Hydrogen metabolism in aerobic hydrogen-oxidizing bacteria.
Bernhard SCHINK and Hans-Giinter SCH~LEGEL. Institut [i~r Mikrobiologie der Gesellscha[t [iir Strahlen- und Umwelt[orschung mbH Miinchen, und der Universitiit G6ttingen, 3400 G6ttingen, Grisebachstr. 8, Federal Republic o[ Gerlnany.
R~sum~. Summary. On rencontre deux types d'hydroq6nases A survey on orqanisms able to use mole- dans les bact6ries a6robies capables d'oxyder cular hydroqen as electron donor in the enerqy- l'hydroq~ne mol6culaire : les unes sont solubles yieldinq process is presented. In the cyroup of et r6duisent le NAD (H~: NAD oxydor~duc- the aerobic hydroqen-oxidizing bacteria so far rases), les autres sont membranaires et ne two types of hydroqenases have been en- r6duisent pas les pyridine-nucl6otides. Les bac- countered, a NAD-reducinq, soluble enzyme t6ries qui oxydeni l'hydroq~ne peuvent ~tre (H2 : NAD oxidoreductase) and a membrane- divis6es en trois qroupes suivant qu'elles con- bound enzyme unable to reduce pyridine tiennent (i) les deux types d'enzymes (A/ca- nucleotides. With respect to the distribution of liqenes eutrophus), (ii) settlement une enzyme both types of hydroqenases three qroups of soluble (Nocardia opaca lb), (iii) seulement hydroqen-oxidizinq bacteria can be diffen- une enzyme membranaire (la majorit~ des tiated containinq (i) both types (Alcaliqenes qenres et des esp~ces). Nous r6sumons les eutrophus), (ii) a soluble enzyme only (Nocar- 6tudes faites sur l'enzyme & NAD de A. eutro- dia opaca lb), and (iii) a membrane-bound phus. Les r6sultats concernant la solubilisation hydroqenase only (majority of genera and et la purification de l'hydroq6nase membra- species). The results of studies on the NAD- naire de A. eutrophus sont pr6sent6s en d6tail. specific hydroqenase of A. eutrophus are L'enzyme a 6t6 solubilis6e & partir de mem- summarized. Results on the solubilization branes purifi6es & raide de Triton X-100 et de and purification of the membrane-bound hydro- d6soxycholate ou de phespholipase D. L'extrait qenase of A. eutrophus are presented in detail. membranaire brut a 6t6 fractionn6 par pr6ci- The enzyme was solubilized from purified pitation au sulfate d'ammonium et chromato- membranes by Triton X-100 and sodium desoxy- qraphie sur carboxym6thylcellulose & pH 5,5. cholate or phospholipase D. The crude mem- L'enzyme est stable en tampon phosphate ; brane extract was fractionated by ammonium sa stabilit6 en milieu oxydant est comparable sulfate precipitation and chromatography on & celle de l'enzyme soluble. Des donn6es bio- carboxymethylcellulose at pH 5.5. The enzyme chimiques et immunoloqiques indiquent cepen- was stable in potassium phosphate buffer ; dant que les deux enzymes poss~dent des struc- it resembles the soluble enzyme with respect tures diff6rentes. to stability under oxidizinq conditions. Further biochemical and immunoloqical data indicate, however, that both enzymes are different with respect to their native structure.
Several metabolic types of organisms are able use hydrogen for aerobic respiration and the to use molecular hydrogen (dihydrogen or H 2) as reduction of carbon dioxide. The second group, an electron donor in the energy-yielding process. which is able to utilize hydrogen under anaerobic Two groups have to be considered. The first group conditions only, comprises various metabolic comprises bacteria utilizing hydrogen under aero- types, chemotrophic as well as phototrophic bac- bic conditions. These are the cheumlithoauto- teria. The chemotrophic hydrogen-utilizing bac- trophic, aerobic hydrogen-oxidizing bacteria teria are the methanogenic bacteria (e.g., Metha- (e.g. Alcaligenes eutrophus); they are able to nosarcina barkeri), the acetogenic bacteria (e.g., 21 298
Ci'ostridium aceticum and Acetobaclerium woodii), energy and on the level, at which the electrons the sulfate-reducing bacteria (e.g., Desul[ovibrio enter the respiratory chain; HJCO 2 ratios of 4 vulgaris) , the nitrate-reducing, denitrifying bac- to 19 have been measured. The average ratio mea- teria (e.g. Paracoccus denitri[icans) and the fuma- sured with Alcaligenes eatrophas is represented rate-reducing bacteria (e.g., Escherichia colt). in equation 3. Among the phototrophic bacteria there are at least 2 H 2 + 02 )'2 H20 (1) some species able to grow with hydrogen as an 2 ~ + C09 ~-
299
as Alcaligenes latus, Aquaspirillum aulolrophicum, (Entner-Doudoroff) enzymes is repressed [11]. Paracoccus den itri[icans, several pseudomonads In contrast, in Paracoccus denitri[icans glucose and the coryneform nitrogen-fixing strains, con- is the dominant substrate, and hydrogenase for- tain only a membrane-bound hydrogenase [8]. mation is repressed E12~. The latter type of regu-
TABLE I. Hydrogen-oxidizing bacteria and the localization o[ hydrogenases.
Hydrogenase Species membrane- nitrogen Gram soluble bound fixation stain
Alcaligenes eutrophus dt- Alcaligenes paradoxus 71_ __ w Alcaligenes ruhlandii .~- Alealigenes latus .ql- -- --
Pseudomonas [acilis -AI- -- -- Pseudomonas saccharophila d[_ -q- -- -- Pseudomonas carboxydooorans q.------Pseudomonas pseudoflava -q.- -- -- Pseudomonas palleronii -q- -- -- Pseudomonas hydrogenooora "JI- -- -- Pseudomonas hgdrogenothermophila
Flavobacterium autothermophilum -31- -- -- Hydrogenomonas thermophilus Aquaspirillum autotrophieum -q- -- __ Paracoccus denitrificans 71- -- __ << Cor!lnebacterium >> autotrophicum + + - Nocardia opaca lb -~- __ -- -q- Nocardia aulotrophica AV __ __ .qt_ Mgcobacterium 9ordonae + -- + Arthrobacter sp. (llX, Pd-I 12) m + -- + Seliberia carboxgdohgdrogena
The regulation of hydrogenase formation in lation seems to predominate among the hydrogen these bacteria is not well known [9]. All aerobic bacteria. In addition, hydrogen can exert an inhi- hydrogen bacteria are facultatively autotrophic bitory effect on the utilization of the organic sub- bacteria and are able to grow under autotrophic strate. In Alcaligenes eutrophus H16 hydrogen conditions with CO 2 + H 2 as well as under hetero- inhibits the utilization of fructose; this effect trophic conditions on a number of organic com- turned out to be due to the inhibition of glucose-6- pounds as substrates, such as sugars, organic phosphate dehydrogenase [13] ; in A. eutrophus acids, and amino acids. Maximum hydrogenase this enzyme is sensitive to ATP and NAD'H2. From activity is found in autotrophically grown ceils. following the pool sizes of glucose-6-phosphate, In heterotrophically grown cells hydrogenase is 6-phosphogluconate, ATP and NA,DH2 during either absent or present in minor specific acti- transition of the cells from air to a hydrogen- vities ; in Alcaligenes eutrophus the specific total oxygen mixture it was concluded that glucose-6- hydrogenase activity depends on the kind of phosphate dehydrogenase is the regulatory en- organic substrate (table Ii) and the growth rate zyme and that in vivo the inhibition is caused [101. When the cells are exposed to both energy by the increase in the NADH 2 concentration [141. donors, hydrogen and an organic substrate, the response is different in different strains ; a kind The presence of two hydrogenases has been of catabolite repression nlay be operative. In described for Pseudomonas saccharophila [6~, Alcaligenes eutrophus hydrogen is the dominant Alcaligenes ruhlandii [15; 5] and Alcaligenes substrate ; in the presence of hydrogen and fruc- eutrophus [4]. The soluble enzyme has been tose the formation of the fructose degrading recently purified and characterized. The results 300 have been published [16; 81 and will only be of a flavin component. It was identified as being summarized here. FMN by thin layer chromatography, fluorescence spectrum and fluorescence quenching by the addi- The soluble hlldrogenase o[ Alcaligenes eutrophus. tion of apoflavodoxin. Eleven strains of Alcaligenes eutrophus were The molecular weight is 2,0.5 t)00.. The isoelectric used to examine the stability of the soluble hydro- point was found to be 4.85. The pH optinmm was genase, and strain HI6 was selected as the enzyme 8.0, the optimal temperature 33°C, and the Mi-
TABLE II. Rate o[ hydrogen oxidation by intact cells o[ Alcaligenes eutrophus H 16 after growth on glutamate, lactate or fructose compared to autotrophi- cally grown cells.
Rate of hydrogen oxidation (l~mol H~/min.mg protein) after growth on Multiplication factor of cell protein Glutamate Lactate Fructose during growth on organic substrate 85per cent N~ 85 per centN~ + 5 par cent O~ + 5 per cent O~ Air + 10 per cent C0~ + 5 per cent CO~
Autotrophic culture 1.36 1.45 1.26 ca. 10 ~ 0.454 0.056 0.651 ca. 104 -- -- 0.484 ca. 10 6 0.160 0.056 i 10 8 -- 0 0.424
Rates of total gas uptake were measured in an atmosphere of 85 per cent H~2 + 5 per cent O2 + 10 per cent COs (10). The rate of hydrogen oxidation was calculated assuming ratio. H~ :O~. :CO~= 6:2:1.
source. The enzyme was purified 68-fold with a chaelis constants for NAD and H 2 were '0.56 and yield of 20 percent and a specific activity (NAD- 0:037 mM, respectively. reduction) of 54 !~mol H 2 oxidized per min and mg protein. The enzyme is not only insensitive to oxygen, its stability can even be increased by The membrane-bound hydrogenase oxygen treatment or the addition of oxidizing of A. eutrophus. agents like ferricyanide (,0.5 raM). The enzyme, The existence of two enzymes for the activation which was homqgeneous by the criterium of poly- of hydrogen in A. eatrophus, a soluble and a acrylamide gel electrophoresis, is not strictly membrane-bound one, raised the question, whether NAD-specific. In its reactive state it has the ability these two hydrogenases were two independent to react in the absence of I'¢AI)(H) with a large isozymes or one enzyme protein exhibiting dif- number of arti,ficial and physiological hydrogen ferent properties due to its localization in diffe- acceptors such as methylene blue, ferricyanide, rent sites in the cell. Preliminary experiments dichlorophenol indophenol, phenazinemethosul- indicated almost coordinate formation of both fate, viologen dyes, FMtN, FAD, ubiquinone, cyto- enzymes [17]. However, these observations con- chrome c and oxygen. In some cases the reduction tradicted the differences in their biochemical rates were even up to 5-fold higher than with behaviour. For further differentiation of both NAD. The enzyme catalyzed the evolution of hydrogenase activities, the inembrane-bound hydrogen from NADH, dithionite-reduced me- hydrogenase had to be purified and compared to thylviologen and benzylviologen. It exhibited dia- the soluble enzyme. Since the results have not yet phorase and b~AD (P)H oxidase activity. been published, the solubilization, purification, In accordance with the low acceptor specificity and some properties of the membrane-bound and with diaphorase activity was the presence hydrogenase will be described in some detail.
301
TABLE III. S'olubilization of the membrane-botmd hydrogenase of A. eutroI)hus by di[ferent agents (usually in KP buffer 50' mM, pH 7..0).
Incubation Incubation Per cent Per cent Agent time temperature protein activity (hours) / "C) solnbilizeO solubilized
EDTA 10 mM 2 4 3.0 2.0 KC1 3 M 2 4 12.0 0.4 KC1 3 M -~- EDTA 10 mM 2 4 12.3 2.2 KCi 3 M -1(- MgC1._, 10 mM 2 4 99 0.3 Glyeine 2 M 22 4 13.0 1.1 KCI 3 M in K acetate buffer pH 4.0 2 20 1 6 -- KCI 3 M in Na citrate buffer pH 5.0 2 20 4.8 4.0 KC1 3 M in Tris-HCl buffer pH 9.0 2 2o 13.0 5.0 n-pentanol j 0.5 4 14.5 3 5 iso-pentanol ( 50 0 5 4 15.8 3.7 per cent tert pentanol ~ o 5 4 19.0 35 n-butanol ~, 0.5 4 36.5 -- NaSCN ! 0 5 4 22 6 -- NaCI04 i 0.5 M in Tris-HCl buffer o 5 4 18.4 -- KNO:, i pH 8.0 -~ 0.25 M sucrose 0.5 4 20.0 3.5 LiBr 0.5 4 14.8 1.3 Guanidine HCI ; o 5 4 16.3 4.8 Phospholipase D (3 U/g wet weight) in K acetate 0 5 30 8.2 14.1 buffer pH 5.5 Brij 35 2 per cent 2 20 11 6 0.2 Lubrol WX 2 per cent 2 20 12.5 1.3 Digitonin 0.1 per cent 2 20 12.0 3.0 Na deoxycholate 0.05 per cent 2 20 9 0 4.8 Na deoxycholate 0.1 per cent 2 2o 13.5 11.0 Na deoxyeholate 0.2 per cent 2 20 22.0 9.0 Triton X-100 0.2 per cent 2 20 20.0 9.7 Triton X-100 0.5 per cent 2 20 30.0 11.8 Triton X-100 1.0 per cent 2 20 36.0 9.2 Triton X-100 2.0 per cent 2 20 300 4 2 Triton X-100 0.5 p. 100 -1- Na deoxychol. 0.1 per 0.5 20 23 8 22.0 cent -~- sucrose 10 per cent -J- EDTA 10 mM
When the investigation on the membrane-bound determined manometrically under hydrogen atmo- hydrogenase was started only a few data were sphere with methylene blue as electron acceptor. known; these concerned the hydrogen oxidase system in intact membranes. The activity ol For rapid deternfination of hydrogenase acti- hydrogen oxidase was stable in frozen membrane vity in membranes and in solubilized preparations preparations under air, and it was inhibited by the following test system was used. A solution of oxygen at concentrations over 70 ~ [18, 19]. 200' ~M methylene blue in '50, mM potassium phos- Quinones and cytochromes of the b, c, a, a3, and phate buffer, pH 7.0, was kept under hydrogen gas o type were involved in the electron transfer and pipetted into 3 nil cuvettes. These were system from hydrogen to oxygen [203. In silu the flushed with hydrogen before and after adding hydrogenase reacted only with artificial electron the buffer. Glucose (0.2 !mml), glucose oxidase (1 acceptors like methylene blue, phenazinemetho- unit) and catalase (1 unit) were added as an oxy- sulfate, and dichlorophenolindophenol ; it did not gen trap. The reaction was started by adding the react with pyridine nucleotides. The physiological membrane suspension or solution containing the electron acceptors were unknown. Hydrogenase hydrogenase, and the absorption decrease was activity of membrane preparations was usually followed photometrically at 570 nm wave-length.
302
The results were in accordance with those deri- The following procedure for solubilizing hydro- ved from manometric tests. genase in a preparative scale has been finally used (after Yu and Wolin, [21~ ; modified) : membranes The cytoplasmic membranes were prepared by prepared by sonication were washed with potas- sonication, treatment by lysozyme followed by sium phosphate buffer and again with this buffer osmotic shoctk, and in the French pressure cell. In supplemented by 0..1;5 M NaC1 and '0.26 M sucrose these preparations the hydrogenase activity was for removing proteins loosely bound to the mem- stable when frozen under air. Due to their rela- brane surface. The membranes were resuspended tively high specific hydrogenase activities, soni- in phosphate buffer to a protein content of 4 to cared membranes were used for solubilization. 5 mg/ml. 10, per cent sucrose, 110 mM EDTA, 0.1 per The membranes were washed with phosphate buf- cent deoxycholate, and ,0.5 per cent Triton X-IO0 fer and resuspended in the same buffer to a pro- (final concentrations) were added, and after stir- tein content of about 3-4 mg/ml. After adding the ring for 30, rain at room temperature, membrane solubilizing agent, the membrane suspension was particles not solubilized were removed by centri- stirred for ,0.5 to 2;2 h. The insoluble debris was fugation. A maximal yield of 22 per cent of the removed by centrifugation at 1,0~)~00~)g. Hydroge- hydrogenase activity of intact membranes was nase activity and protein content were determined solubilized. As explained later, this corresponds in the supernatant. The results of various attempts to ahnost the total amount of hydrogenase present to solubilize the enzyme are summarized in in the membranes. The apparent loss of activity table III. turned out to be due to a change in the pH opti- mum of the hydrogenase after solubilization. The pH optimum change made it possible to follow the solubilization process by measuring the 100 decrease of enzyme activity in the membrane sus-
>_
< ~5 ~> 50 E E 0.4 LIJ rr %--0 O,3 0 I I ~ ~ L ~, I 0 10 20 30 40 50 150 0.2 Time (min) <
FIG. 1. -- Kinetics of the solubilization of the mem- brane-bound hydrogenase of A. eutrophus H 16 by the combination of Triton X-IO0 and Na-deoxgcholate.
Protein content : 2.4 mg/ml. The decrease of hydro- i I i i t i genase activity is the result of the shift of the pH optimum during solubilization. 30 6O Time (days) Fro. 2. -- Decrease of activity of the solubilized membrane-bound hydrogenase during storage at --18°C under air.
Only low hydrogenase activities were released by the incubation of the membranes in the pre- sence of potassium chloride, magnesium chloride pension after the addition of detergents (fig. 1). or E~)TA, in acidic or aIkaline buffers. Alcohols Already after 29 min more than 9,0 per cent of as well as chaotropic agents decreased the hydro- the hydrogenase was solubilized. The velocity of genase activity. Satisfactory results were obtained this process was not influenced by the incubation with Triton X-I'&0, and sodium deoxycholate as temperature between 4 and 3'0°C. well as with phospholipase I). These results indi- cated the hydrogenase enzyme being tightly bound The yellow supernatant tolerated storage under to the membrane. The enzyme protein is apparen- air at --18°C for two months almost without loss tly in close contact to the lipid membrane layer of enzyme activity (~i'g. 2). During storage at 4°C and has to be regarded as an integral, not a peri- under air for two days the loss was 10 per cent, pheral membrane protein. whereas under reducing conditions (addition of
303 mercaptoethanol or hydrogen) the loss was 70 per phate buffer. After dialysis and centrifugation, cent. The membrane-bound hydrogenase turned the supernatant was chromatographed on CM~cel- out to be stable under oxidizing and unstable lulose (Whatman CM 52) equilibrated with 25 mM under reducing conditions ; in this respect it potassium acetate buffer, pH '5.5, in a 1.5 × 30 cm resembles the soluble hydrogen dehydrogenase. column using a 2{)0 ml linear gradient from 0 to 0.5 M KCl. Fractions with a volume of 2 ml were The yellow supernatant, which contained pro- collected. Profiles of hydrogenase and protein teins, lipids, detergents, sucrose, and EDTA, was content are presented in figure 3. fractionated by ammonium sulfate precipitation. In the first step ammonium sulfate was added to A high amount of protein with low hydrogenase give a 25 per cent saturated solution. After centri- activity passed the column. The hydrogenase was fugation the emulsion was separated into an oily, eluted in two adjacent peaks. The fractions of red coloured supernatant layer and a clear both activity peaks were concentrated separately aqueous solution containing the hydrogenase. The by ultrafiltration to a protein content of 1.0 mg/ml supernatant layer could be fixed to the surface by and dialyzed against potassium phosphate buffer, adding light petroleum or hexane before centrifu- pH 7.0. c 21
o ..t- 05 Og '": f J J f 0.4 / 7 / / J /
02 /t~~ m ,¢5
~ / / ol : .../ /~ c LU t4t - ~- -o-4-.6.~. ... 0 ~¢~, n ""'~".'~-'o"'~'er 'e' n I I I -o--Q4 ,O-,I~,D- 0 5o 100 Number of fraction Fro. 3. -- Elution profile of the membrane-bound hgdrogenase during chromatography on CM-cellalose. 9 mg of protein were applied to the column equilibrated with potassium acetate buffer 25. mM pH 5.5 and eluted by a linear gradient of 0 to 0.5 M KC1. Fractions of 2 ml were collected. Symbols : @--@ protein content, measured as extinction at 280 nm ; /X--A hydrogenase activity, measured as the rate of methylene blue reduction.
gation ; the aqueous solution was easily sucked The total purification procedure is summarized off with a pipette. This procedure was repeated in table IV. For comprehension of the low values after increasing the ammonium sulfate concentra- of activity recovered after enzyme solubilization tions to 3.0 per cent saturation. The hydrogenase it has to be mentioned that activity was consisten- remained in the aqueous solution. Hydrogenase tly measured at pH 7.0, which is about the opti- was precipitated by ammonium sulfate between mum pH for the hydrogenase bound to the mem- 4,0 and 5'5 per cent saturation. The precipitate was brane. The solubilized enzyme has a pH optimum dissolved in phosphate buffer and again precipi- at pH 5:5. At pH 7.'0,, the solubilized enzyme has tated by ammonium sulfate between 4~-6() per cent only 24 per cent of the activity measured at saturation. The precipitate was dissolved in phos- pH 5.5. For making out a balance-sheet of the 304
TABLE IV. Solubilization and purification of the membrane-bound hydrogenase of A. eutrophus.
Specific Total Total activity Purili- Yield Step protein activity (units/g cation (mg) (units) protein) {--fold) (per cent)
Washed membranes 1 890 1 870 990 1.0 100 Washing by NaCl/sucrose 1 620 1 810 1 120 1.13 96 Solubilization 360 430 1 200 1.21 23 (1 770) (4 950) (5.0) (95) 1. Precipitation with ammonium 37 286 7 700 7.8 15.3 sulfate (40-55 per cent) (1 180 (31 700) (32.0) (63 0) 2. Precipitation with ammonium 9 240 26 700 27.0 12.8 sulfate (45-60 per cent) (990) (110 000) (111.0) (55.0) CM-cellulose chromatography 4.2 165 41 300 41.7 8.8 (685) (170 000) (172) (35.8)
whole purification procedure, the activities have Concluding remarks. to be multiplied by the factor 4.1 (values in The majority of species of the hydrogen bac- brackets). Comparing the values obtained after teria have a membrane-bound hydrogenase only the first washing of the membranes with those [8]. The possession of a single enzyme is shared of the two purified enzyme fractions obtained with Azotobacter vinelandii [22], Rhizobium bac- after CM-cellulose chromatography, one calculates teroids [1] and many facultatively and strictly a yield of 35.8 per cent, an enrichment factor of anaerobic bacteria as well as phototrophic bac- 172 and a specitff~c activity of 170 i~mol H 2 oxidized teria [23, 24]. The study of these enzymes lends per rain and mg protein. itself as an approach to reveal relationships Both fractions of purified enzyme separated by among the hydrogen-oxidizing bacteria as ~vell as carboxymethyl cellulose chromatography were between this group, the phototrophic and the non- homogeneous by means of polyacrylamide gel autotrophic hydrogenase containing bacteria. electrophoresis at different polyacrylamide con- The progress in the studies on the hydrogenases centrations, only their R~-values differed slightly. of Alcaligenes eutrophus so far revealed similari- However, these ~ere the only differences between ties and differences. Both the NAD-reducing, the two fractions we found. A mixture of both soluble and the membrane-bound hydrogenase turned out to be homogeneous in isoelectric focu- are remarkably stable to oxygen, they are inacti- sing, in gel filtration on Sephadex G 210,0., and in vated when stored under reducing conditions. sucrose gradient centrifugation. Sodium dodecyl- However, the enzymes are different in several sulfate gel electrophoresis revealed no differences basic properties ; these differences concern the in subunit composition. The ¢ greater >> or (( less electron acceptor specificities, the isoelectric polar )> molecule, however, was not stable, but points, the molecular weights and the pH and gradually changed to the other one during storage temperature optima. Immunodiffusion experi- for several days. In fresh crude solubilized prepa- ments done with antisera prepared against both rations almost onIy the (~ greater >> molecule was purified enzymes did not show any cross reactions found as demonstrated by activity staining the between both enzyme proteins. The native hydrogenase in polyacrylamide gel slabs by phe- enzymes are different proteins. nazinemethosulfate and nitroblue tetrazolium, chloride under hydrogen (method according to 16). On the basis of these data it may be assumed that both types of hydrogenuse molecules differ REFERENCES. from each other by the amount of adherent phos- 1. Dixon, R. O. D. (1972) Arch. Mikrobiol., 85, 193-201. 2. Schlegel, H. G. (1975) in <( Marine Ecology >> (Kinne, pholipid or detergent attached to unpolar regions O. ed.), vol. II, part I, pp. 9-60, John Wiley ,~ of the enzyme. Sons, London. 305
3. Mortenson, L. E. ~ Chen, J.-S. (1974) in (( Microbial 15. Vishniac, W. ~ Trudinger, Ph. A. (1962) Bacteriol. Iron Metabolism >> (Neilands, J. B. ed.), pp. 231- Rev., 26, 168-169. 282, Academic Press, New York and London. 16. Schneider, K. ~ Schlegel, H. G. (1976) Biochim. Bio- 4. Eberhardt, U. (1966) Arch. Mikrobiol., 53, 288-3(12. phys. Acta, 452, 66-80. 5. Bone, D. H., Bernstein, S. ~ Vishniac, W. (1963) 17. Schink, B. (1977) Ph.D. Thesis, University of GiSt- Biochim. Biophys. Acta, 67, 581-588. tingen. 6. Bone, D. H. (19'60) Biochem. Biophys. Res. Com- mnn, 3, 211-214. 18. Probst, I. (1975) Ph.D. Thesis, University of GSt- tingen. 7. Aggag, M. ~ Schlegel, H. G. (1974) Arch. MicrobioL, 100, 25-39. 19. Probst, I. ~ Sehlegel, H. G. (1976) Biochim. Biophys. 8. Schneider, K. & Schlegel, H. G. (1977) Arch. Micro- Acta, 440, 412-428. biol., 112, 229-238. 20. Pfitzner, J. (1972) Zentralbl. Bakteriol. Parasitenkd. 9. Schlegel, H. G. & Eberhardt, U. (1972) Adv. Microb. Infektionskr. Hyg. Abt. 1, Orig. Reihe A, 220, Physiol., 7, 205-242. 396-401. 10. Eberhardt, U. (1966) Arch. Mikrobiol., 54, 115-124. 21. Yu, L. ~ Wolin, M. J. (1972) J. Bacteriol., 109, 59-68. 11. Gottschalk, G. (1965) Biochem. Z., 341, 249-259. 22. Hyndman, L. A., Bnrris, R. H. • Wilson, P. W. 12. Vogt, M. (1965) Arch. Mikrobiol., 50, 256-281. (1953) J. Bacteriol., 65, 52~-531. 13. Black,kolb, F. a Schlegel, H. G. (1968) Arch. Mikro- 23. Gitlitz, P. H. ~ Krasna, A. I. (1975) Biochemistry, biol., 63, 177-196. 14, 2561-2567. 14. Bowien, B., Cook, A. M. & Schlegel, H. G. (19'74) 24. Adams, M. W. W. ~ Hall, D. O. (1977) Biochem. Arch. Microbiol., 97, 273-281. Biophys. Res. Commun., 77, 730-737.