Specialized persister cells and the mechanism of multidrug tolerance in Escherichia coli

Bacterial populations produce persisters, cells that neither grow nor die in the presence of bactericidal agents, and thus exhibit multidrug tolerance (MDT). The mechanisms of MDT and the nature of persisters have remained elusive. Our previous research has shown that persisters are largely responsible for the recalcitrance of biofilm infections. A general method for isolating persisters was developed, based on lysis of regular cells by ampicillin. A gene expression profile of persisters contained toxin-antitoxin (TA) modules and other genes that can block important cellular functions such as translation. Bactericidal antibiotics kill cells by corrupting the target function (for example, aminoglycosides interrupt translation, producing toxic peptides). We reasoned that inhibition of translation will lead to a shutdown of cellular functions, preventing antibiotics from corrupting their targets, giving rise to MDT persister cells. Overproduction of the RelE toxin, an inhibitor of translation, caused a sharp increase in persisters. Functional expression of a putative HipA toxin also increased persisters, while deletion of the hipBA module caused a sharp decrease in persisters in both stationary and biofilm populations. HipA is thus the first validated persister-MDT gene. We suggest that random fluctuation in the levels of MDT proteins leads to the formation of rare persister cells. The function of these specialized dormant cells is to ensure the survival of the population in the presence of lethal factors.     

Cloning and nucleotide sequence of the glpD gene encoding sn-glycerol-3-phosphate dehydrogenase of Pseudomonas aeruginosa.

Nitrosoguanidine-induced Pseudomonas aeruginosa mutants which were unable to utilize glycerol as a carbon source were isolated. By utilizing PAO104, a mutant defective in glycerol transport and sn-glycerol-3-phosphate dehydrogenase (glpD), the glpD gene was cloned by a phage mini-D3112-based in vivo cloning method. The cloned gene was able to complement an Escherichia coli glpD mutant. Restriction analysis and recloning of DNA fragments located the glpD gene to a 1.6-kb EcoRI-SphI DNA fragment. In E. coli, a single 56,000-Da protein was expressed from the cloned DNA fragments. An in-frame glpD'-'lacZ translational fusion was isolated and used to determine the reading frame of glpD by sequencing across the fusion junction. The nucleotide sequence of a 1,792-bp fragment containing the glpD region was determined. The glpD gene encodes a protein containing 510 amino acids and with a predicted molecular weight of 56,150. Compared with the aerobic sn-glycerol-3-phosphate dehydrogenase from E. coli, P. aeruginosa GlpD is 56% identical and 69% similar. A similar comparison with GlpD from Bacillus subtilis reveals 21% identity and 40% similarity. A flavin-binding domain near the amino terminus which shared the consensus sequence reported for other bacterial flavoproteins was identified.     

Analysis of membrane stereochemistry with homology modeling of sn-glycerol-1-phosphate dehydrogenase

Different enantiomeric isomers, sn-glycerol-1-phosphate and sn-glycerol-3-phosphate, are used as the glycerophosphate backbones of phospholipids in the cellular membranes of Archaea and the remaining two kingdoms, respectively. In Archaea, sn-glycerol-1-phosphate dehydrogenase is involved in the generation of sn-glycerol-1-phosphate, while sn-glycerol-3-phosphate dehydrogenase synthesizes the enantiomer in Eukarya and Bacteria. The coordinates of sn-glycerol-3-phosphate dehydrogenase are available, although neither the tertiary structure nor the reaction mechanism of sn-glycerol-1-phosphate dehydrogenase is known. Database searching revealed that the archaeal enzyme shows sequence similarity to glycerol dehydrogenase, dehydroquinate synthase and alcohol dehydrogenase IV. The glycerol dehydrogenase, with coordinates that are available today, is closely related to the archaeal enzyme. Using the structure of glycerol dehydrogenase as the template, we built a model structure of the Methanothermobacter thermautotrophicus sn-glycerol-1-phosphate dehydrogenase, which could explain the chirality of the product. Based on the model structure, we determined the following: (1) the enzyme requires a Zn(2+) ion for its activity; (2) the enzyme selectively uses the pro-R hydrogen of the NAD(P)H; (3) the putative active site and the reaction mechanism were predicted; and (4) the archaeal enzyme does not share its evolutionary origin with sn-glycerol-3-phosphate dehydrogenase.     

Biosynthesis in Escherichia coli of sn-glycerol 3-phosphate, a precursor of phospholipid. Palmitoyl-CoA inhibition of the biosynthetic sn-glycerol-3-phosphate dehydrogenase.

Homogeneous biosynthetic sn-glycerol-3-phosphate dehydrogenase (EC 1.1.1.8) of Escherichia coli was potently inhibited by palmitoyl-CoA and other long chain acyl-CoA thioesters. The concentration dependence of this inhibition was not cooperative. Enzyme activity was inhibited 50% at 1 microM palmitoyl-CoA; thus, this inhibition occurred at concentrations below the critical micellar concentration of palmitoyl-CoA. Palmitoyl-CoA was a reversible, noncompetitive inhibitor with respect to both NADPH and dihydroxyacetone phosphate. Palmitoyl-CoA did not affect the quaternary structure of the enzyme. This inhibition could be prevented or reversed by the addition of phospholipid vesicles prepared from E. coli phospholipids. Palmitoyl-CoA did not alter the kinetics of inhibition by sn-glycerol 3-phosphate, which is a proven physiological regulator of this enzyme. Decanoyl-CoA, dodecanoyl-CoA, myristoyl-CoA, palmitoyl-(1,N6-etheno)CoA, stearoyl-CoA, and oleoyl-CoA inhibited sn-glycerol-3-phosphate dehydrogenase at concentrations below their critical micellar concentrations. Palmitate inhibited sn-glycerol-3-phosphate dehydrogenase activity 50% at 200 microM. Palmitoyl-carnitine, deoxycholate, taurocholate, and dodecyl sulfate were more potent inhibitors than Triton X-100, Tween-20, or Tween-80. Palmitoyl-acyl carrier protein at concentrations up to 50 microM had no effect on sn-glycerol-3-phosphate dehydrogenase activity. The possible physiological role of long chain fatty acyl-CoA thioesters in the regulation of sn-glycerol 3-phosphate and phospholipid biosynthesis in E. coli is discussed.     

Identification of sn-glycerol-1-phosphate dehydrogenase activity from genomic information on a hyperthermophilic archaeon, Sulfolobus tokodaii strain 7

sn-Glycerol-1-phosphate dehydrogenase is responsible for the formation of sn-glycerol-1-phosphate, the backbone of membrane phospholipids of Archaea. This activity had never been detected in cell-free extract of Sulfolobus sp. Here we report the detection of this activity on the thermostable ST0344 protein of Sulfolobus tokodaii expressed in Escherichia coli, which was predicted from genomic information on S. tokodaii. This is another line of evidence for the general mechanism of sn-glycerol-1-phosphate formation by the enzyme.     

Characterization of active and latent forms of the membrane-associated sn-glycerol-3-phosphate acyltransferase of Escherichia coli

The intrinsically active, sn-glycerol-3-phosphate acyltransferase present in membranes prepared from both wild type Escherichia coli and from strains which overproduce the enzyme can be kinetically distinguished from a latent enzyme species which is unmasked by solubilization and reconstitution. Both membrane-associated and solubilized/reconstituted enzyme preparations exhibited cooperativity with respect to sn-glycerol-3-phosphate and palmitoyl-coenzyme A substrates; positive cooperativity in membranes toward palmitoyl-coenzyme A (napp = 4) and negative cooperativity toward sn-glycerol-3-phosphate (napp = 0.75) were significantly altered upon solubilization and reconstitution. Since the degree of alteration increased with the amount of sn-glycerol-3-P acyltransferase present in the membranes, a detergent-dissociable homooligomerization of the sn-glycerol-3-phosphate acyltransferase was considered as an underlying mechanism. This possibility was investigated by changing the protein-to-Triton X-100 ratio of homogeneous enzyme prior to reconstitution and then analyzing the subsequent migration of samples on a Sephacryl S-300 sizing column. The elution positions were consistent with monomeric and dimeric polypeptide bound to micelles of Triton X-100. Hill coefficients for monomeric, reconstituted enzyme preparations were comparable to those obtained for the active, membrane-associated sn-glycerol-3-phosphate acyltransferase. The reduced cooperativity of dimeric, reconstituted enzyme preparations correlated closely to the Hill coefficient values obtained for latent, solubilized/reconstituted sn-glycerol-3-phosphate acyltransferase from membranes of Escherichia coli which overproduce the enzyme. The physiological significance of these findings is discussed.     

Transfer of pro-R hydrogen from NADH to dihydroxyacetonephosphate by sn-glycerol-1-phosphate dehydrogenase from the archaeon Methanothermobacter thermautotrophicus

sn-Glycerol-1-phosphate dehydrogenase is responsible for the formation of the sn-glycerol-1-phosphate backbone of archaeal lipids. [4-3H]NADH that had 3H at the R side was produced from [4-3H]NAD and glucose with glucose dehydrogenase (a pro-S type enzyme). The 3H of this [4-3H]NADH was transferred to dihydroxyacetonephosphate during the sn-glycerol-1-phosphate dehydrogenase reaction. On the contrary, in a similar reaction using alcohol dehydrogenase (a pro-R type enzyme), 3H was not incorporated into glycerophosphate. These results confirmed a prediction of the tertiary structure of sn-glycerol-1-phosphate dehydrogenase by homology modeling.     

Kinetic properties of a sn-glycerol-3-phosphate dehydrogenase purified from the unicellular alga Chlamydomonas reinhardtii

A sn-glycerol-3-phosphate dehydrogenase (sn-glycerol-3-phosphate:NAD+ 2-oxidoreductase, EC 1.1.1.8) has been purified from the unicellular green alga Chlamydomonas reinhardtii 3400-fold to a specific activity of 34 mumol/mg protein per min by a simple procedure involving two chromatographic steps on affinity dyes. The pH optimum for reduction of dihydroxyacetone phosphate was 6.8 and for glycerol 3-phosphate oxidation it was 9.5. In the direction of dihydroxyacetone phosphate reduction, the enzyme showed Michaelis-Menten kinetics. The enzyme reacted specifically with NADH and dihydroxyacetone phosphate as substrates with affinity constants of 16 and 12 microM, respectively. Product inhibition as well as competitive inhibition pattern indicated a random-bi-bi reaction mechanism for sn-glycerol-3-phosphate dehydrogenase from C. reinhardtii. The effective control of dihydroxyacetone reduction catalysed via this enzyme by ATP, Pi and NAD gave evidence for a physiological role of the enzyme in plastidic glycolysis.     

sn-glycerol-3-phosphate oxidase and alcohol tolerance inDrosophila melanogaster larvae

The role of sn-glycerol-3-phosphate oxidase (GPO; EC 1.1.99.5) in the variation of ethanol tolerance in Drosophila melanogaster was assessed in isofemale lines derived from individuals collected at the Chateau Tahbilk Winery and Wandin North Orchard of Victoria, Australia. When fed an undefined medium (semolina-treacle) with 6% ethanol (v/v), larvae of lines with high GPO activities survived better than did larvae of lines with low GPO activities. Although GPO was induced to higher activity levels by dietary ethanol in larvae of all the test lines, GPO activity was greater in lines representing the area outside the wine cellar. This implied that the cellar environment selected against individuals with high levels of GPO. These data do not explain the established difference in tolerance between cellar and outside populations. The GPO activities of lines were not dependent upon the activities of the lipogenic enzyme, glycerol-3-phosphate dehydrogenase; the major ethanol-degrading enzyme, alcohol dehydrogenase; or the citric acid cycle enzyme, fumarase. Thus, GPO activity is an important component of the metabolic mechanism of ethanol tolerance in larvae, but the mode of action of GPO has not been defined.     

sn-Glycerol-3-phosphate dehydrogenase and its interaction with nitrate reductase in wild-type and hem mutant strains of Staphylococcus aureus

Staphylococcus aureus has membrane-associated sn-glycerol-3-phosphate dehydrogenase activity that is strongly activated by detergents. The enzyme can be measured spectrophotometrically in intact cells in assay systems containing lauryldimethylamine oxide (Ammonyx LO). The dehydrogenase activity was located exclusively in the membrane fraction of cells grown with glycerol under aerobic conditions or under anaerobic conditions with the addition of nitrate; there was no evidence of multiple forms. Development of sn-glycerol-3-phosphate dehydrogenase activity was studied with suspensions of cells grown previously under semianaerobic conditions with glucose and nitrate. The wild-type strain rapidly formed the enzyme when incubated with glycerol under aerobic conditions or under semianaerobic conditions in the presence of nitrate. Under similar conditions, suspensions of hem mutant H-14 required the addition of hemin. Induction of the enzyme was strongly repressed by glucose with both organisms. A procedure was established to obtain cells of mutant H-14 with sn-glycerol-3-phosphate dehydrogenase and nitrate reductase activities, but which could not link the systems unless supplemented with hemin. The coupled activity could also be reconstructed in vitro by the addition of hemin to the depleted membranes.     

Cloning, sequence, and disruption of the Saccharomyces diastaticus DAR1 gene encoding a glycerol-3-phosphate dehydrogenase.

The Saccharomyces diastaticus DAR1 gene was cloned by complementation in an Escherichia coli strain auxogrophic for glycerol-3-phosphate. DAR1 encodes an NADH-dependent dihydroxyacetone phosphate reductase (sn-glycerol-3-phosphate dehydrogenase [G3PDase; EC 1.1.1.8]) homologous to several other eukaryotic G3PDases. DAR1 is distinct from GUT2, which encodes a glucose-repressed mitochondrial G3PDase, but is identical to GPD1 from S. cerevisiae, a close relative of S. diastaticus. The level of DAR1-encoded G3PDase was increased about threefold in a medium of high osmolarity. Disruption of DAR1 in a haploid S. cerevisiae was not lethal but led to a decrease in cytoplasmic NADH-dependent G3PDase activity, an increase in osmotic sensitivity, and a 25% reduction in glycerol secretion from cells grown anaerobically on glucose.     

Peptergents: peptide detergents that improve stability and functionality of a membrane protein, glycerol-3-phosphate dehydrogenase

Toward enhancing in vitro membrane protein studies, we have utilized small self-assembling peptides with detergent properties ("peptergents") to extract and stabilize the integral membrane flavoenzyme, glycerol-3-phosphate dehydrogenase (GlpD), and the soluble redox flavoenzyme, NADH peroxidase (Npx). GlpD is a six transmembrane spanning redox enzyme that catalyzes the oxidation of glycerol-3-phosphate to dihydroxyacetone phosphate. Although detergents such as n-octyl-beta-D-glucpyranoside can efficiently solubilize the enzyme, GlpD is inactivated within days once reconstituted into detergent micelles. In contrast, peptergents can efficiently extract and solubilize GlpD from native Escherichia coli membrane and maintain its enzymatic activity up to 10 times longer than in traditional detergents. Intriguingly, peptergents also extended the activity of a soluble flavoenzyme, Npx, when used as an additive. Npx is a flavoenzyme that catalyzes the two-electron reduction of hydrogen peroxide to water using a cysteine-sulfenic acid as a secondary redox center. The lability of the peroxidase results from oxidation of the sulfenic acid to the sulfinic or sulfonic acid forms. Oxidation of the sulfenic acid, the secondary redox center, results in inactivation, and this reaction proceeds in vitro even in the presence of reducing agents. Although the exact mechanism by which peptergents influence solution stability of Npx remains to be determined, the positive effects may be due to antioxidant properties of the peptides. Peptide-based detergents can be beneficial for many applications and may be particularly useful for structural and functional studies of membrane proteins due to their propensity to enhance the formation of ordered supramolecular assemblies.     

A gene (plsD) from Clostridium butyricum that functionally substitutes for the sn-glycerol-3-phosphate acyltransferase gene (plsB) of Escherichia coli.

The sn-glycerol-3-phosphate acyltransferase (plsB) of Escherichia coli is a key regulatory enzyme that catalyzes the first committed step in phospholipid biosynthesis. We report the initial characterization of a novel gene (termed plsD) from Clostridium butyricum, cloned based on its ability to complement the sn-glycerol-3-phosphate auxotrophic phenotype of a plsB mutant strain of E. coli. Unlike the 83-kDa PlsB acyltransferase from E. coli, the predicted plsD open reading frame encoded a protein of 26.5 kDa. Two regions of strong homology to other lipid acyltransferases, including PlsB and PlsC analogs from mammals, plants, yeast, and bacteria, were identified. PlsD was most closely related to the 1-acyl-sn-glycerol-3-phosphate acyltransferase (plsC) gene family but did not complement the growth of plsC(Ts) mutants. An in vivo metabolic labeling experiment using a plsB plsX plsC(Ts) strain of E. coli confirmed that the plsD expression restored the ability of the cells to synthesize 1-acyl-glycerol-3-phosphate. However, glycerol-3-phosphate acyltransferase activity was not detected in vitro in assays using either acyl-acyl carrier protein or acyl coenzyme A as the substrate.     

Glycerol 3-phosphate analogues as metabolic inhibitors in Escherichia coli, 3-hydroxy-4-oxobutyl-1-phosphonate, a drug that interferes with normal phosphoglyceride metabolism.

3-Hydroxy-4-oxobutyl-1-phosphonate, the phoshonic acid analogue of glyceraldehyde 3-phosphate, enters Escherichia coli via the glycerol 3-phosphate transport system. There is no differential effect upon the accumulation of deoxyribonucleic acid, ribonucleic acid, or phosphoglycerides, although the accumulation of proteins was less effected. Examination of the phospholipids revealed that phosphatidylglycerol accumulation was most severely inhibited and cardiolipin accumulation was least affected. Concentrations of glyceraldehyde 3-phosphate and its phosphonic acid analogue that markedly inhibit macromolecular and phosphoglyceride biosynthesis have no effect upon the intracellular nucleoside triphosphate pool size. The phosphonate is a competitive inhibitor of sn-glycerol 3-phosphate in reactions catalyzed by acyl coenzyme A:sn-glycerol-3-phosphate acyltransferase and CDP-diacylglycerol:sn-glycerol-3-phosphate phosphatidyltransferase. A Km mutant for the former enzyme was susceptible to the phosphansferase activity. Studies with mutant strains ruled out the aerobic glycerol-3-phosphate dehydrogenase, glycerol-3-phosphate synthase, and fructose-1,6-biphosphate aldolase as the primary sites of action.     

sn-Glycerol-3-phosphate transport in Salmonella typhimurium

Salmonella typhimurium contains a transport system for sn-glycerol-3-phosphate that is inducible by growth on glycerol and sn-glycerol-3-phosphate. In fully induced cells, the system exhibited an apparent Km of 50 microM and a Vmax of 2.2 nmol/min . 10(8) cells. The corresponding system in Escherichia coli exhibits, under comparable conditions, a Km of 14 microM and a Vmax of 2.2 nmol/min . 10(8) cells. Transport-defective mutants were isolated by selecting for resistance against the antibiotic fosfomycin. They mapped in glpT at 47 min in the S. typhimurium linkage map, 37% cotransducible with gyrA. In addition to the glpT-dependent system, S. typhimurium LT2 contains, like E. coli, a second, ugp-dependent transport system for sn-glycerol-3-phosphate that was derepressed by phosphate starvation. A S. typhimurium DNA bank containing EcoRI restriction fragments in phage lambda gt7 was used to clone the glpT gene in E. coli. Lysogens that were fully active in the transport of sn-glycerol-3-phosphate with a Km of 33 microM and a Vmax of 2.0 nmol/min . 10(8) cells were isolated in a delta glpT mutant of E. coli. The EcoRI fragment harboring glpT was 3.5 kilobases long and carried only part of glpQ, a gene distal to glpT but on the same operon. The fragment was subcloned in multicopy plasmid pACYC184. Strains carrying this hybrid plasmid produced large amounts of cytoplasmic membrane protein with an apparent molecular weight of 33,000, which was identified as the sn-glycerol-3-phosphate permease. Its properties were similar to the corresponding E. coli permease. The presence of the multicopy glpT hybrid plasmid had a strong influence on the synthesis or assembly of other cell envelope proteins of E. coli. For instance, the periplasmic ribose-binding protein was nearly absent. On the other hand, the quantity of an unidentified E. coli outer membrane protein usually present only in small amounts increased.     

Emergence of Pseudomonas aeruginosa strains producing high levels of persister cells in patients with cystic fibrosis

The majority of cystic fibrosis (CF) patients succumb to a chronic infection of the airway with Pseudomonas aeruginosa. Paradoxically, pathogenic strains are often susceptible to antibiotics, but the infection cannot be eradicated with antimicrobial therapy. We find that in a majority of patients with airway infections, late isolates of P. aeruginosa produce increased levels of drug-tolerant persister cells. The genomes of a clonal pair of early/late isolates from a single patient have been previously sequenced, and the late isolate (obtained at age 96 months) showed a 100-fold increase in persister levels. The 96-month isolate carries a large number of mutations, including a mutation in mutS that confers a hypermutator phenotype. There is also a mutation in the mexZ repressor controlling the expression of the MexXY-OprM multidrug pump, which results in a moderate increase in the ofloxacin, carbenicillin, and tobramycin MICs. Knocking out the mexXY locus restored the resistance to that of the parent strain but did not affect the high levels of persisters formed by the 96-month isolate. This suggests that the late isolate is a high-persister (hip) mutant. Increased persister formation was observed in exponential phase, stationary phase, and biofilm populations of the 96-month isolate. Analysis of late isolates from 14 additional patients indicated that 10 of them are hip mutants. Most of these hip mutants did not have higher drug resistance. Increased persister formation appears to be their sole mechanism for surviving chemotherapy. Taken together, these findings suggest a link between persisters and recalcitrance of CF infection and identify an overlooked culprit-high-persister mutants producing elevated levels of drug-tolerant cells. Persisters may play a similarly critical role in the recalcitrance of other chronic infections.     

Periplasmic protein related to the sn-glycerol-3-phosphate transport system of Escherichia coli.

Two-dimensional gel electrophoresis of shock fluids of Escherichia coli K-12 revealed the presence of a periplasmic protein related to sn-glycerol-3-phosphate transport (GLPT) that is under the regulation of glpR, the regulatory gene of the glp regulon. Mutants selected for their resistance to phosphonomycin and found to be defective in sn-glycerol-3-phosphate transport either did not produce GLPT or produced it in reduced amounts. Other mutations exhibited no apparent effect of GLPT. Transductions of glpT+ nalA phage P1 into these mutants and selection for growth on sn-glycerol-3-phosphate revealed a 50% cotransduction frequency to nalA. Reversion of mutants taht did not produce GLPT to growth on sn-glycerol-3-phosphate resulted in strains that produce GLPT. This suggests a close relationship of GLPT to the glpT gene and to sn-glycerol-3-phosphate transport. Attempts to demonstrate binding activity of GLPT in crude shock fluid towards sn-glycerol-3-phosphate have failed so far. However, all shock fluids, independent of their GLPT content, exhibited an enzymatic activity that hydrolyzes under the conditions of the binding assay, 30 to 60% of the sn-glycerol-3-phosphate to glycerol and inorganic orthophosphate.     

sn-Glycerol-3-phosphate auxotrophy of plsB strains of Escherichia coli: evidence that a second mutation, plsX, is required.

sn-Glycerol-3-phosphate auxotrophs defective in phospholipid synthesis contain a Km-defective sn-glycerol-3-phosphate acyltransferase. Detailed genetic analysis revealed that two mutations were required for the auxotrophic phenotype. One mutation, in the previously described plsB locus (sn-glycerol-3-phosphate acyltransferase structural gene), mapped near min 92 on the Escherichia coli linkage map. Isolation of Tn10 insertions cotransducible with the auxotrophy in phage P1 crosses revealed that a second mutation was required with plsB26 to confer the sn-glycerol-3-phosphate auxotrophic phenotype. This second locus, plsX, mapped between pyrC and purB near min 24 on the E. coli linkage map. Tn10 insertions near plsX allowed detailed mapping of the genetic loci in this region. A clockwise gene order putA pyrC flbA flaL flaT plsX fabD ptsG thiK purB was inferred from results of two- and three-factor crosses. Strains harboring the four possible configurations of the mutant and wild-type plsB and plsX loci were constructed. Isogenic plsB+ plsX+, plsB+ plsX50, and plsB26 plsX+ strains grew equally well on glucose minimal medium without sn-glycerol-3-phosphate. In addition, plsX or plsX+ had no apparent effect on sn-glycerol-3-phosphate acyltransferase activity measured in membrane preparations. The molecular basis for the plsX requirement for conferral of sn-glycerol-3-phosphate auxotrophy in these strains remains to be established.     

Structural analysis of adult and larval isozymes of sn-glycerol-3-phosphate dehydrogenase of Drosophila melanogaster

Compositional analysis of the soluble tryptic peptides representing about 70% of the 293 residues of sn-glycerol-3-phosphate dehydrogenase in Drosophila melanogaster reveals a single peptide difference between the sn-glycerol-3-phosphate dehydrogenase adult (GPDHF-1) and larval (GPDHF-3) isozymes. This peptide was shown to be the carboxyl terminus by sequence determination and by carboxypeptidase A digestion of the native protein. For GPDHF-1, the sequence of the COOH-terminal tryptic peptide is Asn-His-Pro-Glu-His-Met-Gln-Asn-Leu-COOH, while that of GPDHF-3 is Asn-His-Pro-Glu-His-Met-COOH.