3060 9500 LIBRARY Michigan State University This is to certify that the dissertation entitled EVIVOANDEVITROEFFECTS OF DIFFERENTIATION FACTORS (NERVE GRGVIH FACTOR AND GLIA MA’IURATION FACTOR (N WIS presented by NARAYAN R. RAJU has been accepted towards fulfillment of the requirements for P . hD degree in PATHOILEY 2% Major professor Date February 23, 1988 MS U it an Affirmative Action/Equal Opportunity Institution 0-12771 MSU LlBRARlES .—_—. RETURNING MATERIALS: Place in book drop to remove this checkout from your record. FINES will be charged if book is returned after the date stamped below. I A DEH3ER'IATI‘K‘ i l Sub-erred to . Ht-‘hi‘m Scat. University . ' “vpartlul fulfillment of flu 1‘0“!th . for flu degree a! - a”. ,. ,i DOCTOR or mm ' . V j ‘3 Depart-ant of Pathology ‘ w w Lou ' ‘ .‘q. . .‘ m“’-. I — o‘ - aw ' \ n m AND m m EFFECTS OF DIFFERENTIATION "l ‘PACMS (NERVE cam FACTOR AND CLIA HAMATION FACTOR) ON NEURO-ONCOCENISIS. 1 ~ BY Nereyen R. Reju ‘tunr-r; '? " In H. p A. . ‘ --1nduc¢.’ , A v dtud) ' ~ - baht-diva '.>“- '4 3 A'Dlssmxrm‘ Michigan State University It CWMRI ”relaunch: ‘0! the“ require-ewe '.-‘ T for the degree of .mrinoc-- 't I“! i; L L , ' 41, 5.; ‘x‘ V m OF PHILOSOPHY _ Gavan 'tn".r‘f' ital- . w ;-..':_, ~-'~ ».u (99? run-:- magenta: letteloy‘ ‘ ‘r. "-T’ . "Jinx! \f.‘ Lyw‘ nu: m2 l' \“" 'il. HS" KIL‘US' l6.'-"§ ace of .‘i'lF‘R atzee. "Plnv'fi " steamer: — W’IIL'. . .Mptm 91"i°:lt. ff"; .~.-v;,u;zr, {mien}: a peoi'tw Var ' ' ‘. ,- en the sup-'T"555\‘I‘ can: en. cl ween nrrL- .‘nnaui develop. ‘ ‘ ‘ ABSTRACT m m AND m m EFFECTS OF DIFFERENTIATION FACTORS (NERVE GROWTH FACTOR AND GLIA MATURATION FACTOR) 0N NEURO-ONCOGENESIS. BY Nerayan R . Reju - » ftuh tree the purpose of this study to test the reverse transforma- .71) upbn‘ tuners of the NS (nervous system) induced by EM) (ethylnitro- "i ' , ~13 . fflYt"“I‘n the first experiment in 21.19 effect of NGF on transpla- .? r \ 1': «4 m-induced peripheral nerve neurinomas in rats was investigated ' 3* 3 ‘7': ”~an etudy. 0f the 34 rats trensplacentally exposed to mm and .4! V‘ i by eubcutaneoue administration of NC? on days 12-16, 90-9lo, ‘7", "pestl'pertm, 15 (47%) rats were affected with peripheral nerve Narayan R. Raju In a second experiment the in giyg effects of NGF and GMF on the anaplastic glioma T9 cell line were explored by transmission and scan- ning electron microscopy. After exposure to GMF, cells became slender with long branching processes that formed an interconnecting network. NGF induced flattened multipolar cells with web-like cytoplasmic processes that formed somatic links with adjacent cells via demonstrable junctional devices. Internal cell features, common to both treatments, were character- ized by formation of an elaborated cytoskeleton, intracellular organ- elles and junctional complexes. The nuclear cytoplasmic ratio was markedly reduced and the nuclei often contained single nucleoli. These changes are morphological indicators of differentiation. In a third study ENU-induced NS tumors from the first experiment were characterized by immunohistochemistry. S-100 protein and GFAP (glial fibrillary acidic protein) were shown to be reliable markers for astrocytomas and astrocytic cells within mixed gliomas. Neurinomas and oliogodendrogliomas were negative for GFAP. 8-100 immunoreactivity in neurinomas was inversely correlated with the degree of tumor anaplasia. The results of this study provide evidence that NGF and GMF (not tested in 212g) are potent reverse transformation agents for neuro- epithelial tumor cells possessing proper membrane receptors. tone-w! ' “Vii" QMI thanks it i , ‘ .‘H V‘V‘ ‘| ‘I 0' M .m'.‘ ”iOIlP-“I: '-—‘ Y t ‘ p ' I gl" “ “U.le—i‘r ‘l with I. . Initial.“ . ‘., in BUQD'IT'Z -, ' x~ In: ('t sniw‘.‘ it. NBDIGATION of recbu ugy. fan "'30 uni-fly "no , . Itfy ll‘bt .'.'.‘- I" N" '4 us 3.‘ t ‘ -. ' ' sis- ;iv'i ; .‘e ~ en. 9‘ Ms itulleeuu Adi-u? uuel'r - .gw-i oxygen .5 mil I‘Mo Deparreer m’; foam-"w. A396 fi'“‘-JI'-' Judaism. tor ' L Jepermni. for the new enema .e W W... 1‘0 Du. Karen (Ion-tens and smiley M upside. tor aren't-1p with ‘3an ‘ ll“ u i Mi ‘ 00.19391“! 'c': ’. v"; w‘ -- .c' ;' 23:7 '4'. " ~ ”new 'Wrn .. ACKNOWLEDGEMENT My sincere thanks and appreciation are extended to the members of my guidance committee for their support, advise and collaboration during the entire research program. Special thanks to Dr. Adalbert Koestner who served as the major guide in the overall experiment designs. His research experience and the art of scientific writing were the two most valuable wisdoms. I am extremely indebted to Dr. Yasuko Marushige and Dr. Keiji Marushige, Molecular Biology Laboratory, for sacrificing relentless time and energy in support of this project, and also for demonstrating the meticulous art of scientific research. Sincere thanks to Dr. Kathryn Lovell and Ms. Dorothy Okazaki, Department of Pathology, for the initiation and establishment of the immunohistochemistry laboratory. Ms. Okazaki's devotion and skills need special mention. To Dr. Thomas Mullaney, Animal Health Diagnostic Laboratory, and Dr. James Trosko, Department of Pediatrics and Human Development, for serving on my guidance committee and for the useful comments and criti- cisms on my dissertation. I wish to extend my thanks to Drs. Robert Langham, Robert Leader, Janver Krehbiel, Stuart Sleight, Robert Dunstan, Gary Watson, and Thomas Bell, Pathology Department, for the useful comments and encouragements throughout the program. To Drs. Karen Klomparens and Stanley Flagler, Center for Electron Optics, for their help with ultrastructural iii .y)" i‘Iy 'h‘ 0‘01. of ('r. r.s A I ' \i‘io '1‘» i‘1 'Ald to Mrs. Kay Butcher for resolving financial problems and - m: abinistrative work . fellow graduate students, particularly roomtes Dev ‘11. and John Dillberger for valuable scientific die- if“ A A” md ‘.'c '1 Z,” and " tr ”rimming. .. " lu‘ti-L‘ .\ IQM H!- A ‘A' 1! of AA . ”3;? nu: my”: J .. p -. ‘ AL arut.;v'u~ .« .. . . ,. , . , .fi pom“: 2mg “dated 5 . ' . .1: . :‘mnons . , . . s .55‘ —-‘ yo eaten of ”SI". . . . . . .. . i5 tfirmourea . . ... . '2'. ”them: .. . .. . .. .. . jmle”..,... . .. ‘ A t. of NM o rum is hcuprnr cos five 2 Au". .............. . . -. . ...... bu “Chemistry Detection at i"?! ........ .. . . n ”Anal-yam; .. ... . “‘55 in » gs TABLE OF CONTENTS Page LIST OF TABLES .................................................... LIST OF FIGURES ................................................... ix LIST OF ABBREVIATIONS ............................................. xv CHAPTER 1: A REVIEW OF NERVE GROWTH FACTOR AND ITS EFFECT ON NEURO-ONCOGENESIS ................................................. 1 Nerve Growth Factor - Historical Perspective ................ l Nerve Growth Factor ......................................... 2 Nerve Growth Factor and Differentiation of Neurons .......... 4 NGF Bioassay ................................................ 5 Mechanism of Action ......................................... 5 Nerve Growth Factor Receptor ................................ 6 NGF Gene .................................................... 6 NGF and Neuro-oncology ...................................... 6 NGF and N-ethyl-N-nitrosourea-induced peripheral nerve Neurinomas .................................................. 7 Ethylnitrosourea: Neurocarcinogenesis ...................... 8 ENU and DNA Alkylation ...................................... 9 Repair of Alkylated DNA and Mutation ........................ 10 Role of Oncogenes in N-nitrosourea-induced neurogenic tumors ...................................................... 10 REFERENCES ........................................................ 12 CHAPTER 2: THE EFFECT OF NERVE GROWTH FACTOR (NGF) ON TRANSPLACENTAL ETHYINITROSOUREA (ENU)-INDUCED NEUROGENIC TUMORS IN SPRAGUE-DAWLEY RATS ........................................... 20 SUMMARY ........................................................... 20 INTRODUCTION ...................................................... 22 MATERIALS AND METHODS ............................................. 25 Preparation of NGF .......................................... 25 Ethylnitrosourea ............................................ 25 Animal Experiments .......................................... 25 Histopathology .............................................. 26 Development of Nerve Growth Factor Receptor Positive Control ..................................................... 26 Immunohistochemistry: Detection of NGF-R ................... 27 Statistical Analysis ........................................ 28 viii TABLE OF CONTENTS (cont..) Page RESULTS ........................................................... 29 Incidence of Trigeminal Nerve Neurinoma ..................... 29 Incidence of CNS Tumors ..................................... 34 Classification of Central Nervous System Tumors ............. 34 Oligodendroglioma ........................................... 35 Mixed Glioma ................................................ 35 Astrocytomas ................................................ 35 Glioependymoma .............................................. 35 Meningioma .................................................. 35 Optic Nerve Glioma .......................................... 36 Non-neural Neoplasms ........................................ 36 Nerve Growth Factor-Receptor ................................ 36 DISCUSSION ........................................................ 57 REFERENCES ........................................................ 59 CHAPTER 3: THE m 11139 EFFECTS OF NERVE GROWTH FACTOR (NGF) AND GLIA MATURATION FACTOR (GMF) ON ANAPLASTIC GLIOMA T9 CELL LINE: SCANNING AND TRANSMISSION ELECTRON MICROSCOPY STUDIES ............. 63 SUMMARY .......................................................... 63 INTRODUCTION. .................................................... 64 MATERIALS AND METHODS ............................................ 65 Preparation of NGF and GMF ................................. 65 Cell Culture ............................................... 65 Transmission Electron Microscopy (TEM) ..................... 67 Scanning Electron Microscopy (SEM) ......................... 69 RESULTS .......................................................... 7O Scanning Electron Microscopy ............................... 70 Control T9 Cells ........................................... 70 GMF-treated Cells .......................................... 70 NGF-treated Cells .......................................... 70 Transmission Electron Microscopy ........................... 71 Control T9 Cells ........................................... 71 GMF-treated Cells .......................................... 71 NGF-treated Cells .......................................... 72 DISCUSSION ....................................................... 126 REFERENCES ....................................................... 129 vi LAKE: TABLE OF CONTENTS (cont...) Page CHAPTER 4: IMMUNOHISTOCHEMICAL CHARACTERIZATION OF CENTRAL AND PERIPHERAL NERVE TUMORS INDUCED BY ETHYINITROSOUREA IN RATS UTILIZING ANTI-GLIAL FIBRILLARY ACIDIC PROTEIN (GFAP), ANTI-LEU 7, ANTI—S-lOO PROTEIN ANTIBODIES ................................. 132 SUMMARY .......................................................... 132 INTRODUCTION ..................................................... 133 MATERIALS AND METHODS ............................................ 136 Tumors ..................................................... 136 Immunohistochemistry ....................................... 136 Glial Fibrillary Acidic Protein (GFAP) ..................... 136 Procedure .................................................. 137 S-lOO Protein Stain - ABC Vectastain ....................... 138 Human Natural Killer-1 Monoclonal Antibody (HNK-l MAB) ..... 138 PAP ........................................................ 138 Preparation of Reagents .................................... 140 PBS (Phosphate Buffer Saline) pH 7.4 ....................... 140 Trypsin Solution ........................................... 140 Diaminobenzidine-Hydrogen Peroxide Substrate Solution ...... 140 RESULTS .......................................................... 141 Neurinomas ................................................. 141 S-lOO Protein .............................................. 141 GFAP. . . .‘. .................................................. 141 Astrocytoma ................................................ 141 8-100 ...................................................... 141 GFAP ....................................................... 142 Oligodendrogliomas ......................................... 142 8-100 ...................................................... 142 GFAP ....................................................... 142 Mixed Gliomas .............................................. 143 Glioependymomas ............................................ 143 Meningiomas ................................................ 143 Anti-Leu 7 ................................................. 143 DISCUSSION ....................................................... 162 REFERENCES ....................................................... 166 VITA ............................................................. 170 vii LIST OF TABLES Page [2%. effect of NGF on trigeminal nerve neurinomas and ENPs ngoeurring after ENU adainistration......................... 29 I~~~ effbct of NGF on peripheral nerve neurinomas occurring its: ENU administration................................... 30 ?‘ occurring after ENU administration......................... 31 influx-cuties 32 ‘J A \ L'; (.l u A. - A . ,1... . ... . ..v‘l Tm. . x 40-14. ‘ ’l' ("3.1 n ‘ ' '! ml" ' 3‘ ' ~ ,Vx'afi‘u' L -. ' . .‘ x w... 9'. ~. ..v :' ?r{,&u‘.m "-7 l.u"‘-‘:~. H‘“. L' t. ' ‘. "“ .. - “lii‘n. { '0' $-3DE (3) Mg: .rfl. 214:. up} ..i‘ :- '3 5'... ; int, CL!‘(_.\{1): i . . ‘ 5.71:; s; 2.1‘ .r'. . ' rum. .. )g .c..'.- .5 {pram} , 3:5 ‘£.A 85,59;- .‘ i 5". ‘ -." .4. f viii ’ f e ‘ ~ ~ 5 ‘ r- ‘- . u _ " < , Figure/Page 1-1/37 1-2/37 1-3/39 1-4/39 1-5/41 1-5/41 1-7/43 1-8/43 1-9/45 1-10/45 1-11/47 1-12/49 1-13/49 1-14/51 1-15/51 LIST OF FIGURES Ventral view of rat brain. Notice bilaterally swollen trigeminal nerves (arrows). Dorsal view of rat trigeminal nerve affected with anaplas- tic neurinoma (1); brain reflected dorsally (2). Ventral view of rat brain showing bilateral anaplastic trigeminal nerve neurinoma (l); hemorrhage (2). Photomicrograph of an anaplastic neurinoma illustrating hyperchromatic polyhedral cells and mitoses (arrow). H & E. X 400. Photomicrograph of an anaplastic neurinoma showing areas of necrosis and hemorrhage. H & E. X 400. Photomicrograph of an early neoplastic proliferation (ENP) in a trigeminal nerve. Arrow indicates CNS-PNS junction. H & E. X 160. Gross appearance of an oligodendroglioma in the lumbo-sacral region of a spinal cord. Photomicrograph of an oligodendroglioma. Notice small round cells with scant cytoplasm and dense nuclei arranged in a "honeycomb"-like pattern. H & E. X 160. Photomicrograph of a cerebral mixed glioma illustrating! mixed population of neoplastic cells. H & E. X 400. Photomicrograph of cerebral astrocytoma. H & E. 400 X Photomicrograph of a cerebellar glioependymoma. Neoplastic cells are arranged in short chains and rosettes. H & E. X 400. Gross appearance of'a meningioma involving the entire dorsal cerebral hemisphere. Photomicrograph of the meningioma in Figure 1-12. Notice oval to fusiform neoplastic cells forming nebulous whorls. H & E. 400 X. Ventral view of a rat brain showing bilateral trigeminal nerve neurinoma (1); optic nerve glioma (2); globe (3). Photomicrograph of optic nerve tumor. Glioma (1); intraocular neurinoma (2); retina (arrow). H & E. X 60. ix Figure/Page 1-16/53 1-17/55 1—18/55 2-1/74 2-2/76 2-3/78 2-4/80 2-5/82 2-6/84 LIST OF FIGURES (cont....) Photomicrograph of a neurinoma immunostained for nerve growth factor receptor (NGFR). Notice strong NGFR positive cytoplasmic reactions in neurinoma cells. ABC method. Hematoxylin counter stain. X 100. Photomicrograph of 7-day transected rat sciatic nerve immunostained for NGFR (positive control). Schwann cells have strong positive reaction around the cytoplasmic rim. An unstained blood vessel (center). ABC method. X 100. Photomicrograph of a neurinoma immunostained for NGFR. Notice all neoplastic cells are negative. ABC method. Hematoxylin counterstain. X 100. Scanning electron micrograph of control anaplastic T9 glioma cells grown in HL-l media. The cells are variable in shape and size with disorientated piling-over growth pattern. Scant microvilli are discernible along the cytoplasmic borders. (440 X). Scanning electron micrograph of a single control T9 cell. Notice the broad multipolar cell body with thick foot processes. Variably-sized microvilli are scattered randomly over the cell body and along the cytoplasmic edges. (2200 X). Scanning electron micrograph. The bilayered control T9 cells have specialized surface spherical structures or bulbous excrescences called zeiotic blebs. These exocytes are associated with cells during active mitosis. (960 X) Scanning electron micrograph of anaplastic T9 glioma cells after exposure to GMF for 4 days. Notice cells are markedly reduced in size and have long cytoplasmic pro- cesses which form an interconnecting somatic network with the processes from the neighboring cells. (440 X) Scanning electron micrograph of anaplastic T9 glioma cells after exposure to NGF for 4 days. Notice multipolar cells are relatively reduced in size and have broad cytoplasmic expansions or lamellopodia from which protrude extremely fine thread-like processes, filopodia. The cells are con- nected by broad bands of cytoplasmic projections. (X 720) Scanning electron micrograph. High magnification of T9 glioma cells in Figure 2-5. Cytoplasmic processes appear to blend with processes from adjacent cells forming an intricate network. The free ends of the projections have delicate expansion of lamellopodia (La) and fine filopodia (F1). (1200 X) l -.., Figure/Page 2-7/86 2-8/88 2-9/90 2-10/92 2-11/94 2-12/96 2-13/98 2-14/100 LIST OF FIGURES (cont...) Scanning electron micrograph of a single T9 glioma cell 4 days after exposure to NGF. Notice poles of the cytoplas- mic projections are decorated with extremely fine and deli- cate film of lamellopodia, filopodia and secondary branches. The body surface and cytoplasmic edges are sparsely covered with microvilli. (1560 X) Scanning electron micrograph. High magnification of a cytoplasmic projection of NGF-treated T9 glioma cell as seen in Figure 3-7. Notice elaborate and delicately expanded lamellopodia (La), filopodia (F1) and secondary branches (S). (3600 X) High magnification of terminal portion of cell structure in Figure 2-8. Transmission electron micrograph of control anaplastic T9 glioma cells grown in HL-I media for 4 days. Notice the cells are large and polyhedral in shape with high nuclear: cytoplasmic ratio. The nuclear contour is rough and sometimes jagged. The nucleus contains multiple dense nucleoli. Notice in the cytoplasm there is rudimentary mitochondria, SER, RER and free ribosomes. Microvilli are rare along the cell border. (3400 X) Transmission electron micrograph. High magnification of the anaplastic T9 glioma cell in Figure 2- 10. Notice in the perikaryon 5nm microfilament (mf) dispersed among RER, SER, mitochondria (m), and free ribosomes (r). n; nucleus. (34, 200 X) Transmission electron micrograph of T9 glioma cell after 4 days of exposure to GMF. The cells are relatively reduced in size with low nuclearzcytoplasmic ratio. The nucleus is irregular and contains condensed marginated chromatin material. Abundant dense mitochondria, SER, RER, myelin figures, and free ribosomes are present in perikaryon. Numerous wavy microvilli project from cell edges. Notice several points of somatic attachment with the neighboring cells. (3400 X) Transmission electron micrograph of T9 glioma cells after 4 days exposure to GMF. Notice growth of fine filamentous cytoplasmic processes. (3400 X) Transmission electron micrograph of T9 glioma cells after 4 days exposure to GMF illustrates the intimate intertwin- ing of microvilli between opposing cells. (4500 X) xi Figure/Page 2-15/102 2-16/104 2-17/105 2-18/108 2-19/110 2-20/112 2-21/114 2-22/116 2-23/118 LIST OF FIGURES (cont....) Transmission electron micrograph. High magnification of T9 glioma cells 4 days after exposure to GMF. The perikaryon has abundant parallel rows of 25 nm microtubules (mt), 10 nm intermediate filaments (IF) and wavy bands of microfilaments (mf). These cytoskeletal frameworks are intimately associated with RER. mv; multivesicular body. (27,000 X) Transmission electron micrograph of a cytoplasmic process in T9 glioma cell 4 days after exposure to GMF. Notice stacks of microtubules, intermediate filaments and micro— filaments arranged along the long axis of the body of the cytoplasmic filopodia. (34,200 X) Transmission electron micrograph of secondary branch of a filopodia in Figure 2-16. Equal complements of the cyto- skeletal support extends into the terminal branches. Transmission electron micrograph. High magnification of T9 glioma cell after exposure to GMF. Notice the gigantic mitochondria are closely associated with streaks of micro- filaments and intermediate filaments. (45,000 X) Transmission electron micrograph of T9 glioma cell after 4 days exposure to NGF. The condensed cells have low nuclearzcytoplasmic ratio with smooth nuclear contour and evenly dispersed heterochromatin. A - elonaged mitochondria; L - lipid inclusions. (4500 X) Transmission electron micrograph of T9 glioma cell after 4 days exposure to NGF. Notice profuse growth of tortuous villi from the cytoplasmic borders (arrow). M - myelin figure; A - elongated mitochondrias; R - rough endoplasmic reticulum. (4500 X). Transmission electron micrograph of a T9 glioma cell 4 days after exposure to NGF. The perikaryon has many microfilaments (m) and intermediate filaments (f). Transmission electron micrograph of T9 glioma cells after exposure to NGF. Notice streaks of microfilaments, inter- mediate filaments and microtubules converge around the two centrioles (c). R - rough endoplasmic reticulum; G - dilated golgi apparatus. (19,800 X) Transmission electron micrograph of T9 glioma cells after exposure to NGF. The figure illustrates terminal portion of a cytoplasmic projection (P), lamellopodia (L), and filopodia (F) corresponding to the SEM details of cells in Figure 2-8. (3420 X) xii Figure/Page 2-24/120 2-25/122 2-26/124 2-27 3-1/141 3-2/141 3-3/146 3-4/148 3-5/150 3-6/150 LIST OF FIGURES (cont....) Transmission electron micrograph. High magnification of cytoplasmic projection (P) as in Figure 2-21. The long axis of the process is enriched with parallel bundles of micro- filaments, intermediate filaments and microtubules. Note that these cytoskeletal frameworks are in close association with rough endoplasmic reticulum and mitochondria. (45,000 X). Transmission electron micrograph. High magnification of filopodium as in Figure 2-21. The process contains rich complement of structural support. Transmission electron micrograph of T0 glioma cells after exposure to NGF (Figure 2-26) and GMF (Figure 2-27). Note the junctional complex, zonula occludens (JC) between two opposing cells. The outer leaflets of the opposing cell membrane form a single intermediate dense line. (81,000 X) Photomicrograph of a trigeminal nerve neurinoma immuno- stained for S-100 protein. CNS-PNS junction (arrow). Immunoperoxidase, ABC method; Hematoxylin counter stain. 400 X. Higher magnification of trigeminal nerve neurinoma in Figure 3-1. Notice positive reaction on cell membranes, in cyto- plasms and nuclei. 640 X. Photomicrograph of trigeminal nerve neurinoma immunostained for GFAP. The neurinoma cells are negative, whereas the astrocytes in the CNS are strongly positive. CNS-PNS junction (arrow). Immunoperoxidase, ABC method. Hematoxy- lin counterstain. 400 X. Photomicrograph of an astrocytoma immunostained for S-lOO. Immunoperoxidase, ABC method. Hematoxylin counterstain. 400 X. Photomicrograph of a cerebral astrocytoma immunostained for GFAP. Immunoperoxidase ABC method. Hematoxylin counter- stain. 400 X Photomicrograph of a cerebral undifferentiated astrocytoma immunostained for GFAP. Notice peripheral large reactive astrocytes stained intensely while an occasional astrocytoma cell within the tumor showed a weak reaction. Immunoperoxidase, ABC method. Hematoxylin counterstain. 160 X. xiii Figure/Page 3-7/152 3-8/154 3-9/156 3-10/156 3-11/158 3-12/158 3-13/160 LIST OF FIGURES (cont....) Photomicrograph of an oligodendroglioma immunostained for 8-100. Notice strong positive reaction in the cytoplasm and processes of reactive astrocytes within the tumor. The small oligodendroglioma cells showed weak positive reaction (arrow). Reactive fibrillary astrocytes (arrowhead). Immunoperoxidase, ABC method. Hematoxylin counterstain. 400 X. Photomicrograph of an oligodendroglioma immunostained for GFAP. Notice intense positive reactions in the reactive fibrillary astrocyte within the tumor. The oligodendro- glials are negative for GFAP. Immunoperoxidase, ABC method. Hematoxylin counterstain. 400 X. Photomicrograph of a mixed glioma immunostained for 8100. Notice strong positive reaction in the broad astrocytic cytoplasm and its thin processes. The oligodendroglioma cells do not stain for 8100. Immunoperoxidase, ABC method. Hematoxylin counterstain. 400 X. Photomicrograph of a mixed glioma immunostained for GFAP. The astrocytic component of the tumor shows strong positive reaction, whereas the oligodendroglioma cells are uniformly negative. Immunoperoxidase, ABC method. Hematoxylin counterstain. 400 X. Photomicrograph of a glioependymoma immunostained for S-lOO. Notice many positive cells forming the pseudo- rosettes. Immunoperoxidase, ABC method. Hematoxylin counterstain. 400 X. Photomicrograph of a glioependymoma immunostaind for GFAP. Notice several tumor cells have positive stain in the thin membraneous cell processes. Immunoperoxidase, ABC method. Hematoxylin counterstain. 400 X Photomicrograph of a meningioma immunostained for 8100. Notice relatively strong reaction in cells surrounding the blood vessels. Immunoperoxidase, ABC method. Hematoxylin counterstain. 400 X. LIST OF ABBREVIATIONS NGF ........ nerve growth factor NGF-R ...... nerve growth factor receptor GMF ........ glia maturation factor ENU ........ ethylnitrosourea CNS ........ central nervous system PNS ........ peripheral nervous system CD ......... cessarian derived NACl ....... sodium chloride DAB ........ diaminobenzidine PC-12 ...... pheochromocytoma cell line KD ......... kilo dalton KI-RAS-Z...kirstein-ras gene HA-RAS-l...harvey-ras gene PAP ........ peroxidase-anti-peroxidase ABC ........ avidin-biotin-complex {SW9 1’“. Rants 9‘ - d Swim. . ““19 es flu t .~ ... ‘01 «or - JV ,« 6,»!!! born-w ‘ . m 1 fl tee....e ( ... a . norm-owns Immunisation!“ um . . . A ”...-amass: teem . .. (was... m.» vac-Mm . (a ., . , ....“ ... fpfi““t .' _ ‘ .. m CH! 053'.“ . ."t‘d‘ ‘9‘) o :wt‘C-es 4.. ‘17:: --.,n;. ... . ii; ”314. ”a ‘Qet ”if: m i...» ”w M Uni kw! mate-4h -- may; alu- ‘ if. * W “4 (:AR'LQA9:1“L'W '1 "r.‘5& my'm - ‘4‘ ‘,',:"” 4' 3-3 any. an»: mm wt z.» was: sent WWW La 4“\ '.." We‘d rh- m seams «no. ”W, ‘ ‘34.. v“ vetoes elm seq-ens to d» MWNW ..g L 3'"- c , u- m- an mime-age W ,4 1"qu Ngrxg nggth Factor - Historical perspegtive Nerve growth factor discovery was incidental to a bold and imagina- tive experimental manipulation with fragments of several neoplasms by Elmer Bueker (1948).1 In his experiments, he implanted pieces of mouse sarcoma 180, Rous fowl sarcoma and mouse mammary adenocarcinoma into the body wall of 3-day-old chick embryos to study the effect of foreign grafts on the development of the nervous system. In these experiments, a fragment of wing bud was surgically removed from the implantation site. It had been previously shown that wing bud ablation resulted in a severe hypoplasia of ventral horn motor nerve cells in that spinal cord hemisec- tion.2 It became apparent that, following the ablation, hypoplasia of nerve centers resulted from death of differentiated neurons, and not from failure of precursor cell replacement. He postulated that there were unidentified physiochemical affinities which resided in the peripheral appendages (wing) that were responsible for inducing the outgrowth of nerve fibers from the spinal cord3. Thus, Bueker selected tumor cells of diverse genetic origin for the limb bud substitute, to assess the effect on the development and differentiation of spinal ganglion cells. Within 3-5 days after tumor grafting, dorsal root ganglia sensory nerve fibers innervated the mouse sarcoma cells. Concurrently, the gang- lia increased in volume when compared to the corresponding ganglia inner- vating the intact contralateral wing. Bueker repeated this experiment, placing the tumor fragments on the chorioallantoic membrane. In spite of physical separation from the underlying embryo, implanted tumor cells induced enlargement and profuse growth of the sympathetic nervous system. 2 He postulated that a soluble, diffusable substance in the tumor fragments had triggered the nerve fibers to sprout. Cohen, Levi-Montalcini and Hamburger attempted (1954) to extract and purify from mouse sarcoma cells the fraction responsible for the nerve growth - promoting activity. This led to the unfolding of another impor- tant chapter in the discovery of nerve growth factor. Snake venom con- taining nucleic acid-degrading enzyme phosphodiesterase and other enzymes, was used to degrade the nerve growth-promoting fraction.“ An unexpected result from the experiment revealed that snake venom itself contained a much higher fraction of the nerve growth-promoting molecule. This fact was later confirmed in 6 to 8 day old chick embryos.5 A survey of organs homologous to venom producing tissues revealed that mouse submandi- bular salivary gland was the richest source of the compound which was later named nerve growth factor (NGF).5 ngxg Qrgggh Eggggr. Nerve growth factor (NGF), a polypeptide extracted from salivary glands, exists in two different forms (78 and 2.58 NGF), depending upon purification procedures.6 Both forms have identical physiological significance but differ slightly in molecular composition due to minor proteolytic damage during preparation.7 The high molecu- lar weight protein, (130,000, [or the 78]) isolated by Varon g; 518 is a noncovalently linked complex of two alpha (2.78), one beta-dimer (2.68) and two gamma (2.58) subunits, with the beta-dimer being the active mole- cule.9 The weak association of the three subunits and their different isoelectric points allows separation by ion-exchange chromatography.8 2+ The complex subunits are structurally bound by one or two Zn mole- cules. Removal of Zn molecules or pH < 5 or > 8 causes dissociation of the complex.14'15 3 The biologically active 2.58 NGF form is most commonly prepared from partially purified NGF according to the method of Bocchini and Angeletti.9 The purified 2.58 NGF and intact beta-subunit differ only in an amino-terminal octapeptide and carboxyl-terminal arginine.lo’11 The biological significance of the alpha and the gamma subunits is not known, but it has been postulated that they protect and store NGF in the salivary glands, presumably participating in the formation of the NGF from a larger biosynthetic precursor.12’13 To this day, the mouse submaxillary gland remains the richest source of NGF.6 The NGF-type protein has also been purified to homogeneity from snake venom (MW 28,000, 2.28), and both preparations, although dif- fering minimally in molecular weights and sedimentation coefficient, pos- sess similar biological properties and activities.16’17 The biological role of this salivary protein remains undefined, as well as its potential for access to the nervous system. Thoenen g; 5118 and Green g; £119 ruled against the possible discharge of this NGF into the circulating blood and also demonstrated that a surgical removal of this organ had no adverse effect on the sympathetic and sensory cells. Increased production (10 fold) by male mice has no defined physiologic basis. Recent studies indicate that the synthesis of the protein molecule is controlled by testosterone and thyroxine, and a discharge of unusually high levels of NGF into the circulating blood of male mice during intra- species fighting may be instrumental in triggering a more aggressive defensive or offensive behavior.2o’22 The mechanism of NGF discharge into the blood stream and the associated temperamental change in the male is not known. 4 Nerve growth factor is contained in varied amounts in many tissues of the body. Sympathetic ganglia, adrenal glands, kidneys, blood vessels, vas deferens and peripheral organs innervated by the sympathetic nervous system have higher levels of NGF than thymus, placenta, heart, spleen, liver and muscle.23'26 The protein has been identified in fish, birds, reptiles, mammals, and amphibians.23'27’28 Several cells in culture are capable of producing NGF, including 3T3 cells, SV40 cells, fibroblasts, myoblasts, melanoma cells, glioma cells, neuroblastoma cells and glioblas- toma cells.29'34 NEE and Differengigtion of Neurons; Nerve growth factor is required for the development and survival of sympathetic and sensory neurons in the peripheral nervous system and differentiation of neurons in the central nervous system.6'7’35'37 Levi-Montalcini and Booker demonstrated the absolute NGF requirement of sympathetic neurons by a pharmacological experiment termed immunosympathectomy.38'39 Daily injections of small amounts of anti-NGF antibody into neonatal rodents resulted in an almost complete disappearance of sympathetic chain ganglia.40 Parenteral ad- ministration of NGF to neonatal animals produced an appreciable increase in size of the ganglia and earlier innervation of sympathetic end-organs by nerve fibers.28 Levels of tyrosine hydroxylase and dopamine- hydroxylase, enzymes required for catecholamine biosynthesis, increase dramatically/‘1 During developmental innervation, the target tissue provides NGF, which is retrogradely transported from the synaptic junction to the neuro- nal some, for maintenance of the differentiated state of the neurons.42 5 NEIthgggfigy; The two most common in 21;;2 bioassays for NGF activ- ity are the phenotypic changes elicited in rat pheochromocytoma cells (PC-12),43 and the response of chick embryo ganglia to NGF exposure.44 The chromaffin-derived neoplastichC-12 clonal cells, not requiring NGF for survival, are provoked to sprout neurite-like processes by NGF exposure.45 This phenotypic alteration is characteristic of differentiated neurons biochemically and ultrastructurally.46 The sec- ond assay is based on the graded pattern of neurite outgrowth elicited from the chick embryo dorsal root ganglia after inoculation with NCF.28'44 Although this assay is still in use, it has minor drawbacks in that the quantitation is somewhat subjective in scoring with limited accuracy and requires dissection of the embryonic chick tissues.28 The PC-l2 method, while being a simple and rapid test, offers more accurate results and is also highly sensitive and easily reproducible.45 Egghgnigm 2f gction: The actual mechanism by which NGF exerts its multiple effects on target cells remains ill-defined. However, the ini- tial step of the interaction between NGF and its target cells requires coupling of the protein to specific cell surface receptors-“47'48 This complexation results in several rapid cell conformational and functional changes, with rapid uptake of metabolites, polymerization of tubulin to microtubules and stimulation of anabolic and catabolic pathways.49 Several schemes have been postulated to explain the above NGF-mediated actions, including activation of secondary messengers such as cAMP and 2+ 50 51 intracellular Ca mobilization, 52 protein methylation, phosphati- and by induction of cyclic AMP-Ca+2/phospho- 53 dylinositol turn-over lipid-dependent protein kinases. l" - 6 Within a few hours following coupling of NGF and receptor, the com- plex is internalized by endocytosis, partially engineered by micro- tubules.48 The endocytotic vesicles containing the hormone-receptor complex have two possible fates: (1) fusion with lysosomes leading to degradation or (2) translocation to nuclear receptors.48’54 The latter pathway directly affects subsequent transcriptional events.54 This concept is supported by detection of elevated levels of transcription-dependent enzymes, such as ornithine decarboxylase and tyrosine hydroxylase follow- ing administration of NGF.12'41 NgIyg_§;gg§h_£§g£gr_fig§gp§gr: The nerve growth factor receptor (NGFR) is a glycoprotein with a molecular weight of 70,000.55"56 It is localized uniformly on cell bodies and growing neurites.57 Two forms of NGFR have been identified; high affinity and low affinity NGFR with de of 10'11M and 10'9M, respectively.58 It is believed that the low affinity receptor may give rise to the high affinity receptor.59 NQELQgggz The first genetic information of NGF was achieved from the gene sequencing studies of mouse salivary NGF.60 This discovery led to the identification of NGF cDNA and the cloning of the NGF gene in man and animals.61’62 In humans, the NGF gene is located on the proximal short area of chromosome 1, and codes for 307 amino acid residues which, upon fractionation, yield the essential 118 amino acids that constitute the NGF protein.63 u - c : Several investigations on the possible role of NGF and its antibody in the suppression of tumors arising from NGF responsive neural crest cells have been performed.l‘9'6l"'68 These ex- periments have shown that NGF has both maturation and differentiation 7 influence on neurectodermal tumor cells. Exposure of PC-12 pheochromocy— tome cells to NGF resulted in cessation of mitosis, induction of neurite outgrowth,45 induction of neuron-specific enolase and ornithine decarb- oxylase, and stimulation of amino acid uptake.69'71 NGF also induced differentiation of human neuroblastoma line IMR-32,66 and SH-SY5Y cells.72 In an in 3112 experiment, NGF treatment of recipient rats bearing implanted undifferentiated F-98 glioma clone cells resulted in decreased tumor growth rate and increased life span of the animals.73 Pretreat- ment of the malignant cells with NGF 24 hours prior to implantation showed a similar but less dramatic result.73 In a recently completed in giggg study in our laboratory, it was shown that treatment of anaplastic glioma T9 cells with NGF retarded the growth and induced morphological changes characteristic of a differen- tiated neuroepithelial cell.74 - - - t so ea- uced e s: It has been shown that NGF is capable of suppressing neurinomas in rats and mice transplacentally induced with N-ethyl-N-nitrosourea (ENU).67'68'75’76 ENU selectively induces tumors of the nervous system in rats when exposed transplacentally as fetuses during late gesta- tion.77 A single dose of 50 mg/kg ENU administered via the lateral tail vein to pregnant rats at 20 days of gestation results in the production of neurogenic tumors in nearly 100% of the offspring.77 Neurinomas of the trigeminal nerves appear as early as 20 days post-partum at which time the 8 lesions are classified as early neoplastic proliferations (ENP). By 90 days post-exposure, nearly 100% of the rats are affected.67'76'78 Administration of NGF either postnatally after optimal dose ENU ex- posure or transplacentally prior to optimal dose ENU exposure led to a significant reduction of ENP and trigeminal neurinomas at 90 days of age in the offspring.67’76 Wm: Lu—LMeu o are no e es Ethylnitrosourea (ENU) is a potent resorptive neurocarcinogen in a variety of rat strains.79'80 It is a direct-acting carcinogen belong- ing to the Acyl-Alkyl nitrosomides which non-enzymatically activate to an ultimate carcinogen by dissociating rapidly in a slightly alkaline environment to release the reactive electrophil.81 Transplacental ad- ministration of ENU to pregnant rats in late gestation or to newborns soon after birth selectively results in tumors of the central and peripheral nervous systems.79'80’83 The degree of sensitivity of the developing nervous system to ENU is directly related to the gestational age of the fetuses, the dose and route of ENU administration.79'84 Susceptibility extends from the 12th day of gestation until 2 weeks after birth, with a maximum sensitivity at 15 days of gestation at which time predominantly neural tumors of the cen— tral nervous system will develop.79'85 Fetal rats exposed to ENU after the 15 day gestation develop more peripheral nerve tumors.71’76 A teratogenic effect is observed when ENU is administered before the 12th day of gestation.83’85 The degree of oncogenic effect declines postnatally, and after 30 days of age, the nervous system is relatively resistant to neoplastic transformation by a single dose of ENU.83'86 9 After an optimal transplacental ENU dose, tumors of the peripheral nervous system have the shortest latent period followed by those of the central nervous system.83’85 The peak of sensitivity varies with different structures in the nervous system and is further dictated by age, species and strain.79’82 In the CNS oligodendrogliomas, ependy- momas, or mixed gliomas are primarily located in the hippocampus, subepen- dymal areas, lateral ventricles, cerebral cortical white matter, thoraco- lumbar and lumbosacral region of the spinal cord.86 Neurinomas of the trigeminal nerve are first to appear followed by those of spinal cord nerve roots, lumbar and brachial plexuses, and sciatic nerve. ENH_§nQ_DEA_Alkyl§§ignz The one most important property of N- nitroso compounds is the capability of alkylating DNA bases.87 Although alkylation can occur at various sites on nitrogen and oxygen atoms in the DNA bases, it is the ethylation of the 06-guanine which is one of the prime suspects contributing to oncogenesis.87'88 Other adducts also formed are 02-ethylcytosine, 04- and 02-ethylthymine.88’89 Although the major products derived from ENU reactions with DNA are ethyl- phosphotriesters, their contribution to mutagenic activity is not known.87'89 It is known that most alkylating agents are either carcino- genic, mutagenic or both.88 The precise molecular mechanism directing mutagenesis by alkylating agents is still obscure; however, experimental evidence supports the hypo- thesis that base-substitution mutations arise via the formation of alkyl- DNA adducts that could direct the misincorporation of nucleotides during DNA replication.89 The alkylation of 06 position of guanine could cause a mutation during DNA replication resulting from a miscoding with thymine instead of with the complementary cytosine base via the DNA poly- merase enzyme.90 10 BflPa1I_2jLAlE!l§£§d_DNA_ang_Mu§atign: The persistence of alkylated DNA base(s) during cell replication is necessary for mutation to occur. After an exposure to ENU, brain, liver and kidney have comparable levels of OG-alkylguanine in DNA.90 However, within a few hours, liver and kidney efficiently remove the DNA adduct by a catalytic activity which correlates with the levels of the repair enzyme alkylguanine transferase in these organs.87 In the repair of O6-alkylguanine by the catalytic transferase enzyme, the alkyl group (methyl or ethyl) is transferred to a protein residue without any structural damage to the liberated DNA.91 The rate of 06-alkylguanine loss from the brain (target organ) is con- siderably less efficient and corresponds to the low level of the repair enzyme activity detected in the developing nervous system.90 The fail- ure to correct the DNA damage provides the high risk for the persistence of the pro-mutagenic lesion in rapidly replicating cells such as the glia. It is apparent, therefore, that alkylation of DNA, with persistence of alkylated adducts and subsequent DNA replication and cell multiplica- tion are necessary for early initiation of tumor growth by the N-nitroso compounds. Belg 2f Qnggggngg in N-nitrogourea induggd neurogenic tumgrg Investigations have shown the association of transforming genes (oncogenes) in chemically induced neoplasms in rodents.92'94 Harvey- ras gene (Ha-ras-l), in methylnitrosourea-induced mammary carcinomas in Buf/N female rats,92 the Kirsten ras gene (Ki-ras-2) in renal mesenchy- 93 94 mal tumor in rats, and the N-ras gene in thymic lymphoma in mice have been constantly found to be activated during tumor induction. ..... 11 F studies have attempted to show the possible role of neu in er/gliobleetoma cell lines derived from tumors induced by . retegs and in ENU induced Schwannomae in P344 rate.96 - ' of this transforming gene is believed to be consequential to a 97 3, ~ tion in the transmembrane domain of neu gene product. ‘ .: :9" ' “QR I“. '1 "'. eye... A“, V N .. i. H eentxa' x V. 4 .(‘DH 1 .mlmentn .r 9: , 8, Levi V \ . 1.”. I. a... " W 1015 w M «telcin. 'ni . ”OCIJJ‘HHH. ‘ef" .Q U “3:691! ‘1, 9" - n. “(V. f 31"}. 5'!" ‘ n I Q“. shoorfit V‘l n fecror, AW 34’ - . ’- are. i. I. “We J. St. N. :.'.' 1"4 ET "-1 , 196i , low." vet as: pMLJJ. .Ichelter 0. ~21: Bf an, names new r- :w- a... It”! :5: me» x was. Ian—t. N . V- .‘itku 3 A k . 1 . g > ‘ ‘ u l.” T‘s-- H ; Mutts-me “at; u‘ gun .~. Fee‘s . W”. w “it: tawdu‘ , umnmmm - I I 0“ 4 a, . ' o ..g ,(r , . _ z . I ; "-9 ‘7. \ ; \‘ --' - .'- I n'i'fl'. REFERENCES wanna 1 L ' '. r; \ _‘..‘....1 '0 a. J“. .‘x w, 1 tot prcrwin it. o Mgr «fin n o zs‘u .q.v:. ham‘i...‘ V Angelou! W ‘m- w" “as. age.» {fig-hm“ “tit ,d‘““h “dim «w-«wmwmc w m m» ‘mq‘m: bonnet. W -~ - (scrotum m I“. 'Isfleenc; umgmmm: Md“. F «ed-.511 10. ll. 12. 13. 14. REFERENCES Bucker, ED: Implantation of tumors in the hind limb field of the em- bryonic chick and the developmental response of the lumbosacral ner- vous system. 53;; Egg 102:369-389, 1948. Hamburger V: The effect of wing bud extirpation on the development of the central nervous system in chick embryos. 1 Exp Zool 449-494, 1934. Hamburger V, Levi-Montalcini R: Proliferation differentiation and degeneration in the spinal ganglia of the chick embryo under normal and experimental conditions. J Exp Zool 111:457-501, 1949. Cohen S, Levi-Montalcini R, Hamburger V: Nerve growth stimulating factor isolated from sarcoma 37 and 180. Proc Natl Acad fig; ggg 40:1014-1018, 1954. Levi-Montalcini R, Cohen S: In vitro and in vivo effects of nerve growth stimulating agent isolated from snake venom. Eroc Natl Acad $91 ESL 42:695-699, 1956. Bradshaw RA: Nerve growth factor. Ann Egg fiiochem 47:191-216, 1978. Yanker BA, Shooter EM: The biology and mechanism of action of nerve growth factor. Ann ggy fiioghem 51:845-868, 1982. Varon S, Nomura J, Shooter EM: The isolation of the mouse nerve growth factor protein in a high molecular weight form. fiioghemistry 6:2202-2209, 1967. Bocchini V, Angeletti PU: The nerve growth factor: purification as a 30,000-molecu1ar-weight protein. 219; Egg; Aggg figi HSA 64:787- 794, 1969. Mobley WC, Schenker A, Shooter EM: Characterization and isolation of proteolytically modified nerve growth factor. fligghgmigtzx 15:5543- 5552, 1976. Moore JB, Mobley WC, Shooter EM: Proteolytic modification of the .-nerve growth factor protein. fligghgmifigzx 13:833-840. 1974. Vinores S, Guroff G: Nerve growth factor: mechanism of action. Ann 3:! filanhxg biggng 9:223-257, 1980. Greene LA, Shooter EM, Varon S: Enzymatic activities of mouse nerve growth factor and its subunits. 2:9; Natl Aggg 591 ufiA 60:1383-1388, 1968. Pattison SE, Dunn MF: On the relationship of zinc ion to the struc- ture and function of the 7S nerve growth pattern. 14:2733-2739, 1975. 12 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 13 Baker ME: Molecular weight and structure of 7S nerve growth factor protein. 1 5191 Chem 250:1714-1717, 1975. Cohen S: Purification and metabolic effects of a nerve growth-pro- moting protein from snake venom. ; 5191 Chem 234:1129-1137, 1959. Cohen S: Purification of a growth promoting protein from the mouse salivary gland and its neuro-cytotoxic anti-serum. Egg; Egg; Aged Sci QSA 46:302-311, 1960. Thoenen H, Barde YA: Physiology of nerve growth factor. Physio; Egg 60:1284-1335, 1980. Calissano P, Cattaneo A, Biocca S, Aloe L, Mercanti and Levi-Montalcini R: The nerve growth factor. Exp Cell Egg 154:1-9, 1984. Levi-Montalcini R, Angeletti PU: Salivary glands and their secre- tions. Sreebry 1M and Meyer J, eds. (Paragamon, Oxford, 1964) p 129-141. Aloe L, Levi-Montalcini R: Comparative studies on testosterone and L-thyroxine effects on synthesis of nerve growth factor in mouse sub- maxillary salivary glands. Exp Cell 3g; 125:15-22, 1980. Lakshmann J: Aggressive behavior in adult male mice elevates serum nerve growth factor levels. Am 1 Ehxgigl 250:E386-E392, 1986. Busker ED, Schenkein I, Bane JL: The problem of distribution of a nerve growth factor specific for spinal and sympathetic ganglia. gauge; 3;; 20:1220-28, 1960. Hendry IA: Developmental changes in tissue and plasma concentrations of the biologically active species of nerve growth factor in the mouse, by using a two-site radioimmunoassay. fligghgm 1 128:1265- -1272, 1972. Johnson DG, Gordon P, Kopin IJ: A sensitive radioimmunoassay for 78 nerve growth factor antigens in serum and tissues. J Hggxgghgm 18:2355-2362, 1971. Shine HD, Perez-Polo JR: A recombination assay of high molecular weight nerve growth factor protein complexes. 1 Eggggghgm 26:513- 519, 1976. Berowitz LI, Greene 1A: Nerve growth factor in the goldfish brain: biological assay studies using pheochromocytoms cells. 3;; 162:164-68, 1979. Levi-Montalcini R, Angeletti PU: Nerve growth factor. Ehxgigl Bey 68:534-569. 1968. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 14 Oger J, Arnason BGW, Pantazis N, Lehrich J, Young M: Synthesis of nerve growth factor by L and 3T3 cells in culture. Pree Neel Acad Eel USA 71:1554-1558, 1974. Young M, Oger J, Blanchard MH, Asdourian H, Amos H, Arnason BGW: Se- cretion of a nerve growth factor by primary chick fibroblast cul- tures. fielegee 187:361-362, 1975. Murphy RA, Pantazis NJ, Arnason, BGW, Young M: Synthesis and secre- tion of a high molecular weight form of nerve growth factor by skeletal muscle cells in culture. Bree Neel Aged fiei USA 74:4496- 4500, 1977. Sherwin SA, Sliski AH, Todaro GJ: Human melanoma cells have both nerve growth factor and nerve growth factor-specific receptors on their surfaces. Pree Natl eeeg fie; USA 76:1288-1292, 1979. Murphy RA, Oger J, Saide JD, Blanchard MH, Arnason BGW, Hogan C, Pantazis NJ, Young M: Secretion of nerve growth factor by central nervous system glioma cells in culture. 1 Cell B121 72:769-773. 1977. Murphy RA, Pantazis NJ, Arnason BGW, Young M: Secretion of a nerve growth factor by mouse neuroblastoma cells in culture. Ezee Neg; Aged §e1 USA 72:1895-1898, 1975. Mobley WC, Rutkowski JL, Tennekoon GI, Buchanan K, Johnson MV: Cho— line acetyltransferase activity in striatum of neonatal rats in creased by nerve growth factor. fielegee (fleeh. pg 229:284-287, 1985. Martinez HJ, Dreyfus CF, Jonakait GM, Black I: Nerve growth factor promotes cholinergic development in brain striatal cultures. Bree Hflsl Aggfl 591 USA 82:7777-7781, 1985. levi-Montalcini R, Aloe L: Differentiating effects of murine nerve growth factor in the peripheral and central nervous systems of Xeno- pies Laevis tadpoles. Bree Natl Aeee fiei USA 82:7111-7115, 1985. Levi-Montalcini R, Booker B: Distribution of the sympathetic ganglia in mammals by an antiserum to a nerve-growth protein. Bree Neel Asfld 521 HEA 42:384-391. 1960. levi-Montalcini R, Angeletti PU: Immunosympathectomy. Ehexmeeel Beg 18:619-628, 1966. Levi-Montalcini R: Growth control of nerve cells by a protein factor and its antiserum. Seiegee 143:105-110, 1964. Thoenen H, Angeletti PU, Levi-Montalcini, R, Kettler R: Selective induction by nerve growth factor of tyrosine hydroxylase and dopa- mine -hydroxy1ase in the rat superior cervical ganglia. tree Neel Aged fiei USA 68:1598-1602, 1971. ‘ 42. 43. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 15 Henry IA, Bonyhady RE, Hill CE: The role of target tissues in devel- opment and regeneration-retrophius. In: Haber B, Perez-Polo JR, Hashim GA, Giuttrida-Stella AM (eds): "Nervous System Regeneration, Birth Defects: Original article series, Vol 19, No 4 New York: Alan R. Liss pp 263-276, 1983. Greene CA, Tischler AS: Establishment of a nonadrenergic clonal line of rat adrenal pheochromocytoma cells which respond to nerve growth factor. Bree Neel Aeeg fiel CS5 73:2424-2428, 1976. . Levi-Montalcini R, Meyer H, Hamburger V: ln vltre experiments on the effects of mouse sarcomas 130 and 37 on the spinal and sympathetic ganglia of the chick embryo. Cence; Ree 14:49-57, 1954. Greene CA: A quantitative bioassay for nerve growth factor (NGF) ac- tivity employing a clonal pheochromocytoma cell line. Brele Ree 133:350-353, 1977. Greene LA, Shooter EM: The nerve growth factor: biochemistry, syn- thesis and mechanism of action. Age Reg Neereeel 3:353-402, 1980. Stach RV, Perez-Polo JR: Binding of nerve growth factor to its receptor. J Neegeeel Bee 17:1-10, 1987. Schechter AL, Bothwell MA: Nerve growth receptor on PC-12 cells: evidence for two receptor classes with differing cytoskeletal asso- ciation. Cell 24:867-874. 1981. Connolly JL, Greene LA, Viscarello RR, Riley WD: Rapid sequential changes in surface morphology of PC-12 pheochromocytoma cells in response to nerve growth factor. 1 Cell filel 82:820-827, 1979. Schubert D, LaCorbieve M, Whitlock C, Stallcup: Alterations in the surface properties of cells responsive to nerve growth factor. (lane) 273:718-723, 1978. Pfenninger KH, Johnson MP: Nerve growth factor stimulates phospho- lipid methylation in growing necenites. Pree HEEl Aeee Eel CCA 78:7797-7800, 1981. Lakshmannan J: Nerve growth factor induced turnover of phosphatidy- linositol in rat superior cervical gang1ia. Bleeheg fileehye Ree Cenmgn 82:676~775, 1978. Cremins J, Wagner JA, Halegona g: Nerve growth factor action is mediated by cyclic AMP- and Ca+ /phospholipid-dependent protein kinases. J Cell Blel 103:887-893, 1986. Shooter EM, Yanker BA, Landreth GE, Sutter A: Biosynthesis and mech- anism of nerve growth factor. Ree Exeg; fiery Ree 37:417-446. 1981. Stach RW, Rusenko KW: Covalent attachment of 125I-Beta nerve growth factor to its receptor on sympathetic neurons. J Negxeeel Ree 11:215-229, 1984. 56. 57. 58. 59. 60. 61. 62. 63. 64. 65. 66. 67. 68. 16 Compito GA, Wagner BJ, Stach RW: Covalent linkage of 125I-Beta nerve growth factor to the nerve growth factor receptor from embry- onic sensory neurons. J Neurochem 47(2):554-562, 1986. Kim SU, Rogue-Angeletti R, Gonatas K: Localization of nerve growth factor receptors in sympathetic neurons cultured in vitro. Breln Ree 168:602-608, 1979. Stach RW, Wagner BJ, Ennulat DJ, Perez-Polo JR: Characteristics of the nerve growth factor receptor and a nerve growth factor-nerve growth factor receptor covalent complex. Intern J Neurosci 26:87- 94, 1985. Landreth GE, Shooter EM: Nerve growth factor receptors on PC-12 cells: ligard-induced conversion from low- to high-affinity states. 219; Neel Aged gel QCA 77:4751-4755, 1980. Rogue-Angeletti R, Bradshaw RA: Nerve growth factor from mouse sub- maxillary gland: Amino acid sequence. Eroe Natl Acad Sel CSA 68:2417-2420, 1971. Scott J, Sebly M, Urdea M, Quiroga M, Bell GI, Rutter WJ: Isolation and nucleotide sequence of a cDNA encoding the precursor mouse nerve growth factor. Naguze (Leggen) 302:538-540, 1983. Meier R, Becker-Andre M, Gotz R, Heumann R, Shaw A, Thoenen H: Mole- cular cloning of bovine and chick nerve growth factor (NFC): delinea- tion of conserved and unconserved domains and their relationship to the biological activity and antigenicity of NGF. EMBC J 5:1489- 1493, 1986. Franke U. de Martinville L, Coussens A, Ullrich: The human gene for -subunite of nerve growth factor is located on the proximal short arm of chromosome I. Celenee 222:1248-1251, 1983. Boonstra J, Van Der Saag P, Moolenaar W, DeLaat S: Rapid effects of nerve growth factor on the Na , K -pump in rat pheochromocytoma cells. E33 Cell Ree 131:452-455, 1981. Dichter MA, Tischler AS, Greene 1A; Nerve growth factor-induced increase in electrical excitability and acetylcholine sensitivity of a rat pheochromocytoma cell line. fleeeze 268:501-504, 1977. Reynolds CP, Perez-Polo JR: Induction of neurite outgrowth in the IMR-32 human neuroblastoma cell line by nerve growth factor. J. Heuneeel Ree 6:319-325, 1981. Vinores SA, Koestner A: Reduction of ethylnitrosourea-induced neo- plastic proliferation in rat trigeminal nerves by nerve growth factor. Cenee; Ree 42:1038-1040, 1982. Vinores SA, Perez-Polo JR: The effect of nerve growth factor and antibodies to nerve growth factor on ethylnitrosourea carcinogenesis in mice. J Cegee; Ree Cllg Cneel 98:59-63, 1980. 69. 70. 71. 72. 73. 74. 75. 76. 77. 78. 79. 80. 81. 82. 17 Vinores SA, Maragos PJ, Parma AM, Guroff G: Increased levels of neuron-specific enolase in PC-12 pheochromocytoma cells as a result of nerve growth factor treatment. J Negrochem 37:597-600, 1981. Guroff G, Dickens G, End D: The induction of omithine decarboxylase by nerve growth factor and epidermal growth factor in PC-12 cells. J nenreehen 37:342-349, 1981. Mcguire JC, Greene LA: Rapid stimulation by nerve growth factor of amino acid uptake by clonal PC-12 pheochromocytoma cells. J filel Chen 254:3362-3367, 1979. Perez-Polo JR, Werrbach-Perez K, Tiffany-Castiglioni E: A human clonal cell line model of differentiating neurons. Deg file 71:341- 355, 1979. Vinores SA, Koestner A: The effect of nerve growth factor on un- differentiated glioma cells. Cancer Lee 10:309-318, 1980. Marushige Y, Raju NR, Marushige K, Koestner A: Modulation of growth and of morphological characteristics in glioma cells by nerve growth factor and glio maturation factor. Ceeee; Ree 47:4109-4115, 1987. Vinores SA, Perez-Polo JR: Role of nerve growth factor in ethylni- trosourea-induced neural carcinogenesis. J EEEEQEEI Ree 5:351-361, 1980. Camp RC, Koestner A, Vinores SA, Capen CC: The effect of nerve growth factor and antibodies to nerve growth factor on ethylnitro- sourea-induced neoplastic proliferation in rat trigeminal nerves. 2e; Zeehel 21:67-73, 1984. Koestner A, Swenberg JA, Wechsler W: Transplacental production with ethylnitrosourea of neoplasms of the nervous system in Sprague Dawley rats. Amer J Egghel 63:37-56, 1971. Swenberg H, Wechsler W, Koestner A: The segmental development of transplacentally-induced neuroectodermal tumors. J. Neereeethel Egg Neezel 31:202-203, 1972. Druckrey H, Ivankovic S, Preussman R: Teratogenic and carcinogenic effect in the offspring after single injection of ethylnitrosourea to pregnant rats. fleeeze 210:1378-1379, 1966. Rice JM: An overview of transplacental chemical carcinogenesis. 8:113-125, 1973. Magee PN, Swann PF: Nitroso compounds. Bel; flee Bell 25:240, 1969. Wechsler W, Kleihues P, Matsumoto S, Zulch KJ, Ivankovic S, Preussmann R, Druckrey H: Pathology of experimental neurogenic tumors chemically induced during prenatal and postnatal life. Ann NY Aeee Eel 159:360-408. 83. 84. 85. 86. 87. 88. 89. 90. 91. 92. 93. 94. 95. 18 Koestner A, Swenberg JA, Wechsler W: Experimental tumors of the ner- vous system induced by resorptive N-nitrosourea compounds. 2121 Exp lune; Bee 17:9-30, 1972. Koestner A, Swenberg JA, Wechsler W: Transplacental production with ethylnitrosourea of neoplasms of the nervous system in Sprague Dawley rats. Am J Peshgl 63:37-36, 1971. Druckrey H: Specific carcinogenic and teratogenic effects of in- direct alkylating methyl and ethyl compounds and their dependency on stages of ontogenic developments. K222D12£1£§ 3:271-303, 1973. Koestner A: Transplacental carcinogenesis. Peee lgeh Cen Ceeee; Cee 10:65-75, 1973. Pegg AE: Alkylation and subsequent repair of DNA after exposure to dimethylnitrosamine and related carcinogenes. Rev Bloehem logical 5:83-133, 1983. Kleihues P, Lautos PL, Magee PN: Chemical carcinogenesis in the nervous system. lee Reg Exe Pathol 15:153-232, 1976. Singer B: N-nitroso alkylating agents: formation and persistence of alkyl derivatives in mammalian nucleic acids as contributing factors in carcinogenesis. J Neel Cegee; lnst 62:1329-1339, 1979. Ruggero Motesano: Alkylation of DNA and tissue - specificity in nitrosamine carcinogenesis. J Seeremelee §££B£ Cell Bleehem 17:259-273, 1981. Olsson M, Lindahl T: Repair of alkylated DNA in c methyl group transfer from 0 -methylguanine to a protein cystein residue. J Blel Chem 255:10569-10571, 1980. Sukumar S, Notario V, Martin-Zanca D, and Barbacid M: Induction of mammary carcinomas in rats by nitroso-methylurea involves malignant activation of H-ras-l locus by single point mutations. 306:658-661, 1983. Sukumar S, Perantoni A, Reed C, Rice JM, Wenk ML: Activated K-ras and N-ras oncogene in primary renal mesenchymal tumors induced in F-344 rats by methyl (methoxymethyl) nitrosamine. Mel Cell Blel 6:2716-2720, 1986. Guerrero I, Calzada P, Mayer A, Pellicer A: A molecular approach to leukemogenesis: Mouse lymphomas contain an activated C-ras oncogene. IIQE Natl Aged Eel HCA 81:202-205, 1984. Schechter AL, Stern DF, Vaidyanathan L, Decker SJ, Drebin JA, Greene MI, Weinberg RA: The neu oncogene: an erb-B-related gene encoding a 185,000-Mr tumor antigen. Neeeee 312:513-516, 1984. -‘ 19 so, Rice .121. Reed cn, Watatani u, Wen]: 1a.: Activated gene sequences in primary tmors of the peripheral system induced in rats by transplacental exposure to ‘ trosourea. In: M m m m 84:6317-6321, 1987. l 1’ CI, Hung MC, Weinberg RA: Multiple independent activations . flu neu oncogene by a point mutation altering the transmem- ' domain of p185. Cell 45:649-657. 1986. o '. “SJ-s. . -.ll ‘. ‘5 CHAPTER 2 THE EFFECT OF NERVE GROWTH FACTOR (NGF) ON TRANSPLACENTAL ETHYLNITROSOUREA (ENU)-INDUCED NEUROGENIC TUMORS IN SPRAGUE-DAWLEY RATS W To determine the effect of nerve growth factor (NGF) on the spec- trum, incidence and latency periods of neurogenic tumors, with particular attention to the neurinoma (Schwannoma) incidence, specific pathogen-free, date-mated Sprague-Dawley rats were injected with 50 mg/kg body weight of ethylnitrosourea (ENU) on the 20th day of gestation. Pups were weaned on the 28th day and caged in sex-matched pairs. Forty, 60 and 80 micrograms of NGF was subcutaneously inoculated to 34 offspring on days 12-16, 90-94 and 210-214 post-partum. Thirty-four control rats, not inoculated with NGF, were maintained under the same conditions as the experimental group. In the NGF-treated group, 11/34 rats were affected with trigeminal nerve neurinomas compared to 18/34 in the NGF-untreated controls (p < 0.05). In the peripheral nerves, there were 5 and 11 neurinomas, respectively, in each group. When the total numbers of neurinomas (trigeminal and peri- pheral nerves) between these groups were compared (16/34 versus 29/34), the significance of neurinoma reduction due to NGF treatment showed a p value of < 0.01. The spectrum, incidence and latency period of central nervous sys- tem (CNS) tumors between NGF-treated (20/34) and NGF-untreated (25/34) groups did not vary significantly. A majority of the tumors consisted of differentiated gliomas. These results are consistent with the previous studies in which it was demonstrated that ENU selectively induced differ- entiated tumors in the CNS of rats, and that NGF did not have any in- fluence on these differentiated neoplasms. Five trigeminal and 2 peripheral nerve neurinomas in the NGF-un- treated group were shown to contain nerve growth factor receptor (NGF-R) sites by the avidin and biotinylated horseradish peroxidase complex 20 21 ‘ 1:;'h (IL ‘ ‘n , cried cells. and that this activity is dependent upon the presence ‘ use J". . . ‘ receptor binding sites. jot-tow." .a 4 “3010;1121.1.- ‘ -‘ ‘ v, . ‘. : ' -) mcipl 'J a" a. u . ~ ._ r .-. ._, .+ . ’7. ‘ 13.17 'A-I‘T-L‘ 30'0" ‘ 1 H" ‘d-.:' ' ~ 7." -’ “W" ."3,. 6 "“1" - ., .."v . _ _ ~ . . . _ - . 5"u- In! L. . , ,-\ ‘ *ul » .‘- . “ »o‘ -', f;-v€‘!i.41'\i l4 , 9” PC 15‘ ....13.5.""“ I'. ': .':‘-‘-.-'-r-..- .. 3 9' rt. Hand‘s: I fit; ‘1 - I tnfv Lamina: I-i: ‘Ksrit. ‘3 : single (1.4.; 1;; ‘35) :3/‘15 bcr,1_. \sig...‘ "ca: yum n. pragmn: {dale ‘ the 131‘: 1&1 tail vein 'w 'E.. .2171: is“ o: gestation result-«him . tion or mutagenic ulna-:2: .1- nearly 300‘ of the off~ 22 12W Nerve growth factor (NGF) is a polypeptide composed of alpha, beta, and gamma fractions with a molecular weight of 131,5001'2. The beta subunit, (B—NGF, a dimer of 13,000 molecular weight), is the active mole- cule which possesses the nerve growth promoting activity.3‘6 Recent evidence indicates that NGF also has a neurotropic effect on the cholin— ergic neurons in the brain.7'9 In embryologic development, any tissue which will be innervated produces nerve growth factor. After binding with the specific nerve growth factor receptor (NGF-R) on the nerve fibers, NGF is internalized and retrogradedly transported to the neuronal cells in the ganglia.10’11 Several biochemical changes occur in the cell following the coupling of NGF with its receptor, some of which include an increase in cAMP levels and mobilization of intracellular Ca+2'12 phospholipid methylation,13 and phosphatidylinositol metabolism,14'15 and induc- tion of cyclic AMP- and Ca+2/phospholipid-dependent protein kinases.16 The locations of NGF-R on the neurons have been identified by the immunoprecipitiation technique using monoclonal antibodies to the NGF-R protein.17 This receptor has been identified in sympathetic and sensory neurons,6 Schwann cells,18 19 peripheral neuroglial cells, pheochro- mocytoma PC-12 cells,20 and melanoma cells.21 (Refer to Chapter I for additional information on NGFR.) A single dose of 50 mg/kg body weight ENU given to pregnant female rats via the lateral tail vein on the 20th day of gestation resulted in the formation of neurogenic tumors in nearly 100% of the off- Spring.22'23 Sequential evaluation of these neoplasms indicated that neurinomas of the trigeminal nerves were evident as early as 20 days after 23 exposure, with nearly 100% involvement by 90 days.24 The lesions pro- gressed from early neoplastic proliferations (ENP) or micro-tumors to grossly detectable macro-tumors attaining a peak of development by 7 months. Gliomas of the central nervous system (CNS) and neurinomas of the peripheral nervous system (PNS, spinal nerve) appeared between 6 and 7 months post-exposure. At the time of termination of the one year study, there was an increasing trend in the number and frequency of these neo- plasms.23'24 Several investigations demonstrated that NGF had both maturation and differentiation influences on neuroectodermal tumor cells.25'29 Exposure of PC-12 pheochromocytoma cells to NGF causes a rapid stimulation of Na, K+-pump mediated K+ influx,25 development of excitable mem- 26 neurite outgrowth and cessation of mitosis,27 all biochemi- branes, cal characteristics associated with differentiated sympathetic neu- rons.30 A similar phenotypic alteration was also induced in 1MR-32 human neuroblastoma cell line after NGF treatment.29 Rats implanted with anaplastic glioma cells and treated with NGF had a significant re- duction of tumor growth rate (p < 0.025) and increased survival time (p < 0.00005).31’32 Pretreatment of the malignant cells with NGF 24 hours prior to implantation yielded a similar, but less dramatic result.32 In a recent 13 ELEEQ study,33 using NGF and glia maturation factor (GMF), it was shown that both these factors induced characteristic changes of cell morphology and growth pattern in an anaplastic glioma cell line T9. Nerve growth factor retarded cell growth rate and induced a flattened cytoplasm with numerous protruding processes forming somatic links with adjacent cells. GMF did not affect the cell growth rate. The GMF 24 treatment transformed the plump T9 cells into slender cells with long cytoplasmic processes forming a characteristic interconnecting network. The effect of NGF was persistent following withdrawal of NGF from the medium, whereas the phenotypical alterations induced by GMF were revers- ible. Interestingly, concomitant treatment with NGF and GMF had the com- bined effect of both factors. These experiments supported the hypothesis that both growth factors function as regulators of cell differentiation. In two 90-day termination studies,3l"35 treatment of pregnant rats with NGF, prior to ENU exposure or postnatally following gestational ENU administration, resulted in a significant reduction of ENPs in trigem- inal nerves of the offspring. The present experiment was an extension of these 90-day studies and designed to explore the persistence of the NGF effect on neurogenic tumor development following transplacental ENU expo- sure by determining changes in the incidence, latency period and spectrum of neurogenic tumors. Special attention was to be given to the final effect of NGF upon trigeminal neurinoma development complementing the results obtained from the 90 day termination study.3"'35 Testing of the neurogenic tumors for the presence of NGF receptors would further determine whether prospective changes in tumor spectrum and/or incidence were the consequences of NGF administration and interaction with the tumor cells. 25 WNW Ezeeexeelen efi NCE. B-NGF was isolated from salivary glands of male Swiss Webster mice and purified by the procedure described by Bocchini and Angeletti.36 Samples were tested for biological activity by the PC-12 method.37 §£h1121§I2§22££§I ENUa was dissolved in citrate phosphate saline buffer (3 mM sodium phosphate and 2 mM citric acid in 0.15 M NaCl, pH 4.2) at a concentration of 10 mg/ml, and injected within one hour after pre- paration. WW Eight date-mated 20 day pregnant Sprague-Dawley (CD) ratsb were given a single slow dose of 50 mg/kg body weight ENU via the lateral tail vein as previously described.19 The nursing rats, along with the pups, were randomly divided into 2 groups. Green A: 34 offspring (18 males and 16 females) were inoculated with 40, 60, and 80 ug of NGF (dissolved in sterile 0.15 M NaCl solution) divided in 5 subcutaneous doses on days 12-16, 90-94, and 210-214 as pre- viously described.31'35'38 This method of NGF administration was chosen to facilitate slow absorption and to ensure prolonged action of the hor- mone on the ENU-transformed cells. The incremental dose levels (20 ug) allowed for an increase in age and body weight. Creee B: 34 offspring (15 males and 19 females), which served as a positive control, were exposed to ENU but did not receive NGF. They were ‘ Serva Fine Biochemical, Inc., Garden City Park, New York, 11040 b Charles River Laboratories, Portage, MI 26 maintained under the same conditions as the experimental group. The pups were weaned at 28 days of age and housed in sex-matched pairs. The rats were fed autoclaved Purina Lab Chow 5010C, with water ad libitum. All animals were observed twice daily and weighed weekly throughout the ex- perimental period. Termination of the experiment at one year of age precluded any sig- nificant natural occurrence of neurogenic tumors in CD rats.39 A NGF control was, therefore, not included. NGF is, of course, not considered to be a carcinogen. Rats either died as a consequence of neoplasms or were euthanatized due to progressive neurological signs and weight loss. All animals were necropsied as soon as possible after death or euthan- asia. All lesions, serial sections of the brain, trigeminal nerves, and six selected segments of the spinal cord were fixed in 10% buffered- formaldehyde. For immunocytochemistry, similar tissues, acquired from all animals, were frozen in liquid nitrogen and stored at -70°C. flleeeeeehelegy: Brain, spinal cord, trigeminal nerves, lung, liver, heart, stomach, intestines and kidneys from each rat and all gross lesions were routinely processed, embedded in paraffin, sectioned at 5 u and stained with hematoxylin and eosin (H&E). Special stains used included Masson's trichome, periodic acid-Schiff (PAS) and Giemsa stains. Wefmsmhmwmm: TheNGF-R positive control was developed in 7-day transected sciatic nerve in male Sprague-Dawley rats as previously described with slight modifications.18 The rats were anesthetized by IP injection of Equithesin.c A 3 mm c Prepared by Department of Pharmacology, Michigan State University 27 section of sciatic nerve was removed near the tendon of the obturator internus muscle, and the proximal stump folded beneath the tendon to pre- vent regeneration. Seven days post-surgically, the rats were euthanatized and both ends of the transected sciatic nerve, intact sciatic nerve from the opposite leg, and the surrounding innervated muscles were frozen on dry ice and stored at -70°C. Immehimahssistm; Miss 2f HIE-B. Cryostat sections of ENU-induced neurinomas, transected sciatic nerves and innervated muscles were rapidly fixed in 4Q paraformaldehyde prior to exposure to 192-IgG NGFR monoclonal antibodyd (5 ug/ml). The antibody was suspended in a solution of 100 mM potassium phosphate, 160 mM NaCl/5% heat-inactivated horse serum/0.02% NaN3, adjusted to pH 7.5. After binding for 30 min- utes, sections were washed in PBS (20 mM potassium phosphate/150 mM NaCl, pH 7.5), and incubated for 30 minutes at room temperature with biotiny- lated horse anti-mouse IgG immunoglobulin. The sections were then washed, treated with 0.3% H202 in methanol to quench endogenous peroxidase activity, and, after a 10 minute wash, incubated for 30 minutes with a complex of avidin and biotinylated horseradish peroxidase.8 The sec- tions were washed gently several times, incubated in 0.05% 3, 3-diamino- benzidine-/0.01% H202 for 10 minutes. The sections were rinsed in distilled water (5 minutes), counter stained with Gill's hematoxylin (1-2 minutes), cleaned in distilled water, dehydrated in graded alcohols, and mounted. d A generous gift from Dr. Eugene M. Johnson, Jr., Department of Pharmacology, Washington University School of Medicine, St Louis, MO e Vector Laboratories, Burlingame, CA 28 SIAIISIICAL ANALXSlS: A chi-square test of independence was used to de- termine if there was a significant difference in the incidence, spectrum and latency of neurogenic tumors between the NGF-treated rats (Group A) and the control rats (Group B). 29 W WflWMW In the NGF-untreated group (positive control) 15 of the 34 animals had grossly visible trigeminal neurinomas, and 3 had microsc0pic trigemi- nal ENPs (53%)(Table l-l). Most trigeminal nerves were bilaterally in- volved with the gross lesions varying from edematous and unevenly swollen foci to friable, reddish-brown hemorrhagic tumors (Figures 1-1, 1-2, 1-3). Histologically, the neurinomas varied in their degree of anaplasia. The anaplastic neurinomas consisted of hyperchromatic polyhedral cells with round to oval nuclei containing dense nuclear chromatin and pale cytoplasm. These cells were arranged in undulating sheets, whorls and dense clusters (Figure 1-4). Mitotic figures ranged from one to two per high power field. Scattered pockets of hemorrhage and necrosis were also present (Figure 1-5). The ENPs were classified according to the criteria previously des- cribed.19’20 These foci, typically located in the proximity of the junction between the peripheral nervous system and central nervous system, were characterized by disorganized arrays of hyperchromatic cells haphaz- ardly arranged in irregular sheets (Figure 1—6). In the NGF-treated group, 10 of the 34 rats had trigeminal nerve neurinomas and one had an ENP (32%)(Table 1-1). The difference in inci- dence between control versus treated group was significant at p < 0.05. These data are supportive of NGF effect by a significant reduction in the number of trigeminal nerve neurinomas (p < 0.05). The incidence of peripheral nerve (spinal nerve root) neurinomas between groups did not differ significantly (p < 0.1)(Table 1-2). Five of the 34 treated and 11 of 34 control animals developed neoplasms. adorned 38:: n fl :3 a £83 a 3 en ..8 «2926 undo.» g g .33 ,3eu 958 “when. ems 5.3 was..." nfli use.» 5? Be.» no noel! no .8852 no 3a.: no .8952 H38. .395 .mg «goofing £98 .355.» agggggggfigfigss "slang 31 dove .. Sm menuema 13.3 3 an a «ma HTS” issued m en a oneness one gag gum: dogged REM no g as“: even no g .8852 “mung one 60ng .m 965 among .4965 .8353? Egggggfiayofigg nutmeg .n.o v Q #33 mm «33 3 $an 3 an m .335 3 «33 m can as an e 386...— mg Ben M 5 egg mam: Handgun—on 9P8: 3% g Hg 38. 5.3 33 5.? even Hg no .8852 no g .uupgdgagég .Sflobflfiao Eggguoggflflfiguo 060ng AA 2an 33 Table 1-4: Incidence and spectrum of neural tumors in rats after ENU administration. NGF Treated ENU Alone Total number of rats 34 34 rats with CNS tumors 20 25 Type of Tumor Tumors of CNS Oligodendroglioma 16 20 Mixed Glioma 11 9 Astrocytoma 5 7 Meningioma 2 2 Glioependymoma 0 2 Total 34 40 Rats with PNS Tumors — Neurinoma 16 29 Total number of tumors 50 69 Total number of tumor- bearing rats 32 32 Average number of tumor/rat 1.56 2.16 34 When the total neurinomas (trigeminal and peripheral) were compared between the groups, the effect of NGF on neurinoma development was highly significant (x2 - 3.72, p < 0.01)(Table 1-3). lnsidense 2f 9N5 tuners In the central nervous system, the proportion of rats developing tumors in both groups was similar; 20/34 in Group A versus 25/34 in Group B (Table l-4)(p > 0.5). However, due to multiplicity, the total number of neurogenic tumors in both groups exceeded the number of rats affected (Table 1-4). The number of tumors per animal ranged from 1 to 4 (an average of 1.56 in NGF-treated group and 2.16 in the ENU control). In the experimental group, 8 rats had both neurinomas and CNS tumors, whereas 14 rats had both types of tumors in the control group. The average survival time of rats with the tumors is given in Table 1-4. The spectrum, incidence and latency period of CNS tumors did not differ significantly between the NGF-treated and untreated groups (Table 1-4). A majority of CNS tumors were differentiated gliomas. These results are consistent with previous studies which indicated that ENU selectively induces differentiated tumors in the CNS of rats,22'24 and that NGF did not have any influence on these differentiated neo- plasms.31 leaaifisstign 2f 9N5 Tumors The tumors were categorized as described previously.23'24 The gliomas in both groups were located in the hippocampus, periventricular areas, subcortical white matter, cerebral cortex, basal ganglia, and in- frequently in cerebellum and medulla. 35 Qliggggngzggligmggz These were the most common tumors in the CNS and were equally distributed between brain and spinal cord. Most spinal cord oligodendrogliomas were grossly detectable and preferentially located in the upper cervical, lower thorax, and lumbosacral areas. The lesions were pale, gelatinous to wet, reddish-brown, hemorrhagic foci (Figure 1-7). Histologically, the well-differentiated neoplastic cells were round, with scant cytoplasm and dense nuclei, and were sometimes arranged in small clusters with ”honeycomb-like” patterns supported by thin fibro- vascular stroma (Figure 1-8). The periphery of oligodendrogliomas blended into the adjacent brain neuropil. Mixed gliomas: Tumors in this category consisted of a mixture of differentiated astrocytes and oligodendrocytes (Figure 1-9), with either type predominating a single lesion. These were often located in the cere- bral cortex, the hippocampus and, rarely, in the cerebellum. Aggxggyggmggz The predilection site for astrocytomas corresponded to that of oligodendrogliomas. Most were microscopic in size and con- sisted of well-differentiated cells, rich in cytoplasm with oval to round nuclei (Figure l-lO). Hitoses and anaplasia were mostly inapparent. Glioependxmgmgg: Two of these neoplasms were identified in the ependymal region of the brain and contained proliferating glial cells and ependymal cells, forming vague rosettes.(Figure 1-11). Meningigmgfiz 0f 4 meningiomas, one appeared to involve the entire cerebral hemisphere containing areas of hemorrhage and necrosis (Figure 1-12). Histologically, the neoplastic cells invaded the cerebral cortex and had a moderate rate of mitosis. Isolated areas of necrosis were 36 observed. The remaining three meningiomas only compressed the cerebral hemispheres, but no invasion was recognized (Figure l-13). 9p; g Nerve glioma: The observable lesion progressed from a slight epiphora and conjunctivitis in the right eye at the initial observation to an almost complete proptosis of the eyeball within 5 days (Figure l-14). The tumor consisted of intraocular (neurinoma) as well as extraocular (glioma) cellular proliferation cells (Figure 1-15). Egn-nggxgl nggnlggmg: Non-neural tumors included three mammary gland adenomas, one thymic lymphosarcoma, one thyroid follicular adenoma, three renal myxofibrosarcomas, one nephroblastoma, one renal adenocarci- noma, and two ameloblastic odontomas. The latter is a very rare tumor in animals, which has not been previously described in rats. Herve Growth Eggggxrfieggptggz The hypothesis that suppression of neurinoma development by NGF is dependent upon the presence of NGFR was investigated. Forty-five neurinomas from both groups (29 from untreated and 16 from treated) were tested for NGFR by the immunoperoxidase method. Five trigeminal and 2 peripheral nerve neurinomas (7/29) in the untreated group were positive, while all neurinomas in the treated group were nega- tive. Staining variability ranged from focal positive areas to almost complete staining of the complete sections (Figure 1-16). Elapsed time post-sectioning had no effect on the specificity or consistency of the procedure. There was no nonspecific background staining. The NGF-R positive control (transected sciatic nerve) and NGF-R negative neurinomas are illustrated in Figures 1-17 and 1-18. 37 lFigure l-l Ventral view of rat brain. Notice bilaterally swollen trigeminal nerves (arrows). ‘ Figure 1-2 Dorsal view of rat trigeminal nerve affected with anaplas- tic neurinoma (1); brain reflected dorsally (2). 38 Figure 1—1 Figure 1-2 39 Figure l-3 Ventral view of rat brain showing bilateral anaplastic trigeminal nerve neurinoma (l); hemorrhage (2). Figure 1-4 Photomicrograph of an anaplastic neurinoma illustrating hyperchromatic polyhedral cells and mitoses (arrow). H & E. X 400. 40 METRIC 1 Figure 1-3 C ’ld‘.“ I .W a... ...... -4 Figure l 41 Figure 1-5 Photomicrograph of an anaplastic neurinoma showing areas of necrosis and hemorrhage. H & E. X 400. Figure l-6 Photomicrograph of an early neoplastic proliferation (ENP) in a trigeminal nerve. Arrow indicates CNS-PNS junction. H & E. X 160. 42 ‘x ~M”:‘:4“ z ”:33 '33» ~ """d‘s 0" ~"“' ‘7 . €22». . ' ~ ‘ v/J'N‘s'k‘ m."— ., .- - ' . . ' . .0 fe‘ ._ fl. . > ‘_ "i, - 4 .“ 3‘ . g a ‘ .' A t _- ‘ 11‘ g . I .. L J; . f 5:! .. A . ' X , ' - " ' 'V ‘ " Q {:3 43 Figure 1-7 Gross appearance of an oligodendroglioma in the lumbo-sacral region of a spinal cord. Figure 1-8 Photomicrograph of an oligodendroglioma. Notice small round cells with scant cytoplasm and dense nuclei arranged in a "honeycomb'-like pattern. H & E. X 160. 44 Figure 1-8 45 Figure 1-9 Photomicrograph of a cerebral mixed glioma illustrating mixed population of neoplastic cells. H & E. X 400. Figure 1-10 Photomicrograph of cerebral astrocytoma. H & E. 400 X 47 Figure 1-11 Photomicrograph of a cerebellar glioependymoma. Neoplastic cells are arranged in short chains and rosettes. H & E. X 400. 49 Figure 1-12 Gross appearance of a meningioma involving the entire dorsal cerebral hemisphere. Figure 1-13 Photomicrograph of the meningioma in Figure 1-12. Notice oval to fusiform neoplastic cells forming nebulous whorls. H & E. 400 X. z, «...... .... .....- - 1...... .manes. 3%. Mwo .. 3%... . ... .. . .r/. \‘v.......... ‘ Q.“ ..QAWoV .c Ly... .d .31.. \. I .. 4.19.. p. .. O ‘ ‘ 4.." wt". .0. ......Iah. 3. .... st . 0.2. 0 vote 2 ..c it. .... IA... 1.. ... .. .V o . 1 %9‘9% 0'. 'y1fi . Limo . \ m . O 9 ‘mfl .o O .3fi..¥0 1 ~“1 .I u... . L F ‘3? 9.. Ck ”dame“ \ . a nu. .. ... .. ..u.‘ ...? I’ . ‘ It. ‘J’k «IthoL Figure 1-13 51 Figure 1-14 Ventral view of a rat brain showing bilateral trigeminal nerve neurinoma (1)} optic nerve glioma (2); globe (3). 1) E. x 60. Figure 1-15 Photomicrograph of optic nerve tumor. Glioma H ( intraocular neurinoma (2); retina (arrow). & 53 Figure 1-16 Photomicrograph of a neurinoma immunostained for nerve growth factor receptor (NGFR). Notice strong NGFR positive cytoplasmic reactions in neurinoma cells. ABC method. Hematoxylin counter stain. X 100. 54 Figure 1—16 55 Figure 1—17 Photomicrograph of 7-day transected rat sciatic nerve immunostained for NGFR (positive control). Schwann cells have strong positive reaction around the cytoplasmic rim. An unstained blood vessel (center). ABC method. X 100. Figure 1-18 Photomicrograph of a neurinoma immunostained for NGFR. Notice all neoplastic cells are negative. ABC method. Hematoxylin counterstain. X 100. Figure 1—17 Figure 1-18 57 12133115319! This study supports the hypothesis that NGF reduces neurinoma devel- opment in rats transplacentally exposed to ENU and complements the pre- vious 90 day studies.3“'35 It establishes the persistence of the neurinoma-reducing effect previously recognized in the 90-day studies. Since the mechanism of NGF interaction with neuronal cells is de- pendent upon the presence of NGF-Rs,10’11 the ability of the NGF to induce differentiation of the anaplastic neuroepithelial cells is also dependent upon binding of the hormone to the receptor molecules. In the present study, NGF-R protein was present only in the NGF-untreated group and none of the 16 neurinomas in the NGF-treated group were positive. Since there was an association between NGF-R and the administration of NGF, and since the number of tumors in the NGF-treated group was inversely related to the presence of NGF-R, we suggest that tumors with NGF-R bind- ing sites may respond to the exogenous NGF administration and this treat- ment results in suppression or differentiation of anaplastic cells. Immature Schwann cells of peripheral nerves are the target cells for ENU transformation.23 Studies by Vinores and l(oestner,34’38 and Camp g; g1,35 documenting the reduction of ENPs by NGF, given prior to ENU exposure, might be explained by enhanced maturation of immature Schwann cells, thereby reducing the target cell population. NGF given after ENU exposure suggests that it is also capable of suppressing the phenotypic expression of the transformed cells. One mechanism important in the process of transformation is forma- tion and persistance of premutagenic 06-ethylguanine adducts in DNA.40 The developing nervous system is prone to retain this DNA lesion, since there is a virtual lack of the 06-ethylguanine repair enzyme.40 58 Selective gene suppression via the epigenetic pathway might ex- plain the protective effect of NGF on ENU-initiated cells. The epi- genetic pathway is activated by two possible mechanisms: (1) by means of secondary messengers following the coupling of NGF to the cytoplasmic receptors, (2) internalization and binding of NGF to the nuclear receptor triggering the induction of transcriptional-dependent enzymes via mRNA. In support of this mechanism, NGF induces specific transcription dependent enzymes, including tyrosine hydroxylase,41 dopamine-beta- 42 and ornithine decarboxylase43 in normal and neoplastic hydroxylase, neural cells after NGF exposure. Although there was also a slightly lower incidence of gliomas in the NGF-treated rats (20 versus 25), the difference was not significant sta- tistically; however, a trend cannot be eliminated. A significant effect upon these differentiated gliomas by NGF was not expected since previous studies determined a selective NGF effect solely for anaplastic glioma cells.31’32 In the present experiment, it was not determined whether a single NGF treatment or a combination of all three treatment schedules were responsible for the suppressive effect on neurinoma incidence. Further studies are needed to evaluate the effectiveness of the individual NGF exposure times on neurinoma development. The results of our study provide a promising new regime in the management of tumors derived from neural crest cells. In theory, there- fore, beneficial effects from NGF therapy might be derived when used in pharmacological doses in conjunction with the conventional therapeutic approaches including surgery, radiotherapy and chemotherapy. REFERENCES - CHAPTER 2 10. 11. 12. REFERENCES Cohen S: Purification and metabolic effects of a nerve growth-pro- moting protein from snake venom. l 3121 Chem 234:1129-1137, 1959. Bradshaw BA: Nerve growth factor. Annu Rev Biochem 47:191-216, 1978. Bucker ED, Schenkein I, Bane JL: The problem of distribution of a nerve growth factor specific for spinal and sympathetic ganglia. Cancer Res 20:1220-1228, 1960. Levi-Montalcini R, Angeletti PU: Nerve growth factor Physiol Egg 18:619-628, 1968. Greene LA, Shooter EM: The nerve growth factor: biochemistry, synthesis and mechanism of action. Ann Rev Neuroggi 3:353-402, 1980. Yankner BA, Shooter EM: The biology and mechanism of action of nerve growth factor. Annu Egg Biochem 51:845-868, 1982. Crutcher KA, Collins F: 13 vitro evidence for two distinct hippo- campal growth factors: basis for neuronal plasticity? Science 217:67-70, 1982. Korsching S, Auburger G, Heumann R, Scott J, Thoenen H: Levels of nerve growth factor and its mRNA in the central nervous system of the rat correlate with cholinergic innervation.- EMBO J 4:1389-1393, 1983. Shelton DL, Reichardt LF: Studies on the expression of the B nerve growth factor (NGF) gene in the central nervous system: level and regional distribution of NGF in RNA suggest that NGF functions as a trophic factor for several distinct populations of neurons. roc Natl Acad Sgi flfiA 83:2714-2718, 1986. Taniuchi M, Schweitzer JB, Johnson EM: Nerve growth factor receptor molecules in the brain. Erog Natl Acad Sgi, QSA 83:1950-1954, 1986. Yan Q, Johnson EM: A quantitative study of developmental expression of nerve growth factor (NGF) receptor in rats. 22! Biol 121:139- 148, 1987. Schubert D, LaCorbiere M, Whitlock C, Stallcup W: Alterations in the surface properties of cells responsive to nerve growth factor. Ngggzg 273:718-723, 1978. 59 13. 14. 15. l6. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 60 Pfenninger KH, Johnson MP: Nerve growth factor stimulates phospho- lipids methylation in growing neurites. £12; Natl Atad 5&1, USA 78:7797-7800, 1981. Lakshmannan J: Nerve growth factor induced turn of phosphatidylino— sitol in rat superior cervical ganglia. filflfihfifl fligphxg Egg figmmtg 82:767-775. 1978. Traynor AE, Schubert DS, Allen WR: Alteration of lipid metabolism in response to nerve growth factor. 1 Ngnttthgn 39:1677-1683, 1982. Cremins J, Wagner JA, Halegouazsz Nerve growth factor action is mediated by cyclic AMP and Ca+ /phospholipid-dependent protein kinases. J ggll 5191 103:887-893, 1986. Taniuchi M, Johnson EM: Characterization of the binding properties and retrograde axonal transport of a monoclonal antibody directed against the rat nerve growth factor receptor. J lel £121 101: 1100-1106, 1985. Taniuchi M, Clark RB, Johnson EM: Induction of nerve growth factor receptor in Schwann cells after axotomy. £199 Natl Atad fiti, DEA 83:4094-4098, 1986. Zimmerman A, Sutter A: Beta nerve growth factor (BNGF) receptors on glial cells. Cell-cell interaction between neurons and Schwann cells in culture of chick sensory ganglia. Emtg J 2:879-885, 1983. Schechter AL and Bothwell MA: Nerve growth factor receptors on PC12 cells: evidence for two receptor classes with differing cytoskele- tal association. lel 24:867-874, 1981. Benerjee SP, Synder SH, Cuatrecasas P and Green LA: Binding of nerve growth factor in superior cervical ganglia. £125 Natl 5934 SE1 USA 70:2519-2523, 1973. Swenberg JA, Koestner A, Wechsler W, Denlinger RH: Quantitative aspects of transplacental tumor induction with ethylnitrosourea in rats. flange; Res 32:2656-2660, 1972. Koestner A, Swenberg JA, Wechsler W: Transplacental production with ethylnitrosourea of neoplasms of the nervous system in Sprague- Dawley rats. Am 1 Egthgl 63:37-556, 1971. Swenberg JA, Wechsler W, Koestner A: The sequential development of transplacentally-induced neuroectodermal tumors. 1 thttngthgl E52 Neural 31:202-203, 1972. Boonstra J, Van der Saag P, Moolenaar W, Delatt 8: Rapid effect of nerve growth factor on the Na+, K+-pump in rat pheochromocytoma cells. Exp lel Egg 131:452-455, 1981. Dichter MA, Trischler AS, Greene LA: Nerve growth factor-induced increase in electrical excitability and acetylcholine sensitivity of a rat pheochromocytoma cell line. Ngtgtg 268:401-404, 1977. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 61 Green LA, Tischler AS: Establishment of a noradrenergic clonal line of rat adrenal pheochromocytoma cells which respond to nerve growth factor. Etgg Natl Aggg Egi (Wgsh) 73:2424-2428, 1976. Connolly JL, Greene LA, Viscarello RR, Riley, WD: Rapid sequential changes in surface morphology of PC-12 pheochromocytoma cells in response to nerve growth factor. 1 lel Eigl 82:820-827, 1979. Reynolds CP, Perez-Polo JR: Induction of neurite outgrowth in the IMR-32 human neuroblastoma cell line by nerve growth factor. 1 Ngttggg1_Egg 6:319-325, 1981. Green LA, Shooter EM: The nerve growth factor: biochemistry, syn- thesis and mechanism of action. Ann Egg Neutgsgi 3:353-402, 1980. Vinores SA, Koestner A: The effect of nerve growth factor on undif- ferentiated glioma cells. anggt Lgtt 10:309-318, 1980. Vinores SA, Koestner A: Effect of nerve growth factor producing cells on anaplastic glioma and pheochromocytoma clones: Involvement of other factors. E Neutgggi Egg 6:389-401, 1981. Marushige Y, Raju NR Marushige K, Koestner A: Modulation of growth and of morphological characteristics in glioma cells by nerve growth factor and glia maturation factor. Qggggt Egg 47:4109-4115, 1987. Vinores SA, Koestner A: Reduction of ethylnitrosourea-induced neoplastic proliferation in rat trigeminal nerves by nerve growth factor. ang I Egg 42:1038-1040, 1982. Camp, RC, Koestner A, Vinores SA, Capen CC: The effect of nerve growth factor on ethylnitrosourea-induced neoplastic proliferation in rat trigeminal nerve. ygt Egthgl 21:67-73, 1984. Bocchini V, Angeletti PU: The nerve growth factor: purification as a 30,000 molecular weight protein. Etgg Ngtl Atgd 5&1. DEA 64:787-794, 1969. Greene LA: A quantitative bioassay for nerve growth factor (NGF) activity employing a clonal pheochromocytoma cell line. Etgin Egg 133:350-353, 1977. Vinores SA, Peres-Polo JR: The effect of nerve growth factor and antibodies to nerve growth factor on ethylnitrosourea carcinogenesis in mice. 1 anggt Egg glin anol 98:59-63, 1980. Dugle GE, Zwicker GM, Renne RA: Morphology of spontaneous brain tumors in the rat. ygt,2gthg1 16:318-324, 1979. 40. 41. 42. 43. 62 Goth R, Rajewsky M: Persistence of 06-ethy1guanine in rat brain DNA: Correlation with nervous system-specific carcinogenesis by ethylnitrosourea. Etgg Ngtl Aggg figi, DEA 71:639-643, 1974. Goodman R, Herschman HR: Nerve growth factor-mediated induction of tyrosine by hydroxylase in a clonal pheochromocytoma cell line. 2129 NQEL Aggfl 5&1, ugg 75:4587-4590, 1978. Levi-Montalcini R, Aloe L, Mugnaini E, Oesch F, Thoenen H: Nerve growth factor induces volume increase and enhances tyrosine hydroxylase synthesis in chemically axotomized sympathetic ganglia of newborn rats. Etgg Nat; Acad ggi, ugA 72:595-599, 1975. Guroff G, Dickens G, End D: The induction of ornithine decarboxy- lase by nerve growth factor and epidermal growth factor in PC—12 cells. 1 Nggtgghgg 37:342-349, 1981. CHAPTER 3 THE 1N NIIEQ EFFECTS OF NERVE GROWTH FACTOR (NGF) AND GLIA MATURATION FACTOR (GMF) ON ANAPLASTIC GLIOMA T9 CELL LINE: SCANNING AND TRANSMISSION ELECTRON MICROSCOPY STUDIES figmmgtg: The objective of this study was to explore the ultra- structural components of the anaplastic glioma T9 cells following exposure to glia maturation factor (GMF) and nerve growth factor (NGF). While the internal cytoskeletal elements, as examined by transmission electron microscopy (TEM), were essentially identical in both treatments, scanning electron microscopy (SEM) revealed distinctly different cell surface char- acteristics. GMF exposed cells were characterized by three dimensionally cylindrical and multipolar cell bodies with several long, slender cyto- plasmic processes which appeared to form an interconnecting network with neighboring cells. These processes were sparsely decorated with thin, hair-like structures. NGF induced short and broad cell bodies with wide cytoplasmic "foot” processes. Fine, web-like configurations, or lamelli- podia, sprouted from the ends of these processes which were ramified by filamentous, hair-like filopodia and secondary branches. Internal architecture, common to both treatments, included elaborate arrays of 5 nm microfilaments, 10 nm intermediate filaments and 25 nm microtubules dispersed throughout the perikaryon. An abundant golgi appa« ratus, SER, RER, fat globules, free ribosomes and an extensive network of tortuous mitochondria were evident in treated versus control cells. Junc- tional complexes (zonula occludens) between the adjacent cells were a prominent feature with both treatments. When compared to control cells, the nuclear:cytoplasm ratio was markedly reduced in treated cells. The nuclear contour appeared smooth with occasional irregular indentations. Often, there was a single nucleolus with condensed chromatin material. 63 64 W Glia maturation factor (GMF) is an acidic protein which is capable of promoting growth and differentiation of glioblasts.1 It is endoge- nous to the brain and has recently been purified to homogeneity.2 Studies have shown that GMF can induce proliferation and morphologic dif- 3 1 in cell cultures. ferentiation of Schwann cells and astrocytes Nerve growth factor (NGF) is a basic protein which is necessary for the development and maintenance of the sympathetic neurons and the devel- opment of sensory neurons in the peripheral nervous system and for dif- ferentiation of the neurons in the central nervous system.“'7 (A de- tailed review of NGF is contained in Chapter 1 and 2 of the dissertation.) In gigg and 13 gtttg studies have shown that NGF and GMF have both suppressing and differentiating effects on neuroectodermal tumor cells.8'14 After NGF exposure, PC-12 pheochromocytoma cells cease mito- sis, sprout neuritic processes, increase the rate of metabolite uptake, increase activity of the Na+, KI pump and induce transcription-depen- dent enzymes normally associated with terminally differentiated neu- rons.8'9 A similar conformational change was also observed when neuro- blastoma cell lines were treated with NGF.10'15. Recent studies have shown that GMF is capable of reversing or sup- pressing neoplastic characteristics of Schwannoma and glioma cells.3'14 We demonstrated changes16 in rat T9 cells which developed distinctively different characteristics in growth pattern and cell morphology after treatment with NGF and GMF. This chapter describes the scanning (SEM) and transmission electron microscopic (TEM) evaluation of the changes preci- pitated by NGF and GMF treatments. 65 Mariam Etgggtgtign gt NQE gag ENE. Beta-NGF was isolated and purified from male mouse submaxillary glands by the procedure of Bocchini and Angeletti,17 and the biological activity assayed by the PC-12 method.18 Partially purified GMF was prepared from bovine brains as described by Lim and Miller.19 Cgll ggltgtg. Stock monolayer cultures of rat T9 glioma 20'21 were maintained in DMEM8 containing 10% fetal bovine cells serumb, 4 mM glutamine, 100 units/ml of penicillin and 50 ug/ml of streptomycin in a humidified chamber with 7% C02 at 37°C. A single cell suspension was prepared by trypsinization, and cells (1-3x104/dish) were seeded in 5 ml of the stock culture medium onto Lux culture dishesc, (Permanox, 60 mm-diameter). The dishes were coated with 50 ug/ml of polyD-Lysine (Mr 30,000-70,000) for 5 minutes,22 washed with Hank's balanced salt solution and used immediately. One day after seed- ing, the culture medium was replaced with 5 ml of chemically defined serum-free medium, HL-ld, containing NGF (5 ug/ml) or GMF (5 ug/ml). The control cultures were maintained in the identical medium without NGF or GMF. HL-l contains DMEM:F12 base, HEPES buffer, insulin, transferrin, testosterone, sodium selenite, ethanolamine, saturated and unsaturated fatty acids and stabilizing proteins. It was supplemented with 4 mM glutamine. a Dulbecco's modified Eagle medium. Gibco Laboratories, Grand Island, New York. Armour Pharmaceutical, Kankakee, Illinois. c Miles Scientific, Naperville, Illinois. Ventrex Laboratories, Portland, Maine. 66 T9 cells treated with GMF,in particular, and with NGF tended to par- tially detach from the culture surface and round up with slight mechanical shocks. It became necessary to develop a special procedure for medium ex— change. The cells treated with GMF were fixed one day earlier (at 3 days) in order to keep the cell structures intact during cell culture and to prevent mechanical damage during the process of washing and fixation. After two days of culture in the experimental medium with daily medium change, 2 m1 of the medium was replaced by 2 ml of the fresh medium con- taining either 25 ug of NGF or 25 ug of GMF. On the following day, 1 ml of fresh medium containing 25 ug of NGF was added without withdrawing the existing medium to the culture of cells with NGF, and 1 ml of the medium without the factor was added to the control culture. The control cells which were cultured in the experimental medium for 3 days and 4 days were identical in the morphological characteristics. 67 WWWQEM cells After 4 days of culture, the control and the growth factor treated were processed for TEM by the following procedure: 1. The media was gently extracted and immediately replaced by 5 ml of 0.1M FDA for washing. After two careful washings, the cells were fixed in 4% glutaral- dehyde in 2% (0.1M) sucrose P04 (5 ml of 8% glutaraldehyde + 5 m1 0.2M P04 with 4% sucrose) for 15 minutes at room tempera- ture. The fixative was washed three times with 0.1M P04 + 2% sucrose. This was followed by osmification of the cells with 1% 0504 for 30 minutes. After one wash with 0.1M P04, the cells were rinsed once with deionized distilled water. 1% uranyl acetate was then added to the culture dish for 30 minutes. The cells were dehydrated in graded ethyl alcohol (30, 50, 65, 75, and 95%) for 5 minutes at each change followed by 3 treat- ments with absolute ethanol. The dehydrated cells were rinsed twice for 2 minutes in propy- lene oxide. Finally, the cells were embedded in 1 part propylene oxide + 3 parts resin mixture overnight inside a hood and then poly- merized in an oven for 2 days. The area of epon-embedded cells to be examined by TEM was first ob- served and outlined under a phase contrast microscope. Several 1-2 mm 68 sections were cut by wire-saw and glued to the tip of resin stubs suitable for mounting on the ultra microtomes for sectioning. Thin sections were stained with uranyl acetate and lead citrate, and examined with a Philips 300 transmission electron microscope. 69 WWW oco (SEM) Cells were plated on poly-D-lysine coated Lux Thermanox tissue cul- ture coverslipsc and incubated in Falcon 24-well plates with 0.5 ml/well (2 cm2) of the culture medium. The slightest disturbance during media change tended to float these coverslips in the solution. Although careful direct dropwise addition of the medium to the coverslips minimized the flotation, it did not prevent the dislodging of the cells from the coverslips. The procedure was modified to overcome this problem: after 2 days of culture in the experimental medium, 0.2 ml of the medium was replaced by 0.2 ml of the fresh medium containing either 2.5 ug of NGF or 7.5 ug of GMF; on the following day, 0.1 ml of the fresh medium contain- ing the same amount of NGF or GMF was added without withdrawing the ex- isting medium. After 4 days of culture, the cells were fixed by carefully adding 0.5 m1 of 5% glutaraldehyde in 0.1 M phosphate buffer (pH 7.4) on top of the medium. The cells were postfixed with 1% osmium tetroxide, dehydrated with successively increasing concentrations cf ethanol, and critical-point dried and sputter-coated with a 300-Ao layer of gold- palladium. Cells were observed with a JEOL JSM-35C scanning electron microscope at 15 RV. 70 Emits WWW anttgl 12 ggllg: The cells were large and polyhedral in shape with a tendency to form clusters. Generally, these cells exhibited a haphazard overlapping pattern of growth which often obscured individual cell bodies (Figure 2-1). The cell edges were studded with few minute finger-like processes or microvilli. On higher magnification (Figure 2-2), the cells consisted of relatively broad cell bodies with broad bands of "foot"-processes. Variably-sized microvilli were randomly scattered along cytoplasmic borders as well as over the cell bodies. Cells in the process of division was not an uncommon observation (Figure 2-3). FEE-M92112 Cells exposed to GMF showed remarkable phenotypic alterations characterized by greatly reduced size of cell bodies and development of several long, slender cytoplasmic processes which appeared to form an interconnecting (or communicating) network with neighboring cells (Figure 2-4). These processes were decorated with sparse, thin, hair- like structures. NEE-M93111 NGF treatment, on the other hand, revealed more dramatic changes in the T9 cells. In addition to suppression of cell division, there were several characteristic cytoplasmic configurations. The cell bodies were short and broad with relatively broad cytoplasmic ("foot”) pro- cesses (Figures 2-5, 2-6). These processes, similar to those in GMF- exposed cells, formed somatic links with the processes from the adjacent cells. Whenever a cell was individualized, these processes appeared to 71 serve as an anchoring device (Figure 2-7). Fine, "paintbrush"-like configurations or lamellipodia sprouted from the terminal end of these "foot" processes. These lamellipodia, in turn, were ramified by filamen- tous, hair-like filopodia and secondary branches, giving the entire struc- ture a ”crown-like" configuration (Figures 2-8, 2-9). WWW I2 anttol lelg. As depicted in Figure 2-10, T9 control cells were large, multipolar cells with large nuclei in proportion to the plump cytoplasm devoid of processes. There were scattered, small, round to oval mitochondria, attenuated smooth and rough endoplasmic reticulum (SER, RER), golgi complexes, abundant ribosomes and glycogen granules and peri- nuclear intermediate filaments (Figure 2-11). These cytoplasmic organ- elles appeared clustered close to the perinuclear region and were absent in the lamellae. The nuclear membrane appeared smooth with occasional areas of indentations. The nucleus contained multiple nucleoli with clumped chromatin. ENE-ttggtgg ggllg: These cells had an elongated shape with multiple extensions (processes) and a decreased nuclear size (The nuclear:cytoplas- mic ratio was morphometrically determined (Figure 2-12, 2-13). The nu- clear membrane was smooth with irregular shallow clefts. The heterochro- matin varied from coarsely condensed to irregularly tortuous with margin- ated nuclear chromatin. Dispersed among the elaborate cytoplasmic organ- elles were large, prominent spiralling mitochondria, stacks of distended SER, RER, golgi apparatus, myelin bodies, ribosomes and glycogen gran- ules. The edges of the cells had abundant, variably-sized, slender, "finger"-like or filamentous processes. Cells lying adjacent to one an- other were opposed at the plasma membranes (Figure 2-12). Cells distant 72 to one another were in contact by interdigitation of the filamentous pro- cesses (Figure 2-14). Higher magnification of the perinuclear region re- vealed numerous linear arrays of 10 nm intermediate filaments (IFs), intermingled and intimately connected with the cytoplasmic organelles (Figure 2-15). The IFs were arranged parallel to the larger 25 nm microtubules with occasional 4-5 nm microfilaments also detectable. The microtubules with their complement of intermediate filaments extended to the distal filopodia terminating at the secondary branches (Figures 2-16, 2-17). Figure 2-18 illustrates an example of filamentous mitochondria closely associated with numerous microtubules and IFs. NEE-ttggtgg 12 lelg: Ultrastructural changes induced were elabor- ate in comparison with the changes seen in T9 control and GMF-treated cells. These changes were characterized by reduction of the nuclear size versus an extended cytoplasmic surface, increase in cytoplasmic organ- elles, including mitochondria, SER, RER, golgi complexes, and numerous lipid inclusions with myelin figures (Figure 2-19, 2-20). The nuclear membrane was generally smooth and occasionally interrupted by irregular indentations. Nuclear heterochromatin was well dispersed and nuclei con- tained usually a solitary nucleolus. Higher magnification (Figure 2-21), revealed parallel arrays of tortuous and disoriented microfilaments intermingled with cytoplasmic organelles. The cistern of the RER and SEE were closely associated with these filaments. As seen in Figure 2-22, the cytoplasm was enriched with an elaborate meshwork of slender, long microtubules and IFs which appeared to converge at the centrioles. Figure 2-23 illustrates the terminal filo- podia and the secondary branches radiating in a "fern-like" manner. The body of the filopodia had numerous parallel stacks of rigid microbtubules 73 and IFs in close association with endoplasmic reticulum and mitochondria (Figure 2-24). This cytoplasmic framework terminated in the secondary branches of the filopodia (Figure 2-25). This well-arranged laminated cytoskeletal meshwork extending into the processes was not observed in untreated T9 cells. Intercellular junctional complexes developed between closely opposed plasma membranes of cells in both treatments (Figures 2-26, 2-27). This occurred by fusion of the outer leaflet of the adjacent cell membranes forming a single intermediate continuous dense line which obliterated the intercellular space. A thick band of electron-dense material can be seen on both sides of the cytoplasm at the free end of the complex. The structural appearance of this junctional complex is consistent with a zonula occludens or tight junction.23" Figure 2-1 74 Scanning electron micrograph of control anaplastic T9 glioma cells grown in HL-l media. The cells are variable in shape and size with disorientated piling-over growth pattern. Scant microvilli are discernible along the cytoplasmic borders. (440 X). 75 Figure 2—1 76 Figure 2-2 Scanning electron micrograph of a single control T9 cell. Notice the broad multipolar cell body with thick foot processes. Variably-sized microvilli are scattered randomly over the cell body and along the cytoplasmic edges. (2200 X). 77 78 Figure 2-3 .Scanning electron micrograph. The bilayered control T9 cells have specialized surface spherical structures or bulbous excrescences called zeiotic blebs. These exocytes are associated with cells during active mitosis. (960 X) 0" . 79 Figure 2—3 80 Figure 2-4 Scanning electron micrograph of anaplastic T9 glioma cells after exposure to GMF for 4 days. Notice cells are markedly reduced in size and have long cytoplasmic pro- cesses which form an interconnecting somatic network with the processes from the neighboring cells. (440 X) 81 Figure 2—4 Figure 2-5 82 Scanning electron micrograph of anaplastic T9 glioma cells after exposure to NGF for 4 days. Notice multipolar cells are relatively reduced in size and have broad cytoplasmic expansions or lamelldpodia from which protrude extremely fine thread-like processes, filopodia. The cells are con- nected by broad bands of cytoplasmic projections. (X 720) 83 Figure 2-5 84 Figure 2-6 Scanning electron micrograph. High magnification of T9 glioma cells in Figure 2-5. Cytoplasmic processes appear to blend with processes from adjacent cells forming an intricate network. The free ends of the projections have delicate expansion of lamellopodia (La) and fine filopodia (F1). (1200 X) 85 Figure 2-6 Figure 2-7 86 Scanning electron micrograph of a single T9 glioma cell 4 days after exposure to NGF. Notice poles of the cytoplas- mic projections are decorated with extremely fine and deli- cate film of lamellopodia, filopodia and secondary branches. The body surface and cytoplasmic edges are sparsely covered with microvilli. (1560 X) 87 Figure 2-7 88 Figure 2-8 Scanning electron micrograph. High magnification of a cytoplasmic projection of NGF-treated T9 glioma cell as seen in Figure 3-7. Notice elaborate and delicately expanded lamellopodia (La), filopodia (F1) and secondary branches (8). (3600 X) 89 Figure 2—8 90 Figure 2-9 High magnification of terminal portion of cell structure in Figure 2-8. 91 Figure 2—9 Figure 2-10 92 Transmission electron micrograph of control anaplastic T9 glioma cells grown in HL-I media for 4 days. Notice the cells are large and polyhedral in shape with high nuclear: cytoplasmic ratio. The nuclear contour is rough and sometimes jagged. The nucleus contains multiple dense nucleoli. Notice in the cytoplasm there is rudimentary mitochondria, SER, RER and free ribosomes. Microvilli are rare along the cell border. (3400 X) 93 Figure 2—10 94 Figure 2-11 Transmission electron micrograph. High magnification of the anaplastic T9 glioma cell in Figure 2-10. Notice in the perikaryon 5nm microfilament (mf) dispersed among RER, SER, mitochondria (m), and free ribosomes (r). n; nucleus. (34,200 X) 95 2-11 Figure 96 Figure 2-12 Transmission electron micrograph of T9 glioma cell after 4 days of exposure to GMF. The cells are relatively reduced in size with low nuclear:cytoplasmic ratio. The nucleus is irregular and contains condensed marginated chromatin material. Abundant dense mitochondria, SER, RER, myelin figures, and free ribosomes are present in perikaryon. Numerous wavy microvilli project from cell edges. Notice several points of somatic attachment with the neighboring cells. (3400 X) 97 Figure 2—12 98 Figure 2-14 Transmission electron micrograph of T9 glioma cells after 4 days exposure to GMF illustrates the intimate intertwin- ing of microvilli between opposing cells. (4500 X) 99 Figure 2-14 100 Figure 2-13 Transmission electron micrograph of T9 glioma cells after 4 days exposure to GMF. Notice growth of fine filamentous cytoplasmic processes. (3400 X) 101 Figure 2-15 102 Transmission electron micrograph. High magnification of T9 glioma cells 4 days after exposure to GMF. The perikaryon has abundant parallel rows of 25 nm microtubules (mt), 10 nm intermediate filaments (IF) and wavy bands of microfilaments (mf). These cytoskeletal frameworks are intimately associated with RER. mv; multivesicular body. (27,000 X) 103 2—15 Figure Figure 2-16 104 Transmission electron micrograph of a cytoplasmic process in T9 glioma cell 4 days after exposure to GMF. Notice stacks of microtubules, intermediate filaments and micro- filaments arranged along the long axis of the body of the cytoplasmic filopodia. (34,200 X) 105 106 Figure 2-17 Transmission electron micrograph of secondary branch of a filopodia in Figure 2-16. Equal complements of the cyto- skeletal support extends into the terminal branches. 107 108 Figure 2-18 Transmission electron micrograph. High magnification of T9 glioma cell after exposure to GMF. Notice the gigantic mitochondria are closely associated with streaks of micro- filaments and intermediate filaments. (45,000 X) 109 110 Figure 2-19 Transmission electron micrograph of T9 glioma cell after 4 days exposure to NGF. The condensed cells have low nuclear:cytoplasmic ratio with smooth nuclear contour and evenly dispersed heterochromatin. A - elonaged mitochondria; L - lipid inclusions. (4500 X) lll Figure 2—19 112 Figure 2-20 Transmission electron micrograph of T9 glioma cell after 4 days exposure to NGF. Notice profuse growth of tortuous villi from the cytoplasmic borders (arrow). M - myelin figure; A - elongated mitochondrias; R - rough endoplasmic reticulum. (4500 X). 113 2-20 igure F 114 Figure 2-21 Transmission electron micrograph of a T9 glioma cell 4 days after exposure to NGF. The perikaryon has many microfilaments (m) and intermediate filaments (f). 115 Figure 2-22 116 Transmission electron micrograph of T9 glioma cells after exposure to NGF. Notice streaks of microfilaments, inter- mediate filaments and microtubules converge around the two centrioles (c). R - rough endoplasmic reticulum; G - dilated golgi apparatus. (19,800 X) 117 Figure 2—22 118 Figure 2-23 Transmission electron micrograph of T9 glioma cells after exposure to NGF. The figure illustrates terminal portion of a cytoplasmic projection (P), lamellopodia (L), and filopodia (F) corresponding to the SEM details of cells in Figure 2-8. (3420 X) 119 120 Figure 2-24 Transmission electron micrograph. High magnification of cytoplasmic projection (P) as in Figure 2-21. The long axis of the process is enriched with parallel bundles of microfilaments, intermediate filaments and microtubules. Note that these cytoskeletal frameworks are in close asso- ciation with rough endoplasmic reticulum and mitochondria. (45,000 X) 121 122 Figure 2-25 Transmission electron micrograph. High magnification of filopodium as in Figure 2-21. The process contains rich complement of structural support. 123 124 Figures 2-26 Transmission electron micrograph of T9 glioma cells after 2-27 exposure to NGF (Figure 2-26) and GMF (Figure 2-27). Note the junctional complex, zonula occludens (JC) between two opposing cells. The outer leaflets of the opposing cell membrane form a single intermediate dense line. (81,000 X) 126 21mins This study revealed several cell surface and internal ultrastruc- tural changes induced in T9 anaplastic glioma cell line after treatment with NGF and GMF. While the internal fine components promoted by both these growth factors are essentially identical, the cell surface features are characteristically different. The T9 cells attained a cylindrical and multipolar appearance with reduced cell size after exposure to GMF. Long cytoplasmic processes or filopodia had a tendency to make connections or blend with the processes from other cells. The observed surface morpholo- gical alterations following GMF treatment was similar to glioblast matura- tion as studied by Lim gt g1 24 NGF promoted a totally different cell surface feature with flat- tened, broad cell bodies with "foot" processes ramified with weblike fine branches. This modification of phenotypic expression may be a form of cell maturation or differentiation process unique to NGF influence on the T9 cell line. 4 Greater evidence for cell maturation in the present study is the detailed revelation of the internal cell elements by TEM. Glia maturation factor and NGF promoted extension of elaborate cytoskeletal structures, notably the assembly of microfilaments, IFs and microtubules in the long axis of the cytoplasmic processes. This finding is consistent with other altudies,26‘28 in which it was shown that exposure of ganglion cells to NGF promoted neurite outgrowth and concurrent assembly of cytoplasmic neurofilaments and microtubules forming the basis of the structural sup- Port in the treated cells. 127 GMF may also stimulate the differentiation of glioblasts and activate the synthesis of these structural elements.24 The concentra- tion as well as the integrity of microtubules are essential for the mor- phological differentiation of cells (LimZS). The appearance of a dis- tended golgi apparatus, fat globules, abundant SER, RER, free ribosomes, and the increased number and size, as well as distribution of mito- chondria are all indicative of cell differentiation. The importance of cell junctions between adjacent cells may be viewed as indication of a shift towards a higher state of cell maturation. 23’29 It is believed that these junctions may contribute to contact in- hibition of cell growth.30 Malignant cells, on the other hand, lack contact inhibition which has been linked to the absence or defects in cell junction communication.31 In the present study, cell junctions (zonula occludens) between opposing NGF- and GMF-treated cells were prominent fea- tures. The NGF—mediated action is an event which follows NGF interaction with its receptor binding sites on responsive cells (Chapter II). It is logical, therefore, to assume that the effects of NGF and GMF on the T9 glioma cell were induced via a similar receptor mediated action. Studiesal’32 have elucidated that, following the protein-receptor bind- ing, biochemical events start at the level of the cell membrane, in the cytoplasm and in the nucleus. This molecular mechanism of mediation in- volves secondary messengers such as cyclic AMP and Ca+2.3a The over- all effects of these growth factors in the process of maturation of transformed cells may be conceived as a modulation or regulation of gene expression through an epigenetic mechanism. 128 It should be of great interest to investigate the possibility of cell-cell communication through gap-junction channels in the differen- tiated cells. Equally rewarding would be an in gigg experiment in which the animals could be inoculated with treated and untreated cells to in- vestigate the histologic appearance of the tumor cells in a clinical environment. REFERENCES - CHAPTER 3 10. ll. 12. 13. REFERENCES Lim R: Glia maturation factor. Q31; Igg Dgg 5121 16:305-322, 1980. Lim R, Miller JF, Hicklin DJ, Andresen AA: Purification of bovine glia maturation factor and characterization with monoclonal antibody. Eigghgmigttg 24:8070-8074, 1985. Lim R, Nakagawa S, Arnason BGW, Turriff DE: Glia maturation factor promotes contact inhibition in cancer cells. Etgg Ngtl Aggd figi N§A 78:4373-4377, 1981. Levi-Montalcini R, Angeletti PU: Nerve growth factor. Ehygig Egg 48:534-569, 1968. Bradshaw RA: Nerve growth factor. Ann Egg’fiigghgn, 47:191-216, 1978. Yankner BA, Shooter EM: The biology and mechanism of action of nerve growth factor. Egg Egg Eigghgm 51:845-868, 1982. Levi-Montalcini R, Aloe L: Differentiating effect of murine nerve growth factor in the peripheral and central nervous system of Xenopus laevis tadpoles. Ngtl Aggg figi NEE 82:7111-7115, 1985. Boonstra J, VanderSaag P, Moolenar W, Delaat S: Rapid effect of nerve growth factor on the Na+, Rf pump in rat pheochromocytoma cells. Exp lel Egg 131:452-455, 1981. Connolly JL, Green LA, Viscarello RR, Riley WD: Rapid sequential changes in surface morphology of PC-12 pheochromocytoma cells in response to nerve growth factor. 1 ggll E121 82:820-827, 1979. Reynolds CP, Perez-Polo JR: Induction of neurite outgrowth in the IMR-32 human neuroblastoma cell line by nerve growth factor. 1 Nggtgggi Egg 6:319-325, 1981. Vinores SA, Perez-Polo JR: The effect of nerve growth factor and antibodies to nerve growth factor on ethylnitrosourea carcinogenesis in mice. J Qggggt Egg glin Qnggl 98:59-63, 1980. Vinores SA, Koestner A: Reduction of ethylnitrosourea-induced neo- plastic proliferation in rat trigeminal nerves by nerve growth fac- tor. ggnggt Egg 42:1038-1040, 1982. Camp RC, Koestner A, Vinores SA, Capen CC: The effect of nerve growth factor and antibodies to nerve growth factor on ethylnitro- sourea-induced neoplastic proliferation in rat trigeminal nerves. Ngt Egthgl 21:67-73, 1984. 129 14. 15. 16. 17. l8. 19. 20. 21. 22. 23. 24. 25. 26. 27. 130 Lim R, Hicklin DJ, Ryken TC, Han Xue-Mei, Lui Kang-Nian, Miller JF, Baggenstoss BA: Suppression of glioma growth in gittg and in vivg by glia maturation factor. ang I Egg 46:5241-5247, 1986. Sonnenfeld KH, Ishii DN: Nerve growth factor effects and receptors in cultured human neuroblastoma cell lines. 1 Nggtgggi Egg 8:375- 391, 1982. Marushige Y, Raju NR, Marushige, K, Koestner A: Modulation of growth and morphological characteristics in glioma cells by nerve growth factor and glia maturation factor. anggt Egg 47:4109-4115, 1987. Bocchini V, Angeletti PU: The nerve growth factor: Purification as a 30,000 molecular-weight protein. Etgg Ngtl Acgd §g1 NEE 64:787- 794, 1969. Greene LA: A quantitative bioassay for nerve growth factor (NGF) activity employing a clonal pheochromocytoma cell line. Etgin Egg 133:350-353, 1977. Lim R, Miller JF: An improved procedure for the isolation of glia maturation factor. 1 Qg11 Ehggigl 119:255-259, 1984. X0 L, Koestner A, Wechsler W: Morphological characterization of nitrosourea-induced glioma cell lines and clones. Agtg,Nggtgggthg1 (Egtl) 51:23-31, 1980. K0 L, Koestner A, Wechsler W: Characterization of cell cycle and bio- logical parameters of transplantable glioma cell lines and clones. Am W (Bari) 51:107-111. 1980. Mckeehan WL, Ham RG: Stimulation of clonal growth of normal fibro- blasts with substrata coated with basic polymers. 1 §g11 5121 71:727-734, 1976. Farquhar MG, Palade GE: Junctional complexes in various epithelia. 1 3:11 5121 17:375-412, 1963. Lim R, Troy SS, Turriff DE: Fine structure of cultured glioblasts before and after stimulation by a glia maturation factor. Egg lel Egg 106:357-372, 1977. K0 L, Koestner A: Morphologic and morphometric analyses of butyrate- induced alterations of rat glioma cells 19 vittg. JNQL 65:1017- 1027, 1980. Levi-Montalcini R, Revoltella R, Calissano P: Microtubule proteins in the nerve growth factor mediated response. Interaction between the nerve growth factor and its target cells. Egggnt Etgg flgtm Egg 30:635-669, 1974. Levi-Montalcini R, Caramia F, Luse SA, Angeletti PU: In g1ttg effects of the nerve growth factor on the fine structure of the sensory nerve cells. Etgig Eggggtgh 8:347-362, 1968. 28. 29. 30. 31. 32. 33. 34. 131 Stockel K, Solomon F, Paravicini U, Thoenen H: Dissociation between effects of nerve growth factor on tyrosine hydrolase and tubulin synthesis in sympathetic ganglia. Ngtgtg 250:150-151, 1974. Attenburg B, Somers K, Steiner 8: Altered microfilament structure in cells transformed with a temperature-sensitive transformation mutant of murine sarcoma virus. anggt Egg 36:251-257. 1976. McNutt NS, Culp LA, Black PH: Contact-inhibited revertant cell lines isolated from SV40-transformed cells. IV. Microfilament distribution and cell shape in untransformed, transformed, and revertant balb/c 3T3 cells. E §g11 fi1g1 56:412-428, 1973. Kanno Y: Modulation of cell communication and carcinogenesis. Japan i Ehgg1gl 35:693, 1979. Stach RW, Perez-Polo JR: Binding of nerve growth factor to its receptor. 1 Nggtgggi Egg 17:1-10, 1987. Schechter AL, Bothwell MA: Nerve growth factor recptors on PC-12 cells. Evidence for two receptor classes with differing cytoskeletal association. §g11 24:867-874, 1981. Cremins J, Wagner JA, HalegonaZS: Nerve growth factor action is mediated by cyclic AMP and Ca+ /phospholipid-dependent protein kinases. J_§g11,3191 103:887-893, 1986. CHAPTER 4 IMMUNOHISTOCHEMICAL CHARACTERIZATION OF CENTRAL AND PERIPHERAL NERVE TUMORS INDUCED BY ETHYLNITROSOUREA IN RATS UTILIZING ANTI-GLIAL FIBRILLARY ACIDIC PROTEIN (GFAP), ANTI-LEU 7, AND ANTI-S-lOO PROTEIN ANTIBODIES 132 max: N-nitrosourea-induced central and peripheral nerve tumors in Sprague-Dawley rats were tested for immunoreactivity with glial fibrillary acidic protein (GFAP), S-100 protein and_human natural killer-l (HNK-l, also called Leu-7) isotope antibodies. The avidin-biotin-complex (ABC) method for GFAP and S-100, and the unlabeled antibody immunoperoxidase (peroxidase-antiperoxidase, PAP) method for HNK-l were used. In peripheral nerve neurinomas, S-100 immunoreactivity varied from strongly positive in differentiated tumor cells to weakly positive in ana- plastic neurinomas. None of the neurinomas, irrespective of the degree of differentiation, reacted to GFAP and HNK-l antibodies. In CNS tumors, S-lOO and GFAP were reliable markers for astrocytomas and were especially useful in characterizing the astrocytic cells within mixed gliomas, a feature which is not readily demonstrable with routine hematoxylin and eosin stain preparations. The perikaryon and processes of reactive astrocytes showed intense positivity with-GFAP and S-100 protein compared to normal and neoplastic counterparts. Oligodendrogliomas were consistently negative for GFAP and S-100, except for 3 out of 36 that showed a superficial rim of weak reaction with S-100 protein antibody. S-lOO expression was demonstrable in meningiomas and glioependymomas. The HNK-l antibody stain was not useful in our investigation. 133 W Prior to the advent of immunohistochemistry, the standard method for characterizing tumors included light and electron microscopic studies. Immunochemistry provided an exciting new diagnostic tool for distinguish- ing tumors of diverse histogenetic origin.1'4 The principle is based on the demonstration of antigen in tissue samples utilizing poly or monoclo- nal antibodies produced selectively against the cell antigenic markers. In neuro-oncology, immunohistochemistry has tremendous benefits when used in adjunct with conventional procedures. The nervous system is composed of a variety of cell types. In order to understand the behavior and management of the tumors arising from different cell precursors, it is extremely important to accurately identify the major tumor cell compo- nent and also characterize the degree of anaplasia. The mammalian nervous tissue cytoskeleton consists of intermediate filaments (IF), microfilaments and microtubules. Among the IF are the neurofilaments, glial fibrillary acidic protein (GFAP), and vimentin. While the former is the principal constituent of IF in the neurons,5'8 of the latter two, GFAP is the predominant IF in the glial cells.9'10 Antibodies against GFAP are commonly used to characterize normal, reactive and neoplastic cells of astroglial origin.9'11 Varying degrees of immunoreactivity to GFAP have been demonstrated in rat and human Schwann cells,6’12’13 human ependymal cells,12'13 l4 rat enteric glial th,11’15'16 and rat cells, tumors of human peripheral nerve shea sensory and sympathetic ganglia satellite cells.15 Similar reactions have been demonstrated in non-nervous tissues such as in human 134 17 18 In salivary gland adenomas and human epiglottis cartilage. neuro-oncology, demonstration of GFAP has several distinct objectives: - to identify astrocytes and tumors arising from glial cells. - to distinguish between glial and non-glial tumors. - to demonstrate astrocytic components in mixed CNS tumors such as glioependymomas. - to identify tumor cells outside the CNS, i.e., expression of GFAP may indicate the metastatic nature of astroglial tumors. The next useful antigenic marker is the S-100 protein.19 It was first isolated as a highly acidic soluble protein from rabbit and bovine nervous systems.20'21 The compound is synthesized in glial 22,23 cells, and Schwann cells,24 but its precise function is not known. Almost all cells of necroectodermal origin can be immunohisto- chemically shown to contain varying amounts of S-100 protein.25'28 How- ever, S-100 immunoreactivity is not restricted to the cells of the nervous system, but a diverse spectrum of mammalian tissues have been demonstrated to express this protein.29"32 Despite the wide range of distribution, S-100 is considered a valuable cell marker for normal, reactive and neo- plastic tissue of neuroectodermal origin.19’30 Another useful method of cell identification can be accomplished by the use of monoclonal antibodies (MAB) raised against specific cell- surface antigenic determinants or epitopes. One such MAB is anti-HNX-l, also known as anti-Leu 7, originally raised against the human T-lympho- blastoid cell line, which has been reported to recognize antigens on human cells with natural killer cell activity.32 Subsequent work has revealed that this MAB cross-reacts with a range of human and rodent tissues, 135 including the elements of the nervous system.33"39 The advantage of using HNX-l in determining tumors of central and peripheral nervous sys- tems is based on the principle that this MAB recognizes a single specific cell surface antigen.31 Although a high percentage of oligodendrogli- omas display HNK-l positivity, this immunostain cannot be considered a specific markerfor oligodendrogliomas, since other neuroepithelial tumors also react with this MAB.39 In one other study,40 the cellular morphology of ENU-induced rat brain microtumors were compared with those in macrotumors by deter- mining the levels of GFAP and Leu 7 in both neoplasms. While the micro- tumors were found to be negative for both these stains, macrotumors were positive either for GFAP or Leu 7. The aim of the present test was to characterize, utilizing the HNK-l MAB and antibodies against GFAP and S-100 protein with immunoperoxidase method, all neurogenic tumors obtained from rats after exposure to ENU, and also, to determine the relationship, if any, between the degree of anaplasia or the stage of differentiation and intensity of the reaction by these antibodies. 136 WWW 12111213. ENU-induced tumors of the central (CNS) and peripheral (PNS) nervous systems in Sprague-Dawley rats (described in Chapter II) were investi- gated. Samples were fixed in 10% buffered formalin and embedded in paraf- fin. Several 5 um thick sections were cut, and one section of each tumor was routinely stained with hematolylin and eosin for histopathological classification. The remaining sections were used for immunohistochemical study. u he ammummmnm In order to prevent detachment of the tissue sections from the glass slides during trypsinization (see below), the glass slides were first coated with poly-L-lysinea (25 mg/200 ml distilled water). The avidin- biotin-complex (ABC) method“1 for immunoperoxidase stain was used. A b commercially available ABC kit, Vectastain, was obtained from Vector Laboratories. a Sigma, St. Louis, Missouri b Vector Laboratories, Inc., Burlingame, California (1) (2) (3) (4) (5) (5) (7) (8) (9) 137 Sections were deparaffinized in xylene for 5 minutes and rehy- drated through graded ethyl alcohol concentrations to distilled water for 3 minutes at each change.‘ After a rinse with phosphate buffer saline (PBS), tissues were digested for 3 minutes in prewarmed (37°C) 0.1% trypsin, washed in distilled water, and dipped in PBS for 3 minutes. Two drops of 3% hydrogen peroxide (H202 in methanol) was added for 5 minutes to the sections to block the endogenous peroxide activity. Sections were thoroughly rinsed for 3 minutes in two changes of PBS, and all further incubations were carried out in a humi- difier at room temperature. Two to three drops of normal goat serum was applied for 30 minutes. After tapping off excess serum, the sections were treated with 2-3 drops of 1/300 anti-rabbit GFAPC for 30 minutes. Sections were rinsed for 30 minutes in two changes of PBS and then incubated with 2 drops of biotinylated antibody (goat- anti-rabbit IgG) for 30 minutes. Following two washings in PBS, sections were exposed to 2 drops of avidin-biotinylated conjugate (ABC) for 30 minutes. The sections were rinsed in 2 changes of PBS, and the reaction was developed in freshly prepared 3.3'-diaminobenzidine tetra- hydrochloride (DAB) for 10 minutes. 138 (10) After a brief wash in deionized water, the sections were treated with copper sulfate solution to enhance the DAB de- velopment, then rinsed in deionized water and counter-stained with Gill's hematoxylin for 30 seconds. (11) Finally, the sections were dehydrated in graded alcohol, cleaned in xylene and mounted in permount mounting medium. Positive controls stained in parallel included human cerebral cortex (obtained during autopsy) and rat brain. s—mmm m-mmrainb All sections were immunostained using anti-S-lOO protein anti- body.b The ABC method employed was essentially identical to that applied for GFAP test, except that in step 6, bovine anti-rabbit S-100 antibody (1/200) was substituted for 30 minutes. Positive controls stained in parallel consisted of human normal skin, human cutaneous melanoma, human cerebral cortex (obtained during autopsy), and rat brain. WWW-IWW(m-IM) Sections were immunostained with HNK-l MAB, commercially available as Anti-Leu 7.‘1 The following 4-step peroxidase-anti-peroxidase (PAP) method was based on the procedure previously described.42’43 P P (1) After the sections were deparaffinized in xylene for 10 minutes and delipidized in chloroform for 30 hours, they were rehy- drated through graded methyl alcohol (95% and 70%) to distilled water for 3 minutes at each change. c Dako Corporation, Santa Barbara, California d Becton and Dickson, Mountain View, California (2) (3) (4) (5) (6) (7) (8) 139 Sections were treated for 30 minutes with hydrogen peroxide (H202 in methanol) to block endogenous peroxidase activity. After a 5 minute wash in PBS, sections were exposed to dilute normal rabbit serum,c (1% in PBS) for 20 minutes to inhibit non-specific binding of immunoglobulins. Without rinsing, the sections were incubated overnight at 4°C with Anti-Leu 7 diluted 1:300 in PBS. After washing in buffer for 5 minutes, they were incubated in link antibodyc for 20 minutes (rabbit anti-mouse immunoglo- bulin). Sections were washed in PBS before incubation for 1 hour in peroxidase-anti-peroxidase complex.c After washing for 5 minutes in PBS, sections were immersed in freshly prepared 3.3'-diaminobenzidine tetrahydrochloride anus)“ for 10 minutes. The sections were washed in deionized distilled water, counter stained with Gill's hematoxylin (1 minute), dehydrated in graded series of ethanol, cleaned in xylene, and mounted in permount mounting media. 140 WQEW mnflla Sodium chloride (NaCl) ........................ . ................. 16 gm Sodium phosphate dibasic anhydrous (Na2HP04) ................ 2.5 gm Potassium phosphate, monobasic anhydrous (KH2P04) ........... 0.4 gm Potassium chloride (RC1) ........................................ 0.4 gm Triple distilled water ....................................... 2000 ml Mrs Man PBS, prewarmed 37°C .......................................... 200 ml Trypsin (Sigma-type II from porcine pancreas) .................. 0.5 gm W-WWWW 3.3' Diaminobenzidine tetrahydrochloride (DAB) ................... 5 mg PBS .................................................... 4 ......... 10 m1 3% Hydrogen peroxide (H202) ................................. 3 drops 141 REESE-IS. .I‘Legrincmas fi-1QQ gtgtg1n: All 55 peripheral nerve neurinomas examined showed positive stain for S-100 protein, although the intensity of the immuno- stain indicated an inverse relation with the degree of malignancy. Figures 3-1 and 3-2 illustrate differentiated trigeminal nerve neurinomas with a majority of the tumor cells expressing S-100 protein. Intense positive reaction was generally concentrated in the cell membranes, in cytoplasm and nucleus. A clear halo was sometimes evident between the strongly positive cell membranes and the nuclei. On the other hand, ana- plastic or undifferentiated neurinomas had scattered nests of cells that showed mild to weakly positive stain, while a majority of the anaplastic cells were negative. GEAR: In contrast to S-100 protein reaction, none of the 55 neuri- nomas had any evidence of immunoreactivity against the GFAP antibodies. Figure 3-3 shows a serial section of the neurinoma in Figure 3-1 immuno- stained for GFAP. Note that the section is at the junction of CNS and PNS (trigeminal nerve). The strongly positive GFAP stain is confined to the astroglial cells and their processes in the CNS, providing a distinct demarcation from the GFAP negative trigeminal nerve neurinoma. Wm 5-199: Of the 12 astrocytomas, 9 well-differentiated and 3 moder- ately undifferentiated astrocytomas stained positively for S-100 protein. Figure 3-4 illustrates a typical differentiated astrocytoma in which a majority of neoplastic cells show strong positive reaction on the cell membranes as well as in the cytoplasm, while the thin cytoplasmic 142 processes stain granular. The staining intensity and the number of posi- tive cells were not determined by the degree of the tumor cell differen- tiation. £252: The expression of GFAP in astrocytomas was identical to that of S-100 protein. The reactive astroglial cells within and peripheral to the tumor stained more intensely than the actual neoplastic cells and the normal counterpart. As depicted in Figure 3-5, GFAP positive tumor cells in a differentiated cerebral astrocytoma have strong reactivity in the cytoplasm and processes. The 3 undifferentiated astrocytomas contained scattered individual positive staining cells (Figure 3-6). However, the tumor border was speckled with characteristically strongly immunostained reactive astro- cytes. Winn fi-1QQ: Of the 36 oligodendrogliomas, there was weak positive stain- ing of 3 and all others were negative. As indicated in Figure 3-7, the small oligodendrocytes have a thin rim of cell membrane showing positiv- ity. However, the bulk of the tumor-supporting framework is composed of fibrous processes of reactive fibrillary astrocytes. The cytoplasm and broad processes of these cells stained intensely positive. QEAE: All oligodendrogliomas were negative for GFAP. However, there was positive staining reactive fibrillary astrocytes within the tumors. Figure 3-8 illustrates the densely stained, large, multipolar reactive cells and the tortuous fibrillary processes stretched between the tumor cells. These stellate cells were randomly scattered throughout the tumors . 143 nixed Elissa; In 20 mixed gliomas (astrocytes-oligodendrocytes), both S-lOO and GFAP stained intensely positive in neoplastic or reactive astrocytes. Figures 3-9 and 3-10 depict a typical mixed glioma in which astroglial cells were fairly evenly spread amongst the small oligodendroglial cells. These stellate cells had abundant cytoplasm and often coarse tapering cell processes, consistent with astrocytes. The oligodendroglial cells were consistently negative in all cases. Wm: The ependymal component showed a positive reaction for GFAP and S-100 protein. Figure 3-11 represents a glioependymoma stained for S-100 protein. Note there are several clusters of variably-sized positive, large, ependymomal cells. In Figure 3-12, GFAP positive ependymomal cells have a delicate fibrillary network radiating from the cytoplasm. Maxim None of the 4 meningiomas tested stained positive for GFAP. All 4 had positive staining of variable intensity for S-100. As can be seen in Figure 3-13, a majority of meningioma cells in the vicinity of the blood vessels show varying degrees of immunostain. The negatively stained ery- throcytes and vascular endothelial cells serve as contrast. Anti-1&3 Z (Anti-Mi) Despite several attempts, specimens of all CNS and PNS tumors tested had negative staining with anti-Leu 7. This was felt to be due to an expression of presumably different isotopes in rat cells (Rubinstein, personal communication) not recognized by the Anti-Leu 7 antibodies. 144 Figure 3-1 Photomicrograph of a trigeminal nerve neurinoma immuno- stained for S-100 protein. CNS-PNS junction (arrow). Immunoperoxidase, ABC method; Hematoxylin counter stain. 400 X. Figure 3-2 Higher magnification of trigeminal nerve neurinoma in Figure 3-1. Notice positive reaction on cell membranes, in cyto- plasms and nuclei. 640 X. 145 146 Figure 3-3 Photomicrograph of trigeminal nerve neurinoma immunostained for GFAP. The neurinoma cells are negative, whereas the astrocytes in the CNS are strongly positive. CNS-PNS junction (arrow). Immunoperoxidase, ABC method. Hematoxy- lin counterstain. 400 X. 147 Figure 3-3 148 Figure 3-4 Photomicrograph of an astrocytoma immunostained for S-100. Immunoperoxidase, ABC method. Hematoxylin counterstain. 400 X. 149 Figure 3—4 150 Figure 3-5 Photomicrograph of a cerebral astrocytoma immunostained for GFAP. Immunoperoxidase ABC method.’ Hematoxylin counter- stain. 400 X Figure 3-6 Photomicrograph of a cerebral undifferentiated astrocytoma immunostained for GFAP. Notice peripheral large reactive astrocytes stained intensely while an occasional astrocytoma cell within the tumor showed a weak reaction. Immunoperoxidase, ABC method. Hematoxylin counterstain. 160 X. 152 Figure 3-7 Photomicrograph of an oligodendroglioma immunostained for S-100. Notice strong positive reaction in the cytoplasm and processes of reactive astrocytes within the tumor. The small oligodendroglioma cells showed weak positive reaction (arrow). Reactive fibrillary astrocytes (arrowhead). Immunoperoxidase, ABC method. Hematoxylin counterstain. 400 X. 153 Figure 3—7 Figure 3-8 154 Photomicrograph of an oligodendroglioma immunostained for GFAP. Notice intense positive reactions in the reactive fibrillary astrocyte within the tumor. The oligodendro- glials are negative for GFAP. Immunoperoxidase, ABC method. Hematoxylin counterstain. 400 X. 155 Figure 3-9 Figure 3-10 156 Photomicrograph of a mixed glioma immunostained for 8100. Notice strong positive reaction in the broad astrocytic cytoplasm and its thin processes. The oligodendroglioma cells do not stain for $100. Immunoperoxidase, ABC method. Hematoxylin counterstain. 400 X. Photomicrograph of a mixed glioma immunostained for GFAP. The astrocytic component of the tumor shows strong positive reaction, whereas the oligodendroglioma cells are uniformly negative. Immunoperoxidase, ABC method. Hematoxylin counterstain. 400 X. Figure 3—10 158 Figure 3-11 Photomicrograph of a glioependymoma immunostained for S-100. Notice many positive cells forming the pseudo- rosettes. Immunoperoxidase, ABC method. Hematoxylin counterstain. 400 X. Figure 3-12 Photomicrograph of a glioependymoma immunostaind for GFAP. Notice several tumor cells have positive stain in the thin membraneous cell processes. Immunoperoxidase, ABC method. Hematoxylin counterstain. 400 X Figure 3-11 Figure 3-12 \_ 160 Photomicrograph of a meningioma immunostained for 8100. Notice relatively strong reaction in cells surrounding the blood vessels. Immunoperoxidase, ABC method. Hematoxylin counterstain. 400 X. Figure 3-13 161 Figure 3-13 162 W Glial fibrillary acidic protein, S-100 and Leu 7 antibodies are a few amongst a battery of immunohistochemical stains utilized in the dif— ferential diagnosis of naturally-occurring neurogenic neoplasms. These staining procedures have been specifically helpful in cases of doubtful or questionable tumor cell origin or tumor cell composition. The present work utilizing experimentally induced intra- and extra- cranial neurogenic tumors in rats discloses significant information in respect to the three cell markers employed. (1) The expression of intermediate filaments (GFAP) and S-100 pro- tein correlated with the degree of tumor cell differentiation. (2) Although GFAP and S-100 protein antibodies are reliable markers of carcinogen-induced neurogenic tumors, the anti-Lou 7 anti- bodies utilized did not recognize rat neurogenic tumor cells. (3) Tumors suspected of mixed cell composition can be characteris- tically identified by the use of these immunostains. In peripheral nerve neurinomas, it was evident that S-100 protein was the most reliable cell marker, although the immunostaining intensity and the number of positive cells decreased with increasing degree of tumor anaplasia. The correlation between malignancy and expression of S-100 protein in carcinogen-induced neurinomas was not determined; however, it can be hypothesized that malignant cells which resemble primitive embry- onal stem cells are less endowed with essential molecules necessary for 13’17'18 of the oc- the expression of the protein. Contrary to reports currence of GFAP positive cells in neurinomas, in the present study, neither the differentiated nor the anaplastic carcinogen-induced 163 neurinomas showed any evidence of GFAP reaction. The presence of GFAP in normal and neoplastic Schwann cells is controversial. It has been sug- gested that a subpopulation of Schwann cells with cytoplasmic 10 nm in- termediate filaments could cross-react with GFAP antisera, and this could account for the report of GFAP positive cells in neurinomas.43 The most consistent immunochemical reaction with GFAP and S-100 was observed in astrocytomas. Although the number of undifferentiated astro- cytomas tested was small (3), immunoreactivity with S-100 was not affected by the degree of tumor anaplasia. Our test results indicated that the num- ber of GFAP positive cells and staining intensity of positive cells was lower in undifferentiated astrocytoma cells when compared to the differ- entiated neoplastic cells. This is consistent with previous re- port“2.13.48 that the expression of GFAP in astrocytoma cells decreases with increasing cellular atypia. Since GFAP is an indication of astrocyte maturation (or differentiation), its reported absence in anaplastic tumor cells and progenitor cells may be viewed as an inability of less differen- tiated cells to produce GFAP. Thus, the significance of GFAP staining in different types of astrocytomas is controversial. Tascos gt 3113 docu- mented that all astrocytomas react positively with GFAP, although the staining intensity and the number of positive cells decreased.with advanc- ing stages of tumor anaplasia. For example, gemistocytic astrocytes stained most intensely, followed by poorly differentiated astrocytoma cells (grade III-IV) where only isolated cells were positive. In glio- blastoma multiforme, only the gemistocytic and multinucleated giant cell components showed varying degrees of reaction. However, two recent 43,49 studies showed that all astrocytomas had strong GFAP stain, regard- less of their degree of malignancy. 164 Both GFAP and S-100 stains proved useful in the case of mixed gliomas by staining the astroglial and not the oligodendroglial components of the tumor justifying the classification. It must be emphasized that small numbers of astrocytes can normally be present in oligodendrogliomas either as peripheral reactive cells or cells entrapped in the tumor (Figure 3-6), features that may not be readily detectable in routine hema- toxylin and eosin preparations. The expression of GFAP and S-100 in glioependymoma, and the presence of S-100 positive cells in meningioma observed in our study are in agree- ment with others.13’43 The meningioma cells in the vicinity of the blood vessels exhibited a high degree of S-100 positive reaction. In glioependymoma, S-100 and GFAP expression were detected in the cytoplasm and along the processes of cells forming the pseudorosettes. Contrary to a report42 of strong S-100 immunoreactivity of oligodendrogliomas, in our study only 3 of the 36 oligodendrogliomas showed superficial positive reaction. Our experience was also in dis- agreement with report312.49 of positive GFAP in oligodendrogliomas. The expression of GFAP in oligodendrocytes has been controversial. Some authorsl"13'48 believe oligodendrogliomas are GFAP negative, while otherslz'l‘B'50 described GFAP immunoreactivity in these tumor cells. However, it was not conclusively determined whether the positive cells in the above studies were true neoplastic oligodendroglial cells or small gemistocytic astrocytes, which are normal components of most oli- godendrogliomas. It may also indicate that the positive cells represent a totally different class of glia termed gliofibrillary oligodendrocytes which, during development, can transiently express GFAP.so 165 Results of our investigation differ remarkably in some areas from those obtained with human tumor studies. Studies involving specimens derived from diverse genetic origin (human versus animal), natural versus experimentally induced tumors and laboratory test procedures could poten- tially contribute to inconsistent results. Despite these reservations, the present study further confirmed that GFAP and S-100 protein antibodies have useful application in diagnostic pathology. This investigation was particularly valuable in characterizing rat astrocytomas and components of mixed gliomas. Although the study was limited to 3 immunohistochemical stains, it has set a strong basis for testing of inconclusive and/or tumors of suspected neurogenic origin by including a wider spectrum of cell ”markers” such as vimentin, neuron specific enolase (NSE), myelin basic protein (MBP), carbonic anhydrase C (CA G) and neurofilaments (NF). REFERENCES - CHAPTER 4 10. 11. 12. 13. Taylor CR, Kledzik G: Immunohistologic techniques in surgical patho- logy. A spectrum of “new" special stains. E23 £gth21 12:590-596, 1981. Bonin JM, Rubinstein LJ: Immunohistochemistry of central nervous system tumors. Its contributions to neurosurgical diagnosis. A Nggt2g21g 60:1121-1133, 1984. Osborn M, Weber K: Tumor diagnosis by intermediate filament typing: A novel tool for surgical pathology. ,Lg2,12gggt 48:372-394, 1983. Eng LF, Rubinstein LJ: Contribution of immunohistochemistry to diagnostic problems of human cerebral tumors. 1,fl1gt22hgm Egtgghgm 26:513-522, 1978. Leim RKH, Yen S-H, Salomon GB, Shelanski ML: Intermediate filaments in nervous tumors. 1 9211 E121 79:637-645, 1978. Yen SH, Fields XL: Antibodies to neurofilaments, glial filament, and fibroblast intermediate filament protein bind to different cell types of the nervous system. 1 Qg11 E121 88:115-126, 1981. Schiffer D, Giordana MT, Mauro A, and Migheli A: Immunohistochem- istry in neuro-oncology. Egg 5221 E1gt2ghgm 30:253-265, 1986. Trojanowski JQ, Walkenstein N, Lee VMY: Expression of neurofilament subunits in neurons of the central and peripheral nervous system: An immunohistochemical study with monoclonal antibodies. 1 Nggtgsci 6:650-660, 1986. Eng LF, Vanderhaeghen JJ, Bignami A, Gerstl B: An acidic protein isolated from fibrous astrocytes. E1213 Egg 28:351-354, 1971. Bignami A, Eng LF, Dahl D, Llyeda CT: Localization of the glial fibrillary acidic protein in astrocytes by immunofluorescence. E1213 Egg 43:429-435, 1972. Lazarides E: Intermediate filaments as mechanical integrators of cellular space. Ngtgtg 283:249-256, 1980. DeArmond SJ, Eng LF, Rubinstein L5: The application of glial fibril- lary acidic (GFA) protein immunohistochemistry in neuro-oncology. A progress report. £2th21 Egg Etggt 168:374-394, 1980. Tascos NA, Parr J, Gonatas NK: Immunocytochemical study of the glial fibrillary acidic protein in neoplasm of the central nervous system. Hun Egth21 13:454-458, 1982. 166 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 167 Dahl D, Chi NH, Miles LE, Nguyken BT, Bignami A: Glial fibrillary acidic (GFA) protein in Schwann cells: Fact or artifact? 1 WW 30:912-918. 1932- Jessen KR, Thorpe R, Mirsky R: Molecular identity. distribution and heterogenicity of glial fibrillary acid protein: An immunoblotting and immunohistochemical study of Schwann cells, satellite cells, enteric glia and astrocytes. l Nggt2ggt21 13:187-200. 1984. Jessen KR, Mirsky R: Glial cells in the enteric nervous system contain glial fibrillary acidic protein. (Letter) Ngtntg 286:736-737, 1980. Memoli VA, Brown EF, Gould VE: Glial fibrillary acidic protein (GFAP) immunoreactivity in periphral nerve sheath tumors. E1ttg- gttggt 22th21 7:269-275, 1984. Gould VE, Moll R, Moll I, Lee I, Schwechheimer K, Franke WW: The intermediate filament complement of the spectrum of nerve sheath neoplasms. ,Lgh Ingggt 55:463-474, 1986. Nakazato Y, Ishizeki J, Takahashi K. Yamaguchi H, Kamei T, Mori T: Localization of S-100 protein and glial fibrillary acidic protein- related antigen in pleomorphic adenomas of the salivary gland. L22 Ingggt 46:621-626, 1982. Kepes JJ, Rubinstein LJ, Chiang H: The role of astrocytes in the formation of cartilage in gliomas: An immunohistochemical study of four cases. Am l_£gth21 117:471-483, 1984. Weiss SW, Langloss JM, Enzinger FM: Value of $100 protein in the diagnosis of soft tissue tumors with particular reference to benign and malignant Schwann cell tumors. Lg2,1ngggt 49:299-308, 1983. Moore BW: Soluble protein characteristic of the nervous system. flashes 3.129111: Bea 9.9mm 19:739-744. 1965. Moore BW, McGregor DJ: Chromatographic and electrophonetic frac- tionation of soluble proteins of brain and liver. 1 E121 thm 240:1647-1653, 1965. Pfeiffer SE, Herschman HR, Lightbody J, Sato G: Synthesis by a clonal line of rat glial cells of a protein unique to the nervous system. 2 £211 Ehgg121 75:329-339. 1970. Edstrom A, Haglid KG, Kanje M, Ronnback L, Walum E: Morphological alterations and increse of 8100 protein in cultured human glioma cells deprived of serum. E32 §g11 Egg 83:426-429. 1974. Pfeiffer SE, Wechsler W: Biochemically differentiated neoplastic cloe of Schwann cells. £122 Ngtl A222 521 69:2885-2889, 1972. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 168 Matus A. Maghal S: Immunohistochemical localization of $1200 protein in brain. Ngtgtg 258:746-748. 1975. Ludwin SK, Kosek JC. Eng LF: The topographical distribution of 8100 and GFA proteins in the adult rat brain: An immunohistochemical study using horseradish peroxidase-labeled antibodies. J §2n2 Nggt2l 165:197-208, 1976. Stefansson K, Wollmann RL, Moore BW: Distribution of 8100 protein outside the central nervous system. E1212 Egg 234:309-317, 1982. Tabuchi K, Kirshc WN: Immunocytochemical localization of 8100 protein in nervous and glia of hamster cerebellum. E1212 Egg 92:175-180. 1975. Takahashi K, Isobe T, Ohtsuki Y, Akagi T. Sonobe H. Okuyama T: Immunohistochemical study of the distribution of and subunits of $100 protein in human neoplasm and normal tissues. y1tgh2gg 5122 E 45:385-396. 1984. Kahn HJ, Marks A, Thom H, Baumal R: Role of antibody of 8100 protein in diagnostic pathology. Am 1 £113 Egth21 79:341-347. 1983. Perentes E, Rubinstein LJ: Recent applications of immunoperoxidase histochemistry in human neuro-oncology. Atgh Egth21 L22,M22 111:796-812, 1987. Haimoto H, Hosoda S, Kato K: Differential distribution of immunore- active 8100- and 8100- proteins in normal non-nervous human tissue. ,Lgh lngggt 57:489-498. 1987. Abo T, Balch CM: A differentiation antigen of human NK and NK cells identified by a monoclonal antibody (HNK-I). J Imnnn21 127:1024-1029, 1981. Schuller-Petrovie S, Gebhart W, Lassmann H. Rumpold H, Kraft D: A shared antigenic determinant between natural killer cells and nervous tissue. Ngtgtg 306:179-181, 1983. Lipinski M, Braham K, Caillaud JM, Carlu C, Tursz T: HNK-I antibody detects an antigen expressed on neuroectodermal cells. 1 E32 Egg 158:1775-1780, 1983. Sato 8, Babe H, Tanaka M, Yanagissawa K, Miyatake T: Antigenic determinant shared between myelin-associated glycoprotein from human brain and natural killer cells. ,E12m22 Egg 4:489-494, 1983. Perentes E, Rubinstein LJ: Immunohistochemical recognition of human neuroepithelial tumors by Anti-Leu-7 (HNK-I) monoclonal antibody. Agtg Nggtgggth21 69:227-233. 1986. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 169 Yoshino T, Motoi M. and Ogawa K: Immunohistochemical studies on cellular character of microtumors induced by ethylnitrosourea in the rat brain utilizing anti-Leu 7 and anti-glial fibrillary acidic protein antibodies. Agtg Ngttgggth21 (Egtl) 66:167-169. 1985. Swanson PE, Manuel JC, Wick MR: Immunoreactivity for Len-7 in neurofibroma and other spindle cell sarcomas of soft tissue. Am i Egthgl 126:546-560, 1987. May EE, Perentes E: Anti-Leu-7 immunoreactivity with 83 tumors: Its value in the diagnosis of prostatic adenocarcinoma. E1gt22gt22- 12gg 11:295-304. 1987. Reifenberger G, Szymas J, Wechsler W: Differential expression of glial- and neuronal-associated antigens in human tumors of the central and peripheral nervous system. Agtg Nggtgngthgl (hell) 74:105-123. 1987. Hsu SM, Raine L, Fanger H: The use of anti-avidin antibody and avidin-biotin-peroxidase-complex in immunoperoxidase techniques. Am 1 £113 Egthg1 75:816-821. 1981. Sternberger LA, Hardy PH Jr, Cuculis JJ, Meyer HG: The unlabeled antibody enzyme method of immunohistochemistry. Preparation and properties of soluble antigen-antibody complex (horseradish-anti- horseradish peroxidase) and its use in identification of spiro- chetes. E E1gtgghgm Qgt2ghgm 18:315-333. 1970. Perentes E, Rubinstein LJ: Immunohistochemical recognition of human neuroepithelial tumors by anti-Leu (HNK-I) monoclonal antibody. A223 Hesrgnarhgl (Ball) 59:227-233. 1986. Achtstatter T, Moll R, Anderson A. Kuhn C, Pitz S, Schwechheimer K, Franke WW: Expression of glial filaments protein (GFP) in nerve sheaths and non-neural cells re-examined using monoclonal antibodies, with special emphasis on the co-expression of GFA and cytokeratins in epithelial cells of human salivary gland and pleomorphic adenomas. E1tfigtggt1gt1gn 31:206-227. 1986. Deck JHN, Eng LF, Bigbee J, Woodcock SM: The role of glial fibrillary acidic protein in the diagnosis of central nervous system tumors. Agtg Nggt22gth21 (Egtl) 42:183-190. 1978. Herpers MJHM, Ramaekers FCS, Alderweireldt J, Moesker O, Sloff J: Co-expression of glial fibrillary acidic protein- and vimentin-type intermediate filaments in human astrocytomas. Agtg Ng2122gth21 (Egtl) 70:333-339. 1986. Herpers MJHM and Budka H: Glial fibrillary acidic protein (GFAP) in oligodendroglial tumors: gliofibrillary oligodendroglioma and transitional oligodendrocytoma as subtypes of oligodendroglioma. AQEQ Ns2£22e£h21 (£211) 64:265-272. 1934- VITA The author was born in Singatoka. Fiji and completed primary and secondary education there. He earned his bachelor's degree in biolo- gical sciences from Osmania University (1971) and a bachelor's degree in veterinary science from Andhra Agricultural University (1976). In 1977, he was employed by the government of Fiji as a veterinary pathologist- /technica1 officer. He earned a master's degree in veterinary pathology from Melbourne University (1981). In 1982. he joined Michigan State University as a resident- /instructor in pathology. After a successful completion and board- certification by the American College of Veterinary Pathologists (1985), the author enrolled as a graduate student in a PhD program. He com- pleted his PhD in neuropathology in 1988. During his appointment at MSU, the author was nominated to Phi Zeta and was the recipient of a 1987 Charles Louis Davis foundation scholarship in veterinary pathology. 170