The Effects of Methotrexate on Drosophila Development, Female Fecundit
http://www.100md.com
《毒物学科学杂志》
Department of Biology, Bioscience Complex, Room 2522, Queen's University, Kingston, Ontario, K7L 3N6 Canada
ABSTRACT
Methotrexate (MTX), a synthetic folate analog, is a tight-binding inhibitor of dihydrofolate reductase (DHFR), a key enzyme for the biosynthesis of purines, thymidylate, and several amino acids. As a consequence, MTX decreases titres of reduced folates, interferes with DNA synthesis, and results in the arrest of rapidly proliferating cells, making it a drug of choice for the treatment of a variety of cancers and auto-immune disorders. MTX is also a known teratogen in all higher animals tested, but there is little information about the effects of this drug on invertebrates. Here we show that MTX has little effect on the survival of Drosophila melanogaster adult flies, but severely diminishes female fecundity. Reduced oviposition, coupled with aberrant egg morphologies, resulted in near sterility of MTX-treated females. Rare surviving progeny showed developmental abnormalities including larval tumors, and bristle, wing, eye, and leg defects. To determine if these phenotypes could be attributed solely to DHFR inhibition, microarray analysis was undertaken and included MTX-treated females, ovaries, and cell line samples. Genes encoding transcripts that were perturbed by the drug were verified using quantitative real-time RT-PCR. Many of these genes were involved in cell cycle regulation, signal transduction, transport, defense response, transcription, or various aspects of metabolism. These studies show that MTX treatment has multiple targets and, in addition, provides a new invertebrate model for the study of teratogenesis.
Key Words: methotrexate; dihydrofolate reductase; Drosophila melanogaster; female fecundity; teratogenesis; gene expression.
INTRODUCTION
Methotrexate (8-amino-10-methyl-pteryoglutamic acid; MTX) is a synthetic analogue of dihydrofolate (DHF) and potent inhibitor of dihydrofolate reductase (DHFR), competing with DHF for the enzyme's active site. DHFR is an important housekeeping enzyme as it catalyses the conversion of DHF to tetrahydrofolate (THF), which acts as a cofactor in the one-carbon transfer reactions in the biosynthesis of purine, thymidylate, and several amino acids (Rader and Huennekens, 1973). By depleting the pool of THF and its derivatives, MTX interferes with DNA synthesis and leads to the arrest of rapidly dividing cells. Thus, it has been widely used as a chemotherapeutic agent for a variety of cancers (Huennekens 1994), as well for treatment of ectopic pregnancy (Fernandez et al., 1998), psoriasis (Collins and Rogers, 1992; Zachariae et al., 1990), rheumatoid arthritis (Nakazawa et al., 2001), and systemic lupus (Wise et al., 1996). Toxic effects of MTX are primarily due to direct inhibition of DHFR and thus, indirectly thymidylate synthase and glycinamide ribonucleotide formyltransferase (McGuire, 2003). Thus, not surprisingly, it is also a potent teratogen. Birth defects have been seen in newborns after treatment of pregnant vertebrate mothers. These defects include prenatal-onset growth deficiency, severe lack of ossification of the calvarium, prominent eyes, small low-set ears, micrognathia, distal limb reductions, low birth-weight, and developmental delay (Aviles et al., 1991; Del Campo et al., 1999; Shaw and Steinback, 1968).
Developmentally toxic effects of MTX have been examined in a variety of other vertebrate model organisms such as rats, rabbits, and mice (DeSasso and Goeringer, 1992; Pellizzer et al., 2004; Scmid, 1984), including some studies that demonstrate a contribution of DHFR inhibition to the teratogenic process (DeSasso and Goeringer, 1991, 1992; Sutton et al., 1998). Typically DHFR inhibition and folate metabolism have been the primary focus of developmental studies. However more recently, microarray-based approaches have been used for gene expression profiling in order to understand the cellular response to a wide range of drugs including MTX (Brachat et al., 2002; Takata et al., 2005).
In Drosophila, asymmetrical bristle placement was noted in MTX-treated flies (Badaev et al., 1992), while aminopterin, a similar antifolate, was shown to cause increased larval mortality, developmental time (Le Menn et al., 1983), as well as wing abnormalities (Silber et al., 1985). Although Drosophila melanogaster is currently being used as a model organism for human neural disease, aging, cardiac disease, "memory loss," and cancer (Bier and Bodmer, 2004; Fortini and Bonini, 2000; Grotewiel et al., 2005; Potter et al., 2000; Saitoe et al., 2005), the fruit fly is not currently used to model human birth defects. Here we show the morphological effects on Drosophila due to MTX treatment and demonstrate the remarkable, conserved ability of MTX to produce teratogenic effects across a wide range of organisms. Considering that many signal pathways are conserved between humans and flies this is not a surprising result, and we show many transcripts with conserved function in humans are affected by MTX in Drosophila. This has great implication for understanding mechanistically what occurs at a cellular level after MTX (or other potential teratogen) treatment. Together with this conservation of gene function, the short generation time, sequenced genome, the readily available scientific tools such as microarrays and gene databases, make the fruit fly an excellent organism to study MTX toxicity and teratogenesis without the need to sacrifice and destroy mammalian offspring.
MATERIALS AND METHODS
Fly line culture conditions, mortality, female fecundity, larval and fly development.
Canton S (CS) D. melanogaster were routinely reared on standard yeast-sucrose or molasses medium. To assess mortality, adults of the same age were placed on fresh culture medium (20 flies per bottle) containing 0, 5, 10, 15, or 20 ppm (4.4 x 10–5 M) MTX and maintained in the dark at 25 ± 1°C. Surviving flies were counted after 10 days. Mortality experiments were done in triplicate.
Female fecundity was determined by collecting adult females within 24 h of emergence from the pupal cases (eclosion). Females (4) and males (3) were placed together in 35 ml glass vials containing 2 ml of culture medium (control, 2, 5, 10, or 20 ppm MTX) with a removable cap. Caps were replaced every 12 h for 14 days and the number of eggs deposited in a 24 h period was determined. Experiments were done in triplicate and eggs laid by treated or untreated flies were totaled and averaged for each 24 h period. Surviving females from control and 20 ppm MTX were kept on the appropriate medium for 20 days, and subsequently ovaries were dissected in physiological saline. The number of follicles at an advanced stage of development were determined. Large opaque follicles were considered mature vitellogenic eggs. ANOVA and two sample t-test were used to determine significant difference in the statistical analysis. First instar larvae that hatched on MTX–containing medium were monitored as they developed, as were adults and both these stages were examined for morphological abnormalities.
Scanning electron microscopy.
Samples of eggs and ovaries from flies reared on culture medium, with or without MTX, were collected. CS flies were placed in vials containing culture medium, with or without 5 ppm MTX, for 3, 4, or 5 days. They were then transferred to grape medium (Williamson et al., 1978) for oviposition for 24 h, followed by dissection of ovaries as above. Eggs were also collected from the medium. Tissues were prepared for scanning electron microscopy (SEM) using a modified version of Turner and Mahowald's method (1976). Briefly, tissues were washed in 0.1 M sodium phosphate buffer pH 7.2 and kept in fixative (3% gluteraldehyde, 2% paraformaldehyde, 2.5% DMSO in 0.1 M sodium phosphate buffer, pH 7.2) overnight. Following six 15-min washes in sucrose buffer (0.2 M sucrose in 0.1 M sodium phosphate buffer, pH 7.2), the tissues were post-fixed (1% OsO4 in 0.1 M sodium phosphate buffer, pH 7.2) for 1–2 h, rinsed twice for 5 min with sucrose buffer and dehydrated through a graded series of ethanol solutions. Samples were dried by the critical point technique using CO2 (Anderson, 1951; Horridge and Tamm, 1969), then mounted on stubs using double-sided adhesive tabs, coated with a thin layer of gold palladium utilizing an E5100 sputter coater (Polaron Instruments, England) and finally photographed with an S-450 SEM (Hitachi, Japan) operated at 20 kV.
Gene expression analysis using microarrays.
RNA from whole females, or ovaries from females, reared on culture medium, with or without MTX, as well as Drosophila S3 cells cultured with or without MTX, were collected for microarray analysis. CS flies were placed in vials containing culture medium, with or without 5 ppm MTX, for 5 days. Females (20 per assay) were collected, frozen with liquid nitrogen, and stored at –80°C. Pairs of ovaries (20) were dissected in 0.1 M NaCl, immediately frozen on dry ice and stored at –80°C. S3 cells were cultured in Schneider's Drosophila medium (Life Technologies, Burlington, ON) supplemented with 10% FBS (Life Technologies) with or without 5.2 x 10–8 M (0.024 ppm) MTX, harvested after 4 days, and stored at –80°C. Whole females, ovaries, and S3 cells were thawed on ice and homogenized for 3 intervals of 30 s using a Rotor Stator homogenizer (KikaWerk, Germany). Total cellular RNA (30 μl) was collected using an RNeasy Mini Kit (Qiagen Inc., Mississauga, ON). Since RNA from MTX-treated ovaries was limited, mRNA from both treated and untreated ovaries was amplified. Double stranded cDNA was made from the RNA and used as a template for new mRNA synthesis using a RiboAmp RNA Amplification Kit (Arcturus, Mountain View, CA) following the manufacturer's directions. Quantity and quality of RNA was determined by measuring the absorbance at a wavelength of 260 nm and 280 nm.
Purified RNAs were used to create Fluor-labelled (Cy3 and Cy5) cDNAs using the indirect labeling protocol at Queen's Microarray Facility (Queen's Microarray Facility, Kingston, ON) and used to compete for hybridization sites on a Drosophila 7K 2.0 cDNA microchip (whole flies and ovary samples) or on a Drosophila 12K 1.0 cDNA microchip (S3 cell samples). Both microarrays were purchased from the Canadian Drosophila Microarray Centre (University of Toronto, Mississauga, ON). Flours were "flipped" to ensure transcript abundance was not a product of dye bias and fluor intensity was measured using a ScanArray 4000 system (Perkin Elmer, Montreal, QC, formerly GSI Lumonics). Data from this study is publicly available from the Gene Expression Omnibus (http://www.ncbi.nlm.nih.gov/geo/) as accession GSE3495.
QuantArray Software (v3.0, Perkin Elmer) was used to quantify and normalize scanned images. Further normalization and analysis of S3 hybridizations were done using GeneTraffic Duo (Stratgene, La Jolla, CA). Signal intensities lower than 150 and signal-to-background ratios less than 1.0 were considered low intensity spots and flagged. Flagged spots were not included in analysis and genes with more than 1/8 flagged spots were not considered for analysis.
Quantitative PCR.
RNA was isolated from maternally treated and control (untreated) ovaries as well as S3 cells as described. Total RNA (20 ng) was used for each quantitative reverse transcription PCR (qRT-PCR) reaction. cDNA was created by reverse transcription from isolated RNA, then amplified in a One-step RT-PCR reaction using the QuantiTect SYBR Green RT-PCR kit (Qiagen), cycled as follows; for reverse transcription, one cycle at 50°C for 1800 s and 95°C for 900 s, followed by 45 cycles of amplification at, 94°C for 15 s, 57°C for 30 s, 72°C for 30 s, and 76.8°C for 15 s with the optics option on, in a SmartCycler (Cepheid SmartCycler, Fisher Scientific Company, Ottawa, ON). Genes were selected based on the microarray analysis and specific primers designed from the corresponding GenBank sequence were used to amplify cDNAs. Primers specific to endogenous D. melanogaster actin 5C (DmAct5C; K00667) was used for relative quantification of gene expression levels reactions in all reactions, and D. melanogaster ribosomal protein 49 (Dmrp49; X00848) was used as a second control to verify the actin results. All qRT-PCR reactions were done in triplicate and means and standard deviation determined.
RESULTS
The Effect of MTX on Adult Mortality and Fecundity
Adult fly survival on the antifolate was determined by placing flies of identical age on medium containing 0, 5, 10, 15 or 20 ppm MTX. Survival appeared to be unaffected, even after 10 days on 20 ppm MTX. An equal percentage of flies (80%) survived on the highest MTX concentration, at low concentrations, and controls. Viability was only assessed for 10 days to avoid potential confounding, synergistic effects of depressed fertility and longevity (Sgro and Partridge, 1999).
Although male fertility did not appear to be reduced by MTX treatment (Goldsmith and Frank, 1952; results not shown), female fecundity was reduced (Fig. 1). Control females oviposited approximately 23 eggs per day from day 2–4, and an average of approximately 15–20 eggs per day for the next 8 days. In contrast, MTX-treated flies (2 ppm) laid a maximum of 13 eggs per day on day 2 and <1 egg per day by day 4. As the concentration of MTX was increased to 5, 10, and 20 ppm, the number of eggs oviposited was reduced and by day 4 (with the exception of a single egg laid on 2 ppm MTX) there were no eggs in vials containing any concentration of MTX. Dissection of the females that had been treated with 20 ppm MTX at day 21 showed that the flies were simply not retaining eggs. Rather the number of yolky follicles (stages 11–14) was significantly lower in the MTX-treated flies; with 0.9 ± 1.3 large, vitellogenic follicles per ovary, compared to 39.7 ± 4.0 follicles in untreated females (p < 0.05).
The Effect of MTX on Developmental Morphology
Since oviposition was dramatically affected by MTX treatment, the development and morphology of follicles and eggs from females exposed to 5 ppm MTX was examined further. Eggs were collected from females exposed to MTX and ovaries were dissected. After only 3 days of MTX exposure, females oviposited eggs with an abnormal appearance (48%; n = 46): the chorionic appendages were frequently shorter and malformed and size of the eggs was reduced compared to eggs oviposited by control females (Figs. 2a, 2d–2f), although there was some variability. After 4 days major developmental abnormalities in the chorionic appendages and outer chorion structure were observed in the majority of the MTX-exposed eggs (91%; n = 45) (Figs. 2g and 2j). In these experiments, no oviposition was observed after 5 days of treatment. Ovaries from both treated and control females appeared similar after 3 days of treatment (Figs. 2b and 2c), but after 4 days, ovaries from the treated females were smaller, with fewer large yolky follicles (100%; n = 12) (Figs. 2h and 2i). Many appeared to be similar to those of immature females, but often with abnormal pedicels and oviduct laterals. After 5 days of MTX treatment no visible developing follicles were seen in any ovary examined, and much of the tissue appeared flaccid and adherent (Fig. 2k).
MTX not only affected the egg stage of development, but also post-embryonic development. Although the hatch rate was low, the few larvae that survived maternal MTX-treatment showed abnormalities. Many larvae had melanotic tumors, evidenced by black patches of cells, indicative of cell death (Fig. 3a). The very few adults that successfully emerged from their pupal cases had showed a significantly high frequency of morphological abnormalities compared to control flies (p < 0.05). Approximately 88% (n = 120) of the flies showed abnormal tufts of bristles on the dorsal thorax, evident curvature in the legs, abnormalities of the eyes, and/or wing deformities (Fig. 3b; not shown).
The Effect of MTX on Gene Expression
Since significant abnormalities in follicle development were observed in females after 4–5 days of MTX treatment, the earlier effects of MTX on patterns of gene expression in females were examined. After only 3 days of MTX treatment, when the majority of eggs were morphologically normal, there were few significant changes in gene expression patterns in whole females (results not shown). As a result, RNA from ovaries of flies fed on control medium or medium containing MTX (5 ppm) was used to assess transcript abundance with Drosophila cDNA microarrays. Because the RNA was amplified prior to cDNA synthesis, only transcripts shown to be more or less abundant (with a log2 ratio of 1.8 and –1.4, respectively) for all 4 hybridization spots were considered. "Down-regulated" (8) and "up-regulated" (2) transcripts were chosen for further analysis (Table 1). To confirm these results, qRT-PCR was conducted using primers specific to each of the 10 genes. Most (7/8 less abundant and 2/2 of the more abundant sequences) were verified. Significantly, considering that amplification of RNA could possibly introduce bias into transcript abundance assays, the results are remarkably concordant (Table 1).
Of the "down-regulated" transcripts, four encoded cell cycle regulators (grapes and loki, Cyclins A and E), three encoded proteins involved in embryo development (stathmin, concertina, and Minichromosome maintenance 6), and one was for a heat response gene (Heat shock factor). Of the "up-regulated" transcripts, one encoded a cold response gene that appears after low temperature stress (Frost), and the other was for a specific RNA polymerase II transcription factor (zerknullt) (Table 1).
In an effort to confirm that changes in transcript abundance observed in the ovaries reflected MTX-sensitive reproductive or developmental changes, cell lines (derived from embryos) were examined by microarrays. In these experiments, which did not require the amplification of RNA, genes encoding transcripts that were considered more (19) or less (9) abundant showed log2 ratios of greater than 1.0 or less than –0.9, respectively (results not shown). None of these genes, representing those with the most dramatic changes in cell transcript levels, had been identified in ovary microarray experiments. However, when the cell line was examined by qRT-PCR using primers specific to the transcripts identified from ovary microarrays, 7/8 of the "down-regulated" genes were also confirmed in the cells (Table 3), with the exception of Cyclin E transcripts, which were unchanged.
Since sequences identified using microarrays from MTX-treated cells were largely concordant with transcript abundance in ovaries, sequences identified in cell line microarrays were assessed in ovaries. First, the changes in message abundance after MTX treatment of cultured cells were confirmed by qRT-PCR experiments on 28 chosen "up and down regulated" genes. Analysis showed that 25 were concordant (Tables 2 and 3). Of these 25 sequences, the majority (9) appeared to be involved in defense or transport response (Glutathione S transferase E9, Drosomycin, meiotic 9, Thymine-DNA-glycosylase 1, Juvenile hormone epoxide hydrolase 3, Odorant-binding protein 99a, Cytochrome P450 12c1, Multiple drug resistance 65, and CG8032), 6 were involved in signal transduction or cell cycle regulation (Neuropeptide-like precursor 2, Tetraspanin 42E, innexin 3, cyclin-dependent kinase 2, Stem-loop binding protein, and CG5384), two were involved in protease inhibition (CG8066 and crammer), and the rest were associated with a variety of metabolic functions including transcription (Ets at 21c), amino acid catabolism (CG9363), metamorphosis (Angiotensin converting enzyme), amino acid de-phosphorylation (MKP-like), receptor binding (Niemann-Pick type C-2), dorsal/ventral axis formation (spatzle), peroxisome biogenesis (CG7609), or of unknown function (CG9894). The majority (19/25) of the transcripts that were sensitive to MTX-toxicity in cultured cells also showed changes in message levels in ovarian samples, as assessed by qRT-PCR analysis of ovaries (Tables 2 and 3).
DISCUSSION
MTX directly targets DHFR and decreases folate pools. As a consequence, it indirectly inhibits other folate-dependant enzymes such as thymidylate synthase, glycineamide ribonucleotide transformylase, and aminoimidazole carboxamide ribonucleotide transformylase (Rader and Huennekens, 1973). Thus MTX is a potent inhibitor with toxic and teratogenic side effects (McGuire, 2003). The morphological abnormalities seen after MTX administration in mammals is remarkably conserved in Drosophila, despite the obvious differences in the invertebrate system. Given the increasing concern about the sacrifice of mammalian embryos in toxicological testing, it was our goal to explore the use of the D. melanogaster model to investigate the genes involved in MTX-induced developmental change. In order to demonstrate that D. melanogaster is indeed an excellent model for these studies, as well as other antifolate-induced birth defects, we have observed mortality, female fecundity, and development and examined transcript abundance in response to MTX in ovaries and cell lines.
Exposure of D. melanogaster to MTX (up to 20 ppm) did not appear to limit the life span of adults, but, if present at levels 5 ppm, it had severe effects on the female reproductive system with a reduction from 23 eggs/day on control medium to <1 egg/day after 4 days on 2–20 ppm MTX. The ovaries from treated (5 ppm MTX) females appeared flaccid with small, undeveloped follicles. Deposited eggs were smaller with short or absent chorionic appendages with an irregular surface pattern, which is imprinted by the surrounding follicle cells during oogenesis (Fig. 2). Of the few eggs that hatch to larvae, many developed melanotic tumors and/or die during later development (Fig. 3; not shown). The small number of adult survivors showed tumors, and severe defects including bristle, eye and leg abnormalities (Fig. 3), strikingly similar to those deformities documented in anti-folate treated mammals (Schardein, 2000). These results are perhaps not surprising considering MTX is used to inhibit rapidly dividing cells. The reduction in fertility appears to be specific to females since no reduction in oviposition was seen when males were exposed to aminopterin or MTX, prior to crossing to control females (Goldsmith and Frank, 1954; not shown).
As an approach to understanding the morphological abnormalities and the underlying molecular consequences of MTX treatment, microarray analyses were done on D. melanogaster females, ovaries, and cell lines. Analyses with whole females showed only a few significant changes in transcript abundance (not shown), but this is not unexpected since adult mortality was unaffected by MTX treatment and cells in the adult are largely post-mitotic. Ovary microarrays and confirming qRT-PCR experiments, on the other hand, revealed an interesting set of genes that appear to encode transcripts that were either increased or decreased in abundance as a result of MTX-treatment (Table 1). These transcripts correspond to proteins involved in cell cycle or DNA damage, embryo development and stress response. Analysis of embryonic cell lines confirmed the susceptibility of the majority of the sequences to MTX treatment (Tables 2 and 3).
Microarray and qRT-PCR analysis of S3 cells showed a large number of transcripts with expression levels that were perturbed by MTX treatment (not shown), and the majority showed the same response in ovaries, suggesting that 19 transcripts that encoded proteins with functions including cell cycle regulation, metabolism, and defense are affected by treatment with MTX (Tables 2 and 3). In none of these studies was Dhfr identified (not shown), which is not surprising as DHFR transcripts are only known to increase in abundance after gene amplification due to selection for resistance (Singer et al., 2000).
The aberrant morphologies including female sterility and developmentally induced melanotic tumors and appendage abnormalities are consistent with MTX-mediated inhibition of rapid cell proliferation. Few transcriptional changes could be detected by microarray analysis of whole flies, and since adults are largely post-mitotic, this suggests that only mitotic cells are susceptible to toxicity. During oogenesis, germ cells rapidly divide and follicle cells are continually renewed from mesodermal origin within the ovary. The inhibition of folate metabolism by MTX treatment, appears then to reduce cell proliferation, and this is reflected in the "down-regulation" of many cell cycle and growth related genes, such as grp, CycA, CycE, cdc2, Slbp, and CG5384, which had transcript levels reduced 0.7–2.7 times. For example, transcripts from the cell damage checkpoint gene, lok, were reduced 11-fold, indicating a dramatic reduction in the normal cell cycle in ovaries. As well, Mcm6, involved in chorion gene amplification, had a 4-fold reduction in mRNA levels after MTX treatment and other mRNAs for membrane receptors, transporters and/or translation associated factors also showed abundance changes in both ovaries and cell lines (Tables 2 and 3). Such inhibition, in turn, would be expected to result in a smaller numbers of oocytes, with limited growth support due to a lack of properly functioning nurse and follicle cells resulting in flaccid ovaries and undeveloped eggs (Fig. 1).
Reduction of cell proliferation during the disc development explains the abnormal adult phenotypes with some cell death resulting in larval melanotic tumors. It should be noted that some cell necrosis was observed in rat livers after MTX administration, but gene expression profiling, did not identify cell death markers (Huang et al., 2004). At the levels of MTX used, treated cells showed a decrease in proliferation (Affleck et al., unpublished), consistent with the changes in cell cycle-related transcripts seen here. Certainly MTX treatment must result in some cell stress as indicated by a 4-fold increase in the detoxifying enzyme GST message and a 4-fold decrease in multiple drug resistance transcripts. Microarray analyses of MTX-treated rodent tissues have noted changes in stress genes and cell cycle (including Cyclin E) genes (Ganter et al., 2005; Huang et al., 2004).
How does maternal MTX treatment cause such dramatic morphological abnormalities in progeny, which in vertebrates would be termed birth defects The teratogenic mechanism of MTX is clearly more complex than simple DHFR inhibition. MTX-induced developmental toxicity can be ameliorated, but not completely prevented, in rabbits using 1-(p-tosyl)-3,4,4-trimethylimidazolidine, a functional analog for THF-mediated one-carbon transfer (DeSasso and Goeringer, 1992). As well, an MTX-resistant form of DHFR was able to only partially protect transgenic murine embryos from the teratogenic effects of MTX (Sutton et al., 1998). The small size, prominent eyes, tufts of hair, and abnormal limbs characteristic of mammalian birth defects (Schardein, 2000) are strikingly similar to the small egg size, rough eyes, abberant bristles, and bent legs and wings observed here (Figs. 2 and 3 and not shown). It is likely then, that although DHFR is an important target for MTX-induced birth defects, there are secondary targets of the drug or other biochemical pathways that are also involved. Certainly, gene expression analysis in the experiments reported here reveals an unexpected complexity in the consequences of MTX treatment during development. The further exploration of altered transcript abundance in response to MTX (or other potential teratogens) combined with the use of "knock out" or "over expression" technologies in Drosophila, which cause no suffering to vertebrates, hold promise for the understanding of these developmental abnormalities. We hope that through this model system we can provide insight into mammalian birth defects, which may be expected to increase in frequency in the future, due to the recent elevated use of MTX and other antifolate therapies for ectopic pregnancies and other human disorders (Nguyen et al., 2002).
ACKNOWLEDGMENTS
A special thanks goes to Bob Temkin and Bill Newcomb for their help with the scanning electron micrographs. The microchip hybridization experiments were carried out by Harriet Feilotter (Queen's University Microarray Facility, Kingston, ON). This work was supported by a grant to V.K.W. (NSERC, Canada) and scholarships to J.G.A. (NSERC) and to K.N. (OGS; J. Armand Bombardier Scholarship Foundation).
REFERENCES
Anderson, T. F. (1951). Techniques for the preservation of three-dimensional structure in preparing specimens for the electron microscope. Trans. N.Y. Acad. Sci. 13, 130–134.
Aviles, A., Diaz-Maqueo, J. C., Talavera, A., Guzman, R., and Garcia, E. L. (1991). Growth and development of children of mothers treated with chemotherapy during pregnancy: Current status of 43 patients. Am. J. Hematol. 36, 243–248.
Badaev, S. A., Drozdovskaia, L. N., and Rapoport, I. A. (1992). [The bilateral asymmetry of the number of bristles in Drosophila occurring under the influence of methotrexate]. Izv. Akad. Nauk. SSSR Biol. 2, 291–295. [Abstract, article in Russian].
Bier, E., and Bodmer, R. (2004). Drosophila, an emerging model for cardiac disease. Gene 342, 1–11.
Brachat, A., Pierrat, B., Xynos, A., Brecht, K., Simonen, M., Brungger, A., and Hiem, J. (2002). A microarray-based, integrated approach to identify novel regulators of cancer drug response and apoptosis. Oncogene 21, 8361–8371.
Collins, P., and Rogers, S. (1992). The efficacy of methotrexate in psoriasis–a review of 40 cases. Clin. Exp. Dermatol. 4, 257–260.
Del Campo, M., Kosaki, K., Bennett, F. C., and Jones, K. L. (1999). Developmental delay in fetal aminopterin/methotrexate syndrome. Teratology 60, 10–12.
DeSasso, J. M., and Goeringer, G. C. (1991). Amelioration by leucovorin of methotrexate developmental toxicity in rabbits. Teratology 43, 201–205.
DeSasso, J. M., and Georinger, G. C. (1992). Methotrexate-induced developmental toxicity in rabbits is ameliorated by 1-(p-tosyl)-3,4,4-trimethylimidazolidine, a functional analog for tetrahydrofolate-mediated one-carbon transfer. Teratology 45, 271–283.
Fernandez, H., Yves, V. S. C., Pauthier, S., Audibert, F., and Frydman, R. (1998). Randomized trial of conservative laparoscopic treatment and methotrexate administration in ectopic pregnancy and subsequent fertility. Human Repro. 13, 3239–3243.
Fortini, M. E., and Bonini, N. M. (2000). Modeling human neurodegenerative diseases in Drosophila on a wing and a prayer. TIG 16, 161–167.
Ganter, B., Tugendreich, S., Pearson, C. I., Ayanoglu, E., Baumhueter, S. et al. (In press). Development of a large-scale chemogenomics database to improve drug candidate selection and to understand mechanisms of chemical toxicity and action. J. Biotechnol. 119, 219–224.
Goldsmith, E. D., and Frank, I. (1954). Sterility in the female fruit fly Drosophila melanogaster produced by the feeding of a folic acid antagonist. Am. J. Physiol. 171, 726–727.
Grotewiel, M. S., Martin, I., Bhandari, P., and Cook-Wiens, E. (2005). Functional senescence in Drosophila melanogaster. Ageing Res. Rev. 4, 372–397.
Horridge, G. A., and Tamm, S. L. (1969). Critical point drying for scanning electron microscopic studies of ciliary motion. Science 163, 817–818.
Huang, Q., Jin, X., Gaillard, E. T., Knight, B. L., Pack, F. D. et al. (2004). Gene expression profiling reveals multiple toxicity endpoints induced by hepatotoxicants. Mutat. Res. 549, 147–168.
Huennekens, F. M. (1994). The methotrexate story: A paradigm for development of cancer chemotherapeutic agents. Adv. Enzyme Regul. 34, 397–419.
Le Menn, A., Silber, J., and Goux, J. (1983). etude de l'effet de deux inhibiteurs de la synthese des nucleotides l'aminopterine et la fluorodesoxyuridine sur des souches sauvage et mutantes vestigial chez Drosophila melanogaster. Biol. Cell 49, 213–218.
McGuire, J. J. (2003). Anticancer antifolates: Current status and future directions. Curr. Pharm. Design 9, 2593–2613.
Nakazawa, F., Matsuno, H., Yudoh, K., Katayama, R., Sawai, T., Uzuki, M., and Kimura, T. (2001). Methotrexate inhibits rheumatoid synovitis by inducing apoptosis. J. Rheumatol. 28, 1800–1808.
Nguyen, C., Duhl, A. J., Escallon, C. S., and Blakemore, K. J. (2002). Multiple anomalies in a fetus exposed to low-dose methotrexate in the first trimester. Obstet. Gynecol. 99, 599–602.
Pellizzer, C., Belle, E., Adler, S., Hartung, T., and Bremer, S. (2004). Detection of tissue-specific effects by methotrexate on differentiating mouse embryonic stem cells. Birth Defects Research (Part B) 71, 331–341.
Potter, C. J., Turenchalk, G. S., and Xu, T. (2000). Drosophila in cancer research. An expanding role. Trends Genet. 16, 33–39.
Rader, J. I., and Huennekens, F. M. (1973). In The Enzymes, 3rd ed. (P. Boyer, Ed.), Vol. 9, pp. 197–223. Academic Press, New York.
Saitoe, M., Horiuchi, J., Tamura, T., and Ito, N. (2005). Drosophila as a novel animal model for studying the genetics of age-related memory impairment. Rev. Neurosci. 16, 137–149.
Schardein, J. L. (2000). In Chemically Induced Birth Defects, 3rd ed., pp. 559–611. Marcel Dekker, New York.
Scmid, B. P. (1984). Monitoring of organ formation in rat embryos after in vitro exposure to azathioprine, mercaptopuring, methotrexate or cyclosporin A. Toxicology 31, 9–21.
Sgro, C. M., and Partridge, L. (1999). A delayed wave of death from reproduction in Drosophila. Science 286, 2521–2524.
Shaw, E. B., and Steinback, H. L. (1968). Aminopterin-induced fetal malformation. Survival of infant after attempted abortion. Amer. J. Dis. Child. 115, 477–482.
Silber, J., Coste, A., Bazin, C., and Le Menn, A. (1985). Dihydrofolate reductase activity and resistance to aminopterin in various species of Drosophila. Mol. Gen. Genet. 200, 92–95.
Singer, M. J., Mesner, L. D., Friedman, C. L., Trask, B. J., and Hamlin, J. L. (2000). Amplification of the human dihydrofolate reductase gene via double minutes is initiated by chromosome breaks. Proc. Natl. Acad. Sci. U.S.A. 97, 7921–7926.
Skalko, R. G., and Gold, M. P. (1974). Teratogenicity of methotrexate in mice. Teratology 9, 159–164.
Sutton, C., McIvor, R. S., Vagt, M., Doggett, B., and Kapur, R. P. (1998). Methotrexate-resistant form of dihydrofolate reductase protects transgenic murine embryos from teratogenic effects of methotrexate. Pediatr. Dev. Path. 1, 503–512.
Takata, R., Toyomasa, K., Kanehira, M., Tsunoda, T., Shuin, T., Miki, T., Namiki, M., Kohri, K., Matsushita, Y., Fujioka, T., et al. (2005). Predicting response to methotrexate, vinblastine, doxorubicin, and cisplatin neoadjuvant chemotherapy for bladder cancers through genome-wide gene expression profiling. Clin. Cancer Res. 11, 2625–2636.
Turner, F. R., and Mahowald, A. P. (1976). Scanning electron microscopy of Drosophila embryogenesis. Dev. Biol. 50, 95–108.
Williamson, J. H., Browder, L., and Thompson, A. (1978). A method of collecting Drosophila embryos of known age. Dros. Inform. Serv. 53, 213.
Wise, C. M., Vuyuru, S., and Roberts, W. N. (1996). Methotrexate in nonrenal lupus and undifferentiated connective tissue disease – a review of 36 patients. J. Rheumatol. 23, 1005–1010.
Zachariae, H., Hansen, H. E., Sogaard, H., and Olsen, T. S. (1990). Kidney biopsies in methotrexate-treated psoriatics. Dermatologica 181, 273–276.(Joslynn G. Affleck, Katerina Neumann, Li)
ABSTRACT
Methotrexate (MTX), a synthetic folate analog, is a tight-binding inhibitor of dihydrofolate reductase (DHFR), a key enzyme for the biosynthesis of purines, thymidylate, and several amino acids. As a consequence, MTX decreases titres of reduced folates, interferes with DNA synthesis, and results in the arrest of rapidly proliferating cells, making it a drug of choice for the treatment of a variety of cancers and auto-immune disorders. MTX is also a known teratogen in all higher animals tested, but there is little information about the effects of this drug on invertebrates. Here we show that MTX has little effect on the survival of Drosophila melanogaster adult flies, but severely diminishes female fecundity. Reduced oviposition, coupled with aberrant egg morphologies, resulted in near sterility of MTX-treated females. Rare surviving progeny showed developmental abnormalities including larval tumors, and bristle, wing, eye, and leg defects. To determine if these phenotypes could be attributed solely to DHFR inhibition, microarray analysis was undertaken and included MTX-treated females, ovaries, and cell line samples. Genes encoding transcripts that were perturbed by the drug were verified using quantitative real-time RT-PCR. Many of these genes were involved in cell cycle regulation, signal transduction, transport, defense response, transcription, or various aspects of metabolism. These studies show that MTX treatment has multiple targets and, in addition, provides a new invertebrate model for the study of teratogenesis.
Key Words: methotrexate; dihydrofolate reductase; Drosophila melanogaster; female fecundity; teratogenesis; gene expression.
INTRODUCTION
Methotrexate (8-amino-10-methyl-pteryoglutamic acid; MTX) is a synthetic analogue of dihydrofolate (DHF) and potent inhibitor of dihydrofolate reductase (DHFR), competing with DHF for the enzyme's active site. DHFR is an important housekeeping enzyme as it catalyses the conversion of DHF to tetrahydrofolate (THF), which acts as a cofactor in the one-carbon transfer reactions in the biosynthesis of purine, thymidylate, and several amino acids (Rader and Huennekens, 1973). By depleting the pool of THF and its derivatives, MTX interferes with DNA synthesis and leads to the arrest of rapidly dividing cells. Thus, it has been widely used as a chemotherapeutic agent for a variety of cancers (Huennekens 1994), as well for treatment of ectopic pregnancy (Fernandez et al., 1998), psoriasis (Collins and Rogers, 1992; Zachariae et al., 1990), rheumatoid arthritis (Nakazawa et al., 2001), and systemic lupus (Wise et al., 1996). Toxic effects of MTX are primarily due to direct inhibition of DHFR and thus, indirectly thymidylate synthase and glycinamide ribonucleotide formyltransferase (McGuire, 2003). Thus, not surprisingly, it is also a potent teratogen. Birth defects have been seen in newborns after treatment of pregnant vertebrate mothers. These defects include prenatal-onset growth deficiency, severe lack of ossification of the calvarium, prominent eyes, small low-set ears, micrognathia, distal limb reductions, low birth-weight, and developmental delay (Aviles et al., 1991; Del Campo et al., 1999; Shaw and Steinback, 1968).
Developmentally toxic effects of MTX have been examined in a variety of other vertebrate model organisms such as rats, rabbits, and mice (DeSasso and Goeringer, 1992; Pellizzer et al., 2004; Scmid, 1984), including some studies that demonstrate a contribution of DHFR inhibition to the teratogenic process (DeSasso and Goeringer, 1991, 1992; Sutton et al., 1998). Typically DHFR inhibition and folate metabolism have been the primary focus of developmental studies. However more recently, microarray-based approaches have been used for gene expression profiling in order to understand the cellular response to a wide range of drugs including MTX (Brachat et al., 2002; Takata et al., 2005).
In Drosophila, asymmetrical bristle placement was noted in MTX-treated flies (Badaev et al., 1992), while aminopterin, a similar antifolate, was shown to cause increased larval mortality, developmental time (Le Menn et al., 1983), as well as wing abnormalities (Silber et al., 1985). Although Drosophila melanogaster is currently being used as a model organism for human neural disease, aging, cardiac disease, "memory loss," and cancer (Bier and Bodmer, 2004; Fortini and Bonini, 2000; Grotewiel et al., 2005; Potter et al., 2000; Saitoe et al., 2005), the fruit fly is not currently used to model human birth defects. Here we show the morphological effects on Drosophila due to MTX treatment and demonstrate the remarkable, conserved ability of MTX to produce teratogenic effects across a wide range of organisms. Considering that many signal pathways are conserved between humans and flies this is not a surprising result, and we show many transcripts with conserved function in humans are affected by MTX in Drosophila. This has great implication for understanding mechanistically what occurs at a cellular level after MTX (or other potential teratogen) treatment. Together with this conservation of gene function, the short generation time, sequenced genome, the readily available scientific tools such as microarrays and gene databases, make the fruit fly an excellent organism to study MTX toxicity and teratogenesis without the need to sacrifice and destroy mammalian offspring.
MATERIALS AND METHODS
Fly line culture conditions, mortality, female fecundity, larval and fly development.
Canton S (CS) D. melanogaster were routinely reared on standard yeast-sucrose or molasses medium. To assess mortality, adults of the same age were placed on fresh culture medium (20 flies per bottle) containing 0, 5, 10, 15, or 20 ppm (4.4 x 10–5 M) MTX and maintained in the dark at 25 ± 1°C. Surviving flies were counted after 10 days. Mortality experiments were done in triplicate.
Female fecundity was determined by collecting adult females within 24 h of emergence from the pupal cases (eclosion). Females (4) and males (3) were placed together in 35 ml glass vials containing 2 ml of culture medium (control, 2, 5, 10, or 20 ppm MTX) with a removable cap. Caps were replaced every 12 h for 14 days and the number of eggs deposited in a 24 h period was determined. Experiments were done in triplicate and eggs laid by treated or untreated flies were totaled and averaged for each 24 h period. Surviving females from control and 20 ppm MTX were kept on the appropriate medium for 20 days, and subsequently ovaries were dissected in physiological saline. The number of follicles at an advanced stage of development were determined. Large opaque follicles were considered mature vitellogenic eggs. ANOVA and two sample t-test were used to determine significant difference in the statistical analysis. First instar larvae that hatched on MTX–containing medium were monitored as they developed, as were adults and both these stages were examined for morphological abnormalities.
Scanning electron microscopy.
Samples of eggs and ovaries from flies reared on culture medium, with or without MTX, were collected. CS flies were placed in vials containing culture medium, with or without 5 ppm MTX, for 3, 4, or 5 days. They were then transferred to grape medium (Williamson et al., 1978) for oviposition for 24 h, followed by dissection of ovaries as above. Eggs were also collected from the medium. Tissues were prepared for scanning electron microscopy (SEM) using a modified version of Turner and Mahowald's method (1976). Briefly, tissues were washed in 0.1 M sodium phosphate buffer pH 7.2 and kept in fixative (3% gluteraldehyde, 2% paraformaldehyde, 2.5% DMSO in 0.1 M sodium phosphate buffer, pH 7.2) overnight. Following six 15-min washes in sucrose buffer (0.2 M sucrose in 0.1 M sodium phosphate buffer, pH 7.2), the tissues were post-fixed (1% OsO4 in 0.1 M sodium phosphate buffer, pH 7.2) for 1–2 h, rinsed twice for 5 min with sucrose buffer and dehydrated through a graded series of ethanol solutions. Samples were dried by the critical point technique using CO2 (Anderson, 1951; Horridge and Tamm, 1969), then mounted on stubs using double-sided adhesive tabs, coated with a thin layer of gold palladium utilizing an E5100 sputter coater (Polaron Instruments, England) and finally photographed with an S-450 SEM (Hitachi, Japan) operated at 20 kV.
Gene expression analysis using microarrays.
RNA from whole females, or ovaries from females, reared on culture medium, with or without MTX, as well as Drosophila S3 cells cultured with or without MTX, were collected for microarray analysis. CS flies were placed in vials containing culture medium, with or without 5 ppm MTX, for 5 days. Females (20 per assay) were collected, frozen with liquid nitrogen, and stored at –80°C. Pairs of ovaries (20) were dissected in 0.1 M NaCl, immediately frozen on dry ice and stored at –80°C. S3 cells were cultured in Schneider's Drosophila medium (Life Technologies, Burlington, ON) supplemented with 10% FBS (Life Technologies) with or without 5.2 x 10–8 M (0.024 ppm) MTX, harvested after 4 days, and stored at –80°C. Whole females, ovaries, and S3 cells were thawed on ice and homogenized for 3 intervals of 30 s using a Rotor Stator homogenizer (KikaWerk, Germany). Total cellular RNA (30 μl) was collected using an RNeasy Mini Kit (Qiagen Inc., Mississauga, ON). Since RNA from MTX-treated ovaries was limited, mRNA from both treated and untreated ovaries was amplified. Double stranded cDNA was made from the RNA and used as a template for new mRNA synthesis using a RiboAmp RNA Amplification Kit (Arcturus, Mountain View, CA) following the manufacturer's directions. Quantity and quality of RNA was determined by measuring the absorbance at a wavelength of 260 nm and 280 nm.
Purified RNAs were used to create Fluor-labelled (Cy3 and Cy5) cDNAs using the indirect labeling protocol at Queen's Microarray Facility (Queen's Microarray Facility, Kingston, ON) and used to compete for hybridization sites on a Drosophila 7K 2.0 cDNA microchip (whole flies and ovary samples) or on a Drosophila 12K 1.0 cDNA microchip (S3 cell samples). Both microarrays were purchased from the Canadian Drosophila Microarray Centre (University of Toronto, Mississauga, ON). Flours were "flipped" to ensure transcript abundance was not a product of dye bias and fluor intensity was measured using a ScanArray 4000 system (Perkin Elmer, Montreal, QC, formerly GSI Lumonics). Data from this study is publicly available from the Gene Expression Omnibus (http://www.ncbi.nlm.nih.gov/geo/) as accession GSE3495.
QuantArray Software (v3.0, Perkin Elmer) was used to quantify and normalize scanned images. Further normalization and analysis of S3 hybridizations were done using GeneTraffic Duo (Stratgene, La Jolla, CA). Signal intensities lower than 150 and signal-to-background ratios less than 1.0 were considered low intensity spots and flagged. Flagged spots were not included in analysis and genes with more than 1/8 flagged spots were not considered for analysis.
Quantitative PCR.
RNA was isolated from maternally treated and control (untreated) ovaries as well as S3 cells as described. Total RNA (20 ng) was used for each quantitative reverse transcription PCR (qRT-PCR) reaction. cDNA was created by reverse transcription from isolated RNA, then amplified in a One-step RT-PCR reaction using the QuantiTect SYBR Green RT-PCR kit (Qiagen), cycled as follows; for reverse transcription, one cycle at 50°C for 1800 s and 95°C for 900 s, followed by 45 cycles of amplification at, 94°C for 15 s, 57°C for 30 s, 72°C for 30 s, and 76.8°C for 15 s with the optics option on, in a SmartCycler (Cepheid SmartCycler, Fisher Scientific Company, Ottawa, ON). Genes were selected based on the microarray analysis and specific primers designed from the corresponding GenBank sequence were used to amplify cDNAs. Primers specific to endogenous D. melanogaster actin 5C (DmAct5C; K00667) was used for relative quantification of gene expression levels reactions in all reactions, and D. melanogaster ribosomal protein 49 (Dmrp49; X00848) was used as a second control to verify the actin results. All qRT-PCR reactions were done in triplicate and means and standard deviation determined.
RESULTS
The Effect of MTX on Adult Mortality and Fecundity
Adult fly survival on the antifolate was determined by placing flies of identical age on medium containing 0, 5, 10, 15 or 20 ppm MTX. Survival appeared to be unaffected, even after 10 days on 20 ppm MTX. An equal percentage of flies (80%) survived on the highest MTX concentration, at low concentrations, and controls. Viability was only assessed for 10 days to avoid potential confounding, synergistic effects of depressed fertility and longevity (Sgro and Partridge, 1999).
Although male fertility did not appear to be reduced by MTX treatment (Goldsmith and Frank, 1952; results not shown), female fecundity was reduced (Fig. 1). Control females oviposited approximately 23 eggs per day from day 2–4, and an average of approximately 15–20 eggs per day for the next 8 days. In contrast, MTX-treated flies (2 ppm) laid a maximum of 13 eggs per day on day 2 and <1 egg per day by day 4. As the concentration of MTX was increased to 5, 10, and 20 ppm, the number of eggs oviposited was reduced and by day 4 (with the exception of a single egg laid on 2 ppm MTX) there were no eggs in vials containing any concentration of MTX. Dissection of the females that had been treated with 20 ppm MTX at day 21 showed that the flies were simply not retaining eggs. Rather the number of yolky follicles (stages 11–14) was significantly lower in the MTX-treated flies; with 0.9 ± 1.3 large, vitellogenic follicles per ovary, compared to 39.7 ± 4.0 follicles in untreated females (p < 0.05).
The Effect of MTX on Developmental Morphology
Since oviposition was dramatically affected by MTX treatment, the development and morphology of follicles and eggs from females exposed to 5 ppm MTX was examined further. Eggs were collected from females exposed to MTX and ovaries were dissected. After only 3 days of MTX exposure, females oviposited eggs with an abnormal appearance (48%; n = 46): the chorionic appendages were frequently shorter and malformed and size of the eggs was reduced compared to eggs oviposited by control females (Figs. 2a, 2d–2f), although there was some variability. After 4 days major developmental abnormalities in the chorionic appendages and outer chorion structure were observed in the majority of the MTX-exposed eggs (91%; n = 45) (Figs. 2g and 2j). In these experiments, no oviposition was observed after 5 days of treatment. Ovaries from both treated and control females appeared similar after 3 days of treatment (Figs. 2b and 2c), but after 4 days, ovaries from the treated females were smaller, with fewer large yolky follicles (100%; n = 12) (Figs. 2h and 2i). Many appeared to be similar to those of immature females, but often with abnormal pedicels and oviduct laterals. After 5 days of MTX treatment no visible developing follicles were seen in any ovary examined, and much of the tissue appeared flaccid and adherent (Fig. 2k).
MTX not only affected the egg stage of development, but also post-embryonic development. Although the hatch rate was low, the few larvae that survived maternal MTX-treatment showed abnormalities. Many larvae had melanotic tumors, evidenced by black patches of cells, indicative of cell death (Fig. 3a). The very few adults that successfully emerged from their pupal cases had showed a significantly high frequency of morphological abnormalities compared to control flies (p < 0.05). Approximately 88% (n = 120) of the flies showed abnormal tufts of bristles on the dorsal thorax, evident curvature in the legs, abnormalities of the eyes, and/or wing deformities (Fig. 3b; not shown).
The Effect of MTX on Gene Expression
Since significant abnormalities in follicle development were observed in females after 4–5 days of MTX treatment, the earlier effects of MTX on patterns of gene expression in females were examined. After only 3 days of MTX treatment, when the majority of eggs were morphologically normal, there were few significant changes in gene expression patterns in whole females (results not shown). As a result, RNA from ovaries of flies fed on control medium or medium containing MTX (5 ppm) was used to assess transcript abundance with Drosophila cDNA microarrays. Because the RNA was amplified prior to cDNA synthesis, only transcripts shown to be more or less abundant (with a log2 ratio of 1.8 and –1.4, respectively) for all 4 hybridization spots were considered. "Down-regulated" (8) and "up-regulated" (2) transcripts were chosen for further analysis (Table 1). To confirm these results, qRT-PCR was conducted using primers specific to each of the 10 genes. Most (7/8 less abundant and 2/2 of the more abundant sequences) were verified. Significantly, considering that amplification of RNA could possibly introduce bias into transcript abundance assays, the results are remarkably concordant (Table 1).
Of the "down-regulated" transcripts, four encoded cell cycle regulators (grapes and loki, Cyclins A and E), three encoded proteins involved in embryo development (stathmin, concertina, and Minichromosome maintenance 6), and one was for a heat response gene (Heat shock factor). Of the "up-regulated" transcripts, one encoded a cold response gene that appears after low temperature stress (Frost), and the other was for a specific RNA polymerase II transcription factor (zerknullt) (Table 1).
In an effort to confirm that changes in transcript abundance observed in the ovaries reflected MTX-sensitive reproductive or developmental changes, cell lines (derived from embryos) were examined by microarrays. In these experiments, which did not require the amplification of RNA, genes encoding transcripts that were considered more (19) or less (9) abundant showed log2 ratios of greater than 1.0 or less than –0.9, respectively (results not shown). None of these genes, representing those with the most dramatic changes in cell transcript levels, had been identified in ovary microarray experiments. However, when the cell line was examined by qRT-PCR using primers specific to the transcripts identified from ovary microarrays, 7/8 of the "down-regulated" genes were also confirmed in the cells (Table 3), with the exception of Cyclin E transcripts, which were unchanged.
Since sequences identified using microarrays from MTX-treated cells were largely concordant with transcript abundance in ovaries, sequences identified in cell line microarrays were assessed in ovaries. First, the changes in message abundance after MTX treatment of cultured cells were confirmed by qRT-PCR experiments on 28 chosen "up and down regulated" genes. Analysis showed that 25 were concordant (Tables 2 and 3). Of these 25 sequences, the majority (9) appeared to be involved in defense or transport response (Glutathione S transferase E9, Drosomycin, meiotic 9, Thymine-DNA-glycosylase 1, Juvenile hormone epoxide hydrolase 3, Odorant-binding protein 99a, Cytochrome P450 12c1, Multiple drug resistance 65, and CG8032), 6 were involved in signal transduction or cell cycle regulation (Neuropeptide-like precursor 2, Tetraspanin 42E, innexin 3, cyclin-dependent kinase 2, Stem-loop binding protein, and CG5384), two were involved in protease inhibition (CG8066 and crammer), and the rest were associated with a variety of metabolic functions including transcription (Ets at 21c), amino acid catabolism (CG9363), metamorphosis (Angiotensin converting enzyme), amino acid de-phosphorylation (MKP-like), receptor binding (Niemann-Pick type C-2), dorsal/ventral axis formation (spatzle), peroxisome biogenesis (CG7609), or of unknown function (CG9894). The majority (19/25) of the transcripts that were sensitive to MTX-toxicity in cultured cells also showed changes in message levels in ovarian samples, as assessed by qRT-PCR analysis of ovaries (Tables 2 and 3).
DISCUSSION
MTX directly targets DHFR and decreases folate pools. As a consequence, it indirectly inhibits other folate-dependant enzymes such as thymidylate synthase, glycineamide ribonucleotide transformylase, and aminoimidazole carboxamide ribonucleotide transformylase (Rader and Huennekens, 1973). Thus MTX is a potent inhibitor with toxic and teratogenic side effects (McGuire, 2003). The morphological abnormalities seen after MTX administration in mammals is remarkably conserved in Drosophila, despite the obvious differences in the invertebrate system. Given the increasing concern about the sacrifice of mammalian embryos in toxicological testing, it was our goal to explore the use of the D. melanogaster model to investigate the genes involved in MTX-induced developmental change. In order to demonstrate that D. melanogaster is indeed an excellent model for these studies, as well as other antifolate-induced birth defects, we have observed mortality, female fecundity, and development and examined transcript abundance in response to MTX in ovaries and cell lines.
Exposure of D. melanogaster to MTX (up to 20 ppm) did not appear to limit the life span of adults, but, if present at levels 5 ppm, it had severe effects on the female reproductive system with a reduction from 23 eggs/day on control medium to <1 egg/day after 4 days on 2–20 ppm MTX. The ovaries from treated (5 ppm MTX) females appeared flaccid with small, undeveloped follicles. Deposited eggs were smaller with short or absent chorionic appendages with an irregular surface pattern, which is imprinted by the surrounding follicle cells during oogenesis (Fig. 2). Of the few eggs that hatch to larvae, many developed melanotic tumors and/or die during later development (Fig. 3; not shown). The small number of adult survivors showed tumors, and severe defects including bristle, eye and leg abnormalities (Fig. 3), strikingly similar to those deformities documented in anti-folate treated mammals (Schardein, 2000). These results are perhaps not surprising considering MTX is used to inhibit rapidly dividing cells. The reduction in fertility appears to be specific to females since no reduction in oviposition was seen when males were exposed to aminopterin or MTX, prior to crossing to control females (Goldsmith and Frank, 1954; not shown).
As an approach to understanding the morphological abnormalities and the underlying molecular consequences of MTX treatment, microarray analyses were done on D. melanogaster females, ovaries, and cell lines. Analyses with whole females showed only a few significant changes in transcript abundance (not shown), but this is not unexpected since adult mortality was unaffected by MTX treatment and cells in the adult are largely post-mitotic. Ovary microarrays and confirming qRT-PCR experiments, on the other hand, revealed an interesting set of genes that appear to encode transcripts that were either increased or decreased in abundance as a result of MTX-treatment (Table 1). These transcripts correspond to proteins involved in cell cycle or DNA damage, embryo development and stress response. Analysis of embryonic cell lines confirmed the susceptibility of the majority of the sequences to MTX treatment (Tables 2 and 3).
Microarray and qRT-PCR analysis of S3 cells showed a large number of transcripts with expression levels that were perturbed by MTX treatment (not shown), and the majority showed the same response in ovaries, suggesting that 19 transcripts that encoded proteins with functions including cell cycle regulation, metabolism, and defense are affected by treatment with MTX (Tables 2 and 3). In none of these studies was Dhfr identified (not shown), which is not surprising as DHFR transcripts are only known to increase in abundance after gene amplification due to selection for resistance (Singer et al., 2000).
The aberrant morphologies including female sterility and developmentally induced melanotic tumors and appendage abnormalities are consistent with MTX-mediated inhibition of rapid cell proliferation. Few transcriptional changes could be detected by microarray analysis of whole flies, and since adults are largely post-mitotic, this suggests that only mitotic cells are susceptible to toxicity. During oogenesis, germ cells rapidly divide and follicle cells are continually renewed from mesodermal origin within the ovary. The inhibition of folate metabolism by MTX treatment, appears then to reduce cell proliferation, and this is reflected in the "down-regulation" of many cell cycle and growth related genes, such as grp, CycA, CycE, cdc2, Slbp, and CG5384, which had transcript levels reduced 0.7–2.7 times. For example, transcripts from the cell damage checkpoint gene, lok, were reduced 11-fold, indicating a dramatic reduction in the normal cell cycle in ovaries. As well, Mcm6, involved in chorion gene amplification, had a 4-fold reduction in mRNA levels after MTX treatment and other mRNAs for membrane receptors, transporters and/or translation associated factors also showed abundance changes in both ovaries and cell lines (Tables 2 and 3). Such inhibition, in turn, would be expected to result in a smaller numbers of oocytes, with limited growth support due to a lack of properly functioning nurse and follicle cells resulting in flaccid ovaries and undeveloped eggs (Fig. 1).
Reduction of cell proliferation during the disc development explains the abnormal adult phenotypes with some cell death resulting in larval melanotic tumors. It should be noted that some cell necrosis was observed in rat livers after MTX administration, but gene expression profiling, did not identify cell death markers (Huang et al., 2004). At the levels of MTX used, treated cells showed a decrease in proliferation (Affleck et al., unpublished), consistent with the changes in cell cycle-related transcripts seen here. Certainly MTX treatment must result in some cell stress as indicated by a 4-fold increase in the detoxifying enzyme GST message and a 4-fold decrease in multiple drug resistance transcripts. Microarray analyses of MTX-treated rodent tissues have noted changes in stress genes and cell cycle (including Cyclin E) genes (Ganter et al., 2005; Huang et al., 2004).
How does maternal MTX treatment cause such dramatic morphological abnormalities in progeny, which in vertebrates would be termed birth defects The teratogenic mechanism of MTX is clearly more complex than simple DHFR inhibition. MTX-induced developmental toxicity can be ameliorated, but not completely prevented, in rabbits using 1-(p-tosyl)-3,4,4-trimethylimidazolidine, a functional analog for THF-mediated one-carbon transfer (DeSasso and Goeringer, 1992). As well, an MTX-resistant form of DHFR was able to only partially protect transgenic murine embryos from the teratogenic effects of MTX (Sutton et al., 1998). The small size, prominent eyes, tufts of hair, and abnormal limbs characteristic of mammalian birth defects (Schardein, 2000) are strikingly similar to the small egg size, rough eyes, abberant bristles, and bent legs and wings observed here (Figs. 2 and 3 and not shown). It is likely then, that although DHFR is an important target for MTX-induced birth defects, there are secondary targets of the drug or other biochemical pathways that are also involved. Certainly, gene expression analysis in the experiments reported here reveals an unexpected complexity in the consequences of MTX treatment during development. The further exploration of altered transcript abundance in response to MTX (or other potential teratogens) combined with the use of "knock out" or "over expression" technologies in Drosophila, which cause no suffering to vertebrates, hold promise for the understanding of these developmental abnormalities. We hope that through this model system we can provide insight into mammalian birth defects, which may be expected to increase in frequency in the future, due to the recent elevated use of MTX and other antifolate therapies for ectopic pregnancies and other human disorders (Nguyen et al., 2002).
ACKNOWLEDGMENTS
A special thanks goes to Bob Temkin and Bill Newcomb for their help with the scanning electron micrographs. The microchip hybridization experiments were carried out by Harriet Feilotter (Queen's University Microarray Facility, Kingston, ON). This work was supported by a grant to V.K.W. (NSERC, Canada) and scholarships to J.G.A. (NSERC) and to K.N. (OGS; J. Armand Bombardier Scholarship Foundation).
REFERENCES
Anderson, T. F. (1951). Techniques for the preservation of three-dimensional structure in preparing specimens for the electron microscope. Trans. N.Y. Acad. Sci. 13, 130–134.
Aviles, A., Diaz-Maqueo, J. C., Talavera, A., Guzman, R., and Garcia, E. L. (1991). Growth and development of children of mothers treated with chemotherapy during pregnancy: Current status of 43 patients. Am. J. Hematol. 36, 243–248.
Badaev, S. A., Drozdovskaia, L. N., and Rapoport, I. A. (1992). [The bilateral asymmetry of the number of bristles in Drosophila occurring under the influence of methotrexate]. Izv. Akad. Nauk. SSSR Biol. 2, 291–295. [Abstract, article in Russian].
Bier, E., and Bodmer, R. (2004). Drosophila, an emerging model for cardiac disease. Gene 342, 1–11.
Brachat, A., Pierrat, B., Xynos, A., Brecht, K., Simonen, M., Brungger, A., and Hiem, J. (2002). A microarray-based, integrated approach to identify novel regulators of cancer drug response and apoptosis. Oncogene 21, 8361–8371.
Collins, P., and Rogers, S. (1992). The efficacy of methotrexate in psoriasis–a review of 40 cases. Clin. Exp. Dermatol. 4, 257–260.
Del Campo, M., Kosaki, K., Bennett, F. C., and Jones, K. L. (1999). Developmental delay in fetal aminopterin/methotrexate syndrome. Teratology 60, 10–12.
DeSasso, J. M., and Goeringer, G. C. (1991). Amelioration by leucovorin of methotrexate developmental toxicity in rabbits. Teratology 43, 201–205.
DeSasso, J. M., and Georinger, G. C. (1992). Methotrexate-induced developmental toxicity in rabbits is ameliorated by 1-(p-tosyl)-3,4,4-trimethylimidazolidine, a functional analog for tetrahydrofolate-mediated one-carbon transfer. Teratology 45, 271–283.
Fernandez, H., Yves, V. S. C., Pauthier, S., Audibert, F., and Frydman, R. (1998). Randomized trial of conservative laparoscopic treatment and methotrexate administration in ectopic pregnancy and subsequent fertility. Human Repro. 13, 3239–3243.
Fortini, M. E., and Bonini, N. M. (2000). Modeling human neurodegenerative diseases in Drosophila on a wing and a prayer. TIG 16, 161–167.
Ganter, B., Tugendreich, S., Pearson, C. I., Ayanoglu, E., Baumhueter, S. et al. (In press). Development of a large-scale chemogenomics database to improve drug candidate selection and to understand mechanisms of chemical toxicity and action. J. Biotechnol. 119, 219–224.
Goldsmith, E. D., and Frank, I. (1954). Sterility in the female fruit fly Drosophila melanogaster produced by the feeding of a folic acid antagonist. Am. J. Physiol. 171, 726–727.
Grotewiel, M. S., Martin, I., Bhandari, P., and Cook-Wiens, E. (2005). Functional senescence in Drosophila melanogaster. Ageing Res. Rev. 4, 372–397.
Horridge, G. A., and Tamm, S. L. (1969). Critical point drying for scanning electron microscopic studies of ciliary motion. Science 163, 817–818.
Huang, Q., Jin, X., Gaillard, E. T., Knight, B. L., Pack, F. D. et al. (2004). Gene expression profiling reveals multiple toxicity endpoints induced by hepatotoxicants. Mutat. Res. 549, 147–168.
Huennekens, F. M. (1994). The methotrexate story: A paradigm for development of cancer chemotherapeutic agents. Adv. Enzyme Regul. 34, 397–419.
Le Menn, A., Silber, J., and Goux, J. (1983). etude de l'effet de deux inhibiteurs de la synthese des nucleotides l'aminopterine et la fluorodesoxyuridine sur des souches sauvage et mutantes vestigial chez Drosophila melanogaster. Biol. Cell 49, 213–218.
McGuire, J. J. (2003). Anticancer antifolates: Current status and future directions. Curr. Pharm. Design 9, 2593–2613.
Nakazawa, F., Matsuno, H., Yudoh, K., Katayama, R., Sawai, T., Uzuki, M., and Kimura, T. (2001). Methotrexate inhibits rheumatoid synovitis by inducing apoptosis. J. Rheumatol. 28, 1800–1808.
Nguyen, C., Duhl, A. J., Escallon, C. S., and Blakemore, K. J. (2002). Multiple anomalies in a fetus exposed to low-dose methotrexate in the first trimester. Obstet. Gynecol. 99, 599–602.
Pellizzer, C., Belle, E., Adler, S., Hartung, T., and Bremer, S. (2004). Detection of tissue-specific effects by methotrexate on differentiating mouse embryonic stem cells. Birth Defects Research (Part B) 71, 331–341.
Potter, C. J., Turenchalk, G. S., and Xu, T. (2000). Drosophila in cancer research. An expanding role. Trends Genet. 16, 33–39.
Rader, J. I., and Huennekens, F. M. (1973). In The Enzymes, 3rd ed. (P. Boyer, Ed.), Vol. 9, pp. 197–223. Academic Press, New York.
Saitoe, M., Horiuchi, J., Tamura, T., and Ito, N. (2005). Drosophila as a novel animal model for studying the genetics of age-related memory impairment. Rev. Neurosci. 16, 137–149.
Schardein, J. L. (2000). In Chemically Induced Birth Defects, 3rd ed., pp. 559–611. Marcel Dekker, New York.
Scmid, B. P. (1984). Monitoring of organ formation in rat embryos after in vitro exposure to azathioprine, mercaptopuring, methotrexate or cyclosporin A. Toxicology 31, 9–21.
Sgro, C. M., and Partridge, L. (1999). A delayed wave of death from reproduction in Drosophila. Science 286, 2521–2524.
Shaw, E. B., and Steinback, H. L. (1968). Aminopterin-induced fetal malformation. Survival of infant after attempted abortion. Amer. J. Dis. Child. 115, 477–482.
Silber, J., Coste, A., Bazin, C., and Le Menn, A. (1985). Dihydrofolate reductase activity and resistance to aminopterin in various species of Drosophila. Mol. Gen. Genet. 200, 92–95.
Singer, M. J., Mesner, L. D., Friedman, C. L., Trask, B. J., and Hamlin, J. L. (2000). Amplification of the human dihydrofolate reductase gene via double minutes is initiated by chromosome breaks. Proc. Natl. Acad. Sci. U.S.A. 97, 7921–7926.
Skalko, R. G., and Gold, M. P. (1974). Teratogenicity of methotrexate in mice. Teratology 9, 159–164.
Sutton, C., McIvor, R. S., Vagt, M., Doggett, B., and Kapur, R. P. (1998). Methotrexate-resistant form of dihydrofolate reductase protects transgenic murine embryos from teratogenic effects of methotrexate. Pediatr. Dev. Path. 1, 503–512.
Takata, R., Toyomasa, K., Kanehira, M., Tsunoda, T., Shuin, T., Miki, T., Namiki, M., Kohri, K., Matsushita, Y., Fujioka, T., et al. (2005). Predicting response to methotrexate, vinblastine, doxorubicin, and cisplatin neoadjuvant chemotherapy for bladder cancers through genome-wide gene expression profiling. Clin. Cancer Res. 11, 2625–2636.
Turner, F. R., and Mahowald, A. P. (1976). Scanning electron microscopy of Drosophila embryogenesis. Dev. Biol. 50, 95–108.
Williamson, J. H., Browder, L., and Thompson, A. (1978). A method of collecting Drosophila embryos of known age. Dros. Inform. Serv. 53, 213.
Wise, C. M., Vuyuru, S., and Roberts, W. N. (1996). Methotrexate in nonrenal lupus and undifferentiated connective tissue disease – a review of 36 patients. J. Rheumatol. 23, 1005–1010.
Zachariae, H., Hansen, H. E., Sogaard, H., and Olsen, T. S. (1990). Kidney biopsies in methotrexate-treated psoriatics. Dermatologica 181, 273–276.(Joslynn G. Affleck, Katerina Neumann, Li)