Article

ISSN 0006-2979, Biochemistry (Moscow), 2025, Vol. 90, No. 12, pp. 1929-1943 © The Author(s) 2025. This article is an open access publication.

Published in Russian in Biokhimiya, 2025, Vol. 90, No. 12, pp. 2032-2048.

1929

REVIEW

ATP in Mitochondria: Quantitative Measurement,

Regulation, and Physiological Role

Anna S. Lapashina

1,2

, Danila O. Tretyakov

1,2

, and Boris A. Feniouk

1,2,a

Faculty of Bioengineering and Bioinformatics, Lomonosov Moscow State University,

119234 Moscow, Russia

A. N. Belozersky Institute of Physico-Chemical Biology, Lomonosov Moscow State University,

119992 Moscow, Russia

e-mail: feniouk@belozersky.msu.ru

Received September 23, 2025

Revised November 18, 2025

Accepted November 20, 2025

Abstract— Oxidative phosphorylation in mitochondria is the main source of ATP in most eukaryotic cells.

Concentrations of ATP, ADP, and AMP affect numerous cellular processes, including macromolecule biosynthe-

sis, cell division, motor protein activity, ion homeostasis, and metabolic regulation. Variations in ATP levels

also influence concentration of free Mg

, thereby extending the range of affected reactions. In the cytosol,

adenine nucleotide concentrations are relatively constant and typically are around 5  mM  ATP, 0.5  mM  ADP,

and 0.05 mM AMP. These concentrations are mutually constrained by adenylate kinases operating in the

cytosol and intermembrane space and are further linked to mitochondrial ATP and ADP pools via the ad-

enine nucleotide translocator. Quantitative data on absolute adenine nucleotide concentrations in the mito-

chondrial matrix are limited. Total adenine nucleotide concentration lies in the millimolar range, but the

matrix ATP/ADP ratio is consistently lower than the cytosolic ratio. Estimates of nucleotide fractions show

substantial variability (ATP 20-75%, ADP 20-70%, AMP 3-60%), depending on the organism and experimen-

tal conditions. These observations suggest that the ‘state 4’ – inhibition of oxidative phosphorylation in the

resting cells due to the low matrix ADP and elevated proton motive force that impedes respiratory chain

activity – is highly unlikely in vivo. In this review, we discuss proteins regulating ATP levels in mitochon-

dria and cytosol, consider experimental estimates of adenine nucleotide concentrations across a range of

biological systems, and examine the methods used for their quantification, with particular emphasis on the

genetically encoded fluorescent ATP sensors such as ATeam, QUEEN, and MaLion.

DOI: 10.1134/S0006297925603338

Keywords: mitochondria, ATP, ADP, ATeam, adenine nucleotide translocator (ANT), ATP synthase

* To whom correspondence should be addressed.

INTRODUCTION

Mitochondria are traditionally called the pow-

erhouses of eukaryotic cells, as one of their main

functions is ATP production through oxidative phos-

phorylation. Most biochemistry textbooks explain

that the majority of ATP is synthesized in the mito-

chondrial matrix from ADP and inorganic phosphate

by the H

-transporting ATP synthase (also known as

, ComplexV, F-ATPase, or F

-ATPase) located in

the inner mitochondrial membrane. ATP synthesis is

driven by the proton motive force (pmf) across the

inner mitochondrial membrane. The pmf is generated

by the respiratory-chain enzymes that oxidize NADH

and succinate produced in the tricarboxylic acid

(TCA) cycle, and, depending on the organism or tis-

sue, by the oxidation of additional substrates such as

glycerol-3-phosphate. The adenine nucleotide translo-

cator (ANT; also known as the ATP/ADP carrier, ATP/

ADP antiporter, AAC, or ATP/ADP translocase) exports

ATP from the matrix into the intermembrane space

in exchange for ADP, whereas inorganic phosphate

enters the matrix via the phosphate carrier (PiC,

encoded by the SLC25A3 gene in humans and MIR1

LAPASHINA et al.1930

BIOCHEMISTRY (Moscow) Vol. 90 No. 12 2025

in baker’s yeast). Nucleotide and phosphate exchange

between the cytosol and the intermembrane space

occurs through the mitochondrial porins – voltage-

dependent anion channels (VDAC) – located in the

outer mitochondrial membrane.

This simplified outline captures the basic prin-

ciples of oxidative phosphorylation, but it is insuffi-

cient for understanding the regulatory mechanisms

governing the process or clarifying how its dysfunc-

tion is related to various pathologies.

Decades of in  vitro research have generated an

extensive body of data on the functions of individual

proteins involved in oxidative phosphorylation and

the mechanisms that regulate them. Recent advanc-

es in microscopy have further accelerated structural

studies of mitochondria. Notably, cryo-electron mi-

croscopy has recently enabled high-resolution (2.5  Å)

visualization of mammalian respiratory-chain super-

complexes in their native cellular environment, re-

vealing previously unknown aspects of their organi-

zation and interactions [1]. This is only one of many

studies that have revealed the structure of mitochon-

drial proteins, their relative positions, dynamics, and

roles in the formation of cristae and contacts with

the endoplasmic reticulum. The body of experimental

data on the structure and localization of mitochondri-

al proteins in vivo is also growing rapidly.

Despite the considerable progress in mitochon-

drial biology, surprisingly little is known about the

precise conditions under which oxidative phosphory-

lation proceeds in vivo. We have very little quantita-

tive data on the absolute concentrations of ATP, ADP,

and AMP in the mitochondrial matrix, on how these

concentrations vary under different physiological and

pathological states, and on how such changes relate

to the pmf or to reactive oxygen species (ROS) pro-

duction. Because ATP and ADP are exchanged rapid-

ly between the matrix and the cytosol, mitochondrial

nucleotide concentrations modulate cytosolic pools

and are themselves shaped by them. Thus, shifts in

the intramitochondrial ATP/ADP/AMP ratio influence

both mitochondrial enzyme function and metabolic

processes in the cytosol.

Moreover, changes in ATP concentration in both

the cytosol and the mitochondrial matrix affect cel-

lular physiology by altering free magnesium levels.

ATP binds Mg

roughly an order of magnitude more

strongly than ADP. In solutions with an ionic strength

of 0.1-0.15  M, the logarithms of the magnesium bind-

ing constants are approximately 4.0-4.6 for ATP,

3.1-3.5 for ADP, and 1.9-2.1 for AMP  [2]. Therefore,

ATP hydrolysis to ADP increases the concentration

of free magnesium, which in turn influences numer-

ous cellular processes, including those that are not

directly sensitive to the changes in nucleotide levels.

A notable example is chromatin condensation during

mitosis, which is triggered by the rise in cytosolic

free magnesium from ~0.5 to 1  mM. This increase oc-

curs in parallel with the drop in ATP concentration,

and it has been proposed that the ATP decline results

from the elevated ATP hydrolysis rate [3].

An example of a magnesium-dependent process

in mitochondria is the A/D transition of complex I

in the mammalian respiratory chain. Under hypoxic

conditions, complex I shifts from the active  (A) to the

de-activated  (D) state, and upon restoration of respi-

ration, the reverse transition normally occurs within

minutes. However, in the presence of 1-5 mM Mg

this reactivation is slowed by several orders of mag-

nitude [4]. Given that the free magnesium concentra-

tion in the matrix exceeds that in the cytosol, this

phenomenon is likely to contribute to the regulation

of mitochondrial recovery from ischemia in  vivo.

In this review, we have aimed to summarize the

available literature on adenine nucleotide concen-

trations in the cytosol and mitochondrial matrix of

eukaryotic cells, as well as the methods used to mea-

sure ATP levels within mitochondria.

PROTEINS AFFECTING ATP CONCENTRATION

IN MITOCHONDRIA

The two main ‘players’ controlling the dynamics

of ATP concentration in the mitochondrial matrix are

ATP synthase [5] and adenine nucleotide transloca-

tor ANT [6], both located in the inner mitochondrial

membrane. Each of these proteins accounts for a sig-

nificant proportion of the total mitochondrial mem-

brane protein – up to 12% for ANT [6] and up to

20%for ATP synthase [7, 8]. Analysis of the mitochon-

drial proteome of the HEK293T cell line derived from

human embryonic kidneys showed that, in terms of

the number of molecules per cell, ANT (all isoforms)

is the most abundant mitochondrial protein, with

more than 8 million molecules per cell, while the

number of ATP synthase molecules reaches nearly

2 million per cell [9]. Only about 20 mitochondrial

proteins exceed ATP synthase in copy number. These

include, besides ANT, chaperones HSP60, HSP10, and

mt-HSP70, peroxiredoxin PRDX3; membrane chan-

nels such as the outer membrane porins VDAC1 and

VDAC3; several enzymes of the central carbon me-

tabolism such as malate dehydrogenase (MDH2) and

citrate synthase (CS) from the TCA cycle; phosphate

transporter SLC25A3; and several proteins involved

in mitochondrial translation and protein import [9].

ATP synthase can catalyze not only ATP synthe-

sis but also ATP hydrolysis. It thus links the activity

of the respiratory chain enzymes, membrane pro-

ton conductivity, and the ATP/ADP ratio in the ma-

trix. When pmf decreases due to increased proton

ATP IN MITOCHONDRIA 1931

BIOCHEMISTRY (Moscow) Vol. 90 No. 12 2025

conductivity or reduced activity of respiratory chain

enzymes, ATP synthase switches from synthesizing

to hydrolyzing ATP, pumping protons from the mito-

chondrial matrix into the intermembrane space and

preventing depolarization of the inner mitochon-

drial membrane. If pmf is absent, the enzyme may

exhibit high ATPase activity: it can hydrolyze from

several hundred to thousands of ATP molecules per

second [10].

Under the low-pmf conditions, the ATPase activ-

ity of mitochondrial ATP synthase is suppressed via

non-competitive inhibition by the MgADP complex

(the so-called ADP inhibition [10]) and by the regu-

latory protein IF1, which binds to the enzyme and

blocks its ATP hydrolysis activity without impeding

ATP synthesis [11]. In mammals, IF1 is believed to

prevent ATP depletion in the cell under the ener-

gy-deprived conditions, such as ischemia [11]. Studies

conducted by our group on baker’s yeast have shown

that IF1 helps to resume division more quickly after

starvation [12]. Along with that, attenuated ADP inhi-

bition increased the growth rate in yeast lacking mi-

tochondrial DNA (rho

strain), which are incapable of

oxidative phosphorylation and instead synthesize ATP

in the cytosol via glycolysis [13]. Such cells lack a ful-

ly functional ATP synthase; however, their mitochon-

dria contain the hydrophilic F

subcomplex, which is

encoded by the nuclear genes and can hydrolyze but

not synthesize ATP. For the rho

cells, intensive ATP

hydrolysis in the mitochondrial matrix proves ad-

vantageous, as ANT exchanges ADP produced in the

matrix for cytoplasmic ATP, transferring one negative

charge from the cytosol to the matrix and thereby

maintaining pmf necessary for protein import and

several other mitochondrial functions. Thus, activities

of ATP synthase and its F

subcomplex, along with

regulation of these activities, are important factors

influencing mitochondrial ATP concentration and,

consequently, cell physiology.

ANT is highly specific for ATP and ADP. It can

transport some other nucleotides and small mole-

cules, but only at more than 100 times lower rates,

rendering such fluxes negligible under physiological

conditions. It is ANT that determines the specificity of

oxidative phosphorylation for ADP; ATP synthase can,

although with lower efficiency, phosphorylate other

nucleoside diphosphates, primarily GDP [14]. ANT is

an electrogenic carrier, and under normal conditions,

high pmf across the inner mitochondrial membrane

shifts the equilibrium toward the export of ATP from

mitochondria to the cytosol.

The ATP/ADP exchange is inhibited in the pres-

ence of magnesium because the nucleotides are trans-

ferred in their free form rather than as magnesium

complexes [15, 16] – a rare case for proteins that

bind nucleoside triphosphates.

Adenylate kinases also influence the concentra-

tions of ATP and other nucleotides in the cytosol and

mitochondria. These enzymes catalyze the reversible

transfer of a phosphate group between nucleoside

triphosphates (usually ATP or GTP) and AMP, pro-

ducing two nucleoside diphosphates. Currently, nine

isoforms of adenylate kinase have been characterized

in human cells ([17], see also a recent review [18] on

the physiological role of adenylate kinases in health

and disease). Here, we will consider only two mito-

chondrial enzymes: AK2 of the intermembrane space

and AK3 (GTP:AMP phosphotransferase) of the mito-

chondrial matrix. Another member of the adenylate

kinase family, AK4, is also localized in the mitochon-

drial matrix in some tissues but apparently lacks en-

zymatic activity [19,  20]; therefore, it is not discussed

in detail here. (It should be noted, however, that the

recombinant human AK4 with intact mitochondri-

al localization peptide expressed in Escherichia coli

cells demonstrated the ability to phosphorylate AMP

in the presence of ATP or GTP [21]). Despite its prob-

able lack of enzymatic activity in  vivo, AK4 is an im-

portant regulator of cellular processes that plays a

role in the oxidative stress response, participates in

hypoxic regulation of mitochondria, and is associated

with drug resistance, aggressive tumor progression,

and metastasis in several cancers, making it a prom-

ising diagnostic and therapeutic target [18].

Adenylate kinase AK2 interconverts ATP, ADP,

and AMP in the mitochondrial intermembrane space

and catalyzes the reaction (1):

NTP + AMP ⇆ ADP + NDP. (1)

Both purine and pyrimidine nucleoside triphos-

phates (NTPs) can act as phosphate donors, but only

AMP can be phosphorylated by AK2 [22]. Like the

vast majority of enzymes whose substrates are nu-

cleoside triphosphates, AK2 and all other adenylate

kinases work only with their magnesium complexes.

AMP, which has about two orders of magnitude low-

er affinity for magnesium ions than ATP, binds and

is phosphorylated without magnesium [23]. The ap-

parent equilibrium constant of the adenylate kinase

reaction varies between 0.4 and 1.3 depending on the

magnesium concentration, reaching a maximum at

approximately 0.5 mM of free magnesium [24]. AK2,

which is localized in the intermembrane space, and

cytosolic adenylate kinases establish a strict relation-

ship between the concentrations of cytoplasmic ade-

nine nucleotides (see the section ‘Adenine nucleotides

in the cytosol’ below). AK2 also plays an important

role in the dynamic disequilibrium of nucleotide com-

position between the cytosol and the intermembrane

space, converting AMP from the cytosol into ADP us-

ing ATP exported from the mitochondria by ANT [25].

LAPASHINA et al.1932

BIOCHEMISTRY (Moscow) Vol. 90 No. 12 2025

Fig. 1. Oxidative phosphorylation in human mitochondria. Proteins are labeled in bold: green – ATP synthase, cyan – phos-

phate transporter, red – respiratory chain enzymes, purple – adenine nucleotide translocator (ANT), yellow – adenylate

kinase AK2, brown – adenylate kinase AK3, gray – porin VDAC. Bold arrows indicate transport of compounds; thin arrows

indicate chemical transformations. The diagram depicts the situation when the respiratory chain generates a sufficiently

high proton motive force (pmf ) for ATP synthesis; pH is 8.0 in the matrix and 7.4 in both the cytosol and intermembrane

space. TCA cycle stands for tricarboxylic acid cycle.

Adenylate kinase AK3 (GTP:AMP phosphotrans-

ferase) is localized in the mitochondrial matrix [26]

and catalyzes the reaction (2):

GTP + AMP ⇆ ADP + GDP. (2)

This enzyme is present in mitochondria in a sig-

nificant amount: a proteomic analysis of the human

HEK293T cell line has shown that the number of AK3

molecules per cell is approximately 700,000 [9].

Functions of the proteins described above, as

well as the phosphate transporter SLC25A3 [27], are

shown schematically in Fig. 1.

Experimental studies from the 1970s and 1980s

indicate that the rate of oxidative phosphorylation

is determined by the ATP/ADP ratio in the inter-

membrane space [28, 29], and that the rate-limit-

ing process is ATP/ADP translocation through ANT

[30, 31]. However, a more detailed analysis has re-

vealed that ANT is only one of several key factors

constraining the rate of oxidative phosphorylation,

with its role being most significant when the respi-

ration rate is approximately 80% of the maximum

possible value [32].

Moreover, the ATP-Mg/P

carrier located in the in-

ner mitochondrial membrane also influences nucleo-

tide concentration dynamics in the mitochondrial ma-

trix. This protein facilitates electroneutral transport

of the MgATP

2−

complex and inorganic phosphate

through the inner mitochondrial membrane [33] (not

shown in Fig.  1). According to the proteomic study

mentioned above, the total number of all isoforms

of this membrane protein approaches 300,000 mole-

cules per cell [9].

Outside mitochondria, the phosphocreatine shut-

tle (and, in many invertebrates, the phosphoarginine

shuttle) significantly influences ATP concentration.

Mitochondrial creatine kinase (or arginine kinase), lo-

cated in the intermembrane space, uses ATP to phos-

phorylate creatine (or arginine). Cytosolic isoforms

of the same enzymes catalyze the reverse reaction,

regenerating ATP from phosphocreatine and ADP,

ATP IN MITOCHONDRIA 1933

BIOCHEMISTRY (Moscow) Vol. 90 No. 12 2025

and are associated with the sites of high energy con-

sumption, such as muscle fiber myosin, sarco/endo-

plasmic reticulum Ca

-ATPase, plasma membrane

-ATPase, and other ATP-utilizing proteins.

However, a detailed examination of how these en-

zymes regulate cytosolic ATP concentrations lies be-

yond the scope of this review.

ADENINE NUCLEOTIDES IN THE CYTOSOL

As shown in Fig.  1, oxidative phosphorylation in

mitochondria results in the conversion of cytosolic

AMP, ADP, and inorganic phosphate to ATP. It should

be noted that, with high adenylate kinase activity,

the concentrations of ATP, ADP, and AMP will remain

close to the thermodynamic equilibrium. Thus, if

the rate of ATP consumption in the cell exceeds the

rate of its synthesis, ATP concentration will decrease

while ADP and AMP concentrations will increase.

Importantly, in this situation, the AMP concentration

relative to ATP will increase much more than the

ADP concentration [34]. Indeed, for the adenylate ki-

nase reaction (3):

[ATP] × [AMP]

[ADP]

= K, (3)

where [ATP], [ADP], and [AMP] are the equilibrium

concentrations of the corresponding nucleotides, and

K is the equilibrium constant, we have (4):

[ATP] × [AMP] = K × [ADP]

, (4)

and dividing both sides of the equation by [ATP]

, we

obtain (5):

[AMP]

[ATP]

∝

(

[ADP]

[ATP]

)

, (5)

meaning the AMP/ATP ratio is proportional to the

square of the ADP/ATP ratio. Therefore, it is not sur-

prising that an increase in the AMP concentration sig-

nals energy deficiency in the cell. The hypothesis that

the AMP/ATP ratio regulates the activity of enzymes

involved in energy metabolism was first proposed by

Atkinson in 1964 during his study of yeast phospho-

fructokinase [35]. Atkinson later expanded this hypoth-

esis by introducing a variable reflecting the number

of high-energy phosphoanhydride bonds per adenine

nucleotide in the cell. Since an ATP molecule contains

two such bonds, and ADP contains one, this variable

is calculated by the formula (6):

2[ATP] + [ADP]

[ATP] + [ADP] + [AMP]

, (6)

ranging from zero (pure AMP; no ADP or ATP) to

two (pure ATP; no ADP or AMP). To normalize the

range to unity, Atkinson divided the expression by 2

and called the result the ‘adenylate energy charge’

(AEC) of the cell, which reflects the energy available

in the phosphoanhydride bonds of adenine nucleo-

tides for metabolic reactions [36]. Thus, cellular AEC

is calculated from intracellular concentrations of ATP,

ADP, and AMP using the formula (7):

AEC =

[ATP] + ½[ADP]

[ATP] + [ADP] + [AMP]

(7)

and ranges from zero to one.

Soon after its initial success, the AEC hypothesis

became less popular because the number of enzymes

directly regulated by adenine nucleotides was found

to be quite small. A revival occurred in the late 1980s

with the discovery of the AMP-activated protein ki-

nase (AMPK) cascade [37], which in response to an

increase in AMP initiates a series of reactions that

stimulate catabolism and suppress ATP-consuming

processes both on a short-term (via enzyme phos-

phorylation) and long-term (via gene expression) time

scale [38]. The subsequent extensive body of data on

AMPK activation during hypoxia, ischemia, muscle

activity, and other states of cellular energy depletion

provided a solid experimental basis for the role of

AEC in regulating cellular bioenergetics and broad-

ened its relevance beyond the strictly metabolic reg-

ulation [34].

As previously noted, due to the activity of ade-

nylate kinases, concentrations of adenine nucleotides

in the cytosol are tightly interconnected, governed

by thermodynamics of the adenylate kinase reaction.

Blair [24] calculated these concentrations and the free

magnesium levels equilibrated by adenylate kinase at

different AEC values under conditions close to those

in the cytosol (pH  7.5, 100  mM  K

; Fig.  2).

The data shown in Fig.  2 illustrate the connec-

tion between the ATP hydrolysis and the increase in

free magnesium concentration, as well as the relative

stability of ADP concentration across a wide range

of AEC values.

To date, a significant amount of experimental

data on nucleotide concentrations in the cytosol has

been accumulated. Most of these data are obtained

from measurements using whole cells and, strictly

speaking, reflect not the cytosolic concentrations but

the total, averaged concentrations in both cytosol

and mitochondria. However, since the volume of mi-

tochondria is generally much smaller than that of the

cytosol, the contribution of mitochondrial nucleotides

to the total pool is also small and, as a first approx-

imation, can be neglected, except for cardiomyocytes

and other cells with very high mitochondrial con-

tent. In a large-scale metabolomic study quantifying

the most abundant low-molecular-weight metabolites

in human cells and baker’s yeast, concentrations of

LAPASHINA et al.1934

BIOCHEMISTRY (Moscow) Vol. 90 No. 12 2025

Fig. 2. Relationship between adenylate energy charge (AEC)

and concentrations of free magnesium and adenine nucle-

otides in the presence of adenylate kinase. Data were tak-

en from [24] for the following conditions: total nucleotide

concentration 5mM, total magnesium concentration 5mM,

potassium concentration 100 mM, pH 7.5. The graph shows

total nucleotide concentrations (magnesium-bound plus free)

for ATP, ADP, and AMP, and concentration of free Mg

adenine nucleotides in eukaryotic cells were reported

to fall within the following ranges: ATP, 1.4-7.0  mM;

ADP, 0.43-0.57  mM; and AMP, 0.036-0.103  mM [39].

These values are in good agreement with the majority

of previously obtained data [40, 41] and correspond

to an AEC value of about 0.95 (see Fig. 2).

ADENINE NUCLEOTIDES

IN MITOCHONDRIA

The interconversion of adenine nucleotides in

mitochondria is less understood than that in the cy-

tosol. As noted above, in mammals, the mitochondri-

al matrix apparently lacks adenylate kinase, which

could regenerate ADP from ATP and AMP. However,

AK3 is capable of converting AMP to ADP using GTP.

Mammalian mitochondria lack a carrier protein that

transports GTP across the inner membrane, so the

mitochondrial GTP and GDP pools can be considered

isolated from the cytosolic pool on a short timescale.

GTP in the matrix is synthesized by the succinyl-CoA

synthetase, an enzyme of the TCA cycle. Notably, ani-

mal mitochondria have two isoforms of this enzyme:

one synthesizes GTP and the other ATP [42]. The rel-

ative abundance of each isoform varies among dif-

ferent tissues.

GTP in mitochondria is mainly required for

RNA and protein synthesis, AMP rephosphorylation

by adenylate kinase AK3, and for conversion of ox-

aloacetate to phosphoenolpyruvate by mitochondrial

phosphoenolpyruvate carboxykinase (PEPCK-M) [43].

Nucleoside diphosphate kinases (NDPKs) consti-

tute another important class of enzymes influencing

the relationship between nucleotide concentrations.

They catalyze the exchange of phosphate groups

between various nucleoside di- and triphosphates,

maintaining thermodynamic equilibrium among ATP,

ADP, and other nucleotides in the cytosol. However,

in human cells, only one nucleoside diphosphate ki-

nase, namely NDPK D, has a mitochondrial target-

ing sequence [44, 45]. This enzyme is present in the

cell in significant amounts; proteomic studies of the

HEK293T cells report up to 200,000 NDPKD molecules

per cell [9]. Most experimental data suggest that it is

localized in the intermembrane space and is an-

chored to the inner mitochondrial membrane [46].

(Some studies, however, report nucleoside diphos-

phate kinase activity in the mitochondrial matrix of

vertebrate cells [47, 48]). Additional support for the

absence of NDPK in the matrix comes from the ex-

periments on baker’s yeast. Yeast normally possess-

es the mitochondrial GTP/GDP transporter, which is

absent in the vertebrate mitochondria. After genetic

deletion of this transporter, it was necessary to ex-

press human NDPK with an added matrix localiza-

tion signal to compensate for the deletion [49]. This

observation implies that yeast mitochondria cannot

produce GTP from ATP without NDPK artificially in-

troduced into the matrix.

We have not found studies that address the re-

lationship between guanine and adenine nucleotide

concentrations in the mitochondrial matrix, or that

report whether the adenylate kinase-driven thermo-

dynamic equilibrium observed in the cytosol is like-

wise sustained in the matrix.

Accurate quantification of nucleotide concentra-

tions in the mitochondrial matrix is crucial for eluci-

dating the mitochondrial contribution to the cellular

energy metabolism and its regulation. However, such

measurements are rather difficult, primarily because

of the substantial methodological difficulties. HPLC

and other analytical approaches provide highly ac-

curate measurements of total nucleotide content in

the sample. Yet, when applied to intact cells, they

cannot resolve relative contributions of the cytoso-

lic and mitochondrial pools. Since the mitochondrial

matrix usually occupies only a minor fraction of the

cellular volume, the resulting values predominantly

reflect cytosolic ATP, offering minimal information

about its concentration in the matrix.

Measuring nucleotide levels in the isolated mi-

tochondria provides only limited insight into in  vivo

concentrations of these nucleotides, as the isolation

process subjects mitochondria to substantial stress

and non-physiological conditions. The ATP/ADP/AMP

ratios measured in the isolated mitochondria vary

depending on biological source and experimental

ATP IN MITOCHONDRIA 1935

BIOCHEMISTRY (Moscow) Vol. 90 No. 12 2025

Table 1. Relative proportions of adenine nucleotides in the isolated mitochondrial preparations from bovine

heart and rat liver

Source ATP, % ADP, % AMP, % AEC Reference

Bovine heart mitochondria

12 30 58 0.27 [50]

39 33 28 0.56 [51]

Rat liver mitochondria

46 29 25 0.61 [52]

35 50 15 0.6 [16]

from 10 to 46 from 47 to 75 from 8 to 16 – [53]

20 69 11 0.55 [31]

Note. Absolute nucleotide concentrations were not determined in these studies, with the exception of reference [31]. TheAEC

was calculated as (ATP + ½ADP)/(ATP + ADP + AMP).

conditions, yet they consistently diverge from the

cytosolic values and typically correspond to an AEC

level of approximately 0.6 or less (Table 1).

Furthermore, when isolated mitochondria are

incubated in the media containing phosphate and

magnesium, adenine nucleotides leak out rapidly.

This loss depends on the ATP/ADP/AMP ratio and

could reach as much as 75% of the total nucleotide

pool [52], implying that a substantial and variable

fraction of adenine nucleotides could be lost during

mitochondrial isolation. In light of these observations,

the relative nucleotide contents measured in the iso-

lated mitochondria seem to provide an unreliable ba-

sis for evaluating their matrix concentrations in  vivo.

Nevertheless, as discussed below, available in  vivo es-

timates align reasonably well with the values shown

in Table 1.

One of the few studies that attempted to quantify

the in  vivo ratios of ATP, ADP, and their magnesium

complexes in the cytosol and mitochondria exam-

ined cambial cells of white maple (Acer pseudoplata-

nusL.)[54]. Using the

P-NMR spectra of living cells,

cell extracts, and isolated mitochondria, the authors

estimated nucleotide levels from the areas under the

peaks corresponding to the γ-phosphate of ATP and

the β-phosphate of ADP. The proportions of free nu-

cleotides and their Mg

complexes were determined

from the chemical shifts(δ), which depend on pH and

concentration. ATP concentrations in the cytosol

and mitochondrial matrix were reported to be simi-

lar – approximately 450  μM and 530  μM, respective-

ly – whereas ADP concentrations were about 30  μM

in the cytosol and 220  μM in the matrix. Thus, the

cytosolic ATP/ADP ratio was several-fold higher than

that in the matrix. It should be noted, however, that

the cytosolic ATP and ADP concentrations reported in

this study are far lower than those characteristic of

animal cells and yeast, in which ATP and ADP lev-

els are roughly 5  mM and 0.5 mM, respectively [39].

Moreover, the nucleotide concentrations reported in

the study were not validated by the independent

methods such as HPLC or luciferase assays, raising

concerns about absolute accuracy of these measure-

ments.

The authors also reported that most ATP exists as

a magnesium complex in both compartments – 88% in

the cytosol and 98% in the mitochondrial matrix [54].

In contrast, ADP was found to be predominantly free

in the cytosol (71%) but largely magnesium-bound in

the matrix (80%), implying that Mg

concentration is

roughly an order of magnitude higher in the matrix

than in the cytosol (2.4  mM vs. 250  μM).

Cytosolic ATP levels in the white maple cambial

cells proved to be sensitive to metabolic context. They

rose severalfold after exposure to excess adenine and

decreased when cells were transferred into a phos-

phate-depleted medium or incubated with glycerol or

choline, which are rapidly phosphorylated. In con-

trast, the cytosolic ADP levels remained comparative-

ly stable. From this, the authors suggested that free

ADP in the cytosol acts as a central regulator of respi-

ration and cellular energy metabolism [54]. However,

considering the limited dynamic range of cytosolic

ADP permitted by adenylate kinase (Fig.  2) and the

well-established role of AMP as a principal metabolic

regulator, further experimental support might be nec-

essary to confirm this interpretation.

Finally, the authors of [54] argued that the com-

paratively low Mg

concentration in the cytosol, rel-

ative to that of the mitochondrial matrix, favors the

presence of ADP in its free form, thereby promoting

its uptake by ANT and subsequent transport into the

matrix, where it participates in oxidative phosphory-

lation as MgADP.

The principal advantage of the NMR approach

used in this study is its ability to quantify adenine

LAPASHINA et al.1936

BIOCHEMISTRY (Moscow) Vol. 90 No. 12 2025

nucleotides in living cells. However, the method has

significant limitations, including technical complexity,

requirement for large amounts of biological material,

and low sensitivity to ADP and AMP.

Alternative approaches for assessing adenine

nucleotide levels in animal cells rely on rapid frac-

tionation to obtain a mitochondrial-enriched fraction

within seconds. One method, reported effective for

rat hepatocytes, uses brief exposure (20-40  s) to dig-

itonin, followed by rapid centrifugation (20  s) [55].

Digitonin selectively disrupts the sterol-rich plasma

membrane while leaving the sterol-poor inner mi-

tochondrial membrane intact, which allows, after

centrifugation, to obtain a supernatant containing

cytosolic nucleotides and a pellet containing in-

tact mitochondria. Using this technique, ATP, ADP,

and AMP proportions of 45-75%, 28-31%, and 3-21%

were reported in the mitochondria from rat hepa-

tocytes [55, 56].

A different rapid-fractionation technique subject-

ed rat liver cell suspensions to shear forces by pass-

ing them through a narrow metal needle, rupturing

the plasma membrane while leaving most mitochon-

dria intact [57]. The complete workflow – including

disruption and centrifugation through silicone oil into

fixative – required less than one minute. Using this

method, the ATP/ADP ratio was found to be ~7 in the

cytosolic fraction and close to 1 in the mitochondrial

fraction.

Another study from the mid-1970s reported an

intriguing and unconventional method for determin-

ing adenine nucleotide composition in rat liver mito-

chondria [58]. The technique involved rapid freezing

of the tissue, its homogenization in the frozen state,

lyophilization, and fractionation using a carbon tet-

rachloride–heptane solvent system. Despite appearing

at first unsuitable for separating membrane-bound

organelles, the procedure proved effective: electron

microscopy and enzyme assays confirmed good sep-

aration of cytosolic and mitochondrial fractions. The

authors found ATP, ADP, and AMP proportions of

roughly 85%, 14%, and 1% in the cytosol and 27%,

47%, and 26% in the mitochondria, respectively. Nota-

bly, these ratios correspond well to the matrix values

observed in the isolated mitochondria (Table  1) and

to the cytosolic concentrations determined by other

methods.

The reliability of this new method was further

verified on isolated rat hepatocytes by direct com-

parison between the digitonin permeabilization pro-

tocol and the freezing-lyophilization-organic solvent

fractionation. When both methods were applied to

the same hepatocyte sample, they yielded the same

values for mitochondrial adenine nucleotide content

within experimental error [59]. A related study ana-

lyzing pre- and postnatal rat liver mitochondria with

the organic solvent fractionation method reported the

following values for the matrix adenine nucleotides:

25%, 32%, and 43% for ATP, ADP, and AMP in adult

animals, and 21%, 19%, and 60% in the three-day-old

pups [60].

A combination of the rapid freezing, lyophiliza-

tion, and organic solvent fractionation method with

an ultra-fast (3-second) procedure for extracting rat

liver allowed estimation of the ATP/ADP ratio in mi-

tochondria in vivo. In the control group of rats, this

ratio was 0.92  ±  0.18; in the rats fasted for 48  h be-

fore liver extraction, it increased to 1.04  ±  0.03, and

in the rats subjected to anesthesia, it decreased to

0.55  ±  0.06 (pentobarbital) or 0.86  ±  0.04 (ketamine)

[61]. These data also agree with the experimental es-

timates of the ATP, ADP, and AMP ratios in the isolat-

ed rat liver mitochondria shown in Table  1.

Estimates of the ratio of ATP, ADP, and AMP con-

centrations in the mitochondrial matrix appear to

be reliable, since many experiments using different

objects and different methods yield similar results.

At the same time, the absolute values of these con-

centrations remain difficult to assess. The range of

absolute values reported in the literature is quite

broad (Table 2).

In the study [62], an attempt was made to quan-

titatively evaluate the content of a wide range of me-

tabolites in the mitochondrial matrix of HeLa cells.

To quickly and selectively separate mitochondria

from other cellular components, the authors used an

immunoaffinity method, capturing mitochondria by

an epitope of a recombinant protein localized in the

outer mitochondrial membrane. To quantify metabo-

lites in the mitochondrial preparations, the authors

employed liquid chromatography-mass spectrometry

(LC/MS) and estimated total mitochondrial matrix

volume per cell – required for calculating absolute

concentrations – using confocal microscopy. The ade-

nine nucleotide concentrations reported in this work

(Table  2, bottom row), however, raise substantial

concerns. Most experimental data, including mea-

surements obtained with fluorescent protein sensors

[66-69], consistently show that the ATP levels in the

matrix are in the millimolar range, nearly two orders

of magnitude higher than the 50  μM value reported

in [62]. A likely explanation is an error in the esti-

mation of mitochondrial volume, a parameter that is

difficult to determine precisely.

Taken together, the experimental evidence in-

dicates that in the cytosol of mammalian cells, ATP,

ADP, and AMP concentrations are constrained by the

adenylate kinase equilibrium and typically are esti-

mated as approximately 5  mM, 0.5  mM, and 0.05  mM,

respectively, yielding an AEC of about 0.95. In the

mitochondrial matrix, the total adenine nucleotide

concentration is likewise in the millimolar range,

ATP IN MITOCHONDRIA 1937

BIOCHEMISTRY (Moscow) Vol. 90 No. 12 2025

Table 2. Experimental estimates of absolute adenine nucleotide concentrations in the mitochondrial matrix

Source ATP, mM ADP, mM AMP, mM AEC Reference

Guinea pig heart 5.6-7.2 1.4-2.0 ND – [63]

Rat liver 3.5 12.4 1.9 0.54 [31]

Rat liver (fasted rats) 5.2 7.8 1.5 0.63 [64]

Rat liver (fed rats) 1.4 7.8 3.8 0.41 [64]

Rat liver 12.5-16.4 12.0-14.1 ND – [57]

Rat liver 10.4 5.9 4.3 0.65 [56]

Rat; renal proximal tubular cells 2.62 ND ND – [65]

HeLa cells 0.05 0.07 0.02 0.61 [62]

Note. In the studies cited, concentrations were calculated using approximate estimates of matrix volume per milligram of

protein or per milligram of dry weight, except in the study [62], where the mitochondrial volume was derived from the

microscopy data. AEC denotes the adenylate energy charge; ND indicates cases where no data were available.

but the ATP/ADP/AMP ratios are more variable and

the AEC is lower – generally between 0.4 and 0.65.

It is widely assumed that in the resting state,

when cellular ATP demand is low, the adenylate ener-

gy charge in the mitochondrial matrix rises and ADP

becomes nearly fully phosphorylated to ATP. In this

scenario, ATP synthase is unable either to continue

ATP production or to consume pmf through proton

translocation coupled to ATP synthesis. As a result,

pmf increases and suppresses the activity of respira-

tory chain enzymes, producing the so-called ‘state  4’

[70], extensively characterized in  vitro in isolated mi-

tochondria. Inhibition of respiration elevates the re-

duction state of the respiratory electron carriers and

increases the intracellular O

levels, thereby promot-

ing non-enzymatic formation of reactive oxygen spe-

cies (ROS) and increasing the risk of oxidative dam-

age [71]. Vladimir Skulachev proposed that, in order

to limit ROS production under these conditions, cells

increase proton conductivity of the inner mitochon-

drial membrane (‘mild uncoupling’), thereby lowering

pmf, stimulating respiration, and reducing O

concen-

tration [72].

The evidence summarized in this review, howev-

er, indicates that such complete depletion of matrix

ADP – and thus a true state  4 – is unlikely to occur

in vivo. All available data show that the matrix ADP

concentration remains appreciable and is consistently

higher than the cytosolic one, which itself remains

remarkably stable at ~0.5  mM – more than sufficient

to support continuous ATP synthesis by ATP synthase.

Accordingly, the processes grouped under the term

‘mild uncoupling’ likely serve a physiological role

distinct from the suppression of mitochondrial ROS

production in the resting state.

APPLICATION OF FLUORESCENT

PROTEIN SENSORS TO MEASURING

ATP CONCENTRATION IN MITOCHONDRIA

A powerful strategy for quantifying ATP levels

within both mitochondria and cytosol is the use of

fluorescent protein-based sensors whose spectral

characteristics shift upon ATP binding. The first such

sensor, ATeam, was developed by Hiromi Imamura

and colleagues [69]. ATeam comprises yellow and

cyan fluorescent proteins joined by a small protein

that undergoes a reversible conformational transi-

tion in response to ATP binding. This structural re-

arrangement decreases the distance between the flu-

orophores and enhances Förster resonance energy

transfer (FRET). Consequently, fluorescence spectral

measurements permit accurate monitoring of relative

abundance of the free versus ATP-bound forms of

the probe. A major advantage of this method is that

its readout is independent of probe concentration.

Application of ATeam in HeLa cells revealed mito-

chondrial ATP concentrations in the millimolar range,

although consistently lower than those observed in

the cytosol [69].

Further development of this approach substan-

tiated the initial observation and enabled real-time

assessment of ATP concentration changes within the

mitochondrial matrix in response to various stimuli

[66-68]. Parallel observations of a mitochondria-tar-

geted ATeam probe and its cytosolic counterpart

in cardiomyocytes demonstrated that hypoxia pro-

duces a substantial decline in the matrix ATP over

several hours, whereas cytosolic ATP levels remain

nearly constant. A similar pattern was observed un-

der normoxia following the addition of oligomycin,

LAPASHINA et al.1938

BIOCHEMISTRY (Moscow) Vol. 90 No. 12 2025

an ATP synthase inhibitor, although in this case the

signal changes were developing within minutes [67].

In another study, ATeam probes were employed

to analyze ATP concentration dynamics in the cyto-

sol and mitochondria of HeLa cells following glucose

depletion [73]. Under these conditions, cytosolic ATP

levels remained largely stable, whereas the mito-

chondrial matrix exhibited a transient elevation in

ATP followed by a marked decline. The authors pro-

posed that the initial rise in mitochondrial ATP level

results from the activity of hexokinases associated

with the outer mitochondrial membrane. Although

the hexokinase-catalyzed phosphorylation of glucose

to glucose-6-phosphate is generally regarded as ir-

reversible under physiological conditions, the au-

thors speculate that rapid mitochondrial ATP uptake

in the absence of cytosolic glucose could make the

reverse reaction feasible. This hypothesis, however,

requires further confirmation. Another study com-

bined the fluorescent dye TMRE with a mitochon-

dria-targeted ATeam probe to assess the correlation

between pmf and ATP levels in the neuronal mito-

chondria [74].

ATeam sensors have also been applied success-

fully to monitor mitochondrial ATP dynamics in in-

tact tissues and even whole organs. Analysis using

a mitochondria-targeted probe in the mouse hearts

demonstrated that empagliflozin, an inhibitor of the

sodium–glucose cotransporter 2 (SGLT2), elevates mi-

tochondrial ATP levels [75].

In our laboratory, an ATeam-derived sensor was

successfully employed to quantify ATP concentrations

in yeast cells and in isolated mitochondria. We select-

ed the yAT1.03 variant, optimized for expression in

Saccharomyces cerevisiae and exhibiting reduced pH

sensitivity relative to the original ATeam sensors [76].

We isolated mitochondria from the yeast expressing

this probe fused to the mitochondrial localization sig-

nal. Permeabilization with the channel-forming anti-

biotic alamethicin – which renders the inner mem-

brane permeable to low-molecular-weight solutes (up

to 1.5 kDa) but not to proteins [77] – enabled cali-

bration of the matrix yAT1.03 signal by adding ATP

at defined concentrations to the incubation medium.

These experiments demonstrated that ATP concen-

tration in the matrix of isolated yeast mitochondria

increases to millimolar levels upon the addition of

respiratory substrates [78]. This method also appears

promising for quantitative in  vitro studies of intra-

mitochondrial ATP concentration dynamics under

various conditions.

Besides the FRET-based ATeam family, additional

protein sensors comprising a single fluorescent pro-

tein with ATP-dependent spectral shifts have been de-

veloped. These ‘single-color’ sensors include QUEEN

[79] and MaLion [80]. QUEEN has been applied with

moderate success to quantify ATP levels in yeast mi-

tochondria [81], where – consistent with observations

in HeLa cells – mitochondrial ATP was found to be

slightly lower than cytosolic ATP. Another study em-

ploying QUEEN demonstrated that, in yeast, addition

of the protonophore FCCP (carbonyl cyanide p-trifluo-

romethoxyphenylhydrazone), which collapses the pmf

across the inner mitochondrial membrane, leads to

an increase in ATP within the matrix and a decrease

in the cytosol [82]. To account for this unexpected

behavior, the authors proposed that FCCP stimulates

ATP import from the cytosol into mitochondria via

the ATP-Mg

/HPO

2−

carrier, although the presented

data provide insufficient evidence to support this

conclusion. In a separate investigation, use of the

mitochondria-targeted MaLion probe revealed that

when hepatocytes transition from glucose depriva-

tion to active glucose utilization, the mitochondrial

ATP concentration declines, implying a potential role

of the mitochondrial adenine-nucleotide dynamics

in regulating cellular metabolism [83].

Another family of fluorescent probes includes

the Perceval sensor and its analogs. These proteins

have very high affinity for both ATP and ADP, so in-

side the cells they are always bound to one of these

nucleotides. The spectral properties of Perceval dif-

fer depending on which nucleotide is bound, and

the overall fluorescence signal reliably reflects the

ATP/ADP ratio [84]. However, the signal of this probe

is sensitive to pH changes between 6 and 8, which

makes quantitative interpretation more difficult.

A later variant, PercevalHR, was developed with im-

proved spectral characteristics [85], but we found no

reports of its successful use for measuring ATP/ADP

ratios in mitochondria. Recently, a protein fluorescent

probe capable of detecting the GTP/GDP ratio in mi-

tochondria was developed [86].

Although the studies described above demon-

strate varying degrees of success in applying the

protein-based fluorescent sensors to mitochondrial

ATP measurements, several substantial methodologi-

cal limitations remain. All such probes are influenced

to some extent by pH and by ionic composition, dis-

play partial cross-reactivity with other nucleotides,

and exhibit temperature-dependent fluorescence be-

havior. Moreover, probes of the ATeam family are

subject to asynchronous maturation of their two

fluorescent proteins, potentially leading to the accu-

mulation of an unknown – yet possibly considerable

– population of non-functional probe species within

the cell. These aberrant molecules lack either the

FRET donor or acceptor and could, therefore, com-

promise the composite signal. Consequently, rigorous

calibration of the probe output directly in the exper-

imental sample (e.g., through permeabilization fol-

lowed by the addition of defined ATP concentrations)

ATP IN MITOCHONDRIA 1939

BIOCHEMISTRY (Moscow) Vol. 90 No. 12 2025

and cross-validation of ATP measurements using in-

dependent approaches (distinct fluorescent probes,

luciferin–luciferase assays, HPLC) are critically im-

portant.

CONCLUSION

When discussing adenine nucleotide levels in eu-

karyotic cells, it can be stated that under most condi-

tions, cytosolic ATP concentration lies in the millimo-

lar range, while the concentrations of ADP and AMP

are approximately 10-fold and 100-fold lower, respec-

tively. In the mitochondrial matrix, the proportion of

ATP is lower and the proportions of ADP and AMP

are higher compared with the cytosol; this indicates

that the so-called ‘state 4’, in which oxidative phos-

phorylation is inhibited in the resting cells due to the

lack of ADP in the matrix, is very unlikely to occur

in vivo under physiological conditions.

To clarify causal relationships between the ac-

tivities of bioenergetic enzymes, changes in pmf, and

adenine nucleotide concentrations in the mitochon-

drial matrix and cytosol, reliable methods are needed

for quantitative measurements of ATP concentrations

invivo. Fluorescent protein sensors are excellent tools

for monitoring dynamic changes in ATP concentration

in the cytosol and mitochondria, but they cannot pro-

vide precise values for absolute nucleotide concen-

trations because their calibration in  vivo is difficult.

Therefore, an optimal strategy is to combine dynamic

measurements obtained with several different fluores-

cent protein probes – calibrated using standard ATP

solutions after cell permeabilization – with tradition-

al nucleotide quantification methods (HPLC, lucifer-

in–luciferase assays) and newer analytical approaches

such as single-cell metabolomics, mass-spectrometry

imaging, and high-resolution respirometry within

the same experimental framework.

Abbreviations

AEC adenylate energy charge

AMPK AMP-activated protein kinase

FRET Förster resonance energy transfer

NDPK nucleoside diphosphate kinase

pmf proton motive force

TCA tricarboxylic acid cycle

Acknowledgments

The authors thank Dr. Dmitry Knorre for valuable

discussions and advice during the preparation of this

manuscript, and one of the reviewers for suggesting

that we address whether the low AEC values in the

matrix are compatible with the hypothesis of ‘mild un-

coupling’ as a protective mechanism against oxidative

stress.

Funding

The study was conducted under the State assign-

ment of Lomonosov Moscow State University

(AAAA-A19-119031390114-5) and financially supported

by the Russian Science Foundation (grant no. 20-14-

00268).

Ethics approval and consent to participate

This work does not contain any studies involving hu-

man and animal subjects.

Conflict of interest

The authors of this work declare that they have no

conflicts of interest.

Open access

This article is licensed under a Creative Commons At-

tribution 4.0 International License, which permits use,

sharing, adaptation, distribution, and reproduction

in any medium or format, as long as you give appro-

priate credit to the original author(s) and the source,

provide a link to the Creative Commons license, and

indicate if changes were made. The images or other

third party material in this article are included in the

article’s Creative Commons license, unless indicated

otherwise in a credit line to the material. If materi-

al is not included in the article’s Creative Commons

license and your intended use is not permitted by

statutory regulation or exceeds the permitted use,

you will need to obtain permission directly from the

http://creativecommons.org/licenses/by/4.0/.

REFERENCES

1. Zheng, W., Chai, P., Zhu, J., and Zhang, K. (2024)

High-resolution in situ structures of mammalian

respiratory supercomplexes, Nature, 631, 232-239,

https://doi.org/10.1038/s41586-024-07488-9.

2. Burgess, D. R. (2004) NIST SRD 46. Critically Selected

Stability Constants of Metal Complexes: Version 8.0 for

Windows, National Institute of Standards and Tech-

nology [Internet], https://doi.org/10.18434/M32154.

3. Maeshima, K., Matsuda, T., Shindo, Y., Imamura, H.,

Tamura, S., Imai, R., Kawakami, S., Nagashima, R.,

Soga,T., Noji,H., Oka,K., and Nagai,T. (2018) A tran-

sient rise in free Mg

ions released from ATP-Mg

hydrolysis contributes to mitotic chromosome con-

densation, Curr. Biol., 28, 444-451.e6, https://doi.org/

10.1016/j.cub.2017.12.035.

4. Babot, M., Birch, A., Labarbuta, P., and Galkin, A.

(2014) Characterisation of the active/de-active tran-

sition of mitochondrial complex I, Biochim. Biophys.

Acta, 1837, 1083-1092, https://doi.org/10.1016/j.bbabio.

2014.02.018.

LAPASHINA et al.1940

BIOCHEMISTRY (Moscow) Vol. 90 No. 12 2025

5. Zubareva, V. M., Lapashina, A. S., Shugaeva, T. E.,

Litvin, A.V., and Feniouk, B.A. (2020) Rotary ion-trans-

locating ATPases/ATP synthases: diversity, similar-

ities, and differences, Biochemistry (Moscow), 85,

1613-1630, https://doi.org/10.1134/S0006297920120135.

6. Klingenberg, M. (2008) The ADP and ATP trans-

port in mitochondria and its carrier, Biochim. Bio-

phys. Acta, 1778, 1978-2021, https://doi.org/10.1016/

j.bbamem.2008.04.011.

7. Penin, F., Deléage, G., Godinot, C., and Gautheron,

D. C. (1986) Efficient reconstitution of mitochondri-

al energy-transfer reactions from depleted mem-

branes and F1-ATPase as a function of the amount

of bound oligomycin sensitivity-conferring protein

(OSCP), Biochim. Biophys. Acta, 852, 55-67, https://

doi.org/10.1016/0005-2728(86)90056-3.

8. Matsuno-Yagi, A., and Hatefi, Y. (1988) Estimation of

the turnover number of bovine heart F0F1 complexes

for ATP synthesis, Biochemistry, 27, 335-340, https://

doi.org/10.1021/bi00401a050.

9. Morgenstern,M., Peikert, C.D., Lübbert,P., Suppanz,I.,

Klemm, C., Alka, O., Steiert, C., Naumenko, N.,

Schendzielorz,A., Melchionda,L., Mühlhäuser, W.W.D.,

Knapp, B., Busch, J. D., Stiller, S. B., Dannenmaier, S.,

Lindau, C., Licheva, M., Eickhorst, C., Galbusera, R.,

Zerbes, R.M., Ryan, M.T., Kraft,C., Kozjak-Pavlovic,V.,

Drepper,F., Dennerlein,S., Oeljeklaus,S., Pfanner,N.,

Wiedemann,N., and Warscheid,B. (2021) Quantitative

high-confidence human mitochondrial proteome

and its dynamics in cellular context, Cell Metab.,

33, 2464-2483.e18, https://doi.org/10.1016/j.cmet.

2021.11.001.

10. Lapashina, A. S., and Feniouk, B. A. (2018) ADP-In-

hibition of H+-FOF1-ATP Synthase, Biochemistry

(Moscow), 83, 1141-1160, https://doi.org/10.1134/

S0006297918100012.

11. Campanella, M., Casswell, E., Chong, S., Farah, Z.,

Wieckowski, M. R., Abramov, A. Y., Tinker, A., and

Duchen, M. R. (2008) Regulation of mitochondri-

al structure and function by the F1Fo-ATPase in-

hibitor protein, IF1, Cell Metab., 8, 13-25, https://

doi.org/10.1016/j.cmet.2008.06.001.

12. Galkina, K.V., Zubareva, V.M., Kashko, N.D., Lapashina,

A. S., Markova, O. V., Feniouk, B. A., and Knorre,

D. A. (2022) Heterogeneity of starved yeast cells in

IF1 levels suggests the role of this protein in vivo,

Front. Microbiol., 13, 816622, https://doi.org/10.3389/

fmicb.2022.816622.

13. Lapashina, A. S., Kashko, N. D., Zubareva, V. M.,

Galkina, K. V., Markova, O. V., Knorre, D. A., and

Feniouk, B. A. (2022) Attenuated ADP-inhibition of

FOF1 ATPase mitigates manifestations of mitochondri-

al dysfunction in yeast, Biochim. Biophys. Acta, 1863,

148544, https://doi.org/10.1016/j.bbabio.2022.148544.

14. Pfaff, E., and Klingenberg, M. (1968) Adenine nu-

cleotide translocation of mitochondria. 1. Specific-

ity and control, Eur. J. Biochem., 6, 66-79, https://

doi.org/10.1111/j.1432-1033.1968.tb00420.x.

15. Winkler, H. H., Bygrave, F. L., and Lehninger, A. L.

(1968) Characterization of the atractyloside-sensi-

tive adenine nucleotide transport system in rat liv-

er mitochondria, J. Biol. Chem., 243, 20-28, https://

doi.org/10.1016/s0021-9258(18)99320-8.

16. Matsuno-Yagi,A., and Hatefi,Y. (1990) Studies on the

mechanism of oxidative phosphorylation. Positive co-

operativity in ATP synthesis, J. Biol. Chem., 265, 82-88,

https://doi.org/10.1016/S0021-9258(19)40198-1.

17. Panayiotou, C., Solaroli, N., and Karlsson, A. (2014)

The many isoforms of human adenylate kinases,

Int. J. Biochem. Cell Biol., 49, 75-83, https://doi.org/

10.1016/j.biocel.2014.01.014.

18. Fujisawa,K. (2023) Regulation of adenine nucleotide

metabolism by adenylate kinase isozymes: physio-

logical roles and diseases, Int. J. Mol. Sci., 24, 5561,

https://doi.org/10.3390/ijms24065561.

19. Miyoshi, K., Akazawa, Y., Horiguchi, T., and Noma, T.

(2009) Localization of adenylate kinase 4 in mouse

tissues, Acta Histochem. Cytochem., 42, 55-64, https://

doi.org/10.1267/ahc.08012.

20. Liu,R., Ström, A.-L., Zhai,J., Gal,J., Bao,S., Gong,W.,

and Zhu, H. (2009) Enzymatically inactive adenylate

kinase 4 interacts with mitochondrial ADP/ATP trans-

locase, Int. J. Biochem. Cell Biol., 41, 1371-1380,

https://doi.org/10.1016/j.biocel.2008.12.002.

21. Panayiotou, C., Solaroli, N., Johansson, M., and

Karlsson, A. (2010) Evidence of an intact N-terminal

translocation sequence of human mitochondrial ad-

enylate kinase 4, Int. J. Biochem. Cell Biol., 42, 62-69,

https://doi.org/10.1016/j.biocel.2009.09.007.

22. Wilson, D. E., Povey, S., and Harris, H. (1976) Ad-

enylate kinases in man: evidence for a third lo-

cus, Ann. Hum. Genet., 39, 305-313, https://doi.org/

10.1111/j.1469-1809.1976.tb00134.x.

23. Nam, K., Thodika, A. R. A., Tischlik, S., Phoeurk, C.,

Nagy, T. M., Schierholz, L., Ådén, J., Rogne, P.,

Drescher,M., Sauer-Eriksson, A.E., and Wolf-Watz,M.

(2024) Magnesium induced structural reorganization

in the active site of adenylate kinase, Sci. Adv., 10,

eado5504, https://doi.org/10.1126/sciadv.ado5504.

24. Blair, J. M. (1970) Magnesium, potassium, and

the adenylate kinase equilibrium. Magnesium as

a feedback signal from the adenine nucleotide

pool, Eur. J. Biochem., 13, 384-390, https://doi.org/

10.1111/j.1432-1033.1970.tb00940.x.

25. Gellerich, F. N. (1992) The role of adenylate ki-

nase in dynamic compartmentation of adenine

nucleotides in the mitochondrial intermembrane

space, FEBS Lett., 297, 55-58, https://doi.org/10.1016/

0014-5793(92)80326-c.

26. Tomasselli, A. G., Schirmer, R. H., and Noda, L. H.

(1979) Mitochondrial GTP-AMP phosphotransfer-

ase. 1. Purification and properties, Eur. J. Biochem.,

ATP IN MITOCHONDRIA 1941

BIOCHEMISTRY (Moscow) Vol. 90 No. 12 2025

93, 257-262, https://doi.org/10.1111/j.1432-1033.1979.

tb12818.x.

27. Seifert, E.L., Ligeti,E., Mayr, J.A., Sondheimer,N., and

Hajnóczky, G. (2015) The mitochondrial phosphate

carrier: role in oxidative metabolism, calcium han-

dling and mitochondrial disease, Biochem. Biophys.

Res. Commun., 464, 369-375, https://doi.org/10.1016/

j.bbrc.2015.06.031.

28. Küster, U., Bohnensack,R., and Kunz,W. (1976) Con-

trol of oxidative phosphorylation by the extra-mito-

chondrial ATP/ADP ratio, Biochim. Biophys. Acta, 440,

391-402, https://doi.org/10.1016/0005-2728(76)90073-6.

29. Davis, E. J., and Davis-van Thienen, W. I. (1978) Con-

trol of mitochondrial metabolism by the ATP/ADP ra-

tio, Biochem. Biophys. Res. Commun., 83, 1260-1266,

https://doi.org/10.1016/0006-291x(78)91357-8.

30. Lemasters, J. J., and Sowers, A. E. (1979) Phosphate

dependence and atractyloside inhibition of mitochon-

drial oxidative phosphorylation. The ADP-ATP carrier

is rate-limiting, J. Biol. Chem., 254, 1248-1251, https://

doi.org/10.1016/s0021-9258(17)34194-7.

31. Slater, E. C., Rosing, J., and Mol, A. (1973) The phos-

phorylation potential generated by respiring mi-

tochondria, Biochim. Biophys. Acta, 292, 534-553,

https://doi.org/10.1016/0005-2728(73)90003-0.

32. Groen, A.K., Wanders, R.J., Westerhoff, H.V., van der

Meer,R., and Tager, J.M. (1982) Quantification of the

contribution of various steps to the control of mito-

chondrial respiration, J. Biol. Chem., 257, 2754-2757,

https://doi.org/10.1016/s0021-9258(19)81026-8.

33. Fiermonte,G., De Leonardis,F., Todisco,S., Palmieri,L.,

Lasorsa, F. M., and Palmieri, F. (2004) Identification

of the mitochondrial ATP-Mg/Pi transporter, J. Biol.

Chem., 279, 30722-30730, https://doi.org/10.1074/

jbc.M400445200.

34. Hardie, D.G., and Hawley, S.A. (2001) AMP-activated

protein kinase: the energy charge hypothesis revisit-

ed, Bioessays, 23, 1112-1119, https://doi.org/10.1002/

bies.10009.

35. Ramaiah, A., Hathaway, J. A., and Atkinson, D. E.

(1964) Adenylate as a metabolic regulator. Ef-

fect on yeast phosphofructokinase kinetics, J. Biol.

Chem., 239, 3619-3622, https://doi.org/10.1016/S0021-

9258(18)91181-6.

36. Atkinson, D.E. (1968) Energy charge of the adenylate

pool as a regulatory parameter. Interaction with feed-

back modifiers, Biochemistry, 7, 4030-4034, https://

doi.org/10.1021/bi00851a033.

37. Hardie, D.G., Carling,D., and Sim, A.T.R. (1989) The

AMP-activated protein kinase: a multisubstrate reg-

ulator of lipid metabolism, Trends Biochem. Sci., 14,

20-23, https://doi.org/10.1016/0968-0004(89)90084-4.

38. Hardie, D. G., Ross, F. A., and Hawley, S. A. (2012)

AMPK: a nutrient and energy sensor that maintains

energy homeostasis, Nat. Rev. Mol. Cell Biol., 13,

251-262, https://doi.org/10.1038/nrm3311.

39. Park, J. O., Rubin, S.A., Xu, Y.-F., Amador-Noguez,D.,

Fan,J., Shlomi,T., and Rabinowitz, J.D. (2016) Metab-

olite concentrations, fluxes and free energies imply

efficient enzyme usage, Nat. Chem. Biol., 12, 482-489,

https://doi.org/10.1038/nchembio.2077.

40. Gauthier, S., Coulpier, F., Jourdren, L., Merle, M.,

Beck, S., Konrad, M., Daignan-Fornier, B., and

Pinson, B. (2008) Co-regulation of yeast purine and

phosphate pathways in response to adenylic nu-

cleotide variations, Mol. Microbiol., 68, 1583-1594,

https://doi.org/10.1111/j.1365-2958.2008.06261.x.

41. Traut, T.W. (1994) Physiological concentrations of pu-

rines and pyrimidines, Mol. Cell Biochem., 140, 1-22,

https://doi.org/10.1007/BF00928361.

42. Johnson, J. D., Mehus, J. G., Tews, K., Milavetz, B. I.,

and Lambeth, D. O. (1998) Genetic evidence for the

expression of ATP- and GTP-specific succinyl-CoA syn-

thetases in multicellular eucaryotes, J. Biol. Chem.,

273, 27580-27586, https://doi.org/10.1074/jbc.273.

42.27580.

43. Stark, R., Pasquel, F., Turcu, A., Pongratz, R. L.,

Roden, M., Cline, G. W., Shulman, G. I., and Kibbey,

R. G. (2009) Phosphoenolpyruvate cycling via mito-

chondrial phosphoenolpyruvate carboxykinase links

anaplerosis and mitochondrial GTP with insulin

secretion, J. Biol. Chem., 284, 26578-26590, https://

doi.org/10.1074/jbc.M109.011775.

44. Milon, L., Meyer, P., Chiadmi, M., Munier, A.,

Johansson, M., Karlsson, A., Lascu, I., Capeau, J.,

Janin, J., and Lacombe, M. L. (2000) The human

nm23-H4 gene product is a mitochondrial nucleoside

diphosphate kinase, J. Biol. Chem., 275, 14264-14272,

https://doi.org/10.1074/jbc.275.19.14264.

45. Tokarska-Schlattner, M., Boissan, M., Munier, A.,

Borot, C., Mailleau, C., Speer, O., Schlattner, U., and

Lacombe, M.-L. (2008) The nucleoside diphosphate

kinase D (NM23-H4) binds the inner mitochondri-

al membrane with high affinity to cardiolipin and

couples nucleotide transfer with respiration, J. Biol.

Chem., 283, 26198-26207, https://doi.org/10.1074/

jbc.M803132200.

46. Boissan,M., Dabernat,S., Peuchant,E., Schlattner,U.,

Lascu,I., and Lacombe, M.-L. (2009) The mammalian

Nm23/NDPK family: from metastasis control to cilia

movement, Mol. Cell Biochem., 329, 51-62, https://

doi.org/10.1007/s11010-009-0120-7.

47. Muhonen, W. W., and Lambeth, D. O. (1995) The

compartmentation of nucleoside diphosphate kinase

in mitochondria, Comp. Biochem. Physiol. B Bio-

chem. Mol. Biol., 110, 211-223, https://doi.org/10.1016/

0305-0491(94)00123-c.

48. Lacombe, M.-L., Tokarska-Schlattner, M., Boissan, M.,

and Schlattner, U. (2018) The mitochondrial nucle-

oside diphosphate kinase (NDPK-D/NME4), a moon-

lighting protein for cell homeostasis, Lab Invest, 98,

582-588, https://doi.org/10.1038/s41374-017-0004-5.

LAPASHINA et al.1942

BIOCHEMISTRY (Moscow) Vol. 90 No. 12 2025

49. Gordon, D. M., Lyver, E. R., Lesuisse, E., Dancis, A.,

and Pain, D. (2006) GTP in the mitochondrial matrix

plays a crucial role in organellar iron homoeosta-

sis, Biochem. J., 400, 163-168, https://doi.org/10.1042/

BJ20060904.

50. Mansurova, S. E., Drobyshev, V. I., and Kulaev, I. S.

(1972) Nucleotides of beef heart mitochondria and

submitochondrial particles, J. Bioenerg., 3, 499-507,

https://doi.org/10.1007/BF01539058.

51. Brierley,G., and O’Brien, R. L. (1965) Compartmenta-

tion of heart mitochondria, J. Biol. Chem., 240, 4532-

4539, https://doi.org/10.1016/s0021-9258(18)97095-x.

52. Meisner,H., and Klingenberg,M. (1968) Efflux of ade-

nine nucleotides from rat liver mitochondria, J. Biol.

Chem., 243, 3631-3639, https://doi.org/10.1016/S0021-

9258(19)34186-9.

53. Haynes, R.C.,Jr., Picking, R.A., and Zaks, W.J. (1986)

Control of mitochondrial content of adenine nucleo-

tides by submicromolar calcium concentrations and

its relationship to hormonal effects, J. Biol. Chem.,

261, 16121-16125, https://doi.org/10.1016/s0021-9258

(18)66686-4.

54. Gout, E., Rébeillé, F., Douce, R., and Bligny, R. (2014)

Interplay of Mg2+, ADP, and ATP in the cytosol and

mitochondria: unravelling the role of Mg2+ in cell res-

piration, Proc. Natl. Acad. Sci. USA, 111, E4560-E4567,

https://doi.org/10.1073/pnas.1406251111.

55. Zuurendonk, P.F., and Tager, J.M. (1974) Rapid separa-

tion of particulate components and soluble cytoplasm

of isolated rat-liver cells, Biochim. Biophys. Acta, 333,

393-399, https://doi.org/10.1016/0005-2728(74)90022-x.

56. Akerboom, T.P.M., Bookelman,H., Zuurendonk, P.F.,

Meer, R., and Tager, J. M. (1978) Intramitochondrial

and extramitochondrial concentrations of adenine

nucleotides and inorganic phosphate in isolated he-

patocytes from fasted rats, Eur. J. Biochem., 84, 413-

420, https://doi.org/10.1111/j.1432-1033.1978.tb12182.x.

57. Tischler, M. E., Hecht,P., and Williamson, J.R. (1977)

Determination of mitochondrial/cytosolic metabolite

gradients in isolated rat liver cells by cell disrup-

tion, Arch. Biochem. Biophys., 181, 278-293, https://

doi.org/10.1016/0003-9861(77)90506-9.

58. Elbers, R., Heldt, H. W., Schmucker, P., Soboll, S.,

and Wiese, H. (1974) Measurement of the ATP/ADP

ratio in mitochondria and in the extramitochondri-

al compartment by fractionation of freeze-stopped

liver tissue in non-aqueous media, Hoppe Seylers Z.

Physiol. Chem., 355, 378-393, https://doi.org/10.1515/

bchm2.1974.355.1.378.

59. Sutton, R., and Pollak, J. K. (1978) The increasing

adenine nucleotide concentration and the mat-

uration of rat liver mitochondria during neona-

tal development, Differentiation, 12, 15-21, https://

doi.org/10.1111/j.1432-0436.1979.tb00985.x.

60. Schwenke, W. D., Soboll, S., Seitz, H. J., and Sies, H.

(1981) Mitochondrial and cytosolic ATP/ADP ratios

in rat liver in vivo, Biochem J, 200, 405-408, https://

doi.org/10.1042/bj2000405.

61. Soboll,S., Akerboom, T.P., Schwenke, W.D., Haase,R.,

and Sies, H. (1980) Mitochondrial and cytosolic ATP/

ADP ratios in isolated hepatocytes. A comparison of

the digitonin method and the non-aqueous fraction-

ation procedure, Biochem. J., 192, 951-954, https://

doi.org/10.1042/bj1920951.

62. Chen, W. W., Freinkman,E., Wang, T., Birsoy, K., and

Sabatini, D.M. (2016) Absolute quantification of ma-

trix metabolites reveals the dynamics of mitochon-

drial metabolism, Cell, 166, 1324-1337.e11, https://

doi.org/10.1016/j.cell.2016.07.040.

63. Soboll,S., and Bünger,R. (1981) Compartmentation of

adenine nucleotides in the isolated working guinea

pig heart stimulated by noradrenaline, Hoppe Seylers

Z. Physiol. Chem., 362, 125-132, https://doi.org/10.1515/

bchm2.1981.362.1.125.

64. Soboll, S., Scholz, R., and Heldt, H. W. (1978) Subcel-

lular metabolite concentrations. Dependence of mito-

chondrial and cytosolic ATP systems on the metabolic

state of perfused rat liver, Eur J Biochem, 87, 377-390,

https://doi.org/10.1111/j.1432-1033.1978.tb12387.x.

65. Pfaller,W., Guder, W.G., Gstraunthaler,G., Kotanko,P.,

Jehart, I., and Pürschel, S. (1984) Compartmentation

of ATP within renal proximal tubular cells, Biochim.

Biophys. Acta, 805, 152-157, https://doi.org/10.1016/

0167-4889(84)90162-9.

66. Yoshida,T., Alfaqaan,S., Sasaoka,N., and Imamura,H.

(2017) Application of FRET-based biosensor “ATeam”

for visualization of ATP levels in the mitochondri-

al matrix of living mammalian cells, Methods Mol.

Biol., 1567, 231-243, https://doi.org/10.1007/978-1-

4939-6824-4_14.

67. Kioka,H., Kato,H., Fujikawa,M., Tsukamoto,O., Suzu-

ki,T., Imamura,H., Nakano,A., Higo,S., Yamazaki,S.,

Matsuzaki,T., Takafuji,K., Asanuma,H., Asakura,M.,

Minamino,T., Shintani,Y., Yoshida,M., Noji,H., Kitaka-

ze,M., Komuro,I., Asano,Y., and Takashima,S. (2014)

Evaluation of intramitochondrial ATP levels identifies

G0/G1 switch gene 2 as a positive regulator of oxida-

tive phosphorylation, Proc. Natl. Acad. Sci. USA, 111,

273-278, https://doi.org/10.1073/pnas.1318547111.

68. Morciano,G., Imamura,H., Patergnani,S., Pedriali,G.,

Giorgi,C., and Pinton, P. (2020) Measurement of ATP

concentrations in mitochondria of living cells using

luminescence and fluorescence approaches, Meth-

ods Cell Biol., 155, 199-219, https://doi.org/10.1016/

bs.mcb.2019.10.007.

69. Imamura, H., Nhat, K. P. H., Togawa, H., Saito, K.,

Iino,R., Kato-Yamada,Y., Nagai,T., and Noji,H. (2009)

Visualization of ATP levels inside single living cells

with fluorescence resonance energy transfer-based

genetically encoded indicators, Proc. Natl. Acad.

Sci. USA, 106, 15651-15656, https://doi.org/10.1073/

pnas.0904764106.

ATP IN MITOCHONDRIA 1943

BIOCHEMISTRY (Moscow) Vol. 90 No. 12 2025

70. Gnaiger, E., Aasander Frostner, E., Abdul Karim, N.,

Abdel-Rahman, E. A., Abumrad, N. A., Acuna-Castro-

viejo,D., Adiele, R.C., etal. (2020) Mitochondrial phys-

iology, Bioenerg. Commun., https://doi.org/10.26124/

bec:2020-0001.v1.

71. Korshunov, S. S., Skulachev, V. P., and Starkov, A. A.

(1997) High protonic potential actuates a mechanism

of production of reactive oxygen species in mitochon-

dria, FEBS Lett., 416, 15-18, https://doi.org/10.1016/

s0014-5793(97)01159-9.

72. Skulachev, V. P. (1996) Role of uncoupled and

non-coupled oxidations in maintenance of safely

low levels of oxygen and its one-electron reductants,

Q. Rev. Biophys., 29, 169-202, https://doi.org/10.1017/

s0033583500005795.

73. Depaoli, M. R., Karsten, F., Madreiter-Sokolowski,

C. T., Klec, C., Gottschalk, B., Bischof, H., Eroglu, E.,

Waldeck-Weiermair,M., Simmen,T., Graier, W.F., and

Malli, R. (2018) Real-time imaging of mitochondrial

ATP dynamics reveals the metabolic setting of single

cells, Cell Rep., 25, 501-512.e3, https://doi.org/10.1016/

j.celrep.2018.09.027.

74. Suzuki, R., Hotta, K., and Oka, K. (2018) Transitional

correlation between inner-membrane potential and

ATP levels of neuronal mitochondria, Sci. Rep., 8,

2993, https://doi.org/10.1038/s41598-018-21109-2.

75. Choi, J., Matoba, N., Setoyama, D., Watanabe, D.,

Ohnishi, Y., Yasui, R., Kitai, Y., Oomachi, A.,

Kotobuki, Y., Nishiya, Y., Pieper, M. P., Imamura, H.,

Yanagita, M., and Yamamoto, M. (2023) The SGLT2

inhibitor empagliflozin improves cardiac energy

status via mitochondrial ATP production in diabetic

mice, Commun. Biol., 6, 278, https://doi.org/10.1038/

s42003-023-04663-y.

76. Botman,D., van Heerden, J.H., and Teusink,B. (2020)

An improved ATP FRET sensor for yeast shows het-

erogeneity during nutrient transitions, ACS Sens., 5,

814-822, https://doi.org/10.1021/acssensors.9b02475.

77. Gostimskaya, I.S., Grivennikova, V.G., Zharova, T.V.,

Bakeeva, L. E., and Vinogradov, A. D. (2003) In situ

assay of the intramitochondrial enzymes: use of ala-

methicin for permeabilization of mitochondria, Anal.

Biochem., 313, 46-52, https://doi.org/10.1016/s0003-

2697(02)00534-1.

78. Lapashina, A. S., Galkina, K. V., Markova, O. V.,

Knorre, D. A., and Feniouk, B.A. (2023) Observations

of ATP levels in mitochondria and whole yeast cells

using fluorescent probes, in Receptors and Intracellu-

lar Signaling Collected Articles (Berezhnov A. V., and

Zinchenko, V. P., eds) pp. 352-357 [in Russian].

79. Yaginuma,H., Kawai, S., Tabata, K.V., Tomiyama, K.,

Kakizuka, A., Komatsuzaki, T., Noji, H., and Imamu-

ra, H. (2014) Diversity in ATP concentrations in a

single bacterial cell population revealed by quanti-

tative single-cell imaging, Sci. Rep., 4, 6522, https://

doi.org/10.1038/srep06522.

80. Arai,S., Kriszt,R., Harada,K., Looi, L.-S., Matsuda,S.,

Wongso, D., Suo, S., Ishiura, S., Tseng, Y.-H.,

Raghunath, M., Ito, T., Tsuboi, T., and Kitaguchi, T.

(2018) RGB-color intensiometric indicators to visual-

ize spatiotemporal dynamics of ATP in single cells,

Angew. Chem. Int. Ed. Engl., 57, 10873-10878, https://

doi.org/10.1002/anie.201804304.

81. Takaine,M., Ueno,M., Kitamura,K., Imamura,H., and

Yoshida, S. (2019) Reliable imaging of ATP in living

budding and fission yeast, J. Cell Sci., 132, jcs230649,

https://doi.org/10.1242/jcs.230649.

82. Sawai, A., Taniguchi, T., Noguchi, K., Seike, T.,

Okahashi,N., Takaine,M., and Matsuda,F. (2025) ATP

supply from cytosol to mitochondria is an addition-

al role of aerobic glycolysis to prevent programmed

cell death by maintenance of mitochondrial mem-

brane potential, Metabolites, 15, 461, https://doi.org/

10.3390/metabo15070461.

83. Tsuno, S., Harada, K., Horikoshi, M., Mita, M.,

Kitaguchi, T., Hirai, M. Y., Matsumoto, M., and

Tsuboi, T. (2024) Mitochondrial ATP concentration

decreases immediately after glucose administration

to glucose-deprived hepatocytes, FEBS Open Bio, 14,

79-95, https://doi.org/10.1002/2211-5463.13744.

84. Berg, J., Hung, Y. P., and Yellen, G. (2009) A genet-

ically encoded fluorescent reporter of ATP:ADP ra-

tio, Nat. Methods, 6, 161-166, https://doi.org/10.1038/

nmeth.1288.

85. Tantama, M., Martínez-François, J. R., Mongeon, R.,

and Yellen, G. (2013) Imaging energy status in live

cells with a fluorescent biosensor of the intracellu-

lar ATP-to-ADP ratio, Nat. Commun., 4, 2550, https://

doi.org/10.1038/ncomms3550.

86. Zhang, M., Yang, B., Zhang, J., Song, Y., Wang, W.,

Li,N., Wang,Y., Li,W., and Wang,J. (2022) Monitoring

the dynamic regulation of the mitochondrial GTP-to-

GDP ratio with a genetically encoded fluorescent bi-

osensor, Angew. Chem. Int. Ed. Engl., 61, e202201266,

https://doi.org/10.1002/anie.202201266.

Publisher’s Note. Pleiades Publishing remains

neutral with regard to jurisdictional claims in published

maps and institutional affiliations. AI tools may have

been used in the translation or editing of this article.