Amino Acid Substitutions in the Non-Ordered Ω-Loop 70–85 Affect Electron Transfer Function and Secondary Structure of Mitochondrial Cytochrome c

Round 1

Reviewer 1 Report

Please see attached file.

Comments for author File: Comments.docx

Author Response

Thank you for the carefully reading our manuscript and valuable comments. With your permission, we will give an answer to your review step by step.

1. How is the heme content in the recombinant cytochrome c determined? Some of the defects in cytochrome c function could be explained by decreased heme association in the recombinant protein preparation.

 

The association of heme with the polypeptide part of the obtained mutant cytochromes c can be judged from the absorption spectra in the visible range.

Figure shows spectra of mutant variants of cytochrome c T78N/K79Y/M80I/I81M/G84N (similar spectra were obtained for the rest of the mutants): the sample after first step of purification (1), after oxidation with an equimolar solution of potassium ferricyanide and second step of purification (2), after the three-step dialysis against 10 mM carbonate buffer (3) and after lyophilization and dissolving in 10 mM phosphate buffer (4). One can see that after purification the cytochrome sample is a mixture of oxidized and reduced forms (Amax 411 nm, two peaks in the region of 530-550 nm, which is typical for mixtured forms, as well as for reduced cytochrome c). Absorption spectra of the cytochrome samples after the oxidation with potassium ferricyanide and after the three-fold dialysis for 20 hours are the same and resemble fully oxidized cytochrome c (Amax 409 nm, one peak in the region of 530-540 nm that is typical for fully oxidized cytochrome c).

For both oxidized and reduced forms of cytochrome c in a fully folded state, i.e. with associated heme, the positions of the maxima are characteristic:

Amax 409 nm – oxidized folded cytochrome c

Amax 415 nm – reduced folded cytochrome c

For partially unfolded cytochrome c in oxidized and reduced forms, different positions of the maxima are characteristic:

Amax 406 nm - oxidized unfolded cytochrome c

Amax 418 nm - reduced unfolded cytochrome c

After oxidation with K₃[Fe (CN)₆], we carried out the second stage of purification, then we analyzed each fraction spectrophotometrically. For further work, we selected only those fractions of mutant cytochromes c that had an absorption maximum at Amax 409 nm, which corresponds to normal folding of the completely oxidized protein and, therefore, normal association of heme with the protein part.

Please, see the attachment

Figure. Spectra of the cytochrome c T78N/K79Y/M80I/I81M/G84N: 1 - a mixture of oxidized and reduced cytochrome c (Amax 411 nm) in the sample of protein after first step of purification; 2 – the same sample of cytochrome c T78N/K79Y/M80I/I81M/G84N after the oxidation by K₃[Fe(CN)₆] and second step of purification; 3 – the same sample after dialysis against 10 mM (NH₄)₂CO₃, pH 7.9; 4 - the same sample after lyophilisation and dissolution in 10 mM phosphate buffer, pH 7.0 (Amax 409 nm for 2-4).

 

2. The authors state the Amax in figure 1 A and B are significantly lower than wild-type. But are they significantly different from each other? Can the authors calculate the significance of these Amax values to determine differences between mutants?

 

Statistical analysis did not show any significant difference in Amax values between CytC mutant forms.

 

3. In figure 1 panel A and B are not indicated.

 

Thanks for the comments. Сorrected.

 

Some general points that may be included in the article.

4. Are the amino acids that are mutated conserved in different eukaryotic homologs of cytochrome c?

 

The amino acid sequence of the Ω-loop 70-85 of cytochrome c is highly conserved. According to the alignment data, residues 70–85 completely coincide in higher eukaryotes, for example, in the case of human, horse, and mouse cytochrome c. In comparison with them, cytochrome c from Drosophila has one substitution, V83A, and cytochrome c from Baker's yeast has two substitutions I81A and V83G.

 

5. Would these mutations impact on other functions of cytochrome c – cardiolipin association, peroxidase activity, apoptosome activity? Could there be some recognition of these other cytochrome c functions be included.

Thanks for this comment. Yes, indeed, a very interesting question is how these mutations in the Ω-loop 70-85 of cytochrome c affect its ability to bind to cardiolipin and peroxidase activity. We have recently completed a new experimental work on this topic. However, since we have received quite a lot of experimental data, we are currently writing a separate article, which will be a continuation of this one. In addition, in the nearest future, we plan to investigate the proapoptotic activity of these mutant variants in one or another system.

Author Response File: Author Response.pdf

Reviewer 2 Report

Review:

In this article, the authors analyzed the structural and functional consequences of several amino acid substitutions at the Ω-loop 70-85 of horse cytochrome c. For this, they choose the proper structural techniques (CD, NMR, IR) and use them correctly. However, there are a number of issues regarding data interpretation that they must solve. My main concerns come from the fact that the four cytochrome c mutants studied could have serious heme coordination problems, which could explain the observed differences in terms of electron transfer activity. In consequence, their conclusions regarding conformational mobility of the Ω-loop 70-85 are not properly supported by their results.

Major comments:

1. The paragraph from lines 57-60 of the introduction section is confused: “Hence, a cytochrome c molecule was found to contain the only P76GTKMIFA83 57 site that was characterized by reduced correlation between amino acid residues (with in-58 creased conformational mobility) and was located within the Ω-loop region (residues 70–85)”. It should be rewritten.
2. The authors designed several cytochrome c mutants within loop Ω comprising 70-85 residues. However, they did not mention why they design those four mutants. On what have they been based to make these substitutions?
3. I am struck by the fact that the authors decided to mutate the M80 residue, essential in heme coordination, as well as K79, which can undergo alkaline transition. These mutations alone generate significant changes in the structure and function of the cytochrome c, which were ignored by authors.
4. In the methods section, line 128, it is mentioned that 1 mM NaN₃ was used during cytochrome c Taking into account that NaN₃ is a potent inhibitor of the cytochrome c oxidase, this compound should be avoided in all purification steps. How have the authors ruled out a possible effect of a residual NaN₃ in the cytochrome c samples?
5. The cytochrome c-oxidase activity was measured in a reaction containing 10 mM ascorbic acid, which is a well-known cytochrome c reducing compound. However, the authors said that reaction contained oxidized cytochrome c, but that is not possible due to the presence of ascorbic acid. In fact, the cytochrome c should be in its reduced state to transfer one electron to complex IV.
6. The authors should explain to readers why they chose to measure a combination of complex II and complex III activities, rather recording complex III activity.
7. There is a technical issue in the 1H-NMR spectra of cytochrome c and it is that they were recorded in deionized Н₂О. Why the authors choose water instead of any buffer? There is plenty of examples of NMR spectra of cytochrome c in the literature recoded in many different buffers. I would suggest to use the 10 mM sodium phosphate buffer, for instance.
8. The kinetic parameters Km and Аmax were expressed as mean ± SEM. The authors should use SD.
9. The succinate:cytochrome c-reductase activity is explained in the second paragraph of results, but this activity was already mentioned in the first paragraph.
10. Why the authors give more importance to a decrease in the cytochrome c-oxidase (complex IV) activity, rather than the ubiquinol-cytochrome c oxidorreductase activity (complex III)? The cytochrome c function is between them, so it is equally important the first and the second.
11. Do the authors find any mutant without a decrease in succinate:cytochrome c-reductase activity? It is strange that any mutant tested yielded that effects.
12. The authors attribute the changes in the succinate:cytochrome c-reductase activity in terms of the ability (or decreased ability) of cytochrome c mutants to form active complexes with the corresponding redox partner, e. cytochrome c₁ of complex III. However, they did not take into account the possibility that the redox potential of the Cc mutants could change, which could explain the observed differences in activity. Without the redox potential values of mutants, they cannot discard this possible effect. This is especially important for the T78N/K79Y/M80I/I81M/F82N mutant, which contains the M80I substitution that yield a loss in the heme coordination, essential in the electron transfer function of the protein. Moreover, to demonstrate that the decrease in such activity in all Cc mutants is due to a decreased ability to form complexes with cytochrome c₁, the authors should carry out any binding experiment.
13. Regarding this phrase in results (lines 228-232): “At the same time, the decrease of Km values of the succinate:cytochrome c-reductase reaction was significant for the mutant cytochrome T78N/K79Y/M80I/I81M/F82N (about 2 229 times). Such decrease may indicate that the observed subside in the reaction rate of the reduction of this mutant is caused by worsening its ability to form active complexes with the corresponding redox partner from the ubiquinol:cytochrome c-oxidoreductase.” The authors wrongly claimed that Km value for that mutant decrease and therefore this Cc mutant binds worst to ubiquinol:cytochrome c-oxidoreductase. Please note that Km is defined as the substrate concentration at which the reaction rate is half of its maximal value, and it indicates the affinity of an enzyme for a given substrate: the lower the Km value, the higher the affinity.
14. In table 1 (secondary structure fractions derived from CD spectra) the standard errors are missing.
15. The authors said (266-270): “Hence, CD spectroscopy studies of the secondary structure of mutant cytochrome c variants carrying T78S/K79P, I81Y/A83Y/G84N, T78N/K79Y/M80I/I81M/F82N, and 267 P76I/G77L/I81L/F82L substitutions demonstrated that the introduction of these mutations increases the contents of fractions having an ordered structure (in particular, β-sheets and β-turns), which seems to reduce the conformational mobility.” The authors observe that the helical content of mutants decreased, while the b-sheet fraction increased, they did not observe changes in the disordered fraction. They cannot conclude that mutants have an increase in the ordered fractions.
16. Another important technical issue: As they mutate M80 and K79 residues, they must record visible spectra of mutants to show the coordination of the heme group. Otherwise, all changes observed in such mutants regarding decreased electron transfer activities can be attributed to a loss of the heme coordination.
17. Regarding ¹H NMR spectra of cytochrome c, why the authors perform those analyses with the protein in its oxidized state? Did the authors check the reduced state?
18. Although the authors showed that mutants showed dispersion of chemical shifts of amides, Ha and Ca protons which are indicative of secondary structure, the proper heme coordination must to be checked by NMR (and visible CD). This is important since such coordination is essential for Cc electron transfer function.
19. It is very striking that the authors found that that the intensity of the Met80 C_βH proton signal (12.85 ppm) was significantly reduced or even not detected in the spectra of all the mutants, even those with the M80 intact. How can they rule out that this is due to a possible loss of the heme coordination?
20. The authors detect a shift in the signal attributed to the meso-β proton (at -0.79 ppm) T78S/K79P and 306 T78N/K79Y/M80I/I81M/G84N mutants. Again, how can the authors rule out that this effect is due to a possible loss of the heme coordination? In these two mutants, K79 or M80 were substituted, with the consequent implications for heme coordination. Determination of pH for alkaline transition of Cc mutants would be elucidating.

Minor comments:

• Line 80: “cytochromes c1” should be replaced by “cytochrome c₁”.
• Lines 130 and 135: “SDS-PAAG electrophoresis” should be replaced by “SDS-PAGE electrophoresis”.
• Line 169: the TMPD acronym should be defined. By the way, what is this used for?
• Line 214: the acronym ETC should be defined as electron transport chain.
• Lines 222-223: The phase “Figure 2 shows the Amax and Km values for…” should be replaced by “Figure 1 shows the Amax and Km values for…”.
• Line 303-304: “Here we present description of 1H-NMR spectra (Fig.4) that directly characterize 303 heme molecule in oxidized cytochrome c [29].” Fig. 4 shows IR spectra.
• Reference 11 is not properly formatted.

Author Response

We thank the Reviewer for the careful reading of the manuscript and valuable comments. With Reviewer’s permission, we give answers to comments step by step.

Major comments:

1. The paragraph from lines 57-60 of the introduction section is confused: “Hence, a cytochrome c molecule was found to contain the only P76GTKMIFA83 site that was characterized by reduced correlation between amino acid residues (with increased conformational mobility) and was located within the Ω-loop region (residues 70–85)”. It should be rewritten.

 

We thank the Reviewer for this comment. We rewrote the paragraph.

 

2. The authors designed several cytochrome c mutants within loop Ω comprising 70-85 residues. However, they did not mention why they design those four mutants. On what have they been based to make these substitutions?

 

The goals and principles on the basis of which we carried out the construction of the studied cytochrome с mutant forms we discussed in our previous publication [20. Chertkova, R.V.; Brazhe, N. A.; Bryantseva, T.V.; Nekrasov, A.N .; Dolgikh D.A.; Yusipovich, A.I.; Sosnovtseva, O .; Maximov, G.V.; Rubin, A.B .; Kirpichnikov, M. New insight into the mechanism of mitochondrial cytochrome c function. PLoS One 2017, 12 (5), e0178280]. We refer to this publication in our manuscript.

          The cytochrome c structure was analyzed by an ANIS-method (Analysis of Informational Structure), which made it possible to detect a single site with high conformational mobility (ADD- site), corresponding to the loop section P⁷⁶GTKMIFA⁸³. On the basis of data obtained by ANIS method we designed amino acid substitutions in the region of 70-85 residues to obtain a region with the low conformational mobility (ADD+ site). We supposed that the presence of ADD+ site should prevent the conformational changes of cytochrome c heme which are important for the electron acceptance and donation. Indeed, we observed the decrease in the electron transport activity of cytC mutant forms possessing ADD+site with low conformational mobility. We should note, that the introduction of single mutations did not lead to the formation of the ADD+ site, therefore they were not considered for the following study.

Briefly, ANIS is based on the use of the primary protein amino acid sequence to reveal a hierarchy of the ELements of Information Structure (ELIS). ELIS, or information units, correspond to the variable length sites with an increased density of structural information. The mathematical treatment of E. coli proteome was performed and a subset of information units with a high level of local interactions (structure-forming unit, SF) and a low level of local interactions (non-structure-forming unit, NSF) was obtained. The presence or absence of the high level of local interactions was controlled by the molecular dynamics (MD) method. For the information units of their two subsets, MD trajectories were obtained, the analysis of which showed the presence of an advantageous conformation for the information units from the SF subset and the absence of that for the information units from the NSF subset. Information units included in a subset of SF were priority for replacement in the analyzed area P⁷⁶GTKMIFA⁸³.

          In addition, statistical analysis was made to estimate the probability of occurrence of individual amino acid residues in subsets of SF and NSF information units (Fig. 1). Analyzed subsets of SF and NSF information units were obtained from the E. coli proteome and from the set of non-homologous protein sequences (Fig. 1). Based on the results of the analysis, the frequency of occurrence of different amino acids in a subset of SF of the E. coli proteome decreases in the series: G> A> V> L> S> E> I> R> T> K> D> P> F> N> Q> Y> H> C> M> W. Based on the results of the analysis of the amino acid composition in a subset of SF in non-homologous proteins, the frequency of occurrence decreases in the series: G> V> A> L> S> E> K> I> R> T> D> P> N> F> Q> Y> C> > H> M> W. We can see that the obtained sequences have the pronounced similarity. Since ADD+ sites consist of closely located information units of SF, rules for the desired point mutationsare desirable for the formation of ADD+ sites.

Please see the attachment

Figure 1. Aminoacid frequency at the E.coli proteome (A, B) and nonhomologous proteins (C, D): A, C – nonstructure-forming unit (NSF); B, D – structure-forming unit (SF).

 

When selecting amino acid residues for mutant variants of cytochrome c, the residues Cys, Trp, Glu, His and Met were excluded. The Cys residue was excluded because of its propensity to form disulfide bonds in the molecule changing the tertiary structure of the mutant variants Trp. His and Met amino acids were excluded from the number of possible variants for cytC mutagenesis in order to avoid the formation of atypical coordination bonds in mutant cytC forms. Also an important condition for the design of mutant variants was the minimization of amino acid substitutions in order to form an ADD+ site by introducing the smallest possible number of site-directed mutations in the P76GTKMIFA83 area.

Considering all listed conditions and the data on the preferential amino acid composition of SF and NSF, the following variants of mutations in the region P76GTKMIFA83 were proposed: I81Y/A83Y/G84N, T78N/K79Y/M80I/I81M/F82N, T78S/K79P, K79V/I81L/F82R, T78A/K79A/I81A/F82T/T89A, I81L/F82S/A83S/G84A, I75G/G77R/T78I, T78A/K79A/I81A/F82T.

 

3. I am struck by the fact that the authors decided to mutate the M80 residue, essential in heme coordination, as well as K79, which can undergo alkaline transition. These mutations alone generate significant changes in the structure and function of the cytochrome c, which were ignored by authors.

 

In the answer to the question №2 we discussed the principles and tasks of constructing cytochrome c mutant forms, which underlie our study. Among many mutant variants that satisfy the design goals, there was the mutant in which the residues at positions 80 and 81 were interchanged. It seemed to us interesting to try to obtain such a mutant variant and, in the case of a successful attempt, to analyze its properties, both electron transport and changes in conformational and structural properties.

 

4. In the methods section, line 128, it is mentioned that 1 mM NaN₃ was used during cytochrome c Taking into account that NaN3 is a potent inhibitor of the cytochrome c oxidase, this compound should be avoided in all purification steps. How have the authors ruled out a possible effect of a residual NaN3 in the cytochrome c samples?

 

After oxidation with potassium ferricyanide, cytochrome c samples are dialyzed three times against 10 mM ammonium carbonate buffer, pH 7.9 (line 138-139 in the methods section). At each of three step, we carried out dialysis against ~ 400-fold volume of ammonium carbonate buffer, so the total decrease in the concentration of NaN₃ was about 400³, thus, the residual concentration of NaN₃ in the cytochrome c solution could be about 1.5 x 10^-11 M. We carried out experiments on measuring the cytochrome c-oxidase activity in the presence of commercial cytochrome c and cytochrome c (WT) prepared by us according to the described scheme. As a result we did not see any differences in the activity of cytochrome c oxidase.

 

5. The cytochrome c-oxidase activity was measured in a reaction containing 10 mM ascorbic acid, which is a well-known cytochrome c reducing compound. However, the authors said that reaction contained oxidized cytochrome c, but that is not possible due to the presence of ascorbic acid. In fact, the cytochrome c should be in its reduced state to transfer one electron to complex IV.

 

Yes, we agree with the Reviewer that after the initiation of the reaction, cytochrome c was actually in a reduced state. However, since at the end of the purification of all mutant forms, as well as WT cytochrome, all samples were oxidized with K₃[Fe(CN)₆], they were added to the reaction mixture in the oxidized state. Then the reaction was initiated by adding ascorbic acid to the reaction medium. The ascorbic acid acted as an electron donor for tetramethyl-p-phenylene diamine (TMPD), which, in its turn, transferred electrons to oxidized cytochrome c [Ferguson-Miller S, Brautigan DL, Margoliash E. Correlation of the kinetics of electron transfer activity of various eukaryotic cytochromes c with binding to mitochondrial cytochrome c oxidase. J. Biol. Chem. 1976; 251:1104-1115.]. The reduced cytochrome c then transferred electrons to complex IV that reduced oxygen to water.

 

6. The authors should explain to readers why they chose to measure a combination of complex II and complex III activities, rather recording complex III activity.

 

In our studies complex III activities of rat liver mitoplasts were measured according to the well-known method described in [Vinogradov, A.; Leikin, Yu.; Lipskaya T. Mitochondrial biochemistry. Bioenergetics. In Manual of practical study to animal biochemistry. Lomonosov Moscow State University publishers, Russia, 1977; pp. 19-22.], this link is in the Materials and methods, section 2.14. This techniques was successfully used previously to test a series of cytochrome forms with Lys residue replacements in the universal binding site [Chertkova, R.V.; Brazhe, N.A.; Bryantseva, T.V.; Nekrasov, A.N.; Dolgikh, D.A.; Yusipovich, A.I.; Sosnovtseva, O.; Maksimov, G.V.; Rubin, A.B.; Kirpichnikov, M.P. New insight into the mechanism of mitochondrial cytochrome c function. PLoS One 2017, 12(5), e0178280; Pepelina TYu, Chertkova RV, Dolgikh DA, Kirpichnikov MP, Bioorg Khim. 2010; 36 (1): 98-104 ; Pepelina TY, Chertkova RV, Ostroverkhova TV, Dolgikh DA, Kirpichnikov MP, Grivennikova VG, Vinogradov AD, Biochemistry 2009 (Mosc) 74 (6): 625-32.]. In this method, potassium succinate served as an electron donor, transferring electrons to the respiratory chain to complex II - succinate dehydrogenase - and then to ubiquinone, complex III, and exogenous cytochrome c. Because of the need to first transfer an electron to ubiquinone, succinate: cytochrome c-reductase activity is measured in this system.

For clarification to readers, we have added the following sentence to the Results and Discussion section, 3.1.:

«In succinate: cytochrome c - reductase reaction, potassium succinate served as an electron donor, transferring electrons to the respiratory chain to complex II - succinate dehydrogenase (SDH) - and then to ubiquinone, complex III, and exogenous cytochrome c.»

 

7. There is a technical issue in the 1H-NMR spectra of cytochrome c and it is that they were recorded in deionized Н2О. Why the authors choose water instead of any buffer? There is plenty of examples of NMR spectra of cytochrome c in the literature recoded in many different buffers. I would suggest to use the 10 mM sodium phosphate buffer, for instance.

 

The main aim to use buffer solutions with cytC is the necessity to maintain pH~7. In our experiments obtained cytC WT and mutant forms were lyophilized from the ammonium carbonate buffer and were dissolved in mQ H₂O with pH 5.5. After the cytC dissolution the pH of the cytC solution was ~ 6.8 since there always was some amount of unevaporated ammonium carbonate (20-30% from the initial amount). So, the pH value of the studied cytC solution was in the physiological region.

 

8. The kinetic parameters Km and Аmax were expressed as mean ± SEM. The authors should use SD.

We thank the Reviewer for this comment. Revised.

 

9. The succinate:cytochrome c-reductase activity is explained in the second paragraph of results, but this activity was already mentioned in the first paragraph.

This Reviewer’s comment is not clear and needs clarification. The succinate: cytochrome c-reductase activity is only discussed in section 3.1. of Results and discussion and is not mentioned in section 3.2.

 

10. Why the authors give more importance to a decrease in the cytochrome c-oxidase (complex IV) activity, rather than the ubiquinol-cytochrome c oxidorreductase activity (complex III)? The cytochrome c function is between them, so it is equally important the first and the second.

We completely agree with the Reviewer that both cytochrome c ability to accept an electron from ubiquinol:cytochrome c-oxidoreductase and the ability to transfer an electron to cytochrome c-oxidase are equally important for ETC operation. Perhaps, we are discussing a little more the decrease in cytochrome c-oxidase activity of cytochrome variants P76I/G77L/I81L/F82L due to the fact that all other variants failed to achieve such a decrease in this activity.

 

11. Do the authors find any mutant without a decrease in succinate:cytochrome c-reductase activity? It is strange that any mutant tested yielded that effects.

 

In our previous paper [20. Chertkova, R.V .; Brazhe, N. A .; Bryantseva, T.V .; Nekrasov, A.N .; Dolgikh D.A .; Yusipovich, A.I .; Sosnovtseva, O .; Maximov, G.V .; Rubin, A.B .; Kirpichnikov, M. New understanding of the mechanism of mitochondrial cytochrome c function. PLoS One 2017, 12 (5), e0178280] we describe in detail the properties of a number of mutant variants with a decrease in succinate: cytochrome c-reductase activity to varying degrees - from ~ 50% to ~ 3%. The subject of our study was the conformational changes in hemoporphyrin and the structural features of the protein part in variants with the most reduced succinate: cytochrome c-reductase and cytochrome c-oxidase activities. Therefore, from the entire series, we selected mutant variants T78S/K79P, I81Y/A83Y/G84N, T78N/K79Y/M80I/I81M/G84N, P76I/G77L/I81L/F82L.

 

12. The authors attribute the changes in the succinate:cytochrome c-reductase activity in terms of the ability (or decreased ability) of cytochrome c mutants to form active complexes with the corresponding redox partner, e. cytochrome c1 of complex III. However, they did not take into account the possibility that the redox potential of the Cc mutants could change, which could explain the observed differences in activity. Without the redox potential values of mutants, they cannot discard this possible effect. This is especially important for the T78N/K79Y/M80I/I81M/F82N mutant, which contains the M80I substitution that yield a loss in the heme coordination, essential in the electron transfer function of the protein. Moreover, to demonstrate that the decrease in such activity in all Cc mutants is due to a decreased ability to form complexes with cytochrome c1, the authors should carry out any binding experiment.

 

We thank the Reviewer for this comment. Yes, it would be extremely interesting to investigate the ability of mutant cytochromes to form a complex with cytochrome c1. As far as possible, we will deal with this, however, such a study is a separate and voluminous work. Within the framework of this study, such a task was not set, and, interpreting our results, we only hypothesized that possibly the change in the succinate: cytochrome c-reductase activity of mitoplasts is associated with the decrease in the ability of cytochrome c mutants to form a complex with the corresponding redox partners.

Regarding the Reviewer's comment on "... T78N/K79Y/M80I/I81M/F82N mutant, which contains the M80I substitution that yield a loss in the heme coordination, essential in the electron transfer function of the protein.": at all steps of isolation and purification of mutant variants we monitored the obtained proteins using spectrophotometry in the visible range. For further studies, we selected only those fractions, the spectrum of which corresponded to the fully folded cytochrome c. For more information, see below Answer 16.

Regarding the Reviewer's comment on "... they did not take into account the possibility that the redox potential of the Cc mutants could change, which could explain the observed differences in activity.": indeed, we have not yet investigated the redox potentials of our mutant variants. But, in our earlier works we tested the redox potential of a mutant variant of cytochrome c with eight substitutions of charged amino acid residues. In the presence of this variant, both succinate: cytochrome c-reductase and cytochrome c-oxidase activity of mitoplasts were suppressed almost completely: up to ~ 3% and <1% of the residual activities of reductase and oxidase, respectively [22. Pepelina TY, Chertkova RV, Ostroverkhova TV, Dolgikh DA, Kirpichnikov MP, Grivennikova VG, Vinogradov AD, Biochemistry 2009 (Mosc); 74 (6): 625-32.]. We have shown that even with such a significant change in the total cytochrome charge, the redox potential does not change dramatically: there was a slight decrease in Em at 27 mV compared to WT cytochrome (bioRxiv preprint first posted online February 11, 2015; doi: http://dx.doi.org/10.1101/015131). Thus, taking into account the fact that in this paper all the changes introduced in cytochrome either changed the total protein charge by not more than 1, or did not have any effect at all, we believe that the redox potentials of the mutant variants did not change significantly.

 

13. Regarding this phrase in results (lines 228-232): “At the same time, the decrease of Km values of the succinate:cytochrome c-reductase reaction was significant for the mutant cytochrome T78N/K79Y/M80I/I81M/F82N (about 2 times). Such decrease may indicate that the observed subside in the reaction rate of the reduction of this mutant is caused by worsening its ability to form active complexes with the corresponding redox partner from the ubiquinol:cytochrome c-oxidoreductase.” The authors wrongly claimed that Km value for that mutant decrease and therefore this Cc mutant binds worst to ubiquinol:cytochrome c-oxidoreductase. Please note that Km is defined as the substrate concentration at which the reaction rate is half of its maximal value, and it indicates the affinity of an enzyme for a given substrate: the lower the Km value, the higher the affinity.

 

We thank the Reviewer for this comment. Indeed, this proposal and conclusion is confusing. We rewrote it as follows:

“Such decrease may indicate a change in its ability to form active complexes with the corresponding redox partner from ubiquinol:cytochrome c-oxidoreductase. It should be noted that such a decrease is combined with the observed decrease in the rate of the reduction reaction of this cytochrome c mutant.”

 

14. In table 1 (secondary structure fractions derived from CD spectra) the standard errors are missing.

 

The CD spectra of cytochrome c and its mutant variants were recorded in three sequential measurements, an average of which was calculated by the device software. The intermediate spectra were not stored in the instrument database and could not be extrapolated, which is a common practice for CD measuring. The secondary structure analysis was performed on the basis of the averaged CD spectra using CONTILL software. Briefly, the protocol involves a comparison of the resulting spectra with reference spectra of proteins with known crystal structures for which the true percentages of secondary structures can be determined. In this regard, the common way to represent the accuracy of a CD analysis is to provide a root-mean-square (RMS) deviation and/or normalized root-mean-square displacement, RMSE/data range, as a measure of error. NRMSD refers to the root-mean-square difference between the fitted curve and the actual (measured) data. It is usual to regard a value of NRMSD above 0.25 as constituting an error in the analysis procedure; in practice values of less than 0.1, or ideally 0.05 or lower should be aimed for [Kelly S. M., Jess T. J., Price N. C. (2005). How to study proteins by circular dichroism. Biochim. Biophys. Acta (BBA) - Proteins Proteomics 1751, 119–139. 10.1016/j.bbapap.2005.06.005]. Taken together, we trust the performance of the analysis was presented correctly and in accordance with accepted practices.

 

15. The authors said (266-270): “Hence, CD spectroscopy studies of the secondary structure of mutant cytochrome c variants carrying T78S/K79P, I81Y/A83Y/G84N, T78N/K79Y/M80I/I81M/F82N, and 267 P76I/G77L/I81L/F82L substitutions demonstrated that the introduction of these mutations increases the contents of fractions having an ordered structure (in particular, β-sheets and β-turns), which seems to reduce the conformational mobility.” The authors observe that the helical content of mutants decreased, while the b-sheet fraction increased, they did not observe changes in the disordered fraction. They cannot conclude that mutants have an increase in the ordered fractions.

 

We thank the Reviewer for this comment. Indeed, it is not strictly correct to refer to the 1% change in the disordered fraction as an ‘increase in the contents of fractions having an ordered structure’. In this regard, the sentence was rewritten as follows:

“Hence, CD spectroscopy studies of the secondary structure of mutant cytochrome c variants carrying T78S/K79P, I81Y/A83Y/G84N, T78N/K79Y/M80I/I81M/F82N, and P76I/G77L/I81L/F82L substitutions demonstrated that the introduction of these mutations altered the contents of fractions with different types of structures (in particular, the fraction of α-helix increases while β-layer fraction decreases), which seem to change the conformational mobility.”

 

16. Another important technical issue: As they mutate M80 and K79 residues, they must record visible spectra of mutants to show the coordination of the heme group. Otherwise, all changes observed in such mutants regarding decreased electron transfer activities can be attributed to a loss of the heme coordination.

 

The coordination of the heme group in obtained mutant cytochromes c can be judged from the absorption spectra in the visible range.

Figure 2 (below) shows spectra of mutant variants of cytochrome c T78N/K79Y/M80I/I81M/G84N (similar spectra were obtained for the rest of the mutants): the sample after first step of purification (1), after oxidation with an equimolar solution of potassium ferricyanide and second step of purification (2), after the three-step dialysis against 10 mM carbonate buffer (3) and after lyophilization and dissolving in 10 mM phosphate buffer (4). One can see that after purification the cytochrome sample is a mixture of oxidized and reduced forms (Amax 411 nm, two peaks in the region of 530-550 nm, which is typical for mixtured forms, as well as for reduced cytochrome c). Absorption spectra of the cytochrome samples after the oxidation with potassium ferricyanide and after the three-fold dialysis for 20 hours are the same and resemble fully oxidized cytochrome c (Amax 409 nm, one peak in the region of 530-540 nm that is typical for fully oxidized cytochrome c).

For both oxidized and reduced forms of cytochrome c in a fully folded state, i.e. with associated heme, the positions of the maxima are characteristic:

Amax 409 nm – oxidized folded cytochrome c

Amax 415 nm – reduced folded cytochrome c

For partially unfolded cytochrome c in oxidized and reduced forms, different positions of the maxima are characteristic:

Amax 406 nm - oxidized unfolded cytochrome c

Amax 418 nm - reduced unfolded cytochrome c

After oxidation with K₃[Fe(CN)₆], we carried out the second stage of purification, then we analyzed each fraction spectrophotometrically. For further work, we selected only those fractions of mutant cytochromes c that had an absorption maximum at Amax 409 nm, which corresponds to normal folding of the completely oxidized protein and, therefore, normal association of heme with the protein part.

Please see the attachment

Figure 2. Spectra of the cytochrome c T78N/K79Y/M80I/I81M/G84N: 1 - a mixture of oxidized and reduced cytochrome c (Amax 411 nm) in the sample of protein after first step of purification; 2 – the same sample of cytochrome c T78N/K79Y/M80I/I81M/G84N after the oxidation by K₃[Fe(CN)₆] and second step of purification; 3 – the same sample after dialysis against 10 mM (NH₄)₂CO₃, pH 7.9; 4 - the same sample after lyophilisation and dissolution in 10 mM phosphate buffer, pH 7.0 (Amax 409 nm for 2-4).

         

We have been investigating various cytochrome c mutants for many years and each sample (including WT cytochrome c) is monitored using visible spectrophotometry. However, we believe that spectra similar to the above are technical information and therefore do not include them in the our articles.

 

17. Regarding ¹H NMR spectra of cytochrome c, why the authors perform those analyses with the protein in its oxidized state? Did the authors check the reduced state?

 

Oxidized cytochrome c is the most functionally important form in various processes in which cytochrome c is involved in vivo. Therefore, in our research, we focused our attention on this particular form. In addition, the oxidized form of cytochrome c is sufficiently stable in the absence of strong reducing agents there may be quite a long time. On the contrary, the reduced state of cytochrome c is short-lived and highly unstable; therefore, obtaining ¹H NMR spectra of cytochrome c in the reduced state is associated with technical difficulties.

18. Although the authors showed that mutants showed dispersion of chemical shifts of amides, Ha and Ca protons which are indicative of secondary structure, the proper heme coordination must to be checked by NMR (and visible CD). This is important since such coordination is essential for Cc electron transfer function.

 

We thank the Reviewer for the suggestion and we agree that measuring NMR and CD in the visible range will add additional information to the heme coordination but unfortunately, this suggestion can not be implemented immediately as both NMR and visible CD recording requires a substantial amount of protein which is not in hand now. We believe that heme coordination may be a prospective topic of additional research we would be happy to explore later on in our laboratory. It is also important to mention, that in our previous work (Chertkova et al., PloS One, 2017, 12, e0178280 “New insight into the mechanism of mitochondrial cytochrome c function”) we applied surface-enhanced Raman spectroscopy  to study heme conformation of cytC mutants used in the current manuscript and we did not observe spectral changes attributed to the severe changes in heme coordination or to the cleavage of the coordination bonds between Fe atom and the protein, though we found increased probability of the ruffled heme conformation in mutant cytC forms comparing to WT cytC.

19. It is very striking that the authors found that that the intensity of the Met80 CβH proton signal (12.85 ppm) was significantly reduced or even not detected in the spectra of all the mutants, even those with the M80 intact. How can they rule out that this is due to a possible loss of the heme coordination?

 

At all steps of isolation and purification of mutant variants we monitored the obtained proteins using spectrophotometry in the visible range. For further studies, we selected only those fractions, the spectrum of which corresponded to the fully folded cytochrome c (for more information, see below Answer 16). For this reason, we believe that the coordination of the heme is preserved. We associated the observed features of the NMR spectra, which Reviewer points out, with conformational changes in hemoporphyrin and its microenvironment. Besides, as we mentioned in the previous answer, data of SERS measurements did not demonstrated changes in Raman spectra of mutant cytC forms that could be attributed to the loss of heme coordination.

20. The authors detect a shift in the signal attributed to the meso-β proton (at -0.79 ppm) T78S/K79P and T78N/K79Y/M80I/I81M/G84N mutants. Again, how can the authors rule out that this effect is due to a possible loss of the heme coordination? In these two mutants, K79 or M80 were substituted, with the consequent implications for heme coordination. Determination of pH for alkaline transition of Cc mutants would be elucidating.

We thank the Reviewer's for the advice, we use this method in further work with mutant variants of cytochrome c. Regarding the preservation of heme coordination, the answer to this question was covered in detail in Answer 16.

Minor comments:

We thank the Reviewer's minor comments.

Line 80: “cytochromes c1” should be replaced by “cytochrome c1”.

Сorrected.

Lines 130 and 135: “SDS-PAAG electrophoresis” should be replaced by “SDS-PAGE electrophoresis”.

Сorrected.

Line 169: the TMPD acronym should be defined. By the way, what is this used for?

Сorrected.

The cytochrome c oxidase activity was measured amperometrically using a closed platinum electrode [Vinogradov, A.; Leikin, Yu.; Lipskaya T. Mitochondrial biochemistry. Bioenergetics. In Manual of practical study to animal biochemistry. Lomonosov Moscow State University publishers, Russia, 1977; pp. 19-22.]. The reaction was initiated by adding ascorbic acid to the reaction medium. The ascorbic acid acted as an electron donor for tetramethyl-p-phenylene diamine (TMPD), which, in its turn, transferred electrons to oxidized cytochrome c [Ferguson-Miller S, Brautigan DL, Margoliash E. Correlation of the kinetics of electron transfer activity of various eukaryotic cytochromes c with binding to mitochondrial cytochrome c oxidase. J. Biol. Chem. 1976; 251:1104-1115.]. The reduced cytochrome c then transferred electrons to complex IV that reduced oxygen to water. We monitored the activity of cytochrome c oxidase relatively to the decrease in oxygen concentration in the reaction mixture.

Line 214: the acronym ETC should be defined as electron transport chain.

Сorrected.

Line 222-223: The phase “Figure 2 shows the Amax and Km values for…” should be replaced by “Figure 1 shows the Amax and Km values for…”.

Сorrected.

Line 303-304: “Here we present description of 1H-NMR spectra (Fig.4) that directly characterize 303 heme molecule in oxidized cytochrome c [29].” Fig. 4 shows IR spectra.

Сorrected.

Reference 11 is not properly formatted.

Сorrected.

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

The authors have addressed my comments - thank you.