Mitochondrial reticulum for cellular energy distribution in muscle

 http://www.nature.com/nature/journal/v523/n7562/full/nature14614.html

Brian Glancy, Lisa M. Hartnell, Daniela Malide, Zu-Xi Yu, Christian A. Combs, Patricia S. Connelly, Sriram Subramaniam, and Robert S. Balaban


Cells need energy, especially muscle cells. Mitochondria generate part of this energy and to be able to do this they need to be supplied with various resources. Muscle cells can be quite big, so these resources (that enter the cell at its periphery) need to diffuse through the cell towards the mitochondria, which might take a long time. Is there a better, more efficient way of producing energy in large muscle cells?

In this paper they show that the mitochondrial network provides a conductive pathway for energy distribution.

Mitochondria use their Electron Transport Chain protein complexes to pump protons across their membrane, which creates an electrochemical gradient in which energy is stored. Their ATP synthase then uses this stored energy to make ATP. If many mitochondria are connected in the cell (i.e. if they all have an electrically continuous inner membrane) then a mitochondrion at one end of the cell can use the potential energy created by a mitochondrion at the other end of the cell, to generate ATP. The conduction of electric potential along the mitochondria can be faster than all kinds of resources needing to diffuse through the cell. This idea was first proposed by a Russian scientist Владимир Скулачёв (Vladimir Skulachev).

Here they show that the proteins involved in generating the electrochemical gradient are mainly found at the cell periphery (where the resources enter the cell), while proteins involved in using this energy to create ATP are found in the cell's interior. They also show that the mitochondria are indeed electrically connected to each other. The mitochondria in muscle cells are organized in a way that facilitates energy conduction.

The question remains whether this mitochondrial conductivity plays a role in all cells, or only in cells that are very energy demanding. Skin cells for example, hardly seem to need a fast conducting mitochondrial network. Nevertheless, mitochondrial fusion and network forming is seen in a variety of cells, so the fusion of mitochondria probably has other uses as well.

Cross-strand binding of TFAM to a single mtDNA molecule forms the mitochondrial nucleoid

http://www.pnas.org/content/early/2015/08/20/1512131112.short?rss=1

Christian Kuka, Karen M. Davies, Christian A. Wurm, Henrik Spåhr, Nina A. Bonekamp, Inge Kühl, Friederike Joos, Paola Loguercio Polosa, Chan Bae Park, Viktor Posse, Maria Falkenberg, Stefan Jakobs, Werner Kühlbrandt and Nils-Göran Larsson

Mitochondrial DNA is found to exist in protein-DNA complexes called nucleoids.  The number of mtDNA molecules per nucleoid is an important quantity to know, as it has consequences for how mtDNA is distributed amongst successive generations. The compactness of mtDNA also determines its ability to be transcribed and replicated, so the way it is packaged is also important to understand.

In their first experiment, the authors use electron microscopy in vitro, to observe how mtDNA is compacted, with increasing concentrations of the molecule TFAM (the only protein known to package mtDNA). They find that upon binding to mtDNA, TFAM causes the DNA to bend by 180°. At intermediate concentrations 1 TFAM/30 bp, dense protein-DNA spots were dispersed amongst regions of naked DNA (perhaps being a concentration appropriate for translation or replication of mtDNA). At concentrations of 1 TFAM/6 bp, mtDNA was completely compacted, and a further increase in TFAM concentration had no further effect. This may be the density of TFAM required for storage of mtDNA.

They then continued their analysis in vivo, by studying TFAM-overexpressing mouse embryonic fibroblasts (OE MEFS), which had ~2.5-fold higher mtDNA copy number. Nucleoids can cluster in cells, and confocal microscopy cannot always resolve individual nucleoids. The number of nucleoids detected in wild-type cells using superresolution STED microscopy / confocal was 1.36+/-0.76. By counting nucleoids using STED microscopy, and mtDNA copy number with rt-qPCR, they found that wild-type cells had ~1.1 mtDNA molecules / nucleoid and OE MEFS had ~1.5 mtDNA molecules / nucleoid. Thus, overexpressing TFAM has only minor changes in the mean amount of mtDNA per nucleoid. Instead, the number of nucleoids was observed to increase, relative to wild-type cells.  

Mitotic redistribution of the mitochondrial network by Miro and Cenp-F

http://www.ncbi.nlm.nih.gov/pubmed/26259702

Kanfer G, Courthéoux T, Peterka M, Meier S, Soste M, Melnik A, Reis K, Aspenström P, Peter M, Picotti P, Kornmann B


If a cell divides, how do its mitochondria get segregated into the two daughter cells?  In general, mitochondria often move along microtubules with the help of molecular motors. Specialized adaptors on the mitochondrial outer membrane  recruit these motors, one of them being the GTPase Miro. But how exactly the segregation of mitochondria during mitosis is coordinated, is not well understood.

In this paper, they identify centromeric protein F (Cenp-F) which interacts with Miro. Cenp-F is a large protein that binds to the microtubules. It was found to be strongly recruited to mitochondria at the end of mitosis.  During S/G2, a fraction of Cenp-F is found on mitochondria, located mainly at the mitochondrial tips projecting away from the cell centre. The mitochondrial Cenp-F puncta that were visible in the S/G2 phase were all colocalized with microtubules.

They further find that Cenp-F directly interacts with the GTPase domains of both Miro1 and Miro2, leading to its recruitment to mitochondria. In wildtype cells, during cytokinesis mitochondria were parallel and extended away from the division plane, whereas in cells without Miro (or Cenp-F) the mitochondria were all clustered together around the nucleus.
 
The conclusion was that Cenp-F, recruiting by Miro, connects mitochondria to the tips of growing microtubules. The mitochondria then track the tips of the microtubules. Mitochondria were also seen to influence local microtubule dynamics, it can be that dragging forces caused by the attached mitochondria play a role in this.

Maternal transmission, sex ratio distortion, and mitochondria

http://www.pnas.org/content/112/33/10162.full

Steve J. Perlman, Christina N. Hodson, Phineas T. Hamilton, George P. Opit, Brent E. Gowen

Most (but not all) multicellular organisms have uniparental inheritance of mitochondrial DNA, which is thought to prevent invasion by a more competitive lineage, and maintain compatibility between mitochondrial and nuclear genomes. Mitochondrial DNA is also subject to evolutionary pressures, and in this review the authors discuss potential (deleterious) side-effects of uniparental inheritance. 

An immediate consequence of uniparental inheritance is that one sex is an evolutionary dead-end. Thus, any selective pressure on mtDNA is principally exerted on the transmitting sex (females), and so deleterious mutations for the non-transmitting (males) sex can accumulate. This explains why male infertility is often attributed to mutations in mtDNA. A study by Innocenti et al. [1], established fly lines with variation in their mtDNA. A large fitness variation was observed, but only in males; this caused significant differential expression across the nucleus of male flies. 

The authors suggest three possible consequences of maternal inheritance of mtDNA:

1. MtDNA mutations which are detrimental to males may fixate, if they do not affect female fitness
2. MtDNA mutations which cause the frequency of females to increase, may fixate
3. Nuclear symbionts carried on the nuclear female chromosome which fixate, will also cause their associated mtDNA haplotype to fixate

[1] Innocenti P, Morrow EH, Dowling DK (2011) Experimental evidence supports a sex-specific selective sieve in mitochondrial genome evolution. Science 332(6031):845–848.

An Essential Role of the Mitochondrial Electron Transport Chain in Cell Proliferation Is to Enable Aspartate Synthesis

http://www.sciencedirect.com/science/article/pii/S0092867415008533

Kıvanç Birsoy, Tim Wang, Walter W. Chen, Elizaveta Freinkman, Monther Abu-Remaileh, David M. Sabatini

When mitochondrial respiration is inhibited, it is observed that a cell's ability to proliferate is diminished. It is also known that cells with inhibited oxidative phosphorylation (OXPHOS) are still able to proliferate, if cultured in high concentrations of pyruvate (a metabolite which is the by-product of glycolysis, and feeds the tricarboxylic acid cycle). In this study, the authors screen a library of ~3,000 metabolic enzymes, treated with a low dose of complex I inhibitor phenformin, whose loss causes a severe anti-proliferative phenotype. This would reveal genes which are required in an adaptive response to mild OXPHOS inhibition.

The best scoring gene in their screen was GOT1, a component of the malate-aspartate shuttle, which transfers substrates for OXPHOS into the mitochondrial matrix. They find that GOT1-null cells are hypersensitive to phenformin, causing their proliferation to halt, at doses where wild-type cells do not.

Under normal conditions, GOT1 shuttles the amino acid aspartate, into the mitochondrial matrix. They find that, upon inhibition of complex I, GOT1 reverses its flux and exports aspartate. Aspartate is an amino acid, required for the synthesis of proteins, purines and pyramidines. Normally, it is generated in the mitochondrial matrix; the authors propose that during ETC inhibition, a drop in NAD+/NADH ratio causes aspartate synthesis to shut down, using normal pathways. Thus GOT1 reverses its flux, to become a source of this amino acid.

It is also known that supplementation of pyruvate can overcome the inhibitory effects of several ETC inhibitors. By culturing GOT1-null cells with/without pyruvate, the authors show that there is no benefit to supplementing cells with pyruvate under ETC inhibition, if cells lack GOT1. Thus a key mechanism of pyruvate supplementation is GOT1-catalyzed aspartate synthesis. Interestingly, these conclusions appear to hold in cybrid cell lines harbouring mtDNA mutations (homoplasmic mutations in CYTB and tRNA lysine were tested). When these cells overexpress an aspartate transporter SLC1A3, and cultured in high aspartate medium, their proliferation rate recovers to similar levels as wild-type cells.

Reduced mitochondrial Ca 2+ transients stimulate autophagy in human fibroblasts carrying the 13514A>G mutation of the ND5 subunit of NADH dehydrogenase

http://www.nature.com/cdd/journal/vaop/ncurrent/full/cdd201584a.html

Granatiero V, Giorgio V, Calì T, Patron M, Brini M, Bernardi P, Tiranti V, Zeviani M, Pallafacchina G, De Stefani D, Rizzuto R

Many layers of compensation potentially exist, to compensate for mutations in mitochondrial DNA: mtDNA biogenesis, modifications to transcription, translation, diffusive complementation and so on.

This study aims to determine how one such mutation is compensated for. The authors use primary skin fibroblasts from patients carrying a (phenotypically mild) heteroplasmic point mutation in a mitochondrially-encoded subunit of complex I (13514A>G, affecting the ND5 subunit).

They observed a higher rate of mitochondrial recycling in mutant cells, via mitophagy. In addition, the mutants had a lowered calcium uptake into the mitochondria when exposed to histamine (which causes calcium release from the endoplasmic reticulum into both the cytoplasm and mitochondria), which was associated with higher AMPK levels. Using split GFP technology, they showed that the mutants possessed fewer mitochondrial-ER contact sites.

To determine causality, the authors manipulated mitochondrial calcium uptake by overexpressing components of the mitochondrial calcium uniporter, and also treating cells with Kaempferol. These interventions recovered the calcium response to histamine, reduced AMPK levels and reduced autophagosome number. This intervention, however, reduced cell viability; presumably because of a lowered mitophagy rate.

Thus, the authors propose that, in this system, the cell reduces mitochondrial calcium uptake to increase mitochondrial recycling. They suggest this is mediated via AMPK, which senses the AMP/ATP ratio. This acts as a pro-survival mechanism, in mutated cells.

---------------------------
Thoughts: AMPK correlates with the recycling rate, but it would be interesting to see whether an intervention altering AMPK levels, alters mitophagy rate in the expected way, to demonstrate this causal link?

I wonder whether these cells have lowered respiration and generate less ATP, because of their mutation...? This could then explain their increased AMPK levels, and perhaps also their increased mitophagy rate? Maybe perturbing cells to uptake more calcium, causes more flux through the respiratory chain and so alleviates energy deficiency and lowers mitophagy, but a mutated complex I causes higher levels of ROS to give a pro-apoptosis signal? Interesting system!