drosophila model of human inherited triosephosphate

Copyright � 2006 by the Genetics Society of AmericaDOI: 10.1534/genetics.106.063206

Drosophila Model of Human Inherited Triosephosphate IsomeraseDeficiency Glycolytic Enzymopathy

Alicia M. Celotto,1 Adam C. Frank,1 Jacquelyn L. Seigle and Michael J. Palladino2

Department of Pharmacology, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania 15261 and Pittsburgh Institute forNeurodegenerative Diseases, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania 15260

Manuscript received July 10, 2006Accepted for publication September 1, 2006

ABSTRACT

Heritable mutations, known as inborn errors of metabolism, cause numerous devastating humandiseases, typically as a result of a deficiency in essential metabolic products or the accumulation of toxicintermediates. We have isolated a missense mutation in the Drosophila sugarkill (sgk) gene that causesphenotypes analogous to symptoms of triosephosphate isomerase (TPI) deficiency, a human familialdisease, characterized by anaerobic metabolic dysfunction resulting from pathological missense mutationsaffecting the encoded TPI protein. In Drosophila, the sgk gene encodes the glycolytic enzyme TPI. Ouranalysis of sgk mutants revealed TPI impairment associated with reduced longevity, progressive locomotordeficiency, and neural degeneration. Biochemical studies demonstrate that mutation of this glycolyticenzyme gene does not result in a bioenergetic deficit, suggesting an alternate cause of enzymopathyassociated with TPI impairment.

METABOLIC defects resulting from inherited dis-orders cause numerous human diseases. Gly-

colytic enzymopathies result from a disturbance inanaerobic metabolism; however, these diseases remainpoorly understood. Familial triosephosphate isomerase(TPI) deficiency, an autosomal recessive disorder, hasbeen reported in numerous pedigrees and results inanemia, neuromuscular wasting, and reduced longevity(Schneider et al. 1965; Valentine 1966). The relation-ship of anemia, neuromuscular degeneration, and gly-colytic flux to disease pathogenesis is not clear, and ananimal model that captures salient features of TPIdeficiency has not been reported. Our studies demon-strate the utility of Drosophila sugarkill (sgk) mutants asa model of glycolytic enzymopathy.

TPI is a 26.5-kDa soluble protein responsible for theconversion of dihydroxyacetone phosphate into glycer-aldehyde-3-phosphate in glycolysis (Rieder and Rose

1959). The protein’s structure has been studied in yeast(Alber et al. 1981), chicken (Banner et al. 1975), bac-teria (Noble et al. 1993), trypanosome (Wierenga et al.1991), and human (Lu et al. 1984; Mande et al. 1994;Maquat et al. 1985) with a high degree of structuralsimilarity shared between the various reported struc-tures. TPI is considered a near perfect enzyme due to itscatalytic efficiency; the rate of catalysis is diffusion con-

trolled, suggesting the presence of strong selectivepressure throughout the gene’s evolution (Knowles

1977). TPI is known to exist functionally as a homo-dimer, and the dimer interaction sites have been wellcharacterized (Schneider 2000).

In humans, several mutations have been reportedthat result in TPI deficiency, a progressive disease thateventuates in neuromuscular failure, hemolytic anemia,increased susceptibility to infection, and prematuredeath. At least nine different TPI missense mutationsthat affect various positions throughout the encodedprotein have been identified as the cause of glycolyticenzymopathies (Daar et al. 1986; Daar and Maquat

1988; Chang et al. 1993; Watanabe et al. 1996; Arya

et al. 1997; Valentin et al. 2000) (supplemental FigureS1 at http://www.genetics.org/supplemental/). Hetero-zygosity of null human TPI alleles has been observed in5% of African American and 0.5% of Caucasian pop-ulations, although the disease is typically caused byhomozygous missense mutations (Mohrenweiser andFielek 1982; Mohrenweiser 1987). Although prenataldetection is available, there is no treatment for thisprogressive and devastating neurological disease (Arya

et al. 1996).TPI has been studied in Drosophila and its mRNA

is highly expressed during development and in adultanimals (Shaw-Lee et al. 1991). Interestingly, studies ofnaturally occurring Drosophila populations revealedallozyme polymorphisms, including TpiS (slow) andTpiF (fast) alleles, which indicate the rate of proteinanodal electrophoresis migration (Voelker et al. 1979).Research findings suggest that the TpiF polymorphism

1These authors contributed equally to this work.2Corresponding author: Pittsburgh Institute for Neurodegenerative

Diseases and Department of Pharmacology, University of PittsburghSchool of Medicine, 3501 Fifth Ave., BST3 7042, Pittsburgh, PA 15260.E-mail: [email protected]

Genetics 174: 1237–1246 (November 2006)

(Lys-173-Glu) may alter the ‘‘hinged lid’’ of the enzyme’sactive site; however, the functional and behavioral con-sequences of the fast and slow polymorphisms have notyet been characterized (Hasson et al. 1998).

We report a Drosophila mutation in the sgk gene thatwas originally isolated due to its temperature-sensitivelocomotor impairment (Palladino et al. 2002). Ourstudies demonstrate that reduced longevity and pro-gressive neural degeneration are associated with thetemperature-sensitive sgk1 fly mutation. Interestingly,some of the well-characterized mutations that causehuman TPI deficiency disease have been shown to betemperature sensitive, owing to thermal instability ofthe encoded protein (Daar et al. 1986; Chang et al.1993). Although TPI has a well-characterized biochem-ical function, the mechanism of disease pathogenesis isunclear. Our biochemical studies indicate that the mu-tation does not cause general bioenergetic impairment,suggesting that toxicity is associated with substratecatabolism, a secondary effect of reduced glycolysis in-dependent of total cellular energy availability, or lossof an unappreciated TPI function. The sgk mutant re-ported provides a useful and amenable genetic animalmodel to study anaerobic metabolism and human gly-colytic enzymopathy in vivo.

MATERIALS AND METHODS

Drosophila stocks and culture: Standard cornmeal molassesfly media was used. Flies were maintained at room temper-ature, unless otherwise noted. The sgk1 mutation is maintainedas a homozygous viable ve e sgk1 strain. Wild-type controls areve e homozygotes, unless otherwise noted. Flies were tetracy-cline treated to rid them of microbial pathogens, which wasverified by PCR as previously reported (Celotto et al. 2006).The mutant and control strains used all contained the morecommon TpiS variant (Oakeshott 1984).

Life span and behavior analysis: Life span analysis wasperformed as previously described (Palladino et al. 2003).Stress sensitivity (a.k.a., bang sensitivity) was assayed by vor-texing flies in a standard media vial for 20 sec and measuringthe length of paralysis, similar to a previously describedprotocol (Ganetzky and Wu 1982). Stress sensitivity (rateof recovery) was assessed by assigning a numerical score tothe time in seconds to regain normal locomotion using a10-sec incremented scale, such that 1 indicates ,10 sec,2 indicates 11–20 sec, through 20 indicating .191 sec. Tem-perature sensitivity was assayed as previously described at37� (Palladino et al. 2002, 2003).

Genetics—sugarkill positional cloning, deletion, and trans-genic rescue: The sgk1 mutant, previously named ND14, wasoriginally identified in a screen of mutants with behavioraldeficits for neurodegeneration (Palladino et al. 2002). Can-didate gene sequencing identified the mutation in the sgkgene. Transposon-mediated mutagenesis was utilized to deletethe sgk locus, using standard procedures. Eighty revertants ofPEPgy2EY03361 (located �100 nucleotides 59 relative to the startof exon 1) were identified as w� Dr1 offspring of w1118;;delta2-3Dr/PEPgy2 jump-start males. The white revertants werescreened by PCR to identify deletion and precise excisionevents. Genomic DNA was isolated using the tissue protocol ofQIAamp DNA (QIAGEN, Valencia, CA). The sgk genomic

locus was amplified using a standard 50-ml PCR reaction;amplicons were resolved by electrophoresis with a 1.0%agarose gel and visualized with ethidium bromide staining.We isolated one 1.6-kb deletion (removing �319 to 1288nucleotides relative to the start of exon 1) and numerousapparent precise excision events. Direct sequence of PCRamplicons from four homozygous viable excision events ver-ified the precise nature of these revertants (named sgkR1–4).The deletion amplicon generated from heterozygote animalswas gel purified, extracted, and directly sequenced to identifythe deletion end points. Using standard P-element-mediatedtransgenesis, UAS� sgk transgenes were generated from EcoRI–XhoI PCR amplicons derived either from the ve e sgk1 wild typeor from the ve e sgk1 genomic DNA corresponding to positions57–1371 in FBgn0003738. The second chromosome actinTGAL4 (P{Act5C-GAL4}25FO1) was used to drive expression inrescue experiments.

Quantitative real-time RT–PCR: RNA was isolated from 10flies (performed in triplicate for each genotype), using 200 mlTrizol (Invitrogen, San Diego). The chloroform-extractedsample was precipitated and the RNA pellet was resuspendedin 10 ml dH2O and quantified by spectrophotometer readingat 260 nm. Five micrograms of RNA were used to perform areverse transcription reaction (Superscript RT, Invitrogen).Quantitative real-time PCR [Bio-Rad (Hercules, CA) iCycler]was performed using standard techniques. Briefly, 2 ml ofcDNA, 400 nm each of forward and reverse primers, and 12.5ml 23 iQ SYBR Green Supermix in a total volume of 25 ml wereamplified. PCR specificity was evaluated by melting-curveanalysis and resolution on a 1% agarose gel. All quantitativePCR experiments were repeated three times. The data werenormalized using the mRNA expression levels of RP49. Foldchange (FC) was determined using the Bio-Rad equation,FC ¼ 2�D(DCt).

Paraffin histology: Heads were removed, probosci weredissected away, and heads were fixed overnight in Carnoy’sfixative as previously described (Palladino et al. 2002). Agedspecimens were obtained at the genotype’s median adult age(day of 50% survivorship, see Figure 3). Bodies were dissectedand fixed overnight in Carnoy’s fixative. Tissue was processedinto paraffin and 5-mm-thick sections were stained with he-matoxylin and eosin (H&E) using previously described meth-ods (Palladino et al. 2000).

Transmission electron microscopy: Indirect flight muscleand brain tissue were dissected from mutant and age-matchedcontrol animals and processed similarly to methods describedpreviously (Kawasaki et al. 1998). Briefly, fixation occurredovernight at 4� in a buffered saline solution containing 2.5%paraformaldehyde and 1.5% glutaraldehyde (primary fixa-tive). Tissue was postfixed in 1% osmium tetroxide solution,dehydrated in a graded ethanol series, and embedded in eponresin. Sections (65 nm) were obtained from a Reichert Ultra-cut ultramicrotome and stained with 4% uranyl acetate and2.7% lead citrate. The tissue was imaged on a JEOL (Akishima,Tokyo) 100CX transmission electron microscope. Stereologi-cal protocols were followed in examining tissue samples andnumerous representative images were collected from muscleand central brain tissue.

Bioenergetic analysis: Creatine (Cr), phosphocreatine(PCr), cyclocreatine (CCr), and lactic acid (LA) were mea-sured from whole-animal extracts using high performanceliquid chromatography (HPLC). This method is a modifica-tion of a previously reported method (Matthews et al. 1998).Shimadzu HPLC components were used: SIL-HTc autosampler, LC-20AD pump, CTO-20AC column oven, and SPD-M20A diode array detector. Animals were aged at the ap-propriate temperature, frozen in liquid nitrogen, and keptat �20�. Whole animals (50 per preparation) were counted,

1238 A. M. Celotto et al.

weighed, and homogenized in 200 ml 0.4 m perchloric acidusing an electric homogenizer. Each sample was neutralizedwith 25 ml 2 m K2CO3 and centrifuged (12000 3 g) at 4� for 10min. The supernatant was filtered through a PVDF 0.45-mmfilter and stored at �80� until analysis. Samples were diluted(1:2 in dH2O) and 25 ml was injected and separated on aWaters Atlantis dC18 150 3 4.6-mm 3 mm column with a 4.6 320-mm 3-mm Guard column (changed every 100 injections).Sample separation was achieved using a 10-mm NaH2PO4 pH2.5 mobile phase at a flow rate of 0.5ml/min and a columntemperature of 30� with a detection wavelength at 190 nm. Allstandards were linear over a 10- to 20-fold concentrationrange. Retention times (minutes) were as follows: PCr, 4.2; LA,4.4; Cr, 5.0; and CCr, 6.4.

Supplemental data: Supplemental data are provided athttp://www.genetics.org/supplemental/.

RESULTS

Positional cloning and molecular characterization ofa Drosophila neurodegeneration mutant: We have uti-lized forward genetic screens to identify genes requiredto maintain neuronal viability with age and have iden-tified several mutants with behavioral abnormalitiesand neurodegeneration (Palladino et al. 2002, 2003;Celotto et al. 2006). One such neurodegenerative mu-tant identified in this screen, ND14, was reported toreside proximal to cytological region 3–100 (Palladino

et al. 2002). All of the available transposons and mu-tations in the 99A–F cytological region complementedND14 for its temperature-dependent locomotor im-

pairment and adult longevity phenotypes. To clonethe affected gene we sequenced candidate genes in thecritical region including takR, axn, trp, and tpi. Onlythe tpi locus contained a nonpolymorphic change af-fecting a codon in the open reading frame of the gene(Figure 1). The tpi mutation in the ND14 strain is a Tto C transition that alters a highly conserved methio-nine (M) to a threonine (T) codon at position 80 inthe Drosophila glycolytic TPI protein. Due to the well-characterized function of the TPI protein in carbohy-drate metabolism and the reduced longevity defect, theND14 mutant was named sugarkill1 (sgk1).

To confirm that the sgk1 mutation causes a temperature-dependent defect in locomotion, we generated an in-dependent deletion of the sgk locus (Figure 2). Weisolated 80 transposase-mediated white revertants ofthe homozygous viable EP{gy2} P element. PCR ampli-fication of the sgk genomic locus demonstrated numer-ous apparent precise excisions and a 1.6-kb deletionthat removed much of the sgk genomic locus, includingtwo of its three constitutive exons (named sgkjs10, Figure2). Direct sequence of PCR amplicons verified fourprecise excisions sgkR1–R4 and the deletion end pointsof the sgkjs10 allele. The sgkjs10 mutation failed to com-plement sgk1 while all four molecularly characterizedrevertants and the parent EP{gy2} chromosome fullycomplement the sgk1 mutation and are homozygous vi-able. Unlike the homozygous-viable sgk1 mutation, sgkjs10

is lethal: Adult sgkjs10 homozygotes are not observed in

Figure 1.—Mutation in the sugarkill gene af-fects a conserved methionine in the TPI protein.(A) Direct sequence analysis of genomic PCR-amplified fragments revealed exactly one muta-tion that was not polymorphic in the mutantstrain. The mutation is a T to C transition that al-ters a methionine (M) codon (ATG) to a threo-nine (T) codon (ACG) affecting position 80 inthe TPI protein (CG2171-PB). (B) Amino acid se-quence alignment demonstrates the conserva-tion of coding potential at and near the site ofthe mutation. Red shading indicates a match withconsensus. Orange shading indicates a conser-vative substitution. (C) The TPI dimer depictedas a tube and worm structure on the basis ofthe 2.2-A crystallographic data of the humanTPI protein (Kinoshita et al. 2005). One mono-mer is colored blue and the other magenta. Yel-low indicates the amino acid, M80, affected ineach monomer by the sgk1 mutation. (D) Withthe amino acid side chains added (red is acidic,blue is basic) the location of the mutation atthe interface between the monomers is evident.Cn3D version 4.1 was used to generate structuresin C and D. Species included in the alignmentare: Anopheles gambiae (A.g.), Drosophila melano-gaster (D.m.), Mus musculus (M.m.), Rattus norve-gicus (R.n.), Gallus gallus (G.g.), Xenopus laevis(X.l.), and Homo sapiens (H.s.). The full proteinalignment is provided in supplemental Figure 1at http://www.genetics.org/supplemental/.

Glycolytic Disease in Drosophila 1239

our sgkjs10/TM3, KrTGAL4, UAS-GFP strain and non-fluorescent homozygotes are not identifiable beyondlarval stage 2 (data not shown).

Expression in sugarkill mutant strains: The sgk1 muta-tion is predicted to alter a methionine (M) to threonine(T) at position 80 in the protein, which is likely the basisof the genetic defect. Alternatively, it is possible that thismutation destabilizes sgk mRNA or that a noncodingmutation causes a reduction in expression and thiscauses the recessive sgk1 phenotypes. Semiquantitativereverse transcription real-time PCR (RT–rtPCR) wasused to determine sgk expression levels. This analysisdemonstrated that sgk expression was not reduced insgk1 homozygous animals (Figure 2C). In fact, the anal-ysis revealed a 2.2-fold increase in sgk expression in thesgk1 mutant strain, suggesting an increase in mRNAstability or the presence of a transcriptional compensa-tory mechanism. No significant difference was observedin sgkjs10 heterozygote animals from wild-type animals,consistent with the recessive nature of this mutation.

Progressive dysfunction in sugarkill mutants: Flies ho-mozygous for sgk1 exhibit impaired locomotion that is

antagonized by elevated temperature and physicalstress. To quantify the stress-sensitive impairment phe-notypes and determine if they are progressive, we ex-amined young and aged sgk mutants and age-matchedcontrols for stress-induced locomotor deficits. At alltemperatures examined—room temperature (RT), 25�,and 29�—the locomotor defect arising from mechanicalstress was progressive and increased significantly withthe age of the animals (Figure 2, D–F). Severe stress-induced locomotor impairment is evident much ear-lier in sgk mutant adults when maintained at 25� and29� (Figure 2, D–F, and data not shown). sgk1 homozy-gotes and sgk1/JS10 animals have a similar stress-sensitivephenotype.

Consistent with progressive decline associated withglycolytic enzymopathy, longevity is severely compro-mised in sgk mutants (Figure 3). Life spans of sgk1/sgkjs10

animals were compared to those of revertant and het-erozygous control animals. There was no difference be-tween the two control strains used or between sgk1/sgkjs10

and sgk1 homozygotes. Longevity was analyzed at RT, 25�,and 29�, which represent three normal physiologically

Figure 2.—Genomic organiza-tion, expression, and locomotorfunction of sugarkill. (A) The sgklocus contains three constitutiveexons (shaded boxes) with a pre-dicted translational start at the be-ginning of exon 2. The M80Tmutation resides in exon 2 (aster-isks) and the js10 allele specifi-cally removes the predicted sgkgene. (B) PCR amplicons resolvedon an agarose gel show a 1.6-kbdeletion resulting in a �500-bpproduct (arrowhead) in the js10heterozygote and the normal-sized product obtained from re-vertant strains (R1–R3). Standard(STD) is a 100-bp ladder, and thetop band is 1.5 kb. (C) mRNA ex-pression from sgk mutants andcontrol animals was analyzed us-ing RT-rtPCR. Threshold cycle isthe fractional cycle number atwhich the fluorescence reaches10 times the standard deviationof the baseline for each genotypeand is inversely related to expres-sion. sgk mRNA expression is in-creased 2.2-fold in sgk1 mutantsover wild-type control levels(Student’s t-test). There is no sig-nificant difference between sgkjs10

and control animals. Error shown is SEM, n¼ 3. (D–F) Progressive locomotor impairment in sgk mutants. Locomotor impairmentis quantified as the recovery time from stress-induced paralysis. Mechanical stress/hyperstimulation induces sustained paralysis insgk1/JS10 (shaded bars) mutant strains, which increases significantly with age relative to the sgk1/R2 (solid bars) control. Locomotorimpairment is progressive in flies stored at (D) 29�, (E) 25�, and (F) room temperature. A numerical score, representing time toregain normal locomotion, was assigned to each fly as described in materials and methods. sgk1/R2 and wild-type animals (datanot shown) are extremely tolerant to mechanical hyperstimulation. ***, significantly different from age-matched wild-type con-trols (one-way ANOVA test, determined in Prism 4.0b, P , 0.001). Error is SEM. A total of 15 animals per genotype were tested.sgk1/JS10 mutants were not tested beyond the days reported due to morbidity in the populations.


Figure 3.—Life span analysis and transgenicrescue of sugarkill. Life span curves of sgk [1]/[js10] (A) and sgk [1]/1 (B) at room tempera-ture (yellow), 25� (orange), and 29� (red) indi-cate reduced longevity in sgk mutant flies. (C)Statistical analysis of median age reveals that mu-tant (sgk[1]/[js10]) life spans are significantlyreduced compared to heterozygote (sgk[1]/1)and revertant (sgk[R2]/[js10]) control strains(Student’s t-test, **P , 0.0001). Median age isthe age coincident with 50% survivorship of thegenotype. Survival curves represent 4–12 inde-pendent populations per genotype. Error givenis SEM. (D) Expression of wild-type TPI rescuesthe severely impaired longevity defect of sgk1

mutants. Transgenic UAS-sgk were generated toexpress wild-type or mutant sgk proteins. Ubiqui-tous expression of these transgenes using the ac-tin promotor and the GAL 4 system (Act:GAL4)was used to test rescue of the sgk1 mutation. Ex-pression using sgk1 (green) was effective at rescu-ing sgk1. The GAL4 control (red) and expressionof a UAS-sgk1 transgene bearing the M80T muta-tion (blue) are insufficient to rescue sgk1. Errorrepresents SEM from four independent longevityexperiments (n . 50 animals per genotype).


relevant temperatures for Drosophila, and life span wassignificantly reduced at all temperatures examined (Fig-ure 3C). Life span is also antagonized by temperature:There were 55, 89, and 92% decreases in life span fromthat of control strains at RT, 25�, and 29�, respectively.

Genetic rescue of sugarkill: The sgk1 mutation causes astriking progressive locomotor impairment and markedreduction in longevity. UAS-sgk transgenes were ex-pressed using actinTGAL4, and animals were examinedfor locomotor function and life span in sgk1 homozygousmutants. Animals of the genotype UAS-sgk1; actinTGAL4;sgk1/sgk1 did not exhibit temperature- or stress-inducedlocomotor impairment; however, UAS-sgk1; CyO; sgk1/sgk1

(control lacking GAL4) and UAS-sgk1; actinTGAL4; sgk1/sgk1 (expressing mutant transgene) animals were notdifferent from sgk1 homozygotes (data not shown).

To test the veracity of the transgenic rescue, longevityof animals bearing the UAS-sgk transgenic constructs wasmeasured. sgk1 animals exhibit a striking 92% reductionin median life span when reared at 29�, suggesting thatthis would be a rigorous test of transgenic rescue. UAS-sgk1; CyO; sgk1/sgk1 (control lacking GAL4) and UAS-sgk1;actinTGAL4; sgk1/sgk1 (expressing mutant transgene)animals had severely reduced life spans, whereas UAS-sgk1; actinTGAL4; sgk1/sgk1 had normal longevity at 29�,demonstrating transgenic rescue of the life span defectof sgk1 with wild-type TPI expression (Figure 3D). These

data demonstrate that the M80T mutation is responsiblefor locomotor and life span impairment in sgk1. To-gether with the finding that sgk1 is less severe than sgkJS10,these data demonstrate that sgk1 is a hypomorphic, loss-of-function missense mutation.

Neurodegeneration associated with sugarkill muta-tion: Human TPI deficiency results in complex neuro-muscular dysfunction with neurodegeneration. The sgkmutant and control flies were examined for the pres-ence of neuropathology to determine the extent towhich the fly model of TPI deficiency recapitulated thisimportant characteristic of the human disease. Thebrains of young mutants were uniformly free of pathol-ogy (data not shown); however, sgk mutants developedmarked neuropathology that was evident by lightmicroscopic evaluation at their median age (Figure 4).Control flies did not exhibit neurodegeneration. Inter-estingly, animals reared at all three temperatures (RT,25�, and 29�) exhibited marked neuropathology intheir brain and thoracic ganglion that was not presentin control tissues (Figure 4). We saw no difference inneuropathology between sgk1 homozygotes and sgk1/js10

animals (data not shown). These data demonstrate thatneural dysfunction and degeneration are associatedwith sgk loss-of-function mutation in Drosophila.

Normal mitochondrial ultrastructure in sugarkill: Pre-vious studies have shown that stress-sensitive locomotor

Figure 4.—Neuropathology in Dro-sophila sugarkill mutants. H&E-stainedneural tissue of sgk1/sgkjs10 mutants andsgk1/R2 controls is shown. (A) Midbrainfrontal sections of mutant and controlanimals aged at 29�, 25�, and room tem-perature. sgk mutants show marked vac-uolar pathology throughout the centralbrain and optic lobes. High-magnifica-tion panels are�4.83 higher magnifica-tion of boxed neuropil regions anddemonstrate marked degeneration insgk that is never observed in controlbrain tissues. (B) Thoracic ganglion sec-tions of mutant and control flies aged at29�. Similar degeneration was seen inanimals aged at 25� and room tempera-ture. Mutant sgk tissue is from animalsaged to days 3 (29�), 7 (25�), and 22(RT), respectively. Control tissue is fromanimals aged to days 37 (29�), 70 (25�),and 80 (RT), respectively. Analysis didnot reveal neuropathology in agedsgk1/R2 or wild-type animals. In younganimals of all genotypes there is no evi-dence of neural pathology (data notshown). n $ 15 animals per genotype.Bar, 100 mm.


defects, reduced longevity, and neural dysfunction canresult from mitochondrial ultrastructural defects asso-ciated with the ATP61 mutation (Celotto et al. 2006).Due to the similarity in phenotypes between ATP61 andsgk1 we hypothesized that a similar ultrastructural defectin mitochondria may underlie sgk pathogenesis, possi-bly resulting from inefficient glycolysis and overrelianceon mitochondria. To determine whether the sgk muta-tion causes any obvious ultrastructural pathology, flightmuscle and brain tissue were examined using trans-mission electron microscopy (TEM), as previously de-scribed (Celotto et al. 2006). Interestingly, muscle andbrain mitochondria have normal ultrastructural mor-phology (Figure 5).

Bioenergetic analysis of sugarkill mutants: TPI hasa well-described function in glycolysis as the enzyme

that converts dihydroxyacetone phosphate (DHAP) toglyceraldehydes-3-phosphate (G3P). As predicted fromthis biochemical pathway, loss of TPI function wouldalter the products of glycolysis from 2 ATP, 2 NADH, and2 pyruvate to 1 NADH, 1 pyruvate, and 1 DHAP permolecule of glucose (supplemental Figure S2 at http://www.genetics.org/supplemental/). This suggests thatthe basis of the pathology may be due to bioenergeticimpairment, a toxic effect of DHAP accumulation orone of its catabolites, or a defect arising from reducedglycolytic flux independent of energy production.

To directly examine the hypothesis that sgk1 mu-tants have impaired cellular energy stores, the bioener-getic state of sgk mutants was measured using HPLCto analyze PCr, creatine Cr, and lactic acid LA levels.Phosphocreatine is used to buffer cellular ATP concen-tration and is an excellent indicator of the cellularbioenergetic state (Matthews et al. 1998). Acid lactatelevels were significantly lower in sgk1 mutants vs. age-matched controls, consistent with a reduced rate ofanaerobic metabolism. Surprisingly, the data also re-vealed a significant increase in phosphocreatine in sgk1

mutants (Figure 6). These data argue that the impairedlocomotion does not result from bioenergetic impair-ment and suggest that lack of locomotor function maybe the direct cause of increased PCr and PCr:Cr ratiosin sgk1 mutants. This suggests that a mechanism inde-pendent of general bioenergetic impairment is causingthe poor locomotion, such as toxicity associated withDHAP or its catabolites, or poor glycolytic flux.

DISCUSSION

Forward genetic screening has been used to isolateinformative mutants that model progressive neurode-generative diseases in Drosophila (Palladino et al.2002, 2003; Celotto and Palladino 2005; Celotto

et al. 2006). Here we report the isolation of a novelmutant in Drosophila that displays reduced life span,locomotor defects, and neurodegeneration throughoutthe central nervous system. The affected gene wascloned and we discovered a missense mutation causingan M80T substitution in the highly conserved TPI pro-tein. We have named this mutation sgk1 and believe itserves as a useful model of human genetic TPI defi-ciency disease.

Numerous mutations causing human TPI deficiencyexist, including those that are temperature sensitive andinterfere with the ability of the protein to form a stablehomodimer (supplemental Figure S1 at http://www.genetics.org/supplemental/). The location of the sgk1

mutation affects a conserved methionine residue resid-ing at the dimer interface. (Figure 1, supplemental Fig-ure S1). sgk1 is thought to be a hypomorphic mutation(mild, loss of function) consistent with an amino acidsubstitution from an uncharged, weakly hydrophobic

Figure 5.—Ultrastructural analysis of sgk1. Transmissionelectron microscopy of sgk1 mutants reveals no morphologicalaberrations of indirect flight muscle structure. Myofibrillar Zand M lines appear intact and exist in consistent intervals inboth mutant and wild-type control tissue. Muscle mitochon-dria from sgk1 do not display overt, aberrant morphologyand are indistinguishable from control mitochondria. Mito-chondria from brain also appear normal in mutant and con-trol samples. Muscle micrographs are from day 5 adults. Brainmicrographs are from day 3 sgk mutants and days 3–10 controladults. Bar, 500 nm. N $ 3, per genotype.


residue to an uncharged polar residue. The TPI pro-tein has evolved to a state of catalytic perfection withdiffusion-controlled reaction rates, suggesting a strongselective advantage associated with efficient enzymefunction. Interestingly, disease results from specificrecessive, loss-of-function, missense mutations that ap-pear to be hypomorphic for protein function. Miceseverely deficient for TPI function have been reportedto die at early postimplantation stages of development(Merkle and Pretsch 1989), while heterozygotes ofTPI mutations believed to be null do not appear toeffectively model human enzymopathic disease (Zingg

et al. 1995).Our biochemical analysis indicates that bioenergetic

impairment is not responsible for the decreased lifespan or locomotor deficits associated with sgk1. Specif-ically, phosophocreatine was not reduced in sgk mutantsbut was surprisingly found to be significantly increased,possibly resulting from reduced animal activity (Figure6). Also, sgk1 mutants do not have aberrant mitochon-drial ultrastructure (Figure 5) as observed in anotherstress-sensitive fly mutant (Celotto et al. 2006) andsuggested by previous studies using human tissues(Bardosi et al. 1990). Lactic acid levels were found tobe decreased, suggesting a reduction in the overall rateof glycolysis, but this does not result in a depletion ofphosphocreatine levels that would be consistent withbioenergetic impairment (Figure 6). These data areconsistent with unchanged ATP levels that were seen inerythrocytes derived from human TPI-deficient patients(Eber et al. 1991). Without TPI function there is be-lieved to be no net ATP production from the anaerobic

metabolism of glucose and production of intermediateDHAP (supplemental Figure S2 at http://www.genetics.org/supplemental/). Although increased DHAP levelshave been reported in human erythrocytes (Zanella

et al. 1985; Eber et al. 1991; Hollan et al. 1997; Karg

et al. 2000; Olah et al. 2002, 2005), there is no directevidence that this molecule is toxic and data suggestthat it can be shunted into lipid biosynthetic pathways(Olah et al. 2002). TPI mutation resulting in reducedglycolytic flux is predicted to impair an organism’s abil-ity to function under conditions of oxygen debt andcause overreliance on mitochondrial ATP production.However, lack of ultrastructural changes in brain or mus-cle mitochondria suggests that pathogenesis is likelynot of mitochondrial origin. The mechanistic conse-quences of reduced glycolytic flux and an inability toenter oxygen debt (relying on anaerobic metabolism)are not well understood but appear to underlie thepathogenesis associated with TPI impairment.

Mutations causing human TPI deficiency are com-plex and result in a severe disease pathogenesis markedby progressive neuromuscular impairment and reducedlife expectancy. Human TPI disease conditions are typ-ically caused by specific homozygous missense muta-tions that often encode thermolabile proteins (Daar

et al. 1986; Chang et al. 1993). The recessive nature andtemperature sensitivity of the sgk1 mutant, whose phe-notypes are closely analogous to human disease symp-toms, suggest that this mutant will effectively model TPIenzymopathic disease and provide an amenable geneticmodel system for studying the basic mechanisms ofdisease pathogenesis.

Figure 6.—Bioenergetic analy-sis of sgk1. HPLC assay of creatine(Cr), phosphocreatine (PCr), andlactic acid (LA) in sgk1 mutantsand age-matched control animalsis shown. (A) There is a significantincrease in the nanomoles of PCrper milligram of total protein insgk[1] mutants. (B) A nonsignifi-cant decrease in the nanomolesof Cr per milligram of total pro-tein is seen in sgk[1] mutants.(C) There is a significant decreasein the nanomoles of LA per milli-gram of total protein in sgk[1]mutants. (D) The ratio betweenPCr:Cr is significantly increasedin sgk[1] mutants vs. controls.The ratio in wild-type flies is simi-lar to that seen in mice (Klivenyi

et al. 2004). (E) The TCr (totalnanomoles of creatine, phos-phocreatine, and cyclocreatine)remains constant in both geno-types tested. N $ 4 for each geno-type. Error shown is SEM.Significance is determined by Stu-dent’s t-test, *P , 0.05, **P , 0.01.


We thank Sunil Iyer for assistance generating the js10 allele,Nicholas Ierovante for assistance with histology and life span analyses,University of Pittsburgh Biomedical Research Support Facilities forassistance with sequencing, the Pittsburgh Center for the Environ-mental Basis of Human Disease for support, Simon Watkins and theCenter for Biologic Imaging for access to microscopes and histologyequipment, the Bloomington Stock Center for fly strains, and SusanAmara for use of the real-time PCR machine. We thank the NationalInstitutes of Health (NIH) National Institute of Neurological Disor-ders and Stroke (T32NS 07391-07) (A.M.C.), the Pittsburgh Institutefor Neurodegenerative Diseases [research grant 023RA02 (M.J.P.)],the American Heart Association [award 0630344N (M.J.P.)], thePittsburgh Foundation and Emmerling Fund [grant M2005-0068(M.J.P.)], the NIH [grant AG025046 (M.J.P.)], and the University ofPittsburgh Department of Pharmacology and School of Medicine forfinancial support.

LITERATURE CITED

Alber, T., F. C. Hartman, R. M. Johnson, G. A. Petsko and D.Tsernoglou, 1981 Crystallization of yeast triose phosphateisomerase from polyethylene glycol. Protein crystal formation fol-lowing phase separation. J. Biol. Chem. 256: 1356–1361.

Arya, R., M. R. Lalloz, K. H. Nicolaides, A. J. Bellingham andD. M. Layton, 1996 Prenatal diagnosis of triosephosphateisomerase deficiency. Blood 87: 4507–4509.

Arya, R., M. R. Lalloz, A. J. Bellingham and D. M. Layton,1997 Evidence for founder effect of the Glu104Asp substitutionand identification of new mutations in triosephosphate isomer-ase deficiency. Hum. Mutat. 10: 290–294.

Banner, D. W., A. C. Bloomer, G. A. Petsko, D. C. Phillips, C. I.Pogson et al., 1975 Structure of chicken muscle triose phos-phate isomerase determined crystallographically at 2.5 angstromresolution using amino acid sequence data. Nature 255: 609–614.

Bardosi, A., S. W. Eber, M. Hendrys and A. Pekrun, 1990 Myop-athy with altered mitochondria due to a triosephosphate isomer-ase (TPI) deficiency. Acta Neuropathol. 79: 387–394.

Celotto, A. M., and M. J. Palladino, 2005 Drosophila: a ‘‘model’’model system to study neurodegeneration. Mol. Interv. 5: 292–303.

Celotto, A. M., A. C. Frank, S. W. McGrath, T. Fergestad, W. A.Van Voorhies et al., 2006 Mitochondrial encephalomyopathyin Drosophila. J. Neurosci. 26: 810–820.

Chang, M. L., P. J. Artymiuk, X. Wu, S. Hollan, A. Lammi et al.,1993 Human triosephosphate isomerase deficiency resultingfrom mutation of Phe-240. Am. J. Hum. Genet. 52: 1260–1269.

Daar, I. O., and L. E. Maquat, 1988 Premature translation termi-nation mediates triosephosphate isomerase mRNA degradation.Mol. Cell. Biol. 8: 802–813.

Daar, I. O., P. J. Artymiuk, D. C. Phillips and L. E. Maquat,1986 Human triose-phosphate isomerase deficiency: a singleamino acid substitution results in a thermolabile enzyme. Proc.Natl. Acad. Sci. USA 83: 7903–7907.

Eber, S. W., A. Pekrun, A. Bardosi, M. Gahr, W. K. Krietsch et al.,1991 Triosephosphate isomerase deficiency: haemolytic anae-mia, myopathy with altered mitochondria and mental retarda-tion due to a new variant with accelerated enzyme catabolismand diminished specific activity. Eur. J. Pediatr. 150: 761–766.

Ganetzky, B., and C. F. Wu, 1982 Drosophila mutants with oppos-ing effects on nerve excitability: genetic and spatial interactionsin repetitive firing. J. Neurophysiol. 47: 501–514.

Hasson, E., I. N. Wang, L. W. Zeng, M. Kreitman and W. F. Eanes,1998 Nucleotide variation in the triosephosphate isomerase(Tpi) locus of Drosophila melanogaster and Drosophila simu-lans. Mol. Biol. Evol. 15: 756–769.

Hollan, S., M. Magocsi, E. Fodor, M. Horanyi, V. Harsanyi et al.,1997 Search for the pathogenesis of the differing phenotype intwo compound heterozygote Hungarian brothers with the samegenotypic triosephosphate isomerase deficiency. Proc. Natl.Acad. Sci. USA 94: 10362–10366.

Karg, E., I. Nemeth, M. Horanyi, S. Pinter, L. Vecsei et al.,2000 Diminished blood levels of reduced glutathione and

alpha-tocopherol in two triosephosphate isomerase-deficientbrothers. Blood Cells Mol. Dis. 26: 91–100.

Kawasaki, F., A. M. Mattiuz and R. W. Ordway, 1998 Synapticphysiology and ultrastructure in comatose mutants define anin vivo role for NSF in neurotransmitter release. J. Neurosci.18: 10241–10249.

Kinoshita, T., R. Maruki, M. Warizaya, H. Nakajima andS. Nishimura, 2005 Structure of a high-resolution crystal formof human triosephosphate isomerase: improvement of crystalsusing the gel-tube method. Acta Crystallograph. Sect. F Struct.Biol. Cryst. Commun. 61: 346–349.

Klivenyi, P., N. Y. Calingasan, A. Starkov, I. G. Stavrovskaya,B. S. Kristal et al., 2004 Neuroprotective mechanisms of crea-tine occur in the absence of mitochondrial creatine kinase.Neurobiol. Dis. 15: 610–617.

Knowles, J. R., 1977 Perfection in enzyme catalysis: the energetics iftriosephosphate isomerase. Accounts Chem. Res. 10: 105–111.

Lu, H. S., P. M. Yuan and R. W. Gracy, 1984 Primary structure of hu-man triosephosphate isomerase. J. Biol. Chem. 259: 11958–11968.

Mande, S. C., V. Mainfroid, K. H. Kalk, K. Goraj, J. A. Martial

et al., 1994 Crystal structure of recombinant human triosephos-phate isomerase at 2.8 A resolution. Triosephosphate isomerase-related human genetic disorders and comparison with thetrypanosomal enzyme. Protein Sci. 3: 810–821.

Maquat, L. E., R. Chilcote and P. M. Ryan, 1985 Human triose-phosphate isomerase cDNA and protein structure. Studies of tri-osephosphate isomerase deficiency in man. J. Biol. Chem. 260:3748–3753.

Matthews, R. T., L. Yang, B. G. Jenkins, R. J. Ferrante, B. R. Rosen

et al., 1998 Neuroprotective effects of creatine and cyclocrea-tine in animal models of Huntington’s disease. J. Neurosci. 18:156–163.

Merkle, S., and W. Pretsch, 1989 Characterization of triosephos-phate isomerase mutants with reduced enzyme activity in Musmusculus. Genetics 123: 837–844.

Mohrenweiser, H. W., 1987 Functional hemizygosity in the humangenome: direct estimate from twelve erythrocyte enzyme loci.Hum. Genet. 77: 241–245.

Mohrenweiser, H. W., and S. Fielek, 1982 Elevated frequency ofcarriers for triosephosphate isomerase deficiency in newborn in-fants. Pediatr. Res. 16: 960–963.

Noble, M. E., J. P. Zeelen, R. K. Wierenga, V. Mainfroid, K. Goraj

et al., 1993 Structure of triosephosphate isomerase from Escher-ichia coli determined at 2.6 A resolution. Acta Crystallogr. D Biol.Crystallogr. 49: 403–417.

Oakeshott, J., S. W. McKechnie and G. K. Chambers,1984 Population genetics of the metabolically related Adh,Gpdh, and Tpi polymorphisms in Drosophila melanogaster.Genetica 63: 21–29.

Olah, J., F. Orosz, G. M. Keseru, Z. Kovari, J. Kovacs et al.,2002 Triosephosphate isomerase deficiency: a neurodegenera-tive misfolding disease. Biochem. Soc. Trans. 30: 30–38.

Olah, J., F. Orosz, L. G. Puskas, L. Hackler, Jr., M. Horanyi et al.,2005 Triosephosphate isomerase deficiency: consequences ofan inherited mutation at mRNA, protein and metabolic levels.Biochem. J. 392: 675–683.

Palladino, M. J., L. P. Keegan, M. A. O’Connell and R. A. Reenan,2000 A-to-I pre-mRNA editing in Drosophila is primarily in-volved in adult nervous system function and integrity. Cell 102:437–449.

Palladino, M. J., T. J. Hadley and B. Ganetzky, 2002 Temperature-sensitive paralytic mutants are enriched for those causing neuro-degeneration in Drosophila. Genetics 161: 1197–1208.

Palladino, M. J., J. E. Bower, R. Kreber and B. Ganetzky,2003 Neural dysfunction and neurodegeneration in Drosoph-ila Na1/K1 ATPase alpha subunit mutants. J. Neurosci. 23:1276–1286.

Rieder, S. V., and I. A. Rose, 1959 The mechanism of the triose-phosphate isomerase reaction. J. Biol. Chem. 234: 1007–1010.

Schneider, A. S., 2000 Triosephosphate isomerase deficiency: his-torical perspectives and molecular aspects. Baillieres Best Pract.Res. Clin. Haematol. 13: 119–140.

Schneider, A. S., W. N. Valentine, M. Hattori and H. L. Heins, Jr.,1965 Hereditary hemolytic anemia with triosephosphate isom-erase deficiency. N. Engl. J. Med. 272: 229–235.


Shaw-Lee, R. L., J. L. Lissemore and D. T. Sullivan, 1991 Structureand expression of the triose phosphate isomerase (Tpi) gene ofDrosophila melanogaster. Mol. Gen. Genet. 230: 225–229.

Valentin, C., S. Pissard, J. Martin, D. Heron, P. Labrune et al.,2000 Triose phosphate isomerase deficiency in 3 French fami-lies: two novel null alleles, a frameshift mutation (TPI Alfortville)and an alteration in the initiation codon (TPI Paris). Blood 96:1130–1135.

Valentine, W. N., 1966 Hereditary hemolytic anemia with triose-phosphate isomerase deficiency. Am. J. Med. 41: 27–41.

Voelker, R. A., S. Ohnishi and C. H. Langley, 1979 Genetic andcytogenetic studies of four glycolytic enzymes in Drosophila mela-nogaster: aldolase, triosephosphate isomerase, 3-phosphoglyc-erate kinase, and phosphoglucomutase. Biochem. Genet. 17:769–783.

Watanabe, M., B. C. Zingg and H. W. Mohrenweiser, 1996 Molec-ular analysis of a series of alleles in humans with reduced activity

at the triosephosphate isomerase locus. Am. J. Hum. Genet. 58:308–316.

Wierenga, R. K., M. E. Noble, G. Vriend, S. Nauche and W. G. Hol,1991 Refined 1.83 A structure of trypanosomal triosephosphateisomerase crystallized in the presence of 2.4 M-ammonium sul-phate. A comparison with the structure of the trypanosomal tri-osephosphate isomerase-glycerol-3-phosphate complex. J. Mol.Biol. 220: 995–1015.

Zanella, A., M. Mariani, M. B. Colombo, C. Borgna-Pignatti,P. De Stefano et al., 1985 Triosephosphate isomerase defi-ciency: 2 new cases. Scand. J. Haematol. 34: 417–424.

Zingg, B. C., W. Pretsch and H. W. Mohrenweiser,1995 Molecular analysis of four ENU induced triosephosphateisomerase null mutants in Mus musculus. Mutat. Res. 328: 163–173.

Communicating editor: A. J. Lopez


drosophila model of human inherited triosephosphate

Documents