Analysis of Ras-Induced Overproliferation in Drosophila Hemocytes

Corresponding author: Charles R. Dearolf, Massachusetts General Hospital, 55 Fruit St., Boston, MA 02114., cdearolf{at}partners.org (E-mail)

Communicating editor: K. V. ANDERSON

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

We use the Drosophila melanogaster larval hematopoietic system as an in vivo model for the genetic and functional genomic analysis of oncogenic cell overproliferation. Ras regulates cell proliferation and differentiation in multicellular eukaryotes. To further elucidate the role of activated Ras in cell overproliferation, we generated a collagen promoter-Gal4 strain to overexpress Ras^V12 in Drosophila hemocytes. Activated Ras causes a dramatic increase in the number of circulating larval hemocytes (blood cells), which is caused by cellular overproliferation. This phenotype is mediated by the Raf/MAPK pathway. The mutant hemocytes retain the ability to phagocytose bacteria as well as to differentiate into lamellocytes. Microarray analysis of hemocytes overexpressing Ras^V12 vs. Ras⁺ identified 279 transcripts that are differentially expressed threefold or more in hemocytes expressing activated Ras. This work demonstrates that it will be feasible to combine genetic and functional genomic approaches in the Drosophila hematopoietic system to systematically identify oncogene-specific downstream targets.

THE ras genes encode highly conserved GTP-binding proteins that regulate cell growth, proliferation, and differentiation in almost all multicellular eukaryotes (reviewed in MCCORMICK 1994 ). In addition, ras genes have been the subject of intensive research because they are mutated in almost 30% of human cancers (BOS 1989 ). These cancers include solid tumors and several types of leukemia, including chronic myelomonocytic leukemia, acute myelogenous leukemia, and acute lymphoblastic leukemia (BEAUPRE and KURZROCK 1999 ). Ras proteins function by switching between an active GTP-bound state and an inactive GDP-bound state. Activated Ras proteins bind to and activate several distinct downstream effector pathways, including Raf, Ral-GDS, and PI3-kinase. Many of the oncogenic mutations lock Ras in the activated state, leading to the constitutive activation of downstream effector pathways. The mechanisms by which Ras activation leads to increased cell proliferation and impaired differentiation, as is observed in many cancers, have still not been fully elucidated.

Much of our understanding of Ras-mediated signaling comes from a combination of biochemical experiments conducted in mammalian tissue culture cells and genetic screens conducted in Drosophila and Caenorhabditis elegans (reviewed in GREENWALD and RUBIN 1992 and MCCORMICK 1994 ). In Drosophila, Ras oncogene at 85D (Ras85D, also known as Ras1 and hereafter referred to as Ras) appears to be the ortholog of H-ras, Ki-ras, and N-ras found in mammals (NEUMAN-SILBERBERG et al. 1984 ). Ras oncogene at 64B (Ras64B, also known as Ras2) is the Drosophila counterpart of mammalian R-ras.

The most intensively studied Ras-mediated signaling pathway in Drosophila has been the one that leads to the specification of the R7 photoreceptor in the developing ommatidial clusters of the compound eye (HAFEN 1991 ; VAN VACTOR et al. 1991 ; GREENWALD and RUBIN 1992 ). Since the R7 cell is specified from a group of postmitotic cells, the role of Ras in this pathway may differ from pathways where Ras promotes growth and cell proliferation. Indeed, it has been shown that the expression of activated Ras in Drosophila imaginal discs results in increased growth and cell proliferation (KARIM and RUBIN 1998 ; PROBER and EDGAR 2000 ). However, when clones of cells expressing activated Ras are induced in wing discs, they are eliminated from the intervein regions during differentiation later in development (PROBER and EDGAR 2000 ). Thus, in this aspect at least, their properties differ from mammalian tumor cells in which Ras mutations appear to sustain continued growth and proliferation. One reason for this difference may be that the rate of growth and proliferation decreases in the cells of the wing disc toward the end of larval development, and Ras activation may be toxic to the cells at that time. In contrast, Ras mutations detected in mammalian tumors often arise in populations of cells that are capable of sustained proliferation, such as the cells found in intestinal crypts (STOPERA and BIRD 1992 ; LOSI et al. 1996 ).

As an alternative, the Drosophila hematopoietic system can be utilized to study the proliferative effects of Ras and other oncogenes. In contrast to imaginal disc cells, Drosophila hemocytes appear to be capable of sustained cell proliferation, as a number of mutations can lead to the increased numbers of circulating larval hemocytes (reviewed in DEAROLF 1998 ). Drosophila hemocytes serve several biological functions, including the secretion of extracellular matrix components and a role in the cellular immune response (DEAROLF 1998 ; LANOT et al. 2001 ; VASS and NAPPI 2001 ). Larval hemocytes apparently originate from the lymph glands and circulate freely in the hemolymph. There are three main subsets of larval hemocytes:

1. The plasmatocytes constitute 95% or more of hemocytes and resemble cells of the vertebrate myeloid lineage. They contribute to host defense by engulfing microorganisms and are important during development for the phagocytosis of apoptotic cells.

2. Lamellocytes are larger cells that participate in encapsulating larger foreign invaders. They normally comprise 1–5% of hemocytes, but increase in number when induced by parasites or by abnormally developing tissues in the body cavity.

3. Finally, the crystal cells are characterized by the presence of crystalline inclusions believed to contain prophenoloxidase (RIZKI et al. 1980 , RIZKI et al. 1985 ). When activated and released, the phenol oxidase acitivity causes the melanization of the encapsulated targets.

Here we show that overexpression of activated Ras in Drosophila hemocytes results in sustained cell proliferation in a manner analogous to that observed in human leukemias. We also carry out a microarray analysis to identify changes in gene expression that result from increasing Ras activity in hemocytes. This work demonstrates the utility of the Drosophila hematopoietic system as a model for the genetic and functional genomic study of oncogenic cell overproliferation.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Generation of transgenic flies:
We generated flies that carry a 2.7-kb regulatory region located between the adjacent collagen type IV genes Cg25C and viking (YASOTHORNSRIKUL et al. 1997 ). We PCR cloned an EcoRI-BamHI fragment, using the primers GAATTCCTTCGCCCGCAGACTC and GGATCCGATGCCCTATGCACTTA, and fused it to the AUG-ß-GAL vector (THUMMEL et al. 1988 ). A BamHI-XbaI fragment encoding the lacZ gene was excised from this construct. This was replaced with a BamHI-SpeI fragment encoding the GAL4 gene excised from the pGATB vector (BRAND and PERRIMON 1993 ) to create the Cg-GAL4 construct. This construct was injected into w¹¹¹⁸ embryos to generate transgenic flies carrying P[w, CgGAL4]. Three independent lines all mapping to the second chromosome were obtained. Two of the lines tested gave identical results and were used in this study.

Genetic crosses and phenotypic analyses:
Flies were cultured on a standard medium containing corn meal, molasses, yeast, agar, and supplemented with Tegosept. When a low density of growth was desired, eggs (30–60) were collected from six pairs of parents for 4 hr in a bottle containing 25 ml of food and cultured at 25°. Under normal conditions at least 100 eggs were cultured in each bottle. To overexpress constructs in hemocytes, transgenic flies carrying the target genes under the control of GAL4 responsive elements (UAS; BRAND and PERRIMON 1993 ) were crossed to transgenic flies carrying the CgGAL4 construct. The hemocytes in the larval progeny that specifically express the target gene were then analyzed for any abnormal phenotype using phase-contrast microscopy. e16E, a GAL4 line expressed in the fat body but not in the hemocytes (HARRISON et al. 1995 ), was used as a negative control. The concentration of larval hemocytes from late third instar larvae was determined as described previously (ZINYK et al. 1993 ). The mutants used in this study include Ras1 (85D), phl (Draf) (3A1), rl (41A), hop (10B5-6), and l(3)mbn (65A6) and are described in FLYBASE 1999 .

Immunohistochemistry:
The production and specificity of monoclonal antibodies to the different hemocyte subsets of Drosophila will be described elsewhere (E. KURUCZ, P. VILMOS, I. NAGY, Y. CARTON, I. OCSOVSZKI, D. HULTMARK, E. GATEFF and I. ANDO, unpublished results). H2, the pan-hemocyte antibody, recognizes all hemocytes. Antibodies P1 and L1 recognize plasmatocytes and lamellocytes, respectively. Requests for antibodies should be addressed to I. Ando. The rabbit antiphosphohistone H3 antibody was obtained from Upstate Biotechnology (Lake Placid, NY).

Wandering third instar larvae of the appropriate genotype were carefully pierced by a sharp tungsten wire and the hemolymph was collected in a drop of PBS on a glass slide. The hemocytes were incubated for 30 min in a humidified chamber to allow them to adhere to the glass slide. The buffer was removed and the cells fixed for 6 min in acetone. The cells were briefly air dried and incubated with 2% BSA in PBS for 30 min, followed by incubation overnight with the monoclonal antibody (1:10 dilution) or the phosH3 antibody (5 µg/ml). After three washes with PBS, the cells were incubated for 2 hr at room temperature with a 1:150 dilution of either Cy3-conjugated goat anti-mouse secondary antibody (Jackson Immunoresearch Labs, West Grove, PA) or FITC-conjugated goat anti-rabbit antibody (Sigma, St. Louis). After three washes in PBS, the slides were mounted in 90% glycerol in PBS. The experiment with the phosH3 staining was repeated more than three times and a total of 564 Ras-act hemocytes and 708 wild-type control hemocytes were counted to determine the number of phosH3-staining cells.

Functional assays:
Hemocytes were tested for their ability to phagocytose in an in vitro assay. Hemocytes were seeded in 25 µl of Schneider medium on the spots of multiwell microscopic slides (SM-011, Hendley-Essex, England) and 2 µl of a 10% FITC-labeled Escherichia coli suspension were added immediately. Bacteria were added in saturating amounts, so that all the hemocytes capable of phagocytosis took up bacteria. The slides were incubated for 30 min in a humidified chamber at room temperature. After incubation, 2 µl PI solution (Sigma) was added to quench the fluorescence of nonphagocytosed bacteria and the slides were covered with a coverslip. The number of cells showing fluorescence as well as the number of phagocytosed bacteria per cell were counted. In each experiment 500 hemocytes were analyzed. In some experiments the phagocytic capacity of the cells was correlated with their immunological phenotype by using streptavidin-Cy3 (Amersham Pharmacia Biotech) in combination with biotinylated L1 and/or P1 antibodies.

We tested for increased lamellocyte differentiation induced by the parasitic wasp Leptopilina boulardi, strain G486 (RUSSO et al. 1996 ) in wild-type and mutant larvae. Approximately 40-sec instar larvae of each genotype were infested by exposure to four mature L. boulardi females for 4 hr at 25°. After 72 hr, we scored the total number of hemocytes and the number of L1⁺ cells from these infested larvae. The larvae were bled and the hemolymph was collected on a glass slide in a drop of Drosophila Ringer solution or Schneider medium. The cells were stained as described above.

Injection of hemocytes into adult flies:
We collected the hemolymph from third instar larvae in ice-cold Drosophila Ringer solution and determined the cell concentration in a Bürker chamber. Wild-type, adult virgin females were anesthetized with diethyl-ether. Cell suspensions with the appropriate hemocyte concentrations or control buffer were injected into the abdomen of adult flies using a thin glass needle. The flies were transferred to fresh food, and the number of surviving flies was determined daily. For each genotype, the experiment was repeated at least three times and a total of 50–139 wild-type adult flies were injected with hemocytes. In each experiment, at least 50 adult flies were injected with the control buffer. To retrieve the hemocytes from the injected adults, the flies were anesthetized and perfused with ice-cold Drosophila Ringer medium. Cells were collected on a glass slide and stained as described above.

Analysis of mRNA expression using oligonucleotide microarrays:
Drosophila larvae carrying either UAS-Ras1^V12/CgGAL4 or UAS-Ras1^wt/CgGAL4 were raised at 25°. The larval hemolymph from 80 larvae for those expressing Ras^V12(Ras-act) and from nearly 700 larvae for those expressing Ras^wt(Ras-wt) was carefully collected in PBS and the cells spun down and stored at -70°. Because wild-type larvae contain relatively small numbers of hemocytes, it was not possible to extract sufficient quantities of RNA for microarray analysis. Total RNA (15 µg) was extracted from the larval hemocytes using the Rneasy mini kit (QIAGEN, Chatsworth, CA). Gene expression analysis was performed using the Affymetrix Drosophila GeneChip, according to the laboratory methods described in the Genechip Expression Analysis Technical Manual (Affymetrix, 2001). Briefly, cDNA was synthesized from poly(A)⁺ RNA and cRNA was labeled with biotin during its synthesis using the BioArray high-yield transcript-labeling kit (Enzo). Fifteen micrograms of labeled, fragmented cRNA was hybridized to each array. The arrays were washed and scanned according to the manufacturer's protocol. For each genotype, three independent RNA samples were used and hybridized to microarrays. The Ras-act microarrays were then compared to the Ras-wt microarrays, giving a total of nine comparisons.

GeneChip.DAT files were analyzed using the Affymetrix MAS 5.0 software that includes the statistical algorithms for GeneChip expression data analysis. The image files are scanned using the Affymetrix GeneChip software and then scaled to the same intensity value before they are compared. Each transcript is represented by a probe set on the Drosophila GeneChip. Each probe set has 14 pairs of perfect match (PM) and mismatch (MM) oligonucleotides. The expression level for each gene is determined by calculating the average of differences in hybridization intensities between the PM and MM oligos. The expression profile of genes in Ras-act hemocytes was compared to that of Ras-wt hemocytes, which was taken as the reference. Three comparisons out of nine were excluded from the analysis due to high noise, following the Affymetrix recommendations. Fold change for each transcript was calculated using the Affymetrix MAS 5.0 software and is represented by the average signal log ratio (SLR is the log to the base 2 value of the fold change). Only those genes that are expressed threefold (SLR = 1.6) or more in Ras-act compared to those in Ras-wt are included in Table 2. A more detailed analysis is presented in Table S1 of the supplementary data at http://www.genetics.org/supplemental/, which includes all genes that are increased in Ras-act compared to those in Ras-wt (SLR of 0.05 or more), along with the P values for "Detection" of each gene and "Change" of expression of each gene in the six comparisons. The genes that are decreased in Ras-act compared to those in Ras-wt are shown in Table S2 of the supplementary data (http://www.genetics.org/supplemental/) along with the corresponding P values. The average SLR values and the corresponding standard deviation values are also included in the supplementary data.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Identification of a hemocyte promoter:
Our first goal was to identify a promoter that would enable us to preferentially express genes in the hemocytes of Drosophila. We previously showed that a DNA regulatory region located between two adjacent collagen type IV genes on the second chromosome, Cg25C and viking, can drive lacZ expression specifically in embryonic hemocytes (YASOTHORNSRIKUL et al. 1997 ). In the larva, lacZ expression is observed in the hemocytes and also in the fat body. We therefore generated CgGAL4 transgenic lines, in which the yeast GAL4 transactivator is expressed under the control of this regulatory region. In the CgGAL4 lines, GAL4 is expressed in the embryonic hemocytes from stage 13 until the end of embryonic development (data not shown). In the larva, GAL4 is expressed strongly in the circulating hemocytes (Fig 1A and Fig B), in the anterior-most pair of lobes of the lymph gland (Fig 1C and Fig D), and in the fat body (Fig 1E and Fig F). Expression of GAL4 is undetectable in lamellocytes (data not shown).

View larger version (68K): In this window In a new window Download PPT slide   Figure 1. The collagen promoter is expressed in the larval lymph glands and fat body. A, C, and E are phase-contrast images from third instar larvae carrying both the UAS-GFP and the CgGAL4 transgenes and B, D, and E are the corresponding fluorescence images showing green fluorescent protein expression. (A and B) Circulating hemocytes. (C and D) The lymph gland showing strong collagen expression in the anteriormost pair of lobes. (E and F) Fat body.

To verify the effectiveness of the CgGal4 line, we used it to overexpress a cDNA encoding a hyperactive form of the Drosophila Jak kinase, hopscotch (hop). The gain-of-function mutation hop^TumL has previously been shown to cause an overproliferation of larval hemocytes (HANRATTY and RYERSE 1981 ; HARRISON et al. 1995 ; LUO et al. 1995 ). At higher culture temperatures, hop^Tuml larvae have melanotic tumors and increased numbers of plasmatocytes and lamellocytes. Larvae carrying both the CgGAL4 and the UAS-hop^Tuml transgenes have melanotic tumors and their hemolymph contains an increased number of plasmatocytes and lamellocytes (total number of hemocytes = 14 x 10³/µl at 25°; Fig 2B). In wild-type larvae grown at 25°, we observe 2–5 x 10³ hemocytes/µl. These abnormalities are similar to those observed in hop^Tuml mutant flies. When hop^Tuml is expressed using the GAL4 driver e16E, which is expressed in the larval fat body but not in the hemocytes, the number and morphology of the hemocytes were similar to those of wild-type larvae (number of hemocytes = 2 x 10³/µl at 25°). These results indicate that the CgGal4 lines provide sufficient expression in the hemocytes. They further suggest that the phenotypic abnormalities observed in hop^Tuml mutant larvae are most likely due to a cell-autonomous effect of hop^Tuml in hemocytes.

View larger version (120K): In this window In a new window Download PPT slide   Figure 2. Expression of activated hop or Ras causes abnormalities in the circulating larval hemocytes. (A–C) Phase-contrast images of unfixed hemocytes from third instar larvae of (A) wild type, (B) CgGAL4/+ UAS-hopTuml/CgGAL4, and (C) UAS-RasV12/CgGAL4. Overexpression of both hopTuml and RasV12 causes increases in circulating hemocyte number compared to those in the control. (D) Phase-contrast image of fixed hemocytes from UAS-RasV12/CgGAL4. (E) Same field as in D, stained with H2, a marker expressed by all hemocytes and visualized by Cy3 fluorescence. (F) A merged image of hemocytes stained with phosH3 visualized by FITC (green) and H2 visualized by Cy3 (red).

Activated Ras causes an overproliferation of hemocytes:
Ras has been implicated in regulating cell proliferation and differentiation in many organisms. To examine the effects of increasing Ras activity in hemocytes, we expressed an activated form of Ras, Ras^V12, using the CgGAL4 line. Late third instar larvae carrying both the UAS-Ras^V12 and the CgGAL4 transgenes (hereafter referred to as Ras-act) display a dramatic 40-fold increase in the number of hemocytes (Fig 2C and Fig 4). These cells express H2, an antigen that is expressed by all hemocytes (Fig 2D and Fig E). These cells morphologically resemble plasmatocytes, although they do not stain with the plasmatocyte-specific antigen P1 (data not shown). Under normal culture conditions at 25°, 99% of these animals die as early pupae. However, 33% of these larvae survive to adults when larvae are cultured at very low densities (see MATERIALS AND METHODS). Consistent with earlier data, (KARIM and RUBIN 1998 ), we found that the phenotypes produced by expressing UAS-Ras^V12 are temperature sensitive. When raised at 18°, the Ras-act larvae display a 20-fold higher number of hemocytes than do wild-type larvae and develop into adults (data not shown). We do not observe an increased number of hemocytes in Ras-act embryos. Larvae carrying UAS-Ras^wt and CgGAL4 (referred to as Ras-wt) display a milder increase in the number of hemocytes (18 x 10³/µl at 25°) and develop into adults. In a control experiment, we used e16E as the GAL4 driver (HARRISON et al. 1995 ) to express Ras^V12 in the fat body but not in the hemocytes (e16E also shows expression in the imaginal discs, gut, and cuticle). This leads to pupal lethality but has no effect on the number or morphology of circulating hemocytes (data not shown). Therefore the increased number of hemocytes in Ras-act larvae probably does not cause the observed lethality.

View larger version (32K): In this window In a new window Download PPT slide   Figure 3. Injection of RasV12 hemocytes into wild-type adult females reduces survival. Survival curves of female flies injected with either control buffer or 1000 hemocytes each from OreR, RasV12, or l(3)mbn larvae. A total of 50, 131, or 139 wild-type female flies were injected with hemocytes from OreR, RasV12, or l(3)mbn, respectively. At least 200 wild-type flies were injected with the control buffer. Hemocytes from l(3)mbn or RasV12 larvae reduce the survival of wild-type hosts compared to buffer or hemocytes from OreR larvae.

View larger version (32K): In this window In a new window Download PPT slide   Figure 4. Large increase in hemocyte number caused by hyperactivation of the Ras-Raf-MAPK pathway. The hemolymphs from at least eight different third instar larvae were examined for each genotype, and the average number of circulating hemocytes are presented with the standard deviation. The expression of RasV12 or DRafgof in hemocytes using CgGAL4 causes an 40-fold increase in the number of circulating hemocytes. Co-expression of p21, the CDK inhibitor, with RasV12 causes a significant decrease in hemocyte number compared with using RasV12 alone. Expression of p35, which blocks caspase-mediated cell death, does not increase hemocyte cell number relative to controls. Ras double mutations that disrupt interactions with Raf, as well as reduction of rl MAPK gene dosage, suppress the RasV12-induced proliferation. Student t-tests show a significant difference in hemocyte numbers (P < 0.01) between UAS-RasV12/CgGAL4 and all other genotypes shown except for UAS-DRafgof/CgGAL4 and UAS-RasV12,S35/CgGAL4.

We next examined whether the large increase in Ras-act larval hemocyte numbers is due to an increase in the proliferation of hemocytes or to a block in cell death. We obtained several lines of evidence to suggest that the mutant cells do indeed overproliferate. First, we cultured the Ras-act hemocytes in the abdomen of wild-type adult flies. Previous studies have shown that cells from tumorous lymph glands can be serially passaged in the abdomen of adult flies (HANRATTY and RYERSE 1981 ). Following injection with Ras-act hemocytes, the abdomen of the injected flies became enlarged, and 64% of the injected flies died within 3 days (Fig 3). These changes are associated with the overproliferation of cells within the host (HANRATTY and RYERSE 1981 ). A similar result was obtained when hemocytes from lethal(3) malignant blood neoplasm-1 [l(3)mbn-1] mutant larvae were injected into the abdomen of adult flies (Fig 3). The l(3)mbn mutant is another example of a mutant that has increased numbers of circulating hemocytes (GATEFF 1978 ; KONRAD et al. 1994 ). In control animals injected with the same number of wild-type hemocytes, only 24% of the host animals died (Fig 3) and the abdomen of the injected flies showed no change in size. In control animals injected with the injection buffer, only 10% of the host animals died (Fig 3). The Ras-act hemocytes can be repeatedly passaged in the abdomen of adult flies (data not shown), and during this process they retain their original size and morphology and continue to express the pan-hemocyte marker H2. These results suggest that the mutant hemocytes proliferate in the adult abdomen and cause the death of the animal.

Second, we stained circulating larval hemocytes with anti-phosH3, a marker that stains chromosomes in cells undergoing mitosis. In a population of proliferating cells, only a small proportion of cells undergoes mitosis and stains with phosH3 at any given time (HENDZEL et al. 1997 ). We found that 0.32% (±0.49) of hemocytes stain with phosH3 in control larvae. QIU et al. 1998 obtained a comparable number of phosH3-staining hemocytes from wild-type larvae. In contrast, a significantly larger proportion (P < 0.01) of Ras-act hemocytes (3.93% ± 1.27) stained with this antibody (Fig 2E and Fig F), indicating that Ras-act hemocytes actively divide while in circulation.

Third, the cyclin-dependent kinase inhibitor, p21, has previously been shown to block the entry of cells into S-phase in Drosophila (DE NOOIJ and HARIHARAN 1995 ). If Ras^V12 induced an overproliferation of hemocytes, expression of p21 would be expected to block their division. We found that the increase in Ras^V12 hemocyte numbers produced is indeed largely suppressed by coexpression of p21 in hemocytes (Fig 4), suggesting that Ras^V12 promotes cell proliferation.

Fourth, the expression of the caspase inhibitor p35 in hemocytes does not cause an appreciable increase in hemocyte number. p35 has previously been shown to block apoptotic cell death in Drosophila (HAY et al. 1994 ). If the Ras^V12 phenotype is caused primarily by a block in cell death, we would expect a suppression of cell death in wild-type larval hemocytes to cause an increase in their number. Our results suggest that the Ras^V12 phenotype is not related to alterations in cell death.

Functional tests for Ras^V12-expressing hemocytes:
We next determined whether the Ras-act hemocytes were capable of performing any of the functions associated with wild-type hemocytes, including the phagocytosis of bacteria and the accumulation of lamellocytes in response to an immune challenge. We performed an in vitro assay for phagocytosis, in which 500 hemocytes of each genotype were examined for their ability to engulf FITC-labeled E. coli. The control OreR plasmatocytes engulfed an average of 10 bacteria/cell, while the Ras-act counterparts engulfed an average of 5 bacteria/cell. Therefore, the Ras-act hemocytes are capable of phagocytosing bacteria, but are slightly less proficient than wild-type hemocytes.

We also observe cells that morphologically resemble lamellocytes in the Ras-act larvae. They range in number from 1 to 3.5% and express L1, the lamellocyte-specific marker (data not shown). If wild-type larvae are immuno-challenged by L. boulardi, a parasitic wasp that lays its eggs inside the Drosophila larvae, the lamellocytes increase in number and function in encapsulating the wasp eggs (RUSSO et al. 1996 ). To test whether the Ras-act larvae are capable of an increase in lamellocyte differentiation upon immune challenge, we allowed them to be parasitized by L. boulardi. In two independent experiments, the percentage of L1-positive cells increased in the wasp-infested Ras-act larvae compared to that in unchallenged Ras-act larvae (Table 1). This result indicates that lamellocyte differentiation can occur in Ras-act larvae.

The Ras^V12 overproliferation phenotype is mediated by the Raf-MAPK pathway:
The proliferative effect of Ras could be mediated via the Raf-mitogen-activated protein kinase (MAPK) pathway or could involve other effectors distinct from Raf. We therefore tested the effect of overexpressing a gain-of-function allele of Drosophila Raf (DRaf^gof) in the larval hemocytes. The expression of DRaf^gof in hemocytes also results in a massive increase in hemocyte number (Fig 4), similar to that seen with the expression of Ras^V12 (Fig 4). This indicates that Raf activation alone is sufficient to induce hemocyte proliferation.

We also looked at the effects of reducing the function of genes located downstream of Ras in this pathway. Ras-act larvae that were hemizygous for a DRaf loss-of-function allele (DRaf¹¹²⁹) had fewer hemocytes than wild-type larvae did (data not shown). This is consistent with previous observations that a complete loss of Raf function causes an almost total loss of hemocytes, most likely due to a decrease in cell viability (LUO et al. 2002 ). Thus, expression of activated Ras cannot bypass this requirement for Raf. Ras-act larvae that were heterozygous for a loss-of-function mutation in the rl gene that encodes MAPK (BIGGS et al. 1994 ) had a threefold decrease in hemocyte count compared to that of Ras-act (Fig 4). A complete lack of rl function (rl^R26/rl^R26) in Ras-act larvae results in very few hemocytes, similar to that seen in larvae homozygous for rl^R26 or in larvae lacking Raf function (data not shown). These results indicate that MAPK function is essential for Ras-mediated signaling in cell proliferation.

Activated Ras has been shown to interact directly with other effectors in addition to Raf, including PI3-kinase and Ral.GDS (KATZ and MCCORMICK 1997 ). To test for the involvement of these effectors, we took advantage of three Ras effector loop mutants previously used in mammalian cell culture experiments (WHITE et al. 1995 ; JONESON et al. 1996 ; KHOSRAVI-FAR et al. 1996 ) and in Drosophila wing development (KARIM and RUBIN 1998 ). In addition to the V12 mutation, these effector loop mutants carry a second amino acid substitution in the Ras effector domain. The mutant Ras^{V12 C40} interacts with PI3-kinase but fails to interact with Raf or Ral.GDS. Ras^{V12 G37} can bind to Ral.GDS but does not interact with Raf or PI3-kinase, and Ras^{V12 S35} interacts with Raf but fails to bind Ral.GDS or PI3-kinase. The expression of the first two mutant forms of Ras (Ras^{V12 C40} and Ras^{V12 G37}) in hemocytes results in a significant reduction in their numbers (P < 0.01) compared with the massive hemocyte numbers seen with the expression of Ras^V12 alone (Fig 4). These two mutants also partially suppress the lethality associated with the expression of Ras^V12 in hemocytes. Since both these mutants fail to interact with Raf, these results indicate that Raf plays an important role in Ras1-mediated signaling. In contrast, the expression of Ras^{V12 S35}, a mutant that interacts with Raf but fails to bind Ral.GDS or PI3-kinase, displays an overproliferation of hemocytes similar to that seen in Ras-act larvae (Fig 4). This result shows that an impaired ability of Ras to interact with PI3-kinase or Ral.GDS does not obviously affect Ras-induced proliferation. However, the number of hemocytes observed in the first two mutants (i.e., those that do not activate Raf) is still increased compared to that in wild type. This result suggests that the effector loop mutants may retain some residual activity or that other effectors may also contribute in a minor way to the Ras-induced overproliferation.

Microarray analysis of Ras-expressing hemocytes:
To identify those genes whose expression correlates with oncogenic Ras-induced cell overproliferation, we examined expression profiles of genes using oligonucleotide microarrays. There is a 10-fold increase in the number of Ras-act hemocytes (192 x 10³/µl) compared with that of Ras-wt hemocytes (18 x 10³/µl). Therefore, a comparison of the patterns of gene expression between these two populations of hemocytes is likely to identify those genes whose expression is altered by increasing Ras activity.

Using Affymetrix Drosophila oligonucleotide microarrays, we compared the expression profiles of >13,000 genes between the two populations of hemocytes. Expression of 6123 transcripts was detected as present in Ras-act and 4644 transcripts as present in Ras-wt hemocytes. In a comparative analysis of Ras-act to Ras-wt hemocytes (see MATERIALS AND METHODS) there is at least a threefold increase in the expression of 279 genes (Table 2 and supplementary information at http://www.genetics.org/supplemental/). The expression of 76 genes is decreased threefold or more in Ras-act compared with that in Ras-wt. These genes are not discussed further in this article, but are included as supplementary information at http://www.genetics.org/supplemental/.

Among the genes that are most strongly induced in Ras-act hemocytes are CG16731, a putative prophenoloxidase activator (128-fold), string (27-fold), and Cks (25-fold; Table 2). Crystal cells contain the enzymes and substrates required for melanization (RIZKI et al. 1985 ). Prophenoloxidase, the inactive precursor of phenol oxidase, is activated in the crystal cells by a series of serine proteases, the last of which is a prophenoloxidase activator. string encodes a phosphatase that activates the mitotic kinase, Cdk1 (Cdc2), and promotes the G₂/M transition in the cell cycle (EDGAR et al. 1994 ). The role of Ras in activating string transcription has not previously been documented. Cks encodes a protein that associates with cyclin/cdk complexes, but whose precise function is not known (SPRUCK et al. 2001 ).

Among the 279 genes that had a >3-fold increase in expression in Ras-act, 17 genes have functions related to cell cycle regulation and 16 genes function in DNA replication or mitosis (Table 2). The increased expression of some of these genes may be directly related to the increased proliferative capacity of Ras-act hemocytes. Interestingly, the genes upregulated include both positive and negative regulators of cell cycle progression. Among the positive regulators that were induced are string (27-fold), cyclin E (6.8-fold), cdc2 (7.8-fold), and cyclin B3 (4.7-fold). Negative regulators of cell proliferation upregulated include dacapo (4-fold) and wee1 (5-fold). Genes induced include those that regulate the G₁/S transition (e.g., cyclin E, dacapo) as well as those that regulate G₂/M (string, wee1, cyclin B3). Expression of cyclin B (2-fold), cyclin A (2.9-fold), cdk2 (1.9-fold), and cdk4/6 (2.4-fold) is also increased (supplementary information at http://www.genetics.org/supplemental/). Expression of cyclin D, a cyclin that has been shown to promote growth in some situations (DATAR et al. 2000 ), is not increased.

Among the genes known to regulate hemocyte fates in Drosophila, expression of lozenge (lz) is increased 5.7-fold in Ras-act hemocytes (Table 3). lz expression is necessary for crystal cell fate specification and lz mutants lack crystal cells (RIZKI et al. 1985 ; LEBESTKY et al. 2000 ). Although lz is induced in Ras-act hemocytes, we do not observe a significant increase in morphologically recognizable crystal cells in Ras-act larvae. However, these cells appear to express much higher levels of both a putative prophenoloxidase activator (128-fold) and Black cells (Bc; 3.6-fold), which encodes monophenol oxidase (Table 2 and Table 3). Both these enzymes are believed to function in crystal cells (RIZKI et al. 1985 ). Thus, increased Ras activity may increase the expression of a subset of crystal-cell-specific genes. The transcription factor, serpent (srp), is essential for the development of all embryonic hemocytes (REHORN et al. 1996 ). srp expression is increased in Ras-act hemocytes by 2-fold (Table 3). The transcription factor, gcm, has been shown to be necessary for plasmatocyte fate specification (BERNARDONI et al. 1997 ). Expression of gcm is not detected in either Ras-act or Ras-wt hemocytes (Table 3). This may correlate with our finding that these cells do not express the plasmatocyte-specific marker, P1. However, peroxidasin (Pxn) and croquemort (crq), genes that are expressed in normal plasmatocytes (NELSON et al. 1994 ; FRANC et al. 1996 ), are expressed to similar levels in both Ras-wt and Ras-act hemocytes (Table 3). Thus, increasing Ras activity does not cause a general increase in the expression of genes normally associated with plasmatocytes.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Ras has been well studied as a mammalian oncogene. Mutations that activate Ras represent one of the steps in the formation of many types of human cancers (BOS 1989 ). Activation of Ras alone is rarely oncogenic. Thus the precise contribution of Ras to the multi-step transformation process and the way in which Ras interacts with other oncogenic events is still not well understood. We have shown that overexpression of wild-type Ras or activated Ras alone can induce hemocyte overproliferation in Drosophila. This one-step model of Ras-induced leukemia may thus represent a valuable system to characterize the link between increased Ras activity and uncontrolled cell proliferation.

Previous studies by others have shown that activated Ras can induce growth and cell proliferation in Drosophila (KARIM and RUBIN 1998 ; PROBER and EDGAR 2000 ). Expression of activated Ras in Drosophila imaginal discs results in disc overgrowth. Analysis of patterns of BrdU incorporation and apoptosis showed that some of the consequences of Ras overexpression were the result of nonautonomous mechanisms (KARIM and RUBIN 1998 ). These nonautonomous mechanisms have not yet been elucidated. Expression of activated Ras in clones of cells in the wing imaginal discs showed that increased Ras activity results in increased cell growth (mass accumulation). Increased Ras activity also results in cell death in the intervein regions of the wing when cells have to undergo terminal differentiation (PROBER and EDGAR 2000 ). In contrast, the hemocyte overproliferation that we describe generates a more uniform cell population, most likely resulting from a cell-autonomous effect of Ras in the hemocytes. Therefore, Drosophila hemocytes might represent a simpler system to elucidate the molecular mechanisms underlying Ras-induced neoplastic cell overproliferation.

Ras can induce an overproliferation of functional yet abnormal hemocytes:
We have shown that the overexpression of activated Ras in Drosophila hemocytes results in a leukemia-like overproliferation. These circulating hemocytes appear to be normal by several functional criteria. The majority of the overproliferating hemocytes resemble wild-type plasmatocytes in their morphology and are capable of phagocytosis. Moreover, lamellocytes accumulate when Ras-act larvae are parasitized in the wasp assay. The Ras-act hemocytes also express several markers associated with normal plasmatocytes, including peroxidasin and croquemort.

At the same time, the mutant hemocytes differ from their wild-type counterparts in their expression patterns of known plasmatocyte markers. The mutant hemocytes do not express the epitope, recognized by the P1 monoclonal antibody that is found on all mature plasmatocytes (E. KURUCZ, P. VILMOS, I. NAGY, Y. CARTON, I. OCSOVSZKI, D. HULTMARK, E. GATEFF and I. ANDO, unpublished results). Further, microarray analysis demonstrated that these cells do not express gcm, a transcription factor that is expressed in normal plasmatocytes (BERNARDONI et al. 1997 ). These cells also appear to express several transcripts that are normally expressed in crystal cells, including the transcription factor lozenge, the CG16731 gene encoding a prophenoloxidase activator, and Bc encoding a monophenol monooxygenase. Although we did not observe an increase in morphologically recognizable crystal cells in Ras-act larvae, when these larvae are subject to a heat assay (RIZKI et al. 1985 ; LEBESTKY et al. 2000 ), we find a considerable increase in the number of black cells (data not shown). These findings can be explained in several ways. The cells may represent multi-potential stem or precursor cells that are capable of giving rise to both plasmatocytes and crystal cells. Alternatively, these cells may not correspond to any population of cells found during normal hematopoiesis, as increasing Ras activity beyond physiological levels may result in the expression of inappropriate combinations of genes. The latter possibility is supported by the observation that wild-type plasmatocytes co-express croquemort and gcm (BERNARDONI et al. 1997 ), unlike Ras-act hemocytes.

Our data indicate that activated Ras induces hemocyte overproliferation in Drosophila through a Raf/MAPK pathway. The Ras-induced overproliferation of hemocytes can be suppressed by mutations that reduce signaling via D-Raf or by reduced activity of the rl MAP kinase. Further, the overexpression of an activated Raf construct resulted in a large increase in the concentration of circulating hemocytes (Fig 4).

The Ras-induced overproliferation can also be suppressed by overexpression of the human cdk inhibitor p21. The Drosophila ortholog of p21, dacapo, has been shown to bind and inactivate cyclin E/cdk2 complexes (DE NOOIJ et al. 1996 ; LANE et al. 1996 ). Thus the Ras-induced overproliferation appears to require active cyclin E/cdk2 complexes and cannot bypass their function. Moreover, the levels of cyclin E and string RNA, both of which have been shown to be induced by E2F (NEUFELD et al. 1998 ), are elevated in Ras-act hemocytes. Thus Ras-induced hemocyte overproliferation may occur via increased E2F activity.

Changes in gene expression induced by Ras:
By comparing the RNA samples from Ras-act and Ras-wt hemocytes, we have been able to assess the transcriptional changes over the entire predicted genome that result from an increase in Ras activity in a specific lineage. To our knowledge, this is the first description of a whole-genome, in vivo analysis of Ras-mediated changes in transcription. Other studies have examined the transcriptional changes resulting from an increase in Raf activity in a human breast epithelial cell line, although only a portion of the human genome was analyzed (SCHULZE et al. 2001 ).

One overall finding is that many of the genes that are upregulated in Ras-act cells include genes that function in cell cycle regulation and DNA replication. These genes include both positive and negative regulators of cell proliferation. The cyclin-dependent kinase inhibitor dacapo (which antagonizes the function of cyclin E/cdk2 complexes), as well as the wee1 kinase (which inactivates cdc2), are both induced. There is currently no known function for either gene in promoting cell cycle progression. Thus the induction of these genes may represent a negative feedback mechanism that attempts to reduce cell proliferation under conditions of excessive cell proliferation. Another possibility is that these two genes have roles in promoting cell cycle progression that are currently unknown. Our microarray data also show that regulators that promote all stages of cell cycle progression are induced, not only those that promote the G₁/S transition. Our data therefore suggest that both the G₁/S and G₂/M cell cycle transitions may be influenced by an increase in Ras activity.

A second finding is that many of the transcriptional targets known to be induced by Ras1 in other tissues are not induced in Ras-act hemocytes. Therefore, although the RTK/Ras pathway induces the expression of phyllopod (CHANG et al. 1995 ; DICKSON et al. 1995 ; TANG et al. 1997 ) in the eye disc, mirror in the ovary (JORDAN et al. 2000 ), and blistered and ribbon in the tracheal cells (NUSSBAUMER et al. 2000 ; BRADLEY and ANDREW 2001 ), none of these genes are obviously induced in Ras-act hemocytes. This is consistent with tissue-specific factors acting together with Ras to determine which target genes are expressed in each cell type. Another gene whose expression is modulated by Ras activity is the pro-apoptotic gene, hid (also known as Wrinkled). It is believed that the anti-apoptotic effect of Ras in embryos is mediated in part by a reduction in hid transcription (KURADA and WHITE 1998 ). Our analysis demonstrates that the hid RNA level does not decrease in Ras-act cells, indicating that this mechanism may not be of importance in hemocytes. Ras may still inactivate hid in these cells via MAPK-mediated phosphorylation of the Hid protein. Other pro-apoptotic genes like reaper or grim are not expressed in either Ras-wt or Ras-act hemocytes.

Finally, our data indicate that the large overproliferation of hemocytes in response to activated Ras does not lead to a general activation of the immune response. Among the 134 Drosophila immune-regulated genes induced by septic injury and fungal infection (DE GREGORIO et al. 2001 ), only 6 genes are upregulated and 4 genes are downregulated by a factor of 3 or more in Ras-act hemocytes. The 6 upregulated genes in Ras-act are Tep2 (complement like), -2M receptor like (complement binding), a trypsin-like serine protease (phenol oxidase cascade), a serpin (serine protease inhibitor), spz (antifungal response), and Tl (antifungal response). The 4 downregulated genes include Tep1 (complement like), Rel (transcription factor), Metchnikowin (antimicrobial response), and a lipase.

Concluding remarks:
Activated versions of both Ras and the Hop Jak kinase induce leukemia-like phenotypes in Drosophila larvae. Further, it is possible to isolate sufficient quantities of larval hemocytes to conduct microarray expression studies. By comparing the expression profiles from different oncogene-induced leukemia cells, coupled with mutational analysis of the newly identified targets, it should be possible to systematically characterize the critical, oncogene-specific target genes. This approach could prove beneficial to the treatment of human cancers.

	ACKNOWLEDGMENTS

We are grateful to F. Karim for the UAS-Ras1 transgenic flies, D. Harrison for UAS-hop transgenic flies, the Indiana University Stock Center for providing fly stocks, and to Y. Carton for generously providing the wasps. We thank J. Couget, S. Meng, and T. Aldredge of the Center for Genome Research at Harvard University for help with the Affymetrix microarray analysis. We thank O. Stevaux, J. Walker, and D. Dimova for their generous help with the microarray data analysis, B. Carney and S. Conto for technical help, and K. Baksa for comments on the manuscript. We are grateful to I. Hariharan for helpful discussions and for critically evaluating the manuscript. H.A. was a recipient of an NRSA fellowship from the National Institutes of Health (NIH). Work in the Dearolf lab is supported by the NIH. The Ando lab was supported by grants from the Hungarian National Science Foundation (OTKA grants T035249 and T035074).

Manuscript received June 4, 2002; Accepted for publication October 17, 2002.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

BEAUPRE, D. M. and R. KURZROCK, 1999  RAS and leukemia: from basic mechanisms to gene-directed therapy. J. Clin. Oncol. 17:1071-1079.[Abstract/Free Full Text]

BERNARDONI, R., B. VIVANCOS, and A. GIANGRANDE, 1997  glide/gcm is expressed and required in the scavenger cell lineage. Dev. Biol. 191:118-130.[Medline]

BIGGS, W. H., III, K. H. ZAVITZ, B. DICKSON, A. VAN DER STRATEN, and D. BRUNNER et al., 1994  The Drosophila rolled locus encodes a MAP kinase required in the sevenless signal transduction pathway. EMBO J. 13:1628-1635.[Medline]

BOS, J. L., 1989  ras oncogenes in human cancer: a review. Cancer Res. 49:4682-4689.[Abstract/Free Full Text]

BRADLEY, P. L. and D. J. ANDREW, 2001  ribbon encodes a novel BTB/POZ protein required for directed cell migration in Drosophila. Development 128:3001-3015.[Abstract/Free Full Text]

BRAND, A. H. and N. PERRIMON, 1993  Targeted gene expression as a means of altering cell fates and generating dominant phenotypes. Development 118:401-415.[Abstract]

CHANG, H. C., N. M. SOLOMON, D. A. WASSARMAN, F. D. KARIM, and M. THERRIEN et al., 1995  phyllopod functions in the fate determination of a subset of photoreceptors in Drosophila. Cell 80:463-472.[Medline]

DATAR, S. A., H. W. JACOBS, A. F. DE LA CRUZ, C. F. LEHNER, and B. A. EDGAR, 2000  The Drosophila cyclin D-Cdk4 complex promotes cellular growth. EMBO J. 19:4543-4554.[Medline]

DEAROLF, C. R., 1998  Fruit fly "leukemia.". Biochim. Biophys. Acta 1377:M13-M23.[Medline]

DE GREGORIO, E., P. T. SPELLMAN, G. M. RUBIN, and B. LEMAITRE, 2001  Genome-wide analysis of the Drosophila immune response by using oligonucleotide microarrays. Proc. Natl. Acad. Sci. USA 98:12590-12595.[Abstract/Free Full Text]

DE NOOIJ, J. C. and I. K. HARIHARAN, 1995  Uncoupling cell fate determination from patterned cell division in the Drosophila eye. Science 270:983-985.[Abstract/Free Full Text]

DE NOOIJ, J. C., M. A. LETENDRE, and I. K. HARIHARAN, 1996  A cyclin-dependent kinase inhibitor, Dacapo, is necessary for timely exit from the cell cycle during Drosophila embryogenesis. Cell 87:1237-1247.[Medline]

DICKSON, B. J., M. DOMINGUEZ, A. VAN DER STRATEN, and E. HAFEN, 1995  Control of Drosophila photoreceptor cell fates by phyllopod, a novel nuclear protein acting downstream of the Raf kinase. Cell 80:453-462.[Medline]

EDGAR, B. A., D. A. LEHMAN, and P. H. O'FARRELL, 1994  Transcriptional regulation of string (cdc25): a link between developmental programming and the cell cycle. Development 120:3131-3143.[Abstract]

FLYBASE,, 1999  The FlyBase database of the Drosophila genome projects and community literature. Nucleic Acids Res. 27:85-88. http://flybase.bio.indiana.edu/(.[Abstract/Free Full Text]

FRANC, N. C., J. L. DIMARCQ, M. LAGUEUX, J. HOFFMANN, and R. A. EZEKOWITZ, 1996  Croquemort, a novel Drosophila hemocyte/macrophage receptor that recognizes apoptotic cells. Immunity 4:431-443.[Medline]

GATEFF, E., 1978  Malignant neoplasms of genetic origin in Drosophila melanogaster. Science 200:1448-1459.[Abstract/Free Full Text]

GREENWALD, I. and G. M. RUBIN, 1992  Making a difference: the role of cell-cell interactions in establishing separate identities for equivalent cells. Cell 68:271-281.[Medline]

HAFEN, E., 1991  Patterning by cell recruitment in the Drosophila eye. Curr. Opin. Genet. Dev. 1:268-274.[Medline]

HANRATTY, W. P. and J. S. RYERSE, 1981  A genetic melanotic neoplasm of Drosophila melanogaster. Dev. Biol. 83:238-249.[Medline]

HARRISON, D. A., R. BINARI, T. S. NAHREINI, M. GILMAN, and N. PERRIMON, 1995  Activation of a Drosophila Janus kinase (JAK) causes hematopoietic neoplasia and developmental defects. EMBO J. 14:2857-2865.[Medline]

HAY, B. A., T. WOLFF, and G. M. RUBIN, 1994  Expression of baculovirus P35 prevents cell death in Drosophila. Development 120:2121-2129.[Abstract]

HENDZEL, M. J., Y. WEI, M. A. MANCINI, A. VAN HOOSER, and T. RANALLI et al., 1997  Mitosis-specific phosphorylation of histone H3 initiates primarily within pericentromeric heterochromatin during G2 and spreads in an ordered fashion coincident with mitotic chromosome condensation. Chromosoma 106:348-360.[Medline]

JONESON, T., M. A. WHITE, M. H. WIGLER, and D. BAR-SAGI, 1996  Stimulation of membrane ruffling and MAP kinase activation by distinct effectors of RAS. Science 271:810-812.[Abstract]

JORDAN, K. C., N. J. CLEGG, J. A. BLASI, A. M. MORIMOTO, and J. SEN et al., 2000  The homeobox gene mirror links EGF signalling to embryonic dorso-ventral axis formation through notch activation. Nat. Genet. 4:429-433.

KARIM, F. D. and G. M. RUBIN, 1998  Ectopic expression of activated Ras1 induces hyperplastic growth and increased cell death in Drosophila imaginal tissues. Development 125:1-9.[Abstract]

KATZ, M. E. and F. MCCORMICK, 1997  Signal transduction from multiple Ras effectors. Curr. Opin. Genet. Dev. 7:75-79.[Medline]

KHOSRAVI-FAR, R., M. A. WHITE, J. K. WESTWICK, P. A. SOLSKI, and M. CHRZANOWSKA-WODNICKA et al., 1996  Oncogenic Ras activation of Raf/mitogen-activated protein kinase-independent pathways is sufficient to cause tumorigenic transformation. Mol. Cell. Biol. 16:3923-3933.[Abstract]

KONRAD, L., G. BECKER, A. SCHMIDT, T. KLOCKNER, and G. KAUFER-STILLGER et al., 1994  Cloning, structure, cellular localization, and possible function of the tumor suppressor gene lethal(3)malignant blood neoplasm-1 of Drosophila melanogaster. Dev. Biol. 163:98-111.[Medline]

KURADA, P. and K. WHITE, 1998  Ras promotes cell survival in Drosophila by downregulating hid expression. Cell 95:319-329.[Medline]

LANE, M. E., K. SAUER, K. WALLACE, Y. N. JAN, and C. F. LEHNER, 1996  Dacapo, a cyclin-dependent kinase inhibitor, stops cell proliferation during Drosophila development. Cell 87:1225-1235.[Medline]

LANOT, R., D. ZACHARY, F. HOLDER, and M. MEISTER, 2001  Postembryonic hematopoiesis in Drosophila. Dev. Biol. 230:243-257.[Medline]

LEBESTKY, T., T. CHANG, V. HARTENSTEIN, and U. BANERJEE, 2000  Specification of Drosophila hematopoietic lineage by conserved transcription factors. Science 288:146-149.[Abstract/Free Full Text]

LOSI, L., L. RONCUCCI, C. DI GREGORIO, M. P. DE LEON, and J. BENHATTAR, 1996  K-ras and p53 mutations in human colorectal aberrant crypt foci. J. Pathol. 178:259-263.[Medline]

LUO, H., W. P. HANRATTY, and C. R. DEAROLF, 1995  An amino acid substitution in the Drosophila hopTum-l Jak kinase causes leukemia-like hematopoietic defects. EMBO J. 14:1412-1420.[Medline]

LUO, H., E. ROSE, M. ROBERTS, and R. DEAROLF, 2002  The Hopscotch Jak kinase requires the Raf pathway to promote blood cell activation and differentiation in Drosophila. Mol. Genet. Genomics 267:57-63.[Medline]

MCCORMICK, F., 1994  Activators and effectors of ras p21 proteins. Curr. Opin. Genet. Dev. 4:71-76.[Medline]

NELSON, R. E., L. I. FESSLER, Y. TAKAGI, B. BLUMBERG, and D. R. KEENE et al., 1994  Peroxidasin: a novel enzyme-matrix protein of Drosophila development. EMBO J. 13:3438-3447.[Medline]

NEUFELD, T. P., A. F. DE LA CRUZ, L. A. JOHNSTON, and B. A. EDGAR, 1998  Coordination of growth and cell division in the Drosophila wing. Cell 93:1183-1193.[Medline]

NEUMAN-SILBERBERG, F. S., E. SCHEJTER, F. M. HOFFMANN, and B. Z. SHILO, 1984  The Drosophila ras oncogenes: structure and nucleotide sequence. Cell 37:1027-1033.[Medline]

NUSSBAUMER, U., G. HALDER, J. GROPPE, M. AFFOLTER, and J. MONTAGNE, 2000  Expression of the blistered/DSRF gene is controlled by different morphogens during Drosophila trachea and wing development. Mech. Dev. 96:27-36.[Medline]

PROBER, D. A. and B. A. EDGAR, 2000  Ras1 promotes cellular growth in the Drosophila wing. Cell 100:435-446.[Medline]

QIU, P., P. C. PAN, and S. GOVIND, 1998  A role for the Drosophila Toll/Cactus pathway in larval hematopoiesis. Development 125:1909-1920.[Abstract]

REHORN, K. P., H. THELEN, A. M. MICHELSON, and R. REUTER, 1996  A molecular aspect of hematopoiesis and endoderm development common to vertebrates and Drosophila. Development 122:4023-4031.[Abstract]

RIZKI, T. M., R. M. RIZKI, and E. H. GRELL, 1980  A mutant affecting the crystal cells in Drosophila melanogaster.. Roux's Arch. Dev. Biol. 188:91-99.

RIZKI, T. M., R. M. RIZKI, and R. A. BELLOTTI, 1985  Genetics of a Drosophila phenoloxidase. Mol. Gen. Genet. 201:7-13.[Medline]

RUSSO, J., S. DUPAS, F. FREY, Y. CARTON, and M. BREHELIN, 1996  Insect immunity: early events in the encapsulation process of parasitoid (Leptopilina boulardi) eggs in resistant and susceptible strains of Drosophila. Parasitology 112:135-142.

SCHULZE, A., K. LEHMANN, H. B. JEFFERIES, M. MCMAHON, and J. DOWNWARD, 2001  Analysis of the transcriptional program induced by Raf in epithelial cells. Genes Dev. 15:981-994.[Abstract/Free Full Text]

SPRUCK, C., H. STROHMAIER, M. WATSON, A. P. SMITH, and A. RYAN et al., 2001  A CDK-independent function of mammalian Cks1: targeting of SCF(Skp2) to the CDK inhibitor p27Kip1. Mol. Cell 7:639-650.[Medline]

STOPERA, S. A. and R. P. BIRD, 1992  Expression of ras oncogene mRNA and protein in aberrant crypt foci. Carcinogenesis 13:1863-1868.[Abstract/Free Full Text]

TANG, A. H., T. P. NEUFELD, E. KWAN, and G. M. RUBIN, 1997  PHYL acts to down-regulate TTK88, a transcriptional repressor of neuronal cell fates, by a SINA-dependent mechanism. Cell 90:459-467.[Medline]

THUMMEL, C. S., A. M. BOULET, and H. D. LIPSHITZ, 1988  Vectors for Drosophila P-element-mediated transformation and tissue culture transfection. Gene 74:445-456.[Medline]

VAN VACTOR, D. L., JR., R. L. CAGEN, H. KRAMER, and S. L. ZIPURSKY, 1991  Induction in the developing compound eye of Drosophila: multiple mechanisms restrict R7 induction to a single retinal precursor cell. Cell 67:1145-1155.[Medline]

VASS, E. and A. J. NAPPI, 2001  Fruit fly immunity. BioScience 51:529-535.

WHITE, M. A., C. NICOLETTE, A. MINDEN, A. POLVERINO, and L. VAN AELST et al., 1995  Multiple Ras functions can contribute to mammalian cell transformation. Cell 80:533-541.[Medline]

YASOTHORNSRIKUL, S., W. J. DAVIS, G. CRAMER, D. A. KIMBRELL, and C. R. DEAROLF, 1997  viking: identification and characterization of a second type IV collagen in Drosophila. Gene 198:17-25.[Medline]

ZINYK, D. L., B. G. MCGONNIGAL, and C. R. DEAROLF, 1993  Drosophila awdK-pn, a homologue of the metastasis suppressor gene nm23, suppresses the Tum-1 haematopoietic oncogene. Nat. Genet. 4:195-201.[Medline]

This article has been cited by other articles:

N. Martinek, J. Shahab, M. Saathoff, and M. Ringuette Haemocyte-derived SPARC is required for collagen-IV-dependent stability of basal laminae in Drosophila embryos J. Cell Sci., May 15,&nb