Crystal structure of D-amino acid oxidase: a case of active site mirror-image convergent evolution with flavocytochrome b2

D-amino acid oxidase is the prototype of the FAD-dependent oxidases. It catalyses the oxidation of D-amino acids to the corresponding alpha-ketoacids. The reducing equivalents are transferred to molecular oxygen with production of hydrogen peroxide. We have solved the crystal structure of the complex of D-amino acid oxidase with benzoate, a competitive inhibitor of the substrate, by single isomorphous replacement and eightfold averaging. Each monomer is formed by two domains with an overall topology similar to that of p-hydroxybenzoate hydroxylase. The benzoate molecule lays parallel to the flavin ring and is held in position by a salt bridge with Arg-283. Analysis of the active site shows that no side chains are properly positioned to act as the postulated base required for the catalytic carboanion mechanism. On the contrary, the benzoate binding mode suggests a direct transfer of the substrate alpha-hydrogen to the flavin during the enzyme reductive half-reaction. The active site Of D-amino acid oxidase exhibits a striking similarity with that of flavocytochrome b2, a structurally unrelated FMN-dependent flavoenzyme. The active site groups (if these two enzymes are in fact superimposable once the mirror-image of the flavocytochrome b2 active site is generated with respect to the flavin plane. Therefore, the catalytic sites of D-amino acid oxidase and flavocytochrome b2 appear to have converged to a highly similar but enantiomeric architecture in order to catalvze similar reactions (oxidation of alpha-amino acids or alpha-hydroxy acids), although with opposite stereochemistry.

The toxic and hematologic effects of interleukin-1 alpha administered in a phase I trial to patients with advanced malignancies

PURPOSE

A phase I trial was undertaken because interleukin-1 alpha (IL-1 alpha) possesses antiproliferative, immunostimulatory, antiinfection, myeloprotective, and myelorestorative properties that could be beneficial in cancer treatment.

PATIENTS AND METHODS

In this phase I trial, IL-1 alpha was administered intravenously (IV) during a 15-minute period daily for 7 days to patients with advanced solid malignancies.

RESULTS

The maximum-tolerated dose (MTD) of IL-1 alpha alone was 0.3 microgram/kg. A second group of patients received indomethacin plus IL-1 alpha based on preclinical studies, which indicated that indomethacin could abrogate IL-1 alpha-induced hypotension; however, the MTD of IL-1 alpha plus indomethacin was 0.1 microgram/kg lower than IL-1 alpha alone. Fever, chills, headache, nausea, vomiting, and myalgia were common but were not dose-limiting. Hypotension resulted from a marked decrease in systemic vascular resistance and required pressors at 0.3 and 1.0 micrograms/kg IL-1 alpha. Dose-limiting toxicities included hypotension, myocardial infarction, confusion, severe abdominal pain, and renal insufficiency. IL-1 alpha treatment caused a significant, dose-related increase in the total WBC count (mainly segmented neutrophils and neutrophilic bands). Bone marrow cellularity increased because of enhanced numbers of relatively mature myeloid cells and megakaryocytes. Platelet counts decreased during therapy but were significantly elevated above baseline values 1 to 2 weeks posttreatment; this may have been an effect of IL-6 that was shown to be induced by IL-1 alpha treatment. Significant increases in triglycerides, cortisol, C-reactive protein, thyroid-stimulating hormone and decreases in cholesterol, testosterone, and protein-C were observed with treatment.

CONCLUSION

We conclude that at doses of IL-1 alpha that can be given safely to cancer patients, significant, potentially beneficial hematopoietic effects occur.

The effects of treatment with interleukin-1 alpha on platelet recovery after high-dose carboplatin

BACKGROUND

Thrombocytopenia is a frequent side effect of cancer chemotherapy and commonly limits attempts to escalate drug doses. To determine whether interleukin-1 alpha could ameliorate carboplatin-induced thrombocytopenia, we combined it with high-dose carboplatin in 43 patients with advanced neoplasms.

METHODS

High-dose carboplatin (800 mg per square meter of body-surface area) was administered alone to a control group. Subsequent patients were randomly assigned to receive the same dose of carboplatin with interleukin-1 alpha, administered either before or after carboplatin. Interleukin-1 alpha was given intravenously at a dose of 0.03, 0.1, or 0.3 microgram per kilogram of body weight per day for five days.

RESULTS

Carboplatin alone consistently produced thrombocytopenia with a median nadir of 19,000 platelets per cubic millimeter and a median of 10 days with less than 100,000 platelets per cubic millimeter. All 15 patients receiving interleukin-1 alpha before carboplatin had similar findings. In contrast, 5 of the 15 patients given one of the two higher doses of interleukin-1 alpha after carboplatin had minimal thrombocytopenia (nadir, 91,000 to 332,000 platelets per cubic millimeter). In the 10 patients given 0.3 microgram of interleukin-1 alpha per kilogram after carboplatin treatment, the platelet count recovered to 100,000 per cubic millimeter significantly earlier than in either the control group (P = 0.002) or the patients who received interleukin-1 alpha before carboplatin (P = 0.003), with the median times to recovery in the three groups being 16, 21, and 23 days, respectively. At the highest dose of interleukin-1 alpha, toxicity was substantial (but reversible), requiring inpatient support for hypotension, supraventricular arrhythmias, and pulmonary-capillary leak.

CONCLUSIONS

Interleukin-1 alpha can accelerate the recovery of platelets after high-dose carboplatin therapy and may be clinically useful in preventing or treating thrombocytopenia induced by chemotherapy.

Increased circulating nitrogen oxides after human tumor immunotherapy: correlation with toxic hemodynamic changes

BACKGROUND

Toxicity to interleukin-2 (IL-2) tumor immunotherapy is manifested principally by the vascular leak syndrome, hypotension, and a hyperdynamic response with low systemic vascular resistance. Nitric oxide (.N = O), a recently discovered biological mediator of vascular smooth muscle relaxation, is produced in increased amounts by numerous cell types exposed to a number of inflammatory cytokines.

PURPOSE

Our purpose was to determine if there is an increased production of .N = O in patients receiving IL-2 tumor immunotherapy, and, if so, whether increases in .N = O production correlate with hemodynamic instability.

METHODS

Twelve patients undergoing immunotherapy trials with IL-2 and anti-CD3 monoclonal antibody-activated lymphocytes (T-AK cells) were studied. Plasma levels of nitrate (NO3-), the stable end metabolic product of .N = O synthesis, were measured before and at the end of IL-2 treatment cycles.

RESULTS

We observed a ninefold increase in plasma levels of NO3- in patients after 7 days of treatment (P less than .0001). A significant decrease in both systolic and diastolic blood pressures was observed in all patients (P less than .001).

CONCLUSIONS

We propose that mediated induction of .N = O synthase enzyme leads to progressive increases in .N = O production which, in turn, produces clinically significant hypotension.

IMPLICATIONS

Since .N = O synthesis can be competitively inhibited by L-arginine analogues, a possible pharmacologic modulation of .N = O production could potentially contribute to better management of toxic side effects seen in IL-2 cancer therapies.

Active site plasticity in D-amino acid oxidase: a crystallographic analysis

D-Amino acid oxidase (DAAO) is the prototype of the flavin-containing oxidases. It catalyzes the oxidative deamination of various D-amino acids, ranging from D-Ala to D-Trp. We have carried out the X-ray analysis of reduced DAAO in complex with the reaction product imino tryptophan (iTrp) and of the covalent adduct generated by the photoinduced reaction of the flavin with 3-methyl-2-oxobutyric acid (kVal). These structures were solved by combination of 8-fold density averaging and least-squares refinement techniques. The FAD redox state of DAAO crystals was assessed by single-crystal polarized absorption microspectrophotometry. iTrp binds to the reduced enzyme with the N, C alpha, C, and C beta atoms positioned 3.8 A from the re side of the flavin. The indole side chain points away from the cofactor and is bound in the active site through a rotation of Tyr224. This residue plays a crucial role in that it adapts its conformation to the size of the active site ligand, providing the enzyme with the plasticity required for binding a broad range of substrates. The iTrp binding mode is fully consistent with the proposal, inferred from the analysis of the native DAAO structure, that substrate oxidation occurs via direct hydride transfer from the C alpha to the flavin N5 atom. In this regard, it is remarkable that, even in the presence of the bulky iTrp ligand, the active center is made solvent inaccessible by loop 216-228. This loop is thought to switch between the "closed" conformation observed in the crystal structures and an "open" state required for substrate binding and product release. Loop closure is likely to have a role in catalysis by increasing the hydrophobicity of the active site, thus making the hydride transfer reaction more effective. Binding of kVal leads to keto acid decarboxylation and formation of a covalent bond between the keto acid C alpha and the flavin N5 atoms. Formation of this acyl adduct results in a nonplanar flavin, characterized by a 22 degrees angle between the pyrimidine and benzene rings. Thus, in addition to an adaptable substrate binding site, DAAO has the ability to bind a highly distorted cofactor. This ability is relevant for the enzyme's function as a highly efficient oxidase.

Cross-talk and ammonia channeling between active centers in the unexpected domain arrangement of glutamate synthase

INTRODUCTION

The complex iron-sulfur flavoprotein glutamate synthase catalyses the reductive synthesis of L-glutamate from 2-oxoglutarate and L-glutamine, a reaction in the plant and bacterial pathway for ammonia assimilation. The enzyme functions through three distinct active centers carrying out L-glutamine hydrolysis, conversion of 2-oxoglutarate into L-glutamate, and electron uptake from an electron donor.

RESULTS

The 3.0 A crystal structure of the dimeric 324 kDa core protein of a bacterial glutamate synthase was solved by the MAD method, using the very weak anomalous signal of the two 3Fe-4S clusters present in the asymmetric unit. The 1,472 amino acids of the monomer fold into a four-domain architecture. The two catalytic domains have canonical Ntn-amidotransferase and FMN binding (beta/alpha)8 barrel folds, respectively. The other two domains have an unusual "cut (beta/alpha)8 barrel" topology and an unexpected novel beta-helix structure. Channeling of the ammonia intermediate is brought about by an internal tunnel of 31 A length, which runs from the site of L-glutamine hydrolysis to the site of L-glutamate synthesis.

CONCLUSIONS

The outstanding property of glutamate synthase is the ability to coordinate the activity of its various functional sites to avoid wasteful consumption of L-glutamine. The structure reveals two polypeptide segments that connect the catalytic centers and embed the ammonia tunnel, thus being ideally suited to function in interdomain signaling. Depending on the enzyme redox and ligation states, these signal-transducing elements may affect the active site geometry and control ammonia diffusion through a gating mechanism.

Properties of D-amino-acid oxidase from Rhodotorula gracilis

The flavoprotein D-amino-acid oxidase was purified to homogeneity from the yeast Rhodotorula gracilis by a highly reproducible procedure. The amino acid composition of the protein was determined; the protein monomer had a molecular mass of 39 kDa and contained one molecule of FAD. The ratio between A274/A455 was about 8.2. D-Amino-acid oxidase from yeast showed typical flavin spectral perturbations on binding of the competitive inhibitor benzoate and was reduced by D-alanine under anaerobiosis. The enzyme reacted readily with sulfite to form a covalent reversible adduct and stabilized the red anionic form of the flavin semiquinone on photoreduction in the presence of 5-deazariboflavin; the 3,4-dihydro-FAD form was not detectable after reduction with sodium borohydride. Thus D-amino-acid oxidase from yeast exhibited most of the general properties of the dehydrogenase/oxidase class of flavoproteins; at the same time, the enzyme showed some peculiar features with respect to the same protein from pig kidney.

A study on apoenzyme from Rhodotorula gracilis D-amino acid oxidase

The apoenzyme of D-amino acid oxidase from Rhodotorula gracilis was obtained at pH 7.5 by dialyzing the holoenzyme against 2 M KBr in 0.25 M potassium phosphate, 0.3 mM EDTA, 5 mM 2-mercaptoethanol and 20% glycerol. To recover a reconstitutable and highly stable apoprotein, it is essential that phosphate ions and glycerol be present at high concentrations. Apo-D-amino acid oxidase is entirely present as a monomeric protein, while the reconstituted holoenzyme is a dimer of 79 kDa. The equilibrium binding of FAD to apoprotein was measured from the quenching of flavin fluorescence and by differential spectroscopy: a Kd of 2.0 x 10(-8) M was calculated. The kinetics of formation of the apoprotein-FAD complex were studied by the quenching of protein and flavin fluorescence, by differential spectroscopy and by activity measurements. In all cases a two-stage process was shown to be present with a fairly rapid first phase, followed by a slow secondary change which represents only 4-6% of the total recombination process. In no conditions was a lag in the recovery of maximum catalytic activity observed. The process of FAD binding to yeast D-amino acid oxidase appears to be of the type Apo + FAD in equilibrium holoenzyme, even though the existence of a transient intermediate not detectable under our conditions cannot be ruled out.

Molecular heterogeneity of ferredoxin-NADP+ reductase from spinach leaves

Ferredoxin-NADP+ reductase (NADPH: ferredoxin oxidoreductase, EC 1.6.7.1) from spinach leaves has been purified according to a new procedure. The enzyme shows the presence of five molecular forms as identified by isoelectric focusing, namely a, b, c, d and e with pI values of 6.0, 5.5, 5.2, 5.0 and 4.8, respectively. All the bands are catalytically active and are clearly identifiable after the first steps of the purification procedure. The basic pattern of the ferredoxin-NADP+ reductase forms is the same whether extracted from one or many spinach plants and is not affected by the different purification procedures used. Two distinct classes of molecular weight have been found for the isolated forms b, c and d as measured by sodium dodecyl sulphate electrophoresis, with values of 33 000-34 000 for the first and 36 000-38 000 for the later two forms. Gel electrophoresis in non-denaturing media at different gel concentrations gives the same order of molecular weight values, thus ruling out the possibility that the native enzyme is a dimer, as has been reported by Schneeman, R. and Krogmann, D.W. ((1975) J. Biol. Chem. 250, 4965-4971). No significant kinetic differences were detectable for the isolated forms of ferredoxin-NADP+ reductase.

Hepatobiliary lymphoepithelioma-like carcinoma associated with Epstein-Barr virus

We describe the clinical and pathologic features of a lymphoepithelioma-like carcinoma (LELC) that originated in the hepatobiliary system. A woman, aged 71 years, was first seen with a noncholangiolar adenocarcinoma with lymphoid stroma, which was discovered by open liver biopsy in 1993. In 1995, retroperitoneal and peripancreatic lymph nodes were involved by LELC. There currently is no evidence of distant metastasis outside the hepatobiliary peripancreatic region. Review of the biopsy material revealed a well-differentiated adenocarcinoma with transition into LELC. Epstein-Barr virus (EBV) transcripts were expressed in all histologic phases of the tumor by in situ hybridization using immunoalkaline phosphatase-labeled oligonucleotide probes for EBV-encoded RNA 1 on formalin-fixed, paraffin-embedded sections. Polymerase chain reaction analysis for EBV nuclear antigen 2 was consistent with EBV strain type A. The LMP-1 gene was found to be wild type by polymerase chain reaction analysis. To our knowledge, this is the first report of a primary hepatobiliary adenocarcinoma associated with EBV infection that transformed into an undifferentiated LELC.

Involvement of serine 96 in the catalytic mechanism of ferredoxin-NADP+ reductase: structure--function relationship as studied by site-directed mutagenesis and X-ray crystallography

The crystal structure of ferredoxin-NADP+ reductase (FNR) suggests that Ser96 is directly involved in hydride transfer between the isoalloxazine moiety of FAD and the nicotinamide ring of NADP(H). To probe its role, Ser96 has been mutated to valine (S96V) and glycine (S96G). These mutations primarily affected the interaction of the nicotinamide ring with the flavin. Absorbance, fluorescence, and circular dichroism spectra and the crystal structure of FNR-S96V indicate that this mutant folds properly. FNR-S96V shows only 0.05% of wild-type activity, while the affinities for both ferredoxin and NADP+ are virtually unchanged. However, spectral perturbations induced by NADP+ binding to FNR-S96V strongly resemble those elicited by the binding of 2'-monophosphoadenosine-5'-diphosphoribose, a substrate analog lacking the nicotinamide ring, both to the mutant and wild-type enzymes. Rapid reaction studies on the valine mutant failed to detect charge-transfer intermediates during flavin reduction by NADPH. In addition, no semiquinone formation was seen during photoreduction of FNR-S96V. The three-dimensional structure of the valine mutant shows small, albeit definite, changes only in the isoalloxazine microenvironment. The glycine mutant of FNR displays behavior intermediate between that of wild-type enzyme and that of the valine mutant. It maintains ca. 2% of the wild-type activity as well as the ability to form the charge-transfer species between reduced FNR and NADP+. In photoreduction experiments, the same degree of flavin semiquinone stabilization was observed with FNR-S96G and with the wild-type enzyme. NADP+ binding to the glycine mutant was very similar to that observed in the case of the valine mutant.(ABSTRACT TRUNCATED AT 250 WORDS)

The role of cysteine residues of spinach ferredoxin-NADP+ reductase As assessed by site-directed mutagenesis

To investigate the functional role of the cysteine residues present in the spinach ferredoxin-NADP+ oxidoreductase, we individually replaced each of the five cysteine residues with serine using site-directed mutagenesis. All of the mutant reductases were correctly assembled in Escherichia coli except for the C42S mutant protein. C114S and C137S mutant enzymes apparently showed structural and kinetic properties very similar to those of the wild-type reductase. However, C272S and C132S mutations yielded enzymes with a decreased catalytic activity in the ferredoxin-dependent reaction (14 and 31% of the wild type, respectively). Whereas the C132S was fully competent in the diaphorase reaction, the C272S mutant flavoprotein showed a 35-fold reduction in catalytic efficiency with respect to the wild-type enzyme (0.4 versus 14.28 microM-1 s-1) due to a substantial decrease of kcat. NADP+ binding by the C272S mutant enzyme was apparently quantitatively the same (Kd = 37 microM) but qualitatively different, as shown by the differential spectrum. Stopped-flow experiments showed that the enzyme-FAD reduction rate was considerably decreased in the C272S mutant reductase, along with a much lower yield of the charge-transfer transient species. It is inferred from these data that the charge transfer (FAD-NADPH) between the reductase and NADPH is required for hydride transfer from the pyridine nucleotide to flavin to occur with a rate compatible with catalysis.

Phase I trial of anti-CD3-stimulated CD4+ T cells, infusional interleukin-2, and cyclophosphamide in patients with advanced cancer

PURPOSE

We performed a phase I trial to determine whether in vivo expansion of activated CD4+ T cells was possible in cancer patients. 111Indium labeling was used to observe trafficking patterns of the infused stimulated CD4+ T cells. The influence of cyclophosphamide (CTX) dosing on immunologic outcome was also examined.

PATIENTS AND METHODS

Patients with advanced solid tumors or non-Hodgkin's lymphoma received CTX at 300 or 1,000 mg/m2 intravenously (i.v.). Leukapheresis was performed to harvest peripheral-blood mononuclear cells (PBMCs) either just before the CTX dose, or when the patient was either entering or recovering from the leukocyte nadir induced by CTX. An enriched population of CD4+ T cells was obtained by negative selection. The CD4+ T cells were activated ex vivo with anti-CD3, cultured with interleukin-2 (IL-2) for 4 days, and adoptively transferred. After adoptive transfer, patients received IL-2 (9.0 x 10(6) IU/m2/d) by continuous infusion for 7 days.

RESULTS

The absolute number of CD4+, CD4+/DR+, and CD4+/CD45RO+ T cells increased in a statistically significant fashion in all cohorts after the first course of therapy. The degree of CD4 expansion was much greater than CD8 expansion, which resulted in a CD4:CD8 ratio that increased in 26 of 31 patients. The greatest in vivo CD4 expansion occurred when cells were harvested as patients entered the CTX-induced nadir. One complete response (CR), two partial responses (PRs), and eight minor responses were observed. Trafficking of 111Indium-labeled CD4 cells to subcutaneous melanoma deposits was also documented.

CONCLUSION

CD4+ T cells can be expanded in vivo in cancer patients, which results in increased CD4:CD8 ratios. The timing of pheresis in relation to CTX administration influences the degree of CD4 expansion. Tumor responses with this regimen were observed in a variety of tumors, including melanoma and non-Hodgkin's lymphoma; a high percentage of patients had at least some tumor regression from the regimen that produced the greatest CD4+ T-cell expansion.

Limited proteolysis and X-ray crystallography reveal the origin of substrate specificity and of the rate-limiting product release during oxidation of D-amino acids catalyzed by mammalian D-amino acid oxidase

Limited proteolysis of D-amino acid oxidase holoenzyme with trypsin cleaves the protein at Arg 221 and near the C-terminus, producing stable 25, 13.4, and 2 kDa polypeptides [Torri-Tarelli, G., Vanoni, M. A., Negri, A., & Curti, B. (1990) J. Biol. Chem. 265, 21242-21246]. The 25 and 13.4 kDa polypeptides remain associated to form a nicked D-amino acid oxidase species. This nicked protein form maintains the ability to bind FAD, but exhibits altered catalytic efficiency toward the oxidation of various D-amino acids when compared to native DAAO. Changes in substrate specificity were first monitored by measuring the activity in the presence of different amino acid substrates at various times during proteolysis. Three amino acid substrates were then selected for further analysis of the properties of the nicked D-amino acid oxidase species produced by limited tryptic proteolysis: D-serine, D-arginine, and D-alanine. The three D-amino acids represented limiting cases of the observed changes of enzyme activity on nicking: loss of activity, increase of activity, and minor activity changes, respectively. D-serine was found to be no longer a substrate of D-amino acid oxidase. D-arginine exhibited a 2.5-fold increased apparent maximum velocity although its Km value increased 2-fold with the nicked enzyme in comparison to the native species. D-alanine was oxidized 1.5-fold faster by the nicked D-amino acid oxidase at infinite substrate concentration, and its Km value increased approximately 4-fold. The Kd for benzoate, which was determined kinetically with D-alanine as the enzyme substrate, increased 17-fold in the nicked species. Primary deuterium kinetic isotope effects on V and V/K during the oxidation of D-alanine were also measured. (D)V/K increased from 1.4 +/- 0.2 to 1.8 +/- 0.3 on nicking, while (D)V increased from 1.04 +/- 0.1 to 2.53 +/- 0.5. All the observed changes of the values of the kinetic parameters and of the observed isotope effects are consistent with the hypothesis that nicking of D-amino acid oxidase at position 221 decreases the strength of binding of both substrates and products to the enzyme active site. The information obtained by limited tryptic proteolysis nicely complements that gathered from the analysis of the three-dimensional structure of D-amino acid oxidase in complex with benzoate, which was recently determined [Mattevi, A., Vanoni, M. A., Todone, F., Rizzi, M., Teplyakov, A., Coda, A., Bolognesi, M., & Curti, B. (1996) Proc. Natl. Acad. Sci. U.S.A. 93, 7496-7501]. Arginine 221 is part of the 216-228 loop that covers the active site and contributes residues to substrate binding and catalysis. The limited proteolysis data support the hypothesis that this loop acts as a lid on the active site and controls both substrate specificity and the rate of turnover of D-amino acid oxidase.

Characterization of the flavins and the iron-sulfur centers of glutamate synthase from Azospirillum brasilense by absorption, circular dichroism, and electron paramagnetic resonance spectroscopies

Azospirillum brasilense glutamate synthase has been studied by absorption, electron paramagnetic resonance, and circular dichroism spectroscopies in order to determine the type and number of iron-sulfur centers present in the enzyme alpha beta protomer and to gain information on the role of the flavin and iron-sulfur centers in the catalytic mechanism. The FMN and FAD prosthetic groups are demonstrated to be non-equivalent with respect to their reactivities with sulfite. Sulfite reacts with only one of the two flavins forming an N(5)-sulfite adduct with a Kd of approximately 1 mM. The enzyme-sulfite complex is reduced by NADPH, and the complexed sulfite is competitively displaced by 2-oxoglutarate, which suggests the reactive flavin to be at the imine-reducing site. These data are in agreement with the two-site model of the enzyme active center proposed on the basis of kinetic studies [Vanoni, M.A., Nuzzi, L., Rescigno, M., Zanetti, G., & Curti, B. (1991) Eur. J. Biochem. 202, 181-189]. Each enzyme protomer was found, by chemical analysis, to contain 12.1 +/- 0.5 mol of non-heme iron. Electron paramagnetic resonance spectroscopic studies on the oxidized and reduced forms of glutamate synthase demonstrated the presence of three distinct iron-sulfur centers per enzyme protomer. The oxidized enzyme exhibits an axial spectrum with g values at 2.03 and 1.97, which is highly temperature-dependent and integrates to 1.1 +/- 0.2 spin/protomer. This signal is assigned to a [3Fe-4S]1+ cluster (Fe-S)I. Reduction of the enzyme with an NADPH-regenerating system results in reduction of the [3Fe-4S]1+ center to a species with a g approximately 12 signal characteristic of the S = 2 spin state of a [3Fe-4S]0 cluster. The NADPH-reduced enzyme also exhibits an [Fe-S] signal at g values of 1.98, 1.95, and 1.88, which integrates to 0.9 spin/protomer and is due to a second cluster (Fe-S)II. Reduction of the enzyme with the light/deazaflavin method results in a signal characteristic of [Fe-S] clusters with g values of 2.03, 1.92, and 1.86 and an integrated intensity of 1.9 spin/protomer. This signal arises from reduction of the (Fe-S)II center and from that of the third, lower potential iron-sulfur center (Fe-S)III. Circular dichroism spectral data on the oxidized and reduced forms of the enzyme are more consistent with the assignment of (Fe-S)II and (Fe-S)III as [4Fe-4S] clusters rather than [2Fe-2S] centers.

Purification and characterization of glutamate synthase from Azospirillum brasilense

Growth conditions for Azospirillum brasilense Sp6 were devised for maximal expression of glutamate synthase. The enzyme levels were largely affected by the type and concentration of the nitrogen source. A 10-fold increase in the synthesis of the enzyme was observed at a limiting concentration of ammonia. The enzyme was purified to homogeneity by a procedure which was fairly rapid and allowed a good recovery of enzyme (30%). Azospirillum glutamate synthase is a complex iron-sulfur flavoprotein with a stoichiometry of 1 flavin adenine dinucleotide:1 flavin mononucleotide:8 Fe:8 S per protomer with a molecular weight of 185,000. The protomer is composed of two dissimilar subunits with molecular weights of 135,000 and 50,000. Kinetic parameters were determined. Km values for NADPH, 2-oxoglutarate, and L-glutamine were 6.25, 29, and 450 microM, respectively. The optimum pH was about 7.5. Complete reduction of the enzyme under anaerobic conditions was obtained either by NADPH (in the presence of a regenerating system) or dithionite or by photochemical reduction (in the presence of EDTA and 5-deazariboflavin). No stable long-wavelength intermediates were observed.

Expression in Escherichia coli of ferredoxin:NADP+ reductase from spinach. Bacterial synthesis of the holoflavoprotein and of an active enzyme form lacking the first 28 amino acid residues of the sequence

A cDNA clone for the preprotein of spinach ferredoxin:NADP+ reductase has been modified to allow the expression in Escherichia coli of the mature flavoprotein form the lacks the transit peptide. An expression vector, pFNR1, was constructed by subcloning the fragment into the plasmid pDS12/RBSII, SphI. In the crude extracts of transformed cells after induction, two active holoproteins of 35 kDa and 32 kDa, respectively, were found. The 32-kDa protein, purified by immunoaffinity chromatography, was found to lack the first 28 residues of the spinach protein sequence and to have a methionine as the N-terminal residue instead of Val29. A new expression plasmid, pFNR2, was obtained by in vitro mutagenesis of the codon GTG for Val29 to the synonymous GTT; in this case, only the 35-kDa protein was expressed by transformed cells. Both the 35-kDa and 32-kDa enzymes were purified and characterized. All the properties analyzed of the cloned 35-kDa enzyme were very similar to those of the spinach flavoprotein. The 32-kDa form showed the same catalytic efficiency of the spinach enzyme as a diaphorase but its interaction with oxidized ferredoxin was partially impaired.

Treatment of cancer patients with ex vivo anti-CD3-activated killer cells and interleukin-2

PURPOSE

This study describes the physiologic and biologic effects resulting from the adoptive transfer of ex vivo anti-CD3-stimulated T-killer cells (T-AK) to patients with advanced cancer in combination with interleukin-2 (IL-2).

METHODS

Autologous peripheral-blood mononuclear cells were obtained by leukapheresis and stimulated ex vivo with anti-CD3. The stimulated cells were reinfused at one of three dose levels on the next day (5 x 10(9), 7.5 x 10(9), and 1 x 10(10)). Cell administration was followed by IL-2 given by bolus and continuous infusion (1.5 x 10(6) U/m2 and 3.0 x 10(6) U/m2, respectively) for 7 days, or continuous infusion alone (3.0 x 10(6) U/m2) for 14 days.

RESULTS

Pronounced leukocytosis and atypical lymphocytosis were observed with individual values as high as 80,000 and 50,000 cells/microL, respectively. The other major clinical sequelae included a marked lactic acidosis with bicarbonate levels as low as 4.0 mmol/L in some patients, and prolongation of the prothrombin time (PT) and partial thromboplastin time (PTT) due to decreases in clotting factors VII, IX, and X. Antithrombin III levels were also reduced. Hypotension associated with increased serum nitrate and neopterin levels was observed. These toxicities were accompanied by increases in hepatocellular enzymes and creatinine previously described with IL-2. These events occurred at a time when the number of circulating T-AK cells reached their peak. The amount of bolus IL-2 correlated with increases in WBC count (P = .0311), atypical lymphocytes (P = .0241), PT (P = .0006), and PTT (P = .0122).

CONCLUSION

Substantial in vivo expansion of activated T lymphocytes was induced by a protocol combining ex vivo activation of peripheral-blood cells with anti-CD3 antibody followed by adoptive transfer and IL-2 administration. The synchronous expansion of these T cells superimposed on diminished liver and kidney function from IL-2 can cause profound but reversible metabolic changes.

Interleukin 1 alpha increases serum leptin concentrations in humans

Leptin, the protein product of the ob gene, regulates appetite and body weight in animals. Endotoxin and cytokines, induced by endotoxin, interleukin (IL) 1 and tumor necrosis factor, increase expression of leptin in mice and hamsters. We measured serum leptin concentrations in patients with cancer before and after administration of recombinant human IL-1 alpha. Fourteen patients received IL-1 alpha at one of three dose levels (0.03, 0.1, or 0.3 microgram/kg.day) for 5 days. Serum leptin concentrations increased in all but two patients within 24 h after the first dose. The increase in leptin was correlated directly with IL-1 alpha dose (P = 0.0030). Despite continued administration of IL-1 alpha, serum leptin concentrations returned to pretreatment levels by day 5 of therapy. An increase in serum leptin concentrations may be one mechanism by which anorexia is induced by IL-1 alpha. However, tachyphylaxis of the leptin response suggests that other mechanisms also are involved.