Article

ISSN 0006-2979, Biochemistry (Moscow), 2026, Vol. 91, No. 6, pp. 1006-1022 © Pleiades Publishing, Ltd., 2026.

1006

Sex-Specific Relationships between Cerebral Cortex

Metabolism, Behavior, and ECG Parameters

in Rats under Physiological Conditions

and after Pyruvate Dehydrogenase Inhibition

Anastasia V. Graf

, Alexey V. Kazantsev

, and Victoria I. Bunik

3,4,5,a

Faculty of Biology, Lomonosov Moscow State University, 119991 Moscow, Russia

Faculty of Chemistry, Lomonosov Moscow State University, 119991 Moscow, Russia

Belozersky Research Institute of Physico-Chemical Biology, Lomonosov Moscow State University,

119991 Moscow, Russia

Faculty of Bioengineering and Bioinformatics, Lomonosov Moscow State University,

119991 Moscow, Russia

Department of Biochemistry, Sechenov University, 119991 Moscow, Russia

e-mail: bunik@belozersky.msu.ru

Received February 18, 2026

Revised June 3, 2026

Accepted June 5, 2026

Abstract— Sex-specific interactions between neurosignaling systems, which generate, propagate, and ter-

minate signals in nervous tissue, and metabolic pathways that support these processes may underlie sex

differences in adaptation and therapeutic efficacy. This study aimed to characterize these interactions as

systemic indicators of sex-specific adaptive responses in a rat model of metabolic stress induced by the

inhibition of pyruvate dehydrogenase complex (PDC), which catalyzes the key reaction linking anaerobic

glycolysis to aerobic glucose oxidation. To inhibit brain PDC, we used a single intranasal administration

of methyl acetylphosphinate (MeAcP), a phosphinate analog of pyruvate, or dimethyl acetylphosphonate

(AcPMe

), a membrane-permeable precursor of phosphonate pyruvate analogs. Effects were assessed 24  h

post-administration by measuring biochemical and physiological parameters in the cerebral cortex, includ-

ing glutamate levels, glutamine synthetase (GS) activity, and activities of enzymes in the tricarboxylic acid

(TCA) cycle and affiliated pathways. Neurosignaling was evaluated using surrogate indicators: ECG (electro-

cardiogram) parameters and spontaneous behavior in the open field test. Relationships between measured

parameters were analyzed using Spearman’s rank correlation coefficients, with the correlation strength clas-

sified according to the Chaddock’s scale. In control animals, no sex differences were observed in the mean

values of biochemical or ECG parameters. However, behavioral parameters (e.g., grooming and locomotion)

and the overall structure of correlations between the studied parameters exhibited marked sex dependence.

In control females, strong correlations were detected between ECG parameters and GS activity, whereas in

males, ECG parameters were strongly associated with malic enzyme (ME) activity. Male controls also showed

strong correlations between locomotor/exploratory behavior and activities of ME, PDC, and 2-oxoglutarate

dehydrogenase complex (OGDC). Administration of PDC inhibitors induced a sex-specific reorganization of

relationships between neurosignaling indicators and glutamate metabolism, which eliminated pronounced

sex differences in locomotor activity observed in controls, while revealing new sex-related differences in

glutamate levels, glutamate dehydrogenase (GDH) and ME activities, grooming bout duration, and freezing

time. The reduction in glutamate levels observed in females following PDC inhibition was consistent with

the established decrease in de  novo glutamate synthesis from glucose under conditions of impaired substrate

flux through the TCA cycle. Overall, these findings demonstrate that the relationships among metabolic,

behavioral, and ECG parameters are inherently sex-specific. Moreover, the homeostatic response of the

* To whom correspondence should be addressed.

SEX SPECIFIC RELATIONSHIPS BETWEEN BRAIN METABOLISM AND NEUROSIGNALING 1007

BIOCHEMISTRY (MOSCOW) Vol. 91 No. 6 2026

cerebral cortex to PDC inhibition reshapes these relationships, thereby modifying sex-dependent biochemical

and behavioral characteristics observed under control conditions.

DOI: 10.1134/S0006297926600444

Keywords: tricarboxylic acid cycle, pyruvate dehydrogenase complex, pyruvate dehydrogenase inhibition,

methyl acetyl phosphinate, dimethyl acetylphosphonate, glutamine synthetase, glutamate dehydrogenase, malic

enzyme, 2-oxoglutarate dehydrogenase complex, glutamate, sex differences, heart rate variability, behavior,

correlation analysis

INTRODUCTION

The multienzyme pyruvate dehydrogenase com-

plex (PDC) plays a central role in cellular energy

metabolism by catalyzing irreversible oxidation of

pyruvate, thereby linking cytoplasmic glycolysis with

mitochondrial tricarboxylic acid (TCA) cycle and ox-

idative phosphorylation  [1]. Hereditary PDC deficien-

cies cause severe neurological pathologies  [2,  3], while

reduced PDC activity in the brain correlates with

cognitive deficits and mitochondrial dysfunction in

neurodegenerative disorders such as Alzheimer’s dis-

ease  [4]. These correlations may stem from the asso-

ciation between PDC function and generation of key

neurotransmitters. For instance, PDC-dependent oxida-

tion of pyruvate (glycolysis end product) in the mito-

chondrial TCA cycle is linked to the de  novo synthesis

of glutamate, the principal excitatory neurotransmit-

ter, from glucose  [5]. Selective PDC inhibition reduces

brain glutamate levels  [6]. Glucose-derived acetyl-CoA

generated by PDC is the primary precursor for acetyl-

choline synthesis in cholinergic neurons [7].

Highly specific and efficient in  vivo inhibition

of PDC can be achieved by using synthetic pyruvate

analogs that either directly inhibit pyruvate dehydro-

genase (PDH) or act as membrane-permeable precur-

sors that generate such inhibitors intracellularly  [6,

  9]. Intranasal administration of pyruvate analogs

ensures their delivery to the brain while bypassing

the blood–brain barrier, thus providing a useful mod-

el of metabolic stress resulting from impaired oxida-

tive glucose metabolism [6, 10].

Many physiological responses to stress are sex-de-

pendent. Accumulating evidence suggests that these

differences are closely linked to glutamate metabo-

lism. For example, female rats have been reported to

display higher basal levels of glutamate in the cere-

bral cortex

[11,

12] and higher glutamine synthetase

(GS) activity

[13], compared to males. Sex hormones,

particularly estrogens, modulate the glutamatergic

system

  [14,  15] and activities of glutamate dehydro-

genase (GDH) [16] and GS  [10], while brain glutamate

levels and interactions between glutamate and other

neurotransmitters influence sexual behavior in ani-

mals  [17].

General principles of sexual dimorphism suggest

that male mammals may prioritize metabolic strate-

gies supporting active behavior and motor functions

more than females  [18,  19]. Consequently, inherent

sex-dependent differences in the relationships be-

tween metabolic and neurosignaling systems in males

and females may shape their adaptive responses to

metabolic disorders. Despite this possibility, a com-

prehensive analysis of sex-specific relationships be-

tween brain metabolism, autonomic regulation, and

behavior under both physiological and metabolically

impaired conditions is lacking. These differences are

also typically overlooked during the development of

therapeutic interventions. For example, in TgF344-AD

rats (a model of Alzheimer’s disease), administration

of a cocktail of vitamin activators, including benfoti-

amine (PDC cofactor precursor), increased acetylcho-

line levels in the hippocampus in response to physio-

logical stimulation

  [20]. Pharmacological activation of

PDC has also been explored as a strategy to alleviate

obesity-associated cardiomyopathy and non-alcoholic

fatty liver disease  [21,  22]. However, none of these

studies considered potential sex differences in thera-

peutic responses, despite established associations be-

tween the pharmacological effects of thiamine, which

activates PDC and other thiamine-dependent enzymes,

and biological sex

  [23,  24], which emphasizes the im-

portance of investigating sex-dependent mechanisms

of PDC regulation.

The present study aimed to characterize sex dif-

ferences in the systemic organization of physiologi-

cal and metabolic networks in rats under normal

conditions and 24  h after administration of pyruvate

analogs inhibiting PDC. Because modifications that

increase membrane permeability may enhance the

intracellular availability of PDC inhibitors, we com-

pared the effects of two structurally related pyru-

vate analogs: the charged methyl acetylphosphinate

(MeAcP) and the uncharged dimethyl acetylphospho-

nate (AcPMe

). MeAcP acts as a direct PDC inhibitor

but requires transporter-mediated passage across

biological membranes, similarly to pyruvate, while

AcPMe

is a precursor of PDC inhibitors (acetylphos-

phonate and its methyl ester) that is converted into

these compounds intracellularly by esterases (Fig.  1).

GRAF et al.1008

BIOCHEMISTRY (MOSCOW) Vol. 91 No. 6 2026

Fig. 1. Structures of phosphinate (MeAcP) and phosphonate (AcPMe

) pyruvate analogs used in the study. The figure illus-

trates the conversion of AcPMe

(

uncharged membrane-permeable precursor) into PDC inhibitors methyl acetylphosphonate

(AcPMe) and acetylphosphonate (AcP) via sequential hydrolysis its methyl ester groups. Previously reported [8, 9, 25] inhi-

bition constants (K

) of these pyruvate analogs toward human and animal PDC are indicated.

To address the study objectives, we examined

sex differences both in the intrinsic relationships be-

tween PDC-dependent metabolic parameters of the

cerebral cortex and neurosignaling processes, and in

the systemic reorganization of these relationships in

response to PDC dysfunction. Neurosignaling process-

es were evaluated using behavioral and heart rate

regulation parameters as surrogate indicators.

MATERIALS AND METHODS

Reagents. AcPMe

and MeAcP sodium salt were

synthesized as described in  [6]. The purity of synthe-

sized compounds was confirmed by nuclear magnet-

ic resonance (NMR) spectroscopy analysis, including

the analysis immediately prior to animal experi-

ments. Other reagents were purchased from Sigma-

Aldrich (USA).

Animal experiments. All animal procedures

were conducted in accordance with the EU Directives

86/609/EEC and 2010/63/EU and were approved by the

Bioethics Committee of Lomonosov Moscow State Uni-

versity (protocol no.  137-d; November  11, 2021). Male

and female Wistar rats were housed under standard

conditions with a 12-h light/dark cycle and ad  libitum

access to water and food. Males (n  =  24) aged 9–10

weeks and weighing 320  ±  15  g and females (n = 25)

aged 8.5-10 weeks and weighing 195  ±  10  g were used

in the study.

When assigning animals to experimental groups,

estrous cycle stages (proestrus, estrus, metestrus, di-

estrus) were taken into account to ensure a compa-

rable distribution of estrous cycle stages across all

groups. This approach minimized potential effects of

cyclic hormonal fluctuations on the studied param-

eters. To further reduce variability associated with

circadian rhythms, all experimental procedures, in-

cluding inhibitor administration, behavioral testing,

and sample collection, were performed at the same

time of day in both sexes.

Sample sizes were as follows: female control

group, n  =  8; female AcPMe

group, n  =  9; female

MeAcP group, n  =  8; all male groups (control, AcPMe

MeAcP), n  =  8 for each group.

Pyruvate analogs were dissolved in deionized

water to obtain 1  M solutions and administered in-

tranasally at a dose of 0.1  mmol/kg using a micropi-

pette. The solution was applied alternately to each

nostril, one drop at a time, to ensure uniform distri-

bution and prevent runoff. The volume administered

to each nostril did not exceed 20  μL. Control animals

received an equivalent volume of physiological saline

(0.9%  NaCl) using the same procedure.

This experimental design was based on previous

studies investigating in  vivo effects of selective inhi-

bition of 2-oxoacid dehydrogenases by phosphonate

analogs of their 2-oxo substrates [6,  26,  27]. Intra-

nasal administration was selected as a noninvasive

and easy route for delivering pyruvate analogs to the

brain while bypassing the blood–brain barrier. A sin-

gle administration was used to model acute metabolic

stress. Although neurotoxic agents administered in-

tranasally or intraperitoneally may induce behavioral

alterations through impairment of olfactory function

[28,  29], such effects were unlikely in the present

study because phosphonate analogs of 2-oxoacids

have not exhibited overt toxicity under similar ex-

perimental conditions [6, 26, 27]. Furthermore, the

behavioral parameters assessed in the open field test

were not dependent on olfactory cues.

Pharmacokinetic studies were not conducted;

the effects of treatment were assessed by measuring

physiological and biochemical parameters 24  h after

inhibitor administration. This time point was selected

based on previous studies [6,  26,  27] to exclude po-

tential influences of short-term stress associated with

substance administration.

SEX SPECIFIC RELATIONSHIPS BETWEEN BRAIN METABOLISM AND NEUROSIGNALING 1009

BIOCHEMISTRY (MOSCOW) Vol. 91 No. 6 2026

Following physiological measurements, the ani-

mals were euthanized by decapitation. Brains were

extracted, and the cerebral cortex was dissected

on ice and immediately frozen in liquid nitrogen

for 90  s. Tissue samples were stored at −70°C until

biochemical analysis.

Physiological tests. Neurosignaling processes

were evaluated using surrogate indicators, such as

spontaneous behavior in the open field test and heart

rate variability (HRV) parameters, which reflect the

contributions of autonomic regulation to behavioral

responses.

Open field test was conducted as previously de-

scribed [30]. Animals were observed for 3  min in

complete silence, under illumination with a 15  W red

lamp. The following parameters were recorded: loco-

motor activity (number of grid lines crossed), num-

ber of center entries, total time spent in the center,

number of rearings, total grooming time, number of

grooming bouts, mean duration of individual groom-

ing bout, and total freezing time.

ECG recording was performed for 3  min using

a non-invasive electrode placement. Disposable sil-

ver/silver chloride electrodes for newborns were at-

tached to the shaved ventrolateral chest surface and

secured using a custom harness with built-in ter-

minals to minimize motion artifacts. The terminals

were connected via harness wires to a connector

positioned at the withers and linked to a biopoten-

tial amplifier (10-20  kHz frequency band). The signal

was digitized at 1  kHz with an E14-440 analog-to-dig-

ital converter (L-Card, Russia) connected via USB to

a computer with the PowerGraph software (DISoft,

Russia).

Animals were pre-adapted to the recording con-

ditions. Two to three days before the experiment, rats

were placed in individual dark recording pens: first

without the harness (5-10  min), then with the har-

ness (10-15  min), until calm behavior was achieved.

On the day of the experiment, a brief additional ad-

aptation procedure was conducted before recording.

ECG was recorded in the first standard lead  [30].

Autonomic heart regulation was assessed by ana-

lyzing the time-domain HRV based on RR interval se-

quences as described previously  [30]. Parasympathet-

ic activity markers included mean RR interval (ms),

total variability evaluated as standard deviation (SD)

of all RR intervals (ms), variation range (dX, differ-

ence between maximum and minimum RR intervals,

ms), and RMSSD (root mean square of successive RR

interval differences, ms).

Sympathetic tone was characterized by the stress

index (SI), calculated from the following equation:

SI  =  AMo/(2×Mo×dX), where Mo is mode (most fre-

quent RR interval value,  s) and AMo is mode ampli-

tude (% of RR intervals corresponding to the mode).

Assessment of metabolic parameters in the

cerebral cortex. Glutamate concentrations were

determined in methanol–acetic acid extracts of the

cerebral cortex using an enzymatic assay based on

GDH reaction  [31]. Cerebral cortex homogenates were

prepared, and the activities of GDH, malate dehydro-

genase (MDH), NADP

-dependent malic enzyme (ME),

isocitrate dehydrogenase (IDH), GS, PDC, and OGDC

were measured using established methods  [26,  32].

Enzyme activities were expressed as μmol of sub-

strate converted per minute per gram of wet tissue

(μmol·min

−1

·g

−1

). PDC and OGDC activities were mea-

sured in the presence of thiamine diphosphate.

When analyzing enzyme activities, we account-

ed for the reversibility of PDC inhibitors used, which

precluded reliable assessment of in  vivo PDC inhibi-

tion based solely on activity measurements in tissue

homogenates  [6,  33]. Therefore, the in  vivo effects of

PDC inhibitors were inferred from a composite pro-

file of PDC-related metabolic alterations measured

in  vitro relative to control animals.

Statistical analysis. Data were analyzed using

STATISTICA 12.0 and GraphPad Prism 10.4.1. Results

are presented as mean  ±  standard error of mean

(SEM). A two-way ANOVA was used to evaluate the

effects of “sex” and “inhibitor” factors, as well as

their interaction, followed by Sidak’s post hoc test to

assess differences between groups. Differences were

considered significant at p  ≤  0.05 and indicative of a

trend at p  ≤  0.1.

Correlation analysis was performed used Spear-

man’s rank correlation coefficient. The strength of

associations was interpreted according to the Chad-

dock’s scale  [34]. Strong correlations (r  ≥  0.7) were

characterized by p  ≤  0.05. No correction for multiple

comparisons was applied, as the primary interpreta-

tion relied on correlation strength and because appro-

priate correction is nontrivial in heterogeneous cor-

relation structures  [35]. Since conclusions were based

on convergent correlation patterns, the likelihood of

random associations was substantially reduced.

RESULTS

Effect of AcPMe

and MeAcP on sexual dimor-

phism in behavioral and ECG parameters. Table  1

summarizes two-way ANOVA results for ECG and

open field test parameters in male and female rats

across control and PDC inhibitor groups. No statis-

tically significant main effects of sex or treatment

were observed for the HRV values. However, analy-

sis across all groups revealed trends toward the ef-

fect of sex on mean RR interval duration (F

1,43

  =  3.7;

p  =  0.06) and toward the effect of PDC inhibitors on

RMSSD (F

2,43

  =  2.4; p  =  0.11) (Table  1).

GRAF et al.1010

BIOCHEMISTRY (MOSCOW) Vol. 91 No. 6 2026

In contrast to ECG measures, behavioral parame-

ters in the control groups demonstrated clear sexual

dimorphism. Compared with control males, control

females showed higher locomotor activity, increased

numbers of center entries, and more frequent groom-

ing bouts.

PDC inhibitors modified sex differences in be-

havioral parameters, as evidenced by significant

“sex”  ×  “inhibitor” interactions for freezing time

2,44

  =  3.6; p  =  0.04) and number of grooming bouts

2,44

  =  4.6; p  =  0.02) (Table  1). Females exhibited

greater sensitivity to both compounds; specifically,

AcPMe

significantly reduced grooming bouts, while

MeAcP significantly increased freezing time. MeAcP in-

duced a sex difference in freezing time that was ab-

sent under control conditions. Inhibitor treatment also

amplified sex differences in grooming bout duration

(Table  1). Importantly, the reduction in the frequency

of grooming bouts in females following MeAcP was not

accompanied by changes in the locomotor activity, sug-

gesting that this effect is unlikely to reflect nonspecific

behavioral suppression or motor impairment. The de-

crease in grooming was observed only in females and

was combined with increased freezing time, indicat-

ing a broader reorganization of behavioral patterns.

Consistent with previous reports, reduction in groom-

ing frequency in conjunction with other behavioral

changes may reflect alterations in emotional state and

should be interpreted differently than an isolated de-

crease in the number of grooming bouts  [36].

Unlike duration of grooming bout and freezing

time, where PDC inhibitors induced sex differences

absent in controls, sex differences in locomotor ac-

tivity observed in controls were abolished following

PDC inhibitor treatment, due to opposing directional

effects in males and females (Table  1).

Therefore, PDC inhibitors altered sexual dimor-

phism in rat behavior, primarily due to a greater re-

sponsiveness in females compared with males. Incon-

trast, ECG parameters remained largely unaffected in

both sexes. This difference suggests that intranasally

administered PDC inhibitors may preferentially influ-

ence central neural circuits governing behavior rath-

er than peripheral autonomic mechanisms controlling

cardiac activity.

Effect of AcPMe

and MeAcP on glutamate

levels and activities of enzymes of the pyruvate

and glutamate metabolic nodes. We assessed the

activities of enzymes of the pyruvate and glutamate

metabolic nodes, as well as glutamate levels, in the

cerebral cortex. As shown in Table  2, control males

and females did not differ significantly in any of

the measured parameters. However, administration

of PDC inhibitors revealed sex-dependent differenc-

es in several metabolic indices. Two-way ANOVA re-

vealed significant effects of sex on the ME activity

1,44

  =  14.3; p  =  0.0005), GDH activity (F

1,44

  =  12.1;

p  =  0.001), and glutamate levels (F

1,44

  =  6.8; p  =  0.01).

Trend-level effects of sex (p  ≤  0.1) were observed for

the OGDC and GS activities. A significant “sex”  ×  “in-

hibitor” interaction was established for MDH activity

2,43

  =  3.9; p  =  0.03). In addition, PDC activity was

also significantly affected by the “inhibitor” factor

2,44

  =  4.0; p  =  0.02) (Table  2).

Biochemical alterations relative to the control

were observed exclusively in females and were most

pronounced following administration of the stronger

inhibitor, MeAcP. In these animals, glutamate levels

were significantly reduced compared with controls,

and PDC activity was lower than in the AcPMe

-treat-

ed group. No significant treatment-induced changes

were detected in males (Table  2).

Therefore, PDC inhibitors induced sex-specific

differences in parameters associated with both the

pyruvate metabolic node (ME activity) and the glu-

tamate metabolic node (GDH activity and glutamate

levels), whereas no such differences were present

in control animals. However, significant biochem-

ical changes relative to controls were observed in

females only, particularly a reduction in glutamate

levels after MeAcP treatment. These results indicate

a greater metabolic sensitivity of the female cerebral

cortex to PDC inhibition.

Correlation analysis between physiological

and biochemical parameters in control male and

female rats. Given the close relationship between

physiological functions and cerebral cortex metabo-

lism, correlation analysis was performed to charac-

terize interactions between metabolic and neurosig-

naling systems. Correlation strength was assessed

using the absolute value of the Spearman’s rank cor-

relation coefficient  (r) according to the Chaddock’s

scale  [34]. Strong correlations (r  ≥  0.7) were consid-

ered biologically meaningful only when supported by

statistical significance (p  ≤  0.05), visual inspection of

correlation networks (Fig.  S1, Online Resource  1), and

distinct sex-specific correlation patterns (Table  S1,

Online Resource  1), as the marked differences in

correlation structures between males and females

suggest that these associations are unlikely to have

arisen by chance. Inaddition to statistical and graph-

ical validation, the observed correlations in control

animals were consistent with established physiolog-

ical mechanisms. In both sexes, parameters reflect-

ing parasympathetic regulation (dX, RMSSD, and SD)

positively correlated with one another, whereas each

of them exhibited negative correlations with the

stress index (SI), an indicator of sympathetic activity

(Table  S1, Online Resource  1). The concordant pos-

itive relationships among parasympathetic markers

(dX, RMSSD, SD) and their inverse relationships with

SI are consistent with the physiological antagonism

SEX SPECIFIC RELATIONSHIPS BETWEEN BRAIN METABOLISM AND NEUROSIGNALING 1011

BIOCHEMISTRY (MOSCOW) Vol. 91 No. 6 2026

Table  1. HRV and behavioral parameters in the open field test in females  (♀) and males  (♂) of the control

groups (n  =  8 in each group) and after administration of AcPMe

(females, n  =  9; males, n  =  8) or MeAcP (n = 8

in each group)

Physiological parameter Group Control AcPMe

MeAcP

RR intervals

(Factor “sex” F

1,43

 = 3.7; p = 0.06)

♀ 116.2  ±  2.0 126.0  ±  4.1 117.4  ±  1.3

♂ 125.8  ±  3.2 123.4  ±  2.6 129.0  ±  7.2

SD ♀ 4.7  ±  0.7 6.2  ±  0.9 5.5  ±  0.4

♂ 5.0  ±  0.8 5.8  ±  0.8 7.1  ±  2.1

dX ♀ 38.9  ±  6.7 47.7  ±  5.3 47.9  ±  2.9

♂ 37.9  ±  5.3 63.1  ±  10.9 56.6  ±  18.8

RMSSD

(Factor “inhibitor” F

2,43

 = 2.4; p = 0.11)

♀ 2.9  ±  0.4 4.7  ±  0.8 4.9  ±  0.8

♂ 3.4  ±  0.6 4.3  ±  1.1 6.3  ±  2.2

SI ♀ 2.1  ±  0.4 1.1  ±  0.3 0.9  ±  0.1

♂ 1.8  ±  0.5 0.9  ±  0.2 2.2  ±  0.9

Number of grooming bouts

(Factor “sex” F

1,43

 = 71.0; p < 0.0001;

Factor “inhibitor” F

1,43

 = 3.7; p = 0.06;

Interaction F

2,43

 = 3.2; p = 0.048)

♀ 15.5  ±  1.7*

10.6  ±  0.7*

11.6  ±  1.3*

♂ 2.8  ±  1.0

2.0  ±  0.5

5.6  ±  2.1

Duration of a single grooming bout

(Factor “sex” F

1,43

 = 15.3; p = 0.0003)

♀ 0.9  ±  0.1 1.5  ±  0.3 1.7  ±  0.3

♂ 4.2  ±  1.3 4.3  ±  1.3 5.2  ±  1.6

Total grooming time ♀ 13.8  ±  2.2 15.9  ±  3.6 18.2  ±  3.2

♂ 12.6  ±  5.1 9.0  ±  3.5 21.0  ±  5.2

Freezing time

(Interaction of factors “sex” and “inhibitor”

1,44

 = 12.1; p = 0.001)

♀ 1.6  ±  0.8* 6.9  ±  2.8 12.4  ±  5.5*

♂ 6.9  ±  3.1 3.3  ±  2.0 1.3  ±  0.4

Time in the center ♀ 23.2  ±  2.4 35.8  ±  5.2 26.6  ±  6.1

♂ 28.9  ±  9.5 31.1  ±  4.0 22.2  ±  6.8

Number of rearings

(Factor “sex” F

1,43

 = 2.6; p = 0.11)

♀ 25.4  ±  1.9 16.9  ±  1.8 20.2  ±  3.1

♂ 16.0  ±  2.7 16.9  ±  4.7 17.9  ±  2.8

Number of center entries

(Factor “sex” F

1,43

 = 16.9; p = 0.0002)

♀ 7.9  ±  1.0

6.3  ±  1.3 7.4  ±  1.3

♂ 3.6  ±  0.9

3.8  ±  0.9 3.3  ±  0.9

Locomotor activity

(Factor “sex” F

1,43

 = 8.9; p = 0.005)

♀ 73.3  ±  6.9

65.8  ±  5.9 62.3  ±  8.1

♂ 49.5  ±  6.5

55.4  ±  4.1 51.9  ±  3.4

Note. For each physiological parameter, two-way ANOVA results (F and p values) are provided for the significances of fac-

tors and their interactions with p  ≤  0.1. The results of Sidak’s post hoc test are indicated as follows: *  values significantly

different due to PDC inhibition; #  values significantly different between females and males. Significant (p  <  0.05) factors and

differences are marked in bold; trends (p  ≤  0.1) are indicated in italics.

GRAF et al.1012

BIOCHEMISTRY (MOSCOW) Vol. 91 No. 6 2026

Table 2. Activities of enzymes of the pyruvate and glutamate metabolic nodes and glutamate levels in female(♀)

and male (♂) rats of control groups (n = 8 in each group) and after administration of AcPMe

(females, n = 9;

males, n = 8) or MeAcP (n = 8 in each group)

Biochemical parameter Group Control AcPMe

MeAcP

MDH

(Interaction of factors F

2,43

 = 3.9; p = 0.03)

♀ 456.9  ±  39.9 433.8  ±  31.4 401.5  ±  14.7

♂ 373.7  ±  14.7 354.1  ±  16.4 428.7  ±  23.5

(Factor “sex” F

1,44

 = 14.3; p = 0.0005)

♀ 0.69  ±  0.06 0.63  ±  0.10 0.61  ±  0.07

♂ 0.82  ±  0.08 0.92  ±  0.10 0.95  ±  0.09

PDC

(Factor “inhibitor” F

2,44

 = 4.0; p = 0.02)

♀ 0.58  ±  0.05 0.72  ±  0.07^ 0.36  ±  0.08

♂ 0.54  ±  0.11 0.62  ±  0.10 0.44  ±  0.09

IDH ♀ 1.11  ±  0.12 0.89  ±  0.07 0.97  ±  0.06

♂ 1.08  ±  0.08 0.96  ±  0.07 1.03  ±  0.08

OGDC

(Factor “inhibitor” F

2,44

 = 2.1; p = 0.1)

♀ 0.71  ±  0.07 0.67  ±  0.07 0.59  ±  0.06

♂ 0.91  ±  0.04 0.92  ±  0.04 0.87  ±  0.04

GDH

(Factor “sex” F

1,44

 = 12.1; p = 0.001)

♀ 1.41  ±  0.10 1.84  ±  0.24

1.90  ±  0.14

♂ 1.35  ±  0.10 1.31  ±  0.11

1.23  ±  0.09

(Factor “sex” F

1,44

 = 2.4; p = 0.13)

♀ 72.1  ±  2.3 78.2  ±  2.9 72.9  ±  2.2

♂ 79.3  ±  3.7 78.9  ±  4.1 77.6  ±  4.2

Glutamate

(Factor “inhibitor” F

2,44

 = 4.2; p = 0.02;

Factor “sex” F

1,44

 = 6.8; p = 0.01;

Interaction of factors F

2,44

 = 2.8; p = 0.07)

♀ 14.6  ±  0.5* 14.9  ±  0.7^ 12.4  ±  0.5*

♂ 15.1  ±  0.2 15.3  ±  0.5 14.9  ±  0.4

Note. For each biochemical parameter, two-way ANOVA results (F and p  values) are provided for the significances of fac-

tors and their interactions with p  ≤  0.1. The results of Sidak’s post hoc test are indicated as follows: *  values significantly

different due to PDC inhibition; #  values significantly different between females and males; ^  values significantly different

between AcPMe

and MeAcP groups. Significant (p  <  0.05) factors and differences are marked in bold; trends (p  ≤  0.1) are

indicated in italics.

between parasympathetic and sympathetic branches

of autonomic heart rate regulation. Furthermore, the

overall patterns of correlations between autonomic

regulation parameters in males and females were

similar, reflecting general principles of autonomic

regulation for both sexes.

In contrast to autonomic regulation, correlations

between autonomic and behavioral parameters ex-

hibited pronounced sexual dimorphism, although the

direction of these associations remained physiologi-

cally plausible. For example, exploratory behavior

correlated positively with increased parasympathetic

activity and reduced sympathetic activation. Inmales,

exploratory activity showed strong positive correla-

tions with parasympathetic regulation parameters.

Specifically, the number of rearings correlated posi-

tively with SD (r  =  0.8) and dX (r  =  0.9). The number

of center entries also correlated positively with SD

(r  =  0.8) and dX (r  =  0.8). The locomotor activity was

positively associated with dX (r  =  0.71). Conversely, SI

negatively correlated with exploratory behavior, in-

cluding the number of rearings (r  =  −0.79) and center

entries (r  =  −0.75). In females, these correlations were

absent, whereas several associations not observed in

males emerged. For example, the RR interval dura-

tion correlated positively with the number of groom-

ing bouts (r  =  0.73). These sex-specific relationships

between behavioral and ECG parameters may reflect

sex differences in the coupling between behavior and

autonomic regulation.

SEX SPECIFIC RELATIONSHIPS BETWEEN BRAIN METABOLISM AND NEUROSIGNALING 1013

BIOCHEMISTRY (MOSCOW) Vol. 91 No. 6 2026

Sexual dimorphism was also evident in correla-

tions between behavioral parameters in control ani-

mals (Table  S1, Online Resource  1). In females, freez-

ing time positively correlated with total grooming

time (r  =  0.71), whereas locomotor activity negatively

correlated with the number of rearings (r  =  −0.78).

Inmales, these parameters were positively correlated

(r  =  0.81). These differences are consistent with the

behavioral profile of females, which exhibited great-

er locomotor activity but fewer rearings than males

(Table  1). Moreover, the female-specific positive cor-

relation between rearings and grooming behavior

(r  =  0.75) may indicate a linkage between vertical ex-

ploratory activity and emotional reactivity that is not

characteristic of males.

Tables  3 and  4 present selected correlations be-

tween physiological parameters and metabolic char-

acteristics of the pyruvate (Table  3) and glutamate

(Table  4) nodes extracted from the complete correla-

tion matrix (Table  S1 in the Online Resource  1). Over-

all, behavioral and ECG parameters – indicators of

neurosignaling function – were predominantly asso-

ciated with the pyruvate metabolic node in males and

the glutamate metabolic node in females. In males,

seven behavioral and ECG parameters showed strong

correlations with the activity of ME, a decarboxylat-

ing malate dehydrogenase that generates pyruvate

(Table  3). In the cerebral cortex, the ME activity was

strongly associated with HRV indices, including SD

(r  =  0.81) and dX (r  =  0.71), and negatively associat-

ed with SI (r  =  −0.71). The ME activity also correlat-

ed positively with exploratory behavior parameters,

including the number of rearings (r  =  0.80), number

of center entries (r  =  0.84), and locomotor activity

(r  =  0.71). There were also trends toward correla-

tions between the locomotor activity and PDC activity

(r  =  −0.69) and between the grooming bout duration

and MDH activity (r  =  0.68).

Analysis of activities of enzymes belonging to

the glutamate metabolic node revealed only three

strong correlations in control males: IDH activity

correlated negatively with the time spent in the cen-

ter (r  =  −0.89) and positively with the grooming bout

duration (r  =  0.71), whereas GDH activity correlated

negatively with freezing time (r  =  −0.81). Trends to-

ward positive associations between time spent in the

center and both OGDC and GDH activities were also

observed (r  =  0.68, p  <  0.1) (Table  4).

Unlike control males characterized by the lack

of significant associations between GS activity and

Table  3. Spearman’s correlations between the enzymatic activities of pyruvate node and physiological param-

eters in females  (♀) and males  (♂) of control groups (n  =  8 in each group) and after administration of AcPMe

(females, n  =  9; males, n  =  8) or MeAcP (n  =  8 in each group)

Control

MDH

ME PDC

♀♂♀

♂ ♀♂

RR intervals −0.12 0.10 0.45 0.48 0.29 0.02

SD −0.4 0.36 −0.26 0.81 −0.26 −0.26

dX −0.4 0.26 −0.22

0.71 −0.31 −0.55

RMSSD −0.02 0.05 −0.4 0.36 −0.62 −0.31

SI 0.26 −0.33 0.21 −0.71 0.38 0.21

Number of grooming bouts 0.32 0.17 −0.06 −0.02 −0.26 0.54

Duration of a grooming bout 0.29

0.68 0.33 −0.14 0.10 −0.07

Total grooming time 0.43 0.36 −0.10 −0.17 −0.43 0.60

Freezing time 0.25 −0.19 −0.25 −0.78 −0.30 0.20

Time in center −0.50 −0.07 0.43 0.50 0.50 −0.11

Number of rearings 0.18 0.43 −0.55 0.80 −0.50 −0.54

Number of center entries −0.11 0.57 −0.14

0.84 −0.22 −0.29

Locomotor activity −0.12 0.45 0.46

0.71 0.43 −0.69

GRAF et al.1014

BIOCHEMISTRY (MOSCOW) Vol. 91 No. 6 2026

Table 3 (cont.)

AcPMe

MDH ME PDC

♀ ♂♀♂♀♂

RR intervals −0.63 0.31 −0.20 −0.64 −0.02 0.36

SD −0.54 0.36 −0.44 −0.57 −0.55 0.45

dX −0.47 −0.07 −0.40 −0.12 −0.58 0.33

RMSSD −0.15 −0.24 −0.22 0.17 −0.52 −0.24

0.62 −0.12 0.40 0.62 0.30 −0.57

Number of grooming bouts 0.33 0.48 −0.03 0.18 −0.18 0.05

Duration of a grooming bout −0.20 0.32 0.07 −0.11 0.23 −0.14

Total grooming time 0.13 0.47 0.03 0.02 0.18 −0.28

Freezing time −0.53 0.21 −0.27 0.21 0.02

−0.75

Time in center 0.70 −0.01 0.08 −0.34 0.07 −0.04

Number of rearings −0.25 −0.14 −0.57 0.17 −0.48 0.43

Number of center entries −0.17 −0.02 −0.11 −0.39 0.28 0.61

Locomotor activity −0.05 0.33 0.38 −0.43 0.52 0.52

MeAcP

MDH ME PDC

♀ ♂ ♀ ♂♀♂

RR intervals 0.60 −0.36 0.17 −0.43 0.17 0.31

SD 0.71 0.19 0.67 0.24 0.12 −0.31

dX −0.11 0.26 0.06 0.21 0.50 −0.14

RMSSD 0.33 0.19 0.48 0.26 0.24 −0.29

SI −0.31 −0.21 −0.38 −0.24 −0.14 0.50

Number of grooming bouts 0.19 −0.40 −0.18 −0.32 −0.05

0.72

Duration of a grooming bout 0.19 −0.11 0.07 0.14 0.12 −0.07

Total grooming time −0.17 −0.29 −0.24 −0.10 0.00 0.45

Freezing time 0.19 −0.28 −0.07 −0.49 −0.19 −0.40

Time in center

0.67 0.61 0.93 0.44 0.19 −0.49

Number of rearings −0.10 −0.19 0.48 −0.18 0.14 0.23

Number of center entries 0.66 0.49 0.68 0.49 −0.15 −0.31

Locomotor activity 0.40 0.24 0.50 0.29 0.14 0.02

Note. Each cell shows the Spearman’s rank correlation coefficient  (r) for the respective parameter pair. Significant correlations

(p  <  0.05) are shown in red bold; trends (p  <  0.1) are indicated in red italics. PDC activity was measured in the presence of

thiamine diphosphate. The enzymes and sexes of control or experimental groups, showing significant correlations with the

physiological parameters, are marked in red.

SEX SPECIFIC RELATIONSHIPS BETWEEN BRAIN METABOLISM AND NEUROSIGNALING 1015

BIOCHEMISTRY (MOSCOW) Vol. 91 No. 6 2026

Table  4. Spearman’s correlations between the activities of glutamate node enzymes and physiological parameters in females  (♀) and males  (♂) of

control groups (n  =  8 in each group) and after administration of AcPMe

(females, n  =  9; males, n  =  8) or MeAcP (n  =  8 in each group)

Control

IDH OGDC GDH GS Glutamate

♀

♂ ♀♂♀♂♀♂♀♂

RR interval 0.10 0.10 −0.17 −0.12 0.31 0.48 0.14 −0.55 −0.26 0.45

SD −0.02 −0.38 −0.21 −0.07 0.21 0.52

−0.88 0.14 0.02 −0.05

dX −0.11 −0.43 −0.3 −0.02 0.26 0.38

−0.84 0.24 0.01 −0.24

RMSSD 0.05 −0.14 −0.19 −0.50 −0.1 0.00 −0.62 0.48 0.36 −0.55

SI 0.05 0.33 0.31 0.12 −0.14 −0.57

0.74 0.10 −0.12 −0.14

Number of grooming bouts −0.40 0.17 0.04 −0.56 −0.6 −0.02 0.02 0.34 0.63 −0.22

Duration of a grooming bout 0.33

0.71 0.17 −0.29 0.38 −0.54 0.19 −0.32 −0.05 −0.11

Total grooming time −0.19 0.48 −0.17 −0.45 −0.12 −0.33 −0.05 0.07 0.48 −0.05

Freezing time −0.14 0.23 0.22 −0.58 0.11

−0.81 −0.16 −0.05 0.41 −0.06

Time in center −0.24

−0.89 −0.35 0.68 0.45 0.68 −0.28 −0.04 −0.50 0.50

Number of rearings −0.60 −0.46 −0.47 0.25 −0.37 0.57 0.08 0.11 0.32 −0.07

Number of center entries 0.35 −0.49 0.02 0.47 0.25 0.54

−0.67 0.07 0.08 0.14

Locomotor activity 0.58 −0.07 0.11 0.38 0.40 0.36 0.05 −0.05 −0.43 −0.14

AcPMe

IDH OGDC GDH GS Glutamate

♀

♂♀♂♀♂ ♀♂♀♂

RR interval −0.48 −0.24 −0.25 0.02 0.40 0.26 −0.23 0.07 0.15 0.33

SD −0.30 −0.21 −0.54 0.05 0.36 0.29 0.64 0.12 −0.14 0.05

dX −0.17 −0.10 −0.48 −0.10 0.33 0.19 0.63 −0.05 −0.15 −0.05

RMSSD −0.12

−0.74 −0.30 0.17 0.18 0.45 0.58 0.38 −0.05 0.62

SI 0.57 0.17 0.73 −0.12 −0.40 −0.10 −0.50 0.02 0.08 −0.07

Number of grooming bouts 0.19 −0.48 −0.20 −0.10 0.25

0.70 0.51 0.60 −0.46 0.03

GRAF et al.1016

BIOCHEMISTRY (MOSCOW) Vol. 91 No. 6 2026

Table 4 (cont.)

Duration of a grooming bout 0.10 0.75 −0.40 −0.29 −0.31 −0.21 −0.10 −0.57 −0.40 −0.14

Total grooming time 0.13 0.04 −0.32 −0.27 −0.62 0.33 0.13 0.00 −0.55 0.28

Freezing time −0.37 0.12 −0.52 0.26 0.12 −0.26 0.18 0.16 −0.47 −0.21

Time in center −0.30 −0.47 −0.28 0.48 −0.02 0.07 0.35 0.08 0.12 0.27

Number of rearings −0.52 0.52 −0.33 −0.55

−0.67 0.02 0.72 −0.81 0.22 −0.07

Number of center entries −0.26

0.83 −0.37 −0.73 −0.09 −0.24 0.51 −0.71 0.01 −0.02

Locomotor activity −0.02 0.64 0.17 −0.19 −0.05 −0.31 −0.18 −0.31 0.12 −0.64

MeAcP

IDH OGDC GDH GS Glutamate

♀

♂ ♀ ♂♀♂♀♂♀♂

RR interval −0.36 −0.14 −0.14 0.48 −0.60 −0.12 0.24 −0.36 0.05 −0.29

SD −0.36

−0.83 0.26 0.33 −0.29 −0.33 0.57 −0.05 0.07 −0.05

dX −0.34 −0.57 0.57 0.38 −0.56 −0.17 −0.60 −0.05 −0.06 0.00

RMSSD −0.43 −0.50 0.64 0.48 −0.50 −0.21 0.14 −0.21 0.17 −0.29

SI 0.07 0.67 −0.55 −0.29

0.79 0.14 0.07 0.07 −0.10 0.19

Number of grooming bouts 0.10 −0.04 −0.02 −0.14 0.55 −0.47 −0.63 −0.22 −0.35 0.32

Duration of a grooming bout 0.00 0.21 0.38

0.96 −0.81 −0.07 0.07 −0.39 0.17 −0.61

Total grooming time −0.02 0.31 0.36

0.81 −0.40 −0.29 −0.26 −0.19 −0.07 −0.10

Freezing time −0.50 −0.01 −0.04 −0.15 −0.07 0.38 0.17 0.19 0.13 0.15

Time in center 0.12 0.02 0.43

−0.76 −0.02 0.44 0.17 0.61 −0.52 0.51

Number of rearings 0.43 0.00 0.17 −0.65 0.31 −0.18 −0.05 −0.28 −0.38 0.12

Number of center entries 0.22 −0.53 −0.29

−0.80 −0.05 −0.17 0.22 0.05 −0.07 0.31

Locomotor activity 0.48 −0.05 −0.07 −0.57 −0.12 −0.12 0.19 −0.29 −0.21 −0.12

Note. Each cell shows the Spearman’s rank correlation coefficient(r) for the respective parameter pair. Significant correlations (p  <  0.05) are shown in red bold; trends

(p  <  0.1) are indicated in red italics. The enzymes and sexes of control or experimental groups, showing significant correlations with the physiological parameters,

are marked in red.

SEX SPECIFIC RELATIONSHIPS BETWEEN BRAIN METABOLISM AND NEUROSIGNALING 1017

BIOCHEMISTRY (MOSCOW) Vol. 91 No. 6 2026

physiological parameters, several prominent correla-

tions were identified in control females (Table  4).

In females, GS activity negatively correlated with the

HRV parameters associated with parasympathetic reg-

ulation, including SD (r  =  −0.88), dX (r  =  −0.84), and

RMSSD (r  =  −0.62). Conversely, GS activity positively

correlated with SI (r  =  0.70), indicative of enhanced

sympathetic influence. A trend toward a negative

correlation between GS activity and exploratory be-

havior (number of center entries) was also observed

(r  =  −0.67, p  =  0.07).

Thus, correlation analysis revealed clear sex-spe-

cific relationships between behavior, ECG parameters,

and cerebral cortex metabolism. In males, ME activi-

ty was strongly associated with autonomic regulation

and exploratory behavior, with higher ME activity

linked to parasympathetic predominance and coor-

dinated locomotor activity and rearing. In females,

the strongest associations with autonomic regulation

were observed for GS activity, which was related to

sympathetic predominance. Moreover, distinct cor-

relation patterns of exploratory behavior in females

compared with males suggested a greater emotional

component in female behavior.

Effect of PDC inhibitors on relationships be-

tween physiological and biochemical parameters

and their sexual dimorphism. Although PDC in-

hibitors did not significantly affect mean values of

metabolic, ECG, or behavioral parameters compared

with controls (Tables  1 and  2), they markedly reor-

ganized the relationships among the studied param-

eters, largely eliminating the pronounced sex-specific

patterns observed in control animals (Tables  3 and  4).

Specific changes depended on the inhibitor used,

likely reflecting differences in the pharmacokinetic

properties of AcPMe

and MeAcP, which influence in-

tramitochondrial inhibitor concentrations at the site

of PDC localization. Nevertheless, several common

trends were evident.

In both sexes, PDC inhibition did not funda-

mentally alter the antagonistic relationship between

parasympathetic (RR, SD, dX, RMSSD) and sympathet-

ic (SI) regulatory indices. However, the strength and

pattern of these associations changed, particularly for

individual parameters (Table  S1, Online Resource  1).

In both males and females, PDC inhibitors substan-

tially modified the correlation structure between

ECG and behavioral parameters in a sex-dependent

manner (Table  S1, Online Resource  1). For example,

following inhibitor administration, both males and

females lost characteristic strong correlations ob-

served in controls between parasympathetic activity

indices (SD, dX, RMSSD) and exploratory behavior,

including locomotion, rearing, and center entries.

The most prominent systemic effect of PDC in-

hibitors was the disappearance of strong correla-

tions that were characteristic of the control groups,

namely the correlations between ME activity and

ECG/behavioral parameters in males (Table  3) and

between GS activity and ECG parameters in females

(Table  4). This was accompanied by a reduction in

the dominance of individual biochemical parameters

within the correlation networks, resulting in diffuse

distributions of strong correlations across multiple

network components. In males, strong correlations

of behavioral and ECG parameters predominant-

ly with ME, but also with PDC and OGDC activities

were lost, while new associations involving PDC,

OGDC, GS, and IDH activities, emerged. As a result,

in males, PDC inhibitors shifted strong correlations

from the pyruvate metabolic node (Table  3) to the

glutamate metabolic node (Table  4). In females, the

opposite pattern was observed, with strong correla-

tions shifting from the glutamate metabolic node to

the pyruvate metabolic node. For example, the loss

of strong correlations between GS activity and ECG

parameters upon PDC inhibition was accompanied

by the emergence of strong correlations linking ECG

parameters with MDH and GDH activities, as well

as behavioral parameters with GS, GDH, MDH, and

ME activities. In addition, complete disappearance

of characteristic strong correlations between cortical

glutamate levels and enzymatic activities of pyru-

vate node was observed in females (Table  S1, Online

Resource  1).

Thus, in both sexes, administration of MeAcP

and AcPMe

altered correlation patterns linking be-

havioral and ECG parameters with markers of gluta-

mate metabolism (activities of OGDC, GS, GDH, IDH

and levels of glutamate), as well as with ME and

MDH activities that contribute to the PDC-dependent

generation of the TCA cycle substrate acetyl-CoA.

However, specific sets of affected correlations dif-

fered between sexes. In females, PDC inhibition

shifted the correlations to those with activities of

MDH, GDH, and ME, whereas in males – to the cor-

relations with activities of OGDC, IDH, and levels of

glutamate.

These findings suggest that PDC inhibition in-

duced a sex-specific reorganization of interactions

between metabolism and neurosignaling. In each

sex, the resulting network architecture incorporated

features that, under control conditions, were char-

acteristic of the opposite sex. Such reorganization

maintained homeostasis, as reflected by the min-

imal changes in the mean values of metabolic and

physiological parameters observed in both sexes

(Tables  1 and  2). At the same time, it reduced the

sex specificity of the correlation structure (Table S1

in the Online Resource  1) and altered sexual differ-

ences in physiological and biochemical parameters,

compared to the control groups (Tables  1  and  2).

GRAF et al.1018

BIOCHEMISTRY (MOSCOW) Vol. 91 No. 6 2026

DISCUSSION

This study demonstrates that correlation analysis

of relationships between physiological characteristics

and underlying molecular parameters (Tables  3 and  4;

Table  S1 in Online Resource  1) provides a more sen-

sitive indicator of organismal state than comparison

of parameter mean values (Tables  1 and  2). In con-

trol animals, sex differences in the coordination of

key metabolic and physiological indicators (Tables  3

and 4) reflect sex-specific homeostatic settings. As a

result, males and females can show similar mean val-

ues for most measured parameters (Tables  1 and  2).

Dysregulation of the key system of pyruvate me-

tabolism through the inhibition of PDC, which is a

central enzyme for oxidative metabolism of glucose,

the brain’s primary energy substrate, elicited a ho-

meostatic response aimed at stabilizing essential

metabolic parameters. Consequently, mean values of

biochemical and physiological parameters changed

only modestly in both sexes, with greater effects ob-

served in females, which were more responsive to

the inhibitors (Tables  1 and  2). However, there was a

substantial dynamic reorganization of the networks

linking cortical metabolism to behavioral and ECG

parameters (Tables  3 and  4; Table  S1 in Online Re-

source  1). These network-level changes largely abol-

ished the pronounced sexual dimorphism observed in

the correlation pattern of control animals (Tables  3

and  4; Table  S1 in Online Resource  1) and were ac-

companied by the emergence or attenuation of sex

differences in mean values of physiological and bio-

chemical parameters (Tables  1  and  2).

These findings suggest that the metabolic net-

works of males and females are tuned to maintain

comparable levels of key biochemical parameters, par-

ticularly glutamate content and activities of enzymes

involved in its metabolism. The tuning is sex-specific,

that is reflected in the different relationships linking

metabolism with neurosignaling to support similar

levels of key variables. During metabolic stress, com-

pensatory reorganization of these networks ensures

the maintenance of critical biochemical parameters.

Based on the indicators examined here, this homeo-

static objective appears to be achieved more effec-

tively in males than in females (Tables  1  and  2). Yet

the compensation altering the relationships between

metabolism and physiological parameters, results in

the emergence of sex differences in GDH activity,

cortical glutamate levels (Table  2), and freezing be-

havior (Table  1), while eliminating pre-existing sex

differences in locomotor activity (Table  1). Thus, the

maintenance of biochemical homeostasis during PDC

inhibition is achieved through the network-level reor-

ganization that modifies the pattern of sex differenc-

es characteristic of control animals.

Correlation analysis revealed sex-specific patterns

in both the native organization of networks linking

brain metabolic pathways with neurosignaling pro-

cesses and in the reorganization of these networks in

response to PDC inhibition. Although PDC inhibitors

were directed to the same metabolic node and affect-

ed glutamate metabolism in both males and females,

specific enzymes participating in adaptive responses

differed between sexes (Tables  3 and  4). These dif-

ferences likely reflect different baseline architecture

of interactions between metabolic and neurosignal-

ing pathways in male and female rats. In control fe-

males, neurosignaling surrogate parameters predomi-

nantly associated with GS activity, whereas in control

males, they were primarily associated with ME activ-

ity (Tables  3  and  4; Table  S1 in Online Resource  1).

The two compounds used for PDC inhibition pro-

duced qualitatively similar effects. However, quanti-

tative differences between AcPMe

and MeAcP likely

arise from their distinct pharmacokinetic properties.

For instance, AcPMe

, which serves as a precursor of

PDC inhibitors that are weaker than MeAcP (Fig.  1), may

nevertheless achieve higher intramitochondrial concen-

trations because of its greater membrane permeability.

The observed sex-dependent correlations of corti-

cal GS activity with biochemical and physiological pa-

rameters, as well as the involvement of GS-associated

networks in adaptation to PDC inhibition, are consis-

tent with findings from an independent study of sex

differences in human glioma metabolism involving

the GS-catalyzed reaction. Metabolic reprogramming

in malignant tumors is characterized by reduced PDC

activity and increased utilization of glutamine (GS re-

action product) as an alternative source for replen-

ishing TCA cycle intermediates. Notably, gliomas have

been shown to consume more glutamine in men than

in women  [37].

The close association between GS activity and

autonomic balance, reflected in ECG parameters and

observed specifically in females (Table  4), is consis-

tent with evidence demonstrating the influence of

sex hormones on GS. This enzyme plays a central

role in regulating glutamatergic neurotransmission

through astrocyte–neuron interactions involved in

synaptic glutamate turnover  [15,  17]. Estrogens have

been shown to modulate the activities of both GS  [13]

and GDH  [16], suggesting that sex-specific patterns of

association between GS activity and autonomic regu-

lation identified in our study (Tables  3  and  4) may be

mediated by estrogen-dependent regulation of gluta-

matergic tone  [14]. Previously, we also demonstrated

a relationship between changes in brain glutamate

levels and HRV parameters in female rats  [38].

In contrast, exploratory behavior and autonomic

balance in control males were more strongly associ-

ated with the activities of enzymes involved in the

SEX SPECIFIC RELATIONSHIPS BETWEEN BRAIN METABOLISM AND NEUROSIGNALING 1019

BIOCHEMISTRY (MOSCOW) Vol. 91 No. 6 2026

pyruvate metabolic node, including ME, PDC, and

OGDC (Tables  3  and  4). This pattern may reflect an

evolutionarily established emphasis on efficient ener-

gy production to support active behavioral responses.

Such a role is consistent with reports of sex-depen-

dent regulation of ME activity during ontogenesis  [19].

Moreover, adult male hearts exhibit higher expres-

sion of PDC-associated enzymes than female hearts

[39], while the expression of PDK4, a key regulator of

PDC activity in skeletal muscle, is influenced by both

estradiol levels and sex  [40]. Sexual dimorphism in

PDC regulation and related metabolic pathways may

also be linked to the location of gene encoding the

α-subunit of PDH on the X chromosome, which con-

tributes to more severe phenotypes resulting from

mutations affecting this subunit in males  [41].

Taken together, these findings indicate that, under

physiological conditions, males and females demon-

strate different interplay between metabolic and neu-

rosignaling processes, as reflected by surrogate ECG

and behavioral parameters. These baseline sex-specific

differences likely shape distinct trajectories of reorga-

nization of the metabolic–neurosignaling network un-

der metabolic stress. Although enzyme sets involved in

the response to PDC inhibition differed between sexes

(OGDC, GDH, ME, and MDH in females versus GS, OGDC,

and IDH in males), in both sexes, this reorganization

converged on pathways related to glutamate metabo-

lism. This observation is consistent with the fact that

reduced metabolic flux through the TCA cycle impairs

de  novo glutamate synthesis from glucose [5, 6, 42].

CONCLUSIONS

Correlation analysis of relationships between

cortical brain metabolism and behavioral and ECG

parameters used here as surrogate markers of neu-

rosignaling processes, revealed pronounced sexual di-

morphism. Both under physiological conditions and

upon PDC inhibition by pyruvate analogs, the archi-

tecture of associations between metabolic and neu-

rosignaling parameters differed substantially between

males and females. In control females, the primary

relationships between physiological parameters were

associated with GS, which is critical for the astrocyte–

neuron coupling necessary for the neurotransmitter

action of glutamate during glutamatergic activation.

In contrast, control males exhibited dominant asso-

ciations of physiological parameters with the activ-

ities of ME, PDC, and OGDC, the enzymes involved

in the pyruvate metabolism sustaining the amphibol-

ic role of TCA cycle in both energy production and

biosynthesis. In both sexes, PDC inhibition induced a

marked reorganization of the relationships between

behavioral/autonomic regulation and activities of en-

zymes involved in glutamate metabolism. However,

specific metabolic nodes affected by this reorganiza-

tion were sex-dependent. Established differences be-

tween males and females in the organization of phys-

iological–metabolic interaction networks highlight

the need for sex-specific, personalized approaches to

brain pathologies associated with acquired or geneti-

cally determined PDC dysfunction.

Abbreviations

AcP acetylphosphonate

AcPMe methyl acetylphosphonate

AcPMe

dimethyl acetylphosphonate

GDH glutamate dehydrogenase

GS glutamine synthetase

IDH isocitrate dehydrogenase

MDH malate dehydrogenase

ME malic enzyme (decarboxylating malate

dehydrogenase)

MeAcP methyl acetylphosphinate

OGDC 2-oxoglutarate dehydrogenase complex

PDC pyruvate dehydrogenase complex

PDH pyruvate dehydrogenase

RR intervals average cardiointerval duration

dX range of RR interval values

RMSSD root mean square of successive

differences of RR intervals

SI stress index

Supplementary information

The online version contains supplementary material

available at https://doi.org/10.1134/S0006297926600444.

Acknowledgments

The authors thank undergraduate and graduate stu-

dents of the Metabolic Engineering Group under su-

pervision of V.  I.  Bunik for their technical assistance

in sample preparation and enzymatic assays during

their association with the group.

Contributions

A.V.G. conducted experiments, analyzed results, wrote

and edited the article text; A.V.K. synthesized and con-

trolled the quality of pyruvate analogs, edited the ar-

ticle text; V.I.B. developed the concept, designed the

experiments, supervised the study, analyzed results,

wrote and edited the article.

Funding

This study was conducted under the state assign-

ments of Lomonosov Moscow State University NN

119042590056-2, 121012290046-4, and 121032300071-8.

Ethics approval and consent to participate

All animal procedures were conducted in accor-

dance with EU Directives 86/609/EEC and 2010/63/EU

GRAF et al.1020

BIOCHEMISTRY (MOSCOW) Vol. 91 No. 6 2026

and were approved by the Bioethics Committee of

the Lomonosov Moscow State University (protocol

no. 137-d; November 11, 2022).

Conflict of interest

The authors of this work declare that they have no

conflicts of interest.

REFERENCES

1. Patel, M. S., Nemeria, N. S., Furey, W., and Jordan, F. (2014) The pyruvate dehydrogenase complexes: struc-

ture-based function and regulation, J. Biol. Chem., 289, 16615-16623, https://doi.org/10.1074/jbc.R114.563148.

2. Bunik, V. I. (2024) A challenging interplay between basic research, technologies and medical education

to provide therapies based on disease mechanisms, Front. Med. (Lausanne), 11, 1464672, https://doi.org/

10.3389/fmed.2024.1464672.

3. DeBrosse, S. D., Okajima, K., Zhang, S., Nakouzi, G., Schmotzer, C. L., Lusk-Kopp, M., Frohnapfel, M. B.,

Grahame, G., and Kerr, D. S. (2012) Spectrum of neurological and survival outcomes in pyruvate dehydrogenase

complex (PDC) deficiency: lack of correlation with genotype, Mol. Genet. Metab., 107, 394-402, https://doi.org/

10.1016/j.ymgme.2012.09.001.

4. Bubber, P., Haroutunian, V., Fisch, G., Blass, J. P., and Gibson, G. E. (2005) Mitochondrial abnormalities in

Alzheimer brain: mechanistic implications, Ann. Neurol., 57, 695-703, https://doi.org/10.1002/ana.20474.

5. Santos, S. S., Gibson, G. E., Cooper, A. J. L., Denton, T. T., Thompson, C. M., Bunik, V. I., Alves, P. M., and

Sonnewald, U. (2006) Inhibitors of the α-ketoglutarate dehydrogenase complex alter [1-13C]glucose and [U-13C]

glutamate metabolism in cerebellar granule neurons, J. Neurosci. Res., 83, 450-458, https://doi.org/10.1002/

jnr.20749.

6. Artiukhov, A.V., Aleshin, V.A., Karlina, I.S., Kazantsev, A.V., Sibiryakina, D.A., Ksenofontov, A.L., Lukashev,N.V.,

Graf, A. V., and Bunik, V. I. (2022) Phosphonate inhibitors of pyruvate dehydrogenase perturb homeosta-

sis of amino acids and protein succinylation in the brain, Int. J. Mol. Sci., 23, 13186, https://doi.org/10.3390/

ijms232113186.

7. Gul-Hinc, S., Jankowska-Kulawy, A., and Szutowicz, A. (2026) Cholinergic phenotypes of acetyl-CoA with

ATP-citrate lyase link, Int. J. Mol. Sci., 27, 782, https://doi.org/10.3390/ijms27020782.

8. Nemeria, N. S., Korotchkina, L. G., Chakraborty, S., Patel, M. S., and Jordan, F. (2006) Acetylphosphinate is the

most potent mechanism-based substrate-like inhibitor of both the human and Escherichia coli pyruvate de-

hydrogenase components of the pyruvate dehydrogenase complex, Bioorg. Chem., 34, 362-379, https://doi.org/

10.1016/j.bioorg.2006.09.001.

9. Bunik, V.I., Artiukhov,A., Kazantsev,A., Goncalves, R., Daloso,D., Oppermann,H., Kulakovskaya, E., Lukashev, N.,

Fernie, A., Brand, M., and Gaunitz, F. (2015) Specific inhibition by synthetic analogs of pyruvate reveals that the

pyruvate dehydrogenase reaction is essential for metabolism and viability of glioblastoma cells, Oncotarget, 6,

40036-40052, https://doi.org/10.18632/oncotarget.5486.

10. Aleshin, V. A., Sibiryakina, D. A., Kazantsev, A. V., Graf, A. V., and Bunik, V. I. (2023) Acylation of the rat brain

proteins is affected by the inhibition of pyruvate dehydrogenase in vivo, Biochemistry (Moscow), 88, 105-118,

https://doi.org/10.1134/S0006297923010091.

11. Duarte, J. M. N., Do, K. Q., and Gruetter, R. (2014) Longitudinal neurochemical modifications in the aging

mouse brain measured in vivo by 1H magnetic resonance spectroscopy, Neurobiol. Aging, 35, 1660-1668, https://

doi.org/10.1016/j.neurobiolaging.2014.01.135.

12. Mkrtchyan, G. V., Graf, A., Trofimova, L., Ksenofontov, A., Baratova, L., and Bunik, V. (2018) Positive correla-

tion between rat brain glutamate concentrations and mitochondrial 2-oxoglutarate dehydrogenase activity,

Anal. Biochem., 552, 100-109, https://doi.org/10.1016/j.ab.2018.01.003.

13. Haghighat, N. (2005) Estrogen (17β-estradiol) enhances glutamine synthetase activity in C6-glioma cells, Neuro-

chem. Res., 30, 661-667, https://doi.org/10.1007/s11064-005-2754-5.

14. Goyette, M. J., Murray, S. L., Saldanha, C. J., and Holton, K. (2023) Sex hormones, neurosteroids, and glutama-

tergic neurotransmission: a review of the literature, Neuroendocrinology, 113, 905-914, https://doi.org/10.1159/

000531148.

15. Al-Suwailem, E., Abdi, S., and El-Ansary,A. (2018) Sex differences in the glutamate signaling pathway in juvenile

rats, J. Neurosci. Res., 96, 459-466, https://doi.org/10.1002/jnr.24144.

16. Bunik, V., Aleshin, V., Artiukhov, A., Kaehne, T., and Karlina, I. (2016) Multiple forms of glutamate dehydroge-

nase in animals: structural determinants and physiological implications, Biology (Basel), 5, 53, https://doi.org/

10.3390/biology5040053.

17. Chiang, V. S.-C., and Park, J. H. (2020) Glutamate in male and female sexual behavior: receptors, transporters,

and steroid independence, Front. Behav. Neurosci., 14, 589882, https://doi.org/10.3389/fnbeh.2020.589882.

SEX SPECIFIC RELATIONSHIPS BETWEEN BRAIN METABOLISM AND NEUROSIGNALING 1021

BIOCHEMISTRY (MOSCOW) Vol. 91 No. 6 2026

18. Klinge, C. M. (2020) Estrogenic control of mitochondrial function, Redox Biol., 31, 101435, https://doi.org/

10.1016/j.redox.2020.101435.

19. Sasse, D., and Hoffmann, H. (1982) Postnatal differentiation of sex-specific distribution patterns of G6Pase, G6P-

DH and ME in the rat liver, Histochemistry, 75, 31-42, https://doi.org/10.1007/BF00492531.

20. Viel, C., Brandtner, A., Weißhaar, A., Lehto, A., Fuchs, M., and Klein, J. (2021) Effects of magnesium orotate,

benfotiamine and a combination of vitamins on mitochondrial and cholinergic function in the TgF344-AD rat

model of Alzheimer’s disease, Pharmaceuticals, 14, 1218, https://doi.org/10.3390/ph14121218.

21. Saed, C. T., Tabatabaei Dakhili, S. A., and Ussher, J. R. (2021) Pyruvate dehydrogenase as a therapeutic target for

nonalcoholic fatty liver disease, ACS Pharmacol. Transl. Sci., 4, 582-588, https://doi.org/10.1021/acsptsci.0c00208.

22. Lewis, A. J., Neubauer, S., Tyler, D. J., and Rider, O. J. (2016) Pyruvate dehydrogenase as a therapeutic tar-

get for obesity cardiomyopathy, Expert Opin. Ther. Targets, 20, 755-766, https://doi.org/10.1517/14728222.

2016.1126248.

23. Artiukhov, A. V., Balashova, N. V., Ikonnikova, A. Y., Kazantsev, A. V., Semashko, T. A., Ksenofontov, A. L.,

Graf, A. V., and Bunik, V. I. (2024) Pharmacological doses of thiamine benefit patients with the Charcot–

Marie–Tooth neuropathy by changing thiamine diphosphate levels and affecting regulation of thiamine-depen-

dent enzymes, Biochemistry (Moscow), 89, 1161-1182, https://doi.org/10.1134/S0006297924070023.

24. Hadano, T., Ishibashi, S., Matsuoka, H., Koyama, K., Kuwabara, S., and Yamamoto, T. (2025) Association of sex

difference with thiamine responsive disorders in internal medicine: a pilot case-control study, J. Clin. Med. Res.,

17, 490-498, https://doi.org/10.14740/jocmr6054.

25. Artiukhov, A. V. (2022) Development of Methods for Targeted Regulation of 2-oxo acid Dehydrogenases in

Mammals and Features of Such Regulation in Cells with Different Metabolic Types, PhD Thesis in Chemistry,

Lomonosov Moscow State University, Moscow.

26. Artiukhov, A. V., Graf, A. V., Kazantsev, A. V., Boyko, A. I., Aleshin, V. A., Ksenofontov, A. L., and Bunik, V. I.

(2022) Increasing inhibition of the rat brain 2-oxoglutarate dehydrogenase decreases glutathione redox

state, elevating anxiety and perturbing stress adaptation, Pharmaceuticals, 15, 182, https://doi.org/10.3390/

ph15020182.

27. Artiukhov, A. V., Kazantsev, A. V., Boyko, A. I., Aleshin, V. A., Ksenofontov, A. L., and Bunik, V. I. (2021) Selective

inhibition of 2-oxoglutarate and 2-oxoadipate dehydrogenases by the phosphonate analogs of their 2-oxo acid

substrates, Front. Chem., 8, 18907, https://doi.org/10.3389/fchem.2020.618907.

28. McBride, K. (2003) Does intranasal application of zinc sulfate produce anosmia in the mouse? An olfactometric

and anatomical study, Chem. Senses, 28, 659-670, https://doi.org/10.1093/chemse/bjg058.

29. Slotnick, B., Glover, P., and Bodyak, N. (2000) Does intranasal application of zinc sulfate produce anosmia in

the rat? Behav. Neurosci., 114, 814-829, https://doi.org/10.1037/0735-7044.114.4.814.

30. Graf, A. V., Maslova, M. V., Artiukhov, A. V., Ksenofontov, A. L., Aleshin, V. A., and Bunik, V. I. (2022) Acute

prenatal hypoxia in rats affects physiology and brain metabolism in the offspring, dependent on sex and ges-

tational age, Int. J. Mol. Sci., 23, 2579, https://doi.org/10.3390/ijms23052579.

31. Araújo, W. L., Trofimova, L., Mkrtchyan, G., Steinhauser, D., Krall, L., Graf, A., Fernie, A. R., and Bunik, V. I.

(2013) On the role of the mitochondrial 2-oxoglutarate dehydrogenase complex in amino acid metabolism,

Amino Acids, 44, 683-700, https://doi.org/10.1007/s00726-012-1392-x.

32. Artiukhov, A. V., Grabarska, A., Gumbarewicz, E., Kaca, W., and Bunik, V. I. (2020) Synthetic analogues of 2-oxo

acids discriminate metabolic contribution of the 2-oxoglutarate and 2-oxoadipate dehydrogenases in mammalian

cells and tissues, Sci. Rep., 10, 1886, https://doi.org/10.1038/s41598-020-58803-z.

33. Bunik, V., Artiukhov, A., Aleshin, V., and Mkrtchyan, G. (2010) Behavioral impact of the regulation of the

brain 2-oxoglutarate dehydrogenase complex by synthetic phosphonate analog of 2-oxoglutarate: implications

into the role of the complex in neurodegenerative diseases, Parkinsons Dis., 2010, 749061, https://doi.org/

10.4061/2010/749061.

34. Borovskaya, R., Krivoguz, D., Chernyi, S., Kozhurin, E., and Goryachkin, S. (2022) Surface water salinity eval-

uation and identification for using remote sensing data and machine learning approach, J. Mar. Sci. Eng., 10,

257, https://doi.org/10.3390/jmse10020257.

35. Sullivan, G. M., and Feinn, R. (2012) Using effect size – or why the p value is not enough, J. Grad. Med. Educ.,

4, 279-282, https://doi.org/10.4300/JGME-D-12-00156.1.

36. Estanislau, C., Díaz-Morán, S., Cañete, T., Blázquez, G., Tobeña, A., Fernández-Teruel, A., and Escorihuela, R. M.

(2013) Context-dependent differences in grooming behavior among the NIH heterogeneous stock and the Roman

high- and low-avoidance rats, Neurosci. Res., 77, 187-201, https://doi.org/10.1016/j.neures.2013.09.007.

37. Sponagel, J., Jones, J. K., Frankfater, C., Zhang, S., Tung, O., Cho, K., Tinkum, K. L., Gass, H., Nunez, E.,

Spitz, D. R., Chinnaiyan, P., Schaefer, J., Patti, G. J., Graham, M. S., Mauguen, A., Grkovski, M., Dunphy, M. P.,

GRAF et al.1022

BIOCHEMISTRY (MOSCOW) Vol. 91 No. 6 2026

Krebs, S., Luo, J., Rubin, J. B., and Ippolito, J. E. (2022) Sex differences in brain tumor glutamine metab-

olism reveal sex-specific vulnerabilities to treatment, Med, 3, 792-811.e12, https://doi.org/10.1016/j.medj.

2022.08.005.

38. Graf, A., Trofimova, L., Loshinskaja, A., Mkrtchyan, G., Strokina, A., Lovat, M., Tylicky, A., Strumilo, S.,

Bettendorff, L., and Bunik, V. I. (2013) Up-regulation of 2-oxoglutarate dehydrogenase as a stress response,

Int. J. Biochem. Cell Biol., 45, 175-189, https://doi.org/10.1016/j.biocel.2012.07.002.

39. Vijay, V., Han, T., Moland, C. L., Kwekel, J. C., Fuscoe, J. C., and Desai, V. G. (2015) Sexual dimorphism in

the expression of mitochondria-related genes in rat heart at different ages, PLoS One, 10, e0117047, https://

doi.org/10.1371/journal.pone.0117047.

40. Christianto, A., Baba, T., Takahashi, F., Inui, K., Inoue, M., Suyama, M., Ono, Y., Ohkawa, Y., and Morohashi, K.

(2021) Sex differences in metabolic pathways are regulated by Pfkfb3 and Pdk4 expression in rodent muscle,

Commun. Biol., 4, 1264, https://doi.org/10.1038/s42003-021-02790-y.

41. Dahl, H. H. (1995) Pyruvate dehydrogenase E1 alpha deficiency: males and females differ yet again, Am. J. Hum.

Genet., 56, 553-557.

42. Bunik, V. I., Artiukhov, A. V., Kazantsev, A. V., Aleshin, V. A., Boyko, A. I., Ksenofontov, A. L., and Lukashev, N. V.

(2022) Administration of phosphonate inhibitors of dehydrogenases of 2-oxoglutarate and 2-oxoadipate to

rats elicits target-specific metabolic and physiological responses, Front. Chem., 10, 968903, https://doi.org/

10.3389/fchem.2022.968903.

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.