Cryo-EM structures of mitochondrial respiratory complex I from Drosophila melanogaster

Article activity feed

1. 

   Version published to 10.7554/elife.84424 on eLife

   Jan 9, 2023
2. 

   eLife

   Dec 12, 2022

   Author Response

   Evaluation summary

   This important study advances our understanding of respiratory complex I. The authors present convincing structural data for the enzyme from Drosophila melanogaster although the interpretation of conformational states is still not conclusively settled. This work will be of interest to researchers studying respiratory enzymes, the evolution of respiration, and mitochondrial diseases.

   Thank you for this positive evaluation of our work. Although we have presented a robust and coherent interpretation of the conformational states we observe, we accept that different opinions on this topic still exist in the field.

   Reviewer #1 (Public Review):

   Agip et al. have resolved the first cryoEM structure of the mitochondrial Complex I from Drosophila melanogaster, an important model organism in biology. The …

   Author Response

   Evaluation summary

   This important study advances our understanding of respiratory complex I. The authors present convincing structural data for the enzyme from Drosophila melanogaster although the interpretation of conformational states is still not conclusively settled. This work will be of interest to researchers studying respiratory enzymes, the evolution of respiration, and mitochondrial diseases.

   Thank you for this positive evaluation of our work. Although we have presented a robust and coherent interpretation of the conformational states we observe, we accept that different opinions on this topic still exist in the field.

   Reviewer #1 (Public Review):

   Agip et al. have resolved the first cryoEM structure of the mitochondrial Complex I from Drosophila melanogaster, an important model organism in biology. The structure revealed a 43-subunit enzyme complex that closely resembles the mammalian Complex I. The authors resolved Complex I in three different conformational states at 3.3-4.0 Å global resolution, with an overall resemblance to the active form of the mammalian Complex I, but also with some characteristic conformational changes near the quinone substrate pocket and surrounding subunits that resemble, at least in part, the deactive form of the mammalian enzyme. The third resolved class was considered 'damaged/broken', and a possible artifact arising from the sample preparation. Biochemical assays showed that the Drosophila Complex I does not undergo an active/deactive transition (as characterized by the N-ethylmaleimide sensitivity), although the structures revealed an exposed ND3 loop that has been linked to transition. The authors could also show that conformational change between an alpha and pi form of transmembrane helix (TM3-ND6) is likely to be involved in catalysis, and distinct from the deactivation mechanism of the mammalian isoform. Due to the 3.3 Å global resolution, water molecules could not be experimentally resolved, and how the observed conformational changes link to the proton pumping activity therefore remains an open question and basis for future studies. Overall I find that this work provides an important basis for understanding mechanistic principles of the mitochondrial Complex I and more specifically a starting point for detailed genetic studies on the fruit fly Complex I.

   We thank the reviewer for their positive evaluation of our manuscript.

   We would like to note that in all three conformational states of Drosophila complex I observed in our study, we do not observe an exposed ND3 loop (Cys39 in particular), as outlined in Figures 3 & 6 and Figure 6 – Figure Supplement 1 (as well as in Figures 5 and 7). This observation is fully consistent with the lack of N-ethylmaleimide (NEM) sensitivity observed in our Drosophila preparation.

   We discuss the relevance of the π-bulge/α-helical nature of ND6-TMH3 to catalysis in the Discussion section in the context of an intercalated phospholipid molecule in the Dm1 structure: “Indeed, if ND6-TMH3 converts between its -bulge and -helical structures during catalysis (Agip et al., 2018; Kampjut and Sazanov, 2020; Kravchuk et al., 2022; Parey et al., 2021; Röpke et al., 2021), then the intercalating phospholipid is very unlikely to be present in the -helical state, moving repeatedly in and out.” While our structures are consistent with this helical change being involved in catalysis, they are resting-state structures and therefore do not provide further evidence in this regard.

   Finally, the reviewer is correct in that the resolutions of the structures resolved here are insufficient to model water molecules, and that how the conformational changes observed here contribute to our currently limited knowledge of the coupling mechanism remains to be answered.

   Reviewer #2 (Public Review):

   • Aim of the study:

   Agip et al. studied the structure of respiratory complex I from Drosophila melanogaster, an important model organism with well-developed genetic toolkit and sufficiently close phylogenetic relationship to mammals. They isolated the complex and analyzed its structure by single-particle electron cryo-microscopy (cryo-EM). They also used mass spectrometry to characterize new subunits. So far, the structures of complex I have been reported for several organisms, including mammals, plants, ciliates, fungi and bacteria, but ones from insects have been missing. This study aims to fill this gap and shed light on some of the key questions pertaining complex I biology, such as 1) the conservacy of supernumerary subunits, 2) the mechanisms and physiological relevance of active/deactive transition and 3) the correspondence between the structurally defined closed/open conformations and the biochemically defined active/deactive states.

   We thank the reviewer for clearly summarising the key aims of the study relative to the current status of the complex I field.

   • Strengths:

   The study provides the first structure of complex I from insects, the organisms at an important phylogenetic branch that has diverged from mammals more recently than other eukaryotic species such as plants and fungi. Using purification methods they developed for mammalian enzymes previously, the authors successfully purified the insect enzyme with high quality - a monodisperse peak in gel filtration, the NADH oxidation activity comparable to mammalian enzymes, and the homogenous subunit composition as confirmed by single-particle analyses. It is noteworthy that the authors used state-of-the art tools in model building and validation, such as ISOLDE and MapQ, which makes this model of high standard. In my opinion such careful validation is particularly important for modelling such a gigantic complex, since without cares one can easily misinterpret the density and draw wrong conclusions.

   The resolution is 3.3 Angstrom for the best class (Dm1), which allowed modelling side chains and comparing between the observed 3D classes and to the known structures. The model confirms the presence of 43 subunits, akin to mammalinan enzymes, composed of 14 conserved core subunits, 28 supernumerary subunits that have close homologs in mammals, and one supernumerary subunit CG9034 that has not been predicted. They are also structurally similar to mammalian enzymes except for minor local differences. The two supernumerary subunits (NDUFC1 and NDUFA2) that are present in mammals are missing. The authors discuss evidence that NDUFC1 is absent from the Drosophila genome and NDUFA2 is genomically present but its expression is restricted to the male germline. Together, the overall similarity to the mammalian enzyme underlines the use of Drosophila complex I as a model system.

   One of the remarkable findings is that common biochemical treatments that are used to deactivate mammalian complex I - heat treatment or NEM treatment - did not reveal deactive state of Drosophila complex I. This is in agreement with their observation that most structural elements are in the active state. The major Dm1 conformation shows all structural features in the active conformation, whereas Dm2 state shows two features in the deactive conformations. Here the author raises an interesting point that the structural elements formerly believed to behave in a consorted manner are actually not coupled, providing new perspective in interpreting complex I structures presented so far and in future. Notably, the authors adopted the same purification procedure for bovine and murine samples. This is a particular strength that they applied a similar procedure for but still observed different behaviors for Drosophila (the absence of the deactive state).

   We thank the reviewer for their positive evaluation of the strengths of the paper.

   • Weaknesses:

   As the authors point out in Discussion, the biochemical statuses of the two described conformations, Dm1 and Dm2, are uncertain. If we assume that Dm1 is a ready-to-go active state, Dm2 could represent several of the possible states; a partially broken state due to delipidation by detergent, a meta-stable state during enzyme turnover, an intermediate towards "full deactiving" structural transition (which the authors argue is unlikely to occur), or a fully reversible state that is in equilibrium to Dm1. Despite these uncertainties, the structure will serve as an excellent starting point to address many open questions in the complex I field in future.

   We agree that the biochemical status of Dm2 is uncertain and as the reviewer notes, we made an attempt to address this question in the Discussion section.

   In the final 3D classification the number of classes was set to 3 (K = 3). This is an arbitrary human decision and implicitly forces particles to separate into 3 descrete classes. It would have been great to mention if the authors had tried different classification parameters and, if so, whether that had led to similar classification results. There are different methods available to dissect conformational heterogeneity other than simple 3D classification. For example, focused classification can differentiate local structural features. MultiBody refinement and 3D variabitlity can analyze continuous conformational changes. The simple 3D classification with local angular sampling employed here may lead to over-simplification of the more complex structural heterogeneity.

   First, the number of classes was set to 5 (K = 5) as written in the Materials and methods section (page 20), which is greater than the number of complex I conformations observed in this study. We apologise if this was not clear and we have amended Figure 1 – Figure Supplement 2 to clarify it.

   Second, as the reviewer correctly points out, there are many different methods to dissect conformational heterogeneity, and for this reason we purposefully performed several classification strategies before validating that the Global 3D classification approach used here (with local angular search extending to 0.2º sampling) yielded comparable (or even better) results. These additional classification strategies include:

   (i) Focus-revert-classify (a strategy often used for complex I (Kampjut and Sazanov, 2020; Klusch et al., 2021; Kravchuk et al., 2022; Letts et al., 2019)) in RELION, where the membrane arm of complex I is first subtracted to focus-refine on the hydrophilic arm, the subtraction reverted, and then focus-classification performed without alignment on the membrane arm. Here, we used a regularisation parameter, t = 8, and K = 5, and the process yielded three complex I classes matching Dm1, Dm2, and Dm3 with comparable population distribution to the aforementioned Global 3D classification method, plus two junk classes.

   (ii) A 3D classification without alignment approach (a strategy also used for complex I (Gu et al., 2022)) in RELION. We used t = 20 and up to K = 12 classes, which resulted in two < 4 Å resolution complex I classes, with the major class matching Dm1 and the minor class a likely mixture of Dm2 and Dm3.

   Based on these three classification strategies, we chose to work with the Global 3D classification approach that has previously proven robust for separating complex I heterogeneity in our hands (Agip, 2018; Chung et al., 2022b; Zhu et al., 2016). However, we agree with the reviewer that it would be valuable to provide this extra information. Therefore, we have amended the Materials and methods section on page 20: “The ‘Focus-Revert-Classify’ classification strategy (Letts et al., 2019), applied using the regularisation parameter t = 8 and K = 5, yielded comparable population distributions (three complex I classes matching Dm1, Dm2, and Dm3, plus two junk classes) whilst 3D classification without alignment using t = 20 and K ≤ 12 yielded two < 4 Å complex I classes, with the major class matching Dm1 and the minor class an apparent mixture of Dm2 and Dm3. The 3D classification approach with local angular sampling was therefore employed to give the final set of Dm1, Dm2 and Dm3 particles as described above.” Furthermore, clear cryo-EM densities for Dm2-specific local features, including the ‘flipped’ ND1-TMH4-Tyr149 and the ND6-TMH3 π-bulge, revealed no evidence for Dm1 contamination in the Dm2 population. This is also now noted on page 20.

   Although 37 degrees heat treatment and NEM treatment did not reveal any sign of deactivation in Drosophila complex I, it does not rule out the possibility that insect complex I has different ways to deactivate the enzyme, to prevent ROS production. It is probably the limitation of applying existing assays that are originally for mammalian and fungal enzymes to the study of insect enzymes.

   The reviewer is correct that Drosophila complex I may have a different way to ‘deactivate’ that does not involve an exposure of ND3-Cys39, and it is also possible that the conditions used for deactivation of mammalian mitochondrial membranes (i.e. 37 ºC heat treatment for 30 min) may not be sufficient to deactivate the Drosophila enzyme. Our approach here was to evaluate if Drosophila complex I undergoes the same active/deactive transition as the mammalian enzyme both structurally and biochemically (and our results suggest that it doesn’t). Moving forward, deactivation mechanisms in different phylogenetic lineages will be an important question to address in the complex I field. We have addressed this question in the first paragraph of the Discussion.

   • Whether they achieved the aims and whether the conclusions are supported by the results:

   Overall, they successfully isolated the active enzyme and determined its structure at 3.3 A resolution, which meets the current state-of-the-art for single-particle cryo-EM and provided an atomic picture of the enzyme composition. The study confirms that the Drosophila complex I is structurally similar to mammalian complex I, but biochemically different in that it does not show the deactive state. It still does not exclude the possibility that Drosophila complex I can transition into a currently unknown state that prevents reverse electron transfer. This question however can be tackled in future by mutagenesis analyses as Drosophila is a genetically tractable organism.

   We agree with the reviewer on his evaluation of the study, and the genetic tractability of the Drosophila enzyme will serve as a crucial tool for future studies.

   • Impact to the field and utility of the data to the community:

   Complex I is important not only for human health but also for understanding universal principles of biological respiration, because of its universal presence in most organisms on Earth. This study provides a basis for relating mammalian complex I with those from other branches of organisms. The current structures will allow Drosophila researchers to interpret and design any mutations that affect complex I functions, and relate them to behavioral, developmental and metabolical changes at tissues, organs and individuals levels.

   We agree with the reviewer on his evaluation of the impact of the study, and thank the reviewer for their comments on the manuscript.

   Reviewer #3 (Public Review):

   The mitochondrial NADH dehydrogenase complex (complex I) is of prime importance for cellular respiration. It has been biochemically and structurally characterized for several groups of organisms, including mammals, fungi, algae, seed plants and protozoa. Furthermore, different complex I conformation have been reported, which are considered to possibly represent distinct physiological states of the enzyme complex. E.g. in mammalian mitochondria, two resting states can be distinguished, designated 'ready-to-go' resting state, and 'deactive' resting state. To better understand the physiological relevance of these states, complex I is here investigated from the fruit fly Drosophila melanogaster, which represents a model for insects but beyond for metazoan in general and which can be easily genetically modified.

   Complex I from Drosophila is presented at up to 3.3 Angstrom resolution. It includes 43 of the 45 complex I subunits defined for mammalian complex I. Subunit NDUFA3 has been found in Drosophila complex I for the first time. Overall, Drosophila complex I is remarkably similar in its composition and structure to the mammalian enzyme. Only minor topological differences were found in some subunits. Furthermore, three different complex I states are described, termed Dm1, Dm2 and Dm3. The three states are extensively discussed and compared to the states found in mammalian complex I. Dm1, which is the dominating class, likely represents the active resting state. In Dm2, the two complex I arms are slightly twisted with respect to Dm1. In Dm3, the membrane arm appears to be 'cracked' at the interface between ND2 and ND4. It possibly represents an artefact resulting from detergent-induced loss of stability in the distal membrane domain of the Dm2 state. Both, Dm2 and Dm3 most closely correspond to the mammalian active state. A state resembling the mammalian deactive state could not be found. This result is further supported by biochemical experiments. It is concluded that Drosophila complex I, despite its remarkable similarity to the mammalian enzyme, does not undergo the mammalian-type active/deactive transition.

   In conclusion, complex I structure from Drosophila is of limited value for the better understanding of the states of mammalian complex I (which could be stated more clearly). However, insights into complex I structure and function of an insect is highly interesting. The conclusions are justified by the presented data. The manuscript is well written and the figures are thoroughly prepared. The discussion very much focusses on the interpretation of the three complex I states. The deactivate state, which is interpreted to protect mammalian mitochondria from ROS production during reverse electron transfer, might be dispensable in species characterized by a comparatively short life cycle like Drosophila, which is in the range of weeks.

   We thank the reviewer for clearly summarising the key findings of the study. We agree that Drosophila complex I may have limitations for studying the full active/deactive transition so far observed exclusively in mammalian enzymes, but we argue that the lack of a fully deactivated state also provides a good system to study which local elements in complex I may offer protection against RET. Despite these limitations, Drosophila remains a powerful model system to study complex I mechanism, assembly, and regulation in physiological contexts.

   Read the original source
3. 

   eLife

   Dec 12, 2022

   eLife assessment

   This important study advances our understanding of respiratory complex I. The authors present convincing structural data for the enzyme from Drosophila melanogaster although the interpretation of conformational states is still not conclusively settled. This work will be of interest to researchers studying respiratory enzymes, the evolution of respiration, and mitochondrial diseases.

   Read the original source
4. 

   eLife

   Dec 12, 2022

   Reviewer #1 (Public Review):

   Agip et al. have resolved the first cryoEM structure of the mitochondrial Complex I from Drosophila melanogaster, an important model organism in biology. The structure revealed a 43-subunit enzyme complex that closely resembles the mammalian Complex I. The authors resolved Complex I in three different conformational states at 3.3-4.0 Å global resolution, with an overall resemblance to the active form of the mammalian Complex I, but also with some characteristic conformational changes near the quinone substrate pocket and surrounding subunits that resemble, at least in part, the deactive form of the mammalian enzyme. The third resolved class was considered 'damaged/broken', and a possible artifact arising from the sample preparation. Biochemical assays showed that the Drosophila Complex I does not undergo an …

   Reviewer #1 (Public Review):

   Agip et al. have resolved the first cryoEM structure of the mitochondrial Complex I from Drosophila melanogaster, an important model organism in biology. The structure revealed a 43-subunit enzyme complex that closely resembles the mammalian Complex I. The authors resolved Complex I in three different conformational states at 3.3-4.0 Å global resolution, with an overall resemblance to the active form of the mammalian Complex I, but also with some characteristic conformational changes near the quinone substrate pocket and surrounding subunits that resemble, at least in part, the deactive form of the mammalian enzyme. The third resolved class was considered 'damaged/broken', and a possible artifact arising from the sample preparation. Biochemical assays showed that the Drosophila Complex I does not undergo an active/deactive transition (as characterized by the N-ethylmaleimide sensitivity), although the structures revealed an exposed ND3 loop that has been linked to transition. The authors could also show that conformational change between an alpha and pi form of transmembrane helix (TM3-ND6) is likely to be involved in catalysis, and distinct from the deactivation mechanism of the mammalian isoform. Due to the 3.3 Å global resolution, water molecules could not be experimentally resolved, and how the observed conformational changes link to the proton pumping activity therefore remains an open question and basis for future studies. Overall I find that this work provides an important basis for understanding mechanistic principles of the mitochondrial Complex I and more specifically a starting point for detailed genetic studies on the fruit fly Complex I.

   Read the original source
5. 

   eLife

   Dec 12, 2022

   Reviewer #2 (Public Review):

   - Aim of the study:

   Agip et al. studied the structure of respiratory complex I from Drosophila melanogaster, an important model organism with well-developed genetic toolkit and sufficiently close phylogenetic relationship to mammals. They isolated the complex and analyzed its structure by single-particle electron cryo-microscopy (cryo-EM). They also used mass spectrometry to characterize new subunits. So far, the structures of complex I have been reported for several organisms, including mammals, plants, ciliates, fungi and bacteria, but ones from insects have been missing. This study aims to fill this gap and shed light on some of the key questions pertaining complex I biology, such as 1) the conservacy of supernumerary subunits, 2) the mechanisms and physiological relevance of active/deactive transition …

   Reviewer #2 (Public Review):

   - Aim of the study:

   Agip et al. studied the structure of respiratory complex I from Drosophila melanogaster, an important model organism with well-developed genetic toolkit and sufficiently close phylogenetic relationship to mammals. They isolated the complex and analyzed its structure by single-particle electron cryo-microscopy (cryo-EM). They also used mass spectrometry to characterize new subunits. So far, the structures of complex I have been reported for several organisms, including mammals, plants, ciliates, fungi and bacteria, but ones from insects have been missing. This study aims to fill this gap and shed light on some of the key questions pertaining complex I biology, such as 1) the conservacy of supernumerary subunits, 2) the mechanisms and physiological relevance of active/deactive transition and 3) the correspondence between the structurally defined closed/open conformations and the biochemically defined active/deactive states.

   - Strengths:

   The study provides the first structure of complex I from insects, the organisms at an important phylogenetic branch that has diverged from mammals more recently than other eukaryotic species such as plants and fungi. Using purification methods they developed for mammalian enzymes previously, the authors successfully purified the insect enzyme with high quality - a monodisperse peak in gel filtration, the NADH oxidation activity comparable to mammalian enzymes, and the homogenous subunit composition as confirmed by single-particle analyses. It is noteworthy that the authors used state-of-the art tools in model building and validation, such as ISOLDE and MapQ, which makes this model of high standard. In my opinion such careful validation is particularly important for modelling such a gigantic complex, since without cares one can easily misinterpret the density and draw wrong conclusions.

   The resolution is 3.3 Angstrom for the best class (Dm1), which allowed modelling side chains and comparing between the observed 3D classes and to the known structures. The model confirms the presence of 43 subunits, akin to mammalinan enzymes, composed of 14 conserved core subunits, 28 supernumerary subunits that have close homologs in mammals, and one supernumerary subunit CG9034 that has not been predicted. They are also structurally similar to mammalian enzymes except for minor local differences. The two supernumerary subunits (NDUFC1 and NDUFA2) that are present in mammals are missing. The authors discuss evidence that NDUFC1 is absent from the Drosophila genome and NDUFA2 is genomically present but its expression is restricted to the male germline. Together, the overall similarity to the mammalian enzyme underlines the use of Drosophila complex I as a model system.

   One of the remarkable findings is that common biochemical treatments that are used to deactivate mammalian complex I - heat treatment or NEM treatment - did not reveal deactive state of Drosophila complex I. This is in agreement with their observation that most structural elements are in the active state. The major Dm1 conformation shows all structural features in the active conformation, whereas Dm2 state shows two features in the deactive conformations. Here the author raises an interesting point that the structural elements formerly believed to behave in a consorted manner are actually not coupled, providing new perspective in interpreting complex I structures presented so far and in future. Notably, the authors adopted the same purification procedure for bovine and murine samples. This is a particular strength that they applied a similar procedure for but still observed different behaviors for Drosophila (the absence of the deactive state).

   - Weaknesses:

   As the authors point out in Discussion, the biochemical statuses of the two described conformations, Dm1 and Dm2, are uncertain. If we assume that Dm1 is a ready-to-go active state, Dm2 could represent several of the possible states; a partially broken state due to delipidation by detergent, a meta-stable state during enzyme turnover, an intermediate towards "full deactiving" structural transition (which the authors argue is unlikely to occur), or a fully reversible state that is in equilibrium to Dm1. Despite these uncertainties, the structure will serve as an excellent starting point to address many open questions in the complex I field in future.

   In the final 3D classification the number of classes was set to 3 (K = 3). This is an arbitrary human decision and implicitly forces particles to separate into 3 descrete classes. It would have been great to mention if the authors had tried different classification parameters and, if so, whether that had led to similar classification results. There are different methods available to dissect conformational heterogeneity other than simple 3D classification. For example, focused classification can differentiate local structural features. MultiBody refinement and 3D variabitlity can analyze continuous conformational changes. The simple 3D classification with local angular sampling employed here may lead to over-simplification of the more complex structural heterogeneity.

   Although 37 degrees heat treatment and NEM treatment did not reveal any sign of deactivation in Drosophila complex I, it does not rule out the possibility that insect complex I has different ways to deactivate the enzyme, to prevent ROS production. It is probably the limitation of applying existing assays that are originally for mammalian and fungal enzymes to the study of insect enzymes.

   - Whether they achieved the aims and whether the conclusions are supported by the results:

   Overall, they successfully isolated the active enzyme and determined its structure at 3.3 A resolution, which meets the current state-of-the-art for single-particle cryo-EM and provided an atomic picture of the enzyme composition. The study confirms that the Drosophila complex I is structurally similar to mammalian complex I, but biochemically different in that it does not show the deactive state. It still does not exclude the possibility that Drosophila complex I can transition into a currently unknown state that prevents reverse electron transfer. This question however can be tackled in future by mutagenesis analyses as Drosophila is a genetically tractable organism.

   - Impact to the field and utility of the data to the community:

   Complex I is important not only for human health but also for understanding universal principles of biological respiration, because of its universal presence in most organisms on Earth. This study provides a basis for relating mammalian complex I with those from other branches of organisms. The current structures will allow Drosophila researchers to interpret and design any mutations that affect complex I functions, and relate them to behavioral, developmental and metabolical changes at tissues, organs and individuals levels.

   Read the original source
6. 

   eLife

   Dec 12, 2022

   Reviewer #3 (Public Review):

   The mitochondrial NADH dehydrogenase complex (complex I) is of prime importance for cellular respiration. It has been biochemically and structurally characterized for several groups of organisms, including mammals, fungi, algae, seed plants and protozoa. Furthermore, different complex I conformation have been reported, which are considered to possibly represent distinct physiological states of the enzyme complex. E.g. in mammalian mitochondria, two resting states can be distinguished, designated 'ready-to-go' resting state, and 'deactive' resting state. To better understand the physiological relevance of these states, complex I is here investigated from the fruit fly Drosophila melanogaster, which represents a model for insects but beyond for metazoan in general and which can be easily genetically modified.

   C…

   Reviewer #3 (Public Review):

   The mitochondrial NADH dehydrogenase complex (complex I) is of prime importance for cellular respiration. It has been biochemically and structurally characterized for several groups of organisms, including mammals, fungi, algae, seed plants and protozoa. Furthermore, different complex I conformation have been reported, which are considered to possibly represent distinct physiological states of the enzyme complex. E.g. in mammalian mitochondria, two resting states can be distinguished, designated 'ready-to-go' resting state, and 'deactive' resting state. To better understand the physiological relevance of these states, complex I is here investigated from the fruit fly Drosophila melanogaster, which represents a model for insects but beyond for metazoan in general and which can be easily genetically modified.

   Complex I from Drosophila is presented at up to 3.3 Angstrom resolution. It includes 43 of the 45 complex I subunits defined for mammalian complex I. Subunit NDUFA3 has been found in Drosophila complex I for the first time. Overall, Drosophila complex I is remarkably similar in its composition and structure to the mammalian enzyme. Only minor topological differences were found in some subunits. Furthermore, three different complex I states are described, termed Dm1, Dm2 and Dm3. The three states are extensively discussed and compared to the states found in mammalian complex I. Dm1, which is the dominating class, likely represents the active resting state. In Dm2, the two complex I arms are slightly twisted with respect to Dm1. In Dm3, the membrane arm appears to be 'cracked' at the interface between ND2 and ND4. It possibly represents an artefact resulting from detergent-induced loss of stability in the distal membrane domain of the Dm2 state. Both, Dm2 and Dm3 most closely correspond to the mammalian active state. A state resembling the mammalian deactive state could not be found. This result is further supported by biochemical experiments. It is concluded that Drosophila complex I, despite its remarkable similarity to the mammalian enzyme, does not undergo the mammalian-type active/deactive transition.

   In conclusion, complex I structure from Drosophila is of limited value for the better understanding of the states of mammalian complex I (which could be stated more clearly). However, insights into complex I structure and function of an insect is highly interesting. The conclusions are justified by the presented data. The manuscript is well written and the figures are thoroughly prepared. The discussion very much focusses on the interpretation of the three complex I states. The deactivate state, which is interpreted to protect mammalian mitochondria from ROS production during reverse electron transfer, might be dispensable in species characterized by a comparatively short life cycle like Drosophila, which is in the range of weeks.

   Read the original source
7. 

   Version published to 10.1101/2022.11.01.514700v1 on bioRxiv

   Nov 1, 2022