssfl/iia + + + + E/EeflLi/Efl + + +/..1 + s/ssfl/flz - +3‘ 1’" 9' 1Choroidal pigmentation is present in all eyes, but is patchy in its distribution. Since h/g;Eh/Eh hamsters have degenerative anophthalmia, these data were derived from the examination of embryonic material. 3Pigmentation is present but reduced in quantity and is restricted to the presumptive pupillary margin of the iris. Pigmentation in the uveal tract does not develop until after birth. 171 contrast to hiWh/+, in all of the Eh/Eh eyes examined, some regions of the choroid were pigmented. These pigmented regions of the choroid varied in size in different eyes, but were always normal in their intensity of pigmentation and were always continuous with the pigmented ciliary stroma. Since Eh[Eh causes extreme degenerative anophthalmia, observations on the effects of Eh[Eh on optic pigmentation were made on embryonic material. At 12% days of develop- ment, when Eh[Eh and Eh[Eh embryos have pigmentation throughout the outer wall of the optic cup, Eh/Eh embryos have pigmentation only in the anterior portion of the optic cup in the presumptive pupillary margin of the iris. Dermal-Epidermal Grafts Involving Eh The data in this section were collected to: 1) deter- mine the site of gene action of Eh on the cutaneous melanocytes and 2) allow a comparison of the site of gene action of hiWh and Eh, to investigate the possible homol- ogy of these two mutations. The results of the hamster dermal-epidermal recom- bination grafts are summarized in Table 37. Graft re- covery in this experiment was quite poor. Of the 106 grafts implantmi,only 46 grafts that had developed hair were recovered. Within these 46 grafts, four pigmentation phenotypes could be recognized: 1) yellow hair with black dermal 172 Table 37. Hamster Dermal-Epidermal Recombination Grafts Epidermal Genotype Dermal Genotype s/e;zh/_§ g/gsflh/g_ g/g:flh/_g .E/£3_E/_E YBl(10)2 YB (2) YB (3) Bd (2) Bd (2) Bd (2) 3 (12/15) W (5) W (5) (9/22) (10/18) e/g;_h/_h YB (1)_ w (1) -4 Y (5) (1/10) Bd (2) W (1) (9/24) e/e;y_/_h YB (2) -“ —“ Y (1) Bd (2) (5/17) 1 YB - graft with yellow hair and black dermal melanocytes, Y - graft with yellow hair and no visible dermal melanocytes, Bd - graft with white hair and black dermal melanocytes, W - graft with white hair and no visible dermal melanocytes 2number of grafts recovered with specific pigmentation phenotype 3(number of grafts recovered/ number of grafts implanted) 4grafts of this genetic combination were not made 173 melanocytes, 2) yellow hair only, 3) white hair with black dermal melanocytes, and 4) white hair only. From these pigmentation phenotypes and their pattern of occurrence in the various genotypic recombinations, several conclusions can be made. First, all genotypic combinations produced some grafts with no follicular pigmentation.. The isogeneic recombi- nation (h/g;gh/Eh //.£[EIEE/!E) was least affected with only 2 of 12 grafts with unpigmented hair. The allogeneic combinations that had Eh[_ in the dermis were slightly but not significantly more affected than the (h/h;Eh/Eh // g/gsfi/fl) grafts with 5 of 14 (yeah/2h // gawk/J grafts lacking pigmented hairs (see Table 38). In com- parison with these first three genetic combinations, all of which had Eh/Eh epidermis, the allogeneic recombinations with Eh/_ in the epidermis has significantly more (14 of 19) grafts lacking pigmented hair (see Table 38). Furthermore, among the allogeneic recombinations that had Eh/_ epidermis, the magnitude of the effect of Eh on the presence of unpigmented hairs was not significantly different between Eh[Eh and Eh[Eh epidermis (see Table 38). Thus, it appears that the presence of Eh in the epidermis of the dermal epidermal recombination grafts adversely affects the development of follicular melanocytes. Also, within the allogeneic recombinations with Eh/_ epidermis, Eh appears to act as a simple dominant allele, i.3., the phenotypic effects of Eh/Eh and Eh/Eh on follicular 174 Table 38. Proportion of Hamster Dermal-Epidermal Grafts Without Pigmented Hair Contrast X2 (Ell/Eh // gh/gh)' vs. (fl/yh // gh/__) 1.169 as2 (113/112 // FA/Ell) and (vi/it // 99L) vs. (EM... // 311/311.) 9-644“ (ELI/11.1 // was.) vs. (Eh/Eh // w_h/Eh) 0.147 n.s.2 1all grafts were also 2/2 2comparison by contingency table, x2 05 1 = 3.841 ** 2 comparison by contingency table, x 01 1 = 6.635 175 melanocyte development are not different. This latter con- clusion is consistent with the effects of the Eh allele on the fur pigmentation phenotype of the h/h;Eh/Eh and h/h;Eh[Eh hamsters. Second, black dermal melanocytes appeared in grafts of all genetic combinations except (g/h;Eh/Eh // h[h;Eh[EhL Since only one (h/h;Eh/Eh // h/h;Eh/Eh) graft was recovered the white phenotype of this graft can hardly be considered as a significant datum to support any hypothesis. In an explanation of the unexpected appearence of the dermal melanocytes, three aspects of this phenomenon must be con- sidered: 1) the presence of these dermal melanocytes in grafts whose E42 genotype ordinarily supresses the presence of melanocytes in the dermis of hairy skin, 2 a) the pre- sence of dermal melancytes in some (h/h;Eh/Eh // h[h;Eh/_) grafts, b) the absence of dermal melanocytes in some (g/h;Eh/_ // h/h;Eh/Eh) grafts, and 3) the genotype of the dermal melanocytes in the(g[g;hh[gh //.£/£3HE/_) grafts. First, as described in the Literature Review, one of the effects of 342 is to supress the presence of the dermal melancytes in all of the hairy skin of the hamster except the costovertebral gland. The costovertebral glands of Eli hamster of both sexes contain melanoblasts, which can upon androgenic stimulation, undergo proliferation and melanogenesis (Frost 5h ii., 1973; Hsia and Voight, 1974). Based on histological examination, the dermal-epidermal recombination grafts do not possess either the hair or 176 or sebaceous gland morphology characteristic of the costo- vertebral glands. Therefore, it is probable that these grafts do not represent costovertebral glands with their associated dermal melanocytes. Nevertheless, since these grafts were cultured in the testes, clearly an environment with high androgen levels, and the grafts did produce func- tional dermal melanocytes, it is clear that the 2,3 dermis of these grafts, in contrast to its normal ih high behavior, is capable of supporting the differentiation of functional (and probably androgen-dependent) melanocytes. Second, the presence of dermal melanocytes in 7 of 14 (h/h;Eh/Eh // g/g;Eh/_) grafts indicates that the g/g;Eh/_ dermis of hairy skin, although not as effective as h/E;Eh[g} dermis, is capable of supporting melanocyte development within the dermis itself. Conversely, the presence of dermal melanocytes in only 9 of 19 (h/h;Eh/_ // h/h;Eh/Eh) grafts, in comparison to 12 of 12 (h/h;gh[gh // h[g;Eh/Eh) grafts, indicates that the g/h;Eh/_ epidermis can adversely affect the development of functional melanocytes within the h/g dermis. Third, since the ear and costovertebral gland of the _[h;Eh[gh hamster and the isogeneic (h/h;Eh[Eh // h/h;Eh[EQ grafts all contain dermal melanocytes, it is reasonable to conclude that most or all of the dermal melanocytes of the allogeneic (3/3;Eh/_ // g/g;Eh/Eh) grafts are h/h;Eh/Eh in genotype. However, the genotype of the dermal 177 melanocytes of the (g/h;Eh[Eh // h/h;Eh/_) grafts cannot be as easily decided. If melanoblasts are capable of retrograde migration from the epidermis to the dermis, then the dermal melanocytes of the (h[g;gh[gh // g/g;Eh/_) grafts could be.E4E3EE/!E in genotype. If, however, melanoblasts cannot undergo retrograde migration, then the dermal melanocytes of the (h/h;hh[gh // h[h;Eh[_) grafts would be h/h;Eh/_ in genotype. Since the ear dermis of the g/h;Eh/Eh hamsters does contain melanocytes, the latter possibility, for the g/h;Eh/Eh dermis at least, is possible. However, in the absence of further data, the genotype(s) of these dermal melanocytes cannot be determined. In summary, the results of the hamster dermal-epider- mal recombination grafts lead to the following conclusions. 1. Epidermis from h/h;Eh/_ embryos adversely affects the development of h[h;Eh/Eh follicular and dermal melano- cytes. 2. Dermis from g/h;Eh/_ embryos does not affect the devel- opment 0f.E/Ei!E/!E melanocytes in an g/h;hh[gh epidermis. 3. Dermis from h[h;Eh/_ embryos can, with less efficiency than h/h;Eh/Eh dermis, support the development of dermal melanocytes. 4. The data in Table 37 do not rule out the possibility that h/g;Eh/_ hairy skin dermis may contain melanoblasts capable of development into functional melanocytes. 5. The g/g;Eh/Eh embryonic skin contains melanoblasts 178 capable of development into eumelanogenic dermal melano- cytes. DISCUSSION The three major topics addressed in this discussion are: 1) an examination of the cutaneous pigmentation phenotype and the mode and site of gene action of hiwh, 2) a comparison of the phenotype, and site, specificity and mode of gene action of hiwh and other mutations in the mouse which cause either a reduction in the intensity of cutaneous pigmentation or white spotting, and 3) a comparison of the phenotypes and mode of gene action of hiWh in the mouse and Eh in the hamster. h Phenotype and Mode of Gene Action of hi? /+ Based on the data presented in the Results section of this dissertation, it seems clear that the gray pigmenta- tion of the fur of hiWhl+ mice is caused by: 1) a decrease in the number of melanosomes deposited in the hair (h/h;hiWh/+ hair has 28% of the number of melanosomes/mg of hair as does h/h;+/+ hair, i[h;hiWh/+ hair has 55% of the number of melanosomes/mg of hair as does h[i;+/+ hair), and 2) a decrease in the size of the individual melanosomes. Both of these effects of hiWhl+ lead to a decrease in the quantity of pigment deposited in the hair, and thus a 179 180 decrease in the intensity of pigmentation in the fur of the hth/+ mouse. The decrease in the number of melanosomes deposited in the hiWh/+ hair in turn, appears to be caused by the action of hiWh/+ on the level of function and the number of follicular melanocytes. First, hiWh/+ (on a h/h back- ground) causes a decrease in the average diameter of the follicular melanocytes. The average diameter of h/h;hi/+ follicular melanocytes is 81% that of the h/h;+/+ follicular melanocytes. Assuming the i/h melanocytes to be roughly spherical, then the i/h;hth/+ melanocytes are approximately (0.81)2 or 64% the size of the h/h;+/+ follicular melanocytes. This decrease in size of the fol- licular melanocytes would seem to indicate some sort of subnormal function of the hiWh/+ melanocytes. The concept of subnormal function of the hiWh/+ is also supported by the decreased size of the individual hiyhl+ melanosomes (see Table 21) and the behavior of the hiWh/+ melanocytes in (13 day hiWh/fliyh // 13 day hiWh/+) recombination grafts (see Table 31). Second, hiWh (again on the Eli background) decreases the number of melanocytes in the zigzag follicles. The h[i;hiWh/+ zigzag follicles contain only 85% the number of melanocytes found in the h/h;+/+ zigzag follicles. Although this decrease in melanocyte number is relatively small, slightly more thanunuamelanocyte/follicle for the follicles 181 which were measured, the conclusion that hiwh decreases melanocyte number is also supported by the effects of .hiWh/+ on melanocyte number in the tail and ear epidermis (see Tables 15 and 17b). If it can be assumed that: 1) melanocyte size is pro- portional to melanocyte function, and 2) all four types of i/h;hiWh/+ hair (zigzag, auchene, awl and guard) follicles have proportionally fewer melanocytes (85% of h[i;+/+), then the effects of hiWhl+ on the function and number of follicular melanocytes are sufficient to account for the observed decrease in the number of melanosomes in the h[h;hiWh/+ hair. Consider the following analysis. Let R - the rate of production of melanosomes by an individual follicular melanocyte, N a the number of melanocytes/ follicle, and M a the number of melanosomes/mg of hair. Then: = M , and (1) (R+/+)(N+/+) +/+ (R )(N ) = M . (2) MiWh/+ MiWh/+ MiWh/+ Based on the assumptions listed above, and the data from Tables 14 and 13, 2 R = (.808) (R Miwh/+ NM1Wh/+ = (.850)(N+/+). (4) +/+)9 and (3) Substituting (3) and (4) into (2), ) = M +/+ MiWh/+ (.808)2(R )(.850)(N +/+ 182 and simplifying, (R )(N+/+)(.554) a M h . (5) Miw /+ +/+ Substituting (1) into (5), (M+/+)(.55) - M wh . (6) 33./“r Equation (6) indicates that the h/h;hiWh/+ hair should con- tain 55% the number of melanosomes found in the i[i;+/+ hair. An examination of the independently derived data in Table 19 shows that the h/h;hiWh/+ hair does indeed contain 55% the number of melanosomes found in the i[i;+/+ hair. Thus, if the two assumptions listed above are correct, then the effects of hiWh/+ on melanocyte function and number are sufficient to explain the observed effect of hiwhl+ on ' melanosome number in the hair. Since the comparable data on melanocyte number and size could not be obtained for the E/E;+/+ and E/Q;MiWh/+ genotypes, a similar numerical analysis of the effects of hiWhl+ on melanosome number cannot be made for these genotypes. However, from the data in Table 19 on the num- ber of melanosomes in h/h;+/+ and h/h;hiWh/+ hair, and the data in Tables 15 and 17b on the number of melanocytes in the epidermis of h/h;+/+ and E/E;hiWh/+ tail and ear skin, it is clear that hth/+, acting on a 2/2 background is h capable of reducing the number and function of h/h;hiw /+ epidermal melanocytes. Concerning the mode of gene action of hiWh which h results in decreased numbers and function of the hiw /+ 183 epidermal melanocytes, consider first the decrease in melanocyte number. During each round of hair growth, the melanocytes in the anagen VI zigzag hair follicle are produced by repeated mitosis of one or two stem cell melanoblasts which were present in the telogen zigzag follicle. Thus,the hth/+-mediated reduction in follicu- lar melanocyte number could be caused by: 1) a reduction in the number of stem cell melanoblasts in the telogen follicle, 2) a decreased rate of mitosis of the melanocytes during anagen I to anagen V, 3) a loss of melanocytes due to cell death, or 4) a failure of differentiation of the melanoblasts into functional melanocytes. The first pos- sibility is unlikely. If him” reduced the number of telogen melanoblasts, then one would expect many of the smaller zigzag follicles which normally contain only a single telogen melanoblast to be completely devoid of melanoblasts and thus able to produce only unpigmented hairs. Since the hiWhl+ phenotype does not include any appreciable numbers of unpigmented zigzag hairs, it is un- likely that hiWh/+ acts to decrease the number of melano- blasts in the telogen follicles. Given the currently available data, it is not possible to decide which, if any of the other three possible modes of gene action are correct. However, the "reduced proliferation" hypothesis is attractive and could be readily tested either ih XlEEQ or ih high with pulse-chase 3H-thymidine labeling experiments (Rosdahl and Szabo, 1968). 184 Considering next the effects of hiWh/+ on melanocyte function, there are no available data to suggest why the hiWhl+ follicular melanocytes should be smaller than the normal +/+ follicular melanocytes. However, as discussed earlier, this reduced size does appear to be correlated h/+ with a reduced output of melanosomes from the hiw follicular melanocytes. Furthermore, all three tyrosinase isozymes are present in the hiWh/+ follicular melanocytes, indicating that tyrosinase function is not grossly affected. Finally, within individual melanosomes, the hiyhl+ tyro- sinases appear to permit normal levels of melanogenesis, resulting in stage IV melanosomes with electron densities comparable to those of stage IV +/+ melanosomes. However, the stage IV fith/+ melanosomes are smaller and have a lower eccentricity than the stage IV +/+ melanosomes. Considering that: 1) Breathnach (1969) has suggested that the keratinocytes may control the rate of melanosome transfer from melanocyte to keratinocyte, and 2) the overall rate of melanosome synthesis in the hiWh/+ follicular melanocyte is depressed, the smaller size and rounder shape of the hiWh/+ melanosomes may be a consequence of the transfer of not-fully matured, i.g., undersized, melano- somes from the hiWhl+ melanocytes in the growing anagen VI hair shaft. Alternatively, the changes in size and shape of the hiWh/+ melanosomes may be a consequence of a funda- mental disruption of some aspect of melanosome synthesis or melanogenesis. 185 h Site of Gene Action of hiw /+ Melanocyte The results of several experiments indicate that the gene action of hiWh is at least, in part, intrinsic to post-13 day hiwhl+ melanoblasts. First, in comparison with wh wh (hi /hi // +/+) and (E/E // +/+) grafts, the reduced h wh ”ii h // hi" /+> recovery of pigmented grafts from the (hiw and (E/E // hiWh/+) recombination grafts, indicates that the hiWh/+ melanoblasts derived from 13 day hiWh/+ embryonic dermis are subnormal in their ability to develop into‘func- h tional melanocytes. Second, in those (E/E // hiw /+) grafts which did produce pigmented hair, the overall pigmentation was gray and not black. Since neither the E[E nor E/+ genotypes ordinarily produce a gray pigmentation phenotype, it appears that the gray pigmentation phenotype of the (E/E //hiWh hi“ within the hiw /+) grafts was caused by the action of h/+ melanocytes. Dermis Since both of the sets of recombination grafts dis- h cussed above contained hiWh/+ dermis as well as hi? /+ melanocytes, it might be argued that the phenotypic effects attributed to the intrinsic gene action of hiWh within the hiWh/+ melanocytes, may actually be due to the gene action of hiWh within the dermis, in a fashion analogous to the dermal site of gene action of the alleles of the agouti 186 h locus. However, since the hi? /hiwh dermis had no adverse effects on the pigmentation phenotype of the +/+ melanocytes h in the (+/+ // hiw /hiyh) recombination grafts, it seems unlikely that hiWh in the heterozygous condition, i,h. in the hiWhl+ dermis, would have a stronger effect on melano- cyte function, than would hiWh in the homozygous condition, h i.h., in the hiw /hiWh dermis. However, the available data do not rule out the possibility that hiwh may act within the hivhl+ dermis before 13 days of development. It is conceivable that hiWh may act within the pre-13 day dermis to re-direct the developmental potential of the hiWh/+ melanoblasts from a black pigmentation phenotype to a gray pigmentation phenotype. Under this hypothesis, once the hiWhl+ melano- blasts are re-directed into a gray pigmentation phenotype, the continued presence of the hiWh/+ dermis would not be necessary for the expression of the gray pigmentation phenotype by the hiyhl+ melanocytes. In order to test this hypothesis, +/+ and hiWh/+ melanoblasts would have to be exposed to +/+ and hiyhl+ dermal environments at a much earlier stage of development, i.h., before the melanoblasts begin their migration from their mid-dorsal position into the mesodermal-dermal environment. Such experimental con- ditions could be obtained by the production of hiWh/+ ++ +/+ chimeric mice. 187 Epidermis The experiments in this dissertation do not provide any direct evidence which either confirms or rejects the possibility that hiwh may act in part within the hiWh/+ epidermis to produce the gray pigmentation phenotype of the hiWh/+ mice. All of the (13 day epidermis // 13 day dermis) graft recombinations which contained hth/+ epidermis also contained hiwhl+ melanoblasts. Therefore, any action of hiWh within the hth/+ epidermis which might have affected the pigmentation phenotype could not be distinguished from the effects ofhiWh acting directly within the hiWh /+ melanocytes. The use of (11 day epidermis // 13 day dermis) graft recombinations would have been appropriate to deter- mine the role of the hiWh/+ epidermis in the production of the gray pigmentation phenotype. However, since the first series of grafts was unsuccessful (see Table 32) any graft recombinations using 11 day hiWh/+ epidermis were not made. h wh Phenotype of hiw lhi h It is clear that adult hiw /hiWh mice do not possess any functional (dopa-positive) cutaneous melanocytes. Likewise, it is apparent that the epidermis derived from h/hiWh embryos does not contain any melano- hlhiwh epidermal 13 or 14 day hi? blasts capable of differentiation, in a hiw environment, into functional melanocytes. Since the epidermis derived from 13 day E/E embryos proved to be 188 inimical to the normal melanocyte differentiation, the lack h of pigmented (E/E // hiw /hiWh) recombination grafts cannot be considered as sufficient evidence to rule out the possi- h bility that the dermis from 13 day hiw lhiWh embryos which, under the appropriate epidermal conditions, might produce functional melanocytes. h wh Site of Gene Action of hiw /hi Epidermis h wh Based on the recovery of (hi? /hi // +/+) and wh wh (hi /hi // hiWh/+) grafts which contained only unpig- mented hair, it seems clear that the epidermis derived h from the 13 day hi? /hiWh embryos adversely affects the development of melanocytes from melanoblasts derived from the dermis of 13 day +/+ or hiWh/+ embryos. Silvers and h wh E. Russell(1955) have demonstrated that neonate hi? /hi follicular epidermis will to a limited degree, support normal melanogenesis of melanocytes derived from non- hiWh/hiWh neonate skin. In their experiments, the non- ‘hiWh/hiWh melanocytes which are engaged in apparently h wh normal melanogenesis within the hiw /hi follicular environment may be comparable to the +/+ melanocytes in the h wh 16 of 30 (hiw /hi // +/+) grafts which were able to pro- ceed with melanogenesis within the generally non-permissive wh wh ' hi /hi follicular environment. Alternatively, it is conceivable that either: 1) the non-permissive nature of 189 h the hii“’r‘/_bi_:l';w epidermis derived from 13 day embryos is short wh lived and the neonate hivh[hi epidermal follicular en- vironment is completely permissive to normal melanocyte differentiation, or 2) the neonate melanoblasts are insen- h wh sitive to the adverse effects of the hiw lhi follicular environment, i.e., the hiWh/hiWh genotype in the epidermis only affects some early step in melanoblast development. According to this latter hypothesis, if melanoblasts that had already passed this developmental step were exposed to the hiWh/hiWh epidermal environment then they would be indifferent to the adverse effects of the hiWh/hiWh epidermal environment. Dermis Based on the black and gray pigmentation phenotypes of h h the (+/+ // 3h" nigh) and (awn // a“ /_M_1_“'h) recombi- nation grafts, it is apparent that the dermis derived from h the 13 day hiw lhiWh embryos does not adversely affect the differentiation of either +/+ or hiWhl+ melanoblasts which are already present in the 13 day epidermis. However, as with the hiWh/+ dermis, it is possible that hiwh may act h/hiwh dermis, prior to 13 days of develop- h/EiWh through the hiw ment, to completely block some early step inhiw melanoblast development. If this hypothesis were correct, then the presence of pigmented hairs in the (13 day high/+ // h 13 day hiw /hiWh) recombination grafts (see Table 31) 190 h wh indicates that either the 13 day hiw lhi dermis has lost its inhibitory action on melanoblast development, or that the hiWhl+ melanoblasts in the 13 day epidermis are h wh insensitive to the action of hiw /hi within the dermis. Melanocyte A variety of experiments, specifically, dopa staining of adult skin, dermal-epidermal recombination grafts, and i3 vitro epidermal culture, have demonstrated a lack of h h wh wh /§$ both functional hi? /hi melanocytes, and hi? melanoblasts capable of developing into functional melano- h wh cytes within either the skin of hiw lhi adult mice or embryos. However, in these experiments, the epidermal h wh environment in which the hi? /hi melanocytes would have developed was either hiwfiflhrh or h/h in genotype. Since it has been clearly demonstrated that both of these epidermal environments are inimical to the development of functional h melanocytes, the lack of demonstrable hiw /hiWh melano- cytes under such conditions does not prove that the h hi? /hiWh melanoblasts monot exist. However, the subnor- mality of the hiWh/+ melanocytes, in both proliferative capacity and melanosomal synthetic activity, suggests h that the hiw /hiWh melanoblasts, if they do exist at all, might be very few in number, and possess only slight melanogenic activity. If this were the case, the demon- h stration of the existence of the hiw /hiWh melanoblasts would require the growth of these melanoblasts in as 191 normal and permissive an environment as possible, g.g. h wh within hiw [hi ++ +/+ chimeric mice. In any such chimeric mice, the dermis, epidermis and melanocytes must all be identified with readily distinguishable genetic markers from both parents to allow for the unequivocal h wh demonstration of the presence or absence of the hiw /hi melanocytes. In summary, it has been demonstrated that: 1) the h epidermis derived from 13 dayhiw lhiWh embryos is inimi- cal to the development of +/+ or hiWh/+ melanoblasts de- rived from the dermis of either +/+ or hiWhl+ 13 day embryos, h/hiWh embryos is-per- h and 2) the dermis from 13 day hi? missive to the development of functional +/+ and hiw /+ melanocytes derived from melanoblasts present in the epi- dermis of 13 day +/+ or hiWh/+ embryos. Furthermore, a1- h wh though hiw lhi melanocytes have not been demonstrated to exist, neither has it been demonstrated that the hi /hiWh do not exist. Nor has the mode of gene action of hi“711 been determined by which MiWh/hiwh blocks the devel- opment of functional‘hiWh/hiyh melanocytes. It should also be noted that the data presented in this dissertation do not directly confirm or reject Mintz's hypothesis of preprogrammed clonal cell death as the cause h of the unpigmented phenotype of the hi? /hiWh mice and the white spotting phenotype of the hiyhl+ mice. However, the h wh demonstration that hiw /hi epidermis is clearly inimical to the development of functional melanocytes does 192 contradict Mintz's contention that the site of gene action of white spotting mutations in general, and hiWh in par- ticular, is exclusively intrinsic to the melanocytes. Comparison of hiWh to Other Mutations Affecting the Mouse Pigmentation Phenotype To place hiWh within the general framework of genetic control of cutaneous pigmentation, some aspects of the phenotype and the site and specificity of gene action of hiWh will be compared with the phenotype, and site and specificity of gene action of other mouse pigmentation mutations. Phenotype h/+ Five aspects of the pigmentation phenotype of hiw can be compared with the phenotypes of the pigmentation mutations discussed in the Literature Review, specifically: I) melanocyte number, 2) melanocyte morphology, 3) melano- some number and distribution, 4) melanosome morphology, and 5) melanosome melanization and tyrosinase activity. Melanocyte number: hiWh reduces the number of, but does not completely remove, the epidermal melanocytes within the non-hairy ear and tail skin of the hiWh/+ mice. Within this restricted aspect of the cutaneous pigmentation phenotype, three other genotypes besides hiWh/+ i.g., hy/_, 3/3, and fl/+, 193 are known to have similar effects on the epidermal melano- cytes in the ear or tail skin. However, in other related aspects of the epidermal melanocyte phenotype, 2.3., fol- licular melanocyte function, none of these three genotypes h/+. Ay/- and 3/3 direct the show any similarity to hiw follicular melanocytes to synthesize primarily phaeomelanin, while hfl+<flrects the follicular melanocytes to synthesize apparently normal quantities of eumelanin. hiWh/+ also reduces the number of melanocytes within the hair follicle. This aspect of the pigmentation pheno- type has been examined in only two other mutant genotypes, hy/_ and i/i. In both of these cases, these other genotypes do not seem to alter the number of melanocytes within the hair follicle. ‘hiWh/hiWh mice have no cutaneous melanocytes. Mutant alleles of two other loci, dominant spotting and steel, al- so cause a total absence of cutaneous melanocytes. How- ever, this absence of cutaneous melanocytes caused by these three loci is only a similarity in phenotype since, as will be discussed in a later section, the specificity of gene action of the mutant alleles of dominant spotting and steel are clearly different from that of the hiWh/hiWh genotype. Melanocyte morphology: On a 2/2 background, the hiWh/+ cutaneous melanocytes have a nucleopetal morphology. On a h/h background, all of the cutaneous i/i;hiWh/+ melanocytes have a nucleofugal 194 morphology. However, the i/i;hiWh/+ follicular melanocytes are also smaller in size than the'h/i;+/+ follicular melano- h cytes. The nucleopetal morphology of the h/i;hiy /+ melanocytes is caused by the i/i genotype, but the smaller h size of the follicular melanocytes is due to the hiw /+ genotype. Only two of the genotypes discussed in the Literature lt/ Review, i/i and h have been reported to alter melano- cyte morphology. As noted above, i/i causes all cutaneous melanocytes to assume a nucleopetal morphology. In filt/ mice, the anagen VI follicular melanocytes assume a nucleo- petal morphology and become lodged in the growing hair shaft so that the late anagen VI hair follicle is devoid wh of melanogenic melanocytes. Clearly, the effects of hi /+ on melanocyte morphology, reduced size, is qualitatively quite different from the effects of either i/i or hltl_ Melanosome number and distribution: In h/h;hiWh/+ mice, a reduced number of melanosomes are present in the normal unclumped distribution within the h/+ medullary and cortical cells of the hair. In the g/g;hiw mice, the hair contains more than the normal quantity of phaeomelanin. Three of the loci discussed in the Liter- ature Review, dilute, the b-locus (hlt) and albino have been reported to affect either the number or distribution It of melanosomes within the hair. In i/i and h /_ hair, a reduced number of melanosomes are present in an irregularly 195 clumped distribution. The irregular distribution of these melanosomes is thought to be caused by the nucleopetal morphology of the follicular melanocytes. Several of the mutant alleles of the albino locus (5h, SCh, 2e and 3) do not alter the distribution, but do reduce the number of melanosomes present in the hair. 1t/_ differ from hiWh/+ in their Clearly h/i and h effects on the distribution and quantity of melanosomes within the hair. Within the restricted phenotype of melano- some number and distribution, the albino locus alleles and hiWh/+ have qualitatively similar effects. However, as will be discussed in the section on melanization and tyros- inase activity, these two loci are fundamentally different in their effects on a closely related aspect of the melano- somal phenotype, i.g., the presence and function of the tyrosinase isozymes. Melanosome morphology: hiWh affects the morphology of the eumelanosomes syn- thesized within the follicular melanocytes such that the hivhl+ melanosomes are both smaller and more spherical than the +/+ eumelanosomes. Among the loci discussed in the Literature Review, four loci have been reported to alter the morphology of the follicular eumelanosomes, i.g., beige, pink eye dilute, the b-locus and the c-locus. The hg/hg_genotype causes the fusion of the membranes of the developing melanosomes resulting in the production of large, 196 irregular macromelanosomes. The 2/2 genotype, on an 2(2 background, causes the synthesis of small "shred-like" eumelanosomes. Clearly both of these genotypes affect h melanosome morphology in ways quite different from hiw /+. l ch/gch Both the b-locus (h/h and h t[_) and the c-locus (g and he/ge) direct the synthesis of eumelanosomes which are smaller and more spherical than normal eumelanosomes. Within this aspect of the pigmentation phenotype, the effects of these mutants are similar to the effects of hiWh/+. However, as will be described below, a related aspect of the melanosomal phenotype, i.g., the presence and function of the tyrosinase isozymes, is differentially affected by these three loci. Melanization and tyrosinase activity: The eumelanosomes from hiWh/+ follicular melanocytes have a uniform pattern of melanization. .hiWh/+ also has no effect on the electrophoretic mobility of the tyrosinase isozymes. In contrast, several of the mutations which decrease the intensity of eumelanin pigmentation have con- siderable effects on the level and pattern of melanization and tyrosinase activity. The mutant genotypes R/p, pg/Ri, hg/_ and the mutant alleles of the albino locus, all de- crease the levels of melanization of the individual melano- somes. The first three loci appear to cause their effects on melanization by a lack of substrate or ionic cofactors necessary for melanization, while the mutant alleles of the 197 albino locus presumably cause changes in the primary struc- ture of the tyrosinase molecule which alters its enzymatic activity. Furthermore, mutant alleles of the b-locus and c-locus alter either, or both, the electrophoretic mobility and the dopa oxidase activity of tyrosinase isozymes. In summary, although discrete portions of the hiWh pigmentation phenotype are qualitatively similar to com- parable aspects of pigmentation phenotypes of other muta- tions in the mouse, the overall pigmentation phenotype of hiWh/+ does not closely resemble the pigmentation phenotype of any other pigmentation mutation in the mouse. Specificity 2i gene action As previously noted in the section on melanocyte h wh number, hiw /hi mice do not have any cutaneous melano- cytes. A nullo phenotype does not directly offer many h/HEWh points for phenotypic comparisons. Therefore, hi? will be compared with other genotypes which cause white spotting by a comparison of the specificities and sites of gene action. h wh In addition to its effects on melanocytes, hiw /hi also affects the development of other neural crest deriva- tives, g,g., adrenal medulla, dorsal root spinal ganglia, and the meninges of the central nervous system. Further- more, hiWh also disrupts normal eye and ear morphogenesis. As discussed‘in the Literature Review, it is possible that these latter two effects may also be due to defects in 198 neural crest-derived tissues. Two other white spotting mutations discussed in the Literature Review, splotch and piebald, share some similar- ities in the specificity of gene action with gth. §£j§2 disrupts the normal development of the dorsal root spinal ganglia and piebald decreases the number of neurons in the enteric ganglia, i.g., mutant alleles of both loci affect the development of portions of the nervous system derived from the neural crest. Conversely, mutant alleles of two other white spotting loci, dominant spotting and steel, have a profoundly differ- ent specificity of gene action from gth. Mutant alleles from both the dominant spotting and steel loci cause macrocytic anemia and sterility. Furthermore, h/h skin does not contain mast cells (Kitamura g; _i., 1978). By wh wh comparison, hi [hi mice do not have macrocytic anemia, are fertile, and appear to have mast cells in the dermis. Site _i gene action Based on the behavior of hiyhl+ melanocytes, it is clear that hiWh acts intrinsic to the post-13 day melano- h wh blasts. Likewise, based on the behavior ofhiw /hi epi- dermis, it is clear that hiwh also acts through the post-13 day epidermis to affect follicular melanocyte differentia- tion, i.e., hiWh gene action is extrinsic to the post-13 day melanoblasts. Among the white spotting mutations for which the site of gene action has been determined, two mutations 199 show strong similarities as to the pattern of gene action of hiyh. The site of gene action of both h/h and hi/hid is in part extrinsic to the post-13 day melanoblasts and appears to be within the epidermis. Furthermore, as with h hiWh/hiw , neither h/h or §i/§_d appear to act within the post-13 day dermis to affect epidermal melanocyte differ- entiation. Furthermore, hiWh, h and hi show similar pat- termsof extrinsic gene activity in the post-13 day skin with respect to follicular melanocyte differentiation. Comparison of hi“h and hh As previously discussed in the Literature Review, five classes of data can be utilized to establish or refute gene homology between species. These classes of data in- volve the comparisons of: 1) nucleic acid sequences of the loci or their RNA, 2) the amino acid sequence of the polypeptide gene products of the loci, 3) the site, spe- cificity and mode of gene action, 4) the phenotypic effects of the mutant alleles of the loci, and 5) the epistatic interactions between loci within each species. In this comparison of hiWh and hh, information is available from the last three classes of data, i.g., site and specificity of gene action, phenotypic description, and epistatic interactions. Specifically, the points of comparison are: la) the pigmentation phenotype, 1b) the site of gene action with respect to the follicular melanocytes, 2) the development of the eye, 3) the development of the ear, and 200 4) physiology and anatomy of the pituitary and related target organs. The following sections will briefly review the relevant data. The final section will consider the question of homology between hiWh and hh. Pigmentation phenotype The fur pigmentation of the g/g:gh[gh hamster is a uniform yellow. The fur pigmentation of the g/g;hh/!h hamster is generally a uniform white. On some genetic backgrounds, this genotype occasionally has patches of ex— tremely pale yellow fur. The fur pigmentation of the g/E;hh/hh hamster is always a completely uniform white. The follicles of the unpigmented fur of both the g/g:hh/gh and g/g5hh/hh hamsters are completely devoid of melano- cytes. The fur pigmentation ofthe g/g;+/+ mouse is a dusky yellow, in which the distal tips of otherwise yellow hairs contain small quantities of eumelanin. The fur pigmenta- tion of the g/g;hiwh/+ mouse is a uniform yellow without any eumelanin. The g/£:hiWh/+ fur contains approximately 20% more phaeomelanin than does the g/£;+/+ fur. The fur pigmentation of the g/g;hiWh/hiWh mouse is a uniform white. The follicles of the g/g:hiWh/hiWh fur are devoid of melanocytes. On the 3/3 background, hh behaves, with respect to fur pigmentation, as a simple dominant allele which is epistatic to 3/3 in both the heterozygous (Eh/3h) and 201 homozygous (Eh/Eh) genotypes. On a comparable s/g back- ground, hiWh behaves, with respect to fur pigmentation, as a partial dominant allele with a slight heterosis, h which is epistatic to g/g only in the homozygous (hiWh/fliv ) Histologically, the unpigmented follicles of both hh/_ h wh and‘hiw /hi are identical in their lack of melanocytes. ,Site _i gene action 22 follicular melanocytes Based on the results of the hamster dermal-epidermal recombination grafts: 1) hh/_ epidermis is inimical to the development of functional follicular melanocytes, and 2) hh/_ dermis does not adversely affect the development of functional follicular melanocytes. Based on the results of the mouse dermal-epidermal recombination grafts, the h/hiWh epidermis and dermis on the develop- effects ofhiw ment of functional follicular melanocytes are completely parallel to those described for the hh/_ epidermis and dermis, i.e., a hostile epidermis and a permissive dermis. Eye development and pigmentation In the g/g;hh/!h hamster, eye morphogenesis is essen- tially normal with the exception that the eye is slightly microphthalmic and the choroidal pigmentation is incomplete and patchy in its distribution. In the g/E;hh/hh hamster, the developing embryonic eye undergoes a progressive degen- eration such that the orbital contents of the adult eye are primarily hardarian gland with disorganized remnants 202 of retina and lens (Asher, 1979; Yoon, 1975). Furthermore, in the developing g/g;hh/hh eye, the presumptive retinal epithelium fails to develop any pigmentation except in the pupillary margin of the iris adjacent to the lens vesicle. In the g/g;hiwh/+ mouse, eye morphogenesis is also normal with the exception that the choroidal pigmentation is usually either patchy in its distribution or completely h/gth mouse, the eyes are absent. In the adult g/g:hiw present but severely microphthalmic. The uveal tract is totally unpigmented and the retinal epithelium is also unpigmented except in the pupillary margin of the iris adjacent to the lens. Ear development h wh /M1 mice Both the g/g;hh/hh hamsters and g/g;hiw are deaf. The etiology of the deafness in the hamster has not been examined. In the mouse, the organ of Corti and particularly the stria vascularis are abnormal in structure Physiology The growth rate of juvenile male hh/hh hamsters is significantly less than that of normal KE/ZE littermates. The weight of adult hh/hh males is less than that of Eh/gh males. The basal metabolic rate of hh/hh hamsters is elevated. The adult hh/hh males may have a short period of fertility, but always become permanently sterile. 203 Sperm development in these sterile males is blocked at the FSH and LH dependent transition from spermatid to spermato- zoa (James, 1979). Eh[Eh females are usually but not al- ways sterile. The growth rates and adult body weights of male wh/ h hi hiw mice are not different from growth rates and body weights of +/+ male mice. The food and water con- sumption, and the rectal temperatures also do not differ h wh between male +/+ andhiw /hi mice. Both sexes of hiWh/hiWh mice have apparently normal fertility. Microscopic anatomy 3i adrenal and pituitary In hh/hh hamsters, the size of the adrenal cortex, specifically, the ACTH dependent fasciculata+reticularis zone is decreased in volume. The hh/hh adrenal medulla is normal in size. In the sex zone of the adenohypophysis of the male hh/hh hamsters, the cell number is decreased by approximately 40%. The remaining cells have excessive numbers of gap and occluding junctions and many of the cells also contain clusters of abnormal cilia. In the hiWh/hiWh mice, the adrenal cortex is normal in size, but the adrenal medulla is reduced in size and cell number. The sex zone of the adenohypophysis of male ‘hiWh[hiWh mice contains normal numbers of cells. 204 Homology The points of similarity between hh and hiWh include: 1) the lack of follicular melanocytes in the unpigmented follicles, 2) the site of gene action on the development of follicular melanocytes, 3) the presence and distribution of choroidal and retinal epithelial pigmentation, and 4) h/hiWh animals. the deafness of both the hh/hh and hiw Considering these similarities, first, the lack of fol- licular melanocytes in the unpigmented follicles confirms that both hh and hiwh are bona fide white spotting mutations. Second, all white spotting mutations in the mouse which have been examined with dermal-epidermal recombination grafts, show the same pattern of gene action on follicular melanocytes as does Eh. It is a possibility that this pattern of gene action is a general characteristic of all white spotting mutations and is not therefore an indi- cation of homology between Eh and hiWh. Third, the parallel between the choroidal and retinal epithelial pigmentation phenotypes of the two mutations appears to be a valid point of similarity between hh and gth. It should be noted, however, that the presence of the iridial pigment ring in the eyes of both homozygotes is possibly not due to the action of either hh or hiWh, but may rather be due to the presence of a lens in the eyes of the homozygous mutant animals. Fourth, the deafness itself is clearly a similarity between hh and hiWh. However, lacking 205 information as to the cause of the deafness in the Eh/Eh hamsters, this phenotypic similarity should not be weighed too heavily. The points of difference between Eh and hiWh include: 1) the pigmentation phenotype of the heterozygotes, Eh/Eh and hiWh/+, and the effects of Eh andhiWh on phaeomelano- cytes, 2) the effects of Eh and hiwh on eye development, 3) the differences in growth rate, adult body weight, metabolic rates and fertility, and 4) the microscopic anatomy of the adrenal and pituitary glands. Considering these differences, first, on an E/E background, Eh behaves as a simple dominant allele, while on a similar E/E background, hiWh behaves as a partial dominant with a heterotic effect on phaeomelanin production. It might be argued that the differences in dominance between these two mutations may be a result of the differences in genetic background between the two species. However, the dele- terious effect of Eh on phaeomelanocytes, evident in the h/h;Eh/Eh fur, can also be seen in the white ventral fur of the h/h;Eh/Eh hamsters. The Eh/Eh genotype appears to be hostile to melanocytes destined to produce phaeomelanin, while the hiWhl+ genotype appears to potentiate the pig- ment production of phaeomelanocytes. Second, the differences between the Eh/Eh and hiWh/hiWh eye development appear to be a matter of degree rather than kind. Until the mechanism of the degenerative anophthalmia of the Eh/Eh eyes and microphthalmia of the 206 Mi [hivh eyes are better understood, these phenotypic differences should not be weighed too heavily. Third, the normal endocrine physiology of the hamster. is highly dependent on the presence of functional eyes. It has been demonstrated that experimentally blinded male Eh[Eh hamsters become temporarily sterile, and increase in body weight (Asher, 1979; Reiter, 1969). It has also been shown that experimentally blinded Eh/Eh males have an elevated basal metabolic rate, and reduced adrenal gland weight (see review: Reiter and Fraschino, 1969). There— fore, it is possible that some of the physiological effects of Eh may be subordinate pleiotropic effects of the Eh- induced anophthalmia. However, the known pleiotropic effects of experimentally caused anophthalmia do not include permanent sterility, decreased growth rates or wh depressed adult body weights. Since the male hi h /_M_1w mice have normal growth rates, normal adult body weight and are fertile, it appears that at least these latter three aspects of the physiological phenotype represent fundamental differences between Eh and hiWh. Fourth, as previously mentioned, the decreased size of the adrenal cortex of the Eh/Eh hamsters may be a sub- ordinate pleiotropic effect of Eh/Eh-induced anophthalmia. However, the differences in cell number of the adrenal h/hiWh animals appears to be a significant difference in phenotype between Eh and hiWh. medulla of the Ell/Eh and 91;" Fifth, and finally, the reduced cell number and 207 abnormal cell morphology of the Eh/Eh adenohypophysis do not appear to be subordinate pleiotropic effects of the Eh/Eh-induced anophthalmia. The adenohypophysis of ex- perimentally blinded Eh/Eh hamsters show a slight inorease in cell number (Dekker, 1967), and do not have any of the abnormal cell morphology seen in the Eh/Eh adenohypophyses (James, 1979). It appears therefore, that the abnormal pituitary cell number and cell morphology are character- istic of the Eh[Eh phenotype, and the former, at least is h wh not part of the hi? /hi phenotype. In summary, notwithstanding the similarities in some aspects of the phenotypes of the Eh/_ and hiWh /_ animals, and those differences attributable to species specific genetic background or physiology, there remain several clear differences between the phenotypes of the two muta- tions. Specifically, these two mutations differ in their effects on: 1) phaeomelanocyte differentiation and function, 2) growth rate, adult body weight and fertility, 3) adre- nal medullary cell number, and 4) pituitary cell number. Therefore, based on this author's interpretation of the available phenotypic data, Eh and hiWh do not appear to be mutations at homologous loci. SUMMARY, CONCLUSIONS AND FURTHER RESEARCH The research contained in this dissertation is summarized in the following paragraphs. Within the pigmentary phenotype, hivhl+ causes the following effects: 1) a reduction in the number of melanocytes in the hair follicles, and ear and tail epidermis, 2) a reduction in the number of melanosomes produced by both h/h5hiWh/+ and h/hghiWh/+ follicular melanocytes, 3) a reduction in the size and a change in the shape of the follicular melanosomes, 4) an increase in the quantity of phaeomelanin deposited in the h/h;hiYh/+ hair, and 5) no change in the electrophoretic mobility of the soluble follicular tyrosinase isozymes. Within the non-pigmentary phenotype,hiWh has no detectable effects on: 1) the red cell diameter, reticulocyte count or leukocyte differential count, 2) the growth or metabolic activity, 3) the mast cell density within the ear dermis, or 5) the cell density within the sex zone of the adenohypophysis. The following results were obtained from the mouse dermal-epidermal recombination grafts. 208 209 h 1. The reduced pigmentation of the fur of the hi? /+ mouse is probably not caused by a systemic defect in the availability of substrates or cofactors necessary for normal melanogenesis in the hiWh/+ follicular melanocytes. M1Wh wh/ 2. The dermis from 13 day hi embryos is permissive to the differentiation of +/+ and hiWh/+ melanoblasts present in the 13 day epidermis. hIMiWh 3. a) Epidermis from either 13 dayhiw or 13 day E/E embryos.adversely affects the normal development of melano- blasts derived from either +/+ or hiyhl+ 13 day dermis. Wh/h$Wh b) The deleterious effects of hi epidermis from 13 day embryos (on +/+ melanocyte development) are more severe than the effects of epidermis from 13 day E/E embryos. wh wh 4. The hi /+ melanoblasts from 13 day hi /+ dermis are subnormal in their ability to develop into functional h/Miwh melanocytes within either a hi? __ h or E/E epidermis. /MiWh embryos does not 5. The epidermis from 13 day hiw contain melanoblasts capable of differentiation into fucntional melanocytes, even when combined with the dermis of 13 day E/E embryos. 6. Dermal-epidermal grafts containing epidermis from 11 day embryos and dermis from either 11 or 13 day embryos (with their associated melanoblasts) failed to consistently develop functional follicular melanocytes. 7. The testes do not appear to be an immunologically priviliged site for the development of dermal-epidermal 210 recombination grafts that differ in their hf; genotype from that of the host animal. In the hamster, the following observations were made. 1. The unpigmented regions of the ear and dorsal skin of ‘§[5;Eh/_ hamsters do not contain any identifiable melano- cytes. 2. Within the eye, the choroid of the E/EBKh/Eh hamster has patchy and irregular pigmentation, but the retinal epithelial pigmentation appears normal. From the results of the hamster dermal-epidermal recombination grafts, the following observations were made. 1. Epidermis from g/g:Eh/ embryos adversely affects the I: development of g/ h_h follicular and dermal melanocytes. / 2. Dermis from g m W/ ;Eh/_ embryos does not affect the h/w development of E/ lo :: hmelanocytes within the g/g;wh/ h epidermis. 3. Dermis from 2/23 I? / embryos can, with less efficiency than g/g:gh[gh dermis, support the development of melano- cytes within the dermis. 4. The skin of g/g;Eh/Eh embryos contains melanoblasts capable of development into eumelanogenic dermal melano- cytes. 5. Either the melanoblasts from embryonic g;g/flh/!fl epidermis are capable of retrograde migration into the dermis, or the h/g:Eh[_ embryonic dermis contains melano- blasts capable of development into eumelanogenic melano- cytes. 211 From the results summarized above, the following conclusions have been made. 1. The gray phenotype of the hiWh/+ mouse fur is directly caused by a reduction in the size and number of the melano- somes in the hair. These effects are in turn caused by a reduction in the number and function of the hiWh/+ fol- licular melanocytes. The functional defect in the hiWh/+ melanocytes appears to be intrinsic to the post-13 day melanoblasts. h wh /Mi fur is caused by a 2. The white phenotype of the hiw lack of functional follicular melanocytes that is in turn wh/ h epidermal “on“ caused by either, or both, a hostile hi hiw environment, or an intrinsic defect within the hiw melanoblasts. 3. The mouse mutation, hiWh, and the hamster mutation, Eh, do not appear to be mutant alleles of homologous loci. This conclusion is based on the differential effects of these two mutations on: 1) the survival and function of their respective g/g follicular phaeomelanocytes, 2) the growth and reproductive physiology, and 3) the cell density within the adrenal medulla and the sex zone of the adenohypophysis. Further research on hiWh could profitably be focused on the mode of gene action of hiWh. Specifically, the following problems could be examined. 1. The mode of gene action whereby hiWh/+ reduces the number and function of the epidermal melanocytes. 212 wh/ wh 2. The mode of gene action whereby hi hi embryonic epidermis adversely affects the differentiation of +/+ and hivhl+ melanoblasts. h/MiWh that results in the 3. The mode of gene.action of hiw wh wh total lack of functional hi /hi melanocytes. The solution of these latter two problems will depend, in part, on the resolution of a fourth problem, the identification of an intrinsic cell marker by which melanoblasts can be reliably distinguished from the surrounding mesenchyme- in which they migrate and differentiate. APPENDICES APPENDIX A Analysis of Pigment Volume Data of E. Russell (1948) E. Russell (1948; Table 2, col. 6) computes values pro- portional to the mean size of the pigmented bodies in dilute (h/h) hair according to the following weighted average formula: vave = (vg)<%c> . where Vave - an index of the average volume of a pigmented body, where the term "pigmented body" refers to either a clump of melanosomes, or a single melanosome; Vg - an index of the volume of a single melanosome; Vc - an index of the volume of a pigment clump that is itself composed of an un- specified number of individual melanosomes; Zg - the per- centage of total pigmented bodies which are single melano- somes; Zc = the percentage of total pigmented bodies which are clumps. Russell (1948; Table 3, col. 5) further computes an index of the total pigment volume within a hair according to the following formula: VT = (Vave)(c + M) , where V = an index of the total volume of pigment in a T single hair; C a an index of the total number of melanosomes in the cortex of the hair; M = an index of the total number 213 214 of melanosomes in the medulla of the hair. The calculation of VT is correct only if (C + M) is equal to the number of pigmented bodies that were used to estimate Va , 1.2., in a V2 _ genotype with an unclumped distribution of melanosomes, h.g., h/h;h/h5h/h. However, this method of calculation for VT is an overestimate of the total pigment volume in the hair when (C + M) is less than the number of pigmented bodies, $a£°: in a genotype with a clumped distribution of melanosomes, s-su yea/Bys- APPENDIX B Elliptical Geometry and Eccentricity The shape of a melanosome can be approximated by a prolate spheroid; a solid of revolution formed by the rota- tion of an ellipse around its major axis (see Figure 9a). Therefore, any longitudinal section through a melanosome which includes its major axis, 3.5., an optical section as seen in the electron micrographs of the dispersed melano- somes, would be an ellipse whose axial dimensions correspond to the axial dimensions of the entire melanosome. The equation of an ellipse with focal length c (see Figure 9b) is: xz/a2 + y2/b2 = 1 , where a = the length of the semi-major axis, b = the length of the semi-minor axis, and: c = (az—bz)l5 . (1) The roundness of an ellipse can be expressed by a uni- variate metric, eccentricity (e), which is a measure of the relative lengths of the interfocal distance and the major axis, i.h., e . c/a . (2) Substituting (1) into (2): 215 Figure 9a. Prolate Spheroid Prolate spheroid formed by rotation of ellipse around its major axis ' err—9 <—a——> Figure 9b. Ellipse Ellipse with focal length c, semi—major axis a and semi-minor axis b ‘ 217 e = (a2 -b2)%/a , the eccentricity of an ellipse can be expressed in terms of the semi-major and semi-minor axes. When the ellipse is circular, a = b, and e - 0. As the ellipse becomes more elongate, a > b, and 0 < e < 1. When the ellipse becomes very elongate, a >> b, the eccentricity of the ellipse approaches a value of 1. APPENDIX C Formulations and Procedures Buffers and salt solution 1. Sorensen's Buffer (Sober, 1970): KHZPO4 1.85g NaZHPo4 7.29g adjust to pH 7.4 with NaOH H20 (dist.) up to 1000m1 2. 0.1M phosphate buffer: NaH2P04-H2 adjust to pH 6.8 with NaOH 0 13.80g H O (dist.) up to 1000m1 2 3. Phosphate buffered saline: NaCl 8.775g KHZPo4 6.125g sodium azide 0.05g adjust to pH 7.5 with NaOH H20 (dist.) up to 1000ml 218 219 4. Hank's balanced salt solution (Hanks and Wallace, 1949): NaCl 8.00g KCl 0.40g NaZHPOA'ZHZO 0.06g KHZPO4 0.063 MgSOa°7H201 0.10g CaCl2 (anhyd.) 0.14g Glucose 1.00g MgClz-nzo1 0.10g NaHCO3 0.35g Phenol Red 0.10g H20 (dist.) up to 1000ml 1modification of original formulation by National Institutes of Health as listed in Grand Island Biological Company 1979 catalogue- Formulations section Fixatives 1. Bouin's fluid (Clark, 1973): Picric acid (sat. aq.) 75.0ml Formalin 25.0ml Glacial acetic acid 5.0ml 2. Vandegrift's fixative (Vandegrift, 1942): Ethanol (95%) 80.0ml Formalin 12.0ml Glacial acetic acid 4.5ml Picric acid 4.0g Mercuric chloride 0.2g Urea 0.5g Mix together in listed order the ethanol, formalin and glacial acetic acid. Add and dissolve in the listed order the picric acid, mercuric chloride and urea. 3. Neutral buffered formalin (Clark, 1973): Formalin 100.0ml NazHPOA (anhyd.) 6.5g NaHzPOA-HZO 4.0g H20 (dist.) 900.0ml Stains 1. Harris's hematoxylin (Clark, 1973): Hematoxylin 5.0g (C.I. 75290) Ethanol (100%) 50.0ml Ammonium alum 100.0g H20 (dist.) 1000.0ml HgO 2.5g Glacial acetic acid 4.0ml Dissolve the hematoxylin in the ethanol. Dissolve the alum in the water by heating. Mix the two solutions and bring to a boil as rapidly as possible. Add the HgO. When the solution becomes dark purple, remove it from the heat and cool rapidly. Before use add glacial acetic acid and filter. 2. Alcoholic eosin: Eosin Y (C.I. 45380) 0.2g Ethanol (95%) 100.0ml 221 3. Toluidine Blue 0: Toluidine Blue 0 0.23g (C.I. 52040) H20 (dist.) 1000.0ml To stain the vaginal smears: 1) air dry the smear, 2) dehydrate in 95% ethanol, 3) stain in Toluidine Blue 0 1-2min, 4) rinse in two changes of tap water. 4. New Methylene Blue (Brecher, 1949): New Methylene Blue 0.5g Potassium oxalate 1.6g H20 (dist.) 100.0ml To stain reticulocytes, two parts blood are well mixed with one part stain, allowed to stand 10-15min. The blood- stain mixture is placed on a glass slide, smeared and air dried. 5. Wright's stain: This stain was purchased premixed from Harleco. To stain the blood: 1) Smear a drop of blood on a glass slide and allow to air dry. 2) Add a small quantity of stain and let stand for one minute. 3) Without removing the stain, add 2.0-2.5ml distilled water to the slide and let stand 3-4min. 4) Rinse the slide with distilled water and air dry. Miscellaneous Formulations 1. Depilatory wax (modification of Pusey, 1926): Rosin 50.0g Beeswax 50.0g Melt the rosin over a low heat, add beeswax and stir until completely mixed. Remove from heat, let cool slightly and pour into mold. 222 2. Formvar-coated grids: Formvar 0.5g Ethylene dichloride 100.0ml Dip a clean slide into Formvar solution and allow the slide to air dry. Free the film from the edge of the slide and float the film onto the surface of a water-filled dish. Place the grids on the floating film. Remove film with grids from the surface of the water and allow to dry. 3. Acrylamide gels (modified from Clarke, 1964): Make: 1) gel buffer TRIS 6.055g Glycine 30.027g H20 (dist.) up to 1000ml 2) ammonium persulfate solution (AP) Ammonium persulfate 56.0mg H20 (dist.) 4.0ml 3) acrylamide monomer solution (AM) Acrylamide 45.0g N,N'-methylene 1.0g bisacrylamide 0 H20 (dist.) up to 150.0ml To make the gels, add in order: H20 (dist.) 11.5m1 Gel buffer 2.5ml AM 5.0ml AP 1.0ml N4-tetramethyl- 7.0ul ethylenediamine 223 Mix thoroughly, de-gas and quickly transfer to gel tubes. Overlay the unpolymerized solution with 1-2cm of distilled water. Let the tubes set undisturbed for at least 30min to allow acrylamide polymerization. 4. Avertin (Mayer, 1977b): 2,2,2-tribromoethanol 1.0g tert-amyl alcohol 1.24ml Dissolve the tribromoethanol in the anyl alcohol. Add 1.0ml of this solution to 80.0ml sterile, distilled water. Heat gently and stir until the alcohol is completely mixed. For 2X-avertin, decrease the distilled water to 40.0ml. 5. 1.0% trypsin Trypsin 100.0mg Hanks' BSS 100.0ml Add the trypsin slowly to the Hank's 388. After the trypsin is dissolved, adjust the pH with sterile 7.5% sodium bicarbonate until the phenol red is red-orange. Pre-filter the trypsin solution through Whatman GF/C filters to remove large particulate matter. Sterilize by filtration through Millepore HA filters and freeze in 5.0m1 aliquots until use. Before use, centrifuge the trypsin solution, 10min, 3,000 xg to remove sediment. 6. Agar plates 2X-agar Agar 0.30g (bacteriological grade) H20 (sterile, dist.) 20.0ml Agitate the mixture to disperse the agar and autoclave 15-20min to dissolve the agar. 2X-culture media Eagle's MEM - 10X 2.0ml (no glutamine or bicarbonate) 224 H20 (sterile, dist.) 5.0ml Antibiotic-antimycotic 0.2m1 (100x Grand Island Biological Co.) Glutamine (200mM) 0.2ml Sodium bicarbonate 0.2ml (7.5%) Combine in order and add the sodium bicarbonate until the phenol red is pink-orange. Remove 2X-agar from autoclave and allow to cool slightly. Mix the 2X-agar and the 2X- culture media and pour into 15 x 60mm Petri dishes (8-10ml/ dish). BIBLIOGRAPHY BIBLIOGRAPHY deAberle, 8.1927. A study of hereditary anemia in mice. Amer. J. Anat. £2: 219-247. Abramowitz, J., W.Chavin. 1978. ih vitro effects of hor— monal stimuli upon tyrosinase and peroxidase activities in murine melanomas. Biochem. Biophys. Res. Comm. hi: 1067-1073. Adelmann, H.B. 1925. The development of the neural folds and cranial ganglia of the rat. J. Comp. Neurol. 22: 19-172. Adler, 8.3., and F.E. Trobaugh, Jr. 1978. Pluripotent (CFU-S) and granulocyte-committed (CFU-C) stem cells in intact and 89Sr marrow-ablated‘gi/gifl mice. Cell Tissue Kinet. ii: 555-566. Allen, B. 1922. The oestrous cycle in the mouse. Amer. J. Anat. 29: 297-371. Andrews, E.J., W.J. White and L.P. Bullock. 1975. Spon- taneous aortic aneurysms in blotchy mice. Amer. J. Pathol. 1h: 199-210. Asher, J.H., Jr. 1979. personal communication Attfield, M. 1951. Inherited macrocytic anemias in the house mouse III. Red blood cell diameters. J. Genet. 2h: 250-263. Auerbach, R. 1954. Analysis of the developmental effects of a lethal mutation in the house mouse. J. Exp. 2001. 127: 305-329. Baker, B.L. and D.S. Gross. 1978. Cytology and distribu- tion of secretory cell types in the mouse hypophysis as demonstrated with immunocytochemistry. Amer. J. Anat. iii: 193-216. Bartelmez, G.W. 1922. The origin of the otic and optic primordia in man. J. Comp. Neurol. ii: 201-232. 225 226 Batten, E.H. 1958. The origin of the acoustic ganglion in the sheep. J. Embryol. Exp. Morphol. h: 597-615. Bennett, D. 1956. Developmental analysis of a mutation with pleiotrOpic effects in the mouse. J. Morphol. Eh: 199-233. Bennett, M., G. Cudkowicz, R.A. Foster and D. Metcalf. 1968. Hematopoietic progenitor cells of E anemic mice studied .EB vivo and ih vitro. J. Cell. Physiol. 1i: 211-226. Bergsma, D.R. and K.S. Brown. 1971. White fur, blue eyes, and deafness in the domestic cat. J. Hered. hi: 171- 185. Berman, M.B. and L.J. Pierro. 1969. Lens detachment and choroid fissure closure in the embryonic mouse eye. Amer. 2001. 2:614. Bernstein, S.E. and E.S. Russell. 1959. Implantation of normal blood-forming tissue in genetically anemic mice, without x-irradiation of host. Proc. Soc. Exp. Biol. Med. igi: 769-773. Bielschowsky, M. and G.C. Schofield. 1962. Studies on megacolon in piebald mice. Aust. J. Exp. Biol. Med. Sci. 39: 395-404. Blois, M.S. 1969. Recent developments in the physics and chemistry of melanins. in The Biolggical Effects hi Ultraviolet Radiation with Emphasis 22.£E2 Skin, F. Urbach, ed., pp. 299-304, Permagon Press, New York. Blois, M.S., A.B. Zahlan and J.E. Maling. 1964. Electron spin resonance studies on melanin. Biophys. J. h: 471- 490. Bloom, G. and J. Kelly. 1960. The copper phthalocyanin dye "astrablau" and its staining properties, especially the staining of mast cells. Histochemie i: 48-57. Brandt, E.J., R.W. Elliot and R.T. Swank. 1975. Defective lysosomal enzyme secretion in kidneys of Chediak— Higashi (beige) mice. J. Cell Biol. hl: 774-788. Breathnach, A.S. 1969. Normal and abnormal melanin pig- mentation of the skin. in Pigments ih Pathology, M. Wolman, ed., pp. 353-394, Academic Press, New York. Brecher, G. 1949. New methylene blue as a reticulocyte stain. Amer. J. Clin Pathol. i2:895—896. 227 Brihaye-VanGeertruyden, M. 1963. Contribution a l'etude experimentale de l'origine du melanoblaste choroidien chez le Mammifere. Arch. Biol. 15: 1-49. Bronner, M.B. and A.M. Cohen. 1979. Migratory patterns of cloned neural crest melanocytes injected into host chicken embryos. Proc. Nat. Acad. Sci. USA lg: 1843- 1847. Brumbaugh, J., R. Bowers and K. Lee. 1973. Histochemical evidence that peroxidase does not affect melanin for- mation in feather melanocytes. Yale J. Biol. Med. Eh: 523-534. Bulfield, G. 1974. Private communication. Mouse News Letter 50: 35. Bu'Lock, J.D. and J. Harley-Mason. 1951. Melanin and its precursors Part II: Model experiments on the reac- tions between quinones and indoles, and consideration of a possible structure for the melanin polymer. J. Chem Soc.: 703-712. Burnett, J.B. 1971. The tyrosinases of mouse melanoma. J. Biol. Chem. 246:3079-3091. Burnett, J.B., T.J. Holstein and W.C. Quevedo, Jr. 1969. Electrophoretic variations of tyrosinase in follicular melanocytes during the hair growth cycle in mice. Burnett, J.B. and H. Seiler. 1969. Multiple forms of tyrosinase from human melanoma. J. Invest. Dermatol. ii: 199-203. Burnett, J.B., H. Seller and I.V. Brown. 1967. Separation and characterization of multiple forms of tyrosinase from mouse melanoma. Cancer Res. 31: 880-887. Burtin, P. and M.C. Gendron. 1971. A precipitation reac- tion between certain human G immunoglobulins and sul- fated galactans. Immunochemistry h: 423-429. Burtin, P. and M.-C. Sabine. 1972. Inhibition of the non- specific fixation of fluorescent globulins. Rev. Europ. Etudes Clin. et Biol. i1: 76-77. Charles, D.R. 1938. Studies on spotting patterns IV. Pattern variation and its developmental significance. Genetics 3;: 523-547. Chase, H.B. 1951. Numbers of entities inactivated by X- rays in greying of hair. Science 113: 714-716. 228 Chase, H.B. 1954. Growth of the hair. Physiol. Rev. 2i: 113-126. Chase, H.B., H. Rauch and V.W. Smith. 1951. Critical stages of hair development and pigmentation in the mouse. Physiol. Zool. 33: 1-8. Chen, Y.M. and D.V. Singh. 1974. Tyrosinase activity and isozymes in B-16 mouse melanomas. Oncology fig: 386- 392. Chiquoine, A.D. 1954. Distribution of alkaline phosphatase in the central nervous system of the mouse embryo. J. Comp. Neurol. 100: 415-435. Cizaldo, G. R. and N. H. Granholm. 1978. In vivo development of the lethal yellow (Ay/Ay) mouse embryo at 105 hours post coitum. Genetica 48. 89- 93. Clark, G. 1973. Staining procedures used by the Biological Stain Commission, 3rd ed., Williams and Wilkins, Baltimore. Clarke, J.T. 1964. Simplified "disc" (polyacrylamide gel) electrophoresis. Ann. N. Y. Acad. Sci. 121: 428-436. Cleffmann, G. 1963. Agouti pigment cells in situ and in vitro. Ann. N. Y. Acad. Sci. 100: 749- 761. Cleffmann, G. 1964. Function-specific changes in the metabolism of agouti pigment cells. Exp. Cell Res. .22: 590-600. Cohen, A.M. and I.R. Konigsberg. 1975. A clonal approach to the problem of neural crest determination. DevelOp. Biol. Eh: 262-280. Cole, R.J., T. Regan, S.L. White and E.M. Cheek. 1975. Expression of congenital defects in the haemopoietic micro-environment - erythroid and granulocyte-macrg- phage progenitor cells in prenatal 'steel" (Slj /Sl ) anemic mice. Cell Tissue Kinet. h: 479-489. Coleman, D.L. 1960. Phenylalanine hydroxylase activity in dilute and nondilute strains of mice. Arch. Biochem. Biophys. 2i: 300-306. Coleman, D.L. 1962. Effect of genic substitution on the incorporation of tyrosine into melanin of mouse skin. Arch. Biochem. Biophys. Eh: 562-568. Cotzias, C.G., P.S. Papavasiliou and S.T. Miller. 1964. Manganese in melanin. Nature 201: 1228-1229. 229 Cotzias, G.C., L.C. Tang, S.T. Miller, D. Sladic-Simic and L.S. Hurley. 1972. A mutation influencing the trans- portation of manganese, l-dopa, and l-tryptophan. Science 176: 410-412. Coulombre, J.L. and E.S. Russell. 1954. Analysis of the pleiotropism at the W-locus in the mouse: The effects of W and Wv substitufion upon postnatal development of germ cElls. J. Exp. Zool. 26' 277-296. Cruishank, C.N.D. and S.A. Harcourt. 1964. Pigment donation ih vitro. J. Invest. Dermatol. hi: 183-184. Dekker, A. 1967. Pituitary basophils of the Syrian hamster: An electron microscopic investigation. Anat. Rec. 158: 351-368. Deol, M.S. 1967. The neural crest and the acoustic ganglion J. Embryol. Exp. Morphol. i1: 533-541. Deol, M.S. 1970a. The relationship between abnormalities of pigmentation and of the inner ear. Proc. Roy. Soc. (London) A 175: 201-217. Deol, M.S. 1970b. The origin of the acoustic ganglion and effects of the gene dominant spotting (EV) in the mouse. J. Embryol. Exp. Morphol. Eh: 773-784. Deol, M.S. 1971. Spotting genes and internal pigmentation patterns in the mouse. J. Embryol. Exp. Morphol. Eh: Deol, M.S. 1973. The role of the tissue environment in the expression of spotting genes in the mouse. J. Embryol. Exp. Morphol. hh: 483-489. Detlefsen, J.A. 1921. A new mutation in the house mouse. Amer. Nat. hh: 469-473. DiVirgilio, G., N. Lavenda and J.L. Worden. 1967. Sequence of events in neural tube closure and the formation of the neural crest in the chick embryo. Acta Anat. hh: Doolittle, C.H. and H. Rauch. 1965. Epinephrin and nor- epinephrin levels in dilute-lethal mice. Biochem. BiOphys. Res. Comm. ih: 43-47. Dry, F.W. 1926. The coat of the mouse (Mus musculus). J. Genet. ih: 287-340. Dunn, L.C. 1942. Studies of spotting patterns V. Further analysis of minor spotting genes in the house mouse, 230 Genetics ii: 258-267. Dunn, L.C. and D.R. Charles. 1937. Studies on spotting patterns I. Analysis of quantitative variations in the pied spotting of the house mouse. Genetics ii: 14-220 Dunn, L.C. and W. Einsele. 1938. Studies of multiple allelomorphic series in the house mouse IV. Quantita- tive comparisons of melanins from members of the albino series. J. Genet. ih: 145-152. Edelstein, L.M., N. Cariglia, M.R. Okun, R.P. Patel and D. Smucker. 1975. Inability of murine melanoma melanosomal "tyrosinase" (L-dopa oxidase) to oxidize tyrosine to melanin in polyacrylamide gel systems. J. Invest. Dermatol. hi: 364-370. Erway, L.C., A.S. Fraser and L.S. Hurley. 1971. Prevention on congenital otolith defect in pallid mutant mice by manganese supplementation. Genetics hl: 97-108. Erway, L.C., L.S. Hurley and A. Fraser. 1966. Neurological defect: Manganese in phenocopy and prevention of a genetic abnormality on inner ear. Science 152: 1766- 1768. Essner, E. and C. Oliver. 1973. A hereditary alteration in kidneyslof mice with Chediak-Higashi syndrome. Amer. J. Pathol. ii: 217-232. Essner, E. and C. Oliver. 1974. Lysosome formation in hepatocytes of mice with Chediak-Higashi syndrome. Lab. Invest. Eh: 596-607. Falconer, D.S. 1953. Total sex-linkage in the house mouse. 2. Indukt. Abstamm.-u. Vereb. hh: 210-219. Feldman, H.W. 1922. A fourth allelomorph of the albino series in mice. Amer. Nat. hh: 573-574. Fey, H. 1972. Eriochrome black, a means for reduction of non-specificity in immunofluorescence. Path. et Microbiol. hh: 271-277. Fisk, D.E. and C. Kuhn. 1976. Emphysema-like change in the lungs of the blotchy (hhblo) mouse. Amer. Rev. Resp. Dis. 113: 787-797. Fitzpatrick, T.B., P. Brunet and A. Kukita. 1958. The nature of hair pigment. in The Biology hi Hair Growth, W. Montagna and R.A. Ellis, eds., pp. 255-303, Academic Press, New York. 231 Foster, M. 1963a. Some physiological genetic aspects of mammalian melanogenesis. Ann. N. Y. Acad. Sci. 100: 743-748. Foster, M. 1963b. Tactics in pigment cell research. (discussion). in Methodology ih Mammalian Genetics, W.J. Burdette, ed., pp. 336-339, Holden-Day, San Fran- cisco. ' Foster, M. 1965. Mammalian pigment genetics, Adv. Genet. ih:311-339. Foster, M. 1967. Genetic aspects of mammalian melano- genesis. in Advances ih Biology hi Skin, vol. 8, ihh Pigmentary System, W. Montagna and F. Hu, eds., pp. 467-477, Permagon Press, New York. Foster, M., E. Barto and L. Thomson. 1972. Genetic con- trols of mammalian melanogenesis. in Pigmentation: Its Genesis and Biological Control, V. Riley, ed., pp. 387-400, Appleton-Century-Crofts, New York. Frost, P., J.L. Geigel, G.D. Weinstein and E.C. Gomez. 1973. Biodynamic studies of hamster flank organ growth: hormonal influences. J. Invest. Dermatol. hi: 159-167. Fujita, H., A. Ichiu, C. Asgami, T. Mori and I. Araki. 1971. Electron microscope study of neural crest cells and adjoining cells in mice, with special reference to melanosome formation. Arch. Histol. Japan. ii: 413- 428. Galbraith, D.B. 1969. Cell mass, hair type and expression of the agouti. gene. Nature 222: 288-290. Galbraith, D.B. 1971. Expression of genes at the agouti locus and mitotic activity of the hair bulb of the mouse. Genetics hi: 559-568. Galbraith, D.B. and R.J. Arceci. 1974. Melanocyte popula- tions of yellow and black hair bulbs in the mouse. J. Hered. hh:381-382. Galbraith, D.B. and A.M. Patrignani. 1976. Sulfhydral compounds in melanocytes of yellow (AV/h), nonagouti (h/h) and agouti (h/h) mice. Genetics hh: 587-591. Genis-Galvez, J.M. 1965. Role of the lens in the morpho- genesis of the iris and cornea. Nature 210: 209-210. 232 Gerson, D.E. and G. Szabo. 1969. The effects of a single gene substitution on the mammalian melanocyte system-- a qualitative and quantitative histological study in C57Bl and DBL mice. J. Phys. Anthropol. hi: 363-382. Ghadially, F.N. and J.F. Barker. 1960. The histogenesis of experimentally induced melanotic tumors in the Syrian hamster (Cricetus auratus). J. Pathol. Bacteriol. 12:263-271. Giroud, A. 1957. Phenomenes d'induction et leur perturba- tions chez les Mammiferes. Acta Anat. 22:297-306. Graham, R.C., Jr., and M.J. Karnofsky. 1966. The early stages of absorption of injected horseradish peroxidase in the proximal tubules of mouse kidney: Ultra- structural cytochemistry by a new technique. J. Histo- chem. Cytochem. ih: 291-302. Green, M.C. 1961. Himalayan, a new allele of albino in the mouse. J. Hered. hi: 73-75. Grobman, A.B. and D.R. Charles. 1947. Mutant white mice: A new dominant autosomal mutant affecting coat color in Mus musculus. J. Hered. hh: 381-384. Grossi, C.E. and H. vonMayersbach. 1964. Technical factors affecting immunohistological methods I. The influence of different pH and buffer salts for unspecific reac- tions. Acta Histochemica i2: 382-398. Gruneberg, H. 1942. Inherited macrocytic anaemias in the house mouse. II. Dominance relationships. J. Genet. Gruneberg, H. 1952. The Genetics hi the Mouse, 2nd ed., Nijhoff, The Hague. Gruneberg, H. 1953. The relations of microphthalmia and white in the mouse. J. Genet. hi:359-362. Gruneberg, H., J.M. Gray, and G.M. Truslove. 1965. Con- genital defects as indicators of lifelong abnormal processes. Genet. Res. h: 263-283. Hadley, M.B. and W.C. Quevedo, Jr. 1966. Vertebrate epi- dermal melanin unit. Nature 209:1334-1335. Hadorn, E. 1961. Developmental Genetics and Lethal Factors. J. Wiley and Sons, New York. 233 Halberg, F. and M.B. Visscher. 1950. Regular diurnal physiological variation in eosinophil levels in five stocks of mice. Proc. Soc. Exp. Biol. Med. 1h: 846- 847. - Halley, G. 1955. The placodal relations of the neural crest in the domestic cat. J. Anat. 92‘ 133-152. Hamburgh, M. and M.R. Bornstein. 1970. Myelin synthesis in two demyelinating mutations in mice. Exp. Neurol. ‘ih: 471-476. Hanks, J.B. and R.E. Wallace. 1949. Relation of oxygen and temperature in the preservation of tissues by refrigeration. Proc. Soc. Exp. Biol. Med. 1i: 196-200. Hauschka, T.S., B.B. Jacobs, and B.A. Holdridge. 1968. Recessive yellow and its interaction with belted in the mouse. J. Hered. hi: 339-341. Hearing, V.J. 1973a. Mammalian melanogenesis: Tyrosinase vs. peroxidase involvement, and activation mechanisms. Arch. Biochem. Biophys. 158: 720-725. Hearing, V.J. 1973b. Tyrosinase activity in subcellular fractions of black and albino mice. Nature New Biol. 245:81-83. Hearing, V.J. and T.M. Ekel. 1975. Involvement of tyrosi— nase in melanin formation in murine melanoma. J. Invest.Dermatol. hh: 80-85. Hearing, V.J., P. Phillips and M.A. Lutzner. 1973. The fine structure of melanogenesis in coat color mutants of the mouse. J. Ultrastruct. Res. hi: 88-106. Heidenthal, G. 1940. A colorimetric study of genic effect on guinea-pig coat color. Genetics g§:197-214. Heston, v.13. and c. Vlahakis. 1968. csn-Avy- a high hepatoma and high mammary tumor strain of mice. J. Nat. Cancer Inst. hh: 1161-1166. Hogeboom, C.H. and M.H. Adams. 1942. Mammalian tyrosinase and dopa oxidase. J. Biol. Chem. 145: 273-279. Holland, J.M. 1976. Serotonin deficiency and prolonged bleeding in beige mice. Proc. Soc. Exp. Biol. Med. 151: 32-39. 234 Holstein, T.J., J.B. Burnett and W.C. Quevedo, Jr. 1967. Genetic regulation of multiple forms of tyrosinase in mice: Action of h and h loci. Proc. Soc. Exp. Biol. Med. iih: 415-418. Holstein, T.J., W.C. Quevedo, Jr., and J.B. Burnett. 1971. Multiple forms of tyrosinase in rodents and lago- morphs with special reference to their genetic control in mice. J. Exp. 2001. iii: 173-184. Holstein, T.J., C.P. Stowell, W.C. Quevedo, Jr., R.M. Zarcaro and T.C. Bienieki. 1973. Peroxidase, "pro- tyrosinase", and the multiple forms of tyrosinase in mice. Yale J. Biol. Med. Eh: 560-571. Horch, C., H.B. Chase and N.I. Gonsalves. 1978. Studies on an anophthalmic strain of mice VI. Lens and cup interaction. Develop. Biol. hi: 352-357. Horstadius, S. 1950. The Neural Crest. Oxford University Press, London. Hsia, S.L. and W. Voight. 1974. Inhibition of dihydrotes- tosterone formation: An effective means of blocking androgen action in hamster sebaceous gland. J. Invest. Dermatol. hi: 224-227. Hunt, D.M. 1974. Primary defect in copper transport under- lies mottled mutants in the mouse. Nature 249: 852- 854. Hunt, D.M. and D.R. Johnson. 1972. An inherited deficiency in noradrenaline biosynthesis in the brindled mouse. J. Neurochem. i2: 2811-2819. Hunt, D.M. and D.F, Skinner. 1976. Private communication. Mouse News Letter hh: 42. Hunter, J.A.A., J.H. Mottaz and A.S. Zelickson. 1970. Melanogenesis: Ultrastructural histochemical obser- vations on ultraviolet irradiated human melanocytes. J. Invest. Dermatol. hh: 213-221. Illman, O. and F.N. Ghadially. 1960. Coat colour and experimental melanotic tumor production in the hamster. Brit. J. Cancer ih: 483-488. James, S. 1979. Effects of the gene Eh on reproduction in the Syrian hamster, Mesocricetus auratus. Master's thesis, Michigan State University. 235 Jimbow, K. and T.B. Fitzpatrick. 1973. Characterization of a new melanosomal structural component- the vesiculo- globular body- by conventional transmission, high vol- tage, and scanning electron microscopy.‘ J. Ultra- struct. Res. hh: 269-283. Johnston, M.C. 1966. A radioautographic study of the mi- gration and fate of cranial neural crest cells in the chick embryo. Anat. Rec. 156: 143-156. Johnston, M.C. and M.A. Listgarten. 1972. Observations on the migration, interaction, and early differentia- tion of orofacial tissues. in Developmental Aspects hi Oral Biology, H.C. Slavkin and L.A. Bavetta, eds., pp. 53-80, Academic Press, New York. Johnston, N.W. and J. Bienenstock. 1974. Abolition of non-, specific fluorescent staining of eosinophils. J. Immunol. Methods 5: 189-194. Karlsson, K.-A., G. Nilsson, A. There and B. Morein. 1975. Quantitation of the inhibitory effect of eriochrome black and sodium nitrite on non-specific immunofluo- rescent staining. Acta Path. Microbiol. Scand. Sect. B hi: 482-490. Kitamura, Y., S. Go and K. Hatanaka. 1978. Decrease of mast cells in E/Ev mice and their increase by bone marrow transplantation. Blood hi; 447-452. Klaus, S.N. 1969. Pigment transfer in mammalian epidermis. Arch. Dermatol. 100: 756-762. Knapp, B.H. and S. Polivanov. 1958. Anophthalmic albino: A new mutation in the Syrian hamster, Cricetus auratus Amer. Nat. hi: 317-318. Knisely, A.S., D.L. Gasser and W.K. Silvers. 1975. Expres- sion in organ culture of agouti locus genes of the mouse. Genetics 12‘ 471-475. Konyukhov, B.V. and V.V. Osipov. 1968. Intra-allelic complementation of microphthalmia and white genes in mice. Soviet Genet. h: 65-76. Konyukhov, B.V., V.V. Osipov and M.P. Vachruscheva. 1965. Damage of derivatives of neural crest in mice of mutant lines microphthalmia and white. Arch. Anat. Histol. Embryol. h: 100-107. Kukita, A. 1957. Preliminary and short report: Changes in tyrosinase activity during melanocyte proliferation in the hair growth cycle. J. Invest. Dermatol. ih: 273-274 236 Lamoreux, M.L. and T.C. Mayer. 1975. Site of gene action in the development of hair pigment in recessive yellow (3/3) mice. Develop. Biol. hh: 160-166. Lamoreux, M.L. and E.S. Russell. 1979. Developmental interaction in the pigmentary system of mice I. Inter- actions between effects of genes on color of pigment and on distribution of pigmentation in the coat of the house mouse (Mus musculus). J. Hered. 1h: 31-36. Lane, P.W. 1966. Association of meagcolon with two rec- cessive spotting genes in the mouse. J. Hered. ii: 29-31. Leach, R.M. 1971. Role of manganese in mucopolysaccharide metabolism. Fed. Proc. ih: 991-994. Leamy, L. and 8.8. Sustarsic. 1978. A morphometric analysis of agouti genotypes in C57Bl/6 house mice. Syst. 2001. i1: 49-60. LeDouarin, N.M. 1975. The neural crest in the neck and other parts of the body. in Morphogenesis and Mal- formation hh the Face and Brain, Birth Defects Origipal Article Series, vol. XI(7), D. Bergsma, ed., pp. 19-50, Alan R. Liss, New York. LeLievre, 0.5. and N.M. LeDouarin. 1975. Mesenchymal derivatives of the neural crest: Analysis of chimearic quail and chick embryos. J. Embryol. Exp. Morphol. ii: 125-154. Lerner, A.B. 1953. Metabolism of phenylalanine and tyrosine. Adv. Enzymol. ih: 73-128. Lim, D.J. and L.C. Erway. 1974. Influence of manganese on genetically defective otoliths, a behavioral and morpho- logical study. Ann. Otol. hi: 565-581. Little, G.C. and A.M. Cloudman. 1937. The occurence of a dominant spotting mutation in the house mouse. Proc. Nat. Acad. Sci. USA ii: 535-537. Lopashov, G.V. 1963. Developmental Mechanisms hi Vertebrate Eye Rudiments, Permagon Press, New York. Lopashov, G.V. and 0.G. Stroeva. 1961. Morphogenesis of the vertebrate eye. Adv. Morph. i: 331-377. Lutzner, M.A. and C.T. Lowrie. 1972. Ultrastructure of the development of the normal black and giant beige melanin granules in the mouse. in Pigmentation: Its Genesis and Biolpgical Control, V. Riley, ed., pp. 89- 105, Appleton-Century-Crofts, New York. 237 Lutzner, M.A., C.T. Lowrie and H.W. Jordan. 1967. Giant granules in leukocytes of the beige mouse. J. Hered. hh: 299-300. Lyon, M.F. 1951. Hereditary absence of otoliths in the house mouse. J. Physiol. 114: 410-418. Lyon, M.F. 1953. Absence of otoliths in the mouse: An effect of the pallid mutant. J. Genet. hi: 638-650. MacDowell, E.C. 1950. "Light"- a new mouse color. J. Hered hi: 35-36. Markert, C.L. 1948. The effects of thyroxine and anti- thyroid compounds on the synthesis of pigment granules in chick melanoblasts cultured ih vitro. Physiol. 2001. ii: 309-327. Markert, C.L. and W.K. Silvers. 1956. The effects of geno- type and cell environment on melanoblast differentia- tion in the house mouse. Genetics hi: 429-450. Markert, C.L. and W.K. Silvers. 1959. Effects of genotype and cellular environment on melanocyte morphology. in Eigment Cell Biology, M. Gordon, ed., pp. 241-248, Academic Press, New York. Mason, H.S. 1948. The chemistry of melanin III. Mechanismof theoxidation of dihydroxyphenylalanine by tyrosinase. J. Biol. Chem. 172: 83-99. Mason, H.S. 1959. Structure of melanins. in Pigment Cell Biology, M. Gordon, ed., pp. 563-582, Academic Press, New York. Mauer, I. and M.B. Sidman. 1967. Phenylalanine metabolism in a dilute-lethal strain of mice. J. Hered. hh: 14- 16. Maul, G.C. and J.A. Brumbaugh. 1971. On the possible func- tion of coated vesicles in melanogenesis of the re- generating fowl feather. J. Cell Biol. hh: 41-48. Mayer, T.C. 1965. The development of piebald spotting in mice. Develop. Biol. ii: 319-334. Mayer, T.C. 1967a. Pigment cell migration in piebald mice. Develop. Biol. ii: 521-535. Mayer, T.C. 1967b. Temporal skin factors influencing the development of melanoblasts in piebald mice. J. Exp. 2001. 166: 397-404. 238 Mayer, T.C. 1970. A comparison of pigment cell development in albino, steel, and dominant-spotting mutant mouse embryos. Develop. Biol. ii: 297-309. Mayer, T.C. 1973a. Site of gene action in steel mice: Analysis of the pigment defect by mesoderm-ectoderm recombinations. J. Exp. 2001. 184: 345-352. Mayer, T.C. 1973b. The migratory pathway of neural crest cells into the skin of mouse embryos. Develop. Biol. ih: 39-46. Mayer, T.C. 1977a. Enhancement of melanocyte development from piebald neural crest by a favorable tissue en- vironment. Develop. Biol. hh: 255-262. 'Mayer, T.C. 1977b. Personal communication. Mayer, T.C. and J.L. Fishbane. 1972. Mesoderm-ectoderm interaction in the production of the agouti pigmenta- tion pattern in mice. Genetics ii: 297-303. Mayer, T.C. and M.C. Green. 1968. An experimental analysis of the pigment defect caused by mutations at the E and §i_loci in mice. Develop. Biol. ii: 62-75. Mayer, T.C. and L. Oddis. 1977. Pigment cell differentia- tion in embryonic mouse skin and isolated epidermis: An ih vitro study. J. Exp. 2001. 202: 415-424. vonMayersbach, H. 1967. Principles and limitations of immunohistochemical methods. J. Roy. Micros. Soc. hi: 295-308. McCoshen, J.A. and D.J. McCallion. 1975. A study of the primordial germ cells during their migratory phase in steel mutant mice. Experentia ii: 589-590. McGrath, E.P. and W.C. Quevedo, Jr. 1965. Genetic regula- tion of melanocyte function during hair growth in the mouse: Cellular events leading to "pigment clumping" within developing hairs. in Biology hi the Skin and Hair Growth, A.G. Lyne and B.F. Short, eds., pp. 727- 745, Angus and Robertson, Sydney. Miliare, J. 1974. Histochemical Aspects hi Organoghhesis ih Vertebrates Part i. Hand. der Histochemie, vol III, suppl. (part 3), G. Fischer Verlag, Stuttgart. Mintz, B. 1967. Gene control of mammalian pigmentary differentiation I. Clonal origin of melanocytes. Proc. Nat. Acad. Sci. USA hh: 344-351. 239 Mintz, B. 1969. Developmental mechanisms found in allo- phenic mice with sex chromosomal and pigmentary mosaicism. in Birth Defects Original Article Series i, First Conference hh Clinical Delineation hi Birth hh- fects, Part I- special lectures, D. Bergsma and V. Mc- Kusick, eds., pp. 11-22, National Foundation, New York. Mintz, B. 1970. Gene expression in allophenic mice. in Control Mechanisms in the Expression of Cellular Pheno- types, H.A. Padylkulz, ed., pp. 15-42, Academic Press, New York. Mintz, B. 1971a. Genetic mosaicism ih vivo: Development and disease in allophenic mice. Fed. Proc. ih: 935-943. Mintz, B. 1971b. Clonal basis of mammalian differentiation. in Control Mechanisms hi Growth and Differentiation, 25th Symp. Soc. Exp. Biol., D.D. Davies, M. Balls, eds., pp. 345-369, University Press, Cambridge. Mintz, B. 1974. Gene control of mammalian differentiation. Ann. Rev. Genet. h: 411-470. Mintz, B. and C. Cronmiller. 1978.. Normal blood cells of anemic genotype in teratocarcinoma-derived mosaic mice. Proc. Nat. Acad. Sci. USA ih: 6247-6251. Mintz, B. and E.S. Russell. 1957. Gene-induced embryological modifications of primordial germ cells in the mouse. Mintz, B. and W.K. Silvers. 1970. Histocompatibility anti- gens on melanoblasts and hair follicle cells. Trans- plantation h: 497-505. Miyazaki, K. and N. Ohtaki. 1976. Tyrosinase as Glyco- protein. in Pigment Cell, vol. 3, V. Riley, ed., pp. 113-120, S. Karger, Basel. Morrison, D.E. 1967. Multivariate Statistical Methods, 2nd ed., McGraw-Hill, New York. Morriss, G.M. and P.V. Thorogood. 1978. An approach to cranial neural crest cell migration and differentia- tion in mammalian embryos. in Development ih Mammals, vol. 3, M.H. Johnson, ed., pp. 363-412, North-Holland, New York. Mottaz, J.H. and A.S. Zelickson. 1967. Melanin transfer: A possible phagocytic process. J. Invest. Dermatol. hi: 605-610. 240 Moyer, F. 1961. Electron microscope studies on the origin, development, and genetic control of melanin granules in the mouse eye. in The Structure hi £EE.§Z£’ G.K. Smelser, ed., pp. 469-486, Academic Press, New York. Moyer, F. 1963. Genetic effects on melanosome fine struc- ture and ontogeny in normal and malignant cells. Ann. N.Y. Acad. Sci. 100: 584-606. Moyer, F. 1966. Genetic variations in the fine structure and ontogeny of mouse melanin granules. Amer. 2001. 6: 43-66. Mufson, R.A. 1975. The tyrosinase activity of melanosomes from Harding-Passey melanoma: The absence of a per- oxidase component ih vitro. Arch. Biochem. Biophys. ihi: 738-743. Nichols, S.E. and W.M. Reams. 1960. The occurence and mor- phogenesis of melanocytes in the connective tissue of the PET/MCV mouse strain. J. Embryol. Exp. Morphol. 8: 24-32. Niece, R.L., E.C. McFarland and E.S. Russell. 1963. Eryth- roid homeostasis in normal and genetically anemic mice: Reaction to induced polycythemia. Science 142: 1468- 1469. Noden, D.M. 1975. An analysis of the migratory behavior of avian cephalic neural crest cells. Develop. Biol. hi: 106-130. Ohtaki, N. and K. Miyazaki. 1973. Immunologic homogeneity and electrophoretic heterogeneity of mouse melanoma tyrosinases. J. Invest. Dermatol. hi; 339-343. Ohtaki, N. and K. Miyazaki 1976. Immunologic cross-reaction between human and mouse tyrosinases. J. Invest. Derm- matol. hh: 327-329. Okazaki, K., M. Uzuka, F. Morikawa, K. Toda and M. Seiji. 1976. Transfer mechanism of melanosomes in epidermal cell culture. J. Invest. Dermatol. hi: 541-547. Okun, M.R., L.M. Edelstein, N. Or, G. Hamada, B. Donnellan and W.F. Lever. 1970. Histochemical differentiation of peroxidase-mediated from tyrosinase-mediated melanin formation in mammalian tissues. Histochemie ii: 295- 309. Oliver, C. and E.Essner. 1973. Distribution of anomalous lysosomes in the beige mouse: A homologue of Chediak- Higashi syndrome. J. Histochem. Cytochem. ii: 218-228. 241 Oliver, C. and E. Essner. 1975. Formation of anomolous lysosomes in monocytes, neutrophils, and eosinophils from bone marrow of mice with Chediak-Higashi syndrome. Lab. Invest. ii: 17-27. Packer, 8.0. 1967. The eye and skeletal effects of two mutant alleles at the microphthalmia locus of Mus musculus. J. Exp. 2001. 165: 21-46. Phillips, R.J.S. 1961. 'Dapples", a new allele at the mottled locus in the house mouse. Genet.Res. i: 290- 295. Pierro, L.J. 1963a. Effects of the light mutation of mouse coat color on eye pigmentation. J. Exp. 2001. 153: 81- 87. Pierro, L.J. 1963b. Pigment granule formation in slate, a coat color mutant in the mouse. Anat Rec. 146: 365- 372. Plocher, T.A. and T.L. Powley. 1976. Effect of hypophysec- tomy on weight gain and body composition in the genetically obese yellow (hY/h) mouse. Metabolism ii: 593-602. Pomerantz, S.E. and J. P.-C. Li. 1974. Tyrosinase in the skin of the albino hamsters and mice. Nature 252:241- 243. Poole, T.W. 1974. Dermal-epidermal interactions and the site of action of the yellow (AV) and nonagouti (h) coat color genes in the mouse. Develop. Biol. ih: 208- 211. Poole, T.W. 1975. Dermal-epidermal interactions and the action of alleles at the agouti locus in the mouse. Develop. Biol. hi: 203-210. Poole, T.W. and W.K. Silvers.1976a. The development of regional pigmentation patterns in black and tan (at) mice. J. Exp. 2001. 197: 115-120. Poole, T.W. and W.K. Silvers.1976b. An experimental analysis of the recessive yellow coat color mutant in the mouse. Develop. Biol. hh: 377-381. Potten, C.S. 1968. Radiation depigmentation of mouse hair: A study of follicular melanocyte populations. Cell Tissue Kinet. i:239-254. Potten, C.S. and H.B. Chase. 1970. Radiation depigmentation of mouse hair: Split dose experiments and melanocyte 242 precursors (amelanotic melanoblasts) in the resting Potten, C.S. and A. Howard. 1968. Radiation depigmentation of the hair: Dose-response relationship for follicular melanocyte precursors. Radiat. Res. ii: 544. Prota, G. 1972. Structure and biogenesis of phaeomelanins. in Pigmentation: Its Genesis and Biological Control, V. Riley, ed., pp. 615-630, Appleton-Century-Crofts, New York. Prota, G. and R.H. Thomson. 1976. Melanin pigmentation in mammals. Endeavor 124:32-38. Purichia, N. and L.C. Erway. 1972. Effects of dichloro- phenamide, zinc, and manganese on otoligh development in mice. Develop. Biol. ii: 395-405. Pusey, W.A. 1926. Recipe for epilating wax. J. Amer. Med. Assoc. ii: 663. Quevedo, W.C., Jr. 1957. Loss of clear cells in the hair follicles of X-irradiated albino mice. Anat. Rec. 127: 725-730. Quevedo, W.C., Jr. 1971. Genetic regulation of pigmentation in mammals. in Biology hi Normal and Abnormal Melano- cytes, T. Kawamura, T.B. Fitzpatrick and M. Seiji, eds., pp. 99-115, Univ. of Tokyo Press. Quevedo, W.C., Jr., T.C. Bienieki, F.A. McMorris, and M.J. Hepinstall. 1966. Environmental and genetic influences of radiation-induced tanning of murine skin. in Advances ih Biology hi Skin, vol. 8, ihh Pigmentary System, W. Montagna and F. Hu, eds., pp. 361-377, Permagon Press, Oxford. Quevedo, W.C., Jr. and H.B. Chase. 1958. An analysis of the light mutation of coat color in mice. J. Morphol. 102: 329-345. Quevedo, W.C., Jr. and C.E. McTague. 1963. Genetic in- fluences on the response of mouse melanocytes to ultra- violet light: The melanocyte system of hair-covered skin. J. Exp. 2001. iii: 159-168. Quevedo, W.C., Jr. and J.A. Smith. 1963. Studies on radia- tion-induced tanning of skin. Ann. N. Y. Acad. Sci. 100: 364-388. 243 Quevedo, W.C., Jr., M.C.Youle, D.T. Rovee, and T.C. Bienieki. 1966. The developmental fate of melanocytes in murine skin. in Structure and Control hi the Melanocyte, G. Della Porta and O. Muhlbock eds., pp. 228-241, Springer- Verlag, Berlin. Randall, L.O. 1946. Reaction of thiol compounds with peroxidase and hydrogen peroxide. J. Biol.Chem. 164: Raper, H. 1926.. The tyrosinase-tyrosine reaction. Biochem. J. ih: 735-742. Rappaport, H., T. Nakai and H. Swift. 1963. The fine struc- ture of normal and neoplastic melanocytes in the Syrian hamster with particular reference to carcinogen-induced melanotic tumors. J. Cell Biol. ih: 171-186. Rauch, H. and M.T. Yost. 1963. Phenylalanine metabolism in dilute-lethal mice. Genetics hh: 1487-1495. Raven C.P. and J. Kloos. 1945. Induction by medial and lateral pieces of the archenteron roof, with special reference to the determination of the neural crest Acta Neer. Morphol. Norm. et Pathol. Utrecht. i: 348-362. Rawles, M.E. 1947. Origin of pigment cells from the neural crest in the mouse embryo. Physiol. 2001. ih: 248-266. Reed, S.C. and J.M. Henderson. 1940. Pigment cell migration in mouse epidermis. J. Exp. 2001. hi; 409-418. Reiter, R.J. 1969. Pineal function in long term blinded male and female golden hamsters. Gen. Comp. Endocr. ii: 460-468. Reiter, R.J. and F. Fraschino. 1969. Endocrine aspects of the mammalian pineal gland: A review. Neuroen- docrinology i: 219-255. Reynolds, J. 1954. The epidermal melanocytes of mice. J. Rhodin, J.A. 1974. Histology, Oxford University Press, New ' York. Ris, H. 1941. An experimental study on the origin of melano- phores in birds. Physiol. 2001. ih: 48-66. Rittenhouse, E.W. 1965. Inherited variation in the fine structure of melanin granules in hair bulb melanocyte of the house mouse. Doctoral dissertation, Univ. of Michigan. 244 Rittenhouse, E. 1968a. Genetic effects on fine structure and development of pigment granules in mouse hair bulb melanocytes I. The h and h_loci. 'Develop. Biol. ii: 351-365. . ‘ Rittenhouse, E. 1968b. Genetic effects on fine structure and development of pigment granules in mouse hair bulb melanocytes II. The i and R loci. Develop. Biol. ii: 366-381. Robinson, R. 1955. Two new mutations in the Syrian hamster. Nature 176: 353-354. Robinson, R. 1958. Genetical studies of the Syrian hamster I. The mutant genes cream ruby-eyed and piebald. J. Genet. hh: 85-103. ' Robinson, R. 1962. Anophthalmic whites: A mutant with un- usual morphological and pigmentary properties in the Syrian hamster. Amer. Nat. hh: 183-185. Robinson, R. 1964. Genetical studies of the Syrian hamster VI. Anophthalmic white. Genetica ih: 241-250. Robison, W.C., T. Kuwabara and D.G. Cogan. 1975. Lysosomes and melanin granules of the retinal pigment epithelium in a mouse model of Chediak-Higashi syndrome. Invest. Ophthal. ih: 312-317. Rosdahl, I.K. and G. Szabo. 1978. Mitotic activity of epi- dermal melanocytes in UV-irradiated mouse skin. J. Invest. Dermatol. ih: 143-148. Rowe, D.W., B.B. McGoodwin, G.R. Martin, M.D. Sussman, D. Grahn, B. Paris and C. Franzblau. 1974. A sex-linked defect in the cross-linking of collagen and elastin associated with the mottled locus in mice. J. Exp. Med. iii: 180-192. Ruscetti,F.W., D.R. Boggs, B.J. Torok and 8.8. Boggs. 1976. Reduced blood and marrow neutrophils and granulocyte colony-forming cells in hijhih mice. Proc. Soc. Exp. Biol. Med. iii: 398-402. Russell, E.S. 1946. A quantitative histological study of the pigment found in the coat color mutants of the house mouse I. Variable attributes of the pigment granules. Genetics ii: 327-346. Russell, E.S. 1948. A quantitative histological study of the pigment found in the.coat color mutants of the house mouse II. Estimates of the total volume of pigment. Genetics ii: 228-236. 245 Russell, E.S. 1949a. A quantitative histological study of the pigment found in the coat color mutants of the house mouse III. Interdependence among the variable granule attributes. Genetics ih: 133-145. Russell, E.S. 1949b. A quantitative histological study of the pigment found in the coat color mutants of the house mouse. IV. The nature of the effects of genic substitution in five major allelic series. Genetics ih: 146-166. Russell, E.S. 1949c. Analysis of pleiotropism at the E-locus in the mouse: Relationship getween E and EV substitu- tion on hair pigmentation and on erythrocytes. Genetics ih: 708-723. Russell, E.S. and S.E. Bernstein. 1966. Blood and blood formation. in Biology hi the Laboratory Mouse, 2nd ed., B.L. Green, ed., pp. 351-372, McGraw-Hill, New York. Russell, E.S., S.E. Bernstein, F.A. Lawson and L.J. Smith. 1959. Long-continued function of normal blood-forming tissue transplanted into genetically anemic hosts. J. Nat Cancer Inst. ii: 557-566. Russell, E.S. and E. Fetke. 1958. Analysis of E-series pleiotropism in the mouse: Effect of EV/Ev substitution on definitive germ cells and on ovarian tumorigenesis. J. Nat. Cancer Inst. ii: 365-381. Russell, E.S., L.J. Smith and F.A. Lawson. 1956. Implanta- tion of normal blood-forming tissue in radiated genetically anemic hosts. Science 124: 1076-1077. Russell, L.B. and W.L. Russell. 1948. A study of the physiological genetics of coat color in the mouse by means of the dopa reaction in frozen sections of skin. Genetics ii: 237-262. Sakurai, T., H. Ochiai and T. Takeuchi. 1975. Ultrastructural change of melanosomes associated with agouti pattern formation in mouse hair. Develop. Biol. hi: 466-471. Sarvella, P.A. and L.B. Russell. 1956. Steel, a new dominant gene in the house mouse. J. Hered. hi: 123-128. Schaible, R.H. 1969. Clonal distribution of melanocytes in piebald-spotted and variegated mice. J. Exp. Zool. 17% 181-199. Schenk, B.A. and C.J. Churukian 1974.. Immunoflourescence counterstains. J. Histochem. Cytochem. ii: 962-966. 246 Schulte, H. vonW., F. Tilney. 1915. Development of the neuraxis in the domestic cat to the stage of twenty- one somites. Ann. N. Y. Acad. Sci. ih} 319-346. Schumann, H. 1960. Die entstehung der Scheckung bei der mausen mit weisser blesse. Develop. Biol. i: 501-515. Searle, A.G. 1952. A lethal allele of dilute in the house mouse. Heredity h: 395-401. Searle, A.G. 1968a. Comparative Genetics hi Coat Colour ih Mammals. Logos Press, London. Searle, A.G. I968b. An extension series in the mouse. J. Hered. ii: 341-342. Seller, M.J: 1972. Some observations on the blood levels of phenylalanine in dilute mice and a comparison be- tween human phenylketonuria and the disease of mice homozygous for the dilute lethal gene. Clin. Genet. i: 495-500. Sengal, P. 1974. Morphogenesis hi Skin. Cambridge Univ. Press, Cambridge. Shrader, R.E., L.C. Erway and L.S. Hurley. 1973. Muco- polysaccharide synthesis in the developing inner ear of manganese-deficient and pallid mutant mice. J. Nutrition ihh: 257-266. Sidman, R.L. and R. Pearlstein. 1965. Pink-eyed dilution (p) gene in rodents: Increased pigmentation in tissue culture. Develop. Biol. ii: 93-116. Silver, A.F., H.B. Chase and C.S. Potten. 1969. Melanocyte precursos cells in the hair follicle germ during the dormant stage (telogen). Experentia ii: 299-301. Silver, A.F., R.D. Fleischmann and H.B. Chase. 1977. The fine structure of the melanocytes of the adult mouse hair follicle during their amelanotic phase (telogen and early anagen). Amer. J. Anat. ihh: 653-658. Silvers, W.K. 1956. Pigment cells; occurrence in hair follicles. J. Morphol. ii: 41-55. Silvers, W.K. 1957. Melanoblast differentiation secured from different mouse genotypes after transplantation to adult mouse spleen or to chick embryo coelom. J. Exp. 2001. 135: 221-238. ‘ Silvers, W.K. 1958a. An experimemtal approach to activity of genes at the agouti locus in the mouse II. Trans- plants of newborn £42 ventral skin to ht/g.‘hw/g and 247 hjh hosts. J. Exp. 2001. 137: 181-187. Silvers, W.K. 1958b. An experimental approach to the action of the genes at the agouti locus in the mouse III. Transplants of newborn hV-, h- and ht-.skin to hY-, hV-, h- and 22 hosts. J. Exp. 2001. 137: 189-196. Silvers, W.K. 1958c. Origin and identity of clear cells found in hair bulbs of albino mice. Anat. Rec. 130: 135-144. Silvers, W.K. 1979. The Coat Colors h_ Mice. Springer- Verlag, New York. Silvers, W.K. and E.S. Russell. 1955. An experimental approach to action of genes at the agouti locus in the mouse. J. Exp. 2001. 130: 199-220. Simler, S., S. Essayag, M. Ledig, C. Koehl and P. Mandel. 1977. Free amino acid pool in the brain of mice homozygous for the gene "dilute lethal". J. Neurosci. Res. i: 281-287. Sober, H.A. 1970. Handbook hi Biochemistry, 2nd ed., Che- mical Rubber Co., Cleveland, Ohio. Sorsby, A., J.B. Davey. 1954. Ocular associations of dappling in the coat colour of dags 1. Clinical and genetical data. J. Genet. ii: 425-440. Starrico, R.J. and H. Pinkus. 1957. Quantitative and qualitative data on the pigment cells of adult human epidermis. J. Invest. Dermatol. ih: 33-45. Stroeva, O.G. 1960. Experimental analysis of the eye morphogenesis in mammals. J. Embryol. Exp. Morphol. h: 349-368. Stroeva, O.G. 1967. The correlation of the processes of proliferation and determination in the morphogenesis of iris and ciliary body in rats. J. Embryol. Exp. Morphol. ih: 269-287. Swan, G.A. 1973. Current knowledge of melanin structure. in Pigment Cell, vol. 1, V.J. McGovern and P. Russell, eds., pp. 151-157, S. Karger, Basel. Sweet, S.E. and W.C. Quevedo, Jr. 1968. Role of melanocyte morphology in pigmentation of mouse hair. Anat Rec. 162: 243-254. 248 Teillet, M.-A. 1971. Recherches sur la mode de migration et la differenciation des melanoblastes cutanes chez l'embryon d'ouiseau. Ann. d'Embryol. Morphol. h: 95-109. Teillet, M.-A and N.M. LeDouarin. 1970. La migration des cellules pigmentaires etudiee par la methode des graffes heterospecifiques de tube nerveaux chez d' embryon d'ouiseau. C.R. Acad. Sci. Paris, Ser. D. iih: 3095-3098. Theriault, L.L. and L.S. Hurley. 1970. Ultrastructure of developing melanosomes in C57 black and pallid mice. Develop. Biol. ii: 261-275. Toda, K., M.A. Pathak, J.A. Parrish and T.B. Fitzpatrick. 1972. Alteration of racial differences in melanosome distribution in human epidermis after exposure to ultraviolet light. Nature New Biol. iih: 143-145. Tosney, K.W. 1978. The early migration of neural crest in the trunk region of the avain embryo: An electron microscipic study. Develop. Biol. hi: 317-333. Treiman, D.M. and A. Tourain. 1973. Phenylalanine hydroxylaae in dilute lethal mice. Biochem. Biophys. Acta 313: 163-169. Turner, W.A., J.D. Taylor, and T.T. Tchen. 1975. Melanosome formation in the goldfish: The role of multivesicular bodies. J. Ultrastruct. Res. ii: 16-31. Vandegrift, W.B. 1942. A dehydrating fixative for general use, including a description of techniques and stains for paraffin and celloidin sections. Johns Hopkins Hosp. Bull. ii: 96-111. Waardenburg, P.J. 1951. A new syndrome combining develOpmen- tal anomalies of the eyelids, eyebrows, and nose root with pigmentary defects of the iris, head hair and with congenital deafness. Amer. J. Hum. Genet. i: 195- 253. Watterson, R.L. 1942. The morphogenesis of down feathers with special referece to the developmental history of melanophores. Physiol. 2001. ii:234-262. Webster, W. 1974. Aganglionic megacolon in piebald-lethal mice. Arch. Pathol. ii: 111-117. ' Weiss, S.E.W. 1970.. The fine structure and distribution of melanin granules in hair bulb melanocytes of Peromyscus maniculatus. Doctoral dissertation, Univ. of Michigan. 249 Weston, J.A. 1963. A radioautographic analysis of the migration and localization of trunk neural crest cells in the chick. Develop. Biol. h: 279-310. Weston, J.A. 1970. The migration and differentiation of neural crest cells. Adv. Morph. h: 41-114. Williams, W.J., E. Beutler, A.J. Erslev and R.W. Rundles. 1972. Hematolqu. McGraw-Hill, New York. Winterbourn, C.C, F. Woolf and L.I. Woolf. 1971. Brain lipids in mice homozygous for the gene 'dilute lethal' J. Neurochem. ih: 1077-1086. Wolfe, H.G. and D.L. Coleman. 1964. hi-spotted: A mutation in the mouse. Genet. Res. i: 432-440. Wolfe, H.G. and D.L. Coleman. 1966. Pigmentation. in Biology hi the Laboratorerouse, 2nd ed., EiL. Green, ed., pp. 405-426, McGraw-Hill, New York. Wolff, C.L. and H.C.Pitot. 1972. Variation of hepatic malic enzyme capacity with hepatoma susceptibility in mice of different genotypes. Cancer Res. ii: 1861-1863. Wolff, K. 1973. Melanocyte-keratinocyte interactions ih vivo: The fate of melanosomes. Yale J. Biol. Med. hh: 384- 396. . Woolf, L.I., A. Jakubovic, F. Woolf and P. Bory. 1970. Metabolism of phenylalanine in mice homozygous for the gene 'dilute lethal'. Biochem. J. 119: 895-903. Yoon, C.H. 1975. Total retinal degeneration in apparent anophthalmos of the Syrian hamster. Invest Ophthal. ih: 321-325. Zannoni, V.G. and E. Moraru. 1970. Mechanism of phenyl- alanine hydroxulation and "phenylketonuria" in dilute- lethal mice. in Metabolic Regulation and Enzyme Actiom A. $013 and S. Grisolia, eds., pp. 347-354, Academic Press, London. Zannoni, V.G., W.W. Weber, P. VanValen, A. Rubin, R. Bern- stein and B.N. La Du. 1966. Phenylalanine metabolism and "PKU" in dilute lethal mice. Genetics hh: 1391- 1399. Zimmerman, A.A. and S.W. Becker. 1959. Melanoblasts and melanocytes in fetal negro skin. Illinois Monogr. Med. Sci. h, no. 3, Univ. of Illinois Press, Urbana. IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII 1|11|“WWWWulwulmlmur