Lima Bean Trypsin Inhibitor Research Articles

Two forms of basic trypsin-like arginine amidases in boar sperm.

Acrosin and newly detected basic arginine amidase were separated from boar sperm by affinity adsorption using lima bean trypsin inhibitor (LBTI) and aprotinin columns, respectively. These enzymes differed in various respects, including response against calcium chloride, amidolytic substrate specificity, and reaction against the inhibitor. They also differed widely in the affinity to LBTI. The difference appears to be expressed as the difference in ability of affinity of the enzymes with LBTI, thus leading to their separation.

Two forms of acidic arginine amidases in human kidney.

Two forms of acidic arginine amidases were separated from human kidney extract using the techniques of basic ion exchange adsorption and elution as well as lima bean trypsin inhibitor (LBTI) and aprotinin affinity adsorptions and elutions. The enzymes were tentatively named acidic human renal arginine amidase-L (AHRAA-L, with affinity to an LBTI column) and -A (AHRAA-A, with affinity to an aprotinin column). Both enzymes showed a similar molecular mass of approximately 3.0 x 10(4) daltons, differing from that of human renal kallikrein (HRK, molecular mass of 4.8 x 10(4) daltons). The specific activity of AHRAA-L and -A were 106 and 680 nmol/min/A280 of Val-Leu-Arg-pNA amidolysis, respectively, and they were strongly inhibited by LBTI and human urinary trypsin inhibitor (UTI), while ethylenglycol-bis(beta-amino ethylether)-N,N,N',N'-tetraacetic acid (EGTA) showed a weak or no effect on both enzymes.

Contact factor proteases and the complexes formed with alpha 2-macroglobulin can interfere in protein C assays by cleaving amidolytic substrates.

Plasma from women taking combined oral contraceptives and cold-activated plasma contain proteases which cleave chromogenic substrates in protein C assays in the absence of protein C activators such as Protac. This spontaneous activity makes a background substraction necessary and makes protein C (PC) assays less accurate. We investigated two commonly used substrates < Glu-Pro-Arg-pNA (S-2366) and 2AcOH.H-D-Lys(Cbo)-Pro-Arg-pNA (PC substrate) and found that cold-activated normal and protein C-deficient plasmas gave absorbance values up to 300 times higher than buffer blanks. FXIa cleaves these substrates but activity was not blocked by corn or lima bean trypsin inhibitors, soy bean trypsin inhibitor (SBTI), hirudin or epsilon-amino-n-caproic acid (EACA). Kaolin activation of normal, FXI, FIX, FVIII, FVII and protein C-deficient, but not of FXII or prekallikrein (PKK)-deficient plasmas led to cleavage of chromogenic substrate for protein C. The protein C substrates were cleaved by purified kallikrein and alpha- and beta-FXIIa. Immunoabsorption with alpha 2-macroglobulin (alpha 2M) antibodies removed 60% of the alpha 2M and 70% of the activity on PC Substrate. Gel filtration of normal plasma on Sephadex G-150 gave a single peak of protein C activity and antigen in the included volume. After cold activation of the fractions, a second protein C-like peak appeared in the void volume, but with no detectable protein C antigen. This peak coincided with alpha 2M (chromogenic and ELISA) and plasma kallikrein (S-2302), but FXII (measured with a substrate insensitive to kallikrein) eluted separately.(ABSTRACT TRUNCATED AT 250 WORDS)

Purification and characterization of a putative proenkephalin cleaving enzyme

A putative proenkephalin-cleaving enzyme (PCE) extracted from bovine adrenal chromaffin granules was purified with soybean trypsin inhibitor high-performance affinity chromatography. The 12,600-fold purified enzyme was maximally active at pH 8.0. The enzyme was completely inhibited with lima bean trypsin inhibitor (0.1 mg/ml), soybean trypsin inhibitor (0.1 mg/ml), and p-(chloromercuri)benzenesulfonic acid (1.0 m m), indicating PCE is a serine protease with cysteine residues likely to be involved in its structure or activity. It exhibited significant autoproteolysis without specific substrates present. The substrate specificity and kinetic constants with the enkephalin-containing (EC) peptides Leu-9 and proenkephalin Peptides B, E, and F as substrates were studied. The cleavage patterns were substantially different than with trypsin digestion. PCE specifically recognized the paired basic amino acid residues and predominantly cleaved the peptide bonds between Lys and Arg sites and peptide bonds after Lys-Lys and Arg-Arg sites. Different K m and V max values for the different Lys-Arg sites indicate sequences in addition to the paired basic residues can affect enzyme activity. Also, the lower K m and V max of Peptide E suggest a higher affinity for this peptide but much slower cleavage. The C-terminally located Lys-Arg site appears responsible for this high affinity. Based on these observations, we propose the following: (a) the primary structure of these peptides contains enough information to be processed correctly by PCE and (b) PCE may be regulated by pH and Peptide E to prevent extensive processing of the intermediate EC peptides which are the major opioid peptides found in the adrenal chromaffin granules.

Behavioural and physiological responses of grass grub larvae (Costelytra zealandica) feeding on protease inhibitors

Abstract Growth of first instar Costelytra zealandica larvae was significantly reduced after 6 weeks when reared on an artificial diet containing 0.3 and 1% soybean trypsin inhibitor (SBTI), 0.1% and 0.3% potato inhibitor II, and 0.3% potato protease inhibitor I and cowpea trypsin inhibitor. Limabean trypsin inhibitor at 1% significantly stimulated growth compared with that on diet with corresponding levels of added casein. A direct relationship between increased free-trypsin activity and decreased larval growth was observed. Sequential measurement of enzyme activity in third instar larvae feeding on SBTI was compared with that of larvae feeding on casein. The increase in enzyme activity was observed after 14 days in larvae feeding on SBTI. Larvae preferred to feed on SBTI-free diet when given a choice between diet containing this inhibitor at 0.3% and added casein at 0.3%.

Enzymatic action of basic arginine amidases in human seminal plasma.

Three basic arginine amidases with different affinities to lima bean trypsin inhibitor (LBTI) and aprotinin affinity columns were separated in the middle molecular weight (MMW) preparation obtained from Cellulofine GCL-2000 gel filtration of CM-cellulose adsorbed human seminal plasma and were tentatively called basic human seminal plasma arginine amidase-L (BHSAA-L, with affinity to LBTI), -A (BHSAA-A, with affinity to aprotinin), and -TH (BHSAA-TH, without affinity to either). Some enzymatic properties were measured, including Ki values of LBTI and human seminal plasma proteinase inhibitor (HSP-PI) toward present enzymes. The Ki values of LBTI toward BHSAA-L and -TH were lower than those of HSP-PI and no Ki values for LBTI toward BHSAA-L were observed. The Km values of BHSAA-L and -A to some tripeptidyl-p-nitroanilide substrates seemed relatively lower than that of BHSAA-TH.

Detection and separation of two kinds of acidic arginine amidases from boar sperm using lima bean trypsin inhibitor and aprotinin affinity adsorptions.

Two kinds of acidic arginine amidase activity were found in boar sperm. One enzyme was separated by a treatment consisting of lima bean trypsin inhibitor (LBTI) affinity adsorption and elution. The other enzyme was separated by aprotinin affinity adsorption and elution through the same solutions as those used for first enzyme; the two enzymes provisionally named boar sperm acidic arginine amidases 1 (BSAA-1) and 2 (BSAA-2), respectively. The amidolytic activity of BSAA-1 was increased by high concentrations of calcium chloride, while the activity of BSAA-2 was independent of calcium chloride. Their behavior with LBTI and aprotinin, and profiles of their substrate specificities, were also different. The affinity of LBTI to BSAA-1 was approximately 14 times higher than that to BSAA-2.

Proteolytic conversion of neurohypophyseal peptides by rat thymocytes: involvement of endopeptidases.

The proteolytic conversion of oxytocin and Arg8-vasopressin by purified rat thymocytes was studied at 37 degrees C and physiological pH 7.4. The formed peptide fragments were isolated by high performance liquid chromatography and characterized by amino acid analysis. When oxytocin was incubated with rat thymocytes, oxytocin 1-8 and oxytocin 1-7 were isolated. In contrast, only Arg8-vasopressin 1-8 was found when Arg8-vasopressin was incubated with thymocytes. The formation of oxytocin 1-8, oxytocin 1-7 and Arg8-vasopressin 1-8 was prevented partially by 10(-3) M phenylmethylsulfonyl fluoride and iodoacetamide, and abolished by 0.5 x 10(-3) M Zn2+ and Hg2+ ions and 10(-3) M o-phenanthroline, but not by 10(-5) M leupeptin, lima bean trypsin inhibitor, trasylol, captopril and phosphoramidon. 0.5 x 10(-3) M EDTA was without effect on the formation of oxytocin 1-8 and Arg8-vasopressin 1-8 but increased by about 30% the formation of oxytocin 1-7. The results suggest that proteases capable of metabolizing oxytocin and Arg8-vasopressin are localized in the thymocyte surface membrane. Since oxytocin and vasopressin are synthetized by thymic epithelial cells and exert several actions on thymocytes, these proteases may play a physiological role in the inactivation of neurohypophyseal peptides at the thymocyte level.

Trypsin-like arginine amidases including plasminogen and plasmin in human seminal plasma by affinity adsorption and elution.

An enzyme preparation with affinity to a lysine column was detected from a DEAE-cellulose-adsorbed preparation of human seminal plasma containing plasminogen and plasmin. Two kinds of trypsin-like acidic arginine amidase activity with different affinity to lima bean trypsin inhibitor (LBTI) and aprotinin affinity column were detected from the DEAE-cellulose-adsorbed preparation after treatment of the lysine column. Two kinds of trypsin-like basic arginine amidase activity were also separated by the above-mentioned affinity adsorptions from a CM-cellulose-adsorbed preparation of human seminal plasma. The effect of calcium chloride on these two enzymes was different from human acrosin.

Tryptase from rat skin: purification and properties

Tryptase was purified 13,000-fold to apparent homogeneity from rat skin. The two-step procedure involved ammonium sulfate fractionation of the initial extract followed by combined sequential affinity chromatography on agarose-glycyl-glycyl-p-aminobenzamidine and concanavalin A-agarose. The purified enzyme had a specific activity toward N-benzoylarginine ethyl ester (BzArgOEt) of 170 mumol/min mg-1 and was obtained in a yield of 28% as determined by the specific substrate, H-D-Ile-Pro-Arg-p-nitroanilide. Rat skin tryptase was thermal labile, losing 50% of its activity when preincubated for 30 min at 30 degrees C. The presence of NaCl (1 M) improved thermal stability and was necessary for long-term storage. Heparin did not stabilize the enzyme against thermal denaturation, and heparin-agarose failed to bind the enzyme. Rat skin tryptase was inhibited by diisopropylphosphofluoridate, antipain, leupeptin, and aprotinin but not by alpha 1-antitrypsin, ovomucoid, or soybean or lima bean trypsin inhibitors. Substrate specificity studies using a series of tri- and tetrapeptidyl-p-nitroanilide and peptidyl-7-amino-4-methylcoumarin substrates demonstrated the existence of an extended substrate binding site. Rat skin tryptase hydrolyzed [Arg8]vasopressin, neurotensin, and the oxidized B-chain of insulin at the -Arg8-Gly9-NH2, -Arg8-Arg9-, and -Arg22-Gly23-bonds, respectively. No general proteinase activity was observed toward casein, hemoglobin, or azocoll. Rat skin tryptase had a Mr of 145,000 by gel filtration. The subunit Mr was either 34,000 or 30,000 depending on the electrophoretic technique used. Treatment of the enzyme with peptide N-glycosidase F (N-glycanase) decreased the subunit Mr by 4000. The enzyme exhibited multiple isoelectric forms (pI's of 4.5-4.9). Rat skin tryptase was found to be related statistically to other tryptases on the basis of amino acid composition. The N-terminal amino acid sequence was Ile1-Val2-Gly3-Gly4-Gln5-Glu6-Ala7-+ ++Ser8-Gly9-Asn10-Lys11-Trp12-Pro13- Trp14- Gln15-Val16-Ser17-Leu18-Arg19-Val20- --21-Asp-22Thr23-Tyr24-Typ25-, with a putative glycosylation site at residue 21. This sequence was 72-80% homologous with the N-terminus of other tryptases but only 40% homologous with that of bovine trypsin.

Partial Purification and Characterization of a Putative Prohormone-Processing Enzyme Complex from Bovine Pituitary

A putative prohormone-processing enzyme complex with specificity toward basic residues was partially purified from whole bovine pituitary glands. The complex is basic, binding to S-Sepharose at pH 8.2. The pH optimum of the enzyme is around 8.0. The enzyme is capable of cleaving proenkephalin and is present in at least three forms with relative molecular masses of about 36,000, 58,000, and 90,000 Da. The proteinase complex is inhibited by soybean trypsin inhibitor, limabean trypsin inhibitor, and aprotinin, but not by inhibitors of thiol proteinases or metal chelators. Our results indicate that this proteinase is a trypsin-like serine esterase with properties appropriate to that of a prohormone-processing enzyme.

Purification and characterization of leucyl aminopeptidase and pyroglutamyl aminopeptidase from human s̀keletal muscle

The purification and characterization of leucyl aminopeptidase and pyroglutamyl aminopeptidase from human skeletal muscle are described. The characteristics of leucyl aminopeptidase were as follows: optimum activity was at pH 9.5 in the presence of 5 mmol/l Mg 2+ or 0.5 mmol Mn 2+. No activation of enzyme activity was obtained following addition of other divalent cations or sulphhydryl reagents. Only the leucyl-AMC and methionyl-AMC derivatives were appreciably hydrolysed. The mol mass was estimated as 280 kDa. Approx. 50% inhibition of activity was obtained following addition of p-hydroxymercuriphenyl sulphonate (10 /gmmol/l), N-ethyl maleimide (2 mmol/l), o-phenanthroline (5 mmol/l), bacitracin (1 mmol/l), amastatin (1 μg/ml) and bestatin (0.1 μmol/l); no inhibition of activity was obtained in the presence of phenylmethanesulphonyl fluoride (1 mmol/l), limabean trypsin inhibitor (100 μg/ml) or pepstatin (100 μg/ml). The following oligopeptides were hydrolysed by the enzyme: luliberin 7–10, proctolin and [Leu 5]enkephalin; oligopeptides not appreciably hydrolysed included neurotensin, angiotensin-I, substance-P and bradykinin. Pyroglutamyl aminopeptidase had the following characteristics: optimum activity was at pH 8.5 in the presence of 1 mmol/l dithiothreitol (an absolute requirement for maintenance of enzyme activity). Maximum activity was obtained in the absence of divalent cations. Only the pyroglutamyl-AMC derivative was appreciably hydrolysed. The mol mass of this enzyme was estimated as 22 kDa. Approximately 50% inhibition of activity was obtained on addition of phenanthroline (4 mmol/l) and antipain (7 μg/ml); no inhibition of activity was obtained following addition of phenyl methanesulphonyl fluoride (1 mmol/l), limabean trypsin inhibitor (100 /gmg/ml) or pepstatin (100 /gmg/ml). Only oligopeptides with a pyroglutamyl N-terminal residue (thyroliberin, neurotensin, and luliberin) were hydrolysed by the enzyme.

Human Acrosin: Purification and Some Properties

Human sperm with normal morphology and good viability were obtained by centrifugation using a discontinuous Percoll density gradient with an inner column. Acrosin (E.C.3.4.21.10) was rapidly purified from sperm by ion exchange adsorption and elution and was purified by affinity adsorption on a lima bean trypsin inhibitor (LBTI) Cellulofine column. The final preparation was found to be homogeneous on polyacrylamide gel electrophoresis and to have a molecular weight of about 4 x 10(4) daltons. The enzyme had an esterolytic activity of 3.5 mumol/min/A280 with N-alpha-tosyl-L-arginine methyl ester as the substrate. Human acrosin showed a broad substrate specificity for arginine and lysine derivatives and it seemed to have a somewhat different specificity from trypsin. The optimal pH of this enzyme with amidolytic activity was 9.0. Enzyme activity was stimulated by a high concentration of calcium chloride. LBTI and aprotinin strongly suppressed the amidolytic activity with the D-valyl-L-leucyl-L-arginine-p-nitroanilide (Val-Leu-Arg-pNA) as the substrate, but alpha 1-antitrypsin and soybean trypsin inhibitor were less effective.

Human cytotoxic lymphocyte granzyme B. Its purification from granules and the characterization of substrate and inhibitor specificity.

Granzyme B has been purified to homogeneity from the granules of a human cytolytic lymphocyte line, Q31, in an enzymatically active form by a three-step procedure. Q31 granzyme B hydrolyzed Na-t-butyloxycarbonyl-L-alanyl-L-alanyl-L-aspartyl (Boc-Ala-Ala-Asp) thiobenzyl ester with a kcat of 11 +/- 5 mol/s/mol enzyme and catalytic efficiency kcat/Km of 76,000 +/- 44,000 M-1 s-1. The hydrolysis of Boc-Ala-Ala-Asp thiobenzyl ester by crude Q31 Percoll fractions paralleled the tryptase activity for granule-containing fractions, which showed that granzyme B was associated with granules. When chromatographed on Sephacryl S-300, Q31 granzyme B eluted in two broad bands corresponding to dimer and monomer, both of which electrophoresed at 35 kDa in reducing NaDodSo4 polyacrylamide, and both of which showed a lag phase in assays. The lag phase in assays could be extended with 0.03 mM pepstatin. Upon elution from ion-exchange chromatography Q31 granzyme B electrophoresed at 32 kDa in reducing NaDodSO4 polyacrylamide and did not have a lag phase in assays. The amino-terminal sequence of the 32-kDa Q31 granzyme B was identical to four other human cytotoxic T-lymphocyte granzymes B in 18 of 18 positions sequenced. Purified Q31 granzyme B had a preference for substrates with Glu or Asp as the residue amino-terminal to the scissile bond; little or no activity was noted with oligopeptide substrates for trypsin-like, chymotrypsin-like, and elastase-like proteases. Human plasma alpha 1-protease inhibitor, human plasma alpha 2-protease macroglobulin, soybean and lima-bean trypsin inhibitors, bovine aprotinin, phosphoramidon, and chymostatin inhibited Q31 granzyme B. The inhibition by alpha 1-protease inhibitor was rapid enough to be of physiological significance.

Role of Gastric Juice in Feedback Regulation of Rat Pancreatic Secretion by Luminal Proteases

The role of gastric juice in the intestine on the pancreatic secretory response to intraduodenal infusion of trypsin inhibitors or to diversion of bile and pancreatic juice from the intestine was studied in conscious rats with pylorus ligation and gastric juice drainage. In absence of gastric juice in the intestine, diversion of bile and pancreatic juice from the intestine stimulated pancreatic secretion, but the incremental protein and fluid secretory responses to diversion of bile and pancreatic juice were increased approximately 2.9-fold and 2.5-fold, respectively, by intraduodenal infusion of HCl (60 microEq/h). Intraduodenal infusion of HCl (240 microEq/h) had no effect on the pancreatic secretory response to infusion of lima bean trypsin inhibitor (20 mg). These results support the hypothesis that the inhibitory effect of atropine on the pancreatic secretory response to diversion of pancreatic juice or bile and pancreatic juice is secondary to inhibition of gastric acid secretion. The lack of effect of HCl on the pancreatic response to trypsin inhibitor contradicts the hypothesis that acid in the intestine is important or necessary for the feedback response to loss of intraluminal protease activity. It is proposed that acid in the intestine augments the pancreatic response to diversion of pancreatic juice or bile and pancreatic juice by reducing intraluminal pH and thereby inactivating residual pancreatic proteases.

Purification of Tryptase From a Human Mast Cell Line

The neutral protease tryptase has been isolated from a human mast cell line, HMC-1. The HMC-1 line was established from the peripheral blood of a patient with mast cell leukemia and maintained as continuously proliferating clones in vitro and as solid mast cell tumors in nude mice. HMC-1-derived tryptase was purified by sequential chromatography on Dowex 1, DEAE 5 PW, and heparin-agarose. Purified tryptase has an apparent molecular weight of 150,000, as determined by molecular sieve HPLC, but migrates as a doublet of bands of 32/35,000 on SDS-PAGE gels. Maximal enzymatic activity was observed at pH 8.5. Cleavage of tosyl-L-arginine methyl ester by purified tryptase was inhibited by dansyl-L-glutamyl-glycyl-L-arginine chloromethyl ketone 2 HCl, HgCl2, tosyl-L-lysine chloromethyl ketone, leupeptin, and PMSF but not by benzamidine, aprotinin, tosyl-L-phenyl-alanine chloromethyl ketone, soybean trypsin inhibitor, human plasma, ovomucoid inhibitor, or lima bean trypsin inhibitor. Microsequencing of purified tryptase yielded an amino terminal sequence that was identical to that previously reported for human pituitary-derived tryptase.

Plasma inter‐α‐trypsin inhibitor‐related urinary glycoprotein EDC 1 inhibits the growth of a Burkitt's lymphoma cell line

A homogeneous preparation of a urinary glycoprotein has been isolated from urine of patients with malignant melanoma and advanced adenocarcinomas of colon and lung. This molecule, Mr 30 kDa, is homologous to EDC1, a proteinase inhibitor antigenically related to plasma inter-alpha-trypsin inhibitor (IATI) originally isolated from the urine of a leukemic patient, E.D. The newly isolated EDC1 inhibits cellular proliferation of a Burkitt's lymphoma cell line, Raji, growing in serum-free medium supplemented with insulin, transferrin, selenium, and linoleic acid. This concentration-dependent inhibitory effect was monitored in terms of change in cell number and 3H-thymidine incorporation. The growth of cells treated with approximately 3.3 pmol EDC1/ml was 50% that of the control group by both assays. EDC1 was not cytotoxic to the cells because the EDC1-treated cells excluded trypan blue and resumed normal growth after removal of EDC1. In addition, EDC1 treatment of Raji cells prelabeled with 3H-labeled DNA did not release more radioactivity into the conditioned medium than the untreated labeled cells. EDC1 did not affect the growth of Hs2B2, a B-lymphoblast cell line, and Hs294T, a human malignant melanoma cell line. Equimolar and larger quantities of other proteinase inhibitors with inhibitory profiles similar to that of EDC1 (alpha-1 proteinase inhibitor, soybean trypsin inhibitor, lima bean trypsin inhibitor, and turkey ovomucoid) did not affect the growth of Raji cells. Raji cells have an absolute requirement of transferrin as a nutrient and require insulin to modulate the expression of transferrin receptors. The cells also synthesize interleukin-1 as an autocrine growth stimulator. EDC1 did not form a detectable complex with transferrin, insulin, or any autocrine factor synthesized by the cells.

Purification and characterization of a kallikrein-like T-kininogenase.

A T-kininogenase has been purified to homogeneity from rat submandibular gland extracts by DEAE-Sepharose chromatography and preparative gel electrophoresis. The purified protein has an apparent Mr of 28,000 on sodium dodecyl sulfate-polyacrylamide gel electrophoresis and splits into heavy and light chains with Mr of 22,000 and 6,000 in the presence of dithiothreitol. It is an acidic glycoprotein with pI of 4.65-4.75. The carbohydrate moiety is located on the light chain and binds concanavalin A and wheat germ agglutinin. The active site serine residue of the heavy chain is labeled with [14C]diisopropylfluorophosphate and visualized by fluorography. NH2-terminal amino acid sequences of the light and heavy chains reveal 74-84% identity to rat tissue kallikrein, tonin, and other kallikrein-related enzymes. The enzyme cleaves T-kininogen to release T-kinin which was separated by high performance liquid chromatography on a reverse phase C18 column and identified by a kinin radioimmunoassay. Its T-kininogenase but not N-tosyl-L-arginine methyl ester esterase activity can be enhanced 10-fold in the presence of dithiothreitol. The esterolytic activity of the enzyme is inhibited by soybean trypsin inhibitor, aprotinin, leupeptin, and antipain; whereas lima bean and ovomucoid trypsin inhibitors stimulate its activity. The enzyme is localized at the granular convoluted tubule and striated duct cells in rat submandibular glands by immunohistochemistry. The results indicate that T-kininogenase belongs to the group of structurally similar yet distinct kallikrein-like serine proteases.

Evidence of two forms of basic arginine esterases in human male urine.

Two forms of basic arginine ester hydrolyzing enzymes were found in human male urine, and their partial purification was performed using lima bean trypsin inhibitor (LBTI) affinity adsorption and Cellulofine GCL-2000 gel chromatography. The forms tentatively called basic human urinary arginine esterase-1 (BHUAE-1) and -2 (BHUAE-2), had respective estimated molecular weights of about 4.5 x 10(4) and 1.8 x 10(4) daltons, with pH optima observed at 9.5 and 10.0. These two new enzymes in the human male urine were different from human renal urokinases.

Stress Initiated During Isolation of Rat Renal Proximal Tubules Limits in Vitro Survival

The effects of oxidative damage were assessed in rat proximal tubule fragments (isolated by collagenase perfusion) by monitoring lactate dehydrogenase release (LDH-R) to measure cell viability and thiobarbituric acid (TBA) reactive material to follow oxidative damage. Increasing the oxygen content in the incubation atmosphere from 10 to 95% significantly increased LDH-R and TBA reactants. Addition of butylated hydroxytoluene or deferoxamine (DF) to the medium prevented these changes, but ascorbic acid or mannitol had no positive effect. Lima bean trypsin inhibitor also reduced LDH leakage significantly when added to the medium, but not when added to the perfusion buffers. In contrast, adding DF to the perfusate during tubule isolation produced the most pronounced benefit; net LDH-R after 4 hr was about 10% in tubules prepared this way compared to 20% when DF was omitted. Basal oxygen consumption declined to approximately the same extent as LDH-R increased. Maintenance of nystatin-stimulated respiration, ATP/ADP, GSH content and total adenine nucleotides indicated good cell function. These results suggest that oxidative damage initiated during the tubule isolation procedure limits cell survival but this effect can be counteracted substantially by the addition of DF to the perfusion buffer.