Phosphorylation sites, cytochrome complement, and alternate pathways of coupled electron transport in Euglena gracilis mitochondria.

Abstract Mitochondria from Euglena gracilis have been examined for the possible absence of a phosphorylation coupling site and of b type cytochromes. Mitochondria from cells grown on glutamate + malate as carbon source are shown by site-specific assays to contain coupling sites analogous to Sites I, II, and III of the mammalian respiratory chain. An alternate pathway, relatively specific for oxidation of lactate and NADH, is able to bypass the antimycin-sensitive site. The lowered phosphorylation efficiency associated with this pathway suggests that it involves one less than the normal number of coupling sites. In the absence of antimycin, oxidation of lactate involves at least two coupling sites in these mitochondria. Low temperature difference spectra provide evidence for the presence of seven substrate-reducible cytochrome components in Euglena mitochondria, including components analogous to a and a3 of mammalian cytochrome oxidase, two c type cytochromes, and three b types. One of these may bind CO and cyanide; another resembles mammalian cytochrome b.

Mitochondria from cells grown on glutamate + malate as carbon source are shown by site-specific assays to contain coupling sites analogous to Sites I, II, and III of the mammalian respiratory chain. An alternate pathway, relatively specific for oxidation of lactate and NADH, is able to bypass the antimycin-sensitive site. The lowered phosphorylation efficiency associated with this pathway suggests that it involves one less than the normal number of coupling sites. In the absence of antimycin, oxidation of lactate involves at least two coupling sites in these mitochondria.
Low temperature difference spectra provide evidence for the presence of seven substrate-reducible cytochrome components in Eugleno mitochondria, including components analogous to a and a3 of mammalian cytochrome oxidase, two c type cytochromes, and three b types. One of these may bind CO and cyanide; another resembles mammalian cytochrome b.
Osidative phosphorylation in mitochondria isolated from Euglena gracilis has been described by Ruetow and Buchanan (1) who showed a number of similarities to mammalian mitochondria. Based on their observation of P:O ratios approaching one less than the theoretical values for various substrates, and on the reported inability of Perini,Schiff,and Kamen (2)  suggested that Euglena mitochondria might lack one of the three coupling sites found in mitochondria of higher organisms or be deficient in some component (such as a b type cytochrome) essential for the activity of one such site.
hlitochondria lacking a specific coupling site are of interest because they offer a possible approach to identifying the components that participate in the initial, site-specific reactions of oxidative phosphorylation.
The discovery that the mitochondria of Saccharomyces cerevisiae lack coupling site I, rotenonesensitive NADH oxidase, and an electron paramagnetic resonance active component in the NADH dehydrogenase region of the electron transport chain (3, 4) has stimulated a good deal of work in this area. It would evidently be useful to identify other cases in which phosphorylation sites are missing from the mitochondrial electron transport chain.
The present investigations were undertaken to establish the number and location of coupling sites in Euglena mitochondria and to characterize the respiratory chain further.
Results obtained under one set of growth conditions are described in this paper. These indicate the presence of three coupling sites, but suggest that one of these can be bypassed selectively by an alternate pathway of electron transport.
We present spectroscopic evidence that the respiratory chain involves at least three b type cytochromes in addition to the c and a types already identified (2, 5).
These results serve as a basis for comparing biochemical properties of the mitochondria under different conditions of growth.
In some instances, profound modifications of the biochemical patterns described here occur. One of these is the subject of an accompanying paper (6).
Issue of January 10, 1970 T. K. Xharpless and R. A. Butow 51 a citrate plus salts mixture modified from a medium described by Hutner et al. (7) and containing 6.0 g per liter of dl-malic acid + 6.6 g per liter of sodium glutamate as carbon source. Initial pH was adjusted to 4.1 f 0.1 with NaOH.
Thiamine HCl and Vitamin B12 (Nutritional Biochemicals) were added as filter sterilized solutions just before each growth cycle, to final concentrations of 1 mg and 5 pg per liter, respectively.
The growth conditions gave generation times of about 12 hours and culture densities in late log phase of approximately 0.5 x lo6 cells per ml, corresponding to 3 to 4 g, wet cell weight, per liter. Stocks were maintained at 27" on a semidefined liquid medium and transferred at 2-to 3-week intervals.
The exact compositions of the media used were as follows; all quantities are expressed in milligrams per liter. Preparation of Mitochondria-Mid to late log phase cultures were filtered through several layers of cheesecloth to remove slime and the cells were harvested by centrifugation at 1200 x g for 5 min. From this point all operations were carried out at O-4". The ce 1s were washed once in a medium containing 0.15 M sorbitol, 1 mM EDTA, and 25 mM Tris chloride (pH 7.5), washed again in isolation medium (0.3 M sorbitol, 0.5 mM EDTA, and 25 mM N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid (pH 7.0)), and suspended at a concentration of approximately 50% (w/v) in the same medium.
Twenty to 40 ml of this suspension were shaken with 50 g of acid-washed glass beads (0.5mm diameter) for 20 set at maximum speed in an MSK cell homogenizer (Bronwill Scientific Company). The homogenate was decanted, the beads were rinsed two or three times, and the volume was adjusted to a 10% (w/v) suspension.
The diluted homogenate was centrifuged at 1200 x g for 10 min, yielding a large, firm pellet of starch, unbroken cells, and debris.
The supernatant suspension was centrifuged at 8000 x g for 15 min and the crude mitochondrial pellet was resuspended by swirling and careful pipetting to separate it from a small starch pellet usually present.
The mitochondria were washed once or twice in 20 to 30 ml of isolation medium and resuspended at a concentrat on of 20 to 30 mg protein per ml. When the mitochondria were to be used for phosphorylation experiments the washing and resuspending medium was supplemented with crystalline bovine serum albumin (Sigma) at 0.1 g/100 ml. Final yields of mitochondrial protein determined according to Lowry et al. (8) or by the biuret procedure (9) were 1.5 to 2.5 mg per g of wet cells. Mitochondria were prepared on the day of each experiment since oxidative phosphorylation was found to decay rather rapidly on storage, either at 4" or frozen at -20" or -90".
The data presented in Tables I and II were obtained with mitochondria prepared by the procedure of Buetow and Buchanan (lo), in which the cells were broken by hand grinding in a mortar with glass beads (0.2-mm diameter).
In this case the isolation medium contained 0.25 M sucrose, 0.5 InM EDTA, and 25 mM Tris chloride (pH 7.4). Less consistent results were obtained by this procedure, probably because of the difficulty of reproducing the conditions of breakage.
Measurement of Respiration and Phosphorylation-Rates of oxygen consumption were measured in sorbitol-EDTA-N-2hydroxyethylpiperazine-N-2-ethanesulfonic acid buffer at 30" in a water-jacketed 1.5-ml glass chamber (Gilson Medical Electronics) containing a Clark type polarographic oxygen electrode. Oxidative phosphorylation was assayed either in the same apparatus with 32Pi (New England Nuclear) at specific activities between 2 x 10" and IO5 cpm per pmole or in the Warburg respirometer with 32Pi at specific activities of 0.5 to 2.5 X lo4 cpm per pmole. In e&her case the assay mixtures contained 16.7 or 20 mM Pi, 5 InM Mg"+, 2 mM ATP, 33 mM glucose, 1 mg per ml of bovine serum albumin, and 3 to 5 units per ml of hexokinase (Boehringer, dialyzed).
In the Warburg experiments, flask center wells contained filter paper strips moistened with 10% KOH or with 1 M KCN in 0.1 N KOH when cyanide was added to the medium.
Incubations were terminated after exhaustion of dissolved O2 in the polarographic experiments or after 30 to 45 min in Warburg experiments, by adding 0.1 volume of 50% (w/v) trichloracetic acid, and the precipitated proteins were removed by centrifugation.
Inorganic phosphate was precipitated from 0.2ml aliquots of the deproteinized solutions by the triethylammonium molybdate method of Sugino and Miyoshi (11); after centrifugation the supernatant solutions containing esterified Y' were filtered through small plugs of nonabsorbent cotton in Pasteur pipettes to assure removal of traces of the precipitate. This procedure reliably removed more than 99.9% of the radioactivity from blank mixtures, provided that the 32Pi was of high radiochemical purity. Esterified phosphate was determined by counting samples and suitably diluted standards on aluminum planchets in a Nuclear-Chicago low background gas flow counter.

Site-speciJic Phosphorylation
Assays-Phosphorylation corresponding to coupling Site I was assayed by the method of Schatz and Racker (12) with coenzyme Q1 as acceptor for electron transfer from NADH. Phosphorylation in the span from NADH to oxygen was assayed in the same experiments by omitting coenzyme Q1 and cyanide from the mixture.
The reactions were followed by recording absorbance changes due to the disappearance of NADH at 340 rnp with a Gilford automatic spectrophotometer, with an adjusted extinction coefficient of 6.8 rnM-' cm-l when coenzyme &I was present (12).
Site II phosphorylation was assayed by a modification of the method described by Lee,Sottocasa,and Ernster (13). Reduction of ferricyanide was determined spectrophotometrically at 410 rnp, with an extinction coefficient of 1.0 n1M-l cm-l, and phosphorylation was measured as described above. P/2eratios were calculated on the basis of half of the number of moles of ferricyanide reduced. (D t 1 e ai s are given in Table II   Relative concentrations of the various absorbing species were calculated by reference to the intensity of the fully reduced cytochrome oxidase a-band at 607 mp (reference wave length, 625 mp). For spectra at liquid nitrogen temperature, mitochondria were suspended in 50% (v/v) glycerol containing 0.5 mM EDTA and 25 mM N-2-hydroxyethylpiperazine-N'-2ethanesulfonic acid, adjusted to pH 7.0 with KOH.

Number of Phosphorylation Sites-The
site-specific assays gave results which are consistent with the presence of phosphorylation coupled to electron transport through each of the corresponding coupling sites. Table I shows the results of an experiment measuring Site I phosphorylation.
Here, the effects of cyanide, coenzyme Q1, and rotenone on NADH oxidation show the presence of an active, rotenone-sensitive NADH-COQ1 reductase. Despite the low over-all P/2e-ratios, it is clear that the reaction is coupled to phosphorylation with a P/2e-ratio about one-third as great as the P/2e-ratio for oxidation of NADH by oxygen. Buetow and Buchanan (1) obtained P:O ratios with succinate which were about 1 unit lower than with NAD-linked substrates, implying the presence of a coupling site analogous to Site I which we have shown directly.
Thus if Euglena mitochondria indeed lack one coupling site, the defect must lie in the span bet.ween succinate and oxygen.
We have found that Euglena mitochondria catalyze a rapid reduction of ferricyanide by succinate in the presence of rotenone and cyanide.
The sensitivity of the reaction to antimycin is high (70 t,o 90%) and varies only slightly with ferricyanide concentration in the range 0.2-2 mM. The situation is thus favorable for assay of phosphorylation specific to Site II, as shown in Table II. The rate of ferricyanide reduction was inhibited 70% by antimycin in this case, and the phosphorylation by 74%, suggesting that the major part of the electron flow from succinate to ferricyanide traverses a coupling site analogous to Site II. Table  II also shows the results of the Site III assay in the same preparation of mitochondria.
Again the data suggest the presence of a coupling site on the major electron transport pathway.
The corrected P/2e-ratios (given in parentheses) for the site-specific assays approximate half of the over-all P:O ratio for oxidation of succinate (Table II), but the magnitude of the corrections applied makes this apparent quant,itative agreement somewhat uncertain.
By an extension of the ferricyanide method above it has been possible to make a more convincing quantitative demonst,ration of the presence of two separate coupling sites in the region between succinate and oxygen.
These experiments take advantage of the fact that, in the absence of cyanide, Euglena mitochondria catalyze the reduction of ferricyanide and of oxygen at comparable rates, so that the two acceptors can be made to compete for reducing equivalents derived from succinate. When ferricyanide (0.2 to 1 mM) was added to mitochondria oxidizing cofactors, and bovine serum albumin as described under "Methods " 6.7 mM succinate, 3.3 FM rotenone, and 1.7 mg per ml of mitochohdrial protein.
Ferricyanide was present at 0, 0.28, 0.84, and 0.84 mM in the order plotted; in addition, the last sample contained 0.4 mM KCN.
Total rates of electron transfer in t,he four samples were 95, 77, 72, and 60 mFeq min-1 w-l, respectively.
Oxygen consumption was measured polarographically; ferricyanide remaining was determined spectrophotometrically after deproteinization of the reaction mixtures Control experiments confirmed that ferrocyanide was not oxidized under the conditions used.
succinate in the polarograph, 02 consumption slowed by more than SOTo, returning to its original rate after all of the ferricyanide had been reduced as judged by the disappearance of its yellow color. In the experiment illustrated by Fig. 1, mitochondria were allowed to oxidize measured amounts of ferricyanide and of oxygen, present together, and the fraction of the total electron flux reaching each acceptor was varied by varying the initial concentration of ferricyanide and the time of incubation, and by addition to one sample of a low concentration of cyanide, sufficient to inhibit the oxygen reaction by about 60%. If reducing equivalents must traverse two independent coupling sites betu-een succinate and oxygen, and one of these is between succinate and ferricyanide, then one expects the P/20 ratio (Pi esterified/ total equivalents transferred) to decrease linearly with a decrease in the fraction of the flux reaching oxygen.
The P/2e-ratio extrapolated to zero reduction of oxygen should be equal to the P/2e-ratio for Site II alone. The data of Fig. 1 conform to the prediction of a linear variation of the P/2e-ratio, providing strong evidence for the presence of two separate phosphorylation sites in the span between succinate and oxygen. The extrapolated values for the ferricyanide reaction (0.24) are somewhat less than half of the P/2e-ratio from succinate to oxygen (0.66). If it is assumed that only 75% of the ferricyanide was reduced via the phosphorylating pathway, in agreement with the observed sensitivity of the reaction to antimycin (Table II, and the text above), then the corrected P/2e-value for Site II becomes 0.32. It should be noted that the total rates of electron transfer (see legend to Fig. 1) differed by less than a factor of 1.5 between the various samples. Experiments of this type have yielded consistent results even when the succinate P:O ratio was as low as 0.2. Respiratory rates were determined polarographically as described under "Methods." Succinate was used at a final concentration of 16.7 mM, NADH at a final concentration of 0.93 mM; the concentration of mitochondrial protein was 0.087 mg per ml. Antimycin was added as a solution of 2.0 pg per ml (3.65 PM) in isopropanol.
In addition, we have confirmed that the phosphorylation observed at both sites is completely inhibited by 3.3 PM carbonyl cyanide-m-chlorophenylhydrazone.
Similar results were also obtained with mitochondria from E. gracilis var. bacillaris.

Antimycin-resistant
Respiration-Mitochondria prepared under the present conditions have, as a rule, shown a high degree of inhibition of succinate oxidation by antimycin. On the other hand, we have consistently observed that a significant fraction of the NADH oxidase activity (corrected for cyanide-insensitive NADH oxidase, see "Discussion") was not sensitive to antimycin. Fig. 2 shows titrations of succinoxidase and NADH oxidase activities with antimycin, in a preparation low in cyanide-insensitive NADH oxidase activity. The succinoxidase titration curve shows the sigmoid shape and sharp end point characteristic of antimycin inhibition in mammalian mitochondria (14); the end point lies at about 0.15 mpmoles of antimycin per mg of protein, a value comparable to the estimated cytochrome a content of these mitochondria (see below). While the shape and transition midpoint of the curve for NADH oxidase are identical, it displays about 25% residual activity which is still sensitive to cyanide; this activity persisted up to more than loo-fold higher concenkations of antimycin. All preparations tested, both from Z strain and var. bacillaris cells, have given similar results.
In other experiments, oxidation of malate showed a partial antimycin resistance similar to NADH, and oxidation of n-and L-lactate were even more resistant to antimycin, maximum in-  hibitions being in the range of 10 to 50% when assayed polarographically.

Phosphorylation
Coupled to Antimycin-resistant Respiration-Since its sensitivity to cyanide suggests that the antimycinresistant respiration involves at least part of the normal electron transport chain, it was of interest to determine whether this activity is coupled to phosphorylat,ion. The data summarized in Tables 111 and IV show that this antimycin-resistant respiration is coupled. Table III shows the effect of antimycin on oxidative phosphorylation with NADH as substrate.
The effect of antimycin is to cause a small decrease in the P:O ratio; at the higher concentration there is a small inhibition of respiration as well. In contrast to the effect of antimycin, cyanide inhibits phosphorylating respiration almost completely, although in this particular experiment there is a considerable amount of nonphosphorylating, cyanide-insensitive NADH oxidase activity remaining. Note that the P:O values given in Table III  necessary, little significance can be attached to the absolut)e values of the P:O ratios obtained. Table IV compares results obtained with lactate and succinate. The data have not been corrected for cyanide-resistant activities, which were found to be low and similar for the three substrates. It can be seen that, with succinate as substrate, oxidation and phosphorylation are strongly inhibited by antimycin. The rate of n-lactate oxidation, on the other hand, is unaffected by antimycin, while the P:O ratio is decreased by almost 500/,. Osidation of L-lactate is moderately inhibited by antimycin and the P : 0 ratio is diminished by about one-third.
The finding of P : 0 ratios significantly greater than 1.0 with both lactate isomers distinguishes Euglena from yeast, whose mitochondrial lactate oxidase system includes only the single phosphorylation site associated with cytochrome oxidase (3). The depression of the P:O ratios by antimycin may reflect some uncoupling by the inhibitor at the relatively high concentration used (2.4 mpmoles per mg of protein).
Significant uncoupling by antimycin at concentrations greater than 0.8 mpmoles per mg of protein has been observed in other systems (15,16). However, in another experiment antimycin at only 0.52 mpmole per mg of protein depressed the P:O ratios for both D-and L-lactate by more than 50%; note also in Table III that antimycin apparently had a similar effect on the P:O ratio whether present at 0.09 or at 0.47 mpmole per mg of protein.
CytochromesFig. 3 shows a difference spectrum recorded at liquid nitrogen temperature between anerobic, succinate-reduced mitochrondria and aerobic, oxidized mitochondria. A large a-band complex (Fig. 3B), which at room temperature appears as a perfectly symmetrical peak at 559 rnp, is resolved into at least four components, with shoulders at 552 and 565 rnp and maxima at 558 and 561 mp. The a-band of cytochrome oxidase at 607 ml is smaller and more symmetrical, with a broad shoulder centered at 594 mp. In the Soret region (Fig. 3A) a maximum at 443 rnp and a shoulder at 453 rnp suggest the presence of two cytochrome a type components analogous to the a and a3 of mammalian cyt,ochrome oxidase (17). Mitochondria suspended at 3.0 mg of protein per ml in sorbitol-EDTA=N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid buffer (see "Methods"), containing 9 mM succinate, were deaerated in Thunberg type cuvettes of l-cm light path by evacuating and refilling with Nz several times. After an instrument base line was determined the sample cuvette was evacuated, refilled with CO, and mixed by shaking.
Spectra were recorded at the indicated times after mixing and corrected by subtraction of the base line value. Fig. 4 shows difference spectra recorded at room temperature at various times after addition of carbon monoxide to anaerobic, succinate-reduced mitochondria.
A strong trough at 445 rnp which appears rapidly and changes little with time is charact.eristic of the reaction of an us type of cytochrome with CO (18); the small peak near 593 rnp may be due to the same pigment.
A second component which reacts more slowly with CO produces a strong peak at 421 rnp and a peak and trough in the a-band region, apparently due to the shift to shorter wave length of an absorption band originally centered near 560 mp. Cyanide induces a difference spectrum in substrate-reduced, anaerobic mitochondria which is similar to this slower reacting CO component (data not shown).
In this case the Soret feature of the cyanide-binding cytochrome appears as a symmetrical peak and trough at 422 and 433 rnp, respectively.
Carbon monoxide difference spectra similar to those shown in Fig. 4 have been reported for Astasiu Zonga (5) and for some bacterial systems (19,20).
The use of antimycin for selective reduction of the 6 type cytochromes (21) allows further resolution of the overlapping a-bands in the region around 558 rnp. Fig. 5, Curve A, shows a difference spectrum induced by succinate plus antimycin, recorded at liquid nitrogen temperature.
The maximum at 558 rnp and the pronounced shoulder at 565 rnp indicate the presence of at least two b type cytochromes.
In other experiments a third b type cytochrome at 561 mp has been clearly resolved (compare Fig. 3 and also Fig. 7 of the accompanying paper (6)). Curve B of Fig. 5 is the difference (succinate + cyanide) -(succinate + antimycin), recorded directly, and shows those components which were not reduced in the presence of antimycin.
Here a shoulder at 552 rnp and a peak at 555 rnp indicate the presence of two c type cytochromes.
The features in the region between 555 and 565 rnp are complex, and may be due in part to a shift in the a-band of the cyanide-binding cytochrome mentioned above. It is possible, however, that part of the strong shoulder at 560 rnp is due to a b type cytochrome which did not become ABSORBANCE   I   I  I  I  I  I  I  I   540  560  580  600  Dithionite was added as a few tenths of a milligram of the solid per ml of suspension.
fully reduced in the presence of antimycin; this would be consistent with the finding of an alternate respiratory pathway in these mitochondria which bypasses the antimycin block. The small trough at 565 rnp in Fig. 5, Curve B, almost certainly results from the fact that the corresponding b type cytochrome, like mammalian cytochrome 6 shows a greater absorbance in the presence of antimycin (22) or in dithionite-reduced mitochondria (23) than under anaerobiosis or in the presence of cytochrome oxidase inhibitors.
The effect may be verified by comparing the relative prominence of this band in Fig. 5, Curve C (dithionite -oxidized), and in Fig. 3 (anaerobic -oxidized). Table V summarizes spectral data for the cytochromes which we have been able to detect in Euglena mitochondria.
These data have been varified by comparison of difference spectra made under a variety of conditions, and we believe that they describe the minimum number of components necessary to account consistently for the spectral features observed.
The types of cytochromes present and their relative proportions, as judged from difference spectra, have been very consistent between different lots of mitochondria as well as in mitochondria isolated under different conditions of growth (6). The content of cytochrome oxidase in all of these preparations has averaged 0.25 mpmole per mg of mitochondrial protein (range: 0.12 to 0.56), based on room temperature measurements of the reduced a-band at 608 rnp, with 625 mp as a reference wave length and assuming an extinction coefficient difference of 10 rnM-l cm-l (24).
Comparison of whole cell difference spectra suggests that the c type component at 555 rnp ( The data are for reduced-oxidized difference spectra at liquid nitrogen temperature. Intensities of absorbance were measured from base lines drawn through the 543, 575, and 625 rnp points in dithionite-reduced spectra, and are given relative to the intensity of the cytochrome a a-band at 607 rnp (mean of five determinations). In the pure state this cytochrome was shown to have a double a-band (reduced) with maxima at 554.5 and 558 mp (25). If it retains this property in situ then part of the absorbance at 558 rnp must be contributed by this cytochrome; however, it is not possible to account for all of the differences that we have observed at both 555 and 558 mp in terms of a single component.
We have not detected any qualitative difference between spectra induced by succinate and by D-lactate.
In particular, the same group of three b type cytochromes is reduced to about the same extent in the steady state either substrate. There is, however, a quant,itative difference in the degree of further reduction of the b t,ype cytochromes brought about by addit,ion of antimycin, the increase being 3 to 4 times larger for succinate than for lactate.
With either substrates the three b components respond to antimycin as a group; none of them shows large absorbance changes relative to the others. The 565 rnp b type shows the largest percentage change, however.

DISCUSSION
It is apparent from the results presented in this paper that three phosphorylation sites corresponding to mammalian Sites I, II, and III, lying in the regions of the electron transport chain sensitive to rotenone, antimycin, and cyanide, respectively, are present in Euglena mitochondria.
Moreover, these mitochondria appear to have a full complement of cytochromes but differ from mammalian mitochondria in that at least three b type cyto-  Buetow and Buchanan (1). This conclusion is confirmed by the results of the s:te-specific assays, the simultaneous assay of Sites II and 111, and the spectroscopic data. One obvious explanation for low P:O ratios is that the present methods of preparation yield damaged, partly uncoupled mitochondria.
This was certainly the case in our early experiments, in which P:O ratios less than 1.0 were regularly observed with all substrates, and there was no indication of respira.tory control by ,4DP. In more recent. preparations we have observed stimulations of respiration of the order of 2-fold by phosphate acceptor, and P:O ratios significantly higher than 1.0 with succinate and lactate, suggesting that with improved methods of preparation theoretical P:O ratios might be obtained consistently. The elimination of a nonphosphorylating, cyanide-insensitive NADH oxidase activity (compare Table III), which fractionation studies (not reported in this paper) indicate to be a cytoplasmic contaminant in the mitochondrial preparations, and lead to substantial increases in the apparent P:O ratios for oxidation of NADH and of NAD-linked substrates. Under certain conditions the efficiency of oxidative phosphorylation may be affected by the presence of alternate pathways for electron transport, which bypass one or more coupling sites. Inhibitor-resistant respiration has been observed frequently in mitochondria from plant tissues (27)(28)(29) and in some cases the resistant pathways show considerably altered phosphorylation efficiencies. Hackett, Rice, and Schmid (29) observed that P:O ratios in sweet potato mitochondria with citrate and succinate as substrates were specifically depressed by several respiratory chain inhibitors including carbon monoxide, cyanide, and antimycin.
They suggested that the normal respiratory chain between cytochrome b and oxygen could be bypassed by an alternate pathway containing fewer phosphorylation sites. We have always observed phosphorylating respiration in Euglena mitochondria to be cyanide-sensitive, suggesting that cytochrome oxidase participates in the alternate as well as the normal pathway.
Levels of CO sufficient to inhibit succinate oxidation by more than 40% have no effect on the P:O rati0.l An extreme example, in which all phosphorylation sites between succinate and oxygen are bypassed by an alternate pathway, occurs in Euglena under different growth conditions from those used here, and is described in the following paper (6).
The effects of antimycin on the P:O ratios with NADH, lactate, and even succinate (Table IV) suggest that in the presence of antimycin electron transport bypasses a single phosphorylation site, presumably Site II, thus lowering the P:O ratio by about one-third for NADH and about one-half for succinate or lactate.
It must further be supposed that the alternate pathway carries only a minor fraction of the electron flow in the absence of antimycin.
It is not difficult to imagine how this might be the case for succinate, NADH, or L-lactate, since antimycin inhibits the total rate of oxidation of these substrates by at least 40%.
On the other hand, D-lactate is oxidized at nearly the same rate in the presence or the absence of the inhibitor, only the P:O ratio changing.
One possible explanation of these results is that oxidation of n-lactate (and to a lesser extent of L-lactate or NADH) normally proceeds largely via an alternate electron transport chain containing two phosphorylation sites, i.e. Site III and a site similar to but not identical with Site II of the succinate chain. The Issue of January 10, 1970 T. K. Xharpless and R. A. Butow 57 depression of P:O ratios with other substrates could result from a combination of diversion of electron flow into this alternate pathway and of site specific uncoupling by antimycin. This uncoupling must be specific since, over a wide range of antimycin concentrations, P:O ratios are depressed to a similar extent. Another possibility is that antimycin somehow activates an alternate pathway specific for lactate oxidation, which bypasses the normal antimycin-sensitive locus and coupling Site II, so that net electron flux remains nearly the same in the presence of the inhibitor while the P:O ratio decreases.
In Euglena it is nevertheless clear that in the absence of antimycin the oxidation of lactate must involve at least two coupling sites. This is consistent with the observation of Rutner and Price (30)