The mechanism of heme uptake and release by cell membranes and proteins was investigated using a model system of carbon monoxide heme, human globin, and phosphatidyl choline single bilayer vesicles. The combination of heme with globin involves the rapid (t(, 1/2) << 1 ms) formation of an intermediate complex described by the equation: K(,d) k(,3) heme + globin heme-globin hemoglobin From stopped flow measurements, observed rates were obtained and absolute spectra were constructed for intermediate reaction mixtures. At pH 7.2, 10(DEGREES)C, the K(,d) is 22 (mu)M and k(,3) equals 700 s('-1). The effects of pH, glycerol, organic phosphate, and substituted heme were examined. The results suggested that heme is rapidly adsorbed onto hydrophobic areas of the protein to form a complex exhibiting spectral characteristics of heme dissolved in liposomes. The proper orientation of heme induces a rate limiting conformational change in globin, followed by rapid iron-histidine binding. The reaction of heme with phospholipid vesicles was analyzed as a simple binding phenomenon by calculating equilibrium fractions of bound and unbound heme. The K(,d) is 0.7 (mu)M with 7.8 lipid molecules/binding site. The observed binding rate was defined as k(,obs) = k(,1)(lipid phosphate) + k(,2), where the association rate, k(,1), equals 1.36 (mu)M P('-1)s('-1) and the dissociation rate, k(,2), equals 2-5 s('-1). Multilamellar vesicles and liposomes containing stearylamine exhibited predictable deviations. Fluorescence measurements using pyrene labelled phospholipid indicate that rates of heme transfer between and release from liposomes are similar. The combination of globin with membrane bound heme is much slower (t(, 1/2) (DBLTURN) 200 ms) than with free heme (t(, 1/2) (DBLTURN) 1 ms) due to partitioning between the various phases. A theoretical expression for k(,obs) was derived and evaluated using constants obtained from independent experiments. Our results demonstrate the rapid exchange of heme between liposomes and extraction by apoproteins which occurs by a rate limiting first order step of heme dissociation into the aqueous phase. A carrier protein is not necessary for heme transport but may be required for compartmentalization.

Lactose repressor from Escherichia coli regulates the expression of the lac operon. The protein binds specifically to the operator region and nonspecifically to the rest of genome. The repression is released upon addition of inducer. Modification of repressor with tetranitromethane results in loss of both operator and nonspecific DNA binding activity with maintenance of most of the inducer binding capacity. The tyrosine residues modified are primarily tyrosines 7 and 17. These two residues may be involved in or close to the operator binding site. Modification of core protein also results in loss of operator DNA binding, though to a lesser extent. The charge on the nitro-tyrosine at neutral pH apparently prohibits operator DNA binding. The presence of nonspecific DNA protects tyrosine residues from modification and prevents loss of operator DNA binding activity. Reduction of nitro-tyrosine to amino-tyrosine by sodium dithionite restores the neutral charge; and the amino group does not influence nonspecific DNA binding, but interference with operator DNA binding is observed. Steric restraints at tyrosines 7 and 17 are apparently more stringent for specific vs nonspecific DNA binding. Modification of the protein with dansyl chloride results in loss of operator DNA binding at low molar ratio of reagent/monomer. Loss of nonspecific DNA binding can be observed at higher molar excess, while IPTG binding is not affected. Lysine residues are the only modified amino acids detectable. Four DNS-lysine peptides are observed by high pressure liquid chromatography: one residue modified is in the amino terminus; the other three are in the core region. Energy transfer occurs between a DNS incorporated in the core region and tryptophan 201 in the repressor. Cross-linking studies have been used to explore the contact sites between the repressor and the operator DNA fragment. Cross-linking is observed between p-azidophenacyl bromide-modified repressor and operator DNA fragment. However, analysis of the specific bases involved was not possible. No cross-linking can be observed using p-azidophenylglyoxal modified repressor and operator DNA. Using depurinated operator DNA fragments, effects observed appear due largely to nonspecific DNA binding.

A soluble, stable, L-glutamine:aminocyclitol aminotransferase has been purified 100-fold from mature mycelia of the gentamicin producer Micromonospora purpurea ATCC 15835. An absorbance maximum at 415 nm indicated the presence of pyridoxal phosphate in the enzyme; its involvement in the reaction, and hence its location at the active site, was indicated by abolition of the absorbance maximum upon addition of amino donors, and its restoration upon addition of amino acceptors. In addition, inhibition of the aminotransferase activity by carbonyl reagents was partially reversed by pyridoxal phosphate but not pyridoxal, confirming the importance of pyridoxal phosphate for reactivity. The keto-acid reaction product of transaminations with L-glutamine was identified as 2-ketoglutaramate, which exists in equilibrium with the inactive lactam, 2-hydroxy-5-oxoproline. Substrate specificity studies indicated that this aminotransferase is distinct from the commonly encountered aminotransferases since it did not readily catalyze transaminations with L-glutamate, L-aspartate, L-alanine, 2-ketoglutarate, oxaloacetate, or pyruvate. Active as amino donors were N('3)-methyl-2-deoxystreptamine, 2-deoxystreptamine, streptamine, scyllo-inosamine, and L-glutamine. Active as amino acceptors were scyllo-inosose, D,L-epi-inosose, and 2-ketoglutaramate. This aminotransferase preparation was able to catalyze transaminations which are compatible with both aminotransferase steps in the 2-deoxystreptamine biosynthetic pathway. Appearance of the enzyme during growth of M. purpurea, plus the substrate specificity, indicated that this aminotransferase is an idiophase enzyme, specific for 2-deoxystreptamine biosynthesis. A 2-deoxystreptamine-less idiotroph, M. purpurea ATCC 31119, was examined and found to be an excellent alternative source for the purification of this aminotransferase. The L-glutamine:aminocyclitol aminotransferase from crude extracts of idiophase cells of the neomycin producer, Streptomyces fradiae ATCC 10745, exhibited a substrate specificity very similar to that of the aminotransferase from M. purpurea. In addition, a 2-ketoglutaramate-hydrolyzing amidase activity was identified in both M. purpurea and S. fradiae. A functional relationship may exist between the L-glutamine:aminocyclitol aminotransferases and the amidases, as has been speculated for the L-glutamine:keto-acid aminotransferase and omega-amidase from rat tissues.

This research has investigated the initiation of transcription in E. coli. The main focus has been the assessment of the effects that specific promoter sequences have on RNA polymerase-promoter interactions, both in vivo and in vitro. Previous studies have shown the importance of two partially conserved hexanucleotide sequences near -10 and -35 of the promoter. However, the specific roles of these sequences, constraints on the locations of these sequences, and contributions from other promoter regions have not been determined. The studies described here have addressed several of these questions. These include: characterization in vitro parameters of the (beta)-lactamase promoter, studies utilizing a variant of the trp promoter, construction of hybrid promoters with different -10 and -35 regions, construction of promoters with altered spacing between the -10 and -35 regions, and construction of promoter variants where adjacent DNA sequences have been altered. The hybrid promoters suggest the spacing between the -35 and -10 regions must be 16 to 18-bp with 17-bp the optimum spacing. The -35 and -10 regions can be used interchangeably and promoter strength increases as the sequence homology of each region approaches the consensus sequence. These data suggest the back half of the trp promoter is several fold stronger than either lac or tet, while the front half of trp was the weakest of the three. DNA sequences outside of these two regions must also play a role in promoter function. It is apparent that sequences beyond -39 of the trp promoter do not play an important role in expression, while changes in the transcribed, nontranslated region can alter gene expression. In vitro characterizations have suggested a general relationship between the rate of open complex formation and transition temperature among the promoters studied here. It was further shown that these parameters alone are not accurate predictors of in vivo promoter expression. Several of the promoters described here, trp-lac, trp, and trp-tet, will be very useful in studies requiring transcription or gene expression.

Dopamine beta hydroxylase (DBH) is the terminal enzyme in the biosynthetic pathway of norepinephrine. The enzyme is a monooxygenase and functions to insert a hydroxyl group on the beta carbon of dopamine to yield norepinephrine. Physical and chemical methods were employed to examine structural features of the enzyme, and the contributions of various amino acids to the catalytic activity were investigated with chemical modifying agents. No free sulfhydryl residues were detected upon titration of DBH with 2-chloromercuri-4-nitrophenol. Reaction with iodoacetamide caused a time dependent loss in activity. The activity change was correlated to the loss of approximately three histidine residues. Methionine reaction was also observed. Reaction of DBH with diethylpyrocarbonate or diazonium tetrazole resulted in the inactivation of the enzyme within fifteen minutes. Modification with either reagent was slightly pH dependent over a range of pH 4.5 to 8.0. Analysis of the diethylpyrocarbonate reacted DBH indicated a loss of one to two histidines per DBH monomer. No tyrosine or lysine modification was observed with this reagent. Inactivation of DBH with diazonium tetrazole resulted in the extensive modification of tyrosine and lysine residues. The activity loss observed with diazonium tetrazole appeared due to general denaturation of the enzyme. Incubation of DBH with trinitrobenzene sulfonic acid, methyl acetimidate or dansyl chloride was carried out to investigate the involvement of lysine residues in enzyme catalysis. Dansyl chloride caused a time dependent loss of activity, completely inactivating the enzyme at a 500 fold excess of reagent over DBH monomers. The other reagents had little effect on DBH activity. DBH was found to bind the fluorescent dyes anilinonaphthalene sulfonate (ANS) and fluorescein. The binding of ANS had no effect on the catalytic competence of DBH, while the binding of fluorescein caused the inactivation of the enzyme. The reaction of DBH with low excesses of fluorescamine resulted in the production of fluorescently labeled active enzyme. No significant difference spectra were observed upon incubation of labeled enzyme with substrates or effector compounds.

Periplasmic binding proteins from gram-negative bacteria are essential components in the osmotic shock-sensitive transport systems for carbohydrates, amino acids, and ions. The structure of the leucine, isoleucine, valine-binding protein (LIV-BP), an integral part of the high affinity branched-chain, aliphatic amino acid transport system in Escherichia coli, has been determined by X-ray crystallography. Purified LIV-BP (M(,r) = 36,770) that was devoid of residually-bound amino acid formed large, polyhedral crystals from 12% polyethylene glycol at pH 4.5. The structure was solved by the multiple isomorphous replacement method at 3.0 (ANGSTROM) resolution using five heavy atom derivatives (mean figure of merit = 0.77 for 6653 reflections). A polyalanine model was fit to the electron density map with the modeling program FRODO implemented on an Evans & Sutherland PS 300 interactive graphics system. The model of the polypeptide chain backbone reveals two distinct, globular domains connected by three strands. Between the domains is a wide cleft which is accessible to solvent. Though lacking sequence similarity, each of the two domains has essentially the same secondary structure arrangement: a central, five-strand parallel (beta)-sheet flanked by two major (alpha)-helices on either side running anti-parallel to the sheet. The secondary structure elements are connected in the classic (beta)(alpha)(beta) folding pattern. The amino acid binding site was located by soaking crystals in solutions of 10 or 50 mM L-leucine, L-isoleucine, L-valine, or L-threonine. Difference Fourier maps in all cases displayed a significant peak located in the cleft between the two domains, but confined only to the wall of the N-terminal domain near two loops that connect (beta)-sheets with helices. Despite little sequence homology, the supersecondary structure of the LIV-BP domains is very similar to the corresponding domains of the bilobate L-arabinose-binding protein (ABP). Seventy percent of the (alpha)-carbons from the two proteins were determined to be equivalent. The major difference between the two structures is the orientation of the two domains and the size of the cleft: narrow in ABP and wide-open in LIV-BP. Since the ABP crystal structure contains bound substrate and LIV-BP doesn't, a substrate-induced cleft closure is proposed. The LIV-BP structure with an open cleft may be typical of all unliganded binding proteins.

A comparison of model heme compounds with isolated and with mitochondrial cytochrome b using MCD and EPR showed that the unusual coordination structure of the cytochrome b heme(s) can be represented by hindered bis-imidazole protoheme. The midpoint potentials for the b and c(,1) cytochrome were measured using MCD and EPR. A value of 270 mV was obtained for cytochrome c(,1) while the midpoint potentials found for the two species of cytochrome b varied with temperatures, viz. 62 and -20 mV at 23(DEGREES)C (MCD) compared to 116 and -4 mV at about 10 K (EPR). The midpoint potential of the iron-sulfur center obtained by low-temperature EPR was 286 mV. The potentials of the two half reactions of ubiquinone were measured by EPR at 110 and 296(DEGREES)K. Potentials of 176 and 51 mV were found at 110(DEGREES)K, while values of 200 and 110 mV were observed at 296(DEGREES)K. The potentials of cytochrome b, as that of c, were also independent of pH when titrations were performed in deoxycholate buffers, while a variation of -30 mV per pH unit was observed for both b species in taurocholate buffers. These two detergents also produced different MCD-contributions of the two b hemes. The oxidative reaction mechanism was investigated through the reoxidation of 2-electron, 4-electron (CoO-depleted) and fully-reduced Complex III by ferricyanide. The nonlinear first-order plots of the oxidation of c(,1), were obtained for the reaction of 2-electron reduced sample at all tested ferricyanide concentrations. These data were explained by an extremely rapid intramolecular electron equilibration(> 2000 s('-1)) between the iron-sulfur center and cytochrome c(,1). Neither antimycin-A nor CoO had a substantial effect on the reoxidation kinetics of c(,1). The oxidation of both b and c(,1) cytochromes of 4-electron reduced and fully-reduced samples monitored by the absorbance changes at 561.5 and 553.5 nm exhibited multiphasic kinetic data. The absorbance change at 553.5 nm was preceded by that at 561.5 nm during the first reaction period, indicating a fast depletion of electrons from cytochrome b through c(,1) (or iron-sulfur center) to ferricyanide. Satisfactory simulation for the kinetic data collected at both wavelengths was achieved by a linear scheme:(UNFORMATTED TABLE FOLLOWS) slow fast fast k(,ox) b(,H) b(,L) {2Fe/2S} c(,1) oxidant(TABLE ENDS)

The structure of the sulfate-binding protein, a component for the active transport of sulfate in Salmonella typhimurium, has been solved at 3.0 (ANGSTROM) resolution using multiple isomorphous replacement phases derived from 5 heavy atom derivatives. The mean figure of merit is 0.60 for 5626 reflections. Sulfate-binding protein is an elongated molecule with dimensions 35 x 35 x 60 (ANGSTROM). The polypeptide chain folds into two domains, which are designated the N domain and the C domain; the N domain contains the amino terminus, while the other domain contains the carboxyl terminus. The two domains are connected by three segments of polypeptide chain. Though these three segments are in close proximity in the tertiary structure, they are widely separated in the amino acid sequence. The connecting segments serve as the base of a deep cleft formed between the two domains. The N domain consists of a core of 5 strand (beta) sheet with two (alpha) helices on either side. The C domain is also composed of a core of 5 strand (beta) sheet with two (alpha) helices on one side, but 3 helices occur on the other side. The secondary structure is comprised of 42% (alpha) helices and 32% (beta) strands. Since the (beta) strands and (alpha) helices alternate along the chain except for two helices in the C domain, sulfate-binding protein is an (alpha)/(beta) protein. The overall structure of the sulfate-binding protein is similar to those previously determined for L-arabinose-, D-galactose- and leucine/isoleucine/valine-binding proteins. All have a two domain structure with each domain consisting of a core (beta) sheet flanked on either side by at least two helices. The sulfate binding site is in a cleft between the two domains. The amino-termini of three (alpha) helices, two from the C domain and a third from the N domain, interact with the sulfate anion through helix dipole moments. These interactions contribute to the stability of the anion-protein complex. The binding sites of the other binding proteins whose structures have been solved are also located in the cleft between the two domains. The location of the sulfate binding site is consistent with the proposal that binding proteins may undergo a hinge-bending so that one domain twists or rotates relative to the other upon ligand binding.

Several analogues of creatine have been synthesized, characterized, and fed to animals, and the accumulation and utilization by tissues of their high energy phosphorylated derivatives have been studied. Chicks fed 1-carboxyethy1-2-iminoimidazolidine(homocyclocreatine) accumulated the extremely stable synthetic phosphagen, 1-carboxyethyl-2-imino-3-phosphonoimidazolidine (homocyclocreatine-P) in breast muscle (32 (mu)mol/g wet wt), heart (7 (mu)mol/g), and brain (2.4 (mu)mol/g). Homocyclocreatine-P reacted with creatine kinase (EC 2.7.3.2) to regenerate ATP up to 200,000-fold more slowly than creatine-P, and its Gibbs standard free energy of hydrolysis was approximately 2 kcal/mol lower than that of creatine-P. Muscle, heart, and brain of chicks and mice fed N-ethylguanidinoacetate accumulated N-ethylguanidinoacetate-P, which was found to be the most reactive known analogue of creatine-P, with a similar Gibbs standard free energy of hydrolysis. Mouse Ehrlich ascites tumor cells incubated in vitro accumulated up to 8 (mu)mol/g packed cells of N-ethylguanid-inoacetate-P, which was shown to be utilized for the regeneration of ATP more rapidly and completely than another coexisting synthetic phosphagen, 1-carboxymethyl-2-imino-3-phosphonoimidazolidine(cyclocreatine-P). Dietary N-ethylguanidinoacetate reduced the level of arginine:glycine amidinotransferase (EC 2.1.4.1), the first enzyme of creatine biosynthesis, in liver of chicks. On the basis of V(,max)/K(,m), N-ethylgunidinoacetate was more active than creatine as substrate for bacterial creatinine amidohydrolase (EC 3.5.2.10), and was cyclized to a greater extent than creatine. Hearts of rats fed cyclocreatine accumulated 12 (mu)mol/g of cyclocreatine-P, higher levels of glycogen, and during subsequent ischemia demonstrated a nearly 2-fold delay in development of rigor-contracture, prolonged glycolysis, and a marked delay in depletion of ATP levels. During ischemia cyclocreatine-P continued to be utilized for ATP regeneration long after creatine-P stores had been exhausted.

Aspartase has been purified from E. coli B cells using high performance liquid ion exchange chromatography. This procedure allows rapid purification of the enzyme with >80% recovery of total activity. The kinetic mechanism of aspartase, determined from initial velocity experiments, is best described by the equilibrium ordered addition of a divalent metal cation to free enzyme followed by binding of aspartate. Following interconversion of the central complex, NH(,4)('+) and fumarate are randomly released. Michaelis constants for each substrate are: 85 mM for NH(,4)('+), 0.17 mM for fumarate, 0.6-1.0 mM for aspartate and 9.0 (mu)M for Mg('2+). From the results of inhibition studies using structural analogues of aspartate, it was determined that aspartase recognizes the (beta)-carboxylate group of aspartate. The (alpha)-amino group contributes very little to the binding of aspartate to the enzyme. Compounds having an -OH group in addition to the (beta)-carboxylate had no effect on the rate of deamination of aspartate, suggesting a restricted geometry at the active site. S-2,3-Dicarboxy aziridine was found to be a potent competitive inhibitor of aspartase (K(,i) = 0.1 (mu)M) and fumarase (K(,i) = 0.08 (mu)M). The aziridine did not inactivate either enzyme nor did it exhibit any observable substrate activity. It is likely that it functions as a transition state analogue mimicking the carbanion intermediate found in the normal catalytic reaction. The aziridine inhibited fumarate utilization in ruptured but not intact mitochondria.