Identifying a Protein Central to the Activity of ATP Synthase Much of our knowledge about the steps in the respiratory chain and the mechanism of ATP synthase came about by dissecting the pathway, using various inhibitors and uncouplers (see Table 19-4) and bacterial mutants. In this problem, we see how Robert Fillingame used dicyclohexylcarbodiimide (DCCD) and $E .$ coli mutants resistant to its effects to identify the components that came to be known as the c subunits of the $\mathrm{F}_{\mathrm{o}}$ portion of ATP synthase. DCCD reacts with carboxyl groups in the side chains of Asp and Glu residues. When DCCD is added to a suspension of intact, actively respiring mitochondria, the rate of electron transfer (measured by $\mathrm{O}_{2}$ consumption) and the rate of ATP production dramatically decrease. If a solution of 2,4 -dinitrophenol (DNP) is now added to the preparation, $\mathrm{O}_{2}$ consumption returns to normal, but ATP production remains inhibited.
(a) Explain the effect of DNP on the inhibited mitochondrial preparation.
(b) Which process is directly affected by DCCD, electron transfer or ATP synthesis?
E. coli carries out oxidative phosphorylation with machinery remarkably similar to that in mammals, and $E .$ coli is far more amenable to mutant selection. Addition of DCCD to a culture of wild-type $E .$ coli (strain AN180) growing aerobically blocks further growth in a time- and dose-dependent fashion. Fillingame selected a DCCD-resistant mutant of $E .$ coli (RF-7) for which aerobic growth was only slightly diminished in the presence of DCCD. Next, he needed to demonstrate that the DCCD-resistant component in his $E .$ coli strains was the ATP synthase. He isolated the membrane fraction from the wild-type and RF-7 strains and assayed them for ATPase activity in the presence and absence of DCCD. He found timeand dose-dependent inhibition of the ATPase activity in the membrane fraction of the wildtype, but not in the RF-7 membrane fraction.
(c) Why did Fillingame assay ATPase activity instead of ATP synthase activity?
(d) Is the DCCD-binding protein missing in the mutant RF-7, or just altered?
Fillingame wanted to know whether the DCCD-sensitive protein was an integral part of the membrane or could be solubilized into the fraction that contained the ATPase activity. He prepared "stripped membrane" and "soluble ATPase" fractions from both wild-type cells and RF-7 mutants, by treating intact membranes with dithiothreitol. He measured the ATPase activity in the native membranes, in the stripped membranes, and in systems reconstituted by mixing the stripped membranes with the soluble fraction from the wildtype or RF-7 mutant strain. The native membranes and reconstituted systems all had ATPase activity; the stripped membrane fractions had very little ATPase activity. Having established that all combinations of reconstituted systems had similar ATPase activity, Fillingame then added DCCD to see which combinations were inhibited.
(e) What results would you expect if the DCCD-binding protein were in the stripped membranes? What would you expect if it were in the soluble fraction?
The results were clear. For the stripped membranes from wild-type cells, the reconstituted ATPase was sensitive to DCCD, regardless of the source of the soluble fraction. For the stripped membranes from mutant cells, the reconstituted ATPase was insensitive to DCCD. So, DCCD sensitivity is due to a protein in the stripped membrane fraction, not to a protein in the fraction solubilized with dithiothreitol. To identify the DCCD-sensitive protein, Fillingame exposed intact membranes of the wild-type (AN180) and RF-7 E. coli to $^{14}$ C-labeled DCCD, then used SDS-PAGE to separate the proteins. He cut the gel into thin slices from bottom to top and determined the $^{14}$$\mathrm{C}$ content of each slice, measured as disintegrations per minute (dpm) per $2 \mathrm{mm}$ gel slice, normalized to the amount of protein applied to the gel. The distance migrated is equal to the slice number times $2 \mathrm{mm}$. The results are plotted below. The arrows denote cytochrome $c,$ used as a molecular mass marker; I and II, peaks of interest; and BPB, bromphenol blue, a tracking dye to indicate the front of the sample as it moves through the gel. Many proteins from each sample were labeled with $\left[^{14} \mathrm{C}\right]$ DCCD (measured in dpm, disintegrations per minute). (FIGURE CAN'T COPY)
(f) How did Fillingame know which labeled protein(s) was/were of interest?
(g) When he repeated the experiment using "stripped membranes" prepared as before, he found the same protein was specifically labeled in the wild-type fraction. Why was this step necessary?
(h) The protein of interest proved to have an $M_{\mathrm{r}}$ of about $9 \mathrm{kDa}$, and was readily soluble in a very nonpolar solvent (chloroform/methanol). What could Fillingame deduce about the structure, location, and topology of this protein?
(i) In later work, Fillingame found that the residue that reacts with DCCD in this protein is $\mathrm{Asp}^{61} .$ An $E .$ coli mutant in which this protein has a Ser residue substituted for Ala $^{21}$ is much less sensitive to inhibition by DCCD than is the wild type. What explanation can you give for this observation?
(j) Extensive studies of this DCCD-inhibited protein have shown it to be a central part of the $\mathrm{F}_{\mathrm{o}} \mathrm{F}_{1}$ ATP synthase of bacteria, plants, and animals. What is its role in oxidative phosphorylation?