Synthesis, assembly, and localization of periplasmic cytochrome c.

Abstract The kinetics of synthesis, assembly, and localization of periplasmic cytochrome c-550 was examined in the gram-negative bacterium, Spirillum itersonii. For this purpose, a disc electrophoresis technique was developed to isolate and quantitate radioactive pigment produced from isotopic tracers of amino acids and iron. The synthesis of heme which was not covalently bonded to protein (NCBP-Heme) was also measured for comparison. Pulse amino acid-labeled cytochrome c-550 entered the periplasmic space with no significant lag. Periplasmic radioactivity did not increase or decrease upon chase, or in the presence of chloramphenicol. A cytoplasmic precursor pool could not be demonstrated. It is concluded that the periplasmic hemoprotein is not a precursor to, or a product of, a membrane-bound or soluble cytoplasmic pigment. Indeed, stable localization seems to occur upon completion of the pigment's polypeptide chain, or within a few seconds thereafter. The initial rate of iron incorporation into periplasmic cytochrome c-550 followed an upward-shaped profile, while uptake of tracer into NCBP-Heme was linear. Addition of chloramphenicol completely blocked further iron incorporation into the protein, but stimulated NCBP-Heme accumulation 1.6-fold. These results suggest that assembly of the pigment is tightly coupled with protein synthesis, whereas prosthetic group synthesis is not. The possibility that cytochrome c-550 might be assembled from a pool of iron-tetrapyrrole precursor was tested using the inhibitor of heme biosynthesis, levulinate. When levulinate and radioactive iron were added simultaneously to cells, uptake of isotope into both cytochrome c-550 and NCBP-Heme was inhibited by over 85%. In contrast, when levulinate was added to cells previously labeled with radioactive iron, isotope uptake into cytochrome c-550 continued unaffected for 15 min, and at a decreasing rate for an additional 45 min. This increase was accompanied by a 20% decrease in previously formed NCBP-Heme which was dependent on protein synthesis. These and other results suggest that c-type cytochromes in this microorganism are assembled from an iron-tetrapyrrole precursor pool. The intracellular concentration of the putative intermediate was estimated to be 10 µm. Upon depletion of the iron-tetrapyrrole precursor pool by exposure to levulinate, no evidence was gained for the continued synthesis of cytochrome c-550 protein. These results suggest that the availability of prosthetic group precursor may influence the production of the cytochrome c-550 protein counterpart.


SUMMARY
The kinetics of synthesis, assembly, and localization of periplasmic cytochrome c-550 was examined in the gramnegative bacterium, Spirillum ifersonii. For this purpose, a disc electrophoresis technique was developed to isolate and quantitate radioactive pigment produced from isotopic tracers of amino acids and iron.
The synthesis of heme which was not covalently bonded to protein (NCBP-Heme) was also measured for comparison.
Pulse amino acid-labeled cytochrome c-550 entered the periplasmic space with no significant lag. Periplasmic radioactivity did not increase or decrease upon chase, or in the presence of chloramphenicol.
A cytoplasmic precursor pool could not be demonstrated.
It is concluded that the periplasmic hemoprotein is not a precursor to, or a product of, a membrane-bound or soluble cytoplasmic pigment.
Indeed, stable localization seems to occur upon completion of the pigment's polypeptide chain, or within a few seconds thereafter.
The initial rate of iron incorporation into periplasmic cytochrome c-550 followed an upward-shaped profile, while uptake of tracer into NCBP-Heme was linear. Addition of chloramphenicol completely blocked further iron incorporation into the protein, but stimulated NCBP-Heme accumulation 1.6-fold.
These results suggest that assembly of the pigment is tightly coupled with protein synthesis, whereas prosthetic group synthesis is not.
The possibility that cytochrome c-550 might be assembled from a pool of iron-tetrapyrrole precursor was tested using the inhibitor of heme biosynthesis, levulinate. When levulinate and radioactive iron were added simultaneously to cells, uptake of isotope into both cytochrome c-550 and NCBP-Heme was inhibited by over 85 %. In contrast, when levulinate was added to cells previously labeled with radioactive iron, isotope uptake into cytochrome c-550 continued unaffected for 15 min, and at a decreasing rate for an additional 45 min. This increase was accompanied by a 20% decrease in previously formed NCBP-Heme which was de- These and other results suggest that c-type cytochromes in this microorganism are assembled from an iron-tetrapyrrole precursor pool. The intracellular concentration of the putative intermediate was estimated to be 10 PM.
Upon depletion of the iron-tetrapyrrole precursor pool by exposure to levulinate, no evidence was gained for the continued synthesis of cytochrome c-550 protein.
These results suggest that the availability of prosthetic group precursor may influence the production of the cytochrome c-550 protein counterpart.
Cytochrome c is a hemoprotein which contains a covalently linked prosthetic group.
The porphyrin nucleus is linked to the protein component by two thioether bridges, from reduced vinyl side chains to cysteinyl residues (1,2).
Little is known about the sequence of intermediary steps leading to the assembly of holocytochrome c from its precursors. The nature of the control mechanisms which coordinate the synthesis of the prot.ein moiety with the formation of its prosthetic group are also not clear.
In an effort to define more clearly t.he biosynthesis of cytochrome c, a study of the formation of this pigment in the gramnegative aerobic bacterium, Spirillum itersonii, was performed. In this organism, soluble c-type hemoproteins are localized in the surface layers of the cell, and can be selectively released along with other periplasmic proteins by Tris-EDTA treatment (3). The major pigment, periplasmic cytochrome c-550 (4),' with a molecular weight of 10,800 and an isoelectric point at pH 9.8 (5), was specifically chosen for radioisotope studies. Data are presented which indicate that the assembly and localizat.ion of periplasmic cytochrome c-550 are tightly coupled with protein synthesis; assembly occurs by the ut.ilization of iron-tetrapyrrole molecules from a large precursor pool. It is shown further that prosthetic group production is not dependent upon the formation of the cytochrome protein, but that the availability of prosthetic group may influence the production of the protein. Corp. Materials from the following commercial sources were also used: levulinic acid and 2,6-lutidine (Eastman Organic Chemicals), rifamycin SV (Calbiochem), chlorampherlicol (Sigma Chemical Co.), cyclohexanone (J. T. Baker Chemical Co.), and "Aquasol" (Yew England Nuclear Corp.).
Culture System-Soluble cytochrome c of S. itersonii (ATCC 11331) is maximally produced during the second phase of diauxic growth (6) in nitrate-supplemented medium (4). This condition was reproducibly achieved by first growing cells to stationary phase in iron-limited (0.5 to 2.0 PM iron) glycine-glutamatesuccinate medium (7) containing 20 mM KNOa by using I y0 v/v yeast extract-malate-glutamate culture inocula (6) and incubation at 30" for 24 hours at 175 rpm in Erlenmeyer flasks filled to 80% of their capacity.
This growth condition arrests cells in phase one of diauxic growth and severely limits their cytochrome content (4). After harvesting by centrifugation at 24", cytochrome synthesis was initiated by similar incubation of cells (0.3 mg of protein per ml Of CUhre) in iron-SUffiCient (20 t0 40 pM iron citrate) glycine-glutamate-succinate medium (7) containing 20 mM KNOa.
Cultures prepared in this manner exhibit a reproducible 30.fold increase in hemoprotein content after lo-hours incubation, while t,otal cell protein increases about 2-fold. In this system, t.he rate of periplasmic cytochrome c-550 synthesis is maximal at 6 hours.
Assay of Cytochronze c-550 Synthesis-Labeled cells are treated with Tris-EDTA, selectively releasing the cytochrome. The protein is isolated by acid precipitation and subjected to pH 6.6 disc electrophoresis.
The location of cyt.ochrome c-550 on such gels is visualized directly from carrier purified hemoprotein which has been subjected to co-elect.rophoresis.
After electrophoresis the resulting faint red bands are cut. out for the determination of radioactivity.
Recovery is estimated by adding a k11own amount of differentially labeled cytochrome c-550 before acid precipitation.
Purification of Carrier Cytoclzrome c-550--Purification was accomplished by a modificat,ion of a previously reported procedure (5). Final purification was achieved by linear gradient elutio11 (0.03 to 0.15 RI sodium acetate, pII 5.2) of the material adsorbed to a column of carboxymethylcellulose. Cytochrome c-550 is eluted at about 0.08 M sodium acetat,e. Tl1e protein isolated in this manner was judged to be over 95% pure by the absorption ratio of the reduced a11d oxidized forms at 550 and 280 nm, respectively, and by acrylarnidc gel electrophoresis.
Preparation of ""Fe-labeled Sfandard-For the quantitative det,ermiriation of recovery of j9Fe-or "C-labeled cytochrome c-550 isolated by disc electrophoresis (see below) it was necessary to prepare diffrrent,ially labeled hemoprotein standard. For this purpose, a 40-m] culture (prepared as described above) was incubated for IO hours with jjFeCls (25 PCi per ml of culture, 25 PM final iro11 concentratiol1).
;\fter harvesting at 4", cells were resuspended in 4.0 ml of 50 111~ Tris (pH 8.0)-5 mtr EDTh a11d centrifuged at 20,000 x g for 30 min at 4". I'ortions of the resulting periplasmic extract, were analyzed for labeled cytochrome c-550 by subjecting know11 amounts directly to disc electrophoresis (see below) after mixing with 15% sucrose and 30 pg of purified marker cytochrome c-550. This procedure revealed 2.2 x 10" cpm of cytochrome c-550 per ml of periplasmic extract.
111 some experiments j"Fe-labeled standard was used for estirnating the recovery of "5Fe-or 3H-labeled cytocllronle c-550. This standard was prepared and tested by procedures similar to those out,lined above.
Culture Labeling and Sample Preparation-Radioactive iron dissolved in 0.1 N HCl was normally added t,o cultures 3 to 6 hours after the initiation of cytochrome synthesis by unlabeled iron citrate.
The labeled iron was incorporated at increasingly higher rates the later it was added. This could be due to a decreasing solubility with time of the unlabeled iron. The high rate of iron incorporation which occurs when isotope is added at later times was used as an advantage for rapid kinetic &dies. When it was necessary to know the specific activity of iron, unlabeled iron and isotope were mixed before addition to cells.
Incorporation of "gFe or L-[14C]leucine was terminated by pipetting l.O-ml culture samples into tubes containirlg 3.0 ml of cold 0.5 M Tris (pH 8.5)-0.05 M EDTA-1.0 111M unlabeled L-lew tine. After 30-min incubation at 4", cells were removed by centrifugation at 20,000 X g for 15 min at 4". Supernatant fractions were dispensed into 15.0-ml conical glass centrifuge tubes, and to each tube 50 pg of purified cytochrome c-550 and 1 X lo1 cpm of cytochrome c-550 (j"Fe-labeled standard) were added.
After oxidation of cytochrome by addition of 1 drop of 1.0 mM potassium ferricyanide and vigorous agitatio11, an equal volume of cold 25% (w/v) trichloroacetic acid was added. l'rotein was pelleted after l-hour incubation at, 4" by centrifugation in a swinging bucket rotor for 30 min at 4,000 x g. Resulting supernatants were discarded and pellets were blotted dry of excess acid. To each tube 20 ~1 of 2.5 PIT NaOH were added, and after pellets were dissolved (15 to 30 min at 24") the pH wah adjusted to neutrality by the addition of about 20 ~1 of 1.0 I HCl.
after addition of 20 ~1 of 80% sucrose, sample volume:: were adjusted to 100 ~1, and portions were subjected to pH 6.6 disc electrophoresis (see below). When the incorporation of j;Fe or L-[3H]pheaylalanine 11-a' used to measure cytochrome c-550 synthesis, all of the abox-e operations were the same with the t.wo following exceptions: samples were treated with Tris-EDTA corltaining 1.0 m>\1 1111. labeled L-phenylalanine, and SSFe-labeled standard was used to determine recovery.
Cytochrome c-550 remaining in cells after Tris-EDTA t,reat,ment was also determined.
This measurement was an important, control, especially in cases n-here inhibitors were used and the possibility existed that the pigment might accumulate in the cytoplasm.
Tris-EDTA-treated cells were resuspended, disrupted by sonic oscillation (3), and debris was removed by cellt,rifugation at 105,000 x g for 4 hours at 4". The resulting supernatants were prel)ared for disc electrophoresis as described above.
Isolation of Cytochrome c-560 by pl1 6.6 Disc Electrophoresis-Polyacrylamide lower gels, 8 cm in length and 6 mm in diameter, were composed of the following ingredients:

ATCBP-meme
Synthesis-NCBP-Heme synthesis was measured by 5gFe incorporation into material which was extractable by acid-cyclohexanone (8,9). incubation for 30 min at 4", 3.0 ml of cold cyclohesanone were added, and the resulting two phases were emulsified by vigorous agitation.
After an additional 2 hours at 4" the phases were separated by centrifugation.
Portions of the organic phase were counted using "Aquasol" scintillation fluid to dissolve dr3 residues.
Assay of RNA and Protein Synthesis-RNh and protein synthesis were measured by the incorporation of [3H]uracil and l4Clabeled amino acid into acid-insoluble material. Culture samples were pipetted into cold trichloroacetic acid to give a final coilcentration of 10yO (w/v) acid. When the final amount of protein per sample was less than 0. 1  Discs were placed in vials, dried at 110", and counted after addition of toluene-based scintillat.ion fluid.
Radioactivity 'was det.ermined using a three-channel Beckman liquid scintillation spectrometer.
Protein was determined by the method of Lowryet al. (lo), with crystalline bovine serum albumin used as the standard.
Cytochromes were estimated spectroscopically as previously described (3,7). The purity of carrier cyt.ochrome c-550 was analyzed by disc electrophoresis at pH 4.3 (11) and pH 9.5 (12). Protein was stained with 1 yc Xmido black in 7 c/0 acetic acid. How long does it t.ake to reach the periplasmic space from t,he t,ime it is synthesized on polyribosomcs?

Kinetics
Is periplasmic cytochrome c-550 a 1)recursor to, or a product of, a membrane-associated hemoprotein? In an attempt to answer these questions, a pulse-chase esperimerit, wit,h L-[14C]leucine was performed with cells actively synthesizing both soluble and membrane-bound hemoproteins. At various times periplasmic and remaining cytoplasmic proteins were separated.
The radioactivity present. in periplasmic cytochrome c-550 w-as determined after isolating the protein by disc electrophoresis.
For comparison, the kinetic profiles of L-[W]leucinc incorporation into total periplasmic and remaining cytoplasmic protein were also measured (Fig. 2).
In the pulse stage of incorporation, the rate of appearance of radioactivity into periplasmic cytochromr c-550 followed linear kinetics, without. a significant lag (Fig. 2s).
The The kinetics of n-[l%]leucine incorporation revealed that in less than 15 s material became labeled which migrated with native cytochrome c-550 during electrophoresis ( Fig. 2A). Although this observation suggests that the pigment is rapidly assembled, the possibility exists that a precursor might co-migrate with holoprotein.
A study of cytochrome c-550 synthesis by iron incorporation was therefore performed. The kinetic profile of 55Fe incorporation into periplasmic cytochrome c-550 is shown in Fig. 3A. An exponential type curve was obtained.
In contrast, the same population of cells incorporated n-[%']leucine into cytochrome c-550 protein at a linear rate (Fig. 3B). The upward shape of the iron uptake profile was not a result of a lag in the conversion of a precursor to native cytochrome c-550, since incorporation was prevented completely upon addition of 100 pg per ml of chloramphenicol (Fig. 3A). This antibiotic also completely blocked n-[*4C]leucine incorporation into the hemoprotein (Fig. 3B). These results suggest that cytochrome c-550 is rapidly assembled, but radioactive iron equilibrates relatively slowly with the form of iron used in holoprotein synthesis.
NCBP-Heme Synth&s-NCBP-Heme synthesis was studied to explore further the process of assembly of cytochrome c-550. NCBP-Heme is defined as both free heme and heme associated noncovalently to heme proteins (for example, cytochrome b heme). The prosthetic groups of c-type cytochromes are covalently linked to their protein components, and thus are not included in this classification.
Cells actively producing hemoproteins incorporated 5gFe into NCBP-Heme at a linear rate with no significant lag (Fig. 4). The slight upward shape of the curve was due to a corresponding increase in cell mass. Addition of 100 pg per ml of chloramphenicol at 1; hours or at later times after the addition of isotope resulted in an apparent increase of about 1.8fold in the rate of NCBP-Heme accumulation (Fig. 4). Addition of rifamycin (100 pg per ml) gave similar results.
In contrast, only slight stimulation was observed when chloramphenicol was added 15 min aft,er the addition of tracer (Fig. 4).
The second step in heme biosynthesis is catalyzed by the enzyme &aminolevulinate dehydratase. This enzyme has been purified from S. itersonii and is competitively inhibited in vitro by the substrate analog levulinate (13). When 20 mM levulinate was added to cultures actively producing hemoproteins, at the same time as 5gFe, greater than 85% inhibition of NCBP-Heme synthesis was observed (Fig. 5A). Addition of levulinate at later times resulted in the rapid disappearance of about 20% of the total radioactivity (Fig. 5A). This decline did not occur when chloramphenicol was also present (Fig. 5B). when chloramphenicol was also present (Fig. 5B) . 5A) suggests t,hat a portion of the total SCI<I'-Ilcme might contain ln~ursors used for the covalellt assembly of c-type cytochromcs.
In contrast, when levulinate and S9Fe were added at the same time, the furt.her synthesis of cytochromc c-550 as measured by "9Fe uptake was inhibited by 85% (Fig. 7A).
In the control culture without levulinate, an upward-shaped profile of isotope incorporation into cytochrome c-550 was observed. The extent of the lag is consistent with dilution by a previously formed precursor pool (Fig. TA).
For comparison, XCLZP-Heme synthesis was also measured in the same cultures (Fig. 7B). The kinetics and amount of decline of previously labeled material upon addition of levulinate are consistent with a cytochrome c precursor-product relationship. Inhibition of protein synthesis with chloramphenicol (100 pg by guest on July 8, 2020 http://www.jbc.org/ Downloaded from per ml), in the presence or absence of levulinate (20 m&r), abruptly halted further incorporation of jgFe into cytochromc c-550 (Fig. 7A). This observation confirms the previous expcriment (Fig. 3A) in which complete inhibition of iron uptake was noted during briefer periods of exposure to this antibiotic.
Size of Precursor Heme Pool-The pool size of iron-tetrapyrrole precursor can be estimated from the data of the previous experiment (Fig. 7A). Such cells were making total c-type cytochromes at a rate of 0.18 nmole per hour per ml of culture, as determined by spectroscopic measurements.
When levulinate was added to the previously labeled culture, the resulting increase in cytochrome c-550 radioactivity equaled about 28 min of normal synthesis.
Assuming other c-type cytochromes are assembled from the same pool of precursor, the total pool size is estimated to be 0.084 nmole per ml of culture, or about 25,000 molecules per cell. Given a volume of 4 x lo-l2 ml for the S. itersonii cell, the intracellular concentration of heme precursor would be about 10 PM.
The above method of approximation could result in an underestimate if the precursor prosthetic group pool was not completely utilized during inhibition of heme synthesis; levulinate might cause secondary effects on t.he synthesis of c-type cytochrome polypeptide chains and on their conversion to holocytochromes. On the other hand, the above method could lead to an overestimate since inhibition of heme synthesis by levulinate is never complete; levulinate is a competit,ive inhibitor (13). Considering the above limitations, the precursor pool size was estimated by another technique.
Equal portions of cells, previously grown under conditions of iron limitat'ion, were incubated in media containing varying amounts of iron with known specific activities of jgFe. After 53 hours, levulinate was added, and the decrease in labeled NCRP-Heme was followed with time t,o the plateau values. The amount of the decline occurring in response to levulinate can be considered to represent a minimum estimate for the pool size of the putative int,ermediate.
The average size of the precursor pool determined by this technique was 0.084 nmole of iron per ml of culture (Table II), a value in excellent agreement with the above approximation.
The size of the precursor pool was not significantly influenced by the concentration of iron in the incubation media, although the rate of total NCBP-Heme accumulat.ion varied 1.6.fold (Table  II).

Effect of Inhibition of Heme Synthesis on Incorporation of L-[14C]Leucine into Periplasmic
Cytochrome &JO--To find out what happens to the synthesis of the polgpept'ide chain of periplasmic cytochrome c-550 when heme synthesis is curtailed, the effect of levulinate (20 mM) on the incorporation of QV]leucine into material which migrated with native cytochrome c-550 during electrophoresis was examined. L-1s a control, the effect of this inhibitor on holocytochrome c-550 synthesis was measured in the same population of cells by 5gFe incorporation (Fig. 8). 111 the presence of levulinate, the kinetics and ext.ent of inhibition of r$4C]leucine incorporation were indistinguishable from those measured by 5"Fe uptake (Fig. 8). So evidence for differential inhibition was found. If an apoprotein precursor component of cytochrome c-550 exists, then the interpretat,ion of the above results depends on it,s electrophoretic properties.
Assuming apocytochrome and native llemoprotein undergo co-electrophoresis, theu the above results indicate that only enough polypeptide is made to which prosthetic group can be complexed.
The synthesis of the protein counterpart would be coordinated with the synt.hesis of heme. Five 20.ml cultures were prepared by suspending equal portions of iron-limited cells i n media containing 1.3 X lo6 cpm of S9Fe per ml and various amolmts of unlabeled iron to give the indicated final iron concent,rations.
After 5:.hours incubation, 20 mM sodium levulinate (pII 7.0) was added and duplicate 1.0. ml samples were t,aken immediately and at 15-min intervals for 1 hour. Heme was extracted with acid-cyclohexanone and its amount calculated from the specific radioactivity of iron. might not undergo co-electroyhoresis with the native pigment. During inhibition of prosthetic group synthesis, such an intermediate might accumulate but not be detected.
This latter possibility was tested by several experiments.
The electrophoretic profile of radioactive basic proteins, produced by cells which had their prosthetic group pool depleted, was examined.
Cells were treated for 1 hour with levulinate (20 mM), and then pulse-labeled with L-['4C]lcucine for 10 min. 'I'hc %-labeled periplasmic protein was subjected to disc electrophoresis in the presence of marker ":Fe-labeled cytochromc c-550, and the gel was fractionated for the determination of radioactivity.
No new Y-labeled peaks were observed compared to control cells not treated with levulinate.
In addit,ion, t,hroughout the native cytochrorne c-550 band, the ratio of 'YI to jjFe was constant.
No evidence was gained for the accumulation of a precursor during prosthetic group starvation.
A further experiment was performed in an effort to rescue a hypothetical precursor of cytochrome c-550 which might be produced in the absence of prosthetic group synthesis. Cells were exposed to levulinate for 13 hours, levulinatc was removed, and cytochrome c-550 synthesis was assayed by the incorporation of either S9Fe or L-[%]leucine, in the presence or absence of 100 pg per ml of chloramphenicol. lifter 2 hours of incubation, the illhibition of cytochrome c-550 synthesis by chloramphenicol was 99.1 y0 as judged by r-["Clleucine incorporation, and also 99.1 T/o as determined by SgFe uptake.
No evidence for the rescue of a precursor was found. DlSCUSSIOS 'l'hc possibility that pcriplasmic cytochrome c-550 might be a precursor to, or a product of, a protein firnlly bound to t,he cytoplasmic membrane was eliminated by a,11 amino acid pulse-chase experinient.
A particulate form of cytochrome c-550 might be produced, but its implantation into the cytoplasmic membrane would have to occur rapidly, and without exchange with its periplasmic counterpart.
Examination These observations suggest that if a cytoplasmic precursor pool of the protein exists, its half-life can only be a few seconds. Direct analysis of cytoplasmic fractions for cytochrome c-550 also support t,his conclusion.
The kinet.ics of localization of total nascent periplasmic protein was also rapid.
This observation suggests that the majority of t,he proteins residing in the surface layers of the cell becomes localized in a manner not depending on large cytoplasmic precursor pools. Possibly, membrane-bound ribosomes are involved in the synthesis of such proteins.
Ext,ernalization could be facilitated by the process of peptide chain elongation. Work by ot,her investigators has implicated a membrane localization for ribosomes involved in the synthesis of extracellular prot,eins (14, 15).
Several lines of evidence suggest that a large pool of iron-tetrapyrrolc precursor is an intermediate in the assembly of cytochrome c-550, as well as in the formation of other c-type cytochromes in S. itersonii.
(a) The profile of iron incorporation into cytochrome c-550 followed an upward-shaped curve, although no significant lag was observed for the labeling of NCBP-Heme.
(b) Inhibit'ion of heme synthesis by levulinate allowed synthesis of cytochrome c-550 to proceed for about 30 min from previously formed precursor, but not from newly labeled intermediate.
(c) The apparent rate of NCBP-Heme synthesis was st.imulated upon addition of inhibitors of RNA and protein synthesis.
(d) Addition of levulinate to previously labeled cells resulted in a decrease in NCBP-Heme which was dependent on protein synthesis.
(e) The extent of the decrease agreed well with the amount of increase in cytochrome c-heme, assuming all c-type cytochromes in X. itersonii were assembled in a manner similar to cytochrome c-550.
The chemical nature of the postulated iron-tetrapyrrole precursor has not been elucidated.
Since relatively large amounts of this material are present in NCBP-Heme extracts, elaboration of the structure of the intermediate should be an approachable problem.
Earlier it was suggested that cytochrome c is assembled by first forming a covalent intermediate with protoporphyrinogen, followed by the insertion of iron (16,17). This proposal is not consistent with the observations of the present investigation, nor with those previously reported (18).

cultures.
After an additional 0.8.hour incubation, 20 rnM sodium levulinate, pH 7.0 (LEV) was added to two cultures, and each culture was sampled (1.0 ml) at, the indicated times into Tris-EDTA to prepare periplasmic extract. Isotope incorporation into cytochrome c-%0 was determined by pH 6.6 disc electrophoresis. O--O, control cells; &--A, cells incubated with levlllinate.
Formation of the covalent linkages between the protein component of cytochrome c-550 and its prosthetic group occurs very rapidly.
Since L-[*4C]leucine was incorporated within 15 s into material that underwent co-electrophoresis with native cytochrome c-550, the possibility was considered that a precursor component might have migrated with the completed molecule. However, when protein synthesis was curtailed, radioactive iron incorporation into the protein was also abruptly halted. It can thus be concluded that assembly of holoprotein is tightly coupled with protein synthesis.
A corlsideration of the stage of assembly of cytochrome c-550 must include the time it takes to polymerize the protein component's polypeptide chain.
Previous studies indicat,e that the molecule contains about 94 amino acid residues (4). Assuming the rate of peptide chain elongation in S. itersonii at 30" is similar to those reported in other bacteria at 37" (19,20), the polypeptide chain would take about 6 s to be polymerized.
An indirect estimate of the polymerization time of cytochrome c-550 by another method agrees well with this approximation (4). The sensitivity of the experiment demonstrating curtailment of 55Fe incorporation into cytochrome c-550 upon addition of chloramphenicol was limited to the detection of an apoprotein pool with about a 1 min turnover time.
Thus no conclusions can be drawn as to when the prosthetic group becomes attached to the polypept,ide.
Assembly could occur during peptide chain elongation, during chain termination, or after release of completed apoprotein molecules from polyribosomes.
Nevertheless, the results of the present study are not inconsistent with the possibility that assembly occurs at the translational level. The synthesis and assembly of cytochromes requires the delicate cooperation of two complex biosynthetic processes. A balance in the rates of production of precursor prosthetic groups and polypeptide chains must be maintained by yet unknown regula.tory devices. In an attempt to explore such relationships, the effect of inhibition of synthesis of one component upon the production of its counterpart was examined. Heme synthesis in S. itersonii does not appear to be tightly coupled with the production of cytochrome protein counterparts. Inhibition of either RNA or protein synthesis did not curtail prosthebic group formation.