Escolar Documentos
Profissional Documentos
Cultura Documentos
2016: A Mitochondria
Odyssey
Catherine Cherry,1,2 Brian Thompson,1,2 Neil Saptarshi,1,2
Jianyu Wu,1 and Josephine Hoh1,*
The integration of the many roles of mitochondria in cellular function and the
contribution of mitochondrial dysfunction to disease are major areas of research.
Within this realm, the roles of mitochondria in immune defense, epigenetics, and
stem cell (SC) development have recently come into the spotlight. With new
understanding, mitochondria may bring together these seemingly unrelated
elds, a crucial process in treatment and prevention for various diseases. In this
review we describe novel ndings in these three arenas, discussing the signicance of the interplay between mitochondria and the cell nucleus in response to
environmental cues. While we optimistically anticipate that further research in
these areas can have a profound impact on disease management, we also bring
forth some of the key questions and challenges that remain.
Thus Spoke Mitochondria
Mitochondria are cellular organelles with important roles in signaling and bioenergetics. They are
surrounded by two membranes, the inner mitochondrial membrane (IMM) and the outer
mitochondrial membrane (OMM). In most cell types, mitochondria are not isolated organelles;
they radiate from the cell nucleus in a reticular network, displaying high levels of interconnectivity
and plasticity facilitating their functional roles within the cell [1].
The eld of biology has come a long way in understanding mitochondria since their discovery
over a century ago [2]. In 2015, the UK made changes to legislation allowing the use of
mitochondrial replacement therapies to help prevent the development of mitochondrial diseases
[3]. Despite rapid progress in mitochondrial biology, little emphasis has been placed on
mitochondrial involvement in epigenetics, SC biology, or immune defense. These three areas
are intricately linked by the functional roles of mitochondria. Consequently, by appreciating this
link we may also improve our understanding of the environmental signals that control gene
function and inuence mitochondrial dysfunction and disease.
This review aims to tie together the recent steps forward in these three underrepresented elds
of mitochondrial biology. In addition, to facilitate the development of strategic approaches to
answer complex questions in these elds, we discuss rapidly evolving technologies and
experimental tools to study mitochondria in great detail. Of clinical relevance, we provide
examples of treatments using mitochondria that are either licensed or currently in development
aiming to treat various pathologies.
Trends
Mitochondria play a pivotal role in the
immune system by detecting foreign
invaders through signaling pathways
(e.g., inammasomes) and generating
immune responses. Modulation of this
role might open up new therapeutic
potential.
Methylation by DNA methyltransferases contributes to the epigenetic
modication of mitochondrial DNA.
Dysregulation of the mitochondrial epigenome within cells has been implicated in various diseases.
Mitochondria contribute to tissue
regeneration and integrity, which are
maintained by stem cell renewal and
differentiation. Stem cells present exciting medical possibilities in regenerative
medicine. Understanding specic mitochondrial biology in stem cells is vital.
Novel techniques are allowing the
study of mitochondria in much greater
detail than before.
Possible new therapeutic avenues are
emerging with increased scientic
knowledge linking mitochondria to
immunity, epigenetics, and stem cell
biology.
1
School of Medicine, Departments of
Environmental Health Science and
Ophthalmology, Yale University, New
Haven, CT, USA
2
These authors contributed equally.
*Correspondence:
Josephine.hoh@yale.edu (J. Hoh).
http://dx.doi.org/10.1016/j.molmed.2016.03.009
2016 Published by Elsevier Ltd.
391
responses are triggered and virally infected cells can be eliminated by mitochondria-driven
apoptosis. In these molecular events, protein-signaling complexes that drive the production of
interferons (IFNs) form active complexes on mitochondria [5]. When present, viral RNA forms a
complex with Rig-1-like receptors (see Glossary) and translocates to the mitochondrial
antiviral signaling protein (MAVS) in the OMM. MAVS forms aggregates in the OMM that can
subsequently activate the key signaling mediators IFN regulatory factor 3 (IRF3) and the
transcription factor nuclear factor kappa B (NF-kB) pathway in the cytoplasm (Figure 1A) [6].
It is increasingly recognized that mitochondrial DNA (mtDNA) and mitochondrial reactive
oxygen species (mtROS) play signicant roles in the cellular immune response. mtDNA released
during Bcl-2-mediated apoptosis can bind to cGMPAMP synthase (cGAS) causing the
generation of cGAMP, which in turn activates stimulator of IFN genes (STING). This results
in the production of IFN (Figure 1B) [5]. Caspase-3, -9, and -7 of the apoptotic caspase cascade
Mature IL-1
producon
RAGE
TLR9
TFAM
Acvaon of NLRP3 inammasome
mtDNA
mtDNA
pDC
(D)
Cell
damage/necrosis
(C)
CRIF1
ROS
LEM
(B)
mtDNA
ATP + GTP
cGAS
cGAMP
STING
MAVS
Viral RNA
(A)
IRF3
NF-B
IRF3/7
RIG1
Producon of
IFNs and
cytokines
Figure 1. Mitochondria and the Immune System. (A) Viral RNA forms a complex with RIG1 and binds to mitochondrial
antiviral signaling protein (MAVS) on the outer mitochondrial membrane (OMM). This then stimulates the nuclear factor
kappa B (NF-kB) and interferon (IFN) regulatory factor 3/7 (IRF3/7) pathways resulting in the production of IFNs and
cytokines. (B) Mitochondrial DNA (mtDNA) released from the mitochondria is a stress signal and can activate the stimulator
of IFN genes (STING) pathway. mtDNA binds to cGMPAMP synthase (cGAS) generating cGAMP, which activates STING.
IRF3 can then induce expression of IFN and other IFN-stimulated genes (ISGs). (C) Transcriptional factor A, mitochondrial
(TFAM) is a mtDNA-binding protein. After cell damage/necrosis TFAM acts as a danger signal and enhances the
plasmacytoid dendritic cell (pDC) response by binding to the receptor for advanced glycation end products (RAGE)
and toll-like receptor 9 (TLR9). (D) CR6-interacting factor (CRIF1) generates reactive oxygen species (ROS) through an
interaction with the lymphocyte expansion molecule (LEM). Mitochondrial ROS (mtROS) stimulate the immune system by
activating the Nod-like receptor family, pyrin domain containing 3 (NLRP3) inammasome pathway, which generates
downstream mature IL-1b.
392
Glossary
Acetyl-CoA: metabolic intermediate
produced during fatty acid
metabolism.
Age-related macular
degeneration (AMD): leading cause
of vision loss in elderly populations. In
the dry form, debris or drusen
accumulates. In the wet form, blood
vessels grow from the choroid.
Diabetic retinopathy: complication
of diabetes affecting the eyes and
leading to vision loss.
Genome-scale analysis: analysis of
genomic features such as DNA
sequence and gene expression over
the whole genome. The genome is
searched for small variations called
SNPs that occur more frequently in
people with a particular disease.
Heteroplasmy: the mix of nonmutated and mutated mtDNA that
can exist in a cell. The level of
heteroplasmy can differ between
cells, tissues, and individuals.
Mammosphere: a clump of human
mammary gland cells.
Mitochondrial DNA (mtDNA):
circular genome inside nucleoids in
the inner mitochondrial membrane
that encodes for 13 proteins and 24
RNA molecules.
Mitochondrial ssion: the process
of two mitochondria separating.
Mitochondrial fusion: joining of two
more mitochondria to form a
network.
MT-RNR1: the mitochondrial gene
that encodes 12s RNA.
Nucleoid architecture: pattern by
which DNA is compacted, folded, or
wrapped.
Oxidative phosphorylation
(OXPHOS): metabolic pathway in
which mitochondria produce ATP.
Rig-1-like receptor: a PRR in the
cytoplasm.
Stemness: common molecular
processes underlying the core SC
properties of self-renewal and the
generation of differentiated progeny.
Superoxide: a compound containing
the anion O2.
can silence immune activation [7,8], which may prevent dying cells from producing IFN. Of note,
loss of the caspase cascade in vivo and in vitro leads to elevated IFN-b levels. With the goal of
silencing immune activation, caspase inhibitors have been included in animal preclinical trials
yielding promising results. For example, VX-765, a selective inhibitor of caspase-1 [9,10], is
currently being investigated. However, none of these caspase inhibitors is currently licensed. Of
note, caspase inhibition can amplify IFN production, which is an interesting concept from a
pharmacological standpoint.
The Nod-like receptor family, pyrin domain containing 3 (NLRP3) inammasome can be
activated by a wide range of ligands including bacterial toxins and PAMPs [11]. NLRP3 enables
the activation of caspase-1 which cleaves IL-1b into its mature form. Experiments in murine bone
marrow-derived macrophages (BMDMs) showed that oxidized mtDNA released into the cytoplasm could bind and activate the NLRP3 inammasome during programmed cell death
(Figure 1D) [12]. Another study using murine BMDMs reported that mtDNA release depended
on the NLRP3 inammasome and mtROS and that mtDNA could further amplify inammasome
signaling (caspase-1 activation) [13]. Of note, the autophagy proteins microtubule-associated
protein-1 light chain 3B (LC3B) and Beclin-1 were required to maintain mitochondrial integrity
[13]. mtDNA is a ligand of the NLRP3 inammasome and this system provides a positive
feedback loop to prolong the activation of the NLRP3 inammasome. Notably, dysregulation of
the NLRP3 inammasome has been associated with many diseases, including Alzheimer's
disease and type 2 diabetes [14]. Hence, besides immune responses and inammation, it is
conceivable that the mechanisms and regulation of mtDNA in the context of inammasome
activation could be applicable to many other diseases.
The mtDNA-binding protein transcriptional factor A, mitochondrial (TFAM) regulates nucleoid
architecture, abundance, and segregation [15]. A TFAM heterozygous (TFAM+/) knockout
mouse line has been shown to display 4060% mtDNA depletion and mild mtDNA repair
defects, which can cause an increase in mtDNA mutations [5]. In TFAM+/ mouse embryonic
broblasts (MEFs), mtDNA stress was induced with lack of TFAM and in the absence of major
oxidative phosphorylation (OXPHOS) defects [5]. This resulted in decreased total mtDNA,
creating larger nucleoids and instigating mitochondrial hyperfusion. mtDNA instability and
mitochondria dysregulation have been observed in many human diseases; hence, cells isolated
from this mouse model can be used to study cellular responses to mtDNA stress in vitro.
Specically, challenging TFAM+/ MEFs with herpes simplex virus 1 or vesicular stomatitis virus
demonstrated that the mice were resistant to infection compared with wild-type (WT) MEFs [5].
Depletion of mtDNA in WT MEFs however, reduced the resistance to viral infection, suggesting
that virally induced mtDNA stress boosted the host's antiviral responses, as evidenced by
induced type 1 IFN and IFN-stimulated gene (ISG) responses [5].
Julian and colleagues [16,17] built on recent evidence suggesting that mtDNA is the principal
regulator of systemic inammation in the immune response [18]. The damage-associated
molecular pattern (DAMP) nuclear DNA-binding high-mobility group box protein1 (HMGB1)
can be secreted by immune cells and act as a mediator of inammation [19]. HMGB1 has been
shown to engage the receptor for advanced glycation end products (RAGE), which in turn
induces cytokine secretion through activation of the transcription factor NF-kB and enhance
responses to CpG DNA in murine plasmacytoid dendritic cells (pDCs) [20]. HMGB1 has also
been shown to direct cell migration of murine mesoangioblasts (mesoderm SCs) in a NF-kBdependent manner [21]. TFAM is a HMGB1 structural homolog. Consequently, it has been
postulated that TFAM engages the RAGE to enhance pDC activation via toll-like receptor 9
(TLR9), as shown in Figure 1C [16,20]. pDCs are antigen-presenting cells that promote immune
responses to self-antigens and to self-DNA released from necrotic cells. In these studies,
exposure to TFAM alone did not activate pDCs but did, however, amplify type 1 IFN and tumor
393
necrosis factor alpha (TNF/) responses to CpG-DNA in cultured splenocytes [16,17]. Consequently, TFAM-mediated stimulation of pDCs enhanced their cytokine responses to CpG DNA,
suggesting a strong link between mitochondria and immune activation stemming from necrotized cells. These ndings may have strong implications in pathological conditions where
necrotic or apoptotic cells are present and can trigger an immune response.
From another angle, T lymphocyte proliferation can increase with changes in metabolic respiration and ROS production, which are critical processes in mitochondrial function [22]. Compared
with resting T cells, activated T cells have a different metabolic program that includes the ability to
adapt to changing environments, as has been shown in numerous studies. For instance, one
report identied AMP-activated protein kinase (AMPK) as a metabolic checkpoint that regulates
T cell adaptation and maintains cell viability [23]. In addition, a recently identied mutation in the
lymphocyte expansion molecule (LEM) has been shown to impact T cell immunity and to
modulate mitochondrial function. In one study, LEM mutations were identied via high-throughput exome sequencing [24] in lymphocytic choriomeningitis virus clone13 (LCMV Cl13)-infected
mice (Retro strain). Specic mutations enhanced the production of LCMV-specic cytotoxic
CD8+ T cells (CTLs) as well as long-lived memory T cell numbers [25]. Moreover, LEM in CTLs
was shown to interact with CR6-interacting factor (CRIF1), a protein needed for the translation
and insertion of OXPHOS peptides into the IMM [25,26]. Presumably, LEM interacts with the
OXPHOS protein CRIF1 to increase the levels of mtROS (Figure 1D). The discovery of LEM and
the insights into its role further implicate mitochondria in immunity, in both effector and memory
T cell responses. Whether upregulation of LEM in the mouse Retro strain (where the phenotype
was bred to heterozygosity) can restore CTL immunity and enhance memory in different
contexts may prove to be an exciting opportunity to explore future therapeutic avenues.
Recently, another interesting link between mitochondria and immunity has been reported.
Shahni and colleagues identied signal transducer and activator of transcription 2 (STAT2)
as an activator of mitochondrial ssion [27]. A mutation in STAT2 resulted in complete loss of
expression, leading to severe multiorgan dysfunction with impaired mitochondrial ssion in
three human patients who had received the live-attenuated mumpsmeaselsrubella (MMR)
vaccination: two were siblings presenting with neurological deterioration and one was a STAT2decient patient. The STAT2 deciency had not caused symptoms until exposure to viral
challenge in this patient (the MMR vaccination). Furthermore, in patient broblasts there was
decreased expression of activated dynamin-related protein 1 (DRP1) and increased expression
of inactive DRP1, which the authors deemed responsible for the observed hyperfused, elongated mitochondria [27]. The authors recapitulated this effect by silencing STAT2 in SHSY5Y
neuroblastoma cells, while STAT2 overexpression rescued the phenotype [27]. The link between
mitochondrial dynamics and immunization (memory responses) described here suggests that
disruption of the JAKSTAT signaling pathway may impair mitochondrial dynamics and function
and could potentially provide clues to why patients with mitochondrial diseases are susceptible
to viral infections. Together these examples further illustrate the interconnected nature of
mitochondrial biology in various cellular processes and host responses.
Mitochondrial Epigenetics
Epigenetics the study of heritable changes in gene expression that do not alter DNA sequences
is a major eld of investigation in mitochondrial biology. Epigenetics can determine the
expression of nucleus-encoded genes in accordance with environmental cues. Of relevance,
an increasing number of disorders and complex phenomena such as aging have been associated with mitochondrial dysfunction and epigenetics. Cytosine methylation is an epigenetic
modication of DNA catalyzed by DNA methyltransferases (DNMTs) leading, in principle, to
transcriptional silencing. Epigenetic studies in mitochondrial biology have been mostly focused
on the transcriptional control or modication of nuclear DNA (nDNA), since DNMTs were
394
mtDNA
CpG
(A)
CpG
CpG
DNMTs
(B)
CpG
CpG
Figure 2. Epigenetic Modications of Mitochondrial DNA (mtDNA). (A) Two examples of mtDNA CpG islands in hypermethylated states are shown. DNA
methyltransferases (DNMTs) such as DNMT3a or DNMT1 methylate the 440 identied mtDNA islands. Methyl groups are indicated by a closed circle whereas open
circles represent unmethylated mtDNA. Hypermethylation has been implicated in cancer, amyotrophic lateral sclerosis (ALS), diabetic retinopathy, and the response to
environmental toxicant exposure. (B) Two examples of mtDNA CpG islands in hypomethylated states are shown. Methyl groups are indicated by a closed circle whereas
open circles represent unmethylated mtDNA. Hypomethylation of mtDNA has been detected in patients with, for example, Down's syndrome.
originally thought to be unable to access mitochondria in vertebrates. Also, since mtDNA does not
contain histones, it was thought that mtDNA could not be epigenetically modied, as depicted in
Figure 2 [28]. Moreover, early attempts to utilize mtDNA methylation as a biomarker for cancer
detection across 15 cancer cell lines from patients with gastric and colorectal cancer were hindered
by the lack of DNA methylation reported across all samples tested. Direct sequencing conrmed an
absence of methylated mtDNA, further describing mtDNA methylation as a rare event [29].
However, nDNA modications could not accurately represent the whole picture when considering
overall cellular gene regulation. Thus, several epigenomic hypotheses have been formed that take
into account the regulation and crosstalk of both nDNA and mtDNA in modulating cell function.
The rst breakthrough suggesting a role for epigenetics in mitochondria came from the
identication and characterization of DNMT1 in mitochondria from MEFs and HCT116 human
colon carcinoma cells [30], where immunoprecipitation against 5-methylcytosine (5-mc) or
5-hydroxymethylcytosine (5-hmc) demonstrated signicant enrichment compared with IgG
controls [30]. Mitochondrial DNMT1 identication subsequently gave way to the identication
of DNMT3a, which also localized inside mouse and human central nervous system mitochondrial
fractions [31].
Using bisulte genomic sequencing and next-generation sequencing on mtDNA regions from
human HEK293 and HCT116 cell lines, a study reported that the overall CpG island methylation
frequency was less than 0.1% [32]. This called into question both the methods involved in the
detection of CpG methylation and the overarching physiological signicance of mtDNA methylation in epigenetic regulation. However, the pathophysiological relevance of methylation levels
in disease etiology cannot be ignored. Recent evidence demonstrated a signicant increase in
5-mc and decrease in DNMT3a levels in spinal cord neurons and skeletal muscle myobrils from
transgenic murine amyotrophic lateral sclerosis (ALS) models [33]. The salient message from this
395
study was that mtDNA methylation was tissue specic and could contribute to the degree of
tissue inammation seen in ALS pathology in mice [33]. Mitochondrial genome-scale analysis
has provided a platform where large-scale bisulte sequencing can be mapped to the human
mitochondrial genome and methylation patterns ascertained with an expected degree of
methylation heterogeneity across 39 cell line publicly available datasets [34]. Such studies
suggest that epigenetic modication of mtDNA is more prevalent than previously thought.
When more specically considering disease, DNMT1 hypermethylation might play a role in the
pathogenesis of diabetic retinopathy [35]. In patients with diabetes, nucleus-encoded DNMT1
is translocated into retinal mitochondria, hypermethylating the mtDNA control (D-loop) region
where transcription and replication elements are located [35]. Hypermethylation of this region
causes aberrant transcription of mitochondrial genes crucial to the regulation of the electron
transport chain, thus leading to the generation of superoxide radicals promoting a hyperglycemic superoxide radical milieu [35]. This nding underscores the importance of mtDNA
methylation outside classical CpG sites and highlights the regulatory role of epigenetic modications in mitochondria that can contribute to disease [35,36].
Anecdotally, environmental exposure to toxicants has been shown to have a major impact not
only on nDNA but potentially on mtDNA as well. For example, workers highly exposed to
airborne pollutants (e.g., metals, trafc-derived particles, benzene) have been reported to exhibit
increased mtDNA methylation in the 12S rRNA region (MT-RNR1) compared with workers
exposed to low levels of airborne pollutants [37]. Of signicance, aberrant methylation of
MT-RNR1 could lead to aberrant mitochondrial ribosome function and protein production
[37]. Although further validation is required in these studies, impaired mitochondrial protein
production stemming from epigenetic changes in MT-RNR1 presumably might lead to environment-associated pathologies (e.g., cancer, lung disease).
With continued advances in the understanding of the regulatory role of the mitochondrial
epigenome in mitochondrial function, a novel layer of regulatory crosstalk between the nucleus
and the mitochondrion is emerging. A mitochondria-to-nucleus pathway, shown in Figure 3, can
transmit signals from mtROS to the nucleus and modulate gene expression. An example of this
process has been reported with the inactivation of the histone demethylase Rph1p at subtelomeric heterochromatin [38]. In this study, mtROS signaled through Tel1p and Rad53p
(homologs of the mammalian DNA damage response kinases ATM and Chk2) to ensure yeast
longevity. This pathway subsequently inactivated Rph1p leading to transcriptional silencing of
telomeric genes [38]. Moreover, another study has provided evidence linking mtDNA to mitochondrial metabolites to regulate nuclear gene expression in skeletal muscle SCs (SMSCs) [39].
This work demonstrated that SMSCs undergo a metabolic switch from fatty acid oxidation to
glycolysis (transpiring in mitochondria) when transitioning from quiescence to proliferation. This
led to decreases in both intracellular NAD+ levels and the activity of the histone deacetylase
sirtuin 1 (SIRT1), resulting in elevated histone H4K16 acetylation and activation of muscle
gene transcription [39]. Therefore, such changes in metabolic state can inuence metabolites
and the epigenetic regulation of gene expression [39].
Bidirectional regulation of gene expression between nDNA and mtDNA presents an additional
layer of complexity with the presence of miRNA. miRNAs that reside on both the OMM and IMM
can affect epigenetic regulation and in turn be themselves epigenetically regulated. Early studies
revealed the presence of 15 nucleus-encoded miRNAs present in mitochondria of murine liver
tissues [40]. This initially raised the question of the function of miRNAs in mitochondria, bringing
forth the possibility of an alternative mechanism of nuclear control of mtDNA function. The
miRNA miR-1 has been found to enter mitochondria and stimulate the translation of mtDNA in
muscle cells [41]. In addition, miRNAmtDNA crosstalk was also suggested in a rat model of
396
Cytosol
Nucleus
mtROS
Rph1p
NAD+
Sirt1
miRNA
traumatic brain injury (TBI) [42]. Animals with TBI a leading cause of cognitive defects in
humans exhibited mitochondrial dysfunction and dysregulation of a set of miRNAs expressed
in the hippocampus region of the brain [42]. Collectively, these studies are timely, reinforcing an
epigenetic coregulatory role of nDNA, mtDNA, and, presumably, miRNA. However, whether
miRNA dysregulation is a cause or a result of mitochondrial dysfunction in these contexts
remains to be determined. Filling this gap in understanding should be a major focus of future
research.
The epigenetics of both nDNA and mtDNA are clearly important regulatory processes for proper
cell function. Studies currently positioned to obtain a better understanding of mtDNA epigenetic
modications and the crosstalk between the two epigenomes are now at the forefront of
biomedical research.
Mitochondria in SC Biology
SCs have become one of the most potentially promising therapeutic avenues for regenerative
medicine. Recent studies in SC research have demonstrated that SCs isolated from human
blastocysts not only show conventional hallmark characteristics of naive pluripotency but also
exhibit additional functional features such as mitochondrial respiration [43]. SCs have two
dening qualities: self-renewal by the production of identical daughter cells and the ability to
produce independent daughter cells that can differentiate into many different cells. Researchers have faced multiple challenges associated with studying SCs, some of which have been
overcome with the development of iPSCs. This methodology is progressively allowing the
scientic community to understand various complex factors involved in regulating the maintenance of SCs, with the role of mitochondria just beginning to be woven into this complex
picture.
397
Asymmetric division allows SCs to generate daughter cells with differing fates [44]. There is
evidence suggesting that damaged proteins are inherited asymmetrically [45] and this trend has
also been observed in Saccharomyces cerevisiae [46]. However, there is limited supportive
evidence to indicate that mitochondria asymmetrically divide in mammalian systems. To address
this, recent work was conducted in SCs using tag techniques. Photoactivatable GFP was
tagged to MOM protein 25 (paGFP-Omp25) and Snap-tag was used to track apportioned
mitochondria in daughter cells originating from stem-like cells [44] recently identied from
immortalized human mammary cells [47]. The data indicated that mitochondria were not evenly
distributed among daughter cells. Interestingly, through uorescence-activated cell sorting
(FACS) and replating of daughter cell populations it was observed that daughter cell populations
that received younger mitochondria displayed stronger stem-like characteristics such as a
mammosphere-forming ability (Figure 4). This study illustrates the power of controlling the
apportioning of mitochondria into daughter cells [44].
SC differentiation is a tightly regulated process that is crucial for both animal development and
tissue homeostasis [48]. However, little is known about the intrinsic cellular mechanisms
governing this process. One recent study using in vivo RNAi in Drosophila melanogaster
discovered that mitochondrial ATP synthase, a protein that chemically synthesizes ATP from
ADP and Pi [49], plays an important role in regulating germ cell differentiation and ensuring
germline development [48]. Furthermore, knockdown studies of other members of the OXPHOS
system demonstrated that ATP synthase acts during differentiation through a mechanism that is
separate from its role in OXPHOS. ATP synthase expression was observed to be specically
regulated during differentiation [48]. With the use of electron micrographs, native polyacrylamide
Young
Mitochondria
Division 0
Old
Division 1
Division 2
Division 3
Figure 4. Mitochondria in Stem Cell Differentiation. Asymmetrically dividing stem cells acquire young (yellow) and old
(blue) mitochondria. Stem cells that acquire more young mitochondria have an enhanced proliferation advantage over stem
cells that acquire a greater number of old mitochondria. This has important implications in senescence.
398
gel electrophoresis, and in-gel ATPase assays, it was demonstrated that ATP synthase dimerization is required for mitochondrial crista formation during differentiation [48]. Combining these
ndings with previous results demonstrating differences between SCs and differentiated cells
(e.g., cardiomyocytes, follicle cells) in mitochondrial fusion and ssion as well as IMM
structure, [50,51], it is becoming increasingly apparent that mitochondria could play important
roles in the regulation of SC differentiation and function.
Reinforcing these ndings, recent research has suggested that mitochondrial proteins such
as mitofusion-1 and -2 (Mfn1/2), which are known to be intimately involved in controlling
mitochondrial dynamics and energy production, play an extensive role in regulating cell fate
transition [52]. It was observed that during the early stages of reprogramming, around day 7,
mitochondrial function was downregulated with an associated decrease in Mfn1/2 levels.
Mfn1/2 genetic ablation or pharmacological inhibition of mitochondrial fusion in both human
ESCs and mouse MEFs resulted in reprogramming changes and, furthermore, demonstrated
a glycolytic bioenergetic transition to meet the energy demands of proliferating pluripotent
cells [52]. Other research has shown that mitochondrial uncoupling protein 2 (UCP2), a
protein that regulates mitochondrial respiration by controlling metabolite transportation out of
mitochondria [53], appears to play an important role in the regulation of SC differentiation by
blocking the shift from glycolysis to cellular respiration [54]. Furthermore, recent work has
shown that in both human and mouse ESCs, bioenergetics processes, namely glycolysis, are
crucial in the maintenance of the pluripotent state [55]. This study employed high-resolution
NMR and 13C glucose-tracing using mass spectrometry in pluripotent SCs to document that
glycolysis-mediated changes in acetyl-CoA occurred with differentiation (decreased glycolysis) and also led to H3K9/K27 histone acetylation [55]. In addition, increases in ROS
production have been associated with SC aging or loss of regenerative capacity [56],
suggesting that ROS production may play a regulatory role in stemness and SC proliferation.
Taken together, these results demonstrate that through the control of mitochondrial dynamics
and bioenergetics, novel approaches to promoting somatic cell reprogramming may be
obtained.
iPSCs can be created from somatic cells through forced expression of reprogramming
factors. iPSCs have different gene expression patterns [57], differentiation potentiality [58],
and DNA methylation patterns [59] compared with ESCs. These differences have led
researchers to develop a technique known as somatic cell nuclear transfer (SCNT). SCNT
allows the transfer of a somatic cell nucleus into an oocyte with subsequent reprogramming to
convert it into a pluripotent cell [60]. Recent evidence has emerged showing that mouse ESCs
with different mtDNA haplotypes display differential expression of genes associated with DNA
methyltransferases and processes of energy metabolism and pluripotency [61]. However, a
problem surfaces: mismatching mitochondria between donor and recipient during ESC
nuclear transfer (NT-ESC) might lead to immunoreactivity [62]. For example, mouse NT-ESCs
were generated with mismatched C57BL/6J mitochondria and BALB/c nuclei. On injection of
mismatched NT-ESCs into BALB/c mice, there was an increase in helper T cell activation in
addition to NT-ESC-directed antibody production [62]. This result poses a challenge to the
development of SCNT therapy as the heterogeneity level in human mitochondria is higher than
that in mice.
399
Application
Principle
Refs
Mitoash
Measurement of superoxide
production
[7981]
MitoParaquat
Measurement of superoxides
[82]
Seahorse Bioscience
Measures extracellular ux in
living cells
[83]
Gradient centrifugation
Mitochondrial purication
[8487]
LCESI-MS/MS
[88,89]
mitochondria for the quantication of mtDNA methylation and mapping of 5-mC and 5-hmC.
Moreover, existing nDNA tools are providing exciting prospects that can be applied to mtDNA.
Bisulte sequencing and liquid chromatographyelectrospray ionization tandem mass spectrometry (LCESI-MS/MS) (Table 1) as well as afnity methods and restriction methods are being
used, but the best way seems to involve a combination of several methods. For instance, the
combination of bisulte sequencing and methylated mtDNA immunoprecipitation assays has
accurately shown methylated and hydroxymethylated cytosines in mtDNA [34,36].
This progression is leading towards novel therapeutics and improvements of existing treatments
that may achieve what was previously inconceivable. Summarized in Table 2 are examples of
promising new treatments that utilize our knowledge of mitochondria: MitoC, phenformin,
mitochondria-targeted transcription activator-like effector nucleases (mitoTALENs), and SBI0206965. In particular, compelling research is being conducted into developing mitochondrial
replacement therapies (MRTs). It is hoped that with the combined utilization of in vitro fertilization
and MRTs, mutated maternal mtDNA and unhealthy mitochondria can be replaced with
unmutated donor mtDNA and healthy mitochondria. Macaque- and human-based studies have
demonstrated that MRT may be a viable mitochondrial disease prevention strategy [63,64]. A
study with human oocytes has shown that following nuclear genome exchange, the swapped
pluripotent cells and derived broblasts exhibited normal metabolic proles and respiratory chain
enzyme activity compared with ESCs and ESC-derived broblast controls [63]. Another study
showed that replacing mutant mtDNA with healthy mtDNA using spindlechromosomal complex transfer (ST) gave rise to healthy rhesus macaque monkeys [64]. To improve fertility
potential, AUGMENTSM treatment (Table 2) has been marketed. This involves the transfer of
healthy energy-producing mitochondria (AUGMENTSM processed) from a woman's own progenitor egg cells taken from the ovarian lining (EggPCs) into her mature egg cells in combination
with sperm during in vitro fertilization procedures; it is licensed in only some countries [65].
Despite these promising results, there has been much resistance towards using MRTs. Many
opponents of MRT cite that studies performed on invertebrates and mice have demonstrated
altered parameters of health such as energy production, fertility, and learning [6668]. After
taking into consideration the risks of MRT, in early 2015 the UK was the rst country to change
legislation allowing the use of MRT on parents who want to conceive a child but are at risk for
having a child with a mitochondrial disease [3]. The lessons learned from these earlier pioneering
studies will be pivotal in setting the course for MRT-granting legislation to be passed on to other
nations.
400
Outstanding Questions
Name
Type
Mechanism
Application
Refs
MitoC
Antioxidant
(ascorbate)
conjugated
to TPP
Mitochondria-targeted
antioxidant and tool to
explore the role of
ascorbate in mitochondria
[90]
Mitochondrial
inhibitor
Metabolism-based
therapeutic for
LKB1-decient tumors
[91]
SBI-0206965
ULK1 kinase
inhibitor
[92,93]
AUGMENTSM
treatment
Mitochondrial
transfer
Transfer of mitochondria
from a woman's own
immature EggPCs to
supplement the existing
mitochondria in her
mature eggs
http://www.
augmenttreatment.com
[65]
Targeted to mtDNA
mutation
Keeping heteroplasmy
below threshold levels
[94,95]
Phenformin
mitoTALENs
Nuclease
We came to the eld of mitochondrial biology through genetic studies of one of the most
prevalent eye diseases in aging populations, age-related macular degeneration (AMD)
[6972]. Mitochondrial dysfunction had already been documented in age-related diseases
including AMD [7274]. Further studies on the same AMD-associated gene family implicated
the mitochondrial protein high temperature-dependent serine peptidase 2 (HtrA2) in AMD
disease [7578]. As mentioned above, MRT is currently being used to eliminate dysfunctional
mitochondria in rare inherited diseases. For common disorders, a new theme has emerged from
recent work: mitochondria can play a pivotal role in immunity, epigenetic regulation, and SC
development. Deciphering the interplay between mitochondria and nuclear processes will be
critical in understanding the mitochondrial role in cellular function in these three areas in health
and disease. From a public health point of view, it will be signicant to follow these lines of
investigation, potentially providing further clues to the participation of mitochondria in mediating
responses to environmental cues, infection, tissue transplantation, aging, and cellular dysfunction as in the case of autoimmune disorders and neurodegenerative diseases, among others.
Answers to some of these queries (see Outstanding Questions) may yield novel approaches to
better manage or prevent severe disease and/or to facilitate tissue regeneration strategies.
When contemplating the treatment of mitochondria-related disorders, the most important task is
to understand how to better translate our experimental knowledge to patients and human
populations while accounting for heteroplasmy within individuals. Despite recent signicant
progress, scientists need to continue to focus on the multiple hurdles and challenges that remain
ahead and that need to be overcome.
Acknowledgments
The authors are grateful to Professor Steve Waxman for the opportunity and for advice in writing this review. They are also
grateful to the anonymous reviewers for their input and comments. This work is funded by the Sackler Foundation and the
Rosebay Medical Foundation.
References
1. Burte, F. et al. (2015) Disturbed mitochondrial dynamics and
neurodegenerative disorders. Nat. Rev. Neurol. 11, 1124
401
37. Byun, H.M. et al. (2013) Effects of airborne pollutants on mitochondrial DNA methylation. Part. Fibre Toxicol. 10, 18
38. Schroeder, E.A. et al. (2013) Epigenetic silencing mediates mitochondria stress-induced longevity. Cell Metab. 17, 954964
14. Lamkan, M. and Dixit, V.M. (2012) Inammasomes and their roles
in health and disease. Annu. Rev. Cell Dev. Biol. 28, 137161
15. Kasashima, K. et al. (2011) Human mitochondrial transcription
factor A is required for the segregation of mitochondrial DNA in
cultured cells. Exp. Cell Res. 317, 210220
16. Julian, M.W. et al. (2013) Mitochondrial transcription factor A, an
endogenous danger signal, promotes TNF/ release via RAGEand TLR9-responsive plasmacytoid dendritic cells. PLoS ONE 8,
e72354
17. Julian, M.W. et al. (2012) Mitochondrial transcription factor A
serves as a danger signal by augmenting plasmacytoid dendritic
cell responses to DNA. J. Immunol. 189, 433443
18. Zhang, Q. et al. (2010) Circulating mitochondrial DAMPs cause
inammatory responses to injury. Nature 464, 104107
22. van der Windt, G.J. and Pearce, E.L. (2012) Metabolic switching
and fuel choice during T-cell differentiation and memory development. Immunol. Rev. 249, 2742
47. Chaffer, C.L. et al. (2011) Normal and neoplastic nonstem cells can
spontaneously convert to a stem-like state. Proc. Natl. Acad. Sci.
U.S.A. 108, 79507955
23. Blagih, J. et al. (2015) The energy sensor AMPK regulates T cell
metabolic adaptation and effector responses in vivo. Immunity 42,
4154
48. Teixeira, F.K. et al. (2015) ATP synthase promotes germ cell
differentiation independent of oxidative phosphorylation. Nat. Cell
Biol. 17, 689696
49. Walker, J.E. (2013) The ATP synthase: the understood, the uncertain and the unknown. Biochem. Soc. Trans. 41, 116
25. Okoye, I. et al. (2015) T cell metabolism. The protein LEM promotes CD8+ T cell immunity through effects on mitochondrial
respiration. Science 348, 9951001
50. Kasahara, A. et al. (2013) Mitochondrial fusion directs cardiomyocyte differentiation via calcineurin and Notch signaling. Science
342, 734737
26. Kim, S.J. et al. (2012) CRIF1 is essential for the synthesis and
insertion of oxidative phosphorylation polypeptides in the mammalian mitochondrial membrane. Cell Metab. 16, 274283
51. Mitra, K. et al. (2012) DRP1-dependent mitochondrial ssion initiates follicle cell differentiation during Drosophila oogenesis. J. Cell
Biol. 197, 487497
27. Shahni, R. et al. (2015) Signal transducer and activator of transcription 2 deciency is a novel disorder of mitochondrial ssion.
Brain 138 (Pt 10), 28342846
402
55. Moussaieff, A. et al. (2015) Glycolysis-mediated changes in acetylCoA and histone acetylation control the early differentiation of
embryonic stem cells. Cell Metab. 21, 392402
76. Lesage, S. et al. (2016) Loss of VPS13C Function in AutosomalRecessive Parkinsonism Causes Mitochondrial Dysfunction and
Increases PINK1/Parkin-Dependent Mitophagy. Am. J. Hum.
Genet. 98, 500513
59. Lister, R. et al. (2011) Hotspots of aberrant epigenomic reprogramming in human induced pluripotent stem cells. Nature 471,
6873
61. Kelly, R.D. et al. (2013) Mitochondrial DNA haplotypes dene gene
expression patterns in pluripotent and differentiating embryonic
stem cells. Stem Cells 31, 703716
62. Deuse, T. et al. (2015) SCNT-derived ESCs with mismatched
mitochondria trigger an immune response in allogeneic hosts. Cell
Stem Cell 16, 3338
63. Paull, D. et al. (2013) Nuclear genome transfer in human oocytes
eliminates mitochondrial DNA variants. Nature 493, 632637
81. Cheng, H. et al. (2014) Cheng et al. reply. Nature 514, E14E15
65. Woods, D.C. and Tilly, J.L. (2015) Autologous Germline Mitochondrial Energy Transfer (AUGMENT) in Human Assisted Reproduction. Semin. Reprod. Med. 33, 410421
86. Clayton, D.A. and Shadel, G.S. (2014) Isolation of mitochondria from
tissue culture cells. Cold Spring Harb. Protoc. Published online
October 1, 2014. http://dx.doi.org/10.1101/pdb.prot080002
66. Sackton, T.B. et al. (2003) Cytonuclear coadaptation in Drosophila: disruption of cytochrome c oxidase activity in backcross genotypes. Evolution 57, 23152325
94. Bacman, S.R. et al. (2013) Specic elimination of mutant mitochondrial genomes in patient-derived cells by mitoTALENs. Nat.
Med. 19, 11111113
74. Ma, B. et al. (2016) Simultaneous determination of 8-oxo-20 deoxyguanosine and 8-oxo-20 -deoxyadenosine in human retinal
DNA by liquid chromatography nanoelectrospray-tandem mass
spectrometry. Sci. Rep. 6, 22375
403