The role of extra fragment at the C-terminal of cytochrome b (Residues 421-445) in the cytochrome bc1 complex from Rhodobacter sphaeroides.

Sequence alignment of cytochrome b of the cytochrome bc1 complex from various sources reveals that bacterial cytochrome b contain an extra fragment at the C terminus. To study the role of this fragment in bacterial cytochrome bc1 complex, Rhodobacter sphaeroides mutants expressing His-tagged cytochrome bc1 complexes with progressive deletion from this fragment (residues 421-445) were generated and characterized. The cytbDelta-(433-445) bc1 complex, in which 13 residues from the C-terminal end of this fragment are deleted, has electron transfer activity, subunit composition, and physical properties similar to those of the complement complex, indicating that this region of the extra fragment is not essential. In contrast, the electron transfer activity, binding of cytochrome b, ISP, and subunit IV to cytochrome c1, redox potentials of cytochromes b and c1 in the cytbDelta-(427-445), cytbDelta-(425-445), and cytbDelta-(421-445) mutant complexes, in which 19, 21, or all residues of this fragment are deleted, decrease progressively. EPR spectra of the [2Fe-2S] cluster and the cytochromes b in these three deletion mutant bc1 complexes are also altered; the extent of spectral alteration increases as this extra fragment is shortened. These results indicate that the first 12 residues (residues 421-432) from the N-terminal end of the C-terminal extra fragment of cytochrome b are essential for maintaining structural integrity of the bc1 complex.

The cytochrome bc 1 complex is an essential energy transduction electron transfer complex in mitochondria and many aerobic and photosynthetic bacteria (1). The complex catalyzes electron transfer from ubiquinol to cytochrome c 1 with concomitant translocation of protons across the membrane to generate a membrane potential and proton gradient for ATP synthesis. All the cytochrome bc 1 complexes contain three core subunits, cytochrome b, cytochrome c 1 , and Rieske iron-sulfur protein (ISP), 1 which house two b-type hemes (b L and b H ), one c-type heme (heme c 1 ), and a high potential [2Fe-2S] cluster, respec-tively. In addition to these three core subunits, the cytochrome bc 1 complex also contains varying numbers (one to eight) of non-redox containing subunits, known as supernumerary subunits (2,3).
Because the bacterial complexes contain no (or one) supernumerary subunit, it is unlikely that the structures of the core subunits in these complexes are, as suggested for the mitochondrial complex (4), stabilized through interactions between core subunits and their neighboring supernumerary subunits. Perhaps interactions between a part of a core subunit and another part of the same subunit or another core subunit contribute to the stability of a core subunit in the bacterial complex. This speculation finds some support from the fact that core subunits in bacterial complexes are generally bigger than their counterparts in the mitochondrial complex.
Sequence alignment of cytochrome b, cytochrome c 1 , and ISP in bacterial complexes with their counterparts in mitochondrial complexes reveals four extra fragments in bacterial cytochrome b and one each in bacterial cytochrome c 1 and ISP (5). These extra fragments are modeled into the structure of the Rhodobacter sphaeroides bc 1 complex by using coordinates of mitochondrial supernumerary subunits (5). These findings encouraged us to suggest that these extra fragments may possess mitochondrial supernumerary subunit function in stabilizing the structure of the core subunits in the bacterial complex. This suggestion is further supported by the recent finding that ISP is lost from the R. sphaeroides bc 1 complex if the extra fragment of ISP is deleted or substituted with alanine (6). Of course, confirmation of this suggestion will have to wait until the function of extra fragments in cytochrome b and c 1 are established.
Cytochrome b holds a central role in the cytochrome bc 1 complex because it houses two ubiquinone binding sites, Q o and Q i , and two redox centers, heme b L and heme b H . According to the Q-cycle mechanism (7), electrons from ubiquinol are bifurcated at the Q o site. The first electron of ubiquinol is transferred through the so-called "high potential" chain consisting of [2Fe-2S] and heme c 1 . The second electron of ubiquinol is passed through the "low potential" chain consisting of hemes b L and b H . Thus, maintaining the structural stability of cytochrome b in the bc 1 complex is crucial for the electron and proton transfer functions of this complex.
In the structure model of R. sphaeroides cytochrome bc 1 complex (5), the four extra fragments of cytochrome b are located at the N terminus (residues 2 to 12), the connecting loop between helices D and E (residues 232 to 239), the connecting loop between helices E and F (residues 309 to 326), and the C terminus (residues 421-445) (see Fig. 1). We started functional studies of these cytochrome b extra fragments with the one at the C terminus because this extra fragment is in close proximity to subunit IV and ISP. Herein we report generation of * This work was supported by Grant MCB0077650 (to L. Y.) from National Science Foundation and Grant GM30721 (to C.-A. Y.) from the National Institutes of Health and Oklahoma Agricultural Experiment Station Projects 1819 and 2372, Oklahoma State University. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
The pRKD418/fbcFB m C H Q plasmid in E. coli S17-1 cells was mobilized into R. sphaeroides BC-17 cells by a plate-mating procedure (10). The presence of engineered mutations was confirmed twice by DNA sequencing of the 962-base pair BstEII-XbaI fragment before and after photosynthetic growth as previously reported (10). DNA sequencing and primer synthesis were performed by the Recombinant DNA/Protein Core Facility at Oklahoma State University.
Enzyme Preparations and Activity Assay-Chromatophores were prepared from frozen cell paste, and cytochrome bc 1 complexes with a His 6 tag placed at the C terminus of cytochrome c 1 were purified from chromatophores as previously reported (11). To assay the cytochrome bc 1 complex activity, chromatophores or purified cytochrome bc 1 complexes were diluted with 50 mM Tris-Cl, pH 8.0, containing 200 mM NaCl and 0.01% N-dodecyl-␤-D-maltostide (DM) to a final concentration for cytochrome b of 3 M, unless otherwise specified. Appropriate amounts of the diluted samples were added to 1 ml of assay mixture containing 100 mM Na ϩ /K ϩ phosphate buffer, pH 7.4, 1 mM EDTA, 100 M cytochrome c, and 25 M Q 0 C 10 BrH 2 . Activity was determined by measuring the reduction of cytochrome c (the increase of absorbance at 550 nm) in a Shimadzu UV 2101 PC spectrophotometer at 23°C, using a millimolar extinction coefficient of 18.5 for calculation. The nonenzymatic oxidation of Q 0 C 10 BrH 2 , determined under the same conditions, in the absence of enzyme, was subtracted. Although the chemical properties of Q 0 C 10 BrH 2 are comparable with those of Q 0 C 10 H 2 , it is a better substrate for the cytochrome bc 1 complex (8). Specific activity is defined in the legend to Reductive titrations were carried out by addition of sodium dithionite solution to the ferricyanide-oxidized sample; oxidative titrations were carried out by addition of ferricyanide solution to the dithionite-reduced sample. At indicated E h values during the redox titration absorption spectra, from 600 to 500 nm were taken with a Shimadzu model UV-2100 spectrophotometer. The optical density at 562 nm, minus that at 575 nm, is used for cytochrome b and that at 552 minus 540 nm for cytochrome c 1 . The midpoint potentials of cytochrome b L and b H were calculated by fitting the redox titration data, obtained for cytochrome b, the Nernst equation for a one-electron carrier (n ϭ 1) with two components using Kaleidagraph, and that of cytochrome c 1 was fitted for a one-electron carrier with one component.
Other Biochemical and Biophysical Techniques-The contents of cytochrome b (14) and cytochrome c 1 (15) were determined according to published methods. SDS-polyacrylamide gel electrophoresis was performed according to Laemmli (16) using a Bio-Rad Mini-Protean® 3 Cell. Western blotting used rabbit polyclonal antibodies raised against R. sphaeroides ISP and subunit IV. The polypeptides separated in the SDS-PAGE gel were transferred to a 0.22-m nitrocellulose membrane for immunoblotting. Protein A conjugated to horseradish peroxidase was used as the second antibody. EPR spectra were recorded with a Bruker EMX spectrometer, equipped with a liquid helium flow cryostat, at 7 K. Instrument settings are detailed in the legends of relevant figures.
The mass spectrometry determination of molecular weights of cytochrome b in the complement and mutant bc 1 complexes was performed with an Applied Biosystems DE-PRO MALDI-TOF mass spectrometer operated in delayed-extraction positive-ion liner mode according to the method of Ghaim et al. (17) with modifications. Samples (ϳ30 l) were mixed with 6 volumes of 10% trichloroacetic acid, chilled for 5 min on ice, centrifuged, and the precipitate was rinsed briefly with 95% ethanol to remove salts. The resulting pellet was redissolved in 30 l of 99% formic acid and a 1:1 dilution with 99% formic acid to a final volume of 60 l, which gave the best crystals and signal. As matrix for MALDI-TOF analyses, 1% of 2,5-dihydroxybenzoic acid in 30% acetonitrile, 0.1% trifluoroacetic acid and 1% 5-methoxysalicilic acid in 30% acetonitrile, 0.1% trifluoroacetic acid were mixed 9:1 (v/v). This matrix solu- tion was mixed 1:1 (v/v) with protein samples. Matrix and sample were spotted onto the sample plate, and were co-crystallized by evaporation in a covered Petri dish under ambient temperature and pressure (17).

Comparison of Electron Transfer Activity, Subunit Composition, and Detergent Lability of Cytochrome bc 1 Complexes in Chromatophore Membranes from Complement and C-terminal Truncated Cytochrome b Mutants-
The C-terminal extra fragment of R. sphaeroides cytochrome b corresponds to residues 421-445 with a sequence of PATIEEDFNAHYSPATGGTKTV-VAE (see Fig. 2). To probe the role of this fragment, R. sphaeroides mutants with progressive deletion of amino acid residues from the C terminus of cytochrome b were generated and characterized. These are: cytb⌬-(433-445), cytb⌬-(427-445), cytb⌬-(425-445), and cytb⌬-(421-445) with deletion of 13, 19, 21, and all residues from this C-terminal extra fragment of cytochrome b, respectively.
When these four C-terminal-truncated cytochrome b mutants were subjected to anaerobic photosynthetic growth conditions, all of them grew at a rate comparable with that of the complement cells. Chromatophores prepared from these mutant cells have the content and absorption spectral properties of cytochrome b and cytochrome c 1 ϩ c 2 similar to those in the complement chromatophores (data not shown). The amounts of ISP and subunit IV in these mutant chromatophores, determined by Western blotting using antibodies against R. sphaeroides ISP and subunit IV, respectively, are also comparable with those in complement chromatophores. These results indicate that the C-terminal extra fragment of cytochrome b is not required for assembly of the cytochrome bc 1 complex protein subunits (cytochromes b and c 1 , ISP, and subunit IV) into the chromatophore membrane.  It should be noted that the effectiveness of DM in solubilizing the bc 1 complex from these deletion mutant chromatophores is comparable with that for the complement chromatophores (   When various chromatophores were treated with 1.2% DM and centrifuged at 200,000 ϫ g for 90 min, about 80% of cytochrome b, cytochrome c 1 , ISP, and subunit IV in the complement and mutant chromatophores was recovered in the supernatant fractions. When these supernatant fractions were applied to Ni-NTA columns, most of cytochrome c 1 in all of these supernatant fractions were absorbed on the Ni-NTA gel, and were recovered in column eluates using an eluting buffer containing 200 mM histidine. This is as expected because the His 6 tag is placed at the C terminus of cytochrome c 1 . The b/c 1 ratios in column eluates of the complement and mutants cytb⌬-  Table  I Fig. 4B). It should be noted that subunit IV of the bc 1 complex produced by R. sphaeroides BC-17 cells carrying the pRKD418/fbcFBCQ plasmid includes chromosomal and plasmid copies, whereas cytochrome b, c 1 , and ISP have only the plasmid copy. These results indicate that the binding affinities of ISP and subunit IV to the complex are affected by the first 12 residues (residues 421-432) at the N-terminal end of the C-terminal extra fragment of cytochrome b; binding affinity decreases as the fragment size decreases.
The finding that the binding of ISP and subunit IV to cytochrome c 1 are affected by the C-terminal extra fragment of cytochrome b is rather surprising, because they are not in the same subunit as cytochrome b. Perhaps interactions between the C-terminal extra fragment of cytochrome b and the Nterminal portion of ISP or subunit IV, located on the cytoplasmic side of the chromatophore membrane, are required for maintaining the structures of ISP and subunit IV for binding to cytochrome c 1 . Thus, the mutations on the C-terminal extra fragment of cytochrome b induce conformational changes on ISP and subunit IV that weaken their binding affinities for cytochrome c 1 . Alternatively, the binding of ISP or subunit IV to cytochrome c 1 is through cytochrome b and mutation of the extra fragment not only decreases the binding affinity of cytochrome b to cytochrome c 1 , but also to ISP and subunit IV.
The Abnormality of Electrophoretic Mobilities of C-terminal Truncated Cytochrome b Mutants- Fig. 5 shows electrophoretic mobility of wild-type and C-terminal-truncated cytochrome b in separation gels containing two concentrations of acrylamide (T) and bisacrylamide (C). In a separating gel having T ϭ 12.5% and C ϭ 3%, a system used routinely for SDS-PAGE analysis of the R. sphaeroides bc 1 complex, wild-type and mutant bc 1 complexes of cytochrome b of cytb⌬-(433-445), (433-445), cytb⌬-(427-445), cytb⌬-(425-445), and cytb⌬-(421-445). Purified complement and mutant bc 1 complexes (230 M cytochrome b) were treated with a small excess of ascorbate solution to fully reduce cytochrome c 1 and frozen in liquid nitrogen. Panel A, spectra obtained from samples scanning once from 3200 to 4000 G magnetic field; panel B, enlarged spectra at the g x region. The g y signal amplitude in mutant complexes have been adjusted to about the same as that in the complement complex by scanning the cytb⌬-(433-445), cytb⌬-(427-445), and cytb⌬-(425-445) complexes, through the magnetic field, once, twice, and five times, respectively. EPR spectra were recorded at 7 K with the following instrument settings: microwave frequency, 9.3 GHz; microwave power, 2.  decreased by 1200, 1947, 2206, 2588, respectively, from that of wild-type cytochrome b. It should be noted that the larger than calculated R f values observed are not because of reduction of the molecular mass of mutant b proteins by proteolytic enzyme digestion, because the molecular mass of cytochrome b in the cytb⌬-(421-445) mutant complex, determined by MALDI-TOF mass analysis, is 47,264, which corresponds to the calculated value.

FIG. 6. EPR spectra of the [2Fe-2S] cluster of the Rieske iron-sulfur protein in purified bc 1 complexes from the complement and mutants cytb⌬-
To further confirm that C-terminal-truncated cytochrome b have abnormal electrophoretic mobility in SDS-PAGE, purified wild-type and mutant cytochrome bc 1 complexes were subjected to SDS-PAGE using a separating gel having T ϭ 16% and C ϭ 4% (see Fig. 5B). The electrophoretic mobility of wild-type R. sphaeroides cytochrome b, relative to cytochrome c 1 , is decreased when the concentrations of T and C in a separation gel is increased. This increased distance between wildtype cytochrome b and c 1 enables us to obtain the R f values for the cytochrome b mutants, relative to cytochrome c 1 . The values are: 0.70, 0.76, 0.81, 0.84, and 0.86, respectively. These values are larger than those calculated. Thus, cytochrome b with decreasing lengths of the C-terminal extra fragment exhibits abnormal electrophoretic mobility in SDS-PAGE. Perhaps removal of the C-terminal extra fragment in cytochrome b makes cytochrome b protein assume a molecular shape more globular than the wild-type protein, in the presence of SDS, which moves faster in SDS-PAGE than expected.
Effect of Mutations on the Rieske Iron-Sulfur Cluster-Western blotting and SDS-PAGE analysis indicate that the amount of ISP in the bc 1 complex decreases as the C-terminal extra fragment of cytochrome b decreases; no ISP is detected in the complex with cytochrome b lacking this extra fragment. To see whether or not the mutations on the C-terminal extra fragment of cytochrome b also affect the microenvironments of the ironsulfur cluster, EPR spectra of the [2Fe-2S] cluster in complement and mutant bc 1 complexes were determined and compared (see Fig. 6).
When complement and mutant complexes of cytb⌬-(433-445), cytb⌬-(427-445), cytb⌬-(425-445), and cytb⌬-(421-445), at a cytochrome b concentration of 230 M, were reduced by a small excess of ascorbate, the EPR signal of the Rieske ironsulfur cluster in the complement complex is essentially the same as that previously reported for the wild-type R. sphaeroides bc 1 complex (18,19), with resonance at g x ϭ 1.80, g y ϭ 1.90, and g z ϭ 2.02 (see Fig. 6A). The signatures of g y and g z of [2Fe-2S] cluster in the cytb⌬-(433-445), cytb⌬-(427-445), and cytb⌬-(425-445) mutant complexes are the same as those detected in the complement complex, but with signal amplitudes decreasing as the C-terminal extra fragment of cytochrome b shortens (see Fig. 6A). No EPR spectrum of the [2Fe-2S] cluster is detected in the cytb⌬-(421-445) mutant complex. These results are consistent with Western blot results showing that the amount of ISP in the bc 1 complex decreases as the C-terminal extra fragment of cytochrome b decreases and no ISP is detected in the complex that has cytochrome b lacking the entire C-terminal extra fragment.
In contrast to g y and g z , the g x signal of [2Fe2S] in mutant complexes of cytb⌬-(433-445), cytb⌬-(427-445), and cytb⌬-(425-445) mutant complexes change progressively from a relatively sharp peak with g ϭ 1.80 to a broadened peak with g ϭ 1.76 (see Fig. 6B). Whereas the g x signal from the cytb⌬-(433-445) mutant complex is quite sharp with g ϭ 1.80 peak, that from the cytb⌬-(425-445) mutant complex is broader with g ϭ 1.76 peak. These results indicate that the alteration of the microenvironments of the [2Fe-2S] cluster increases as the C-terminal extra fragment of cytochrome b decreases. This finding is somewhat surprising, because in the model structure of R. sphaeroides cytochrome bc 1 complex (5), constructed by using the coordinates of subunits from beef heart mitochondrial bc 1 complex, the Rieske iron-sulfur cluster is located at the head domain of ISP on the periplasmic side of the chromatophore membrane (positive side), whereas the C-terminal extra fragment of cytochrome b is located at the cytoplasmic side of this chromatophore membrane. Thus, the effect of shortening the C-terminal extra fragment of cytochrome b on the EPR signature of the Rieske [2F-2S] center would appear to be long range.
The line shape of the g x signature of [2Fe-2S] clusters is thought to be sensitive to the redox state of ubiquinone present in the Q o center (18 -23). The g x of bc 1 from wild-type R. sphaeroides is at g ϭ 1.80 when oxidized ubiquinone is present but shifts to 1.76 and becomes much broader when ubiquinol is present. In a study of the effect of extraction of ubiquinone, from chromatophore membranes, on the iron-sulfur cluster, Ding et al. (23) found that the g x signal became very broad and was located at ϳ1.765 upon deletion of ubiquinone from the R. capsulatus chromatophore membrane. Although the broadened, g x ϭ 1.76 resonance observed in the cytb⌬-(427-445) and cytb⌬-(425-445) mutant complexes resembles the "reduced state" or the "depleted state" spectrum, it is not because of changes the redox state of Q or a decrease of Q in the mutant complex, because no EPR spectrum of [2Fe-2S cluster] is detected in these two mutant complexes without treatment with ascorbate and the amount of Q in these two mutant complexes is the same as that in the complement complex. It should also be noted that upon complete reduction by addition of dithionite, the broadened, g x ϭ 1.76 signal observed in the mutant complexes remains unchanged, whereas the g x ϭ 1.80 signals becomes broadened and shifts to a g x ϭ 1.76 signal, FIG. 8. EPR spectra of cytochromes b H and b L in purified complement and mutant cytochrome bc 1 complexes. Sample preparations and instrument settings were the same as those described in the legend to Fig. 6, except that the microwave power used was 108.1 milliwatts and modulation amplitude was 20 G. similar to the changes observed in the complement complex (18,19).
The broadened, g x ϭ 1.76 EPR signal observed in the cytb⌬-(427-445) and cytb⌬-(425-445) mutant complexes is similar to the g x signal observed for the substitution of Leu for Phe-144 (F144L) in the cytochrome b from R. capsulatus (22) and of serine for Thr-160 in cytochrome b from R. sphaeroides (10). The F144L bc 1 complex in R. capsulatus and the T160S mutant complex in R. sphaeroides chromatophores were reported to have a decreased turnover rate with a broadened, redox stateinsensitive g x value at 1.765. It was suggested that these properties of the F144L and T160S complexes resulted from a reduced affinity for quinone and quinol at the Q o site of the mutated complex. One possibility, which could account for the decreased turnover rate of the cytb⌬-(427-445) and cytb⌬-(425-445) mutant complexes and their reduced state or the high field shift of the g x EPR signal, is that shortening the C-terminal extra fragment to less than 6 residues induces conformational changes at the Q o site, which raise the effective redox potential of bound ubiquinol beyond the optimal range for transfer to the [2Fe-2S] cluster.
Effect of the Mutations on Redox potentials and EPR Characteristics of Cytochrome b in the bc 1 Complex- Fig. 7 shows redox titration curves of cytochrome b in the complement and C-terminal truncated cytochrome b mutant complexes. Two redox components (b L and b H ) are resolved from each of the titration curves and their redox potentials were calculated (see Fig. 7 and Table II complement complex are Ϫ87 and 41 mV, respectively, similar to previously reported values (19). The redox potentials of b L and b H in the cytb⌬-(433-445) mutant complex are Ϫ73 and 49 mV, respectively, indicating that deleting 13 residues has little effect on redox potentials of cytochrome b. In contrast, the redox potential of b L in mutant complexes of cytb⌬-(427-445), cytb⌬-(425-445), and cytb⌬-(421-445) decreases by 28, 58, and 73 mV, respectively, and that of b H decreases by 16, 82, and 111 mV, respectively, compared with counterparts in the complement complex. Thus residues 421-432 in the C-terminal extra fragment are essential for maintaining redox potentials of b cytochromes. The effect is larger in b H than in b L . It should be noted that the amplitudes of the reductive and oxidative titrations, shown in Figs. 7 and 9, match so well because no protein denaturation occurs during the titrations, as a freshly thawed sample from the same batch of a given bc 1 complex was used for each trial of reductive and oxidative titrations. Fig. 8 shows EPR spectra of the b cytochromes from the complement and mutant complexes, taken after the samples were reduced with sodium ascorbate to eliminate the overlapping signal from cytochrome c 1 . The complement complex has features at g ϭ 3.53 and g ϭ 3.77 previously assigned to cytochrome b H and b L , respectively, in the wild-type bc 1 complex (18). The g ϭ 4.3 signal is thought to be nonspecific bound iron(III). Similar EPR spectra were reported for the mitochon- Effect of the C-terminal Extra Fragment of Cytochrome b on Cytochrome c 1 - Fig. 9 shows redox titration curves of cytochrome c 1 in complement and mutant complexes. The redox potentials of cytochrome c 1 in these bc 1 complexes, calculated to be 237, 227, 165, 170, and 187 mV, respectively (see Fig. 9 and Table II), indicate that the redox potential of cytochrome c 1 is affected by deleting more than 13 residues from the C terminus of cytochrome b. In the proposed structural model of the R. sphaeroides cytochrome bc 1 complex, the C-terminal extra fragment of cytochrome b is located on the cytoplasmic side of the chromatophore membrane where N-terminal portions of cytochrome c 1 , ISP, and cytochrome b also reside. Heme c 1 is located on the periplasmic side of the chromatophore membrane. Perhaps interactions between the first 12 residues from the N-terminal end of the C-terminal extra fragment of cytochrome b and residues of cytochrome b, cytochrome c 1 , or ISP on the cytoplasmic side stabilize the structure of cytochrome c 1 and maintains its redox potential. Thus, the effect on the redox potential of cytochrome c 1 by the C-terminal extra fragment of cytochrome b is a long range effect. It has been previously reported that redox midpoint potentials of cytochrome c 1 are affected by substitutions in cytochrome b (27).
It should be noted again that the C-terminal extra fragment of cytochrome b is located on the cytoplasmic side of the chromatophore membrane, yet its deletion affects redox components, such as heme b H , iron-sulfur cluster, and heme c 1 , located on the periplasmic (opposite) side of the chromatophore membrane. Therefore, these effects cannot be explained by direct interaction between the extra fragment of cytochrome b and the redox components or their ligands or their vicinity peptides. These long range effects are probably because of globular changes in the deleted mutant complex. That deleted cytochrome b proteins have electrophoretic mobilities greater than those of comparable molecular mass, in SDS-PAGE, suggests that the deleted cytochrome b proteins are more globular than the unaltered protein.