Aging: A mitochondrial DNA perspective, critical analysis and an update

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4237642/

Inna N Shokolenko, Glenn L Wilson, and Mikhail F Alexeyev

 

This review discusses the recent criticisms of the mainstream view that a vicious cycle of ROS-induced mtDNA damage induces further ROS production and mtDNA damage. For instance, the authors cite evidence that chronic exposure of cells to rotenone (a complex I inhibitor and ROS generator) and hydrogen peroxide causes no significant increase in mtDNA mutation. Indeed, the superoxide radical inhibits the enzyme aconitase, suppressing the Krebs cycle and reducing the supply of NADH and FADH2, which reduces the electron flow through the ETC. Thus there may exist a negative feedback loop for ROS production in oxidative phosphorylation. The authors suggest that ROS may in fact contribute to adaptive signalling to mitigate the effects of ageing. For instance, naked mole-rats live almost 8 times longer than mice, and yet have a much higher oxidative burden, especially in their mtDNA.

Ischaemic accumulation of succinate controls reperfusion injury through mitochondrial ROS

http://www.nature.com/nature/journal/v515/n7527/full/nature13909.html

Edward T. Chouchani, Victoria R. Pell, Edoardo Gaude et al.

The authors investigate the mechanism of ischaemia-reperfusion (IR), which occurs when the blood supply to a tissue is disrupted and then restored, such as during a heart attack or stroke. It is already well-established that tissue damage occurs once blood supply is restored, as this causes the electron transport chain to run in reverse: using ATP to pump protons into the mitochondrial matrix, generating reactive oxygen species (ROS) and inducing cell death. This study shows that accumulation of succinate is a universal feature of ischaemia, due to complex II reversal at low oxygen concentration. After reperfusion, the large amount of succinate drives conventional electron transport in complex III and IV, whilst driving complex I to run in reverse, to generate ROS. The authors show that inhibition of complex II ameliorates reperfusion injury. This cardiprotection is lost again when dimethyl succinate is added back, indicating that succinate elimination is the key protective feature of complex II inhibition.

Compartmentalization of the protein repair machinery in photosynthetic membranes

http://www.pnas.org/content/111/44/15839.full

Sujith Puthiyaveetil, Onie Tsabari, Troy Lowry, Steven Lenhery, Robert R. Lewis, Ziv Reich and Helmut Kirchoff

Photosynthesis relies on the functioning of the PSII complex, which is embedded in the thylakoid membranes of plant chloroplasts. This complex is large and composed of multiple subunits, of which the D1 subunit bears most of the brunt of the damage from photosynthesis. The complex undergoes a repair cycle which reduces the amount of protein synthesis required by only replacing the D1 subunit through a series of reactions designed to remove and degrade the damaged D1 subunit without destroying the entire complex, before synthesising a new D1 subunit. This repair process occurs in the stroma lamellae (which connect the stacked grana), however PSII is concentrated in the stacked grana. PSII phosphorylation triggers its disassembly before it is transported through the grana margin to the stroma lamellae for D1 replacement. The repair cycle is involved and must happen rapidly - the entire PSII complement of a plant can be turned over in just 1 hour. This requires an efficient process with minimal back-reactions.

The authors investigated the organisation of the PSII repair system through fractionation of thylakoid membranes and electron microscopy (EM) investigation of stacked grana. High-light (HL) treatment, which causes increased damage to PSII, led to growth in grana margins at the expense of grana core and stroma lamellae. By quantifying levels of key proteins involved in PSII repair in the grana core, margins and stroma lamellae, the authors deduced that localising different components in different areas allows the PSII repair system to improve its throughput by minimising back-reactions. In addition, localisation of proteolytic steps also reduces the amount of unnecessary protein degradation, by ensuring that access to proteasomes is restricted.

The authors identify discovering factors that govern the localisation of enzymes to the appropriate compartment as an important future step.

Transcription could be the key to the selection advantage of mitochondrial deletion mutants in aging

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3939916/pdf/pnas.201314970.pdf

Axel Kowald and Thomas B. L. Kirkwood

In this paper they discuss a mechanism that could be responsible for the observed "clonal expansion" of mtDNA mutations. Clonal expansion means that in individual cells, the mitochondrial population is often taken over by a single type of mutant (as opposed to many different mutants). There have been various suggestions as to how clonal expansion occurs but none of these suggestions can account for all the experimental observations. For example, the hypothesis that clonal expansion of a single mutant happens merely by chance (random drift) can explain clonal expansion in long lived, but not short lived species.

In this paper, they propose a new mechanism of clonal expansion. The idea they propose is that normally, mtDNA replication rate decreases if concentrations of some of its products are high (a negative feedback loop). Mutants that miss that part of the genome that encodes for proteins involved in this feedback loop will not be able to suppress mtDNA replication rate through this feedback mechanism. These deletion mutations therefore have a replicative advantage over wild-type mitochondria.

By studying the position in the genome of mtDNA deletions that were observed to be clonally amplified, they found that most deletions overlap a subunit of complex I of the respiratory chain. This means that it could be that this subunit is involved in the negative feedback loop, and missing this subunit gives a replicative advantage that allows for clonal expansion.

They then provide an ODE model describing the dynamics of wildtype and mutant mtDNA as well as ATP concentration. In their model they assume that mutants have a 50% higher replication rate because they lack the negative feedback loop.  Because of this, the mutant population increases exponentially until the cell collapses (collapse is defined by a certain drop in ATP concentration and each extra mutant present lowers the ATP concentration by some amount).  Their model predicts that, if one starts with 1000 wildtype mtDNAs and 1 mutant mtDNA, at the moment of collapse the number of mutants is 9-fold higher than the number of wildtypes. This is in agreement with experimental results.

They continue to make a stochastic version of their ODE model, which can predict clonal expansion of a single type of mutant (rather than accumulated of various mutants) in both long- and short-lived species.

A Mitochondrial ATP Synthase Subunit Interacts with TOR Signaling to Modulate Protein Homeostasis and Lifespan in Drosophila

http://www.cell.com/cell-reports/abstract/S2211-1247(14)00685-8

Xiaoping Sun, Charles T Wheeler, Jason Yolitz, Mara Laslo, Thomas Alberico, Yaning Sun, Qisheng Song, Sige Zou

The authors studied a Drosophila model in which ATP synthase d (ATPsyn-d) was knocked down after reaching adulthood using a drug-inducible gene switch system (simply knocking down ATPsyn-d led to lethality before pupation). ATPsyn-d knockdown after reaching adulthood increased lifespan in female but not male Drosophila with no alteration to food intake, however overexpression did not have any effect. The lifespan extension was diet-dependent, and enhanced lifespan in high protein, low sugar diets but decreased them when the proportions were reversed.

ATPsyn-d knockdown paradoxically led to increased ATP levels through a decrease in UCP expression, but only on a protein-rich diet. It also caused significant transcriptional changes, including proteins involved in a range of protective, proteostatic and energy-generating roles.

The authors also explored the effects of ATPsyn-d knockdown on TOR and MAPK pathways, and found that overexpression of the TOR signalling suppressor Tsc2 cancelled out the longevity-promoting effects of ATPsyn-d knockdown. Feeding with rapamycin had a similar but less pronounced effect, reducing the lifespan extension due to ATPsyn-d knockdown. This is attributed to overlapping pathways of lifespan extension between ATPsyn-d and rapamycin, which are likely to include TOR signalling.

Finally, a decrease in protein polyubiquitination and polyubiquitinated protein aggregates was observed. These are biomarkers for proteostasis, leading the authors to conclude that the lifespan extension due to ATPsyn-d knockdown is at least partially accomplished through improved proteostasis. The authors suggest that ATPsyn-d and other mitochondrial proteins regulate TOR signalling, and that perhaps the beneficial effects exist in high protein, low carbohydrate diets because there TOR levels are high, whereas in low protein, high carbohydrate diets TOR levels are low so there is nothing to knock down.

Kinetic evidence against partitioning of the ubiquinone pool and the catalytic relevance of respiratory-chain supercomplexes

James N. Blaza, Riccardo Serreli, Andrew J. Y. Jones, Khairunnisa Mohammed, and Judy Hirst

http://www.pnas.org/content/111/44/15735.full

The electron transport chain (ETC) consists of protein complexes, which pump protons across a membrane to store electrochemical potential energy. This potential energy is then converted by another element of the ETC into an energy currency: ATP. Recent studies have shown that the complexes of the electron transport chain exist in supercomplexes, where multimers of the ETC components combine, which is thought to increase their efficiency. 

The authors present evidence disputing the importance of these supercomplexes (but not their existence). Ubiquinone (Q) is a protein which ferries electrons through the lipid membrane of the ETC, and exists in a pool between the leaflets of the membrane. It has been proposed that the Q pool is partitioned such that supercomplexes have a separate pool to the TCA-cycle. However, the authors show that altering the levels of substrate for the ETC and TCA causes global changes in the Q pool, which is evidence against partitioning.

Supercomplexes are also widely cited to exist because of the observation that flux through the ETC is almost entirely controlled by both complex I and III of the ETC, suggesting they exist in a complex together. The authors here, however, question this finding because these studies rely on a particular inhibitor, rotenone, to inhibit complex I. Upon replication of previously published protocols, the authors find flux control coefficients of >100%, which is impossible. Using alternative inhibitors, the authors measured more modest control coefficients, which the authors interpret as evidence against the catalytic importance of supercomplexes.

Feedback regulation via AMPK and HIF-1 mediates ROS-dependent longevity in Caenorhabditis elegans

http://www.pnas.org/content/111/42/E4458.short

There's a wealth of stuff in this paper and it's all pretty cool. The idea is to examine the finding that inhibiting mitochondrial respiration extends lifespan. These guys pick apart a feedback mechanism involving HIF-1 and AMPK that responds to, and balances, ROS levels. They find that HIF-1 and AMPK cross-repress and regulate ROS levels in different directions, potentially allowing for fine control over ROS levels. The feedback system is coupled to free iron homeostasis (free iron can lead to the production of ROS) and intriguingly seems to modulate resistance to bacterial pathogens (suggesting that mitochondrial ROS lowers rates of infection).

(By the way, they're using ROS probes DCF-DA and Amplex Red -- not cpYFP, which we have issues with)

A Mitofusin-2 – dependent inactivating cleavage of Opa1 links changes in mitochondria cristae and ER contacts in the postprandial liver

  http://www.pnas.org/content/early/2014/10/28/1408061111.long

Aditi Sood,Danny Vijey Jeyaraju, Julien Prudent, Alexandre Caron, Philippe Lemieux, Heidi May McBride, Mathieu Laplante, Katalin Tóth, and Luca Pellegrini; PNAS

In this paper, they study the liver of two groups of mice: 1) the first group was killed 2 hours after feeding 2) the second group was killed 5h after feeding. Nutrient levels become limited in the 5h group, they then studied the shape of cristae and the amount of ER-mitochondria contact in the 5h group as compared to the 2h group.

Changes in cristae structure

The number of cristae per mitochondrion had decreased in liver cells with limited nutrients. The average length per cristae remained the same.

 
Changes in mitochondria-ER contact
About 1 in 4 mitochondria in both groups was in contact with the ER, but in the nutrient limiting group the mitochondrial surface area in contact with the ER had increased. The ER thus wrapped more extensively around mitochondria when nutrients were limited.

OPA1 cleavage observed which is MFN2 dependent
They then continue to investigate what caused the cristae remodelling. Total OPA1 expression levels were the same in both groups, but two new forms of OPA1 were identified in nutrient-limiting conditions. These two forms were the result of OPA-1 cleavage by an unknown cysteine protease.The sites of cleavage were named C1 and C2. Cleavage at either C1 or C2 is likely to inactive the dynamin-like activity of OPA1, but does not interfere with mitochondrial dynamics. Mfn2 was required for the observed OPA1 cleavage, further suggesting a link between changes in cristae shape and changes in mitochondria-ER contact (because MFN2 is involved in mitochondria-ER tethering).

Conclusions 
 Mitochondria adapt to changes in nutrient availability by remodelling their cristae and changing the amount of mitochondria-ER contact. The cristae remodelling is mediated through MFN2-dependent OPA1 cleavage by a cysteine protease. The study suggests that cristae remodelling and changes in mitochondria-ER contact are coupled during nutrient depletion.