Lysine metabolism in the human and the monkey: Demonstration of pipecolic acid formation in the brain and other organs

Metabolism ofl-[U-¹⁴C]lysine was studied in the human autopsy tissues and the intact monkeys through intracerebroventricular and intravenous injections. The human tissues were more active in the metabolism ofl-[¹⁴C]lysine to [¹⁴C]pipecolate than the rat tissues previously reported. This metabolism was equally active in the phosphate (pH 7) and the glycyl-glycine (pH 8.6) buffers with the brain and the kidney having higher activity than the liver. Besides [¹⁴C]pipecolate, traces of [¹⁴C]saccharopine and -[¹⁴C]aminoadipate were also detected in the liver incubation. Twenty-four hr after intraventricular injection ofl-[¹⁴C]lysine to the monkey, substantial labeling of pipecolate and -aminoadipate was observed in the brain and spinal cord, with the kidney, liver and the plasma having much reduced levels. Radioactivity levels of these two compounds were found low in the organs and plasma of the intravenously injected monkey. The urine of both monkeys contained only traces of [¹⁴C]pipecolate, even though it contained high levels ofl-[¹⁴C]lysine and -[¹⁴C]aminoadipate. It was concluded thatl-lysine is actively metabolized to pipecolate and -aminoadipate in the human and the monkey, that this reaction is most active in the brain whenl-lysine is intraventricularly administered, and that in contrast to the rat, the monkey may have an effective renal reabsorption for pipecolate which is similar to the human.     

Conversion of Pipecolic Acid into Lysine in Penicillium chrysogenum Requires Pipecolate Oxidase and Saccharopine Reductase: Characterization of the lys7 Gene Encoding Saccharopine Reductase

Pipecolic acid is a component of several secondary metabolites in plants and fungi. This compound is useful as a precursor of nonribosomal peptides with novel pharmacological activities. In Penicillium chrysogenum pipecolic acid is converted into lysine and complements the lysine requirement of three different lysine auxotrophs with mutations in the lys1, lys2, or lys3 genes allowing a slow growth of these auxotrophs. We have isolated two P. chrysogenum mutants, named 7.2 and 10.25, that are unable to convert pipecolic acid into lysine. These mutants lacked, respectively, the pipecolate oxidase that converts pipecolic acid into piperideine-6-carboxylic acid and the saccharopine reductase that catalyzes the transformation of piperideine-6-carboxylic acid into saccharopine. The 10.25 mutant was unable to grow in Czapek medium supplemented with alpha-aminoadipic acid. A DNA fragment complementing the 10.25 mutation has been cloned; sequence analysis of the cloned gene (named lys7) revealed that it encoded a protein with high similarity to the saccharopine reductase from Neurospora crassa, Magnaporthe grisea, Saccharomyces cerevisiae, and Schizosaccharomyces pombe. Complementation of the 10.25 mutant with the cloned gene restored saccharopine reductase activity, confirming that lys7 encodes a functional saccharopine reductase. Our data suggest that in P. chrysogenum the conversion of pipecolic acid into lysine proceeds through the transformation of pipecolic acid into piperideine-6-carboxylic acid, saccharopine, and lysine by the consecutive action of pipecolate oxidase, saccharopine reductase, and saccharopine dehydrogenase.     

Pipecolic acid biosynthesis in Rhizoctonia leguminicola. I. The lysine saccharopine, delta 1-piperideine-6-carboxylic acid pathway

The biosynthesis of pipecolic acid from L-lysine in the fungal parasite, Rhizoctonia leguminicola has been reinvestigated. Pipecolate is then utilized to form the toxic octahydroindolizine alkaloids, slaframine and swainsonine. Incorporation studies of L-versus D-[U-14C]lysine into R. leguminicola metabolites confirmed earlier findings that L-lysine is the predominant substrate for pipecolate formation and D-lysine for alpha-N-acetyllysine (concerned in lysine catabolism). However [alpha-15N]lysine, not [epsilon-15N]lysine as previously reported, labeled pipecolate. Such findings implied that delta 1-piperideine-6-carboxylate, not delta 1-piperideine-2-carboxylate, was formed from lysine and was the immediate precursor of pipecolate. Evidence from cell-free enzyme systems established the following biosynthetic events: L-lysine A----saccharopine B----delta 1-piperideine-6-carboxylate C----pipecolate. Products of reactions A and C were identified from biological and chemical considerations. Reaction B was carried out by a previously undescribed flavin enzyme termed saccharopine oxidase. The product of reaction B, which reacted with p-dimethylaminobenzaldehyde, was reduced with Na-CNB2H3. Its NMR spectrum was identical with that of deuteriated pipecolate prepared from authentic delta 1-piperideine-6-carboxylate, but not from authentic delta 1-piperideine-2-carboxylate. Reaction B represents a branching of primary lysine metabolism from saccharopine to a secondary pathway leading to pipecolate and to octahydroindolizine alkaloids in R. leguminicola.     

Metabolism of lysine in α-aminoadipic semialdehyde dehydrogenase-deficient fibroblasts: Evidence for an alternative pathway of pipecolic acid formation

The mammalian degradation of lysine is believed to proceed via two distinct routes, the saccharopine and the pipecolic acid routes, that ultimately converge at the level of α-aminoadipic semialdehyde (α-AASA). α-AASA dehydrogenase-deficient fibroblasts were grown in cell culture medium supplemented with either l-[α-¹⁵N]lysine or l-[ε-¹⁵N]lysine to explore the exact route of lysine degradation. l-[α-¹⁵N]lysine was catabolised into [¹⁵N]saccharopine, [¹⁵N]α-AASA, [¹⁵N]Δ¹-piperideine-6-carboxylate, and surprisingly in [¹⁵N]pipecolic acid, whereas l-[ε-¹⁵N]lysine resulted only in the formation of [¹⁵N]saccharopine. These results imply that lysine is exclusively degraded in fibroblasts via the saccharopine branch, and pipecolic acid originates from an alternative precursor. We hypothesize that pipecolic acid derives from Δ¹-piperideine-6-carboxylate by the action of Δ¹-pyrroline-5-carboxylic acid reductase, an enzyme involved in proline metabolism.     

Reciprocal Control of Thyroid Binding and the Pipecolate Pathway in the Brain

Thyroid hormones have long been known to play an essential role in brain growth and development, with cytoplasmic thyroid hormone binding proteins (THBPs) playing a critical role in thyroid hormone bioavailability. A major mammalian THBP is μ-crystallin (CRYM), which was originally characterized by its ability to strongly bind thyroid hormones in an NADPH-dependent fashion. However, in 2011 it was discovered that CRYM is also an enzyme, namely ketimine reductase (KR), which catalyzes the NAD(P)H-dependent reduction of –C=N– (imine) double bonds of a number of cyclic ketimine substrates including sulfur-containing cyclic ketimines. The enzyme activity was also shown to be potently inhibited by thyroid hormones, thus suggesting a novel reciprocal relationship between enzyme catalysis and thyroid hormone bioavailability. KR is involved in a number of amino acid metabolic pathways. However, the best documented biological function of KR is its role as a ?¹-piperideine-2-carboxylate (P2C) reductase in the pipecolate pathway of lysine metabolism. The pipecolate pathway is the main l-lysine degradation pathway in the adult brain, whereas the saccharopine pathway predominates in extracerebral tissues and in infant brain, suggesting that KR has evolved to perform specific and important roles in neural development and function. The potent regulation of KR activity by thyroid hormones adds further weight to this suggestion. KR is also involved in l-ornithine/l-glutamate/l-proline metabolism as well as sulfur-containing amino acid metabolism. This review describes the pipecolate pathway and recent discoveries related to mammalian KR function, which have important implications in normal and pathological brain functions.     

Lysine metabolism in mammalian brain: an update on the importance of recent discoveries

The lysine catabolism pathway differs in adult mammalian brain from that in extracerebral tissues. The saccharopine pathway is the predominant lysine degradative pathway in extracerebral tissues, whereas the pipecolate pathway predominates in adult brain. The two pathways converge at the level of ?¹-piperideine-6-carboxylate (P6C), which is in equilibrium with its open-chain aldehyde form, namely, α-aminoadipate δ-semialdehyde (AAS). A unique feature of the pipecolate pathway is the formation of the cyclic ketimine intermediate ?¹-piperideine-2-carboxylate (P2C) and its reduced metabolite l-pipecolate. A cerebral ketimine reductase (KR) has recently been identified that catalyzes the reduction of P2C to l-pipecolate. The discovery that this KR, which is capable of reducing not only P2C but also other cyclic imines, is identical to a previously well-described thyroid hormone-binding protein [μ-crystallin (CRYM)], may hold the key to understanding the biological relevance of the pipecolate pathway and its importance in the brain. The finding that the KR activity of CRYM is strongly inhibited by the thyroid hormone 3,5,3′-triiodothyronine (T₃) has far-reaching biomedical and clinical implications. The inter-relationship between tryptophan and lysine catabolic pathways is discussed in the context of shared degradative enzymes and also potential regulation by thyroid hormones. This review traces the discoveries of enzymes involved in lysine metabolism in mammalian brain. However, there still remain unanswered questions as regards the importance of the pipecolate pathway in normal or diseased brain, including the nature of the first step in the pathway and the relationship of the pipecolate pathway to the tryptophan degradation pathway.     

Lysine catabolism in Haemonchus contortus and Teladorsagia circumcincta

Catabolism of lysine through the pipecolate, saccharopine and cadaverine pathways has been investigated in L3 and adult Haemonchus contortus and Teladorsagia circumcincta. Both enzymes of the saccharopine pathway (lysine ketoglutarate reductase (LKR) and saccharopine dehydrogenase (SDH)) were active in L3 and adult worms of both species. All three enzymes which catabolise lysine to α-amino adipic semialdehyde via pipecolate (lysine oxidase (LO), Δ(1)-piperideine-2-carboxylate reductase (Pip2CR) and pipecolate oxidase (PipO)) were present in adult worms, whereas the pathway was incomplete in L3 of both species; Pip2CR activity was not detected in the L3 of either parasite species. In adult worms, the saccharopine pathway would probably be favoured over the pipecolate pathway as the K(m) for lysine was lower for LKR than for LO. Neither lysine dehydrogenase nor lysine decarboxylase activity was detected in the two parasite species. Enzyme activities and substrate affinities were higher for all five enzymes in adult worms than in L3. An unexpected finding was that both LKR and SDH were dual co-factor enzymes and not specific for either NAD(+) or NADP(+), as is the case in other organisms. This novel property of LKR/SDH suggests it could be a good candidate for anthelmintic targeting.     

Pipecolic acid biosynthesis in Rhizoctonia leguminicola. II. Saccharopine oxidase: a unique flavin enzyme involved in pipecolic acid biosynthesis

The fungal parasite Rhizoctonia leguminicola produces two indolizidine alkaloids, slaframine and swainsonine, of physiological interest. These alkaloids are biosynthesized from pipecolic acid which in turn is derived from L-lysine in this fungus as shown in the accompanying paper (Wickwire, B.M., Harris, C.M., Harris, T.M., and Broquist, H.P. (1989) J. Biol. Chem. 265, 14742-14747): L-lysine----saccharopine----delta 1----piperideine-6- carboxylate----pipecolate. This paper concerns the discovery, purification, and properties of a flavoenzyme, termed saccharopine oxidase, which carries out the oxidative cleavage of saccharopine as follows: Saccharopine + O2----delta 1-piperidine-6-carboxylate + glutamate + H2O2 The enzyme was purified 2,000-fold to homogeneity (polyacrylamide gel electrophoresis) in 14% yield from R. leguminicola mycelia, and had a native molecular mass of about 45,000 daltons by gel filtration (fast protein liquid chromatography Superose). Evidence for the presence of a flavin in the enzyme was drawn from these considerations: (a) the enzyme, while oxidatively cleaving saccharopine, concomitantly reduces 2,6-dichlorophenolindophenol; (b) the purified enzyme has a fluorescence spectrum typical of flavins; and (c) the enzyme requires oxygen and produces hydrogen peroxide. Good correlation was shown with purified saccharopine oxidase between disappearance of saccharopine with the concomitant appearance of delta 1-piperideine-6-carboxylate plus glutamate. The enzyme has a pH optimum about 6 and a Km for saccharopine of 0.128 mM. The enzyme apparently exists in R. leguminicola to shunt saccharopine, a major lysine metabolite, into a secondary pathway of lysine metabolism leading to pipecolate and subsequently to slaframine and swainsonine.     

Enhancement of hexobarbital-induced sleep by lysine and its metabolites

Lysine has been shown to be metabolized in the rat brain to pipecolic acid which is a precursor of piperidine. Lysine and its proposed metabolites in this pathway were studied for the first time for their effect on the sleeping time induced by hexobarbital in the rat. Only $\text{L}$-lysine and $\text{D}$-lysine were found to prolong sleeping time significantly without toxic effect. A 3-day pretreatment with $\text{L}$-lysine produced an even more profound sleep prolongation. In most cases sleep enhancement was accompanied by a significant shortening of the time of sleep onset. Quantification of brain hexobarbital levels in the control and treated rats indicates that prolongation of sleeping time was not produced by inhibition of hexobarbital metabolism. The sleep prolonging effect of lysine, therefore, may be a direct action of lysine, or the metabolite(s) derived $\text{in}$$\text{vivo}$ from lysine, on the central nervous system.     

Lysine Catabolism in Barley (Hordeum vulgare L.)

Lysine catabolism in seedlings of barley (Hordeum vulgare L. var. Emir) was studied by direct injection of the following tracers into the endosperm of the seedlings: aspartic acid-3-(14)C, 2-aminoadipic acid-1-(14)C, saccharopine-(14)C, 2,6-diaminopimelic acid-1-(7)-(14)C, and lysine-1-(14)C. Labeled saccharopine was formed only after the administration of either labeled 2,6-diaminopimelic acid or labeled lysine to the seedlings. The metabolic fate of the other tracers administered also supported a catabolic lysine pathway via saccharopine, and apparently proceeding by a reversal of some of the biosynthetic steps of the 2-aminoadipic acid pathway known from lysine biosynthesis in most fungi. Pipecolic acid seems not to be on the main pathway of l-lysine catabolism in barley seedlings.     

Potential role for saccharopine reductase in swainsonine metabolism in endophytic fungus, Undifilum oxytropis

Locoweed plants in the southwestern United States often harbour a slow-growing endophytic fungus, Undifilum oxytropis (Phylum: Ascomycota; Order: Pleosporales), which produces a toxic alkaloid, swainsonine. Consumption of U. oxytropis by grazing animals induces a neurological disorder called locoism for which the toxic alkaloid swainsonine has been reported to be the causal agent. Little is known about the biosynthetic pathway of swainsonine in endophytic fungi, but previous studies on non-endophytic ascomycetous fungi indicate that pipecolic acid and saccharopine are key intermediates. We have used degenerate primers, Rapid amplification of cDNA ends (RACE)-PCR and inverse PCR to identify the gene sequence of U. oxytropis saccharopine reductase. To investigate the role of this gene product in swainsonine metabolism, we have developed a gene deletion system for this slow-growing endophyte based on our recently established transformation protocol. A strain of U. oxytropis lacking saccharopine reductase had decreased levels of saccharopine and lysine along with increased accumulation of pipecolic acid and swainsonine. Thus, saccharopine reductase influences the accumulation of swainsonine and its precursor, pipecolic acid, in U. oxytropis.     

Catabolism of lysine in Penicillium chrysogenum leads to formation of 2-aminoadipic acid, a precursor of penicillin biosynthesis.

Penicillium chrysogenum L2, a lysine auxotroph blocked in the early steps of the lysine pathway before 2-aminoadipic acid, was able to synthesize penicillin when supplemented with lysine. The amount of penicillin produced increased as the level of lysine in the media was increased. The same results were observed in resting-cell systems. Catabolism of [U-14C]lysine by resting cells and batch cultures of P. chrysogenum L2 resulted in the formation of labeled saccharopine and 2-aminoadipic acid. Formation of [14C]saccharopine was also observed in vitro when cell extracts of P. chrysogenum L2 and Wis 54-1255 were used. Saccharopine dehydrogenase and saccharopine reductase activities were found in cell extracts of P. chrysogenum, which indicates that lysine catabolism may proceed by reversal of the two last steps of the lysine biosynthetic pathway. In addition, a high lysine:2-ketoglutarate-6-aminotransferase activity, which converts lysine into piperideine-6-carboxylic acid, was found in cell extracts of P. chrysogenum. These results suggest that lysine is catabolized to 2-aminoadipic acid in P. chrysogenum by two different pathways. The relative contribution of lysine catabolism in providing 2-aminoadipic acid for penicillin production is discussed.     

Assay of Δ1-piperideine-2-carboxylate and synthesis of l-[14C]pipecolate from dl-[14C]pipecolate

Four assay methods were tested for the measurement of Δ¹-piperideine-2-carboxylate, a proposed alicyclic ketimino acid intermediate in the pathway of lysine metabolism to l-pipecolate, and the product of d-amino acid oxidase on d-pipecolate. The method using Δ¹-piperideine-2-carboxylate reductase from Pseudomonas putida was found to be most sensitive and specific. Measurement of Δ¹-piperideine-2-carboxylate by reduction with NaBH₄ and ninhydrin assay of the resultant pipecolate, by direct acidic ninhydrin assay, and by o-aminobenz-aldehyde assay were less desirable because of lower sensitivity and specificity. Two synthetic methods for preparing l-[¹⁴C]pipecolate from the racemic dl-[¹⁴C]pipecolate were investigated. Incubation of dl-[¹⁴C]pipecolate with a combination of d-amino acid oxidase and Δ¹-piperideine-2-carboxylate reductase or d-amino acid oxidase and NaBH₄ totally inverted the d-isomer to the l-isomer, with $\text{Δ}{}^1\text{-[}{}^{14}\text{C]piperideine-2-carboxylate}$ as an intermediate in each cycle of interconversion. No purification except desalting through a Dowex 50 (H⁺) column was necessary in order to recover l-[¹⁴C]pipecolate in pure form. The yield was 95–97% compared to <50% in the conventional method.     

Preferential production of rapamycin vs prolylrapamycin by Streptomyces hygroscopicus

A trace of prolylrapamycin is often produced in rapamycin fermentations carried out by strains of Streptomyces hygroscopicus. Prolylrapamycin was produced as the major rapamycin when L-proline was added to the fermentation medium. Addition of proline plus thiazolidine-2-carboxylic acid (T2CA), a sulfur-containing proline analog, prevented rapamycin production and stimulated prolylrapamycin production, thereby resulting in an even greater selective production of prolylrapamycin. T2CA addition inhibited rapamycin production even in the presence of L-lysine which is converted into pipecolic acid intracellularly and normally stimulates rapamycin formation. Addition of the rapamycin precursor, DL-pipecolic acid, surprisingly failed to stimulate rapamycin production. However, when DL-pipecolic acid was added with L-proline, it reduced the formation of prolylrapamycin and stimulated rapamycin production; this was evident especially in the presence of T2CA. The evidence suggests that T2CA suppresses rapamycin production by inhibiting intracellular conversion of L-lysine into pipecolate. Furthermore, the data suggest that uptake of pipecolate into the cell is stimulated or induced by growth in the presence of L-proline and/or T2CA. Received 24 December 1997/ Accepted in revised form 12 May 1998     

Accumulation and metabolism of pipecolic acid in the brain and other organs of the mouse

The long-term accumulation of pipecolic acid, as well as its disappearance following exogenous administration was studied in brain and other organs of the mouse. Mice were pulse-injected intraperitoneally or intravenously with 1Ci[³H]D,l-pipecolic acid (6.9 nmol/mouse=2.9 g/kg). The total radioactivity retained in tissues was measured in brain, liver, and kidney, as well as in plasma during the period 1 min to 24 hr. TLC separation of DNP-derivatives was performed. Three features of the pattern of retention of pipecolic acid are most salient; first the rapid accumulation in brain, second the rapid secretion of this compound in the urine, and third the long-lasting steady levels of radioactivity maintained in brain.Sixty minutes after i.v. injection, the brain/plasma ratio is approximately 0.2 and approaches unity only at 5 hr. Following intraperitoneal injection the percent recovered as pipecolic acid in brain is 78% at 30 min and 71% at 120 min, suggesting a slow metabolic activity. Liver shows a different trend than brain with a slower accumulation and a faster disappearance. Kidney shows a pattern similar to plasma with a rapid secretion of radioactivity into urine which correlates well with the exponential decrease in plasma and urine. The administration of probenecid significantly increases radioactivity due to pipecolic acid in brain, liver, and urine. Formation of -aminoadipic acid, a known metabolite of pipecolic acid, can be demonstrated in kidney 30 min after intraperitoneal injection. The present data together with results obtained previously with intracarotid injections suggest that pipecolic acid is taken up in the mouse brain from the circulation. Most of the pipecolic acid taken up is rapidly removed through the circulation and secreted in the urine; however, a small part is retained and probably metabolized by brain and kidney.     

Conversion of Lysine to Pipecolic Acid by Rumen Ciliate Protozoa

Addition of lysine to the culture medium of rumen ciliates increased the amount of pipecolate in the medium to the level over the control.

l-Lysine- U-¹⁴C was partly incorporated into ciliate protein unchanged, and partly converted to radioactive pipecolate. The autoradiogram of amino acids in supernatant fluid of the medium revealed the traces of two unidentified spots other than lysine and pipecolate as a main metabolite.

Since the radioactive pipecolate was not changed further more by ciliates, it seemed to be the end product in lysine degradation.

After addition of l-lysine-α,ε-¹⁵N to the medium, ¹⁵N was detected mainly in the fractions of pipecolate and ammonia and a little in that of glutamate, alanine and aspartate of the medium.     

Analysis of swainsonine and its early metabolic precursors in cultures of Metarhizium anisopliae

The alpha-mannosidase inhibitor swainsonine is produced by the filamentous fungus Metarhizium anisopliae. The primary metabolite pathway from which it is derived is known to be that leading to lysine. In order to effect improvements in the yield of swainsonine it is of interest to study the changes in the intracellular levels of lysine and its biosynthetic intermediates, as well as swainsonine itself, which accompany changes in culture conditions or in the genetics of the microbe. Czapek-Dox defined medium has been used for these studies. A reversed-phase, high performance liquid chromatography procedure was developed for the analysis of lysine, saccharopine, alpha-aminoadipic acid and pipecolic acid in mycelial extracts. The method is based upon precolumn derivatization with 9-fluorenylmethyl chloroformate (FMOC), a reagent known to be useful for the derivatization of amino-containing compounds. Elution with an acetate buffer/acetonitrile gradient effected separation of the four metabolites which were quantified by UV absorption at concentrations from 1 to 20 μg ml-1. Swainsonine concentrations were determined using a previously described enzyme-based method, but applied now to intracellular as well as extracellular samples. Analysis of mycelial extracts from the end of swainsonine accumulation in medium supplemented with L-lysine revealed the accumulation of pipecolic acid and to a lesser extent lysine compared to control mycelium. Controlling the culture medium pH to 9.0 resulted in a drop in swainsonine yield accompanied by an increase in intracellular pipecolic acid levels. Spontaneous mutants tolerant to the presence of the toxic lysine analogue 2-aminoethylcysteine (AEC) were isolated in an attempt to generate lysine over-producers, which might be expected to produce more swainsonine. Surprisingly, four independently isolated mutants produced lower yields of swainsonine, but accumulated higher levels of saccharopine. The tolerance to AEC therefore appears to be due to a reduction in the diversion of saccharopine into swainsonine biosynthesis, allowing the biosynthesis of sufficient lysine to overcome AEC competition. This revised version was published online in November 2006 with corrections to the Cover Date.     

The conversion of L-lysine to saccharopine and alpha-aminoadipate in mouse

Saccharopine [?-N-(l-glutaryl-2)-l-lysine] has been found to occur in normal, untreated mouse liver. The pool of saccharopine as well as that of α-aminoadipate become labeled shortly after the administration of l-lysine-U-¹⁴C into intact mouse. In vitro experiments using the mouse liver homogenate have shown that l-lysine is converted to saccharopine in the presence of α-ketoglutarate and NADPH, and saccharopine to α-aminoadipate in the presence of NAD⁺. The oxidation of α-aminoadipic-δ-semialdehyde (Δ¹-piperideine-6-carboxylate), the proposed reaction product of saccharopine cleavage, to α-aminoadipate is effected by either NAD⁺ or NADP⁺.     

Inactivation of the lys7 gene, encoding saccharopine reductase in Penicillium chrysogenum, leads to accumulation of the secondary metabolite precursors piperideine-6-carboxylic acid and pipecolic acid from alpha-aminoadipic acid

Pipecolic acid serves as a precursor of the biosynthesis of the alkaloids slaframine and swainsonine (an antitumor agent) in some fungi. It is not known whether other fungi are able to synthesize pipecolic acid. Penicillium chrysogenum has a very active alpha-aminoadipic acid pathway that is used for the synthesis of this precursor of penicillin. The lys7 gene, encoding saccharopine reductase in P. chrysogenum, was target inactivated by the double-recombination method. Analysis of a disrupted strain (named P. chrysogenum SR1-) showed the presence of a mutant lys7 gene lacking about 1,000 bp in the 3'-end region. P. chrysogenum SR1- lacked saccharopine reductase activity, which was recovered after transformation of this mutant with the intact lys7 gene in an autonomously replicating plasmid. P. chrysogenum SR1- was a lysine auxotroph and accumulated piperideine-6-carboxylic acid. When mutant P. chrysogenum SR1- was grown with L-lysine as the sole nitrogen source and supplemented with DL-alpha-aminoadipic acid, a high level of pipecolic acid accumulated intracellularly. A comparison of strain SR1- with a lys2-defective mutant provided evidence showing that P. chrysogenum synthesizes pipecolic acid from alpha-aminoadipic acid and not from L-lysine catabolism.