Malabsorption of Folate Polyglutamates in Tropical Sprue

Malabsorption of folate polyglutamates prepared from yeast has been shown in eight patients with untreated tropical sprue and in three out of six patients receiving therapy for sprue. The absorptive defect for folate polyglutamates among these 14 patients occurred more frequently and in all but one patient more severely than for folic acid.Folate polyglutamates, the principal dietary form of folate, probably require deconjugation by the jejunal enzyme, folate conjugase, before absorption. The mean concentration of jejunal folate conjugase of 21 patients with untreated sprue and of 13 patients with sprue receiving therapy were both significantly less than the mean concentration in a control group. Nevertheless, all but five of the 34 patients had jejunal folate concentrations within the control range. There was no correlation in the individual patients between the jejunal folate conjugase concentration measured in vitro and the ability to absorb folate polyglutamates—nine patients having normal jejunal folate conjugase levels despite showing malabsorption of folate polyglutamates.     

Comparison of pteroylpolyglutamate hydrolase (folate conjugase) from porcine and human intestinal brush border membrane

1. A comparative study of pteroylpolygluatamte hydrolase (folate conjugase) of brush border membrane vesicles from human and porcine intestine was conducted. 2. The enrichment of conjugase activity during membrane isolation was 5-fold greater for the human than the pig. 3. Porcine and human conjugases exhibited similar Km values and could completely hydrolyze pteroyltriglutamate (PteGlu3) to PteGlu1 via an exohydrolytic process. 4. Pteroic acid, PteGlu1 and anionic polysaccharides did not inhibit human or porcine conjugase. 5. Apparent mol. wts for detergent-enzyme complexes were 237,000 (pig) and greater than 500,000 (human). 6. These results indicate similar kinetic properties and mode of action but differences in physical behavior between the intestinal brush border folate conjugases of human and pig.     

The possible role of pancreatic proteases in the turnover of intestinal brush border proteins.

1. Intestinal brush border enzymes have heterogeneous rates of turnover, the largest proteins having the fastest turnover. Since the membrane faces the intestinal lumen, the effects of pancreatic factors were examined in mediating this turnover. Surgical subtotal pancreatectomy was used as an experimental model to study the turnover of brush border proteins in the absence of most pancreatic secretions. 2. Subtotal (95%) pancreatectomy of rats was found to cause elevations by about 50% of total activity and specific activities of certain brush border enzymes (maltase, sucrase, lactase), but not of others (alkaline phosphatase, trehalase). Rats were judged to be functionally deficient in pancreatic proteolytic enzymes (a) by demonstration of vitamin B-12 malabsorption, which was corrected by trypsin, and (b) by the finding of only about 20% of proteolytic activity appearing in the lumen after a test meal when compared to control. 3. To measure protein turnover in vivo the method of double labelling was used, where [3H]- and [14C]valine were administered intraduodenally in sequence 10 h apart. With this technique, a high 3H/14C ratio is correlated with rapid turnover. Proteins with apparent molecular weights of about 200 000-270 000 were found to turn over more rapidly than smaller proteins. 3H/14C ranged from 4.7 to 6.2 in animals without pancreatic insufficiency. In the face of decreased pancreatic proteolysis, the 3H/14C ratio was 2.3-3.1, similar to that of proteins with a slow half life. 4. Estimates of relative synthetic rates of large brush border proteins were lower than normal in pancreatectomized animals, but were constant over the period of the labelling experiment. The high enzyme levels in the face of lower synthetic rates confirms that, at the new steady rate, degradation rates must be slower for large brush border proteins in pancreatic insufficiency. 5. In vitro, using purified brush borders, unfractionated pancreatic enzymes were found to remove sucrase, maltase and lactase, but not alkaline phosphatase and trehalase. The enzyme most potent in this respect was the pancreatic protease, elastase. Non-proteolytic enzymes (amylase, lipase, phospholipase A) were inactive in removing enzyme from the brush border. The addition of elastase to pancreatectomized animals in vivo restored the rapid turnover rate of large brush border proteins. 6. A model is thus proposed for the normal catabolism of some large intestinal brush border proteins. It is suggested that the surface of intestinal absorptive cells is being constantly remodelled, and that certain surface enzymes are in part removed from the membrane by the action of pancreatic proteases. A possible special role for elastase is suggested.     

Pteroylpolyglutamate hydrolase from human jejunal brush borders. Purification and characterization

Pteroylpolyglutamate hydrolase was solubilized with Triton X-100 from human jejunal mucosal brush borders and purified approximately 5,000-fold using organomercurial affinity chromatography, DEAE-cellulose chromatography, and gel filtration. The apparent molecular weight of the purified enzyme in the Triton micelle was estimated as 700,000 using Bio-Gel A-1.5m gel filtration. Sodium dodecyl sulfate/urea-polyacrylamide gel electrophoresis followed by Coomassie stain demonstrated two polypeptide bands at 145,000 and 115,000 daltons. The purified enzyme had an isoelectric point of 7.2, was maximally active at pH 5.5, and was stable above pH 6.5 and at temperatures up to 65 degrees C for at least 90 min. Human jejunal brush-border pteroylpolyglutamate hydrolase is an exopeptidase which liberated [14C]Glu as the sole labeled product of PteGlu2[14C]Glue (where PteGlun represents pteroylpolyglutamate), failed to liberate a radioactive product from PteGlu2[14C]GluLeu2, and released all possible labeled PteGlun products during incubation with Pte[14C]GluGlu6 with the accumulation of Pte[14C]Glu. PteGlu2, PteGlu3, and PteGlu7 were substrates, each with Km = 0.6 microM, whereas PteGlu was a weak inhibitor of the hydrolysis of PteGlu3 with Ki = 20 microM. Components of the pteroyl moiety, Glu, and short chain Glun in alpha or gamma linkages were not inhibitory. The enzyme was activated by Zn2+ or Co2+. The properties of brush-border pteroylpolyglutamate hydrolase are different from those described for the soluble intracellular pteroylpolyglutamate hydrolase in other species and in human mucosa, yet are consistent with previous data on the process of hydrolysis of PteGlun in the intact human intestine.     

Down-regulation of reduced folate carrier may result in folate malabsorption across intestinal brush border membrane during experimental alcoholism

Folate plays a critical role in maintaining normal metabolic, energy, differentiation and growth status of all mammalian cells. The intestinal folate uptake is tightly and diversely regulated, and disturbances in folate homeostasis are observed in alcoholism, attributable, in part, to intestinal malabsorption of folate. The aim of this study was to delineate the regulatory mechanisms of folate transport in intestinal absorptive epithelia in order to obtain insights into folate malabsorption in a rat model of alcoholism. The rats were fed 1 g.kg(-1) body weight of ethanol daily for 3 months. A reduced uptake of [(3)H]folic acid in intestinal brush border membrane was observed over the course of ethanol administration for 3 months. Folate transport exhibited saturable kinetics and the decreased intestinal brush border membrane folate transport in chronic alcoholism was associated with an increased K(m) value and a low V(max) value. Importantly, the lower intestinal [(3)H]folic acid uptake in ethanol-fed rats was observed in all cell fractions corresponding to villus tip, mid-villus and crypt base. RT-PCR analysis for reduced folate carrier, the major folate transporter, revealed that reduced folate carrier mRNA levels were decreased in jejunal tissue derived from ethanol-fed rats. Parallel changes were observed in reduced folate carrier protein levels in brush border membrane along the entire crypt-villus axis. In addition, immunohistochemical staining for reduced folate carrier protein showed that, in alcoholic conditions, deranged reduced folate carrier localization was observed along the entire crypt-villus axis, with a more prominent effect in differentiating crypt base stem cells. These changes in functional activity of the membrane transport system were not caused by a general loss of intestinal architecture, and hence can be attributed to the specific effect of ethanol ingestion on the folate transport system. The low folate uptake activity observed in ethanol-fed rats was found to be associated with decreased serum and red blood cell folate levels, which might explain the observed jejunal genomic hypomethylation. These findings offer possible mechanistic insights into folate malabsorption during alcoholism.     

Release of intestinal surface-membrane glycoproteins associated with enzyme activity by brief digestion with papain

Rat intestinal surface-membrane glycoproteins were labelled by intraperitoneal injection of [1-(14)C]glucosamine 4h before the animals were killed. At this time, density-gradient centrifugation of disrupted brush borders indicated that glycoprotein radioactivity was distributed identically with sucrase, a plasma-membrane marker. Labelled brush borders were digested by papain for brief time-intervals known to release surface-enzyme particles without disruption of the unit membrane. Digestion for 5min released 90% of the surface sucrase, and almost one-half of the brush-border glycoprotein and label. On Sepharose 4B column chromatography most of the glycoprotein and label emerged as a single peak. This peak contained the most actively labelled glycoprotein in the brush border and was closely associated with maltase, sucrase, beta-naphthylamidase and alkaline phosphatase. The peak was partially resolved on polyacrylamide-gel electrophoresis into three bands. Each band contained a distinctive enzyme or enzyme pair, and was labelled by [1-(14)C]glucosamine. No periodic acid-Schiff-negative protein was observed in the peak material. Glycoproteins susceptible to brief digestion with papain are therefore closely linked to released surface-enzyme particles. Intestinal surface glycoproteins are heterogeneous with respect to molecular weight, electrophoretic mobility and function.     

Effect of anions on folate binding by isolated brush border membranes from rat kidney

The characteristics of folate binding by brush border membranes from rat kidney homogenates were investigated. At pH 7.4, binding of [3′, 5′, 9-³H]-pteroylglutamic acid to membranes containing endogenous folate is inhibited by anions, with chloride being most effective followed by bromide, thiocyanate, iodide, phosphate and sulfate. A maximum inhibition of 70–75% is attained at a concentration of 0.1 M chloride and an incubation time of 30 min. The inhibition diminishes with increased incubation time and at 24 h is negligible. The binding of [3′,5′,9-³H]pteroylglutamic acid to brush border membranes stripped of endogenous folate by acid treatment is not inhibited by anions. Anion sensitivity can be restored to these treated membranes by reconstitution with membrane-derived folate, particularly 5-methyltetrahydropteroylglutamic acid, or by preincubation with synthetic 5-methyltetrahydropteroylglutamic acid. Inhibition of [3′,5′,9-³H]pteroylglutamic acid binding by anions in membranes with endogenous folate is best explained by an anion-induced stabilization of endogenous folate-binding protein complex resulting in a decreased rate of exchange with exogenous [3′,5′,9-³H]pteroylglutamic acid.     

Rat intestinal ceramidase: purification, properties, and physiological relevance

Neutral ceramidase activity has previously been identified in the intestinal mucosa and gut lumen and postulated to be important in the digestion of sphingolipids. It is found throughout the intestine but has never been fully characterized. We have purified rat intestinal neutral ceramidase from an eluate obtained by perfusing the intestinal lumen with 0.9% NaCl and 3 mM sodium taurodeoxycholate. Using a combination of acetone precipitation and ion-exchange, hydrophobic-interaction, and gel chromatographies, we obtained a homogenous enzyme protein with a molecular mass of approximately 116 kDa. The enzyme acts on both [14)]octanoyl- and [14C]palmitoyl-sphingosine in the presence of glycocholic and taurocholic acid and the bile salt analog 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonate but is inhibited by 2 mM or more of other bile salts. It is a glycosylated protein stable to trypsin and chymotrypsin exposure, is not influenced by Ca2+, Mg2+, or Mn2+, and is inhibited by Zn2+ and Cu2+. Mass fragmentographic analysis identified 12 fragments covering 17.5% of the sequence for neutral/alkaline ceramidase 2 purified (Mitsutake S, Tani M, Okino N, Mori K, Ichinose S, Omori A, Iida H, Nakamura T, and Ito M. J Biol Chem 276: 26249-262459, 2001) from rat kidney and located in apical membrane of renal tubular cells. Intestinal and kidney ceramidases also have similar molecular mass and ion dependence. Intestinal ceramidase thus is a neutral ceramidase 2 released by bile salts and resistant to pancreatic proteases. It is well suited to metabolize ceramide formed from dietary and brush border sphingolipids to generate other bioactive sphingolipid messengers.     

Ramipril inhibition of rabbit (Oryctolagus cuniculus) small intestinal brush border membrane angiotensin converting enzyme

1. Rabbit small intestinal brush border membranes possessed prominent angiotensin converting enzyme (ACE) activity. 2. Intestinal ACE was located on the lumen surface, as verified by ACE co-enrichment with brush border membrane marker enzymes. 3. Hydrolysis kinetics of rabbit intestinal ACE were comparable to the lung, utilizing the substrate (N-[3-(2-furyl)acryloyl]-L-phenylalanylglycylglycine; the Vmax = 543 +/- 51 mumol/min/g and Km = 0.62 +/- 0.09 mmol/l. 4. Intestinal brush border ACE activity was strongly inhibited by the antihypertensive drug Ramipril, which yielded an IC50 value of 5 nmol/l; the ACE activity remained completely inhibited during 15 days after a single dose of 10 mumol/l Ramipril.     

Binding of nicotinamide adenine dinucleotide by the renal brush border membrane from rat kidney cortex

The characteristics of nicotinamide adenine dinucleotide (NAD) binding on brush border membranes prepared from rat renal cortex were investigated with the use of radioactively labelled NAD, [adenine-2,8-3H]NAD+, as a ligand. (1) We found that NAD binds on brush border membrane and that the extent of NAD binding is linearly proportional to the brush border membrane protein, and progressively increases with concentration of NAD in the medium. (2) The rate of NAD binding was dependent on temperature. At 20 degrees C, the equilibrium binding was obtained at 15 min, while NAD binding at 0 degree C was slower, but the final level of binding reached at 120 min was similar to that plateau of binding observed at 20 degrees C. Brush border membrane inactivated by heating at 95 degrees C for 3 min did not bind NAD. Binding of NAD on brush border membranes was reversed by simple dilution or by the addition of unlabelled NAD. Both alpha-NAD and beta-NAD stereoisomers displaced bound [3H]NAD. Reduced NAD (NADH) caused less displacement of bound NAD than oxidized NAD+. Adenine, nicotinamide, pyrophosphate, of 5'-AMP did not displace bound NAD. (3) The NAD binding to brush border membranes was nearly saturable, approximating saturation at 10(-4) M NAD. Kinetic analysis by Scatchard plot indicates two sets of NAD binding sites in brush border membranes: a high-affinity binding site (Kd = 1.9 . 10(-5) M) and a low-affinity binding site (Kd = 2.2 . 10(-3) M). (4) Unlike concentrative uptake of D-[14C]glucose by brush border membrane vesicles, binding of NAD was not dependent on the presence of an outside-in sodium gradient [Na+0 greater than Na+i], nor was it abolished by repeated freezing and thawing of brush border membranes. Unlike D-[14C]glucose uptake, NAD binding by brush border membranes did not change upon decrease of intravesicular volume in hypertonic media. These observations indicate that NAD association with brush border membranes is true binding rather than intravesicular uptake of this compound. (5) The presence of specific binding sites in renal brush border membrane capable of binding of NAD with a high degree of affinity suggests that such sites may be involved in previously observed (Kempson, S.A., Colon-Otero, G., Ou, S.L., Turner, S.T. and Dousa, T.P. (1981) J. Clin. Invest. 67, 1347) modulatory effect of NAD on sodium-gradient-dependent uptake of phosphate across luminal brush border membrane of proximal tubules.     

The hydrolysis of nicotinamide adenine nucleotide by brush border membranes of rat intestine.

The hydrolysis of NAD by rat intestine was studied to determine the subcellular site of this hydrolysis and to identify the niacin-containing products that are formed. Using [nicotinamide-14C]NAD as substrate, and high pressure liquid chromatography for identification and quantification of products, the present study demonstrates two independent reactions for the hydrolysis of NAD; one that forms nicotinamide through hydrolysis of the ribosyl-pyridinium bond and one that forms nicotinamide mononucleotide through the hydrolysis of the pyrophosphate bond. The nicotinamide mononucleotide is subsequently dephosphorylated to nicotinamide riboside. Enzymes which release nicotinamide mononucleotide and nicotinamide riboside are associated with the brush border membrane as determined by analysis of fractionated intestinal homogenates. The enzyme activity which releases nicotinamide from NAD is associated with the brush border membrane fraction and also with a second cellular particulate fraction. Between pH5 and pH6 NAD is hydrolysed principally to nicotinamide. At pH 7.0 rates of nicotinamide and nicotinamide mononucleotide formation are the same. Above pH 7.0 the formation of nicotinamide mononucleotide is preferred.     

Characterization of brush border membrane-bound alkaline phosphatase activity in different segments of the porcine small intestine

This study was conducted to characterize enterocyte apical membrane-bound alkaline phosphatase activity in different segments of the porcine small intestine. Duodenal, jejunal, and distal ileal segments were isolated from three 26-kg pigs and enterocyte brush border membrane, enriched between 19- and 24-fold in sucrase specific activity, was prepared by Mg(2+) precipitation and differential centrifugation. With P-nitrophenyl phosphate as substrate, the optimum pH for porcine brush border membrane-bound alkaline phosphatase activity was defined to be 10.5 for all three segments. At the optimal pH, the kinetics of membrane-bound alkaline phosphatase were determined for the three intestinal segments. The affinity of this enzyme (K(m), mM) in the jejunum (0.64 +/- 0.07) was four times greater than that in the duodenum (2.75 +/- 0.59) and the distal ileum (2.71 +/- 1.14). These results indicate that different isomers of membrane-bound alkaline phosphatase might have been expressed in different segments of porcine small intestine. The maximal specific activity (V(max), micromol/mg protein . min) of this enzyme was highest in the duodenal (7.74 +/- 0.95), intermediate in the jejunal (4.31 +/- 0.18), and lowest in the distal ileal (3.53 +/- 0.84) brush border membrane. Therefore, the maximal specific activity of brush border membrane-bound alkaline phosphatase along the intestinal longitudinal axis in growing pigs decreases from the duodenum toward the distal ileum.     

Prednisolone enhances aminopeptidase turnover in adult rat small intestine

In adult male rats, fed prednisolone (0.75 mg/kg/day) for 7 days, brush border aminopeptidase activity was increased (P < 0.001) by 106% compared to pair-fed controls. [¹⁴C]Tyrosine was injected intraperitoneally 16 h and [³H]tyrosine 6 h before death. The ³H/¹⁴C ratio was 1.79 ± 0.21 (S.D.) in purified microvillus membranes from treated rats compared to 1.30 ± 0.16 (P < 0.01) in controls. Polyacrylamide gel electrophoresis of brush border membranes under denaturing conditions showed that the increased double-isotope ratio in membranes from treated rats was mainly in the high molecular weight protein subunits (> 80 kDa) Detergent-solubilized aminopeptidase was purified after in vivo labeling by protein A-Sepharose-antiaminopeptidase affinity chromatography. The ³H/¹⁴C ratio in aminopeptidase was 2.42 ± 0.15 (P < 0.05) in treated rats compared to 1.63 ± 0.13 in controls. Over the experimental period steady-state isotope reutilization and protein labeling was demonstrated and there was no isotope metabolism. Total microvillus membrane lipid content was unaffected by prednisolone. We conclude that prednisolone increases brush border aminopeptidase activity by increasing enzyme turnover. Other high molecular weight brush border proteins were similarly affected.     

Folate deficiency and pancreatic acinar cell function

The present study was designed to determine the effect of folate deficiency on pancreatic acinar cell function. In the first series of experiments, three groups of rats were fed ad libitum regular rat feed, folate-deficient diet, or an equivalent amount of folate-sufficient diet. In the second series of experiments, rats were either fed ad libitum or rendered folate deficient by a purified folate-deficient diet; half of the folate-deficient group was replenished with oral folate. Body weight, pancreatic weight, DNA [methyl-14C]thymidine incorporation into DNA, RNA, [8-14C]adenine incorporation into RNA, protein content, synthesis of proteins, amylase content, and basal and bethanechol-stimulated amylase secretion were determined. The parameters were the same in the rats fed a folate-sufficient diet as in those fed a regular rat feed. Feeding a folate-deficient diet resulted in impaired DNA synthesis as evidenced by diminished incorporation of [methyl-14C]thymidine into DNA. There was no change in secretion of amylase. Similar results were obtained in the second series of experiments. These studies indicate that folate deficiency (rather than antibiotic content of the diet) impaired pancreatic function. Folate deficiency may therefore contribute to pancreatic injury in malnutrition and alcoholism.     

Similarities between a dipeptide hydrolase from brush-border and cytosol fractions of guinea-pig intestinal mucosa.

A dipeptide hydrolase from the brush border of guinea-pig intestinal mucosa was purified. The enzyme resembles another dipeptide hydrolase isolated from the cytosol fraction of intestinal mucosa. Studies on the binding of cytosol peptide hydrolase to brush-border membranes indicate that the enzyme found in the brush border may be a cytoplasmic contaminant.     

Effect of antiserum to a 99 kDa polypeptide on the uptake of taurocholic acid by rat ileal brush border membrane vesicles

A 99 kDa polypeptide in rat ileal brush border membrane (BBM), regarded as a component of the active bile acid transport system on account of photoaffinity labeling, has been purified by affinity chromatography and preparative gel electrophoresis and utilized as an immunogen for raising polyclonal antibody. Immune serum, but not preimmune serum, specifically recognized a single band of 99 kDa protein on immunoblots of ileal and renal BBM. In contrast, no reactivity was observed with proteins in jejunal BBM. This polyclonal antibody, compared with preimmune serum and anticytosolic bile acid binding protein (14 kDa) serum, significantly inhibited the Na+ dependent uptake of [3H] taurocholate by BBM vesicles (p less than 0.01). [14C] D-glucose uptake by BBM vesicles was not influenced by the immune serum (p less than 0.01). Thus, these studies provide further support for the specific role of a 99 kDa protein in ileal BBM bile acid transport.     

Calcium uptake by isolated intestinal brush border membranes following dietary calcium restriction

The uptake of ⁴⁵Ca by isolated rat small intestinal brush border membranes was measured during the process of adaptation to dietary calcium deficiency. Uptake by membranes from the duodenum of calcium deficient rats was elevated compared to uptake by membranes prepared from control animals although no differences were seen comparing jejunal uptake rates. The results suggest that part of the adaptation producing increased intestinal transport by rats deprived of dietary calcium involves an increase in uptake by the duodenal brush border independent of other components of the transport system.     

Folate binding protein from kidney brush border membranes contains components characteristic of a glycoinositol phospholipid anchor.

A number of cell surface proteins have been shown to be anchored to the plasma membrane by a covalently attached glycoinositol phospholipid (GPL) in amide linkage to the C-terminus of the mature protein. We applied several criteria to establish that folate binding protein (FBP) in brush border membranes of rat kidney contains a GPL anchor. Brush border membranes were isolated and labeled with [3H]folate, and the complex of FBP and [3H]folate was shown to be released to the supernatant by incubation with purified bacterial phosphatidylinositol-specific phospholipase C (PIPLC) but not by incubation with a purified bacterial phosphatidylcholine-specific phospholipase C. The FBP-[3H]folate complex both in crude extracts and after FBP purification by ligand-directed affinity chromatography interacted with Triton X-114 micelles, and prior incubation with PIPLC prevented this detergent interaction. Individual residues characteristic of GPL anchors were found to be covalently associated with FBP following polyacrylamide gel electrophoresis in sodium dodecyl sulfate. These included glucosamine and ethanolamine, which were radiolabeled by reductive methylation and identified by chromatography on an amino acid analyzer, and inositol phosphate, which was inferred by Western blotting with an anti-CRD antisera. This antisera gave positive immunostaining only after FBP had been cleaved by PIPLC, a reliable diagnostic of a GPL anchor. The relationship between GPL-anchored FBP in biological membranes and soluble FBP in biological fluids also is discussed.     

Characterization of a processing protease that converts the precursor form of gamma-glutamyltranspeptidase to its subunits

Previously it was found that the proteolytic processing of precursors of gamma-glutamyltranspeptidase takes place on the brush border membrane of the kidney. The activity of the processing protease in purified brush border membranes was examined using endogenous substrates labeled with [3H]fucose and [35S]methionine. On incubation with brush border membranes in vitro, the precursors were converted stoichiometrically to two subunits, and the reaction followed first order kinetics with a rate constant k of -0.048 min-1. The enzyme responsible for this conversion was membrane-bound, had a weakly basic optimum pH and was inhibited by serine protease inhibitors. These results suggest that the precursor of gamma-glutamyltranspeptidase is processed to the mature form by a serine protease bound to the brush border membrane of kidney.     

Stereochemical action of mouse brain glutamate decarboxylase

4-[4-²H]Aminobutyrate was prepared by incubation in ²H₂O of glutamate with a partially purified glutamate decarboxylase from mouse brain. The 4R configuration was assigned to the compound on the basis of ¹H nmr analysis of the ω-camphanoylamide of its methyl ester in the presence of Eu(dpm)₃. Moreover 4-[4(S)4-³H,U-¹⁴C]aminobutyrate was shown to be formed from [2(S)2-³H,U-¹⁴C]glutamate by the same enzyme fraction. It is therefore demonstrated that glutamate decarboxylation catalyzed by this enzyme preparation occurs with retention of configuration.