Amyotrophic lateral sclerosis patient antibodies label Ca2+ channel alpha 1 subunit

Sporadic amyotrophic lateral sclerosis is an idiopathic human degenerative disease of spinal cord and brain motor neurons. Prior studies demonstrated that most patients with amyotrophic lateral sclerosis possess immunoglobulins that bind to purified L-type voltage-gated calcium channels, that titers of anti-voltage-gated calcium channel antibodies correlate with disease progression rates, and that amyotrophic lateral sclerosis patient-derived antibodies (ALS IgG) produce electrophysiological changes in the function of voltage-gated calcium channels. Using Western transfer immunoblots and enzyme-linked immunosorbent assays, the calcium ionophore-forming alpha 1 subunit of the voltage-gated calcium channel is now identified as the major voltage-gated calcium channel antigen to which ALS IgG binds. Additionally, the binding of an L-type voltage-gated calcium channel alpha 1 subunit-directed monoclonal antibody, which itself mimics the effects of ALS IgG on skeletal muscle voltage-gated calcium channel currents, is selectively prevented by preaddition of ALS IgG. Voltage-gated calcium channel-binding IgG from patients with Lambert-Eaton myasthenic syndrome appears to be differentiated from ALS IgG by the reactivity of the former to both alpha 1 and beta subunits of the calcium channel. These assays provide further evidence linking amyotrophic lateral sclerosis to an autoimmune process, and suggest one means to differentiate immunoglobulins from patients with amyotrophic lateral sclerosis from those of patients with another autoimmune disease expressing calcium channel antibodies.

Stable co-expression of calcium channel alpha 1, beta and alpha 2/delta subunits in a somatic cell line

1. The high-voltage-activated L-type calcium channel is a multi-protein complex of alpha 1, alpha 2/delta, beta and gamma subunits. The alpha 1 subunit contains the voltage-dependent calcium-conducting pore. Chinese hamster ovary (CHO) cells were stably transfected with the complementary DNA of the alpha 1, beta and alpha 2/delta subunits. These subunits were not detected in wild-type CHO cells. 2. The alpha 1 (CaCh2b) subunit itself directed the expression of functional calcium channels which bound calcium channel blockers and showed voltage-dependent activation and inactivation. 3. The co-expression of the alpha 1 subunit with the beta subunit (CaB1 gene) enhanced the density of the dihydropyridine binding sites 2- to 3-fold and increased dihydropyridine-sensitive barium inward currents (IBa) up to 3.5-fold from -13.3 microA/cm2 (alpha 1 subunit) to -46.7 microA/cm2 (alpha 1 and beta subunits). 4. Co-expression of the beta subunit did not change the sensitivity of IBa towards dihydropyridines, but accelerated current activation and inactivation and shifted the half-maximal steady-state activation and inactivation to slightly more hyperpolarizing potentials. 5. The co-expression of the alpha 2/delta subunit together with alpha 1 and beta subunits accelerated the inactivation kinetics of the channel without a major effect on the other parameters. 6. These results indicate that the beta and alpha 2/delta subunit interact with the alpha 1 subunit and modulate thereby the properties of the alpha 1 subunit-dependent inward current.

Serum antibodies to L-type calcium channels in patients with amyotrophic lateral sclerosis

BACKGROUND AND METHODS

Sporadic amyotrophic lateral sclerosis is a chronic, progressive degenerative disease of the motor neurons of the spinal cord and motor cortex. The cause is unknown. Recent electrophysiologic studies in animals indicate that immunoglobulins from patients with this disease alter presynaptic voltage-dependent calcium currents and calcium-dependent release of neurotransmitters. To determine whether similar interactions might be identified biochemically, we used an enzyme-linked immunosorbent assay (ELISA) to detect the reaction of serum IgG with purified complexes of L-type voltage-gated calcium channels from rabbit skeletal muscle. The results from patients with amyotrophic lateral sclerosis were compared with those obtained from patients with other types of motor neuron disease, patients with autoimmune and non-autoimmune neurologic diseases, and normal subjects.

RESULTS

Serum samples from 36 of 48 patients with sporadic amyotrophic lateral sclerosis (75 percent) contained IgG that reacted with L-type calcium-channel protein, and serum reactivity on ELISA correlated with the rate of disease progression (Spearman rank-correlation coefficient, 0.62). Reactive serum was present in only 1 of 25 normal subjects and 1 of 35 control patients with no motor neuron disease. Antibodies to L-type voltage-gated calcium channels were identified in 6 of 9 patients with Lambert-Eaton syndrome, and in 3 of 15 patients with Guillain-Barré syndrome.

CONCLUSIONS

Antibodies to L-type voltage-gated calcium channels are present in the serum of patients with amyotrophic lateral sclerosis, and antibody titers correlate with the rate of disease progression. Together with previous data, these results suggest a role for autoimmune mechanisms in the pathogenesis of sporadic amyotrophic lateral sclerosis.

Target size analysis of prostaglandin endoperoxide synthase. Radiation inactivation of both cyclooxygenase and peroxidase correlated with the monomer of 72 kDa

To determine the size of the functional catalytic unit of prostaglandin endoperoxide (prostaglandin H) synthase, radiation inactivation experiments were performed. Both microsomes from ovine seminal vesicles and purified enzyme were irradiated with 10 MeV electrons. The enzymic activities of prostaglandin H synthase, cyclooxygenase and peroxidase, showed mono-exponential inactivation curves dependent on radiation dose, indicating molecular masses of approximately 72 kDa. The enzyme in microsomes, in its native environment, as well as in its purified state after solubilisation with nonionic detergent showed identical molecular masses. The results clearly demonstrate that the monomer of the enzyme with an apparent molecular mass of 72 kDa (SDS/PAGE) is the functional unit for catalysis of both activities. Hence the two active sites of cyclooxygenase and peroxidase reside on the same polypeptide chain.

Identification of the site of interaction of the dihydropyridine channel blockers nitrendipine and azidopine with the calcium-channel alpha 1 subunit

The dihydropyridine binding site of the rabbit skeletal muscle calcium channel alpha 1 subunit was identified using tritiated azidopine and nitrendipine as ligands. The purified receptor complex was incubated either with azidopine or nitrenidpine at an alpha 1 subunit to ligand ratio of 1:1. The samples were then irradiated by a 200 W UV lamp. The ligands were only incorporated into the alpha 1 subunit, which was isolated by size exclusion chromatography and digested either by trypsin (azidopine) or endoproteinase Asp-N (nitrendipine). Each digest contained two radioactive peptides, which were isolated and sequenced. The azidopine peptides were identical with amino acids 13-18 (minor peak) and 1428-1437 (major peak) of the primary sequence of the skeletal muscle alpha 1 subunit. The nitrendipine peptides were identical with amino acids 1390-1399 (major peak) and 1410-1420 (minor peak). The sequence from amino acids 1390 to 1437 is identical in the alpha 1 subunits of skeletal, cardiac and smooth muscle and follows directly repeat IVS6. These results indicate that dihydropyridines bind to an area that is located at the putative cytosolic domain of the calcium channel.

The dihydropyridine-sensitive calcium channel of the skeletal muscle: biochemistry and structure

The dihydropyridine-sensitive calcium channel of the rabbit skeletal muscle is the first voltage-gated calcium channel which has been purified and biochemically characterized. The alpha 1-subunit, a 165 kDa protein, of the purified dihydropyridine receptor contains all regulatory sites of a L-type calcium channel and the calcium conducting unit. The purpose of this review is to summarize and discuss recent findings on the structure and possible function of the skeletal muscle calcium channel subunits.

Evidence for coordinate, selective regulation of eicosanoid synthesis in platelet-derived growth factor-stimulated 3T3 fibroblasts and in HL-60 cells induced to differentiate into macrophages or neutrophils

We used Swiss 3T3 fibroblasts stimulated with platelet-derived growth factor and HL-60 cells induced to differentiate into macrophages or neutrophils to study the regulation of prostaglandin and leukotriene synthesis. Addition of platelet-derived growth factor to quiescent 3T3 fibroblasts led within 4 h to a dramatic and preferential increase in prostacyclin synthesis from endoperoxide prostaglandin H2, and microsomal assays showed a strong platelet-derived growth factor-dependent increase in the maximal velocities (Vmax) of both prostaglandin H synthase and prostacyclin synthase. In contrast, addition of phorbol ester to HL-60 cells to induce differentiation into macrophages led within 4 h to a strong and preferential increase in thromboxane synthesis from prostaglandin H2, and microsomal assays disclosed a major rise in Vmax for both prostaglandin H synthase and thromboxane synthase. No comparable changes occurred in HL-60 cells that were differentiating into neutrophils, though upregulation of 5-lipoxygenase pathway enzymes occurred in both differentiation systems. Actinomycin D and cycloheximide prevented the appearance of all of these enzymes of eicosanoid synthesis in all three model systems. Thus, the distinctive patterns of eicosanoid synthesis that are seen in replicating fibroblasts and in differentiating macrophages and neutrophils appear to depend on a coordinate, selective upregulation of several enzymes of eicosanoid biosynthesis that is specific for each cell system.

Site-specific phosphorylation of the purified receptor for calcium-channel blockers by cAMP- and cGMP-dependent protein kinases, protein kinase C, calmodulin-dependent protein kinase II and casein kinase II

Five protein kinases were used to study the phosphorylation pattern of the purified skeletal muscle receptor for calcium-channel blockers (CaCB). cAMP kinase, cGMP kinase, protein kinase C, calmodulin kinase II and casein kinase II phosphorylated the 165-kDa and the 55-kDa proteins of the purified CaCB receptor. The 130/28-kDa and the 32-kDa protein of the receptor are not phosphorylated by these protein kinases. Among these protein kinases only cAMP kinase phosphorylated the 165-kDa subunit with 2-3-fold higher initial rate than the 55-kDa subunit. Casein kinase II phosphorylated the 165-kDa and the 55-kDa protein of the receptor with comparable rates. cGMP kinase, protein kinase C and calmodulin kinase II phosphorylated preferentially the 55-kDa protein. The 55-kDa protein is phosphorylated 50 times faster by cGMP kinase and protein kinase C than by calmodulin kinase II or casein kinase II and about 10 times faster by these enzymes than by cAMP kinase. Two-dimensional peptide maps of the 165-kDa subunit yielded a total of 11 phosphopeptides. Four or five peptides are phosphorylated specifically by cAMP kinase, cGMP kinase, casein kinase II and protein kinase C, whereas the other peptides are modified by several kinases. The same kinases phosphorylate 11 peptides in the 55-kDa subunit. Again, some of these peptides are modified specifically by each kinase. These results suggest that the 165-kDa and the 55-kDa subunit contain specific phosphorylation sites for cAMP kinase, cGMP kinase, casein kinase II and protein kinase C. Phosphorylation of these sites may be relevant for the in vivo function of the CaCB receptor.

cAMP-dependent protein kinase rapidly phosphorylates serine- 687 of the skeletal muscle receptor for calcium channel blockers

The cAMP-dependent phosphorylation of the 165-kDa subunit of the receptor for organic calcium channel blockers (CaCB-receptors) was studied. Tryptic peptide analysis showed that cAMP-dependent protein kinase phosphorylates rapidly a serine in one peptide. Up to three peptides containing phosphoserine and -threonine are phosphorylated in a 2-h incubation. The isolated 165-kDa subunit was digested with trypsin and the endoproteinase Lys-C and Glu-C. The rapidly phosphorylated peptide was isolated from each digest. The amino acid sequence was determined by Edman degradation and compared with the deduced amino acid sequence of the CaCB-receptor from rabbit skeletal muscle (Tanabe, T., Takeshima, H., Mikami, A., Flockerzi, V., Takahashi, H., Kangawa, K., Kojima, M., Matsuo, H., Hirose, T., and Numa, S. (1987) Nature 238, 313-318). Phosphoserine was determined as the phenylthiohydantoin-derivative of dithiothreitol-dehydroalanine. The phosphorylated serine was identified as Ser-687 which is localized between the transmembrane regions II and III. A second phosphopeptide was isolated into which phosphate was incorporated into Ser-1617 with a slow time course. This peptide is located in the COOH-terminal cytoplasmic domain of the 165-kDa subunit. It is anticipated that phosphorylation of serine 687 affects the opening probability of the calcium channel.

Higher oxidation states of prostaglandin H synthase. EPR study of a transient tyrosyl radical in the enzyme during the peroxidase reaction

Purified prostaglandin H synthase (EC 1.14.99.1), reconstituted with hemin, was reacted with substrates of the cyclooxygenase and peroxidase reaction. The resulting EPR spectra were measured below 90 K. Arachidonic acid, added under anaerobic conditions, did not change the EPR spectrum of the native enzyme due to high-spin ferric heme. Arachidonic acid with O2, as well as prostaglandin G2 or H2O2, decreased the spectrum of the native enzyme and concomitantly a doublet signal at g = 2.005 was formed with maximal intensity of 0.35 spins/enzyme and a half-life of less than 20 s at -12 degrees C. From the conditions for the formation and the effect of inhibitors, this doublet signal was assigned to an enzyme intermediate of the peroxidase reaction, namely a higher oxidation state. The doublet signal with characteristic hyperfine structure was nearly identical to the signal of the tyrosyl radical in ribonucleotide reductase (EC 1.17.4.1). Hence the signal of prostaglandin H synthase was assigned to a tyrosyl radical. Electronic spectra as well as decreased power saturation of the tyrosyl radical signal indicated heme in its ferryl state which coupled to the tyrosyl radical weakly. [FeIVO(protoporphyrin IX)]...Tyr+. was suggested as the structure of this two-electron oxidized state of the enzyme. A hypothetical role for the tyrosyl radical could be the abstraction of a hydrogen at C-13 of arachidonic acid which is assumed to be the initial step of the cyclooxygenase reaction.

Higher oxidation states of prostaglandin H synthase. Rapid electronic spectroscopy detected two spectral intermediates during the peroxidase reaction with prostaglandin G2

The reaction of prostaglandin H synthase with prostaglandin G2, the physiological substrate for the peroxidase reaction, was examined by rapid reaction techniques at 1 degree C. Two spectral intermediates were observed and assigned to higher oxidation states of the enzymes. Intermediate I was formed within 20 ms in a bimolecular reaction between the enzyme and prostaglandin G2 with k1 = 1.4 x 10(7) M-1 s-1. From the resemblance to compound I of horseradish peroxidase, the structure of intermediate I was assigned to [(protoporphyrin IX)+.FeIVO]. Between 10 ms and 170 ms intermediate II was formed from intermediate I in a monomolecular reaction with k2 = 65 s-1. Intermediate II, spectrally very similar to compound II of horseradish peroxidase or complex ES of cytochrome-c peroxidase, was assigned to a two-electron oxidized state [(protoporphyrin IX)FeIVO] Tyr+. which was formed by an intramolecular electron transfer from tyrosine to the porphyrin-pi-cation radical of intermediate I. A reaction scheme for prostaglandin H synthase is proposed where the tyrosyl radical of intermediate II activates the cyclooxygenase reaction.

Phosphorylation of the purified receptor for calcium channel blockers by cAMP kinase and protein kinase C

The dihydropyridine receptor purified from rabbit skeletal muscle contains three proteins of 165, 55 and 32 kDa. cAMP kinase and protein kinase C phosphorylate the 165-kDa and the 55-kDa proteins. At identical concentrations of each protein kinase, cAMP kinase phosphorylates the 165-kDa protein faster than the 55-kDa protein. Protein kinase C phosphorylates preferentially the 55-kDa protein. cAMP kinase incorporates up to 1.6 mol phosphate/mol protein into the 165-kDa protein and 1 mol/mol into the 55-kDa protein upon prolonged incubation. At a physiological concentration of cAMP kinase 1 mol phosphate is incorporated/mol 165-kDa protein within 10 min, suggesting a physiological role of this phosphorylation. Protein kinase C incorporates up to 1 mol phosphate/mol into the 55-kDa protein and less than 1 mol/mol into the 165-kDa protein. Tryptic phosphopeptide analysis reveals that cAMP kinase phosphorylates two distinct peptides in the 165-kDa protein, whereas protein kinase C phosphorylates a single peptide in the 165-kDa protein. cAMP kinase and protein kinase C phosphorylate three and two peptides in the 55-kDa protein, respectively. Mixtures of the tryptic phosphopeptides derived from the 165-kDa and 55-kDa proteins elute according to the composite of the two elution profiles. These results suggest that the 165-kDa protein, which contains the binding sites for each class of calcium channel blockers and the basic calcium-conducting structure, is a specific substrate for cAMP kinase. The 55-kDa protein apparently contains sites preferentially phosphorylated by protein kinase C.

The 165-kDa peptide of the purified skeletal muscle dihydropyridine receptor contains the known regulatory sites of the calcium channel

The dihydropyridine receptor purified from rabbit skeletal muscle yields in the presence of dithiothreitol and sodium dodecyl sulfate on polyacrylamide gels bands of apparent molecular mass 165 +/- 5, 130 +/- 5, 55 +/- 3, 32 +/- 2 and 28 +/- 1 kDa (chi +/- SEM, n = 12). Under nonreducing conditions, the 130 kDa and 28-kDa peptides migrate as a single peptide of 165 kDa. These peptides were separated on a HPLC size-exclusion column. The specific absorption coefficients of the isolated peptides were determined. From these a stoichiometry of 1:1.7 +/- 0.2:1.4 +/- 0.3 (chi +/- SEM of 12 experiments with three different preparations) was calculated for the 165-kDa, 55-kDa and 32-kDa peptides. The relative amount of the 130/28-kDa peptide varied with different preparations. Tryptic, chymotryptic and V-8 protease peptides of the isolated proteins suggested that the 130/28-kDa peptide was not related to the 165-kDa peptide. The dihydropyridine photoaffinity analog (+/-)-azidopine was specifically incorporated only into the 165-kDa peptide with an efficiency of about 2.4%. The azido analog of desmethoxyverapamil, LU 49888, was specifically incorporated into the same peptide with an efficiency of 1.5%. These results suggest that only the 165-kDa peptide contains the regulatory sites detected so far in the voltage-operated L-type calcium channel. They suggest further that the 130/28-kDa peptide, which migrates as a 165-kDa peptide under nonreducing conditions, does not contain high-affinity binding sites for the calcium channel blockers.

EPR study of ferric native prostaglandin H synthase and its ferrous NO derivative

Purified prostaglandin H synthase (EC 1.14.99.1) apoprotein, a polypeptide of 72 kDA, was titrated with hemin and EPR spectra of high-spin ferric heme were observed at liquid-helium temperature. With up to one hemin per polypeptide, a signal at g = 6.6 and 5.4, rhombicity 7.5%, evolved owing to specifically bound, catalytic active heme. At higher heme/polypeptide ratios signals at g = 6.3 and 5.9 were observed which were assigned to non-specific heme with no catalytic function. In microsomes from ram seminal vesicles the native enzyme showed the signal at g = 6.7 and 5.2 which could not be increased by the addition of hemin. Cyanide, an inhibitor of the enzyme, reacted at lower concentrations with the specific heme abolishing its signal at g = 6.6 and 5.4. Higher concentrations of cyanide were needed for the disappearance of the signal of non-specific heme. The reduced enzyme reacted with NO and formed two types of NO complexes. A transient complex, with a rhombic signal at gx = 2.07, gz = 2.01 and gy = 1.97, was assigned to a six-coordinate complex. The final, stable complex showed an axial signal at g = 2.12 and g = 2.001 and was assigned to a five-coordinate complex, where the protein ligand was no longer bound to the heme iron. Neither type of signal showed a hyperfine splitting from nitrogen of histidine indicating the absence of a histidine-iron bond in the enzyme. From these results and the similarity of the EPR signal at g = 6.6 and 5.4 to the signal of native catalase (EC 1.11.1.6) we speculated that tyrosinate might be the endogenous ligand of the heme in prostaglandin H synthase.

Binding of reactive metabolites of CCl4 to specific microsomal proteins

The covalent binding of [14C]carbon tetrachloride to microsomal proteins in rat liver microsomes under anaerobic conditions was investigated by SDS-polyacrylamide slab gel electrophoresis and fluorography. Most of the labeled proteins were observed in the molecular weight range of 52-61 kDa, indicating that cytochrome P-450 forms (EC 1.14.14.1) were labeled. Protein bands at the position of the NADPH-cytochrome P-450 reductase (78 kDa) (EC 1.6.2.4) and NADH-cytochrome b5 reductase (33 kDa) (EC 1.6.2.2) also showed radioactivity. The fluorographic pattern of the protein labeling was cytochrome P-450-dependent, as was demonstrated by CO and metyrapone inhibition as well as by pretreatment of rats with inducing drugs such as 3-methylcholanthrene, benzo(a)pyrene, phenobarbitone and Aroclor 1254. Immuno-precipitation with a purified anti-P-450 immunoglobulin against cytochrome P-450 PB-B (52 kDa) of rat liver indicated that this protein contained about 10-20% of the total bound radioactivity in an average ratio of 0.8 mol [14C]CCl4-metabolite/mol cytochrome P-450 PB-B.

Spectral intermediates of prostaglandin hydroperoxidase

Microsomes from ram seminal vesicles or purified prostaglandin H synthase supplemented with either arachidonic acid or prostaglandin G2 formed an unstable spectral intermediate with maxima at 430 nm, 525 nm and 555 nm and minima at 410 nm, 490 nm and 630 nm. At -15 degrees C the band at 430 nm disappeared within 4 min whereas the trough at 410 nm increased three fold. At higher temperatures (10-37 degrees C) spectral complex formation and decay were observed in less than 1 s. An apparent KS-value of about 3 microM was determined for the titration of purified prostaglandin synthase with prostaglandin G2 at -20 degrees C. Substrates for cooxidation reactions of prostaglandin synthase such as phenol, hydroquinone and reduced glutathione as well as the peroxidase inhibitors cyanide and azide inhibited the prostaglandin G2-induced spectral complex formation. The oxene donor iodosobenzene and hydrogen peroxide formed a spectral intermediate analogous to the complex observed with prostaglandin G2 or arachidonic acid in ram seminal vesicle microsomes as well as with the purified prostaglandin synthase. These results are interpreted as the formation of a ferryl-oxo complex (FeO)3+ of the heme of prostaglandin synthase with prostaglandin G2 analogous to the formation of compound I of horseradish peroxidase.

EPR titration of ovine prostaglandin H synthase with hemin

To characterize further the prosthetic group of PGH synthase (EC 1.14.99.1), titrations of the apoenzyme with hemin were investigated by EPR. The first hemin bound per polypeptide showed an EPR signal at g = 6.7 and 5.3 (rhombicity 9%) and was tentatively assigned to the hemin effective as prosthetic group of PGH synthase. Additional hemin bound showed a less rhombic signal (g = 6.3 and 5.8, rhombicity 3%) presumably due to nonspecific hydrophobic binding sites not effective in catalysis.

Singlet oxygen formation detected by low-level chemiluminescence during enzymatic reduction of prostaglandin G2 to H2

Addition of arachidonic acid to a suspension of ram vesicular gland microsomes or purified prostaglandin synthase, causes a rapid burst of light emission in the range 600-750 nm, as detected by single-photon counting. Maximal light emission intensity is obtained within 15-30 s after the addition of arachidonic acid and is followed by a rapid decay to the background level. The intensity of chemiluminescence is dependent on the amount of ram vesicular gland microsomes or isolated prostaglandin synthase and arachidonic acid concentration (Km about 6 microM). Spectral analysis of arachidonic acid-induced photoemission of isolated prostaglandin synthase in the range 600-750 nm showed two distinctive peaks at about 634 and 703 nm. The similar relative intensities of these peaks, along with the lower intensity at about 668 nm is indicative of singlet oxygen dimol emission. Chemiluminescence with arachidonate is enhanced by 1,4-diazabicyclo[2,2,2]octane and inhibited by azide, indomethacin, acetylsalicylic acid and beta-carotene. Cooxygenation substrates such as phenol, hydroquinone and reduced glutathione, inhibited the arachidonic acid-induced chemiluminescence. Dioxygen is a requirement for the observation of singlet oxygen dimol emission with arachidonic acid as a substrate for ram vesicular gland microsomes or purified prostaglandin synthase. However, when prostaglandin G2 is substituted for arachidonic acid, light emission is not dependent on oxygen. Thus, singlet oxygen can be formed in the dismutation reaction, 2 PGG2 leads to 2 PGH2 + 1O2, catalysed by prostaglandin hydroperoxidase.

[Prevention by metyrapone of halothane induced liver necrosis in rats (author's transl)]

Liver necrosis and an increase of serum sorbitol dehydrogenase activity can be produced in rats by halothane anesthesia (1% v/v in oxygen) following pretreatment with polychlorinated biphenyls. Using this model, it was shown that administration of metyrapone (2-methyl-1,2-bis(3-pyridyl)-1-propanone) (100 mg/kg b.w.) 1 h prior to anesthesia prevents liver necrosis and the concomitant increase in serum sorbitol dehydrogenase activity. No significant differences were observed when halothane was applied in air instead of oxygen, but the protective effect of metyrapone was abolished.

The mechanism of reductive dehalogenation of halothane by liver cytochrome P450

The reductive dehalogenation of halothane leading to 2-chloro-1,1,1-trifluoroethane (CTE) and 2-chloro-1,1-difluoroethylene (CDE) has been investigated in vitro using at liver microsomes under anaerobic conditions. The stimulation of NADPH oxidation by halothane as well as the formation of the products were dependent upon cytochrome P450 as indicated by their CO and metyrapone inhibition. After replacement of NADPH by sodium dithionite as a reducing agent CDE was the only product of the enzymatic reaction. The product pattern was influenced by pretreatment with 3-methylcholanthrene, benzo(a)pyrene, phenobarbitone and Arochlor 1254 and by addition of anti-cytochrome P450-PB immunoglobulin. The CTE:CDE ratio was shifted by addition or inhibition of cytochrome b5 and by pH variation indicating a crucial role of the second electron donation to cytochrome P450 in determining the product pattern. The intermediate complex of cytochrome P450 with a Soret band at 470 nm formed with halothane in reduced liver microsomes was shown to decompose spontaneously to give CDE. Therefore we propose the 470 nm peak to represent a cytochrome P450 Fe3+----CHCl-CF3 carbanion complex. From these results a reaction pathway could be derived which includes radical and carbanion intermediates as reactive precursors of CTE and CDE, respectively.

The reductive metabolism of halogenated alkanes by liver microsomal cytochrome P450

Under anaerobic conditions various polyhalogenated alkanes (CCl3-CCl3, HCl2C-CCl3, CF3-CCl3, CCl4, CF-CHCIBr) stimulate the oxidation of NADPH by liver microsomal fractions. The participation of cytochrome P450 in the NADPH oxidation was shown by inducers and inhibitors of the monooxygenase system. The products of the reductive pathway of hexachloroethane were tetrachloroethene (99.5%) and pentachloroethane (0.5%). From pentachloroethane as substrate trichloroethene (96%) and tetrachloroethane (4%) were produced. The stoichiometry of NADPH oxidation and product formation was close to 1:1. There was a synergistic effect in the presence of NADPH and NADH for both hexa- and pentachloroethane. The influence of dioxygen and radical traps (RSH) on the formation of products from hexachloroethane with reduced cytochrome P450 has been investigated. The results indicate the possibility of a reductive in vivo metabolism of polyhalogenated alkanes even at physiological dioxygen concentrations. For the reductive dehalogenation of polyhalogenated alkanes by microsomal cytochrome P450 a reaction scheme is proposed: the reduction proceeds by two subsequent one electron reductions forming first a radical and then a carbanion. The carbanion can form an alkene via beta-elimination of chloride.