pmc logo image
Logo of geneticsJournal URL: http://www.genetics.org/
Journal List > Genetics > v.172(3); Mar 2006

Formats:
Genetics. 2006 March; 172(3): 1665–1674.
doi: 10.1534/genetics.105.052753.
PMCID: PMC1456268
Copyright © 2006 by the Genetics Society of America
Analysis of Pyrimidine Catabolism in Drosophila melanogaster Using Epistatic Interactions With Mutations of Pyrimidine Biosynthesis and β-Alanine Metabolism
John M. Rawls, Jr.1
Molecular and Cellular Biology Group, Department of Biology, University of Kentucky, Lexington, Kentucky 40506
1Address for correspondence: Department of Biology, 101 T. H. Morgan Bldg., University of Kentucky, Lexington, KY 40506. E-mail: jrawls/at/uky.edu
Communicating editor: J. A. Birchler
Received October 24, 2005; Accepted November 20, 2005.
Small right arrow pointing to: This article has been cited by other articles in PMC.
Abstract
The biochemical pathway for pyrimidine catabolism links the pathways for pyrimidine biosynthesis and salvage with β-alanine metabolism, providing an array of epistatic interactions with which to analyze mutations of these pathways. Loss-of-function mutations have been identified and characterized for each of the enzymes for pyrimidine catabolism: dihydropyrimidine dehydrogenase (DPD), su(r) mutants; dihydropyrimidinase (DHP), CRMP mutants; β-alanine synthase (βAS), pyd3 mutants. For all three genes, mutants are viable and fertile and manifest no obvious phenotypes, aside from a variety of epistatic interactions. Mutations of all three genes disrupt suppression by the rudimentary gain-of-function mutation (rSu(b)) of the dark cuticle phenotype of black mutants in which β-alanine pools are diminished; these results confirm that pyrimidines are the major source of β-alanine in cuticle pigmentation. The truncated wing phenotype of rudimentary mutants is suppressed completely by su(r) mutations and partially by CRMP mutations; however, no suppression is exhibited by pyd3 mutations. Similarly, su(r) mutants are hypersensitive to dietary 5-fluorouracil, CRMP mutants are less sensitive, and pyd3 mutants exhibit wild-type sensitivity. These results are discussed in the context of similar consequences of 5-fluoropyrimidine toxicity and pyrimidine catabolism mutations in humans.
 
URACIL is degraded to β-alanine in three enzymatic steps: dihydropyrimidine dehydrogenase (DPD; E.C. 1.3.1.2), dihydropyrimidinase (DHP; E.C. 3.5.2.2), and β-alanine synthase (βAS; E.C. 3.5.1.6) (Figure 1Figure 1.). As a catabolic pathway, these enzymes are important to maintenance of proper pyrimidine levels in cells. In addition, these three enzymes constitute a significant pathway for biosynthesis of β-alanine and related compounds from uracil. The need to understand the catabolic and biosynthetic roles of this pathway is demonstrated by recent discoveries of human pathologies that appear to be associated with these enzymes. DPD deficiency among patients is associated with toxic reactions to 5-fluorouracil (5-FU), one of the most commonly prescribed chemotherapeutic agents used to treat cancer (van Kuilenburg et al. 2000; Gross et al. 2003; Kubota 2003; van Kuilenburg 2004). Furthermore, deficiencies for DPD, DHP, or βAS have been reported among individuals exhibiting a variety of clinical presentations, including convulsive and other neurological disorders (Hamajima et al. 1998; van Kuilenburg et al. 1999, 2002, 2003, 2005). The metabolic mechanisms that underlie these abnormalities are poorly understood, however, and it is unclear whether they result from perturbed homeostasis of pyrimidines, β-alanine, or β-aminoisobutyrate (derived by catabolism of thymine by these same enzymes) (van Kuilenburg et al. 2004).
Figure 1.
Figure 1.
Figure 1.
Interrelationships of uracil and β-alanine metabolic pathways in Drosophila. Loss-of-function mutations of de novo pyrimidine biosynthesis genes (e.g., rudimentary) produce a truncated wing phenotype that is normalized by enhanced pyrimidine salvage (more ...)
Erik Bahn and colleagues at the University of Copenhagen demonstrated that modified pyrimidine degradation can be studied in Drosophila through epistatic interactions with mutations of de novo pyrimidine biosynthesis (overview in Figure 1Figure 1.). The suppressor of rudimentary mutation [su(r)1; Bahn 1972] was identified as a recessive suppressor of the truncated wing phenotype exhibited by rudimentary (r) mutants in which de novo pyrimidine biosynthesis is blocked (Nørby 1970; Rawls and Porter 1979). Furthermore, su(r)1 animals are unusually sensitive to high dietary levels of pyrimidines and they are unable to convert uracil to dihydrouracil in vivo, suggesting that they lack the first enzymatic step of pyrimidine degradation, DPD (Bahn 1972; Strøman 1974). The Drosophila melanogaster genome database indicates a single gene for DPD (CG2194, Reg-3; Misra et al. 2002), the predicted map position of which corresponds closely to that of su(r) (Bahn 1972). One objective of this study was to establish that su(r), CG2194, and Reg-3 are the same gene and to determine the phenotype of clearly null, loss-of-function su(r) mutations.
Perturbations of uracil catabolism also can be studied through epistatic interactions with mutations affecting levels of the pathway end product, β-alanine. Uracil catabolism is the major source of β-alanine in flies (Ross and Monroe 1972) in which adult cuticle pigmentation is qualitatively dependent upon β-alanine levels in cuticle-forming cells (Wright 1987). For example, black (b) mutant animals display a uniformly black cuticle phenotype, low levels of β-alanine (Hodgetts 1972), and an abnormal increase in the catabolic enzyme β-alanine transaminase (Weber et al. 1992). The b phenotype is reversed by injection of uracil, dihydrouracil, ureidopropionate, or β-alanine (Hodgetts and Choi 1974; Jacobs 1974). Genetically, the b mutant phenotype is suppressed by a semidominant gain-of-function mutation of the r gene, rSu(b) (Pedersen 1982; Bahn and Søndergaard 1983). rSu(b) is a UTP feedback-insensitive mutation of the CAD protein that results in overproduction of pyrimidines, apparently enhancing β-alanine levels and thereby normalizing the pigmentation of rSu(b) b flies (Piskur et al. 1993; Simmons et al. 1999; Figure 1Figure 1.). Because rSu(b) suppression of b must be mediated by the pyrimidine degradation pathway, mutations of this pathway should be epistatic to the rSu(b) b interaction, a prediction that is confirmed by the black cuticle phenotype of su(r)1 rSu(b) b animals (Pedersen 1982).
The D. melanogaster genome database indicates a single gene for DHP (CG1411, CRMP; Gojkovic et al. 2000; Celniker et al. 2002). This gene was named CRMP on the basis of the sequence similarity of a CG1411 cDNA with chick CRMP-62, a protein implicated in neuronal growth cone dynamics (Goshima et al. 1995). In animals, a closely related gene family encodes DHP and CRMPs (Takemoto et al. 2000). A second objective of this study was to confirm that the single Drosophila DHP/CRMP gene, CG1411, encodes DHP and to develop an experimental system with which to dissect the role of this gene in the disparate processes of pyrimidine catabolism and neurogenesis.
A final objective was to comprehensively assess the roles of pyrimidine catabolism in animal development. Therefore, βAS mutations were sought at the D. melanogaster pyd3 gene (Gojkovic et al. 2000; Celniker et al. 2002), thereby permitting comparisons of loss-of-function mutations of all three pyrimidine catabolic enzymes.
MATERIALS AND METHODS
Drosophila strains and transgenes:
Most genes, markers, balancers, and strains are described at FlyBase (Drysdale et al. 2005; http://flybase.bio.indiana.edu). The EY04394 strain was provided by the Szeged Drosophila Stock Centre and contains a copy of the P{EPgy2} transposon inserted into the X chromosome, ~0.6 kb from the 3′-end of the DPD open reading frame (ORF) in CG2194 (Figure 2Figure 2.; Bellen et al. 2004). Exelixis provided the EP(3)3238 strain in which a copy of the P{EP} transposon (Rørth 1996) is inserted near the 5′-end of the CRMP gene (Figure 5Figure 5.; Liao et al. 2000). The Kyoto Drosophila Genetics Resource Center provided the NP2343 strain (Hayashi et al. 2002), containing a copy of the P{GawB} transposon (Brand and Perrimon 1993) inserted near the 5′-end of the pyd3 gene (Figure 8Figure 8.). Each of these transposons contains a w+ marker that was used to follow the element in crosses. The Df(3R)noi-B deficiency is deleted for noi and other centromere-proximal genes, including CRMP (Meyer et al. 1998). The Df(3R)dsx10M deficiency is deleted for pyd3 and neighboring genes (Baker et al. 1991). TM3, Sb P{r+} e is a standard TM3 balancer chromosome into which a copy of the P{r+} transposon has been inserted (described below).
Figure 2.
Figure 2.
Figure 2.
The su(r) gene region. Above the double horizontal line are locations of exons for the su(r) (FlyBase Reg-3–RC transcript) and CG12118 (FlyBase CG12118–RA transcript) genes, as well as the insertion site of the P{EPgy2}04394 (more ...)
Figure 5.
Figure 5.
Figure 5.
The CRMP gene region. Above the double horizontal line are locations of exons for the rheb (FlyBase Rheb–RA transcript) and CRMP (FlyBase CRMP–RA transcript) genes and the insertion site of the P{EP}3238 transposon (Rørth (more ...)
Figure 8.
Figure 8.
Figure 8.
The pyd3 gene region. Above the double horizontal line are locations of exons for the pyd3 gene (FlyBase pyd3–RA tanscript) and the insertion site of the P{GawB}2343 transposon (Celniker et al. 2002; Hayashi et al. 2002; FlyBase). (more ...)
Two P-element transposons containing rudimentary DNA were used. P{r+} [FlyBase designation P{rcSa}] contains a 11.5-kb r genomic DNA fragment from which r introns 3–7 have been deleted; thus, it encodes a normal r ORF under control of r 5′ and 3′ DNA sequences. P{rSu(b)} [FlyBase designation P{rSu(b).cSa}] is identical to P{r+} except that it contains the nucleotide substitution of the rSu(b) mutation; thus, it encodes a feedback-insensitive form of the r protein. Both of these transposons carry the w+ marker and are described elsewhere (Simmons et al. 1999).
The P{PYD1+} transgene (Figure 2Figure 2.) was created by inserting into the pCaSpeR4 vector (Thummel et al. 1988) a 7358-bp NsiI–KpnI wild-type DNA fragment of the BACR39L04 clone (Hoskins et al. 2000) that spans CG2194. The P{PYD2+} transgene (Figure 5Figure 5.) contains a 10,859-bp EagI–BamHI wild-type DNA fragment from the BACR24O24 clone inserted into pCaSpeR4. The P{PYD3+} transgene (Figure 8Figure 8.) contains a 3847-bp XhoI–PstI fragment from BACR07N24 inserted into pCaSpeR4.
P-element mobilization screens:
Deletion mutations of the su(r) gene were generated by P{Δ2-3}-mediated imprecise excision of the P{EPgy2} transposon in the EY04394 strain. y w P{EPgy2}EY04394 rC/Y; In(3LR)TMS, Sb P{Δ2-3}99B/+ males were crossed to C(1)DX, y f/Y females and the wings of non-Sb male progeny were examined. To avoid isolating clusters of identical mutations, these crosses were performed using two males per vial and no more than one event was eventually isolated from any one vial. Of 324 vials, approximately one-half (171) yielded at least one male with visibly normalized wings; these males were individually crossed to attached-X C(1)DX, y f/Y females to retest and establish new mutant lines. Forty-five of these lines produced normal- or virtually normal-winged males, indicating full suppression of r by the excision product, and were studied further; a few lines were isolated that produced partially normalized wings, but these were discarded.
Imprecise excision of the P{EP} transposon within the EP(3)3238 strain was used to create deletion mutations of the CRMP gene. w/Y; P{EP}3238/In(3LR)TMS, Sb P{Δ2-3}99B males were crossed to w rC; Df(3R)noi-B, e/TM3, Sb P{r+} e females [the P{r+} transgene fully complements the rC mutation, restoring female fertility] and the wings of non-Sb male progeny (i.e., w rC/Y; P{EP}?/Df(3R)noi-B, e) were examined for individuals with normalized wings. Exceptional males were mated with w rC; Df(3R)noi-B, e/TM3, Sb P{r+} e females to test for transmission of the suppression phenotype (seen as normalized wings of non-e progeny) and to establish strains. Because the wings of exceptional flies were invariably less than fully normalized (e.g., only partial suppression of r; see results) and because the r phenotype is inherently variable, this screen was clearly inefficient; numerous false positives were selected (i.e., failed to transmit the suppression phenotype) and subtle mutations were undoubtedly overlooked. However, a total of six independently derived suppressor mutations were isolated, three of which were characterized further: CRMPsupI2, CRMPsupI3, and CRMPsupB3.
The CRMPsupA4 mutation was obtained by EMS mutagenesis, using a similar screen strategy. w/Y; ri pp males (isogenic chromosome 3) were fed 25 mm EMS (Lewis and Bacher 1968) and crossed to w rC; Df(3R)noi-B, e/TM3, Sb P{r+} e females. Among the non-Sb male progeny of this cross, individuals bearing normalized wings were selected and crossed to w rC; Df(3R)noi-B, e/TM3, Sb P{r+} e females to retest the suppression mutations and to create mutant strains.
Mutations of pyd3 were created by imprecise excision of the P{GawB}2343 element of the NP2343 strain, selecting for mutations that blocked suppression of the black cuticle phenotype in rSu(b) b animals. w/Y; b; P{GawB}3238/In(3LR)TMS, Sb P{Δ2-3}99B males were crossed to w; b; Df(3R)dsx10M, P{rSu(b)}/TM3, Sb females and non-Sb progeny were examined for rare dark-cuticle individuals. Each presumptive new mutant fly was crossed to the maternal strain to retest and to create a balanced stock of the new mutation. Three new mutations, designated pyd3L2, pyd3L5, and pyd3L10, were obtained in this manner.
Determining lesions in mutant DNAs:
To determine the nature of the transposon excision events that generated each mutation, PCR analysis of mutant fly DNA was carried out using primers flanking the original insertion sites as well as primers from the P element borne by the progenitor chromosome. A battery of PCR reactions was performed to determine whether P-element DNA had been removed and, if so, the length of genomic DNA remaining at the site. PCR-amplified mutant genomic fragments were gel purified and sequenced using primers flanking the sites of the mutations (DNA Sequencing Facility of Cincinnati Childrens Hospital).
5-fluorouracil toxicity assay:
Ten pairs of parental adults were placed in vials containing standard culture medium (Lewis 1960) supplemented with 5-fluorouracil. After 3 days, the parents were discarded and emerging adult progeny were scored. Cultures were maintained at 25° throughout. For su(r) mutant tests, parents were su(r)/Y or P{EPgy2}EY0439/Y [non-su(r)] males and C(1)DX, y f/Y (su(r)+) females, producing progeny of the same genotypes, which should occur in equal numbers. For CRMP tests, parents were CRMP/TM3 or CRMP+/TM3 males and Df(3R)noi-B/TM3 females, producing hemizygous mutant [CRMP/Df(3R)noi-B] and heterozygous wild-type [CRMP/TM3 and Df(3R)noi-B/TM3] progeny (controls). These two classes of zygotes should be produced in a 1:2 ratio. pyd3 tests were performed similarly by crossing pyd3/TM3 or P{GawB}2343 (non-pyd3)/TM3 males and Df(3R)dsx10M/TM3 females, yielding pyd3 hemizygotes [pyd3/Df(3R)dsx10M] and heterozygous controls [pyd3/TM3 and Df(3R)dsx10M] at a predicted ratio of 1:2.
RESULTS
su(r)1 is a mutation of the Drosophila DPD gene:
The D. melanogaster genome database indicates a single DPD gene (CG2194) located within cytogenetic region 8D (Gojkovic et al. 2000; Celniker et al. 2002), a site that corresponds closely to the recombination map position (1-27.7) reported for the su(r)1 mutation by Bahn (1972). To test whether CG2194 and su(r) are the same gene, an attempt was made to “rescue” the su(r)1 mutant phenotype with a transgene containing a 7358-bp segment of wild-type DNA that spans the CG2194 genomic region (Figure 2Figure 2.). The P{PYD1+} construct was introduced into flies by germline transformation and crosses were performed to create su(r)1 rC animals with and without the transgene. Flies without the transgene had entirely normal wings [i.e., complete suppression of r by su(r); Figure 3BFigure 3. ], whereas their transgene-bearing siblings displayed severe rudimentary wing phenotypes, similar to that in Figure 3AFigure 3.. Complementation of the su(r)1 mutation by P{PYD1+} confirmed that CG2194 is the su(r) gene.
Figure 3.
Figure 3.
Figure 3.
Suppression of the rudimentary wing phenotype by su(r) and CRMPsup mutations. (A) Typical severe rudimentary wing of a rC animal. (B) Entirely wild-type wing of a su(r)1 rC animal. (C) Moderate rudimentary wing of a rC; CRMPsupB3 animal.
The su(r)1 mutation results from substitution of a highly conserved amino acid residue within DPD:
To identify the su(r)1 mutation, genomic DNA of mutant flies was PCR amplified and a 4591-bp segment was sequenced, extending from 596 bp 5′ to the apparent translation initiation codon for DPD to 609 bp 3′ to the translation termination codon (Figure 2Figure 2.). Comparison of su(r)1 DNA and the published sequence for wild-type CG2194 (Celniker et al. 2002) revealed 30 nucleotide differences, presumably including silent polymorphic differences between the two strains as well as the su(r)1 mutation itself. Of these, 20 single-nucleotide differences occur within introns and 3′-UTR sequence; none of these differences involve obviously significant residues (i.e., canonical sequences required for RNA splicing or polyadenylation). Of the 10 nucleotide differences identified within protein open reading frame sequences, 9 result in fully synonymous codons. The sole missense nucleotide difference was a G → A substitution that converts the codon for Gly203 of wild-type DPD sequence to Glu203 in su(r)1.
Gly203 is conserved among all known DPD protein sequences (Figure 4Figure 4.). Within the crystallographic structure of pig DPD (Dobritsch et al. 2001), the corresponding residue (pig Gly206) lies in a turn between β-strand and α-helical stretches within an NADPH-binding site of the protein. This residue lies in close proximity (within 4 Å) to the phosphates and ribose of substrate NADPH. Thus, a reasonable explanation of the su(r)1 mutation is that substitution of anionic glutamate in this region disrupts NADPH binding and DPD catalysis. However, these results do not distinguish whether su(r)1 is a partial or complete loss-of-function mutation.
Figure 4.
Figure 4.
Figure 4.
su(r)1 DNA contains a nonconservative substitution of an invariant amino acid residue. A segment of the DPD protein from a variety of species is aligned: Dictyostelium discoideum (AAQ11981), Caenorhabditis elegans (Q18164), pig (B54718), mouse (AAH39699), (more ...)
Isolation and properties of su(r) null mutations:
To isolate unambiguously null mutations of su(r), deletions were created by imprecise excision of a nearby P-element insertion, the P{EPgy2}04394 transposon that is inserted 639 bp 3′ to the translation termination codon of the DPD ORF (Figure 2Figure 2.; Bellen et al. 2004). The element was mobilized in a y w P{EPgy2}04394 rC chromosome (strongly rudimentary wing animals) and new su(r) mutations were identified among progeny as individuals with normal wings (i.e., suppression of the r phenotype). A total of 31 independently derived lines were isolated in most of which the y+ and w+ markers of the P{EPgy2} element were lost, indicating extensive rearrangement or loss of the P element from the chromosome. Subsequent PCR analysis of genomic DNA from these mutant lines confirmed that both ends of the transposon were excised in 16 lines, 15 of which display deletions ranging from 0.4 to 3.5 kb, all of which extend into su(r) DNA (data not shown). Four of these deletions, representing a variety of deletion lengths, were PCR amplified and sequenced (Figure 2Figure 2.). Each of these deletions contains a common end that corresponds to the insertion site of the P{EPgy2}04394 element, as well as a unique end that in each case is located within the su(r) gene. Each removes the DPD termination codon and varying extents of the open reading frame upstream from that codon. The longest deletion, su(r)C11, removes 56% of the DPD open reading frame.
For each deletion, hemizygous males and homozygous females are viable, fertile, and morphologically normal. A variety of crosses were performed to confirm that these new mutations [designated su(r)new] are recessive, loss-of-function alleles of su(r)1: su(r)1 rC/su(r)new rC females have normal wings, whereas su(r)+ rC/su(r)new rC females invariably display severe r wings; su(r)1/su(r)new; b/b; P{rSu(b)}/+ females have typical black cuticle, whereas su(r)+/su(r)new; b/b; P{rSu(b)}/+ females have normal cuticle pigmentation (suppressed black). Finally, the P{PYD1+} transgene invariably complements all su(r)new mutations: su(r)new rC/Y; P{PYD1+}/+ males have severely rudimentary mutant wings and su(r)new/Y; P{PYD1+} b/b; P{rSu(b)}/+ males have normal cuticle pigmentation.
Isolation of CRMP mutations that suppress rudimentary phenotypes:
DHP mutations were screened on the basis of the prediction that, like su(r) mutations, loss-of-function CRMP mutants might also accumulate elevated levels of pyrimidines and thereby suppress the r wing phenotype. Accordingly, genetic screens were carried out using the Df(3R)noi-B deficiency chromosome (which lacks the CRMP gene) to screen for new CRMPsup mutations. Screens of EMS-treated chromosomes resulted in recovery of one mutation, CRMPsupA4. Mobilization of the P{EP}3238 transposon, inserted near the predicted 5′-end of the CRMP gene (Figure 5Figure 5.; Liao et al. 2000), resulted in the recovery of three mutations: CRMPsupB3, CRMPsupI2, and CRMPsupI3. For each of these mutations, r; CRMPsup animals are viable, fertile, and display weakly rudimentary wings (Figure 3CFigure 3.) that are distinctly more normal than the wings of r; CRMP+ siblings (Figure 3AFigure 3.). Thus, in contrast to the complete suppression seen with su(r) mutations, CRMPsup mutations only partially suppress the rudimentary wing phenotype.
To confirm that these suppressor mutations are alleles of the CRMP gene, the P{PYD2+} transgene that includes 10.8 kb of wild-type genomic DNA spanning the CRMP gene was created (Figure 5Figure 5.). Each suppressor strain was crossed to create r; CRMPsup/Df(3R)noi-B flies with and without the transgene. For each suppressor mutation, flies receiving the transgene showed distinctly more severe r wings (indistinguishable from the wing in Figure 3AFigure 3.) than did siblings without the transgene (Figure 3CFigure 3.), confirming that these suppressors are mutations of the CRMP gene.
Sequencing of CRMP DNA identified each of these mutations. CRMPsupA4 contains a single A → T nucleotide substitution that corresponds to a Asp430 → Val change in the mutant DPD protein. The Asp430 residue lies within a strongly conserved segment of the wild-type protein and is invariant in all known DHP proteins, as well as in all known CRMP proteins (Figure 6Figure 6.). In both bacterial hydantoinase (Abendroth et al. 2002) and mouse CRMP-1 (Deo et al. 2004), the corresponding aspartate residue participates in ionic interactions that form a peptide loop structure within a dimer interface of the protein, suggesting that the CRMPsupA4 mutation may affect polypeptide dimerization.
Figure 6.
Figure 6.
Figure 6.
The CRMPsupA4 mutation encodes a nonconservative substitution of an invariant amino acid residue. A segment of the DHP and CRMP proteins from a variety of species is aligned: Pseudomonas aeruginosa (AAG03830), Saccharomyces kluyveri (AAF69237), Arabidopsis (more ...)
Each of the mutations created by mobilization of the P{EP}3238 transposon are deletions of the transposon and varying extents of flanking genomic DNA (Figure 5Figure 5.). All are deleted for the first apparent exon of the CRMP gene transcript, with the longest (CRMPsupI2) also removing the entire intergenic region between the neighboring rheb gene and CRMP. The CRMPsupI2 lesion fully complements lethal rheb mutations (data not shown), indicating that it affects only CRMP.
CRMPsup mutations are epistatic to rSu(b):
Crosses were performed to test whether CRMPsup mutations block suppression of the black body phenotype by the rSu(b) mutation. rSu(b) b CRMPsupA4 animals (Figure 7CFigure 7.) display cuticle pigmentation similar to that of b animals (Figure 7AFigure 7.), but invariably darker than that of rSu(b) b CRMP+ siblings (Figure 7BFigure 7.). The same result was obtained using the CRMPsupI2 allele. Cuticle pigmentation of rSu(b) b CRMPsup flies was indistinguishable from su(r) rSu(b) b flies, suggesting that the conversion of uracil to β-alanine in cuticle-forming cells is similarly impaired in su(r) and CRMPsup mutations.
Figure 7.
Figure 7.
Figure 7.
CRMPsup mutations block suppression of the black cuticle phenotype by rSu(b). (A) Typical black body exhibited by a w; b; CRMPsupA4/TM3 fly. (B) Suppressed black phenotype of a w; b; CRMPsupA4 P{rSu(b)}/TM3 fly. (C) The genotype w; b; (more ...)
Isolation of pyd3 mutations:
Whereas suppression of the r wing phenotype was a successful criterion for selection of su(r) and CRMP mutants, similar screens to isolate mutations in the pyd3 gene were unsuccessful. Thus, an alternative strategy was developed on the basis of the ability of pyrimidine degradation pathway mutations to block the interaction of rSu(b) and b. By screening for imprecise excision events of the P{GawB}2343 transposon in a b P{rSu(b)} background, excision derivatives were obtained that produced black cuticle flies in Df(3R)dsx10M hemizygotes. The pyd3La2 and pyd3Lb5 mutations are deletions extending from the transposon insertion site and removing, respectively, 1088 and 1338 bp of 5′ flanking sequences of the pyd3 gene (Figure 8Figure 8.); each of these mutations retains a portion of the P element at its original site. The pyd3Lb10 mutation is a deletion of 1732 bp extending 792 bp 5′ to the apparent pyd3 transcription start site and 941 bp into the transcribed region of the gene; thus the entire transcription initiation region is removed in the pyd3Lb10 mutation, including the N-terminal 60 codons of the βAS open reading frame.
Each of the new pyd3 mutations is viable and fertile as homozygotes and no morphological changes are evident in these flies. The ability of each mutation to disrupt the suppression of b by rSu(b) was fully complemented by a single copy of the P{PYD3+} transgene containing wild-type DNA spanning the pyd3 gene: b/b; P{rSu(b)} pyd3/pyd3 animals display dark cuticle, whereas P{PYD+} b/b; P{rSu(b)} pyd3/pyd3 animals have normalized cuticle. Thus, these mutations are indeed loss-of-function alleles of pyd3. However, pyd3 mutations do not suppress the r mutant wing phenoptype: the wings of rC/Y; pyd3/pyd3 and rC/Y; pyd3/+ males are indistinguishable and similar to those in Figure 3AFigure 3..
su(r), CRMP, and pyd3 mutants differ in their sensitivities to dietary 5-fluorouracil:
Experiments were performed to determine whether DPD, DHP, and βAS deficiencies in Drosophila result in enhanced toxicity to 5-FU by rearing animals on standard medium containing a range of concentrations of the compound and measuring survival to adulthood. To test DPD deficiency, su(r)/Y males were mated with attached-X females, crosses that yield the same male [su(r)] and female [su(r)+] zygotes. P{EPgy2}04394 flies [non-su(r)] were insensitive to 5-FU, displaying similar survival at all concentrations (Figure 9AFigure 9.). Substantial numbers of su(r) mutant males were observed in absence of the drug; however, very few su(r)1 and su(r)C11 males survived to adulthood in the presence of 10 or 30 μm 5-FU (Figure 9AFigure 9.). At the higher levels of the drug, the few su(r) escapers displayed thin, short bristles, whereas female siblings [su(r)+] displayed normal bristles. Clearly, su(r) mutations confer hypersensitivity to dietary 5-FU.
Figure 9.
Figure 9.
Figure 9.
Sensitivities of pyrimidine catabolism mutants to dietary 5-fluorouracil. Animals were reared on varying concentrations of 5-FU as described in materials and methods, and emerging adult progeny were counted. The mean number of surviving experimental (hatched (more ...)
To test DHP deficiency, heterozygous CRMPsup/TM3 and Df(3R)noi-B/TM3 parents were mated, a cross that should yield hemizygous CRMPsup/Df(3R)noi-B (mutant) and heterozygous TM3 controls (CRMP+) (TM3/TM3 zygotes die). At 0 and 10 μm 5-FU, CRMPsup mutants emerged in numbers similar to those of CRMP+ hemizygotes (Figure 9BFigure 9.); however, survival of CRMPsup mutant flies was diminished at 30 μm 5-FU. Most CRMPsup mutant animals that survived to adulthood on 30 μm 5-FU displayed thin, short bristles, similar to the phenotype observed for su(r) escapers grown on 5-FU. These results show that CRMPsup mutants are sensitive to 5-FU, but less sensitive than su(r) mutants.
βAS deficiency was assayed by crosses of pyd3/TM3 and Df(3R)dsx10M/TM3 flies, creating pyd3/Df(3R)dsx10M hemizygotes and TM3 heterozygotes. As shown in Figure 9CFigure 9., ,5-FU5Figure 5.-FU was not differentially toxic to these animals over the tested concentrations of the compound. Therefore, pyd3 mutants are distinctly less sensitive to dietary 5-FU than either su(r) or CRMPsup mutants.
DISCUSSION
Results provided here unambiguously identify su(r) to be the Drosophila DPD gene, as originally proposed by Bahn (1972). Loss-of-function su(r) mutations are readily identified by their epistatic interactions with mutations of de novo pyrimidine biosynthesis (e.g., suppression of the r mutant wing phenotype) (Bahn 1972; Figure 3Figure 3.) and cuticle pigmentation (e.g., blocking rSu(b) suppression of the b mutant phenotype) (Pedersen 1982). Also, dietary 5-FU is highly toxic to su(r) animals (Figure 9Figure 9.). Although more cryptic phenotypes such as behavior have not been critically measured, the normal morphology and fertility of DPD-deficient flies indicate that uracil conversion to β-alanine (and thymine conversion to β-aminoisobutyrate) is not essential for largely normal development. Either imbalances in these compounds are inconsequential or alternate homeostatic mechanisms that compensate for those imbalances exist. For example, reduced β-alanine levels result in the black mutant cuticle phenotype (Hodgetts 1972; Hodgetts and Choi 1974; Jacobs 1974) and over half of β-alanine in early housefly puparia has been shown to derive from uracil catabolism (Ross and Monroe 1972). The normal pigmentation of su(r), CRMPsup, and pyd3 mutant animals suggests that β-alanine derived from nonuracil sources (e.g., aspartate decarboxylation) is sufficient for normal pigmentation.
The consequences of DPD deficiency in flies are comparable to those seen in humans in which genetically determined DPD deficiency and numerous functionally significant polymorphisms have been described. Most cases have been identified among patients showing severe toxic reactions to pyrimidine analogs, especially 5-FU, which is one of the most frequently prescribed chemotherapeutic agents for the treatment of cancer (Gross et al. 2003; van Kuilenburg 2004). DPD deficiency also has been discovered among patients with clinical presentations ranging from convulsive disorders, motor and mental retardation, to other growth and development problems (van Kuilenburg et al. 1999, 2002). The causal relationships between DPD deficiency and 5-FU toxicity are clearly established (Heggie et al. 1987); however, the consequences of human DPD deficiency upon normal development and metabolism is unclear. It has been proposed that the etiology of DPD deficiency may be perturbations in uracil, β-alanine, and/or β-aminoisobutyrate homeostasis, but direct evidence is lacking (van Kuilenburg et al. 2004). Some DPD-deficient individuals are asymptomatic and the broad range of the clinical presentations of others suggests that DPD deficiency alone does not determine abnormality; rather, DPD deficiency may synergistically interact with other genetic and metabolic factors to yield clinically manifested disorders.
Transcripts of the D. melanogaster su(r) gene exhibit daily cycles of abundance in fly heads under the control of photoperiod, diet, and function of the period gene (called Dreg-3; Van Gelder et al. 1995). Similar rhythmic, cyclic variations have been reported for DPD and its mRNA in a variety of animals (Grem et al. 1997; Porsin et al. 2003). The biological role of such oscillations is unknown; however, their wide conservation suggests an important process. Drosophila offers a model system for understanding the basis and role of this phenomenon in animals.
The consequences of DHP deficiency and βAS deficiency in Drosophila appear to be similar to, although quantitatively different from, DPD deficiency. Like su(r) mutants, CRMPsup and pyd3 mutants develop essentially normally although, again, cryptic phenotypes may have been overlooked so far. CRMPsup and pyd3 mutations block rSu(b) suppression of the b phenotype to a degree indistinguishable from the action of su(r) mutations, a similarity that agrees with the prediction that all of these mutations of pyrimidine catabolic steps should equally block β-alanine formation from uracil (Figure 1Figure 1.). In contrast, CRMPsup and pyd3 mutants display distinctly milder phenotypes than do su(r) mutants for the other criteria described here: CRMPsup mutations only partially suppress the r wing phenotype (Figure 3Figure 3.) and pyd3 mutants display no obvious suppression; CRMPsup mutants show reduced 5-FU toxicity and pyd3 mutants are essentially unaffected (Figure 9Figure 9.). These differences are predicted by considering the specific metabolic lesions in these mutants: su(r) mutants should accumulate higher levels of uracil (the substrate of DPD) than CRMPsup mutants, which should accumulate a combination of dihydrouracil (DHP substrate) and some uracil (by reversible catalysis of DPD). Similarly, su(r) animals fed 5-FU should accumulate 5-FU only, whereas at least a fraction of ingested 5-FU should be converted to 5-fluorodihydrouracil in CRMPsup mutants; thus, 5-FU accumulation should be relatively reduced in CRMPsup animals. pyd3 animals should accumulate little or no 5-FU, thereby exhibiting essentially wild-type sensitivity to the drug. These results with Drosophila predict that similar differential 5-FU responses may arise in humans deficient for DPD, DHP, and βAS, possibly explaining why DPD deficiency is the prevalent pharmacogenetic disorder encountered in 5-FU therapy (van Kuilenburg 2004).
DHP is a member of a closely related protein family that includes bacterial hydantoinases and animal CRMP proteins (Takemoto et al. 2000; Gojkovic et al. 2003). CRMPs modulate axonal growth cone dynamics in cultured neurons through mediation of signal transduction and/or microtubule dynamics (reviewed in Castellani and Rougon 2002; Deo et al. 2004). In vertebrates, separate genes encode DHP and at least four CRMPs; however, the D. melanogaster genome contains a single gene (CRMP). Thus, it will be interesting to determine if and how this gene encodes both DHP and CRMP, two structurally similar yet functionally divergent proteins.
The CRMPsup mutants described here exhibit no obvious neurogenic or neurological defects and their phenotypes are entirely explained as consequences of DHP deficiency. However, a role for this gene in neurogenesis cannot be ruled out at present for at least two reasons. First, all of these CRMPsup mutations were obtained in screens in which DHP deficiency (suppression of the r wing phenotype) was the foremost criterion; thus, it is conceivable that these mutations might selectively disrupt DHP activity without disrupting CRMP function. A second possibility is that CRMP function is impaired in one or more of these mutations but that the CRMP loss-of-function phenotype is relatively cryptic. Elucidating DHP and CRMP functions of CRMP will require additional discerning mutations (e.g., unambiguous knockouts, gain-of-function mutations) and measurements of more subtle phenotypes among those mutations (e.g., growth cone dynamics in embryogenesis, behavior, and longevity).
Acknowledgments
Thanks go to the Bloomington Stock Center, the Kyoto Drosophila Genetic Resource Center, the Szeged Drosophila Stock Centre, Exelixis, E. Bahn, and D. Pauli for providing fly strains; to the BACPAC Resources Center for BAC clones; to Kentucky Biomedical Research Infrastructure Network students Tommy Antony, Joey Kolb, Jersy Napoleon, Shingirirai Nyamwanza, Lindsay Poling, Emily Steinmetz, Angie Thacker, and Ali Wright for assistance in the mutant screens. This work was partially supported by grant 2 P20 RR-16481 from the National Institutes of Health.
Notes
Sequence data from this article have been deposited with the EMBL/GenBank Data Libraries under accession nos. DQ027826 and DQ257367–DQ257377.
References
  • Abendroth, J., K. Niefind and D. Schomburg, 2002. X-ray structure of dihydropyrimidinase from Thermus sp. at 1.3 A resolution. J. Mol. Biol. 320: 143–156. [PubMed]
  • Bahn, E., 1972. A suppressor locus for the pyrimidine requiring mutant rudimentary. Dros. Inf. Serv. 49: 98.
  • Bahn, E., and L. Søndergaard, 1983. Suppression of the semidominant Suppressor of black by rudimentary mutants in Drosophila melanogaster. Hereditas 99: 309–310. [PubMed]
  • Baker, B., G. Hoff, T. Kaufman, M. Wolfner, and T. Hazelrigg, 1991. The doublesex locus of Drosophila melanogaster and its flanking regions: a cytogenetic analysis. Genetics 127: 125–138. [PubMed]
  • Bellen, H., R. Levis, G. Liao, Y. He, J. Carlson et al., 2004. The BDGP gene disruption project: single transposon insertions associated with 40% of Drosophila genes. Genetics 167: 761–781. [PubMed]
  • Brand, A., and N. Perrimon, 1993. Targeted gene expression as a means of altering cell fates and generating dominant phenotypes. Development 118: 401–415. [PubMed]
  • Castellani, V., and G. Rougon, 2002. Control of semaphorin signaling. Curr. Opin. Neurobiol. 12: 532–541. [PubMed]
  • Celniker, S., D. Wheeler, B. Kronmiller, J. Carlson, A. Halpern et al., 2002. Finishing a whole genome shotgun: release 3 of the Drosophila melanogaster euchromatic genome sequence. Genome Biol. 3: Research0079.1–0079.14. [PubMed]
  • Deo, R., E. Schmidt, A. Elhabazi, H. Togashi, S. Burley et al., 2004. Structural bases for CRMP function in plexin-dependent semaphorin3A signaling. EMBO J. 23: 9–22. [PubMed]
  • Dobritsch, D., G. Schneider, K. Schnakerz and Y. Linqvist, 2001. Crystal structure of dihydropyrimidine dehydrogenase, a major determinant of the pharmacokinetics of the anti-cancer drug 5-fluorouracil. EMBO J. 20: 650–660. [PubMed]
  • Drysdale, R., M. Crosby and FlyBase Consortium, 2005. FlyBase: genes and gene models. Nucleic Acids Res. 33: D390–D395. [PubMed]
  • Gojkovic, Z., K. Jahnke, K. Schnackerz and J. Piskur, 2000. PYD2 encodes 5,6-dihydropyrimidine amidohydrolase, which participates in a novel fungal catabolic pathway. J. Mol. Biol. 295: 1073–1087. [PubMed]
  • Gojkovic, Z., L. Rislund, B. Andersen, M. Sandrini, P. F. Cook et al., 2003. Dihydropyrimidine amidohydrolases and dihydroorotases share the same origin and several enzymatic properties. Nucleic Acids Res. 31: 1683–1692. [PubMed]
  • Goshima, Y., F. Nakamura, P. Strittmatter and S. Strittmatter, 1995. Collapsin-induced growth cone collapse mediated by an intracellular protein related to UNC-33. Nature 276: 509–514.
  • Grem, J., L. Yee, D. Venzon, C. Takimoto and C. Allegra, 1997. Inter-and intra-individual variation in dihydropyrimidine dehydrogenase activity in peripheral blood mononuclear cells. Cancer Chemother. Pharmacol. 40: 117–125. [PubMed]
  • Gross, E., T. Ullrich, K. Seck, V. Mueller, M. De Wit et al., 2003. Detailed analysis of five mutations in dihydropyrimidine denydrogenase detected in cancer patients with 5-fluorouracil-related side effects. Hum. Mutat. 22: 498–506.
  • Hamajima, N., M. Kuwaki, P. Vreken, K. Matsuda, S. Sumi et al., 1998. Dihydropyrimidinase deficiency: structural organization, chromosomal localization, and mutation analysis of the human dihydropyrimidinase gene. Am. J. Hum. Genet. 63: 717–726. [PubMed]
  • Hayashi, S., K. Ito, Y. Sada, M. Tanigushi, A. Akimoto et al., 2002. GETDB, a database compiling expression patterns and molecular locations of a collection of Gal4 enhancer traps. Genesis 34: 58–61. [PubMed]
  • Heggie, G., J. Sommadossi, D. Cross, W. Hurster and R. Diasio, 1987. Clinical pharmacokinetics of 5-fluorouracil and its metabolism in plasma, urine, and bile. Cancer Res. 47: 2203–2206. [PubMed]
  • Hodgetts, R., 1972. Biochemical characterization of mutants affecting the metabolism of β-alanine in Drosophila. J. Insect Physiol. 18: 937–947. [PubMed]
  • Hodgetts, R., and A. Choi, 1974. β-alanine and cuticle maturation in Drosophila. Nature 252: 710–711. [PubMed]
  • Hoskins, R., C. Nelson, B. Berman, T. Laverty, R. George et al., 2000. A BAC-based physical map of the major autosomes of Drosophila melanogaster. Science 287: 2271–2274. [PubMed]
  • Jacobs, M., 1974. β-alanine and adaptation in Drosophila. J. Insect Physiol. 20: 859–866. [PubMed]
  • Kubota, T., 2003. 5-Fluorouracil and dihydropyrimidine dehydrogenase. Int. J. Clin. Oncol. 8: 127–131. [PubMed]
  • Lewis, E., 1960. A new standard food medium. Dros. Inf. Serv. 34: 117–118.
  • Lewis, E., and F. Bacher, 1968. Method for feeding ethyl methane sulfonate (EMS) to Drosophila males. Dros. Inf. Serv. 43: 193.
  • Liao, G., E. Rehm and G. Rubin, 2000. Insertion site preferences of the P transposable element in Drosophila melanogaster. Proc. Natl. Acad. Sci. USA 97: 3347–3351. [PubMed]
  • Meyer, V., B. Oliver and D. Pauli, 1998. Multiple developmental requirements of noisette, the Drosophila homolog of the U2 snRNP-associated polypeptide Sp3a60. Mol. Cell. Biol. 18: 1835–1843. [PubMed]
  • Misra, S., M. Crosby, C. Mungall, B. Matthews, P. Hradecky et al., 2002. Annotation of the Drosophila melanogaster euchromatic genome: a systematic review. Genome Biol. 3: Research0083.1–0083.22. [PubMed]
  • Nørby, S., 1970. A specific nutritional requirement for pyrimidines in rudimentary mutants of Drosophila melanogaster. Hereditas 66: 205–214. [PubMed]
  • Pedersen, M., 1982. Characterization of an X-linked semi-dominant Suppressor of black Su(b) (1–55.5) in Drosophila melanogaster. Carlsberg Res. Commun. 47: 391–400.
  • Piskur, J., D. Kolbak, L. Søndergaard and M. Pedersen, 1993. The dominant Suppressor of black indicates that de novo pyrimidine biosynthesis is involved in the Drosophila tan pigmentation pathway. Mol. Gen. Genet. 241: 335–340. [PubMed]
  • Porsin, B., J. Formento, E. Filipski, M. Etienne, M. Francoual et al., 2003. Dihydropyrimidine dehydrogenase circadian rhythm in mouse liver: comparison between enzyme activity and gene expression. Eur. J. Cancer 39: 822–828. [PubMed]
  • Rawls, J., and L. Porter, 1979. Organization of the rudimentary wing locus in Drosophila melanogaster. I. The isolation and partial characterization of mutants induced with ethyl methanesulfonate, ICR-170 and X rays. Genetics 93: 143–161.
  • Rørth, P., 1996. A modular misexpression screen in Drosophila detecting tissue-specific phenotypes. Proc. Natl. Acad. Sci. USA 93: 12418–12422. [PubMed]
  • Ross, R., and R. Monroe, 1972. β-Alanine metabolism in the housefly, Musca domestica: studies on anabolism in the early puparium. J. Insect Physiol. 18: 1593–1597. [PubMed]
  • Simmons, A., J. Rawls, J. Piskur and J. Davidson, 1999. A mutation that uncouples allosteric regulation of carbamyl phosphate synthetase in Drosophila. J. Mol. Biol. 287: 277–285. [PubMed]
  • Strøman, P., 1974. Pyrimidine-sensitive Drosophila wing mutants: withered (whd), tilt (tt) and dumpy (dp). Hereditas 78: 157–168. [PubMed]
  • Takemoto, T., Y. Sasaki, N. Hamijima, Y. Goshima, M. Nonaka et al., 2000. Cloning and characterization of the Caenorhabditis elegans CeCRMP/DHP-1 and -2: Common ancestors of CRMP and dihydropyrimidinase? Gene 261: 259–267. [PubMed]
  • Thummel, C., S. Boulet and H. Lipshitz, 1988. Vectors for Drosophila P-element-mediated transformation and tissue culture transfection. Gene 74: 445–456. [PubMed]
  • Van Gelder, R., H. Bae, M. Palazzolo and M. Krasnow, 1995. Extent and character of circadian gene expression in Drosophila melanogaster: identification of twenty oscillating mRNAs in the fly head. Curr. Biol. 5: 1424–1436. [PubMed]
  • van Kuilenburg, A., 2004. Dihydropyrimidine dehydrogenase and the efficacy and toxicity of 5-fluorouracil. Eur. J. Cancer 40: 939–950. [PubMed]
  • van Kuilenburg, A., P. Vreken, N. Abeling, H. Bakker, R. Meinsma et al., 1999. Genotype and phenotype in patients with dihydropyrimidine dehydrogenase deficiency. Hum. Genet. 104: 1–9. [PubMed]
  • van Kuilenburg, A., J. Haasjes, D. Richel, L. Zoetekouw, H. Van Lenthe et al., 2000. Clinical implications of dihydropyrimidine dehydrogenase (DPD) deficiency in patients with severe 5-fluorouracil-associated toxicity: identification of new mutations in the DPD gene. Clin. Cancer Res. 6: 4705–4712. [PubMed]
  • van Kuilenburg, A., D. Dobritzch, R. Meinsma, H. Waterham, M. Nowaczyk et al., 2002. Novel disease-causing mutations in the dihydropyrimidine dehydrogenase gene interpreted by analysis of the three-dimentional protein structure. Biochem. J. 364: 157–163. [PubMed]
  • van Kuilenburg, A., R. Meinsma, B. Zonnenberg, L. Zoetekouw, F. Baas et al., 2003. Dihydropyrimidinase deficiency and severe 5-fluorouracil toxicity. Clin. Cancer Res. 9: 4363–4367. [PubMed]
  • van Kuilenburg, A. Stroomer, H. Van Lenthe, N. Abeling and A. Van Gennip, 2004. New insights in dihydropyrimidine denydrogenase deficiency: A pivotal role for β-aminoisobutyric acid? Biochem. J. 379: 119–124. [PubMed]
  • van Kuilenburg, A., R. Meinsma, E. Beke, B. Assmann, A. Ribes et al., 2005. β-Ureidopropionase deficiency: an inborn error of pyrimidine degradation associated with neurological abnormalities. Hum. Mol. Genet. 13: 2794–2801.
  • Weber, J., R. Bolin, M. Hixon and A. Sherald, 1992. β-Alanine transaminase activity in black and suppressor of black mutations of Drosophila melanogaster. Biochim. Biophys. Acta 1115: 181–186. [PubMed]
  • Wright, T., 1987. The genetics of biogenic amine metabolism, sclerotization, and melanization in Drosophila melanogaster. Adv. Genet. 24: 127–222. [PubMed]
Articles from Genetics are provided here courtesy of
Genetics Society of America

PubMed articles by these authors
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph