CA2388343A1 - Methods for inhibiting neurofibromatosis type 1 (nf1) - Google Patents

Abstract

A method is provided to identify inhibitors of any one of Raf, MEK, extracellular signal regulated kinase 1 (ERK1), extracellular signal regulated kinase 2 (ERK2), phosphatidylinositol triphosphate kinase (PI3K), p21 activated kinase 2 (PAK2), p21 activated kinase 3 (PAK3), Rac, CDC42, p21 activated kinase 1 (PAK1), protein kinase B (PKB), Akt, and c-jun N-terminal kinase (JNK), or crosstalk between the PI3-K an the Ras-Raf-MEK-ERK pathway, useful to treat disorders such as neurofibromatosis type 1 or at least one clinical manifestation thereof.

Description

METHODS TO DETECT AGENTS USEFUL TO INHIBIT
NEUROFIBROMATOSIS TYPE 1 (NFl) Statement of Government Rights This invention was made at least in part with a grant from the Government of the United States of America (grant CA 74177-O1 from the National Institutes of Health). The Government may have certain rights in the invention.
Claim of Priority This application claims priority under 35 U.S.C. ~ 119(e) from U.S.
application Serial No. 60/165,218, filed November 12, 1999.
Background of the Invention Neurofibromatosis type 1 (NF1) is one of the most common genetic disorders in man and is caused by a mutation in the gene NFl. NF1 is a panethnic, autosomal dominant disorder with an incidence of 1 in 3000.
Patients with this disorder have multiple clinical sequelae including the development of benign and malignant tumors, particularly in the skin and brain, e.g., tumors in the skin called neurofibromas which are infiltrated with large numbers of mast cells, hyperpigmentated areas of skin, bone disorders, e.g., curvature of the spine (scoliosis), learning disabilities, myeloid malignancies, high blood pressure, and numerous other complications in several organ systems. Nonmalignant complications of NF1 are very common, frequently debilitating, and a cause of decreased productivity and well-being to the affected patients. No treatments have been available for the bone disorders found in NF1. Moreover, there are no specific treatments for the learning disorders associated with NF1, although some patients have been treated with anticonvulsants and others have been treated with medications that modulate hyperreactivity.
And although mast cell stabilizing agents are the only current therapy for the treatment of the growth of neurofibromas, it is controversial whether this treatment offers any efficacy at all. Tumors in NF1 patients, e.g., tumors that impinge on other tissues such as the plexiform fibromas, have only been treated surgically. This therapy has sequelae because of difficulty in removing the tumor, and the surgery is oftentimes associated with morbidity.

Neurofibromin, the protein encoded by NFl, is a tumor suppressor and functions within the cell (at least in part) through its influence of Ras proteins.
Ras proteins cause cell growth in many different cell types. Neurofibromin converts Ras from its active to its inactive form, thereby limiting cell growth, S i.e., it negatively regulates Ras activity in mammalian cells. In particular, it functions as a GTPase activating protein (GAP) for p2lras by accelerating the conversion of active p2l~as-GTP to inactive p21'~-GDP (Viskochil et al., 1990;
Wallace et al., 1990). Cutaneous neurofibromas are slow-growing tumors that are pathognomonic for NF1, and though mast cells have been implicated in the pathogenesis of neurofibroma formation as well as other cutaneous tumors (Hirota et al., 1993; Ryan et al., 1994; Coussens et al., 1999), the cellular and biochemical consequences of alterations of p2lras activity secondary to partial loss of neurofibromin is unknown.
Thus, what is needed is a method to identify agents useful to treat the clinical manifestations of NF 1.

Summary of the Invention The invention provides a method to inhibit, prevent or treat the clinical manifestations of NF1, e.g., in cells that are heterozygous or homozygous for a mutant NFl allele, by administering an agent that restores normal intracellular signaling as well as growth control. Preferred agents inhibit the activity at least one protein selected from the group consisting of Raf, MEK, extracellular signal regulated kinase 1 (ERK1), extracellular signal regulated kinase 2 (ERK2), phosphatidylinositol triphosphate kinase (PI3K), p21 activated kinase 2 (PAK2), p21 activated kinase 3 (PAK3), Rac, CDC42, p21 activated kinase 1 (PAK1), protein kinase B (PKB), Akt, and c-jun N-terminal kinase (JNK). The agents are thus useful to inhibit, prevent or treat clinical manifestations of NF1 including but not limited to neurofibromas, cutaneous melanocyte hyperplasia, lytic bone lesions or other skeletal abnormalities, pruritis, plexiform fibromas, brain tumors, myeloid malignancies, e.g., juvenile myelomonocytic leukemia (JMML), and learning disorders or other disorders associated with alterations in small GTPases.
Nfl +/ mice have increased numbers of peritoneal mast cells compared to wildtype littermates. Bone marrow derived mast cells (BMMCs) cultured from Nfl +/ mice demonstrate increased proliferation and survival relative to wildtype cells in response to stem cell factor (SCF), a known mitogenic and survival factor for mast cells (Ingram et al., 2000). While these studies suggest that alterations in p21'~S activity can alter mast cell fates in vitro and in vivo, little is known about the specific role of p21'~S in controlling distinct mast cell functions. As described hereinbelow, utilizing Nfl +/ mice and primary mast cells to investigate the effect of p21'~ on specific mast cell functions and biochemical pathways, a distinct signaling pathway downstream of p2lras was identified that is directly responsible for alterations of multiple mast cell functions in vitro and in vivo. Specifically, it was found that hyperactivation of the p21'as-PI-3 kinase pathway alters proliferation, survival, chemotaxis and actin cytoskeletal dynamics of Nfl +/ mast cells. In addition, it was found that activation of the classical p2lr~-Raf Mek-ERK pathway is dependent on co-stimulatory signals from PI-3 kinase which directly contributes to the Nfl +l mast cell phenotype, and identifies p21 activated kinase (PAK) as a signaling molecule involved in this crosscascade activation. Thus, the results described herein offer the first physiological evidence that alterations in both signaling through the p21«-PI-3 kinase pathway and crosstalk between PI-3 kinase and the p21'as-Raf Mek-ERK pathway can be linked to a cellular phenotype implicated in the pathogenesis of a human genetic disease.
Thus, the invention provides a therapeutic method for treating the clinical manifestations of NF1 in a mammal. The method comprises administering to a mammal in need to such treatment an effective amount of an agent that directly or indirectly inhibits the activity at least one protein selected from the group consisting of Raf, MEK, extracellular signal regulated kinase 1 (ERK1), extracellular signal regulated kinase 2 (ERK2), phosphatidylinositol triphosphate kinase (PI3K), p21 activated kinase 2 (PAK2), p21 activated kinase 3 (PAK3), Rac, CDC42, p21 activated kinase 1 (PAKl), protein kinase B (PKB), Akt, and c-jun N-terminal kinase (JNK), e.g., so as to inhibit abnormal cellular growth or pruritus due to NF-1. Abnormal cell growth includes but is not limited to abnormal growth characterized by neurofibromas, lytic bone lesions, plexiform fibromas, brain tumor, myeloid malignancy, optic glioma or pheochromocytoma.
Preferred agents for use in the method of the invention include an agent that inhibits Raf, PI-3 kinase, e.g., LY294002, PAK3, PAK2, PAK1, ERK, MEK, e.g., PD98059, or an agent that inhibits crosstalk between the PI3-K and the Ras-Raf MEK-ERK pathway.
Also provided is a method of determining the ability of an agent to inhibit abnormal mammalian cellular growth associated with NF-1. The method comprises measuring, detecting or determining the ability of the agent to block or inhibit stem cell factor (SCF)-induced hyperproliferation and/or survival and/or degranulation of Nfl +/ mammalian mast cells. Preferably, the cells are derived from skin, fetal liver or from bone marrow of a mammal. In one embodiment of the invention, the agent is contacted with a mouse comprising Nfl +/ mast cells.
Further provided is a method of determining the ability of an agent to treat abnormal cellular growth. The method comprises measuring, detecting or determining the ability of the agent to reduce ERK2 activity in Nfl +/
mammalian cells following stimulation with SCF. For example, the ERK1 phosphorylation of an ELK-1 fusion protein is measured. Preferably, the cellular growth which is treated is mast cell growth. Also provided is the use of an agent 5 that alters the Ras-Raf ERK-MEK pathway in a mammal for the preparation of a medicament for the treatment or prevention of at least one clinical manifestation of NF1, e.g., to prevent or treat cutaneous neurofibromas or learning disorders.
Further provided is the use of an agent that directly or indirectly suppresses Ras signalling cascade hyperactivation for the preparation of a medicament for the treatment or prevention of indications associated with abnormal intracellular signalling and mast cell growth in a mammal.
Brief Description of the Figures Figure 1. CFU-GM colony growth expressed as percent of maximal colony formation in Nfl +l+, Nfl +/- and Nfl -/- fetal liver cells over a range of GM-CSF (A) and IL-3 (B) concentrations. Cells from Nfl -/- embryos form more colonies in the presence of low concentrations of GM-CSF.
Figure 2. Dose response curves of progenitor colony growth in response to cytokines. Growth of Scal+ lin-/dim cells from wildtype (Nfl +l+), heterozygous (Nfl +/-) and homozygous (Nfl -/-) mice is shown at the indicated concentrations of mGM-CSF (a), mIL-3 (b), mGM-CSF and 10 ng/ml SCF (c), mIL-3 and 10 ng/ml mSCF (d) or mSCF (e). The data are expressed as percent of maximal numbers of CFU-GM colonies (Zhang et al., 1998).
Figure 3. ERK kinase activity in Nfl -/- and Nfl +/+ primary c-kit+ cells.
C-kit+ bone marrow cells from mice previously reconstituted with Nfl -/- or Nfl +/+ fetal liver cells were isolated, cultured with either MEK inhibitor or vehicle for 1 hour prior to stimulation with a maximum stimulating capacity of SCF and IL3. Cells were lysed S minutes after stimulation with cytokines. The autoradiographs and activities of 1 of 3 representative experiments is shown.
Cytokine and MEK inhibitor treatment of cells is indicated by + or -.
Figure 4. Effect of haploinsufficiency of Nfl on coat color and total numbers of cutaneous and peritoneal mast cells. (A) Coat color pattern of a representative mouse from each of the following genotypes: +l+, +/+, Nfl +l-, +/+, +/+, yy4l/W 1, Nf1 +/-, W 1/W 1, Heterozygosity at Nfl partially corrects the coat color deficiency in mice homozygous for the T~1 allele in a C57BL/6 genetic background. (B) Representative cytospins from peritoneal lavages stained for mast cells from individual mice of the four Nfl and W genotypes.
Peritoneal cells were stained with toluidine blue to quantify the total number of mast cells per peritoneal lavage. A 200 X magnification of a representative mast cell is shown in the insert of the wildtype mouse. The bar of the insert is equivalent to 10 microns. The bar on the far right panel represents 30 microns.
(C) Representative ear biopsies stained for cutaneous mast cells from individual mice of the four Nfl and W genotypes. Specimens were stained with hematoxylin-eosin to assess routine histology and Giemsa to identify mast cells.
Ear biopsies were stained with Fontana-Masson to differentiate melanin-containing cells from mast cells. Cutaneous mast cells (Giemsa-positive, Fontana-Masson-negative) were quantitated in a blinded fashion by counting the distal 5 mm of ears. Dark arrows indicate Giemsa-positive mast cells and open arrows indicate Fontanna-Masson melanin containing cells. The bar on the far right panel represents 35 microns.
Figure 5. Effect of haploinsufficiency of Nfl and W on the survival and proliferation of bone marrow derived mast cells (BMMCs) in response to stem cell factor (SCF). (A) Proliferation of BMMCs from mice of the four Nfl and W
genotypes in response to recombinant murine SCF. Following deprivation of growth factors for 24 hours, 2 x 105 cells/ml were plated in triplicate in 24 well dishes in RPMI containing 1 % glutamine, 10% fetal bovine serum and 100 ng/ml of SCF in a total volume of 1 ml. After 1 and 3 days, viable cells were counted using a hemocytometer, and expressed as the percent of input cells. Asterisk indicates P < 0.05 for comparison of Nfl +/-; W 1/W 1 to +/+;
W I/T~1 and Nfl +/-; +/+ to +/+; +/+ cells by Student's paired t-test. (B) Percent survival of bone marrow derived mast cells (BMMCs) of the four Nfl and W
genotypes. Following deprivation of growth factors for 24 hours, 3 x 105 cells of each genotype were plated in RPMI containing 1% bovine serum albumin and 100 ng/ml of recombinant murine SCF. The number of surviving cells was determined by trypan blue exclusion and expressed as a percentage of input cells. Asterisk indicates P < 0.05 for comparison of Nfl +/-; W 1/T~1 to +/+;
W 1/W4~ and Nfl +/-; +/+ to +/+; +/+ cells by Student's paired t-test.
Figure 6. Analysis of ERK kinase activity from bone marrow derived mast cells stimulated with SCF in the four Nfl and W genotypes. Activation of S ERK kinase was determined by depriving cells of growth factors for 24 hours followed by stimulation with SCF for 5 and 15 minutes. Autoradiography and quantitative densitometry of the phosphorylation of Elkl fusion protein by ERK
kinase from lysates obtained from SCF stimulated BMMCs are shown. Data represents 1 of 3 independent experiments.
Figure 7. Analysis of Akt kinase activity from bone marrow derived mast cells stimulated with SCF in Nfl +/- and Nfl +/+ genotypes. Activation of ERK kinase was determined by depriving cells of growth factors for 24 hours followed by stimulation with SCF for 5 and 15 minutes. Autoradiography and quantitative densitometry of the phosphorylation of histone 2b fusion protein by Akt kinase from lysates obtained from SCF stimulated BMMCs are shown. One of 3 independent experiments with similar results are shown.
Figure 8. Schematic representation of c-kit mediated Ras-Raf ERK and PI-3-kinase signaling pathways and potential cross talk between them.
Figure 9. Effect of LY294002, a specific PI-3 kinase inhibitor, on DNA
synthesis of Nfl +/- and Nfl +/+ BMMCs. Data represents the mean of triplicate cultures in one representative experiment. Six other experiments were conducted with similar results. Asterisk indicates significant reduction in DNA
synthesis after addition of LY294002 to the cultures compared to cultures incubated with vehicle alone.
Figure 10. Effect of LY294002 on ERK2 activity in Nfl +/- and Nfl +/+
BMMCs following stimulation with SCF.
Figure 11. Effect of PD98059, a specific inhibitor of ERK activation, on DNA synthesis of Nfl +/- and Nfl +/+ BMMCs. Asterisk indicates significant reduction in DNA synthesis after addition of PD98059 to the cultures compared to cultures incubated with vehicle alone. A 15% reduction in DNA synthesis was observed in Nfl +/+ cells cultured with PD98059, while a 2.5 fold greater reduction in DNA synthesis was noted in Nfl +/- mast cells cultured with PD98059.
Figure 12. Analysis of p21 Activated Kinase (PAK) activity in Nfl +/
and Nfl +/+ BMMCs stimulated with SCF. Activation of PAK was determined by serum starving cells for 18 hours followed by stimulation with 10 ng/ml SCF
for 30, 60, and 120 seconds. Autoradiography and quantitative densitometry of the phosphorylation of myelin basic protein (MBP) by PAK from lysates obtained from SCF stimulated BMMCs are shown. Nfl +/- cells have a 2-3 fold higher PAK activity at all time points tested.
Figure 13. Analysis of p2lras activity, co-immunoprecipitation of p21'as with p1 lOS and Akt kinase activity ofNfl +l and wildtype BMMCs stimulated with SCF. (A) p21'as activity in Nfl +l and wildtype BMMCs. GTP-bound p21'~S levels were determined by incubating cell lysates with Raf 1 p21'~
binding domain (RBD) agarose beads, fractionating immunoprecipitations by SDS-PAGE and probing with anti-p21'as antibody. Immunoblots, quantitative densitometry for p2lr~-GTP levels, and Western blots for total p2lr~ are shown.
(B) Co-immunoprecipitation of PI-3 kinase p110b with p2lras in BMMCs. Co-immunoprecipitation of p1108 with p2lr~ was measured by immunoprecipitating p 1 l OS from cell lysates, separating immune complexes by SDS-PAGE gels and probing with p21'~ antibody. Western blots for total p1108 are shown. (C) Akt kinase activity in BMMCs. Akt kinase activity in the presence or absence of 100 nM of wortmannin was determined from Akt immunoprecipitates and an in vitro kinase assay was performed using histone 2b as a substrate. Autoradiography and quantitative densitometry of the phosphorylation of histone 2b are shown. Western blots for total Akt are shown.
Data are representative of 4 independent experiments for each assay. Similar inhibition of Akt kinase was observed in cells preincubated in 25 p,M of LY294002, a second PI-3 kinase inhibitor (data not shown).
Figure 14. Effect of heterozygosity of Nfl on mast cell survival, DNA
synthesis, and activation of downstream effectors of c-kit in response to SCF
in the presence or absence of MEK or PI-3 kinase inhibitors. (A) Survival of Nfl +/ and wildtype BMMCs. After deprivation of growth factors for 24 hours, 3 x 105 cells of each genotype were plated in RPMI containing 1% BSA and SCF in the presence of 0, 5, 10, or 25 ~M of LY294002. The number of surviving cells after 24 hours of culture was determined by trypan blue exclusion and expressed as the percentage of input cells. *P <0.41 for Nfl +/+ vs Nfl +/ ; **p<p.01 for Nfl +/+ vs Nfl +/ with LY294002. (B) DNA synthesis of BMMCs in response to SCF in the presence or absence of either PD98059 (MEK inhibitor) or LY294002 (PI-3 kinase inhibitor). After deprivation of growth factors for 24 hours, 1 x 105 cells were plated in triplicate, fully enriched cultures with ng/ml SCF in the presence or absence of either 5 ~M of LY294002 or 50 p,M of PD98059. After 48 hours of culture, 3H thymidine was added for 6 hours prior to cell harvest. Cells were harvested on glass fibers and (3 emission was measured. *P <0.01 for Nfl +/+ vs Nfl +/ ; **p<0.001 for Nfl +/+ vs Nfl +/+
with LY294002 or PD98059; ***P <0.01 for Nfl +/ vs. Nfl +/ with LY294002 or PD98059. (C) JNK activity in BMMCs after stimulation with SCF for 2 or S
minutes in the presence or absence of 100 nM of wortmannin. JNK activity was determined from immunoprecipitates from cell lysates and by in vitro kinase using ELK-1 fusion protein as the substrate. Western blots for total JNK are shown. (D) ERK activity in BMMCs after stimulation with SCF for 2 and 5 minutes in the presence or absence of 100 nM wortmannin. ERK activity was determined from immunoprecipitates from cell lysates and by in vitro kinase assay using an ELK-1 fusion protein as the substrate. For both JNK and ERK
activation, autoradiography and quantitative densitometry of the phosphorylation of ELK-1 are shown. Western blots for total ERK are shown. Data are representative of 5 independent experiments for each kinase assay with similar results. Similar inhibition of JNK and ERK kinase was detected in the presence of 25 p,M of LY294002 (data not shown).
Figure 15. Analysis of MEK and Raf 1 kinase activity and phosphorylation of MEK 1/2 serine residues 217/221 in Nfl +/ and wildtype BMMCs following stimulation with SCF in the presence or absence of a PI-3K
inhibitor. (A) MEK (a) and (B) Raf 1 activity in Nfl +/ and wildtype mast cells following stimulation with SCF for 2 and 5 minutes in the presence or absence of 100 nM wortmannin. Raf l and MEK kinase activities were determined from immunoprecipitates from cell lysates and in vitro kinase assays using GST-MEK
as the substrate for Raf 1 and GST-ERK p42 fusion protein as the substrate for MEK. For both kinase assays, autoradiography and quantitative densitometry of the phosphorylated substrates are shown. Western blots for total Raf 1 and 5 MEK are shown. (C) Phosphorylation of MEk Ser217/221 in BMMCs following stimulation with 10 ng/ml SCF in the presence or absence of 100 nM
wortmannin. MEK was immunoprecipitated from cell lysates and immune complexes were fractionated by SDS-PAGE. Immunoblots were probed with anti-phospho-MEK Ser217/221 antibody. The phosphorylation levels were 10 determined by densitometric scanning of the immunoblots. Western blots for total MEK are shown. Data are representative of 5 independent experiments for each assay.
Figure 16. Effect of heterozygosity of Nfl on PAK, Cdc42 and Rac activity and PAK phosphorylation sites on Raf 1 and MEK in BMMCs following stimulation with SCF. (A) PAK activity in Nfl +/ and wildtype mast cells following stimulation with SCF for 2 minutes. PAK activity was determined in PAK immunoprecipitates from cell lysates by in vitro kinase assay using myelin basic protein as the substrate. Western blots for total PAK are shown. Autoradiography and quantitative densitometry of the phosphorylated substrate are shown. (B) Phosphorylation of Raf 1 Ser338 and MEK Ser298 in BMMCs following stimulation with 10 ng/ml SCF in the presence or absence of 100 nM of wortmannin. MEK and Raf 1 were immunoprecipitated from cell lysates with specific polyclonal antibodies. Immune complexes were then fractionated by SDS-PAGE, and immunoblots were probed with either anti-phospho-Raf 1 Ser338 or anti-phospho-MEK Ser 298, antibodies respectively.
Phosphorylation levels were determined by densitometric scanning of the immunoblots. Western blots for total Raf 1 and MEK are shown. (C) Cdc42 and (D) Rac activity in Nfl +/ and wildtype BMMCs. GTP-bound Cdc42 or Rac levels were determined by incubating cell lysates with PAK binding domain agarose beads, fractionation by SDS-PAGE and probing with either anti-Cdc42 or anti-Rac antibodies. Immunoblots and quantitative densitometry for the respective Cdc42-GTP and Rac-GTP levels, and Westerns of total Rac and Cdc42, are shown. Data are representative of 3 independent experiments for each assay.
Figure 17. Effect of heterozygosity of Nfl on F-actin content, phosphorylation of cofilin and chemotaxis of BMMCs in response to SCF. (A) F-actin content in BMMCs. Mast cells were stimulated with 10 ng/ml SCF and fixed, at the time points indicated, by addition of formaldehyde. Wildtype and Nfl +/ cells were examined in triplicate. Results are expressed as mean channel fluorescence (MCF) and represent the mean MCF of 4 independent experiments per genotype. *P<0.001 for all comparisons between genotypes using a Student's unpaired t-test. (B) Fluorescent photomicrograph of F-actin content in SCF stimulated BMMCs. Mast cells were stimulated with SCF or vehicle for 1 minute prior to fixation in paraformaldehyde. Cells were then permeabilized and stained with rhodamine phalloidin for F-actin. A representative photomicrograph showing increased phalloidin staining for F-actin in unstimulated and stimulated Nfl +/ mast cells compared with wildtype cells is presented. White arrows indicate phalloidin staining of F-actin. (C) Phosphorylation of cofilin in BMMCs following stimulation with SCF. Whole cell lysates from stimulated BMMCs were separated by SDS-PAGE and immunoblots were probed with anti-phospho-cofilin antibody. Phosphorylation levels were determined by densitometric scanning of the immunoblots. Data are representative of 4 independent experiments. (D) Migration of BMMCs across 8 pM pore polycarbonate filter in response to a gradient of SCF or vehicle in the lower well. 2 x 105 cells were loaded per upper well and incubated at 37°C for 3 hours. After incubation, the total number of migrated cells were counted in the lower chamber. Data represent the mean number of total migrated cells +/- SEM
of 9 independent experiments. *P<0.001 for comparison of Nfl +/+ vs Nfl +l using a Student's paired t-test. (E) Sequential exposures of Nfl +/ and wildtype BMMCs taken at 2.5 minute intervals during exposure to a 10 ng/ml SCF
gradient (from left to right) for 1 hour. BMMCs were allowed to adhere to fibronectin-coated glass coverslips for 15 minutes before being loaded into the chemotactic gradient of a Zigmond chamber. Nfl +/ BMMCs demonstrated greatly enhanced movement to the SCF gradient compared with wildtype cells.

(F) Percent ofNfl +l or wildtype cells moving at various speeds during videomicroscopy. The average speed of directed movement was calculated from data collected from the second 30 minute interval of videomicroscopy from three pairs of samples (n >100 cells analyzed per genotype).
Figure 18. Effect of heterozygosity of Nfl on accumulation of cutaneous mast cells in response to local administration of SCF in vivo. Representative skin biopsy stained for cutaneous mast cells taken from the site of SCF
administration via a micro-osmotic pump on the midorsum of (A) wildtype and (B) Nfl +/ mice. Specimens were stained with hematoxylin-eosin to assess routine histology, and with Giemsa to identify mast cells. Cutaneous mast cells were quantitated in a blinded fashion by counting 2 mm2 sections. A higher magnification of areas outlined in black boxes of the skin biopsy are shown to illustrate the Giemsa positive mast cells. Black arrows indicate Giemsa positive mast cells. C) Number of cutaneous mast cells per mm2 in Nfl +/ and wildtype mice in response to local administration of various concentrations of SCF.
Osmotic pumps loaded with varying concentrations of SCF or PBS were placed subcutaneously in the dorsal skin and removed 7 days later. Specimens were processed for quantification of mast cells as described in a-b. *P <0.001 for Nfl +/+ vs Nfl +l .
Figure 19. Schematic representation of the biochemical pathways operative in wildtype and Nfl +/ mast cells in response to SCF.
Detailed Description of the Invention Mutations of NFI cause NF1, a common autosomal dominant disorder (Xu et al., 1990; DeClue et al., 1992; Menon et al., 1990; Legius et al., 1993;
Skuse et al., 1989). Affected individuals are predisposed to the development of a wide array of malignant and nonmalignant clinical manifestations. Malignant manifestations occur in up to 3-4% of NF1 patients and include optic gliomas, neurofibrosarcomas, solid tumors and a predisposition to develop childhood myeloid leukemias. Consistent with Knudson's "two hit" model of tumor suppressor genes (Knudson, 1985), the leukemia and malignant solid tumors are frequently acquired after a somatic mutation and loss of function of the normal NFI allele. Similarly, approximately 10% of heterozygous Nfl +/- mice spontaneously develop a JMML-like myeloproliferative disorder (MPD) during the second year of life with loss of the wildtype Nfl allele (Bourne et al., 1991).
In addition, although homozygous Nfl knockout mice (Nfl -/-) die in utero around day E13.5 from complex cardiovascular defects (Stokoe et al., 1994;
S Bourne et al., 1991), adoptive transfer of E13.5 Nfl -/- fetal liver hematopoietic stem cells into irradiated syngeneic recipients consistently induces the JMML-MPD (Largaespada et al., 1996; Zhang et al., 1998).
Though the genetic analysis of the malignant complications of NF 1 are consistent with the Knudson paradigm for classifying tumor suppressor genes, the cellular and biochemical pathobiology of the more prevalent heterozygous, non-malignant complications of neurofibromatosis have not been understood.
Common heterozygous stigmata include cutaneous melanocyte hyperplasia (> 99% of patients), dermal neurofibromas (> 99% of adult patients), learning disabilities (30-40% of patients), skeletal abnormalities, particularly short stature and lytic bone lesions (30%), and pruritus (30-40%) (Riccardi, 1992).
Neurofibromas are composed of Schwann cells, endothelial cells, fibroblasts, and mast cells (reviewed, Riccardi, 1992). Though numerous investigators have examined different aspects of neurofibroma biology (Stark et al., 1995; Sawada et al., 1996; Colman et al., 1995; Badache et al., 1998;
Hirota et al., 1993; Rosenbaum et al., 1995), a comprehensive understanding of the genetic, biochemical and cellular mechanisms leading to the formation of neurofibromas has not been clear. Some neurofibromas have a loss of heterozygosity of the normal NF1 allele (Colman et al., 1995) consistent with the "two hit" tumor suppressor model, however, the majority of these lesions retain the normal NFI allele (Sawada et al., 1996; Colman et al., 1995; Stark et al., 1995). Many features of neurofibroma biology including the very large numbers of lesions found in some patients, their self limited growth, and the low propensity of these tumors to undergo malignant degeneration are more consistent with a dosage effect on cell growth. One possible explanation for these conflicting genetic analyses is that neurofibroma samples are comprised of mixed populations of cells where only a small fraction of cells have loss of heterozygosity (LOH). Given the possibility that heterozygous effects of NFI

on cellular physiology are important in the pathogenesis of neurofibroma formation, a detailed study of the biological effects of heterozygous NFI
inactivation in specific cells found in neurofibromas may provide insight into the pathobiology of these prevalent lesions. Furthermore, specific cellular interactions between different cell types found in neurofibromas could lead to LOH in some cell lineages consistent with prior genetic analyses in neurofibromas demonstrating LOH. Therefore, detailed characterization of the genetic, cellular and biochemical contributions of specific cell types to the formation of these tumors is necessary to understand the pathogenesis of these tumors and possibly the cellular mechanism leading to LOH in a subpopulation of cells.
Mast cells have been implicated in the pathogenesis of neurofibromas and other skin malignancies in both human and marine systems (Hirota et al., 1993; Galli et al., 1993; Ryan et al., 1994; Coussens et al., 1999; Galli 1993;
Kerdel et al., 1987). Mast cells release mediators of inflammation including histamine, serotonin, proteoglycans, and leukotrienes subsequent to crosslinking of the high affinity IgE receptor (FcsRI) and the c-kit receptor (Galli et al., 1993;
Serve et al., 1995). In addition, mast cells release inflammatory mediators secondary to integrin attachment to the extracellular matrix (Serve et al., 1995).
Since FceRI, integrins and c-kit signal through Ras proteins, aberrations in the release of mast cell molecules could exist where Ras in hyperactivated. Thus, NF1 mast cells could release abnormal levels of mediators that act locally upon Schwann cells, endothelial cells and fibroblasts to influence their biologic functions (Skuse et al., 1989). Interestingly, in at least two independent clinical series, a significant number of neurofibromas in NF 1 patients were associated with pruritus at the site of development (Huson et al., 1994). Furthermore, treatment with mast cell stabilizers is associated with a reduction in pruritus in some patients with NF 1, and represents the only known medical treatment that alters the growth of neurofibromas (Huson et al., 1994).
Children with NF1 are at markedly increased risk of developing malignant myeloid disorders, particularly JMML, and comprise as many as 10%
of de novo cases of preleukemia in the pediatric age group (Johnson et al., 1993;

Shannon et al., 1992; Neubauer et al., 1991; Miles et al., 1995; Brodeur 1994;
Shannon et al., 1994). These disorders are characterized by onset in infancy and early childhood, dysregulated myeloid proliferation with a peripheral leukocytosis, hepatosplenomegaly, absence of the Philadelphia chromosome, S and poor prognosis. A hallmark of low density blood and bone marrow cells from children with JMML is that there is a selective hypersensitivity of myeloid progenitors (colony-forming unit-granulocyte macrophage, CFU-GM) to the growth factor of GM-CSF (Emanuel et al., 1991; Gualtiera et al., 1989).
Therefore, when bone marrow cells of children with JMML are cultured in vitro 10 for growth of CFU-GM, the dose response curve of colony growth in response to GM-CSF concentrations is shifted to the left in JMML as compared to normal bone marrow cells or progenitors derived from patients with adult CML, while patients in all groups show a similar sensitivity to interleukin-3 (IL-3).
Neurofibromin, the protein encoded by Nfl negatively regulates Ras 15 activity in NFl -/- murine myeloid hematopoietic cells in vitro through the c-kit receptor tyrosine kinase (dominant white spotting locus, T~. SCF, and its receptor c-kit, are components of a signaling pathway that is essential for murine hematopoiesis, melanogenesis and gametogenesis. These proteins are encoded by the Dominant White Spotting (V~ and Steel (5I) loci, respectively, and ligand binding to c-kit activates Ras in myeloid lineage cells. The c-kit/SCF
signaling pathway influences the development of cell lineages such as mast cells and melanocytes, which are implicated in the common, nonmalignant sequelae of NF 1.
Ras proteins regulate the growth and differentiation of many cell types by acting as molecular switches that transduce signals from the extracellular environment to the nucleus (Marshall 1995; Satoh et al., 1991; Hall 1992;
Bourne et al., 1990; Hill et al., 1995; Satoh et al., 1992; Stokoe et al., 1994; Hall 1990; Bourne et al., 1991). The biochemical output of Ras proteins is tightly regulated by their ability to cycle between a guanosine triphosphate bound state (Ras-GTP), and an inactive guanosine diphosphate bound state (Ras-GDP) (Hall 1992; Bourne et al., 1990; Bourne et al., 1991). Ras activation is an essential component of proliferative responses induced after receptor binding by a variety of growth factors including interleukin-3 (IL-3), granulocyte macrophage colony stimulating factor (GM-CSF), and SCF. Stimulation of receptors for these growth factors induces an increase in the percentage of Ras-GTP in the target .
cell (Marshall 1995; Satoh et al., 1991; Hill et al., 1995; Satoh et al., 1992).
S Ras-GTP recruits the serine-threonine kinase Raf 1 to the plasma membrane and Raf l, in turn, activates a series of downstream effectors such as ERKs (p42/p44 MAPkinase) (Sokoe et al., 1994; Leevers et al., 1994) that are important in the mitogenic response to extracellular stimuli. In addition, Ras-GTP has recently been shown to activate other downstream effectors, including phosphatidylinositol triphosphate kinase (PI3K) which influences multiple cell functions including survival. GTPase activating proteins (GAPS) regulate Ras output by stimulating the slow intrinsic Ras GTPase (Satoh et al., 1992;
Rodenhuis 1992; Miyauchi et al., 1994). Because Ras-GTP transduces signals to multiple downstream effectors, GAPS act (at least in part) as important negative regulators of Ras function. Two GAPs, p120 GAP and neurofibromin (the protein encoded by the NFI gene) regulate Ras output in mammalian cells by promoting the conversion ofRas-GTP to Ras GDP (Satoh et al., 1991).

I. Identification of Therapeutic Agents Falling within the Scone of the Invention Agents useful in the practice of the invention include agents that inhibit, reduce or prevent one or more of the clinical manifestations of NF1, e.g., neurofibromas, cutaneous melanocyte hyperplasia, lytic bone lesions or other skeletal abnormalities, pruritis, plexiform fibromas, brain tumors, myeloid malignancies, e.g., juvenile myelomonocytic leukemia (JMML), and learning disorders or other disorders associated with alterations in small GTPases.
These agents can be identified by in vitro or in vivo assays, such as the assays described hereinbelow.
1. In vitro BMMC culture BMMCs can be cultured as described by Serve et al., 1995.
Homogeneity of BMMCs can be determined by Giemsa staining. Cells can also be stained with alcian blue and safranin to confirm they are mast cells.
1 S 2. In vitro mast cell survival assay To determine whether an agent inhibits at least one protein selected from the group consisting of Raf, MEK, ERK1, ERK2, PI3K, PAK1, PAK3, Rac, CDC42, PKB, Akt or JNK, a survival assay may be conducted in which, before the addition of a growth factor varying amounts of an agent are mixed in the presence of mast cells previously deprived of growth factors. For example, a range of known concentrations of an agent, e.g., a kinase inhibitor, are contacted with a defined number (e.g., 104 - 106) of marine BMMCs and plated in individual wells of a tissue culture dish in serum-free RPMI containing 1%
glutamine and 100 ng/ml of recombinant marine SCF in a total volume of 1 ml.
Cells are then incubated for 48, 72, and 96 hours in a 37°C, 5% CO2, humidified incubator. Any method suitable for counting cells can be used, for example, counting with a hemocytometer or FACS analysis. The number of surviving cells are determined by trypan blue exclusion. The number of apoptotic cells are calculated using a TIJNEL assay as described in other studies (Haneline et al., 1998).
3. In vitro mast cell proliferation assay Agents may also be screened in a proliferation assay in which mast cells previously deprived of growth factors are mixed with or without growth factors.
Proliferation of mast cells is determined adding tritiated thymidine to the culture and measuring (3 emission. For example, a defined number of cells (e.g., 104-106) are plated in 96 well dishes in 200 ml of RPMI containing 1 % glutamine, 10% fetal bovine serum (Hyclone Laboratories) and 100 ng/ml of rmSCF or no growth factors and incubated in a 37°C, 5% COZ, humidified incubator.
Cells were cultured for 48 hours, and tritiated thymidine (New Life Science Products, Inc.) is added to cultures 6 hours prior to harvest. Cells were harvested on glass fiber filters (Packard, Meriden, CT) and ~i emission was measured.
4. In vitro mast cell chemotaxis assay In addition, agents may be screened in a chemotaxis assay, in which varying amounts of a growth factor are mixed with cells in the presence of a known chemoattractant. For example, murine BMMCs are serum starved for a period of time, e.g., 16-18 hours. A defined number (e.g., 104 - 106) of BMMCs are resuspended in a volume of serum free RPMI containing 1 % glutamine and 0.5% BSA (Sigma). BMMCs are then loaded onto each transwell filter (8-p.m pore filter Transwell, 24-well cluster; Costar, Boston, MA). Filters are then placed in wells containing 600 ~1 of serum free RPMI supplemented with a range of concentrations of a growth factor, e.g., 0, 2.5, 5, and 10 ng/ml of SCF
placed in the upper and lower chambers of the transwell in a formal checkerboard analysis as previously described (Kim et al., 1998; Meininger et al., 1992). After incubation, the upper chamber is removed and the cells in the bottom chamber are counted using a hemocytometer.
5. In vitro mast cell motility assay Mast cell motility assays, in which cellular motility is observed by time lapse imaging of cells exposed to a gradient of growth factor, may also be employed to screen agents. For example, BMMC motility is directly observed by time lapse imaging of cells exposed to a gradient of 10 ng/ml rmSCF in a Dunn chemotaxis chamber (Dunn and Z'icha, Dunn & Jones, 1997; Weber Scientific Ltd., Surrey, England). A defined number (e.g., 104 - 106) of BMMCs are resuspended in a volume of medium, e.g., 101 RPMI, and are applied to fibronectin fragment H-296 covered glass coverslips.
Cells are allowed to adhere for 10 minutes at 37°C. The coverslips are mounted on the Dunn chamber, the inner well of which is filled with RPMI and the outer well is filled with RPMI/rmSCF. The chamber is sealed and mounted on the stage of a Nikon Diaphot 300 inverted microscope equipped with DIC optics.
Chamber temperature is maintained at 37°C using a stage heater (Instec Instruments Inc., Boulder, CO). A stable gradient to form, and images are recorded digitally at 15-second intervals using a 40x LWD objective and a Spot RT cooled CCD camera and analyzed using Metamorph software (Universal Imaging, Brandywine, PA). The centroid coordinates of cells at each timepoint is determined on calibrated images and is used to calculate the speed and direction of migration.
6. In vitro quantification F-actin in Mast Cells F-actin quantitation may also be performed to screen agents of the invention.
Briefly, flow cytometry is used to quantitate the amount of filamentous actin per mast cell. A defined number of mast cells (e.g., 106) are resuspended in 1 ml HBSS in polypropylene tubes and warmed at 37° for 3 minutes prior to addition of growth factor, e.g., SCF, or Garner. After a specified time, cells are fixed by the addition of 10 volumes of PBS containing 4.6% formaldehyde and 0.1%
BSA. Fixed cells are treated with 0-1% Triton X-100 (Sigma) in PBS for 5 minutes at 22°C, washed and then incubated with 160 nm FITC-phallodoin (Sigma) for 20 minutes at 22°C prior to FACS analysis by flow cytometry.
7. In vitro mast cell kinase assays.
Assays measuring the activation of Raf 1, MEK, ERK, PAK and Akt kinases may also be utilized to screen agents of the invention. Mast cells deprived of serum and growth factors are stimulation with a growth factor for varying lengths of time. For example, marine BMMCs are deprived of serum and growth factors for 24 hours. BMMCs are then stimulated by contacting then with an amount of growth factor, e.g., 10 ng/ml rmSCF, for various amounts of time. Cells are washed with PBS containing 1 mM sodium orthovanadate. Cells are then lysed in nonionic lysis buffer (20 mM Tris/HCI, 137 mM NaCI, 1 mM

EGTA, 1% Triton X-100, 10% Glycerol, 1.5 mM MgCl2, and COMPLETE
protease inhibitors (Amersham) as previously described (Bollag et al., 1996a).
Lysate protein concentration is quantitated, e.g., using the BCA assay (Pierce Chemical Co.). A volume of each lysate is subjected to SDS-PAGE, and equal 5 loading of kinases in these assays is confirmed by Western blot. Gels are transferred to PVDF filters. Gels and PVDF filters are dried and subjected to autoradiography.
ERK and PAK kinase immunoprecipitations can be carried out with an anti-ERK-2 (C-14) antibody and an anti-PAK(C-19) antibody (Santa Cruz), 10 respectively. Raf 1 and Mek kinase immunoprecipitations can be carried out with an anti-Raf 1 (C-12) antibody (Santa-Cruz) and an anti-Mek-1 (C-18) antibody (Santa Cruz), respectively. Akt kinase immunoprecipitations can be carried out with an anti-Akt antibody (Cell Signaling Technology).
The kinase immune complex assay can also be performed as previously 15 described (Bollag et al., 1996a). For example, Elk-1 fusion protein (New England Biolabs) can be used as phosphorylation substrates for JNK and ERK
kinase, histone 2B (Roche) can be used as phosphorylation substrate for Akt, myelin basic protein (Sigma) can be used as phosphorylation substrate for PAK
kinase, and recombinant MEK1 fused with GST at the N-terminus (Upstate 20 Biotechnology) can be used as phosphorylation substrate for Raf 1 kinase.
8. In vitro mast cell Ras, Cdc42 and Rac activation assays.
Agents of the invention may also be screened by Ras, Cdc42 and Rac activation assays, in which mast cells are stimulated with a growth factor and activation is determined via a commercially available assay kit. For example, BMMCs are deprived of serum and growth factors for 24 hours and stimulated at various amounts of time with a known concentration of growth factor, e.g., 10 ng/ml rmSCF. Ras, Cdc42 and Rac activation is determined using Ras, Cdc42, and Rac activation assay kits (Upstate Biotechnology) according to the manufacturer's protocol and as previously described in Gille and Downward (1999).

9. In vivo A rapid method to determine whether an agent of the invention inhibits at least one protein selected from the group consisting of Raf, MEK, ERKl, ERK2, PI3K, PAK1, PAK3, Rac, CDC42, PKB, Akt or JNK, is to inject a selected growth factor into the skin of an animal in the presence of absence of an agent of the invention. At some later point in time, animals are sacrificed and the number of mast cells in animals exposed to the agent is compared to the number of mast cells in animals exposed to growth factor alone, e.g., by quantitative immunofluorescence, relative to control animals.
For example, a continuous infusion of various doses of a growth factor, e.g., rmSCF, or vehicle (PBS) may be delivered from micro-osmotic pumps (Alzet) placed under the dorsal back skin of adult Nfl +/ or wildtype mice.
Osmotic pumps may be surgically placed under light avertin anesthesia. Growth factor or vehicle is released over a period of time at a set rate, e.g., 7 days at 0.5 p.l/hr. Osmotic pumps are then surgically removed following sacrifice. To accurately identify cutaneous sections for quantitating changes in mast cell numbers in response to growth factor, the dorsal skin of the animal is stained with a drop of India ink at the point of exit of growth factor from the osmotic pump prior to removal of the pump. Sections of skin marked with India ink are removed, fixed in buffered formalin and processed in paraffin-embedded sections. Specimens are stained with hematoxylin-eosin to assess routine histology and with Giemsa stain to identify mast cells. Cutaneous mast cells can be quantitated in a blinded fashion by counting 2 mm2 sections in proximity to the India ink stain.
II. Dosages. Formulations and Routes of Administration of the Agents of the Invention The therapeutic agents of the invention are preferably administered so as to inhibit, treat or prevent at least one of the clinical manifestations of NF
1. The amount administered will vary depending on various factors including, but not limited to, the agent chosen, whether prevention or treatment is to be achieved, and if the agent is modified for bioavailability and in vivo stability.

Administration of the therapeutic agents in accordance with the present invention may be continuous or intermittent, depending, for example, upon the recipient's physiological condition, whether the purpose of the administration is therapeutic or prophylactic, and other factors known to skilled practitioners.
The administration of the agents of the invention may be essentially continuous over a preselected period of time or may be in a series of spaced doses. Both local and systemic administration is contemplated.
One or more suitable unit dosage forms comprising the therapeutic agents of the invention, which, as discussed below, may optionally be formulated for sustained release, can be administered by a variety of routes including oral, or parenteral, including by rectal, buccal, vaginal and sublingual, transdermal, subcutaneous, intravenous, intramuscular, intraperitoneal, intrathoracic, intrapulmonary and intranasal routes. The formulations may, where appropriate, be conveniently presented in discrete unit dosage forms and may be prepared by any of the methods well known to pharmacy. Such methods may include the step of bringing into association the therapeutic agent with liquid carriers, solid matrices, semi-solid carriers, finely divided solid Garners or combinations thereof, and then, if necessary, introducing or shaping the product into the desired delivery system.
When the therapeutic agents of the invention are prepared for oral administration, they are preferably combined with a pharmaceutically acceptable Garner, diluent or excipient to form a pharmaceutical formulation, or unit dosage form. The total active ingredients in such formulations comprise from 0.1 to 99.9% by weight of the formulation. By "pharmaceutically acceptable" it is meant the carrier, diluent, excipient, and/or salt must be compatible with the other ingredients of the formulation, and not deleterious to the recipient thereof.
The active ingredient for oral administration may be present as a powder or as granules; as a solution, a suspension or an emulsion; or in achievable base such as a synthetic resin for ingestion of the active ingredients from a chewing gum.
The active ingredient may also be presented as a bolus, electuary or paste.

Formulations suitable for vaginal administration may be presented as pessaries, tampons, creams, gels, pastes, douches, lubricants, foams or sprays containing, in addition to the active ingredient, such Garners as are known in the S art to be appropriate. Formulations suitable for rectal administration may be presented as suppositories.
Pharmaceutical formulations containing the therapeutic agents of the invention can be prepared by procedures known in the art using well known and readily available ingredients. For example, the agent can be formulated with common excipients, diluents, or Garners, and formed into tablets, capsules, suspensions, powders, and the like. Examples of excipients, diluents, and Garners that are suitable for such formulations include the following fillers and extenders such as starch, sugars, mannitol, and silicic derivatives; binding agents such as carboxymethyl cellulose, HPMC and other cellulose derivatives, alginates, gelatin, and polyvinyl-pyrrolidone; moisturizing agents such as glycerol; disintegrating agents such as calcium carbonate and sodium bicarbonate; agents for retarding dissolution such as paraffin; resorption accelerators such as quaternary ammonium compounds; surface active agents such as cetyl alcohol, glycerol monostearate; adsorptive carriers such as kaolin and bentonite; and lubricants such as talc, calcium and magnesium stearate, and solid polyethyl glycols.
For example, tablets or caplets containing the agents of the invention can include buffering agents such as calcium carbonate, magnesium oxide and magnesium carbonate. Caplets and tablets can also include inactive ingredients such as cellulose, pregelatinized starch, silicon dioxide, hydroxy propyl methyl cellulose, magnesium stearate, microcrystalline cellulose, starch, talc, titanium dioxide, benzoic acid, citric acid, corn starch, mineral oil, polypropylene glycol, sodium phosphate, and zinc stearate, and the like. Hard or soft gelatin capsules containing an agent of the invention can contain inactive ingredients such as gelatin, microcrystalline cellulose, sodium lauryl sulfate, starch, talc, and titanium dioxide, and the like, as well as liquid vehicles such as polyethylene glycols (PEGS) and vegetable oil. Moreover, enteric coated caplets or tablets of an agent of the invention are designed to resist disintegration in the stomach and dissolve in the more neutral to alkaline environment of the duodenum.
The therapeutic agents of the invention can also be formulated as elixirs or solutions for convenient oral administration or as solutions appropriate for parenteral administration, for instance by intramuscular, subcutaneous or intravenous routes.
The pharmaceutical formulations of the therapeutic agents of the invention can also take the form of an aqueous or anhydrous solution or dispersion, or alternatively the form of an emulsion or suspension.
Thus, the therapeutic agent may be formulated for parenteral administration (e.g., by injection, for example, bolus injection or continuous infusion) and may be presented in unit dose form in ampules, pre-filled syringes, small volume infusion containers or in mufti-dose containers with an added preservative. The active ingredients may take such forms as suspensions, solutions, or emulsions in oily or aqueous vehicles, and may contain formulatory agents such as suspending, stabilizing and/or dispersing agents.
Alternatively, the active ingredients may be in powder form, obtained by aseptic isolation of sterile solid or by lyophilization from solution, for constitution with a suitable vehicle, e.g., sterile, pyrogen-free water, before use.
These formulations can contain pharmaceutically acceptable vehicles and adjuvants which are well known in the prior art. It is possible, for example, to prepare solutions using one or more organic solvents) that is/are acceptable from the physiological standpoint, chosen, in addition to water, from solvents such as, acetone, ethanol, isopropyl alcohol, glycol ethers such as the products sold under the name "Dowanol", polyglycols and polyethylene glycols, C1-C4 alkyl esters of short-chain acids, preferably ethyl or isopropyl lactate, fatty acid triglycerides such as the products marketed under the name "Miglyol", isopropyl myristate, animal, mineral and vegetable oils and polysiloxanes.
The compositions according to the invention can also contain thickening agents such as cellulose and/or cellulose derivatives. They can also contain gums such as xanthan, guar or carbo gum or gum arabic, or alternatively polyethylene glycols, bentones and montmorillonites, and the like.

It is possible to add, if necessary, an adjuvant chosen from antioxidants, surfactants, other preservatives, film-forming, keratolytic or comedolytic agents, perfumes and colorings. Also, other active ingredients may be added, whether for the conditions described or some other condition.
5 For example, among antioxidants, t-butylhydroquinone, butylated hydroxyanisole, butylated hydroxytoluene and a-tocopherol and its derivatives may be mentioned. The galenical forms chiefly conditioned for topical application take the form of creams, milks, gels, dispersion or microemulsions, lotions thickened to a greater or lesser extent, impregnated pads, ointments or 10 sticks, or alternatively the form of aerosol formulations in spray or foam form or alternatively in the form of a cake of soap.
Additionally, the agents are well suited to formulation as sustained release dosage forms and the like. The formulations can be so constituted that they release the active ingredient only or preferably in a particular part of the 15 intestinal or respiratory tract, possibly over a period of time. The coatings, envelopes, and protective matrices may be made, for example, from polymeric substances, such as polylactide-glycolates, liposomes, microemulsions, microparticles, nanoparticles, or waxes. These coatings, envelopes, and protective matrices are useful to coat indwelling devices, e.g., stems, catheters, 20 peritoneal dialysis tubing, and the like.
The therapeutic agents of the invention can be delivered via patches for transdermal administration. See U.S. Patent No. 5,560,922 for examples of patches suitable for transdermal delivery of a therapeutic agent. Patches for transdermal delivery can comprise a backing layer and a polymer matrix which 25 has dispersed or dissolved therein a therapeutic agent, along with one or more skin permeation enhancers. The backing layer can be made of any suitable material which is impermeable to the therapeutic agent. The backing layer serves as a protective cover for the matrix layer and provides also a support function. The backing can be formed so that it is essentially the same size layer as the polymer matrix or it can be of larger dimension so that it can extend beyond the side of the polymer matrix or overlay the side or sides of the polymer matrix and then can extend outwardly in a manner that the surface of the extension of the backing layer can be the base for an adhesive means.
Alternatively, the polymer matrix can contain, or be formulated of, an adhesive polymer, such as polyacrylate or acrylate/vinyl acetate copolymer. For long--term applications it might be desirable to use microporous and/or breathable backing laminates, so hydration or maceration of the skin can be minimized.
Examples of materials suitable for making the backing layer are films of high and low density polyethylene, polypropylene, polyurethane, polyvinylchloride, polyesters such as polyethylene phthalate), metal foils, metal foil laminates of such suitable polymer films, and the like. Preferably, the materials used for the backing layer are laminates of such polymer films with a metal foil such as aluminum foil. In such laminates, a polymer film of the laminate will usually be in contact with the adhesive polymer matrix.
The backing layer can be any appropriate thickness which will provide the desired protective and support functions. A suitable thickness will be from about 10 to about 200 microns.
Generally, those polymers used to form the biologically acceptable adhesive polymer layer are those capable of forming shaped bodies, thin walls or coatings through which therapeutic agents can pass at a controlled rate.
Suitable polymers are biologically and pharmaceutically compatible, nonallergenic and insoluble in and compatible with body fluids or tissues with which the device is contacted. The use of soluble polymers is to be avoided since dissolution or erosion of the matrix by skin moisture would affect the release rate of the therapeutic agents as well as the capability of the dosage unit to remain in place for convenience of removal.
Exemplary materials for fabricating the adhesive polymer layer include polyethylene, polypropylene, polyurethane, ethylene/propylene copolymers, ethylene/ethylacrylate copolymers, ethylene/vinyl acetate copolymers, silicone elastomers, especially the medical-grade polydimethylsiloxanes, neoprene rubber, polyisobutylene, polyacrylates, chlorinated polyethylene, polyvinyl chloride, vinyl chloride-vinyl acetate copolymer, crosslinked polymethacrylate polymers (hydrogel), polyvinylidene chloride, polyethylene terephthalate), butyl rubber, epichlorohydrin rubbers, ethylenvinyl alcohol copolymers, ethylene-vinyloxyethanol copolymers; silicone copolymers, for example, polysiloxane-polycarbonate copolymers, polysiloxane-polyethylene oxide copolymers, polysiloxane-polymethacrylate copolymers, polysiloxane-alkylene copolymers (e.g., polysiloxane-ethylene copolymers), polysiloxane-alkylenesilane copolymers (e.g., polysiloxane=ethylenesilane copolymers), and the like;
cellulose polymers, for example methyl or ethyl cellulose, hydroxy propyl methyl cellulose, and cellulose esters; polycarbonates;
polytetrafluoroethylene;
and the like.
Preferably, a biologically acceptable adhesive polymer matrix should be selected from polymers with glass transition temperatures below room temperature. The polymer may, but need not necessarily, have a degree of crystallinity at room temperature. Cross-linking monomeric units or sites can be incorporated into such polymers. For example, cross-linking monomers can be incorporated into polyacrylate polymers, which provide sites for cross-linking the matrix after dispersing the therapeutic agent into the polymer. Known cross-linking monomers for polyacrylate polymers include polymethacrylic esters of polyols such as butylene diacrylate and dimethacrylate, trimethylol propane trimethacrylate and the like. Other monomers which provide such sites include allyl acrylate, allyl methacrylate, diallyl maleate and the like.
Preferably, a plasticizer and/or humectant is dispersed within the adhesive polymer matrix. Water-soluble polyols are generally suitable for this purpose. Incorporation of a humectant in the formulation allows the dosage unit to absorb moisture on the surface of skin which in turn helps to reduce skin irntation and to prevent the adhesive polymer layer of the delivery system from failing.
Therapeutic agents released from a transdermal delivery system must be capable of penetrating each layer of skin. In order to increase the rate of permeation of a therapeutic agent, a transdermal drug delivery system must be able in particular to increase the permeability of the outermost layer of skin, the stratum corneum, which provides the most resistance to the penetration of molecules. The fabrication of patches for transdermal delivery of therapeutic agents is well known to the art.

For administration to the upper (nasal) or lower respiratory tract by inhalation, the therapeutic agents of the invention are conveniently delivered from an insufflator, nebulizer or a pressurized pack or other convenient means of delivering an aerosol spray. Pressurized packs may comprise a suitable propellant such as dichlorodifluoromethane, trichlorofluoromethane, dichlorotetrafluoroethane, carbon dioxide or other suitable gas. In the case of a pressurized aerosol, the dosage unit may be determined by providing a valve to deliver a metered amount.
Alternatively, for administration by inhalation or insufflation, the composition may take the form of a dry powder, for example, a powder mix of the therapeutic agent and a suitable powder base such as lactose or starch.
The powder composition may be presented in unit dosage form in, for example, capsules or cartridges, or, e.g., gelatine or blister packs from which the powder may be administered with the aid of an inhalator, insufflator or a metered-dose inhaler.
For intra-nasal administration, the therapeutic agent may be administered via nose drops, a liquid spray, such as via a plastic bottle atomizer or metered-dose inhaler. Typical of atomizers are the Mistometer (Wintrop) and the Medihaler (Riker).
The local delivery of the therapeutic agents of the invention can also be by a variety of techniques which administer the agent at or near the site of disease. Examples of site-specific or targeted local delivery techniques are not intended to be limiting but to be illustrative of the techniques available.
Examples include local delivery catheters, such as an infusion or indwelling catheter, e.g., a needle infusion catheter, shunts and stems or other implantable devices, site specific carriers, direct injection, or direct applications.
For topical administration, the therapeutic agents may be formulated as is known in the art for direct application to a target area. Conventional forms for this purpose include wound dressings, coated bandages or other polymer coverings, ointments, creams, lotions, pastes, jellies, sprays, and aerosols, as well as in toothpaste and mouthwash, or by other suitable forms, e.g., via a coated condom. Ointments and creams may, for example, be formulated with an aqueous or oily base with the addition of suitable thickening and/or gelling agents. Lotions may be formulated with an aqueous or oily base and will in general also contain one or more emulsifying agents, stabilizing agents, dispersing agents, suspending agents, thickening agents, or coloring agents.
The S active ingredients can also be delivered via iontophoresis, e.g., as disclosed in U.S. Patent Nos. 4,140,122; 4,383,529; or 4,051,842. The percent by weight of a therapeutic agent of the invention present in a topical formulation will depend on various factors, but generally will be from 0.01 % to 95% of the total weight of the formulation, and typically 0.1-25% by weight.
When desired, the above-described formulations can be adapted to give sustained release of the active ingredient employed, e.g., by combination with certain hydrophilic polymer matrices, e.g., comprising natural gels, synthetic polymer gels or mixtures thereof.
Drops, such as eye drops or nose drops, may be formulated with an 1 S aqueous or non-aqueous base also comprising one or more dispersing agents, solubilizing agents or suspending agents. Liquid sprays are conveniently delivered from pressurized packs. Drops can be delivered via a simple eye dropper-capped bottle, or via a plastic bottle adapted to deliver liquid contents dropwise, via a specially shaped closure.
The therapeutic agent may further be formulated for topical administration in the mouth or throat. For example, the active ingredients may be formulated as a lozenge further comprising a flavored base, usually sucrose and acacia or tragacanth; pastilles comprising the composition in an inert base such as gelatin and glycerin or sucrose and acacia; mouthwashes comprising the composition of the present invention in a suitable liquid carrier; and pastes and gels, e.g., toothpastes or gels, comprising the composition of the invention.
The formulations and compositions described herein may also contain other ingredients such as antimicrobial agents, or preservatives. Furthermore, the active ingredients may also be used in combination with other therapeutic agents, for example, oral contraceptives, bronchodilators, anti-viral agents, steroids and the like.

The invention will be further described by the following non-limiting examples, which employed the following methodologies.
Example 1 5 Hematopoietic Colony Growth in Nfl -/- Fetal Liver Cells Because leukemic cells often contain multiple genetic alterations, the relationship between inactivation of NFI and aberrant growth in response to GM-CSF was uncertain. Nfl +/- mice were used to examine the effects of homozygous disruption of Nfl. Nfl -/- embryos die in utero at d12.5-13.5 p.c.
10 from apparent hydrops fetalis secondary to complex developmental abnormalities of the cardiac system, including overgrowth of the ventricular outflow tracts (Jack et al., 1994). Fetal liver hematopoietic cells were isolated from Nfl -/- embryos prior to their death at day 12.5-13.5 p.c. and myeloid progenitors established in methylcellulose medium containing a range of 15 recombinant murine GM-CSF or murine IL-3 concentrations. Figure 1B shows that cultures of Nfl -/- cells yielded more CFU-GM colonies at low GM-CSF
concentrations than cells from wild-type (Nfl +/+) or heterozygous (Nfl +/-) littermates. In contrast, the dose-response relationship for myeloid colony formation was similar for Nfl +/+, Nfl +/-, and Nfl -/- cells in the presence of 20 IL-3 (Figure 1 A). These experiments show that disruption of Nfl confers an aberrant pattern of hematopoietic progenitor growth that is remarkably similar to that seen in JMML (Figure 1).
Since hematopoiesis is hierarchical and has multiple compartments, it is possible that neurofibromin is only required to regulate growth of specific 25 subpopulations of cells in response to a restricted number of hematopoietic growth factors. Previous studies in human and murine cells examined only the responsiveness of relatively differentiated lineage restricted progenitors from mixed populations of hematopoietic cells, and cultures were established using single growth factors. Therefore, highly purified hematopoietic cells enriched 30 for primitive hematopoietic progenitors were isolated using fluorescence cytometry (Scal+ lin-~d"") and the effects of stimulating Scal+ 1n ~d'm cells with a low concentration of mSCF in combination with either GM-CSF or IL-3 were examined to ascertain if the effects were additive or synergistic.
Similar to mixed populations of cells, the dose-response curve for CFU-GM colony formation from Nfl -/- Scal+ lin ~d"" cells was significantly left-s shifted relative to the growth curve for wild-type cells in cultures stimulated with GM-CSF (Figure 2A), but not IL-3 (Figure 2B). Surprisingly, the addition of SCF to cultures stimulated with IL-3 resulted in a hypersensitive pattern of growth in Nfl -/- cells that was not seen with IL-3 alone (Figure 2D), and a low concentration of SCF in combination with GM-CSF further shifted the hypersensitivity of Nfl -/- cells to the left compared to cells stimulated with GM-CSF alone (Figure 2C). These data indicate that GM-CSF and SCF have a synergistic effect in these cells. Although SCF is relatively inefficient at increasing colony formation in the absence of other growth factors, a significant left shift was detected in the dose-response curve of Nfl -/- versus Nfl ++
cells cultured in the presence of SCF alone (Figure 2E). Taken together, these data suggest that loss of Nfl deregulates the growth of hematopoietic progenitors in response to SCF, which acts on more immature populations of progenitors than IL-3 or GM-CSF, and demonstrates that SCF can synergize with either factor to induce a hypersensitive pattern of in vitro growth in Nfl -/- progenitors.
Based on these in vitro findings, it was hypothesized that loss of neurofibromin in Nfl -/- hematopoietic cells augments signaling through the c-lot receptor tyrosine kinase following SCF stimulation by increasing Ras activity. Stimulation with a variety of growth factors including SCF, GM-CSF
or IL-3 activates the Ras-Raf ERK kinase cascade in cultured cell lines (Hall, 1992; Bourne et al., 1990; Stokoe et al., 1994; Bourne et al., 1991). If neurofibromin is a regulator of Ras activity in response to these cytokines in primary hematopoietic progenitors, then loss of Nfl function might result in an augmented ERK activity following stimulation. To test this possibility, c-kit+
cells were isolated from the bone marrows of mice previously irradiated and reconstituted with fetal liver cells from d13.5 gestation Nfl -/- and Nfl +/+
littermates. It was determined that the Nfl -/- c-kit+ cells isolated from transplant recipients had a similar hyperresponsiveness to GM-CSF, SCF, or SCF and IL-3 in clonogenic assays as Nfl -/- fetal liver cells (data not shown).
Following isolation using fluorescence-activated cell sorting, c-kit+ cells were tested for activation of ERKs following growth factor stimulation. Cells were maintained in liquid cultures overnight without exogenous growth factors to induce quiescence. In addition, aliquots of cells were incubated with either a MEK inhibitor, which inhibits ERK activation, or vehicle alone to confirm that ERK activation results from activation of the Ras-MEK-ERK pathway. Cells were then stimulated with SCF and IL-3 for 5 or 60 minutes. Three independent experiments yielded similar results. Data from one of these is shown in Figure 3.
Interestingly, an elevation in ERK activity was observed in unstimulated c-kit+ cells isolated from the Nfl -/- recipients. In addition, this kinase activity increased significantly in Nfl -/- cells after exposure to SCF + IL-3 and persisted above baseline at S minutes. In contrast, c-kit+ cells isolated from Nfl +l+
recipients had lower baseline activity and a much less pronounced increase in ERK activity 5 minutes after stimulation compared to Nfl -/- cells. A high concentration of a MEK inhibitor, which in turn blocks activation of ERKs, inhibited ERK activity in both Nfl -/- and +/+ cells. Thus, the absence of neurofibromin in c-kit+ hematopoietic progenitors results in a constitutive activation of Ras-Raf MEK-ERK signaling pathway which can be further stimulated in response to SCF and IL-3. In subsequent kinase assays, Nfl -l- c-kit+ cells had elevated and prolonged activation of ERK in response to SCF
alone.
Example 2 Generation of Mice with c-kit Receptor Tyrosine Kinase (W Locus) And Nfl Mutations Because of the in vitro observations in Nfl -/- myeloid progenitor cells following stimulation with SCF and the involvement of myeloid, mast cell, and melanocyte lineages in some of the pathologic complications of NF1, mice were generated with mutations at both W and Nfl to examine potential genetic and biochemical interactions.in vivo and in vitro, and to understand the biological effect of heterozygous inactivation of Nfl in these specific cell lineages that are involved in the pathogenesis of NF 1. C-kit is known to be important in skin pigmentation and mast cell biology, and mast cells and skin melanocytes are implicated in the nonmalignant complications of NF 1.
W mice display varying degrees of sterility, mast cell hypoplasia, anemia, and coat color deficiency correlating with the residual kinase activity of the mutant receptor (Blouin et al., 1993; Nocka et al., 1990). Mice homozygous for a point mutation in the cytoplasmic domain of c-kit (YI~1) have a partial inactivation of the kinase activity of the c-kit receptor resulting in an abnormal mottled, white coat color (Nocka et al., 1990). When YY4~ and Nfl mice, both of which had been backcrossed onto a C57BL/6 background for greater than 10 generations, were crossed, a very surprising result was observed. Mice heterozygous at Nfl and homozygous for the W ~ mutation (Nfl +/-; y~l/~1) displayed a 60-70% restoration of coat color (Figure 4A). This finding was consistent in > 150 F2 progeny from multiple founders carrying mutations at 1 S both loci. Thus, heterozygosity at Nfl partially corrects the aberrant pigmentation pattern of YY4~lW 1 mice.
Mice homozygous for different mutant W alleles have reduced numbers of peritoneal and cutaneous mast cells correlating with the residual kinase activity of the mutant receptor (Blouin et al., 1993; Paulson et al., 1996).
To investigate whether heterozygous inactivation of Nfl could modulate the deficiency of mast cells in T~1 mice, numbers of peritoneal mast cells harvested from Nfl +/-; y~l/W 1 animals were compared to those taken from singly mutant mice (+/+; y~l/W 1). Peritoneal mast cell numbers in Nfl +/-; +/+ and wildtype mice were also compared. Cells isolated by peritoneal lavage were stained with toluidine blue to identify mast cells. Representative cytospins from individual mice from each of the four Nfl and W genotypes are shown in Figure 4B.
Peritoneal mast cell numbers in Nfl +/-; W ~/y~'I mice were 40-fold higher than in +/+; W 1/W 1 littermates (Table 1). Results represent mean numbers of mast cells ~ standard error of the mean (s.e.m.) from 6 animals in each genotype.
Importantly, in addition to restoring peritoneal mast cell numbers to about 10%
ofwildtype levels in W41/W 1 mice, heterozygous inactivation ofNfl significantly increased mast cell numbers in animals that were wildtype at the W
locus (Table 1).
Similar experiments were performed to determine if heterozygosity at Nfl increased numbers of cutaneous mast cells in the mutant YY41/T~' and wildtype backgrounds. Giemsa-stained ear biopsies from mice of the four W and Nfl genotypes are shown in Figure 4C. Ear biopsies from Nfl +/-; W 1/T~~ mice showed a 2-fold increase in numbers of cutaneous mast cells compared to +/+;
Yl~l/YY41 mice (Table 1). Nfl +/-; +/+ mice also had a modest, though statistically insignificant, increase in cutaneous mast cell numbers compared to wildtype mice (Table 1). Giemsa-stained biopsies obtained from a second site (dorsal skin) revealed similar differences in cutaneous mast cell numbers between the genotypes (data not shown). The increased numbers of peritoneal and cutaneous 'mast cells in Nfl +/- mice provides additional evidence that haploinsufficiency at Nfl augments signaling through the c-kit receptor in vivo.
Table 1. Effect of W and Nfl Genotypes on Mast Cell Numbers and Mast Cell Colony Growth Genotype No. of Peritoneal Mast Cells (x 102) 780 ~ 10 1000 ~ 14* 2 ~ 1 81 ~ 17*
No. of Cutaneous Mast Cells (mmz) 25.6 ~ 7 31 ~ 2 6 ~ 1 14 ~ 2*
No. of Mast Cell Colonies (x 102) 79 ~ 6 117 ~ 9* 0 8 ~ 0.8 Asterisk indicates P < 0.05 for comparison of Nfl +/+ and Nfl +/- in both W4~/W 1 and wildtype genetic backgrounds by Student's paired t-test.
Example 3 Evaluation of Proliferation in Nfl +/- Heterozygous Mast Cells C-kit receptor activation in BMMCs mediates diverse biological responses including proliferation. To assess whether heterozygosity at Nfl augments the proliferative responses of a homogenous population of mast cells to SCF, the growth kinetics of Nfl +/- BMMCs were compared with Nfl +/+
cells in mice with a normal or mutant (W4~) c-kit receptor. BMMCs from mice of each genotype were cultured in 10% FCS without added growth factors for 24 hours prior to the addition of SCF. Cell numbers were determined at the time SCF was added (day 0) and after 24 and 72 hours in culture. Nfl +/- cells of both W genotypes showed greater proliferative responses to SCF after 24 and 72 hours in culture than the corresponding Nfl +/+ populations (Figure SA).
5 Utilizing the same experimental design, thymidine incorporation assays gave similar results (data not shown). In addition, a total of 9 independently derived heterozygous and wildtype lines were evaluated with similar results. Thus, haploinsufficiency at Nfl is associated with an increased proliferative response of wildtype and W 1/ W 1 BMMCs in liquid cultures containing 10% FCS and 10 exogenous SCF.
Mast cells derived from different lines of W mutant mice display reduced or no survival when cultured in the presence of SCF alone, which correlates with residual receptor tyrosine kinase activity (Nocka et al., 1990; Nakahata et al., 1982). Therefore, the effect of heterozygous inactivation of Nfl on the survival 15 of BMMCs isolated from wildtype and YY~1/W 1 mice was examined (Figure SB).
Nfl +/- BMMCs of both W genotypes demonstrated increased survival compared to Nfl +/+ cells after 48 hours of culture in serum free media in response to exogenous Steel factor. Since this assay has been shown to correlate with c-kit receptor tyrosine kinase activity (Nocka et al., 1990), surface expression of c-kit 20 was quantitated for the four W and Nfl genotypes by fluorescence cytometry to ensure that these differences in cell survival were not explained by variable levels of receptor expression. No differences were observed (data not shown).
These data suggest that heterozygosity of Nfl enhances the survival of mast cells in vitro.
25 The Ras-Raf ERK signaling pathway is an important downstream target of c-kit receptor activation, and neurofibromin negatively regulates Ras signaling by functioning as a GAP for Ras (Chabot et al., 1996; Sawada et al., 1996). The phenotypic data presented above suggest a model whereby heterozygous inactivation of Nfl enhances c-kit induced Ras activity by reducing 30 neurofibromin levels (and GAP activity for Ras) in susceptible cell lineages. To test this hypothesis, primary BMMCs were stimulated with SCF and ERK
activities measured (Riccardi 1992; Hemesath et al., 1998). Nfl +/-; W 1/W I

BMMCs demonstrated a 5 fold greater increase in ERK activity from baseline 5 minutes after the addition of SCF relative to +/+; W 1/W 1 BMMCs (Figure 6).
Indeed, heterozygosity of Nfl restored the ability of the mutant W ~ c-kit receptor to activate ERK to wildtype levels at the 5 minute time point (Figure 6).
Similarly, Nfl +/-; +/+ BMMCs had a 2-fold greater increase in ERK activity from baseline in comparison to wildtype mast cells that was sustained at both tested time points (Figure 6). These biochemical data indicate that the phenotypic effects in Nfl +/- mast cells correlates with enhanced signaling through a major downstream effector of Ras-GTP.
Example 4 Mice with Mutations at N 1 and W
There is an increasing body of evidence that mast cells, fibroblasts, endothelial cells, and neuronal cells form a homeostatic regulatory unit in the 1 S skin as well as other organs (Galli et al., 1993; Coussens et al., 1999;
Bienenstock et al., 1991; Bienenstock et al., 1988; Iemura et al., 1994;
Lorenz et al., 1996). Specifically, numerous authors have suggested that mast cells and sensory neurons interact in the axon reflex to generate the flare reaction to noxious stimuli in skin (Galli 1993; Mecheri et al., 1997; Costa et al., 1996).
Mast cell/nerve cell interactions have been well characterized morphogenically in a number of disease states affecting the skin in both human diseases and in marine models (Coussens et al., 1999; Galli 1993; Kerdel et al., 1987;
Bienenstock et al., 1988; Mecheri et al., 1997). In a recent study in a marine model of squamous epithelial cell carcinoma, exclusion of mast cells from the experimental model by utilizing a genetic intercross with mast cell deficient mice ( W mice) prevented the development of the progressive anaplastic changes in the skin leading to malignancy (Coussens et al., 1999). These studies emphasize the important role of mast cells in locally influencing the growth and differentiation of multiple cell type.
In preliminary studies in mice expressing either a normal or mutant (T~
c-kit receptor, Nfl +/- mice were shown to have increased basal numbers of peritoneal mast cells in vivo compared to Nfl +/+ littermates. In addition, NFI

+/- mast cells had increased proliferation and enhanced survival in response to SCF compared to wildtype mast cells in vitro. Though Nfl heterozygous mice on a mutant W background have an increased number of cutaneous mast cells, unstressed, singly mutant, Nfl +/- mice have only a modest increase in cutaneous mast cells, analogous to NF1 patients. Interestingly, investigators have shown that mast cells increase dramatically in neoplastic lesions that involve Schwann cells, such as neurofibromas and Schwannomas (Hirota et al., 1993; Ryan et al., 1994), and mast cells cluster around normal peripheral nerves, particularly during times of nerve injury and repair (Iemura et al., 1994). Recently, Kitamura et al. have demonstrated that the amount of SCF mRNA was greater in neurofibroma tissue, suggesting that SCF and the c-kit receptor are associated with the increase in mast cells found in neurofibroma tissues (Hirota et al., 1993). Schwann cells and fibroblasts, two principal components of neurofibromas, secrete SCF in response to many different stimuli (Iemura et al., 1 S 1994). While the role of the c-kit/SCF signaling pathway in mast cells is not understood in neurofibroma formation, it is interesting that mast cell stabilizing drugs have been preliminarily shown to decrease the size and pruritus associated with neurofibromas (Huson et al., 1994). This increase in the number of mast cells in vivo could be the result of increased proliferation, survival or a combination of both proliferation and survival in NF1 patients.
In addition to these data, there are clinical data in patients with other diseases that infer the importance of c-kit signaling and mast cell proliferation/survival in cutaneous pathologic processes in two other disease states. First, there are recently described mutations in human patients who have a gain of function mutation in the c-kit receptor. These patients with point mutations in c-kit have acquired cutaneous mastocytosis leading to hyperplastic and neoplastic cutaneous lesions with infiltrations of large numbers of mast cells (Kettelhut et al., 1987). Secondly, there are rare reports of patients that have both neurofibromatosis and piebaldism (Chang et al., 1993; Tay 1998).
Piebaldism is an autosomal dominant genetic disorder that occurs secondary to a mutation in the c-kit receptor and results in a decrease in c-kit receptor tyrosine kinase activity. The most distinctive clinical features of patients with piebaldism is cutaneous hypopigmentation, though some patients also have learning disorders and hypoplastic anemia. Interestingly, a striking feature of rare patients diagnosed with both piebaldism and neurofibromatosis type 1 was that none of the patients had neurofibromas.
In both the marine and human systems, apoptosis is a common mechanism for reducing the size of mast cell populations in vivo (Iemura et al., 1994). Cells undergoing apoptosis are efficiently recognized by macrophages and are cleared from the tissue without the significant release of inflammatory mediators (Iemura et al., 1994). In other mast cell studies in marine systems, activation of c-kit has resulted in enhanced cell survival by inhibiting apoptosis (Yee et al., 1994). Biochemically, increased mast cell survival and proliferation are associated with c-kit activation of the PI3K/Akt pathway (Serve et al., 1995;
Vosseller et al., 1997). The in vitro observations in Nfl +/- mast cells indicating greater basal and cytokine activated levels of Akt, compared to wildtype cells, 1 S suggest that Nfl +/- cells will have both enhanced survival and proliferation in vivo. Since the PI3K/Akt pathway has been implicated in controlling degranulation, Nfl +/- mast cells may also release abnormal amounts of inflammatory mediators influencing the local growth of fibroblasts, Schwann cells and endothelial cells in vivo.
Because of the known involvement of mast cells in the pathogenesis of other systemic and dermatologic diseases as well as the potential contribution of mast cells to the pruritus, dermatitis and neurofibromas common in NF 1 patients, a series of in vitro and in vivo experiments were conducted to evaluate the proliferation, survival, and degranulation of mast cells using the Nfl +/-marine model.
PI-3 kinase (PI3K) is a major downstream effector of c-kit activation which has an important role in both proliferation and survival of wildtype mast cells. In addition, Ras-GTP has been shown to activate PI3K by numerous investigators. Since in vitro studies demonstrated that Nfl +/- cells have both increased proliferation and survival compared to wildtype cells, Akt activity, a downstream effector of PI3K, was measured in Nfl +/- and wildtype mast cells, as an indication of PI3K activity. To test the hypothesis that Akt activity is elevated in Nfl +/- mast cells in response to SCF, primary Nfl +/+ and Nfl +/-BMMCs were stimulated with SCF and basal and cytokine activated Akt activity measured. These data are shown in Figure 7. The data indicate that Nfl heterozygous mast cells demonstrate a 2-fold increase in Akt activity in comparison to wildtype mast cells at both time points tested. Surprisingly, Nfl +/- mast cells also had at 3-fold higher intrinsic basal Akt activity compared to wildtype cells. Collectively, these studies indicate the feasibility of utilizing this genetic intercross to better understand the biochemical and cellular biology of Nfl in modulating susceptible cell lineages that are implicated in the pathobiology of NF1 in a gene dosage-dependent fashion.
The c-kit/SCF signaling pathway is known to be essential to the regulation of hematopoiesis (including mast cell development), melanogenesis and gametogenesis. A large number of independent experiments in murine models of either receptor deficiency (YT~ or ligand deficiency (St7 have shown that studies of mast cell development and function are critically dependent on the c-kit receptor tyrosine kinase and its ligand, SCF. Nfl may modulate c-kit signaling and activated downstream effectors of c-kit in a gene dose dependent fashion. Moreover, this increased responsiveness of Nfl +/- mast cells to SCF
stimulation may mediate diverse alterations in the function of these cells including proliferation, survival, differentiation, degranulation, and cytoskeletal rearrangement. The purpose of these studies is to evaluate the cellular and biochemical consequences of heterozygosity of Nfl in mast cells in vitro and in vivo.
Evaluation of in vivo proliferation of cutaneous mast cells in Nfl +/- and +/+
mice. Mast cells have a complex developmental pattern of differentiation in that they are derived from progenitor cells found in the bone marrow (reviewed, Galli 1993). Mast cell precursors migrate via the circulation to interstitial tissues, serosal cavities and epithelial cells, and then complete their program of maturation, and differentiation within these tissues. Preliminary data indicate that heterozygosity of Nfl modulates the proliferation of BMMCs in vitro in response to stem cell factor. To investigate these findings in vivo, a murine model was employed that was originally developed by Galli who demonstrated that local subcutaneous injection of SCF increases the proliferation (Tsai et al., 1991) and survival (Iemura et al., 1994) of mast cell precursors located within the skin. Though other cytokines have been shown to influence the proliferation of mast cells (Lu-Kuo et al., 1996; Matsuda et al., 1991; Zhang et al., 1997), 5 SCF injections were employed in these experiments for reasons directly related to neurofibroma biology. Since fibroblasts and Schwann cells, two principal components of neurofibromas, have been shown to secrete substantial amounts of SCF to the local tissues (Badache et al., 1998; Bernex et al., 1996; Motro et al., 1991), examination of the heterozygosity of Nfl on mast cell proliferation, 10 survival and potentially migration to local concentrations of SCF in vivo may provide insight into one potential mechanism of accumulation of mast cells in neurofibromas ofNFl patients.
Cohorts of 8-12 week old Nfl +/- or Nfl +/+ mice on both a wildtype and W ~lYY4~ genetic background (wildtype, Nfl +/+; yy4l/T~~, Nfl +/-; y~~/yy'~, Nfl 15 +/-; +/+ receive for 3 weeks a daily subcutaneous injection of recombinant murine SCF (0, 30, or 100 ng) or vehicle alone (sterile saline with 0.1%
fraction V, bovine serum albumin) analogous to previous studies by Galli and colleagues (Tsai et al., 1991). Seven to 8 mice are enrolled into each cohort. Injections are performed with light anesthesia and delivered to approximately the same site on 20 the back for 21 consecutive days including the day of sacrifice. On the contralateral side of the back in the same animal, vehicle alone is administered.
In order to evaluate the proliferation of tissue mast cells proliferating in vivo, 5 bromo-2'-deoxyuridine is injected intraperitoneally 1 hour before death.
After sacrifice, the cutaneous injection site is excised, fixed in Carnoy's 25 fixative, and embedded in paraffin. Four micron sections are cut, placed on polylysine-coated slides and stained for counting of both total numbers of mast cells and the percentage of total cells that are actively proliferating. Two types of dermal mast cells are readily identified by conventional staining methods.
Mast cells that retain a capacity for local tissue proliferation are identified by 30 alcian blue staining, while terminally differentiated mast cells are identified by staining with the heparin-binding fluorescent dye berberine sulfate (alcian-blue negative). Both types of mast cells are quantitated to compare total numbers of mast cells at the local injection site between the four genotypes. In addition, slides that are stained for alcian blue, identifying mast cells that retain a proliferative potential, are subjected to immunohistochemical staining for BrdU-labeled nuclei using an anti-BrdU-mAb (Becton Dickinson, Mountain View, S Ca). By staining BrdU labeled nuclei, accurate quantification of the percentage of immature mast cells that are actively proliferating in the dermis is determined.
Quantification of both total numbers and the percentage of proliferating cells aids in determining the mechanism of the local accumulation in the dermis. An additional section is stained with hematoxylin and eosin to qualitatively evaluate increased endothelial cell, fibroblast, and cutaneous neuronal cell proliferation.
Administration of recombinant murine SCF induces a local proliferation of dermal mast cells in vivo in a dose-dependent fashion in all groups as well as a recruitment of immature mast cell precursors. If heterozygosity of Nfl enhances c-kit signaling in mast cells in vivo, then the tissues surrounding the injection site of Nfl +/- mice on either a W ~lT~l or wildtype genetic background will contain higher total numbers of mast cells (alcian blue and berberine sulfate positive cells) compared to tissues isolated from Nfl +/+ mice. One definitive assessment of localized mast cell proliferation used previously is measurement of the proportion of Brd-U positive/alcian blue positive cells. If increased signaling through c-kit in N. f'1 +/- mast cells confers a proliferative advantage in vivo, then Nfl +/- cells will have a higher percentage of BrdU positive mast cells than Nfl +/+ cells on both a mutant (W) and wildtype c-kit receptor background.
Since SCF has been shown to be important for mast cell survival as well as proliferation, it is possible that differences in total numbers of mast cells at local injection sites between genotypes could be in part a result of differences in survival. If differences in total numbers of mast cells at local injection sites are secondary to increased mast cell survival, then the percent of BrdU positive cells (proliferating cells) would be equivalent between experimental groups despite differences in total numbers of mast cells.
Evaluation of whether heterozygosity of Nfl reduces apoptosis and enhances survival of mast cells.

Survival of Mast Cells in Yitro. Nfl +/- BMMCs of both W genotypes is compared to Nfl +/+ cells in two previously used methods (Nocka et al., 1990).
First, mast cells from each respective genotype are assayed in serum-free media in response to exogenous SCF only. Survival in this assay has been shown to S correlate previously with c-kit receptor tyrosine kinase activity (Nocka et al., 1990). The mast cell survival assay as previously established is conducted as follows: BMMCs from each genotype are deprived of growth factors for 24 hours and 3 x 105 cells are plated in 24 well dishes in serum-free RPMI
containing 1% glutamine and 100 ng/ml of recombinant murine SCF in a total volume of 1 ml. The number of surviving cells are determined by trypan blue exclusion at 48, 72, and 96 hours of culture in a 37°C, S% C02, humidified incubator, and the number of apoptotic cells are calculated using a TUNEL
assay as described in other studies (Haneline et al., 1998).
A second method of evaluating the intrinsic resistance of Nfl +/- cells independent of either growth factor or serum is to evaluate their survival in serum-free and growth factor-free conditioned media. Cells are cultured as described above in 24 well dishes and evaluated for TUNEL positive cells 8, 16, 24, 36, and 48 hours after initiation of cultures and adjusted as indicated from preliminary studies. At least 4-5 independent experiments are conducted for each in vitro measure of apoptosis, and differences between genotypes will be examined using Student's T-test.
Evaluation of Survival Advantage~Nfl +/- mast cells in vivo.
Cohorts of mice receive a daily subcutaneous injection of either vehicle or 100 ng recombinant murine SCF for 3 weeks analogous to previous studies by Galli et al. (Iemura et al., 1994). Cohorts of mice (7-8 per group of each respective genotype) are treated as previously described above with vehicle or 100 ng/day of subcutaneous polyethylene glycol prepared SCF, and sacrificed at 1, 2, 4, and 7 days after daily injections of this long-acting SCF to quantitate total numbers of mast cells remaining at the injection site. In multiple prior experiments, Galli has shown that increased numbers of mast cells at local injection sites quickly return to normal by 2-4 days after cessation of SCF
subcutaneous treatment in wildtype mice (Iemura et al., 1994). Furthermore, this decline in mast cells directly correlated with an increase in apoptosis. Given that Nfl +/- mast cells have a survival in vitro, this design should be a sensitive measure of evaluating the effect of heterozygosity of Nfl on mast cell survival in vivo.
One hour prior to sacrifice, the mice receive BrdU as described above.
Sections are also prepared as described above, to quantitate total numbers of mast cells on days 1, 2, 4, and 7 after cessation of treatment. By comparing total numbers of mast cells on days 2, 4, and 7 post-treatment to total numbers of cells present on day 1 following the last injection of SCF, the proportionate decline in absolute numbers of cells is determined and compared between genotypes. In addition, slides are prepared for evaluation of apoptosis using the TUNEL
assay.
In addition, evaluation of apoptosis is conducted on sections stained with alcian blue and berberine sulfate, using characteristic morphologic indicators including condensed nuclear chromatin, pyknotic nuclei, and swelling or fusion of cytoplasmic mast cell granules.
If heterozygosity of Nfl augments signaling through SCF, there will be a decreased rate of apoptosis in dermal mast cells from the Nfl +/- mice. There are three measures that are used to test this: (1) morphologic characterization of alcian blue and berberine sulfate stained cells for apoptosis; (2) evaluation of apoptosis using TUNEL by determining the proportion of cells that are found in the tissues at the time of examination as compared to the number examined immediately after sacrifice; and (3) determining the proportionate decline in absolute numbers of cells between genotypes as a function of time. Total numbers of mast cells decrease more rapidly in wildtype mice as compared to Nfl +/- mice after cessation of daily injections of SCF. However, it is possible that a small residual amount of subcutaneous SCF may allow limited proliferation of dermal mast cells in the Nfl +/- mice. Evaluation of the proportion of BrdU positive cells directly measures any residual proliferation in either genotype. If there is minimal residual proliferation at the early time points, the other two measures of apoptosis will be used exclusively and subsequent experiments would be conducted where lower concentrations of subcutaneous SCF are administered.

Evaluation of whether heterozy osit~Nfl enhances mast cell degranulation.
To examine whether heterozygosity of Nfl alters the secretory function of mast cells, measurement of in vitro degranulation using previously established methods (Vosseller et al., 1997) is conducted. The release of [3H] serotonin (purchased from Amersham, Arlington Heights, IL) is measured. One to 1.5 x 106 of a homogeneous population of BMMCs from Nfl +/- and Nfl +/+ mice previously cultured in IL-3 and complete RPMI medium containing 10% fetal calf serum are incubated with the tritiated serotonin. Sensitization of mast cells with IgE increases the amount of serotonin that is released when a subsequent inflammatory stimulus is administered in vitro or in vivo. Therefore, serotonin release is measured from mast cells that have been prestimulated in either IgE
(10 mcg/ml DNP IgE, Sigma, St. Louis, MO) or vehicle and given a 2 minute pulse stimulation of SCF (3 ng/ml, 10 ng/ml, or 100 ng/ml). The cells are incubated for 10 minutes at 37°C, pelleted at 4°C, and the supernatants transferred to the scintillation tubes. Cells will then be lysed, and placed in separate scintillation tubes. The percentage of serotonin release is determined by the following equation:
100 supernatant ~ supernatant of spontaneous release specimens - (supernatant ~ supernatant of spont . release) + cells of pellet) At least 4 independent experiments are conducted. SCF as a single stimulant can also induce degranulation of connective tissue derived mast cells found.
Therefore, Nfl +/- and Nfl +/+ mast cells are pulse-treated with SCF alone in the absence of IgE sensitization, to evaluate the release of serotonin.
Since c-kit activation of PI3K and Rc~RI activation of Ras has been shown to be important in mast cell degranulation (Vosseller et al., 1997), Nfl +/-mast cells will have greater serotonin release than wildtype cells in response to SCF and IgE alone or in combination. To determine the effect of hypersecretion of mast cell mediators on the proliferation and/or survival of fibroblasts, coculture experiments of mast cells and fibroblasts in vitro. Since fibroblasts secrete growth factors such as SCF, the potential of Nfl deficient fibroblasts to alter the survival, proliferation and degranulation of Nfl +/- mast cells is determined in coculture experiments. Interestingly, in preliminary studies, the stimulation of Nfl +/- and Nfl -/- fibroblasts with low concentrations of histamine tremendously enhanced the proliferation in vitro as compared to wildtype cells in the presence of low concentrations of cytokines (unpublished results).

Example 5 The Biochemical Mechanisms Leading to Abnormal Cell Function in Nfl -l-, ~l +/ and Nfl +/+ Mast Cells in Response to Stem Cell Factor Loss of neurofibromin leads to elevated levels of activated Ras which 10 potentially results in the hyperactivation of multiple downstream effectors. Ras-GTP binds and activates the serine/threonine kinase Raf 1 which leads to activation of the MEK and ERK kinases (Ras-Raf ERK kinase pathway) (Marshall, 1996). Most proliferative stimuli in mammalian cells activate ERK
kinases following growth factor stimulation. While Ras activation of the ERK
15 kinase pathway is the best characterized biochemical pathway, Ras-GTP also binds and activates other downstream effectors via Raf independent mechanisms which are important for coordinating cellular responses to a wide variety of extracellular stimuli (White et al., 1995) (Figure 8). Recently, activated Ras has been shown to bind and activate PI3-kinase (PI3K) (Marshall, 1996). Ras 20 activation of PI3K leads to the activation of another MAP kinase family member, jun-N-terminal kinase (JNK), through a Raf 1 independent mechanism (White et al., 1995). PI3K activation of JNK requires the small G proteins Rac and Cdc42, but these pathways are currently not well defined (Minden et al., 1995; Coso et al., 1995). However, simultaneous activation of multiple 25 downstream effectors by Ras-GTP underscores the importance of examining the coordinated response of these downstream pathways in specific cell types, particularly in primary cells where Ras is hyperactivated. The biological importance of multiple effectors in Ras signaling has been recently illustrated using new Ras point mutants which bind and selectively activate only subsets of 30 Ras effectors (White et al., 1995). These studies have shown that these mutants are deficient in signaling when tested alone but cooperate to induce Ras transformation of cells when introduced together (White et al., 1995; Joneson et al., 1996; Rodriguez-Vicuna et al., 1997; White et al., 1996). Taken together, these studies have stressed the complexity and functional importance of simultaneous activation of multiple effectors following Ras activation.
While signaling pathways are often depicted in a linear fashion, an increasing body of evidence has now emerged that emphasizes the importance of cross talk between signaling pathways (Robinson et al., 1997; Garnngton et al., 1999; King et al., 1998). Specifically, many reports have documented significant cross talk between the PI3K and Ras-Raf ERK pathways (Robinson et al., 1997; Garrington et al., 1999). These studies have demonstrated inhibition of ERK activation by pharmacological inhibitors of PI3K following stimulation with various growth factors (Weenstrom et al., 1999). However, determination of the biological significance of cross talk between these different pathways has been limited because of the use of immortalized cell lines, and the wide variability in the strength and duration of stimulants used in these assays.
Recently, detailed descriptions of the mechanism of cross-cascade activation of ERKs by the Rho family of small GTP binding proteins (Rac and Cdc42) which are activated upon stimulation of the PI3K pathway have emerged (Robinson et al., 1997). Frost et al. (Frost et al., 1997) have recently demonstrated in NII-I 3T3 cells that p21 activated protein kinases (PAKs) may mediate the cross talk between the PI3K and Ras-Raf ERK pathways.
Specifically, these investigators have shown that PAK1 becomes activated via Rac following PI3K activation. Furthermore they showed that activated PAK1 phosphorylates and activates MEK1 on serine 298, which is the immediate upstream effector of ERK. The net effect of PAKI activation in these studies was 2-fold (1) greatly enhanced ERK activation via phosphorylation of MEK on serine 298 and (2) increased transcription of key growth related genes secondary to hyperactivated ERKs (Frost et al., 1997). In addition, Tang et al. (Tang et al., 1999) have shown that PAK1 activation via PI3K is necessary for Ras transformation of Rat-1 fibroblast, and other groups have described similar results utilizing different cell types and growth factor stimulation. Taken together, these studies strongly suggest that MEK1 activation by PAK1 may be the focal point of cross-cascade activation of ERK upon Ras-GTP activation of PI3K.
Recently, Tang et al. examined the role of PAK1 in Schwann cell transformation utilizing a neurofibrosarcoma cell line derived from an NF 1 S patient (Tang et al., 1998). They demonstrated that transfection of this cell line with a dominant negative PAK1 mutant inhibited transformation by preventing Ras activation of the ERK pathway. These data provide compelling evidence for the importance of PAKl in mediating cross-cascade activation of ERK from the PI-3 kinase pathway in pathogenesis of neurofibrosarcoma formation. While this study emphasizes the potential importance of PAKs in transformation of malignant cells where Ras is hyperactivated, little is known about the role of PAKs in regulating cell growth and differentiation in nonmalignant cells.
Furthermore, the role of PAKs in regulating the growth of Nfl +/- cells is completely unknown.
1 S To initially investigate the biochemical pathways responsible for differences in Nfl +/- and Nfl +/+ mast cell proliferation, pharmacologic inhibitors of two well characterized mitogenic pathways (the PI3K and the Ras-Raf ERK pathways ) were used to evaluate the contribution of each pathway to the proliferation of each genotype to SCF. 1 x 105 Nfl +/- and Nfl +/+ BMMCs were growth factor starved for 24 hours and then cultured in complete RPMI
media containing 10% fetal calf serum with 100 ng/ml SCF, in the presence of either LY294002 or vehicle, in triplicate cultures for 48 hours. Six hours prior to cell harvest, tritiated thymidine was added to the culture to quantitate DNA
synthesis as a measure of proliferation. Cells were harvested, and (3 emission was measured. Consistent with prior studies in wildtype mast cells, a specific inhibitor of PI3K, LY294002, significantly inhibited the proliferative response of both Nfl +/- and +/+ cells to SCF (Figure 9).
Because PI3K activation of ERKs via cross talk from the PI3 pathway has been shown to be important in some cell lineages where Ras is hyperactivated (Robinson et al., 1997), ERK activity in Nfl +/- and +/+ BMMCs was compared after stimulation with SCF in the presence of 25 p.M LY294002 or vehicle. These data are shown in Figure 10. Equivalent numbers of Nfl +/-and Nfl +/+ BMMCs were serum starved overnight. BMMCs were then incubated with LY294002 or vehicle for 1 hour and stimulated with SCF for 2 and 5 minutes. BMMCs were lysed and ERK activity was measured in a kinase activity assay as previously described (Bollag et al., 1996). Autoradiography and quantitative densitometry of the phosphorylation of Elk-1 fusion protein by ERK2 from lysates obtained from SCF stimulated BMMCs are shown. ERK2 activity in Nfl +/- mast cells was higher at baseline and 2 and 5 minutes after stimulation with SCF compared to wildtype cells. Data represents 1 of 3 independent experiments with similar results. Interestingly, preincubation of Nfl +/- and Nfl +/+ BMMCs with Ly294002 prior to stimulation significantly inhibited ERK activity in both genotypes at both time points tested.
Given the biochemical evidence of cross-cascade activation of ERK from PI3K, the effect of PD98059, a specific inhibitor of ERK, on the proliferation of Nfl +/- and +/+ mast cells was examined using thymidine incorporation assays to determine whether cross-cascade activation of ERK is important in mast cell proliferation. 1 x 105 Nfl +/+ and Nfl +/- BMMCs were growth factor starved for 24 hours, and cultured in complete RPMI media containing 10% fetal calf serum, 100 ng ml SCF, and either 50 ~M PD98059 or vehicle. Cells were cultured for 48 hours in triplicate wells and 6 hours prior to harvest tritiated thymidine was added to cultures. Cells were harvested on glass fiber filters and ~3 emission was measured. Data represents the mean of triplicate cultures in one representative experiment. Six other experiments were conducted with similar results. Using the same thymidine incorporation assay as outlined above, PD98059 was found to reduce the proliferation of Nfl +/- mast cells after stimulation with SCF to wildtype levels, while a modest (15-20%) inhibitory effect on wildtype cells was observed. These data are shown in Figure 11.
Since PD98059 markedly reduced the proliferation of Nfl +/- mast cells in comparison to wildtype cells, studies were initiated to determine the role of PAKs in mediating cross-cascade activation of ERK from PI3K in both genotypes. Using a similar method for measuring Akt activity and ERK activity, PAK activity was determined using a kinase activity assay in Nfl +/- and Nfl +/+ cells after stimulation with a low concentration of SCF (10 ng/ml). This dose was selected because it is closer to the range (3-5 ng/ml) observed in vivo in normal animals as compared to traditional saturating cytokine concentrations used in in vitro studies. Interestingly, PAK activity was greatly enhanced in Nfl +/- mast cells compared to wildtype cells at all time points tested. Thus, PAKs S may confer a growth advantage to Nfl +/- BMMCs following growth factor stimulation.
Evaluation of p21 activated kinases (PAKs) activation kinetics in Nfl -l-, N I
+l-and +/+ primary bone marrow derived mast cells following stimulation with stem cell factor.
Primary bone marrow derived mast cells (BMMCs) lines from Nfl +l-and +/+ mice are generated as previously described (Serve et al., 1995), and Nfl -/- mast cells are derived from primary d13.5 gestation fetal liver cells. At least 3-4 independently derived primary mast cell lines of the respective genotypes are tested. Fetal liver and bone marrow derived mast cells behave similarly in 1 S culture (unpublished results).
To determine both the level and kinetics of activation of PAKs after stimulation with SCF, BMMCs from each of the three genotypes are serum starved overnight to induce quiescence and stimulated with a range of concentrations (0, 3, 10, and 100 ng/ml) of SCF. Prior to stimulation, BMMCs are equalized for cell number and viability. Cells are stimulated for the following time course: 0, 15 seconds, 30 seconds, 1 minute, S minutes, and 15 minutes. Several concentrations of SCF are tested because of recent studies linking the strength of the stimulus and subsequent activation of PI3K to cross-cascade activation of MAPkinase. Wennstrom et al. (1994) has shown in COS-7 cells that basal levels of PI-3 kinase activity may be more important in cross-cascade activation of ERKs rather than pharmacologic growth factor stimulated PI-3 kinase activity. Similarly, Duckworth et al. 1997 has shown in Chinese hamster ovary (CHO) cells that activation of ERKs through PI3K occurs preferentially at low concentrations of platelet derived growth factor (PDGF), which PDGF is in the same growth factor superfamily as SCF, suggesting that cross talk between pathways is dependent on signal strength. , Following stimulation with SCF, BMMCs are lysed and PAK activity is determined utilizing a PAK activity assay. Three isoforms of PAKs have been identified which include PAK1, PAK2, and PAK3. Most studies have characterized the ability of PAK1 to mediate cross talk from the PI3K pathway 5 by phosphorylating MEK1 on serine 298. Thus, a specific antibody to PAK1 is used for immunoprecipitation after cell lysis to test PAK1 activity in these assays. Equal amounts of protein from each test group are used by performing a Bradford assay prior to immunoprecipitation. If differences are observed in PAK kinase activity between the three genotypes, total levels of PAK protein in 10 BMMCs are detected to ensure that differences are not secondary to increased levels of protein in Nfl deficient cells. However, this is unlikely because in other studies using leukemic cells of Nfl -/- myeloid cells the absolute amount of signaling proteins such as Ras and ERKs were similar (Zhang et al., 1998;
Bollag et al., 1996).
15 In addition, antibodies have been generated that specifically recognize the phosphorylated sites on serine residue 298 on MEK1, the specific phosphorylation site of PAK1. In addition to assaying for PAK1 activity, lysates from SCF stimulated BMMCs of the three genotypes are immunoprecipitated using a specific MEK1 antibody. Following immunoprecipitation, Western blots 20 are performed utilizing antibodies that recognize phosphorylated sites on serine-298 to evaluate the amount of phosphorylation by PAK1 in lysates generated from the 3 experimental groups. The amount of phosphorylation is quantitated using densitometry.
While much attention has focused on the role of PAK1 in mediating 25 cross-cascade activation of MAP kinase, recently PAK3 kinase was shown to specifically phosphorylate serine 338 on Raf 1 following PI3 kinase activation (King et al., 1998). Phosphorylation on this site enhanced Raf 1 kinase activity (King et al., 1998). Thus, PAK3 may be another potential mediator of cross talk between the PI3 kinase pathway and the Ras-Raf ERK pathway. To test whether 30 PAK3 may also contribute to the hyperproliferative phenotype observed in Nfl deficient cells, PAK3 activity is determined following stimulation with SCF.
Following stimulation and lysis of BMMCs from the 3 experimental groups, a specific PAK3 antibody generated for PAK3 immunoprecipitation is used.
Following PAK3 immunoprecipitation, PAK3 activity assays are performed to compare levels and time course of activation between the 3 genotypes.
In addition, specific antibodies have been generated that recognize phosphorylated sites on serine 338 on Raf 1 kinase in murine cells. Thus, in addition to assaying for PAK3 activity, lysates from the SCF stimulated BMMCs are immunoprecipitated using a specific Raf 1 kinase antibody. Following immunoprecipitation, Western blots are performed utilizing antibodies that recognize serine 338 on Raf 1 kinase to evaluate the amount of phosphorylation at different time points after stimulation with SCF in the three experimental groups. The amount of phosphorylation is quantitated using densitometry.
Most studies have shown that PAKs are directly activated by either Rac or Cdc42 which are downstream effectors of activated PI3K (Robinson et al., 1997). However, some investigators have shown that activated Ras-GTP can directly activate Rac and possibly other downstream effectors independent of PI3K (Robinson et al., 1997). Though data in Nfl +/- BMMCs suggests signaling through the PI3K pathway is important for cross-cascade activation of ERKs, it remains possible that elevated levels of Ras-GTP may also directly activate Rac or other downstream effectors independent of PI3K. If PAK
activation in any of the 3 experimental groups is dependent on PI3K
activation, then preincubation of BMMCs with LY294009 prior to stimulation with SCF
should inhibit PAK activation. In addition, it remains possible that different isoforms of PAK utilize PI3K independent pathways for cross-cascade activation of kinases in the Ras-Raf ERK pathway. By performing specific PAK1 and PAK3 kinase assays in the 3 genotypes in the presence of LY294009 or vehicle, this possibility can be tested. Finally, Rac (or other downstream effectors) and PI3 kinase may be simultaneously activated by Ras-GTP following SCF
stimulation converging upon the activation of PAKs. If this is true, preincubation of BMMCs with LY294009 will produce a partial reduction of PAK following SCF stimulation.

Evaluation of transfection of dominant ne;~ative PAKl or PAK3 on cellular proliferation of Nfl -l-, NCI +/-, and +/+ mast cells in response to SCF.
Stimulation of Nfl +/- BMMCs with 10 ng/ml of SCF results in elevated levels of PAK activity as compared to wildtype mast cells. In addition, a MEK
S inhibitor reduced the proliferation of Nfl +/- mast cells to wildtype levels, while addition of a MEK inhibitor to wildtype cells had a minimal effect on proliferation. The effect of a dominant negative PAK1 or PAK3 on the proliferative responses of Nfl -l-, Nfl +/-, or +/+ mast cells is determined.
Proliferative assays provide a sensitive measure of the biological effects of PAK
activation in BMMCs.
Primary BMMCs from each of the 3 experimental genotypes are transfected using standard lipofection transduction or alternatively transduced with retroviral vectors having PAK1 or PAK3 cDNAs (MSCV-pac) (Hawley et al., 1994; Clapp et al., 1995a; Clapp et al., 1995b). MSCV based vectors have been shown to strongly promote expression of encoded sequences in vivo in myeloid cells without evidence of transcriptional silencing (Clapp et al., 1995c).
The internal sequences of this bicstronic vector are under the transcriptional control of the myeloproliferative sarcoma retrovirus promoter. In addition, the retrovirus encodes the selectable marker gene Pac which renders cells resistant to puromycin. Thus, following transduction, transduced cells expressing the dominant negative PAK1 or PAK3 are selected by adding puromycin to the cultures. This backbone has been used with the same promotor sequences to express genes in murine fetal liver hematopoietic stem/progenitor cells for up to 20 months in vivo (Clapp et al., Blood, 86, 94 (1995)). In addition, this virus was used to transduce the NFl GRD sequences into Nfl -/- mast cells (unpublished results). Aliquots of cells from each line are also transduced with an MSCV retrovirus that expressed Pac sequences only in an identical transduction and selection protocol.
Initially, proliferative responses to SCF of BMMCs containing either a DN PAKl or a DN PAK3 from each genotype are compared to BMMCs from each experimental group transduced with the constructs encoding only the reporter genes. Mast cells are growth factor starved overnight in complete mast cell media containing 10% FCS, 2% penicillin/streptomycin, 1% glutamine.
Following growth factor deprivation, 3 x 105 viable BMMCs are counted using a trypan blue exclusion assay and plated in triplicate cultures in 24 well plates in complete mast cell media containing either 10, 25, 50, or 100 ng of SCF. Mast S cells are then placed in a humidified incubator at 37°C and S% COZ
and counted on days 1, 3, and 5 following incubation by trypan blue exclusion. Numbers of viable mast cells are expressed as a percentage of input cells to assess the proliferative response. Though 100 ng/ml is considered to be the maximal proliferative dose of SCF for BMMCs, the proliferation of mast cells in response to lower doses of SCF is tested to generate a dose response curve and to approximate more physiologic concentrations of SCF. Thymidine incorporation assays are also performed to confirm any proliferative differences observed in the experimental.
If PAKI or PAK3 is mediating cross-cascade activation of ERK via PI3K in Nfl -/- and Nfl +/- mast cells, then elevated levels of PAK1 or PAK3 activity in Nfl deficient mast cells are found compared to wildtype cells following SCF stimulation. Similarly, increased phosphorylation on serine residue 298 on MEK in Nfl deficient cells is observed if PAK1 activity is elevated, and an increase in phosphorylation of serine 338 on Raf 1 is found if PAK3 activity is elevated.
If PAK1 or PAK3 is mediating cross-cascade activation of ERK via PI3K to confer a proliferative advantage to Nfl +/- and Nfl -/- mast cells, then transduction of the dominant negative PAKl or PAK 3 cDNAs into Nfl deficient (Nfl -/- or Nfl +/-) cells results in a restoration of normal cytokine responsiveness to SCF in proliferation assays, similar to that observed in wildtype cells.
If transduction of Nfl deficient cells with a dominant negative PAKl or PAK3 results in only a partial correction of hyperproliferation to SCF, retroviruses are employed so that PAKl and PAK3 are coexpressed in the same cell. Following coexpression of both cDNAs within Nfl deficient mast cells, a similar pattern of proliferation to wildtype cells in response to SCF is noted if PAKl and PAK3 both contribute to the hyperproliferation of Nfl deficient cells.

Example 6 Heterozygosity at the Nfl Gene Alters Multiple Mast Cell Functions Methods Animals. Nfl +/ mice were obtained from Tyler Jacks at the Massachusetts Institute of Technology (Cambridge, MA) in a C57BL/6.129 background and backcrossed for 13 generations into the C57BL/6 strain. Wildtype C57BL/6 mice were obtained from Harlan Laboratories (Indianapolis, III. These studies were conducted with a protocol approved by the Indiana University Laboratory Animal Research Center. The Nfl allele was genotyped as described in Bollag et al. (1996) and Zhang et al. (1998).
Bone marrow mast cell culture and survival assay. Six to nine bone marrow derived mast cell lines (BMMCs) from each marine genotype were generated and used for survival, proliferation, chemotaxis, F-actin quantification and biochemistry assays. All experiments were conducted using at least 3 lines from each genotype. BMMCs were cultured as described in Serve et al. (1995) with minor modifications, and homogeneity of BMMCs was determined by Giemsa staining. Aliquots of cells were also stained with alcian blue and safranin to confirm that they were mast cells. Furthermore, FACS analysis revealed similar forward and side light scatter characteristics and the same percentage of c-kit+
expression in BMMCs of both marine genotypes (data not shown). The mast cell survival assay was done as follows: BMMCs from each genotype were deprived of growth factors for 24 hours and 3 x 105 cells were plated in 24-well dishes in serum-free RPMI containing 1% glutamine, 1.5% HEPES
(BioWhittaker, Walkersville, MD), 2% penicillin/streptomycin (BioWhittaker, Walkersville, MD) and 100 ng/ml recombinant marine stem cell factor (rmSCF) (Peprotech) in a total volume of 1 ml. In some experiments, LY294002 (Sigma) or PD98059 (New England Biolabs) was added to the cultures 1 hour before addition of rmSCF. The number of surviving cells was determined by trypan blue exclusion at 24 hours of culture in a 37°C, 5% CO2, humidified incubator.
Apoptosis was determined as described in Haneline et al. (1998) by examining SS
DNA fragmentation using the TUNEL assay according to the manufacturer's protocol. Assays were performed in triplicate.
Proliferation assays. BMMCs were deprived of growth factors for 24 hours and 1 x 105 cells were plated in 96 well dishes in 200 ml of RPMI containing 1%
S glutamine, 10% fetal bovine serum (Hyclone Laboratories) and 100 ng/ml of rmSCF or no growth factors, as indicated, in a 37°C, 5% COZ, humidified incubator. Cells were cultured for 48 hours, and tritiated thymidine (New Life Science Products, Inc.) was added to cultures 6 hours prior to harvest. Cells were harvested on glass fiber filters (Packard, Meriden, CT) and ~3 emission was measured. Assays were performed in triplicate.
Chemotaxis assays. Migration of BMMCs was evaluated using a transwell migration assay as previously described (Kim et al., 1998; Roberts et al., 1999).
Briefly, BMMCs were serum starved for 16-18 hours and 2.5 x 105 BMMCs were resuspended in 100 ~tl of serum free RPMI containing 1 % glutamine and 0.5% BSA (Sigma). BMMCs were then loaded onto each transwell filter (8-p.m pore filter Transwell, 24-well cluster; Costar, Boston, MA). Filters were then placed in wells containing 600 p,1 of serum free RPMI supplemented with 10 ng/ml rmSCF. After 3 hours of incubation at 37°C in S% COZ, the upper chamber was removed and the cells in the bottom chamber were counted using a hemocytometer. Assays were performed in triplicate. Varying concentrations of rmSCF (0, 2.5, 5, and 10 ng/ml) were placed in the upper and lower chambers of the transwell in a formal checkerboard analysis as previously described (Kim et al., 1998; Meininger et al., 1992).
Motility assay. Mast cell motility was directly observed by time lapse imaging of cells exposed to a gradient of 10 ng/ml rmSCF in a Dunn chemotaxis chamber (Dune and Zicha, Dunn & Jones, 1997; Weber Scientific Ltd., Surrey, England).
Cells (2-S x 104 cells in 10 p1 of RPMI) were applied to fibronectin fragment H-296 covered glass coverslips and allowed to adhere for 10 minutes at 37°C. The coverslips were mounted on the Dunn chamber, the inner well of which was filled with RPMI and the outer well was filled with RPMI/rmSCF. The chamber was sealed and mounted on the stage of a Nikon Diaphot 300 inverted microscope equipped with DIC optics. Chamber temperature was maintained at 37°C using a stage heater (Instec Instruments Inc., Boulder, CO). The chamber was allowed to equilibrate for 20 minutes to allow a stable gradient to form.
Images were recorded digitally at 15-second intervals using a 40x LWD
objective and a Spot RT cooled CCD camera and analyzed using Metamorph software (Universal Imaging, Brandywine, PA). The centroid coordinates of cells at each timepoint were determined on calibrated images and used to calculate the speed and direction of migration.
F-Actin duantitation. F-actin quantitation was performed as previously described (Roberts et al., 1999). Briefly, flow cytometry was used to quantitate the amount of filamentous actin per mast cell. Mast cells (106) were resuspended in 1 ml HBSS in polypropylene tubes and warmed at 37° for 3 minutes prior to addition of SCF or carrier. Cells were fixed after the specified time by addition of 10 volumes of PBS containing 4.6% formaldehyde and 0.1%
BSA. Fixed cells were treated with 0-1% Triton X-100 (Sigma) in PBS for 5 minutes at 22°C, washed and then incubated with 160 nm FITC-phallodoin (Sigma) for 20 minutes at 22°C prior to FACS analysis by flow cytometry. A
minimum of 10,000 mast cell events were recorded routinely and the results are reported as the mean cellular fluorescence (MCF). Assays were performed in triplicate.
Fluorescence microscopy. Cells were fixed with 3.7% paraformaldehyde in PBS
for 10 minutes at room temperature, then permeabilized with 0.01% Triton X-100 in PBS for S minutes. Coverslips were treated with block solution (wash buffer PBS with 1 mg/ml BSA, 0.5 mM MgCl2, 2.5 mM calcium chloride containing 50 mM ammonium chloride, 25 mM glycine, 25 mM L-lysine and 10% goat serum) for 30 minutes and then incubated in block solution containing 0.1 pg/ml rhodamine phalloidin (Molecular Probes, Eugene, OR) for 1 hour.
Coverslips were washed briefly with wash buffer and mounted on 50% glycerol-PBS containing 100 mg/ml diamino-bicyclo-[2.2.2] octane (Sigma). Slides were imaged using a Zeiss LSM S 10 laser scanning confocal microscope using a 100x, 1.4 N.A. oil immersion objective.
Raf 1. Mek, ERK. PAK, Akt in vitro kinase assays. Activation of Raf l, Mek, ERK, PAK and Akt kinases was determined by depriving BMMCs of serum and growth factors for 24 hours and followed by stimulation with 10 ng/ml rmSCF
for various amounts of time. Cells were washed twice with PBS containing 1 mM sodium orthovanadate and lysed in nonionic lysis buffer (20 mM
Tris/HCI, 137 mM NaCI, 1 mM EGTA, 1% Triton X-100, 10% Glycerol, 1.5 S mM MgCl2, and COMPLETE protease inhibitors (Amersham) as previously described(Bollag et al., 1996a). The protein lysates were equalized for protein concentration using the BCA assay (Pierce Chemical Co.) and equal loading of kinases in these assays was confirmed by Western blot. ERK and PAK kinase immunoprecipitations were carried out with an anti-ERK-2 (C-14) antibody and an anti-PAK(C-19) antibody (Santa Cruz), respectively. Raf 1 and Mek kinase immunoprecipitations were carried out with an anti-Raf 1 (C-12) antibody (Santa-Cruz) and an anti-Mek-1 (C-18) antibody (Santa Cruz), respectively. Akt kinase immunoprecipitations were carried out with an anti-Akt antibody (Cell Signaling Technology). The kinase immune complex assay was performed as previously described (Bollag et al., 1996a) and the following proteins were used as phosphorylation substrates: Elk-1 fusion protein (New England Biolabs) for JNK and ERK kinase, histone 2B (Roche) for Akt, myelin basic protein (Sigma) for PAK kinase and recombinant MEK1 fused with GST at the N-terminus (Upstate Biotechnology) for Raf 1 kinase. For Mek, Raf 1, and ERK, kinase reactions were resolved on 10% SDS-PAGE gels (Novex). For PAK and Akt, kinase reactions were resolved on 4-20% gradient SDS-PAGE gels (Novex), and transferred to PVDF filters. Gels and PVDF filters were dried and subjected to autoradiography. Densitometry of individual bands was conducted using NIH
Image software.
Ras, Cdc42 and Rac activation assays. Bone marrow derived mast cells were deprived of serum and growth factors for 24 hours and stimulated at various amounts of time with 10 ng/ml rmSCF. Ras, Cdc42 and Rac activation was subsequently determined using Ras, Cdc42, and Rac activation assay kits (Upstate Biotechnology) according to the manufacturer's protocol and as previously described in Gille and Downward (1999).
Immunoprecipitation and Western blotting_ BMMCs were lysed in nonionic lysis buffer as previously described (Bollag et al., 1996). Cleared lysates were normalized for protein content using the BCA assay (Pierce Chemical Co.) and proteins were immunoprecipitated with protein A sepharose beads (Amersham) coupled with polyclonal antibodies for Raf 1, Mek, PI-3 kinase p1 108 (H-219) (Santa Cruz), PI-3 kinase p1 10a (H-201) (Santa Cruz), or c-kit (C-19) (Santa Cruz) for 2 hours at 4°C. Immunoprecipitates were washed three times in lysis buffer, resuspended in sample buffer, boiled for 5 minutes, subjected to SDS-PAGE and transferred to nitrocellulose. The membranes were blocked in TBS-Tween containing 5% BSA overnight. Membranes were incubated for 1-3 hours at room temperature with following antibodies: anti-phospho-Mek 1/2 (1:1000) (New England Biolabs), anti-phosho-Raf 1 338 (1:5000) (Upstate Biotechnology), anti-phospho-Mek 298 (gift from Dr. Mark Marshall), anti-p21'~S (Upstate Biotechnology), anti-phosphotyrosine antibody (Transduction Laboratories) anti-phospho-cofilin antibody (gift from Dr. Simon Atkinson).
Secondary antibodies used were either anti-rat, anti-rabbit, or anti-mouse Ig horseradish peroxidase conjugated (Transduction Laboratories). Proteins were visualized by enhanced chemiluminescence (Santa Cruz).
Studies in vivo. Adult Nfl +/ or wildtype mice received a continuous infusion of various doses of rmSCF or vehicle (PBS) from micro-osmotic pumps (Alzet) placed under the dorsal back skin. Osmotic pumps were surgically placed under light avertin anesthesia. rmSCF or vehicle was released over 7 days at a rate of 0.5 ~.l/hr, and osmotic pumps were surgically removed on day 7 following sacrificed by cervical dislocation. To accurately identify cutaneous sections for quantitating changes in mast cell numbers in response to rmSCF, the dorsal skin was stained with a drop of India ink at the point of exit of rmSCF from the osmotic pump prior to removal of the pump. Three cm sections of skin marked with India ink were removed, fixed in buffered formalin and processed in.
paraffin-embedded sections. Specimens were stained with hematoxylin-eosin to assess routine histology and with Giemsa stain to identify mast cells.
Cutaneous mast cells were quantitated in a blinded fashion by counting 2 mm2 sections in proximity to the India ink stain.

Results Stimulation of Nfl +/ mast cells with stem cell factor. SCF binding to its receptor, c-kit, causes a rapid increase in p21'~ activity in primary BMMCs (Serve et al., 1995). To investigate whether heterozygosity at the Nfl allele alters p2lr~ activity in mast cells, Nfl +/ ~d wildtype BMMCs were stimulated with SCF and assayed for changes in active p2lras-GTP levels. Following stimulation of BMMCs, levels of GTP-bound p21'~ in cellular lysates were determined by precipitating the active GTPase with a GST-fusion of the p2lr~
binding domain of Raf 1 kinase. Even though Nfl +/ mast cells had detectable, but reduced, levels of neurofibromin as determined by Western blot (data not shown), Nfl +/ mast cells had higher basal and SCF stimulated p2lr~-GTP
levels compared to wildtype cells (Figure 13A). Importantly, preincubation of both mast cell genotypes with PI-3 kinase inhibitors, wortmannin or LY294002, did not alter p21'as activation (data not shown). In contrast to other cell types, these data position p21'~S either in parallel or upstream of PI-3 kinase in mast cells. Thus, heterozygosity at Nfl alters p2lTas activity in primary BMMCs.
In mast cells, p1108 is the predominantly expressed class IA PI-3 kinase catalytic subunit isoform, and active p2l~as-GTP can directly bind to the p21'as binding domain of p1 108 to increase PI-3 kinase activity (Luo-Kuo et al., 2000, and Vanhaesebroeck et al., 1997). To examine whether p2lras co-immunoprecipitates with p 1108 in primary BMMCs after stimulation with SCF, Nfl +/ and wildtype BMMCs were stimulated with SCF for S minutes and p1108 was immunoprecipitated from lysates using a specific polyclonal antibody. Following immunoprecipitation, immune complexes were separated on SDS-PAGE gels and Western blotting was performed with a specific antibody for p21'as. While p21'~S was identified in the p1108 immune complex in both mast cell genotypes following SCF stimulation, p21'~S was also present in the immune complex of unstimulated Nfl +/ BMMCs at a level equivalent to that in SCF stimulated wildtype BMMCs (Figure 13B). Thus, identification of increased levels of p2l~as in the p1108 immune complex in unstimulated Nfl +/
BMMCs is consistent with the elevated basal p21'~S activity observed in these cells.

Direct binding of c-kit to the p85a regulatory subunit of class IA PI-3 kinase is important for kinase activation following SCF stimulation (Serve et al., 1995). However, the effect of hyperactivation of p21'~ on PI-3 kinase activation is unknown in primary mast cells. To determine the effect of increased p21'~-5 GTP levels on PI-3 kinase activity in Nfl +/ mast cells, the activation of the serine/threonine protein kinase B (PKB), also known as Akt, in N_ fl +/ and wildtype BMMCs was compared after stimulation with SCF. Activation of Akt is PI-3 kinase dependent and provides a sensitive measure for PI-3 kinase activity (Alessi and Cohen, 1998). Nfl +l BMMCs demonstrated greater basal 10 and SCF- stimulated Akt activity compared to wildtype cells (Figure 13C).
In addition, inhibitors of PI-3 kinase inhibited Akt activity in both mast cell genotypes (Figure 13C) confirming that the p21'~ contribution to Akt activation is via activation of PI-3 kinase. Taken together, these data demonstrate that increased p21'as activity in Nfl +/ BMMCs following SCF stimulation is 15 biochemically linked to increased activation of the PI-3K-Akt kinase pathway.
Increased Survival and Hvoernroliferation of mast Cells is Mediated Through Hyneractivation of PI-3 kinase. SCF-stimulated activation of PI-3 kinase is critical for mast cell survival (Serve et al., 1995). Ingram et al. (2000) showed a relative increase in survival of Nfl +/ BMMCs compared to wildtype cells in 20 response to SCF. To determine whether hyperactivation of PI-3 kinase could account for this phenotype, SCF-induced survival of Nfl +/ and wildtype BMMCs was compared after 24 hours of culture in serum free media in the presence or absence of varying concentrations of LY294002, a specific PI-3 kinase inhibitor. Significantly greater concentrations of LY294002 were 25 required to inhibit survival of Nfl +/ BMMCs compared with wildtype BMMCs, which correlates with hyperactivation of Akt in Nfl +/ BMMCs (Figure 14A). In addition, when both mast cell genotypes were cultured in serum-free media without SCF, wildtype mast cells undergo apoptosis at a faster rate compared with Nfl +/ mast cells (data not shown) which is consistent with 30 increased basal Akt activity observed in Nfl +/ BMMCs. Thus, increased survival of Nfl +/ BMMCs in response to SCF occurs at least in part via hyperactivation of the p21'as-PI-3 kinase pathway.

Proliferation of wildtype BMMCs in response to SCF is dependent on PI-3 kinase activation and subsequent stimulation of downstream effectors including the protein kinase JNK (c-jun N-terminal kinase) and Akt (Lu-Kuo et al., 2000; Timokhina et al., 1998). While it was previously shown that Nfl +/
mast cells have increased proliferation in response to SCF, the biochemical mechanisms responsible for this phenotype remain unclear. To test whether SCF-induced hyperproliferation of Nfl +l BMMCs is mediated through hyperactivation of PI-3 kinase, DNA synthesis was measured in Nfl +/ and wildtype BMMCs when incubated with SCF or medium alone, in the presence or absence of LY294002. Thymidine incorporation was determined forty-eight hours after addition of SCF. Nfl +/ BMMCs showed a 2-fold greater increase in DNA synthesis compared with wildtype cells, as previously observed (Figure 14B). DNA synthesis of both mast cell genotypes was greatly inhibited in the presence of LY294002 (Figure 14B), though Nfl +/ mast cells consistently have 1 S higher proliferation compared with wildtype BMMCs over a range of LY294002 concentrations (data not shown). Similar results were obtained when mast cells were counted and viability was assessed with trypan blue (data not shown).
Thus, increased SCF-induced proliferation of Nfl +/ mast cells is mediated through hyperactivation of the p21'as-PI-3 kinase pathway.
Since activation of JNK, through PI-3 kinase, is important for proliferation of BMMCs (Timokhina et al., 1998), JNK kinase activity in Nfl +/
and wildtype BMMCs was compared following stimulation with SCF. Nfl +/
BMMCs had a 2-fold greater increase in JNK kinase activity from baseline after stimulation compared to wildtype cells (Figure 14C). Thus, differences in proliferation between wildtype and Nfl +/ BMMCs correlate with differences in activation of the PI-3 kinase-JNK pathway. While activation of JNK is dependent on PI-3 kinase in wildtype BMMCs, p2l~as-induced activation of JNK
has been shown to be independent of PI-3 kinase in other cell types. To test whether hyperactivation of JNK in Nfl +/ mast cells occurred independent of PI-3 kinase, JNK activity in both mast cell genotypes was compared in the presence or absence of PI-3 kinase inhibitors following stimulation with SCF.
Both wortmannin (Figure 14C) and LY294002 (data not shown) inhibited JNK

activity in both mast cell genotypes indicating that JNK activation is dependent on signals from PI-3 kinase.
H ,y~erproliferation of Nfl +/ Mast Cells is Independent on Crosscasade Activation of ERK from PI-3 Kinase. While hyperactivation of the PI-3 kinase -JNK pathway provides a simple model for the hyperproliferation of Nfl +/
BMMCs, other downstream effectors of PI-3K could contribute to the proliferation of Nfl +/ BMMCs. In other cell types, signals from PI-3K
converge upon the p2l~as-Raf Mek-ERK pathway to activate ERK and promote transcription of genes that control mitogenesis (Frost et al., 1997; Tang et al., 1999). Nfl +/ BMMCs have elevated ERK activity relative to wildtype cells following stimulation with SCF, but the contribution of ERK activation to the proliferation of mast cells is unclear (Ingram et al., 2000). To test whether ERK
activity contributes to BMMC proliferation, DNA synthesis was measured in Nfl +/ and wildtype mast cells when incubated with SCF or medium alone, in the presence or absence of the MEK inhibitor PD98059, which blocked ERK
activation in both genotypes (data not shown). Forty eight hours after the addition of SCF, thymidine incorporation was determined. Surprisingly, in multiple independent experiments (n=6), addition of PD98059 reduced the proliferative response of wildtype BMMCs by approximately 30% and that of Nfl +/ BMMCs by 50% (Figure 14B). Thus, ERK activation contributes to proliferation of both mast cell genotypes. Importantly, hyperactivation of ERK
confers a distinct proliferative advantage to Nfl +/ BMMCs, especially in light of the observation that addition of PD98059 reduced the SCF-induced proliferation of Nfl +/ mast cells to that of wildtype levels.
Given that PI-3 kinase inhibitors significantly block the proliferation of both genotypes and MEk inhibitors partially inhibit proliferation, it was hypothesized that PI-3 kinase is biochemically linked to ERK activation in BMMCs. To test this hypothesis, both mast cell genotypes were stimulated with SCF in the presence or absence of PI-3 kinase inhibitors and measured ERK
activity. Similar to previous studies, Nfl +/ BMMCs demonstrated a greater increase in ERK activity from baseline compared with wildtype cells (Figure 14D). However, ERK activity was inhibited in both genotypes in the presence of PI-3 kinase inhibitors (Figure 14D). In contrast to other cell types, where ERK
activation is completely dependent on the classical p2lras-Raf Mek pathway, these data demonstrate that activation of ERK in mast cells, in response to SCF, is dependent on biochemical signals generated through PI-3 kinase converging on the p2lras-Raf MEK-ERK pathway.
Crosscascade Activation of ERK from PI-3 kinase Occurs at the Level of MEK
and Raf 1 kinase in Nfl +/- and Wildtype Mast Cells. In other experimental systems, crosscascade activation of ERK from PI-3 kinase occurs at the level of either MEK or Raf 1 kinase (Frost et al., 1998; Garrington et al., 1999; Tang, et al 1999). To initially test whether crosstalk between PI-3 kinase and the p21'~-Raf MEK-ERK pathway occurs at the level of MEK, Nfl +/ and wildtype BMMCs were stimulated with SCF in the presence or absence of a, PI-3 kinase inhibitor and measured MEK activity. Nfl +/ BMMCs demonstrated a greater increase in MEK activity from baseline compared to wildtype cells and MEK
activity was inhibited in the presence of PI-3 kinase inhibitors in both mast cell genotypes (Figure 15A).
Since Raf 1 functions as the immediate upstream effector of MEK by phosphorylating Ser217/221 (Alessi et al., 1994), Nfl +/ and wildtype mast cells were stimulated with SCF in the presence or absence of wortmannin and Raf 1 activity and phosphorylation of these residues measured. While wildtype BMMCs demonstrated minimal Raf 1 activation after stimulation with 10 ng/ml of SCF, Nfl +/ BMMCs had a significant increase in Raf 1 activity from baseline in multiple independent experiments (n=5) which was completely inhibited in the presence of the inhibitor (Figure 15B). In addition, Nfl +/
BMMCs had increased phosphorylation of Mek Ser217/221 which correlates with increased Raf 1 activity in these cells (Figure 15C). Importantly, phosphorylation of these residues was also inhibited by a PI-3K inhibitor in both mast cell genotypes (Figure 15C).
Taken together, these data suggest a biochemical model where PI-3 kinase activation of ERK occurs at the level of either Raf 1 alone or at the level of both Raf l and MEK in BMMCs after stimulation with SCF. Importantly, increased Raf 1 and MEK activity observed in Nfl +l mast cells is consistent with the hypothesis that the hyperproliferative phenotype of Nfl +/ BMMCs is secondary to increased crosstalk from PI-3 kinase to the p21 '~S-Raf MEK-ERK
pathway.
Different isoforms of p21-activated kinase (PAK) can mediate crosstalk between PI-3 kinase and the p21 gas-Raf MEK-ERK pathway by phosphorylating Ser298 on MEK1 and Ser338 on Raf 1 (Frost et al., 1997; King et al., 1998).
Phosphorylation of Ser298 increases MEK1 binding to Raf 1 to enhance the kinase activity of the complex and phosphorylation of Ser338 is critical for Raf 1 activation in a number of experimental systems (Chaudhary et al., 2000;
Frost et al., 1997; King et al., 1998). Since PAK may mediate crosstalk between PI-3K and ERK in BMMCs through these specific residues and potentially account for the Nfl +/ proliferative phenotype, PAK activity was measured in both mast cell genotypes in response to SCF. While both mast cell genotypes demonstrated PAK activation after stimulation with SCF, Nfl +/ BMMCs had dramatically increased basal and SCF stimulated PAK activities compared with wildtype cells (Figure 16A). In addition, hyperactivation of PAK in Nfl +/
BMMCs correlated with increased phosphorylation of Ser298 on MEKl and Ser338 on Raf 1 (Figure 16B). Since PAK has been shown to be located either upstream or downstream of PI-3 kinase in different cell types (Chaudhary et al., 2000; Sun et al., 2000; Yablonski et al., 1998), the PAK activity assay was repeated in the presence or absence of a PI-3K inhibitor. In both mast cell genotypes, wortmannin inhibited both PAK activation (data not shown) and phosphorylation of Ser298 on MEK1 and Ser338 on Raf 1 following stimulation with SCF (Figure 16B).
PAK interacts specifically with GTP bound forms of its direct upstream effectors, the small Rho GTPases Rac and Cdc42. The catalytic activity of PAK
is regulated by binding of Rac-GTP and Cdc42-GTP to the highly conserved N-terminus, known as the CRIB binding domain (Bagrodia and Cerione, 1999;
Daniels and Bokoch, 1999). To test whether heterozygosity of Nfl altered the direct upstream effectors of PAK, accounting for hyperactivation of the kinase, Cdc42-GTP and Rac-GTP levels were compared in Nfl +/ and wildtype BMMCs after stimulation with SCF. Consistent with elevated PAK activity, Nfl +/ BMMCs had higher basal and SCF stimulated levels of both active Cdc42-GTP and Rac-GTP compared with wildtype cells (Figures 16C-D). Thus, though other effectors may mediate crosstalk between PI-3 kinase and ERK in both mast cell genotypes, these data offer evidence that hyperactivation of PAK
5 may directly contribute to elevated ERK activity and enhanced proliferation in Nfl +/ mast cells.
Heterozvgosit~Nfl Increases F-actin content, Cofilin phosphorylation and Chemotaxis in Mst Cells in Response to SCF. Alterations in Cdc42, Rac, and PAK activity have been linked to dramatic changes in actin cytoskeletal 10 dynamics and actin based cell functions including chemotaxis (Bagrodia and Cerione, 1999; Daniels and Bokoch, 1999). Given increases in Cdc42/Rac-GTP
levels and PAK activity in Nfl +/ mast cells, F-actin content was compared in Nfl +/ and wildtype BMMCs following stimulation with SCF. After stimulation, both mast cell genotypes were stained with phalloidin to quantitate 15 levels of F-actin by fluorescence cytometry and fluorescence microscopy. In both assays, Nfl +/ BMMCs displayed increased F-actin compared with wildtype cells (Figures 17A-B). Surprisingly, Nfl +/ BMMCs had greater F-actin content at baseline compared to F-actin levels obtained in wildtype cells stimulated with SCF (Figures 17A-B). Thus, increased F-actin content in Nfl 20 +/ BMMCs is consistent with hyperactivation of proteins known to regulate the actin cytoskeleton. To test whether increases in F-actin content were dependent on activation of PI-3 kinase, these experiments were repeated in the presence of PI-3 kinase inhibitors. Preincubation of both mast cell genotypes with either LY294002 or wortmannin inhibited increases in F-actin content after stimulation 25 with SCF (data not shown).
Recently PAK has been linked to activation of the LIM-kinase >ADF/cofilin > F-actin pathway (Edwards et al., 2000). ADF/cofilin is a class of small actin binding proteins that promote depolymerization of actin by increasing the off rate for actin monomers at the pointed end (slow growing end) 30 and by promoting filament fragmentation (Moon and Drubin, 1995; Theriot, 1997). ADF/cofilin is phosphorylated by LIM-kinase. The ability of cofilin to disassemble actin filaments is reduced following its phosphorylation by LIM-kinase resulting in increased actin content (Arber et al., 1998; Yang et al., 1998).
Since Nfl +/ mast cells have increased PAK activity and F-actin content, the phosphorylation of cofilin was compared in both mast cell genotypes after stimulation with SCF by Western blot. Nfl +/ BMMCs had higher levels of phosphorylated cofilin at both baseline and after stimulation with SCF
compared to wildtype cells (Figure 17C) consistent with hyperactivation of a signaling cascade which controls F-actin polymerization.
Since PAK activation and subsequent changes in F-actin polymerization have been linked to the ability of cells to move towards a chemical gradient (Bagrodia and Cerione, 1999; Daniels and Bokoch, 1999), the effect of heterozygosity at Nfl on mast cell chemotaxis was examined. Chemotaxis was assessed using a transwell migration assay in which cells in the upper compartment migrate through a porous membrane towards a chemoattractant in the lower chamber. SCF was placed in the lower well as a chemoattractant since SCF has previously been shown to stimulate the chemotaxis of BMMCs (Meininger et al., 1992). At concentrations of SCF (10 ng/ml) where mast cell chemotaxis was maximal, Nfl +/ BMMCs showed a two fold increase in chemotaxis compared to wildtype cells (Figure 17D). These findings were verified using single cell level video microscopy and quantitating range of speeds for both mast cell genotypes. (Figures 17E-F). Importantly, varying the concentration of SCF in the upper and lower chamber of the transwell in a formal checkerboard analysis (data not shown) revealed that mast cell movement was directional towards the SCF gradient (chemotaxis) and not random (chemokinesis). In addition, preincubation of both mast cell genotypes with PI-kinase inhibitors greatly reduced the chemotactic response to SCF confirming the importance of hyperactivation of PI-3 kinase to the Nfl +l mast cell phenotype (data not shown).
Nfl +l Mice Have Greater Accumulation of Mast Cells in Response to Local Cutaneous Administration of Stem Cell Factor. Local cutaneous injection of SCF into wildtype mice results in a dramatic accumulation of mast cells at the site of injection (Iemura et al., 1994; Tsai et al., 1991). Local expansion of mast cells is dependent on mast cell proliferation, survival, and chemotaxis in response to SCF in this experimental model (Tsai et al., 1991). To test whether heterozygosity at Nfl alters accumulation of mast cells to SCF in vivo, slow release micro-osmotic pumps were filled with a range of concentrations of SCF
or vehicle and placed into the dorsal back skin of either wildtype or Nfl +l mice. Seven days after placement, pumps were removed and cutaneous skin biopsies were taken from the site of pump insertion and stained with Giemsa stain to identify mast cells (Figure 18A-B). Nfl +/ mice demonstrated greater accumulation of mast cells at the site of SCF release compared with wildtype mice, even at a dose (2 pg/kg/day) which failed to elicit a significant mast cell response in wildtype mice (Figure 18C). Interestingly, Nfl +l mice also had a greater percentage of degranulating mast cells at this dose (28% vs 8%) compared to wildtype mice at the site of SCF release. Thus, these studies clearly demonstrate that the biochemical mechanisms identified in vitro for the Nfl +l mast cell phenotype are operative in a physiologic system.
Discussion Neurofibromatosis type I is a complex genetic disease that affects multiple cell lineages. Alterations in p21'as activity have been linked with LOH
in tumors generated from individuals with NF 1 and Nfl knockout mice, consistent with the classification of NFl as a tumor suppressor gene (Bollag et al., 1996; DeClue et al., 1992; Largaespada et al., 1996). However, the frequent nonmalignant manifestations found in patients with NF1 infer a gene dosage effect. Consistent with this hypothesis is the slow and self limited growth of most neurofibromas, combined with the observation that genetic analysis of many neurofibromas fail to detect LOH. Thus, a major gap in the understanding of the pathogenesis of NF 1 has been the identification of altered biochemical pathways responsible for these heterozygous effects, including biochemical abnormalities that may predispose patients to a somatic loss of the normal NFI
allele.
In primary mast cells, activation of PI-3 kinase following c-kit binding to SCF is important for controlling proliferation, survival, and chemotaxis (Meininger et al., 1992; Vosseller et al., 1997). The present study shows that Nfl +/ mast cells have a 2-fold increase in each of these cellular functions in response to SCF as compared with wildtype cells in vitro. In addition, these functions were dramatically reduced with PI-3 kinase inhibitors, confirming the importance of this pathway in Nfl +/ cells. Importantly, placement of micro-osmotic pumps containing SCF into the skin of Nfl +/ mice resulted in a greater accumulation of mast cells at the site of insertion compared to wildtype mice.
Others have previously shown that in vivo mast cell expansion in response to local injection of SCF in wildtype mice occurs secondary to local proliferation, increased survival, and chemotaxis of mast cells to SCF (Iemura et al., 1994;
Tsai et al., 1991). Thus, the in vivo observations suggest that the biochemical mechanisms identified in vitro for the Nfl +/ cell phenotype are valid and biologically operative in a more physiologic system. Interestingly, compared with wildtype mice, Nfl +/ mice also presented a dramatic increase in extracellular matrix deposition, neovascularization, and inflammation which are hallmark features of neurofibromas (Jaakkola et al., 1989; Metcalfe et al., 1999).
Taken together, these studies strongly suggest that hyperactivation of the p2lras-PI-3 kinase pathway in Nfl +/ mast cells alters multiple mast cell functions in vivo and in vitro.
A second line of evidence that hyperactivation of p2lras alters signaling through PI-3 kinase in Nfl +/ mast cells is provided by measuring activation of Akt and JNK, kinases important for mast cell proliferation (Lu-Kuo et al., 2000;
Timokhina et al., 1998). Nfl +/ mast cells have increased Akt and JNK activity consistent with the hyperproliferative phenotype. These data are intriguing since prior studies in mast cells have focused on the requirement of binding of the p85oc regulatory subunit of PI-3 kinase to the phosphorylated c-kit receptor allowing activation of these kinases. Complementary studies in p85a -/ mast cells and in mast cells where c-kit is mutated at a point that interferes with its interaction with p85a demonstrate that this binding is necessary for Akt and JNK activation (Lu-Kuo et al., 2000; Timokhina et al., 1998). Since no differences in c-kit receptor expression were detected between Nfl +/ and wildtype BMMCs following stimulation with SCF (data not shown), another mechanism of Akt and JNK activation must exist in Nfl +/ mast cells to account for the biochemical phenotype.

p2l~as-GTP can specifically bind the p110a and the p1108 catalytic subunits of Class IA PI-3 kinase to increase protein kinase activity (Khwaja et al., 1997; Kodaki et al., 1994; Rodriguez-Vicuna et al., 1996; Rodriguez-Viciana et al., 1997; Vanhaesebroeck et al., 1999). Importantly, these p 110 catalytic subunits have been shown to control cellular functions necessary for accumulation of inflammatory cells at the site of stimulus (Vanhaesebroeck et ' al., 1999). In macrophages, blocking antibodies directed at the p1 10a subunit inhibited the proliferation and survival of these cells in response to colony stimulating factor and antibodies raised against the p 1108 subunit inhibited migration and F-actin polymerization in response to the same stimulus (Vanhaesebroeck et al., 1999). In the present studies, p110a (data not shown) and p1 108 were not co-immunoprecipitated with p2lras in both mast cell genotypes following stimulation with SCF. Utilizing a p21'~ effector mutant, another group recently correlated levels of Akt phosphorylation with p21'~
signaling to p1108 in primary mast cells (Kinashi et al., 2000). In the present study, coimmunoprecipitation of p1108 with p2lr~ was observed following SCF
stimulation in both mast cell genotypes, but co-immunoprecipitation of these proteins was also observed in the absence of SCF stimulation in Nfl +/ mast cells. Given high basal and SCF stimulated p21'as activity, this result could provide a mechanistic link between p21'as and hyperactivation of Akt and JNK
in Nfl +/ mast cells.
While the described studies focused on the role of the p21~-PI-3 kinase pathway in the Nfl +/ cellular phenotype, several unexpected findings led to the examination of the potential contribution of p21 activated kinase (PAK) to the phenotype of these cells. First, Nfl +/ mast cells showed increased chemotaxis and F- actin content in response to SCF. Interestingly, increases in F-actin content were greater at baseline in Nfl +/ cells than at any stimulated timepoint in wildtype cells. Hyperactivation of PAK has been linked to increased motility and F-actin polymerization in fibroblasts (Sells et al., 1999). In addition, PAK
can phosphorylate LIM- kinase which results in increased phosphorylation of the actin regulatory protein, cofilin. When cofilin is phosphorylated, its F-actin depolymerizing activity is inhibited leading to the accumulation of actin filaments (Moon and Drubin, 1995; Theriot, 1997). In the present studies, Nfl +/ mast cells displayed increased endogenous PAK activity and elevated levels of phosphorylated cofilin both at baseline and in response to SCF when compared to wildtype cells. Finally, phosphorylation levels of cofilin observed S in Nfl +/ mast cells were consistent with elevated F-actin content. Taken together, these experiments link the alterations observed in motility and F-actin content in Nfl +/ mast cells to hyperactivation of proteins critical for regulating these processes.
The second cellular function where hyperactivation of PAK could 10 contribute to the Nfl +/ phenotype is proliferation. Ingram et al. (2000) reported that Nfl +/ mast cells elicit increased ERK activation compared to wildtype mast cells after stimulation with SCF. While hyperactivation of PI-3 kinase provides a simple model for the increased proliferation of Nfl +/ mast cells, unexpectedly it was found that activation of ERK is completely dependent 1 S on signals from PI-3 kinase in both mast cell genotypes and is linked to the hyperproliferative phenotype of Nfl +/ mast cells. Several lines of evidence support these conclusions. PI-3 kinase inhibitors completely blocked the activation of ERK in both mast cell genotypes linking ERK to PI-3 kinase. In addition, physiological relevance of crosstalk between these pathways is shown 20 by the action of a specific MEK inhibitor which reduced the SCF-stimulated proliferation of both mast cell genotypes. Since a MEK inhibitor reduces proliferation of Nfl +/ mast cells to wildtype levels, hyperactivation of ERK
from PI-3 kinase confers a distinct growth advantage to these cells. Given the importance of this finding to the proliferative phenotype of Nfl +/ mast cells, 25 PAK was focused on as a biochemical mediator of this crosstalk.
Different isoforms of PAK have been shown to increase signaling through the Raf MEK-ERK pathway by phosphorylating specific residues on Raf and MEK (Chaudhary et al., 2000; Frost et al., 1997; Sun et al., 2000).
Specifically, activated PAK1 can phosphorylate MEK1 at Ser298 to enhance the 30 stable association between Raf 1 and MEK1 (Frost et al., 1997). PAK1 and PAK3 can also phosphorylated Ser338 in the Raf 1 catalytic domain essential for activation of Raf 1 kinase (Chaudhary et al., 2000; King et al., 1998; Sun et al., 2000). Importantly, in these prior studies, PAK has been positioned downstream of PI-3 kinase (Chaudhary et al., 2000; Sun et al., 2000). In the current studies, Nfl +l mast cells had higher Raf 1 and MEK activity compared with wildtype cells. The present study data suggests that PAK contributes to these differences in activation for two reasons. First, phosphorylation levels of Ser 338 on Raf and Ser298 on MEKI were greatly increased in Nfl +l mast cells. Second, phosphorylation of these sites as well as the respective kinase activities of both Raf l and MEK were inhibited by a PI-3 kinase inhibitor. Though these studies cannot eliminate the possibility that other effectors may contribute to crosstalk between PI-3 kinase and the p21'as-Raf MEK-ERK pathway, the data suggest that hyperactivation of PAK could account for the hyperproliferation of mast cells. These findings are especially intriguing since placement of a dominant negative PAK has been shown to reverse the transformation of a neurofibrosarcoma cell line taken from an individual with NF1 (Tang et al., 1998). In addition, hyperactivation of Rac, the immediate upstream effector of PAK, could potentially cooperate with p21'~S in promoting cellular transformation. Further characterization of these pathways in Nfl +l and Nfl -l primary cells could provide insight into the early events leading to LOH and the subsequent malignant manifestations in NF 1.
In summary, these studies show that heterozygous inactivation of Nfl can alter multiple functions in a cell lineage which is implicated in the pathogenesis of NF 1 and can be used as a model to identify specific biochemical targets for potential therapeutic intervention. This report emphasizes the complexity of p2l~as signaling by showing that activation of the Raf MEK-ERK pathway is dependent on co-stimulatory signals generated through PI-3 kinase in a specific cellular context after growth factor stimulation. Finally, while several groups have demonstrated the importance of interactions between p21'~S and PI-3 kinase, these data offer the first evidence that alterations in this interaction can be linked to the pathogenesis of a human genetic disease.

References Alessi, D. R., and Cohen, P. , Mechanism of activation and function of protein kinase B., Curr. Opin. Genet. Dev., 8, SS-62 (1998).
Alessi, D. R., Saito, Y., Campbell, D. G., Cohen, P., Sithanandam, G., Rapp, U., Ashworth, A., Marshall, C. J., and Cowley, S. (1994). Identification of the sites in MAP kinase kinase-1 phosphorylated by p74raf 1. EMBO J. 13, 1610-1619.
Artier, S., Barbayannis, F. A., Hanser, H., Schneider, C., Stanyon, C. A., Bernard, O., and Caroni, P. (1998). Regulation of actin dynamics through phosphorylation of cofilin by LIM-kinase. Nature 393, 805-809.
Badache A, Muja N, De Vires G: Expression of Kit in neurofibromin-deficient human Schwann cells: role in Schwann cell hyperplasia associated with Type 1 Neurofibromatosis. Oncogene, 17, 795-800 (1998).
Bagrodia, S., and Cerione, R. A. (1999). Pak to the future. Trends Cell Biol 9, 350-355.
Bernex F, De Sepulveda P, Kress C, Elbaz C, Delouis C, Panthier JJ:
Spatial and temporal patterns of c-kit-expressing cells in WlacZ/+ and lacZ/WlacZ mouse embryos. Development, 122, 3023-3033 (1996).
Blouin R, Bernstein A: The white spotting and steel hereditary anemias of the mouse, in Feig S, Freedman M (eds): Clinical Disorders and Experimental ' Models of Erythropoietic failure. Boca Raton, CRC Press, 1993, p 1257-1173.
Bienenstock J, MacQueen G, Sestini P, Marshall JS, Stead RH, Perdue MH: Mast celUnerve interactions in vitro and in vivo. Am. Rev. Respir. Dis., 143, S55-58 (1991).
Bienenstock J, Perdue M, Blennerhassett M, Stead R, Kakuta N, Sestini P, Vancheri C, Marshall J: Inflammatory cells and the epithelium. Mast celUnerve interactions in the lung in vitro and in vivo. Am. Rev. Respir.
Dis., 138, S31-34 (1988).
Bollag, G., Adler, F., elMasry, N., McCabe, P. C., Conner, E., Jr., Thompson, P., McCormick, F., and Shannon, K. (1996). Biochemical characterization of a novel KRAS insertion mutation from a human leukemia. J.
Biol.Chem. 271, 32491-32494.

Bourne H, Sanders D, McCormick F: The GTPase superfamily: a conserved switch for diverse cell functions. Nature, 348, 125-132 (1990).
Bourne HR, Sanders DA, McCormick F: The GTPase superfamily:
conserved structure and molecular mechanism. Nature, 349, 117-127 (1991).
S Brodeur G: The NF1 gene in myelopoiesis and childhood myelodysplastic syndromes. N. Engl. J. Med., 330, 637-638 (1994).
Chabot B, Stephenson D, Chapman V, Besmer P, Berstein A: The proto-oncogene c-kit encoding a trans-membrane tyrosine kinase receptor maps to the mouse W locus. Nature, 335, 88-89 (1988).
Chang T, McGrae J, Hashimoto K: Ultrastructural study of two patients with both Piebaldism and Neurofibromatosis 1. Pediatric Dermatolo~y, 10, 224-234 (1993).
Chaudhary, A., King, W. G., Mattaliano, M. D., Frost, J. A., Diaz, B., Morrison, D. K., Cobb, M. H., Marshall, M. S., and Brugge, J. S. (2000).
Phosphatidylinositol 3-kinase regulates rafl through pak phosphorylation of serine 338. Curr. Biol. 10, 551-554.
Clapp DW, Freie B, Srour E, Yoder M, Fortney K, Gerson S:
Myeloproliferative sarcoma virus directed expression of beta-galactosidase following retroviral transduction of murine hematopoietic cells. Exp. Hemat., 23, 630-638 (1995a).
Clapp DW, Freie B, Lee WH, Zhang YY: Molecular evidence that in situ-transduced fetal liver hematopoietic stem/progenitor cells give rise to medullary hematopoiesis in adult rats. Blood, 86, 2113-2122 (1995b).
Clapp DW, Williams DA: The use of umbilical cord blood as a cellular source for correction of genetic diseases affecting the hematopoietic system.
Stem Cells, 13, 613-621 (1995c).
Colman SD, Williams, CA, Wallace MR: Benign neurofibromas in type 1 neurofibromatosis (NFl) show somatic deletions of the NFl gene. Nature Genetics, 11, 90-92 (1995).
Coso OA, Chiariello M, Yu JC, Teramoto H, Crespo P, Xu N, Miki T, Gutkind JS: The small GTP-binding proteins Racl and Cdc42 regulate the activity of the JNK/SAPK signaling pathway. Cell, 81, 1137-1146 (1995).

Costa JJ, Demetri GD, Harrist TJ, Dvorak AM, Hayes DF, Merica EA, Menchaca DM, Gringeri AJ, Schwartz LB, Galli SJ., Recombinant human stem cell factor (kit ligand) promotes human mast cell and melanocyte hyperplasia and functional activation in vivo, J.Exp.Med., 183, 2681-2686 (1996).
Coussens LM, Raymond WW, Bergers G, Laig-Webster M, Behrendtsen O, Werb Z, Caughey GH, Hanahan D: Inflammatory mast cells up-regulate angiogenesis during squamous epithelial carcinogenesis. Genes Dev., 13, 1382-1397 (1999).
Daniels, R. H., and Bokoch, G. M. (1999). p21-activated protein kinase:
a crucial component of morphological signaling? Trends Biochem. Sci. 24, 350-355.
DeClue, J., Papageorge, A., Fletcher, J., Diehl, S., Ratner, N., Vaas, W., and Lowry, D. (1992). Abnormal regulation of mammalian p21 ras contributes to malignant tumor growth in von Recklinghausen (Type 1) neurofibromatosis.
Cell 69, 265-273.
Duckworth BC, Cantley LC: Conditional inhibition of the mitogen-activated protein kinase cascade by wortmannin. Dependence of signal strength.
J. Biol. Chem., 272, 27665-27670 (1997).
Edwards et al., Nature Cell Biol., 1, 253 (2000).
Emanuel P, Bates, L, Castleberry R, Gualtieri R, Zuckerman K: Selective hypersensitivity to granulocyte-macrophage colony-stimulating factor by juvenile chronic myeloid leukemia hematopoietic progenitors. Blood, 77, 925-929 (1991).
Frost, J., Steen, H., Shapiro, P., Lewis, T., Ahn, N., Shaw, P., and Cobb, M. (1997). Cross-cascade activation of ERKs and ternary complex factors by Rho family proteins. EMBO J. 16, 6426-6438.
Frost, J. A., Khokhlatchev, A., Stippec, S., White, M. A., and Cobb, M.
H. (1998). Differential effects of PAK1-activating mutations reveal activity-dependent and -independent effects on cytoskeletal regulation. J. Biol.Chem.
273, 28191-28198.
Galli S, Tsai M, Wershil B: The c-kit receptor, stem cell factor, and mast cells. American Journal of Patholo~y, 142, 965-974 (1993).

Galli SJ: New concepts about the mast cell. N. Engl. J. Med., 328, 257-265 (1993).
Garrington TP, Johnson GL: Organization and regulation of mitogen-activated protein kinase signaling pathways. Curr. Olin. Cell Biol., 11, 211-S (1999).
Gille, H., and Downward, J. (1999). Multiple ras effector pathways contribute to G(1) cell cycle progression. J. Biol. Chem. 274, 22033-22040.
Gualtiera R, Emanual P, Zuckerman K, Martin G, Clark S, Shadduck R, Dracker R, Akabutu J, Nitschke R, Heterinton M, Dickerman J, Hakami N, 10 Castleberry R:, Granulocyte-Macrophage Colony-Stimulating Factor is an Endogenous Regulator of Cell Proliferation in Juvenile Chronic Myelogenous Leukemia. Blood, 74, 2360-2367 (1989).
Hall A: The Cellular Functions of Small GTP-Binding Proteins. Science, 249, 635-640 (1990).
15 Hall, A: Signal Transduction through Small GTPases-A Tale of Two GAPS. Cell, 69, 389-391 (1992).
Haneline, L. S., Broxmeyer, H. E., Cooper, S., Hangoc, G., Carreau, M., Buchwald, M., and Clapp, D. W. (1998). Multiple inhibitory cytokines induce deregulated progenitor growth and apoptosis in hematopoietic cells from Fac-/-20 mice. Blood 91, 4092-4098.
Hawley R, Lieu F, Fong A, Hawley T: Versatile retroviral vectors for potential use in gene therapy. Gene Therany, 1, 136-138 (1994).
Hill, C, Treisman R: Transcriptional regulation by extracellular signals:
Mechanisms and specificity. Cell, 80, 199-211 (1995).
25 Hirota, S., Nomura, S., Asada, H., Ito, A., Morii, E., and Kitamura, Y.
(1993). Possible involvement of c-kit Receptor and its ligand in increase of mast cells in neurofibroma tissues. Arch. Pathol.Lab. Med. 117, 996-999.
Huson SM, Upadhyaya M. Neurofibromatosis 1: Clinical management and genetic counseling. In The Neurofibromatoses: A pathogenic and clinical 30 overview. Husan SM, Hughes R.A. (eds), Chapman and Hall, London, Chapman and Hall, pp 355381 (1994).
Iemura A, Tsai M, Ando A, Wershil B, Galli S: The c-kit ligand, stem cell factor, promotes mast cell survival by suppressing apoptosis. The American Journal of Patholo~y, 144, 321-328 (1994).
Ingram, D. A., Yang, F. C., Travers, J. B., Wenning, M. J., Hiatt, K., New, S., Hood, A., Shannon, K., Williams, D. A., and Clapp, D. W. (2000).
Genetic and Biochemical Evidence that Haploinsufficiency of the Nfl Tumor Suppressor Gene Modulates Melanocyte and Mast Cell Fates In Vivo. J. Exp.
Med. 191, 181-188.
Jaakkola, S., Peltonen, J., Riccardi, V., Chu, M. L., and Uitto, J. (1989).
Type 1 neurofibromatosis: selective expression of extracellular matrix genes by Schwann cells, perineurial cells, and fibroblasts in mixed cultures. J.
Clin.Invest.
84, 253-261.
Jacks T, Shih TS, Schmitt EM, Bronson RT, Bernards A, Weinberg RA:
Tumour predisposition in mice heterozygous for a targeted mutation in NFI.
Nat Genet, 7, 353-361 (1994).
Johnson M, Look A, Declue J, Valentine M, Lowy D: Inactivation of the NF1 gene in human melanoma and neuroblastoma cell lines without impaired regulation of GTP Ras. Proc. Natl. Acad. Sci. USA, 90, 5539-5543 (1993).
Joneson T, White MA, Wigler MH, Bar-Sagi D: Stimulation of membrane ruffling and MAP kinase activation by distinct effectors of RAS.
Science, 271, 810-812 (1996).
Kerdel F, Belsito D, Scotto-Chinnici R, Soter N: Mast cell participation during the elicitation of murine allergic contact hypersensitivity. The Journal of Investigative Dermatolo~y, 88, 686-690 (1987).
Kettelhut VB, Metcalfe DD: Plasma histamine concentrations in evaluation of pediatric mastocytosis. J. Pediatr., 111, 419-421 (1987).
Khwaja, A., Rodriguez-Viciana, P., Wennstrom, S., Warne, P. H., and Downward, J. (1997). Matrix adhesion and Ras transformation both activate a phosphoinositide 3-OH kinase and protein kinase B/Akt cellular survival pathway. EMBO J. 16, 2783-2793.
Kim, C. H., Pelus, L. M., White, J. R., and Broxmeyer, H. E. (1998).
Differential chemotactic behavior of developing T cells in response to thymic chemokines. Blood 91, 4434-4443.

Kinashi, T., Katagiri, K., Watanabe, S., Vanhaesebroeck, B., Downward, J., and Takatsu, K. (2000). Distinct mechanisms of alpha5betal integrin activation by Ha-ras and R-ras. J. Biol. Chem. 275, 22590-22596.
King AJ, Sun H, Diaz B, Barnard D, Miao W, Bagrodia S, Marshall MS:
he protein kinase Pak3 positively regulates Raf 1 activity through phosphorylation of serine 338. Nature, 396, 180-183 (1998).
Kodaki, T., Woscholski, R., Hallberg, B., Rodriguez-Vicuna, P., Downward, J., and Parker, P. J. (1994). The activation of phosphatidylinositol kinase by Ras. Curr. Biol. 4, 798-806.
Knudson AJ: Hereditary cancer, oncogenes, and antioncogenes. Cancer Res., 45, 1437-1443 (1985).
Largaespada D, Brannan C, Jerkins N, Copeland N: Nfl deficiency causes Ras-mediated granulocyte/macrophage colony stimulating factor hypersensitivity and chronic myeloid leukaemia. Nat. Genet., 12, 137-143 (1996).
Leevers SJ, Paterson HF, Marshall CJ: Requirement for Ras in Raf activation is overcome by targeting Raf to the plasma membrane. Nature, 369, 411-414 (1994).
Legius E, Marchuk D, Collins F, Glover T: Somatic deletion of the neurofibromatosis type 1 gene in a neurofibrosarcoma supports a tumour suppressor gene hypothesis. Nat Genet, 3, 122-126 (1993).
Lorenz U, Bergemann AD, Steinberg HN, Flanagan JG, Li X, Galli SJ, Neel BG: Genetic analysis reveals cell type-specific regulation of receptor tyrosine kinase c-Kit by the protein tyrosine phosphatase SHP1. J. Exp. Med., 184, 1111-1126 (1996).
Lu-Kuo J, Austen K, Katz H: Post-transcriptional stabilization by interleukin-lbeta of interleukin-6 mRNA induced by c-kit ligand and interleukin-10 in mouse bone marrow-derived mast cells. The Journal of Biological Chemistry, 271, 22169-22174 (1996).
Lu-Kuo, J. M., Fruman, D. A., Joyal, D. M., Cantley, L. C., and Katz, H.
R. (2000). Impaired kit- but not FcepsilonRI-initiated mast cell activation in the absence of phosphoinositide 3-kinase p85alpha gene products. J. Biol. Chem.
275, 6022-6029.
Marshall C: Specificity of receptor tyrosine kinase signaling: Transient versus sustained extracellular signal-regulated kinase activation. Cell, 80, 185 (1995).
Marshall CJ: Ras effectors. Curr. Opin. Cell Biol., 8, 197-204 (1996).
Matsuda H, Kannan Y, Ushio H, Kiso Y, Kanemoto T, Suzuki H, Kitamura Y: Nerve growth factor induces development of connective tissue-type mast cells in vitro from murine bone marrow cells. J. Exp. Med., 174, 7-14 (1991).
Mecheri S, David B: Unravelling the mast cell dilemma: culprit or victim of its generosity. Immunology TodaX, 18, 212-215 (1997).
Meininger, C., Yano, H., Rottapel, R., Bernstein, A., Zsebo, K., and Zetter, B. (1992). The c-kit receptor ligand functions as a mast cell chemoattractant. Blood 79, 958-963.
Menon A, Anderson K, Riccardi V, Chung R, Whaley J, Yandell D, Farmer G, Freiman R, Lee J, Li F, Barker D, Ledbetter D, Kleider A, Martuza R, Gusella J, Seizinger B: Chromosome 17p deletions and p53 gene mutations associated with the formation of malignant neurofibrosarcomas in von Recklinghausen neurofibromatosis. Proc. Natl. Acad. Sci. USA, 87, 5435-5439 ( 1990).
Metcalfe, D. D., Baram, D., and Mekori, Y. A. (1997). Mast cells.
Physiol. Rev. 77, 1033-1079.
Miles D, Freedman M, Stephens K, Pallavicini, Sievers E, Leppig K, Grunberger T, Thompson P, Shannon K: Allelic loss of NF1 locus and hematopoietic lineage involvement in malignant bone marrows from children with neurofibromatosis, type 1. Blood, 86, 59 (1995).
Minden A, Lin A, Claret FX, Abo A, Karin M: Selective activation of the JNK signaling cascade and c-Jun transcriptional activity by the small GTPases Rac and Cdc42Hs. Cell, 81, 1147-1157 (1995).

Miyauchi J, Asada M, Sasaki M, Tsunematsu Y, Kojima S, Mizutani S:
Mutations of the N-ras gene in juvenile chronic myelogenous leukemia. Blood, 83, 2248-2254 (1994).
Moon, A., and Drubin, D. G. (1995). The ADF/cofilin proteins: stimulus-S responsive modulators of actin dynamics. Mol. Biol. Cell 6, 1423-1431.
Motro B, Van der Kooy D, Rossant J, Reith A, Bernstein A: Contiguous patterns of c-kit and steel expression: analysis of mutations at the W and SI
loci.
Development, 113, 1207-1221 (1991).
Nakahata T, Ogawa M: Identification in culture of a class of hemopoietic colony-forming units with extensive capability to self renew and generate multipotential hemopoietic colonies. Proc. Natl. Acad. Sci. USA, 79, 3843-3847 (1982).
Neubauer A, Shannon K, Liu E: Mutations of the ras proto-oncogenes in childhood monosomy 7. Blood, 77, 594-598 (1991).
Nocka K, Tan J, Chiu E, Chu T, Ray P, Traktman P, Besmer P:
Molecular bases of dominant negative and loss of function mutations at the murine c-kit/white spotting locus: W37, Wv, W41 and W. EMBO J, 9, 1805-1813 ( 1990).
Paulson RF, Vesely S, Siminovitch KA, Bernstein A: Signalling by the W/Kit receptor tyrosine kinase is negatively regulated in vivo by the protein tyrosine phosphatase Shpl. Nat. Genet., 13, 309-315 (1996).
Riccardi VM: Neurofibromatosis: Phenotype, natural history and pathogenesis (ed 2nd). Baltimore, The Johns Hopkins University Press, 1992.
Roberts, A. W., Kim, C., Zhen, L., Lowe; J. B., Kapur, R., Petryniak, B., Spaetti, A., Pollock, J. D., Borneo, J. B., Bradford, G. B., et al. (1999).
Deficiency of the hematopoietic cell-specific Rho family GTPase Rac2 is characterized by abnormalities in neutrophil function and host defense.
Immunity 10, 183-196.
Robinson MJ, Cobb MH: Mitogen-activated protein kinase pathways.
Curr. Opin. Cell. Biol., 9, 180-186 (1997).
Rodenhuis S: ras and human tumors. Semin Cancer Biol, 3, 241-247 (1992).

Rodriguez-Viciana, P., Warne, P., Vanhaesebroeck, B., Waterfield, M., and Downward, J. (1996). Activation of phosphoinositide 3-kinase by interaction with Ras and by point mutation. EMBO J. I S, 2442-2451.
Rodriguez-Viciana P, Warne PH, Khwaja A, Marte BM, Pappin D, Das S P, Waterfield MD, Ridley A, Downward J: Role of phosphoinositide 3-OH
kinase in cell transformation and control of the actin cytoskeleton by Ras.
Cell, 89, 457-467 (1997).
Rosenbaum T, Boissy YL, Kombrinck K, Brannan CI, Jenkins NA, Copeland NG, Ratner N: Neurofibromin-deficient fibroblasts fail to form 10 perineurium in vitro, Development, 121, 3583-3592 (1995).
Ryan JJ, Klein KA, Neuberger TJ, Leftwich JA, Westin EH, Kauma S, Fletcher JA, DeVries GH, Huff TF: Role for the stem cell factor/KIT complex in Schwann cell neoplasia and mast cell proliferation associated with neurofibromatosis. J. Neurosci. Res., 37, 415-432 (1994).
15 Satoh T, Nakafuku M, Miyajima A, Kaziro Y: Involvement of ras p21 protein in signal-transduction pathways from interleukin 2, interleukin 3, and granulocyte/macrophage colony-stimulating factor, but not from interleukin 4.
Proc. Natl. Acad. Sci. USA, 88, 3314-3318 (1991).
Satoh T, Uehara Y, Kaziro Y: Inhibition of interleukin 3 and 20 granulocyte-macrophage colony-stimulating factor stimulated increase of active ras GTP by herbimycin A, a specific inhibitor of tyrosine kinases. J. Biol.
Chem., 267, 2537-2541 (1992).
Sawada S, Florell S, Purandare S, Ota M, Stephens K, Viskochil D:
Identification of NF1 mutations in both alleles of a dermal neurofibroma. Nat.
25 Genet, 14, 110-112 (1996).
Sells, M. A., Boyd, J. T., and Chernoff, J. (1999). p21-activated kinase 1 (Pakl) regulates cell motility in mammalian fibroblasts. J. Cell Biol. 145, 849.
Serve H, Yee N, Stella G, Sepp-Lorenzino L, Tan J, Besmer P:
30 Differential roles of P 13-kinase and kit tyrosine 821 in kit receptor-mediated proliferation, survival and cell adhesion in mast cells. EMBO J, 14, 473-483 (1995).

Shannon K, Watterson J, Johnson P, O'Connell P, Lange B, Shah N, Steinherz P, Kan Y, Priest J: Monosomy 7 myeloproliferative disease in children with neurofibromatosis, type 1: epidemiology and molecular analysis. Blood, 79, 1311-1318 (1992).
S Shannon K, O'Connell P, Martin G, Paderanga D, Olson K, Dinndorf P, McCormick F: Loss of the normal NFI allele from the bone marrow of children with type 1 neurofibromatosis and malignant myeloid disorders. N. Engl. J.
Med., 330, 597-601 (1994).
Skuse G, Kosciolek B, Rowley P: Molecular genetic analysis of tumors in von Recklinghausen neurofibromatosis: loss of heterozygosity for chromosome 17. Genes Chromosomes Cancer, 1, 36-41 (1989).
Stark M, Assum G, Krone W: Single-cell PCR performed with neurofibroma Schwann cells reveals the presence of both alleles of the neurofibromatosis type 1 (NFl) gene. Hum Genet, 96, 619-623 (1995).
Stokoe D, Macdonald SG, Cadwallader K, Symons M, Hancock JF:
Activation of Raf as a result of recruitment to the plasma membrane [see comments] [published erratum appears in Science 1994 Dec 16; 266(5192), 1792-3]. Science, 264, 1463-1467 (1994).
Sun, H., King, A. J., Diaz, H. B., and Marshall, M. S. (2000). Regulation of the protein kinase Raf 1 by oncogenic Ras through phosphatidylinositol 3-kinase, Cdc42/Rac and Pak. Curr. Biol. 10, 281-284.
Tang Y, Marwaha S, Rutkowski JL, Tennekoon GI, Phillips PC, Field J:
A role for Pak protein kinases in Schwann cell transformation. Proc. Natl.
Acad.
Sci. USA, 95, 5139-5144 (1998).
Tang Y, Yu J, Field J: Signals from the Ras, Rac, and Rho GTPases converge on the Pak protein kinase in Rat-1 fibroblasts. Mol. Cell Biol., 19, 1881-1891 (1999).
Tay Y: Neurofibromatosis 1 and Piebaldism: A case report.
Dermatology, 197, 401-402 (1998).
Theriot, J. A. (1997). Accelerating on a treadmill: ADF/cofilin promotes rapid actin filament turnover in the dynamic cytoskeleton. J. Cell Biol. 136, 1165-1168.

Timokhina, L, Kissel, H., Stella, G., and Besmer, P. (1998). Kit signaling through PI 3-kinase and Src kinase pathways: an essential role for Racl and JNK
activation in mast cell proliferation. EMBO J. 17, 6250-6262.
Tsai M, Shih LS, Newlands GF, Takeishi T, Langley KE, Zsebo KM, S Miller HR, Geissler EN, Galli SJ: The rat c-kit ligand, stem cell factor, induces the development of connective tissue-type and mucosal mast cells in vivo:
Analysis by anatomical distribution, histochemistry, and protease phenotype.
J.
Exp. Med., 174, 125-131 (1991).
Vanhaesebroeck, B., Welham, M. J., Kotani, K., Stein, R., Warne, P. H., Zvelebil, M. J., Higashi, K., Volinia, S., Downward, J., and Waterfield, M. D.
(1997). P110delta, a novel phosphoinositide 3-kinase in leukocytes. Proc.
Natl.
Acad. Sci. USA 94, 4330-4335.
Vanhaesebroeck, B., Jones, G. E., Allen, W. E., Zicha, D., Hooshmand-Rad, R., Sawyer, C., Wells, C., Waterfield, M. D., and Ridley, A. J. (1999).
Distinct PI(3)Ks mediate mitogenic signalling and cell migration in macrophages. Nature Cell Biol. l, 69-71.
Viskochil, D., Buchberg, A. M., Xu, G., Cawthon, R. M., Stevens, J., Wolff, R. K., Culver, M., Carey, J. C., Copeland, N. G., and Jerkins, N. A.
(1990). Deletions and a translocation interrupt a cloned gene at the neurofibromatosis type 1 locus. Cell 62, 187-192.
Vosseller K, Stella G, Yee N, Besmer P: c-kit Receptor Signaling through Its Phosphatidylinositide-3 -Kinase-binding Site and Protein Kinase C:
Role in Mast Cell Enhancement of Degranulation, Adhesion and Membrane Ruffling. Molecular Biology of the Cell, 8, 909-922 (1997).
Wallace, M. R., Marchuk, D. A., Andersen, L. B., Letcher, R., Odeh, H.
M., Saulino, A. M., Fountain, J. W., Brereton, A., Nicholson, J., and Mitchell, A.
L. (1990). Type 1 neurofibromatosis gene: identification of a large transcript disrupted in three NF1 patients. Science 249, 181-186.
Weenstrom S, Downward J: Role of phosphoinositide 3-kinase in activation of ras and mitogen-activated protein kinase by epidermal growth factor. Mol. Cell Biol., 19, 4279-4288 (1999).

Wiestler O, Radner H: Pathology of neurofibromatosis 1 and 2: The Neurofibromatoses. London, Chapman and Hall Medical, 1994, p. 135-159.
White MA, Nicolette C, Minden A, Polverini A, Van Aelst L, Karn M, Wigler MH: Multiple Ras functions can contribute to mammalian cell transformation. Cell, 80, 533-541 (1995).
White MA, Vale T, Camonis JH, Schaefer E, Wigler MH: A role for the Ral guanine nucleotide dissociation stimulator in mediating Ras-induced transformation. J. Biol. Chem., 271, 16439-16442 (1996).
Xu G, O'Connell P, Viskochil D, Cawthon R, Robertson M, Culver M, Dunn D, Stevens J, Gesteland R., White R, Weiss R: The neurofibromatosis type 1 gene encodes a protein related to GAP. Cell, 62, 599-608 (1990).
Yablonski, D., Kane, L. P., Qian, D., and Weiss, A. (1998). A Nck-Pakl signaling module is required for T-cell receptor-mediated activation of NEAT, but not of JNK. EMBO J. 17, 5647-5657.
Yang, N., Higuchi, O., Ohashi, K., Nagata, K., Wada, A., Kangawa, K., Nishida, E., and Mizuno, K. (1998). Cofilin phosphorylation by LIM-kinase 1 and its role in Rac-mediated actin reorganization. Nature 393, 809-812.
Yee NS, Paek I, Besmer P: Role of kit-ligand in proliferation and suppression of apoptosis in mast cells: basis for radiosensitivity of white spotting and steel mutant mice. J. Exp. Med., 179, 1777-1787 (1994).
Zhang Y, Vik T, Ryder J, Srour E, Jacks T, Shannon K, Clapp D: Nfl regulates hematopoietic progenitor cell growth and Ras signaling in response to multiple cytokines. J Exp Med, 187, 1893-1902 (1998).
Zhang C, Baumgartner R, Yamada K, Beaven M: Mitogen-activated Protein (MAP) kinase regulates production of tumor necrosis factor-alpha and release of arachidonic acid in mast cells: Indications of communication between p38 and p42 MAP kinases, J. Biol. Chem. 272, 13397-13402 (1997).
All publications, patents and patent applications are incorporated herein by reference. While in the foregoing specification, this invention has been described in relation to certain preferred embodiments thereof, and many details have been set forth for purposes of illustration, it will be apparent to those skilled in the art that the invention is susceptible to additional embodiments and that certain of the details herein may be varied considerably without departing from the basic principles of the invention.

Claims (37)

WHAT IS CLAIMED IS:

1. The use of an agent that directly or indirectly inhibits at least one protein selected from the group consisting of Raf, MEK, extracellular signal regulated kinase 1 (ERK1), extracellular signal regulated kinase 2 (ERK2), phosphatidylinositol triphosphate kinase (PI3K),), p21 activated kinase 2 (PAK2), p21 activated kinase 3 (PAK3), Rac, CDC42, p21 activated kinase 1 (PAK1), protein kinase B (PKB), Akt, and c-jun N-terminal kinase (JNK), for the preparation of medicament to inhibit or treat abnormal cellular growth or pruritus associated with neurofibromatosis Type 1 (NF-1).

2. The use as claimed in claim 1 wherein the mammal is a human.

3. The use as claimed in any one of claims 1 to 2 wherein said abnormal cell growth comprises neurofibromas, lytic bone lesions, plexiform fibromas, brain tumor, myeloid malignancy, optic glioma or pheochromocytoma.

4. The use as claimed in any one of claims 1 to 3 wherein said enzyme is Raf.

5. The use as claimed in any one of claims 1 to 4 wherein the agent is a PI-3 kinase inhibitor.

6. The use as claimed in claim 5 wherein the agent is LY294002.

7. The use as claimed in any one of claims 1 to 3 wherein the agent is a PAK3, PAK2 or PAK1 inhibitor.

8. The use as claimed in any one of claims 1 to 3 wherein the agent is a MEK inhibitor.

9. The use as claimed in any one of claims 1 to 3 wherein the agent is PD98059.

10. The use as claimed in any one of claims 1 to 3 wherein the agent inhibits crosstalk between the PI3-K and the Ras-Raf MEK-ERK pathway.

11. The use as claimed in claim 10 wherein the agent is an inhibitor of MEK
or ERK.

12. The use as claimed in any one of claims 1 to 11 wherein the inhibition of abnormal cellular growth or pruritus is due to the suppression of Ras signalling cascade hyperactivation.

13. The use as claimed in any one of claims 1 to 12 wherein the agent alters the proliferation of mast cells.

14. The use as claimed in any one of claims 1 to 12 wherein the agent alters the survival of mast cells.

15. The use as claimed in any one of claims 1 to 12 wherein the agent alters mast cell chemotaxis.

16. The use as claimed in any one of claims 1 to 12 wherein the agent alters mast cell actin cytoskeletal dynamics.

17. The use as claimed in any one of claims 1 to 12 wherein the agent alters mast cell motility.

18. A method of determining the ability of an agent to inhibit abnormal mammalian cellular growth associated with NF-1 comprising detecting of determining the ability of an agent to block or inhibit SCF-induced hyperproliferation, survival, degranulation, or any combination thereof of Nfl +/- mammalian mast cells.

19. The method of claim 18 wherein the cells are derived from skin, fetal liver or from bone marrow.

20. The method of claim 18 wherein the cells are murine mast cells.

21. The method of claim 18 wherein the agent is contacted with cultured cells in vitro.

22. The method of claim 18 wherein the agent is administered to a mouse comprising Nfl +/- mast cells.

23. The method of claim 18 wherein the cells also comprise a mutation in the c-kit gene.

24. A method of determining the ability of an agent to treat abnormal cellular growth comprising detecting or determining the ability of the agent to reduce ERK2 activity in Nfl +/- mammalian cells following stimulation with SCF.

25. The method of claim 24 wherein the ERK1 phosphorylation of an ELK-1 fusion protein is detected or determined.

26. The method of claim 24 wherein the cellular growth is mast cell growth.

27. The method of claim 18 or 24 wherein the agent is a PI-3 kinase inhibitor.

28. The method of claim 18 or 24 wherein the abnormal cellular growth is characteristic of NF-1.

29. The use of an agent that alters the Ras-Raf ERK-MEK pathway in a mammal for the preparation of a medicament for the treatment or prevention of cutaneous neurofibromas associated with neurofibromatosis Type 1.

30. The use of an agent that alters the Ras-Raf ERK-MEK pathway in a mammal for the preparation of a medicament for the treatment or prevention of learning disorders associated with neurofibromatosis Type 1.

31. The use of an agent that directly or indirectly suppresses Ras signalling cascade hyperactivation for the preparation of a medicament for the treatment or prevention of indications associated with abnormal intracellular signalling and mast cell growth in a mammal.

32. The use of claim 35, wherein the ras signalling cascade is the Ras-Raf Mek-ERK pathway.

33. The use of claim 36, wherein the Ras-Raf Mek-ERK pathway is blocked.

34. The use of claim 35, wherein the ras signalling cascade is the PI3K
pathway.

35. The use of claim 38, wherein the PI3K pathway is blocked.

36. The use of claim 38, wherein the Ras signalling cascade hyperactivation comprises cross-talk between the Ras-Raf Mek-ERK and PI3K
pathways.

37. The use of claim 35, wherein the indications associated with abnormal intracellular signalling or mast cell growth comprise the development of benign and malignant tumors, bone disorders, learning disabilities, myeloid malignancies, neurofibromas, lytic bone lesions, plexiform fibromas, brain tumors, optic gliomas, and pheochromocytomas.