Article

ISSN 0006-2979, Biochemistry (Moscow), 2026, Vol. 91, No. 6, pp. 939-955 © Pleiades Publishing, Ltd., 2026.

939

Impact of Genetically Induced Metabolic Syndrome

on the Outcomes of Photothrombotic Stroke

in Leptin-Deficient and Leptin Receptor-Deficient Mice

Alexey D. Bocharnikov

1,2

, Polina A. Abramicheva

, Elmira I. Yakupova

Olga A. Averina

, Olga O. Grigoryeva

, Anastasia V. Priymak

Svetlana A. Garmash

, Oleg A. Permyakov

, Irina B. Pevzner

1,4

Anna A. Brezgunova

, Nadezda V. Andrianova

, Petr V. Sergiev

and Egor Y. Plotnikov

1,4,a

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

119992 Moscow, Russia

Faculty of Bioengineering and Bioinformatics, Lomonosov Moscow State University,

119992 Moscow, Russia

Institute of Biomedical Problems of the Russian Academy of Sciences, 123007 Moscow, Russia

Kulakov National Medical Research Center of Obstetrics, Gynecology and Perinatology,

117997 Moscow, Russia

e-mail: plotnikov@belozersky.msu.ru

Received January 18, 2026

Revised April 30, 2026

Accepted May 30, 2026

Abstract— Diabetes and obesity are associated with poorer outcomes after ischemic stroke; however, it re-

mains unclear whether this results from increased neuronal susceptibility to injury or from vascular dys-

function induced by metabolic syndrome. To minimize the contribution of vascular factors, we used a

model of photoinduced thrombosis (PT) in cortical vessels, which generates lesions of reproducible size

and is less dependent on collateral blood flow. PT was induced in wild-type (WT) mice, as well as in ob/ob

(leptin-deficient) and db/db (leptin receptor-deficient) mice. Magnetic resonance imaging (MRI) revealed that

PT produced comparable infarct volumes in all mouse groups. Several genes associated with inflammation

and activation of microglia and macroglia in the peri-infarct area (Cst7, Ccl3, Tlr2, Gfap) exhibited similar

expression patterns across all three mouse strains, while transcriptional response to cerebral ischemia of

Tnfa, Cxcl9, Il6, Cox2, Mmp3, and Bdnf genes depended on the genotype. Overall, despite individual dif-

ferences in the expression profiles of certain genes, disruption of leptin signaling (whether due to leptin

deficiency or leptin receptor deficiency) caused no genotype-specific exacerbation of stroke-induced injury.

Assessment of post-stroke neurological deficits revealed substantial differences in absolute scores between

WT and ob/ob or db/db mice, attributable to baseline disparities in body weight and motor activity. In db/db

mice, normalization of post-stroke neurological status scores to pre-injury values revealed a more pro-

nounced relative functional decline compared to ob/ob mice, suggesting impairments in early compensatory

mechanisms and an important role of leptin signaling in neuroplasticity rather than in the extent of acute

neuronal damage. Thus, under conditions that minimize vascular complications, neither leptin deficiency

nor leptin receptor deficiency exacerbated acute ischemic brain damage or neuroinflammation.

DOI: 10.1134/S0006297926600134

Keywords: db/db mice, ob/ob mice, stroke, photothrombosis, inflammation, leptin, leptin receptors, neurological

deficit

* To whom correspondence should be addressed.

BOCHARNIKOV et al.940

BIOCHEMISTRY (MOSCOW) Vol. 91 No. 6 2026

INTRODUCTION

The risk of ischemic stroke depends on many

factors, including obesity and diabetes mellitus. Both

conditions are well-established risk factors for stroke

in humans  [1] and are associated with stroke sever-

ity in both animal models  [2] and patients  [3]. Since

diabetes and ischemic stroke are strongly interrelated

and often co-occur in clinical practice, research on the

treatment of ischemic stroke in patients with diabe-

tes mellitus is of particular importance. Such studies

commonly employ animal models in which diabetes

is induced through dietary interventions or genetic

modifications. However, in most preclinical stroke

models, particularly middle cerebral artery occlusion

(MCAO), the outcome severity in animals with meta-

bolic syndrome is determined not only by neuronal

damage but also by alterations in the cerebral vascu-

lature and blood–brain barrier (BBB) integrity, which

complicates the assessment of the effect of diabetes

on the ischemic tolerance of nervous tissue [4, 5].

Among animal models of diabetes, ob/ob and

db/db mice are most extensively studied. Both strains

exhibit impaired leptin signaling, but through fun-

damentally different mechanisms: ob/ob mice do not

express leptin, while db/db mice lack the main leptin

receptor (LepR), resulting in phenotypic differences

between these strains  [6]. Despite a complete absence

of leptin production in ob/ob mice, their phenotype is

less severe than that of db/db mice, which are defi-

cient only in the long isoform of the leptin receptor

(LepRb), even though at least six splicing isoforms of

the LepR gene have been described  [7]. Furthermore,

db/db mice also show a compensatory increase in

leptin levels, which may induce secondary effects be-

yond those caused by impaired receptor signaling  [8].

Importantly, in this study, we utilized a mouse model

generated by our group that carries a different LepR

mutation, namely, a frameshift in the early part of

the gene shared by all receptor isoforms, resulting in

the inactivation of the entire LepR family rather than

the long isoform only. Consequently, this model en-

ables assessment of the consequences of a complete

shutdown of LepR signaling. Thus, although both

ob/ob and db/db mice have been designed to simulate

the same pathology, the differences in the underlying

genetic defects may lead to different responses to cer-

tain factors  [9]. This consideration is particularly rel-

evant to the present study, since the LepR mutation

used here affects a region common to all receptor

isoforms, potentially resulting in a phenotype that

differs substantially from that of conventional db/db

strains.

The molecular basis underlying the pathogen-

esis of ischemic damage in animals with metabolic

syndrome remains a subject of particular interest.

To investigate stroke in transgenic db/db and ob/ob

mice, previous studies have primarily employed mid-

dle cerebral artery occlusion (MCAO) [2, 10, 11]. This

model is regarded as the gold standard for experi-

mental ischemic stroke because it reproduces the

blockage of a large vessel and closely resembles

thrombotic stroke in clinical practice settings  [12,  13].

Stroke caused by MCAO is characterized by the for-

mation of a pronounced ischemic penumbra; how-

ever, in animals with metabolic syndrome, diabetes,

or obesity, vascular dysfunction and disruption of

the BBB can accelerate the progression of damage in

this penumbra, making it difficult to interpret wheth-

er observed outcomes result from altered ischemic

tolerance of nervous tissue cells or from vascular

pathology. Unlike MCAO, photothrombosis  (PT) does

not produce a pronounced penumbra, but offers a

minimally invasive model with a high reproducibili-

ty of the size and location of the ischemic focus  [14].

Consequently, PT enables a direct assessment of the

effects of leptin signaling on the nervous tissue sus-

ceptibility to ischemic injury, largely independent of

the vascular component.

Although metabolic syndrome, obesity, and diabe-

tes are consistently associated with poorer outcomes

after ischemic stroke, it remains unclear whether

these effects are driven primarily by increased sus-

ceptibility of nervous tissue to ischemia or by sys-

temic vascular impairments accompanying these

pathologies. The aim of the present study was to de-

termine whether the deficiency of leptin or its recep-

tor affects the size of ischemic lesion, as well as early

functional and molecular responses of the brain in

standardized focal stroke. To assess the tolerance of

nervous tissue to ischemic damage independently of

alterations in cerebral blood flow and BBB integrity,

we employed the model of cortical PT. Comparative

analysis was performed between wild-type (WT) mice

and ob/ob and db/db knockout strains modeling dif-

ferent disruptions in leptin signaling. The db/db mice

used in this study differ from the conventional db/db

strains, as they carry an LepR mutation causing an

early frameshift and premature translation failure,

likely resulting in a complete loss of leptin signaling.

This allowed us to assess whether the acute resis-

tance of nervous tissue to ischemia is altered by the

deficiency of leptin or its receptor, and to identify po-

tential genotype-dependent differences in post-stroke

recovery.

MATERIALS AND METHODS

Animals. The study used male WT mice and

mice carrying targeted db/db (B6-Lepr

d10

) and ob/ob

(B6-Lep

d10

) mutations on the C57BL/6 background.

STROKE IN LEPTIN-DEFICIENT MICE 941

BIOCHEMISTRY (MOSCOW) Vol. 91 No. 6 2026

Fig.  1. Genotyping of db/db and ob/ob mice. Representative Sanger sequencing chromatograms for db/db (B6-LepR

d10

)  (a)

and ob/ob (B6-Lep

d10

)  (c) mouse strains and corresponding melting curves obtained by HRM analysis  (b and  d, respectively).

From top to bottom in each chromatogram: wt, heterozygous mutant (hetero), and homozygous (mut) genotypes.

The ob/ob strain was generated by introducing a

10-base pair deletion into the Lep gene, resulting in

complete loss of leptin production. The db/db strain

was generated by introducing a 10-base pair deletion

into the LepR gene, causing an early frameshift in

a region common to all known LepR isoforms. This

mutation is predicted to inactivate all LepR isoforms

and, consequently, abolish leptin signaling. Homozy-

gous ob/ob and db/db mice were obtained by crossing

corresponding heterozygous animals; WT littermates

derived from heterozygous breeding pairs were used

as controls. All experiments were performed on ani-

mals at 6 months of age.

Genotyping of ob/ob, db/db, and WT mice was

performed using Sanger sequencing; high-resolution

melting (HRM) assay after PCR amplification was used

for routine genotype screening. Sanger sequencing re-

action was performed using a BigDye Terminator v3.1

Cycle Sequencing Kit (Thermo Fisher, USA) according

to the manufacturer’s protocol using a SeqStudio se-

quencer (Thermo Fisher, USA) with a SeqStudio  v2

cartridge and a BDx Short Read module (injection at

2  kV, 10  s) using the following cycling regime: 96°C

for 1  min, 25  cycles of 96°C for 10  s, 55°C for 5  s,

and 60°C for 4  min. Primers for PCR amplification

and sequencing are listed in Table  S1 in the Online

Resource  1. Sequencing products were purified with

an iX-Pure Dye Terminator Cleanup Kit (Nimagen,

Netherlands) according to the manufacturer’s instruc-

tions. Chromatograms (.ab1 files) were analyzed us-

ing SnapGene v8.0.3 software (GSL Biotech, 2025) and

Benchling (Benchling Inc., 2025). Sanger sequencing

BOCHARNIKOV et al.942

BIOCHEMISTRY (MOSCOW) Vol. 91 No. 6 2026

Fig. 2. Scheme of the experiment (created with BioRender.com; license: h6sxe55).

confirmed the presence of 10-bp deletions in LepR in

db/db mice and in Lep in ob/ob mice (Fig.  1, a  and  b).

Sequence alignment confirmed that the deletion in

the LepR gene causes an early frameshift leading to

the functional loss of all receptor isoforms (Fig.  1a).

We used melting curves obtained by HRM analy-

sis of PCR products to distinguish between the three

genotypes (wt, mut, and hetero) of db/db and ob/ob

mice. Homozygous animals (mut) were used in the

experiments, while WT mice (wt) served as controls.

Heterozygous mice (hetero) were used to establish

breeding pairs. Genomic DNA was rapidly extracted

using alkaline lysis-based protocol  [15]. Quantitative

PCR (qPCR) was performed using a Gentier 96E am-

plifier (Tianlong, China). Fragments of the Lep and

LepR genes were amplified using SYBR® Green  I in-

tercalating dye and 5×  qPCRmix-HS (Evrogen, Russia)

(Table  S2 in the Online Resource  1). Optimal anneal-

ing and elongation temperatures were determined

using gradient PCR: the annealing temperature for

Lep and LepR primers was 54°C and 58°C, respec-

tively; the elongation temperature was 72°C for both

primer sets. Representative melting curves corre-

sponding to different mouse genotypes are shown in

Fig.  1,  b  and  d.

Photothrombotic stroke model. The animals

were housed at a conventional animal facility in stan-

dard M-6 cages under a 12  h light/12  h dark cycle at

22  ±  2°C with ad  libitum access to food and water. All

animal procedures were approved by the Bioethics

Commission of the Belozersky Moscow State Univer-

sity Research Institute (protocol no.  015-8/10/2024;

approved June 9, 2025). The animals were assigned

into three groups according to genotype (WT, ob/ob,

and db/db). The number of animals used in each

experiment is shown in Table  S3 in the Online Re-

source  1. All animals underwent neurological evalua-

tion before and after PT. To  analyze gene expression

in the injured brain tissue, samples obtained from

intact (Int) and stroke-affected  (PT) mice were com-

pared for all three genotypes. The experimental de-

sign is shown in Fig.  2.

Standard PT model [16] with minor modifications

was used to induce focal cerebral ischemia. Mice

were anesthetized with isoflurane (5% for induction

and 3% for maintenance) delivered via a SomnoSuite

inhalation anesthesia system (Kent Scientific, USA).

Bengal Rose dye (1%  solution) was injected into the

jugular vein at a dose of 40  mg/kg. The skull was ex-

posed through a midline scalp incision and cleared of

periosteal tissue. The animal was secured in a stereo-

taxic frame (RWD Life Sciences, China). Seven min-

utes after dye injection, selected region of the right

hemisphere was irradiated with a 50-mW green laser

(wavelength, 532  nm) for 15  min at the following coor-

dinates: 0.5  mm anterior to the bregma along the A-P

axis and 2.7  mm lateral along the M-L axis. The laser

beam was focused on the selected area of the right

hemisphere, producing illumination spot 1.5-2  mm in

diameter. No significant differences in linear brain

dimensions, brain weight, and craniometric param-

eters were observed among the mouse strains used

in the study (see Fig.  S1 in the Online Resource  1).

During the procedure, animal’s body temperature was

STROKE IN LEPTIN-DEFICIENT MICE 943

BIOCHEMISTRY (MOSCOW) Vol. 91 No. 6 2026

maintained at37°C. Following irradiation, the incision

was closed with a polyglycolide-based suture (diam-

eter, 4-0; Medical Technology Laboratory, Russia),

and the wound site was treated with the antiseptic

Fukortsin. Mice were then placed under an infrared

heating lamp until full recovery from anesthesia,

with body temperature maintained at 37°C.

Neurological deficit assessment. Three behav-

ioral tests were performed to assess the severity of

neurological deficits following damage to the senso-

rimotor cortex. The setup for the beam walking test

included a dark box (house) connected to a double

beam tapered toward the house. The beam consist-

ed of a lower, wider section and a narrower upper

section. Animals were trained for three consecutive

days before testing. During training, mice learned to

traverse the beam toward the box, while the start-

ing distance was gradually increased. For testing,

each mouse was placed in the dark box filled with

sawdust for 5  min. Next, the animal was placed at

the distal end of the beam and allowed to return to

the box independently. Neurological deficits were as-

sessed by scoring paw placement: 2  points for normal

limb placement on the upper track, 1  point for partial

slipping, and 0 points for complete paw placement on

the lower track. Partial slipping was defined as hold-

ing the paw on the lateral surface of the upper track

or placing the foot at an angle on the edge of the

upper track. For each animal, the average score from

three passes was calculated. Testing was performed

before PT and on days  3 and 6 after PT.

Neurological deficit was estimated using the fol-

lowing formula (1):

Neurological deficit =

total slips + 0.5 × partial slips

total number of steps

. (1)

Motor coordination and limb strength were eval-

uated using a rotarod device (Ugo Basile, Italy) con-

sisting of a rotating drum on which a mouse should

remain during the test, while avoiding falls. The

maximum test duration was 300  s, with the rotation

speed gradually increasing from 4 to 40  rpm over

240  s. Before the experiment, mice were trained for

three consecutive days; on day  4, their baseline per-

formance was recorded (intact). The effect of stroke

on rotarod performance was evaluated 3 and 6  days

after PT. Since the mutant mice were severely obese,

we used an integral index to minimize the effect of

body weight on test results. This index was calculated

as body weight  ×  retention time  =  impulse, providing

a combined measure of endurance and motor control.

For the grid walking (foot fault) test, we used

a metal grid with a mesh size of 19.7×19.5  mm, en-

closed by borders on all four sides. During testing,

mice were allowed to move freely on the grid for

5  min. The number of paws slips into the grid cells

was recorded. Testing was performed before PT (in-

tact) and on days 3 and 7 after PT to avoid potential

confounding effects of fatigue from other neurologi-

cal tests. All sessions were video-recorded and ana-

lyzed by reviewers blinded to the study conditions.

Analysis of gene expression after PT-induced

stroke. Brain tissue samples were collected to study

gene expression on day  10 after PT. The animals were

euthanized by decapitation, and cortical tissue from

the right hemisphere was dissected from the ischemic

focal region corresponding to the stereotactic coordi-

nates AP 0.5, ML 2.7. Tissue samples were immedi-

ately placed in IntactRNA™ stabilizing reagent (Evro-

gen, Russia) to prevent RNA degradation and stored

at −80°C. Total RNA was isolated and treated with

duplex-specific nuclease using commercial RNA-Solo™

kit (Evrogen, Russia). The quantity and purity of iso-

lated RNA were assessed spectrophotometrically with

an Eppendorf Biophotometer plus (Eppendorf, USA)

by measuring absorption ratios at 260/280  nm and

260/230  nm. Reverse transcription was performed

using an MMLV RT kit (Evrogen, Russia). qPCR was

carried out with a Gentier 96E PCR system (Tianlong,

China) using the intercalating dye SYBR® Green  I

for detecting double-stranded amplification products.

Amplification was performed using 5×  qPCRmix-HS

optimized reaction mixture (Evrogen, Russia). Primers

were designed with the cloud-based Benchling plat-

form (Benchling Inc., 2025, USA) and Primer-BLAST

(NCBI, USA). Primer efficiency for each target gene

was calculated using a standard curve generated for

fivefold serial dilutions of cDNA. Relative mRNA ex-

pression levels were calculated using the formula  (2):

(Rplp0)

(gene of interest)

, (2)

where E is the primer efficiency and Ct is the thresh-

old cycle. Expression was normalized to the reference

gene Rplp0 (60S acidic ribosomal protein). All oligo-

nucleotide primers were synthesized by DNA Synthe-

sis Company (Moscow, Russia). Primer sequences are

shown in Table  S4 in the Online Resource  1.

Assessment of lesion volume by MRI. Brain

lesion volume was assessed using a high-field (7.0

Tesla) BioSpec 70/30 MRI system (gradient strength,

105  mT/min) (Bruker, Germany). Animals were anes-

thetized with 3% isoflurane and positioned in a

temperature-controlled holder device. MRI examina-

tion was performed using a standard mouse brain

imaging protocol, including rapid acquisition of

T2-weighted images with relaxation enhancement

(RARE) spin-echo sequence with the following param-

eters: repetition time (TR), 6000  ms; echo time (TE),

63.9  ms; slice thickness 0.4  mm with an interslice

distance of 0.1  mm; matrix size, 256×384; resolution,

0.164×0.164  mm

/pixel. The total imaging time for

BOCHARNIKOV et al.944

BIOCHEMISTRY (MOSCOW) Vol. 91 No. 6 2026

each animal was ~25  min. Lesion volume was deter-

mined from serial MRI scans on day 7 after PT us-

ing ImageJ software (NIH, USA). Lesion areas were

measured in ImageJ after calibration of image scale

according to the MRI acquisition parameters.; the le-

sion area  (S) in each slice was manually delineated

and calculated. The lesion volume  (V) was estimated

using the formula  (3):

V=ΣS × H, (3)

where H is the increment between the slices.

Western blotting. Brain samples were collected

on day 10 after PT and homogenized in ice-cold phos-

phate-buffered saline (PBS) containing 1  mM phenyl-

methylsulfonyl fluoride (Thermo Fisher Scientific,

USA). Total protein concentration was determined

using a bicinchoninic acid (BCA) protein assay (Sigma

Aldrich, USA). Protein samples (10  μg of total protein

per lane) were fractionated in 10% polyacrylamide

gel (TGX FastCast Acrylamide Kit, Bio-Rad, USA) un-

der denaturing conditions and then transferred to a

PVDF membrane (Amersham Pharmacia Biotech, UK)

using a semi-dry Trans Blot Turbo Transfer System

(Bio-Rad). Membranes were blocked for 45  min at

room temperature with 5%  fat-free milk (Serva, Ger-

many) in Tris-buffered saline (TBS) containing 0.1%

Tween-20 (Helicon, Russia), then washed three times,

followed by overnight incubation at 4°C with pri-

mary rabbit antibodies against ERK1/2 (p44/p42)

(dilution, 1  :  1000; #9102, Cell Signaling, USA), phos-

pho-p44/42 MAPK (Erk1/2)(Thr202/Tyr204) (dilution,

1  :  1000; #9101, Cell Signaling), Akt (dilution, 1  :  1000;

#9272, Cell Signaling), and phospho-Akt (Ser473) (di-

lution, 1  :  1000; #4060, Cell Signaling) diluted in TBS

containing 0.2%  BSA. The membranes were then

washed and incubated for 60  min at 37°C with sec-

ondary goat anti-rabbit horseradish peroxidase-con-

jugated antibodies (dilution, 1  :  5000; IMTEK, Russia).

After washing, the membranes were developed with

Advansta ECL Bright chemiluminescent solution

(Advansta, USA) and scanned using a ChemiDoc MP

Imaging System (Bio-Rad). Luminescence signal in-

tensities for target proteins were quantified using

ImageLab 6.0.1 software (Bio-Rad) and normalized

to the total protein load estimated using TGX® gel

densitometry in accordance with the manufacturer’s

recommendations.

Glucose tolerance test. Glucose levels were

measured in all mouse strains in blood collected from

the tail vein after fasting and after glucose admin-

istration. Animals received an intraperitoneal injec-

tion of 40% glucose solution at a dose of 1  g/kg body

weight. Blood glucose concentration was determined

at 15, 30, 60, and 120  min after injection using a Sat-

tellit Express glucose meter (Company ELTA, Russia).

Overall glucose tolerance was evaluated from area

under the curve (AUC) using GraphPad Prism  8

(GraphPad Software, USA).

Statistical analysis was performed using Graph-

Pad Prism  8 (GraphPad Software). The Shapiro–Wilk

test was used to assess the normality of data distri-

bution. For normally distributed data, differences be-

tween the groups at three time points were estimated

with a two-way ANOVA followed by the Tukey’s test

for multiple comparisons. Small-size samples were

assessed with the Kruskal–Wallis test followed by the

Dunn test. In cases when the homogeneity of variance

was violated, the Holm–Sidak test with the Welch cor-

rection was applied. In all comparisons, outliers were

identified and excluded using the ROUT test (Q  =  1%).

All data are presented as mean ± standard error of

the mean (SEM). Detailed statistic data are presented

in the Table  S5.

RESULTS

Manifestations of metabolic syndrome in ge-

netically altered mice. Both ob/ob and db/db mice

exhibited a pronounced metabolic phenotype char-

acteristic of leptin signaling deficiency, including

obesity and impaired carbohydrate metabolism. The

body weight of ob/ob and db/db mice was more than

twofold higher than that of WT animals (Figs.  S2

and  S3 in the Online Resource  1). Mice with the

metabolic phenotype showed a marked increase in

the weight of epididymal and perirenal adipose tis-

sue (Fig.  S2, b  and  c). In contrast, brain size and

weight, as well as skull linear dimensions did not

differ significantly between ob/ob, db/db, and WT

mice (Fig.  S1 in the Online Resource  1). Therefore,

the elevated body weight of animals with metabol-

ic syndrome is largely attributable to fat accumula-

tion. Glucose tolerance test revealed that in WT mice,

blood glucose levels increased significantly 15  min

after administration of 40% glucose solution and

then returned to baseline values. In contrast, ob/ob

and db/db mice exhibited elevated blood glucose lev-

els for up to 60  min, with normalization occurring

only 120  min after glucose administration (Fig.  S2f).

We also measured glucose levels in animals under

fasting conditions and upon ad libitum access to

food and found no significant difference in this pa-

rameter among the groups (Fig.  S1d in the Online Re-

source  1). When analyzing the area under the curve

(AUC), the values were 1228  ±  80  mmol  ×  min/l for

WT mice, 1829  ±  220  mmol  ×  min/l for ob/ob mice,

and 1615  ±  147  mmol  ×  min/l for db/db mice. Analysis

of the total AUC revealed significant differences be-

tween WT mice and both genetically altered strains

(Fig. S2e in the Online Resource 1).

STROKE IN LEPTIN-DEFICIENT MICE 945

BIOCHEMISTRY (MOSCOW) Vol. 91 No. 6 2026

Fig.  3. Ischemic lesion volume in mice of different genotypes on day 7 after PT-induced stroke. a)  Representative T2-weighted

MRI images of brain sections from WT, ob/ob, and db/db mice, showing the location and morphology of ischemic lesions in

the sensorimotor cortex of the right hemisphere. b)  Quantitative analysis of infarct volume (mm

) based on measurements

of the hyperintense signal area across consecutive MRI sections.

The effect of leptin deficiency and LepR defi-

ciency on the lesion size in the PT model. To as-

sess the extent of brain damage following PT-induced

stroke, MRI was performed on day  7 after surgery

in WT, ob/ob, and db/db mice (Fig.  3). T2-weighted

images revealed a hyperintense signal region corre-

sponding to the ischemic focus in the sensorimotor

cortex of the right hemisphere (Fig.  3a). Quantita-

tive analysis showed that the mean lesion volume

was 23.94  ±  2.91  mm

in WT mice, 15.19  ±  2.4  mm

in ob/ob mice, and 17.45  ±  3.09  mm

in db/db mice

(Fig.  3b). Therefore, despite pronounced metabolic

differences between the strains, infarct volumes were

comparable, suggesting that the cortical sensitivity to

standardized PT damage is largely unaffected by im-

paired leptin signaling (including complete inactiva-

tion of all LepR isoforms in db/db mice). No mortality

was observed in the WT group following PT induc-

tion, while mortality in the ob/ob and db/db groups

reached 20% and 40%, respectively. Infarct volume

was analyzed only in surviving animals.

Assessment of neurological deficit. To assess

neurological deficits after PT, mice of the WT, ob/ob,

and db/db groups were subjected to a series of be-

havioral tests, including beam walking, rotarod, and

grid walking tests, before surgery (baseline) and on

postoperative days 3, 6, and 7. Because of pronounced

intergroup differences in body weight and baseline

motor activity, absolute behavioral test values may

not accurately reflect the extent of neurological defi-

cit. Therefore, in several tests, we also analyzed rel-

ative changes in behavioral indices in each group

compared to individual pre-stroke baseline values

taken as 100%.

In the grid walking test, all three groups showed

a significant increase in the number of foot faults

made by the left (affected) forelimb after PT com-

pared with the baseline, indicating the development

of a persistent sensorimotor deficit (Fig.  4a). The

number of left forelimb slips within 5 min in WT

mice was 65.00  ±  10.57 on day  3 and 45.00  ±  4.75 on

day  7. In ob/ob mice, these values were 19.67  ±  5.23

and 24.67  ±  7.83, and in db/db mice, 23.00  ±  4.72 and

21.00  ±  1.73, respectively (Fig.  4a). Although the abso-

lute number of slips was lower in ob/ob and db/db

mice than in WT animals, this likely reflected their

reduced motor activity rather than superior motor

coordination.

BOCHARNIKOV et al.946

BIOCHEMISTRY (MOSCOW) Vol. 91 No. 6 2026

Fig. 4. Assessment of sensorimotor deficits after PT in mice of different genotypes. a)  Absolute number of slips of forelimbs

and hindlimbs on the affected (left) side in the grid walking test. b)  Locomotor performance in the rotarod test calculated

from the time the animal remained on the rotarod drum and its body weight. c)  Overall neurological deficit score in the

beam walking test. d)Time required to traverse the beam in beam walking test. *  p  <  0.05; **  p  <  0.01; ***  p  <  0.001, statis-

tically significant differences (two-way ANOVA with post hoc Tukey’s test for multiple comparisons; Kruskal–Wallis test in

the grid walking test; and Holm–Sidak test with the Welch correction for track time in the beam walking test).

Since ob/ob and db/db mice are obese, maintain-

ing balance on the rotarod drum is more challeng-

ing for them than for WT mice, and therefore they

remained on the apparatus for significantly shorter

time periods. To minimize the impact of body weight

when assessing the severity of neurological damage,

we used an integrated “retention impulse” parameter

(see Materials and Methods). PT caused a significant

decline in this parameter in all groups. On day  3 af-

ter PT, the retention impulse was 4235  ±  533.1  s  ×  g

in WT mice, in 1022  ±  436.8  s  ×  g in ob/ob mice, and

2150  ±  440.2  s  ×  g in db/db mice (Fig. 4b) (p  <  0.005

vs. baseline values). The retention impulse of ob/ob

mice was significantly lower than that of WT mice

(p  <  0.001). Moreover, WT mice demonstrated signifi-

cantly better performance on day  3 post-stroke than

both mutant groups (p  <  0.001 vs. ob/ob and p  <  0.05

vs. db/db).

In the beam walking test, there was also a ten-

dency toward development of neurological deficits af-

ter PT in all groups (Fig.  4c). On day  3, the number of

left limb slips was comparable between the groups:

48.14  ±  8.63 in WT mice, 41.25  ±  5.28 in ob/ob mice,

and 55.85  ±  8.56 in db/db mice. On day  6, the corre-

sponding values were 35.91  ±  9.05, 40.67  ±  4.73, and

43.12  ±  3.43, respectively (Fig.  4c). However, the time

required to traverse the beam differed significantly

between WT and mutant mice. On day  3, WT mice

STROKE IN LEPTIN-DEFICIENT MICE 947

BIOCHEMISTRY (MOSCOW) Vol. 91 No. 6 2026

Table 1. Relative change in sensorimotor activity pa-

rameters in the grid walking and rotarod test after PT

in mice of different genotypes compared to pre-stroke

baseline

WT ob/ob db/db

Grid walking test;

number of front left paw slips

Intact 100% 100% 100%

day 3 1769.23%* 761.9% 6666.67%**

day 7 1423.08% 1142.86% 6500%**

Grid walking test;

number of hind left paw slips

Intact 100% 100% 100%

day 3 509.67% 1125%** 805.56%

day 7 274.19% 791.66%* 527.77%

Grid walking test;

total number of slips on the left side

Intact 100% 100% 100%

day 3 881.81%* 893.93% 1642.85%**

day 7 613.63% 1015.15% 1380.95%**

Rotarod test

Intact 100% 100% 100%

day 3 64.58%** 29.52%** 37.85%*

day 6 87.87% 71.54% 59.41%

Note. * p < 0.05, ** p < 0.005, significant differences with

baseline (Kruskal–Wallis test for the grid walking test and

two-way ANOVA for the rotarod test).

crossed the beam in 21.33  ±  0.6  s, whereas ob/ob mice

required 49.5  ±  4.76 s (Fig. 4d) (p < 0.05). It should

be noted that the travel time primarily reflects the

overall movement speed, which was reduced in ob/ob

and db/db mice due to diabetes and obesity, and may

not reflect post-stroke neurological deficits in these

strains.

When analyzing intragroup dynamics (i.e., chang-

es relative to each animal’s pre-stroke baseline), all

three genotypes exhibited significant functional de-

terioration after PT (Table  1). In all behavioral tests,

performance markedly declined on post-stroke days

3 and/or 7. However, db/db mice demonstrated the

greatest relative decline in performance, as evi-

denced by the marked increase in number of slips

in the grid walking test (1642.9% of baseline on day  3

and 1381.0% on day  7) and poor performance in the

rotarod test (37.9% of baseline on day  3 and 59.4%

on day  6). In contrast, ob/ob mice exhibited a degree

of relative impairment comparable to that observed

in WT animals. Thus, despite similar lesion volumes,

db/db mice demonstrated the most pronounced func-

tional deterioration, suggesting possible differences in

the mechanisms of early post-stroke functional com-

pensation in leptin-deficient and LepR-deficient ani-

mals with metabolic syndrome.

Analysis of gene expression after PT. To assess

the extent of damage to the sensorimotor cortex and

the severity of post-ischemic inflammatory response,

we analyzed expression profiles of several proinflam-

matory cytokines and markers of macrophage polar-

ization and activation of microglia and astrocytes

in brain samples. On day  10 after PT, expression of

genes for the proinflammatory markers cystatin  F

(Cst7) (Fig.  5a) and chemokine (C-C motif) ligand  3

(Ccl3) (Fig.  5b) was upregulated in all animal groups.

In contrast, pronounced stroke-induced increase in

the mRNA level of the classical proinflammatory cy-

tokine tumor necrosis factor  α (TNFa) was found only

in WT and ob/ob groups (Fig.  5c), while the expression

of the proinflammatory marker C-X-C motif chemo-

kine ligand  1 (Cxcl1) remained unchanged (Fig.  5d).

At the same time, expression of monokine induced

by gamma-interferon (Cxcl9) (Fig.  5e) increased only

in the WT group. Baseline expression of cyclooxy-

genase  2 (Cox2) was higher in both genetically mod-

ified strains and decreased after the stroke (Fig.  5g).

A similar trend was observed for interleukin-6 (Il6)

expression in db/db mice (Fig.  5f). The damage to

the sensorimotor cortex in WT mice led to a signifi-

cant upregulation of genes for the Toll-like receptor  2

(Tlr2) and cluster of differentiation 36 (Cd36), which

are markers of M1 and M2 macrophages, respectively

(Fig.  5, h  and  i). However, increased Cd36 expression

was observed only in WT and db/db mice, but not

in ob/ob mice (Fig.  5i). The expression of markers

associated with nerve tissue damage, such as glial

fibrillar acid protein (Gfap) and matrix metallopro-

teinase 3 (Mmp3), increased significantly after stroke

in db/db mice (Fig.  5, j  and  k). WT and ob/ob mice

also showed a significant increase in Gfap expres-

sion, while upregulation of Mmp3 in response to PT

in these animals did not reach the level of statistical

significance (Fig.  5k). The baseline expression of the

brain-derived neurotrophic factor (Bdnf ) (Fig.  5l) was

higher in db/db mice and decreased significantly after

the stroke only in this group.

Analysis of phosphorylation of signaling kinas-

es after PT. To assess the level and phosphorylation

status of key kinases associated with leptin signaling

in the cerebral cortex after PT, we analyzed the total

BOCHARNIKOV et al.948

BIOCHEMISTRY (MOSCOW) Vol. 91 No. 6 2026

Fig.  5. Expression of key markers associated with tissue damage and post-ischemic responses in the cerebral cortex 10 days

after PT: Changes in expression of are shown for genes of proinflammatory factors Cst7  (a), Ccl3  (b), Tnfa  (c), Cxcl1  (d),

Cxcl9  (e), Il6  (f), Cox2  (g), macrophage markers Tlr2  (h) and Cd36  (i), astrocyte marker Gfap  (j), neurotrophic factor Bdnf  (l),

and matrix metalloproteinase Mmp3  (k). *  p  <  0.05, **  p  <  0.005, ***  p  <  0.001, ****  p  <0.0001, statistically significant differ-

ences with baseline (two-way ANOVA with Tukey’s post  hoc test).

STROKE IN LEPTIN-DEFICIENT MICE 949

BIOCHEMISTRY (MOSCOW) Vol. 91 No. 6 2026

Fig.  6. Analysis of key signaling pathways and their components involved in leptin-mediated signaling and responsible for

neuroprotection in the cerebral cortex. a)  Total Akt and phosphorylated Akt (p-Akt) levels. b)  Total ERK1/2 (p44/p42) and

phosphorylated ERK1/2 (p-ERK1/2) levels. Protein expression was normalized to the total protein load (see TGX® gels shown

below the corresponding immunoblots).

levels of Akt (protein kinase  B) and ERK1/2 (extra-

cellular signal-regulated kinase 1/2) and their phos-

phorylated forms by Western blotting. Quantitative

analysis revealed no significant differences in the

phosphorylation levels of Akt or ERK1/2 among the

studied genotypes on day  10 after PT (Fig.  6), indicat-

ing the absence of sustained changes in the activi-

ty of PI3K/Akt and MAPK/ERK signaling pathways in

the subacute phase of damage, despite the observed

differences in functional parameters. The lack of

differences in these signaling pathways at the stud-

ies time point suggests that the observed functional

impairments in db/db mice are not associated with

sustained dysregulation of these signaling cascades in

the subacute phase or may reflect earlier or transient

changes in their activity.

DISCUSSION

It has been well established that ischemic stroke

is more severe in patients with obesity, diabetes,

and other metabolic disorders  [1,  3]. This phenome-

non has often been studied in mouse models with

impaired leptin signaling [2, 10]. However, the mech-

anisms underlying an increased vulnerability of ner-

vous tissue to ischemia in these conditions, partic-

ularly the respective contributions of diabetes and

leptin signaling, remain poorly understood, partly due

to limitations inherent to the available experimental

models. In this study, we employed the photoinduced

cerebral vascular thrombosis model that generates a

volume-standardized lesion while minimizing the im-

pact of collateral blood flow on the penumbra  [14].

Using this approach, we demonstrated that neither

leptin deficiency (ob/ob mice) nor complete inactiva-

tion of all LepR isoforms (our db/db line) increased

the primary infarct volume.

Despite the known metabolic differences between

these mutant strains and WT animals, stroke severity,

as assessed morphologically, was comparable across

all mouse strains (Fig.  3). In contrast, the functional

consequences of stroke differed markedly, as db/db

mice showed more pronounced sensorimotor deficits

following stroke (Fig.  4, Table  1). This suggests that

leptin signaling is unlikely to play a major role in

the mechanisms of acute nervous tissue damage but

may contribute to other pathogenic processes that

influence post-stroke outcomes. We found no signifi-

cant differences in the phosphorylation levels of Akt

BOCHARNIKOV et al.950

BIOCHEMISTRY (MOSCOW) Vol. 91 No. 6 2026

or ERK1/2 on day 10 after PT, indicating the absence

of persistent changes in these signaling pathways

during the subacute phase of damage. Therefore, the

observed genotype-dependent differences in the func-

tional recovery between genotypes are unlikely to re-

sult from prolonged dysregulation of these cascades

and may instead be related to other mechanisms,

including impaired neuroplasticity. It should be not-

ed that mortality was higher in the knockout lines,

probably reflecting increased systemic susceptibility

of these animals to ischemia. This observation does

not contradict the lack of differences in infarct vol-

ume among surviving animals.

It should be emphasized that most previously

published studies investigating the effect of diabetes

mellitus on brain resistance to ischemia have used

the classical transient MCAO model. In this model,

ob/ob and db/db mice consistently exhibit significant-

ly larger infarct volume and more severe functional

outcomes than WT animals [11, 17-20]. However, in-

terpretation of these findings requires consideration

of the model’s specifics. MCAO and other models in-

volving mechanical vascular occlusion mimic throm-

bosis of a major cerebral artery, and the severity

of ischemic damage in such experiments is strongly

influenced by systemic and cerebral hemodynamics.

Critical factors include the capacity for blood supply

compensation through collateral vessels and the cir-

cle of Willis, as well as the functional state of the en-

dothelium  [21]. Moreover, unlike PT, MCAO produces

extensive regions of ischemic penumbra. Functional

outcomes, therefore, depend largely on the survival

of neurons in this area, which is governed by a com-

plex interplay of hemodynamic and metabolic factors.

Therefore, the greater infarct volume observed in

ob/ob and db/db mice following MCAO may primar-

ily be due to the vascular components of pathogene-

sis, including endothelial dysfunction, increased BBB

permeability, and impaired angiogenesis [19, 22-24].

Together, these vascular abnormalities promote expan-

sion of the infarct area in MCAO, making it difficult

to distinguish their contribution from that of intrin-

sic neuronal susceptibility to ischemia. Nevertheless,

available evidence suggests that neuronal survival

in diabetes remains compromised even when the in-

fluence of vascular factors is minimized, which has

been attributed to chronic systemic inflammation and

changes in inflammatory responses in brain tissue  [2,

25,  26]. Unlike MCAO, PT induces local thrombosis in

distal vessels without affecting major arteries. Lesion

size and location are determined primarily by the pa-

rameters of laser irradiation and are only minimally

affected by collateral blood flow and other vascular

factors. As a result, vascular abnormalities play much

less prominent role in this model, allowing a more

direct assessment of the resistance of neurons and

astrocytes to ischemia, which has been proposed to

be impaired under diabetic conditions based on find-

ings from MCAO studies in ob/ob and db/db mice  [27].

Our data indicate that when the lesion is standard-

ized, the diabetic phenotype alone does not increase

the extent of primary damage, even in the complete

absence of LepR signaling, which supports the idea

that the exacerbation of stroke in diabetic animals in

the MCAO model is driven predominantly by vascular

factors.

Analysis of the subacute inflammatory response

on day 10 after PT revealed increased expression of

markers associated with microglial and astroglial ac-

tivation (Cst7, Ccl3, and Gfap) in animals of all geno-

types (Fig.  5, a,  b, and  j). These genes are commonly

used as conventional markers of brain damage  [28,

29]. The upregulation of Cst7 was consistent with its

role as a mediator of microglial reactivity  [30], while

the increase in Ccl3 expression reflected the develop-

ment of a pro-inflammatory response  [31]. The simi-

larity of expression patterns of these markers among

the experimental groups indicates that the overall

microglial response to damage was largely preserved

in mice with impaired leptin signaling. We also ob-

served upregulated expression of the proinflammato-

ry cytokine Tnfa gene in WT and ob/ob mice (Fig.  5c),

together with the elevated expression levels of mac-

rophage phenotype-associated markers Tlr2 and Cd36

(Fig.  5, h  and  i). The absence of expression activation

for Tnfa and Cxcl9 in the brains of db/db mice may

be due to suppression of the acute inflammatory

response in this strain, possibly resulting from im-

paired immune response or chronic inflammation,

as indicated by an elevated baseline Tnfa expression.

Upregulated Tlr2 expression has previously been as-

sociated with poor stroke outcomes, while increased

Cd36 expression in the prolonged post-stroke period

may be associated with the resolution of inflamma-

tion and activation of reparative processes  [32,  33].

In this context, a less pronounced increase in Cd36

expression on day  10 after PT in the ob/ob mice may

indicate a disruption of the balance between pro- and

anti-inflammatory signaling and a predominance of

pro-inflammatory processes during the late post-

stroke phase  [34]. The observed upregulation of Gfap

and Mmp3 is consistent with the well-established post-

stroke mechanisms of reactive gliosis and extracellu-

lar matrix remodeling [35-39]. A more pronounced

increase in Mmp3 expression in the db/db mice (but

not in ob/ob mice) may lead to the increased BBB per-

meability in these animals. The baseline expression

of the neurotrophic factor Bdnf gene was elevated

in db/db mice, possibly representing a compensatory

response to chronic inflammatory stress. However,

PT induced a significant decrease in Bdnf expres-

sion exclusively in this group, whereas no significant

STROKE IN LEPTIN-DEFICIENT MICE 951

BIOCHEMISTRY (MOSCOW) Vol. 91 No. 6 2026

Fig. 7. Proposed mechanisms by which impaired leptin signaling affects the severity of PT-induced stroke in ob/ob and db/db

mice. PDK1, 3-phosphoinositide-dependent protein kinase-1; Raf, Raf (rapidly accelerated fibrosarcoma) kinase. Created in

https://BioRender.com (license: cue7l0x).

changes were detected in WT and ob/ob mice. The

downregulation of Bdnf during the late post-stroke

stages may indicate impaired neuroplasticity in db/db

mice. Interestingly, reduced plasma BDNF levels were

found in stroke patients with type  2 diabetes mellitus

[40,  41], despite evidence that baseline BDNF concen-

trations may be higher in diabetic individuals than

in healthy controls  [41]. Therefore, the pattern of

changes in Bdnf expression in our model is broadly

consistent with clinical observations. Despite compa-

rable infarct volumes and similar inflammatory re-

sponses, db/db mice exhibited the greatest relative

decline in motor functions (Table  1). This observation

is unlikely to reflect decreased tolerance to acute in-

jury itself, but rather suggests that impaired leptin

signaling may adversely affect post-stroke recovery

mechanisms, particularly those involving brain neu-

roplasticity. It should be noted that in our study,

neuroplasticity was not evaluated directly, and this

conclusion is therefore based on indirect data, in-

cluding functional outcomes and changes in Bdnf ex-

pression.

It has already been established that neurogene-

sis in db/db mice is disrupted  [42,  43], and these an-

imals are often referred to as a model of impaired

cognitive plasticity  [44]. In addition, axonal shorten-

ing has been observed in neurons derived from db/db

mouse  [45], along with changes in the expression of

neuroplasticity-related genes  [46].

In addition to its role in regulating metabolism,

leptin is a key modulator of neuronal plasticity and

inflammatory processes in the CNS  [47]. Its effects

on NMDA receptors, as well as on the MEK (mito-

gen-activated protein kinase kinase)/ERK and PI3K

(phosphatidylinositol  3-kinase)/Akt signaling pathways

[48,  49], are central to adaptive responses following

brain damage. However, only the long leptin recep-

tor isoform (LepRb) possesses the full spectrum of

signaling activity and activates the JAK/STAT, MAPK,

and PI3K cascades  [7] associated with neuropro-

tection and synaptic plasticity. Beyond the nervous

system, leptin signaling via the JAK/STAT pathway

in macrophages and T-cells modulates immune re-

sponses and inflammation, including secretion of

BOCHARNIKOV et al.952

BIOCHEMISTRY (MOSCOW) Vol. 91 No. 6 2026

pro-inflammatory cytokines IL6 and TNFα, in particu-

lar, by the microglia  [50,  51]. In contrast, short recep-

tor isoforms (LepRa, LepRc, LepRe, LepRf) perform

other functions, such as leptin transport across the

BBB (LepRa, LepRc) and regulation of leptin bioavail-

ability in the brain (LepRe)  [52,  53]. In the db/db mice

(B6-LepR

d10

strain generated for this study), a muta-

tion in LepR causes a frameshift early in the coding

sequence, leading to inactivation of all LepR isoforms.

Consequently, both LepRb-dependent signaling and

the functions of short LepR isoforms (leptin transport

and accumulation) were abolished in this model. This

allowed us to consider this strain as a model of com-

plete leptin signaling ablation, equivalent to leptin

deficiency, but without the influence of isoform-spe-

cific factors.

Unlike the db/db strain, ob/ob mice do not pro-

duce leptin due to a mutation in the Lep gene  [54,  55].

In these animals, LepRb and other isoforms remain

structurally and functionally intact; however, the ab-

sence of their ligand prevents the activation of leptin

signaling despite preserved receptor architecture. In

contrast, the db/db strain carries a mutation that caus-

es complete inactivation of entire LepR family and,

therefore, fully suppresses leptin signaling even in the

presence of ligand excess. Together, these models pro-

vide a complementary framework for distinguishing

the contributions of leptin deficiency and LepR defi-

ciency to post-stroke processes, unlike conventional

db/db mouse strains with preserved activity of short

receptor isoforms. In ob/ob mice, leptin deficiency

likely disrupts neuroplasticity by limiting activation

of LepRb and related PI3K/Akt-dependent mechanisms

of neuronal survival and axonal sprouting, whereas

complete loss of LepR in the db/db strain leads to a

more profound disruption of leptin-dependent trophic

support and reduces the capacity of neural networks

for functional compensation after damage.

Therefore, disruptions in leptin signaling in the

pathogenesis of post-stroke injury occur via two pos-

sible mechanisms (Fig.  7): leptin deficiency in ob/ob

mice restricts neurotrophic support by preventing ac-

tivation of LepR-dependent processes, while complete

loss of LepR signaling due to the absence of all recep-

tor isoforms in db/db mice promotes chronic metabol-

ic syndrome, which compromises post-stroke recovery.

At the same time, LepR signaling-mediated activation

of neuroinflammation was not observed in our db/db

model, as confirmed by transcriptional analysis. It is

important to note that complete inactivation of all

LepR isoforms in these mice is an important meth-

odological advantage, because it eliminates residual

signaling activity characteristic of conventional db/db

strains and enables evaluation of the leptin pathway

involvement without contribution of individual LepR

isoforms. These findings also highlight the potential

of therapeutic approaches aimed at modulating in-

flammation or alternative neuroplasticity pathways.

CONCLUSION

Our study demonstrates that leptin signaling

affects stroke outcomes by modulating post-stroke

neuroplasticity rather than exacerbating primary

ischemic damage. In the used PT model providing

standardized lesion size, neither leptin deficiency nor

LepR inactivation increased infarct volume. However,

both conditions impaired functional recovery after

the stroke. Importantly, disruption of leptin signaling

occurred via two different mechanisms: ob/ob mice

were deficient for leptin while retaining function-

al leptin receptors, whereas generated db/db mice

lacked all functional LepR isoforms, which eliminat-

ed potential isoform-specific effects and strengthened

the conclusion that LepR signaling does not directly

affect infarct volume. Our findings also highlight the

importance of adapting neurological assessment pro-

tocols for obesity and diabetes models, since altered

locomotor performance in these animals makes abso-

lute values of behavioral indicators unreliable crite-

ria for neurological deficits. Furthermore, the differ-

ences between the PT and MCAO models emphasize

the critical importance of model selection when in-

vestigating stroke mechanisms. Unlike MCAO, PT min-

imizes the contribution of vascular factors, thereby

enabling identification of direct effects of metabolic

disorders on the neural tissue.

Abbreviations

BBB blood–brain barrier

BDNF brain-derived neurotrophic factor

Ccl3 chemokine (C-C motif) ligand 3

(macrophage inflammatory protein 1a)

Cox2 cyclooxygenase 2

Cst7 cystatin F

Cxcl9 gamma interferon-induced monokine

GFAP glial fibrillar acid protein

IL6 interleukin 6

LepRb leptin receptor long isoform

MCAO middle cerebral artery occlusion

Mmp3 matrix metalloproteinase 3

PT photothrombosis

qPCR quantitative polymerase chain reac-

tion

Tlr2 Toll-like receptor 2 (CD282)

TNFa tumor necrosis factor α

WT wild-type

Supplementary information

The online version contains supplementary material

available at https://doi.org/10.1134/S0006297926600134.

STROKE IN LEPTIN-DEFICIENT MICE 953

BIOCHEMISTRY (MOSCOW) Vol. 91 No. 6 2026

Contributions

E.Y.P. and P.V.S. developed the concept and supervised

the study; A.D.B., P.A.A., E.I.Y., O.A.A., O.O.G., A.V.P.,

S.A.G., O.A.P., and I.B.P. conducted experiments; E.Y.P.,

P.V.S., A.D.B., P.A.A., I.B.P., N.V.A, and A.A.B. discussed

research results; E.Yu.P., A.D.B., and P.A.A. wrote the

text of the article; E.Y.P., A.D.B., P.A.A., I.B.P., N.V.A, and

A.A.B. edited the manuscript.

Funding

The work was supported by the Ministry of Science

and Higher Education of the Russian Federation

(agreement no. 075-15-2025-489).

Ethics approval and consent to participate

All animal procedures complied with the ethical stan-

dards of the Belozersky Research Institute of Physi-

co-Chemical Biology and other institutions and legal

acts of the Russian Federation and international or-

ganizations. Animal research protocol was approved

by the Ethics Committee of the Belozersky Research

Institute of Physico-Chemical Biology (protocol no.

015-8/10/2024; June 09, 2025).

Conflict of interest

The authors of this work declare that they have no

conflicts of interest.

REFERENCES

1. Baird, T. A., Parsons, M. W., Barber, P. A., Butcher, K. S., Desmond, P. M., Tress, B. M., Colman, P. G., Jerums, G.,

Chambers, B. R., and Davis, S. M. (2002) The influence of diabetes mellitus and hyperglycaemia on stroke in-

cidence and outcome, J. Clin. Neurosci., 9, 618-626, https://doi.org/10.1054/jocn.2002.1081.

2. Terao, S., Yilmaz, G., Stokes, K. Y., Ishikawa, M., Kawase, T., and Granger, D. N. (2008) Inflammatory and injury

responses to ischemic stroke in obese mice, Stroke, 39, 943-950, https://doi.org/10.1161/STROKEAHA.107.494542.

3. Lau, L. H., Lew, J., Borschmann, K., Thijs, V., and Ekinci, E. I. (2019) Prevalence of diabetes and its effects on

stroke outcomes: a meta-analysis and literature review, J. Diabetes Invest., 10, 780-792, https://doi.org/10.1111/

jdi.12932.

4. Ding, H., and Triggle, C. R. (2010) Endothelial dysfunction in diabetes: multiple targets for treatment, Pflugers

Arch., 459, 977-994, https://doi.org/10.1007/s00424-010-0807-3.

5. Sallam, N. A., and Laher, I. (2020) Redox signaling and regional heterogeneity of endothelial dysfunction in

db/db mice, Int. J. Mol. Sci., 21, 6147, https://doi.org/10.3390/ijms21176147.

6. Suriano, F., Vieira-Silva, S., Falony, G., Roumain, M., Paquot, A., Pelicaen, R., Régnier, M., Delzenne, N. M., Raes, J.,

Muccioli, G. G., et al. (2021) Novel insights into the genetically obese (ob/ob) and diabetic (db/db) mice: two

sides of the same coin, Microbiome, 9, 147, https://doi.org/10.1186/s40168-021-01097-8.

7. Gorska, E., Popko, K., Stelmaszczyk-Emmel, A., Ciepiela, O., Kucharska, A., and Wasik, M. (2010) Leptin receptors,

Eur. J. Med. Res., 15, 50, https://doi.org/10.1186/2047-783X-15-S2-50.

8. Wauman, J., Zabeau, L., and Tavernier, J. (2017) The leptin receptor complex: heavier than expected? Front.

Endocrinol., 8, 30, https://doi.org/10.3389/fendo.2017.00030.

9. Sahai, A., Malladi, P., Pan, X., Paul, R., Melin-Aldana, H., Green, R. M., and Whitington, P. F. (2004) Obese and

diabetic db/db mice develop marked liver fibrosis in a model of nonalcoholic steatohepatitis: role of short-

form leptin receptors and osteopontin, Am. J. Physiol. Gastrointest. Liver Physiol., 287, G1035-G1043, https://

doi.org/10.1152/ajpgi.00199.2004.

10. Valerio, A., Dossena, M., Bertolotti, P., Boroni, F., Sarnico, I., Faraco, G., Chiarugi, A., Frontini, A., Giordano, A.,

Liou, H. C., et al. (2009) Leptin is induced in the ischemic cerebral cortex and exerts neuroprotection through

NF-kappaB/c-Rel-dependent transcription, Stroke, 40, 610-617, https://doi.org/10.1161/STROKEAHA.108.528588.

11. Jiang, Y., Lin, L., Liu, N., Wang, Q., Yuan, J., Li, Y., Chung, K. K., Guo, S., Yu, Z., and Wang, X. (2020) FGF21

protects against aggravated blood-brain barrier disruption after ischemic focal stroke in diabetic db/db male

mice via cerebrovascular PPARγ activation, Int. J. Mol. Sci., 21, 824, https://doi.org/10.3390/ijms21030824.

12. Fluri, F., Schuhmann, M. K., and Kleinschnitz, C. (2015) Animal models of ischemic stroke and their application

in clinical research, Drug Des. Dev. Ther., 9, 3445, https://doi.org/10.2147/DDDT.S56071.

13. Espinosa, A., Meneses, G., Chavarría, A., Mancilla, R., Pedraza-Chaverri, J., Fleury, A., Bárcena, B., Pérez-

Osorio, I. N., Besedovsky, H., Arauz, A., et al. (2020) Intranasal dexamethasone reduces mortality and brain

damage in a mouse experimental ischemic stroke model, Neurotherapeutics, 17, 1907, https://doi.org/10.1007/

s13311-020-00884-9.

14. Uzdensky, A. B. (2018) Photothrombotic stroke as a model of ischemic stroke, Transl. Stroke Res., 9, 437-451,

https://doi.org/10.1007/s12975-017-0593-8.

15. Truett, G. E., Heeger, P., Mynatt, R. L., Truett, A. A., Walker, J. A., and Warman, M. L. (2000) Preparation

of PCR-quality mouse genomic DNA with hot sodium hydroxide and tris (HotSHOT), Biotechniques, 29, 52-54,

https://doi.org/10.2144/00291bm09.

BOCHARNIKOV et al.954

BIOCHEMISTRY (MOSCOW) Vol. 91 No. 6 2026

16. Watson, B. D., Dietrich, W. D., Busto, R., Wachtel, M. S., and Ginsberg, M. D. (1985) Induction of reproduc-

ible brain infarction by photochemically initiated thrombosis, Ann. Neurol., 17, 497-504, https://doi.org/10.1002/

ana.410170513.

17. Yang, Z., Huang, C., Huang, W., Yan, C., Wen, X., Hu, D., Xie, H., He, K., Tsang, C. K., and Li, K. (2024) Exacer-

bated ischemic brain damage in type 2 diabetes via methylglyoxal-mediated miR-148a-3p decline, BMC Med.,

22, 557, https://doi.org/10.1186/s12916-024-03768-3.

18. Iwatani, Y., Hayashi, H., Yamamoto, H., Minamikawa, H., Ichikawa, M., Orikawa, H., Masuda, A., Tada, N.,

Moriyama, Y., and Takagi, N. (2024) Pathogenic role of NAMPT in the perivascular regions after ischemic

stroke in mice with type 2 diabetes mellitus, Exp. Neurol., 371, 114584, https://doi.org/10.1016/j.expneurol.

2023.114584.

19. Cui, X., Chopp,M., Zacharek, A., Ye, X., Roberts, C., and Chen, J. (2011) Angiopoietin/Tie2 pathway mediates type2

diabetes induced vascular damage after cerebral stroke, Neurobiol. Dis., 43, 285-292, https://doi.org/10.1016/

j.nbd.2011.04.005.

20. Vannucci, S. J., Willing, L. B., Goto, S., Alkayed, N. J., Brucklacher, R. M., Wood, T. L., Towfighi, J., Hurn, P. D.,

and Simpson, I. A. (2001) Experimental stroke in the female diabetic, db/db, mouse, J. Cereb. Blood Flow Metab.,

21, 52-60, https://doi.org/10.1097/00004647-200101000-00007.

21. Akamatsu, Y., Nishijima, Y., Lee, C. C., Yang, S. Y., Shi, L., An, L., Wang, R. K., Tominaga, T., and Liu, J. (2015)

Impaired leptomeningeal collateral flow contributes to the poor outcome following experimental stroke in the

type 2 diabetic mice, J. Neurosci., 35, 3851-3864, https://doi.org/10.1523/JNEUROSCI.3838-14.2015.

22. Kumari, R., Willing, L. B., Patel, S. D., Baskerville, K. A., and Simpson, I. A. (2011) Increased cerebral matrix

metalloprotease-9 activity is associated with compromised recovery in the diabetic db/db mouse following a

stroke, J. Neurochem., 119, 1029-1040, https://doi.org/10.1111/j.1471-4159.2011.07487.x.

23. Yukami, T., Yagita, Y., Sugiyama, Y., Oyama, N., Watanabe, A., Sasaki, T., Sakaguchi, M., Mochizuki, H., and

Kitagawa, K. (2015) Chronic elevation of tumor necrosis factor-α mediates the impairment of leptomeningeal

arteriogenesis in db/db mice, Stroke, 46, 1657-1663, https://doi.org/10.1161/STROKEAHA.114.008062.

24. Kumari, R., Bettermann, K., Willing, L., Sinha, K., and Simpson, I. A. (2020) The role of neutrophils in mediat-

ing stroke injury in the diabetic db/db mouse brain following hypoxia-ischemia, Neurochem. Int., 139, 104790,

https://doi.org/10.1016/j.neuint.2020.104790.

25. Haley, M. J., Mullard, G., Hollywood, K. A., Cooper, G. J., Dunn, W. B., and Lawrence, C. B. (2017) Adipose tissue

and metabolic and inflammatory responses to stroke are altered in obese mice, Dis. Model Mech., 10, 1229-1243,

https://doi.org/10.1242/dmm.030411.

26. Kumari,R., Willing, L. B., Patel, S.D., Krady, J. K., Zavadoski, W. J., Gibbs, E.M., Vannucci, S. J., and Simpson, I. A.

(2010) The PPAR-gamma agonist, darglitazone, restores acute inflammatory responses to cerebral hypoxia-isch-

emia in the diabetic ob/ob mouse, J. Cereb. Blood Flow Metab., 30, 352-360, https://doi.org/10.1038/jcbfm.

2009.221.

27. Kim, B., Sullivan, K. A., Backus, C., and Feldman, E. L. (2011) Cortical neurons develop insulin resistance and

blunted Akt signaling: a potential mechanism contributing to enhanced ischemic injury in diabetes, Antioxid.

Redox Signal., 14, 1829-1839, https://doi.org/10.1089/ars.2010.3816.

28. Ianni, M., Corraliza-Gomez, M., Costa-Coelho, T., Ferreira-Manso, M., Inteiro-Oliveira, S., Alemãn-Serrano, N.,

Sebastião, A. M., Garcia, G., Diógenes, M. J., and Brites, D. (2024) Spatiotemporal dysregulation of neuron-Glia

related genes and pro-/anti-inflammatory miRNAs in the 5xFAD mouse model of Alzheimer’s disease, Int. J. Mol.

Sci., 25, 9475, https://doi.org/10.3390/ijms25179475.

29. Pallarés-Moratalla, C., and Bergers, G. (2024) The ins and outs of microglial cells in brain health and disease,

Front. Immunol., 15, 1305087, https://doi.org/10.3389/fimmu.2024.1305087.

30. Wang, Y.-T., Li, Q., Liu, J.-C., Chen, C., Ding, H.-X., Zha, X., and Zhang, K. (2025) Cystatin F-a key player in central

nervous system disease, J. Neuroinflammation, 22, 203, https://doi.org/10.1186/s12974-025-03526-z.

31. Pelisch, N., Rosas Almanza, J., Stehlik, K. E., Aperi, B. V., and Kroner, A. (2020) CCL3 contributes to secondary

damage after spinal cord injury, J. Neuroinflammation, 17, 362, https://doi.org/10.1186/s12974-020-02037-3.

32. Bai, M., Sun,R., Cao, B., Feng,J., and Wang,J. (2023) Monocyte-related cytokines/chemokines in cerebral ischemic

stroke, CNS Neurosci. Ther., 29, 3693, https://doi.org/10.1111/cns.14368.

33. Woo, M.-S., Yang, J., Beltran, C., and Cho, S. (2016) Cell surface CD36 protein in monocyte/macrophage contrib-

utes to phagocytosis during the resolution phase of ischemic stroke in mice, J. Biol. Chem., 291, 23654, https://

doi.org/10.1074/jbc.M116.750018.

34. Hu, X., Li, P., Guo, Y., Wang, H., Leak, R. K., Chen, S., Gao, Y., and Chen, J. (2012) Microglia/macrophage polariza-

tion dynamics reveal novel mechanism of injury expansion after focal cerebral ischemia, Stroke, 43, 3063-3070,

https://doi.org/10.1161/STROKEAHA.112.659656.

STROKE IN LEPTIN-DEFICIENT MICE 955

BIOCHEMISTRY (MOSCOW) Vol. 91 No. 6 2026

35. Abdul,Y., Jamil, S., Li, W., and Ergul,A. (2023) Cerebral microvascular matrix metalloproteinase-3 (MMP3) contrib-

utes to vascular injury after stroke in female diabetic rats, Neurochem. Int., 162, 105462, https://doi.org/10.1016/

j.neuint.2022.105462.

36. Jickling, G. C., Liu, D., Stamova, B., Ander, B. P., Zhan, X., Lu, A., and Sharp, F. R. (2014) Hemorrhagic trans-

formation after ischemic stroke in animals and humans, J. Cereb. Blood Flow Metab., 34, 185-199, https://

doi.org/10.1038/jcbfm.2013.203.

37. Yang, Y., and Rosenberg, G. A. (2011) Blood-brain barrier breakdown in acute and chronic cerebrovascular

disease, Stroke, 42, 3323, https://doi.org/10.1161/STROKEAHA.110.608257.

38. Liu, Z., Li, Y., Cui, Y., Roberts, C., Lu, M., Wilhelmsson, U., Pekny, M., and Chopp, M. (2014) Beneficial effects of

gfap/vimentin reactive astrocytes for axonal remodeling and motor behavioral recovery in mice after stroke,

Glia, 62, 2022-2033, https://doi.org/10.1002/glia.22723.

39. Li, Y., Chen, J., Zhang, C. L., Wang, L., Lu, D., Katakowski, M., Gao, Q., Shen, L. H., Zhang, J., Lu, M., et al. (2005)

Gliosis and brain remodeling after treatment of stroke in rats with marrow stromal cells, Glia, 49, 407-417,

https://doi.org/10.1002/glia.20126.

40. Gao, Q., Hei, B., Hu, S., Zhou, J., Wang, B., Wang, D., Wang, D., Liu, J., and Fan, J. (2025) Plasma BDNF is re-

duced in acute ischemic stroke patients with type 2 diabetes mellitus and associated with fibrin-rich thrombi,

J. Diabetes Res., 2025, 6693363, https://doi.org/10.1155/jdr/6693363.

41. Chaturvedi, P., Singh, A. K., Tiwari, V., and Thacker, A. K. (2020) Brain-derived neurotrophic factor levels in

acute stroke and its clinical implications, Brain Circ., 6, 185-190, https://doi.org/10.4103/bc.bc_23_20.

42. Bracke,A., Domanska, G., Bracke, K., Harzsch, S., van den Brandt,J., Bröker, B., and von Bohlen Und Halbach, O.

(2019) Obesity impairs mobility and adult hippocampal neurogenesis, J. Exp. Neurosci., 13, 1179069519883580,

https://doi.org/10.1177/1179069519883580.

43. Bonds, J. A., Shetti, A., Stephen, T. K. L., Bonini, M. G., Minshall, R. D., and Lazarov, O. (2020) Deficits in hippo-

campal neurogenesis in obesity-dependent and -independent type-2 diabetes mellitus mouse models, Sci. Rep.,

10, 16368, https://doi.org/10.1038/s41598-020-73401-9.

44. Yermakov, L. M., Griggs, R. B., Drouet, D. E., Sugimoto, C., Williams, M. T., Vorhees, C. V., and Susuki, K. (2019)

Impairment of cognitive flexibility in type 2 diabetic db/db mice, Behav. Brain Res., 371, 111978, https://

doi.org/10.1016/j.bbr.2019.111978.

45. Yermakov, L. M., Drouet, D. E., Griggs, R. B., Elased, K. M., and Susuki, K. (2018) Type 2 diabetes leads to axon

initial segment shortening in mice, Front. Cell. Neurosci., 12, 146, https://doi.org/10.3389/fncel.2018.00146.

46. Xiang, Q., Tao, J.-S., Dong, S., Liu, X.-L., Yang, L., Liu, L.-N., Deng, J., and Li, X.-H. (2024) Heterogeneity and

synaptic plasticity analysis of hippocampus based on db mice induced diabetic encephalopathy, Psychoneuro-

endocrinology, 159, 106412, https://doi.org/10.1016/j.psyneuen.2023.106412.

47. Horvath, T. L. (2006) Synaptic plasticity in energy balance regulation, Obesity (Silver Spring), 14, 228S-233S,

https://doi.org/10.1038/oby.2006.314.

48. Signore, A. P., Zhang, F., Weng, Z., Gao, Y., and Chen, J. (2008) Leptin neuroprotection in the CNS: mechanisms

and therapeutic potentials, J. Neurochem., 106, 1977-1990, https://doi.org/10.1111/j.1471-4159.2008.05457.x.

49. Shanley, L. J., Irving, A. J., and Harvey, J. (2001) Leptin enhances NMDA receptor function and modulates hip-

pocampal synaptic plasticity, J. Neurosci., 21, RC186, https://doi.org/10.1523/JNEUROSCI.21-24-j0001.2001.

50. Santos-Alvarez, J., Goberna, R., and Sánchez-Margalet, V. (1999) Human leptin stimulates proliferation and acti-

vation of human circulating monocytes, Cell. Immunol., 194, 6-11, https://doi.org/10.1006/cimm.1999.1490.

51. Tang, C. H., Lu, D.Y., Yang, R. S., Tsai, H. Y., Kao, M.C., Fu, W.M., and Chen, Y. F. (2007) Leptin-induced IL-6 pro-

duction is mediated by leptin receptor, insulin receptor substrate-1, phosphatidylinositol 3-kinase, Akt, NF-kappaB,

and p300 pathway in microglia, J. Immunol., 179, 1292-1302, https://doi.org/10.4049/jimmunol.179.2.1292.

52. Golden, P. L., Maccagnan, T. J., and Pardridge, W. M. (1997) Human blood-brain barrier leptin receptor. Binding

and endocytosis in isolated human brain microvessels, J.Clin. Invest., 99, 14-18, https://doi.org/10.1172/JCI119125.

53. Boado, R. J., Golden, P. L., Levin, N., and Pardridge, W. M. (1998) Up-regulation of blood-brain barrier short-form

leptin receptor gene products in rats fed a high fat diet, J. Neurochem., 71, 1761-1764, https://doi.org/10.1046/

j.1471-4159.1998.71041761.x.

54. Rehni, A. K., Liu, A., Perez-Pinzon, M. A., and Dave, K. R. (2017) Diabetic aggravation of stroke and animal

models, Exp. Neurol., 292, 63, https://doi.org/10.1016/j.expneurol.2017.03.004.

55. Zhang, Y., Proenca, R., Maffei, M., Barone, M., Leopold, L., and Friedman, J. M. (1994) Positional cloning of the

mouse obese gene and its human homologue, Nature, 372, 425-432, https://doi.org/10.1038/372425a0.

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.