Article

ISSN 0006-2979, Biochemistry (Moscow), 2026, Vol. 91, No. 6, pp. 967-978 © The Author(s), 2026. This article is an open access publication.

967

HMGB1 Protein in Alcohol-Related

Psychopathological Disorders at Various Stages

of Alcohol Dependence

Tamara P. Vetlugina

1,a

*, Elena V. Epimakhova

, Ivan V. Voevodin

Valentina D. Prokopieva

, and Nikolay A. Bokhan

Mental Health Research Institute, Tomsk National Research Medical Center, Russian Academy of Sciences,

634014 Tomsk, Russia

e-mail: vetluga21@mail.ru

Received January 26, 2026

Revised May 12, 2026

Accepted May 15, 2026

Abstract— Background. Chronic alcohol consumption in alcohol dependence (AD) is associated with various

molecular and cellular dysfunctions. One of the key mechanisms involved is the extracellular release of

HMGB1 (high-mobility group box  1), a nuclear protein that acts as a proinflammatory signaling molecule.

However, the dynamics of peripheral HMGB1 levels across different stages of AD and its association with

psychopathological symptoms remain insufficiently understood. Objective. To investigate peripheral HMGB1

concentrations in patients with AD during different disease stages – alcohol withdrawal syndrome (AWS),

post-abstinence state (PAS), and remission – and to assess their associations with clinical characteristics of

the disease. Patients and Methods. The study included 53 men aged 30-60 years diagnosed with AD and 19

conditionally healthy men who served as controls for biological analyses. Clinical and biological assessments

were performed at three time points: (1)  during AWS; (2)  after completion of standard therapy (PAS); and

(3)  during remission following hospital discharge. The severity of psychopathological symptoms was evalu-

ated using validated clinical rating scales. Serum HMGB1 concentrations were measured by enzyme-linked

immunosorbent assay (ELISA). Results. Patients with AD demonstrated significantly elevated serum HMGB1

levels during AWS and PAS compared with controls, whereas HMGB1 concentrations decreased to control

values during remission. Correlation and ROC analyses revealed that higher HMGB1 levels were associ-

ated with a less pronounced reduction in psychopathological symptoms following treatment. Conclusion.

Peripheral HMGB1 may represent a promising biomarker for monitoring treatment response and predicting

relapse risk in AD.

DOI: 10.1134/S0006297926600237

Keywords: HMGB1 protein, alcohol dependence, remission, relapse

* To whom correspondence should be addressed.

INTRODUCTION

Alcohol dependence (AD) is a chronic relapsing

condition characterized by recurrent cycles of remis-

sion and relapse, with stable remission often difficult

to achieve despite treatment. The limited effectiveness

of current therapeutic approaches is largely attribut-

able to the multifaceted effects of ethanol, which dis-

rupts physiological processes at both molecular and

cellular levels. These effects include alterations in

specific brain structures involved in the emergence

of dependence syndrome, as well as systemic toxic

effects that contribute to the development of comor-

bid somatic pathologies  [1, 2].

The identification of peripheral biomarkers for

the response to treatment and risk of relapse remains

an important objective for improving therapeutic strat-

egies in patients with AD. Potential biomarkers may

be identified based on their association with clinical

manifestations observed at different disease stages.

VETLUGINA et al.968

BIOCHEMISTRY (MOSCOW) Vol. 91 No. 6 2026

Chronic alcohol consumption, as well as the

use of other psychoactive substances, induce long-

term alterations in the brain’s neurotransmitter

systems, leading to dysregulation of neural net-

works responsible for emotional processing, psy-

chophysiological status, and overall behavioral con-

trol  [1,  3-6].

The toxic effects of ethanol are closely linked

to its ability to induce excessive production of re-

active oxygen species, resulting in the disruption

of cellular redox homeostasis, development of oxi-

dative stress, mitochondrial DNA damage, and sub-

sequent organ dysfunction  [7-9]. In chronic alcohol

exposure, oxidative stress impairs autophagy and

mitophagy, promotes the accumulation of oxidatively

modified biomacromolecules, and contributes to neu-

rotoxicity, neuronal injury, and neurodegenerative

processes  [10,  11].

Disruption of cellular homeostasis promotes the

formation of endogenous damage-associated molec-

ular patterns (DAMPs). A prototypical DAMP is the

non-histone nuclear protein high-mobility group box  1

(HMGB1), which is ubiquitously expressed across var-

ious cells and tissues and exhibits a wide range of

biological functions. HMGB1 has been extensively

studied over the past several decades (see reviews

[12-14]). The functional diversity of HMGB1 is de-

termined by its structural characteristics, post-trans-

lational modifications, redox state, and intracellular

localization. Under physiological conditions, HMGB1

primarily functions in the nucleus, where it acts

as a DNA chaperone involved in essential cellular

processes, including transcription, DNA replication,

and DNA repair.

When cellular integrity is disrupted, HMGB1 is

released to the extracellular space through two prin-

cipal mechanisms: passive release from necrotic or

damaged cells and active secretion by innate immune

cells in response to various stimuli, such as pro-in-

flammatory mediators, stress, and ethanol. Extracel-

lular HMGB1 functions as a DAMP or alarmin  [12,

15,  16]. It interacts with several pattern recognition

receptors (PRRs), in particular, members of the Toll-

like receptor (TLR) family (TLR2 and TLR4) and the

receptor for advanced glycation end-products (RAGE),

thereby triggering innate immune response and the

production of pro-inflammatory mediators  [17,  18].

Although RAGE and TLRs activate inflammatory re-

sponses through different signaling cascades, analy-

sis of the published data suggests potential func-

tional interactions between RAGE and TLR4 at the

cell surface during the development of inflamma-

tion [19].

In  AD, exogenous ethanol readily crosses the

blood–brain barrier (BBB) and induces HMGB1 re-

lease from brain cells, including microglia, neu-

rons, and astrocytes. HMGB1 subsequently mediates

pro-inflammatory signaling through TLRs and RAGE,

promoting neuroinflammation that adversely affects

brain regions involved in behavior and emotional

regulation, increases anxiety, and contributes to the

AD development [20].

Additionally, alcohol and its metabolites com-

promise intestinal barrier integrity, promoting the

translocation of pathogen-associated molecular pat-

terns (PAMPs), including lipopolysaccharide (LPS),

a structural component of the outer membrane of

gram-negative bacteria, into the bloodstream. Elevat-

ed circulating LPS levels have been reported in pa-

tients with AD  [21,  22]. In response to LPS exposure,

innate immune cells release HMGB1, thereby trigger-

ing signaling cascades that stimulate the production

of pro-inflammatory cytokines, thus contributing to

the development of systemic inflammation and pro-

moting organ damage, which might lead to liver dys-

function  [23,  24].

Studies investigating peripheral HMGB1 levels

in AD have yielded inconsistent results. In patients

with alcoholic hepatitis, serum HMGB1 concentrations

were significantly elevated compared with controls

both at diagnosis and after 30 days of follow-up, al-

though no associations between HMGB1 and clinical

outcomes were observed  [25]. In another study, in-

creased serum HMGB1 levels were found in patients

with alcoholic liver disease (ALD) compared with AD

patients without liver pathology, and HMGB1 concen-

trations positively correlated with ALD severity  [26].

A pilot study evaluating the clinical utility of circu-

lating biomarkers in 25  AD  patients during a two-

week period of alcohol abstinence found elevated

plasma HMGB1 levels only in patients with cogni-

tive impairment, whereas no increase was detected

in the overall AD group. The authors also reported

a positive correlation between HMGB1, soluble RAGE

(sRAGE), and duration of alcohol consumption  [27].

Elevated plasma HMGB1 levels were identified in

AD patients during alcohol withdrawal syndrome

(AWS); however, no associations were found between

HMGB1 concentrations and either clinical character-

istics of the disorder or comorbid liver disease  [28].

Most existing studies have assessed HMGB1 levels

only at isolated AD stages. Consequently, the longi-

tudinal dynamics of this inflammatory mediator and

its relationship with psychopathological symptoms

across different stages of AD remain poorly under-

stood.

The aim of the present study was to investi-

gate changes in peripheral HMGB1 concentrations

in patients with AD during AWS, post-abstinence pe-

riod, and remission, as well as to examine associa-

tions between HMGB1 levels and clinical AD charac-

teristics.

HMGB1 AND ALCOHOL DEPENDENCE 969

BIOCHEMISTRY (MOSCOW) Vol. 91 No. 6 2026

PATIENTS AND METHODS

The study included 53 male patients aged 30-60

years (mean age, 48.29  ±  7.51 years) with AD who

were admitted for treatment at the clinic of the

Mental Health Research Institute, Tomsk National

Research Medical Center. Diagnoses were established

according to the ICD-10 criteria: Mental and behavior-

al disorders due to use of alcohol (dependence syn-

drome – F10.21). Patients were admitted to the clinic

in a withdrawal state (F10.30). The average duration

of the disease defined by the presence of the formed

AWS was 16.55  ±  8.90 years. Exclusion criteria in-

cluded refusal to participate in the study, severe co-

morbid psychiatric disorders classified under ICD-10

categories F0, F2, or F3, and decompensated somatic

diseases. All patients underwent clinical evaluation

and received pharmacological treatment in accor-

dance with current standards of psychiatric care.

Clinical and biological assessments of the patient

cohort were performed longitudinally at three time

points:

1. Time point  1 (53 patients): during AWS on days

2-3 of alcohol withdrawal after the relief of acute

withdrawal manifestations.

2. Time point  2 (53 patients): prior to the discharge

from the hospital, after 14-17  days of standard

therapy, corresponding to the post-abstinence

state (PAS) and the initial stage of therapeutic

remission formation.

3. Time point  3 (20  patients): during remission, at

a mean follow-up period of 5.36  ±  3.45 months

after hospital discharge.

The maintenance of patient sobriety in remission

was controlled by the absence of ethanol vapors in

exhaled air during repeated clinic visits, as well as

conversations with relatives. Patients were excluded

from the study after discharge from the hospital due

to refusal to further participate, relapse, or moving

out of the city. The average age and duration of the

disease of patients who continued to participate in

the study were 47.00  ±  7.92 years and 15.39  ±  9.77

years, respectively, and did not differ statistically sig-

nificantly from the initial values.

The control group for biological studies consist-

ed of 19 conditionally healthy men who were not on

dispensary records, had no chronic somatic diseases

in the acute stage, and had no signs of acute infec-

tions at the time of examination. The average age

(42.70  ±  10.54 years) did not differ statistically from

the average age in the patient group.

The severity of the patients’ clinical condition was

assessed using the following standardized clinical scales:

• Anxiety and depression symptoms were evalu-

ated using the Hospital Anxiety and Depression

Scale (HADS) [29].

• Post-abstinence symptoms were assessed using

the Quantified Comprehensive Psychopathologi-

cal Rating Scale (CPRS) in Remission (QCPRSR).

According to the developers’ recommenda-

tions, this scale is intended for evaluating psy-

chopathological symptoms after resolution of

acute abstinence state and enables assessment

of symptom dynamics during remission forma-

tion  [30].

• The severity of alcohol craving was evaluated

using the Pennsylvania Alcohol Craving Scale

(PACS)  [31].

Venous blood samples were collected from the

cubital vein in the morning after overnight fasting

using a sterile single-use Vacutainer system contain-

ing a clotting activator for serum preparation. Serum

samples were aliquoted into Eppendorf tubes, frozen,

and stored at −80°C until analysis

Serum HMGB1 concentrations were measured

using enzyme-linked immunosorbent assay (ELISA)

with a Human HMGB1 ELISA Kit (FineTest, China),

according to the manufacturer’s instructions. The as-

say had a coefficient of variation below  10%, a de-

tection range of 31.25-2000  pg/mL, and a sensitivity

of 18.75  pg/mL. Absorbance was measured with a

Varioskan LUX multimode reader (Thermo Scientific,

USA) at the Medical Genomics Shared Research Cen-

ter, Tomsk National Research Medical Center. HMGB1

concentrations were expressed in pg/mL.

Statistical processing of data was carried out us-

ing Statistica, v.  13.0 and MedCalc, v.  18.9.1. Descrip-

tive statistics are presented as the median (Me) and

interquartile range [25%  Q1; 75%  Q3]. Age of par-

ticipants and disease duration are presented as the

mean ± standard deviation (M ± SD).

Intergroup comparisons of quantitative vari-

ables were conducted using the Mann–Whitney

U  test for independent samples, the Wilcoxon signed-

rank test for paired samples, and the Kruskal–Wal-

lis test for comparisons among three independent

groups. Correlation analyses were performed using

the Spearman’s rank correlation coefficient. Non-

parametric statistical methods were applied because

clinical variables were expressed as ordinal (rank)

data and biological variables did not follow a nor-

mal distribution. Normality was assessed using the

Shapiro–Wilk test and the Kolmogorov–Smirnov

test with Lilliefors correction. The potential utility

of HMGB1 protein as a biomarker for clinical im-

provement was evaluated using receiver operating

characteristic (ROC) curve analysis. The area under

the ROC curve (AUC) with a 95% confidence interval

(95% CI), the optimal cut-point value for the Youd-

en index, sensitivity, and specificity were evaluated.

Differences were considered statistically significant

at p < 0.05.

VETLUGINA et al.970

BIOCHEMISTRY (MOSCOW) Vol. 91 No. 6 2026

RESULTS

Assessment of patients’ clinical condition

(Table  1) demonstrated that, at study point  1,

the QCPRSR scores for psychopathological symptoms

and PACS scores for craving exceeded threshold

values by 2-3 times. Following the course of thera-

py (point  2) and during remission (point 3), clinical

parameters showed a positive trend toward normal-

ization and approached threshold levels. However,

patients in remission exhibited a significant increase

in PACS scores compared with point  2 (p

  =  0.041),

along with a trend toward higher QCPRSR scores

  =  0.062). These findings may reflect a resurgence

of relapse-related symptoms.

Despite the overall positive therapeutic effect,

residual anxiety symptoms (8 or more points) at

discharge persisted in 5.8% of patients. Incomplete

normalization of psychopathological symptoms (6 or

more points) was observed in 28.8% cases, while

craving (8 or more points on the PACS scale) re-

mained elevated in 5.8% patients. During remission,

elevated QCPRSR scale scores were detected in 42.1%

patients, and 21.1% continued to exhibit increased

craving levels.

Serum HMGB1 concentrations in AD patients

were also evaluated (Table  2).

As shown in Table  2, HMGB1 concentrations in

patients were significantly elevated compared with

the control group at both point  1 (p  =  0.030) and

point  2 (p  =  0.009). In contrast, during the remission

stage (point  3), HMGB1 levels do not differ signifi-

cantly from those observed in the control group.

Spearman correlation analysis was performed to

assess the relationship between HMGB1 levels and

clinical characteristics of the disease. Table  3 sum-

marizes the correlations between HMGB1 concentra-

tions measured at three study time points and clini-

cal indicators assessed at hospital discharge (point  2).

A higher concentration of HMGB1 in patients at

point  1 was associated with more pronounced asthe-

nia at discharge, as assessed by the QCPRSR (r

  =  0.33;

t  =  2.42; p  =  0.019). In addition, HMGB1 concentration

at point  2 showed a positive correlation with the dy-

namics of the total QCPRSR score during hospital-

ization (r

  =  0.31; t  =  2.20; p  =  0.033). Given that the

therapeutic goal was a reduction in the total scale

score over time, this positive association indicates

reduced treatment effectiveness in patients with el-

evated HMGB1 levels.

Table 1. Clinical characteristics of AD patients across study stages; Me [Q1; Q3]

Index Point 1 (AWS),

n=53

Point 2 (PAS),

n=53

Point 3 (remission),

n=20

Threshold

value

Anxiety, HADS score 8 [5; 11] 2 [1; 4]

 < 0.001

2 [1; 6]

 < 0.001 p

 = 0.363

≤7

Depression, HADS score 6 [4; 10] 2 [1; 4]

 < 0.001

3 [0; 5]

 = 0.003 p

 = 0.489

≤7

Psychopathological symptoms,

total QCPRSR score

15 [13; 16] 3 [2; 6]

 < 0.001

5 [3; 7]

 = 0.001 p

 = 0.062

≤5

Craving, PACS score 16 [10; 20] 0 [0; 3]

 < 0.001

2 [0; 5]

1

= 0.002 p

 = 0.041

≤7

Note. n, number of examined individuals; p

, statistically significant compared to data at point 1, Wilcoxon criterion;

, statistically significant compared to data at point 2, Wilcoxon criterion.

Table 2. HMGB1 concentration in the blood serum of AD patients at different disease stages

Indicator Study points Control group,

n = 19

Point 1 (AWS),

n = 53

Point 2 (PAS),

n = 53

Point 3 (remission),

n = 20

HMGB1,

pg/mL

1417.2

[1228.2; 1681.8]

= 0.030

1551.3

[1309.5; 1720.9]

= 0.009

1307.8

[992; 1435.2]

= 0.923

1286.6

[1160.9; 1447.0]

Note. The data are shown as Me [Q1; Q3]; p

, statistically significant relative to the control.

HMGB1 AND ALCOHOL DEPENDENCE 971

BIOCHEMISTRY (MOSCOW) Vol. 91 No. 6 2026

Table 3. Spearman rank correlations between HMGB1 concentration at different study time points and clinical

indicators at discharge (point 2)

Clinical indices

Spearman correlation coefficient (r

)

HMGB1, point 1 HMGB1, point 2 HMGB1, point 3

HADS indices at point 2

Anxiety −0.03 0.06 −0.13

Depression 0.20 0.11 −0.14

Changes in anxiety index between points 1 and 2 −0.02 −0.15 −0.11

Changes in depression index between points 1 and 2 0.17 0.10 0.05

QCPRSR indices at point 2

Emotional component of alcohol craving 0.21 0.12 −0.20

Autonomic component of alcohol craving 0.16 −0.03 0.00

Ideational component of alcohol craving −0.16 0.00 −0.15

Behavioral component of alcohol craving 0.13 0.03 −0.16

Attitude toward disease 0.04 0.06 −0.18

Commitment to sobriety 0.08 0.09 −0.13

Depression 0.02 0.13 0.16

Anxiety 0.09 0.21 0.06

Dysphoria 0.08 0.02 −0.12

Asthenia 0.33* −0.02 0.10

Psychoorganic syndrome 0.14 0.01 0.00

Psychopath-like syndrome 0.00 0.02 0.01

Total QCPRSR score 0.16 0.13 −0.18

Changes in total QCPRSR score between points 1 and 2 0.26 0.31** −0.12

PACS indices at point 2

Total craving score −0.02 −0.05 −0.11

Changes in craving score between points 1 and 2 0.26 −0.00 0.29

Note. Statistically significant correlations: * p = 0.019; ** p = 0.033.

In patients in remission (Table  4), analysis re-

vealed a direct correlation between HMGB1 levels

at point  1 and dysphoric manifestations (accord-

ing to QCPRSR) during remission (r

  =  0.46; t  =  2.20;

p  =  0.041). Elevated HMGB1 at point  1 was also as-

sociated with higher anxiety scores (HADS) during

remission (r

  =  0.48; t  =  2.17; p  =  0.045).

Next, we evaluated HMGB1 as a potential pre-

dictor of efficacy of treatment, the main goal of

which is achieving high-quality remission. The as-

sessment was conducted using two approaches:

(i)  comparative analysis of HMGB1 levels and their

dynamics in groups with different treatment out-

comes (high vs. low efficacy), and (ii)  ROC analysis.

Patient groups were stratified according to the se-

verity of psychopathological symptoms measured

by the QCPRSR. The total QCPRSR score of ≤5 was

used as a threshold for classification into the

VETLUGINA et al.972

BIOCHEMISTRY (MOSCOW) Vol. 91 No. 6 2026

Table 4. Spearman rank correlations between HMGB1 concentration at different study time points and clinical

indicators in remission (point 3)

Clinical indices

Spearman correlation coefficient (r

)

HMGB1, point 1 HMGB1, point 2 HMGB1, point 3

HADS indices at point 2

Anxiety 0.36 0.10 −0.13

Depression 0.33 −0.15 0.19

Changes in anxiety index between points 1 and 2 0.48* 0.31 −0.12

Changes in depression index between points 1 and 2 0.29 −0.13 0.20

QCPRSR indices at point 2

Emotional component of alcohol craving −0.13 −0.16 0.03

Autonomic component of alcohol craving −0.13 −0.11 −0.07

Ideational component of alcohol craving −0.09 0.12 −0.10

Behavioral component of alcohol craving 0.06 −0.10 0.11

Attitude toward disease 0.14 −0.09 −0.06

Commitment to sobriety −0.06 −0.06 0.10

Depression 0.03 0.06 −0.12

Anxiety 0.25 0.12 −0.16

Dysphoria 0.46** 0.05 −0.11

Asthenia 0.40 0.17 0.17

Psychoorganic syndrome −0.19 −0.11 0.30

Psychopath-like syndrome −0.20 0.13 0.11

Total QCPRSR score 0.08 −0.18 −0.11

Changes in total QCPRSR score between points 1 and 2 −0.01 −0.12 −0.02

PACS indices at point 2

Total craving score −0.04 −0.18 −0.16

Changes in craving score between points 1 and 2 −0.15 −0.13 −0.15

Note. Statistically significant correlations: * p = 0.045; ** p = 0.041.

high- efficacy group, indicating efficient symptom

reduction. Patients whose score at time point  2 re-

mained above this threshold were assigned to the

low-efficacy group. No statistically significant age

differences were observed between the groups.

Comparative analysis demonstrated (Fig.  1) that

elevated HMGB1 concentrations at time point  2 were

associated with a lower likelihood of achieving thresh-

old values of the total QCPRSR score by the time of

hospital discharge, reflecting reduced treatment effi-

cacy (p  =  0.037).

The results of ROC analysis are presented in Fig. 2.

If normalization of the total QCPRSR score was

used as an outcome variable (Fig.  2a), the AUC was

0.692 (95% CI: 0.543-0.815). At the protein level

≤1593  pg/mL, 70.3% of patients achieved normalization

of QCPRSR scores, indicating high treatment efficacy.

In contrast, at protein levels >1593  pg/mL, 69.2% of

HMGB1 AND ALCOHOL DEPENDENCE 973

BIOCHEMISTRY (MOSCOW) Vol. 91 No. 6 2026

Fig. 1. HMGB1 levels in the blood serum of AD patients from groups with different therapy efficacy at study point  2. Com-

parison groups:  0,  high efficacy (reduction of score to threshold values ≤5); 1,  low efficacy, QCPRSR scores above threshold

values; statistically significant differences in HMGB1 levels: Z  =  −2.09; p  =  0.0371 (Mann–Whitney U  test).

Fig. 2. ROC analysis of HMGB1 levels at study point  2 as a biomarker of treatment efficacy. a)  ROC curve for the classifying

variable “normalization of QCPRSR score”; b)  ROC curve for the classifying variable “complete remission of psychopatho-

logical symptoms.”

patients failed to achieve normalization of clinical

indicators.

When complete remission of all psychopatho-

logical symptoms, including QCPRSR, HADS, and

PACS scores, was considered as the outcome variable

(Fig.  2b), the AUC was 0.706 (95%  CI: 0.560-0.826).

At the protein level ≤1590.6  pg/mL, 72.7% patients

achieved full normalization of the clinical status,

VETLUGINA et al.974

BIOCHEMISTRY (MOSCOW) Vol. 91 No. 6 2026

whereas at protein levels >1590.6  pg/mL, 70.6% pa-

tients continued to exhibit residual symptoms, re-

flected by QCPRSR, HADS, or PACS scores remaining

above threshold values

DISCUSSION

The study included longitudinal observation of

the same group of patients with AD, with assessment

of HMGB1 dynamics at three time points. Elevated

serum HMGB1 concentrations were detected in pa-

tients on days  2-3 after hospital admission during

AWS (point 1). HMGB1 levels remained elevated after

a two-week therapy course at the PAS stage (point  2)

and normalized only during remission after several

months of follow-up (point  3) (Table  2).

Chronic alcohol consumption in AD is associated

with molecular and cellular dysfunctions in multi-

ple organ systems. One of the mechanisms underly-

ing these alterations is the passive or active release

of HMGB1 to the extracellular space. Extracellular

HMGB1 acts as a mediator of neuroimmune inflam-

mation and metabolic processes  [16]. In the alcohol-

affected brain, ethanol exposure induces transcrip-

tional changes characterized by activation of genes

encoding pro-inflammatory mediators  [20].

Elevated HMGB1 concentrations observed in our

study may be related to the ethanol-induced release

of HMGB1 from brain cells. This hypothesis is sup-

ported by experimental studies in rodents and cell

cultures, as well as analyses of postmortem brain

tissue from individuals with a history of alcohol

abuse  [20,  32,  33]. These studies have demonstrated

increased HMGB1 levels in multiple brain regions

and in peripheral blood during ethanol intoxication,

which persisted during abstinence. In a rat model of

chronic alcohol exposure, increased HMGB1 mRNA

expression in the striatum was detected on both

day  1 and day  14 of ethanol withdrawal  [34].

The increase of HMGB1 in peripheral blood

is associated with its ability to cross the BBB. In a

mouse model of LPS-induced endotoxemia, HMGB1

was detected in the bloodstream 8-32  h after LPS

administration [35]. An experimental study using

HMGB1labeled with radioactive iodine demonstrated

that this protein can cross the BBB bidirectionally,

thereby providing a potential mechanism for neuro-

immune signaling between the brain and peripheral

tissues  [36].

In AD, HMGB1 release into the circulation is

also promoted by impaired intestinal barrier integ-

rity. Increased intestinal permeability facilitates LPS

translocation into the bloodstream, stimulates HMGB1

release by innate immune cells, activates pro-inflam-

matory signaling, and contributes to the development

of systemic inflammation [21-23]. Current evidence

supports the existence of a bidirectional commu-

nication between the intestine and the brain, and

disruption of the gut–brain axis has been implicat-

ed in the pathogenesis of AD  [37]. The mechanisms

underlying this interaction are complex and involve

metabolic, immune, neural, and endocrine pathways

[38-40]. Circulating cytokines are considered key me-

diators in the gut–brain communication, reaching the

central nervous system (CNS) (through several routes:

via brain regions with a deficient BBB, through cyto-

kine-specific transport systems, and through afferent

fibers of the vagus nerve, thereby promoting neu-

roinflammation and neuroadaptive changes in the

CNS [41-43].

Analysis of associations between HMGB1 levels

and clinical manifestations revealed several signifi-

cant relationships. Elevated HMGB1 concentrations

during AWS were associated with more severe as-

thenic symptoms at the PAS stage, more pronounced

dysphoric symptoms, and higher anxiety indicators

during remission (Tables  3 and  4). In addition, elevat-

ed HMGB1 levels after therapy at time point  2 were

associated with a less pronounced reduction in total

psychopathological symptom scores to the threshold

values (Fig.  1), suggesting lower treatment efficacy,

poorer establishment of therapeutic remission, and

increased risk of relapse. ROC analysis (Fig.  2a and  b)

further supported these findings, demonstrating that

HMGB1 levels below the threshold at time point  2

predicted high treatment efficacy in 70.3-72.7% cases.

Since HMGB1 can cross the BBB bidirectionally

[35,  36], changes in its blood concentration may re-

flect neuroadaptive alterations occurring in the brain.

Evidence suggests that in women with alcohol use

disorder, elevated circulating levels of HMGB1 and

other pro-inflammatory markers are associated with

poorer performance in tests evaluating episodic mem-

ory and executive functioning  [44]. Increased plasma

HMGB1 concentrations have also been reported in

patients with AD who exhibited severe cognitive im-

pairments  [27].

Crews et  al.  [20] proposed a hypothesis stating

the central role of HMGB1 and TLR signaling in the

alcohol-induced neuroimmune activation  [20]. Accord-

ing to this model, in healthy brain, neurons, astro-

cytes, and microglia release trophic factors that main-

tain neuronal and glial functioning. Chronic alcohol

exposure induces transcriptional reprogramming

characterized by increased expression of pro-inflam-

matory genes. HMGB1 released from damaged or

activated cells is thought to stimulate TLR-mediat-

ed pro-inflammatory signaling, thereby promoting

alterations in neural networks, heightened anxiety,

cognitive dysfunction, and ultimately reinforcing al-

cohol consumption.

HMGB1 AND ALCOHOL DEPENDENCE 975

BIOCHEMISTRY (MOSCOW) Vol. 91 No. 6 2026

Interpretation of data on the involvement of

HMGB1 in neuroinflammatory and systemic inflam-

matory processes in AD is complicated by sever-

al limitations. Among the most significant are the

functional heterogeneity of HMGB1, whose activity

depends on its structural form and cellular localiza-

tion  [14], as well as the influence of comorbid somatic

disorders.

HMGB1 is currently considered as a promising

therapeutic target in the treatment of inflammato-

ry diseases and AD. Consequently, a range of neu-

roimmune signaling inhibitors is under develop-

ment, including antibodies targeting HMGB1/IL-1β

complexes  [45], recombinant HMGB1–Box  A protein,

acetylcholine  [46], glycyrrhizin  [47], and carbenoxol-

one  [48].

CONCLUSION

Overall, remission formation in AD patients is

characterized by distinct changes in clinical indicators

and peripheral HMGB1 levels. During AWS, HMGB1

concentrations are significantly elevated compared to

controls. Following therapy, HMGB1 levels remain in-

creased, likely reflecting residual psychopathological

symptoms, but subsequently normalize during remis-

sion. Correlation, comparative, and ROC analyses of

clinical and laboratory parameters demonstrated that

poorer therapeutic outcomes during inpatient treat-

ment, as well as greater severity of relapse-related

symptoms during remission, are associated with ele-

vated HMGB1 levels. Therefore, HMGB1 may serve as

a promising peripheral biomarker of treatment effica-

cy and risk of relapse in AD.

Abbreviations

AD alcohol dependence

AWS alcohol withdrawal syndrome

BBB blood–brain barrier

DAMPs damage-associated molecular patterns

ELISA enzyme-linked immunosorbent assay

HADS Hospital Anxiety and Depression Scale

HMGB1 high-mobility group box  1

QCPRSR Quantified Comprehensive Psycho-

pathological Ranking Scale in Remis-

sion

PACS Pennsylvania Alcohol Craving Scale

PAS post-abstinence state

RAGE receptor for advanced glycation

end-products

ROC receiver operating characteristic

TLRs Toll-like receptors

Contributions

T.P.V. and N.A.B. developed the concept and supervised

the study; E.V.E. and V.D.P. conducted experiments;

I.V.V. performed clinical examination of patients;

T.P.V., E.V.E., and I.V.V. discussed research results and

wrote the text of the article; T.P.V., V.D.P., and N.A.B.

edited the manuscript.

Funding

The study was supported by the Russian Science

Foundation (project 25-25-00188; https://rscf.ru/en/

project/25-25-00188)/.

Ethics approval and consent to participate

All procedures performed in studies involving human

participants were in accordance with the ethical stan-

dards of the national research ethics committee and

the 1964 Helsinki Declaration and its later amend-

ments or comparable ethical standards. Informed

voluntary consent was obtained from each partici-

pant included in the study. The study was approved

by the Local Ethics Committee at the Mental Health

Research Institute, Tomsk National Research Medi-

cal Center (Protocol no. 181; February 11, 2025; case

no. 181/1.2025).

Conflict of interest

The authors of this work declare that they have no

conflicts of interest.

Open access

This article is licensed under a Creative Commons

Attribution 4.0 International License, which permits

use, sharing, adaptation, distribution, and reproduc-

tion in any medium or format, as long as you give

appropriate credit to the original author(s) and the

source, provide a link to the Creative Commons li-

cense, and indicate if changes were made. The im-

ages or other third party material in this article are

included in the article’s Creative Commons license,

unless indicated otherwise in a credit line to the

material. If material is not included in the article’s

Creative Commons license and your intended use

is not permitted by statutory regulation or exceeds

the permitted use, you will need to obtain permis-

sion directly from the copyright holder. To view a

copy of this license, visit http://creativecommons.org/

licenses/by/4.0/.

REFERENCES

1. Anokhina, I. P. (2017) The basic biological mechanisms of substance use disorders [in Russian], J.Addict. Prob-

lems, 2-3, 15-41.

VETLUGINA et al.976

BIOCHEMISTRY (MOSCOW) Vol. 91 No. 6 2026

2. Peregud, D. I., and Gulyaeva, N. V. (2024) Contribution of visceral systems to the development of substance

use disorders: translational aspects of interaction between central and peripheral mechanisms, Biochemistry

(Moscow), 89, 1868-1888, https://doi.org/10.1134/S0006297924110026.

3. Koob, G. F., and Volkow, N. D. (2016) Neurobiology of addiction: a neurocircuitry analysis, Lancet Psychiatry,

3, 760-773, https://doi.org/10.1016/S2215-0366(16)00104-8.

4. Abrahao, K. P., Salinas, A. G., and Lovinger, D. M. (2017) Alcohol and the brain: neuronal molecular targets,

synapses, and circuits, Neuron, 96, 1223-1238, https://doi.org/10.1016/j.neuron.2017.10.032.

5. Polunina, A. G., and Bryun, E. A. (2018) Neural network activity against the background of acute and chronic

alcohol intoxication, [in Russian], Narcology, 17, 82-93, https://doi.org/10.25557/1682-8313.2018.06.82-93.

6. Gulyaeva, N. V., and Peregud, D. I. (2024) An intricated pas de deux of addicted brain and body is or-

chestrated by stress and neuroplasticity, Biochemistry (Moscow), 89, 1863-1867, https://doi.org/10.1134/

S0006297924110014.

7. Hoek, J. B., Cahill, A., and Pastorino, J. G. (2002) Alcohol and mitochondria: a dysfunctional relationship,

Gastroenterology, 122, 2049-2063, https://doi.org/10.1053/gast.2002.33613.

8. Li, S., Tan, H. Y., Wang, N., Zhang, Z. J., Lao, L., Wong, C. W., and Feng, Y. (2015) The role of oxidative stress

and antioxidants in liver diseases, Int. J. Mol. Sci., 16, 26087-26124, https://doi.org/10.3390/ijms161125942.

9. Prokopieva, V. D., and Vetlugina, T. P. (2023) Features of oxidative stress in alcoholism, [in Russian], Biomed.

Chem., 69, 83-96, https://doi.org/10.18097/PBMC20236902083.

10. Solonsky, A. V., Prokopieva, V. D., and Yarygina, E. G. (2018) Neuromorphological and molecular effects of eth-

anol, [in Russian], Sib. J. Psychiatry Narcol., 2, 28-32, https://doi.org/10.26617/1810-3111-2018-2(99)-28-32.

11. Ruiter-Lopez, L., Khan, M. A. S., Wang, X., and Song, B. J. (2025) Roles of oxidative stress and autophagy in

alcohol-mediated brain damage, Antioxidants (Basel), 14, 302, https://doi.org/10.3390/antiox14030302.

12. Andersson, U., Yang, H., and Harris, H. (2018) High-mobility group box 1 protein (HMGB1) operates as

an alarmin outside as well as inside cells, Semin. Immunol., 38, 40-48, https://doi.org/10.1016/j.smim.

2018.02.011.

13. Chikhirzhina, E. V., Polyaničko, A. M., and Starkova, T. Yu. (2020) Extranuclear functions of the non-histone

protein HMGB1 [in Russian], Cell Tissue Biol., 62, 716-725, https://doi.org/10.31857/S0041377120100016.

14. Tang, D., Kang, R., Zeh, H. J., and Lotze, M. T. (2023) The multifunctional protein HMGB1: 50 years of discovery,

Nat. Rev. Immunol., 23, 824-841, https://doi.org/10.1038/s41577-023-00894-6.

15. Frank, M. G., Weber, M. D., Watkins, L. R., and Maier, S. F. (2015) Stress sounds the alarmin: the role of

the danger-associated molecular pattern HMGB1 in stress-induced neuroinflammatory priming, Brain Behav.

Immun., 48, 1-7, https://doi.org/10.1016/j.bbi.2015.03.010.

16. Chen, R., Kang, R., and Tang, D. (2022) The mechanism of HMGB1 secretion and release, Exp. Mol. Med., 54,

91-102, https://doi.org/10.1038/s12276-022-00736-w.

17. Yu, M., Wang, H., Ding, A., Golenbock, D. T., Latz, E., Czura, C. J., Fenton, M. J., Tracey, K. J., and Yang, H.

(2006) HMGB1 signals through Toll-like Receptor (TLR) 4 and TLR2, Shock, 26, 174-179, https://doi.org/10.1097/

01.shk.0000225404.51320.82.

18. Ling, Y., Yang, Z. Y., Yin, T., Li, L., Yuan, W. W., Wu, H. S., and Wang, C. Y. (2011) Heparin changes the con-

formation of high-mobility group protein 1 and decreases its affinity toward receptor for advanced glycation

end products in vitro, Int. Immunopharmacol., 11, 187-193, https://doi.org/10.1016/j.intimp.2010.11.014.

19. Prantner, D., Nallar, S., and Vogel, S. N. (2020) The role of RAGE in host pathology and crosstalk between

RAGE and TLR4 in innate immune signal transduction pathways, FASEB J., 34, 15659-15674, https://doi.org/

10.1096/fj.202002136R.

20. Crews, F. T., Coleman, L. G., Jr., Macht, V. A., and Vetreno, R. P. (2024) Alcohol, HMGB1, and innate immune

signaling in the brain, Alcohol Res., 44, 04, https://doi.org/10.35946/arcr.v44.1.04.

21. Patel, S., Behara, R., Swanson, G., Forsyth, C., Voigt, R., and Keshavarzian, A. (2015) Alcohol and the intestine,

Biomolecules, 5, 2573-2588, https://doi.org/10.3390/biom5042573.

22. Bishehsari, F., Magno, E., Swanson, G., Desai, V., Voigt, R. M., Forsyth, C. B., and Keshavarzian, A. (2017) Alcohol

and gut-derived inflammation, Alcohol Res., 38, 163-171, https://doi.org/10.35946/arcr.v38.2.02.

23. Kawase, A., Irie, K., Matsuda, N., Takai, Y., Shimada, H., and Iwaki, M. (2022) Distinct roles of HMGB1 in the

regulation of P-glycoprotein expression in the liver and kidney of mice with lipopolysaccharide-induced inflam-

mation, Mol. Med. Rep., 26, 342, https://doi.org/10.3892/mmr.2022.12858.

24. Ge, X., Han, H., Desert, R., Das, S., Song, Z., Komakula, S. S. B., Chen, W., Athavale, D., Lantvit, D., and Nieto, N.

(2024) A protein complex of liver origin activates a pro-inflammatory program that drives hepatic and intes-

tinal injury in alcohol-associated liver disease, Cell. Mol. Gastroenterol. Hepatol., 18, 101362, https://doi.org/

10.1016/j.jcmgh.2024.05.010.

HMGB1 AND ALCOHOL DEPENDENCE 977

BIOCHEMISTRY (MOSCOW) Vol. 91 No. 6 2026

25. Laursen, T. L., Støy, S., Deleuran, B., Vilstrup, H., Grønbaek, H., and Sandahl, T. D. (2016) The damage-associated

molecular pattern HMGB1 is elevated in human alcoholic hepatitis, but does not seem to be a primary driver

of inflammation, APMIS, 124, 741-747, https://doi.org/10.1111/apm.12568.

26. Vannier, A. G. L., Wardwell, B., Fomin, V., PeBenito, A., Wolczynski, N., Piaker, S., Kedrin, D., Chung, R. T.,

Schaefer, E., Goodman, R., Patel, S. J., and Luther, J. (2021) Serum HMGB1 associates with liver disease and

predicts readmission and mortality in patients with alcohol use disorder, Alcohol, 95, 37-43, https://doi.org/

10.1016/j.alcohol.2021.05.003.

27. Rodriguez De Fonseca, F., Medina-Paz, F., Sapozhnikov, M., Hurtado-Guerrero, I., Rubio, Letici., Martin-

de-Las-Heras, S., Requena-Ocaña, N., Flores-López, M., Del Mar Fernández-Arjona, M., Rivera, P., Serrano, A.,

Serrano, P., Zapico, S. C., and Suárez, J. (2024) Plasma concentrations of high mobility group box 1 proteins

and soluble receptors for advanced glycation end-products are relevant biomarkers of cognitive impairment in

alcohol use disorder: a pilot study, Toxics, 12, 190, https://doi.org/10.3390/toxics12030190.

28. Tolmacheva, A. S., Bobrik, D. V., Melamud, M. M., Efremov, I. S., Nevinsky, G. A., Buneva, V. N., Akhmetova, E. A.,

Asadullin, A. R., and Ermakov, E. A. (2025) High-mobility group protein B1 (HMGB1) level in plasma of al-

cohol use disorder patients in withdrawal state, Middle East Curr. Psychiatry, 32, 41, https://doi.org/10.1186/

s43045-025-00537-7.

29. Zigmond, A. S., and Snaith, R. P. (1983) The hospital anxiety and depression scale, Acta Psychiatr. Scand., 67,

361-370, https://doi.org/10.1111/j.1600-0447.1983.tb09716.x.

30. Dubinina, L. A., Rybakova, K. V., Rybakova, T. G., Bakman, I. Yu., Kisilev, A. S., Erychev, O. F., and Krupitsky, E. M.

(2012) Clinical and Socio-Psychological Criteria for Evaluating the Readaptation of Patients with Alcohol Depen-

dence at the Stages of Remission Formation: A Guide for Physicians, V. M. Bekhterev National Research Institute

of Psychiatry and Neurology, St. Petersburg, Russia.

31. Flannery, B. A., Volpicelli, J. R., and Pettinati, H. M. (1999) Psychometric properties of the Penn alcohol craving

scale, Alcohol Clin. Exp. Res., 23, 1289-1295, https://doi.org/10.1111/j.1530-0277.1999.tb04349.x.

32. Crews, F. T., Qin, L., Sheedy, D., Vetreno, R. P., and Zou, J. (2013) High mobility group box 1/Toll-like receptor

danger signaling increases brain neuroimmune activation in alcohol dependence, Biol. Psychiatry, 73, 602-612,

https://doi.org/10.1016/j.biopsych.2012.09.030.

33. Coleman, L. G., Jr., Zou, J., and Crews, F. (2017) Microglial-derived miRNA Let-7 and HMGB1 contribute to eth-

anol-induced neurotoxicity via TLR7, J. Neuroinflammation, 14, 22, https://doi.org/10.1186/s12974-017-0799-4.

34. Airapetov, M. I., Eresko, S. O., Bychkov, E. R., Lebedev, A. A., and Shabanov, P. D. (2021) Expression of the

HMGB1 gene changes in the striatum and amygdala of rat brain during long-term alcoholization and ethanol

withdrawal [in Russian], Biomed. Chem., 67, 95-99, https://doi.org/10.18097/PBMC20216701095.

35. Wang, H., Yang, H., Czura, C. J., Sama, A. E., and Tracey, K. J. (2001) HMGB1 as a late mediator of lethal sys-

temic inflammation, Am. J. Respir. Crit. Care Med., 164, 1768-1773, https://doi.org/10.1164/ajrccm.164.10.2106117.

36. Banks, W. A., Hansen, K. M., Erickson, M. A., and Crews, F. T. (2023) High-mobility group box 1 (HMGB1) crosses

the BBB bidirectionally, Brain Behav. Immun., 111, 386-394, https://doi.org/10.1016/j.bbi.2023.04.018.

37. Shukla, S., and Hsu, C. L. (2025) Alcohol use disorder and the gut-brain axis: a narrative review of the

role of gut microbiota and implications for treatment, Microorganisms, 13, 67, https://doi.org/10.3390/

microorganisms13010067.

38. Leclercq, S., de Timary, P., Delzenne, N. M., and Stärkel, P. (2017) The link between inflammation, bugs, the

intestine and the brain in alcohol dependence, Transl. Psychiatry, 7, e1048, https://doi.org/10.1038/tp.2017.15.

39. Didkovsky, N. A., Malashenkova, I. K., Ogurtsov, D. P., Krynskiy, S. A., and Babaeva, L. A. (2018) Neuroimmune

mechanisms of alcoholism [in Russian], Allergolog, Immunol., 19, 215-221.

40. Ru, Y., Jia, J., Yang, X., Ba, X., Yan, J., Zhang, H., and Zhang, Y. (2025) From bugs to behavior: targeting the

gut-brain interface for addiction therapeutics, Gut Microbes, 17, 2569740, https://doi.org/10.1080/19490976.

2025.2569740.

41. Banks, W. A. (2015) The blood-brain barrier in neuroimmunology: tales of separation and assimilation, Brain

Behav. Immun., 44, 1-8, https://doi.org/10.1016/j.bbi.2014.08.007.

42. Erickson, E. K., Grantham, E. K., Warden, A. S., and Harris, R. A. (2019) Neuroimmune signaling in alcohol use

disorder, Pharmacol. Biochem. Behav., 177, 34-60, https://doi.org/10.1016/j.pbb.2018.12.007.

43. Nutt, D., Hayes, A., Fonville, L., Zafar, R., Palmer, E. O. C., Paterson, L., and Lingford-Hughes, A. (2021) Alcohol

and the brain, Nutrients, 13, 3938, https://doi.org/10.3390/nu13113938.

44. Orio, L., Anton, M., Rodriguez-Rojo, I. C., Correas, A., García-Bueno, B., Corral, M., Rodríguez de Fonseca, F.,

García-Moreno, L. M., Maestú, F., and Cadaveiraet, F. (2018) Young alcohol binge drinkers have elevated

blood endotoxin, peripheral inflammation and low cortisol levels: neuropsychological correlations in women,

Addict. Biol., 23, 1130-1144, https://doi.org/10.1111/adb.12543.

VETLUGINA et al.978

BIOCHEMISTRY (MOSCOW) Vol. 91 No. 6 2026

45. Coleman, L. G., Jr., Zou, J., Qin, L., and Crews, F. T. (2018) HMGB1/IL-1beta complexes regulate neuroimmune

responses in alcoholism, Brain Behav. Immun., 72, 61-77, https://doi.org/10.1016/j.bbi.2017.10.027.

46. Yang, H., Wang, H., and Andersson, U. (2020) Targeting inflammation driven by HMGB1, Front. Immunol., 11,

484, https://doi.org/10.3389/fimmu.2020.00484.

47. Ghosh, D., Singh, A., Kumar, A., and Sinha, N. (2022) High mobility group box 1 (HMGB1) inhibition attenuates

lipopolysaccharide-induced cognitive dysfunction and sickness-like behavior in mice, Immunol. Res., 70, 633-643,

https://doi.org/10.1007/s12026-022-09295-8.

48. Zhu, C. S., Wang, W., Qiang, X., Chen, W., Lan, X., Li, J., and Wang, H. (2021) Endogenous regulation and

pharmacological modulation of sepsis-induced HMGB1 release and action: an updated review, Cells, 10, 2220,

https://doi.org/10.3390/cells10092220.

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.