Pathways leading to and from serine during growth of Pseudomonas AM1 in C1 compounds or succinate

1. The following enzymes of the phosphorylated pathway of serine biosynthesis have been found in methanol- and succinate-grown Pseudomonas AM1: phosphoglycerate dehydrogenase, phosphoserine-alpha-oxoglutarate aminotransferase and phosphoserine phosphohydrolase. Their specific activities were similar in the organism grown on either substrate. 2. A procedure for preparation of auxotrophic mutants of Pseudomonas AM1 is described involving N-methyl-N'-nitro-N-nitrosoguanidine as mutagen and a penicillin enrichment step. 3. A mutant, M-15A, has been isolated that is unable to grow on methanol and that lacks phenazine methosulphate-linked methanol dehydrogenase. The mutant is able to grow on methylamine, showing that the amine is not oxidized by way of methanol. 4. Loss of methanol dehydrogenase activity in mutant M-15A led to loss of phenazine methosulphate-linked formaldehyde dehydrogenase activity showing that the same enzyme is probably responsible for both activities. 5. A mutant, 20B-L, has been isolated that cannot grow on any C(1) compound tested but can grow on succinate. 6. Mutant 20B-L lacks hydroxypyruvate reductase, and revertants that regained the ability to grow on methanol, methylamine and formate contained hydroxypyruvate reductase activity at specific activities similar to that of the wild-type organism. This shows that hydroxypyruvate reductase is necessary for growth on methanol, methylamine and formate but not for growth on succinate. 7. The results suggest that during growth of Pseudomonas AM1 on C(1) compounds, serine is converted into 3-phosphoglycerate by a non-phosphorylated pathway, whereas during growth on succinate, phosphoglycerate is converted into serine by a phosphorylated pathway.     

Acetyl-CoA production and utilization during growth of the facultative methylotroph Pseudomonas AM1 on ethanol, malonate and 3-hydroxybutyrate.

In Pseudomonas AM1, conversion of 3-hydroxybutyrate to acetyl-CoA is mediated by an inducible 3-hydroxybutyrate dehydrogenase, an acetoacetate: succinate coenzyme A transferase (specific for succinyl-CoA) and an inducible beta-ketothiolase. Ethanol is oxidized to acetate by the same enzymes as are involved in methanol oxidation to formate. An inducible acetyl-CoA synthetase has been partially purified and characterized; it is essential for growth only on ethanol, malonate and acetate plus glyoxylate, as shown by the growth characteristics of a mutant (ICT54) lacking this enzyme. Free acetate is not involved in the assimilation of acetyl-CoA, and hydroxypyruvate reductase is not involved in the oxidation of acetyl-CoA to glyoxylate during growth on 3-hydroxybutyrate. A mutant (ICT51), lacking 'malate synthase' activity has been isolated and its characteristics indicate that this activity is normally essential for growth, of Pseudomonas AM1 on ethanol, malonate and 3-hydroxybutyrate, but not for growth on other substrates such as pyruvate, succinate and C1 compounds. The growth properties of a revertant (ICT51R) and of a mutant lacking malyl-CoA lyase (PCT57) indicate that an alternative route must exist for assimilation of compounds metabolized exclusively by way of acetyl-CoA.     

Cloning, mutagenesis, and physiological effect of a hydroxypyruvate reductase gene from Methylobacterium extorquens AM1.

The gene encoding the serine cycle hydroxypyruvate reductase of Methylobacterium extorquens AM1 was isolated by using a synthetic oligonucleotide with a sequence based on a known N-terminal amino acid sequence. The cloned gene was inactivated by insertion of a kanamycin resistance gene, and recombination of this insertion derivative with the wild-type gene produced a serine cycle hydroxypyruvate reductase null mutant. This mutant had lost its ability to grow on C-1 compounds but retained the ability to grow on C-2 compounds, showing that the hydroxypyruvate reductase operating in the serine cycle is not involved in the conversion of acetyl coenzyme A to glycine as previously proposed. A second hydroxypyruvate-reducing enzyme with a low level of activity was found in M. extorquens AM1; this enzyme was able to interconvert glyoxylate and glycollate. The gene encoding hydroxypyruvate reductase was shown to be located about 3 kb upstream of two other serine cycles genes encoding phosphoenolpyruvate carboxylase and malyl coenzyme A lyase.     

Genetics of the serine cycle in Methylobacterium extorquens AM1: identification, sequence, and mutation of three new genes involved in C1 assimilation, orf4, mtkA, and mtkB.

In a recent paper we reported the sequence of the beginning of a serine cycle gene cluster on the Methylobacterium extorquens AM1 chromosome, containing the genes encoding serine glyoxylate aminotransferase (sgaA), hydroxypyruvate reductase (hprA), and 5,10-methylenetetrahydrofolate dehydrogenase (mtdA) (L. V. Chistoserdova and M. E. Lidstrom J. Bacteriol. 176:1957-1968, 1994). Here we present the sequence of the adjacent downstream region containing three full and one partial open reading frames. The first of the full open reading frames (orf4) remains unidentified, while the other two (mtkA and mtkB) code for the two subunits of malate thiokinase, and the fourth, a partial open reading frame (ppcA), apparently encodes phosphoenolpyruvate carboxylase. Mutants containing insertion mutations in orf4, mtdA, and mtdB all were unable to grow on C1 compounds, showing that these three newly identified genes are indispensable for the operation of the serine cycle. Mutants in orf4 were also unable to grow on C2 compounds, but growth was restored by glyoxylate, suggesting that orf4 might be required for the conversion of acetyl coenzyme A to glyoxylate.     

The biosynthesis of serine and glycine in Pseudomonas AM1 with special reference to growth on carbon sources other than C1 compounds

1. A mutant, 20S, of Pseudomonas AM1 was obtained that requires a supplement of serine to grow on succinate, lactate or ethanol. This mutant lacks phosphoserine phosphatase and revertants to wild-type phenotype regained this enzymic activity showing that the phosphorylated pathway of serine biosynthesis is necessary for growth on these three substrates. 2. The requirement for supplemental serine by mutant 20S could be met by glycine, suggesting that Pseudomonas AM1 can obtain C(1) units from glycine. 3. Mutant 20S grows on C(1) compounds at a lower rate compared with the wild type. Supplementation with serine stimulated the growth rate of the mutant suggesting that the phosphorylated pathway of serine biosynthesis plays some role, but not an essential role, during growth on C(1) compounds. 4. A mutant, 82G, was obtained that requires a supplement of glycine to grow on succinate, lactate or ethanol. When grown in such supplemented media, the mutant lacks serine hydroxymethyltransferase and revertants to wild-type phenotype regained enzymic activity showing that during growth on succinate, lactate or ethanol, glycine is made from serine via serine hydroxymethyltransferase, and that the organism can obtain C(1) units from glycine. 5. Mutant 82G grew on methanol and then contained serine hydroxymethyltransferase suggesting that this enzyme is necessary for growth on C(1) compounds and that Pseudomonas AM1 may synthesize two such enzymes, one used in growth on C(1) compounds, the other used in growth on other substrates. Mutant 82G might lack the latter enzyme. 6. Phosphoglycerate dehydrogenase is specifically inhibited by l-serine and the regulatory implications of this are discussed.     

Metabolism of C2 compounds inAcetobacter aceti

The presence of isocitrate lyase and malate synthase was detected in cell-free extracts ofAcetobacter aceti, grown in a mineral medium with acetate as sole carbon source. The presence of these enzymes explains the ability of this strain to grow with ethanol or acetate as sole carbon source, which is an important characteristic in Frateur's classification system forAcetobacter. In addition to isocitrate lyase and malate synthase, these cell-free extracts were found to contain glyoxylate carboligase, tartronicsemialdehyde reductase and glycerate kinase. The induction of these enzymes during growth on acetate is thought to be caused by the very high activity of isocitrate lyase, which may lead to an accumulation of glyoxylate. The importance of this pathway in cells growing with acetate as sole carbon source for the synthesis of their carbohydrate components is discussed. The presence of the enzymes from the pathway from glyoxylate to 3-phosphoglycerate explains the ability of this strain to grow with ethyleneglycol and glycollate as sole carbon source.     

Aspects of glycine and serine biosynthesis during growth of Pseudomonas AM1 on C1 compounds

1. Methanol or formate can replace serine or glycine as supplements for growth on succinate of the auxotrophic mutants 20S and 82G of Pseudomonas AM1, showing that the organism can synthesize glycine and serine in net fashion from C(1) units. 2. Double mutants of Pseudomonas 20S and 82G have been prepared (20ST-1 and 82GT-1) that are unable to grow on succinate+1mm-glyoxylate, succinate+2mm-methanol or methanol alone. 3. Mutants 20ST-1 and 82GT-1 lacked serine-glyoxylate aminotransferase activity, and revertants to the phenotype of 20S and 82G regained serine-glyoxylate aminotransferase activity. A total revertant of 82GT-1 to wild-type phenotype regained activities of serine hydroxymethyltransferase and serine-glyoxylate aminotransferase. 4. The activity of serine-glyoxylate aminotransferase in methanol-grown Pseudomonas AM1 is eightfold higher than in the succinate-grown organism. 5. The combined results show that in Pseudomonas AM1 serine-glyoxylate aminotransferase is necessary for growth on C(1) compounds and is involved in the conversion of methanol into glycine via glyoxylate. 6. It is suggested that the phosphorylated pathway of serine biosynthesis from phosphoglycerate replenishes the supply of alpha-amino groups necessary for the flow of glyoxylate through the main assimilatory pathway during growth on C(1) compounds.     

Conversion of serine to glycerate in intact spinach leaf peroxisomes: role of malate dehydrogenase

In photorespiration, leaf peroxisomes convert serine to glycerate via serine-glyoxylate aminotransferase and NADH-hydroxypyruvate reductase. We isolated intact spinach leaf peroxisomes in 0.25 M sucrose, and characterized their enzymatic conversion of serine to glycerate using physiological concentrations of substrates and coenzymes. In the presence of glycolate (glyoxylate), and NADH and NAD alone or together in physiological proportions, the rate of serine-to-glycerate conversion was enhanced and sustained by the addition of malate. The rate was similar at 1 and 5 mM serine, but was two to three times higher in 50 mM than 5 mM malate. In the presence of NAD and malate, there was 1:1 stoichiometric formation of glycerate and oxaloacetate. Addition of 1 or 5 mM glutamate resulted in a negligible enhancement of the conversion of hydroxypyruvate to glycerate. Intact peroxisomes produced glycerate from either serine or hydroxypyruvate at a rate two times higher than osmotically lysed peroxisomes. These results suggest that under physiological conditions, the peroxisomal malate dehydrogenase operates independent of aspartate-alpha-ketoglutarate aminotransferase in supplying NADH for hydroxypyruvate reduction. This supply of NADH is the rate-limiting step in the conversion of serine to glycerate. The compartmentation of hydroxypyruvate reductase and malate dehydrogenase in the peroxisomes confers a higher efficiency in the supply of NADH for hydroxypyruvate reduction under a normal, high NAD/NADH ratio in the cytosol.     

Purification and characterization of hydroxypyruvate reductase from the facultative methylotroph Methylobacterium extorquens AM1.

Hydroxypyruvate reductase was purified to homogeneity from the facultative methylotroph Methylobacterium extorquens AM1. It has a molecular mass of about 71 kDa, and it consists of two identical subunits with a molecular mass of about 37 kDa. This enzyme uses both NADH (Km = 0.04 mM) and NADPH (Km = 0.06 mM) as cofactors, uses hydroxypyruvate (Km = 0.1 mM) and glyoxylate (Km = 1.5 mM) as the only substrates for the forward reaction, and carries out the reverse reaction with glycerate (Km = 2.6 mM) only. It was not possible to detect the conversion of glycolate to glyoxylate, a proposed role for this enzyme. Kinetics and inhibitory studies of the enzyme from M. extorquens AM1 suggest that hydroxypyruvate reductase is not a site for regulation of the serine cycle at the level of enzyme activity.     

Enzymes involved in the assimilation of one-carbon units by Pseudomonas MS.

Pseudomonas MS can grow on methylamine and a number of other compounds containing C1 units as a sole source of carbon and energy. Assimilation of carbon into cell material occurs via the "serine pathway" since enzymes of this pathway are induced after growth on methylamine, but not malate or acetate. A mutant has been isolated which is unable to grow on methylamine or any other related substrate providing C1 units. This mutant is also unable to grow on acetate. Measurment of enzyme activities in cell-free extracts of wild-type cells showed that growth on methylamine caused induction of isocitrate lyase, a key enzyme in the glyoxylate cycle. The mutant organism lacks malate lyase, a key enzyme of the serine pathway, and isocitrate lyase as well. These results suggest that utilization of C1 units by Pseudomonas MS results in the net accumulation of acetate which is then assimilated into cell material via the glyoxylate cycle.     

Metabolism and decarboxylation of glycollate and serine in leaf peroxisomes

The linked utilization of glycollate and L-serine has been studied in peroxisomal preparations from leaves of spinach beet (Beta vulgaris L.). The generation of glycine from glycollate was found to be balanced by the production of hydroxypyruvate from serine and similarly by 2-oxoglutarate when L-glutamate was substituted for L-serine. In the presence of L-malate and catalytic quantities of NAD⁺, about 40% of the hydroxypyruvate was converted further to glycerate, whereas with substrate quantities of NADH, this conversion was almost quantitative. CO₂ was released from the carboxyl groups of both glycollate and serine. Since the decarboxylation of both substrates was greatly in creased by the catalase inhibitor, 3-amino-1,2,4-triazole, and abolished by bovine liver catalase, it was attributed to the nonenzymic attack of H₂O₂, generated in glycollate oxidation, upon glyoxylate and hydroxypyruvate respectively. At 25–30° C, about 10% of the glyoxylate and hydroxypyruvate accumulated was decarboxylated, and the release of CO₂ from each keto-acid was related to the amounts present. It is suggested that hydroxypyruvate decarboxylation might contribute significantly to photorespiration and provide a metabolic route for the complete oxidation of glycollate, the magnitude of this contribution depending upon the concentrations of glyoxylate and hydroxypyruvate in the peroxisomes.     

Microbial growth on oxalate by a route not involving glyoxylate carboligase

1. The metabolism of oxalate by the pink-pigmented organisms, Pseudomonas AM1, Pseudomonas AM2, Protaminobacter ruber and Pseudomonas extorquens has been compared with that of the non-pigmented Pseudomonas oxalaticus. 2. During growth on oxalate, all the organisms contain oxalyl-CoA decarboxylase, formate dehydrogenase and oxalyl-CoA reductase. This is consistent with oxidation of oxalate to carbon dioxide taking place via oxalyl-CoA, formyl-CoA and formate as intermediates, and also reduction of oxalate to glyoxylate taking place via oxalyl-CoA. 3. The pink-pigmented organisms, when grown on oxalate, contain l-serine–glyoxylate aminotransferase and hydroxypyruvate reductase but do not contain glyoxylate carboligase. The converse of this obtains in oxalate-grown Ps. oxalaticus. This indicates that, in contrast with Ps. oxalaticus, synthesis of C₃ compounds from oxalate by the pink-pigmented organisms occurs by a variant of the `serine pathway' used by Pseudomonas AM1 during growth on C₁ compounds. 4. Evidence in favour of this scheme is provided by the finding that a mutant of Pseudomonas AM1 that lacks hydroxypyruvate reductase is not able to grow on oxalate.     

Purification and properties of two glyoxylate reductases from a species of Pseudomonas

1. Two enzymes that catalyse the reduction of glyoxylate to glycollate have been separated and purified from a species of Pseudomonas. Their molecular weights were estimated as 180000. 2. Reduced nicotinamide nucleotides act as the hydrogen donators for the enzymes. The NADH-linked enzyme is entirely specific for its coenzyme but the NADPH-linked reductase shows some affinity towards NADH. 3. Both enzymes convert hydroxypyruvate into glycerate. 4. The glyoxylate reductases show maximal activity at pH6·0–6·8, are inhibited by keto acids and are strongly dependent on free thiol groups for activity. 5. The Michaelis constants for glyoxylate and hydroxypyruvate were found to be of a high order. 6. The reversibility of the reaction has been demonstrated for both glyoxylate reductases and the equilibrium constants were determined. 7. The reduction of glyoxylate and hydroxypyruvate is not stimulated by anions.     

Genetics of the serine cycle in Methylobacterium extorquens AM1: identification of sgaA and mtdA and sequences of sgaA, hprA, and mtdA.

In a previous paper, we reported identification of the 5' part of hprA of Methylobacterium extorquens AM1, which encodes the serine cycle enzyme hydroxypyruvate reductase (L. V. Chistoserdova and M. E. Lidstrom, J. Bacteriol. 174:71-77, 1992). Here we present the complete sequence of hprA and partial sequence of genes adjacent to hprA. Upstream of hprA, the 3' part of an open reading frame was discovered, separated from hprA by 263 bp. This open reading frame was identified as the gene encoding another serine cycle enzyme, serine glyoxylate aminotransferase (sgaA). Cells containing an insertion mutation into sgaA were unable to grow on C1 compounds, demonstrating that the gene is required for C1 metabolism. Sequencing downstream of hprA has revealed the presence of another open reading frame (mtdA), which is probably cotranscribed with hprA. This open reading frame was identified as the gene required for the synthesis of 5,10-methylenetetrahydrofolate dehydrogenase. Our data suggest that this enzyme plays an integral role in methylotrophic metabolism in M. extorquens AM1, either in formaldehyde oxidation or as part of the serine cycle.     

The pathway of glycollate utilization in Chlorella pyrenoidosa

1. Exogenous glycollate was rapidly metabolized in both the light and the dark by photoautotrophically grown Chlorella pyrenoidosa. 2. The incorporation of (14)C from [1-(14)C]glycollate by these cells was inhibited by the tricarboxylic acid-cycle inhibitors monofluoroacetate, diethylmalonate and arsenite, and also by alpha-hydroxypyrid-2-ylmethanesulphonate and isonicotinylhydrazine. 3. Short-term kinetic experiments showed over 80% of the total (14)C present in the soluble fraction from the cells to be in glycine and serine after 10s. This percentage decreased with time whereas the percentage radioactivity in glycerate increased for up to 30s then remained steady. The percentage of the total radioactivity present in citrate increased over the experimental period. Malate was the only other tricarboxylic acid-cycle intermediate to become labelled. 4. The kinetic and inhibitor experiments supported the following pathway of glycollate incorporation: glycollate --> glyoxylate --> glycine --> serine --> hydroxypyruvate --> glycerate --> 3-phosphoglycerate --> 2-phosphoglycerate --> phosphoenolpyruvate --> pyruvate --> acetyl-CoA. 5. The specific activities of the enzymes catalysing this metabolic sequence in cell-free extracts were great enough to account for the observed rate of glycollate metabolism of 0.25mumol/h per mg dry wt. of cells in the light.     

Glutamate and serine as competing donors for amination of glyoxylate in leaf peroxisomes

When provided with glycollate, peroxisomal extracts of leaves of spinach beet (Beta vulgaris L. cv.) converted L-serine and L-glutamate to hydroxypyruvate and 2-oxoglutarate respectively. When approximately saturating concentrations of each of these amino acids were incubated separately with glycollate, the utilization of serine was greater than that of glutamate. The utilization of glutamate was substantially reduced by the presence of relatively low concentrations of serine in the reaction mixture, whereas even high concentrations of glutamate caused only small reductions in serine utilization. Over the entire range of concentrations of amino acids examined, serine was invariably the preferred amino-group donor, but this preference was abolished at higher concentrations of glyoxylate. Serine not only competed favourably for glyoxylate but also inhibited L-glutamate: glyoxylate aminotransferase (GGAT), the degree of inhibition depending upon the glyoxylate concentration. Studies of L-serine: glyoxylate aminotransferase (SGAT) and GGAT in partially purified extracts from spinach-beet leaves confirmed that serine competitively inhibited GGAT but glutamate did not affect SGAT. Both enzymes were inhibited by high glyoxylate concentrations, the inhibition being relieved by suitably high concentrations of the appropriate amino acid. It is concluded that at the low glyoxylate concentrations likely to occur in vivo, the preferential utilization of serine would ensure flux through the glycollate pathway to glycerate, but at higher concentrations of glyoxylate, both enzymes could be fully active in glyoxylate amination.Abbreviations SGAT L-serine: glyoxylate aminotransferase - GGAT L-glutamate: glyoxylate aminotransferase     

Growth of Pseudomonas C on C1 compounds: enzyme activites in extracts of Pseudomonas C cells grown on methanol, formaldehyde, and formate as sole carbon sources.

Pseudomonas C can grow on methanol, formaldehyde, or formate as sole carbon source. It is proposed that the assimilation of carbon by Pseudomonas C grown on different C1 growth substrates proceeds via one of two metabolic pathways, the serine pathway or the allulose pathway (the ribose phosphate cycle of formaldehyde fixation). This contention is based on the distribution of two key enzymes, each of which appears to be specifically involved in one of the assimilation pathways, glycerate dehydrogenase (serine pathway) and hexose phosphate synthetase (allulose pathway). The assimilation of methanol in Pseudomonas C cells appears to occur via the allulose pathway, whereas the utilization of formaldehyde or formate in cells grown on formaldehyde or formate as sole carbon sources appears by the serine pathway. When methanol is present together with formaldehyde or formate in the growth medium, the formaldehyde or formate is utilized by the allulose pathway.     

Regulation of Enzymes Associated with C-1 Metabolism in Three Facultative Methylotrophs

The levels of the oxidation enzyme methanol dehydrogenase and the serine pathway enzymes, hydroxypyruvate reductase, glycerate kinase, serine transhydroxymethylase, serine-glyoxylate aminotransferase, phosphoenolpyruvate carboxylase, and malyl-coenzyme A lyase, were studied in cells of the facultative methylotrophs Pseudomonas AM1, Pseudomonas 3A2 and Hyphomicrobium X grown on different substrates. Induction and dilution curves for these enzymes suggest they may be regulated coordinately in Hyphomicrobium X, but not in Pseudomonas AM1 or 3A2. Glyoxylate stimulated the serine transhydroxymethylase activity in methanol-grown cells of all three organisms. A secondary alcohol dehydrogenase activity was detected at low levels in Pseudomonas AM1 and Hyphomicrobium X, but not in Pseudomonas 3A2.     

The enzymic reduction of glyoxylate and hydroxypyruvate in leaves of higher plants

Glyoxylate and hydroxypyruvate are metabolites involved in the pathway of carbon in photorespiration. The chief glyoxylate-reducing enzyme in leaves is now known to be a cytosolic glyoxylate reductase that uses NADPH as the preferred cofactor but can also use NADH. Glyoxylate reductase has been isolated from spinach leaves, purified to homogeneity, and characterized kinetically and structurally. Chloroplasts contain lower levels of glyoxylate reductase activity supported by both NADPH and NADH, but it is not yet known whether a single chloroplastic enzyme catalyzes glyoxylate reduction with both cofactors. The major hydroxypyruvate reductase activity of leaves has long been known to be a highly active enzyme located in peroxisomes; it uses NADH as the preferred cofactor. To a lesser extent, NADPH can also be used by the peroxisomal enzyme. A second hydroxypyruvate reductase enzyme is located in the cytosol; it preferentially uses NADPH but can also use NADH as cofactor. In a barley mutant deficient in peroxisomal hydroxypyruvate reductase, the NADPH-preferring cytosolic form of the enzyme permits sufficient rates of hydroxypyruvate reduction to support continued substrate flow through the terminal stages of the photosynthetic carbon oxidation (glycolate/glycerate) pathway. The properties and metabolic significance of the cytosolic and organelle-localized glyoxylate and hydroxypyruvate reductase enzymes are discussed.     

Compartmentation studies on spinach leaf peroxisomes : evidence for channeling of photorespiratory metabolites in peroxisomes devoid of intact boundary membrane

In concurrence with earlier results, the following enzymes showed latency in intact spinach (Spinacia oleracea L.) leaf peroxisomes: malate dehydrogenase (89%), hydroxypyruvate reductase (85%), serine glyoxylate aminotransferase (75%), glutamate glyoxylate aminotransferase (41%), and catalase (70%). In contrast, glycolate oxidase was not latent. Aging of peroxisomes for several hours resulted in a reduction in latency accompanied by a partial solubilization of the above mentioned enzymes. The extent of enzyme solubilization was different, being highest with glutamate glyoxylate aminotransferase and lowest with malate dehydrogenase. Osmotic shock resulted in only a partial reduction of enzyme latency. Electron microscopy revealed that the osmotically shocked peroxisomes remained compact, with smaller particle size and pleomorphic morphology but without a continuous boundary membrane. Neither in intact nor in osmotically shocked peroxisomes was a lag phase observed in the formation of glycerate upon the addition of glycolate, serine, malate, and NAD. Apparently, the intermediates, glyoxylate, hydroxypyruvate, and NADH, were confined within the peroxisomal matrix in such a way that they did not readily leak out into the surrounding medium. We conclude that the observed compartmentation of peroxisomal metabolism is not due to the peroxisomal boundary membrane as a permeability barrier, but is a function of the structural arrangement of enzymes in the peroxisomal matrix allowing metabolite channeling.